Fundamentals of fragrance chemistry 9783527819768, 3527819762

Table of contents :
Preface xi Introduction xiii 1 The Structure of Matter 1 The Route to the Atomic Theory 1 The Atomic Theory, Atomic Number, and Atomic Weight 4 Atomic Structure 7 Isotopes 8 The Electronic Structure of Atoms 9 Electronic Structure of Transition Metals 11 Hybridisation of Orbitals 11 Chemical Bonding, Ions, Cations, Anions, and Molecules 12 Review Questions 16 2 Carbon 1 - Hydrocarbons 17 Ethane: Conformational Isomers 17 Alkanes: Structural Isomers 20 Alkenes: Geometric Isomers 22 Alkynes 26 Cyclic Structures 26 Polycyclic Structures 28 Greek Letters 30 Aromatic Rings 31 Stereoisomerism 33 Rules for Hydrocarbon Nomenclature 36 Quick Rules for Isomers 37 Stereoisomers 37 Review Questions 38 3 Carbon 2 - Heteroatoms 39 Hydrogen Bonding 39 Alcohols 40 Phenols 43 Ethers 44 Aldehydes 45 Ketones 46 Carboxylic Acids 47 Esters 49 Acid Anhydrides and Chlorides 50 Acetals and Ketals 50 Peroxy Compounds 52 Nitrogen-Amines and Ammonium Salts 53 Nitrogen-Imines, Schiff's Bases, and Enamines 54 Nitrogen-Amides/Peptides 55 Nitrogen-Nitriles 56 Nitrogen-Nitro Compounds 57 Sulfur 58 Heterocyclic Compounds 60 Review Question 66 4 States of Matter 67 Solids 67 Liquids 71 Gases 71 Phase Changes 71 Two-Phase Systems 73 Solubility 74 Surfactants 75 Emulsions 79 Micelles 81 Detergency 81 Bilayers 82 Colloids 84 Review Questions 84 5 Separation and Purification 85 Distillation 85 Sublimation 93 Crystallisation 93 Solvent Extraction 94 Recent Developments in Commercial Extraction of Natural Fragrance Ingredients 95 Chromatography 96 Paper Chromatography 98 Thin Layer Chromatography 98 Column Chromatography 99 High Performance Liquid Chromatography 100 Gas Chromatography 100 Review Questions 105 6 Analysis 107 Physical Methods of Analysis 108 Density 108 Melting Point 108 Boiling Point 108 Refractive Index 109 Optical Rotation 109 Flashpoint 109 Viscosity 109 Colour 109 Chemical Methods of Analysis 110 Titration 110 Acid Content 111 Base Content 111 Peroxide Content 111 Ester Value 111 Aldehyde/Ketone Content 112 Phenol Content 112 Chemical Oxygen Demand (COD) 112 Water Content 112 Atomic Absorption 113 Spectroscopic Methods of Analysis 113 Ultraviolet (UV) 114 Infrared (IR) 118 Nuclear Magnetic Resonance (NMR) 120 Mass Spectrometry (MS) 124 Gas Chromatography-Mass Spectrometry (GC-MS) 127 Eugenol as an Example of Spectroscopic Techniques 127 Quality Control 131 Review Questions 132 7 Chemical Reactivity 133 The Three Laws of Thermodynamics 133 Free Energy 135 Chemical Reactions 136 The Principle of Microscopic Reversibility and Chemical Equilibrium 137 Reaction Profiles 138 Catalysts 140 Types of Organic Reactions 140 Review Questions 145 8 Chemistry and Perfume 1: Acid/Base Reactions 147 Acids and Bases 147 Strong and Weak 149 pH 150 Electrophiles and Nucleophiles 152 Esterification and Ester Hydrolysis 154 The Aldol Reaction and Aldol Condensation 155 Acetals and Ketals 158 Schiff's Bases and Enamines 160 Nitriles 161 Alcohol Dehydration 162 Acid-Catalysed Addition to Olefins 163 The Michael Reaction 164 The Grignard Reaction 165 The Friedel-Crafts Reaction 167 Electrophilic Substitutions in Aromatic Molecules 168 Review Questions 170 9 Oxidation and Reduction Reactions 171 Review Questions 185 10 Perfume Structure 187 Notes, Chords, and Discords 187 Ingredients 187 Odour Families and Top, Middle, and Base Notes 188 Persistence/Tenacity 191 Threshold 192 Impact 192 Radiance/Bloom 193 Physical and Chemical Factors 194 Review Questions 196 11 Chemistry in Consumer Goods 197 Introduction 197 Acids in Consumer Goods 198 Bases in Consumer Goods 199 Nucleophiles in Consumer Goods 200 Oxidants in Consumer Goods 201 Reductants in Consumer Goods 202 Surfactants in Consumer Goods 204 Chelating Agents in Consumer Goods 205 Photoactive Agents in Consumer Goods 206 Antibacterial Agents in Consumer Goods 207 Other Reactive Ingredients in Consumer Goods 208 Types of Consumer Goods 209 Fine Fragrance 209 Cosmetics and Toiletries 210 Personal Wash 210 Laundry 211 Household 212 Review Questions 214 12 The Chemistry of Living Organisms 215 Molecular Recognition 215 Classes of Natural Chemicals 218 Carbohydrates 218 Nucleic Acids 221 Lipids 223 Proteins 225 Toxicity and Product Safety 230 Review Questions 239 13 The Mechanism of Olfaction 243 The Role of Olfaction in Biology 243 The Organs Used in Olfaction 244 The Process of Olfaction 246 Transport to the Receptors 246 The Receptor Event 247 The Combinatorial Nature of Olfaction 249 The Perception of Odour 252 Review Questions 256 14 Natural Fragrance Ingredients 257 Why Does Nature Produce Odorous Chemicals? 257 Basic Principles of Biosynthesis: Enzymes and Cofactors 258 General Pattern of Biosynthesis of Secondary Metabolites 261 Polyketide Biosynthesis 262 Lipid Biosynthesis 263 The Shikimic Acid Pathway 265 Terpenoids 267 Degradation Products 277 Malodours 279 Review Questions 281 15 Synthetic Fragrance Ingredients 283 Why the Industry Uses Synthetic Fragrance Ingredients? 283 The Economics of Fragrance Ingredient Manufacture 284 Production of Fragrance Ingredients from Polyketides and Shikimates 288 Terpenoid Production 290 Production of Fragrance Ingredients from Petrochemicals 302 What Is Required of a Fragrance Ingredient? 320 How Novel Fragrance Ingredients Are Designed? 322 Review Questions 326 16 Chemical Information 329 How New Chemical Information Is Generated and Published? 329 Patents 329 Reviews and Books 331 Abstracts 331 How to Find Chemical Information? 333 17 Towards a Sustainable Future 335 What Is Sustainability? 335 Commercial Feasibility 337 Safety in Use 337 Natural Fragrance Ingredients 340 Synthetic Fragrance Ingredients 341 Synthetic Fragrance Ingredients A: Use of By-Products 341 Synthetic Fragrance Ingredients B: Environmental Impact 342 Synthetic Fragrance Ingredients C: Biotechnology 344 Synthetic Fragrance Ingredients D: Finding the Balance 345 The Symrise Route 347 The Takasago Route 347 The BASF Route 348 Menthol Sustainability 349 Pro-fragrances 351 Social and Health Factors 353 Understanding Olfaction 353 Malodour Management 354 Health and Well-Being 355 Information 356 Conclusion 356 Answers to Review Questions 357 Glossary 371 Bibliography 379 Index 381

Citation preview

Fundamentals of Fragrance Chemistry

­Fundamentals of Fragrance Chemistry Charles S. Sell

Author Charles S. Sell

Aldington, Ashford, Kent United Kingdom

All books published by Wiley‐VCH are carefully produced. Nevertheless, authors, editors, and publisher do not warrant the information contained in these books, including this book, to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate. Library of Congress Card No.: applied for British Library Cataloguing‐in‐Publication Data

A catalogue record for this book is available from the British Library. Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at . © 2019 Wiley‐VCH Verlag GmbH & Co. KGaA, Boschstr. 12, 69469 Weinheim, Germany All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form – by photoprinting, microfilm, or any other means – nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Print ISBN: 978‐3‐527‐34577‐9 ePDF ISBN: 978‐3‐527‐81976‐8 ePub ISBN: 978‐3‐527‐81978‐2 Cover Design  Adam-Design, Weinheim, Germany Typesetting  SPi Global, Chennai, India Printing and Binding

Printed on acid‐free paper 10 9 8 7 6 5 4 3 2 1

v

Contents Preface  xi Introduction  xiii 1 The Structure of Matter  1

The Route to the Atomic Theory  1 The Atomic Theory, Atomic Number, and Atomic Weight  4 Atomic Structure  7 Isotopes  8 The Electronic Structure of Atoms  9 Electronic Structure of Transition Metals  11 Hybridisation of Orbitals  11 Chemical Bonding, Ions, Cations, Anions, and Molecules  12 Review Questions  16

2 Carbon 1 – Hydrocarbons  17

Ethane: Conformational Isomers  17 Alkanes: Structural Isomers  20 Alkenes: Geometric Isomers  22 Alkynes  26 Cyclic Structures  26 Polycyclic Structures  28 Greek Letters  30 Aromatic Rings  31 Stereoisomerism  33 Rules for Hydrocarbon Nomenclature  36 Quick Rules for Isomers  37 Stereoisomers  37 Review Questions  38

3 Carbon 2 – Heteroatoms  39

Hydrogen Bonding  39 Alcohols  40 Phenols  43 Ethers  44 Aldehydes  45

vi

Contents

Ketones  46 Carboxylic Acids  47 Esters  49 Acid Anhydrides and Chlorides  50 Acetals and Ketals  50 Peroxy Compounds  52 Nitrogen–Amines and Ammonium Salts  53 Nitrogen–Imines, Schiff ’s Bases, and Enamines  54 Nitrogen–Amides/Peptides  55 Nitrogen–Nitriles  56 Nitrogen–Nitro Compounds  57 Sulfur  58 Heterocyclic Compounds  60 Review Question  66 4 States of Matter  67

Solids  67 Liquids  71 Gases  71 Phase Changes  71 Two‐Phase Systems  73 Solubility  74 Surfactants  75 Emulsions  79 Micelles  81 Detergency  81 Bilayers  82 Colloids  84 Review Questions  84

5 Separation and Purification  85

Distillation  85 Sublimation  93 Crystallisation  93 Solvent Extraction  94 Recent Developments in Commercial Extraction of Natural Fragrance Ingredients  95 Chromatography  96 Paper Chromatography  98 Thin Layer Chromatography  98 Column Chromatography  99 High Performance Liquid Chromatography  100 Gas Chromatography  100 Review Questions  105

Contents

6 Analysis 107

Physical Methods of Analysis  108 Density  108 Melting Point  108 Boiling Point  108 Refractive Index  109 Optical Rotation  109 Flashpoint  109 Viscosity  109 Colour  109 Chemical Methods of Analysis  110 Titration  110 Acid Content  111 Base Content  111 Peroxide Content  111 Ester Value  111 Aldehyde/Ketone Content  112 Phenol Content  112 Chemical Oxygen Demand (COD)  112 Water Content  112 Atomic Absorption  113 Spectroscopic Methods of Analysis  113 Ultraviolet (UV)  114 Infrared (IR)  118 Nuclear Magnetic Resonance (NMR)  120 Mass Spectrometry (MS)  124 Gas Chromatography–Mass Spectrometry (GC–MS)  127 Eugenol as an Example of Spectroscopic Techniques  127 Quality Control  131 Review Questions  132

7 Chemical Reactivity  133

The Three Laws of Thermodynamics  133 Free Energy  135 Chemical Reactions  136 The Principle of Microscopic Reversibility and Chemical Equilibrium  137 Reaction Profiles  138 Catalysts  140 Types of Organic Reactions  140 Review Questions  145

8 Chemistry and Perfume 1: Acid/Base Reactions  147

Acids and Bases  147 Strong and Weak  149 pH  150

vii

viii

Contents

Electrophiles and Nucleophiles  152 Esterification and Ester Hydrolysis  154 The Aldol Reaction and Aldol Condensation  155 Acetals and Ketals  158 Schiff ’s Bases and Enamines  160 Nitriles  161 Alcohol Dehydration  162 Acid‐Catalysed Addition to Olefins  163 The Michael Reaction  164 The Grignard Reaction  165 The Friedel–Crafts Reaction  167 Electrophilic Substitutions in Aromatic Molecules  168 Review Questions  170 9 Oxidation and Reduction Reactions  171

Review Questions  185

10 Perfume Structure  187

Notes, Chords, and Discords  187 Ingredients  187 Odour Families and Top, Middle, and Base Notes  188 Persistence/Tenacity  191 Threshold  192 Impact  192 Radiance/Bloom  193 Physical and Chemical Factors  194 Review Questions  196

11 Chemistry in Consumer Goods  197

Introduction  197 Acids in Consumer Goods  198 Bases in Consumer Goods  199 Nucleophiles in Consumer Goods  200 Oxidants in Consumer Goods  201 Reductants in Consumer Goods  202 Surfactants in Consumer Goods  204 Chelating Agents in Consumer Goods  205 Photoactive Agents in Consumer Goods  206 Antibacterial Agents in Consumer Goods  207 Other Reactive Ingredients in Consumer Goods  208 Types of Consumer Goods  209 Fine Fragrance  209 Cosmetics and Toiletries  210 Personal Wash  210 Laundry  211

Contents

Household  212 Review Questions  214 12 The Chemistry of Living Organisms  215

Molecular Recognition  215 Classes of Natural Chemicals  218 Carbohydrates  218 Nucleic Acids  221 Lipids  223 Proteins  225 Toxicity and Product Safety  230 Review Questions  239

13 The Mechanism of Olfaction  243

The Role of Olfaction in Biology  243 The Organs Used in Olfaction  244 The Process of Olfaction  246 Transport to the Receptors  246 The Receptor Event  247 The Combinatorial Nature of Olfaction  249 The Perception of Odour  252 Review Questions  256

14 Natural Fragrance Ingredients  257

Why Does Nature Produce Odorous Chemicals?  257 Basic Principles of Biosynthesis: Enzymes and Cofactors  258 General Pattern of Biosynthesis of Secondary Metabolites  261 Polyketide Biosynthesis  262 Lipid Biosynthesis  263 The Shikimic Acid Pathway  265 Terpenoids  267 Degradation Products  277 Malodours  279 Review Questions  281

15 Synthetic Fragrance Ingredients  283

Why the Industry Uses Synthetic Fragrance Ingredients?  283 The Economics of Fragrance Ingredient Manufacture  284 Production of Fragrance Ingredients from Polyketides and Shikimates  288 Terpenoid Production  290 Production of Fragrance Ingredients from Petrochemicals  302 What Is Required of a Fragrance Ingredient?  320 How Novel Fragrance Ingredients Are Designed?  322 Review Questions  326

ix

x

Contents

16 Chemical Information  329

How New Chemical Information Is Generated and Published?  329 Patents  329 Reviews and Books  331 Abstracts  331 How to Find Chemical Information?  333

17 Towards a Sustainable Future  335

What Is Sustainability?  335 Commercial Feasibility  337 Safety in Use  337 Natural Fragrance Ingredients  340 Synthetic Fragrance Ingredients  341 Synthetic Fragrance Ingredients A: Use of By‐Products  341 Synthetic Fragrance Ingredients B: Environmental Impact  342 Synthetic Fragrance Ingredients C: Biotechnology  344 Synthetic Fragrance Ingredients D: Finding the Balance  345 The Symrise Route  347 The Takasago Route  347 The BASF Route  348 Menthol Sustainability  349 Pro‐fragrances  351 Social and Health Factors  353 Understanding Olfaction  353 Malodour Management  354 Health and Well‐Being  355 Information  356 Conclusion  356

Answers to Review Questions  357 Glossary  371 Bibliography  379 Index  381

xi

Preface Chemistry can be a difficult subject and may seem far removed from the glitter of the fragrance business. However, it is the essential science behind the latter. Some chapters, especially the first, contain more of the basic principles of chemistry and may seem less relevant than those with regard to fragrance at first sight. But these basic concepts are important because they lay the groundwork on which fragrance chemistry is founded. The reader is advised to bear with them, study them, and refer to them when appropriate while reading the more obviously relevant chapters. Thoughts and opinions expressed in this book, especially in Chapter 17, are those of the author and hence are not necessarily in agreement with those of the industry, the publisher, or any individual company. 2 January 2019

Aldington, Kent, England

xiii

Introduction To the layman, the world of perfumery conjures up images of glamour, dreams, romance, expensive oils extracted from exotic plants, and so on. The names that spring to mind are those of the great perfumers and fashion houses such as Jean Patou, Francois Coty, Chanel, Christian Dior, and so on. These names and images are part of our fascinating industry, but, in addition, behind all of this allure is a modern industry with a strong scientific basis, and the core science is chemistry. Ernest Beaux, the perfumer who created Chanel No. 5, said, ‘One has to rely on chemists to find new aroma chemicals creating new, original notes. In perfumery, the future lies primarily in the hands of chemists’. And his words are as true today as in 1921 when he created his famous masterpiece. Many Nobel Prize winners mentioned fragrance chemistry in their prizewinning lectures. It is also significant that the times of strongest growth of a fragrance company are associated, more often than not, with the presence of a first‐rate, practicing chemist in a senior position. Thus, to the names of the perfumers, we can add great chemists such as Yves‐Rene Naves (Givaudan), Ernst Theimer (IFF), Leopold Ruziča (Firmenich), Ernst Günther (Fritzsche, Dodge, and Olcott), Ernest Polak (Polak’s Frutal Works), Paul José Teisseire (Roure Bertrand Dupont), Günther Ohloff (Firmenich), and George Fráter (Givaudan) as key figures in the history of perfumery. Not everyone needs to be a chemist of such a calibre as these, but for all of those individuals working in the fragrance business and in the consumer goods industries that it serves, knowledge of chemistry is invaluable in understanding how fragrance is produced, how it works, and the factors that control its performance in products. Perfume molecules are compounds of carbon and hence come under the general heading of organic chemistry. Our bodies are also composed of organic chemicals and so are most of the components of consumer goods such as soaps and detergents. This book therefore concentrates on those aspects of organic chemistry, which are of particular importance to the fragrance industry. It is intended for those who have little or no previous training in chemistry and who would like to know enough in order to improve their understanding of perfume and its interactions with the wide variety of products in which it is used. Chapter  1 covers the nature of matter, the building blocks from which it is made, and how these building blocks are held together. Chapter 2 describes the basic concepts of how carbon atoms join together to form the backbones of organic chemicals. It also describes the various shorthand

xiv

Introduction

methods that chemists use to indicate the composition of materials and the structure of their molecules and so will enable participants to make sense of the ‘fried eggs and spiders’ that chemists draw. It also gives an insight into the language that chemists use and the names they give chemicals. Chapter 3 introduces organic materials that contain oxygen, nitrogen, or sulfur as well as carbon and hydrogen. The vast majority of fragrance ingredients fall into this class. Chapter 4 describes the three states of matter and how one may be converted into another. It also describes how surface‐active agents behave at interfaces between immiscible liquids and this behaviour leads on to cover the basis of detergency and the structure of mammalian cell walls. In order to analyse and manipulate materials, it is important to be able to isolate them from mixtures and obtain them in pure form. The various methods by which purification can be achieved both for analytical and manufacturing purposes are described in Chapter 5. Chapter 6 concerns the methods used to identify and characterise perfume molecules, an activity of vital importance for everything from purchasing of raw materials to studying the fate of fragrance materials after use. Chapter 7 outlines the factors controlling chemical reactivity and provides a basis for understanding of the chemistry to be described in the subsequent chapters. The chemistry of acids and bases and the relevance of this chemistry to perfume chemistry is the subject of Chapter 8, while Chapter 9 covers oxidation and reduction reactions. Chapter 10 describes the structure of a fragrance and the effects of this on performance in consumer goods. Chapter  11 discusses the chemical interactions that occur between perfume ingredients and the other materials present in consumer goods. Chapter 12 gives a very basic introduction to the chemistry of living organisms, and this paves the way for a discussion of the mechanism of olfaction in Chapter 13. Chapter 14 moves on to describe the variety of chemicals made by plants and animals and, in particular, those that constitute the essential oils and other fragrant extracts. Chapter 15 follows on by describing how we copy and improve upon the perfume ingredients of nature in order to provide the perfumers with the palette available to them today. Chapter  16 provides a brief introduction to chemical literature, and it also contains a list of recommended reading. Thus, it serves as a guide for the reader who wishes to pursue the subject in more detail. The last chapter, Chapter 17, surveys the trends that are likely to affect the industry in the future and how we can respond to these to make the industry as sustainable as possible.

1

1 The Structure of Matter ­The Route to the Atomic Theory Chemistry is a subject of vital importance to human society. We even measure the progress of civilisation by the chemical technology that our ancestors possessed at various stages in history. Thus, the earliest phase of civilisation is known as the Stone Age, when humans used readily available materials such as stone to form tools. In chemical terms, the stone was used as it was found. The only manipulation was to shape it by physical means into knives, axes, and so on. The discovery of bronze moved civilisation forward significantly and gave birth to the Bronze Age. As an example of this technological advancement, bronze axes could be made with much more acute angles at the cutting edge of the blade than can stone axes, and so fewer strokes were required to cut through a tree trunk. Now chemistry was involved, since ores such as malachite had to undergo a chemical conversion to release the copper metal that they contained. Heating the ore to a high temperature brought about this chemical change. The temperature required to release iron from its ores, such as haematite, is even higher, so it was not until furnace technology had reached the required level that the Iron Age began. Chemistry is important to all industries to some extent, but to perfumery, it is absolutely central. The odorous substances that produce the sensation of smell, whether of natural or synthetic origin, are chemicals. The receptors in our noses that perceive them are chemicals. Smell begins with the process of chemical recognition of the odorant by the olfactory receptor, and therefore smell is very much a chemical sense. To understand fragrance perception, we must understand chemistry. The products into which perfumes are incorporated are also composed of chemicals and chemical interactions can occur between the perfume and the product. Thus, in order to understand the interaction of perfume with products such as soaps and detergents, we must understand chemistry. Chemistry is very much a practical science and people were practising it long before theories about the nature of matter and of these chemical processes were developed. Metallurgy, which is one branch of chemistry, started in the Nile Delta in ancient Egypt. Because of the colour of the rich alluvial soil, the Greeks knew this region as ‘The Black Country’. Metallurgy was considered to be the art of Egypt, the Black Country, and hence became to be known as the ‘Black Art’. Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

2

1  The Structure of Matter

The debate about the nature of matter began in Greece around the fourth century. Democritus (460–370 bce) and Epicurus (341–270 bce) argued that matter was made of small indivisible particles that they called atoms. The word atom is derived from the Greek verb τομεω (tomeo), which means ‘I cut’, and ατομος (atomos), meaning ‘uncuttable’ or ‘indivisible’. On the other hand, Empedocles (c. 450 bce) and later Aristotle (384–322 bce) believed that matter was continuous and composed of four basic ingredients or elements: earth, air, fire, and water. In order to distinguish living from inanimate matter, Aristotle invoked a fifth element or quintessence, which he called spirit. The legacy of his erroneous theory still survives in our language today. Adherents of the Aristotelian philosophy believed that by heating plant material, they were removing the spirit (or quintessence) of the plant and so the oil they obtained was called the quintessential (later shortened to essential) oil. Similarly, we refer to other distillates, such as whisky, gin, or brandy, as spirits. With these two philosophical schools came the first theories of how the sense of smell worked. Epicurus believed that odours were made up of atoms that travelled through the air from the source to the nose. Smooth, rounded atoms gave rise to sweet smells and pointed ones to sharp odours. Aristotle believed that odours radiated from the source to the nose, just as heat radiated from the sun to the earth. In ce 50, Dioscorides produced a book called De Materia Medica in which he listed all the known facts about herbal medicines. The compilation of what was known about the physical universe gained further momentum in ce 866 when Razi began a systematic collection of facts. Around ce 1000 the Arabs invented distillation, which meant that liquids could be produced in a pure state. New solvents for distillation such as alcohol were used in addition to water and therefore allowed for a great increase in the ability to manipulate materials. For instance, the odorous components of plants had previously been capable only of being extracted into fats and oils through the process of enfleurage (see Chapter 4, for details). With distillation, the volatile oils could be extracted directly from the plant material. The availability of alcohol as a solvent meant that the odorous principles could also be extracted from the fatty concretes by dissolving them in ethanol. (Again, more detail will be found in Chapter 4.) The alchemists of medieval Europe searched for a method to turn base metals into gold. We now know that this is a futile endeavour but, in their work, they built up a fund of experimental evidence about interconversions of substances. In the thirteenth century, Roger Bacon, an English Franciscan friar and scientist, laid the foundations of what we now call ‘The Scientific Method’. Scientific method uses five steps in developing theories about the physical universe. These steps are observe, correlate, postulate, test, and revise. Thus, true science begins with the observation of facts. It then seeks to find relationships between them and to devise theories to account for them. The next step is to devise experiments that will test the theories. If the theory passes the test, it remains valid. If not, the theory must be abandoned or revised until a new theory is developed that passes all known practical tests. We must always remember that in science nothing is ever established beyond doubt; every theory, every model is only accepted, while no exceptions are known. The possibility always exists of an inconvenient fact turning up and forcing us to revise our theories again – hence

­The Route to the Atomic Theor

the saying ‘The exception proves the rule’, the verb prove here being used in the sense of tests. Armed with Bacon’s powerful scientific method, the scientists of the Age of Enlightenment were able to start interpreting the growing body of facts in a more rational way, and, in one sense, the opposing theories of Democritus and Aristotle began to come together to form a more accurate picture of the universe. Democritus had seen each type of matter as being composed of characteristic particles or atoms. Aristotle saw different forms of matter as being composed of combinations of four basic elements. Gradually, a new picture began to emerge in which atoms of a larger number of elements came together in different ways to form other substances. As an illustration, let us look at some chemical relationships between iron and sulfur. These two substances appear in various guises, and so the suspicion arose that they might be elements, basic building blocks of matter. Heating iron ore produces iron, which can be purified by heating to burn off some of the contaminants present and then pouring the molten iron away from the more refractory minerals around it. Sulfur was collected from the rims of volcanoes, hence its former name of brimstone. If iron powder and sulfur are mixed together, they can easily be separated again with a magnet. However, if they are heated together, they form a new substance that turns out to be identical to the mineral known as pyrites or ‘fool’s gold’. Burning sulfur in air produces an irritant gas that is referenced in Homer’s Odyssey, when Odysseus burnt sulfur in his house to cleanse it from the traces of those who had occupied it during his famous return journey from Troy. If pyrites is burnt in air, we drive off the same acidic gas and obtain iron. So iron and sulfur can be chemically combined to form a new substance. However, they are not lost, and both can be recovered from the combination. Therefore, we can conclude that they are both elements, as opposed to pyrites, which is a compound of iron and sulfur. Of course, another element, oxygen, is involved in the above conversions. However, oxygen is difficult to characterise and it was not identified as an element until much later. In this way, a number of elements were identified and then laws about the way they combined began to be discovered. The first was the law of definite proportions that was first defined by J.B. Richter in 1792. This law states, ‘The ratios of the weights of elements which are present in a given chemical compound are constant’. So, taking our example of pyrites, the ratio of the weights of iron to sulfur in any given sample will be the same. Then, came the law of equivalent proportions, which states: ‘The proportions in which two elements separately combine with the same weight of a third element are also the proportions in which the first two elements combine together’. For example, if we find that 3 g of carbon combined with 1 g of hydrogen to form methane and 3 g of carbon combined with 8 g of oxygen to form carbon dioxide, then we can predict that water, a compound of hydrogen and oxygen, will contain 8 g of oxygen for every 1 g of hydrogen. Mixtures are combinations of substances from which the components can be separated by purely physical means. Elements are pure substances that cannot be broken down further into other chemicals. They are made up of atoms, which are the smallest possible pieces of that element that will still retain its chemical

3

4

1  The Structure of Matter

properties. Chemical compounds are substances that are composed of atoms of different elements but in which the atoms are held together by a force known as a chemical bond. The smallest unit of a compound that still retains all the chemical properties of that compound is called a molecule.

­The Atomic Theory, Atomic Number, and Atomic Weight Consideration of these and other laws and observations led the English chemist John Dalton to develop his atomic theory in 1806. In this theory, Dalton proposed that the elements were composed of indivisible particles called atoms, each with a characteristic weight, and that chemical compounds were composed of atoms joined together in some way. The ability of atoms to join together is known as valence, and each type of atom has a specific number of valencies or combining power. In 1810, J.J. Berzelius observed that sometimes two elements could combine with each other in different ways. The weight ratios of the elements in these different compounds led him to define the law of multiple proportions that states: ‘When two elements combine to form more than one compound, the amounts of them which combine with a fixed amount of the other exhibit a simple multiple relation’. For example, iron can combine in two ways with oxygen to form two different oxides. In one of them 7 g of iron combines with 2 g of oxygen, and in the other 7 g of iron combines with 3 g of oxygen. So the ratio between the weights of oxygen in the two is 2 : 3. So, elements each seem to have a characteristic weight, known as the atomic weight, and also characteristic valencies. The atomic weights were first expressed in relation to that of hydrogen, the lightest element. Thus, if the weight of a hydrogen atom is defined as one unit, currently called the atomic mass unit or the Dalton, then helium has an atomic weight of 4, lithium 7, and so on. In 1819, the Swedish chemist J.J. Berzelius devised a convenient shorthand system for describing the elements by using the first, or first two, letters of their Latin names. Thus, hydrogen is symbolised by H, carbon by C, iron by Fe (for its Latin name ferrum), sodium by Na (for natrium), and so on. In 1869, the Russian chemist D.I. Mendeleyev noticed that if the known elements are arranged in order of their atomic weights, a pattern or periodicity about their chemical properties is shown. The periodic interval initially is 8. Thus, for example, the third element, lithium, has similar properties to the 11th, sodium; the fourth, beryllium, to the 12th, magnesium; and so on. The elements were assigned atomic numbers based on their places in this series. Thus, the lightest element, hydrogen, has an atomic number of 1; the next, helium 2; then lithium with 3; and so on. Mendeleyev laid this pattern out in tabular form, thus presenting us with the most complete piece of scientific information, which exists, the periodic table. So powerful is the periodic table that Mendeleyev was able to use it not only to predict the existence of elements unknown at the time but also to describe what their chemical properties would be like. A simple representation of the periodic table is shown in Figure 1.1. The elements are arranged left to right in order of their atomic numbers. The first row

­The Atomic Theory, Atomic Number, and Atomic Weigh IA

IIA

IIIB IVB VB

V1B VIIB

VIIIB

IB

IIB

IIIA IVA

VA

VIA VIIA VIIIA

H

He

Li

Be

B

C

N

O

F

Na

Mg

Al

Si

P

S

Cl

Ar

K

Ca

Br

Kr

Sc

Ti

V

Cr

Mn

Fe

Co

Ni

Cu

Zn

Ga

Ge

As

Se

Ne

Rb

Sr

Y

Zr

Nb

Mo

Tc

Ru

Rh

Pd

Ag

Cd

In

Sn

Sb

Te

I

Xe

Cs

Ba

#1

Hf

Ta

W

Re

Os

Ir

Pt

Au

Hg

Tl

Pb

Bi

Po

At

Rn

Fr

Ra

#2 #1

La

Ce

Pr

Nd

Pm

Sm

Eu

Gd

Tb

Dy

Ho

Er

Tm

Yb

Lu

#2

Ac

Th

Pa

U

Np

Pu

Am

Cm

Bk

Cf

Es

Fm

Md

No

Lr

Figure 1.1  The periodic table of the elements.

(or period) contains only two elements. The next two rows have eight elements each, and are followed by two rows of 18. The last two rows contain 32 elements each but are usually drawn as in the figure with two blocks of 15 elements shown separately in order to prevent the table becoming so wide as to be unwieldy. The column of elements in darker shaded boxes are the inert or noble gases, so called because of their very low chemical reactivity. The vertical columns are normally referred to as groups and the group numbers are shown in the bar across the top of the figure. The inert gases thus belong to group VIIIA. One of the things Mendeleyev had noticed was the similarity in chemical properties in each group. The elements of group VIIA, for instance are known as the halogens, or saltforming elements. Group IA are known as the alkali metals and are the most reactive metals. The next group is known as the alkaline earths. The alkali metals all form salts with the halogens in the ratio of one metal atom to one halogen atom, giving formulae of the type MX, where M represents the metal and X the halogen. Examples would include common salt or sodium chloride, NaCl. The alkali earths, on the other hand, form halide (or halogen) salts in which there are two halogen atoms for each metal, for instance magnesium bromide is MgBr2. The elements in lighter grey shaded boxes are non-metals, those in clear boxes are metals. The metals in group VIIIB are very interesting. The first row contains iron, cobalt, and nickel. These elements are all important in forming catalysts including natural catalysts such as cytochrome P450 and vitamin B12, which contain iron and cobalt, respectively. The other six metals in this group are known as the platinum metals (including platinum), and these are of considerable importance as catalysts in the manufacture of fragrance ingredients. The heaviest naturally occurring element is uranium, number 92. The transuranic elements, those with a higher atomic number than 92, are only formed in nuclear reactors and are unstable, breaking down quickly into lighter elements. A list of the elements is shown in Table 1.1. The table includes their names, symbols, atomic numbers, and atomic weights. For sake of completeness, all of the elements are shown in both Figure 1.1 and Table 1.1. This list is not intended to discourage the reader, as this book will concentrate on only a small number of

5

6

1  The Structure of Matter

Table 1.1  The elements. Atomic no. Name

Atomic Symbol weight

Atomic no. Name

Atomic Symbol weight

1

Hydrogen

H

1.0079

53

Iodine

I

126.9045

2

Helium

He

4.0026

54

Xenon

Xe

131.29

3

Lithium

Li

6.941

55

Cesium

Cs

132.9054

4

Beryllium

Be

9.01218

56

Barium

Ba

137.33

5

Boron

Be

10.81

57

Lanthanum

La

138.9055

6

Carbon

C

12.011

58

Cerium

Ce

140.12

7

Nitrogen

N

14.0067

59

Praseodymium Pr

140.9077

8

Oxygen

O

15.9994

60

Neodymium

Nd

144.2

9

Fluorine

F

18.9984

61

Promethium

Pm

145a)

10

Neon

Ne

20.179

62

Samarium

Sm

150.36

11

Sodium

Na

22.98977

63

Europium

Eu

151.96

12

Magnesium

Mg

24.305

64

Gadolinium

Gd

157.25

13

Aluminium

Al

26.98154

65

Terbium

Tb

158.9254

14

Silicon

Si

28.0855

66

Dysprosium

Dy

1262.5

15

Phosphorus

P

30.97376

67

Holmium

Ho

164.9304

16

sulfur

S

32.06

68

Erbium

Er

167.26

17

Chlorine

Cl

35.453

69

Thulium

Tm

168.9342

18

Argon

Ar

39.948

70

Ytterbium

Yb

173.04

19

Potassium

K

39.0983

71

Lutetium

Lu

174.967

20

Calcium

Ca

40.08

72

Hafnium

Hf

178.49

21

Scandium

Sc

44.9559

73

Tantalum

Ta

180.9479

22

Titanium

Ti

47.88

74

Tungsten

W

183.85

23

Vanadium

V

50.9415

75

Rhenium

Re

186.207

24

44 Chromium Cr

51.996

76

Osmium

Os

190.2

25

Manganese

Mn

54.938

77

Iridium

Ir

192.22

26

Iron

Fe

55.847

78

Platinum

Pt

195.08

27

Cobalt

Co

58.9332

79

Gold

Au

196.9665

28

Nickel

Ni

58.69

80

Mercury

Hg

200.59

29

Copper

Cu

63.546

81

Thallium

Tl

204.383

30

Zinc

Zn

65.38

82

Lead

Pb

207.2

31

Gallium

Ga

69.72

83

Bismuth

Bi

208.9804

32

Germanium

Ge

72.59

84

Polonium

Po

209a)

33

Arsenic

As

74.9216

85

Astatine

At

210a)

34

Selenium

Se

78.96

86

Radon

Rn

222a)

35

Bromine

Br

79.904

87

Francium

Fr

223a)

­Atomic Structur

Table 1.1  (Continued) Atomic no. Name

Atomic Symbol weight

Atomic no. Name

Atomic Symbol weight

36

Krypton

Kr

83.8

88

Radium

Ra

226.0254b)

37

Rubidium

Rb

85.4678

89

Actinium

Ac

227.0278b)

38

Strontium

Sr

87.62

90

Thorium

Th

232.0381b)

39

Yttrium

Y

88.9059

91

Protactinium

Pa

231.0359b)

40

Zirconium

Zr

91.22

92

Uranium

U

238.0289

41

Niobium

Nb

92.9064

93

Neptunium

Np

237.0482b)

42

Molybdenum Mo

95.94

94

Plutonium

Pu

244a)

43

Technetium

Tc

98a)

95

Americium

Am

243a)

44

Ruthenium

Ru

101.07

96

Curium

Cm

247a)

45

Rhodium

Rh

102.9055

97

Berkelium

Bk

247a)

46

Palladium

Pd

106.42

98

Californium

Cf

251a)

47

Silver

Ag

107.868

99

Einsteinium

Es

252a)

48

Cadmium

Cd

112.41

100

Fermium

Fm

257a)

49

Indium

In

114.82

101

Mendelevium

Md

258a)

50

Tin

Sn

118.69

102

Nobelium

No

259a)

51

Antimony

Sb

121.75

103

Lawrencium

Lr

260a)

52

Tellurium

Te

127.6

a) Mass number of longest‐lived isotope. b) Atomic weight of most commonly available long‐lived isotope.

these elements. One thing to note about the elements listed in Table 1.1 is that, for the majority of them, their atomic weights are close to whole numbers. This quality provides an important clue about the structure of the atom.

­Atomic Structure The structure of atoms was elucidated in the early part of the twentieth century. For the purposes of this book, we can assume that atoms are composed of three more fundamental particles, namely, protons, neutrons, and electrons. Protons and neutrons each have an atomic mass of 1 Da. Protons carry a positive electrical charge, and neutrons, as their name suggests, are neutral. Electrons carry one unit of negative electrical charge each and have no mass. Atoms have a structure rather like that of a planetary system. At the centre is a nucleus composed of neutrons and protons, and the electrons orbit around the nucleus similar to the way planets orbit around their stars. In order to maintain electrical neutrality, the number of electrons orbiting the nucleus equals the number of protons in the nucleus. The factor controlling the chemistry of an element is the number of

7

8

1  The Structure of Matter

protons in its nucleus. The simplest atom therefore, the hydrogen atom, has a nucleus containing one proton only. This proton is balanced by one electron. Since the electron has no mass and the proton has a mass of 1 Da, the hydrogen atom has an atomic mass, or atomic weight, of 1 Da. This fact is the case for most hydrogen atoms. However, some hydrogen atoms have one neutron also in their nucleus. The charge in the nucleus is still one positive charge, and so there is still one electron in orbit and the chemical properties are still those of hydrogen. However, the atomic weight is now 2 Da. When atoms exist with the same atomic number (i.e. the same number of protons) but with different atomic weights (i.e. different numbers of neutrons), we call them isotopes.

­Isotopes The word isotope comes from Greek and means ‘same place’. The atoms have the same place in chemistry as each other. The hydrogen isotope having a mass of 2 Da is known as deuterium, and the symbol D is often used to denote it. More properly, it should be identified by the symbol 2H, while 1H would then specify the more common isotope of hydrogen. Both isotopes will have the same chemical properties. It is also possible to have two neutrons in a hydrogen nucleus, and this isotope is called tritium, 3H. However, in this case, the nucleus is unstable and breaks down, or decays, into smaller fragments, emitting radiation in the process. Such unstable isotopes are called radioactive isotopes or radioisotopes because of the radiation they emit. In Table 1.1, some of the elements do not have an atomic weight like the others but are shown with the weight of the most stable or longest‐lived isotope, because such elements, like radium, are intrinsically unstable and undergo radioactive decay into lighter elements. Normal hydrogen contains a mixture of its three isotopes. In a natural sample, there are far fewer deuterium atoms than protium (as 1H is also known) and even fewer atoms of tritium. If we calculate the atomic weight based on a proton or neutron weighing 1.0000 Da, the result will be the average of some atoms with a weight of 2, some with a weight of 3, and most of them with a weight of 1. This average is why the atomic weight of hydrogen is shown as being 1.0079 in Table 1.1. Carbon is the element that concerns us most in perfumery. It has three isotopes, and all of these are important to us in different ways. The atomic number of carbon is 6, and so each atom of carbon has six electrons and six protons. The most common isotope, and hence the most important, has six neutrons in the nucleus. The atomic weight of carbon is therefore close to 12, 12.011 to be precise. Some carbon atoms have seven neutrons and therefore an atomic weight of 13. This is therefore known as 13C or carbon‐13 and is important in spectroscopy, as we will see in Chapter 5. If there are eight neutrons in a carbon nucleus, then it is designated 14C or carbon‐14. This isotope is unstable and therefore radioactive. This isotope and its radioactive decay are the basis of so‐called carbon dating of archaeological specimens and, in our industry, give us one tool in determining the ‘natural’ status of fragrance and flavour ingredients as will be seen in Chapter 6.

­The Electronic Structure of Atom

­The Electronic Structure of Atoms The electrons orbiting a nucleus are not distributed randomly but are confined to volumes of space around the nucleus that we call orbitals. It is the pattern of these orbitals and their occupancy by electrons that determine the chemical properties of atoms. In order to understand the nature of an electron, we can picture it as either a wave or a particle. In reality, it is neither, but sometimes it is easier to make sense of its properties if we picture it as a one or the other. If we picture the electron as a particle, the orbital therefore becomes a probability distribution in space of where the particle might be. If we picture the electron as a wave, then the orbital becomes a standing wave of negative electricity around the nucleus. In either case, an electron in an orbital can be viewed as something possessing a definite distribution in space, a negative charge, and as something that can be distorted by electrical charges around it. In other words, the surface of an atom or molecule is not hard like a miniature billiard ball, but is more like a balloon or a cloud, which is affected by other charges around it. It is attracted by opposite charges and repelled by similar charges. The orbitals are considered in order of the energy required to keep an electron in them. The first or lowest orbital energy orbital has a capacity for only two electrons and is spherical in shape with the nucleus at its centre. It is called the 1s orbital. The names of the orbitals are derived from the number of the shell and the quality of the lines associated with them in their atomic spectra. Thus, s stands for sharp, p for principal, d for diffuse, and f for fundamental. The hydrogen atom therefore has one electron in its 1s orbital. Similarly, the helium atom has two electrons in its 1s orbital, and the orbital is full. Electrons have a property called spin. We can picture this as the way the electron, as a particle, will spin about its axis. There are two directions of spin and as a simple picture; we can see this as left‐handed and right‐handed spin. Each electron likes to pair up with another with the opposite spin. So, in the helium atom, the electrons are happy in that they are paired up and the orbital is full. Chemistry involves electrons moving from one atom to another. The electrons in helium have no desire to do this and so the helium atom is very unreactive chemically. The hydrogen atom, on the other hand, has only one electron in an orbital designed for two, and the single electron has no spin partner. Hydrogen therefore wants to do something with its electron to rectify the situation and consequently is chemically reactive. The 1s orbital completes what is called the first valence shell, and the electrons of the next eight elements populate the second valence shell. We now see the physical basis behind Mendeleyev’s arrangement of the periodic table, with two elements on the first row and eight on the second. The second valence shell contains four orbitals, each capable of holding two electrons. One of these is another s‐type orbital, the 2s orbital. The other three are known as p orbitals and have a shape reminiscent of a dumbbell. The three p orbitals are arranged at right angles to each other in space, all with their centres on the atomic nucleus. The shapes of s and p orbitals are shown in Figure 1.2. As the orbitals are filled, each additional electron fits into the next empty orbital. When all four of the two orbitals are occupied, the next electron pairs up with the one already in the 2s orbital, thus filling it. This then continues across the 2p orbitals.

9

10

1  The Structure of Matter

An s orbital

An sp3 orbital

A p orbital

Figure 1.2  Shapes of orbitals.

For example, the first element of the second row of the periodic table, lithium, has three protons in its nucleus and therefore has two electrons in the 1s shell (as do all subsequent elements) and one electron in its second shell. This latter electron occupies the 2s orbital. The next element is beryllium and has one of its electrons in the 2s orbital and the other in one of the 2p orbitals. Boron has one electron in the 2s orbital and one in each of two of the 2p orbitals. Carbon has one electron in each of its four 2 orbitals. As we move on to nitrogen, we now see the second shell electrons doubling up. Two of nitrogen’s electrons are in the 2s orbital, and the remaining three are distributed across the three 2p orbitals. Oxygen has four of its second valence shell electrons paired up and two single electrons. Fluorine has only one unpaired electron and neon has none. This process of building up the valence shell by adding each new electron to the available orbitals in order of increasing energy is known as the ‘Aufbau principle’. (Aufbau is German for building up.) Figure 1.3 shows this schematically with the electrons being represented by arrows with an upward pointing arrow indicating one spin direction and a downward pointing arrow representing an electron with the opposite spin. The three 2p orbitals are designated x, y, and z to represent the three orthogonal axes. As stated above, unpaired electrons are unhappy and need to do something to find a partner; this process is called chemical bonding. On inspecting Figure 1.3, it is clear that lithium has one unpaired electron, beryllium two, boron three, carbon four, nitrogen three, oxygen two, and fluorine one and neon – like helium – has no unpaired electrons. These numbers are the same as the common valencies of 2pz 2py 2px 2s 1s

Protons in nucleus

H

He

Li

Be

B

C

N

O

F

Ne

1

2

3

4

5

6

7

8

9

10

Figure 1.3  The electronic configurations of the first 10 elements.

­Hybridisation of Orbital

the elements, and now we have the explanation for the periodicity that Mendeleyev noticed. Lithium has one unpaired electron and a valence of one just like sodium, the first element in the next row. Fluorine has a valence of 1, and so one lithium atom will bind to one fluorine atom to give lithium fluoride, LiF. On the other hand, beryllium with its valence of 2 requires two fluorine atoms to form beryllium fluoride, BeF2. Compounds with oxygen are referred to as oxides, and we can now predict from Figure 1.3 that lithium oxide will have the formula Li2O, whereas beryllium oxide will be BeO. Formulae of the type LiF, BeF2, Li2O, and BeO are known as an empirical formula and the subscript numbers tell us the proportions of the various atoms in the basic unit of the chemical compound.

­Electronic Structure of Transition Metals The third valence shell contains three types of orbitals. In addition to the s and p orbitals that we found in the first two shells, it contains a new type, the d orbitals. The five d orbitals of the third shell can hold a total of 10 electrons. However, they are slightly higher in energy than the 4s orbitals and so do not start filling until after the latter. This accounts for the broadening of the periodic table in the fourth row. Having filled the 3s and 3p orbitals at the inert gas argon, the next electron goes into the 4s orbital to give potassium, and the next electron completes the 4s orbital giving calcium. Instead of going into the 4p orbital, the next electron goes into the first 3d orbital to give scandium, and it is not until the 3d orbitals are full (at zinc) that we start feeding into the 4d orbitals at gallium. The fourth shell also contains f orbitals – which are slightly higher in energy than the 5p orbitals – and so we see another broadening of the periodic table after ­element 57, lanthanum. There are seven f orbitals in each shell, and so there are places for 14 elements in these series. These elements are called lanthanides and actinides after the elements that immediately precede them. The fact that the orbitals of the higher valence shells overlap each other in energy means that electrons can be moved relatively easily from one to another and thus the elements in these central blocks exhibit variable valency, depending on how the electrons are arranged. The importance of this property will be seen in Chapter 8.

­Hybridisation of Orbitals As stated above, the carbon atom has two electrons in its first valence shell and four in the second. These latter are distributed with one in each of the 2s, 2px, 2py, and 2pz orbitals. However, carbon can combine these orbitals though a ­process known as hybridisation. Hybridisation of one s orbital with one p gives two sp orbitals, one s with two p orbitals gives three sp2 orbitals, and one s with three p orbitals gives four sp3 orbitals. This phenomenon is a very important feature of carbon chemistry, and since fragrance ingredients are all compounds of carbon, it is imperative to understand for all that follows in this book. The shape of an sp3 orbital is shown in Figure 1.2; it is essentially a dumbbell in which one side is much larger than the other.

11

12

1  The Structure of Matter sp3

p

p

sp2 sp

sp p

C with sp orbitals

sp2

sp3 sp2

C with sp2 orbitals

sp3

sp3 C with sp3 orbitals

Figure 1.4  Carbon atoms with hybridised orbitals.

The spatial distribution of these hybridised orbitals is also very important and is shown in Figure 1.4. A carbon atom with sp hybridisation has two sp orbitals arranged in a straight line on opposite sides of the nucleus and two p orbitals mutually at right angles to each other and to the line of the sp orbitals. Thus, in Figure 1.4, the vertical p orbital and the two sp orbitals should be viewed as lying in the plane of the page, and the other p orbital has one lobe projecting directly upwards from the page and the other downwards behind it. In the sp2 hybridised carbon atom, the p orbital and the left‐hand sp2 orbital are both in the plane of the paper, while one of the other two sp2 orbitals projects forward from the page and other lies behind it. If this atom is viewed from directly above the p orbital, there is an angle of 120° between each of the three sp2 orbitals. In the case of the sp3 hybridised atom, the four sp3 orbitals point towards the corners of a regular tetrahedron with an angle of approximately 119° between any two of them. Thus, in Figure 1.4, the upper and left‐hand orbitals lie in the plane of the paper, and one of the right‐ hand orbitals projects forward, while the other recedes behind the page.

­Chemical Bonding, Ions, Cations, Anions, and Molecules We have already seen that only the inert gases have atoms that are happy as they are; all other atoms contain unpaired electrons that seek to form a pair with another electron. This pairing can be achieved by forming a chemical bond. There are two main types of bonding: ionic bonds and covalent bonds. If we look at the two ends of the third row of the periodic table, we will find the alkali metal sodium at the left and the halogen chlorine at the right, just before the inert gas argon. Sodium and chlorine atoms both have two full shells of electrons, the first shell with two electrons and the second with eight. In the third shell, sodium has one electron and chlorine has seven. So, each has one unpaired electron, and each would really like to achieve the configuration of the nearest inert gas. For sodium, this would be most easily achieved by losing its one unpaired electron to give the same electronic structure as neon. On the other hand, chlorine would like to gain an electron, thus pairing up its single odd one and gaining the electronic structure of argon. So, if a sodium atom comes across a chlorine atom, it can donate its unpaired electron to the chlorine atom, and both can achieve inert gas‐like electron shells. However, the sodium atom now has only 10 electrons, but it still has 11 protons in its nucleus. So overall, it has a

­Chemical Bonding, Ions, Cations, Anions, and Molecule

surplus positive charge equivalent to that of one proton. Such charged species are called ions, and those with positive charges are called cations. Similarly, the chlorine atom now has 18 electrons but only 17 protons and therefore has a net negative charge of one unit. Negative ions are called anions. The new substance produced is called sodium chloride, also known as common salt, the stuff we use to season food. The bond between sodium and chlorine in common salt is called an ionic bond. The process is depicted in Figure 1.5 where the shaded circle represents the electronic core equivalent to that of neon, the outer circle represents the third valence shell, and the dots represent the electrons of that shell. Similarly, when magnesium reacts with chlorine to give magnesium chloride, the magnesium atom donates one of each of its two unpaired electrons to one of each of two chlorine atoms to give a magnesium cation carrying a double positive charge and two chlorine atoms, each with a single negative charge. Thus, magnesium chloride contains one magnesium cation, Mg2+, and two chloride anions, Cl−. Because ions carry electrical charges, ionic compounds (substances made up of ions, for example, common salt) tend to be solids (for reasons that will be clear later) and are usually more soluble in polar solvents. Polar solvents are those whose component molecules contain some areas that are positive and some that are negative and/or in which the charge distribution in the molecule is easily distorted. In the latter case, the molecule is said to be readily polarisable. It is easy to see how the charged ions of a salt are more easily taken up and supported in a polarised or polarisable solvent. Water, H2O, is a polar solvent as the hydrogen atoms in each water molecule carry partial positive charges and the oxygen atoms carry negative charges. Thus, when salt, NaCl, dissolves in water, each sodium cation will be surrounded by negative oxygen atoms from the water molecules, and conversely, the chloride anions will be held by the positively polarised hydrogen atoms of the water. In the solid state, the cations and anions will be held in a fixed array. Opposite charges attract, and so each cation will be surrounded by anions and vice versa. The exact way in which the ions are arranged relative to each other will determine the shape of the crystals that the salt will form. A small part of the crystal structure of sodium chloride is shown in Figure 1.6. Each sodium cation is surrounded by six chloride anions arranged at right angles to each other, at the corners of a regular octahedron. Similarly, each chloride anion is surrounded by six sodium cations. This lattice extends indefinitely in all directions. The basic unit of the structure, called the unit cell, is a cube, and so the overall shape of the entire assembly is a cube. If you grow salt crystals from a brine solu+

Na

Cl

Na

Figure 1.5  Formation of an ionic bond between sodium and chlorine.

–

Cl

13

14

1  The Structure of Matter

Cl–

Na+

Na+

Cl–

Cl–

Na+

Na+

Cl–

Cl–

Na+

Na+

Cl–

Cl–

Na+

Na+

Cl–

Figure 1.6  A fragment of the sodium chloride crystal lattice.

tion, you will indeed observe that the crystals are cubic in shape. This is an example of how the bulk properties of matter are dependent on their atomic and molecular properties. The other form of bonding involves sharing electrons to form covalent bonds. For example, two hydrogen atoms can share their unpaired electrons as shown in Figure 1.7. The electrons become paired and inhabit what is called a molecular orbital. This orbital covers the space around both nuclei and therefore holds them together. There are two protons, one in each nucleus, and two electrons, so the electrical charges are balanced, and the hydrogen molecule is electrically neutral. Similarly, one carbon atom can form covalent bonds with four hydrogen atoms to form CH4, a gas known as methane. This occurrence is shown in Figure 1.8 where the solid circles around the hydrogen nuclei now represent a complete first electronic shell, as in the helium atom. The outer circle around the carbon nucleus represents its complete second valence shell. The carbon orbitals used in forming methane are sp3 orbitals, and so the methane molecule will have a hydrogen atom at each corner of a regular tetrahedron. There are various ways of depicting this as shown in Figure 1.9. The most common method used in written figures is to depict the bonds as lines between the symbols representing the nuclei. In Figure 1.9, the bonds are shaded to show perspective. Regular lines are used for bonds in the plane of the paper, solid shading for bonds coming forward out of the plane of the paper and hatched shading for bonds that recede behind the plane of the paper. The formula CH4 is known as the empirical formula of methane. The first drawing shown in Figure 1.9a,

H

H

H

Figure 1.7  Formation of a covalent bond between two hydrogen atoms.

H

­Chemical Bonding, Ions, Cations, Anions, and Molecule

H

H H

C

C

H

H

H H

H

Figure 1.8  Formation of covalent bonds in methane.

H C

H

H H (a)

(b)

(c)

Figure 1.9  Structural formulae of methane.

which shows how the atoms are connected to each other in a molecule, is known as a structural formula. The second representation (Figure 1.9b) shows the nuclei as balls with sticks representing the bonds between them. In this picture, carbon atoms are coloured black and hydrogen white. The third representation (Figure 1.9c) is called a space‐filling model and shows the total space occupied by the orbitals of the molecule, and again, black is carbon and white hydrogen. Electrically neutral molecules such as methane do not like polar solvents such as water and prefer to dissolve in other electrically neutral media such as ­hydrocarbons and vegetable oils. Contrast this with ionic materials that prefer to dissolve in polar solvents as described earlier. Thus, we know that if we add salt to a salad dressing made of olive oil and vinegar, the salt will dissolve in the vinegar layer since it is composed mostly of water and acetic acid, both of which are polar liquids. On the other hand, if we crush a carrot in the same dressing, the oil layer will pick up the orange colour of the carrot, as carotene – the dye in carrots – is an electrically neutral molecule. The distribution of a material between polar and non‐polar liquids is very important in fragrance chemistry. The standard measure is a parameter known as log  P. To determine log  P, we take a mixture of water and octanol, which

15

16

1  The Structure of Matter

serves as a standard for an oily/non‐polar solvent, and add some of the test material to it. We then mix it all thoroughly and allow it to settle before separating the layers and determining how much of the test material dissolved in the octanol and how much in the water. We then take the logarithm (to base 10) of the ratio, and this number is defined as log P. A high positive value for log P indicates a material that prefers to dissolve in oil, and a negative value for one that prefers water. Materials preferring oil are called hydrophobic (water‐hating), and those that prefer water are called hydrophilic (water‐loving). Water phases are usually referred to as aqueous phases and oily phases as non‐polar or organic phases.

Review Questions 1 What is the difference between a mixture and a chemical compound? 2 What is the difference between a chemical compound and a chemical element? 3 What determines the atomic number of an element? 4 What determines the atomic weight of an element? 5 What determines the valency of an element? 6 Would you expect cesium iodide to be more soluble in water or in liquid paraffin?

17

2 Carbon 1 – Hydrocarbons Carbon is a unique element in terms of its ability to form large stable molecules containing chains and rings made up of carbon atoms. The number of possible different structures that can be made from carbon is infinite. This diversity means that carbon is the one element of the periodic table that can be used to create the range of molecules necessary for advanced forms of life. Because living organisms are based on carbon, the chemistry of carbon became known as organic chemistry, and compounds of carbon – other than some basic substances such as diamond and graphite – are known as organic compounds. We humans are made up of organic compounds, the sensors by which we detect odorous molecules are made of organic compounds, and the molecules responsible for odour are also organic compounds. An understanding of organic chemistry is therefore essential for an understanding of perfume. Molecular shape is very important, particularly in relation to biological prop­ erties. Two‐dimensional drawings can only go so far towards showing the overall shape of a molecule. Those who seriously wish to understand organic chemistry are well advised to acquire a set of molecular models so that they can develop a sense of the three‐dimensional shape of the various molecules under discussion. Sets of molecular models are available from various chemical suppliers such as Aldrich.

­Ethane: Conformational Isomers In Chapter 1, we came across the methane molecule, one carbon atom attached to four hydrogen atoms. The hydrogen atoms are arranged at the corners of a regular tetrahedron with the carbon atom at its centre. When two s orbitals, each containing one electron, combine to form a new orbital that contains two elec­ trons and holds two nuclei together in a chemical bond, the new orbital is called a σ orbital. (σ is the Greek letter corresponding to the English letter s and is pro­ nounced ‘sigma’.) These orbitals are shaped like a sausage containing the two nuclei near its ends, as shown in Figure 2.1, representing a hydrogen molecule formed from two hydrogen atoms. Similarly, σ orbitals can also be used to form bonds from sp, sp2, or sp3 hybridised atomic orbitals.

Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

18

2  Carbon 1 – Hydrocarbons

H

H

H

H

Figure 2.1  Formation of a hydrogen molecule.

If one of the hydrogen atoms of a methane molecule is replaced by a second carbon atom and that carbon atom carries three further hydrogen atoms, we come to the next member of the series that is called alkanes or paraffins. Alkanes are compounds composed only of carbon and hydrogen atoms and with only single bonds (i.e. 2‐electron bonds) joining any two atoms. This second member of the series is called ethane, and, like methane, it is a gas at normal temperature and pressure. The bonds around both carbon atoms in ethane are arranged tet­ rahedrally. The carbon atoms can rotate relative to each other, which means that the hydrogen atoms attached to one of the carbon atoms can take up different orientations relative to the hydrogen atoms attached to the other carbon atom. If we view the molecule from one end, looking along the bond between the two carbon atoms, we will see that the hydrogen atoms can, at one point, line up with those on the next carbon atom. This conformation is known as the eclipsed con­ formation because the hydrogens ‘eclipse’ each other. If we now rotate the car­ bon–carbon bond by 60°, the hydrogen atoms on one carbon lie equidistant between those of the next. This is known as the staggered conformation. Rotation by another 60° in the same direction brings us to another eclipsed conformation, and so on. Both the staggered and the eclipsed conformations are, in general, preferred to the intermediate conformations, known as skew conformations, and the staggered conformation is much preferred to the eclipsed. The eclipsed and staggered conformations of ethane are shown in two differ­ ent ways in Figure 2.2. One way of depicting them is via perspective drawings as was used in Figure 1.9 for methane. Another way of drawing them is the ‘end‐on’ view known as a Newman projection. In Newman projections, the nearer carbon atom is shown as a circle, and the bonds coming from it are shown as lines origi­ nating from the centre of the circle. The bonds coming from the further atom are H

H

H

H

H H

H

H

Eclipsed

Figure 2.2  Conformations of ethane.

H

H

H

H

Staggered

Ethane: Conformational Isomers

shown as originating from the edge of the circle, indicating that they are hidden from view by the nearer atom. It is usual to draw eclipsed bonds as very slightly skew; otherwise the drawing becomes very unclear. Molecules differing only in conformation are known as conformers or conformational isomers. The general term isomer will be defined in the paragraph after next. The energy differences between conformational isomers are relatively small, and, unless something else restricts the rotation around the carbon–carbon bond, they will interconvert easily at normal temperature. In Figure 2.2, the carbon atoms are not shown with the letter C but rather just as the intersection point of the bonds attached to them. This standard practice is used to make drawings of larger molecules clearer. If letters were used to mark all of the carbon atoms, structural drawings would become too cluttered and difficult to visualise. In Figure 2.3, the structures are shown using two other shorthand forms. In one of them, rather than showing all the bonds to hydrogen, the hydrogens are indicated next to the carbon atoms to which they are bonded. Again this helps to clarify complex structures. The other shorthand shown in Figure 2.3 is even neater and is the most common one found in scientific publications. It may look odd at first, but, once familiar with it, the reader will find it the clearest of all when dealing with more complex structures. This shorthand will be the standard format used from now on in this book. Using this system, every angle point, line junction or line end indi­ cates a carbon atom. All unspecified valences of the carbon atoms are assumed to be occupied by a hydrogen atom. So, for example, looking at the lower left structure in Figure 2.3, we can see four carbon atoms, one at each end point and two at the angles between them. The terminal carbons have only one other atom shown to be bonded to them, and so they each have three spare valences (remember that the total valence of carbon is four), and these are then taken to imply three hydrogens. The second and third atoms of the chain each have two other carbons attached and, therefore, have two spare valences and two hydro­ gens attached to each. The structure on the lower right of Figure 2.3 has three carbon atoms at the ends of lines, since each of these is shown as bonded to only one other atom, and so each must also carry three hydrogen atoms. The fourth carbon atom, the one at the intersection of the three lines, has three carbon atoms attached to it and, therefore, only has one spare valence and car­ ries one hydrogen atom. CH3

H2 C H3C

C H2

CH3

H3C

Figure 2.3  The isomeric butanes.

C H

CH3

19

20

2  Carbon 1 – Hydrocarbons

­Alkanes: Structural Isomers Joining three carbon atoms in a line produces the third member of the series, propane. When we come to add a fourth carbon atom to give butane, it could be added to either the first or second of the existing carbon atoms, as shown in Figure 2.3. Addition to the third would, because of the symmetry around the central carbon atom, be exactly the same as adding it to the first. Adding it to the first carbon of the propane structure produces a butane molecule with all four carbon atoms in a straight line. This molecule is referred to as normal butane or n‐butane for short. The isomer with the fourth carbon attached to the second atom of the three‐carbon chain is called iso‐butane or i‐butane for short. i‐Butane could also be considered to be a propane molecule carrying a methyl substituent on the second carbon, and so it is also referred to as 2‐­methylpropane. n‐Butane is referred to as a straight chain material, whereas the backbone of the i‐butane molecule is referred to as a branched chain. Molecules such as these two, which have the same empirical formula but different patterns of bonding between atoms, that is, different molecular structures, are known as isomers. Both n‐butane and i‐butane have the empirical formula C4H10 indicating that one molecule of each contains four carbon and ten hydrogen atoms. There are many different forms of isomerism, and we shall cover all of those most relevant to perfumery in this chapter and the next. The two butane isomers we have just encountered are examples of structural isomers where the basic skeleton of the molecule differs in the way the carbon atoms are connected together. In the n‐alkane series, if we draw the atoms out in a straight line format as is shown in Figure 2.4 for the fifth member of the series – n‐pentane, it is clear that there are two hydrogen atoms for each carbon and two extra, one at each end. If we move any carbons from one to another, as for example if we take one of the end carbons from n‐butane in Figure 2.3 and reattach it carbon two to give i‐ butane, we can see that the total number of hydrogen atoms will remain the same as one must be removed to allow the carbon to be introduced at the new position and then attached at the point where the carbon was removed. The formula of any alkane containing n carbon atoms is therefore CnH2n+2. It is impossible to add more hydrogen to an alkane without breaking a carbon–­ carbon bond and thus forming two new, smaller alkanes. Therefore, alkanes are said to be saturated with respect to hydrogen. This term is almost always abbre­ viated to ‘saturated’. As we have just seen, the fifth member of the alkane family is called pentane. From pentane upwards, to make the name of an alkane, we simply take the stem

H

H

H

H

H

H

C

C

C

C

C

H

H

H

H

H

Figure 2.4  n‐Pentane.

H

Alkanes: Structural Isomers

Table 2.1  Numerical prefixes. Number

Prefix

½

hemi‐

1

mono‐

2

di‐ or bi‐

2½

sester‐ or hemipenta‐

3

tri‐

4

tetra‐

5

penta‐

6

hexa‐

7

hepta‐

8

octa‐

9

nona‐

10

deca‐

11

undeca‐

12

dodeca

15

pentadeca‐

20

eicosa‐

of the Greek number and add the ending ‘ane’. So the next members of the series are hexane, heptane, and octane, respectively. Table 2.1 shows some of the more common numerical prefixes used in chemical names. The series of n‐alkanes can be used to illustrate the fact that molecular struc­ ture affects the physical properties of a substance. Table 2.2 shows the boiling point and physical state of a number of n‐alkanes. The physical state is given at standard temperature and pressure (STP) as both temperature and pressure affect the boiling point of a liquid. STP is defined as a temperature of 0 °C and a pressure equal to that of a column of mercury 760 mm high (the average atmos­ pheric pressure at sea level). It is immediately obvious from the table that the boiling point rises steadily with increasing molecular size. The first four mem­ bers of the series are gases at room temperature and atmospheric pressure. Eventually, the higher members of the series become solids at room temperature. This pattern relates to the amount of energy required to keep the materials in a gaseous or liquid state. A simple way of picturing this concept is to see the lighter molecules as balloons that float in the air, the medium weight molecules as balls that fall to the ground but are free to move around easily on the ground, and the heaviest molecules as being like chains that are heavy and tangled and therefore settle into fixed positions, packed together like sardines in a tin. These illustra­ tions equate roughly with the behaviour of molecules in gases, liquids, and sol­ ids, respectively, and will be discussed further in Chapter 4. The importance of this very basic relationship between molecular structure and properties will be shown clearly in later chapters.

21

22

2  Carbon 1 – Hydrocarbons

Table 2.2  Boiling points of n‐alkanes. Name

Formula

Boiling point (°C)

Physical state at STP

Molecular weight

Methane

CH4

−161

Gas

16

Ethane

C2H6

−88

Gas

30

Propane

C3H8

−46

Gas

44

Butane

C4H10

−1

Gas

58

Pentane

C5H12

36

Liquid

72

Hexane

C6H14

69

Liquid

86

Heptane

C7H16

98

Liquid

100

Octane

C8H18

126

Liquid

114

Decane

C10H22

174

Liquid

142

Pentadecane

C15H32

271

Liquid

212

Eicosane

C20H42

—

Solid

282

An alkyl residue (a fragment of an alkane structure) from which one hydrogen atom is missing is known as a radical. Common alkyl radicals are given abbrevia­ tions that are often used to save drawing the whole in a structural a formula. Thus, Me is a shorthand for the methyl radical –CH3; Et for ethyl, –CH2CH3; iPr for iso‐propyl, –CH(CH3)2; tBu for tertiary‐butyl, –C(CH3)3; and so on. When writing formulae or drawing structures, if we wish to indicate that an unspecified radical of some sort is present, we use the capital form of the letter R. Multiple alkyl residues can be indicated by using R, R′, R″, R‴, and so on for different radi­ cals. In these instances, the word radical is used to indicate what type of frag­ ment is attached via a bond at a specific point in a molecule. If the fragment is not contained in a molecule, but instead has an unpaired electron in place of a two‐electron bond, then it is called a free radical. Free radicals are very reactive species, as we will see in later chapters. They are represented using a dot to indi­ cate the unpaired electron. For example, Me˙ represents a methyl radical.

­Alkenes: Geometric Isomers In the alkane series, all of the carbon atoms are connected by single bonds. It is also possible for two carbon atoms to form a double bond between them. To do this, the atoms are sp2 hybridised. The 2s orbital combines with two of the 2p orbitals, and the third 2p orbital remains as a p‐type orbital. The three sp2 orbit­ als lie in the same plane with an angle of 120° between each, and the p orbital stands in a plane at right angles to that of the other three orbitals. One of the sp2 orbitals forms a bond to another sp2 hybridised carbon atom, and the other two form bonds either to hydrogen atoms or to other carbon atoms. Using the sim­ plest example – that is, a molecule containing just two carbon atoms and four hydrogen atoms – gives a species with the configuration shown on the left‐hand side of Figure 2.5. The p orbitals on the adjacent carbon atoms are close and lined

Alkenes: Geometric Isomers

H H

H H

H

H

H

H

Figure 2.5  Formation of a C–C double bond.

up with each other. They transform themselves into a new orbital containing two electrons (one from each of the p orbitals that formed the bond) that is shaped rather like two sausages: one above the plane of the carbon atoms and the other below it, as shown on the right‐hand side of Figure 2.5. This bond is called a π bond. (π is the Greek letter corresponding to the English letter p and is pro­ nounced the same as the English word ‘pie’.) Thus, the double bond between two carbon atoms is composed of one σ and one π bond. A C─C π bond is weaker than a C─C σ bond, and so the double bond between two carbon atoms is less than twice the strength of a single bond. The π bond requires that the two p orbitals from which it was formed lie in the same plane, and this requirement passes on to the π bond. In other words, the two carbon atoms are no longer free to rotate around the bond joining them. Materials containing double bonds are known as alkenes, and their names are formed in the same way as those of alkanes but using the suffix ‐ene instead of ‐ ane. Thus, an alkene containing four carbon atoms is known as butene. Alkenes are also called olefins. Since it is possible to chemically add more hydrogen atoms to the structure (thus forming an alkane), alkenes are said to be unsaturated. When we come to draw the structure of a butene, we find that we now have four possibilities, as shown in Figure 2.6. Structures 1 and 2 both have the four carbon atoms in a straight line but differ in the location of the double bond. In order to distinguish between them, we number the bonds starting at the end of the chain and then use the number of the bond that is a double bond in the name. Therefore, structure 1 is known as 1‐butene or but‐1‐ene and 2 is 2‐butene or but‐2‐ene. Structure 3 is also 2‐butene, but since the molecules cannot rotate around the double bond, these two molecules are different, and we must be able to distinguish between them. Isomers of this type are known as geometric iso­ mers. They are distinguished by looking at the way the substituents attached to the double bond lie relative to each other. In 2, they are across the bond on oppo­ site sides, and we call this arrangement trans. In 3, the two substituents are on the same side of the double bond, and we call this arrangement cis. Thus 2 is known as trans‐2‐butene and 3 as cis‐2‐butene. Because 4 contains four carbon 1

2

3

4

Figure 2.6  Isomeric butenes.

23

24

2  Carbon 1 – Hydrocarbons

atoms, it is often referred to as a butene, and to distinguish it from the straight chain butenes, it is called iso‐butene or isobutylene. However, there are rules for naming compounds agreed by the International Union of Pure and Applied Chemists (IUPAC), which state that the name should be taken from the longest straight chain. In the case of 4, the longest straight chain is only three carbons long, and so the molecule is a derivative of propene and is more properly called 2‐methylpropene. It is important to realise that systematic names, which are derived using IUPAC rules to connect a specific name with a specific structure, are used alongside older names, often derived from the first source in which the com­ pound was identified. These older names are known as trivial names, and, since they are often much shorter, they are often used more frequently than their systematic counterparts. For example, the hydrocarbon that comprises 80–90% of all citrus oils is commonly referred to as limonene, since this name is much easier to say or write than its IUPAC name of 4(R)‐(+)‐1‐methyl‐4‐(1′‐methyl­ ethenyl)cyclohex‐1‐ene. The chemist therefore needs to learn the common synonyms for the compounds he uses frequently. It is possible to have more than one double bond in a molecule. The number of double bonds is indicated in the name by the use of the Greek numerical prefixes di, tri, tetra, penta, and so on. The geometry of the double bonds can be indi­ cated by placing the terms cis or trans next to the number indicating the position of the bond along the chain. Thus, the alkene 5 shown in Figure 2.7 is known as 1,3‐trans, 5‐cis‐undecatriene. This compound is not just an academic example but is found in galbanum and contributes significantly to the odour of the oil. The other three molecules in Figure 2.7 are all based on a 12‐carbon chain. Structure 6 is dodecane and the other two are the isomeric 6‐dodecenes. Structure 7 is trans‐6‐dodecene and 8 is cis‐6‐dodecene. The zigzag line shown for structure 6 is the favoured shape for long chain materials and makes the mol­ ecules look rather like caterpillars. One thing to notice is that the overall shapes

5

6 7

8

Figure 2.7  An olefin found in galbanum and shapes of long chains.

Alkenes: Geometric Isomers

of alkane chains, or chains containing trans‐double bonds, is essentially a straight line. Introduction of a cis double bond into the chain has a dramatic effect on the overall shape and causes it to form a bend at the site of the cis double bond. This effect is important as will be seen in later chapters. When only one substituent is present at each end of a double bond, deciding which isomer is cis‐ and which trans‐ is easy. In cases such as structure 9, it is more difficult. One solution might be to refer to the longer carbon chain at each end, but this is also not always helpful either. It is possible to state clearly which two substituents are cis or which two are trans, but a more general system would be preferable. The system used employs atomic numbers to decide priorities and uses the initial letters of the German words entgegen (opposite) and zusammen (together) to describe the result. Thus E‐ corresponds to trans and Z‐ to cis. To decide which substituent has priority, we work outward from the double bond until a difference in atomic number is found. Then the higher atomic number takes precedence. So, with the substituents on the right‐hand end of 9, the first atoms we come to are either carbon or oxygen. Oxygen has the higher atomic number and so takes precedence. On the left‐hand end of the double bond, we come initially to a carbon atom on either side. Continuing out on one arm, we find only hydrogens, whereas on the other arm, there are two hydrogens and one carbon. So, the difference is a carbon over a hydrogen, and thus the carbon takes precedence. Thus, the precedence goes to the longer chain on the left‐hand side and to the oxygen substituent on the right. The double bond is therefore in the Z‐configuration as these two groups lie on the same side of the double bond. Structure 9 would therefore be called (Z)‐3‐methoxy‐4‐methylhept‐3‐ene. Structure 9

O

9

Another shorthand system is in use to save drawing out carbon chains. For example, in structure 9 in addition to the oxygenated group, it contains a one‐ carbon, a two‐carbon, and a five‐carbon chain attached to the double bond. We refer to these as radicals, and their names are derived from the parent hydrocar­ bon from which they are derived. Thus, the CH3 group is called a methyl radical, the CH2CH3 group as an ethyl group, and the CH2CH2CH3 as a propyl group. These groups are then abbreviated to Me, Et, and Pr, respectively. Thus structure 9 can also be drawn as shown in structure 10. Structure 10 Me Et Pr 10

O Me

25

26

2  Carbon 1 – Hydrocarbons

­Alkynes Two carbon atoms can also be joined by a triple bond. To do this, each carbon becomes sp hybridised. The s orbital combines with one of the p orbitals to form two sp orbitals, and the remaining two p orbitals stay as they are. The two sp orbitals lie in the same plane facing directly away from each other, and the p orbitals sit in planes at right angles to this and to each other. The two carbon atoms then form a σ bond using one sp orbital from each. This leaves the two p orbitals on each carbon lined up with those on the other, and so two π bonds can form between them as shown in Figure 2.8. The overall shape is linear with both carbons and both of the atoms attached at each end, all in a straight line. In Figure 2.8, the upper sketches show a perspective view from one side of the acet­ ylene bond. The lower sketches show the end view, looking along the line of the bond. The second π bond is weaker than the first, and so a C–C triple bond is less stable than a C–C double bond. Also, as the bond order increases, the distance between the two carbon atoms becomes less; thus a C–C double bond is shorter than a C─C single bond, and a C–C triple bond is shorter than a double. Compounds with C–C triple bonds are known as alkynes or acetylenes. The sys­ tematic name is formed by taking the numerical prefix and adding ‐yne. Therefore, the first member of the series, commonly known as acetylene and shown in Figure 2.8, is properly called ethyne.

­Cyclic Structures In addition to forming chains, carbon atoms can form rings. The angle between any two bonds around an sp3 hybridised carbon atom is 109°28′. Figure 2.9 shows the internal angles in the series of regular polygons. The regular pentagon has an internal angle very close to the bond angle of an sp3 carbon atom. So, five such carbons linked together in a ring are quite happy, and the ring shape is close to being planar. Since the molecule contains five carbons, it is a pentane. To indi­ cate the ring structure, it is called cyclopentane. The empirical formula is C5H10.

H

H

H

H

H

Figure 2.8  Formation of a C–C triple bond.

H

Cyclic Structures

60°

90°

108°

120°

128°34′

135°

Figure 2.9  Internal angles of regular polygons.

Thus, we see that introducing a ring reduces the empirical formula by two hydro­ gens compared to a saturated open chain structure, as does introducing a double bond. Thus, by looking at the empirical formula of a hydrocarbon, we can imme­ diately tell how many rings and/or double bonds it contains, one for every two hydrogens short of CnH2n+2, which is the formula for a standard saturated open chain alkane. For example, a hydrocarbon with an empirical formula of C15 H26 must contain three rings or double bonds. The prefixes cis and trans are used with cyclic structures to denote whether substituents lie on the same side or opposite side (respectively) of the plane of the ring, just as they are for substitu­ ents around double bonds. If we try to form a smaller ring than cyclopentane, the bond angles are forced to contract, introducing angle strain into the molecule. Thus, cyclobutane is a less favourable molecule than cyclopentane and is therefore chemically more reactive. The ring strain in cyclopropane is so great that its chemical properties become close to those of an alkene and it is relatively easy to break open the ring. Going the other way, i.e. to larger rings than cyclopentane, rather than strain the bond angle, the atoms pucker in order to adopt the tetrahedral bond angle of 109°28′. There are two preferred ways for cyclohexane to do this, shown in Figure 2.10. These two conformations are known as chair and boat conformations for obvious reasons, based on the shapes of the two. The chair is the most preferred and is the form that the molecule will adopt unless forced to do otherwise. In the lower part of the figure, the chair conformation is shown together with the hydrogen atoms attached to the carbons. It can easily be seen

Boat

Chair Axial

H H H

H H

H H H

H

H

H Equatorial

H

Figure 2.10  Conformations of cyclohexane.

27

28

2  Carbon 1 – Hydrocarbons

that the hydrogens fall into two categories, those that are directed above or below the plane of the ring and those that radiate outward from around its perimeter. The former are referred to as axial and the latter as equatorial. If we build a molecular model of the chair conformation of cyclohexane, it becomes clear that it has a beautiful symmetry about it and all of the C─C bonds are in the preferred staggered conformation. Larger rings lose this, and although they can adopt favourable configurations, they are not as favourable as cyclohexane. Medium rings, i.e. those containing 8–12 carbons, have another problem that models show very clearly. The hydrogen atoms attached to carbons on the opposite side of the ring actually touch each other across the ring. Since the electron clouds around atoms, which are not bonded to each other, repel each other, this leads to another form of strain, called steric strain. Basically, two hydrogen atoms try to occupy the same space, and this crowding stresses the structure. Large rings such as the 15–18 rings of macrocyclic musks are free of this problem. Rings can also contain double or triple bonds. cis double bonds are obviously easier to incorporate into a ring than trans in most cases because they pull the groups attached to them together into the ring. Triple bonds are more of a prob­ lem because they introduce a linear section, which is hard to incorporate into a ring unless it is large. Under a certain ring size, an acetylenic bond will introduce considerable ring strain into the molecule containing it.

­Polycyclic Structures A molecule may contain more than one ring, which may be at different sites in the molecule, fused along a common side or bridged. Structures with two rings are called bicyclic; those with three, tricyclic; and so on. The generic term polycyclic is used for any molecule with two or more rings in its molecular structure. For the moment, two examples will suffice to illustrate perfume chemicals con­ taining more than one ring. The molecule shown in structure 11 is called caryo­ phyllene, and it occurs naturally in the oil of cloves. There is a four‐ membered ring fused along one side to a nine‐membered one. The caryophyllene molecule is a very strained one. The four‐membered ring suffers from ring strain, and the nine‐­membered from steric strain. In addition, the nine‐membered ring con­ tains a trans double bond. Attempts to build a molecular model of caryophyllene will soon show how difficult it is to incorporate a trans double bond into a ring of this size. Since the ring is attached to opposite sides of the planar double bond, it is difficult to close the ring using only another seven carbon atoms. The molecule shown in structure 12 is α‐pinene. It is a major component of turpentine, and many different perfume ingredients are manufactured from it. α‐Pinene is an example of a material with a bridged ring system. In bridged rings, the links between the two rings are not on adjacent carbon atoms as they are in fused sys­ tems such as that of caryophyllene.

Polycyclic Structures Structures 11 and 12

2 1

3 7 4

6 5

11

12

In the drawing of structure 12, the carbon atoms of α‐pinene are numbered in order to demonstrate one systematic way of naming bridged structures. Firstly, we identify the two bridgehead carbon atoms. In α‐pinene these are the carbons numbered one and five. We then count the number of atoms in each bridge. In this case it contains one three‐carbon bridge and two one‐­ carbon bridges. In the case of caryophyllene, it has a seven‐membered bridge, a two‐carbon bridge, and one bridge containing no carbon atoms. We now start at one of the bridgehead atoms and give it the number 1. We then pro­ ceed around the longest bridge, numbering the atoms consecutively as we go. Then we go round the next largest ring and so on. In order to decide which bridgehead to start from, the rule is that it should be selected to give the low­ est figure when we come to number substituents. The pinane ring system con­ tains seven atoms in total. The ring total will always be the sum of the numbers of atoms in the bridges plus the number of bridgehead atoms. As is the case with α‐pinene, it holds 3 + 1 + 1 bridge atoms and two bridgeheads giving a total of seven. The bridge sizes are placed in square brackets. The systematic name for α‐pinene is therefore 2,6,6‐trimethylbicyclo[3.1.1]hept‐2‐ene. When it has a third ring, the number of carbons in the last bridge appears in the brackets alongside the others, but it is now necessary to add two superscript numbers to indicate the position of the bridge. The numbers are obtained from the bicyclic material lacking this third bridge, so one first constructs the bicyclic molecule, numbers its skeleton, and then adds the third bridge across between the appropriate numbers. For non‐chemists in the fragrance indus­ try, the important things to remember about such apparently complex names are that the name corresponds exactly to the structure and that each character and punctuation mark in the name has a precise significance. From the name we can reconstruct the molecule, but, if there is a typographical mistake in the name, then it is likely that a wrong structure would result. Many natural fra­ grance ingredients, especially in the terpenoid family (see Chapter 14), con­ tain bridged ring structures. Molecules with bridged rings are more compact than open chain counterparts and therefore usually have lower boiling points. Polycyclic structures are also more rigid than open chains, which means that recognition of them will be more specific and they will also tend to be slower to biodegrade. The importance of these features will be discussed further in Chapters 12 and 13.

29

30

2  Carbon 1 – Hydrocarbons

­Greek Letters Once again, we have come across the use of Greek characters in chemistry. Greek characters are used as we saw above to describe bonding orbitals, and they are also used in NMR spectroscopy, as we will see in Chapter 5. In naming organic chemicals, the significance of Greek letters lies in their order in the Greek alpha­ bet. In this context the most important ones are the first five and the last one. These letters and their English equivalents are shown in Table 2.3. They are used to indicate position in a sequence. However, various sequences are involved, and so we have to judge by the context as to what the significance is. In all cases, sim­ ply learning the names will serve to distinguish between different molecules. In Table 2.3  Greek letters. Greek letter Small

Capital

English name

English equivalent

α

Α

Alpha

a

β

Β

Beta

b

γ

Γ

Gamma

g

δ

Δ

Delta

d

ε

Ε

Epsilon

e (short, as in set)

ζ

Ζ

Zeta

z

η

Η

Eta

e (long, as in meet)

θ

Θ

Theta

th

ι

Ι

Iota

i

κ

Κ

Kappa

k

λ

Λ

Lambda

l

μ

Μ

Mu

m

ν

Ν

Nu

n

ξ

Ξ

Xi

x

ο

Ο

Omicron

o (short, as in hot)

π

Π

Pi

p

ρ

Ρ

Rho

r

σ

Σ

Sigma

s

τ

Τ

Tau

t

υ

Υ

Upsilon

u

φ

Φ

Phi

ph

χ

Χ

Chi

ch (as in loch)

ψ

Ψ

Psi

ps

ω

Ω

Omega

o (long, as in owe)

Aromatic Rings

most cases, the molecules in question are isomers. So, the name α‐pinene helps us to distinguish between two isomeric alkenes found in turpentine, the other being β‐pinene. The former is present in higher amounts and so is given the first letter, α, and the latter is then the second in the series. Sometimes the sequence depends on relative abundance, sometimes on order of discovery, and sometimes it is used to indicate the position of a double bond. The last letter in the Greek alphabet is ω, and so this letter is used to indicate something at the end of a series or at the end of a chain. For example, the last atom in a carbon chain is often referred to as the ω carbon, irrespective of the length of the chain. The use of Greek letters is only semi‐systematic, and, if in doubt, we should always use a systematic numerical way of identifying carbon atoms in a molecule.

­Aromatic Rings If a six‐membered ring contains three double bonds, two ways of placing them are shown in Figure 2.11, using the structures with numbered atoms for clarity. The double bonds can be interchanged by moving the double bond positions one step round the ring. In doing so, none of the atoms making up the ring or, any of those attached to it, will move. If, instead of drawing the double bonds, we simply draw the p orbitals from which they were formed, we see that all of these p orbitals are lined up with each other in a ring, as also shown in Figure 2.11. What actually happens is that instead of forming three separate double bonds, the six orbitals merge to form a ring of electrons delocalised across all six carbon atoms with one electron cloud above the ring of carbon atoms and one below it. Such rings are called aromatic rings or benzene rings. The latter is named after benzene that is the simplest hydrocarbon containing such a system. The term aromatic was given to these compounds because benzene and related hydrocar­ bons tend to have stronger odours than saturated or olefinic hydrocarbons. They are sometimes drawn with three separate double bonds and sometimes with a circle, as shown in Figure 2.11. Both representations have the same significance, and it must be understood that the three double bonds do not behave as isolated 1 6 5 4

1 2

6

3

5

2 3 4

Figure 2.11  Representations of benzene.

31

32

2  Carbon 1 – Hydrocarbons

double bonds. The ability to spread the electrons around the ring results is favourable in energy terms and makes the system much more stable than iso­ lated double bonds would be. Benzene is therefore more stable, hence less reac­ tive, than olefinic hydrocarbons. This phenomenon of sharing electrons in a ring with resultant chemical stability is known as aromaticity, and the π‐electrons in the ring are said to be delocalised. This equilibrium movement of electrons with­ out moving nuclei is known as resonance, and the individual ‘frozen’ structures are known as canonical structures. This phenomenon also occurs with charged canonical structures as will be seen in the next chapter. Aromaticity is so important in organic chemistry that the words alicyclic and aliphatic have been coined to identify those materials that do not contain aro­ matic rings. Alicyclic materials do contain rings, but not aromatic ones; aliphatic materials do not contain rings. The abbreviation Ph is often used to denote a benzene ring substituent (i.e. the –C6H5 radical) when drawing structures. The Greek letter phi, φ, is also sometimes used in this way to denote a benzene ring. Obviously, when there is more than one substituent on a benzene ring, there is the possibility for structural isomerisation. The relative positions of substituents can be identified by numbers, but a common alternative is the system using the prefixes ortho‐ (o‐), meta‐ (m‐), and para‐ (p‐) to refer to 1,2‐, 1,3‐, and 1,4‐, respectively. For example, xylene is the trivial name for dimethylbenzene, and so the three isomeric xylenes are known as o‐, m‐, and p‐xylene, respectively, as shown in Figure 2.12. The position of groups relative to each other in aromatic rings is very important as the electronic effects of one group on another very considerably from o‐ or p‐ to m‐ as will be discussed in Chapter 8. Benzene rings can be fused together into polycyclic aromatic systems. The next simplest example is naphthalene, which contains two benzene rings fused along one side. The π‐electrons of naphthalene are shared across all 10 carbon atoms. There are several ways of numbering or naming the carbon atoms in naphthalene; the two most common are shown in Figure 2.13. It should be noticed that the two carbon atoms at the ring fusion cannot carry substituents

ortho-xylene

meta-xylene

para-xylene

Figure 2.12  Structures of the xylenes.

8 7 6

1 8a

β

3

4a

5

α 2

4

Figure 2.13  Numbering systems for naphthalene.

Stereoisomerism

(unless the ring becomes saturated), and so they are not given numbers initially; the numbers 4a and 8a are only necessary in the case of saturated or partially saturated naphthalene derivatives. The Greek letters α or β are usually only used with naphthalene rings bearing a single substituent. In the same way that R is used in formulae to indicate the presence of an alkyl radical, Ar is used to indicate the presence of an aromatic ring with unspecified substituents, which could possibly be additional benzene rings fused onto the one referred to.

­Stereoisomerism There is one more form of isomerism that is important in fragrance chemistry. Stereoisomerism is very much a three‐dimensional property, which makes three‐ dimensional molecular models particularly helpful in understanding it. Stereo­ isomerism is also known as chirality. This word comes from the Greek word for hand, since hands are good examples of the phenomenon. If the mirror image of an object is not superimposable on the object, then it is said to be chiral, and the mirror images are said to be enantiomers of each other. By looking at our hands, we will see that they are mirror images of each other and that they cannot be superimposed on each other. We also know that it is difficult to put a right‐hand glove onto the left hand or for two people to shake hands if one uses the right hand and the other the left. In short, a very important property of chiral objects is that they can be recognised by other chiral objects. If a carbon atom has four different groups attached to it, it becomes chiral. Such carbon atoms are known as asymmetric carbon atoms, and they are said to constitute an asymmetric centre in a molecule. As with hands, we refer to them as right handed or left handed, and the non‐superimposability of such isomers on each other is illustrated in Figure 2.14. In this figure we see an asymmetric carbon atom labelled (S) that is sitting to the left of a mirror. The molecule on the right of the mirror is labelled (R), and (R) is identical to the mirror image of (S). If (S) is moved to the far side of the mirror and rotated by 180°, it is not superimposable on its mirror image, (R). Enantiomeric mole­ cules are identical in every physical and chemical property except when they come into contact with another chiral object or with a chiral force. The chiral force most relevant here is plane polarised light. Most people are familiar with polarising lenses in sunglasses. Normal light is not polarised and con­ tains waves vibrating at all angles. Polarising filters only allow light waves vibrating in one particular plane to pass through. To demonstrate this effect, take two pairs of polarising sunglasses, and hold the lens of one in front of the lens of the other. If the glasses are held at the same angle, then light will be able to pass through both. However, if one lens is rotated the amount of light passing through will be reduced, until at 90°, no light will pass through the pair. Chiral molecules have the property of rotating the plane of polarised light. So, if we place a solution of a chiral material between the two sunglass lenses, we will find that the angle at which the maximum light is allowed through is no longer 0°. If the material has twisted the plane of the light to the

33

34

2  Carbon 1 – Hydrocarbons Move

Rotate 1

1 4

4

2

1 3 2

3

3 (S)

(R)

2

4 (S)

Mirror

Figure 2.14  Non‐superimposability of asymmetric carbon atoms.

right, the material is said to be right handed or dextrorotatory; if to the left, then it is left handed or laevorotatory. The angle of rotation depends on the molecular structure around the chiral centre, and when measured under standardised conditions, it is referred to as the specific rotation of the mate­ rial. Because of this property of interfering with polarised light, stereoisomers are also sometimes referred to as optical isomers. One system for identifying enantiomers is to use the prefix d‐ for dextrorota­ tory and l‐ for laevorotatory. Limonene, the major component of citrus oils, con­ tains a chiral centre in its structure and therefore exists in two stereoisomeric forms. The isomer that occurs in orange oil is dextrorotatory and hence is known as d‐limonene. Its enantiomer, l‐limonene, occurs in some pine oils. In order to show these bonds in drawings, we often use shaded lines to indicate them. Heavy, solid shading indicates a bond projecting up from the plane of the paper, whereas a hatched line represents a bond that recedes behind the plane of the paper. The other two bonds attached to the chiral carbon atom are taken to lie in the plane of the paper. Thus, the enantiomers of limonene can be drawn as shown in the top pair of structures in Figure 2.15. Normally, if one of the substituents is hydro­ gen, as is the case with limonene, we usually just draw the other and imply that hydrogen will occupy the missing place. For comparison, this is shown in the lower pair of structures in Figure 2.15. Three other systems are used when identifying enantiomers. One uses (+)‐ instead of d‐ and (−)‐ instead of l‐. This classification is now regarded as more correct, and, it is normally used in modern work. However, it is still very com­ mon to find the older d‐ and l‐ labels. In both of the above systems, the prefixes refer to the direction of rotation of the plane of polarised light. However, they do not tell us about the absolute con­ figuration of the atoms in space. In order to do this, we use one of two other systems.

Stereoisomerism

H

H

l-Limonene

d-Limonene

(–)-Limonene

(+)-Limonene

Figure 2.15  Representations of the enantiomers of limonene.

The older system uses the capital letters, D‐ and L‐, as prefixes. They show the relationship of the chiral carbon atom in question to the chiral centre of a mol­ ecule called glyceraldehyde. Thus, the chiral carbon atom in a molecule that is labelled by the D‐ prefix will have the same configuration as the chiral carbon in dextrorotatory or D‐glyceraldehyde. This material was chosen as the standard because the early work on stereoisomers was carried out on sugars, and it is relatively easy to degrade different sugars to glyceraldehyde. This system of nomenclature is now obsolete, but one still comes across it from time to time, and, on doing so, it is important to realise that it does not necessarily correlate with the d‐/l‐system. In other words, a material correctly labelled D‐ could be either d‐ or l‐. The modern way of indicating the absolute configuration of a carbon atom is to use the prefixes (R)‐ and (S)‐. These are derived from the Latin words rectus and sinister, meaning right and left, respectively. In order to determine which label to use, we first define the priorities of the four groups attached to the chiral carbon atom using atomic numbers, in exactly the same way as was used for determining E‐ and Z‐geometric isomers. The group with highest priority is given the number 1, the second is 2, and so on. We then hold the molecule with the lowest priority group (labelled 4) pointing away from the viewer. If following round the other three groups from 1 to 2 to 3 takes us in a clockwise direction, then the centre is said to be right handed or (R)‐, and if anticlockwise, it is (S)‐. This arrangement is shown in Figure 2.16. It is possible for a molecule to contain more than one asymmetric centre. Each centre has two possible configurations, and so a molecule with n chiral carbon atoms will exist in 2n different stereoisomeric forms. Thus, molecules with two chiral centres will contain 4 isomers; with three, 8; with four, 16; and so on. Stereoisomers containing multiple asymmetric centres are known as diastereom­ ers. Menthol is an example of an important fragrance and flavour ingredient with

35

36

2  Carbon 1 – Hydrocarbons 2

3

4

(R)-

1 3 2 4

(S)-

1

Figure 2.16  Determining (R)‐ and (S)‐nomenclature.

three chiral centres, and hence eight stereoisomeric forms. A molecule with n chiral centres will have 2n/2 pairs of enantiomers, and each enantiomer, as before, will have identical physical and chemical properties to its antipode (opposite enantiomer), except when confronted by a chiral object or chiral force. However, a set of diastereomers will contain molecules that are no longer mirror images and will therefore exhibit different physical and chemical properties. For exam­ ple, in a molecule with two chiral centres, the four isomers will have the configu­ rations (R)(R), (R)(S), (S)(R), and (S)(S). (R)(R) and (S)(S) will be enantiomers of each other, and also (R)(S) and (S)(R) will be enantiomers of each other. However, (R)(R) and (R)(S) will not be enantiomers and will have different properties. For example, the boiling points of (R)(R) and (R)(S) will be different. Of course, the boiling points of (R)(R) and (S)(S) will be identical, as will those of (R)(S) and (S) (R), since these are both pairs of enantiomers. In some cases, the number of pos­ sible stereoisomers will be reduced by internal symmetry in the molecule. The classic case is tartaric acid. This natural product has two chiral centres connected to each other, and each half of the molecule is a mirror image of the other. Thus the (R)(S) form becomes identical to the (S)(R). This form is referred to as a meso‐ isomer and is said to be internally compensating since it is not optically active. A mixture of equal parts of two enantiomers is known as a racemate or racemic mixture. The process of separating individual enantiomers from a racemate is known as resolution. In some cases, it is possible to convert one enantiomer into its antipode. If such a process is allowed to continue without any interference, the natural end point will be a racemic mixture, and this process is known as racemisation. This detail on stereoisomers may all seem unnecessarily complex and esoteric to the reader, but it is very important in analysis of essential oils, in synthesis of fragrance ingredients, and in odour perception, as will be seen in later chapters.

­Rules for Hydrocarbon Nomenclature ●● ●● ●● ●●

Stem name is taken from parent alkane using Greek numbers as in Table 2.1. Saturated materials have suffix ‐ane. Double bonds have suffix ‐ene. Triple bonds have suffix ‐yne.

Stereoisomers

●● ●●

●● ●●

For acyclic materials: Find the longest unbranched chain; number the carbons in that chain. Number starting from the end, which makes the first feature number the lowest, unsaturation takes precedence over substituents. Side chains are numbered using primes. For monocyclic compounds: Add the prefix cyclo‐ to the stem number for the ring. Rings take precedence over chains. For bicyclic compounds: Find the total number of carbons in the ring to derive the stem. Find the two key bridgeheads. Count the numbers of carbons between bridgeheads, and write these, in decreasing numerical order, in square brack­ ets after the prefix. The prefix will be made up of the number of rings (di‐, tri‐, tetra‐, etc.) followed by cyclo‐. For aromatics: Use the basic ring system as the stem. Arabic numerals: Used to identify atoms in rings or chains  –  in numerical order. n‐ = normal, i.e. unbranched i‐ = iso, i.e. branched – usually a methyl group on the penultimate carbon of a chain t‐ = tertiary, i.e. doubly branched at the same position sec‐ = secondary, i.e. attached at a position other than the end of the chain. Greek letters: Used to identify relative positions α = next to, β = next but one to, etc., and ω = last (ω−1 = next to last). The Greek capital delta Δ‐ is used with a number to indicate the position of a double bond

­Quick Rules for Isomers Structural/positional: Differentiate by numerical prefixes. Geometrical: These can be about rings or double bonds. For either, cis‐ = longest carbon chains on the same side of the double bond or ring, or trans‐ = longest carbon chains on opposite sides of the double bond or ring. Alternatively, for double bonds, precedence of substituents is set by atomic numbers, starting at the groups attached to the double bond and working out stepwise until a difference is found. ●●

●●

Z‐ (zusammen) = the two highest precedence substituents are on the same side of the double bond. E‐ (entgegen) = the two highest precedence substituents are on opposite sides of the double bond.

­Stereoisomers + or d (dextro) = rotates the plane of polarised light to the right. − or l (laevo) = rotates the plane of polarised light to the left. For R and S, use the rule shown in Figure 2.16.

37

38

2  Carbon 1 – Hydrocarbons

Review Questions 1 How many isomeric pentanes (i.e. saturated hydrocarbons containing five carbon atoms) are there? 2 How many isomeric hexenes (i.e. hydrocarbons containing six carbon atoms and one double bond) are there? 3 Identify any chiral carbon atoms in the following structures.

4 Write the structural formulae for the following compounds. a) 1,2‐Dimethylbenzene b) 2‐Methylbuta‐1,3‐diene c) 2,2,4‐Trimethylpent‐4‐ene d) (E)‐2‐methylene‐6,10,10‐trimethylbicyclo[7.2.0]undec‐5‐ene e) Tricyclo[5.2.1.02,6]deca‐3,8‐diene

39

3 Carbon 2 – Heteroatoms ­Hydrogen Bonding Carbon not only has an immense ability to form complex structures by making carbon–carbon bonds, but it is also able to increase the variety of possible struc­ tures by forming bonds to other atoms as well. Since the basic element of organic chemistry is carbon, the other atoms are usually referred to as heteroatoms (from the Greek word for ‘different’) in organic structures. The most important heteroatoms from the point of view of the fragrance industry are oxygen, sulfur, and nitrogen. The introduction of heteroatoms makes organic chemistry much more varied, and the resultant possibilities for different structures and reactions are essential for the chemistry of complex life forms such as plants and humans. Odorous molecules tend to contain heteroatoms. The majority of fragrance ingredients contain only carbon, hydrogen, and oxygen in their molecules. If we list typical pleasant smells such as rose, jasmine, sandalwood, musk, grass, hay, vanilla, fresh air, and so on, the typical chemicals responsible for these odours are all CHO (carbon, hydrogen, oxygen) compounds. If we are asked to name malodours, we are much more likely to come up with compounds of sulfur and nitrogen. Sulfur compounds are important contributors to the smell of putre­ fying animal or vegetable matter. Nitrogenous (i.e. nitrogen‐containing) mole­ cules also add to the smell of protein degradation, and so they are the key components of fishy odours and significant contributors to the smell of urine and faeces. However, there are plenty of exceptions to this generalisation. For example, carboxylic acids, which are CHO compounds, are significant contribu­ tors to the smell of sweat and vomit. Conversely, the key component of the pleasant passion fruit odour is a sulfur molecule, while methyl anthranilate, a nitrogenous compound, contributes to the heavy sweetness of the scent of ylang‐ylang blossom. For each of the chemical classes given below, I will give some examples with some of the different names that are in use for each. To give extensive instructions for naming compounds would take more space than could be justified here, but the examples provided should be sufficient for the reader to understand the significance of chemical names and to extrapolate the naming patterns to other compounds. It is important to remember that in the fragrance industry and elsewhere, systematic names, semi‐systematic names, and trivial names are all in Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

40

3  Carbon 2 – Heteroatoms

common use. One must be prepared to deal with this and to realise that any one compound may be referred to by several different names. Each distinct grouping of heteroatom containing structural units is known as a functional group, sometimes abbreviated to function. So, for example, chem­ ists will talk about the C–O–H group as an alcohol functional group or alcohol function.

­Alcohols If one carbon atom forms a bond to one oxygen atom and all spare valencies (oxygen is divalent, and carbon is tetravalent) are occupied by hydrogen, we form the molecule known as methanol. Figure  3.1 shows several representations of the  methanol molecule. The top row contains the normal ways of writing the structure, one showing all the hydrogen atoms; the second, which is more common, only shows the hydrogens connected to the oxygen and indicates the carbon atom only as the terminus of a line indicating the bond. The textual representation at the right of the top line uses the abbreviation Me for the methyl radical and then shows both the oxygen atom and the hydrogen connected to it. The drawings in the lower row illustrate some important properties of methanol. When two carbon atoms form a bond to each other, the electrons in the bond are shared equally between them. However, when a carbon atom forms a bond to a heteroatom, the sharing is not necessarily so even‐handed. For example, in a single bond between carbon and oxygen, the electrons of the σ‐ bond are drawn towards the oxygen since it has a greater number of protons in its nucleus than carbon. This results in a partial negative charge developing over the oxygen atom and a complementary positive charge over the carbon atom. This is referred to as polarisation of the bond. The partial charges are usually written using the Greek letter δ together with the sign of the charge. Thus, the drawing at the left of the lower row in Figure 3.1 shows the polarisation of the carbon–oxygen bond. The central figure in the lower row shows the polarisation in the oxygen–hydrogen bond. As with the oxygen–carbon bond, the greater nuclear charge of the oxygen atom pulls the electrons to itself and away from the hydrogen atom, which therefore becomes partially positively charged. The

H

H

O

OH

H

MeOH

H

H

H δ+

H

δ–

O

H

H

H

δ–

O

H

δ+

H

Figure 3.1  Representations of methanol.

H

H

H

O

H

­Alcohol

difference in nuclear charge between oxygen and hydrogen is even greater than that between oxygen and carbon, so the O–H bond is more highly polarised, and the compound becomes a weak acid. (For definitions of acids and details on their chemistry, see Chapter 7.) The polarisation of the hydroxyl (–OH) group is very important as will be seen in later chapters. As we saw in Chapter 1, the outer shell of electrons in oxygen contains two pairs that are not involved in bonding. These electrons are known as lone pairs, and they are shown in the drawing at the bottom right of Figure 3.1. The overall shape of the oxygen atom in methanol can be taken as tetrahedral, like a carbon atom but with only two of the arms being used for bonding. Although the lone pairs are not used for whole bond formation, they do contain electrons and therefore are areas of electron density, hence exhibiting a negative charge. As such, they can be used in forming electrostatic interactions with other molecules. For example, the positively charged hydrogen atom of one methanol molecule can form an electrostatic interaction with a lone pair of electrons on another as shown in Figure 3.2. This interaction is known as a hydrogen bond and is a very important feature of biological chemistry, as will be seen when we discuss the mechanism of olfaction. The –OH group is known as a hydroxyl, hydroxy, or alcohol group, and compounds containing hydroxyl groups are known as alcohols. Their systematic names are derived from the name of the corresponding hydrocarbon but use the suffix ‐ol instead of ‐ane. It is usually necessary to add a number in front of the ‐ol suffix to indicate the carbon to which the hydroxyl group is attached. Compounds with two alcohol groups are known as diols, those with three as triols and so on. A semi‐systematic way of naming alcohols is to take the stem of the name of the carbon radical to which the oxygen is attached and use it as one word, with the word ‘alcohol’ following it. Thus, the most common alcohol – the one in beer, wine, and spirits – is known either as ethanol or ethyl alcohol. As an example of naming of alcohols, Figure 3.3 shows the structures of some alcohols containing five carbon atoms. In some instances, several possible names are shown. An alco­ hol in which the hydroxyl group is attached to a carbon that also carries one other carbon atom and two hydrogens is called a primary alcohol. If the carbon has two other carbons attached and only one hydrogen, the alcohol is said to be secondary, and if the carbon atom has three other carbon atoms and no hydro­ gens attached to it, the alcohol is tertiary. The alcohols in the left‐hand column of Figure  3.3 are primary alcohols; those in the second column are secondary H

H

δ–

O

H

δ+

H

H

H

O

H

H

Figure 3.2  Hydrogen bonding in methanol.

41

42

3  Carbon 2 – Heteroatoms Primary alcohols

Secondary alcohols

Tertiary alcohol

OH

OH

HO 1-Pentanol Pentan-1-ol n-Amyl alcohol

Diol

OH

2-Pentanol 2-Methyl-2-butanol Pentan-2-ol 2-Methylbutan-2-ol Secondary amyl alcohol Tertiary amyl alcohol s-Amyl alcohol t-Amyl alcohol OH

OH

Pentane-1,3-diol

HO

3-Methyl-1-butanol 3-Methylbutan-1-ol Isoamyl alcohol

3-Methyl-2-butanol 3-Methylbutan-2-ol Methyl isopropyl carbinol

Figure 3.3  Names of some C5 alcohols.

alcohols; then there is one tertiary alcohol and one diol shown. In each case, the top two names are systematic names using the International Union of Pure and Applied Chemistry (IUPAC) rules. The word amyl is common in older literature and refers to a molecule containing five carbon atoms. The term carbinol is not commonly used nowadays, but it will be encountered in older perfumery litera­ ture. The chemical properties of alcohols vary depending on whether they are primary, secondary, or tertiary, as will be seen in later chapters.

OH

Geraniol rose

OH

Citronellol rose

OH

2-Phenylethanol rose

HO HO OH Dihydromyrcenol fresh, floral, citrus

α-Terpineol lilac

Figure 3.4  Some alcohols used as perfume ingredients.

nor-Patchoulenol patcholuli

­Phenol

Alcohols are an important class of materials in perfumery. Some of the most significant are shown in Figure 3.4. The alcohols geraniol, citronellol, and pheny­ lethanol are the major constituents of rose oils, and a simple blend of these three will give a good impression of the rose odour. Dihydromyrcenol is a synthetic material. Its first use in fine fragrance was in Drakkar Noir where it started a new fashion trend in masculine freshness. This trend was given extra impetus through use of an overdose of dihydromyrcenol in Cool Water. α‐Terpineol is the major component of pine oil and is responsible for its characteristic smell, often associ­ ated with disinfectants. It is one of the highest tonnage perfume ingredients. nor‐Patchoulenol is present in fermented patchouli leaves and in the oil extracted from them. It is not the largest component of the oil in weight terms, but it is very important in odour terms. Other odour types associated with alcohols include mint and sandalwood.

­Phenols If the atom to which a hydroxyl group is attached is part of a benzene ring, then the compound is known as a phenol. The generic name phenol is derived from that of the first member of the series, hydroxybenzene or phenol. The aromatic ring system helps to stabilise charges, and so the hydrogen attached to the oxy­ gen of a phenol is more acidic than that of an aliphatic alcohol. In fact, the old name for phenol was carbolic acid. It has good antibacterial properties, but its acidic properties and solubility in organic media make it rather corrosive to liv­ ing tissues. Phenol was the first antiseptic to be used as such in surgery, and it also found wide use as a disinfectant. In names, the phenol group usually takes precedence, and other substituents are numbered relative to it. Numbers or the OH

OH

OH

CI

CI

CI Phenol Carbolic acid

2-Methylphenol o-Cresol

OH

2,4,6-Trichlorophenol TCP

OH OH

1,2-Dihydroxybenzene o-Hydroxyphenol Catechol

5-Methyl-2-(1-methyleth-1-yl)phenol 5-Methyl-2-i-propylphenol thymol

Figure 3.5  Structures and alternative names for some phenols.

43

44

3  Carbon 2 – Heteroatoms O O O

HO

O

O OH Eugenol clove oil carnation

HO

Vanillin vanilla

OH Methyl everniate treemoss

Figure 3.6  Some phenols present in essential oils.

prefixes o‐, m‐, and p‐ can be used to indicate the position of other substituents relative to a phenolic hydroxyl group in naming a molecule. The structures and names of some typical phenols are shown in Figure 3.5. Because of their acidic properties, simple phenols are not used in perfume, but there are a number of more complex phenols that occur in essential oils. For example, thymol, shown in Figure 3.5, is found in essential oils such as thyme. Some other phenolic components of natural products are shown, together with their key sources, in Figure 3.6.

­Ethers If an oxygen atom is connected to two different carbon atoms, the compound is called an ether. Ether names are usually derived from the alkyl radicals attached to the oxygen atom. The most common ether, often referred to simply as ‘ether’, is formed from two ethyl radicals and is more properly called diethyl ether. An alternative method of naming ethers is to see them as a hydrocarbon in which one CH2 group has been replaced by oxygen. The chain name is then taken as that of the ‘parent’ hydrocarbon with the prefix oxa‐ and a number to indicate the nature and location of the substitution. Thus, diethyl ether could also be called 3‐oxapentane. Ethers are generally quite stable in acid and alkali media but can be subject to oxidation by air as will be seen in later chapters. Some simple ethers are shown in Figure 3.7. Their chemical stability makes ethers useful as ingredients in harsh product bases such as acid cleaners and bleaches. However, in general, ethers are less likely to have desirable odours than members of the other classes of oxygenated

O

Diethyl ether ether 3-Oxapentane

O

O

Methyl tert-butyl ether Methyl t-butyl ether

Diphenyl ether diphenyl oxide

Figure 3.7  Structures and alternative names of some ethers.

­Aldehyde

O

O

O

H H

Phenylethyl i-amyl ether Anther ®

Methyl cedryl ether

Floral, rose, chamomile

Cedarwood

Ambrox ® Ambrofix ® Ambroxan® Amberlyn ® Ambergris

Figure 3.8  Some ethers of importance as perfume ingredients.

compounds discussed in this chapter. Nonetheless some ethers are important in perfumery. The geranium‐scented diphenyl ether, shown in Figure 3.7 is one, and some others are shown in Figure 3.8 together with their odour types.

­Aldehydes When carbon forms a double bond to oxygen, the resulting bond is polarised more strongly than the single C─O bond of an alcohol, which opens up a huge range of interesting chemistry. The C═O bond is referred to as a carbonyl group. A variety of classes of compounds containing carbonyl groups exists; the differ­ ences between them depend on what else is attached to the carbonyl carbon atom. If the carbonyl carbon atom also carries a hydrogen atom, the compound belongs to the class of aldehydes. Aldehydes are very important in perfumery as they often have intense odours. The best known of all fragrances, Chanel No. 5, owes its famous top note to aldehydes. Its launch in 1921 started a fashion trend for this top note, and the family of aldehydic fragrances was born, taking their name from the class of ingredients responsible. In systematic nomenclature, aldehydes are named in the same way as alcohols but using the suffix ‐al instead of ‐ol. Since there must always be a hydrogen atom attached to the aldehydic carbon atom, the aldehyde group must always be at the end of a chain, and so no number is necessary to identify its location. Aldehydes are also referred to by a name derived from the corresponding acids (q.v.) using the acid name stem and the suffix ‐aldehyde. Some simple aldehydes are shown in Figure  3.9 together with systematic and trivial names for them. When the aldehyde group is attached to a ring, it is often named from the ring using the term carboxaldehyde to refer to the –CHO group. Some examples of perfumery aldehydes are shown in Figure 3.10. Straight‐ chain aliphatic aldehydes have an odour that is a combination of fatty and fresh air/marine, giving the classic ‘aldehydic’ character. Bourgeonal® and related aldehydes have a fresh smell reminiscent of lily of the valley (muguet). Aromatic aldehydes often have sweeter smells, such as the hawthorn charac­ ter of anisaldehyde, while alicyclic aldehydes like Ligustral® have intense green notes.

45

46

3  Carbon 2 – Heteroatoms O O

O Ethanal Acetaldehyde

3-Methylbutanal iso-Valeraldehyde i-Valeraldehyde

2-Phenylethanal Phenylacetaldehyde O

O

Cyclohexylmethanal Cyclohexane carboxaldehyde

Phenylmethanal Benzaldehyde

Figure 3.9  Structures and names of some simple aldehydes. O

Undecanal aldehyde C11 undecylic

O

10-Undecenal aldehyde C11 undecenyl 2-Methylundecanal methyl nonyl acetaldehyde aldehyde MNA O

O

O

O O

3-(4-t-butyl)Phenylpropanal Bourgeonal®

4-Methoxybenzaldehyde 1,3-Dimethylcyclohex-1-ene-4-carboxaldehyde anisaldehyde Ligustral®

Figure 3.10  Some important perfumery aldehydes.

­Ketones If the carbon atom of a carbonyl group is attached to two other carbon atoms, the functional group is called a ketone, and the compound belongs to the family of ketones. Ketones have some chemical properties in common with aldehydes, but there are a few important differences as we shall see later. Systematic names of ketones are derived from those of the parent hydrocarbon using the suffix ‐one and a number to indicate which carbon is the carbonyl carbon. A semi‐systematic nomenclature uses the names of the two radicals either side of the carbonyl group, followed by the word ketone. Hence, as shown in Figure 3.11, 5‐methyl­ hexan‐2‐one is often also called methyl isoamyl ketone. Sometimes the semi‐sys­ tematic name is easier to use as is the case with diphenyl ketone, also shown in Figure 3.11. But for this material, the trivial name of benzophenone is the one most commonly used in perfumery or other chemical literature.

­Carboxylic Acid O 5-Methylhexan-2-one 5-Methyl-2-hexanone methyl iso-amyl ketone O 1,1-Diphenylmethanone diphenyl ketone benzophenone

Figure 3.11  Naming of simple ketones. O

O

O

cis-Jasmone jasmine

5-Methylhept-2-ene-4-one Filbertone Hazelnut

α-Ionone Violet

O

O O

H Methyl cedryl ketone cedarwood

2-Heptanone methyl amyl ketone fruity, spicy

Camphor camphorwood

Figure 3.12  Some ketones of importance in perfumery.

Methyl ketones often have fruity smells. Other ketones have varied odour types including nutty, woody, violet, jasmine, and sweet. Some ketones with large globular molecules possess woody or amber odour types. Ketones with about 10 carbon atoms and a globular shape are often camphoraceous in odour. Figure 3.12 shows some ketones that are used in perfumery.

­Carboxylic Acids When a carbonyl carbon is attached by a single bond to a second oxygen atom and that oxygen atom is in turn bonded to a hydrogen atom, the resulting com­ pound is known as a carboxylic acid. The systematic names use the suffix ‐oic and the word acid. Obviously, the acid group must always be at the end of a chain, so a number is not needed to identify its position. If an acid group comes at both ends of a chain, the compound is a dioic acid. The trivial names for the acids are

47

48

3  Carbon 2 – Heteroatoms

Table 3.1  Some carboxylic acids. Number of carbons

Systematic name

Trivial name

Odour type

1

Methanoic acid

Formic acid

Acidic, irritant

2

Ethanoic acid

Acetic acid

Vinegar

3

Propanoic acid

Propionic acid

Sour milk

4

Butanoic acid

Butyric acid

Rancid butter

4

2‐Methylpropanoic acid

iso‐Butyric acid

Sour, buttery

5

Pentanoic acid

Valeric acid

Pungent sweaty

5

3‐Methylbutanoic acid

iso‐Valeric acid

Stale sweat

6

Hexanoic acid

Caproic acid

Rancid, sweat

7

Heptanoic acid

Pelargonic acid

Sweat, fatty or oenanthic acid

7

(E)‐3‐Methylhex‐2‐enoic acid

Schizophrenic acid

Stale sweat

8

Octanoic acid

Caprylic acid

Rancid, sweaty

11

10‐Undecenoic acid

Undecylenic acid

Oily, peach, sour

13

Tridecanedioic acid

Brassylic acid

Odourless

in much more common use than the systematic ones, and it is worth learning them as derivatives of the trivial names that are used extensively in other classes of compounds. Table 3.1 lists some of the more common carboxylic acids of rel­ evance to perfumery, together with their odour types that are obviously not in the category of pleasant odours. Many of these acids change their odour charac­ ter depending on dilution. The concentrated acid is usually unpleasant and sweaty or rancid, but, on dilution, they become less unpleasant and often display fruity character at high dilution. The structures of some of these acids are shown in Figure 3.13. The acid known as schizophrenic acid is an important component of the odour of stale human O O

OH

OH

O

O

OH Acetic acid

OH

Schizophrenic acid O

Brassylic acid OH O

OH

Benzoic acid

Cinnamic acid

Figure 3.13  Some carboxylic acids.

O OH Phenylacetic acid

­Ester

sweat. Its name comes from the fact that it was first isolated from the sweat of schizophrenics, but it is, in fact, a component of all stale human sweat. It is not present as such in fresh sweat but is bonded to a protein in sweat. The action of bacteria on the protein releases the free acid that makes a significant contribution to the overall odour that we all recognise clearly as stale sweat. One of the uses of perfume, particularly in the past, is to cover up this odour, and much research has been invested in finding ‘deodorant’ perfumes that will either cover or help pre­ vent it. Brassylic acid is important as a raw material for the manufacture of the musk ethylene brassylate, which is an example of the next class of compounds we will discuss, the esters. Only a handful of acids are used in perfumes. The most important of these are benzoic acid and cinnamic acid, which occur in benzoin, and phenylacetic acid, which has a strong odour reminiscent of honey.

­Esters When a carbon atom is attached to two oxygen atoms, one by a double bond and the other by a single bond, and the latter is also attached to another carbon atom, the compound containing this function is known as a carboxylic ester, or ester for short. Esters are most easily thought of as being formed from an alcohol and a carboxylic acid with loss of one molecule of water, as shown in Figure 3.14 for the formation of ethyl acetate from ethanol and acetic acid. Esters of other types of acids are also possible, for example, sulfate esters from sulfuric acid and phos­ phate esters from phosphoric acid. Normally when the term ester is used without any qualification, the intended meaning is a carboxylic ester. When other esters are intended, the type is usually specified. Esters are named in the same way as we have just seen for ethyl acetate. The name is in two parts: the first word is derived from the radical from which the alcohol is formed, and the second is derived from the acid, replacing the suffixes ‐ic or ‐oic by ate. Thus ethyl acetate is the ester formed from ethanol and acetic acid. The ester function is usually associated with fruity odours. If the ester group occurs in a freely accessible part of the molecule, then the compound will prob­ ably have some fruity character in its odour profile. Some esters used in perfum­ ery are shown in Figure 3.15. It is possible to have more than one type of functional group in a molecule. Methyl jasmonate is as an example as it has both a ketone and an ester group. If the ester group forms part of a ring, that is, if the acid and alcohol groups that form it are also attached by another set of bonds in the same molecule, then the ester is known as a lactone. Greek letters are used to indicate the size of the ring by indicating the position of the alcohol group relative to the O

+ OH Ethanol

HO

O

Acid or alkali As catalyst

Acetic acid

Figure 3.14  Formation of ethyl acetate.

O Ethyl acetate

+

H2O Water

49

50

3  Carbon 2 – Heteroatoms O O

O

O

O Amyl acetate fruity, pear, banana

O Benzyl acetate fruity Geranyl acetate fruity, rosy

O O

O O

O

Methyl jasmonate jasmine

O O

iso-Bornyl acetate fruity and woody

γ-Undecalactone peach

Figure 3.15  Some esters used in perfumery.

carbonyl. The most common lactone ring sizes are five‐ and six‐­membered rings, which are called γ‐lactones and δ‐lactones, respectively, as the hydroxyl groups are attached to the third and fourth carbons after the carbonyl group. An exam­ ple is shown in Figure 3.15. The name of the compound illustrated is γ‐undeca­ lactone. The letter γ‐ indicates a five‐membered ring, and the stem undeca‐ indicates the total chain length of 11 carbon atoms. γ‐Lactones are sometimes also called butenolides. Large ring lactones are particularly important in the musk odour area, as will be seen later.

­Acid Anhydrides and Chlorides Removal of one molecule of water from two molecules of a carboxylic acid pro­ duces an acid anhydride, or anhydride, for short. Anhydrides are reactive mole­ cules and are not used as such in perfumery, but are important as intermediates in the production of fragrance ingredients, such as esters. If one of the acid groups of an anhydride is replaced by a chlorine atom, the compound is called an acid chloride. Acid chlorides are usually prepared from the corresponding acid by treatment with thionyl chloride (SOCl2) or a chloride of phosphorus. The chemistry of anhydrides and chlorides are similar, and the two types of com­ pounds are used in the same way as intermediates. Figure 3.16 shows the prepa­ ration and structures of anhydrides and chlorides.

­Acetals and Ketals An alcohol function can add across the double bond of an aldehyde or ketone to produce compounds called hemiacetals and hemiketals, respectively. If the

­Acetals and Ketal O

O

O +

R OH

R

O

R

R

Acid

Acid

+

H2O

O

HO

Water

Acid anhydride

O

O

SOCl2 or PCl3 or PCl5

R OH

R CI Acid chloride

Acid

Figure 3.16  Acid anhydrides and chlorides.

R′

H

R

O

R″ O Ketone

H

R′

ROH R

O OH

ROH

R′ R

R″ O OH

Hemiketal

H O O

R

Acetal

Hemiacetal

Aldehyde R′

R′

ROH

R′

ROH R

R″ O O R Ketal

Figure 3.17  Formation of acetals and ketals.

hydroxyl group of a hemiacetal of hemiketal is replaced by a second molecule of alcohol, the resulting compounds are called acetals and ketals, respectively. Acetals and ketals can be formed using two molecules of the same alcohol or one each of two different alcohols. Figure  3.17 shows how acetals and ketals are derived. Acetals and ketals are named from the parent carbonyl compound and the alcohols that they contain. The examples shown in Figure  3.18 will suffice to illustrate the basic principle. Hemiacetals and hemiketals form spontaneously from aldehydes/ketones and alcohols under either acid or base catalysis. Acidic conditions are required to either form or break down acetals and ketals. Their stability in alkaline conditions makes acetals and ketals very useful as perfumery ingredients. Cyclic acetals and ketals are derived from diols and are particularly stable as will be seen in later chapters. They can be named either in the same way as for open chain analogues, or, more commonly, they are named based on the hetero­ cyclic ring systems that they contain. Some acetals and ketals of importance in perfumery are shown in Figure 3.18. The simplest of these is phenylacetaldehyde dimethyl acetal, commonly known by the acronym PADMA, in which both alco­ hol residues are the same. Efetaal® is formed from an aldehyde and two different alcohols. Herboxane® is an example of a cyclic acetal formed from a diol and an

51

52

3  Carbon 2 – Heteroatoms

O

O

O

O Phenylacetaldehyde dimethyl acetal

Efetaal® Acetaldehyde ethyl phenylethyl acetal

O O Herboxane® Pentanal 2-methylpentan-2-4-diol acetal 2-Butyl-4,4,6-trimethyl-1,3-dioxane

O

O

Ysamber K®

Figure 3.18  Some acetals and ketals of use in perfumery.

aldehyde. Three names are shown for it. The first is the trade name, and the sec­ ond identifies it as an acetal and shows the aldehyde and diol from which it was formed. The third name is derived from the heterocyclic ring system of the mol­ ecule, in this case 1,3‐dioxane. More detail about heterocyclic systems will be given later in the chapter. Ysamber K® is an example of a ketal. It is formed from a ketone and in this case, a simple diol, ethylene glycol, or 1,2‐ethanediol. The systematic names for Ysamber K are rather cumbersome, and so it is usually referred to by its trade name.

­Peroxy Compounds Peroxy compounds are those containing a single bond between two oxygen atoms. The peroxide bond is weak and breaks easily, liberating reactive species that can lead to chain reactions, as will be seen in a later chapter. There are three types of peroxy species that are important in perfumery, as shown in Figure 3.19. O

R

O

R

O

O

H

Hydroperoxide

R

Peroxide

O R

O

OH

Peracid

Figure 3.19  Peroxy functions of importance to perfumery.

­Nitrogen–Amines and Ammonium Salt

­Nitrogen–Amines and Ammonium Salts Many different possible types of nitrogen‐containing compounds exist in organic chemistry, but we only need to consider a few of them for the purposes of this book. Those compounds of greatest importance in the perfumery business are described below. In organic chemistry, the most common valence state of nitrogen is that which forms three covalent bonds and has one lone pair of electrons. If the three bonds are to hydrogen, the resultant compound is ammonia, NH3. If the nitrogen atom is attached to carbon molecules, the resultant species are called amines. If the nitrogen is attached to just one carbon atom, the amine is said to be primary; if it is attached to two, the amine is secondary; and if it is attached to three, the amine is tertiary. These basic structures are shown in Figure 3.20. The lone pair of elec­ trons of nitrogen is more reactive than those of oxygen, and amines can be pro­ tonated easily to form ammonium salts. They can also be alkylated to produce what are known as quaternary ammonium salts, in which the nitrogen atom is attached to four carbon atoms and carries a positive charge, becoming a cation. Naturally, it always contains an accompanying anion in order to balance the charge. The nature of the anion will depend on how the salt was produced. In the perfumery, cosmetics, soap, and detergent businesses, quaternary ammonium salts are often known as ‘quats’. These are very important as surface‐active agents as will be seen in a later chapter. Amines usually have unpleasant, fishy odours. Quaternary ammonium salts are usually odourless. The hydrogen and/or carbon atoms attached to nitrogen in amines, together with the lone pair of electrons on the nitrogen atom, sit approximately at the corners of a tetrahedron just as do the four hydrogen atoms of methane. However, there is a significant difference. The ammonia molecule can invert by going through a planar intermediate, and so nitrogen atoms do not form chiral centres as do appropriate carbon atoms, except when they are locked in a ring system or are quaternary ammonium salts with four different radicals attached to the nitrogen atom. This phenomenon is shown in Figure 3.21. On the top line of the R

R

NH2

Primary amine

H N

Secondary amine

R

R R

N

R

R N

R

Tertiary amine

R

+

X–

Quaternary ammonium salt

R

Figure 3.20  Structures of amines and ammonium salts.

53

54

3  Carbon 2 – Heteroatoms R′

R′ R″

R′ N

N R″

R′′′

R′′′

R′

R′ +

R″

R″ R′′′

N R′′′′

R′′′

R′′′′

+

N

R″ R′′′

Figure 3.21  Inversion of amines.

figure, we see how a nitrogen atom attached to three different substituents can flip from one tetrahedral configuration to its mirror image and back again. In the lower line, the quaternary ammonium ion cannot do this as the planar interme­ diate is impossible. Thus, a quaternary ammonium ion containing four different substituents is, like a chiral carbon atom, locked into one enantiomeric form.

­Nitrogen–Imines, Schiff’s Bases, and Enamines When a primary amine comes in contact with an aldehyde or ketone, the nitro­ gen atom of the amine forms a bond to the carbonyl carbon atom, giving an α‐ hydroxy amine that can eliminate a molecule of water in one of two ways. One way results in a compound with a double bond between what was the carbonyl carbon atom and the nitrogen atom. Such compounds are known as imines or Schiff ’s bases. The other possibility is for the product to contain a nitrogen atom bonded to an olefinic carbon atom. These compounds are known as enamines. Which compound is formed will depend on the nature of the aldehyde or ketone and on the reaction conditions. Secondary amines can form enamines but not imines, and tertiary amines can form neither. A general scheme for the formation of imines and enamines is shown in Figure 3.22. Imines and enamines are useful as intermediates in synthesis of perfumery ingredients. As shown in Figure 3.22, if a hydrogen atom is on the carbon atom adjacent to the carbonyl carbon of the aldehyde that formed the enamine or Schiff ’s base, then the enamine and Schiff ’s base are isomers of each other, and simple movement of a hydrogen atom through acid/base interactions will interconvert one to the other. Such isomers are known as tautomers, and in consumer products where water and an acid or base present (and even a glass surface would serve as an acidic catalyst in this instance), an equilibrium will be set up between the two species. If the nitrogen atom of an imine is attached, not to another carbon atom but to the oxygen atom of a hydroxy group, the compound is called an oxime. There are a few oximes in use in perfumery, and they usually have very intense odours. Imines, better known in perfumery as Schiff ’s bases, are of great value as they are essentially masked aldehydes and ketones. When used in perfumery, the amine component is almost always methyl anthranilate. In fact, when perfumers

­Nitrogen–Amides/Peptide R R

O

H2N

R

NH2OH

R

H2N

–H2O

R

–H2O

R R

R

–H2O

N

R

OH

Oxime

R R

N H

R

Enamine

Imine or Schiff’s base

R

R R

N

N

R

H

R

N H

R

Figure 3.22  Imines (Schiff’s bases), enamines, and oximes.

talk of a Schiff ’s base, they will almost certainly be referring to an imine formed from methyl anthranilate and a fragrant aldehyde. Such Schiff ’s bases are more chemically stable than the aldehyde and, because of their higher molecular weight, are less volatile and more substantive. In use, they break down to release the aldehyde and methyl anthranilate. This stability means that aldehydic notes can be obtained in applications where the free aldehyde would not survive. But methyl anthranilate has a very distinctive heavy, sweet floral odour, so the nega­ tive aspect of this method of protecting aldehydes is that the final fragrance will have the methyl anthranilate note present whether it is desired or not. An exam­ ple of a fragrance that achieved success using a Schiff ’s base is Giorgio, which combines Helional® with methyl anthranilate. Helional has a soft, fresh floral odour that blends well with the sweetness of methyl anthranilate. However, if aldehydes such as Ligustral are used to form the Schiff ’s base, there will be a dis­ cord in that the sweetness of methyl anthranilate will clash with the green note of Ligustral. These two Schiff ’s bases are shown in Figure 3.23. Of course, as will be seen in a later chapter when the mechanism of olfaction is discussed, the odour of a mixture is often different from the sum of the odours of the compo­ nents of that mixture. Furthermore, discords can be used in an artistically attrac­ tive way in perfumery just as they are in music, and Dior’s Poison is a good example of this dissonance.

­Nitrogen–Amides/Peptides Reaction between a primary or secondary amine and a carboxylic acid gives a product known as an amide or peptide. The amide link is the basis of protein structure and so is of vital importance in living organisms. The formation of an amide is illustrated in Figure 3.24. The amide bond is highly polarised, as shown

55

56

3  Carbon 2 – Heteroatoms O O NH2 Methyl anthranilate

O

O

O

O Ligustral®

Helional®

O

N

O

N O

O

O

O

Figure 3.23  Two Schiff’s bases. O + R

OH

O H2N

R′

δ– O

O R

N H

R′

N H

R

R

R′

O– δ+ R′ N H

R

+ R′ N H

Figure 3.24  Formation and canonical forms of an amide.

in the lower part of Figure 3.24, and one effect of this polarisation is to make it essentially planar and rigid. The lower part of Figure 3.24 shows two ‘canonical forms’ of an amide. Canonical forms are those in which all valences are satisfied or else there are charges on atoms to indicate an excess or deficit of electrons. Thus, in one of the canonical forms, the oxygen atom has only one bond but car­ ries a negative charge, whereas the nitrogen atom has four bonds and thus carries a positive charge. In between the two canonical forms is a structure showing par­ tial charges, which is more representative of what actually exists in the molecule.

­Nitrogen–Nitriles When a carbon atom is joined to a nitrogen atom by a triple bond, the compound is known as a nitrile or cyanide. The latter name stems from the fact that nitriles can be synthesised from halides by reaction with the cyanide anion. Thus, methyl

­Nitrogen–Nitro Compound

N

Citronellyl nitrile lemon

N

Hypo-Lem® lemon

N

Frutonile® peach

Figure 3.25  Some perfumery nitriles.

chloride and sodium cyanide would give methyl cyanide, better known as ace­ tonitrile. The acetonitrile name is derived from that of the corresponding acid and illustrates the general rule for naming nitriles. To name a nitrile, one takes the name of the acid with the same structure and replaces the ‐oic acid suffix by ‐onitrile. However, many nitriles are given trivial names that often look decep­ tively like systematic names. Nitriles are important in perfumery as they often have odours similar to those of the corresponding aldehydes but are much more chemically stable. For example, decanonitrile has an odour similar to that of decanal. Some nitriles of use in perfumery are shown in Figure 3.25. Citronellyl nitrile and Hypo‐Lem® (3,7‐dimethyloctanonitrile) are useful because of their lemon notes, similar to those of citral. Citronellyl nitrile is an example of a name that is not strictly correct. Based on citronellic acid, the nitrile should be called citronellonitrile; however the name citronellyl nitrile is the one in common use.

­Nitrogen–Nitro Compounds The last class of nitrogen‐containing materials we need to discuss are the nitro compounds. The nitro group cannot be represented in simple terms using only covalent bonds between neutral atoms. The nitrogen atom has to be written with a positive charge, and one of the oxygen atoms has to carry a negative charge. These are canonical forms, and in reality, the two oxygen atoms share the nega­ tive charge. More usually, the nitro group is simply written as –NO2. Both repre­ sentations of nitrobenzene are shown in Figure 3.26. The family known as nitro O–

+ N

O

O NO2

O2N

NO2

Alternative representations of nitrobenzene Musk ketone

Figure 3.26  Nitrobenzene and musk ketone.

57

58

3  Carbon 2 – Heteroatoms

musks were once important in perfumery, but these materials are now essentially obsolete because of hazards in manufacturing and, in some cases, in use also. Musk ketone is one example of a nitro musk, and its structure is also shown in Figure 3.26.

­Sulfur Sulfur comes under oxygen in group six of the periodic table, so its chemical properties might be expected to resemble those of oxygen, and to some extent they do. So, for instance, sulfur will form thiols and thioethers that are counter­ parts of alcohols and ethers, respectively, in which the oxygen atom is replaced by sulfur. However, sulfur is different in that it can be oxidised and, in doing so, forms more bonds to oxygen atoms than an oxygen atom could do. In this book, we will only deal with those compounds of sulfur that have most relevance to the fragrance business. We can group sulfur compounds into three classes depend­ ing on the number of bonds that the sulfur atom makes to oxygen atoms. At the lowest level, the sulfur atom forms two bonds to other atoms; hence we can call it divalent sulfur. Some classes of compounds in this category are shown in Figure 3.27. Compounds with a sulfur atom bonded to one carbon and one hydrogen atom are known as thiols or mercaptans. The first name is derived from the word alcohol with the prefix thi‐ (from the Greek word for ‘sulfur’  –  theion) being used to indicate sulfur. The second is derived from Latin, mercurius captans, which means mercury capturing, a reference to the affinity of these materials for heavy metals such as mercury. The thiol group is also associated with very strong smells. For example, the smell of rotten eggs is due to hydrogen sulfide, while methanethiol, also known as thiomethanol, is one of the key components R

SH

Thiol or mercaptan

R

S

Thioether R′

R

S

Disulfide S

R′

O Thioester or thiol ester

R S

R′

Figure 3.27  Some classes of divalent sulfur compounds.

­Sulfu O H3C

S

CH3

Figure 3.28  Dimethyl sulfoxide.

of the smell of bad breath. The defensive secretion of the skunk and the warning smell in natural gas are also based on thiols. Conversely, some thiols have pleas­ ant smells at very high dilution. For example, thioterpineol is the key material responsible for the aroma of grapefruit, where it is present at very low concen­ tration. Some disulfides are organoleptically important components of essential oils, but always at very low levels, as they also tend to smell unpleasant at higher concentrations. Disulfides are very different in properties from peroxides, their oxygen counterparts. The single bond between two oxygen atoms is weak, mak­ ing peroxides potentially explosive. However, sulfur atoms are quite happy to bond to each other, and the bond is reasonably strong. Disulfides are particu­ larly important in protein chemistry as will be seen in later chapters. The sulfur atom of a thioether can form a double bond to an oxygen atom using one of its lone pairs of electrons. The resulting compound is called a sul­ foxide, and the structure of a typical sulfoxide, dimethyl sulfoxide, or DMSO for short, is shown in Figure 3.28. DMSO is used as a solvent but not in perfumery. Sulfoxides are essentially odourless, though traces of strong smelling sulfurous impurities often give the impression that they do have an odour. Sulfoxides, as a class, are of minimal importance in perfumery, and the only reason for including them here is that they are odourless, so a malodour arising from a thioether can be destroyed by oxidising it to the corresponding sulfoxide. Of more importance to the perfumery industry are sulfur compounds in which both lone pairs of the sulfur atom are used to form double bonds to oxygen atoms. Some typical classes of materials in this category are shown in Figure 3.29. In terms of shape, the sulfur atom in all of the materials described in this chapter can be considered to be tetrahedral, with the corners of the tetrahedron being occupied by carbon, hydrogen, or oxygen atoms or a lone pair of electrons, depending on the oxidation state of the sulfur atom. Like sulfoxides, sulfones are not of relevance to the fragrance industry except as the oxidation products of thioethers and hence as a by‐product from malo­ dour removal. Sulfonic acids and their derivatives and sulfuric acid are much important to us. In sulfuric acid, the sulfur atom is bonded only to oxygen atoms, whereas sulfonic acids contain one sulfur to carbon bond. The structure of sulfuric acid is therefore represented by the structure for a sulfate ester shown in Figure 3.29, but in which both R and R′ are hydrogen atoms. Sulfuric acid and sulfonic acids are strong acids. The meaning of the term strong when applied to acids will be explained in Chapter 8. Like carboxylic acids, they can form esters with alcohols, and generic structures for sulfonate and sulfate esters are shown in Figure 3.29. As with carboxylic acid chlorides, sulfonyl chlorides are very reactive species and are used only as intermediates in the preparation of other materials, sulfonate esters, for example. Sulfates in which both R and R′ are alkyl radicals are quite reactive materials and are used mainly as reagents in

59

60

3  Carbon 2 – Heteroatoms O

O S

R O

O S

R O

O

O

R′

Sulfonate ester

R′

Sulfate ester

O S

O

Sulfonyl chloride

Cl

O S

R

Sulfonic acid

O

O

R

OH

S

R

Sulfone

R′

O

Figure 3.29  Classes of compounds with hexavalent sulfur.

O

O H3C

S

OH

Methanesulfonic acid

O

O S

OH

p-Toluenesulfonic acid

Figure 3.30  Two sulfonic acids.

synthesis. Half esters of sulfuric acid, in which one R is an alkyl radical and the other is hydrogen, are less reactive except as acids, and some categories of salts of these are used as surfactants in a variety of consumer goods. More detail will be given in Chapter 11. Salts of some sulfonic acids are also used as surfactants. The two sulfonic acids most commonly encountered as acid catalysts are meth­ anesulfonic acid and p‐toluenesulfonic acid. The structures of these are shown in Figure 3.30.

­Heterocyclic Compounds So far, we have looked only at heteroatoms in open chain, or acyclic, structures, but, of course, it is also possible to incorporate atoms other than carbon into rings. Materials with a ring structure containing heteroatoms are known as het­ erocyclic compounds. The names for heterocyclic compounds are usually based on the names of a basic ring structure. Too many of these compounds exist to give a comprehensive list here, and such a list would not be warranted anyway as

­Heterocyclic Compound Epoxide or oxirane

O

R

O

O

R′ O

R

Glycidate

S

O

O Furan

O

Thiophene

Pyran

O O

1,3-Dioxolane

O

1,3-Dioxane

O

S

1,3-Oxathiane

O O

O

1,3,5-Trioxane

O

O Chroman

Isochroman

Figure 3.31  Some O‐ and S‐containing heterocyclic rings.

we will only come across a small percentage of them in this book. The basic het­ erocyclic ring systems that are most frequently encountered in perfumery are shown in Figures 3.31 and 3.32. Figure 3.31 shows those compounds containing only oxygen and/or sulfur atoms. A three‐membered ring containing one oxygen atom and two carbon atoms is known as an epoxide or an oxirane. Sometimes the term oxide is also used. Thus, the simplest member of the series is known as ethylene oxide, ethylene epoxide, epoxy‐ethylene, or oxirane. A special class of epoxides that are particularly important in perfumery is known as glycidates. They are basi­ cally ethylene oxide with a carboxylic ester function attached to one of the carbon atoms and one or two substituents on the other carbon. They are pre­ pared by a reaction known as the Darzens reaction and usually have intense fruity odours. Figure 3.31 shows the basic structures for furan, thiophene, and pyran rings. The saturated versions of these ring structures are also common and are known simply as tetrahydrofuran, tetrahydrothiophene, and tetrahydropyran, respec­ tively. The next row down in the figure shows the structures of 1,3‐dioxolane and 1,3‐dioxane, which we have already encountered above under the heading acetals and ketals. However, we now also meet the 1,3‐oxathiane structure, which is the same as that of 1,3‐dioxane, except that one of the oxygen atoms has been replaced by a sulfur. In numbering the positions of atoms and substitu­ ents around heterocyclic rings, we must always start with the atom with the highest atomic number. So, in the case of 1,3‐oxathiane, the number one refers to the sulfur and the three to the oxygen. In the bottom row, the basic unit of

61

62

3  Carbon 2 – Heteroatoms H N

H N

H N

Pyrrolidine

Pyrrole

N H Piperidine

N

Indole

N

Pyridine

Quinoline

N HN

N Isoquinolene N

NH

Imidazole

O

N Pyrazine

N

S

Thiazole

Oxazolidine HN

S

Thiazolidine

Figure 3.32  Some N‐containing heterocyclic rings.

1,3,5‐­trioxane is shown. This example is important because aldehydes will form trimeric structures of this type when exposed to acidic conditions, and it there­ fore represents a mechanism whereby aldehydes can be lost from fragrance compositions. Sometimes when a benzene ring is fused to a heterocyclic ring, the name of the resultant polycyclic compound is formed by adding the prefix benzo‐ to the name of the heterocyclic ring. For example, benzofuran indicates a benzene ring fused to a furan. In other cases, the polycyclic system has a name of its own, and two examples of importance to perfumery, chroman and isochroman, are shown in Figure 3.31. Figure 3.32 shows some ring systems containing nitrogen that are of impor­ tance in the perfume and flavour business. Pyrrole and pyridine rings display aromaticity as does benzene, but in the case of pyrrole, since two of the six π‐electrons come from the lone pair of the nitro­ gen, the chemistry of pyrrole is rather different from that of benzene. The nitro­ gen atom in pyridine has a lone pair of electrons that is not involved in the bonds of the ring, so pyridine is a base as will be seen in later chapters. Oxazolidine and thiazolidine, like acetals and ketals, contain a masked carbonyl compound. The carbon between the two heteroatoms is a carbonyl function, and in the case of the oxazolidine and thiazolidine molecules shown in Figure  3.32, the hidden aldehyde is formaldehyde (methanal). In later chapters, we will see how this can be used. Imidazole and thiazole, as well as their derivatives, have very interesting chemical properties that nature uses to good effect in biochemical reactions. Again, this use will be seen in a later chapter.

­Heterocyclic Compound

With heterocyclic systems, as with carbocyclic and acyclic ones, the class of compounds often takes the name of the simplest member of the class, for exam­ ple, phenol/phenols, pyridine/pyridines, and indole/indoles. Figures 3.33–3.40 will serve as quick references to enable the reader to identify and name functional groups. Words in brackets are often omitted in common use.

R

C H2

O

H

R′

O

Primary alcohol

O

R

R H O C H

Secondary alcohol

R O

R′

Tertiary alcohol

H

R″

O

Ketal

OH O O

R

R′

Ether

R

R′

O

R

R″

H O

R

Phenol

Acetal

R′″

O

R′ R

R″

Hemi-acetal

R′

H O

Hemi-ketal

R

Figure 3.33  Functional groups with carbon–oxygen single bonds.

O

O Aldehyde

O

O R

(Carboxylic) acid OH

O (Carboxylic) (acid) anhydride

O

R′ O

O R

R

Ketone

R′

R

R

O

R′ (Carboxylic) (acid) ester

R O

Lactone

R O R

(Carboxylic) (acid) chloride C1

Figure 3.34  Functional groups with carbon–oxygen double bonds.

63

64

3  Carbon 2 – Heteroatoms

R

R

O

O

O

H

Hydroperoxide

R′

Peroxide

O

O O

R

Peracid

OH

Figure 3.35  Functional groups with oxygen–oxygen single bonds. SH

R

S

R

R

Thiol or mercaptan

S

R′

S

Thioether

R′

Disulfide

O S

R

Thioester

R′

Figure 3.36  Functional groups containing divalent sulfur. O R

S

R′

Sulfoxide

Figure 3.37  Functional groups containing tetravalent sulfur. O

O R

S

Sulfone

R′

R

O

S

OH

Sulfonic acid

Cl

Sulfonyl chloride

O O

O R

O

S

R′

R

S O

O

O

O

Sulfate ester R

S

O

R′

Sulfonate ester

O

Figure 3.38  Functional groups containing hexavalent sulfur.

­Heterocyclic Compound

R

R

NH2

H N

Primary amine

Secondary amine

R′

R N

R′

Tertiary amine

R″

R

R′ + N

Quaternary ammonium salt

R″

R″′

Figure 3.39  Functional groups with carbon–nitrogen single bonds.

R R′ R

Nitrile or cyanide N

R″ O

Imine or Schiff’s base R

N

R′

R″

R′ R

N

N H

R″

Amide or peptide O

Enamine R

N

+ –

O

Nitro

Figure 3.40  Other nitrogen‐containing functional groups.

65

66

3  Carbon 2 – Heteroatoms O

O OH Mefrosol®

Jasmatone®

O Ortholate

O

O

O

Hedione®

N

O

N

O Traseolide®

Aurantion®

N

Jasmacyclene®

OH

Sandela®

Exaltolide®

Boronal®

Precyclemone B®

OH

O

O O

OH Vanillin

Mayol®

O

O

Frescile®

O

Gardamide®

O Aldehyde C18®

O

Rose oxide

O

OH O

O

O

O

®

O

Nerolin Bromelia®

HO N

Buccoxime®

Figure 3.41  Selected fragrance ingredients.

Review Question 1 Which functional groups are present in the selection of fragrance ingredients shown in Figure 3.41?

67

4 States of Matter Matter exists in three states: solid, liquid, and gas. We learned the basic macroscopic differences between them during childhood, but now we need to understand what is happening at the molecular level that causes the properties with which we are all so familiar.

­Solids In the solid state, matter retains its physical shape, unless of course it is subjected to a force sufficient to deform it. A piece of iron has a shape that it retains even if it is moved or placed in a container. It is a typical solid, and we all understand this. The reason that the iron retains its shape is that its atoms are all connected to each other by forces that hold them in place and maintain the distances between them within well‐defined limits. In the case of a block of iron, the forces are actually the chemical bonds between the iron atoms. In the case of an ionic solid such as a salt, the components of the solid are not atoms but ions. The forces holding them together are therefore electrostatic rather than chemical bonds. Since the ions carry electrical charges, these electrostatic forces are relatively strong. If we take common salt as an example, the ions are those of sodium and chlorine, positive sodium cations and negative chloride anions. The way these are held together on the molecular scale determines the shape of the crystals on the macroscopic scale. If you look at crystals of table salt, you will see that they are basically cubic in shape as the sodium and chloride ions pack together in a simple cubic lattice. Each sodium ion is surrounded by six chloride ions, equidistant in three‐dimensional space, and equally, each chloride ion is surrounded by six sodium ions. This arrangement gives a regular cubic crystal lattice as shown in Figure 4.1. Figure 4.1a shows part of a single plane in the crystal. For clarity, the ions in front of the plane of the paper and those behind it are not shown. Figure 4.1b is a representation of the basic unit of the three‐ dimensional structure. This is known as a unit cell of the crystal. This regular structure extends in all directions, and so the overall shape of a sodium chloride crystal is a cube. When in the solid form, neutral molecules can also crystallise into regular shapes through regular packing. In this case, the forces holding the molecules Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

68

4  States of Matter

Na+

Cl–

Cl–

Na+

Cl–

Na+

Na+

Cl–

Na+

Cl–

Cl–

Na+

Na+

Cl–

Na+

Na+

Cl–

Cl–

Na+

Na+

Cl–

Cl–

Na+

Cl–

Cl–

Na+

Cl–

Na+

Cl–

Na+

Cl–

Na+

Cl–

Na+

Na+

Cl–

Cl–

Na+

Cl–

Na+

Na+

Cl–

(a)

Na+

Cl– Na+

(b)

Cl–

Cl–

Na+

Figure 4.1  Crystal structure of sodium chloride.

together are the weaker forces of van der Waals interactions, hydrogen bonds, and other electrostatic interactions (weaker than those involving ions). For a description of hydrogen bonds, see Chapter  3. van der Waals interactions or forces are those due to the long‐range attraction that one molecule feels for another, similar to the force of gravity between heavenly bodies such as the Earth and the Sun. As an example of a crystalline solid formed from a neutral molecule, we can look at the structure and crystal packing of the compound shown in Figures 4.2– 4.4. This material belongs to a class known as chalcones, and for convenience we will refer to it by the trivial name of Biradex N. Figure 4.2 shows a simple two‐ dimensional representation of its molecular structure. Figure 4.3 is drawn from an X‐ray crystallographic study and shows that the ring marked A (in Figure 4.2) A O

B

O

O O

O

O

Figure 4.2  Structure of Biradex N.

 ­Solid

Figure 4.3  Actual shape of one molecule of Biradex N .

lies in the same plane as the double bond and the ketone group and the ring marked B in Figure 4.2 sits at right angles to this plane. We now have a good three‐dimensional picture of Biradex N. Note that in Figure 4.3, only the carbon and oxygen atoms are shown. In Figure 4.4, we can see how the individual molecules of Biradex N stack together in the crystal. This neat packing leads to a minimum of free space in the crystal and allows electrostatic interactions between the benzene rings and ether groups of adjacent molecules. Methanol was used as the solvent for preparation of these crystals, and the keen eye will be able to pick out molecules of methanol trapped in the crystal lattice. In fact, these methanol molecules will help to stabilise the crystal by forming hydrogen bond bridges between Biradex N molecules. The crystal packing of such substances, and hence the overall shape of their crystals, will depend on the molecular shape and the way in which the individual molecules can form hydrogen bonds, van der Waals interactions, and so on between each other. In turn, the shape of the crystal will depend on the way the molecules pack together. Equally, the molecular structure of a solid will affect its bulk physical properties. Carbon is a very nice example. Pure carbon can form two different crystalline solid forms and one solid form that is not crystalline. The non‐crystalline form is known as soot or carbon black, and its non‐crystalline structure results in its taking the form of fine particles. In one crystalline form, the carbon atoms are all sp3 hybridised and form strong covalent single bonds with four neighbouring carbon atoms. This structure extends in all directions, giving a very strong three‐dimensional structure that is not easily deformed. This material is

69

70

4  States of Matter

Figure 4.4  Crystal packing of Biradex N.

Figure 4.5  Unit cell of the diamond crystal structure.

known as diamond and is one of the hardest substances known. A unit cell of the diamond crystal structure is shown in Figure 4.5. The other crystalline form of carbon is known as graphite. In graphite, each atom is sp2 hybridised. The atoms bond together in an array that looks like an endless succession of fused benzene rings. These arrays take the form of flat sheets. Each atom is covalently bound to its neighbours in the sheet, but the bonds between sheets are weak non‐bonded interactions. This makeup enables one sheet to move easily across another, making graphite a good lubricant and

  ­Phase Change

Figure 4.6  Fragment of one sheet of graphite.

useful as a writing material in pencils. As the point of the pencil is dragged across paper, sheets of graphite are pulled off the main body and left as a black trace on the paper. Earlier, we considered the ring current of the electrons in a benzene ring. In graphite, the ring current extends across the whole sheet, which makes graphite a conductor of electricity. A fragment of one sheet of graphite is shown in Figure 4.6.

­Liquids Liquids have no shape of their own and will assume the shape of the container in which they are stored. The molecules making up the liquid interact weakly with each other and have no long‐range order, regular structure, or pattern of orientations of molecules. However, short‐range interactions between the molecules of a liquid will hold them together, and so a liquid will, under gravity, settle to the bottom of its container. Since the forces holding the molecules together are relatively weak, it is possible to separate them. A solid object dropped onto the surface of a liquid will penetrate it and fall to the bottom of the liquid, assuming it is denser than the liquid. The weak nature of the intermolecular forces in a liquid will also allow it to flow.

­Gases Gases are very mobile and will spread equally throughout the space allowed to them. No interaction occurs between the individual molecules making up a gas.

­Phase Changes Substances can be transformed from one state of matter to another. We all know that water is normally a liquid, but in winter or in a freezer, it will become solid and is known as ice. Similarly, when heated on a stove, the water in a pot will

71

72

4  States of Matter

6 C

Liquid 1

2

Pressure Solid

5

3

4

P Gas

Temperature

Figure 4.7  A simple phase diagram.

form steam, which is a gas, and evaporate. Of course, a gas close to the point of liquefaction is often referred to as a vapour. These examples illustrate the effect of temperature on states of matter. The other thing that affects state of matter is pressure. A heavy weight placed on a block of ice will melt it, even if it is as cold as the ice. Water boils at 100 °C at sea level, but on top of a mountain, where the air pressure is lower, it will boil at a lower temperature. The relationships between the solid, liquid, and gas phases of a substance can be presented in what is known as a phase diagram, as shown in Figure 4.7. In Figure 4.7, the temperature increases from left to right along the horizontal axis, and the pressure from bottom to top along the vertical axis. The three phases of matter for our test substance are shown in the figure, and the solid lines show the transition from one phase to another. The point P where the three lines meet is called a triple point, because at this point all three phases exist together. Point C is known as the critical point, and the temperature and pressure associated with it are known as the critical temperature and critical pressure, respectively. Above the critical point, no distinction exists between liquid and gas, and the substance is said to be in the supercritical state. The phase transition we are most familiar with is the transition from liquid to gas as a substance is heated, as in the boiling of water to form steam. This transition is represented by the line running from point 1 to point 2. Here the liquid is heated at constant pressure and passes the boiling point (at that pressure) to form a gas. If the pressure is first reduced from point 1 to point 3 and then the liquid heated, the boiling point is lower, and the substance enters the gas phase at a lower temperature as shown on the line from point 3 to point 4. This fact is very important to perfumery as we will see in the next chapter. Some molecules of a liquid will pass into the vapour phase even below the boiling point, and thus an equilibrium exists between the liquid and its vapour at all temperatures. The pressure exerted by the vapour is known as the vapour

  ­Two‐Phase System

pressure; and it is dependent on factors such as the molecular weight of the liquid and the strength of the attractive forces between the molecules. Energy is required to move a molecule from the liquid to the vapour state. This energy is known as the latent heat of vaporisation. The latent heat of vaporisation is dependent on the strength of the attractive forces between the molecules. Thus, a liquid such as hexane, whose molecules are only weakly attracted to each other, will have a low latent heat of vaporisation. On the other hand, water displays relatively strong hydrogen bonds between individual molecules and has a very high latent heat of vaporisation. Another familiar phase transition is that from solid to liquid and vice versa as temperature changes. Water placed in a freezer solidifies into ice cubes as its temperature falls below 0 °C. Similarly, when the ice cubes are removed and placed in a drink at room temperature, the ice will melt as its temperature rises past 0 °C on its way up to the ambient temperature. It is also possible to melt a solid such as ice by applying pressure. This process is shown in the line from point 5 to point 6. Here, the temperature is held constant, but as the pressure increases, the solid is converted to liquid. Just as with the liquid to vapour transition, energy is required to move from the solid to the liquid state, which is known as the latent heat of fusion. Again, as with the latent heat of vaporisation, the latent heat of fusion is dependent on the strength of the forces holding the molecules together in the solid as opposed to those in the liquid. The various phase transitions have names as follows. The transition from solid to liquid is called meting or fusion; from liquid to solid, solidification or  –  if appropriate  –  crystallisation; from liquid to gas, boiling, evaporation, or vaporisation; from gas to liquid, condensation; from solid to gas, sublimation; and from gas to solid, solidification.

­Two‐Phase Systems So far, we have been discussing pure substances. What happens if two different substances come into contact, assuming that there is no chemical reaction between them? If the two are gases, then they will mix to form a gaseous mixture in which the two starting gases are uniformly distributed. If they are both solids, they will not penetrate each other unless sufficient force is applied to shatter or cut one of them, but the fragments will remain distinct. If either of two substances brought into contact with each other is a liquid, then there are two possibilities. Either the two substances could remain distinct and separate, or they might become one single phase. To take an example from the kitchen, consider bringing various combinations of water, cooking oil, and salt into contact with each other. Salt added to water will dissolve in the water and become a single phase. Salt added to oil will remain as salt crystals giving a two‐phase mixture, solid salt and liquid oil. Water added to oil will give a mixture of two liquid phases. If we mix all three, we will obtain two liquid phases, one of oil and the other of salt in water. If the water is left to stand in the air, oxygen from the air will dissolve into it, giving a solution of a gas in a liquid. If we add ethanol (a common perfumery solvent) to water, the two will mix (unlike oil and water) to form a single phase.

73

74

4  States of Matter

­Solubility Solids, liquids, and gases can dissolve in a liquid to form a solution. The carrier liquid is called the solvent; the dissolved material is called the solute. A given liquid will be able to hold only a certain amount of a given solid, which is known as the solubility of that solid in that liquid. If a solution contains as much solute as possible, it is known as a saturated solution. If a solvent is allowed to evaporate from a solution, the amount of solid it contains will then exceed the solubility, as the volume of liquid has been reduced. A metastable state will then exist, and the liquid is known as a supersaturated solution. A trigger such as shock or contact with a new surface – such as dust falling into the liquid – will prompt precipitation of the solid. The solubility of a solid in a liquid is dependent on temperature, so precipitation or crystallisation of the solid will occur if the temperature is reduced. This precipitation provides a method of purification of substances as will be seen in the next chapter. The factor controlling whether or not a substance will dissolve in a given liquid is based on the polarity of the substance relative to that of the liquid. Polarity stems from the distribution of electrons in the molecules making up the substance. If the electrons are spread evenly around the molecule, there is no separation of electrical charge, and the molecule is said to be non‐polar. If the electrons are distributed unevenly around the molecule, areas of relatively positive charge and areas of relatively negative charge will occur, so the molecule is said to be polar. We can illustrate this using the example of salt, water, and oil that we encountered earlier. Water molecules, as described in Chapter 3, have areas of positive charge around the hydrogen atoms and negative charge around the oxygen. Water is therefore a polar solvent. Cooking oil, on the other hand, contains mostly unpolarised carbon–carbon and carbon–hydrogen bonds and is therefore non‐polar. Thus, it will not mix with water but will remain as a separate liquid phase. Salt is sodium chloride and, as described in Chapter 1, comprises positively charged sodium ions and negatively charged chloride anions. The polarised water molecules can organise themselves appropriately around these ions. The positive areas of the water molecules will group around the negative chloride ions, and the negative areas of the water molecules will cluster round the positive sodium ions. Salt will therefore dissolve readily in water. No such stabilisation of the charges on the ions of salt is possible in the electrically neutral non‐polar environment of cooking oil, and so salt will not dissolve in it. Added to a two‐phase mixture of oil and water, salt will dissolve entirely in the water phase. Ethanol is a relatively polar molecule due to its hydroxy group, so it will dissolve freely in water. 1‐Butanol is less polar than ethanol but still polar enough to dissolve to some extent in water. However, its polarity is now sufficiently low that it will also dissolve in oil. Therefore, when added to an oil–water mixture, some butanol will dissolve in the water and some in the oil. We say that the butanol is partitioned between the water and oil. The distribution of a substance between two immiscible liquid phases is very important in biology and in the chemistry of consumer goods, as we will see in later chapters.

 ­Surfactant

The overall polarity of a substance will determine its behaviour when it is placed in a mixture of water and a non‐polar phase. A key measure that we use for this property is known as log P. n‐Octanol is chosen as a standard non‐polar phase, and the test material is added to a mixture of it and water. The ratio of the test substance that dissolves in the octanol to that which dissolves in the water is measured, and the logarithm of that ratio is known as log  P. It is possible to estimate log P values by calculation, which are known as clog P values. Different ways of doing this are used, and each method is likely to give a different result. So, when comparing log P values of materials, it is important to check that all were determined using the same method. The lower the log P value, the more of the material will favour the water, and the higher the log P, the more it will favour the oil. Perfume molecules are mostly non‐polar and have log  P values in the range from 3 to 7. Materials that prefer to dissolve in water are said to be hydrophilic (water loving) or lipophobic (oil hating). Conversely, those that prefer to be in the oil phase are said to be hydrophobic (water hating) or lipophilic (oil loving).

­Surfactants Some molecules have a polar part and a non‐polar part. A typical soap made by hydrolysis of animal fat or vegetable oil is one example. Such fats (also known as lipids) are composed of a molecule of glycerol – a trihydric alcohol – meaning it contains three alcohol groups. As we will see in more detail in Chapter 12, each alcoholic group is esterified with a fatty acid to produce the fat molecule. Hydrolysis of fats with aqueous alkali, such as sodium hydroxide, produces a mixture of glycerol and the fatty acid that is then neutralised by the alkali, giving the corresponding salt of the fatty acid. Since the acid salt contains a cation, such as sodium or potassium, and an anion derived from the fatty acid, it is polar and therefore water soluble. However, the long hydrocarbon chain of the remainder of the molecule is hydrophobic and would prefer to dissolve in oil. If such a molecule comes into contact with a water/oil mixture, it will distribute itself so that its polar end is in the water phase and its non‐polar end in the oil. Because such molecules affect the properties of the oil/water interface, they are said to be surface active and are known as surface‐active agents or surfactants, for short. Figure 4.8 shows an oil/water mixture with surfactant molecules aligned at the interface. In the figure, the hydrophilic head groups are represented by circles, and the hydrophobic hydrocarbon chain by zigzag lines. It is the surfactant property of soaps that makes them useful as cleaning agents as will be seen below. Soaps were probably first made by accident. When meat was roasted over an open fire, some of the fat would melt and run into the fire. If it escaped being burnt, it would mix with the ash from the wood fire. Wood ash is alkaline, and therefore if this mixture became wet, the ester groups in the fat would be hydrolysed to release glycerol and fatty acid salts. Esterification by aqueous alkali is often called saponification. By the Middle Ages, soap making had progressed to the stage where animal fat or vegetable oil was deliberately

75

76

4  States of Matter

Water phase

Oil phase

Hydrophilic head groups

Hydrophobic tail groups

Figure 4.8  Surfactant molecules at an oil/water interface.

boiled with wood ash and the resultant mixture filtered, a very messy process. With advances in electrochemistry in the nineteenth century, sodium hydroxide became available from the electrolysis of brine, and thus soap manufacture became a much cleaner and more selective process. Nowadays, soap is also made by hydrolysing the fat with high temperature steam and then neutralising the resultant acid fraction with alkali. Modern soaps are mostly made from palm oil (PO) or palm kernel oil (PKO). In all soaps derived from natural sources, there is a mixture of fatty acids present. Each source has a characteristic distribution of fatty acid chain length and unsaturation pattern. In PO and tallow (animal fat), the most common chain lengths are 18 carbons, whereas PKO and coconut oil contain mostly 16 carbon acids. Soap suffers from two disadvantages as a surfactant for cleaning purposes. Firstly, if the water used is hard (that is, it contains calcium), the fatty acid is precipitated out of solution as its calcium salt. The calcium salts of fatty acids (other than formic and acetic acids) are insoluble. The scum that forms around the bath in hard water regions is composed largely of the calcium salts of the soaps used. Secondly, if the solution of soap in water becomes acidic, the free fatty acid will be formed, and it will also separate out from the water phase. Soaps must therefore contain an excess of alkali to maintain the fatty acid in the anionic state. The alkalinity makes the soap product somewhat aggressive to the skin, so many alternatives to soaps have been sought out in order to enable products to be made with the surfactant properties of soap, but which are not affected by calcium ions and are kinder to the skin. Such materials are known as non‐soap detergents or NSDs, detergent being another word for surfactant. One end of a molecule can be made soluble in water in various ways, and we can categorise surfactants according to the nature of the hydrophilic group (or hydrophile) each contains. Figure 4.9 shows a selection of some typical surfactants containing an anionic group as the hydrophile. These surfactants are similar to soaps in that each has a hydrophobic part (or hydrophobe) and a polar group comprising the sodium salt

 ­Surfactant LAS Linear alkylbenzene sulfonate

SO3– Na+

R

SO3– Na+

O

O

SCS Sodium cumenesulfonate

n

S

O

O R

O O

O Na+

S O

O– O

Na+

SLES Sodium lauryl ether sulfate

DEFI Distilled ethylene fatty isethionate

Figure 4.9  Anionic surfactants.

of an acid. The difference is that the acids are sulfonic rather than carboxylic. Sulfonic acids are much stronger acids (see Chapter 8) than carboxylic acids, and their calcium salts are more water soluble. These surfactants are therefore capable of being used in neutral solution and will not precipitate out in hard water. A simple example of this type of surfactant is the group known as linear alkylbenzene sulfonates or LAS. The alkyl chains are usually quite long, but one example of a short chain LAS is sodium cumenesulfonate or SCS. Sodium lauryl ether sulfate (or SLES) illustrates the use of ethoxylation in the making of surfactants. In the figure, the brackets indicate a unit that repeats in the structure; in this case it is the fragment – CH2CH2O – which repeats. This common feature in surfactants results from the addition of an alcohol or acid to ethylene oxide (a cyclic ether containing two carbon atoms and one oxygen in a three‐membered ring). The alcohol adds to one ethylene oxide molecule, which results in the formation of a new alcohol and which can then add to a second ethylene oxide and so on. This process is known as ethoxylation. The number of ethylene oxide units in each product molecule will depend on the ratio of the original alcohol to ethylene oxide and the reaction conditions. The letter ‘n’ behind the brackets in the figure indicates that the number of ethylene oxide units in the molecule is not specified. The ether group is much more hydrophilic than a hydrocarbon link, so the repeated ether groups add to the hydrophilicity of the charged end of the molecule. The ethylene isethionates contain a carboxylic ester function in the β‐position relative to the sulfonate. These surfactants, known as distilled ethylene fatty isethionate (DEFI), are solids with a physical feel similar to that of soap but are neutral and therefore make very good alternatives to toilet soap. As discussed above, a polyether chain is hydrophilic, so an ethoxylated fatty alcohol will also behave as a surfactant. These materials, known as alkyl ethoxylates (AEs), constitute one family of electrically neutral, (and therefore also pH neutral) surfactants. Several examples, such as Laureth‐2, Synperonics, and

77

78

4  States of Matter O

O O

n

OH

Synperonics OH e.g. A7, n = 7

Laureth-2 Lutensols – R-O(CH2CH2O)7H e.g. R = C13H27

O OH GMS glycerol monostearate

O OH

O O

O O O

IPM isopropyl myristate

EGDS ethylene glycol distearate

O R R R R R RR R Silicones – the side chains are sometimes functionalised Si etc Si Si Si O O O O O O OSA octenyl succinic acid Starch OH C8H15 O

Figure 4.10  Neutral surfactants.

Lutensols, are shown in Figure 4.10. This figure also shows some other neutral molecules that have some surface‐active properties. Glycerol monostearate (GMS) and isopropyl myristate (IPM), like the AEs, have a polar head group and a hydrophobic tail. Ethylene glycol distearate (EGDS) has its polar group in the centre of the molecule, but if it folds as shown in the figure, the overall effect is similar to that of the other surfactants with a polar region at one end and a double width hydrophobic tail. This structure is not unlike some natural surfactants, as we will see later. Silicones are somewhat different as the hydrophilic areas are distributed along the polymeric backbone. Many examples of natural surfactants, which are neutral molecules, often contain a sugar or higher carbohydrate group as the hydrophile and a terpenoid or steroid fragment as the hydrophobe. Octenylsuccinic acid (OSA) bears some resemblance to this type of natural surfactant. It is a starch that has been modified by addition of a substituted succinic acid. (The bond running from the C8 chain in the figure across the central bond of the succinic acid unit employs a drawing technique used to depict mixtures in which some components will have the group attached to one of the succinate carbons and some to the other.) OSA is not surface active in the same way as the ionic surfactants employed in detergents but is used as an emulsifying agent. Cationic surfactants are mostly based on quaternary ammonium salts or quats. Some typical examples are shown in Figure 4.11. Arquad and Hamburg ester quat (HEQ) both have two hydrophobic tails, similar to those of EGDS. Each of these contains a balancing anion, usually chloride. Quats are mostly used as conditioning agents. When cloth, the skin, or the hair are washed using anionic detergents, the detergent tends to remove the hydrogen atoms from alcohol and acid groups on the surface and so leaves an anionic surface with

  ­Emulsion O Saccharide

N+

O

CAT Guar

Arquad

N+ O R O N+

HEQ Hamburg ester quat

O R O

Figure 4.11  Cationic surfactants.

cations, usually sodium or calcium, to balance the charge. The result is rough or harsh surfaces. Addition of a quat will deposit the bulkier, more hydrophobic quaternary ammonium ion on the surface in place of the sodium or calcium ion, which gives a softer, better lubricated surface, hence the term conditioning. In CAT guar, the saccharide is derived from guar gum (hence the name). The quaternary ammonium salt helps it to adhere to fabrics and thus take advantage of the ability of the starch fragment to retain water, again lubricating the surface. Compounds containing both anionic and cationic groups in the same molecule are known as zwitterions. If the opposite charges are on adjacent carbon atoms, the zwitterion is called an ylide. If the charged centres are β‐ to each other, then the zwitterion is called a betaine. Cocamidopropyl betaine (CAPB) is so called because the decanoic acid unit is derived from coconut oil; the amines stem from 1,3‐diaminopropane, which is attached to the hydrophobe by an amide group. This structure resembles even more closely that of the phospholipids or lecithins, which are the surfactants in mammalian cell walls as will be seen shortly. The structures of CAPD and a typical generic amine oxide are shown in Figure 4.12 as examples of zwitterionic surfactants.

­Emulsions If two immiscible liquids are brought into contact with each other, they may simply continue to exist as a two‐phase system with the less dense layer floating on top of the denser one. Common examples of this can be seen in the kitchen. For example, in French salad dressing, the less dense oil layer floats on the vinegar, and when meat is boiled, the molten fat floats on the broth beneath it.

79

80

4  States of Matter O O–

N+

N H

O

CAPB Cocamidopropyl betaine

O–

R N+

Amine oxides

Figure 4.12  Zwitterionic surfactants.

In some circumstances, two immiscible liquids can be treated in such a way as to form what appears to be a single phase, known as an emulsion. Everyday life provides us with many examples of emulsions, such as paint, cosmetic creams, and butter. In emulsions, tiny droplets of one liquid phase are dispersed in another. In cosmetic creams, droplets of an oily phase are dispersed in water. On the other hand, with butter, tiny droplets of water are dispersed in the fat. The cosmetic creams are referred to as ‘oil‐in‐water’ emulsions, whereas butter is an example of a ‘water‐in‐oil’ emulsion. If water is added to an ‘oil‐in‐water’ emulsion, the emulsion will be diluted, whereas addition of water to a ‘water‐in‐oil’ emulsion will not break the integrity of the emulsion. A piece of butter dropped into cold water will remain a piece of butter, whereas addition of water to a cosmetic cream will produce a thinner and probably less appealing mixture. Figure 4.13 gives a schematic representation of emulsions as opposed to a simple two‐phase system. The liquid in the droplets is known as the disperse phase, and the liquid surrounding the droplets as the continuous phase.

Two-phase system

Oil-in-water emulsion

Oil

Figure 4.13  Structure of emulsions.

Water-in-oil emulsion

Water

  ­Detergenc

Emulsions are of great importance to the fragrance industry. Many products into which we put perfume, especially cosmetic products, exist as emulsions. Emulsions are also involved in dirt and stain removal and hence the chemistry of soaps and detergents, two more product categories of importance to us.

­Micelles Emulsions are usually prepared by vigorous shaking or stirring of a mixture of the two liquids. The process of emulsification and the stability of the resultant emulsion can be improved by the addition of suitable surfactants. This practice leads to the formation of micelles, and the droplets of disperse phase in a micellar solution are much less likely to coalesce to form a second phase (i.e. to separate out from the emulsion) than are those in a simple emulsion. For example, shaking a bottle of French salad dressing will produce an emulsion, but it will separate out quickly once it is left to stand, whereas a cosmetic cream that contains surfactants will remain as an emulsion for a long period of time. Figure 4.8 showed surfactant molecules aligned at an interface between two immiscible liquids. It is easy to see how surfactant molecules can arrange themselves around the outer surface of a liquid droplet as shown in Figure 4.14. In this figure, the surfactants extend their fatty tails into an oil phase and thus make it more stable in water by presenting a hydrophilic surface to the surrounding water phase.

­Detergency Seeing how micelles are stabilised by surfactants now leads us into an understanding of how detergency works.

Figure 4.14  Surfactants stabilising a droplet in a micelle.

81

82

4  States of Matter

(a)

(b)

(c)

(d)

Figure 4.15  Detergency in action.

The left‐hand side of Figure 4.15 contains a representation of an oily stain on a surface, indicated by the letter A. It could be a grease stain on a piece of cotton or linen, a bead of sebum on a hair, or oil on the skin. The surfaces in question are all somewhat hydrophilic in nature as they contain protein and carbohydrate (e.g. cellulose) molecules with hydroxyl and other polar groups on their surfaces. However, these surfaces are not as polar as water, so if we immerse the surface (fabric, hair, or skin) in water, the oily stain will prefer to remain on the surface than to dissolve in water. However, if a surfactant is added to the water, the hydrophobic part of the surfactant will dissolve in the oily stain giving it a more hydrophilic surface and therefore reducing the repulsion between the stain and the water. This situation is indicated by the letter B. As more surfactant is added, a greater surface area can be stabilised between the stain and the water around it. Thus, as the stain‐to‐water repulsion falls, the stain will move more into the water in order to reduce the area exposed to the cloth (or skin or hair whichever it is), which gives a situation as indicated by the letter C. Eventually, the oily droplet will detach from the surface as a micelle and is therefore removed from the surface into the wash water as shown by D.

­Bilayers Figure 4.8 shows a layer of surfactants at an interface with oil on one side and water on the other. It is also possible to form a double layer of surfactants with hydrophilic groups at each side and a hydrophobic layer between them. This set‐up is known as a lipid bilayer and is shown schematically in Figure 4.16. A lipid bilayer, such as that shown in Figure 4.16, has water on both sides and a fatty zone in the middle. If the hydrophilic head groups are ionic in nature, neither hydrophobic nor hydrophilic molecules can pass through the barrier. Hydrophilic molecules will remain in the water layer rather than moving into the fatty centre, and hydrophobic molecules will not be able to cross the polar outer face of the bilayer. The bilayer can extend in three dimensions, and if the surface is curved, it can close round on itself and make a spherical particle. Figure 4.17

  ­Bilayer Water layer

Water layer

Figure 4.16  A simple lipid bilayer.

Lipid bilayer

Figure 4.17  Cross section of a mammalian cell wall.

shows a cross section of such a particle. The particle now exists in an aqueous environment and contains an aqueous centre, but the centre is chemically isolated from the environment. This forms the basic structure of the mammalian cell wall. Some of the components of the mammalian cell wall are shown in Figure 4.18. Phosphatidylcholine is a member of the family of chemicals known as lecithins, and sphingomyelin is a ceramide‐derived lipid. Both contain two hydrophobic chains and a zwitterionic head group. The surface of the mammalian cell wall is therefore stabilised by electrostatic attractions between the opposite charges on the head groups of neighbouring molecules as well as by the hydrophobic/hydrophilic interactions, steric effects, and hydrogen bonding. Most of the hydrophobic chains are 16 or 18 carbons in length, but some are twice that and therefore stabilise the bilayer by bridging right across it. Cholesterol, which is a steroid, sits in the central hydrophobic area and provides rigidity.

83

84

4  States of Matter

O O

–O O P O O

+

N

O

O

Phosphatidylcholine

O HN

N+

PO O–

O

Sphingomyelin

OH

HO

Cholesterol

Figure 4.18  Some components of the mammalian cell wall.

­Colloids Finely divided solids can be suspended in liquids in which they are not soluble. Normally the particles in a suspension will settle to the bottom unless they are of similar density to the liquid or if the liquid is viscous enough to hold the particles. Special cases where the particles remain in suspension are called colloidal solutions or colloids. Colloids are sometimes used to create special visual effects in consumer products, but they are much less important to perfumery than are emulsions, so they will not be considered further here.

Review Questions 1 Why is it not good practice to leave perfume in a hot environment (for example, in direct sunlight)? 2 Your company has facilities near New York City and also Mexico City. You have a process that, at one point, involves removing solvent from an extract. Will the conditions be the same at both sites? 3 Why do we perspire more when the weather is humid? 4 Your customer wants to know how compatible your perfume and his product will be when stored for a period of three months. The difficulty is that he wants to know the answer in two weeks’ time. What do you do?

85

5 Separation and Purification In order to analyse and manipulate materials, it is important to be able to isolate them from mixtures and obtain them in pure form. The various methods by which purification can be achieved will be described in this chapter. In fragrance chemistry, the two principal properties we use to achieve separation of materials are volatility and solubility. The volatility of a material is its ability to become a gas rather than a solid or liquid. Distillation is a purification process that relies on volatility. Crystallisation, solvent extraction, and chromatography rely on a material’s solubility.

­Distillation In the simplest application of distillation, liquids can be separated from the solids dissolved in them by heating the liquid above its boiling point so that it is removed as vapour and then condensing the vapour by contact with a cold surface so that it returns to the liquid state. An example of this might be the separation of water from brine. The liquid that distils is known as the distillate, and the material that remains in the still pot is known as the residue. So, in the case of brine distillation, the distillate would be pure water, and the residue would be salt. The simplest form of still therefore comprises a pot (in which the liquid to be distilled is heated and evaporated), a condenser (in which the vapour is cooled to return it to the liquid state), and a receiver (in which the distillate is collected). Figure 5.1 shows a drawing of such a still. The heat source could be an open flame or an electrical heater. Gas and electrical heaters sometimes use water or oil as a heat transfer medium. The heat source is applied to a thermally stable oil or to water, and the heated liquid is then brought into contact with the outer surface of the still pot. This system has the advantage that localised overheating (or hot spots) is avoided. Hot spots on the pot surface can lead to uneven boiling, generating mechanical shocks, and to degradation of material in the pot. The condenser may be air‐ cooled or, more often, have a supply of cold water running around it as shown in Figure 5.1. Even a simple still is usually equipped with two t­ hermometers – one to measure the temperature in the pot and one to measure the ­temperature of

Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

86

5  Separation and Purification

Condenser

Thermometers

Water inlet

Water outlet Vent to atmosphere Receiver

Pot

Heat source

Figure 5.1  A simple still.

the vapour at the top of the still before it comes into contact with the c­ ondenser (the head temperature). These thermometers give the operator valuable information about the progress of the distillation. It is also usual to provide a stirrer or some other means of helping vapour bubbles to form in the liquid. A piece of wood or boiling stones (small chips of porous ceramic material) will serve this purpose. If no such boiling aid is present, it is possible for the liquid to become superheated – in other words, heated above its boiling point but still in the liquid phase. Superheating can have disastrous consequences when something then triggers boiling since a large part of the liquid can suddenly enter the vapour phase, greatly increasing the pressure in the still and possibly causing mechanical damage and/or loss of hot liquid into the immediate environment. Most fragrance materials have boiling points well above that of water and contain less robust molecules. Heating them to their boiling point can therefore cause decomposition. Even a small amount of decomposition is very likely to have an adverse effect on the odour of the distillate, since traces of volatile decomposition products will often significantly affect the odour of the whole. In the previous chapter, we looked at a typical phase diagram (Figure 4.7) and saw how the boiling point of a liquid could be altered by changing the pressure. Lowering the pressure above a liquid reduces its boiling point and makes it possible to distil the liquid at a lower temperature. Therefore, most fragrance ingredients are distilled at reduced pressure in order to lower the possibility of decomposition or degradation. Of course, it also has the beneficial effect of reducing the amount of heat required and hence the energy consumed in the process. For example, α‐pinene, the major component of turpentine, boils at 156 °C at atmospheric pressure (760 mmHg or 101.3 kPa), but when the pressure is reduced to 100 mmHg (13.33 kPa), the boiling point is 89 °C.

  ­Distillatio

1 mmHg is the pressure exerted by a column of mercury 1 mm high, kPa = 1000 Pa. 1 Pa is a pressure of 1 N/m2. 1 N is the force required to impart an acceleration of  1 m/s to an object weighing 1 kg. Both mmHg and kPa are used as units of ­pressure in distillation.

The reduced pressure is achieved by connecting a vacuum pump to the still, and this, in turn, means that several other additions or modifications to the still construction are necessary. A stirrer is often used to ensure even boiling, but usually an aspirator or bleed is also added. The bleed allows a small continuous flow of a gas to enter the still, with the rate of flow controlled by a valve. The gas needs to be non‐reactive, so nitrogen is usually used for this purpose as it is the least expensive gas that will not react with the still contents. The gas flow promotes boiling and also allows the operator to return the pressure inside the still to atmospheric pressure when the distillation is over. A pressure gauge is used to measure the pressure in the still. In order to be able to remove distillate from the still while it is running under vacuum, it is necessary to install a device known as a Perkin triangle. A Perkin triangle can be constructed various ways; one example is shown in Figure 5.2. In this case, it has two receivers. The upper tap connects one receiver at a time to the condenser and simultaneously connects the other to the atmosphere. The lower tap is a three‐way tap that connects one receiver at a time to the vacuum pump. Thus, while one reservoir is filled with distillate from the still, the other is open to the atmosphere and can be drained out through the drain tap. When the second receiver is full, the taps can be reversed so that the distillate now flows into the first receiver and the second can be drained. Each batch of distillate is Pressure gauge

Thermometers

Water outlet

Condenser

Water inlet Aspirator

Upper tap Threeway tap Receivers

Heat source

Drain tap

Drain tap Vacuum pump

Figure 5.2  Distillation under reduced pressure.

87

88

5  Separation and Purification

known as a fraction, and, using the set‐up in Figure 5.2, as many fractions as desired can be collected without having to shut the still down between them. Obviously, the total volume of the fractions will be no more than the volume charged to the pot. Liquids exist in equilibrium with their vapours. In the space above a mixture of two liquids, the vapours of the two will be present in the ratio of their vapour pressures. Therefore, there will be more of the vapour of the liquid with the higher vapour pressure and hence lower boiling point. So, if a mixture of two liquids is distilled, the distillate will initially contain more of the lower boiling component. The degree of separation will depend on the difference between the boiling points. If the difference is sufficiently large, one single distillation could be sufficient to separate the two components. However, if the boiling points are closer together, only a partial separation will be achieved in a single evaporation/condensation cycle. If this initial distillate is condensed and then brought to boiling again, the lower boiling component will be further enriched in the vapour of the second distillation. Thus, with repeated distillations, it is possible to separate even liquids that have very similar boiling points. In practice, this separation is achieved using a fractionating still. In such a still, the ascending vapour is taken through a column packed with an inert material shaped so that as high a surface area as possible is presented to the vapour. In the laboratory, glass rings or glass helices are often used. Custom designed ceramic or stainless steel packings are used in commercial scale stills. As the vapour ascends through the packed column, it meets a flow of condensate running back down over the surface of the packing. This process enables re‐equilibration to take place with more of the lower boiling component entering the vapour phase and more of the higher boiling component condensing and joining the flow back down towards the pot. The longer the column and the more efficient the liquid/vapour exchange, the better the separation will be. The efficiency of a column is measured in units called theoretical plates. One theoretical plate will achieve one complete equilibration between ascending vapour and descending condensate. The efficiency of the packing in a column is expressed as the height equivalent per theoretical plate (HETP); in other words, the height of column necessary to achieve separation equivalent to one theoretical plate. Columns are very well insulated to keep the liquid/vapour equilibria going throughout the length of the column. Efficient operation of a commercial scale fractionating scale requires considerable skill, as factors such as the rate of boil up (i.e. the rate at which liquid is boiled and fed into the bottom of the still) and the reflux ratio (i.e. the ratio between the amount of material being taken off at the top of the still and passed into the condenser to the amount of material being returned down the column) all play a part in the degree of separation that will be achieved. Of course, all of these parameters are also affected by the temperature and pressure in the various parts of the still. Figure 5.3 shows the principal parts of a fractionating still. There are three significant differences between this and the simpler still that is shown in Figure 5.2. Most importantly, the empty column has been replaced by one packed with inert material that has a high surface area. As described above, this process is

  ­Distillatio Thermometers Reflux ratio control

Water out Water in

Pressure gauge Packing Insulation Pressure gauge

Water outlet

Condenser

Water inlet

Upper tap

Thermometers Threeway tap Aspirator

Heat source

Receivers

Drain tap

Drain tap Vacuum pump

Figure 5.3  A fractionating still.

what allows for separation of components with close boiling points. The packing and the flow of liquid returning down through it create a pressure difference across the length of the column. It is necessary to add an additional pressure gauge so the pressure at both the top and bottom of the column can be measured and the operating conditions adjusted if the pressure in the pot becomes too high. The third difference is the addition of a reflux ratio control device at the top of the column. The control device can function either with vapour or liquid. The one shown in Figure 5.3 is a vapour splitting device. When lowered, it shuts off the path to the receiver and thus directs it to the condenser above where it is condensed to liquid and returned to the column. When raised, it opens the path to the main condenser and hence to the receiver. The ratio of time raised to time lowered therefore determines the reflux ratio. When a liquid splitter is used, all of the liquid is condensed at the top of the column, and a tilting bucket or similar device is used to partition the liquid between column and receiver. In either case, since the splitter is in the evacuated still, it is often moved by electromagnets placed on the outside, thus allowing controlled movement without compromising the vacuum in the still. Reflux is the material that is returned down the column rather than being distilled over into the receiver. The reflux ratio is therefore the ratio between the material flowing back down the column to that distilling across. Control of the reflux ratio is necessary in order to optimise the performance of the still.

89

90

5  Separation and Purification

Sometimes two or more liquids with different boiling points do not separate on fractional distillation but form what is called an azeotrope or constant boiling mixture. The best known example is that of ethanol and water. Ethanol boils at 78.4 °C and water at 100 °C. If a mixture of the two is distilled, the first mixture to come across through the still is an azeotrope that boils at 78.1 °C and contains 95.6% ethanol and 4.4% water. This azeotrope will continue to distil until one of the components is exhausted from the pot. Only then will a pure liquid be distilled, in this case whichever was in excess (relative to the composition of the azeotrope). Azeotropes containing two components are known as binary azeotropes. Those azeotropes with three components are known as ternary azeotropes, those with four as quaternary azeotropes, and so on. The formation of azeotropes can cause some difficulty. For example, it is impossible to produce pure ethanol by distillation because of the presence of excess water (e.g. from a fermentation). This explains why ethanol is usually available as a mixture containing some water. This mixture is known as absolute ethanol or absolute alcohol and is the most common solvent for fine fragrances. On the other hand, azeotrope formation can be used to an advantage in other circumstances. For example, the most common solvent for extraction of natural oils (see section ‘Solvent Extraction’) was once benzene. However, when it became known that this use of benzene presented an unacceptable safety hazard for the workers in the essential oil industry, alternatives were sought. It was not an easy task as it was necessary to find a liquid with similar properties as a solvent and a similar boiling point to those of benzene. No single liquid provided the right combination of properties, but it was found that an azeotrope (bp 65 °C) containing 60% hexane (bp 69 °C) and 40% ethyl acetate (bp 77 °C) came close to matching the properties of benzene. The advantage of using an azeotrope rather than any other mixture of two components is that when the solvent is recovered by distillation, its composition will always be the same. Therefore, it can be recycled without having to check its composition. Other uses for azeotropes include drying of liquids and mixtures. For example, water (bp 100 °C) forms an azeotrope with toluene (bp 110 °C), which boils at 87 °C and contains 16% water and 84% toluene. Therefore, by distilling out the water/toluene azeotrope, a mixture can be dried. This distillation is usually done in a Dean and Stark apparatus such as that illustrated in Figure 5.4. The azeotrope distils up through the apparatus and condenses at the top. On condensation, the mixture separates out as toluene and water that are very poorly soluble in each other. The water, being denser, falls to the bottom of the trap from where it can be drawn off. The toluene floats on the water layer and spills over, back into the pot, allowing it to be recycled. In the preparation of essential oils, it is common practice to use steam distillation. In this technique, the plant material containing the oil is heated in the presence of water. Either liquid water is added to the still pot, or steam may be generated in a separate boiler and injected into the pot. A combination of both can also be used. A steam still has very little fractionating power, and so the essential oils that distil contain a large variety of components with very different boiling points. The principal reason for using steam rather than dry (also called empyreumatic) distillation is that the presence of water restricts the

  ­Distillatio

Water outlet Water inlet

Less dense liquid

Heat source

Liquid mixture

Denser liquid

Figure 5.4  A Dean–Stark apparatus.

temperature to 100 °C and therefore limits decomposition of the oil components. It also helps to prevent hot spots from developing in the pot around the areas where external heating is applied. The oil distils as a mixture with steam and separates from the water when the vapours are condensed. Normally very little oil is lost in the water layer; however, the water is usually recycled to minimise such loss. It also conserves water in areas where it is in short supply. However, some oils do contain components that are moderately water soluble. The most important example is 2‐phenylethanol, which is a major component of rose oil. The water layer from steam distillation is known as the ‘waters of cohobation’. In the case of distilled rose oil, it is called rose water and can be quite fragrant. The oil and water phases are usually then separated in a Florentine flask. Figure 5.5 shows a steam still with a Florentine flask attached to it. In the figure, the Florentine is arranged for an oil that is less dense than water and therefore floats on the water layer. This arrangement is by far the most common case, but, in the case of an oil that is denser than water, it is easy to construct a Florentine that takes the lower layer off as oil and returns the upper (water) layer to the pot. Since the yield of oil in plant material is low, necessitating a relatively large volume of water for handling purposes, steam distillation is an energy‐intensive process. Essential oils also tend to have much higher boiling points than water, so much more water is distilled than oil. Furthermore, water has a very high latent heat of vaporisation (the amount of heat required to convert water at 100 °C to steam at the same temperature). All this means that a large amount of energy is consumed to distil essential oils. An alternative process is known as hydrodiffusion. In hydrodiffusion, steam is introduced at the top of the pot and flows downwards through the plant material, releasing the oil and carrying it out with the water that condenses. This uses less energy than stream distillation but gives an oil with a different quality, since the water solubility of the oil components becomes more important and their volatility less so than in conventional steam distillation.

91

92

5  Separation and Purification

Water outlet

Water inlet

Water and plant material

Essential oil Steam in Heat source

Water

Figure 5.5  A steam still.

Products that are distilled from plant materials are known as essential oils. Essential oils often contain significant quantities of terpenoid hydrocarbons (see Chapter  14 for a definition of terpenoids). These terpenoids often contribute relatively little to the odour character of the oil, so are sometimes removed by distillation or other techniques to provide what is known as a terpeneless oil or folded oil. The process of removing the terpenes is known as deterpenation. An outline of the processes is shown in Figure 5.6 along with other techniques for isolating fragrance ingredients from plant material. These methods will be discussed later in this chapter. Distillation is the most important purification technique used in perfumery. This is not surprising as perfume ingredients must be volatile in order to be smelt, and distillation is a very efficient and practical way of purifying volatile substances. All major perfumery companies and essential oil producers are

Essential oil

Distillation

Source material

Deterpenation

Terpeneless or folded oil

Solvent extraction

Enfleurage

Pomade

Other

Ethanol

Tincture

Concrete (Resinoid)

Ethanol extraction

Ethanol extraction

Absolute

Figure 5.6  Natural extracts and isolates.

  ­Crystallisatio

therefore highly skilled in distillation techniques, and the distillery is a vital part of any fragrance house.

­Sublimation It is possible to purify solids by a process similar to distillation. This technique is known as sublimation. For solids with a suitable phase diagram, it is possible to convert the solid to a gas by heating and then, on cooling the gas, to produce the pure solid. This method is a much less common practice than distillation, especially in perfumery. The fragrance industry is very skilled at distillation and usually prefers to distil solid perfume ingredients and keep them as liquids by holding the condenser at an appropriate temperature, rather than to use sublimation with all the resultant problems of handling solids. One interesting historical example is the sublimation of lead sulfide from the mineral galena. This process gives a very finely divided black solid that was used as eyeshadow by the ancient Egyptians. (No health and safety concerns about the toxicity of lead in those days!) The Egyptian name for the lead sulfide thus produced was ‘kohl’. From this we get English words such as coal and charcoal. When the Arabs invented distillation in the eleventh century, they saw the similarity between the processes of sublimation and distillation and called the product of distillation of fermented liquor ‘al kohl’ (al being the definite article in Arabic). From this, of course, we get the word alcohol, originally meaning the alcohol formed on fermentation, i.e. ethyl alcohol, but later being extended to all members of the same chemical class.

­Crystallisation Crystallisation is a process by which a pure material is obtained from a solution of it in a solvent. The solid is forced out of solution slowly by either reducing the temperature or reducing the volume of solvent. The solvent can be the liquid form of the material to be crystallised, but for purification purposes, it is better to use another liquid. The most common practice is to dissolve the solid to be purified in a minimum amount of hot solvent. The best solvent is one that has a poor solubility for the target solid when cold, a high solubility for it when hot, and also a high solubility for the impurities in the solid. Thus, a relatively large volume of solid dissolves in the hot liquid, and, as the liquid cools, the major component of the solid passes the limit of solubility at the lower temperature and forms crystals. As crystals grow, they prefer to absorb more of the same type of molecule onto the sites around their surfaces. This tendency creates an additional driving force for the major dissolved component to separate out. The impurities are present at a lower level, and so, if the conditions have been chosen well, they do not pass their solubility limits and remain in solution. The crystals formed are therefore purer than the starting solid. For example, if it is desired to purify a non‐polar solid containing a more polar impurity, it might be a good idea to look for the most polar solvent in which the solid will dissolve, knowing that the more polar impurity is likely to stay soluble in the chosen solvent as it is cooled than is the solid to be purified.

93

94

5  Separation and Purification

Often crystallisations are seeded with crystals of the material to be purified as this will cause that material to preferentially crystallise around the seeds. This practice is particularly important in fractional crystallisation where it is desired to crystallise one component out of an equal, or nearly equal, ratio of components. An important example of this process in the flavour and fragrance industry is menthol. Pure l‐menthyl benzoate can be crystallised from a racemic mixture by seeding the solution with crystals of l‐menthyl benzoate. This step is key in one of the commercial routes to l‐menthol.

­Solvent Extraction Solvent extraction is the removal of material from a mixture by bringing it into contact with a solvent in which the desired material is soluble and the undesired is not. Filtration or separation of two liquid phases, whichever is appropriate, is then followed by evaporation of the solvent to leave the desired extract. In organic synthesis, it is common practice when a reaction is complete to add the complete resultant mixture to water and an immiscible organic solvent such as ether or hexane. In this way the desired product (assuming it to be water insoluble) will dissolve in the organic phase, and unwanted water‐soluble materials can be removed in the water phase. For example, in the preparation of an ester from an alcohol and a carboxylic acid, the final reaction mixture might be poured into water and extracted with a water‐insoluble solvent. The acid catalyst used in the esterification would be removed in the wash water. In producing natural extracts for perfumery, the starting mixture is usually plant material that is normally chopped into small pieces to increase the available surface area. Extraction with an organic solvent dissolves out the non‐polar components such as the odorous chemicals, which are required for perfumery. Fats were the earliest solvents used for extraction of perfume oils from plants. In a process known as enfleurage, plant material (such as petals) was pressed into fat, and the plant oils were allowed to diffuse into the fat. The fat was then melted and filtered to remove the plant debris. The fat then would solidify again to give a product known as a pomade. In ancient Egypt, the pomade was shaped into a cone and placed on the head as a way of perfuming the body. After the invention of distillation, ethanol became available as a solvent, but this was actually the ethanol/water azeotrope. Because of the water content in the azeotrope, fat will not dissolve in it, and so the perfume ingredients, which are generally more polar than fat, could be extracted from the pomade. Removal of ethanol from the extract then produces what is known as an absolute. Direct extraction from the plant material using ethanol is also possible, but the ethanol will also extract all of the water present in the plant material. For this reason, direct ethanol extracts are not common. They are used in a few instances, and in these cases, the alcoholic solution is known as a tincture. The most important tincture in perfumery is tincture of ambergris. Ambergris is a material produced by the sperm whale as is described in Chapter 14. It contains very little water, and therefore ethanol extraction is a good technique for concentrating the odour components, leaving behind the various ethanol‐insoluble materials such

  ­Chromatograph

as steroids and other higher terpenoids (see Chapter  14 for definition). An extraction using solvents other than fats or ethanol, followed by solvent removal by distillation, gives what is known as a concrete or resinoid. Concretes contain all the solvent‐soluble materials present in the plant material, and, as the solvents used are mostly non‐polar, this will include all of the fats, waxes, and chlorophyll present in the plant. Concretes are usually therefore fatty in texture and dark in colour. As with pomades, they can be extracted again using aqueous ethanol to give an absolute, free of the fats, waxes, and pigments such as chlorophyll. A summary of all of these processes, together with those involving distillation, is shown in Figure 5.6.

­ ecent Developments in Commercial Extraction R of Natural Fragrance Ingredients Benzene was once the most common solvent used in preparation of concretes, but when its toxicity became known, others were sought to replace it. For the reasons given above, the hexane/ethyl acetate azeotrope is now one of the most common in use. As one would expect, research continues into finding better ways of removing fragrant oils from plant material. Probably the most important of these is the use of carbon dioxide, either in liquid or supercritical form, as an extraction solvent. This substance has the advantages of being easily removed at normal temperature and pressure and of not leaving any solvent residues in the final product. The polarity of carbon dioxide varies with pressure, and therefore different pressures can be used to obtain extracts of different compositions. However, considerable pressure is necessary to liquefy carbon dioxide, and it is costly in terms of both energy and capital investment in extraction plant. Other solvents that can be used for extraction include monoterpene hydrocarbons such as limonene, ethyl lactate, hydrofluorocarbons such as 1,1,1,2‐tetrafluoroethane, ionic liquids such as methyl ethyl 2‐hydroxyethyl salts of carboxylic acids, and subcritical water. Subcritical water is liquid water heated under pressure to below its critical temperature. Its polarity can be adjusted by addition of other solvents such as ethanol. A more recent set of techniques involves the use of microwave heating. In these extraction methods, the plant material is not dried before extraction as in normal steam distillation. The frequency of radiation used is the same as that of domestic microwave ovens, and so it heats water molecules. Thus, the microwave heating of the water in the plant breaks open the structures holding the oil in the plant and enables ‘steam distillation’ of it. This can be used with or without the help of reduced pressure to assist in the evaporation of the essential oil.

­Chromatography Chromatography is the most important technique used for purification of fragrance ingredients for analytical, as opposed to production, purposes. In all forms of chromatography, the purification process relies on differences in

95

96

5  Separation and Purification

v­ olatility and/or solubility of different materials between two different carriers that are moving relative to each other. The various forms of chromatography are paper chromatography, thin layer chromatography (TLC), column chromatography, high performance liquid chromatography (HPLC), and gas chromatography (GC). The last of these is, by far, the most important for the fragrance industry and is also known as gas/liquid chromatography (GLC) and vapour phase chromatography (VPC). In chromatography, materials are partitioned between two media. One is called the stationary phase, and the other is called the mobile phase since it flows across the stationary phase. The first recorded experiments in chromatography were carried out on plant pigments using paper as the stationary phase. This work was done by Pliny the Elder, whose laboratory at Pompeii was destroyed in the famous eruption of Vesuvius in 79 ce. Separating plant dyes on paper gives bands of different colours and hence the name chromatography, which is Greek for ‘writing with colours’. However, the ‘father of chromatography’ is generally considered to be Mikhail Tswett who, between 1903 and 1906, also separated plant pigments. But instead of using paper sheets as Pliny did, Tswett used columns of chalk. The main forms of chromatography in use today are shown in Table 5.1. Figure 5.7 shows the basic principle of chromatography, which is the same for all of the forms described in Table 5.1. For simplicity, we can consider it to represent an experiment such as Pliny’s in which three different plant pigments; one red (represented by grey dots), one yellow (represented by white dots) and one blue (represented by black dots); are separated on a sheet of paper using water as solvent. The figure shows five “snapshots,” each taken at a different time, of the separation in progress. In each, the stationary phase (paper in our example) is shown as the lower block and it is darker in colour than the upper block which represents a layer of solvent (water) that is flowing through the paper in the direction shown by the arrows at each end. The top snapshot shows the system Table 5.1  Types of chromatography. Mobile phase

Chromatography

Stationary phase

Paper

Paper

Liquid solvent

TLC

Silica gel or alumina on glass or aluminium plates

Liquid solvent

Column

Silica gel or alumina powder in a glass tube

Liquid solvent

HPLC

Very finely powdered silica gel or alumina under high pressure in a stainless steel tube

Liquid solvent

GC

High boiling liquid or wax coated onto either a very finely powdered solid support or onto the walls of a capillary tube (capillary GC)

Gas

  ­Chromatograph Mobile phase Stationary phase Time = 1

Time = 2

Time = 3

Time = 4

Time = 5

Figure 5.7  Basic principle of chromatography.

before the experiment starts. In the second snapshot, a mixture of the three pigments is added to the upstream end of the paper. The molecules are initially randomly distributed, and the three colours appear together. However, the red molecules are more soluble in water than the yellow, and the yellow in turn are more soluble than the blue. As the water layer moves across the paper, the red molecules spend more time in the water phase than they do adsorbed onto the paper surface, and so they tend to move fairly quickly with the flow of water. The yellow molecules spend a little less time in the water than the reds and so move more slowly than the latter. The blue really prefer to be adsorbed to the paper than to dissolve in water and so they move more slowly than either the red or yellow. Thus, when we see the third ‘time snapshot’, we see all of the colours moving but the red molecules, on average, are moving ahead of the yellow and the blue ones are tending to be left behind. By the time of the fourth snapshot, the reds and yellows have really moved ahead of the blues and are beginning to separate out from each other, and by the fifth, we can see three distinctly separate bands of colour. With plant pigments moving across white paper, it is easy to see where the different components are at any point. However, with colourless materials or when the stationary phase is not visible, for example, because it is enclosed in a steel tube or inside an oven, we need some other way of detecting where the analytes (materials being analysed) are. This process is known as visualisation, and the way of achieving it will depend on the nature of the analyte and/or the technique being employed. Visualisation techniques will therefore be discussed separately for each chromatographic system in the following paragraphs.

97

98

5  Separation and Purification

­Paper Chromatography In paper chromatography, the stationary phase is a sheet of paper. The solvent flows across it by capillary action if the paper is either horizontal with solvent being added at one side or vertical with the bottom edge of the paper immersed in a reservoir of the solvent. If the paper is vertical and the top edge is held in a tray of solvent, the solvent will flow under gravity. Visualisation of colourless materials on paper chromatography is best achieved by spraying the paper with a reagent that will form a coloured product with the analyte but not with the paper. It is not always easy to achieve desired results and is one of the reasons why paper chromatography is not used much in our industry.

­Thin Layer Chromatography In TLC, the stationary phase is a thin layer of finely divided material, usually silica or alumina, coated onto a support, usually either a glass plate or a sheet of aluminium metal. The plate is stood vertically in a tank with a layer of solvent in the bottom. The solvent rises up across the stationary phase on the plate through capillary action. A horizontal line is drawn across parallel to the bottom of the plate and high enough above it to ensure that the line is above the solvent level when the plate is placed in the tank. The analyte and reference substances are then spotted onto the line by means of adding drops of a solution of the analyte or reference in a suitable solvent. Each drop is allowed to dry before adding the next so that repeated addition builds up a small spot containing sufficient material to allow it to be visualised after development. Development is achieved by allowing the solvent to flow up the plate (by capillary action) across the starting line of analyte(s) and reference materials. When the solvent front reaches near the top of the plate, the plate is removed from the tank and a line drawn on it to mark the position reached by the solvent front. If the materials being analysed are colourless organic materials, they are usually visualised by spraying with a suitable reagent and, if necessary, heating the plate to develop a colour. One common visualisation technique is to spray the plate with sulfuric acid and then heat it. This process has the effect of charring the organic material and hence producing a dark colour where it was. Another common reagent used in the same way is phosphomolybdic acid. The distance travelled by the analyte can be expressed as a fraction of the distance travelled by the solvent front, which is known as the retention factor or Rf. Figure 5.8 shows a TLC plate before development and after visualisation. The plate has been spotted with three materials, two test materials (analytes) labelled A and B, respectively, and one reference sample, marked R. After development and visualisation, we can see that sample A has moved to the same extent (Rf = 0.75) as the reference material and is therefore possibly the same substance. On the other hand, sample B is clearly not the same as the reference material and, moreover, is not a single material but a mixture of two different substances – one with an Rf of 0.5 and one with an Rf of 0.3. It is clear that TLC has uses as a measure of purity of a sample but is more limited in terms of identifying it beyond

  ­Column Chromatograph Set-up at start of development

Plate after visualisation

Development tank

Highest level reached by solvent

TLC plate

Start line

Start line

A

B

R

Solvent level

Figure 5.8  Thin layer chromatography.

doubt. The possibility that two different substances could have the same Rf value always exists. Since the plate was visualised by charring, there is no possibility of carrying out any further identification. The amount of material in any spot will determine the size and darkness of the mark after visualisation, so some degree of quantitative analysis is possible. For example, in Figure 5.8, we might conclude that the faster‐moving component in sample B is present at about twice the level of the slower.

­Column Chromatography Column chromatography usually also uses a stationary phase of finely divided alumina or silica, but, in this case, instead of being coated onto a plate, it is held in a glass column. The starting mixture is added in solvent at the top of the column and fresh solvent passed though the column under gravity, sometimes assisted by pressure of a non‐reactive gas such as nitrogen. The components of the mixture pass down the column at different rates, and each emerges at the bottom of the column at different times. By collecting aliquots (i.e. samples) of the eluent (i.e. material eluting from the column) from the column, it is possible to isolate the separated components as solution and then remove the solvent from them by evaporation. This process allows preparative separation of components from a mixture on laboratory scale. Usually, a TLC plate is run first to identify a good solvent system for column chromatography, and TLC will also be used to identify which fractions of eluent contain which components and to check the degree of success of the separation. The volumes of stationary phase and solvent used in column chromatography make it unattractive on larger scale for our industry.

99

100

5  Separation and Purification

­High Performance Liquid Chromatography HPLC is essentially the same as column chromatography except that the ­stationary phase is even more finely divided, which necessitates the use of high pressure to force the solvent through the column. Detection of the substrate in the effluent is usually by means of ultraviolet absorption or refractive index1 of the material as it leaves the column. This places limitations on the technique as far as fragrance ingredients are concerned, since many of them have little or no ultraviolet absorption in practical wavelength ranges. The material leaving the column is mostly solvent, and so the refractive index method of detection is not very sensitive as even materials with a very different refractive index from that of the solvent will be diluted by large amounts of solvent, thus resulting in a relatively small change in the total.

­Gas Chromatography GC is the most important separation technique used in the analysis of perfumes, essential oils, and fragrance ingredients. In GC, the mobile phase is a gas, usually hydrogen, nitrogen, or argon, or a mixture of these. Nitrogen and argon are used because of their lack of chemical reactivity. Hydrogen is almost always a component of the carrier gas (as the mobile phase is known) for reasons that we will see shortly. The stationary phase is usually a very high boiling liquid; silicones and polyesters are the most common ones in use. Originally the liquid was adsorbed onto a very fine powder and packed into tubes (known as columns), varying from about half of one metre to several metres in length. Modern GC columns use capillary glass tubing, about the diameter of a human hair, in which the inner wall is coated with the stationary phase. The columns are usually between 10 and 50 m in length and are wound into a coil about 15–20 cm in diameter. The coil of tubing is placed in an oven, the temperature of which is controlled and can vary as the analysis is carried out. It is usual to start an analysis at low temperature, for example, between 40 and 80 °C, and then increase it at a set rate up to a maximum of 200 °C or more. Sometimes the temperature will be held steady for a given period at the start, at selected intermediate temperatures, or even during the whole of the programme. In order to be smelt, materials have to be volatile enough at normal temperatures and pressures to reach the nose so that when placed in a gas stream, the materials will tend to move with the carrier gas. The more volatile a material is, the faster it will move along the tube. Therefore, in its simplest form, GC will separate materials according to their boiling points. If the stationary phase is non‐polar (e.g. a silicone), its boiling point will be the prime factor affecting the 1  When light enters one medium from another at an angle other than 90 °C, the beam of light is refracted (i.e. deflected trough an angle). The refractive index (RI) of a material is the degree to which it is refracted.

  ­Gas Chromatograph

time taken for each material to reach the exit point of the chromatography column. However, if a more polar stationary phase (such as a polymeric ester) is used, the more polar analytes will be held back by adsorption in the stationary phase and will therefore take longer to emerge than would be accounted for by their volatility. The choice of stationary phase to be used will therefore depend on what type of separation is required. If two components of a mixture have similar boiling points but different polarities, then they will be separated to a greater extent using a polar stationary phase. Under the same conditions, a material will always take the same time to elute (emerge from the end of the column), so this time can be used as a means of identification. The time taken to elute is known as the retention time, Rt. As with Rf values in TLC, we can conclude that if two materials have different Rts, they are not the same. However, if two materials have the same Rt, they are not necessarily identical, as it is always possible that two different materials could have the same Rt values. When specifying Rt values, it is always important to give details of the column dimensions, the nature and thickness of the stationary phase, the composition of the carrier gas, the flow rate of the carrier gas, and the temperature programme of the oven, as all of these will affect the Rt. It is also common practice to use internal standards in GC. Internal standards are materials deliberately added (usually in a calculated weight ratio) to the mixture to be analysed. The Rt values of the components of the mixture can then be expressed relative to the Rt of the internal standard material, which is more accurate than using direct Rt measurements, since it makes allowances for any unplanned or unnoticed variation in operating conditions. A chemist named Kovats developed the most common system in use, and consequently, the values are known as Kovats indices. Temperature ramping (i.e. raising the temperature of the oven as the analysis proceeds) during a GC analysis is very useful with complex mixtures such as essential oils and perfumes. By starting at a lower temperature, the more volatile components travel more slowly and thus elute with greater time gaps between them. If the starting temperature is too high, the most volatile components will all travel essentially with the gas and elute together. Samples are usually introduced into the instrument in a solvent, so with too high a starting temperature, the more volatile components will in fact just pass through the column with the solvent. If the low temperature is maintained throughout the analysis, the highest boiling components will take an unacceptably long time to elute and thus not produce a sharp peak on the chromatogram. Temperature ramping therefore allows high boiling materials to elute cleanly in a reasonable time. Many methods of identifying when a material is eluting from a GC column are used. The most common is known as a flame ionisation detector (FID). In an FID, the effluent gas stream is burnt continuously in a small flame. In order to burn the gas, it must be flammable, and the most common gas used for this ­purpose is hydrogen as it produces only water on burning, and this will not interfere with products of combustion of the analytes. The hydrogen flame is a clean one, but when an organic compound (i.e. one containing carbon) is burnt, the

101

102

5  Separation and Purification

flame contains a high level of ions, thus enabling the flame to conduct electricity. By placing a pair of electrodes in the flame, its electrical conductivity can be ­measured. Thus, an electrical signal can be generated and recorded in graphical form when an organic compound elutes. The vertical axis of the graph indicates the size of the signal, and the horizontal axis, the time of elution. The area under the peak is proportional to the amount of material which eluted. Such a graph is known as a chromatogram. Some variation in the sensitivity of the detector to one compound relative to another is possible, depending on the molecular structure of each. One important factor controlling the sensitivity of the detector is the number of carbon atoms in the molecule of the analyte, but it is not the only factor. This sensitivity gives us another reason for using an internal standard. If we know the relative response of the detector to the standard and to the analyte and we know how much standard was added, then simple arithmetic will enable us to calculate the actual amount of analyte present in the sample from the relative peak areas for the two. The FID is the most important form of detector used in GC analysis of perfumes. The other two most important detectors are the mass spectrometer (MS) and the human nose. Both of these detectors are almost invariably used in conjunction with an FID. The effluent stream can be split between the two detectors – that is, between the FID and the other – be it MS or nose. This split enables a graphical trace to be produced to show how much material eluted at any given time (via the FID) as well as using a mass spectrum to identify each compound or using a nose to determine how each compound smells. The next chapter gives details of what an MS does and how it can be used to identify chemical compounds, so further discussion here is unnecessary. Use of an MS as a GC detector is known as GC‐MS. The use of a nose seems obvious, but GC‐ sniffing, as the technique is called, is a skill and requires practice and understanding of what is involved. For example, if analysts smell a mixture of musks, they must be aware of the fact that the nose is prone to fatigue where musks are concerned. Thus, if a very strong musk elutes at Rt = x minutes, then a weaker musk that elutes with an Rt of x + 1 minutes may appear to be odourless. A second run is therefore necessary in which the analyst does not attempt to smell the first compound but rather comes to the smelling port with a fresh nose when the second compound elutes. The great value of GC‐sniffing is that it enables the analyst to identify those components that are of importance to the odour of the mixture. The combination of GC‐MS and GC‐sniffing is extremely powerful in the analysis of essential oils, other natural products, and perfumes. GC‐MS alone revolutionised perfumery in that the traditional secrecy of the industry was destroyed. Previously, perfumers were very secretive with their perfume formulae as perfumes were difficult to analyse and copy. Nowadays, with GC‐MS, a perfume can be analysed to identify all of its chemical components in a matter of an hour or so. However, powerful as it is, identification of all of the chemical components is only the start of the process of perfume analysis, as will be shown in the section on GC‐MS in Chapter 6.

  ­Gas Chromatograph Norm. 5000 4000 3000 2000 1000 0

1

1.5

2

2.5

3

3.5

4

4.5

5

min

Figure 5.9  A GC chromatogram of lavender oil.

Figure 5.9 shows a GC chromatogram of a sample of lavender oil. The scale on the horizontal axis shows the retention time in minutes, and that on the vertical axis indicates the relative heights of the peaks. The first peak occurs at about 1.56 minutes and is due to acetone, the solvent in which the sample was dissolved for analysis. This peak is much bigger than any of the others and thus has run off the scale and been truncated at the top of the graph. The first main groups of lavender components elute after 2–2.5 minutes. These more volatile components include monoterpene hydrocarbons such as myrcene and ocimene. Just before 2.6 minutes, we come across the largest component of the oil, the monoterpenoid alcohol called linalool. The groups of peaks between 2.8 and 3 minutes are cyclic monoterpenoid alcohols such borneol and terpineol. The second most significant component of the oil is linalyl acetate that elutes after 3.1 minutes. It is followed at 3.2 minutes by lavandulyl acetate, another monoterpenoid acetate. At 3.7–3.8 minutes we find sesquiterpenoid hydrocarbons such as caryophyllene and farnesene. The small blips behind this tell us that there are oxygenated sesquiterpenoids in the oil. (See Chapter  14 for an explanation of terms such as terpenoid and sesquiterpenoid.) In Figure 5.10, we see the same chromatogram as in Figure 5.9, but the vertical scale has now been enlarged more than five times. The most obvious consequence is that in addition to the solvent peak, the five largest peaks arising from

900 800 700 600 500 400 300 200 100 0

1

1.5

2

2.5

3

3.5

4

4.5

Figure 5.10  GC chromatogram of lavender oil with enlarged vertical axis.

5

min

103

104

5  Separation and Purification

lavender have also run off the top of the scale and been truncated. More importantly, in Figure 5.10, we can see many smaller components appearing along the lower levels that we could not see before. For example, now a group of small peaks is showing around 1.8–1.9 minutes. These materials are clearly very volatile and will have a significant effect on the top note (see Chapter 10 for definition) of the oil. Many essential oils contain tiny traces of volatile sulfur‐containing materials such as methanethiol and dimethyl sulfide, which will elute in this region. They are at very low levels and so are not always easy to see, but their very powerful odours have a significant effect on the total odour of the oil. High impact trace ingredients can occur anywhere along the chromatogram, which the main reason why GC‐sniffing is so important in analysing essential oils and perfumes. In Figure 5.10 we can see that there are several examples of peaks that have not been completely separated from each other. For example, there are two peaks overlapping significantly either side of 2.74 minutes. The computer has added the vertical line between them in order to enable it to estimate the relative areas of the two. Sometimes a small peak can be completely hidden under a larger one. For example, the large peak at 3.74 minutes has a rather asymmetric shape, which suggests that there is a smaller peak immediately behind it. We can employ various methods in order to resolve (that is, to separate) overlapping peaks such as these. These techniques would include using a longer column, using a more retentive stationary phase, using a slower gas flow, using a lower oven temperature, or reducing the rate at which the oven temperature is increased. All of these methods will result in a longer time for the analysis, but this might prove necessary in order to obtain the required information. Sometimes, it is useful to be able to detect the presence of molecules containing certain heteroatoms. sulfur, nitrogen, and halogens are of particular interest in our industry. Sulfur and nitrogen are important because some of their compounds have very intense odours and therefore exert a significant effect on the total odour of a fragrance, even when they are present at levels too low to be easily detected using an FID. Halogenated materials are undesirable in fragrances, and sometimes it is necessary to check that they are absent. Halogens are detected using an electron capture detector. This detector consists of two electrodes. One is made of a radioactive isotope of nickel that emits β‐rays. These are in fact electrons, and they are detected as an electrical signal at the other electrode. Halogen nuclei capture these electrons and reduce the current from the nickel source to the electrode around it, hence giving an electrical signal. There are special GC detectors that can be used to detect sulfur‐, nitrogen‐, or halogen‐containing materials. Sulfur is detected by flame photometry. As sulfur‐containing molecules burn in the flame, they generate sulfur species that are then electrically excited and luminesce on return to the ground state. Measurement of the light emitted then constitutes a measure of the amount of sulfur present. Nitrogen detection is achieved by passing the eluent stream over a heated rubidium silicate bead. Nitrogen in the sample causes emission of ions from the bead and hence generation of an electrical signal. The disadvantage of this technique is that the bead gradually erodes and requires regular replacement.

  Review Questions

Review Questions 1 A drum of perfume was sent to a customer by air freight. When it arrived at the customer’s premises, there were crystals present in the drum. The perfume house was convinced that the perfume was a homogenous liquid when it was dispatched. What is the most likely explanation for this? 2 A perfume has been contaminated by a red dye. What technique would be most useful for isolating a sample of the dye in order to determine its composition? 3 A new chemical compound has been isolated from an essential oil. Which spectroscopic technique will be of most help in determining its molecular structure? 4 It is suspected that lavender oil has been added to a perfume in place of rosemary oil. Which analytical technique would give a definitive answer in the shortest time? If it were found that the suspicion was correct, would it be possible to rectify the mistake by distilling the perfume?

105

107

6 Analysis Chemical analysis  –  that is, the ability to elucidate the chemical composition and/or molecular structure of a substance – is of vital importance to our industry. Analysis is used for research purposes: for example, to discover the molecular structure of a new odorant isolated from a natural source, to confirm that a chemical synthesis has given the desired product, or to determine the nature of an unexpected and unwanted contaminant in a product. It is also the mainstay of quality control (QC) and quality assurance. In the simplest forms of analysis, we measure a property of the analyte (material being analysed) and compare it with a reference standard. If the measured property of the analyte is not identical to that of the standard, then the analyte is not identical to the standard. It must always be remembered that such analysis does not confirm identity. For example, two substances could, by chance, have identical densities but not identical compositions. However, such techniques are usually quick and cheap, and if the material is unlikely to be grossly different from the standard, then it gives a quick check on purity, e.g. for QC purposes. Chemicals can be characterised by examining their physical and/or chemical and/or biological properties. This chapter deals only with the methods of analysis used in the fragrance industry and describes them very much from the viewpoint of our industry. Measurement of physical properties is usually the simplest and cheapest method of analysis but, for the reasons given above, is used only for QC. Chemical analysis tells us a little more about chemical composition of an analyte than physical analysis does, but it is still mostly a tool of the QC department. For investigation of samples of unknown or uncertain composition, the most important analytical tools at our disposal are those coming under the heading of spectroscopic methods. Spectroscopy is the analysis of chemicals by studying how they absorb electromagnetic radiation, principally, infrared (IR), ultraviolet/visible (UV/vis), and radio frequencies. These techniques all give us information about the chemical structure of molecules. We will also look at mass spectrometry (MS), which is extremely useful in the fragrance industry, because of its ability to identify molecules through their MS fingerprints. Analysis of an unknown material is a piece of chemical detective work. Many patterns of analytical tests have been devised in order to produce the maximum Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

108

6 Analysis

amount of information at each stage, so as to find the fastest route to identification. In our industry, the usual procedure is to use a combination of gas chromatography and mass spectrometry. An analytical approach that couples two techniques is known as a hyphenated technique, and GC‐MS is by far the most important analytical technique in the world of fragrance. However, the limitation of GC‐MS is that, while it is very powerful when all of the components of a fragrant mixture are already known, it is not very good when dealing with new chemical compounds that are unknown to the analytical world. In this case, the most powerful tool is nuclear magnetic resonance (NMR) spectroscopy. The disadvantage of NMR spectroscopy is that a pure specimen of the analyte is required. Therefore, if a new chemical compound is discovered as a minor component in an essential oil, the analyst must first separate the new compound away from all the other essential oil components before determining its structure by NMR. MS is very sensitive and can detect mere picograms of material. NMR is less sensitive and requires milligram quantities.

­Physical Methods of Analysis Density Density is the weight of a fixed volume of material, usually expressed as grams/ cubic centimetre. Specific gravity is the density of a material relative to that of water, which is 1 g/cc. This test is easy to carry out but gives relatively little information since many different liquids and liquid mixtures will have the same density. If the density of a sample is not the same as that of a standard, then the sample does not have the same composition as the standard. However, if it does have the same density as the standard, the converse does not necessarily stand, and the sample might or might not be identical to the standard. For this reason, it is normally used in conjunction with other tests. Density is usually measured by accurately weighing an accurately dispensed volume of the liquid. Density varies with temperature, and so the temperature at which the measurement was carried out must be recorded. Melting Point All pure solids have a sharp melting point that is characteristic; the wider the  melting range, the less pure the sample. However, different compounds might have the same melting point, so this test also is used in conjunction with others. Boiling Point Boiling point is much the same for liquids as melting point is for solids. Since a liquid’s boiling point varies considerably with pressure, it is always important to record both the boiling point and the pressure at which it was measured.

­Physical Methods of Analysi

Refractive Index A substance’s refractive index is the degree to which a transparent material will refract (bend) the angle of an incident light beam. Similar comments apply as those for density, and again, the temperature of the measurement is important. Refractive index is measured in an instrument called a refractometer. Optical Rotation Optical rotation is the direction and extent to which a material will rotate the plane of polarised light. It only applies to materials that possess chirality. These materials are said to be optically active. Many essential oil components are optically active, so measurement of optical rotation is a quick check for adulteration of essential oils by synthetic materials as these are usually racemic. However, it is relatively easy for a substandard supplier to readjust an oil’s overall rotation to give a reading that falls in an acceptable range, so a correct optical rotation reading must not be taken as a guarantee of purity unless it is supported by other analytical methods. Flashpoint A material’s flashpoint is the temperature at which a sample of the substance will ignite if exposed to an open flame. It is used for safety in transport and storage. In order to measure the flashpoint of a liquid, a sample is placed in a closed vessel and heated to a set temperature. The lid of the vessel is opened, and simultaneously, a flame is directed into the vapour headspace above the liquid. If the liquid is above its flashpoint, the vapour will ignite, usually with sufficient force to extinguish the pilot flame. This process is repeated at different temperatures until the flashpoint is reached. Viscosity Viscosity is a measure of the resistance of a liquid to flow. It is not usually a QC method but is important to engineers designing a plant for handling of the liquid. It is also of significance for various consumer goods. For example, a thick liquid bleach consists of regular bleach (i.e. sodium hypochlorite) solution to which a thickener (usually a starch) has been added to make it flow more slowly. It is important that any fragrance added to the thick bleach should not reduce the viscosity significantly. Indeed, the fragrance house will endeavour to ensure that the fragrance will not adversely alter any properties of the customer’s product. Colour Fragrance oils, especially essential oils, often exhibit a colour, usually yellowish. This colour is measured by visual comparison with standards.

109

110

6 Analysis

­Chemical Methods of Analysis Titration The most common of the chemical analyses used in our industry involves a technique known as titration. Titrations are based on chemical reactivity. The usual properties being measured are either acid/base reactions (see Chapter 8) or oxidant/reductant, abbreviated red‐ox, reactions (see Chapter  9). An indicator is used to determine the neutral position, known as the end point of the titration. For example, in the simplest system, the acid/base titration, the indicator is a dye such as litmus, which changes colour depending on whether it is in an acidic or an alkaline solution. For example, if we wish to measure the amount of acid present in a sample, we can dissolve a weighed amount in a known volume of pure water and then add an indicator. If the indicator is litmus, the solution will now be red. We then slowly add an alkaline solution of known strength from a graduated vessel known as a burette. At some point, all of the acid is neutralised by the alkali, and then the addition of one more drop of the alkali will turn the solution blue. The result is the end point of the titration, and since we know the amount of alkali used, we can calculate the amount of acid present in the original sample. A back titration is one where we deliberately add an excess of the reagent used in the analysis and then back titrate with another reagent of the same type as the material being analysed. For example, an acid could be determined by adding an excess of alkali and then back titrating with another acid. This method is useful, for example, if the acid being analysed is weak or is poorly soluble and reacts slowly, resulting in an end point that is easy to miss. Another form of titration is the indirect titration. Indirect titrations are those in which a property is analysed indirectly by adding a reagent that will react with the material to be analysed and then titrating to see how much reagent remains unused. Figure 6.1 shows the equipment used in a titration. The burette has a tap at the bottom to allow the standard reagent to be added slowly until the end point is reached. Its side is marked with graduated lines indicating the volume of reagent added. Usually the solution of the analyte is stirred throughout the titration to ensure that mixing is complete and thus that a smooth end point is observed.

Burette

Reagent of known concentration

Tap Solution to be analysed

Figure 6.1  A titration in progress.

­Chemical Methods of Analysi

Acid Content Essential oils and perfumes tend to react with oxygen in the air (see Chapter 9) to form oxidised species, and one consequence of this occurrence is that levels of acidic materials build up. These acids are capable of catalysing further degradation reactions, and so it is important to prevent their build‐up. This degradation is one reason why we need to be able to determine the level of acid in a fragrance or oil. The acid content can be determined by straightforward titration against a base. However, since the perfume oils and the acids they contain are not very water soluble, it is often better to add an excess of base to the oil, stir it vigorously to ensure that all of the acid has reacted, and then determine the amount of base left unreacted by titration against a standard acid. Base Content This analysis is of less importance to our industry but if required can be done either by simple titration against a standard acid or by back titration. Peroxide Content The oxidised species resulting from reaction of atmospheric oxygen with perfumes and essential oils always include peroxides of various forms. These materials will cause further deterioration of the oil, and, even worse, if their content reaches too high a level, they present a safety hazard. This scenario is particularly true if peroxides have built up in the oil before distillation. Higher boiling peroxides will remain in the still pot as the distillation proceeds and can reach a point where their concentration is high enough (and remember the temperature in the pot will be high) to detonate. It is therefore very important to check peroxide levels, especially before distilling an essential oil or other perfume ingredient. This assessment is done by carrying out a red‐ox titration against a reductant such as sodium thiosulfate. (See Chapter 9 for a description of oxidation and reduction and explanations of the various terms involved.) The most common indicator in red‐ox titrations is a mixture of starch with sodium or potassium iodide. The iodide anion is colourless; however, in the presence of an oxidant, it is oxidised to elemental iodine. Iodine is a violet coloured solid but forms a yellow‐brown solution in solvents such as ethanol. In the presence of starch, the iodine molecules migrate into the starch molecules (see Chapter 12), and this complex takes on a very intense blue colour. In the presence of excess reductant, the iodine is reduced to iodide, and the colour disappears. Thus, if an oil containing peroxides is added to a solution of starch and iodide, the solution will turn dark blue. When enough sodium thiosulfate has been added to reduce the peroxides, the solution will suddenly become colourless, and the analyst will know that the end point has been reached and the amount of peroxide present originally can be calculated. Ester Value One measure of the quality of some essential oils is the amount of ester present. For example, lavender oil should contain a certain level of the ester linalyl

111

112

6 Analysis

acetate. Esters are neutral and non‐oxidising, and so straightforward acid/base or red‐ox titration is not possible. However, esters can be hydrolysed by treatment with aqueous base, and the acid produced will neutralise the base. This method, an example of an indirect titration, gives us the potential of measuring the amount of ester present. First the analyte is heated with an excess of aqueous base, such as sodium hydroxide. The esters present are hydrolysed, and each molecule of acid generated is neutralised by one of the base. Titration of the remaining base against a standard acid then tells us how much acid was generated and hence how much ester was present. Of course, if esters of different molecular weights were present, the figure obtained will only be an average. Thus, this test is mostly used only as a QC method, in that various essential oils will have typical ester contents and values obtained can be checked against previous samples. Indeed, the test is almost obsolete nowadays. Aldehyde/Ketone Content Like esters, aldehydes and ketones cannot be titrated directly. To use titration to determine their levels in essential oils and perfumes, hydroxylamine (NH2OH) is added. This reagent reacts rapidly and stoichiometrically (i.e. on an exact molecular ratio) with any aldehydes or ketones present to form oximes (see Chapter 3). Since hydroxylamine is a base, the excess can be titrated with acid. The amount of hydroxylamine consumed, and hence the amount of aldehyde and/or ketone in the sample, can then be calculated. As with the ester titration, this method is essentially obsolete. Phenol Content Phenols are weak acids and so will react with strong bases, but not with weak bases. Thus, acids can be neutralised with a weak base such as bicarbonate, and then titration with a strong base can be used to determine the phenol content. Chemical Oxygen Demand (COD) This analysis is an important measure for effluent (i.e. waste material produced in a factory) and, hence, for any site that manufactures fragrance ingredients. It is related to biological oxygen demand (BOD). BOD is the measure of the amount of oxygen required by bacteria to turn a sample into carbon dioxide and water, thus indicating the ease with which sewage treatment will eliminate a material. BOD determination is time consuming and expensive, so chemical oxygen demand (COD) is used as a quick test for effluent since bacteria use chemical reactions to decompose their food, and thus COD gives a good guide to BOD. It is also a titrimetric analysis. The effluent sample is digested with chromic acid, a very strong oxidising material. Unused chromic acid is then titrated in a red‐ox system against a standard reductant (see Chapter 9). Water Content Water can be determined by physical removal, for instance, by co‐distillation with a solvent, such as toluene, which azeotropes with it. However, often such

­Spectroscopic Methods of Analysi

treatment will fail to remove all of the water or will produce more by dehydrating alcohols. Thus, water is usually determined by a complicated titration known as the Karl Fischer method, of which we need not go into detail here. Atomic Absorption The wavelengths of light that are absorbed when a beam of white light passes through a flame will be affected by any metal ions present in the flame. (see ultraviolet/visible spectroscopy in this chapter for more detail of the absorption mechanism.) By burning a material, and scanning the spectrum of the flame, we can accurately determine very small traces of metal present in the sample. Applications in the fragrance industry include the determination of traces of iron and nickel in samples. Determination of the iron content is important for patchouli oil samples. Iron is a common metal and likely to turn up in all sorts of places. It causes deterioration of patchouli and so must be removed from it. Iron detection by atomic absorption is therefore a good way of ensuring that iron removal treatments have been effective. Nickel is used in many catalysts, but it is a strong skin sensitiser, so we must check that no traces of it get through into perfumes.

­Spectroscopic Methods of Analysis Electromagnetic radiation is a source of energy. Energy is not continuous but exists in discrete amounts called quanta. Each frequency of electromagnetic radiation contains a certain number of quanta of energy. If the energy content of a given wavelength of radiation corresponds exactly to an amount of energy that a molecule can absorb, the molecule will absorb the radiation and convert the energy to another form. If not, the radiation will pass through. Therefore, each molecule has a selected number of wavelengths of radiation that it will absorb. These wavelengths can be measured, and the product of this measurement is called an absorption spectrum. The spectrum will be characteristic of the compound and will give us clues to its molecular structure. Analysis in this way is referred to as spectroscopy or spectrometry. The terms are essentially interchangeable, but spectroscopy is more commonly used, except in the case of mass spectrometry. The instruments used to measure spectra are known as spectrometers. The alternative word, spectroscope, is not used nowadays but does occasionally crop up in older documents. The frequency and wavelength ranges of the electromagnetic spectrum are shown in Figure 6.2. The energy range across the spectrum is huge, ranging from the relatively low energy radio waves at the right‐hand side of the figure to the high energy γ‐rays at the left. The effects of radiation on molecules vary depending on the energy content. Radio waves, as we will see later, affect the magnetic alignment of certain nuclei when they are placed in a magnetic field. Microwaves speed up rotation and motion of molecules, which is what happens in a microwave oven. The microwaves in a domestic oven are tuned to specific frequencies that affect movement of water molecules. As the water molecules absorb the microwave radiation, they move around faster; in other words, they

113

114

6 Analysis

Wavelength (m)

10–10 10–9 10–8 10–7 10–6 10–5 10–4 10–3 10–2 10–1 100

Frequency (MHz)

3 × 1010

Type of radiation

Energy levels of appropriate transitions

3 × 108

X-rays

Ultraviolet

γ-rays

Visible

Atomic electronic transitions

Atomic and molecular electronic transitions

3 × 106

3 × 104

101

3 × 102

102

103

3

Infrared

Microwaves

Radio waves

Molecular vibrations

Molecular rotations

Nuclear magnetic energy levels

Decreasing energy

Figure 6.2  The electromagnetic spectrum.

become heated. Absorption of infrared radiation causes atoms within a molecule to vibrate more vigorously relative to each other, and this method is another way of raising the temperature of the sample. Visible light and ultraviolet close in frequency to visible (hence known as near ultraviolet) are capable of moving electrons from one molecular orbital to another of higher energy. In some ­circumstances, chemical reactions will occur. In other cases, the molecule will return to the ground state and, in doing so, will lose energy to its environment, usually in the form of heat. Far ultraviolet, X‐rays and γ‐rays are even higher in energy, and molecules absorbing such frequencies are likely to break apart, forming ionic species. For this reason, the radiation of these three frequency ranges is known as ionising radiation. In the perfume industry, we use four main spectroscopic techniques to give us information about the chemical structure of materials. A brief summary of these is shown in Figure 6.3. Mass spectrometry is not really a form of spectroscopy. However, the initial output information resembles a spectrum, so the terms mass spectrum and mass spectrometry are now in general use for the technique. Each of these four techniques will be described in more detail below. Ultraviolet (UV) In spectroscopy of organic molecules, we use the term ultraviolet spectroscopy to refer to spectroscopy across the visible part of the spectrum and into the near ultraviolet (i.e. the part of the ultraviolet spectrum closest in frequency to visible light). Any material that absorbs light in the visible spectrum will be either coloured or opaque. Most fragrance materials do not absorb visible light (the exception being Schiff ’s bases of methyl anthranilate that are yellow or red in colour) and so have spectra that show either no absorption or an absorption only in the UV range.

­Spectroscopic Methods of Analysi Spectroscopic technique

Physical process Wavelength

Effect on molecule

Type of information

Strengths

Weaknesses

Infrared

Increased rate of vibration of compounds

Functional groups, rings, skeletons

Fingerprints of known compounds

Little information on connections between structural features

Ultraviolet

Raises electrons from ground state to high energy orbitals

Number of double bonds in conjugation

Very sensitive

Relatively little structural information

NMR

Alignment of magnetic dipole of nuclei

Location of atoms in molecules and presence of neighbouring atoms

Very good for unknown compounds

Open to misinterpretation with mixtures

Mass spec

Ionisation and fragmentation

Molecular weight fragments

Fast, very Limited structural sensitive, information can use with GC, fingerprints of known compounds

Figure 6.3  Spectroscopic techniques.

Everyone is familiar with the fact that white visible light can be split into colours as it passes through media of different refractive indices, for example, as shown in Figure 6.4, on passing from air through a glass prism and back into air again. The same effect in water droplets in the air gives us rainbows. The pattern White light

Red Orange Yellow Green Blue Indigo Violet

Figure 6.4  Splitting of white light into a spectrum.

115

116

6 Analysis Sample containing red dye White light

Red Orange Yellow Green

Figure 6.5  Spectrum of a red dye.

of different frequencies is called a spectrum. The visible spectrum consists of light corresponding to red, orange, yellow, green, blue, indigo, and violet colours. What is happening either in the prism or in the rainbow is that white light is being separated out into light of different frequencies from the lower energy red light to the higher energy violet light. The same thing happens to other frequency ranges such as infrared and ultraviolet radiation. If a sample of dye is placed into the beam of white light, then any frequencies absorbed by the sample will be missing from the spectrum. Thus, as shown in Figure 6.5, a sample that absorbs blue, indigo, and violet light will produce a spectrum without these frequencies. If this light is viewed without splitting it into a spectrum, it will appear red in colour because it is missing the higher energy, higher frequency, and longer wavelengths of light. Figure 6.6 shows the basic elements of a spectrometer. The condensing lens concentrates the white light from the source; then after passing through a fine slit, it is focused into a beam of parallel light by the collimating lens. The sample is placed in this light beam, and the light that passes through is broken into individual frequencies by a prism. Then a detector measures how much light of each wavelength has been absorbed by the sample. The results are plotted out in the form of a graph, called an absorption spectrum or spectrum for short. Figure 6.7 shows what a spectrum of the red dye in Figure 6.5 might look like. The basic principles described here for visible light apply to all forms of spectroscopy (but with some modification in the case of mass spectrometry). Slit

Source of white light

Sample

Red

Violet

Condensing lens

Collimating lens

Figure 6.6  Basic components of a spectrometer.

Prism

Object lens

Detector

­Spectroscopic Methods of Analysi 100

% light absorbed

% light transmitted

0

100

0 420

220 Wavelength (nm)

Figure 6.7  Spectrum of red dye of Figure 6.5.

Spectra of substances are characteristic of the substance and can thus be used for comparison with previous samples or reference standards. If the spectrum of a substance does not match that of a reference sample, then the substance being tested is not the same as the reference. However, spectroscopy can be used to much greater effect than simple comparisons. As described above, the various frequencies of light bring about changes in the molecules that absorb it, and by understanding the detail of these changes, the absorption spectrum can tell us much about the structure of the molecule. Of the three true spectroscopic techniques used in fragrance chemistry, ultraviolet uses the highest energy radiation. The energy content of ultraviolet and visible light corresponds to the amount required to raise electrons in molecules from bonding to anti‐bonding orbitals. In Chapter 1, we saw how electrons in molecules are shared between atoms to form bonds that hold the atoms together. For every such bonding orbital, a corresponding anti‐bonding orbital exists. The names of anti‐bonding orbitals are formed by adding an asterisk to the name of the corresponding bonding orbital. For example, the anti‐bonding orbital corresponding to a σ orbital is known as a σ* orbital. The anti‐bonding orbitals are of higher energy than the bonding orbitals, and the gap between the two matches the energy content provided by a specific frequency of ultraviolet light. Thus, if ultraviolet light of the correct frequency hits a hydrogen molecule, it can promote one of the electrons in the molecular orbital from the ground state bonding orbital to the corresponding anti‐bonding orbital. Clearly this promotion weakens the bond considerably as now only one unpaired electron is holding the two nuclei together. The energy required to move an electron from a σ orbital to a σ* orbital is high and corresponds to far ultraviolet light. Of more use to fragrance chemists are the transitions from π to π* orbitals where the energy gap is lower and corresponds to near‐ultraviolet and visible light. Even so, for isolated double bonds, the corresponding ultraviolet frequencies are too short in wavelength to be easily handled in the laboratory. The more double bonds that are conjugated together in a molecule, the further up towards the red end of the spectrum will be the maximum absorption. Molecules with

117

118

6 Analysis

two or three double bonds in conjugation will still absorb well into the ultraviolet. A molecule such as benzophenone – in which a ketone group is sandwiched between two benzene rings, making a total of seven double bonds in the conjugated system – absorbs ultraviolet light close to the violet end of the visible spectrum. The dyes that are used in consumer goods have much larger systems of extended conjugation, which often include the presence of strongly electron‐ donating substituents such as amino groups or even stabilised carbocations (cations where the positive charge is located on a carbon atom). As we will see in Chapters 9 and 11, this is important in consumer products since breaking of the extended conjugation, for example, by oxidation (bleaching), will destroy the colour of a dye or stain. The energy levels between bonding and anti‐bonding orbitals are affected by the presence of differing vibrational and rotational states of the molecule, and consequently, the peaks in ultraviolet spectra are usually rather broad. They give us information about the extent of conjugation of double bonds in molecules and sometimes about substituents attached to the conjugated system. However, this information is of limited use, and the technique is not one of the most important to fragrance chemistry today. Infrared (IR) Molecules are not rigid billiard‐ball‐like structures. The atoms in them move about relative to each other in various ways as if connected by springs, rather than hard rods. In other words, the molecules vibrate. Figure 6.8 shows some possible ways in which the three atoms in a water molecule can move relative to each other. In the top row, we see the bond between the oxygen atom on one of the hydrogen atoms shortening and lengthening as the ‘spring’ holding them together contracts and stretches. The middle row shows the angle between the two oxygen–hydrogen bonds changing as the molecule undergoes a pincer‐like motion. The bottom line shows one of the hydrogen atoms wagging above and below the plane of the paper, which represents the average plane of the molecule. If we think of this in simple mechanical terms, the energy required to set such vibrations in motion will depend on the weights of the atoms, which are in OH

H

H

O

H

O

H

O

O H

H

H

O H

H

H

H

H

Figure 6.8  Some vibrations in water molecules.

O H

O H

H

H

O

H

­Spectroscopic Methods of Analysi

motion, and the strengths of the bonds holding them together. Each specific energy requirement will correspond to a specific frequency of infrared radiation. The usual measure for infrared radiation is wave numbers, expressed as cm−1. This figure represents the number of waves in a centimetre. If we shine radiation of 2000 cm−1 onto water, it will pass straight through as the energy content of such radiation does not correspond with any vibrational energy levels in water. However, IR radiation with a wave number of 3000 cm−1 will cause the O─H bonds to vibrate more energetically, and this frequency will be strongly absorbed by the water molecules. Therefore, the absorption spectrum of water will show a large absorption at around 3000 cm−1. An infrared spectrometer has a basic set‐up similar to that of an ultraviolet spectrometer but with one major difference. The optical components (lenses and prisms) of a UV spectrometer are made of quartz, which allows all near‐­ultraviolet and visible light to pass through. However, glass and quartz are opaque to infrared radiation, so for an IR spectrometer, it is necessary to use potassium bromide for the lenses and prisms. Because potassium bromide is very water soluble, infrared spectrometers must be kept rigorously dry to prevent deterioration of the optics. The frequencies of infrared light absorbed are very characteristic for  functional groups. For example, carbon–oxygen double bonds absorb very strongly around 1700 cm−1. In fact, it is even more specific since esters generally absorb at about 1730 cm−1, whereas α,β‐unsaturated ketones will absorb at about 1680 cm−1. Figure 6.9 shows some infrared frequencies that are associated with functional groups found in fragrance molecules. In the region between 1200 and 650 cm−1, we find vibrational frequencies that are related to vibrations in the carbon framework of the molecule’s structure. These frequencies are complex and difficult to interpret but are very characteristic of the specific molecular structure, so this region of the spectrum is known as the ‘fingerprint’ region, since infrared spectra can serve for the analyst in the way that fingerprints do for police detectives. If the infrared fingerprints of two samples do not match, then the two samples are not the same. (The converse does not necessarily apply, i.e. two different materials could have very similar fingerprints.)

Wave number (cm–1) 4000

3000

2000

C H

1500

650

C C C Cl

C C O H C N N H

1000

C O C O

S H

Figure 6.9  Some characteristic infrared frequencies.

119

120

6 Analysis

Infrared spectroscopy is more useful than ultraviolet in fragrance chemistry because it can give information about the functional groups present in the molecule of the analyte. However, it still does not tell us how these individual components are assembled in the molecular structure. Nuclear Magnetic Resonance (NMR) NMR spectroscopy uses radio frequency radiation and the effects of strong magnetic fields on certain nuclei. Because they are charged and spinning, nuclei in atoms and molecules behave like tiny bar magnets. Not all nuclei exhibit these properties. Of those that do, the most important to fragrance chemistry are hydrogen (1H), deuterium (2H), and carbon‐13 (13C). Unfortunately for us, the most common isotope of carbon, carbon‐12 (12C), does not exhibit the physical properties required to make it suitable for NMR spectroscopy. Normally, these little magnets are oriented randomly, but, when a strong external magnetic field is applied, they all align with the field. Some align parallel to the field, and some align in exactly the opposite direction as shown in Figure 6.10. The lower energy, more stable state is to align with the field, but given the correct amount of energy, nuclei can be flipped from this position to the higher energy state in which they are aligned opposed to the applied field. The necessary quanta of energy can be supplied by radio frequency radiation. The actual magnetic field felt by an atom in a molecule is affected by the other atoms that are close to it in the molecule. The differences between the fields felt by different atoms in a molecule are tiny compared with the field produced by the electromagnet in an NMR spectrometer; however, we can measure both radio frequency and the size of the applied field very accurately, so these differences between nuclei can be clearly seen. In principle, either the applied magnetic field can be held steady, while the radio frequency can be changed or vice versa. In practice, it is easier to hold the radio frequency steady and vary the applied field by varying the current to the electromagnet. The fields required are very strong, enough to stop a heart pacemaker or wipe the details off a credit card at a range of 1 m. In order to achieve such field strengths, it is necessary to hold the magnet at very low temperature, which is usually done using a bath of liquid nitrogen (−199.5 °C) or liquid helium (−268 °C). Clearly, NMR spectrometers are expensive instruments to purchase and to run, but the information they can give is well worth the investment. The exact position of a hydrogen

No magnetic field

Magnetic field on

Figure 6.10  Alignment of magnetic nuclei in the presence of an external magnetic field.

­Spectroscopic Methods of Analysi

atom in an NMR spectrum is measured relative to that of a selected standard, tetramethylsilane (TMS), Si(CH3)4. This substance shows only one single line in its NMR spectrum since all of the hydrogen atoms are equivalent and it occurs at a higher field than the vast majority of hydrogen signals. Thus, NMR spectral positions are rated on a scale of parts per million downfield relative to TMS. The units are known by the Greek letter δ (delta). In earlier books and journals, the unit τ (tau) is sometimes found. To convert one to the other is simple, τ = 10 − δ. The displacement of an NMR signal relative to that of TMS is known as chemical shift since it depends on the chemical environment of the atom in the molecule. If we take ethanol as an example, the oxygen atom exerts a major effect on the field experienced by all of the hydrogen atoms in the molecule. Those closest to the oxygen atom are shifted furthest downfield. So, if we run a low‐resolution NMR spectrum of ethanol, we obtain a spectrum such as that shown in Figure 6.11. The hydrogen attached directly to the oxygen is shifted most, to just over 5δ. The two hydrogen atoms attached to the carbon that carries the oxygen are next most affected and are found at about 3.6δ. The hydrogens of the methyl group at the far end of the molecule are least affected and show a signal at about 1δ. As is clear from this example, the chemical shift tells us quite a bit about the molecular environment of each individual atom in a molecule. Tables such as that shown in Figure 6.12 are used to enable us to start to work out the structure of an unknown compound from its proton NMR spectrum. Additionally, the area under the peak is proportional to the number of atoms producing it. So, the ratio of the areas of three peaks in the spectrum shown in Figure 6.11 is 1 : 2 : 3 reading from the furthest downfield upwards. As well as feeling the effect of their chemical environment, nuclei also feel the effects of other nuclei close to them and connected through bonds, rather than simple proximity in space. If a hydrogen atom has one close neighbour, it could be aligned with or against it, thus modifying the field upwards or downwards. Statistically, half of the atoms will have one set‐up; the other half will have one down, and so the peak will appear as a doublet rather than a single peak. This structure is shown in Figure 6.13. For a nucleus with two identical neighbours, there are four possible combinations, and two of them will have the same overall effect on the field. Therefore the signal will be split into three peaks in the ratio –CH3 –CH2

Energy absorbed

–OH

6.0

5.0

4.0 3.0 Downfield shift

2.0

Figure 6.11  Low‐resolution NMR spectrum of ethanol.

1.0

0.0 δ

121

122

6 Analysis

R

CH3

O R

O

H H

H O R

R

R

H2 C

R

H

O R

O

H R

H

H

12

11

10

9

8

7

6

5

4

3

2

1

0

ppm

Figure 6.12  Proton NMR chemical shifts for some common functional groups.

One neighbour Two neighbours

Three neighbours

Figure 6.13  Effect of neighbouring bonded atoms.

of 1 : 2 : 1. With three identical neighbouring atoms, as shown in Figure 6.13, there will be four peaks in the ratio 1 : 3 : 3 : 1. If a hydrogen atom has two identical neighbours on one side and three on the other, then its signal will be split into 12, a triplet each of which is further split into a quartet. Thus, if we obtain a high‐resolution spectrum of ethanol, it will appear as shown in Figure 6.14. The chemical shifts are slightly different from those shown in Figure 6.11 as the spectrum in Figure 6.11 is of a pure sample of ethanol. The spectrum shown in Figure 6.14 is in deuterochloroform (CDCl3) solution. It is the standard solvent for high‐resolution spectra. Chloroform dissolves most

­Spectroscopic Methods of Analysi

H H

O H H

Energy absorbed

H

H

TMS 6.0

5.0

4.0

3.0

2.0

1.0

0.0

Downfield shift

Figure 6.14  High‐resolution NMR spectrum of ethanol.

organic materials, and the deuterated analogue has no proton signals of its own to interfere with the spectrum of the analyte. Solvents do affect the spectrum, and so it is important to know which solvent has been used when analysing an NMR spectrum. The hydrogen attached directly to the oxygen has no neighbouring hydrogens close enough to split its signal, so it still appears as a single peak. However, the two hydrogens on the carbon next to the oxygen each have three identical neighbouring hydrogens (those on the methyl group), and so the signal is split into a quartet. Similarly, the hydrogens of the methyl group are split into a triplet by the two hydrogens of the methylene group. The distance between the peaks of the triplet and those of the quartet is identical in size and is called the coupling constant. If a number of pairs of coupled peaks are shown in a spectrum, we can work out which belong to which by measuring the coupling constants and finding those with the same values. As said earlier, it is unfortunate that the most common isotope of carbon does not produce NMR spectra, but we can obtain spectra from the 13C isotope. This isotope is much less abundant than 12C, and so a larger sample is needed to obtain a 13C spectrum than for a 1H spectrum. If a large enough sample is available, a great deal of information can be obtained from its 13C spectrum. The  chemical shifts are spaced out much further than in proton spectra as in Figure 6.15. A technique known as distortionless enhancement by polarisation transfer (DEPT) in which the peaks due to carbons carrying odd numbers of hydrogens appear above the baseline and those with even numbers of hydrogens below it is also used. Those molecules with no hydrogens attached disappear. Using NMR spectroscopy, it is possible to determine how many carbon and hydrogen atoms are present in a compound, which atoms are connected to which, which are their neighbours, and so on. Putting all of the NMR techniques together, it is usually (and for the types of molecules found in fragrances, almost invariably) possible to determine the molecular structure of the analyte. In fact, computer programs that predict the spectrum of a given structure, or conversely work out the structure from the NMR spectrum, are available and usually give quite accurate results.

123

124

6 Analysis

Type of carbon

Chemical shift (δ)

RCH3

0–40

RCH2R

15–55

R3CH

20–60

R3COR

40–80

RC═CR

65–85

R2C═CR2

100–15

Benzene ring

110–160

RCOOR

160–180

RCO2H

175–185

RCHO/RCOR

180–210

Figure 6.15  Chemical shifts in 13C NMR spectra.

NMR is the most powerful spectroscopic tool for structural determination of organic molecules. Its only weaknesses are that it requires pure samples (those of mixtures can lead to confusion and mistakes) and it requires at least milligram amounts, especially for carbon spectra. Mass Spectrometry (MS) The basic components of a mass spectrometer are shown in Figure 6.16. All of the parts shown in the figure are held under a high vacuum. The analyte is introduced either as a solid or liquid on a probe or in a gas stream. The latter is particularly useful as will be described below under the heading ‘Gas Chromatography–Mass Spectrometry’. The molecules of the analyte vaporise under the high vacuum employed (with the assistance of heating, if required) and are hit by a stream of charged particles. The particles used most commonly are electrons. The effect of electron impact on a neutral gas phase molecule is to knock another electron out of the molecule and leave it as a cation.

Electron beam

Ion accelerating voltage + – OC

Sample inlet

Ion beam Collimating slits

Magnetic field

Figure 6.16  Basic elements of a mass spectrometer.

Detector

­Spectroscopic Methods of Analysi

Close to the source of ionisation is a pair of electrodes with the negatively charged one closer to the source of ions than is the positive one. Because the newly formed cations are positively charged, they accelerate towards this negative electrode, and some pass through a narrow slit in the electrode. Thus, a fine beam of fast‐moving positively charged particles has been generated. This beam then passes through a magnetic field, which has the effect of deflecting the particles in the beam to one side. The degree of deflection depends on the mass‐to‐charge ratio of the particles and the strength of the magnetic field. Since the vast majority of particles carry only a single charge, the ‘spectrum’ that results at the detector consists of a series of lines, each corresponding to a specific molecular mass. The mass ‘spectrum’ can be recorded in one of two ways. One way would be to move the detector along the line of the ‘spectrum’ so that it meets one mass value after another. However, the more easy mechanical, and hence more common, method is to vary the electric current to the electromagnet generating the field so that the ‘spectrum’ moves across the position of the detector. If one electron is knocked out of a fragrance molecule, its weight is barely changed since electrons have essentially no mass. Thus pinene, with a molecular weight of 136, will generate a cation with a mass of 136 Da. This cation is called the molecular ion, and its detection will tell us the molecular weight of the analyte. However, the cations generated in the mass spectrometer, in addition to their intrinsic instability, also contain an excess of energy as a result of the collision with the electron beam. They are therefore very unstable and break into smaller fragments. The fragmentation reactions follow standard mechanistic pathways. With most fragmentation reactions, the ion splits into two parts, and since it contains only one charge, one of the fragments will be positively charged, and the other will be electrically neutral, either a neutral molecule or a free radical. (For a definition of the term free radical, see Chapter 2 under the heading, ‘Alkanes, Structural Isomers’.) Since the fragmentation reactions occur in the area between the source and the accelerating electrodes, only the positively charged fragments are detected. Of course, positively charged fragment ions can also break down again to generate yet smaller fragments. The way in which the ions fragment and hence the masses of the cations that result will tell something about the way in which the molecule was put together in the first place. For example, we can look at the fragmentation of ethanol molecules in the mass spectrometer. Five possible fragmentations are shown in Figure 6.17. In each case, there is a free radical and a cation, and the mass of the cation is shown. Therefore, in a mass spectrum of ethanol, we might expect to find, along with the molecular ion with a mass/charge ratio (m/e) of 46, peaks corresponding to masses of 45, 31, 29, 17, and 15. Obviously, some reactions will be more common than others, and so not all of the signals will be of equal strength. In fact, a mass spectrum of ethanol will look rather like that shown in Figure 6.18. The highest peak observed is usually the molecular ion, though sometimes the molecular ion is so unstable that it does not exist long enough to be detected, and, in such cases, the molecular weight must be guessed from the fragments that it produces. The strongest peak is called the base peak, and in the case of ethanol, this peak indicates an m/e of 31 – in other words, the formation

125

126

6 Analysis + H•

+

H3C

CH

45

OH

•

+ H3C

H2 C

CH3

+

+ CH2

OH 31

OH

46 H3C

• CH2 +

+

OH 17

•

•

OH + 17

H3C

CH2

OH

+

+ CH

3

15

+ CH2 29

Figure 6.17  Fragmentation of ethanol in the mass spectrometer.

of the +CH2OH cation resulting from the loss of a methyl radical from the molecular ion. In Figure 6.18, there is a tiny peak at m/e 47. This is due to the tiny amounts of higher isotopes of carbon (13C), hydrogen (2H), and oxygen (17O) that will be present in the substance. This observation will be found for all of the peaks in the mass spectrum. 00

50

0 0

10

15 20 29 30 31 40 Ratio of mass to charge

Figure 6.18  Mass spectrum of ethanol.

46

50

­Gas Chromatography–Mass Spectrometry (GC–MS

Although the mass spectrum of a substance does give information about the molecular structure, it is not usually sufficient to enable us to decide without doubt on what the structure is. However, the fragmentation pattern of a molecule is very characteristic of it and, like the infrared spectrum, can be used as a fingerprint of the molecule. Mass spectrometry has two other great advantages. Firstly, it is extremely sensitive, and only minute traces, in the picogram region or lower, are required to give a good spectrum. Secondly, mass spectra are easily stored in digital form, and libraries of spectra of known substances can be searched and compared very quickly. For instance, a mass spectrum can be compared against 20 000 reference spectra in about one second. Its sensitivity and comparability to known substances make it ideally suited for combination with gas chromatography.

­Gas Chromatography–Mass Spectrometry (GC–MS) The combination of gas chromatography with mass spectrometry (GC‐MS) is probably the most powerful single tool in the analysis of fragrances. In a GC‐MS instrument, the sample is introduced to the mass spectrometer through a GC instrument. Thus, complex mixtures of organic chemicals such as essential oils or fragrances can be passed through a GC and hence separated into individual components, which are then fed into the source of a mass spectrometer and compared against libraries of spectra of known compounds. Since most of the chemical components will already be known to the analyst, comparison of their GC retention time and mass spectrum with those of standards will usually provide a positive identification. The speed of acquisition of mass spectra and comparison against a library is so fast that it can be done as the samples elute from the GC column. Thus, in the space of less than an hour, a fragrance or essential oil can be broken down into its hundreds of components with each of them identified and their relative proportions quantified from the peak areas in the GC. When analysing a perfume, this process is just the start of the exercise. The identification of all of individual chemical components does not tell us the formula of the perfume. For example, what do we learn if limonene is detected in a fragrance? Limonene is a common component of perfumes and is found in them because it is a very common constituent of essential oils (see Chapter 14). All citrus oils contain substantial amounts of limonene (up to 80%), and so the likeliest source of significant levels of limonene in a perfume would be a citrus oil. But which citrus oil is it? That is the question. The analyst will answer that by looking for other components that would be indicative of a specific citrus oil. For example, the presence of nootkatone would strongly imply grapefruit oil. Comparison of the relative proportions of nootkatone and limonene would then tell if grapefruit were the only citrus oil present or if a mixture of citrus oils had been used in the formula. Eugenol as an Example of Spectroscopic Techniques As an example of the type of information that each spectroscopic technique gives about a molecule, we can look at the spectra of eugenol, the major component of

127

6 Analysis 100 90 % Transmittance

128

80 70 60

O

50 HO

40 30 4000

3500

3000

2500

2000

1500

1000

500

Wave numbers

Figure 6.19  Infrared spectrum of eugenol.

clove oil. The structure of eugenol is shown on each spectrum to assist in correlating one with the other. The IR spectrum of eugenol is shown in Figure 6.19. The broad, double peak between 3400 and 3500 cm−1 tells us that both C─H and O─H bonds are present. The latter usually appear as broad peaks because hydrogen bonding increases the number of vibrational energy transitions. The absence of absorption in the 1650–1750 cm−1 region shows that no carbonyl functions are present (i.e. no aldehydes, ketones, acids, or esters). However, the peak at 1600 cm−1 does indicate the presence of a C═C double bond. The complex pattern of peaks between 1500 and 500 cm−1 gives relatively little information, though a pattern is present in the 650–950 cm−1 region, which is suggestive of a 1,2,4‐trisubstituted aromatic ring. However, this data cannot be taken as indicative without other supporting evidence, as other structural features could account for some of the peaks. This amount of information falls far short of enabling us to determine the structure. The UV spectrum of eugenol is shown in Figure 6.20. It gives us even less information than the IR did. The large absorption at the violet end of the spectrum is called end absorption. Most organic molecules will show absorption here, and hence it gives us little useful information. The main absorption maximum is at just over 280 nm, which is consistent with a single isolated aromatic ring. Detailed comparison with spectra of other aromatic systems might suggest oxygenation in the ring, but this would be a far from certain conclusion. Figure 6.21 shows the 1H NMR spectrum of eugenol. The curves above the main trace of the spectrum are the integral curves (i.e. curves derived mathematically by integration of the spectral trace) that show the area under the peaks and hence the number of protons responsible for each. The vertical heights of the steps in the integral line are always in simple multiples of each other, and quick comparison in this case shows us that the peaks are due to 1, 2, 1, 1, 2, 3, and 2 protons, respectively, reading from left to right. There is a tiny peak at about 7.3δ that is due to traces of chloroform in the deuterochloroform used as solvent. The

­Gas Chromatography–Mass Spectrometry (GC–MS 2.000

Absorbance

1.600

O

1.200

HO

0.800

0.400

0.000 220.00

260.00

300.00

340.00

380.00

420.0

nm

Eugenol

Figure 6.20  Ultraviolet spectrum of eugenol.

c

O

b

b

g

e f

HO a

c

d

c f

g e

d

COCl3 PPM

10

9

8

7

6

a

5

4

3

2

1

Figure 6.21  Proton NMR spectrum of eugenol.

next peak is at 6.95δ and corresponds to the proton para‐ to the methoxy group and labelled d in the figure. Just above it, at about 6.8δ, are two peaks very close together that correspond with the other two aromatic protons, labelled c in the figure. The next peak, at about 6.5δ, is a triplet, indicating that it is coupled to two neighbours. Its position tells us that it is attached to an olefinic bond but is not at the end of a chain. This corresponds to the proton labelled f in the figure. The other peak showing the same coupling is the doublet at just under 5.2δ. This peak integrates for two protons and corresponds to protons at the end of a terminal double bond, thus confirming its association with the triplet. In other words, these are the protons labelled g in the figure. Between these two is a broad singlet at about 6.7δ that corresponds with a phenolic proton (labelled a). The

129

130

6 Analysis

strong singlet, at about 3.95δ and integrating for three protons, is highly characteristic of a methoxy group (labelled b). The last signal, at about 3.4δ, has a shift in keeping with that of the protons of a methylene group sandwiched between an aromatic ring and a double bond, consistent with those labelled e in the figure. By piecing all of these molecular jigsaw pieces together, we can derive the structure of eugenol uniquely from its proton NMR spectrum. The peak assignments are indicated on the spectrum in the figure. Similarly, the 13C spectrum would also provide sufficient information to assign the structure unambiguously from the spectrum. The DEPT spectrum is shown in Figure 6.22. Two peaks, which would be present in the simple 13C spectrum, are missing from the spectrum as carbon atoms with no protons attached are eliminated in DEPT spectra. The two peaks corresponding to carbon atoms with two protons attached appear below the line. The methylene between the double bond and the aromatic ring shows at about 40 ppm, and the terminal carbon of the double bond at about 114 ppm. The carbon of the methoxy group appears at about 57 ppm, and, since it has an odd number (three) of protons attached, it appears above the line. The other double bond carbon and the three of the aromatic ring, which do carry hydrogens, all appear as peaks above the line between 110 and 140 ppm. The MS of eugenol is shown in Figure 6.23. The base peak is that of the molecular ion with a m/e of 164. The most significant fragment ions have m/e values of 149, 131, 103, 77, 55, and 39. The highest of these ions shows a loss of 15 mass O

HO

PPM 220 210 200 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10

Figure 6.22  13 Carbon DEPT spectrum of eugenol.

0

­Gas Chromatography–Mass Spectrometry (GC–MS 164

MeO 77 55

103 131

39

25

50

149

CH2CH=CH2

HO

75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500

Figure 6.23  Mass spectrum of eugenol.

units from the parent ion and thus suggests the loss of a methyl radical, in this case from the methoxy group. Loss of 33 mass units to give the peak at 131 probably arises by loss of a methoxy radical and two hydrogen atoms. Ions of mass 77 are usually found in the mass spectra of aromatic compounds, and so on. As with IR and UV spectra, we have insufficient information here to give an unambiguous structural determination, but the strength of the MS lies, as discussed earlier, in producing a ‘fingerprint’ that is readily amenable to automated comparison with reference spectra. Quality Control The two key techniques for quality control of fragrances are organoleptic evaluation and gas chromatography. In both cases, the new batch of fragrance is compared against a standard. It should match in both odour and GC profile. Of these two, the odour is the more important, since the fragrance is designed to produce an odour effect and minor variations in chemical composition often have relatively little effect on the odour. In contrast, the odour can be significantly changed by a variation in a powerful odorant even if this change is too small to be seen by GC. Odour assay requires highly trained individuals with an acute sense of smell, the ability to detect very subtle differences in odour, and a vast experience of smelling. Maintaining odour references is a problem as all organic materials degrade over time. Therefore, standards are kept in a fridge and replaced only by something, which approaches them as closely as possible. Synthetic ingredients are relatively easy to QC. They are mostly bought from large, reputable companies, and so a quick check of some simple physical properties, such as refractive index, along with an odour check, will prove all that is needed. Essential oils and other natural products are much more difficult to assay. Some variation in composition will always occur due to weather, soil, and other such factors, and therefore analytical standards are usually set in broad ranges rather than in fine tolerance limits. Dishonest traders will therefore find some scope for adulteration of oils. For example, an oil containing linalool could be diluted with synthetic linalool, which costs a mere fraction of the price of the oil. The QC analyst must therefore find methods of detecting such attempts at fraud. In the case of linalool, one way is to look for the presence of

131

132

6 Analysis

dihydrolinalool, which does not occur in nature but is a common trace component in synthetic linalool. Natural oils usually contain chiral components and therefore display optical properties. Thus, the value of the rotation of plane polarised light might indicate if a sample had been diluted with a racemic synthetic material. However, the optical rotatory power of essential oils does vary, and so the technique is not fool proof. The analyst must always remain alert and try to anticipate what might be done to adulterate the oil and then develop methods to detect it. In the flavour business, naturalness is important and is usually determined by looking at the isotopic composition of ingredients. The simplest thing is to measure the radioactivity of the sample. All natural fragrance and flavour materials show a low level of radioactivity because the plants that produced them take their carbon feedstock from the air in the form of carbon dioxide. Cosmic radiation in the upper atmosphere ensures that atmospheric carbon dioxide contains a level of 14C, the radioactive isotope of carbon. Once incorporated into living matter, it begins the process of radioactive decay, and the level of radioactivity falls with time. This process is the basis of ‘carbon dating’ of ancient remains. If a sample of a flavour or fragrance ingredient does not display a high enough level of radioactivity, then it has not been obtained from a recently living source but probably from petrochemicals. Plants also pick up their feedstocks from the environment where they grow, and so the isotopic composition can also tell us where the material originated. For instance, by looking at the distribution of 12C, 13 C, 14C, 1H, and 2H atoms in a vanilla molecule, it is possible to tell whether it is synthetic or natural and if natural, whether the bean grew in Mexico, Reunion, or Madagascar.

Review Questions 1 A drum labelled ‘linalool’ has arrived at your site. The QC laboratory reports that a sample of the contents had a density of 0.854 g/cc. Your records give the density of linalool as 0.861 g/cc. Do you accept the drum or carry out another analytical check? 2 You suspect that the contents of a drum of an aldehydic fragrance ingredient have been exposed to air and that, as a result, some of the aldehyde has been oxidised to the corresponding acid. Which analytical technique would you use to investigate? 3 You have isolated a new compound with an interesting odour from an essential oil. Which technique would you use first to gain information about its molecular structure? 4. Your customer complains that an off‐odour has appeared in a bar of soap containing your fragrance. Which analytical technique would you turn to first in trying to discover what has happened?

133

7 Chemical Reactivity We live in an ever‐changing universe, and many of the changes around us are chemical in nature. Everything in the universe is composed of chemicals, so when we add perfume to a bar of soap, a bottle of shampoo, or a packet of laundry powder, we are mixing chemicals together, and reactions can occur. Chemistry is used by plants to make essential oils, and chemists use the same chemical processes to make synthetic ingredients. In order to understand how perfume ingredients are made and what happens to them when they are incorporated into consumer goods, we need to understand the forces driving chemical reactions, which is what we will cover in this chapter.

­The Three Laws of Thermodynamics Thermodynamics is the study of the relationship between heat and energy. Energy is defined as the ability to do work, and heat is actually a form of energy since it can be used as a source of power, for example, in a steam engine. Three basic laws govern thermodynamics, the first two of which have significant implications for the study of chemical reactions. The third law of thermodynamics states that the absolute zero of temperature cannot be attained. This law is not really of relevance to us here, so we will concentrate on the first two laws. The first law of thermodynamics states that energy can neither be created nor destroyed, but can only be transformed from one form to another. For example, if infrared radiation (a form of energy) strikes a water molecule in the sea and causes it to heat, the energy of the infrared light has now been converted to heat energy. This conversion might cause the water molecule to evaporate from the sea and rise into the air. It cools, but as it does so, its energy is transformed into potential energy, because it has been moved further away from the earth and has worked against gravity in doing so. If it now falls as rain on a mountain, it loses some of its potential energy (which is dissipated as heat when it hits the rocks of the mountain) but still has potential energy that can be released as it travels down the mountainside in a stream. If a hydroelectric dam is placed across the river into which the stream flows, then some of the potential energy can be converted into electrical energy. This energy might be used to power a heater in Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

134

7  Chemical Reactivity

someone’s house, so the electrical energy is converted back into heat. The heat released eventually goes back into space as infrared radiation, and so the cycle continues. The energy has been changed from one form to another but is never destroyed. As far as chemistry is concerned, an important form of energy is enthalpy, the form of energy that is stored in molecules. For example, plants use sunlight as a form of energy to convert carbon dioxide and water into organic molecules and oxygen. Wood is made of a mixture of cellulose and lignin polymers that contain stored chemical energy. When we heat a piece of wood in the presence of oxygen, it burns – that is, it is converted back into carbon dioxide and water – and the stored chemical energy (enthalpy) is released as heat energy. Matter can be converted to energy and vice versa. However, this conversion is not part of normal chemical processes. As far as fragrance chemistry is concerned, we can also accept that matter cannot be created or destroyed. The second law of thermodynamics states that heat will not flow from a cooler body to a hotter one without other complementary changes. We all know that if a hot object is placed in contact with an ice cube, the hot object will become colder and the ice cube will melt as it becomes warmer. The opposite process – in which the hot object would grow hotter and the ice cube colder – just does not happen. So, heat flows from the hotter body to the colder one. In other words, events happen in a natural direction. To give an example that does not involve heat flow, we could imagine two glass vessels connected to each other by a tap. If the air pressure inside one of them is reduced, it will contain a partial vacuum. If then the tap between the two is opened, air will flow from the vessel at higher pressure into the one with lower pressure so that the final pressure inside both becomes equal. We never see the air from the low pressure vessel flow into the other to increase its pressure further and leave a total vacuum in the other. Similarly, if each vessel is filled with a different gas, say, hydrogen in one and nitrogen in the other, the gases will flow between the two when the tap is opened, so that the final state is one where both vessels contain the same concentrations of hydrogen and nitrogen. This phenomenon is known as entropy. In trying to picture what entropy is at the molecular level, we can see it as a drive towards randomness or disorder. So, in our last example, having all the nitrogen in one vessel and all the hydrogen in the other is a more ordered state than the final one where both hydrogen and nitrogen atoms are equally distributed between the two. In order to achieve the reverse of the natural flow, we must expend energy. In a refrigerator, we do take heat out of a body (the food in the refrigerator) and move it to a warmer one (the air in the room), but we must consume energy in doing so, as the electricity metre will verify. Similarly, we can create a vacuum in a glass vessel but only with the expenditure of energy. One summary of the first two laws of thermodynamics is therefore that, while the energy of the universe remains constant, the entropy of the universe always increases.

­Free Energ

Some Key Facts About Chemical Reactions Energy is conserved; it cannot be created or destroyed. Matter is conserved; it cannot be created or destroyed in a chemical reaction. In principle, all chemical reactions are reversible. The extent to which a chemical reaction proceeds is governed by two factors: enthalpy and entropy. Crudely speaking, enthalpy relates to the energy stored in a molecule, and entropy to the molecular disorder of the system. The speed of a chemical reaction is controlled by the energy needed to deform the reagents into what is known as a transition state. Catalysts accelerate reactions by making it easier to reach the transition state. All organic reactions fall into four basic categories – cationic, anionic, free radical, and electrocyclic (also known as pericyclic). In the first two, electrons move in pairs, in the third they move singly, and in the last, they move in unison.

­Free Energy Enthalpy and entropy are the two main forces that drive chemical reactions. Enthalpy is referred to using the capital letter H and entropy by a capital S. As in the examples above, it is the balance between the two that determines what will happen, and in the case of chemical reactions, the final outcome is known as free energy, represented by the capital letter G. This term was introduced by Josiah Willard Gibbs and is sometimes referred to as Gibbs’ free energy. The enthalpy/ entropy balance is given by the Gibbs equation shown below, in which change is indicated by the Greek capital letter Δ (delta) and T represents the temperature in degrees absolute. (The absolute scale of temperature is also known as the Kelvin scale. The lowest point is 0 °K and is also known as the absolute zero of temperature, and it is equivalent to −273 °C.) In other words, the change in free energy is equal to the change in enthalpy minus the product of the change in entropy and the temperature (in degrees absolute): The Gibbs equation

G

H T S



One obvious implication of this equation is that enthalpy and entropy work against each other. The entropy term is multiplied by temperature, and therefore entropic effects become more significant as the temperature rises. So, for ­example, if we have a reaction in which two starting molecules come together to form a single molecule as the product, this result would be unfavourable in entropy since order would increase. Such a reaction would become less favourable as the temperature increased. Conversely, a reaction in which one molecule broke into two fragments would be more favourable at higher temperatures. We will see further examples of the Gibbs equation at work as we look at some examples of chemical reactions in this and later chapters.

135

136

7  Chemical Reactivity

­Chemical Reactions Let us now look at some of these principles in action in simple chemical reactions. Because matter is neither created nor destroyed in them, all chemical reactions can be written as a balanced equation, with the same number of atoms on each side, and in practice this result can be measured by a mass balance. A mass balance is a record of weights in and out of a reaction, and, of course, the total weight in must equal the total weight out. So, if we consider the neutralisation of sulfuric acid (H2SO4) by sodium hydroxide (NaOH) to give sodium sulfate (Na2SO4) and water (H2O), we can write it as a balanced equation as shown in Eq. (7.1). In order to balance the equation, the total number and type of atoms is the same on each side, that is, there must be four hydrogen atoms, one sulfur atom, five oxygen atoms, and two sodium atoms on each side. This illustration shows us that two equivalents of sodium hydroxide will be required for every equivalent of sulfuric acid and that the reaction will produce one equivalent of sodium sulfate and two of water. Since the weights will be balanced in reality, the equation also tells us that 98 g of sulfuric acid will react with 80 g of caustic soda to give 142 g of sodium sulfate and 36 g of water. That is, 178 g of starting materials go in and 178 g of product come out: H 2 SO 4

2NaOH Na 2 SO 4

2H2 O (7.1)

Anyone who has lived in an area with hard water will be familiar with the white limescale that builds up on the inside of kettles. Our second example, shown in Eq. (7.2), looks at the process involved when a kettle is descaled by treating it with vinegar. The limescale is calcium carbonate (CaCO3); the active ingredient in the vinegar is acetic acid (CH3CO2H). When the kettle is free from scale, the calcium has been converted to calcium acetate (Ca(CH3CO2)2), which is water soluble and can be rinsed out. If you weigh the kettle and the vinegar before and after the treatment, it will seem that weight has been lost. However, the weight has only been lost from the kettle and its contents; matter has not been destroyed. The ‘missing’ weight is actually carbon dioxide (CO2) that has escaped into the atmosphere. Thus, 100 g of calcium carbonate reacts with 120 g of acetic acid to give 158 g of calcium acetate, 44 g of carbon dioxide, and 18 g of water – 220 g in to the reaction and 220 g come out: CaCO3 2CH3 CO2 H Ca CH3 CO2

2

CO2

H2 O (7.2)

The equation by which the limescale formed in the kettle is slightly more complex. Water that is collected in regions with chalk or limestone rocks contains calcium bicarbonate (also known as calcium hydrogen carbonate or Ca(HCO3)2). This is formed when carbon dioxide from the air dissolves in the water to form carbonic acid (H2CO3), and this then dissolves the calcium from the rock. When the water is boiled in a kettle, the reaction shown in Eq. (7.3) occurs. The calcium cation now forms calcium carbonate (CaCO3), and one of the carbonic acid units of the original calcium bicarbonate decomposes into carbon dioxide and water.

­The Principle of Microscopic Reversibility and Chemical Equilibriu

One molecular species has become three in the reaction and thus entropy has increased. Since this example is an entropically favoured reaction, it is temperature dependent, which is why, when cold, the calcium bicarbonate remains as it is but, when heated, the reaction takes place. Another factor driving the reaction forward is the fact that the carbon dioxide is driven out of the system as a gas and so is lost from the reaction medium and is therefore incapable of recombining with the other reagents in a reverse reaction: Ca HCO3

CaCO3 CO2

2

H2 O (7.3)

­ he Principle of Microscopic Reversibility and Chemical T Equilibrium This principle dictates that every microscopic step in the course of a chemical reaction can be reversed. Therefore, in principle, every chemical process is reversible. Therefore, by mixing starting materials together, we generate a mixture of these and the reaction products. So, in the example of Eq. (7.4) where A and B react together to form C and D, C and D can also react together to give A and B. So, the result is a mixture of all four compounds. This result is called an equilibrium mixture, and the system is said to be in chemical equilibrium. Rather than writing it with an equal sign as before, it is more appropriate to write it with a double‐headed arrow as shown in Eq. (7.5). The balance of components in the mixture will depend on the relative rates of the forward and reverse reactions. In Eq. (7.5), if the reaction from A + B to C + D is faster than the reaction from C + D to A + B, then there will be more C and D in the equilibrium mixture than A and B: A B C D (7.4) A B  C D (7.5) The equilibrium position can be changed either by removing one of the components from the system or by adding an excess of one of the components. To illustrate this equation let us look at the equilibrium between an alcohol, a carboxylic acid, an ester, and water as shown in Figure 7.1. +

Alcohol

OH +

PTBCHol

Acid

O

+

Ester

OH

O O

Acetic acid

Water

PTBCHA

Figure 7.1  An esterification/ester hydrolysis equilibrium.

+

H2O

Water

137

138

7  Chemical Reactivity

The top line in the figure gives a generic scheme of alcohol and acid reacting to give ester and water, and vice versa. The drawings show a specific reaction of interest in fragrance. The alcohol in question is para‐tertiary‐butylcyclohexanol or PTBCHol, for short. Esterification with acetic acid gives para‐tertiary‐butylcyclohexyl acetate (PTBCHA) and water. The alcohol is odourless and acetic acid smells of vinegar; however, the ester has a pleasant fruity and woody odour and is an important fragrance ingredient. One way of making PTBCHA is by the process shown. However, the reaction is reversible and so, if carried out in a typical reactor, the ester will be formed in a mixture with the other three components. If this result were all that could be achieved, then the manufacturer would be faced with time‐consuming and costly separation and recycle operations. In order to drive the reaction to completion towards a pure ester, the simplest thing is to add an inert solvent such as hexane or toluene that azeotropes with water and add a Dean–Stark trap to the reactor (see Chapter 5). Thus, if the reaction is carried out at or above the boiling point of the azeotrope, all water will be removed as it is formed, and the reaction will be driven to completion. When PTBCHA is used in a perfume and incorporated into a functional product containing water, a shampoo, for example, then the right‐hand side of Eq. (7.5) will be overloaded with water, and there will be a serious risk of most of the PTBCHA being hydrolysed back to the odourless alcohol. This result is especially true if there is a suitable catalyst, such as an acid or base, present. As we will see in Chapter 11, many of the consumer goods that contain perfume also contain acids or bases, which accounts for the relatively poor performance of esters in these products.

­Reaction Profiles A useful tool in helping to visualise what goes on in a chemical reaction is the reaction profile. In this graph, the horizontal axis represents the time course of the reaction, and the vertical axis, the energy content of the species involved. A typical reaction profile is shown in Figure 7.2. In the figure, we see the reagents at the left. Initially, energy must be put into the reagents in order to form the transition state. This energy is known as the energy of activation. The transition state can either go back to starting materials or go on to products. The amount of energy released is determined by the difference between the starting point and the final position. This difference is known as the heat of reaction and is obtained from the chemical energy stored in the starting molecules, i.e. from the enthalpy. The amount of energy required to form the transition state controls the rate of the reaction. If we compare the reaction course to mountaineering, then the higher the mountain, the harder it is to climb. By providing energy, we can accelerate reactions by helping systems to climb the mountain of activation energy. The simplest way of doing this is to heat the reaction. A quick rule of thumb is that the rate of an average reaction doubles with every 10 °C rise in temperature. (This rule of thumb is very approximate but works reasonably well for the types of reactions we will cover in this book.) If the

­Reaction Profile

Transition state

Energy content

Activation energy Reagents Heat of reaction

Products Reaction

Figure 7.2  A typical reaction profile.

amount of heat liberated is high, especially if it is higher than the energy of activation, and there is no mechanism for removal of the heat of reaction, then the reaction will accelerate once it has started. We call this phenomenon a thermal runaway. If fast enough, it becomes an explosion. In the reaction shown in Figure 7.2, the heat of reaction is positive, that is, heat is given out by the reaction and is called an exothermic reaction. Not all reactions are exothermic. Endothermic reactions consume heat, and their reaction profile is different from that shown in Figure 7.2, since the final state is at a higher level than the starting point. Figure 7.3 shows reaction profiles for an exothermic and an endothermic reaction. Spontaneous endothermic processes may seem counterintuitive, but they do occur. A simple example, though not a chemical one, is the evaporation of water. Water stored in a porous earthenware jar will stay cool if stored in a dry climate, as long as it is not kept in direct sunlight or has some other source of heat available. This process occurs because the water seeps through the porous jar and evaporates from the surface. Evaporation is an endothermic process, very much so in the case of water where the latent heat of evaporation (see Chapter 4) is high and the necessary heat is taken from the body of the water, thus cooling it. The process takes place because of entropy

(a)

(b)

Figure 7.3  (a) Exothermic and (b) endothermic reactions.

139

140

7  Chemical Reactivity

TNT

Coal + air CO2 + H2O + N2

(a)

CO2 + H2O + N2 (b)

Figure 7.4  (a)‘Burning’ of TNT versus (b) burning of coal.

since water molecules are less ordered in the vapour state than they are in liquid. Figure 7.4 shows some of these effects in operation and explains the difference between ‘burning’ of trinitrotoluene (TNT) and burning of coal. The activation energy for ‘burning’ of TNT is much lower than that of coal, but the heat of reaction of coal combustion is much greater. Therefore, TNT will decompose much faster than coal will burn, but coal will produce more heat than the same weight of TNT. The speed of production of gases (carbon dioxide and nitrogen) is what makes TNT a good explosive, and the amount of heat released from controllable burning is what makes coal a good fuel.

­Catalysts A catalyst is defined as a material that speeds up a chemical reaction without being changed itself in the process. The catalyst does actually undergo change, but that change is reversed as the reaction proceeds and the catalyst is returned to its original state. In other words, it undergoes no overall change. What the catalyst does is to reduce the energy of activation by providing an easier path by which the reagents can achieve the transition state. The most superb catalysts are those found in nature. These catalysts are called enzymes, and we will learn more about them in Chapter  12. By the principle of microscopic reversibility, catalysts will accelerate both forward and reverse reactions. Thus, addition of a catalyst to a mixture will mean that the equilibrium is reached more quickly, but the final ratios of components will not be changed.

­Types of Organic Reactions Fragrance ingredients are organic chemicals (i.e. chemicals with structures based on carbon), and so their chemistry is part of organic chemistry. Chemical reactions basically occur when instability or imbalance exists in molecules. Stable, neutral atoms such as helium, or molecules such as nitrogen (N2), require considerable persuasion to undergo chemical reactions. Organic reactions can be classified into four groups, cationic, anionic, free radical, and pericyclic. Figure 7.5

­Types of Organic Reaction H

H

H

H

–H+

H – H

·

–H

H

H

H

H

Carbanion

Free radical

–H–

H

+ H H

Carbocation

Figure 7.5  Formation of ions and radicals from methane.

shows three ways of removing a hydrogen atom from an organic molecule, methane in this case, in order to produce an unstable, and hence reactive, species. On the left‐hand side of Figure 7.5, only the hydrogen nucleus has been removed. Therefore, the resulting carbon atom has an orbital with two electrons in it, but no other atom attached. A quick count of protons and electrons will reveal that it now has one more electron than protons, so the species is an anion carrying one negative charge. Such species are known as carbanions. In the centre, the hydrogen nucleus has been removed together with one electron leaving a carbon species that is neutral but contains an unpaired electron. Such a species is known as a free radical. On the right, the hydrogen nucleus and both electrons of the carbon–hydrogen bond have been removed, thus leaving a positively charged species known as a carbocation. These three species are all unstable and hence reactive. The carbanion and carbocation shown in Figure 7.5 are very unstable and are rarely seen in real chemistry. The carbanions and carbocations that we come across in fragrance chemistry are usually stabilised in some way or other relative to their methyl counterparts, but they are still reactive and form the basis of the chemistry we will discuss in subsequent chapters. Chemical reactions involve transfer of electrons from one entity to another, which are represented in figures by curled arrows. A normal double‐headed arrow is used when two electrons are involved, and a single‐headed arrow when only one electron is involved. Using this convention, we can rewrite the reactions of Figure 7.5 as shown in Figure 7.6, now also including the reagents used to carry out the abstraction of the various hydrogen species. In Figure 7.6a, we see the electron pair of an anion forming a bond between the anion and one of the hydrogen atoms of the methane molecule. The electron pair that formed the bond between that hydrogen atom and the carbon is then displaced onto the carbon atom to form a methyl carbanion. In Figure 7.6b, one electron from a radical and one from the C─H bond join together to form a new bond between R and the hydrogen atom. The other electron of the C─H bond moves onto the carbon atom to form a free methyl radical. In Figure 7.6c, a cation removes the hydrogen atom and the pair of electrons that bonded it to the carbon atom to leave the methyl carbocation.

141

142

7  Chemical Reactivity

H

X–

H

H

H

H

H – H H

(a)

H

H

H

H H

H

X+

H

H + H

H

H (b)

R

H

(c)

H

Figure 7.6  Mechanistic illustration of formation of ions and radicals from methane.

In Chapter 3, we saw how carbon–oxygen bonds are polarised with a partial negative charge over the oxygen atom and a corresponding partial positive charge over the carbon atom. This polarisation applies to carbon–oxygen double bonds also, and consequently all aldehydes, ketones, and esters have a partial positive charge on the carbonyl carbon. (The adjective carbonyl refers to any carbon– oxygen double bond.) The inductive effect occurs when the effects of charges are felt through bonds. Therefore, the positive charge on a carbonyl carbon is felt not just on that carbon atom but also on adjacent carbon atoms and on the hydrogen atoms attached to them. Thus, as shown in Figure 7.7, the hydrogen on the ­carbon α (i.e. immediately adjacent) to an aldehyde group is positive enough to be able to be removed by an anion, and the electrons are pulled onto the oxygen atom of the aldehyde. The resulting species is known as an enolate anion since it could also be considered to have been formed by removal of a proton from the enol form of the aldehyde. The oxygen atom and the α‐carbon share the negative charge on the enolate anion, and so, reactions can occur at either of these atoms. This resonance between the two canonical forms (see Chapter 2) also provides stabilisation for the anion. In practice, this process is the most important way in which carbanions are formed. If we consider a carbon–hydrogen bond to be completely polarised, then we end up with a positive charge on the hydrogen atom and a negative charge on the carbon. Based on this concept, which is called hyperconjugation, alkyl residues tend to donate electrons to their neighbours. Therefore, a primary carbanion is more favoured than a secondary; a secondary is, in turn, more favoured than a tertiary; and so on. As the number of alkyl groups increases, so does the addition of more negative charge from hyperconjugation. It is obviously not favourable to push more electronegativity onto a negative charge. Conversely, carbocations are O

O

H X–

Figure 7.7  Formation of an enolate anion.

–

–

O

­Types of Organic Reaction

+

+

is preferred to H H

H H

H

H

H

H

H

H

H

H

H –

is preferred to

–

H H

H

+

+

is preferred to

H

H+

is preferred to

–

Figure 7.8  Stabilisation of anions by hyperconjugation.

stabilised by having more alkyl groups around them. This example is illustrated in Figure 7.8. In the centre of the figure, two canonical forms of a tertiary‐butyl carbocation are shown to show how hyperconjugation stabilises the positive charge. Charges, particularly positive charges, are also stabilised by conjugation with double bonds and aromatic rings. Figure 7.9 shows such stabilisation in action using both line drawings of bonds and also a representation of the orbitals involved. From the point of view of fragrance chemistry, the two most important ways of forming carbocations are by addition of a proton to a double bond or by elimination of water from a protonated alcohol. Both of these processes are shown in Figure 7.10. In the upper part of the figure, the electron pair of the π‐bond forms a new bond with the incoming proton, which leaves one of the carbon atoms of +

+

+

+

+

+

Figure 7.9  Stabilisation of a carbocation by conjugation.

H

H

H+

H

H

+ H

H

H

H

H H

H+

H H3C

O

H CH3

CH3

H3C

+

H

O

CH3

CH3

Figure 7.10  Formation of carbocations.

O

H

H3C + CH3 CH3

143

144

7  Chemical Reactivity

Figure 7.11  The Diels–Alder reaction.

the starting olefin as a carbocation. In the lower case, a lone pair of electrons on the oxygen atom forms a bond to the proton, leaving the oxygen atom carrying a positive charge. The electrons of the carbon–oxygen bond of the alcohol then leave and move onto the oxygen to form a new lone pair, thus leaving the positive charge on the carbon atom. Pericyclic (or electrocyclic) reactions involve p orbitals moving in a ring to  form new bond systems. These reactions are essentially synchronous in nature. An example is the Diels–Alder1 reaction in which a butadiene reacts with an ethylene to form a cyclohexene, as shown in Figure 7.11, again using both line drawings of bonds and also a representation of the orbitals involved. The Diels– Alder reaction is classified as a 4+2 reaction since one of the components, the diene, has four electrons that take part in the reaction while the other, the dienophile, has two. An example of a Diels–Alder reaction of use in perfumery is shown in Figure 7.12. The top scheme in the figure shows the reaction of methylpentadiene with acrolein (propenal) to form Ligustral® that provides green top notes in fragrances. Another useful 4+2 reaction is the ‘ene’ reaction shown here being used to form isoprenol from isobutylene (2‐methylpropene) and formaldehyde. In the Diels–Alder reaction, all of the bonds involved in the reacting species are double bonds, but the ‘ene’ reaction is different in that one of the reacting bonds is a C─H single bond. Isoprenol is an important feedstock for the preparation of terpenoid compounds, as we will see in Chapter 15. The third reaction shown in Figure 7.12 is a 2+2 reaction since only four electrons are involved. The formation of a cyclobutane from two ethylene molecules is an analogy of the Diels–Alder reaction; it builds a ring from acyclic precursors. This reaction is the reverse. In other words, a ring is broken with the formation of two double bonds. This reaction is the thermal cracking, or pyrolysis, of pinanol to give linalool, and it requires high temperatures. It is an important route to linalool and other terpenoid fragrance ingredients as will be seen in Chapter 15.

1  Many organic chemistry reactions are named after the chemists who discovered them. In this case, the reaction was discovered by two chemists working together, Otto Diels and Kurt Alder.

Review Questions

O

O Diels–Alder Ligustral

O

OH

‘ene’

H

Isoprenol OH

OH

Pinanol

Linalool

Figure 7.12  Some pericyclic reactions used to prepare fragrance ingredients.

Review Questions 1 Why should drums of perfume oil not be left exposed to the sun (for example, on a dockside in the tropics)? 2 Which of the following perfume ingredients is most likely to cause problems when used in a perfume in an alkaline product? O

O O

Methyl cinnamate

O O

Ethyl acetoacetate

O

O

Linalyl acetate

3 Why do perfumes and perfume ingredients disappear from a perfumer’s blotter when left to stand at room temperature? 4 In the Diels–Alder reactions shown in Figures 7.11 and 7.12, some heat is usually applied to cause the reactions to proceed and form the products. However, as the temperature is increased further, the balance between products and starting materials shifts back to favour the starting materials. Why?

145

147

8 Chemistry and Perfume 1: Acid/Base Reactions This chapter concerns those chemical changes brought about by contact with acids or alkalis, which are used in our industry or to which perfumes are subjected when in use. First, we will define what is meant by acids and bases and what is meant by pH. Then we will consider what is meant by the terms electrophiles and nucleophiles and the concept of hard and soft. Then, putting all of these concepts together, we will explore some elementary acid/base reactions in perfumery chemistry. We will see how we can use these reactions to make fragrance ingredients and how the presence of acids and bases in our customers’ products can destroy fragrance ingredients.

­Acids and Bases Three definitions of acids and bases are used, and each is named after the chemist who first formulated them. Bases were originally called alkalis, and the two terms are often used interchangeably. According to the Brønsted definition, an acid is a proton donor and a base is a proton acceptor, whereas in the Lewis definition, an acid is an acceptor of a pair of electrons and a base is a donor of a pair of electrons. The Franklin definition is more complex and states that, when dissolved in a self‐dissociating solvent, an acid generates the same cation as that generated by the solvent, while a base generates the same anion as that generated by the solvent. All three definitions are useful, each being more appropriate in various situations. In aqueous solutions, which cover most of the consumer goods into which perfume is added, the Brønsted definition is usually the easiest to visualise and also the most useful. Indeed, in aqueous solution, the Brønsted and Franklin definitions are the same, and Lewis acids and bases essentially become Brønsted acids and bases by reacting with water. (Chemists describe a substance as a Lewis acid when it behaves as an acid according to the Lewis definition. Similar terminology is used for the other types of acids and bases.) Figure 8.1 shows a simple Brønsted acid/base neutralisation. Hydrogen chloride is an acid by Brønsted’s definition since it can donate a proton by dissociating into H+ and Cl− ions. (We often write H+ for an aqueous acid, but, in reality, the free protons associate with water molecules to give H3O+ ions.) Similarly, Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

148

8  Chemistry and Perfume 1: Acid/Base Reactions HCI + Acid

NaOH Base

NaCl

+ H2O

Salt

Water

Figure 8.1  Brønsted acid/base neutralisation reaction.

sodium hydroxide, or to be more precise the hydroxide anion, is a base since it can accept a proton. In the neutralisation reaction, the acid and base react to give a salt. In this case, the salt is common table salt, sodium chloride. Neutralisation of sulfuric acid by calcium hydroxide would generate calcium sulfate. Figure 8.2 shows some reactions of Lewis acids. In the top of the figure, we see a simple neutralisation reaction. Boron trifluoride is a Lewis acid. Boron has five electrons. Two of these are found in the 1s orbital, and the other three are the valence electrons in the second shell. Each of these electrons is used to form a bond with one of the fluorine atoms, but this still leaves only six electrons in the second shell of the boron. The boron is therefore able to accept an electron pair to complete the second electron shell. The fluoride anion is able to donate a pair of electrons to the boron trifluoride to generate the BF4− anion, so fluoride acts as a Lewis base. The second equation in the figure shows what happens when a Lewis acid is added to water. Aluminium chloride is chosen as the example since it is important to our industry, being the active component of many antiperspirant formulations. Like boron, trivalent aluminium can accept a pair of electrons, so aluminium chloride is a Lewis acid. When added to water, it accepts electrons from an oxygen atom of a water molecule, which has the effect of releasing a proton. Thus, in general, Lewis acids become Brønsted acids when dissolved in water. In nonaqueous systems, Lewis acids have an important role to play in synthetic chemistry. An example is shown in the box in the figure. When an acid chloride comes in contact with a Lewis acid, the latter removes the chloride anion from the former to generate an acylium cation (that is, a cation where the positively charged carbon atom carries a double bond to an oxygen atom). The example shows CH3CO+ being generated from acetyl chloride by the action of BF3 + Na+F –

BF4– Na+

=

AlCl3 + H2O = H+ + HOAlCl3– O Cl

O +

+

+

O

AlCl3

+

+ AlCl4–

O

+ H+

Traseolide®

Figure 8.2  Reactions of Lewis acids.

 ­Acids and Base

In water H2O

H+

+

−OH

HCl

H+

+

Cl−

Acid

−OH

Base

Na+ +

NaOH

In ammonia

In sulphuric acid H2SO4 FH H2O

H3SO5+ + H3O+

+

NH4

+

+

NH2

NH4Cl

NH4+

+

Cl–

H2O

NH4+

+

–

NH3

H3SO5+ + HSO4– F–

Acid

HSO4

Base

NaNH2

Na+

+

–

Acid

OH

NH2

–

Acid Base

Figure 8.3  Franklin acids and bases.

aluminium chloride. Acylium ions are very reactive and, for example, will add to aromatic rings to produce, after loss of a proton, an aromatic ketone. The example in the figure shows the formation of the musk Traseolide® in a Friedel–Crafts reaction, which uses this Lewis acid chemistry. Figure 8.3 shows an example of the Franklin definition in action. In the top box, we see how water dissociates into H+ and HO− ions. When HCl dissolves in water, it generates H+ cations, the same cations as the solvent, and is therefore an acid. Similarly, NaOH, because it produces the same anions as the solvent, is a base. In sulfuric acid, things are rather different. Sulfuric acid dissociates into H3SO4+ and HSO4− ions. Thus, hydrogen fluoride is an acid in sulfuric acid solution, but water now becomes a base. In the same way, in liquid ammonia, ammonium chloride and water are acids, and sodamide, NaNH2, is a base. Strong and Weak Acids and bases can be classified as strong and weak, and it is important not to confuse these terms with concentrated and dilute. A strong acid or base is one that is extensively dissociated into ions in solution. A weak acid or base is one in which the ions are in equilibrium with the neutral species. Examples are shown in Figure 8.4. Hydrogen chloride dissolves in water to give hydrochloric acid. In hydrochloric acid, the HCl is extensively dissociated into H+ cations and Cl− anions with very few neutral HCl molecules to be found. Its substantial dissociation makes it a strong acid. Acetic acid, however, produces a mixture of H+ cations, CH3CO2− anions, and neutral CH3CO2H molecules. Thus, it is a weak acid. As it is diluted further, more acetic molecules become dissociated because of entropy, and it becomes more difficult for an H+ cation to find a CH3CO2− anion with which to partner and reform into a neutral molecule. In other words, as a weak acid is diluted, it becomes stronger. Similarly, sodium hydroxide is extensively dissociated in aqueous solution and is therefore a strong base. Aniline, C6H5NH2,

149

150

8  Chemistry and Perfume 1: Acid/Base Reactions Strong acid

HCl

H+

+

H+

+

Na+

+

Cl– O

O Weak acid

OH

Strong base

NaOH

NH3+

NH2

Weak base

Strong

Concentrated

Weak

Dilute

O– OH– +

OH–

Figure 8.4  Strong and weak acids and bases.

is a Brønsted base because it will accept a proton from water to form the anilinium cation and thus generate free hydroxyl anions in solution. However, as with acetic acid, this reaction does not go to completion, and there will always be plenty of neutral aniline molecules around. Therefore, it is a weak base. pH Water dissociates, to some extent, into H+ and HO− ions. In pure water the numbers of each are equal, and so the solution is regarded as neither acid nor basic. This dissociation is shown in the line at the top of Figure 8.5. The lower part of the figure represents three beakers of water. The centre beaker contains pure water where one of the molecules has dissociated into H+ and HO− ions. In the beaker on the left, one molecule of hydrogen chloride has been added. This substance is a H+

H2O

+

+HCl

H2 O H O H2O 2 O H H2O 2 H2O H2O + H2O H2O – Cl H H2O H2O H 2O HO HO H 2O 2 H 2O 2 – OH H2O + H 2O H H 2O

–OH

+NaOH

H 2O H O H2O 2 O H H 2O 2 H2O H 2O + H 2 O H 2 O H H2O H2O H2O H2O H O H 2O 2 H 2O – OH H 2O H2O H 2O

Figure 8.5  Acids and bases in water.

H 2O H O H2O 2 O H H2O 2 H2O H2O + H2O H2O – OH H H2O H2O H 2O HO HO H2O 2 H2O 2 – OH H2O H2O Na+ H2O

 ­Acids and Base

strong acid and so has dissociated completely into H+ and Cl−. The water still dissociates as in the centre beaker, but because of the dissolved hydrogen chloride, it now contains one more H+ ion than HO−. The water in the left beaker is therefore acidic, hydrochloric acid to be precise. In the beaker on the right, the water dissociates as in the other two, but some sodium hydroxide has been added, which is a strong base that dissociates completely into Na+ and HO− ions. Thus, an excess of HO− ions exist and so the water is basic or alkaline. As more sodium hydroxide is added, the H+ ions are neutralised by the hydroxide, and so the concentration of H+ ions will fall as the self‐dissociation of water is altered. We can now see that the concentration of H+ ions is higher in acidic solutions and lower in basic solutions. The shorthand for hydrogen ion concentration is [H], and we can now see that [H] is a measure of the degree of acidity or basicity of an aqueous solution. The absolute measure of [H] lies on a very large scale, so a more convenient scale is given by defining a measurement, pH, such that pH = −log10 [H]. On this scale, pure water has a value of 7, concentrated acid has a value of 1, and concentrated sodium hydroxide (base) has a value of 14. This value gives a convenient measure by which we can compare the acidity or basicity of solutions, including the consumer goods into which perfumes are added. Figure 8.6 shows this scale with concentrated sulfuric acid at one end, sodium hydroxide solution at the other, and the neutral point of 7 half way between. Also shown on the scale are a number of consumer goods and two other values of particular interest to our industry, which fall at pH 5.5. This value is the pH of both distilled water and human skin. Distilled water is acidic because it dissolves carbon dioxide from the air and becomes carbonic acid, a weak acid that lowers the pH and thus acidifies the otherwise pure water. Skin is maintained at pH 5.5 by the balance of acids and other skin chemicals. As can be seen from the figure, perfumers are asked to create fragrances for products with pH values ranging from one end of the scale to the other. This puts Sulfuric acid

Sodium hydroxide

Neutral

Neutral ‘soap’

Antiperspirant Skin

1

2

3

4

5

6

Fabric conditioner Acid lavatory cleaner

Liquid bleach

Soap

7

8

9

10

11

Laundry powder Distilled water

Figure 8.6  pH in consumer products.

12

13

14

Dishwashing powder

151

152

8  Chemistry and Perfume 1: Acid/Base Reactions % Stable ingredients

pH

Typical product

1

Acid lavatory cleaner

3

Fabric conditioner

65

7

Perfume

100

10 13

Laundry powder Dishwashing powder

45 5

25

Figure 8.7  Stability of perfume ingredients at various pH values.

limitations on the choice of ingredients for fragrances for functional products. At the ends of the scale, that is, with products such as acid cleaners or alkaline dishwashing powders, these constraints seriously restrict the creative palette. Figure 8.7 shows the average pH level of some typical consumer goods and the consequent effects of this pH on the percentage of perfume ingredients that can be used in them. In neutral conditions, such as in a fine fragrance, all ingredients are stable enough to be used. However, as the pH moves away from neutral, the number of stable ingredients drops so that only 65% of the palette can be used by the perfumer creating a fragrance for an antiperspirant, 45% in a laundry powder, 25% in an acid lavatory cleaner, and a mere 5% in a dishwashing powder. Some of the mechanisms by which acids and bases can degrade fragrance ingredients will be discussed below. The more rapid loss of stability on the alkaline side of the scale is partly due to the fact that laundry powders, bleach, and machine dishwashing products also contain oxidising agents, which can also degrade fragrance ingredients by other mechanisms as we will see in the next chapter. As we discovered in Chapter 4, soaps – that is, the salts of fatty acids – are alkaline in nature and will have a pH somewhere in the region of 9–10. This high pH can damage the natural fats on the skin and lead to skin dryness. Therefore, newer products have been developed that use surfactants that act like soaps in neutral conditions or even when the pH of the product is made just slightly acidic (pH 5.5) to mimic the acidity of the skin. If a weak acid, such as acetic acid, or a weak base, such as ammonia, is dissolved together with a salt of the same acid or base (as appropriate) in aqueous solution, a system is set up that will resist changes in pH. For example, a mixture of sodium acetate and acetic acid will be slightly acidic because of the excess of acetic acid. If a stronger acid is added, some of the acetate ions will be protonated to form more acetic acid, and therefore the effect of the stronger acid is counteracted. Similarly, if a strong base is added, some of the acetic acid will be neutralised, and once again, the effect of the added material is resisted. Such a solution is called a buffer solution. If a cosmetic or household product needs to be maintained at constant pH, a buffer can be used to achieve this.

­Electrophiles and Nucleophiles As their names suggest, an electrophile is a species that seeks electron‐rich centres, whereas, a nucleophile is a species that seeks electron‐poor centres. It is easy to see that acids are a subset of electrophiles and bases are a subset of

  ­Electrophiles and Nucleophile

­ ucleophiles. For example, the H+ ion of an acid will seek out and react with n species that are electron rich. If the electron‐rich centre is a hydroxyl ion, then the result will be neutralisation of the acid, but if the electron‐rich species is the oxygen atom of a ketone, then a different train of events will be set in motion. Similarly, a hydroxide ion could react with the electron‐poor proton of an acid, but it could also react with the electron‐deficient carbonyl carbon of an ester. The acidic/electrophilic and basic/nucleophilic behaviour of species is always in balance. The position of this balance can be affected by the nature of the species it is reacting with and by other factors such as the solvent. Electrophiles and nucleophiles are classified as hard (not polarisable) and soft (polarisable). If the charge clouds around them are easily distorted, they are said to be polarisable since the charge distribution will change as the orbitals are distorted. Soft electrophiles prefer to react with soft nucleophiles and hard with hard. Thus, this phenomenon also helps determine whether a species acts as a base/nucleophile or as an acid/electrophile. As was discussed in previous chapters, the electrons in a bond between two dissimilar atoms are not shared equally between these atoms. The atom with the higher number of protons in its nucleus will tend to pull electrons away from the other atom and therefore build up a negative charge over itself, leaving a partial positive charge over the other atom. For example, oxygen or chlorine atoms will draw electrons away from a carbon atom to which they are attached, making the oxygen or chlorine end of the bond slightly negatively charged and the carbon end slightly positively charged. This effect can extend along to the next bond in a chain also. It is known as the inductive effect. In Figure 8.8, we see how the hydroxide ion can act as either a base or a nucleophile. At the top of the figure is a drawing of an ester function showing how the δ– O R

δ+ R O

H H δ+ δ+ Hydroxide as base

Hydroxide as nucleophile

O R H

O O

R

R

H

O H H HO–

HO–

O R

– H

R

O– OH O

R

R H H

Figure 8.8  Hydroxide as base or nucleophile.

O

R

153

154

8  Chemistry and Perfume 1: Acid/Base Reactions

carbon–oxygen double bond is polarised with a partial negative charge on the oxygen atom and a partial positive charge on the carbon. The positive charge on the carbon has an effect on the hydrogen atoms on the adjacent carbon and makes them slightly positive also, as shown in the figure. If the hydroxide ion approaches the α‐hydrogens, it can act as a base by removing one of these to form the carbanion shown in the left of the figure. Alternatively, as shown on the right of Figure 8.8, the hydroxide ion can approach the carbonyl carbon atom and add to it, thus acting as a nucleophile. The first course is likely to lead to condensation reactions, whereas the second will probably result in ester hydrolysis.

­Esterification and Ester Hydrolysis In Figure 8.9, we see the complete mechanism for ester hydrolysis. This process starts with the addition of a hydroxide anion to an ester as was shown also in Figure 8.8. This course of events generates the intermediate in which the carbon atom carries a hydroxy group, an alkoxy group, and a negatively charged oxygen atom. If the negative charge flows back to reform a carbon–oxygen double bond, it can do so by displacing either a hydroxide anion (thus returning to the starting materials) or an alkoxide anion (the salt of an alcohol, formed by removing a proton from it) thus moving on to the acid and alkoxide pair as shown. So far all of these reactions are reversible, as shown in the figure. However, once the alkoxide and acid pair are produced, the alkoxide, which is a relatively strong base, can remove the acidic hydrogen atom of the acid and thus generate an alcohol and the salt of an acid. In practical terms, the equilibrium for this reaction is so far to one side that it is essentially irreversible, so assuming that sufficient hydroxide is present, the equilibria to the left of the figure will eventually be displaced and all the materials will end up as alcohol and the salt of the acid, as shown to the extreme right. Many consumer goods are alkaline in nature, and esters will be hydrolysed in them, thus reducing the usefulness of ester‐based fragrance ingredients in such applications. As can be seen in Figures 8.6 and 8.7, soaps, laundry powders, and dishwashing cleaners all fall into this category. Esterification and ester hydrolysis can also be catalysed by acids, as shown in Figure 8.10. One important difference is that the acid is genuinely catalytic (i.e. recovered), in that it has an H+ ion at both ends of the reaction scheme. In base‐ catalysed hydrolysis, the catalyst is consumed by the acid produced. The neutralisation of bases by carboxylic acids means that base‐catalysed esterification is not favourable. Therefore, when we want to form an ester, it is usual to employ O R +

O

R

R

HO–

Figure 8.9  Ester hydrolysis.

O

O

O

O– OH R

R

OH

R

O–

+ R

O–

+ R

OH

  ­The Aldol Reaction and Aldol Condensatio H+ OH

O R

O H

O

R

R

+

H

O

O

O

OH R

R

H

H

O

H

O

R

R H+

+ R +

HO

OH H+

Figure 8.10  Acid‐catalysed esterification/ester hydrolysis.

OH +

Wanted

H+ O

O

Unwanted

O

HO

Soap O

O

–H2O

NaOH

OH +

O– Na+

O

Figure 8.11  Formation and loss of PTBCHA.

an acid catalyst (usually driving the reaction to completion by removing the water as it is formed and thus preventing the reverse reaction), and when we want to hydrolyse an ester, it is more efficient to use an excess of base. Figure 8.11 shows the (wanted) formation of para‐tertiary‐butylcyclohexyl acetate (PTBCHA) using acid catalysis in the factory and its (unwanted) hydrolysis in soap. PTBCHA is a useful fragrance ingredient with a sweet, woody floral scent. It is important in soap fragrances despite the fact that some of it will be lost by hydrolysis in the product.

­The Aldol Reaction and Aldol Condensation In Figure 8.8 and associated text, we saw how the hydrogen atoms next to carbonyl groups are acidic because of the fact that the oxygen atom pulls electrons away from them. Thus, if a carbonyl compound is treated with a base, an enolate anion can be formed, which will exist in two tautomeric forms as shown in Figure 8.12. The negative charge is a nucleophile and is softer when it reacts via carbon than when through oxygen. The soft carbon‐centred nucleophile gives rise to a set of reactions that are very important in fragrance chemistry. The parent of this family of reactions is known as the aldol reaction. The course of the aldol reaction and subsequent aldol condensation is shown in Figure 8.13. The reaction involves two carbonyl‐containing molecules. They could be identical molecules or, as in this example, different. The reaction starts when a proton is removed from the

155

156

8  Chemistry and Perfume 1: Acid/Base Reactions O R

R

H H

HO–

O R

–

O– R

R

H

R H

Figure 8.12  Tautomeric forms of an enolate anion. O R

+

O

R Base O

R

R

–

O

R

O

H

R

Aldol reaction

OH

H+

Base O Aldol condensation

R

O R

R –

R

OH

Figure 8.13  The aldol reaction and aldol condensation.

α‐­position of one of them. The carbanion thus formed then behaves as a nucleophile and approaches the positively polarised carbonyl carbon atom of the other molecule. A bond is formed between these two carbon atoms with the simultaneous formation of an alkoxide anion (as the anion formed by removal of a proton  from an alcohol is called). Protonation of this alkoxide anion then gives a β‐hydroxy carbonyl compound, in this case a β‐hydroxy ketone. Such compounds are known generically as aldols, and this reaction is known as the aldol reaction. In the presence of acid or base, aldols can easily eliminate water. In the  figure, another α‐proton is removed by base, and then a hydroxyl anion is eliminated, thus completing the formal elimination of water. This sequence gives an α,β‐unsaturated ketone. This overall reaction, i.e. from the two starting

  ­The Aldol Reaction and Aldol Condensatio O

O +

R

R Aldol reaction

Tautomerisation

O R

H R

R

Aldol condensation

H

O

O R

R

R

OH

O

H+

H+

Figure 8.14  The acid‐catalysed aldol reaction and aldol condensation.

compounds to the α,β‐unsaturated ketone, is known as the aldol condensation. A condensation reaction is one in which two molecules combine together with the elimination of a small molecule (water in the case of the aldol condensation). As can be seen from Figure 8.14, the aldol reaction and aldol condensation can also be catalysed by acids. Figure 8.15 shows an example of wanted and unwanted aldol condensations. At the top of the figure, we see the formation of amylcinnamic aldehyde (ACA) from benzaldehyde and heptanal. These compounds are both readily available and inexpensive feedstocks, and the aldol condensation is an easy and i­ nexpensive

Wanted

O Benzaldehyde

O + O

Heptanal

Unwanted

ACA

O + HO

Amylcinnamic aldehyde

Soap components

O Vanillin

Figure 8.15  Wanted and unwanted aldol condensations.

Brown colour

157

158

8  Chemistry and Perfume 1: Acid/Base Reactions

reaction to carry out on large scale. ACA is a useful fragrance ingredient with a fatty, jasmine odour, and this method of synthesis means that it can be produced for a low price and therefore can be used in fragrance applications where cost is an important factor, for example, in perfumes for laundry powder and soap. Structure of Ultravanil®

O OH

In the lower half of the figure, we see that vanillin, which is an aldehyde, can undergo aldol condensations with any other carbonyl compounds present in the perfume or with the actual product to which the perfume is added. One aldol condensation adds one more double bond to the conjugated π‐electron system of vanillin. If scope for a further aldol condensation in the product exists, then a further double bond will be added, and so on. As we saw in Chapter 6, the more double bonds there are in conjugation in a molecule, the longer will be the wavelength of the maximum absorption of UV/visible light. So, repeated aldol ­condensations with vanillin are one mechanism by which an unwanted brown colour is added to the product. This effect with vanillin is easily demonstrated by adding a drop of a vanillin solution in ethanol to a bar of white soap. Within an hour, a brown colour will have formed and will be spreading from the original site into the surrounding areas of the soap. This notorious property of vanillin seriously limits its use in soap. However, the soft, sweet odour of vanillin is very appealing to the consumer, so the perfumer seeks alternative molecules that are not prone to such reactions. One such molecule is Ultravanil®, which has a vanilla odour but a methyl group rather than the aldehyde of vanillin and so cannot undergo aldol condensations.

­Acetals and Ketals Aldehydes are also susceptible to oxidation, as we will see in the next chapter, yet they are very useful fragrance ingredients because many of them have strong and pleasant odours. Fragrance chemists have therefore sought ways of protecting them or replacing them by other, less reactive but similarly scented materials. One way of protecting aldehydes is to form acetals of them. This process is accomplished by treating the aldehyde with an alcohol in acidic conditions. The reaction sequence is shown in Figure 8.16. It is seen that two molecules of alcohol are needed for every one of aldehyde. The hydrolysis of acetals back to the aldehyde follows exactly the reverse of the sequence in the figure. We can see that the first step of this hydrolysis requires acid and cannot be affected by base. Acetals are therefore stable in basic conditions but can be hydrolysed in acid. So, an acetal will be stable in soap that is basic, but if transferred to the skin during washing, it can then release the aldehyde because of the natural acidity of the

 ­Acetals and Ketal H+

Hemi-acetal

OH

O O+ HO

R

R

R

OH

–H+

R O

H H+

Aldehyde

O

R

O

–H+ R

O

R O+ H

Acetal

–H2O

H

O+

H R

R

O HO

R

Figure 8.16  Formation of acetals.

skin. Acetals sometimes have odours that are reminiscent of the parent aldehyde, but often they are much greener in character. Many acetals are important fragrance ingredients in their own right, especially for soap and laundry products, because of their good stability in base. An analogous class of products, which are called ketals, can be made from ketones. Acetal formation is not always desirable. Figure 8.17 shows examples of desired and undesired acetal formation reactions. Figure 8.17a shows the formation of Karanal® from Ligustral® and the appropriate diol, in this case 2‐methyl‐2‐ (1′ethylprop‐1‐yl)propan‐1,3‐diol. As can be seen in the figure, the use of diols, rather than simple alcohols, leads to cyclic acetals. Cyclic acetals are particularly stable. This stability exists because, if one half of the acetal function is hydrolysed, the aldehyde and alcohol fragments are still tied together by the other carbon–oxygen bond and the broken bond therefore has a good chance of re‐forming before the second is broken. In the case of Karanal, the product does not have an odour anything like that of the starting aldehyde. Whereas Ligustral has a strong leafy green note, Karanal has an intense amber character. In Figure 8.17b, we see two less desirable acetal formation reactions. Soap is made by hydrolysing fats to produce fatty acid salts and the trihydric alcohol glycerol. Glycerol can form five‐ and six‐membered cyclic acetals with aldehydes. In either case, one free hydroxyl group will still exist, and therefore these acetals are generally too polar to possess odours (see Chapter 13, for an explanation). Although soaps are basic, acetals and hemiacetals do form from glycerol and aldehydes in soaps that are glycerol rich. This development therefore represents a means by which perfume aldehydes are effectively lost in such soaps. Aldehydes can form cyclic trimers with a ring system known as a 1,3,5‐trioxane. This ring system essentially contains three acetal functions. This problem is of more concern with pure aldehyde ingredients than it is in perfume since it requires a high concentration of aldehyde. Very often, one can see a ring of crystalline material around the rim of a bottle containing an aldehyde or even a deposit of colourless crystals at the bottom of the container. It is likely that such

159

160

8  Chemistry and Perfume 1: Acid/Base Reactions

O

O

O

+

OH

Ligustral®

OH

Karanal®

(a) OH R

O

OH

O

+ OH

O

R

OH

Glycerol from soap

Fragrant aldehyde

Odourless

R 3

R

O

O R

Fragrant aldehyde

O O

R

Odourless

(b) Figure 8.17  (a) Wanted and (b) unwanted acetal formation.

crystalline deposits are the trimer of the aldehyde. Because of their higher molecular weight, these trimers are almost always odourless, and thus valuable odour contribution is lost from the aldehyde.

­Schiff’s Bases and Enamines Another way of protecting aldehydes in fragrances is to convert them into Schiff ’s bases or enamines as described in Chapter 3. Figure 8.18 shows the mechanism of how they are formed. The figure shows the acid‐catalysed reactions, but base‐ catalysed mechanisms also exist. With primary amines, the Schiff ’s base is formed initially, but, if there is a hydrogen atom on the carbon next to the carbonyl carbon, then the Schiff ’s base can be isomerised to the corresponding enamine, and the two tautomers will exist in equilibrium. The reactions are reversible, and so the Schiff ’s base or enamine can be used to provide a delayed release of the aldehyde by slow hydrolysis of the adduct. Most of the primary and secondary amines have unpleasant odours, usually fishy and ammonia like. But methyl anthranilate (which is used as the example in Figure 8.18) is a primary amine that occurs in a number of essential oils such as broom (genet) and orange flower (neroli). It has a strong, sweet smell and so is more amenable to perfume use than most amines. It has therefore become the standard amine component of Schiff ’s bases in perfumery, to such an extent that when perfumers talk about

 ­Nitrile H+

H+

O R

OH

H

NH2

R

N

O

–H2O

O

O

O O

O

H+

H

R

N

R

N+ O

R

N H

H

O

O

R

N

O

Base

O

O

Schiff′s base

Enamine

Figure 8.18  Mechanism of formation of Schiff’s bases and enamines.

Schiff ’s bases, they will almost invariably be referring to the Schiff ’s bases of methyl anthranilate. Two problems arise with methyl anthranilate Schiff ’s bases. Firstly, when hydrolysed, the amine has a strong characteristic smell that will then always be a part of the resulting fragrance and hence impose a creative constraint on the perfumer. Secondly, methyl anthranilate Schiff ’s bases are coloured, ranging from pale yellow to deep red, depending on the overall degree of extended conjugation in the molecule. These colours are not always acceptable in fragrances.

­Nitriles Nitriles often have odours similar to the corresponding aldehydes and are more stable to acid, base, and oxidation conditions. They have therefore become an important group of perfumery ingredients as already mentioned in Chapter 3. The simplest method of preparation of nitriles is as shown in Figure 8.19. Treatment of the aldehyde with hydroxylamine produces the oxime, which is then reacted with acetic anhydride to produce the oxime acetate. Elimination of O R

O

NH2OH R

N

OH

O

O R

N

O

O –HOCOCH3

Figure 8.19  Preparation of nitriles from aldehydes.

R

N

161

162

8  Chemistry and Perfume 1: Acid/Base Reactions

acetic acid from the oxime acetate can be achieved by heating in the presence of either an acidic or basic catalyst (acetic acid is the most convenient), and this produces the corresponding nitrile.

­Alcohol Dehydration In addition to esterification, ester hydrolysis, and other reactions described above, we need to consider two classes of acid‐catalysed reactions here. These reactions are dehydration of alcohols and addition to olefins. Let us first look at alcohol dehydration, the mechanism of which is shown in Figure 8.20. The alcoholic oxygen atom is protonated by the acid, and then water is lost leaving a carbocation. Loss of a proton from the neighbouring carbon atom then regenerates the acid catalyst and forms an olefin. The ease with which this reaction will occur depends on the stability of the intermediate carbocation. Because carbon hydrogen bonds tend to be polarised with the hydrogen more positive and the carbon more negative, carbon substituents are able to feed electrons in towards adjacent carbon atoms in a chain or ring. So, as shown in Figure 8.20, a carbocation with three other carbons attached to it receives some negative charge from each of these carbon atoms, which helps to stabilise the positive charge on it. A carbocation with only two other carbons attached is therefore less able to stabilise its charge than one with three substituents, and a primary carbocation is even less stable. Carbocations are also stabilised by conjugation to double bonds as shown in Figure 8.20. If a double bond is next to the carbocation, then the π‐electrons of the double bond will spread across to the cation and generate a delocalised orbital, which can be represented by the two canonical structures shown in the third line of the figure. If the carbocation is adjacent to a benzene ring, then the degree of delocalisation is even greater, and the cation will be even more stable. H+

H

O H

H

+

+

O+ H

H

+

H

+

+

+

+

+

Figure 8.20  Dehydration of alcohols.

H

+

is preferred to

+

O H

H

H

is preferred to

+

O

+

 ­Acid‐Catalysed Addition to Olefin

If only one carbon is attached to the benzene ring, then the carbocation cannot eliminate a proton as there are none. In this case, a different set of reactions will ensue. Overall then, by looking at the carbocation that would be formed by removing the hydroxyl group, we can make an assessment of how easily an alcohol can be dehydrated. This differentiation is important, since many perfumery ingredients are alcohols, whereas the olefins derived from them have much weaker odours. Dehydration of alcohols therefore represents a significant pathway by which perfume performance can be lost in those consumer goods where the pH is low.

­Acid‐Catalysed Addition to Olefins Protons can be added to olefins to generate carbocations. As with alcohol elimination, the stability of the resultant carbocation determines how readily the reaction will occur. It also determines which end of the double bond will be protonated. In general, the less substituted end of the double bond will be protonated in order to generate the most heavily substituted carbocation. Once generated, the carbocation can react with any nucleophiles present in the medium. The most common such nucleophile in functional products is usually water, so we see the reverse of alcohol dehydration, that is, alcohol formation. In fine fragrances, ethanol is likely to be added since that is the solvent used. If chloride ions are present (e.g. in products containing hypochlorite bleach), then these could also add to the carbocation. The general mechanism of acid‐catalysed addition to olefins is shown in Figure 8.21. Olefins are mostly lower in odour impact than are oxygenated molecules such as aldehydes and esters. We might therefore suspect that acid‐catalysed addition to olefins might be less of a problem than alcohol dehydration, but this is not necessarily the case. Many fragrance ingredients have double bonds as well as X

+

H+

X–

OH

9-Decenol

OH OH

OH

Limonene

Figure 8.21  Acid‐catalysed addition to olefins.

α-Terpineol

163

164

8  Chemistry and Perfume 1: Acid/Base Reactions H+ X X– H+

X X–

Figure 8.22  Acid‐catalysed addition to cyclopropanes.

oxygenated functions. Addition of a second polar function will reduce their volatility and increase their water solubility and therefore lead to loss of odour. Alternatively, an olefin with a low odour value could be converted to an oxygenated molecule with a higher odour impact, which will result in a distortion of the perfume from the perfumer’s original concept. Examples of both are shown in Figure 8.21. Addition of water to 9‐decenol, a very potent odourant, will result in formation of the essentially odourless diol. The impact of the former will therefore be lost from the fragrance. Alternatively, addition of water to limonene will generate the much more highly odoured α‐terpineol, thus distorting the fragrance composition. Cyclopropanes are currently becoming popular as fragrance ingredients. They are slightly more stable than olefins, but it must always be borne in mind that they do have reactivity rather like that of olefins and will also undergo acid‐­ catalysed additions, opening the ring in the process. Two simple examples are shown in Figure 8.22.

­The Michael Reaction Earlier in this chapter, we discussed how atoms are polarised in carbonyl compounds. This polarisation can be extended along a carbon chain if double bonds are located in conjugation with the carbonyl group. Figure 8.23 shows the pattern of such polarisation in a simple α,β‐unsaturated ketone. To help explain why the carbon two atoms away from the carbonyl carbon becomes positive, a canonical structure is also shown in which the charges have been separated completely. From the figure, it is clear that nucleophiles can attack the compound at two positions. Reaction path A shows the normal nucleophilic addition to the carbonyl group, and path B shows the addition to the more remote carbon atom. In each case, the anion produced is quenched by a proton. In the case of path A, the nucleophile and proton add to atoms next to each other, and so this reaction path is referred to as 1,2‐addition. Similarly, the addition in path B is named 1,4‐addition. The balance between the two pathways depends on the nature of the two reagents and the reaction conditions. In general, softer nucleophiles (such as carbanions) undergo 1,4‐addition. This reaction is important in synthetic ­

  ­The Grignard Reactio δ+

δ+ δ–

Oδ–

+

O–

X– O

O X–

Path B

Path A

X

X

X

O–

O–

OH Tautomerisation X O

OH X 1,2-Addition

1,4-Addition

Figure 8.23  Nucleophilic addition to α,β‐unsaturated carbonyls.

c­ hemistry (see Chapter 15), but it is also significant in toxicological terms, as we will see in Chapter 12.

­The Grignard Reaction So far in this book, we have only considered bonds between carbon and non‐ metallic elements such as oxygen, nitrogen, and sulfur. However, carbon also forms bonds with some metals, and the resulting compounds are very useful in organic synthesis. The best known example is the Grignard reaction in which carbon is bonded to magnesium. Grignard reagents are formed when a halogenated organic compound is treated with magnesium metal. The overall effect is essentially the insertion of a magnesium atom into the carbon–halogen bond as shown in Figure 8.24, using methyl bromide as an example. The reaction is carried out in an ethereal solvent (that is, one containing an ether function, such as diethyl ether or tetrahydrofuran), and the magnesium atom of the Grignard reagent becomes coordinated to two molecules of the solvent. Br

Mg

Mg Br

Figure 8.24  Formation of a Grignard reagent.

165

166

8  Chemistry and Perfume 1: Acid/Base Reactions

The carbon–magnesium bond is highly polarised with the magnesium carrying the positive partial charge and carbon the negative. Grignard reagents, like most organometallic compounds, are therefore strongly nucleophilic in character and not stable to water. Methyl magnesium bromide, for instance, if brought into contact with water will react violently to produce methane and magnesium hydroxy bromide. When added to a carbonyl compound, the Grignard reagent will add to the carbonyl bond to generate an alkoxide. Water is usually added at the end of a Grignard reaction, and the alkoxide is therefore hydrolysed to the corresponding alcohol. If the carbonyl compound is an ester or acid, then two molecules of Grignard reagent will add to form a tertiary alcohol. If an α,β‐unsaturated carbonyl compound is treated with a Grignard reagent, then either 1,2‐ or 1,4‐addition can take place. The more usual is 1,2‐addition, but addition of a copper catalyst will reverse the selectivity and lead to a predominance of 1,4‐addition. All of these reactions are shown in Figure 8.25. The hydrogen on the end of a terminal acetylenic group is more acidic than normal for one attached directly to carbon. It is therefore easy to form Grignard reagents from acetylenes. Because of their acidity, it is even possible to form sodium derivatives of acetylene, for instance, by treating an acetylene with sodamide in liquid ammonia. This reaction is important for the preparation of fragrance ingredients, especially in the terpenoid series, as we will see in Chapter 15. An important example, the addition of acetylene itself to methylheptenone, is shown in Figure 8.26.

R′ R

O

(1) R′MgBr (2) H2O

R

R R

O

O R

OH

R

(1) R′MgBr (2) H2O

R

OH

R′

R O

R′

(1) R′MgBr (2) H2O

R

R′

OH

R O (1) R′MgBr (2) H2O

R

O R

OH

R′ O

R′ (1) R′MgBr/Cu cat. (2) H2O

R

Figure 8.25  Grignard reactions of carbonyl compounds.

  ­The Friedel–Crafts Reactio

NaNH2

–

O

Na+

NH3

OH

Figure 8.26  Acetylide anion as a nucleophile.

­The Friedel–Crafts Reaction The Friedel–Crafts reaction is one that is ‘catalysed’ by a Lewis acid. We talk about catalysts in this reaction, but in one of the two basic types of Friedel– Crafts reactions, the Lewis acid is a stoichiometric reagent rather than a catalyst, since it cannot be recovered or reused at the end of the reaction and, indeed, in many instances, an excess of the Lewis acid must be used. In this first reaction, the Friedel–Crafts alkylation, a Lewis acid is used to remove a halide anion from an alkyl halide, thus generating a carbocation. These carbocations are electrophiles and will react with double bonds and aromatic rings. This reaction produces a new carbocation from which a proton is lost to complete the reaction. Since the starting organic halide is an alkyl halide, the net result is to add an alkyl group to the unsaturated component. In these reactions, the Lewis acid is a catalyst as it survives the reaction unchanged. If the organic halide is an acyl halide, in other words, an acid chloride or bromide, then the overall process is the formation of a ketone, and the reaction is called a Friedel–Crafts acylation. The product ketone has a greater affinity for the Lewis acid than does the starting acid chloride, and therefore the Lewis acid remains complexed to the product and will not induce any further reaction. This outcome is why the acylation reaction requires at least one molecule of Lewis acid per molecule of product. The complex of the Lewis acid with the product is broken up by quenching it with water at the end of the reaction, thus destroying the Lewis acid in the process. Aluminium chloride is the most common Lewis acid in Friedel–Crafts reactions, but boron trifluoride, tin tetrachloride, iron trichloride, and titanium tetrachloride are all used in different cases. The mechanisms of the alkylation and acylation reactions are shown in Figures 8.27 and 8.28, respectively. In both cases, the net overall effect is the replacement, or substitution, of a hydrogen atom by an alkyl or acyl radical. For this reason, these, and similar reactions, are known as substitution reactions. Friedel–Crafts acylation reactions can also be catalysed by Brønsted acids since AlCl4– R

δ+ Cl R

Cl AlCl3

H

δ– AlCl3 R

R + + HCl + AlCl3

Figure 8.27  The Friedel–Crafts alkylation reaction.

167

168

8  Chemistry and Perfume 1: Acid/Base Reactions AlCl4– δ+ Cl

Cl O

AlCl3

H

δ– AlCl3

+

O O

H2O O

O

AlCl3

+ HCl

Figure 8.28  The Friedel–Crafts acylation reaction.

a proton can add to the carbonyl group to generate a carbocation analogous to that generated by the action of a Lewis acid on an acid chloride. An example of a Friedel–Crafts acylation by a Brønsted acid will be shown in Figure 8.31.

­Electrophilic Substitutions in Aromatic Molecules Friedel–Crafts reactions with aromatic compounds are important in perfumery, and this introduces the topic of electrophilic substitution into aromatic rings. Benzene rings are electron rich and react readily with strong electrophiles such as Friedel–Crafts reagents. If the ring already carries some substituents, will the substitution occur randomly with respect to the substituents already present, or will there be a repeatable and predictable pattern? The answer is that the substituents already present in the ring will exert an electronic effect on the distribution of the electrons in the ring and thereby direct the incoming positively charged species to one or more specific locations. Substituents that donate electrons to the ring will have the opposite directing effect to those that withdraw electrons from the ring. First let us look at the effects of electron‐donating groups such as alkyl chains, ether groups, and halogens. Figure 8.29 shows how an ether group will push electrons into the ring generating centres of higher electron density in the ring and leaving the ether group with a partial positive charge. By drawing all of the O+

O

O+

–

O+ –

–

Figure 8.29  Directive effects of electron‐donating groups in aromatic rings.

  ­Electrophilic Substitutions in Aromatic Molecule

­ ossible canonical forms, we can see that the positions ortho‐ and para‐ to the p ether group carry more electron density than in an unsubstituted benzene, and therefore anisole (methoxybenzene) is more reactive than benzene towards electrophiles, and the attack will take place at the carbons ortho‐ or para‐ to the methoxy group. Exactly the same reactivity applies to alkylated or halogenated aromatics. In exactly the same way, electron‐withdrawing groups, such as aldehydes, ketones, esters, carboxylic acids, sulfonic acids, and nitro groups, will pull electrons out of benzene rings as shown in Figure 8.30. This fact makes the materials much less likely to undergo electrophilic substitution than an unsubstituted aromatic. Under forcing conditions, substitution may occur, but it will take place at those positions that are least deactivated – in other words at the meta‐positions. Thus, while electron‐donating groups are ortho‐/para‐directing, electron‐­ withdrawing groups are meta‐directing. Figure 8.31 shows an example of this directing effect. It shows a Brønsted acid‐ catalysed addition of acetic acid to anisole. The reaction takes place predomiO–

O

O–

O–

+

+ +

Figure 8.30  Directive effects of electron‐withdrawing groups in aromatic rings.

O

H+

O

H

+ O

OH

O H

H O

H H

O+ O

O

O

O

–H2O

O O

Figure 8.31  Friedel–Crafts acetylation of anisole.

H H

169

170

8  Chemistry and Perfume 1: Acid/Base Reactions

nantly at the para‐position because of a steric effect that hinders the reaction at the ortho‐positions simply because of the steric bulk, or size, of the methoxy group and the reagents.

Review Questions 1 Which of the fragrance ingredients in Figure 8.32 would be most stable in a laundry powder? 2 Place the alcohols shown in Figure 8.33 in order of stability in acidic products. 3 Your customer complains that, when your perfume is added to his acid lavatory cleaner, a smell of vinegar develops in the product. What type of fragrance ingredient might you suspect to be responsible? 4 Your customer loves the soft, gentle smell of your perfume with its hint of vanilla ice cream. However, he thinks it is turning his white soap brown. Is this reasonable? OH O O Benzyl acetate Tetrahydrolinalool

O

O

O

Acetaldehyde ethyl phenylethyl acetal Citral

Figure 8.32  Fragrance ingredients for a laundry powder perfume.

OH Mefrosol

OH

Dihydromyrcenol

OH

OH Geraniol

Figure 8.33  Acid‐stable alcohols.

Citronellol

171

9 Oxidation and Reduction Reactions This chapter concerns those chemical changes brought about through oxida­ tion or reduction. Reduction and oxidation reactions are generally referred to as red‐ox reactions for short. As in the previous chapter, we will concentrate on those reactions that are used in our industry or to which perfumes are sub­ jected when in use. But first we need to understand what is meant by oxidation and reduction, and to help in this, we will look at some inorganic reactions. As with the acid/base reactions of Chapter 8, we will see how we can use red‐ox reactions to make fragrance ingredients and also how the presence of oxidants and reductants in our customers’ products can degrade fragrance ingredients. This latter problem is more serious with red‐ox reactions than with acid/base reactions, since the customer’s product can also be damaged in the red‐ox reac­ tion. Oxygen in the atmosphere is an oxidant, and reaction with it, known as autoxidation, is another hazard for perfumes. Autoxidation is also responsible for forming malodours in some products, but it can also be used in the synthe­ sis of perfume ingredients. Everyone is familiar with the fact that iron rusts when exposed to air and water. This reaction is an example of oxidation. Atoms of metallic iron, Fe, give up their outer valence electrons and are changed into cations. If the iron loses two of its outer electrons, a dipositive cation, Fe2+, results. This cation is capable of losing another electron and becoming Fe3+. We refer to these three different states of iron as oxidation states. Fe, commonly known as metallic iron, is in the ground state and has an oxidation level of 0. The other two oxidation states are called +2 and +3 and are known as ferrous and ferric, respectively. Thus, FeCl2 is known as ferrous chloride, and FeCl3 as ferric chloride. Rust is a mixture of iron in both +2 and +3 oxidation states, in equal amounts. The oxides of these two states of iron are FeO and Fe2O3, so a mixture of the two in equal amounts is Fe3O4, which is the formula for rust. Increasing the oxidation state is known as oxidation, and the reverse process, i.e. the lowering of oxidation state, is known as reduction. This cycle for iron is shown in Figure 9.1. From the figure it is clear that oxidation involves losing electrons and reduction involves gaining electrons. Figure 9.2 shows a part of the more complex red‐ox cycle of chromium. In the figure, we see the oxidation of chromium in the +3 oxidation state to the +6. Chromium in the +3 oxidation level has a blue‐green colour, whereas, in the +6 state, which includes the salts known as dichromates, it has a deep yellow colour. Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

172

9  Oxidation and Reduction Reactions Oxidation –2e–

–1e–

Fe

Fe2+

+2e–

Fe3+

+1e– Reduction

Figure 9.1  Oxidation and reduction of iron. –6e–

2Cr3+

Cr2O72–

+6e–

Figure 9.2  A red‐ox cycle of chromium.

Therefore, reduction of dichromate results in a change of colour from yellow to green, which is the basis of the police breath test for alcohol. So as chromium gains electrons on going from the +6 down to the +3 level, the electrons must come from somewhere. In practice, they come from another molecule or ion, and this species must then lose electrons in the process. In other words, it is oxi­ dised as the chromium is reduced. This reaction leads us to an important princi­ ple. In every red‐ox reaction, for every material oxidised, something else is reduced, and for every material reduced, something else is oxidised. In the case of the breath test, the chromium is reduced and changes colour, while the alcohol on the breath is oxidised, to acetic acid. The oxidation of alcohol in the breath test introduces us to oxidation and reduction of organic compounds. However, before we go into detail about this subject, we should consider two other inorganic red‐ox systems of importance to the fragrance industry, those of sulfur and chlorine. Four different oxidation states of sulfur are shown in Figure 9.3. Elemental sul­ fur is found in the rims of volcanoes and has been mined from these places throughout recorded history. Rather than exist as single atoms, sulfur forms rings of eight atoms in this ground state (i.e. with an oxidation level of 0). In ancient texts such as Homer’s Odyssey and The Bible, we often come across the use of burning sulfur as a means of purification. When sulfur is burnt in air, it is oxidised by the oxygen in the air to produce sulfur dioxide, SO2. This pungent, irritant gas dissolves in water to form sulfurous acid, H2SO3. This acid is lethal to bacteria and fungi, and so this aspect of it is the basis of its use in purification.

Oxidation and Reduction Reactions Oxidation Hydrogen sulfide

Sulfur

H 2S

S8

–2

0

Sulfurous acid

Sulfuric acid

H2SO3

H2SO4

+4

+6

Reduction

Figure 9.3  Oxidation levels of sulfur.

Sulfur dioxide is still used as a fumigant today, for example, in preservation of dried fruit. Further oxidation of the sulfur atom in sulfur dioxide produces sulfur trioxide, SO3. When SO3 dissolves in water, it produces sulfuric acid, H2SO4. The oxidation states of sulfur in sulfurous and sulfuric acids are +4 and +6, respec­ tively. At the other end of the scale, sulfur can be reduced to the −2 oxidation state. The simplest example of a material in this state is hydrogen sulfide, H2S, the main malodorous component of rotten eggs and many other sources also. Figure  9.4 shows oxidation of organosulfur compounds. In this figure, [O] is  used to illustrate oxidation without specifying the oxidant used, as is the convention in reaction schemes. The top line of Figure 9.4 shows the oxidation of dimethylsulfide, (CH3)2S, to dimethylsulfoxide, DMSO. The middle line of Figure 9.4 shows how thiols (also known as mercaptans because of their affinity for mercury) are oxidised to sulfonic acids. Finally, the lower line shows oxida­ tion of a thioether to the corresponding sulfoxide and then on to the corre­ sponding sulfone. These oxidations are important in perfumery because thiols and thioethers often have very intense odours, whereas the oxidised products are usually odourless. Many malodours are due to thiols and thioethers, for example, sewage, bad breath, gas, rotting vegetation, cat urine, and so on. One O [O]

S

O S

R

O

[O] S R

O

[O]

SH

R

S

R

[O]

O

O S

S R

OH

R

Figure 9.4  Oxidation of organosulfur compounds.

R

R

173

174

9  Oxidation and Reduction Reactions

Oxidation Hydrogen chloride

Chlorine

Sodium hypochlorite

Sodium chlorate

Sodium perchlorate

HCI

Cl2

Na+ ClO–

Na+ ClO3–

Na+ ClO4–

–1

0

+1

+5

+7

Reduction

Figure 9.5  Oxidation levels of chlorine.

way of dealing with these malodours is therefore to oxidise them, and even mild oxidants will do the trick. However, some desirable odours are also due to thiols and thioethers. Examples include the key odour components of grapefruit and passion fruit. These odours are often used in perfumery to give tropical fruit notes, but they, like all materials with sulfur in the −2 oxidation level, are very prone to oxidation, and so their odour contribution is easily lost if the product to which the perfume is added contains oxidants. Indeed, even atmospheric oxygen is capable of destroying such odorants. The five most important oxidation states of chlorine are shown in Figure 9.5. The lowest is the −1 oxidation state, and hydrogen chloride is the simplest exam­ ple of this. When dissolved in water, it produces hydrochloric acid, which disso­ ciates into protons and chloride anions. The chloride anion is therefore also an example of chlorine in the −1 oxidation state. In HCl, the chlorine atom has one bond to hydrogen just as, in the −2 oxidation state, sulfur has two bonds to hydrogen. The chloride anion also has one electron more than does a chlorine atom. As we go to the higher oxidation states of chlorine, we find that the chlo­ rine atom has bonds to oxygen. Figure 9.5 shows the salts of the oxo‐acids (i.e. acids in which the acidic hydrogen is bonded to an oxygen atom that is in turn bonded to the heteroatom in question) of chlorine, as these salts are more com­ monly encountered than the parent acids. Sodium hypochlorite is the active ingredient of so‐called ‘chlorine bleaches’, and sodium chlorate is used as a weed killer. The chlorine atom in sodium perchlorate is at the highest oxidation level observed for chlorine. For the perfumery industry, sodium hypochlorite is the most important of the salts of chlorine oxo‐acids. Chlorine gas is easily generated by electrolysis of brine. When dissolved in water, the chlorine molecule, Cl2, is hydrolysed to give one molecule of hydrogen chloride and one of hypochlorous acid, HOCl. Hypochlorous acid is unstable and, in the presence of acids, readily reverts to chlorine. (This instability is the reason you should never mix products based on chlorine bleach with acidic products, e.g. by mixing a chlorine bleach lavatory cleaner with an acidic lava­ tory cleaner, as it will release the dangerous chlorine gas.) Therefore, in the man­ ufacture of chlorine bleach or eau de Javel as it is known in French, chlorine is dissolved instead in sodium hydroxide solution. This solution neutralises both

Oxidation and Reduction Reactions

Cl

H2O

Cl

H

Cl

+

H

O

Cl

2NaOH

Na+

Na+ Cl−

−O

Cl +2H2O

Figure 9.6  Formation of sodium hypochlorite.

the hydrochloric acid and the hypochlorous acid and generates a solution of sodium hypochlorite as shown in Figure 9.6. In order to keep the sodium hypo­ chlorite from reverting to chlorine, an excess of sodium hydroxide is always present, so these bleach solutions are strongly alkaline. The combination of high alkalinity and the oxidising power of hypochlorite makes the solution very effective in destroying dirt and killing bacteria, but it also makes it effective at bleaching colours (e.g. dyes on clothes) and degrading fragrance molecules (not to mention being corrosive to skin and fabrics). The problems that this powerful combination creates for the fragrance industry will be described later in this chapter and also in Chapter 11. Figure 9.7 shows the structures of the oxo‐acids of sulfur and chlorine, and by comparison with Figures 9.3 and 9.5 and the oxidation levels shown there for the sulfur and chlorine atoms in these species, it is clear that the oxidation level cor­ responds to the number of bonds that the atom makes to oxygen or to hydrogen. Thus, if we count the number of bonds an atom has to oxygen and then subtract H

H

H

O

O

O

S O

O

Sulfurous acid

H O

S O

Sulfuric acid H H

H Cl

O

O

Cl

O

O Hypochlorous acid

Chloric acid

O

O Cl O

O

Perchloric acid

Figure 9.7  Structures of sulfur and chlorine oxo‐acids.

175

176

9  Oxidation and Reduction Reactions

the number to hydrogen, we arrive at the oxidation state. For example, perchloric acid has three double bonds and one single bond to oxygen, so (3 × 2) + 1 = 7. Similarly, hydrogen sulfide (shown in Figure 9.3) has no bonds to oxygen but two to hydrogen, and so 0−2 = −2. At the start of the chapter, oxidation and reduction were defined, respectively, as losing or gaining electrons. Considering the simple examples of sulfur and chlorine oxidation states in both Figures 9.3 and 9.5, we can see two alternative definitions, and these definitions will be of help when we come to consider the oxidation levels of carbon compounds. So, we can now define oxidation and reduction in three different ways: Oxidation

Reduction

Losing electrons Gaining bonds to oxygen Losing bonds to hydrogen Gaining electrons Losing bonds to oxygen Gaining bonds to hydrogen

The highest oxidation state of chlorine is perchloric acid and its salts, the per­ chlorates. The prefix ‘per’ is used to indicate high oxidation levels. We will come across it in consumer goods when it is used to indicate high oxidation levels of oxygen itself. The simplest peroxy species is hydrogen peroxide, H2O2. This sub­ stance and the main classes of organic peroxides are shown in Figure 9.8. All of the peroxy species have an oxygen–oxygen bond in them. The oxygen–oxygen bond is high in energy and unstable and will therefore readily break to set in motion an oxidation reaction of other molecules near it. These reactions can be catalysed by acid or base and, in the absence of either, will proceed by a free radi­ cal mechanism. These free radical reactions represent the hazard in organic per­ oxides, since they can be set in motion by heat, particularly when the peroxide is concentrated. For this reason, alternative routes to peroxide bleaches are used, as will be seen in Chapter 11, in which the peroxide is generated at the point where it is used and therefore does not have to be stored. Let us now look at oxidation and reduction in carbon compounds – in other words, red‐ox reactions in organic chemistry. H

O

O

H

Hydrogen peroxide

R R

O

O

H

Alkyl hydroperoxide

R

O

O

R

Dialkyl peroxide

Figure 9.8  Peroxy species in organic chemistry.

O

O

H

O Percarboxylic acid or (for short) peracid

Oxidation and Reduction Reactions Oxidation

H

H H

H

H

H

H H

Ethane

H

H

H

H

etc. etc.

Ethyne

Ethene

etc. etc.

Diamond

Reduction

Figure 9.9  Red‐ox states of hydrocarbons.

Figure  9.9 shows ethane, ethene, ethyne, and diamond (a polymer formed entirely from sp3 hybridised carbon atoms bonded via single bonds in a three‐ dimensional array). Using our definitions of oxidation as ‘losing bonds to hydro­ gen’ and reduction as ‘gaining bonds to hydrogen’, we can see that ethane is the most reduced, i.e. the lowest oxidation state, of carbon in these compounds and diamond is the highest. Since in diamond the carbon atoms are all bonded only to other carbon atoms, it is considered to be the 0 oxidation level. Hydrogenation of a double bond, as in the case of formation of ethane from ethene, constitutes a reduction reaction. Saturated hydrocarbons are the most reduced members of the hydrocarbon family and so might be expected to be the ones to react most readily with oxi­ dants. However, as we saw in Chapter 7, for a reaction to proceed, the reagents must be able to reach the transition state. For saturated hydrocarbons, the transi­ tion state is much higher than that for olefins or aromatic compounds, so it is actually these latter groups that are more susceptible to oxidation. Molecules containing double bonds are prone to oxidation by hypochlorite (chlorine bleach) and peracids (oxygen bleach). Aromatic molecules are less susceptible but can still be attacked. Figure 9.10 shows the oxidation levels of some organic materials containing oxygen. The lowest oxidation level is that of the alcohols, and the highest is carbon dioxide. Oxidation of a primary alcohol (as shown in the first column) produces the corresponding aldehyde, whereas oxidation of a secondary alco­ hol (as shown in the second column) produces a ketone. Tertiary alcohols are resistant to oxidation. Aldehydes are readily oxidised to acids. In fact, this oxi­ dation is easier than the oxidation of a primary alcohol to an aldehyde. It is therefore difficult to produce an aldehyde from the corresponding primary alcohol because the product is usually further oxidised directly under the reac­ tion conditions to give the acid. One way of producing the aldehyde is to dehy­ drogenate the alcohol in a vacuum using a hydrogenation/dehydrogenation catalyst. The aldehyde is usually the most volatile of the three components and is removed from the catalyst surface as soon as it is formed. It is also possible to choose a catalyst, such as copper chromite, which will attack alcohols more easily than aldehydes. Ketones can be oxidised to esters (hence cyclic ketones to lactones) in a reaction known as the Baeyer–Villiger reaction, which is usu­ ally achieved with a peracid as the oxidant. Given vigorous enough oxidation

177

178

9  Oxidation and Reduction Reactions Reduction R Alcohols

R

OH

R

R OH

R

R OH

R Aldehydes/ ketones

R

O

R

O

OH Acids

Carbon dioxide

O

R

O

O

O

R

R O

Oxidation

Figure 9.10  Oxidation levels of oxygenated organics.

conditions, all organic compounds will end up as carbon dioxide. This fact is the basis of the chemical oxygen demand (COD) determination discussed in Chapter  6. The reagent used in COD determinations is chromic acid, which will effectively oxidise any organic matter to carbon dioxide. Aldehydes and ketones, aldehydes in particular, often have desirable odour characteristics, and many key fragrance ingredients contain these functionali­ ties. However, their instability to oxidants limits their usefulness in fragrances for products containing bleaches. In Chapter 8 we saw how acetals, ketals, nitriles, and Schiff ’s bases are used to protect aldehydes from acids and bases, and the same applies to oxidation reactions, although the Schiff ’s bases are less success­ ful than the others in this case since Schiff ’s bases can also be oxidised. In general, ketones are more stable to oxidants than are aldehydes, but one oxidation reaction of ketones should be noted. When treated with hypochlorite, methyl ketones undergo the chloroform reaction, so‐called because it produces chloroform, CHCl3. The mechanism of this reaction is shown in Figure 9.11. The methyl ketone is first enolised, and the enol reacts with base and HOCl as shown in the top line of the figure. This reaction produces the monochlorinated ketone. Repetition of the process with the monochlorinated ketone gives the dichlorok­ etone, and similarly, the dichloroketone is converted to the trichloroketone. As explained in Chapter 8, the chlorine atoms exert an inductive effect on the neigh­ bouring carbon atom, and this withdrawal of charge from it means that an anion centred on the carbon atom is relatively stable. Therefore, when hydroxide adds to the carbonyl carbon, the bond between it and the adjacent carbon atom, which carries three chlorine atoms, breaks to form the CCl3 anion and a carboxylic acid. The CCl3 anion is relatively stable because of the combined inductive effect of the three chlorines, and it is said to be a good leaving group because of the relative ease with which the bond holding it is broken. However, this anion is a

Oxidation and Reduction Reactions HO–

H

O

O

O

H

R

H

R

H H

Cl

H HO–

H

O

H Cl

Cl HO–

O

H

O

Cl

R

H

H

Cl

O Cl

R

H Cl

H H

O

Cl

R

H

O

R

H H

Cl

R

O

Cl

R

O

Cl

Cl

O

Cl

R

H

Cl

Cl O

O O

–OH

O–

Cl

R Cl

OH

R Cl

Cl

OH

R

Cl

+

+

Cl Cl

O–

R

– Cl

Cl

Cl H

Cl Cl

Figure 9.11  The chloroform reaction.

stronger base than water and so extracts a proton from water to form chloro­ form. Since hypochlorite solution is alkaline, the acid will be neutralised to form the corresponding carboxylate anion. The final products are therefore the salt of the acid and chloroform, CHCl3. Reaction between hypochlorite, or peroxide bleach, and perfume ingredients is undesirable for various reasons. First, the perfume is destroyed. Second, the reacting bleach is consumed, making the customer’s product less effective than it should be. In the case of the chloroform reaction, a third problem is the forma­ tion of chloroform. Chloroform presents a toxic hazard, although the dose will be low, and therefore the risk is relatively low. Lastly, formation of any chlorin­ ated organic substance presents an environmental hazard. It is therefore best not to use perfume ingredients containing a methyl ketone in products containing hypochlorite bleach. The so‐called oxygen bleaches rely on peroxy bonds for their activity. The first generation to be used in laundry powders used sodium perborate as the active oxidant. This oxidant is less reactive than hypochlorite, and these early products required high temperatures to activate the oxidation system. Nowadays, per­ borate is used with an activator system. These activators react more readily with perborate than do stains, and they produce a more active oxidant that can then attack the stain. The most common activator is tetraacetylethylenediamine

179

180

9  Oxidation and Reduction Reactions O

O

N

Sodium perborate

O

N

O

O

+ etc.

OH

O TAED

Figure 9.12  Formation of peracetic acid from TAED/perborate.

O R

OH

H O

R

O

O

O

Figure 9.13  Reaction of a peracid with an olefin.

(TAED for short), which reacts with perborate to generate peracetic acid – the active bleaching agent. This reaction is shown in Figure 9.12. Peracetic acid attacks aldehydes, ketones, and esters, and it also attacks double bonds to form epoxides, three‐membered rings with two carbon atoms and one oxygen atom. Epoxides can be opened by hydrolysis or attack by nucleophiles. In acidic conditions, they can rearrange to ketones. All of these species are subject to further attack by oxidants, so peracetic acid is not good news for fragrance ingredients containing olefinic bonds. The mechanism of the reaction is shown in Figure 9.13. Oxygen is an oxidising agent, and, since it comprises about one fifth of the air, it is always a potential danger to perfume ingredients unless we take rigorous steps to ensure that the two are kept separate. Although we normally write the oxygen molecule, O2, as having a double bond between the two oxygen atoms, much of the chemistry of oxygen is easier to understand if we consider it to have only one double bond and one unpaired electron on each oxygen atom (i.e. it is a diradical). Oxidation by atmospheric oxygen gas is known as autoxidation, and it is essentially a free radical process. The cosmetics industry is much concerned with damage to the skin caused by autoxidation, and the layman will probably associate terms such as free radical and antioxidant with skin ageing. Of course, the food industry is also interested in natural antioxidants, such as the flavonoids found in red wine or green tea, as a way of helping to counteract oxidative dam­ age to other organs. As discussed in Chapter  2, free radicals are species containing an unpaired electron and are therefore very reactive. Like carbocations and carbanions, unpaired electrons are stabilised by adjacent double bonds, and in the case of free radicals, this classification includes carbon–oxygen as well as carbon–­ carbon double bonds. They are also stabilised by adjacent ether groups. The more stable a radical is, the more easily it will form, so molecules containing functions that will stabilise free radicals are more readily attacked by other radi­ cals, including oxygen. Therefore aldehydes, ethers, olefins, aromatics, and so on are all prone to autoxidation.

Oxidation and Reduction Reactions

O

O

H H

+

O

O

Figure 9.14  Autoxidation of tetralin – initiation.

Free radical reactions proceed with three different steps: initiation, propaga­ tion, and termination. To illustrate these stages, we will look at the autoxida­ tion of 1,2,3,4‐tetrahydronapthalene, or tetralin as it is more commonly called. Figure  9.14 shows the initiation step of this reaction. The oxygen molecule, behaving as a diradical, abstracts a hydrogen atom from tetralin at the position next to the benzene ring. The carbon radical thus produced is stabilised by the adjacent aromatic system but is still a very reactive species. The other product of the reaction is the hydroperoxy radical, another very reactive species. Fortunately, the initial hydrogen abstraction reaction has a high energy of activation; other­ wise organic matter, including us, would quickly be oxidised on exposure to air. However, once this step has occurred, two very reactive species are liberated into the environment. The most damaging stage of autoxidation is the propagation stage, which is shown in Figure 9.15. Many different reactions can take place by way of propaga­ tion, and only some of them are shown in the figure. The key feature of propaga­ tion is that after each propagation reaction, at least one free radical will still be left to keep the chain going, and so more and more molecules will be involved in O

O

+

O

H O

O H

O H H

O

O

+

H

etc. etc.

Figure 9.15  Autoxidation of tetralin – propagation.

O

etc.

O

H

etc.

181

182

9  Oxidation and Reduction Reactions

the overall reaction. On the left of the figure, we see the initial tetralin radical reacting with a molecule of oxygen to generate an alkyl peroxy radical, which is then capable of removing a hydrogen atom from another molecule of tetralin. In the central column of Figure 9.15, the hydroperoxy radical abstracts a hydrogen atom from another tetralin molecule to give a tetralin radical and a molecule of hydrogen peroxide. The hydrogen peroxide can then cleave to give two hydrop­ eroxide radicals. Thus, from our one initiation reaction, we now have three tetra­ lin molecules that have already been attacked and four more free radicals that are ready to continue the chain of oxidation. Free radical chain reactions will continue until they reach a termination stage. The only way to stop a free radical chain is for two radicals to combine to form a neutral species. This combination is shown in Figure  9.16 where two tetralin radicals react with each other to form an octahydrobinaphthyl molecule. The hydrogen atom attached to the carbonyl carbon of aldehydes is also suscep­ tible to abstraction by radicals, and some of the possible ensuing reactions are shown in Figure 9.17. The initiation reaction forms a carbonyl radical and a hydrop­ eroxy radical. The latter can, of course, start another train of radical reactions,

Figure 9.16  Autoxidation of tetralin – termination.

O R

H

O +

O

O

O R O

O

Fragrance material

Oxidised fragrance material

O

O

O R

R

OH

O

O

H

R

+ Oxidised fragrance material

Fragrance material

O R

O

OH

R

Figure 9.17  Autoxidation in a perfume containing an aldehyde.

etc.

Oxidation and Reduction Reactions H O O R

+

O

O

O

+

O

–CO

R

R

O

R

R

+ OH

O

HO

R

H

R

O R

O

O

R

(1)

R OH

O

(2) RH

O

R

HO

R O

OH O

Figure 9.18  Oxidative decarbonylation of an aldehyde.

while the former can pick up an oxygen molecule and then abstract a hydrogen atom from another fragrance molecule to form a peracid and another radical and hence cause a chain of reactions. The peracid can oxidise other fragrance materials and generate the acid corresponding to the original aldehyde. Radicals derived from aldehydes can also lose a molecule of carbon dioxide, and one important reaction sequence induced by this fragmentation is shown in Figure  9.18. On the top line of the figure, we see the initial abstraction of the carbonyl hydrogen atom and then loss of carbon monoxide from the resultant radical. The alkyl radical thus produced can pick up an oxygen molecule and then a hydrogen atom to give a hydroperoxide. By oxidising other molecules in its environment, this substance is then reduced to the corresponding alcohol. Alcohols, like ethers, can lose the hydrogen atom next to the oxygen through abstraction by a radical. The hydroxy radical thus generated can pick up an oxy­ gen molecule to give, after hydrogen abstraction, a peroxy acetal that can elimi­ nate hydrogen peroxide to give a ketone. Such ketones usually have significant odours. This reaction is observed in products, even solid products such as soap, and the problem of losing the aldehydic note from the fragrance is compounded by the appearance of that of the ketone. Unsaturated fats are also subject to autoxidation as shown in Figure 9.19. As with other autoxidations, the initial step is a hydrogen atom abstraction by oxy­ gen. The radical thus formed then adds to another oxygen atom, and the hydrop­ eroxy radical produced abstracts a hydrogen atom from another molecule to give the corresponding hydroperoxide. Cleavage of the oxygen–oxygen bond in a six‐ membered cyclic reaction then results in the breaking of the carbon chain of the fatty acid to give an unsaturated aldehyde as one of the products. If this aldehyde is formed from the part of the fatty acid molecule that does not carry the acid

183

184

9  Oxidation and Reduction Reactions R

R

R

H O

O

O

O

O

O

RH

R +

R + O

O H2O

O

R

H

H

Figure 9.19  Autoxidation of fats.

function, it will be volatile and thus likely to possess a strong odour. Aldehydes formed in this way are a key component of the odour of rancid fat. They can also be formed from any material that is made from fatty acids, such as soaps. Autoxidation of fats therefore produces many of the malodours that perfumes are asked to cover, but some autoxidation reactions are also used to produce feedstocks for synthesis of fragrance ingredients, as will be seen in Chapter 15. Phenols are easily autoxidised in alkaline solution. The resulting phenoxy radi­ cals can dimerise (when two identical molecules add together, they are said to dimerise) in a termination reaction. The bis‐phenols (molecules containing two phenolic groups) thus produced can be further oxidised to diketones as shown in Figure 9.20. These diketones are reactive molecules and undergo aldol‐type con­ densation reactions to generate products with a red colour. Phenolic compounds and air oxidation are therefore suspected when alkaline products develop pink tints. The ancient Egyptians knew about the damage that can be done to perfumes by autoxidation. Perfume bottles found in tombs were filled to the brim and then sealed with wax to keep air out. Keeping perfume in oxygen‐free conditions, cool, and in the dark (light can accelerate autoxidation) will keep it stable for thousands of years, as evidenced by these perfumes found in Egyptian tombs. However, it is not always possible to achieve such conditions, so other methods of protecting perfumes are needed. This safekeeping is achieved through the use of antioxidants. Antioxidants are materials that are readily attacked by radicals to produce relatively stable, hence

O– 2

[O]

OH OH

Figure 9.20  Oxidative coupling of phenols.

[O]

O O

Review Questions  OH

O

RH

R

+

O

O

OH

OH

O BHA ~ butylated hydroxyanisole

BHT ~ butylated hydroxytoluene

HO O Vitamin E ~ tocopherol

Figure 9.21  Antioxidants.

long‐lived, radicals that do not take part in chain propagation reactions but serve as radical chain termination agents. Very often such antioxidants contain phenolic groups that are sterically hin­ dered (i.e. prevented from reacting because of large components of molecular structure around the potentially reactive site, which physically block other species from approaching close enough to react). Some examples are shown in Figure  9.21. Butylated hydroxyanisole (BHA) and butylated hydroxytoluene (BHT) are two inexpensive synthetic antioxidants. The figure also shows the for­ mation of the relatively stable BHA radical. Vitamin E, also known as tocopherol, is a natural antioxidant, and, as can be seen from the figure, it also contains a hindered phenolic group. The long hydrocarbon tail makes it very oil soluble, so it is now the preferred antioxidant for perfumes and cosmetics.

Review Questions 1 Which of the following two alcohols will be more stable in a product contain­ ing hypochlorite bleach? OH OH

Mefrosol®

Rossitol®

185

186

9  Oxidation and Reduction Reactions

2 Would Lilial® be stable in a laundry powder containing TAED/perborate? O

Lilial®

3 Tertiary alcohols are stable to sodium hypochlorite and peroxy bleaches, but they are oxidised by chromic acid in the COD determination. Why is this so? 4 Which of the following musk ingredients would be best for use in a product containing hypochlorite?

O

Tonalid®

Galaxolide®

O

O O

Cyclopentadecanolide

187

10 Perfume Structure ­Notes, Chords, and Discords A perfume is an artistic composition, like a painting or a symphony. Perfumes use essential oils and aroma chemicals in the way that a painting uses colours and a symphony uses musical notes. Indeed, the language of perfumery resembles the language of music in that perfumers talk of notes (the individual odour compo­ nents), chords or accords (pleasing combinations of notes), and discords (combi­ nations of notes that produce a jarring effect). As one would expect, chords are much more common than discords in perfumery as in music, yet perfumers can use odour discords just as effectively as composers use musical discords.

­Ingredients The primary requirement of a perfume is to have an attractive, aesthetically pleasing odour. The success of this pursuit, of course, is a matter of taste, as what will be pleasing to one person may be repugnant to another. The perfumer will therefore try to create perfumes that appeal to as large a percentage of the population as possible. Fashion and cultural background will also affect what is seen as pleasing, and the perfumer must be sensitive to these factors when cre­ ating a perfume. Learning which ingredients work well together and creating a pleasing combination constitutes a large part of a perfumer’s training. With no easy way to do this, a good sense of what works can only be learned through experience. The original ingredients of perfumery were extracts from odorous natural oils, and these are still used today. The methods for extracting them from the plant sources are described in Chapter 5. Essential oils and other natural extracts are mixtures of odorous chemicals. Occasionally they will contain relatively few chemical components, but the vast majority are complex mixtures containing hundreds of different chemicals. Nature builds these chemicals from carbon dioxide and water, as will be described in Chapter 14. With the development of synthetic organic chemistry in the nineteenth century, it became possible to make odorous chemicals from feedstocks such as coal, min­ eral oil, and turpentine. These synthetic aroma chemicals provide a more secure Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

188

10  Perfume Structure

supply, are usually less expensive, perform better in hostile product bases, and, contrary to some lay opinion, are often safer than their natural counterparts. The methods by which they are produced are outlined in Chapter 15. In addition to natural extracts and aroma chemicals, perfumers also use simple formulae called bases or accords as ingredients. These formulae are used to help perfumers develop a complex formula more easily by providing a premade, balanced odour note that can be incorporated as a component on its own into the full perfume formula. They often provide an alternative to a natural oil. For example, a rose accord could be used to provide an odour profile similar to that of natural rose oil but at a much lower price. Sometimes an accord will be developed for specific product applications. For example, a non‐discolouring jasmine accord might be developed specifically for use in soap applications since the natural oil will produce a brown colouration in any soap to which it is added, as will be explained in Chapter 11.

­Odour Families and Top, Middle, and Base Notes The notes used in perfumery are usually grouped into categories based on the botanical source of the original oils that typify that note. So, for example, lemon, orange, lime, bergamot, and grapefruit are grouped together as citrus notes; rose, jasmine, carnation, and lily of the valley (muguet) as floral notes; and sandalwood and cedarwood as woody notes (see Tables 10.1–10.3 for more examples). One commonly used group that does not have a simple natural counterpart is the aldehydic family. These materials are exactly what their name implies, aldehydes. More specifically, they are aliphatic aldehydes with fatty, slightly ozonic char­acters. Chanel No. 5 uses aldehydes to give it its unique top note. It was such a huge success when it was launched in 1921 that it created a fashion for aldehydic top notes and this is now a mainstream class of fragrance ingredients. The odour families are considered to fall into three larger groups, based largely on their volatility. These three categories are known as top notes (or head notes), middle notes (or heart notes), and base notes (or end or bottom notes). The top notes are the most volatile and represent the first 15 minutes Table 10.1  Top notes. Citrus

Orange, lemon, mandarin, grapefruit, bergamot, lime, petitgrain

Herbal

Pine, rosemary, basil, oregano, tarragon (estragon)

Aldehydic

Decanal, 2‐methylundecanal (or aldehyde MNA or methyl nonylacetaldehyde), undecanal, 10‐undecenal, dodecanal

Marine/ozone

Calone®, Ozonal®, Algol®

Green

Galbanum, cis‐3‐hexenol and its esters, hyacinth (jacinthe)

Fruit

Apple, pear, banana, peach, plum, mango, passionfruit, red fruits – e.g. raspberry, blackberry, blackcurrant (cassis)

­Odour Families and Top, Middle, and Base Note

Table 10.2  Middle notes. Floral

Rose, jasmine, lily of the valley (muguet), ylang‐ylang, carnation (oeillet), daffodil (jonquil), narcissus (narcisse), cyclamen, chrysanthemum, freesia, lilac, lime tree (tilleul, linden), orange flower (neroli bigarade), tuberose, verbena, wallflower (giroflée), hawthorn (aubepine)

Spice

Anise, cinnamon, clove, nutmeg

Table 10.3  Base notes. Wood

Cedarwood, sandalwood, patchouli, vetiver, agarwood (also known as eaglewood, aloes wood, or oud)

Animala)

Civet, musk, castoreum (from beaver = castor)

Amber

Ambergris, labdanum, myrrh

Vanillic

Benzoin, tonka, vanillin, ethylvanillin, ethylmaltol

a) For ethical reasons, natural animalic notes are not used in modern perfumery; they have been replaced by synthetic equivalents.

or so of evaporation, the middle notes last for several hours, and the base notes will persist for days or weeks on a perfumer’s blotter. The first sniff of a freshly opened bottle of perfume will therefore contain more of the top notes, and a perfume should be judged on an extended dry‐out period rather than a first impression. However, we must remember that some middle and end notes have very high impact and will create a significant effect even on the ‘top’ of the fragrance. A triangle is usually used to illustrate the top, middle, and base notes present in a fragrance, as shown in Figure 10.1. The basic outline is shown on the left of the figure, and its application to Chanel No. 5 (as judged by Edwards 1998) is  shown on the right. In their book on perfumery, Calkin and Jellinek (see

Fresh flowery

Top or head

Ylang neroli aldehydes

Middle or heart

Jasmine rose de Mai orris, muguet

Base or end

Sandalwood vetiver, musk, vanilla, civet, oakmoss

Basic structure

Figure 10.1  Perfume structure.

Chanel 5

Floral

Woody

189

190

10  Perfume Structure

bibliography for details) suggest that the balance between the three categories should be 15–25% top notes, 30–40% middle notes, and 45–55% base notes. However, these guidelines are broad and vary according to taste, especially as the definitions of what constitutes a top, middle, or end note vary with differ­ ent experts. The balance will also depend on the use to which the perfume will be put. Calkin and Jellinek’s ratios are characteristic of a fine fragrance, whereas, for example, a fragrance for a shampoo or shower gel will contain a much higher percentage of top notes, since the consumer will want a burst of fragrance when the shampoo or gel is used but will probably not want the per­ fume to last and interfere with a fine fragrance or aftershave that will be worn after the shower. As can be seen from the figure, according to Edwards, Chanel No. 5 can be described as having a fresh flowery top note and a floral heart on a woody base. The freshness in the top comes from the aliphatic aldehydes that its creator, Ernest Beaux, used to such dramatic effect. This use of aldehydes was responsi­ ble for the immediate and enormous success of Chanel No. 5 and led to the development of the whole family of aldehydic fragrances. The other main com­ ponents of the top note are flower oils, but both also have fresh top notes. They are neroli (or orange flower) and ylang‐ylang (the flower oil of a tropical tree, which has a wintergreen like top note). The floral heart contains the three classic flowers of perfumery, rose, jasmine, and muguet. In the case of Chanel No. 5, the rose is actually rose de Mai, which is harvested in May in the Grasse region of Provence. Muguet is the French name for lily of the valley. The other note that Edwards puts into the heart notes is orris. This extract is from the iris plant’s rhizome rather than the flower. The base contains a number of woody notes and is sweetened by musk and vanilla. A well‐constructed perfume will have notes that blend well and run into each other successfully as the perfume evaporates. For instance, aldehydes and muguet notes are related in odour terms and also chemically, as the structures of many of the best muguet aroma chemicals are aldehydes themselves and retain a hint of the aldehydic character. Similarly, muguet and sandalwood are related odours, and small changes in the molecular structure of a muguet ingredient often give a sandalwood chemical and vice versa. Thus, in the example of Chanel No. 5, the muguet notes in the heart provide a link between the aldehydic top notes and the sandalwood in the base. Essential oils often contain chemically related materials with different volatilities, which is part of the reason why they work so well as fragrances. Ylang‐ylang has a very complex fragrance, and many notes appear and disappear as the oil dries out on a blotter. They blend wonderfully well together, and the complexity of the oil makes it an excellent blender in fragrances as it helps to link different notes together. A very good training exercise for the would‐be perfumer or connoisseur of perfume is to study closely the dry‐out of a blotter of this beautiful oil. Tables 10.1–10.3 show some of the most important families of top, middle, and base notes, respectively. Some of the most common members of each fam­ ily are also shown in the tables. The names are given in English, and, where appropriate, the French equivalents are given in brackets. Many ingredients are better known by their French names, so it is worthwhile learning the French

­Persistence/Tenacit

terms. For the most part, the names are those of the natural products, but it should be understood that synthetic (or isolated) single ingredients would also be considered to fall into these categories. So, for example, the term rose will include natural rose oils, nature identical materials such as citronellol and phenylethanol, and also synthetics with no natural counterparts, such as Mefrosol® and Florosa®. (Further details of the botanical sources and methods of extraction of natural ingredients can be found in The Chemistry of Fragrance, The Essential Oils, Perfume Materials of Natural Origin or Flower Oils, and Floral Compounds in Perfumery and those of synthetic ingredients in The Chemistry of Fragrance or Common Fragrance and Flavour Materials. These books are all listed in the bibliography.)

­Persistence/Tenacity Persistence and tenacity refer to the length of time that a perfume or perfume ingredient will remain detectable. It depends to a considerable degree on its volatility. The more volatile (lower boiling) a substance is, the more easily it will evaporate and thus be lost by diffusion in the air. Two main factors determine volatility. The first is molecular weight; the heavier a molecule is, the more difficult it is to move it from the liquid phase to the gas phase. Molecules with fewer than eight carbon atoms in their structure are mostly too volatile to be of use in perfume. Similarly, those with more than 18 carbon atoms are usually not volatile enough to reach the olfactory receptors in the nose. So, as a general rule, perfume ingredients have between 8 and 18 carbon atoms in their empirical formulae. The top note ingredients are usually the most volatile, and the base notes the least volatile and most substantive. Volatility is also affected by the ability of molecules to form non‐bonded inter­ actions between each other and to any surface or medium on which they are placed. The more polar a molecule is, the more easily will it form electrostatic bonds, such as hydrogen bonds, to other molecules around it, whether they are other fragrance molecules, cellulose (e.g. in paper or cotton fibres), or proteins (e.g. skin or hair). Thus, for example, an alcohol will be higher boiling than the corresponding aldehyde, which, in turn, will be higher boiling than the corre­ sponding hydrocarbon. Therefore, we can look at the size of a fragrance molecule and its polarity and from these estimates how long it will persist on a perfumer’s blotter. However, in terms of product applications such as laundry detergent or shampoo, substantiv­ ity also implies the ability of a perfume or perfume ingredient to stick to fabrics or hair after use. In these cases, we are dealing with partition of the material between an aqueous phase (the water in the washing machine or shower) and a solid phase (the fabric or hair). The most important factor is therefore the log P (see Chapter 4 for definition of log P) of the material. The less a perfume ingredi­ ent wants to be in the water phase, the more it will want to stick to cloth, hair, skin, or whatever solid surface presents itself as an alternative. Other factors come into play also, in particular the ability of the compound to recognise the fabric or hair surface.

191

192

10  Perfume Structure

Perfume Types Perfumers’ jargon includes a variety of terms describing types of perfume as defined by the combinations of notes used in them. The most obvious term is floral, which describes a perfume centred on floral notes. Classical fragrances of this type would include Joy and L’Air du Temps. The floral aldehydic type is a reference to floral fragrances with an aldehydic top note. The first and defining type is Chanel No. 5. Another classic example is Je Reviens. Colognes take their name from the town of Cologne where the first and archetypical cologne, 4711, was made. Colognes are dilute, containing less than 10% of fragrance oil in the alcoholic solution. They have citrus top notes and a floral heart. The classic combination is bergamot for the citrus note and lavender for the floral. Oriental fragrances are heavy and long lasting and rich in base notes, especially amber, incense, spice, and balsamic ingredients. They have a heavy floral middle, containing such flowers as rose, jasmine, tuberose, and gardenia. Classic examples include Jicky and Shalimar. Floriental is a term used to describe fragrances resembling orientals but with a lighter, more floral top. The chypre family takes its name from the fragrance Chypre, which is the French name for the island of Cyprus. These fragrances are characterised by a citrus top, floral (rose and jasmine) heart, and a base of oakmoss, sandalwood, and musk. The first in the family was Le Chypre, and more recent examples include Ma Griffe and Eau Sauvage. The fougère family of fragrances is from the French word for fern, first used in the classic fragrance Fougère Royale, and is characterised by the use of lavender in combination with coumarin (5,6‐benzopyran2‐ one) and oakmoss. Azzaro pour Homme is another example of this family.

When creating fragrances, perfumers will select materials with the tenacity and substantivity required for the final application in which the perfume will be used. For example, for an air freshener, the aim will be to use ingredients that are volatile and diffuse into the air quickly, whereas for a laundry powder fragrance, the aim will be to produce a fragrance that deposits well onto cloth and is not washed away with the soil.

­Threshold The lowest concentration at which a perfume ingredient can be perceived is known as the threshold of detection. Similarly, the lowest concentration at which the ingredient cannot only be detected, but also identified in odour terms, is known as the recognition threshold. Clearly, the detection threshold is impor­ tant in perfumery since the lower the detection threshold of an ingredient, the less is required to give a noticeable effect.

­Impact Impact is not the same thing as threshold, although the two are often confused. Impact is the intensity of perceived sensation and is a very subjective phenomenon.

­Radiance/Bloo

log intensity

Compound B Compound A

a

b c log concentration

d

Figure 10.2  Odour intensity and concentration.

Like threshold, it can only be measured using human subjects, but in this case, no simple physical measurement such as concentration exists, and impact can only be rated by comparison with that of another material. The human brain is very good at unconsciously adjusting scales, which makes comparisons difficult. The way in which intensity varies with concentration is very important and can be shown in graphical form. Such graphs are usually drawn using logarith­ mic rather than arithmetic scales in order to make them more manageable and more easily understood. They are known as the psychophysical curve or psycho­ physical function of a material. Figure 10.2 shows such curves for two different materials, A and B. Compound A has a lower threshold than compound B as is demonstrated by the fact that its psychophysical curve crosses the horizontal axis at a much lower concentration. Point a tells us the odour detection thresh­ old for compound A, and similarly, point b is the odour detection threshold of compound B. The two lines have different slopes, which tells us that the rate at which the perceived intensity of each varies with concentration is different. At concentration c, compound A is perceived to be more intense than compound B. However, at concentration d, compound B is now perceived as being more intense, even though it has a much higher threshold of detection than does com­ pound A. Like all sensory attributes, odour impact is very subjective, and a graph such as that in Figure 10.2 will vary from one person to another. All sensory data should therefore be collected from large groups rather than from individuals. The results can then be expressed in terms of the average person. However, we must still remember that no one is average. For example, the average family in Britain today has 1.8 children, but I do not know anyone who has 1.8 children. Sensory data therefore is only a guide and should be treated as such.

­Radiance/Bloom There is much talk about fragrance properties known as radiance and bloom; however everyone seems to have his own idea as to what these properties are.

193

194

10  Perfume Structure

The following are therefore my ideas and might not coincide with those of other people. Radiance is the ability of a perfume or perfume ingredient to fill space. Methyl dihydrojasmonate is an example of a radiant material. Smelt from a perfumer’s blotter, it does not seem to have a high impact, yet if that blotter is left in a room, its delicate jasmine character is immediately discernable to anyone entering the room. It seems to me that two properties are crucial in making a material radiant. First, it needs to have a volatility in a certain range. If it is not volatile enough, it will not reach the nose, but if it is too volatile, it will move too quickly into the vapour phase and then be carried away on air currents. Materials with just the right rate of evaporation will sit as a reservoir on the paper, cloth, or whatever medium they were placed on and evaporate slowly to produce a steady flow of molecules into the air over a protracted period. Thus, radiant materials tend to fall into the base note category or at the lower end of the middle notes. Since not much material will be in the air at any given time, the second requirement must therefore be that they have very low odour detection and recognition thresholds. They should also have a flat psychophysical curve so that the perceived intensity is changed very little by concentration and a small amount in air will give much the same impression as smelling a concentrated sample. Another important property is known as trail (English) or silage (French). This characteristic is the phenomenon of a scent trail being left by a moving object (usually a person, when we are talking of perfumery). In my mind, it is closely related to radiance. The same properties that will make a fragrance fill a room will lead to formation of an ‘atmosphere’ of the perfume around its wearer, and as the person moves around, some of this atmosphere will be left behind him or her. Bloom is a slightly more complex phenomenon. Bloom is the ability of a fra­ grance to perfume a room when the fragrance is introduced, not directly as an oil or aerosol, but in a product such as soap. In the case of soap, dry bloom is the fragrance effect of a dry bar of soap left in a room, and wet bloom is the fragrance effect in the room when the soap is in use, i.e. when it is wet. Bloom requires the same basic properties as radiance, but since it is applied to products such as soap, it also relates to the rate of release of the scent from the product. A radiant molecule that remains trapped inside a soap bar hence will not perfume the air around it.

­Physical and Chemical Factors Perfume oils are mostly used in a product medium in the majority of applications and only occasionally used as neat oils. Air fresheners would be an example of such use. Figure 10.3 illustrates some of the issues in delivering the right perfume to the nose of the user. The customer of the fragrance house adds perfume oil to the consumer product. The chemistry of the product varies enormously from one application to another. In the case of fine fragrance, the product base is aqueous ethanol that is relatively perfume friendly. Soaps are a more aggressive

­Physical and Chemical Factor

Headspace Fragrance ingredients in bottle or drum

Degradation

Release

Product matrix containing various active components

Figure 10.3  From bottle to nose.

medium because of their higher pH, and machine dishwashing products contain even more aggressive alkali and bleaches on top of that. The fragrance must sur­ vive in the product through manufacture, distribution, and storage. Manufactur­ ing processes are not always kind to fragrances, for instance, high temperatures might be used. Distribution of the product could include low temperatures (e.g. in an unheated, unpressurised aircraft hold) to quite high temperatures (e.g. standing on a tropical dockside in the midday sun). The container in which the product is sold can also be of significance. Many products (e.g. shampoos, lava­ tory cleaners, fabric conditioners) are sold in plastic bottles. Even this type of container can cause problems with perfumes since the fragrance could leach plasticiser out of the bottle, for example, making the bottle brittle, or, at the other extreme, the perfume could plasticise the bottle to the point where it would lose its shape. Eventually, the perfume must be released from the product into the air so that it can reach the consumer’s nose. All of these parameters must be taken into consideration when creating a fra­ grance and will have an effect on the way the perfumer structures the fragrance. The market often demands that a fine fragrance is followed by a range of prod­ ucts, e.g. soap or hand cream, with the same smell, a phenomenon known as trickle down. Trickle down represents quite a challenge for the perfumers since they cannot just take the original formula and use it in another product category. In addition to price considerations, all of the factors described above come into play, so the available palette of ingredients will be different. Yet it is up to the perfumer to come up with the same odour effect. The effect of consumer goods chemistry on fragrances will be discussed in the next chapter. The perfumer’s skill in recreating the right odour would be a topic for another book on the art of perfumery, rather than this one.

195

196

10  Perfume Structure

Review Questions 1 Would the balance of a perfume intended for use in a fabric conditioner be biased towards top notes or end notes? 2 In what way would the heart of a feminine fine fragrance be most likely to vary from that of an aftershave? 3 What advantage would there be to an ingredient having a low recognition threshold?

197

11 Chemistry in Consumer Goods ­Introduction This chapter builds on Chapters 8 and 9 and shows how the chemical reactions described there affect perfume when it is added to consumer goods. Firstly, we will review the types of ingredients in consumer goods and why they are used, and then we will look at which materials are used in which product categories. Most countries have legal requirements for product labelling in order to let consumers know what the product contains. This chapter will help the reader to understand the terms used in such product labels and why these ingredients are present. Each manufacturer has its own blend of ingredients and often has unique ingredients that differentiate its products from those of its competitors. What follows is there­ fore a general guide to the chemistry of consumer goods rather than a compre­ hensive catalogue. It deals only with the chemical aspects of consumer products and ignores other factors such as commercial and legal constraints. It will be obvious from the first part of the chapter that consumer goods con­ tain a wide variety of chemical species, some of which can be quite aggressive towards fragrance ingredients. With simple acids and bases, the damage is mostly one way, that is, it is the fragrance that is degraded. However, in some systems, by damaging the fragrance, the active ingredient of the product is also destroyed and therefore loses its efficacy. Clearly the perfumers and applications specialists in fragrance houses must learn to design fragrances that will be stable in the product and not interfere with its active ingredients. In order to assess its stability, fragrances are usually put into the product and stored under various conditions for lengthy periods of time. It is not practicable to run tests on every fragrance, especially since customers tend to demand a faster response to briefs than would allow for full‐storage testing. So, all the ingredients on the palette will be storage tested before introduction to the pal­ ette, and a record of their performance made available in order to guide the per­ fumer in selecting the best ingredients for a given product application. These storage conditions are usually more severe than a perfume would be expected to encounter in real life. For example, the temperature will be well above that of a supermarket shelf. Chemical reactions are faster at higher temperatures (as explained in Chapter 7), and a rough rule of thumb is that the rate of a reaction will double with every 10 °C temperature rise. So, for example, the chemical Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

198

11  Chemistry in Consumer Goods

change after storage for 4 weeks at 50 °C would roughly equate to that of storage at 20 °C for 32 weeks. Storage conditions might also include high humidity (to simulate a tropical environment) and UV irradiation (to simulate storage in a clear glass or plastic bottle in bright sunlight). The perfume stability can be eval­ uated either organoleptically (i.e. by its odour properties), by chemical analysis, or both. Organoleptic analysis is the more important of the two, since produc­ tion of tiny amounts of highly odorous materials would have a significant effect on the overall odour profile, even if the levels are too low to accurately assess by analysis. Active ingredients in the product can be analysed chemically to deter­ mine whether or not they are affected by the perfume. The most likely case of product activity loss is with bleaches. This loss is easily measured using a red‐ox titration, as described in Chapters 6 and 9.

­Acids in Consumer Goods Acids can be added to consumer goods for a variety of reasons, and a range of acids are employed depending on the application. Most obviously, they can be used as pH control agents where a low pH is required, such as acid lavatory cleaners and limescale removal products. In areas where limestone and chalk are prevalent, water supplies tend to contain calcium bicarbonate as a result of con­ tact between water and the rocks. At high pH, calcium hydroxide will precipitate out. Heating of calcium bicarbonate solutions leads to the formation of insoluble calcium carbonate (limescale) in kettles, boilers, and washing machines. Eva­ poration of water also leads to precipitation of calcium carbonate, and so baths, showers, and so on build up scale deposits. Calcium salts of many acids are insol­ uble, and the fatty acids used in soaps are of particular concern for consumer goods. The scum formed around the edges of baths is composed mainly of cal­ cium salts of fatty acids used in soap. Acids that form water‐soluble calcium salts are often added to cleaning products in order to ensure best performance of the products, since they are able to wash away the insoluble deposits. The most common acids used for either of these purposes are citric, acetic, sulfamic, and hydrochloric acids, as shown in Figure 11.1.

OH Acetic acid

HO

OH

HO

O

O

HO HO

O

Citric acid

NH2 S

O

O

O

Sulfamic acid

H

Cl

Hydrochloric acid

Figure 11.1  Acids used for prevention or removal of limescale.

­Bases in Consumer Good

In laundry applications such as detergents, the object of the product is to remove stains from cloth. These stains often contain fats and vegetable oils that are poorly water soluble and sometimes resistant even to detergent action. Proteinaceous stains, such as blood, can be even more difficult to remove. Laun­ dry detergents and other such fabric care products nowadays often contain enzymes. Enzymes are nature’s catalysts (see Chapter 12 for more detail), and the enzymes used in laundry products are acid and base catalysts, which are effective in hydrolysing fats and proteins. Those enzymes that are used to hydrolyse fats are known as lipases, and those that hydrolyse proteins are known as proteases or peptidases. In practice, lipases usually have some activity against proteins, and proteases against fats. However, they will also attack ester and amide bonds in fragrance molecules. Antiperspirants rely on aluminium or zirconium chlorides for their activity. Aluminium is the more common of the two and the less expen­ sive. Anhydrous aluminium chloride, AlCl3, is a powerful Lewis acid. The form used in antiperspirants is the hydrated form, which is known as activated alu­ minium chlorohydrate or AACH. Its empirical formula is AlCl3·nH2O where a typical value of n would be 1.5. The Lewis acidity of aluminium chloride is trans­ lated into Brønsted acidity in the aqueous conditions of an antiperspirant. This transformation reduces the pH in the product to about 3–3.5, which is quite acidic, so many perfume ingredients are found to be unstable in antiperspirants. Zirconium chloride is usually used in combination with aluminium chloride, and the amino acid glycine is often added. This combination of actives is known as  AZAG or activated zirconium aluminium glycine complex (ZrCl4 + AlCl3 + H2NCH2CO2H). It is even more aggressive towards fragrance ingredients as the glycine molecule (and its amino group in particular) provides chemical reactivity towards carbonyl functions, in addition to the acidity of the metal halides. Most consumer goods either contain water or will be used in water, so the acids that they contain will catalyse hydrolysis reactions of esters and can also catalyse aldol‐type reactions of aldehydes and ketones. All of these reactions will lead to loss of fragrance. More importantly, the loss will not be uniform across the various fragrance ingredients in the formula, so, in addition to loss of total impact, the odour impression will be distorted, and malodorous components can be generated. For example, hydrolysis of a fruity smelling ester could produce a carboxylic acid with a distinctly sweaty or rancid character.

­Bases in Consumer Goods As with acids, bases are added to consumer goods for pH control. The most com­ mon bases used for this purpose are sodium hydroxide, sodium bicarbonate (NaHCO3), sodium carbonate (Na2CO3), calcium carbonate (CaCO3), urea (also known as carbamide, CO(NH2)2, which is a weak base acting mostly as an acid scavenger), and alkanolamines. This last class of organic bases is, as the name suggests, alkyl alcohols that also contain amine functions. The most common is ethanolamine, HO(CH2)2NH2. They have the advantage of being soluble in aque­ ous systems and also some organic solvents. In addition, some enzymes also have basic activity and, like the acidic enzymes, are added because their lipase and/or

199

200

11  Chemistry in Consumer Goods

protease activity is useful in removal of stains from cloth. Carbonates and bicar­ bonates are used as deodorisers as well as for pH control, but, of course, they are still bases irrespective of the reason for their incorporation into the product. Bases will catalyse a similar range of reaction to those catalysed by acids and so can cause a similar loss of fragrance. One advantage of bases is that any malodor­ ous acids produced will be converted to their salts and so will not appear so strik­ ingly in the odour profile. However, base hydrolysis is often faster than acidic, and the formation of these salts will reduce the level of base in the product, therefore lowering the pH from that which the manufacturer intended.

­Nucleophiles in Consumer Goods As explained in Chapter 8, all bases are also nucleophiles, so all of the bases men­ tioned in the preceding paragraph are therefore capable of undergoing nucleo­ philic reactions with esters, aldehydes, ketones, and so on. The balance between basic and nucleophilic activity varies with conditions and also with the nature of the base/nucleophile. The carbonate anion is much more of a base than a nucleo­ phile, but, on the other hand, the alkanolamines are much better nucleophiles. Primary and secondary alkanolamines, such as ethanolamine or diethanolamine {HN(CH2CH2OH)2}, are capable of forming Schiff ’s bases or enamines with alde­ hydic or ketonic perfume ingredients and therefore represent a particular prob­ lem for the perfumer. Sodium bisulfite (NaHSO3) is used as a reducing agent, but it is also a nucleophile and will add to aldehydes and ketones (particularly to aldehydes) to form addition products. In principle, this reaction is reversible, but the overall effect is to reduce the amount of available aldehyde in a fragrance by formation of the relatively insoluble and non‐volatile bisulfite adduct, hence reducing the perfume impact. Another active ingredient is thioglycolic acid, which is used in permanent wave products as will be explained later in this chap­ ter. Its structure is HSCH2CO2H, and the reader will notice the presence of the thiol group, which is a strong nucleophile capable of attacking perfume ingredi­ ents containing aldehyde, ketone, or ester functions. Figure 11.2 shows some of the ways in which nucleophiles react with fragrance ingredients. The first is addition to the carbonyl carbon of an aldehyde. Initially Nu

OR′

Nu R

R

O

Nu Nu

OH

R

O

R

Nu Nu R

O

R

Figure 11.2  Reaction of nucleophiles with perfume ingredients.

O

O

­Oxidants in Consumer Good

an alkoxy anion is produced, but this anion picks up a proton from water or some other proton source to give the OH group as shown in the figure. It is through this reaction that bisulfite, for example, would add to an aldehyde. Thiols such as thioglycolic acid could also add in this way to form a hemithioacetal. When amines add in this manner, the addition is usually followed by elimination of water to form an imine (Schiff ’s base) or an enamine. With esters, nucleophiles can displace the alcohol part of the ester to produce an amide (if the nucleophile is an amine), a thioester (if the nucleophile is a thiol), or a ketone (if the nucleo­ phile is a carbanion). Unsaturated aldehydes, esters, and ketones in which the carbon–carbon double bond is conjugated to the carbonyl bond are susceptible to attack at the β‐carbon by nucleophiles. This type of addition (described in Chapter 8) is known as Michael addition, after the chemist who first described it. The carbonyl group of the product is still available for reaction with nucleo­ philes, and so a second addition can occur. (This capacity for double addition is important in biological chemistry as we will see in Chapter 12.)

­Oxidants in Consumer Goods Oxidants will react with perfume ingredients through the various routes described in Chapter 9. The earliest, simplest, and least expensive oxidant to be used in modern consumer goods is sodium hypochlorite, a powerful and fast‐act­ ing oxidant that is too strong for most applications. Its use is restricted to liquid bleaches and the more aggressive household cleaners. Sodium hypochlorite must be used in strongly alkaline solution in order to prevent formation of chlorine gas. It is very damaging to perfume ingredients, both through oxidising power and high alkalinity. It will attack aldehydes, ketones, esters, alcohols (except ter­ tiary), double bonds (olefins), and a variety of aromatic (benzene ring) com­ pounds. Hypochlorite is a particular problem with methyl ketones, as explained in Chapter 9, because of the chloroform reaction. Secondary alcohols are attacked faster than primary, but tertiary alcohols are relatively inert towards hypochlo­ rite, and so ingredients such as tetrahydrolinalool can be used in products con­ taining bleach (see Figure 11.3). A few fragrance ingredients contain sulfur atoms R R

R OH

R

O

R R

OH

R

O

R

OH

Figure 11.3  Alcohol stability to hypochlorite.

R

OH

201

202

11  Chemistry in Consumer Goods SO3–

O

Na2CO4

Na+ O

H2O

O O

OH

SNOBS Sodium nonanoyloxybenzenesulfonate

Figure 11.4  SNOBS/percarbonate bleach system.

in a low oxidation state, for example, as thiols, thioesters, and thioethers. Such materials are oxidised very rapidly indeed by hypochlorite and so will not be use­ ful for perfuming products containing hypochlorite. On a somewhat positive side, any malodours containing thiol groups (such as the key odour component of cat urine) are also easily destroyed by hypochlorite. Hydrogen peroxide (H2O2) is used, diluted with water, as an oxidant in some products such as hair bleaches and antibacterial cosmetic products. It is less aggressive than hypochlorite but is a liquid and relatively unstable, so for applica­ tions such as laundry powders, it is more usual to employ a precursor that releases a peroxy species during use. The first precursor of this family was sodium perbo­ rate (NaBO3), which releases hydrogen peroxide in aqueous solution. However, this solvolysis (cleavage by the solvent, water in this case) reaction requires rela­ tively high temperatures. The combined use of PERborate as a bleach and SILicate as a base led to one of the best known European laundry powder trademarks, Persil. It was then found that the temperature necessary to release an active oxi­ dant could be lowered by using a second component in the precursor system. The most common of these is tetraacetylethylenediamine (TAED). In Chapter 9 we saw how TAED and perborate react to release peracetic acid, which has oxida­ tive properties similar to those of hydrogen peroxide but is somewhat more reac­ tive. Percarbonate salts, e.g. sodium percarbonate (Na2CO4), can react in similar ways, and their use is growing at the expense of perborates. Also, the by‐product of percarbonate is carbonate that is ubiquitous in nature and therefore more environmentally friendly than borate, which is the corresponding by‐product from perborates. Sodium nonanoyloxybenzenesulfonate (SNOBS) is a bleach precursor, which is used with percarbonate bleach. As shown in Figure  11.4, SNOBS reacts with percarbonate in aqueous solution to generate pernonanoic acid, which is then the active bleaching agent, analogous to peracetic acid in the TAED/perborate system. The higher lipophilicity of pernonanoic acid compared with that of peracetic acid will make it even more effective at penetrating oily stains.

­Reductants in Consumer Goods When reductants are added to consumer goods, it is usually to prevent oxida­ tion, for example, by oxygen in the atmosphere. Examples include ascorbic acid, also known as vitamin C, and sodium bisulfite. Sodium hydrogen sulfite (also known as sodium bisulfite), NaHSO3, can be seen as the salt formed by partial

­Reductants in Consumer Good OH O

R

S R′

O– O–

R

S R′

O

OH O

O

Figure 11.5  Reaction of bisulfite with a carbonyl compound.

neutralisation of sulfurous acid, H2SO3. This acid is formed by reaction between sulfur dioxide and water. Sulfur dioxide and bisulfite are used as antibacterial agents to prevent spoilage by bacterial action. Butylated hydroxytoluene (BHT), butylated hydroxyanisole (BHA), and tocopherol (vitamin E) are used as antioxi­ dants to prevent autoxidation of products. Of these, tocopherol is becoming the most important because of its nature identical status and hence greater con­ sumer acceptability. Of all of these reductants and antioxidants, the only one that is really likely to interfere with fragrance ingredients is bisulfite, which will add to aldehydes and some ketones to form odourless adducts as shown in Figure 11.5. This reaction involves the formation of a carbon–sulfur bond using the lone pair of electrons on the sulfur atom of the bisulfite anion. The structures of the organic reductants are shown in Figure 11.6. Thioglycolic acid is used in permanent wave lotions. Its role is to reduce disulfide bridges in the protein structure of hair. As we will see in Chapter 12, disulfide bridges are used by nature to hold protein molecules in a given shape. The disulfide bridge is formed by oxidising the sulfhydryl groups of two cysteine units, cysteine being one of the amino acids used to build proteins. When two cysteines are coupled in this way, they are known as cystine, and clearly the affected parts of the protein chain will be held in proximity in the molecule. Since such disulfide bridges are involved in holding hair fibres in a given shape, the idea behind the use of thioglycolic acid in permanent wave lotions is to break the bridges, then move the hair to a different shape, and subsequently form new H

OH

HO

O

HO

OH

O

H HO

OH

O

Ascorbic acid (vitamin C)

BHT

O HO

Tocopherol (vitamin E)

Figure 11.6  Reductants and antioxidants in consumer goods.

BHA

203

204

11  Chemistry in Consumer Goods R

O

H N

H2N

N H

R

O

R

R

O

R H2N O

O

R N H

R R

O

H N O

R

O

R

N H SH

O

OH

S S

R N H

R N H

H N O

R

O

S N H

O

OH +

R

O

O

H N O

SH H N

S

O

H N

H2N

N H

R

O

H N

O

H N

H2N

H N

O

OH

HS

OH R

O

R

O

H N

HS OH

O OH

+

O

OH

R

O

O

N H

H N O

H N

R

O

O OH

R

Figure 11.7  Making and breaking disulfide bridges using thioglycolic acid.

disulfide bridges to retain that new shape. The overall process is shown in Figure 11.7. Thioglycolic acid presents a serious problem for perfuming products that contain it. The thiol function will react with aldehydes and ketones and also can inter‐esterify with esters to form thioesters; in other words, ester plus thiol gives thioester plus alcohol. Moreover, thioglycolic acid is not the most stable of molecules and breaks down to release other divalent sulfur compounds. These by‐products can appear in many different forms, most of them highly malodor­ ous. The perfumer’s difficulties are therefore twofold. First, the range of perfume ingredients that can be used is limited, and second, the product base malodour is very difficult to cover.

­Surfactants in Consumer Goods Surface active agents are added to consumer goods for a variety of purposes. The most obvious is for washing, either in personal applications such as soap, sham­ poo, or bath gel or in laundry detergent, dishwashing detergent, or hard surface cleaners. Other applications include stabilisation of emulsions, foam control, and conditioning (of skin or fabrics). In general, surface‐active agents are not aggressive towards fragrances, but they can increase or decrease exposure of perfume components to less perfume‐friendly species. For example, in a hard surface cleaner, the perfume might exist as droplets dispersed in the product as

­Chelating Agents in Consumer Good

an emulsion. By solubilising the perfume in the aqueous phase, surfactants could increase the exposure of perfume ingredients to conditions of high pH (neces­ sary because of the hypochlorite bleach content) and thus lead to hydrolysis of ester‐containing ingredients. The chief exception to the general rule is soap, which is alkaline in nature and therefore will attack base‐sensitive perfume ingredients such as aldehydes and esters. The chemistry of surfactants is described in Chapter  4, and so it may be referred to as needed. The main surfactants used for detergency are soaps and anionic and non‐ionic surfactants. Cocamidopropyl betaine and alkyldiethanol­ amides are used for foam control. Glycerol monostearate and alkylethoxylates are used as emulsifiers.

­Chelating Agents in Consumer Goods Metal ions do not exist as bare cations in solution. They attract other species, usually either anions or solvent molecules, which then bind to the cation. This binding is known as coordination, and the coordination number of a metal ion is the number of points on the molecule at which such attachment will occur. For example, iron has a coordination number of 6, so that when ferric chloride (FeCl3) is dissolved in water, each iron cation will have three chloride ions and three water molecules bound to it. The binding is not as strong as, for example, that in a carbon–carbon bond, so the coordinated species will be able to disso­ ciate and exchange with other species in the solution. This quality is an impor­ tant part of the reactivity of the metal ion. Iron is a good example since it is found in all sorts of products. Perfumed consumer goods will more often than not contain traces of iron from either the container or from the machinery used in production. Iron III (ferric iron) is an oxidant, and iron II (ferrous iron) is a reductant. Traces of iron will therefore cause a variety of problems in consumer goods. Chelating agents derive their name from the Greek word chele, meaning ‘a claw’, since they have more than one point at which they can attach to a metal cation and therefore can form a strong multipoint grip around it. If the chelating agent can block most or all of the coordination points of the metal, it will sup­ press the chemistry of the metal. Thus, consumer goods manufacturers may add a chelating agent to their product to eliminate unwanted effects of trace metals such as iron. The two most common chelating agents used in this way are citric acid and ethylenediaminetetraacetic acid (EDTA), and their structures are shown in Figure  11.8. EDTA is usually added as a sodium salt. It can coordinate to a metal by each of the carboxylic acid groups and by both of the nitrogen atoms. One EDTA molecule can therefore fill all six coordination sites of iron and com­ pletely hide the iron ion in doing so. In principle, citric acid could bind via its three acid groups and its alcohol; though once the three acids have coordinated to a metal ion, the shape of the molecule prevents the alcohol from binding to the same ion. Far from posing a threat to perfume molecules, chelating agents help protect them from reactive metal ions.

205

206

11  Chemistry in Consumer Goods O O

OH Citric acid

O

HO

OH

HO

O EDTA

O HO

OH N

OH

N

HO

Ethylenediaminetetraacetic acid

O O

Figure 11.8  Chelating agents.

­Photoactive Agents in Consumer Goods Products that are sold in glass or clear plastic bottles will be exposed to visible and ultraviolet light during storage and use. These frequencies of electromag­ netic radiation, and UV in particular, can induce chemical changes in organic molecules. To prevent this problem, dyes are often added that absorb UV light and dissipate its energy harmlessly as heat. Dyes or pigments are sometimes added to impart a colour to the product. The two main classes of dyes are the azo dyes and the triphenylmethane dyes. The structures of the core of these dyes are shown in Figure  11.9. In both cases, a variety of substituents are placed around the rings of the molecules. The substituents will serve one of two purposes: either they will serve as groups to bind to surfaces and thus ‘fix’ the dye onto a substrate (more impor­ tant when used for dyeing fabrics than cosmetics and cleaning products) or will fine‐tune the frequencies of light absorbed by the dye and hence the col­ our it will appear.

N

N

Azo dyes

C+ Triphenylmethane dyes

SO3– SO3–

Figure 11.9  Photoactive ingredients in consumer goods.

Whitener 10

­Antibacterial Agents in Consumer Good

Another class of photoactives is the opacifiers, which, as their name suggests, render an otherwise transparent product opaque. These ingredients are usually inorganic materials such as silica or titanium dioxide that absorb all frequen­ cies of visible light. Related to the opacifiers are the pearlisers. These materials absorb light but also reflect it from one crystal face. For example, a solid fatty acid with a flat crystal shape can give an opalescent effect in a shampoo, by reflecting light from one crystal after another as the liquid is moved around in the bottle. Optical brighteners are materials that fluoresce, that is, they absorb light of one frequency and re‐emit the energy as light of a different, longer frequency. Brighteners that absorb UV light and emit blue light are added to laundry prod­ ucts. The reason is that the brightener will absorb onto the cloth where, in day­ light, it will emit blue light. This effect counteracts the yellowing effect of time on white cloth and will keep it looking bright and white for longer. Two possible complications for fragrance molecules arise from the use of these photoactive materials. The simplest one is loss of available fragrance because of adsorption onto the surface of opacifier particles or into the crystals of pearlis­ ers. The other danger comes from the dyes. Normally the dye molecule, having received energy from the light, will simply lose it as heat, but it is possible that, in the presence of oxygen, the excited state of the dye molecule could transfer energy to another molecule (such as a perfume ingredient) and trigger what is known as dye‐sensitised photo‐oxidation. However, neither of these problems is anywhere near as serious as some of the other chemical attacks suffered by per­ fumes in products.

­Antibacterial Agents in Consumer Goods Antibacterial agents are sometimes added to consumer goods to act as preserva­ tives by preventing bacterial degradation. (It must be remembered that soaps, oils, creams, and so on will constitute food for bacteria.) Sometimes antibacte­ rial agents are added for the direct benefit of the consumer, for example, as con­ tributors to the hygiene value of soaps. Figure 11.10 shows the structures of two of the most common antibacterial agents (triclosan, also known as Irgasan, and Cl

Cl

Triclosan (Irgasan)

O Cl

Cl

Cl

O N H

N H

Cl

Figure 11.10  Antibacterial agents.

Triclocarban

207

208

11  Chemistry in Consumer Goods

triclocarban) added to consumer products. In soap, triclocarban can break down to release chlorinated anilines, and these, being primary amines, will then react with any fragrance ingredients containing carbonyl groups.

­Other Reactive Ingredients in Consumer Goods Other materials added to products include solvents (such as alcohols, glycols, and glycol ethers), emollients, vitamins (e.g. B vitamins and panthenol), abra­ sives, fillers, thickeners, hydrotropes, and salts. Generally, the main effect of these materials on perfume ingredients is the loss of available perfume onto or into the other active ingredient. This effect will result in a loss of intensity of the perfume in the product per se, but in many cases, the ingredient will be released in use, giving a burst of fragrance. For example, perfume could be absorbed onto the surface of silica particles added as an abrasive to a hard surface cleaner. However, when used, the mixture will be exposed to a surplus of water that will displace the perfume from the silica surface. Calcite, which is also used as an abrasive, is similarly capable of adsorbing perfume molecules. In fact, an effect similar to this example is used to protect perfume. An encap­ sulating agent – usually starch but other polymers are also used – can be added to the perfume, so that its molecules become trapped in the polymer matrix. The encapsulated perfume is added in powdered form to the product, and the per­ fume molecules are protected from the rest of the product formulation by the polymer matrix. In use, the polymer dissolves in water and releases the perfume. Conversely, encapsulation of other active ingredients could also serve to protect the perfume by keeping the two apart. Alkylethoxylates and sodium chloride are used as thickeners. These materials present little difficulty for perfumes, unlike sodium carbonate and bicarbonate, which are used as deodorisers and can damage fragrance ingredients because of their basicity. One other active ingredient that is damaging to perfume is dihydroxyacetone (DHA), which is used as a sunless tanning agent. Figure 11.11 shows how DHA can enolise and then reketonise as dihydroxypropionaldehyde. All of these tau­ tomers are reactive molecules and can react with aldehydes, ketones, esters, and amines in perfume to give the range of products shown in Figure 11.11. O HO

Aldehydes Ketones

Acetals, ketals

OH OH

HO

Aldehydes Ketones Esters Aldol products

OH OH

Esters

Transesterification

Figure 11.11  Reactions of DHA with perfume ingredients.

HO

Amines

Schiff’s bases

O

­Fine Fragranc

­Types of Consumer Goods Different fragrance houses and consumer goods companies have different ways of categorising the range of products into which perfume is incorporated. Here, we will classify them into five categories – fine fragrance, cosmetics and toilet­ ries, personal wash, laundry, and household.

­Fine Fragrance To most people outside the industry, the words perfume and fragrance mean fine fragrance, that is, perfume for application to the skin of the user. Apart from liquid air fresheners, the level of perfume oil in these products is the highest of all of the perfumed products described here. The major component of a fine fragrance is the aqueous alcohol used as solvent. Figure 11.12 shows the various terms used to describe fine fragrances, with different levels of oil in the product and of alcohol in the solvent. Fragrance oils vary from colour­ less to pale yellow. The colour of the natural ingredients, such as essential oils, will vary over time, and this variance will affect the colour of the final product. If the fashion house, or other marketing company, wish to do so, they will add a dye to give a consistent and distinctive colour to the final product. Sometimes other additives are used, e.g. astringents in aftershave, but the major threats to the fragrance ingredients come from the alcohol, air, and light. The ethanol used as solvent can transesterify with esters in the fragrance. It can also oxi­ dise to give acetaldehyde, which can then undergo aldol reactions with car­ bonyl‐containing perfume ingredients. All of these reactions can result in changes in odour over time. Air can oxidise perfume ingredients, which can be accelerated by light through the process of photo‐oxidation. Of course, all of these reactions will speed up if the temperature is increased. The best advice to consumers is therefore to keep perfume in the refrigerator where it is pro­ tected from heat and light and to screw lids on tightly so as to minimise expo­ sure to air.

% Oil in product

% Alcohol in solvent

extrait

~20

>90

eau de parfum

~12

80–90

eau de toilette

~8

60–70

eau de cologne

~4

50–60

Aftershave

~2

40–60

Splash cologne

~2

40–60

Figure 11.12  Composition of fine fragrance.

209

210

11  Chemistry in Consumer Goods

­Cosmetics and Toiletries The many products in this category fall into two classes as far as risk to perfume is concerned. In products such as hair conditioners, deodorants, hair gels, hair colorants, moisturising creams, and talcum powder, the main risk to perfume is through retention in the substrate and hence reduced impact to the user. The perfume is not actually damaged in such instances, but it is not immediately perceptible. Consumers often choose products on the basis of its odour when they open the cap, before purchase. Consequently, a reduction of odour impact from the bottle could lead to a reduction in sales. In some cases, the odour will be lost throughout the use cycle of the product. Talcum powder is a good exam­ ple of this latter phenomenon since perfume molecules seem to have a consider­ able propensity for being trapped in the layered structure of the talc crystals. On the other hand, antiperspirants, with their acidic nature, and permanent wave products with thioglycolic acid represent much more serious challenges to perfume stability. The chemistry of these products has been discussed in this chapter and in earlier chapters and so will not be repeated here. Figure 11.13 shows a table of likely contents of these various products and the nature of the problems they cause to perfumes.

­Personal Wash In the personal wash category, soaps and bath crystals both have moderately alkaline pH levels, and therefore base‐sensitive perfume ingredients such as esters are at risk. On the other hand, some perfume ingredients can have an adverse effect on the product. Indole and vanillin both cause discoloration in Product

Actives

Conditioner

Quats

Deodorant

Fragrance

Antiperspirant

Aluminium chloride and/or zirconium chloride

Hair colorant

Dyes

Hair gel

Polymers

Permanent wave

Thioglycolic acid

Nucleophilic attack, Sulfurous malodours

Talcum powder

Talc

Loss of perfume into talc

Moisturising cream

Oils

Figure 11.13  Cosmetics and toiletries.

Problems for perfume

Acid-catalysed reactions

­Laundr Product

Actives

Problems for perfume

Hard soap

Soap or NSD

Alkaline

Liquid soap

NSD

Shower gel

NSD

Bath gel

NSD

Shampoo

NSD

Bath crystals

Sodium carbonate

Alkaline

Figure 11.14  Personal wash.

soap. This staining happens remarkably quickly, turning the soap dark brown. Indole is an important component of jasmine and vanillin of vanilla, both desir­ able notes in soap perfume, and so perfumers and fragrance chemists have had to work hard to find alternative molecules and formulations that give the same effect but without the colour problems. In all of the other personal wash prod­ ucts, the detergency stems from neutral surfactants, so these products are all mild as far as perfume is concerned. Figure 11.14 shows this situation in tabular form. Some bar soaps use non‐soap detergents (NSDs) such as distilled ethylene fatty isethionate (DEFI) and therefore are also kind to fragrance. Such ‘soaps’ usually have an advertising platform that emphasises their mildness to the skin as well as perfume. The levels of perfume in all of these products will be in the 1.0–1.5% range.

­Laundry In some countries, hard bar ‘soaps’ are popular for washing of clothes. The active ingredient in such bars is an NSD, so the NSD bars are relatively kind to fragrance. This situation is not true for the remainder of products in this cate­ gory. As can be seen from Figure 11.15, these products are either acidic or basic to levels that restrict the perfumers’ palette. Laundry powders have pH levels above 9, and many of them also contain bleach systems and enzymes. Fragrance ingredients are therefore under threat from alkaline hydrolysis, aldol chemistry, and oxidation. The object of the fragrance is to leave a smell on the cloth that communicates ‘fresh and clean’ and such messages to the consumer. However, the product is designed to remove organic materials from the fabric, so the perfume has to fight against the product containing it. In order to achieve dep­ osition onto cloth in the wash, it is best if perfume ingredients have a high log P – in other words, that they are poorly soluble in water. Unfortunately, it is this very property that makes it difficult to decompose them in a sewage treat­ ment plant. Hence perfumes for laundry use are at the forefront of the search for improved biodegradability. The level of perfume in a laundry powder is in the low range 0.3–0.5%, so whatever perfume is used must perform effectively.

211

212

11  Chemistry in Consumer Goods Product

Actives

Problems for perfume

NSD bar

NSD

Laundry powder

Soap, NSD, enzymes?

pH >9

Laundry powder with bleach

Soap, NSD, oxygen bleach, enzymes?

pH >9, oxidising

Fabric conditioner

Quats

pH 3

Figure 11.15  Fabrics.

Fabric conditioners are at the other end of the pH scale with values around 3, well into the acid region. The primary aim of the product is to condition the fabrics by depositing quats onto the surface. However, perfume deposition at this stage is also important since the product is used at the last stage of the laundry process. Obviously, the perfumer must choose acid‐stable ingredients, and this limitation leaves some gaps in his palette, for instance, in the muguet (lily of the valley) odour area that contains relatively few acid‐stable perfume materials. Laundry powders also contain builders. These complexing agents trap cal­ cium salts and prevent scum from attaching to the fabric being washed. Similarly, anti‐redeposition agents prevent micelles containing soil material from return­ ing to the fabric and adhering to it. The builders can be inorganic materials, such as phosphates or water‐soluble acrylate polymers that bind calcium. The anti‐­ redeposition agents are water‐soluble polymers. Neither ingredient poses any serious problems for fragrance.

­Household Household products cover a very wide range, and a few of the key examples are shown in Figure  11.16. As in the other categories, some household products, such as hand dishwashing liquid, furniture polish, and rinse aid for machine dishwashing, place relatively little stress on the perfumes in them. In air freshen­ ers, the perfume is the active product, and delivery is the main concern. A quick inspection of the appropriate shelves in a supermarket will reveal the ingenious array of devices for delivery of air freshening fragrances. Some leave the fra­ grance at room temperature, but others involve electrical heating to help vapor­ ise the perfume; this second type is known as an electrically heated liquid air freshener or ELAF for short. ELAFs require that the perfume ingredients are thermally stable up to the temperature achieved in the device. Scented candles are another way of perfuming the air. Since the medium is a simple wax or hydro­ carbon, there is no danger to the perfume from the product, but the perfume can seriously affect the way the candle burns, and so the perfumer must choose

­Househol Product

Actives

Problems for perfume

General-purpose cleaner

NSD, bleach, abrasives

Oxidation, adsorption

Air freshener

Fragrance

Hand dishwashing liquid

NSD

Machine dishwashing

Silicates, bleach, enzymes?

Strong alkali, bleach

Lavatory cleaners

Acid or base, bleach

Acid/base reactions, oxidation

Rinse aid

Non-ionic detergent

Furniture polish

Waxes

Bleach

Sodium hypochlorite

Strong alkali, oxidation

Figure 11.16  Household.

ingredients carefully for this application. General‐­purpose cleaners or hard sur­ face cleaners contain detergents and usually bleach, probably hypo­ chlorite. Often an abrasive such as silica or calcite is included, so the risks to perfume are those of oxidation, alkali, and adsorption. The fragrance concentration is about 0.3% for a trigger cleaner (i.e. one delivered from the container by means of a trigger device) and 0.01% for a diluted general purpose cleaner (GPC). Machine dishwashing powders and gels are among the most aggressive of con­ sumer goods as far as fragrance is concerned. They contain low foaming non‐ ionic detergents, but much of the cleaning power comes from sodium metasilicate, a powerful base. The first bleach precursors released chlorine bleaches during the wash cycle. More recently, perborates and percarbonates have replaced these precursors, and some products also contain manganese catalysts, which bleach using oxygen from the air. Since many users neglect to recharge the machine’s built‐in water softener with salt, some dishwashing products also contain build­ ers to ensure freedom from calcium deposits. Enzymes such as proteases and lipases are also used, and some products have glass protection via polymers and/ or zinc salts. Lavatory cleaners vary from one country to another. In the United Kingdom, they contain hypochlorite bleach and are therefore strongly oxidising and alkaline. In other European countries, they contain acids designed to remove limescale. These acids include citric, acetic, poly‐phosphoric, sulfamic, and even hydro­ chloric. Clearly, perfume ingredients must be selected accordingly with the appropriate pH and oxidative stability. Liquid bleach is basically a solution of sodium hypochlorite in sodium hydrox­ ide solution. Some bleaches contain polymers to make them stick to surfaces. In

213

214

11  Chemistry in Consumer Goods

any case, the problems for fragrances added to them are those of stability in alka­ line conditions and to oxidation.

Review Questions 1 A customer complains that your fragrance has turned his white soap choco­ late brown in colour. What would you do? 2 A customer complains that your perfume has turned his clear shower gel a bright blue colour. What would you do? 3 Which of the ingredients shown in Figure 11.17 would be best suited for use in an antiperspirant? 4 Which of the ingredients shown in Figure 11.18 would be best suited for use in a fragrance for a machine dishwashing powder? OH OH

OH

OH

Linalool

Mefrosol®

Hydroxycitronellol

Figure 11.17  Stable in antiperspirants.

O O O Citronellal

Terpinyl acetate

O

N Frescile®

Anther®

Figure 11.18  Stable in machine dishwashing powder.

215

12 The Chemistry of Living Organisms This chapter introduces the chemistry of living organisms. An understanding of this discipline is important for comprehension of the structural patterns of natural fragrance ingredients, the basics of toxicity (hence safety of perfumes), and the working of the sense of smell. Perfume safety will be covered in this chapter, but more detailed accounts of the sense of smell and the way plants build fragrant molecules will be found in Chapters 13 and 14, respectively.

­Molecular Recognition So far, in terms of chemical bonding, we have discussed only the bonds that hold atoms together in molecules. These bonds are strong and require enormous physical force to pull them apart. However, molecules associate with each other through weaker forces, known as non‐bonded interactions. These interactions can be van der Waals forces, electrostatic forces, hydrogen bonds, hydrophobic bonding, and so on  –  all of which will be explained below. Each individual interaction is weak, making it relatively easy to break and thus separate the molecules. However, if the molecules are large and have many of these weak bonds between them, then the overall result is a strong association between the molecules. The effect is rather like that of a zipper with many weak forces adding together to make a strong one. Also as with a zipper, the bonds can be broken one at a time to separate the molecules and then remade again. It is forces such as these that actually hold our bodies together. If two molecules have structures that hold the components of such bonds in the right geometrical relationship to make the formation of multiple bonds between them easy, then they are said to recognise each other. The phenomenon is known as molecular recognition and is very important in biological chemistry. Of particular relevance to our industry are chemoreception and the immune response, both of which involve molecular recognition. Atoms and molecules exhibit long‐range attraction for each other, but this attraction turns into repulsion at short range. This longer range attraction is known as van der Waals attraction. Charged species attract those of opposite charge, and this force is known as electrostatic attraction. Partial charges work in

Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

216

12  The Chemistry of Living Organisms δ− O

H H

H

δ+

H

Figure 12.1  Polarisation of the O─H bond in methanol.

the same way and give rise to the most important of these intermolecular forces, the hydrogen bond. Figure 12.1 shows the polarisation of the O─H bond in methanol. The oxygen atom has eight protons in its nucleus, whereas hydrogen has only one. The oxygen nucleus therefore pulls the electrons of the O─H bond towards itself and away from the hydrogen nucleus. This event results in a partial negative charge over the oxygen and a corresponding positive charge over the hydrogen. In Figure  12.2, we then see how methanol molecules will ‘stick’ to each other through electrostatic bonds between the alcoholic hydrogen atom of one molecule and the oxygen of another. The same phenomenon is observed with sulfur and nitrogen (and indeed any other electronegative element) and hydrogen. Since the common feature is the electropositive hydrogen atom, this type of bonding is known as hydrogen bonding. Hydrogen bonding in alcohols, such as methanol, is the cause of their having higher boiling points than the corresponding aldehydes that have almost the same molecular weights. The hydrogen bonding between the alcohol molecules makes it harder to pull one free from the liquid and move it into the gas phase. This difficulty does not happen with the aldehyde, and so it will require less energy (i.e. a lower temperature) for the transformation from liquid to vapour. Hydrogen bonding in water also accounts for its very special physical properties. Most substances contract steadily as the temperature drops. Water contracts as the temperature drops to 4 °C but then expands again until its freezing point (0 °C) is reached; at which point, it begins to contract again. The explanation lies in hydrogen bonding. In liquid water, the hydrogen bonds between water molecules are disorganised, whereas in ice, they are organised in the crystal structure. Between 4 and 0 °C, this organisation is beginning to take place, and this phenomenon causes the water to expand as the individual molecules move into the places they will occupy in the crystal. The fact that ice at 0 °C is less dense than water at the same temperature means that ice floats on water, and this phenomenon makes aquatic life possible.

H

O

H

H

H H H

O H

H

H O H

H

H

Figure 12.2  Hydrogen bonding in methanol.

  ­Molecular Recognitio H H

O

H

H

Figure 12.3  Hydrogen bonding between methanol and benzene.

Hydrogen bonding can also happen with electron‐rich centres, such as the benzene ring, as shown in Figure 12.3. Hydrogen bonds to benzene rings are weaker than those to oxygen or nitrogen atoms. Figure 12.4 shows hydrogen bonding between two acetic acid molecules. The acidic hydrogen of each forms a hydrogen bond to the carbonyl oxygen of the other because the relative positions of the various atoms and interatomic distances favour such an interaction. In this simple example of molecular recognition, the two molecules ‘recognise’ each other. In water, acetic acid forms hydrogen bonds with the water molecules around it, but, in a non‐polar solvent, such as benzene, the acetic acid molecules will pair up as shown in Figure 12.4. Indeed, the measured molecular weight of acetic acid appears to double in such solvents relative to that in water. Chirality (see Chapter 2 for details) can radically affect molecular recognition, and so optical purity can be important in fragrance chemistry. To illustrate chiral recognition, first try shaking hands with someone, with both of you using the right hand. Now try again but with one using the right hand and one the left. The second is much more difficult because during the first, the shapes of the two right hands complement each other when brought together. Similarly, when a small molecule approaches a binding pocket in a larger molecule, if the shape of the small molecule, including its chirality, does not complement the shape and chirality of the pocket, it will not be recognised by the larger molecule. Another force that is important in recognition of fragrance molecules is known as hydrophobic bonding. Imagine the light grey shape at the top left of Figure 12.5 O O

H H

O O

Figure 12.4  Hydrogen bonding in acetic acid.

Figure 12.5  Hydrophobic bonding.

217

218

12  The Chemistry of Living Organisms

to be a large protein molecule sitting in an aqueous environment. Since it is surrounded by water molecules, the protein will adopt a shape such that its outer surface is hydrophilic. The small black shapes represent water molecules adhering to the outer surface of the protein. To the right of the protein molecule is a pocket, and, for the sake of argument, we will say that this pocket is hydrophobic. Since the system is at atmospheric pressure, no vacuum will exist in the pocket, but, instead, water molecules will be forced into it as shown. This energetically unfavourable situation will cause the water molecules and the hydrophobic surface of the pocket to repel each other. If we now introduce a small hydrophobic molecule (shown as darker grey) that could fit into the binding pocket and it does so, as shown in the lower right part of the figure, then the total area of hydrophobic surface now exposed to water is significantly reduced. An energy gain results and therefore becomes a force holding the two species together. This force is known as hydrophobic bonding. The complementarity of shape between the small molecule and the binding pocket is important in determining the strength of the hydrophobic bond, and consequently shape is one of the important factors in determining the odour of molecules as will be seen in Chapter 13.

­Classes of Natural Chemicals Plants and animals, like everything else in the universe, are made up of chemicals. The chemicals of which living organisms are comprised can be divided into two main classes, namely, primary metabolites and secondary metabolites. Primary metabolites are defined as those that are present in all living organisms and are essential for the basic processes of life. On the other hand, secondary metabolites are each likely to be present in only a limited range of species and, while important to those species in which they are present, are not part of the key processes of life. In this chapter, we will look at the primary metabolites, and the secondary metabolites will be covered in Chapter 14. The four classes of primary metabolites are carbohydrates, nucleic acids, lipids, and proteins. The molecules of life Primary metabolites Present in all living organisms Essential for the basic processes of life Carbohydrates Proteins Nucleic acids Lipids

Secondary metabolites Present in limited range of species Important to those species but not part of key life processes Terpenoids Shikimates Polyketides Alkaloids

­Carbohydrates The term carbohydrate is a contraction of carbon hydrate since materials of this class are compounds containing carbon, hydrogen, and oxygen, in which the ratio of hydrogen to oxygen is 2 : 1. They are also known as saccharides, from the

 ­Carbohydrate

Greek word for ‘sweet’. Monosaccharides contain three, five, or six carbons. One of the oxygen atoms is present as an aldehyde or ketone, and the others as alcohols. Monosaccharides can join together to form oligosaccharides (i.e. those containing a few saccharide units) or polysaccharides (i.e. those containing many saccharide units). Mono‐ and oligosaccharides are usually known as sugars, and polysaccharides as starches. Table sugar, sucrose, is a disaccharide that contains one glucose and one fructose unit. Saccharides are used as food sources. Sugars are burnt in plants and animals to provide energy. Starches are used as a way of storing energy and are broken down to sugars when required. Polysaccharides are also used structurally. For example, cellulose (a polymer of glucose) is a key component in the fibrous structure of plants. Saccharides are also used for recognition. For instance, undesired alcohols in the body can be esterified with glucuronic acid (the oxidation product of glucose). These glucuronates will be recognised as ‘foreign’ by the kidneys, filtered out of the bloodstream, and excreted in the urine. There are a number of conventional ways of showing the structures of sugars as can be seen from Figure 12.6. Here, we see four representations of α‐d‐glucose and three of d‐fructose. In each name, the prefix d‐ refers to the direction of rotation of plane polarised light. In the case of glucose, the prefix α‐ refers to the orientation of the OH group of the hemiacetal function of the cyclic structures. Reading from the left in the upper box, the first drawing of glucose shows the linear form. From this example, it is clear that glucose contains an aldehyde function, and it is therefore known as an aldose. It contains six carbon atoms and so is also known as a hexose, and, combining the two, it becomes an aldohexose. The straight chain form does not exist other than transiently, so it is common nowadays to use the cyclic representations of the structure. The two central structures use different ways of showing the absolute stereochemistry of each of the asymmetric carbon atoms in the molecule, and the structure on the right gives an impression of the actual shape of the molecule. Fructose is shown similarly in the lower box. Like glucose, it has six carbons, but, unlike glucose, its carbonyl group is a ketone. Fructose is therefore classified as a ketohexose. O HO HO OH

α-d-glucose HO

HO

O

HO

OH

OH

H H OH OH H

OH OH

H

H HO

O H

HO

H

H

OH OH

O H OH

OH H

OH HO O OH d-fructose

HO OH

O

OH OH

HO HO

OH

OH

Figure 12.6  Representations of glucose and fructose.

HO

OH

O H

H

OH

OH

H

OH

OH

219

220

12  The Chemistry of Living Organisms HO

HO O H

H OH

H OH

H

H

OH

+

HO

H

R

O H

OH

OH

Glucose

H OH

H

H

OH

Alcohol

+ O

H2O

R

Glycoside

Figure 12.7  Formation of a glycoside.

The aldehyde function of glucose exists as a hemiacetal, and, in the presence of an alcohol, this can be converted into an acetal as shown in Figure 12.7. Acetals formed from glucose and an alcohol are known as glycosides. Many fragrant alcohols and aldehydes are present in nature as glycosides, and the fragrant species is only detected when the acetal function is hydrolysed. An example is benzaldehyde, which occurs almost invariably as the glycoside of its cyanohydrin, a compound known as amygdalin. A cyanohydrin is the product of addition of hydrogen cyanide (HCN) to an aldehyde or ketone to give an α‐cyano‐alcohol. In the case of amygdalin, benzaldehyde cyanohydrin forms a glycoside with one molecule of glucose, and this structure then forms a glycoside with a second molecule of glucose. The structure is shown in Figure 12.8, in which the abbreviation Glu is used to represent glucose. The figure also shows the release of benzaldehyde and HCN on hydrolysis of amygdalin. Many explanations have been proposed to account for the similarity in odour between benzaldehyde and HCN. The author believes that the similarity in perception is due to the fact that the two molecules almost invariably occur together in nature (from hydrolysis of amygdalin) and so both have been learned as having the distinctive smell of almonds, the most common source of amygdalin. Other sources of amygdalin include cherries, peach kernels, and cassava. The structure of amygdalin shows that the alcohol groups of one sugar can be glycosylated in the same way as any other alcohol. Another simple example is the formation of sucrose from glucose and fructose as shown in Figure 12.9. Glucose and fructose are known as monosaccharides and sucrose as a disaccharide since it contains two monosaccharide units linked through an acetal junction. Clearly, N N

Hydrogen cyanide

H

O

Excess H2O

Glu Glu

H+

+ + O

Amygdalin Benzaldehyde

Figure 12.8  Hydrolysis of amygdalin.

2 glucose

  ­Nucleic Acid HO

OH

H

O H

H OH OH H

H

+

OH OH

Glucose

OH

O H OH OH

OH H H

OH

HO O H

H OH

H OH

H

H

OH

Fructose

OH

O O

H

OH H

OH

H

+

H2O

Sucrose

Figure 12.9  Formation of sucrose from glucose and fructose.

more monosaccharide units can be added, and this process forms a polysaccharide. Starches and cellulose are examples of polysaccharides that play important roles in nature. Cellulose consists of chains of polyglucose held together in sheets by hydrogen bonding between the hydroxy groups of neighbouring chains. Starches are polymers containing long polysaccharide chains that can be either straight or branched. They tend to tangle together in solution, and a section of one chain can recognise a section of another, thus tying them together into a three‐dimensional network. This property leads them to form gels, a property that is used in jams and in thickening agents for household products. Certain starch structures have a tendency to coil into helices. The space through the centre of the helix is just the right size to accommodate molecules of iodine (I2), and, in this environment, the iodine molecules take on a deep blue colour, the basis for the starch/iodide indicator system for red‐ox titrations as described in Chapter 6. This space in the centre of starches can also trap small organic compounds, which is particularly true with a family of starches known as cyclodextrins. Cyclodextrins are used commercially in deodorising products because of this ability to trap small malodorous molecules and hence remove them from the air.

­Nucleic Acids Nucleic acids are the means by which information is carried from one generation to another in plants and animals. The four basic components are the organic bases adenine, thymine, guanine, and cytosine, as shown in Figure 12.10. These components can be seen as a four letter genetic alphabet comprising the letters A, T, G, and C respectively. The four bases are known as nucleotides. The nucleotides recognise each other in pairs as shown in Figure 12.11. Adenine and thymine recognise each other through formation of two hydrogen bonds; and guanine and cytosine recognise each other through three hydrogen bonds. In order to hold the ‘letters’ together as words and eventually sentences, the bases are linked to the sugars ribose or deoxyribose through aminal bonds (like acetal or ketal formation but with one N atom and one O atom in place of two O atoms) to give compounds known as nucleosides. The sugar groups of the nucleosides are then linked together by phosphate esters to produce a polymer with a built‐in sequence of the four letters. Two such molecules containing complementary sequences of bases then link together through the base pair recognition, and the whole molecule forms into a double helix shape. If the sugar is ribose, the polymer

221

222

12  The Chemistry of Living Organisms NH2 N

O

N

A N H

N

N H

NH2

O N

N

NH

N H

O

Thymine

Adenine

G

T

NH

N

N H

O

NH2

Guanine

C

Cytosine

Figure 12.10  The four bases of nucleic acids.

H Adenine

N

N N H

H N

O

H

Thymine H

N

N O

N

H N

H Guanine

O

N N H

N N

NH

N H

N

O

Cytosine

H

H

Figure 12.11  Recognition between nucleotides.

is known as ribose nucleic acid (RNA), and if it is deoxyribose, the polymer is deoxyribose nucleic acid (DNA). The letters (C, A, G, T) are read as three‐letter words, each coding for a specific amino acid and hence the building instructions for protein synthesis. If the two strands of a nucleic acid are separated, molecular recognition can be used to produce new complementary chains for each, and, of course, each new chain will be identical to the original partner. Hence the genetic information can be replicated and passed on to future generations.

 ­Lipid

­Lipids The structures of the lipids are built around the fatty acids. Fatty acids are built up in nature from acetate units. A simple schematic is shown in Figure 12.12, and the synthesis will be discussed in more detail in Chapter 14. In essence, one acetate unit is added to another by an aldol type reaction, and the ketone group of the resulting β‐keto acid is subsequently reduced to give butyric acid. Repetition of the process adds another two carbon atoms and so on. Thus, all natural fatty acids are produced initially with even‐numbered carbon chains. Sometimes fatty acid chains contain double bonds. These bonds are always formed in the cis‐configuration. Some examples are shown in Figure 12.13. All three contain 18 carbon atoms and their names indicate good natural sources of O

O

OH

OH

O

O OH Reduction O OH

Figure 12.12  Biosynthesis of fatty acids. O OH Stearic acid O OH Oleic acid O OH Linoleic acid

Figure 12.13  Some representative fatty acids.

223

224

12  The Chemistry of Living Organisms

each. Stearic acid (from beef tallow) is a saturated acid, oleic acid (from olive oil) contains one double bond and is thus a monounsaturated fatty acid, and linoleic acid (from linseed oil) with its two double bonds is a polyunsaturated fatty acid. In living organisms, lipids are used as an energy store, as water barriers to keep living organisms from dissolving in their environment and as electrical insulators in, for example, nerves. The simplest class of lipids are the triglycerides. Members of this family contain one molecule of glycerol (1,2,3‐trihydroxypropanol) esterified to three fatty acids. All three fatty acids can be different and, indeed often are, in nature. Palmitic acid (from palm oil) contains 16 carbon atoms in its chain. In the upper part of Figure 12.14, we see three molecules of glycerol tripalmitate, and it is clear from the figure how they can stack together easily. Consequently, triglycerides of the higher saturated fatty acids are usually solid. Beef tallow is an example of a solid triglyceride as it is composed mostly of glycerol tristearate. Introduction of a cis‐double bond has a dramatic effect on the shape of a triglyceride as can be seen in the lower part of figure where the glycerol molecule is esterified to two molecules of palmitic acid and one of palmitoleic acid (a monounsaturated, 16‐ carbon fatty acid). It is clear from the figure that this molecule will not stack so easily into a crystal structure, and so triglycerides containing unsaturated fatty acids are usually liquids. Examples of liquid fatty acids include sunflower oil and olive oil. The other two main groups of lipids are the phospholipids (also known as lecithins) and the sphingolipids, such as ceramides. In phospholipids, only two of the alcohol groups of glycerol are esterified to fatty acids, and the third is esterified to phosphoric acid. Other groups can then be bonded to the phosphate. Phosphatidylcholine, shown in Figure 12.15, is a typical example of a phospholipid. The sphingolipids are based on the amino alcohol sphingosine. In the ceraO O

OO O

OO

O O

O

OO O

O

OO

O

O O OO O

Figure 12.14  Triglycerides.

O O

  ­Protein

N

+

–O

O

O P

O O

O

O

O Phosphatidylcholine

O HN HO Ceramide

OH O N

+

HN O O

P

O O

OH

Sphingomyelin

Figure 12.15  Phospholipids and sphingolipids.

mide family, sphingosine forms an amide with a fatty acid, as shown in Figure 12.15. Sphingomyelin is a ceramide to which a choline phosphate has been added, similar to that of phosphatidylcholine. Lipids of all these classes have two long fatty tails and a polar head group that gives them surface‐active properties and, as described in Chapter 4, that enables nature to use them to build mammalian cell walls.

­Proteins Proteins are used as food energy, but much more important are their uses as structural materials (for example, in hair, bone, and collagen) and catalysts and in recognition. For example, olfactory receptor proteins are responsible for our sense of smell as will be described in detail in Chapter 13. Proteins are polymers made by linking α‐amino acids (usually abbreviated to amino acids) together through amide bonds. The amide linkage is also known as the peptide bond, and so proteins are also known as polypeptides. Twenty‐one natural amino acids are used in making proteins. If we try to synthesise a hexapeptide (i.e. a protein containing six amino acid units) and restrict ourselves to only 10 amino acids, we will have 10 choices for the first unit, 10 for the second, and so on. Therefore, one million possibilities exist. So, when we consider making proteins, some of which contain many hundreds of amino acid units, using the whole palette of 21 amino acids, the number of possibilities is vast. The structures of 6 of the 21 amino acids are shown in Figure 12.16. Each has a carboxylic acid group and a primary amino group on the adjacent carbon. The

225

226

12  The Chemistry of Living Organisms O

OH

O

HO

OH

OH

NH2

OH

NH2

Leucine

NH2

Threonine

H2N

Glutamic acid NH

O

H2N

O

N H

OH Lysine

O

O

NH2

OH Arginine

NH2

O HS

OH

Cysteine

NH2

Figure 12.16  Some amino acids.

remainder of the molecule can contain a variety of functional groups. Some indication of the possibilities is given in the examples in the figure: leucine has an alkyl chain; threonine has an alcohol function; glutamic acid has a second acid function; however, lysine has a second amino group; arginine has a guanidine group; and cystine has a thiol function. These side chain functional groups are very important in determining the activity of proteins, as we will see when we discuss tertiary structure. The SH groups of cystine are important in holding the tertiary structure in place through disulfide bridges as was explained in Chapter 11 and shown in Figure 11.7. Proteins have a wide range of functions in living organisms. Some proteins provide physical structure to the organism, e.g. keratin in hair and nails and collagen in muscle, sinew, and bone. Some other proteins serve to transport smaller molecules around the organism. An example is haemoglobin, which contains the red pigment haem and is responsible for the colour of blood. The iron atom in the haem is used to carry oxygen from the air in the lungs, via the blood, to the points in the body where sugars and so on are being burnt to provide energy. A family of proteins, known as lipocalins are specialised carriers of small molecules, often as a means of removing foreign molecules from the body. A subgroup of the lipocalin family is the so‐called odour binding proteins, which will be discussed further in Chapter 13. The basic structure of the mammalian cell wall was described in Chapter 4. Some proteins sit in the cell wall or span right across it. Those that make contact with both the interior of the cell and the environment surrounding the cell are known as transmembrane proteins. These proteins play a crucial role in chemical communication since they are the means by which the interior of the cell can interact with the environment of the cell. For example, hormone receptors are the means by which chemical signals from one part of an animal’s body can be

  ­Protein

interpreted by another part. Some receptor proteins, including the olfactory receptor proteins, allow animals to learn about the chemistry of the external environment. Other transmembrane proteins contain channels through which charged species can travel from inside the cell to outside or vice versa. These proteins are known as ion channels, and they play a part in the process of olfaction as will be seen in Chapter 13. Figure 12.17 shows a section of a lipid membrane with a transmembrane protein spanning the lipid bilayer. Proteins have three levels of molecular structure, viz. primary, secondary, and tertiary. The primary structure is the sequence of amino acids in the chain and the position of disulfide bridges between cysteine residues in the chain. Two ways of illustrating this structure are shown in Figure 12.18. At the top of the figure, a structural diagram shows how six amino acids can be linked together to form a polypeptide chain. In the lower part of the figure is a sequence of six specific acids (in this case, actually those shown in Figure 12.16) using three‐letter abbreviations for the names of the amino acids. Obviously, these structures have the amino function at one end of the chain and the acid function at the other. These ends are known as the amino and carboxyl termini, respectively. The secondary structure of a protein refers to the way in which the chain sits. The most common form is the α‐helix, similar to the helix of the nucleic acids. A left‐handed α‐helix protein structure is shown in Figure 12.19a. Another com-

Lipid bilayer

Transmembrane protein

Figure 12.17  A cross section of a lipid bilayer and transmembrane protein. R

O

H N

H2N O

R

R N H

O

H N O

R

R N H

O

H N O

OH R Carboxyl terminus

Amino terminus Leu-Thr-Glu-Lys-Arg-Cys

Figure 12.18  Primary structure.

227

228

12  The Chemistry of Living Organisms

βR RH H

R

HR βR O

H

R

HR βR O

H

O

βR RH R N HR βR (a)

O

H

N

H

O

H

R N

R

R

N R βR RH

H

βR RH

HR βR O

O

N HR βR

βR RH

R N

H

R

N

N

βR RH

O

H

O

N HR βR

(b)

Figure 12.19  Secondary structure of proteins.

mon form is the pleated sheet, as shown in Figure 12.19b. In the example shown in this figure, its alternate protein chains run in opposite directions, and it is therefore known as an antiparallel sheet; had all the chains run in the same direction, it would be a parallel sheet. Proteins that fold into distinct shapes such as spheres, ellipsoids, or columns, rather than existing as threads or extensive sheets, are known as globular proteins. The basic chain will probably contain sections that have α‐helix structure or may have regions of sheets and so on. However, the exact shape of the protein is known as its tertiary structure. This shape is determined partly by non‐bonded interactions between different parts of the basic peptide chain and partly by disulfide bridges between cystine units at different points in the chain. It is sometimes possible to ‘denature’ a protein by heating it. This denaturing happens because heat disrupts the non‐bonded interactions and allows the protein chain to break free from the shape imposed on it by a network of such forces. Denaturing means that the protein loses its original shape and thus its key biological properties (such as catalytic activity for an enzyme) since these properties depend on the shape of the protein for their selectivity. Figure 12.20 shows schematically how a folded protein chain loses its shape when heated but regains it again upon cooling as the non‐bonded interactions fall back into place. This result does not always happen; sometimes a denatured protein will not regain its original shape. The tertiary structure of a protein is very important in determining its biological properties. An example of this structure is shown in Figure 12.21 for an enzyme active site. Enzymes are nature’s catalysts. They drive the chemical reactions that make living cells function and are very selective in the substrates they will accept and the products they produce. The enzyme shown in the figure is one that reduces the pyruvate anion (CH3COCO2−) to lactate (CH3CHOHCO2−). The pyruvate ion about to be reduced can be seen at the middle of the left edge

  ­Protein

Figure 12.20  Tertiary structure of proteins. His196

Lys250

H

+

N O O NH2+ H2N Arg171

NH

Arg101

N H H

H2N

NH2

HN

N H

NH2

NH2+

O

O H H

O– O

N

O

HO O

N O P

P O O HO

N

Glu98

O

O

Tyr85

N

O

OH O

H

O O

HO O

N

OH

O Asp30

H

+

H N

O–

Asp53

H

Lys58

Figure 12.21  Molecular organisation in an enzyme active site.

of the figure. The various amino acids of the protein involved are shown as three figure abbreviations, e.g. His for histidine, Lys for lysine, Arg for arginine, and so on. The functional group of the amino acid side chain is also shown, and it is apparent how these form non‐bonded interactions to the various reaction components. This enzyme, like all red‐ox enzymes, requires a coenzyme or cofactor in order to carry out the reaction. In this case, the cofactor is the reduced form of nicotinamide adenine dinucleotide phosphate or NADPH, for short. This molecule can be seen running across the centre of the figure. The hydrogen atom, which will be added to the ketone group of the pyruvate, is attached to a reduced pyridine ring. The remainder of the molecule is used mainly for recognition, and it is clear how it has no fewer than eight hydrogen bonds holding it in place. The pyruvate substrate has two hydrogen bonds holding it in place. The important thing to note is the range of numbers added as subscripts to the amino acid abbreviations. These numbers indicate the position of that amino acid in the primary structure of the protein. It is clear from the figure that amino acid residues from all over the protein molecule are involved. The tertiary structure (that is the overall shape of the protein) is therefore vital in bringing all of these key functions together at exactly the right distance to recognise both the cofactor and the substrate and to hold them together ready for reaction. It is this ability of

229

230

12  The Chemistry of Living Organisms

the protein to organise the reacting molecules that is responsible for its remarkable selectivity (if a substrate doesn’t fit exactly, it won’t take part in the reaction) and also in speeding up the reaction by lowering the entropy (in other words by introducing the necessary order into the system). The centre where reactions take place in an enzyme is known as its active site. The so‐called binding pocket or binding site, where small molecules ‘dock’ in a receptor protein or lipocalin, has a similar degree of organisation to that of an active site in an enzyme.

­Toxicity and Product Safety So far in this chapter, we have been introduced to the types of molecules of which living organisms are composed and had a very brief introduction to the complex chemical processes on which we, and all living creatures, rely. Needless to say, anything that interferes with these processes is potentially harmful to the organism and is described as a poison or toxin. An example is cyanide, either in the form of HCN or as the cyanide anion (−CN). When dissolved in water, HCN dissociates into protons and cyanide anions, thus constituting a way of delivering the cyanide anion to the body. It is the cyanide anion that results in the toxic effect because of its high affinity for metal ions. Many enzymes and other key body chemicals depend on metal ions for their activity, and complexation with cyanide anions destroys this ability. For example, haemoglobin, the protein that gives blood its red colour, serves to carry oxygen from the air in the lungs to the tissues of the body where it is needed. Cyanide binds more strongly to the iron atoms in haemoglobin than oxygen does, and so cyanide in the bloodstream will prevent the haemoglobin from transporting oxygen and consequently result in the body ‘suffocating’ for lack of oxygen. The study of poisons is known as toxicology, and toxicologists seek to understand poisons and to determine the hazards associated with them. Using their data on hazards, the industry and governmental bodies assess the risks of using materials. Risk is a combination of hazard and exposure. As an example of this model, we can consider water and sulfuric acid. The hazards of handling sulfuric acid are much greater than those of water because of its strongly acidic and oxidising nature. However, in daily life, our exposure to water is very much greater, and so water constitutes the greater risk. Indeed, many more people have died as a result of contact with water (e.g. by drowning) than from contact with sulfuric acid. Later in the chapter, we will see the approach taken to risk assessment by government regulators and by the fragrance industry. The most important single principle of toxicology is that of the fifteenth century Swiss scientist, Paracelsus who stated that, ‘All substances are poisonous; the dose alone determines the poison’. For instance, we tend to think of pure water as safe, but half a litre in the lungs will kill, by preventing oxygen uptake by haemoglobin in the blood. Three litres of water drunk in a short space of time is also potentially lethal because of the effect on dilution of the body’s electrolytes. Those materials, which we normally think of as poisons, are those substances in which only a low dose is necessary to produce an adverse effect. This dose varies from species to species and from individual to individual within a species. In

 ­Toxicity and Product Safet

previous times, acute toxicity was measured as the LD50, which is defined as the dose that will kill 50% of animals exposed to it. It is usually expressed as milligrams of substance per kilogram of the animal’s body weight, mg/kg. Thus, the lower the LD50, the less of the substance is required to produce the effect; in other words, the smaller the LD50 value, the more toxic is the substance. The LD50 is species specific and the test species is named with the result. The route of administration is also important (e.g. the example of water given above), and so the test will also include the means of administration. The most common are oral (i.e. by mouth) and dermal (i.e. by skin adsorption). For legal purposes, the European Union has defined various bands as follows. ‘Harmful’ indicates a substance with an LD50 between 2000 and 200 mg/kg, ‘toxic’ indicates a substance with an LD50 between 200 and 20 mg/kg, and ‘very toxic’ indicates a material with an LD50 below 20 mg/kg. The LD50 figures relate to acute toxicity, that is, the immediate effects of a single dose. Chronic toxicity refers to the effects of continued exposure over a long period of time. It must therefore be measured against exposure over a longer time, and the critical figure is known as the no observable effect level (NOEL) or, alternatively, the no observable adverse effect level (NOAEL). These terms are self‐explanatory. Although the LD50 test is now obsolete, the results can still be used to illustrate some principles of risk assessment. Figure 12.22 shows the LD50 values of a number of chemicals for rats and mice. At the top of the LD50 list is sucrose, the major component of cane sugar. This substance has a high LD50 of 33 000 mg/kg, i.e. 33 g/kg body weight. This result means that, if the LD50 were the same for a human as for a rat, a person weighing 80 kg stands a 50/50 chance of being killed by a dose of 33 × 80 g (2640 g). In imperial measure, this equates to 5.8 lb for someone weighing 176 lb (12 stone, 8 lb). If we build in a safety margin by allowing for the possibility that sucrose might be 10 times as toxic to humans as to rats, we can see that adding one teaspoon of sugar (about 5 g) to a cup of coffee is safe, as far as acute toxicity is concerned. A little arithmetic will show that the caffeine content of the coffee constitutes a higher risk. To put perfume in context, Figure 12.22 includes some typical fragrance ingredients, viz. farnesol, limonene, geraniol, citronellol, and linalool. It is clear from the table that these substances are in a similar category to common kitchen ingredients such as salt and alcohol and therefore do not represent a high risk as far as acute toxicity in normal use is concerned. Eating fugu (puffer fish) in a sushi bar involves a higher risk than sniffing a perfume (granted that the route of administration is different). We can also learn two other important general points about toxicology from the data in Figure 12.22. Firstly, it shows that the idea of ‘natural’ equating with ‘safe and healthy’ is a misconception since all of the materials listed as harmful, toxic, or very toxic are natural and specific natural sources are indicated for a number of them. Secondly, we know that we can drink coffee and survive, thus proving that the body has ways of dealing with toxins. In fact, some toxins are used as medicines, such as digitalis. Digitalis is the toxic component of foxgloves, but it is used, at the correct dose level, in treatment of some heart conditions. (Of course, in keeping with the principle of Paracelsus, all medicines have a toxic dose.) Indeed, some toxins are essential for our health. An example is vitamin D2, without which we would suffer from rickets.

231

232

12  The Chemistry of Living Organisms

Compound

LD50 (mg/kg)

Sucrose Alcohol Vitamin B1 (thiamine) Farnesol Limonene Geraniol Citronellol Salt

33 000 13 000 8 224 6 000 4 400 3 600 3 450 3 000 2 790 375 250

Linalool Oxalic acid (rhubarb) Caffeine Digitoxin (foxglove)

200

Vitamin D2

42

Strychnine Saxitoxin (red tide) Tetrodotoxin (puffer fish) Batrachotoxin (frogs) Palytoxin (coral)

5 0.063 0.01 0.002 0.00015

Harmful

Toxic Very toxic

Figure 12.22  Selected LD50 values.

It is clearly not in the interest of the fragrance industry to harm its customers, and so it has invested considerable research into understanding any potential hazards to find ways of removing these or reducing them as much as possible. The subjects that we need to give most thought to include skin sensitisation, skin irritation, eye irritation, phototoxicity, mutagenicity/carcinogenicity, and ecotoxicity. Skin sensitisation is a form of allergic reaction and results from activity of the immune system. It is a direct physiological chain of events, and a blood test can reveal if a subject has become allergic to a substance in this way. It is also possible for an allergic response to be initiated by the brain. In this case, if a subject believes a substance to be a threat, his/her brain can trigger an allergic response in the form of a skin rash, asthma attack, or such. The immune reaction is vitally important to us since it allows our bodies to protect themselves from attack by foreign organisms and chemicals. The key part of the immune system is a group of proteins known as antibodies. These proteins are large and highly variable with a Y‐shaped tertiary structure. If a foreign protein, an antigen, enters the body, it will be recognised as foreign by an antibody, and the antibody will complex with it. This process then sets in motion a series of other biochemical processes that will ensure mobilisation of the body’s defence against the intruder. New antibodies are generated by exposure to a new antigen, and so meeting a foreign protein will lead to the development of specific antibodies against it. This process is the principle behind vaccination. Sometimes small molecules can react with a protein present in the body to produce a ‘foreign’ protein, i.e. an antigen, and this reaction will trigger the release of antibodies and an immune response. Such small molecules are known as haptens. Some fragrance

 ­Toxicity and Product Safet

i­ngredients can act as haptens in some individuals. Thus, one exposure to the fragrance ingredient will cause antibodies to be created against it, and then subsequent exposure to the same material will cause an immune response, which, if it gets out of hand, produces a rash. This outcome is known as skin sensitisation. Dermatologists suggested to the regulatory authorities of the European Union that cosmetic products containing the most common potential sensitisers be labelled so that consumers who are allergic to any of these can avoid buying products containing them. This led to the 7th Amendment to the European Cosmetics Directive, and 26 suspected allergens now have to be declared on product labels. There are 26 substances on the list, 19 of which are common natural components of essential oils and extracts. Skin sensitisers therefore tend to be substances that can modify proteins. As far as fragrance ingredients are concerned, the two main properties that enable this are electrophilicity and the ability to generate oxidising species, usually via autoxidation. The nitrogen, oxygen, and sulfur atoms in proteins are particularly susceptible to attack by soft electrophiles and, in particular, by Michael acceptors (or double bonds carrying electron‐withdrawing groups such as aldehydes, esters, and ketones). Figure 12.23 shows how a nitrogen atom in a protein could react with benzyl chloride or methyl vinyl ketone thus modifying the structure of the protein. Methyl vinyl ketone is known as a Michael acceptor because it can act as the electron pair acceptor in a Michael reaction (see Chapter 8). Michael acceptors are of course capable of reacting with two nucleophiles: the first one can add to the double bond in a 1,4‐addition (Michael addition), and the second can then add to the electron‐withdrawing group whether it be aldehyde, ketone, or ester. This ability to undergo double addition seems to increase the likelihood of a substance being a sensitiser. Peroxides and hydroperoxides are capable of modifying proteins by cleaving to form free radicals that can then trigger oxidation of the protein. Substances that readily undergo autoxidation to give peroxides and hydroperoxides are therefore also potential sensitisers. Knowing these mechanistic factors enables the fragrance chemist to design new molecules that avoid having such properties. Computer models are also used in predicting skin sensitisation. There are two sets of properties that are used in predictions, whether computer or expert based. First of all, properties such as log P and molecular weight are used to determine whether or not a substance is likely to penetrate the outer layers of skin and reach underlying tissue where the allergic response would arise. A second set of properties is concerned NH

NH

Cl

O

N

N

O

Figure 12.23  Reaction of soft electrophiles with an amino function in a protein.

233

234

12  The Chemistry of Living Organisms

with the reactivity of the molecule and hence the likelihood of it causing the sort of structural modification that would lead to a protein being seen as foreign by the immune system. Skin irritation is different from skin sensitisation in that the reaction occurs immediately upon first contact. The reaction is localised and does not involve the immune system. The mechanism is less clear than that of skin sensitisation, and so predictions are less accurate. However, the concentration of fragrance ingredients in consumer goods is usually low enough that there is little risk of irritation. Perfume is never intended to come into contact with the eyes, but, in a product such as shampoo, it is possible that accidental eye contact may occur. Eye irritants tend to have good water solubility, unlike the majority of fragrance ingredients. Phototoxic materials are those that, when exposed to sunlight, absorb ultraviolet radiation and produce new species that are reactive and cause either irritation or sensitisation. The best known examples are extracts from figs (leaves, wood, or sap) and bergamot. The reaction with fig extracts is severe enough to have led the fragrance industry to voluntarily ban their use. In bergamot, the molecule responsible is known as bergaptene (see Figure 12.24), which has a higher boiling point than the fragrant components of the oil, and so can be removed by careful distillation. Mutagenic materials are those that are capable of reacting with nucleic acids in such a way that they lead to the production of nucleic acids that are not identical to the original but contain some changes, meaning that the genetic code has been altered and the organism is changed. In some instances, a cell can be modified so that it becomes cancerous. Materials that do this are known as carcinogens. A very simple test, employing bacteria (bacteria are much more sensitive to mutagens than are humans), is used to identify mutagens. It is called the Ames test after the person who devised it. Thus, novel materials can be easily screened for mutagenic activity. Another screening technique that is used is to look for chromosome aberration is the human lymphocyte test. It sounds very frightening if a substance is declared to be a mutagen or carcinogen. However, we must remember that these are very complex processes in reality and many different factors have to occur together for a mutation or cancer to develop. For example, allyl isothiocyanate (see Figure 12.24) has been shown to cause damage to chromosomes. However, the dose at which this occurs is one fifty‐thousandth that found in a single cabbage leaf, and we know that we can eat cabbages and suffer no ill effects. As with skin sensitisation, soft electrophiles (and O

S N

O

O Bergapten

O Allyl isothiocyanate

Figure 12.24  Bergaptene and allyl isothiocyanate.

 ­Toxicity and Product Safet

­ articularly Michael acceptors) are more likely to display mutagenic or carcinop genic properties. Obviously, the industry does what it can to avoid using ingredients that fall into this category, and such structure/activity relationships are important tools for us. The modern fragrance industry also considers ecotoxicity to be important as part of the ‘cradle to grave’ approach to product safety. Ecotoxicity is the potential of a material to damage the environment when it is released into it after use by the consumer. For most fragrance applications, it means disposal via the sewage system. Biodegradability is becoming increasingly important to us. Organisms can break down chemicals by the reverse of the process they used to make them in the first place. Thus, all the primary metabolites are readily broken down by all organisms. However, most fragrance materials are secondary metabolites or similar, and organisms might resort to using oxidative enzyme systems called cytochromes for breaking down any organic chemicals they wish to discard. Bacteria are used in the first stage of sewage treatment works, and fungi can be used in later parts of the process. The initial stage is carried out in water, and so materials that are not very water soluble might not be accessible to the bacteria. However, they might have just enough water solubility to allow traces to escape into the outflow and hence into the environment. The industry is therefore currently actively seeking to design materials that will degrade readily in the first stage of sewage treatment in order to improve the environmental position of our products. The legal definitions of biodegradability are based on bacterial treatment in sewage works. The range of standard tests for biodegradability was designed by the Organisation for Economic Co‐operation and Development and is therefore known as the OECD test. In this test, the substance being investigated is added to sewage sludge, and the rate at which oxygen is consumed or carbon dioxide is produced is measured. The aim is to determine how long it takes for the bacteria to completely oxidise the carbon in the substance to carbon dioxide and water. One complication is that the bacteria will use some of the carbon atoms from their ‘food’ source to build components of their own cells, and this carbon will not appear as carbon dioxide. Two tests are used, one of which sets more rigorous criteria. The first is known as the ready test, and a substance that passes can be described as ‘readily biodegradable’. The other test uses conditions that are more amenable to biodegradation, and substances that pass it can be labelled ‘inherently biodegradable’ or ‘ultimately biodegradable’. Both tests are very stringent, and, since they ignore degradation by fungi, sunlight, and so on, they always underestimate the potential of a material to degrade in the real environment. As with skin sensitisation, structure/property relationships are used to predict the biodegradability of chemical substances. For biodegradability, two different approaches are used, and often both will be used together to complement each other. The first approach, which is essentially statistical in nature, is a so‐called ‘fragment‐based’ method. In this, molecular fragments such as functional groups (ester, ketone, aldehydes, and so on), linear chains, branched chains, and rings are rated for biodegradability based on known values for molecules containing them. A new molecule can then be rated for biodegradability based on the fragments of which it is composed. The second approach seeks to

235

236

12  The Chemistry of Living Organisms ω-oxidation

CO2H –2H

R

R

via CH2OH, CHO

H2O

O R + CH3COOH

OH

CO2H

R

OH

O

H2O R

CO2H

–2H

R

CO2H

Figure 12.25  ω‐Oxidation and β‐oxidation.

understand the mechanisms by which substances are broken down by microorganisms and thus predict how (and therefore how easily) any individual molecule will be degraded in the environment. Figure 12.25 shows the processes known as ω‐oxidation and β‐oxidation. The former is so‐named because it is the terminal (or ω – omega) carbon that is oxidised. The term β‐oxidation refers to the oxidation of a carbon atom next but one to a carboxylic acid or ketone. These processes are used by bacteria to oxidise alkyl chains, and, in doing so, the bacteria gain both energy (from oxidising organic material and releasing the stored energy of the C─H and C─C bonds) and molecular building material (namely, acetic acid  –  see Chapter  14 for a description of how plants use acetic acid to build larger molecules). In the first step, the terminal carbon atom of the chain is oxidised through the alcohol and aldehyde to the acid. Two hydrogen atoms are then removed by further oxidation to give an α,β‐unsaturated acid. Michael addition of water to this formula produces a β‐hydroxy acid that is then oxidised to a β‐keto acid, and this substance undergoes a reverse aldol reaction to give acetic acid and an acid with two fewer carbon atoms in the chain than there were in the original. The process can then be repeated over again to reduce the chain length further until eventually the whole chain has been degraded. Any substituents along the chain that hinder or prevent any of the intermediate steps in the sequence will obviously slow down the rate at which the chain is degraded. Molecules containing benzene rings can be broken down by bacteria in two different ways. The first of these methods is shown in Figure 12.26. In this route, one of the double bonds of the benzene ring is epoxidised (epoxidation is addition of an O atom to a double bond to form a three‐membered ring) to give a very strained structure with a six‐membered ring fused to a three‐membered one. This hydrolyses rapidly to give a diol that can be oxidised to give a catechol (a benzene ring with phenolic substitution on two adjacent carbon atoms). This phenol can be broken down further by one of two routes, each involving cleavage of one of the bonds of the ring structure. The two main options are shown in the figure in which cleavage of the bond marked A results in path A and, similarly, cleavage of bond B gives path B. In either case, the ring is opened to a highly oxygenated chain that is readily degraded further by a β‐oxidation pathway similar to that of the ω‐oxidation route. The other route is shown in Figure 12.27. In this case the oxidation process starts at a carbon atom adjacent to the ring and then involves oxidation of an

 ­Toxicity and Product Safet R

R

[O]

H2O

R

OH

O OH

–2H B R

OH A

R

O

O

A

OH

B

OH

O OH

R

O

OH

OH

Figure 12.26  Oxidative degradation of benzene rings via epoxidation. R

CO2H

CO2H

OH OH

OH

Figure 12.27  Degradation of aromatics by initial side chain oxidation.

adjacent ring carbon to give a salicylic acid derivative, which can then be decarboxylated to a catechol. This catechol can then be degraded by the route shown in Figure 12.26. Any organic (i.e. carbon containing) chemical substance that is released into the environment, whether by human activity, by plants, or by any other means, will eventually degrade to carbon dioxide and water as a result of oxidation reactions. These oxidations could involve autoxidation, enzyme‐catalysed oxidations in living organisms, photocatalytic reactions (i.e. reactions catalysed by sunlight), or a variety of other processes. The crucial questions are: ‘How long will the degradation take?’ and ‘Will the substance cause any harm before it has been degraded?’. The tests and structure/property correlations discussed in this chapter attempt to answer the first question. In order to address the second question, substances are tested for their toxicity to aquatic organisms such as algae, fish, and daphnia. Any potential to inhibit the bacteria in sewage sludge from degrading other substances is also determined. As with biodegradability, legal definitions have been developed in connection with labelling of ingredients. Substances that harm fish and daphnia at a concentration of 10–100 mg/l are defined as harmful to aquatic organisms, those that do so at 1–10 mg/l are defined as toxic to aquatic organisms, and those that do so at less than 1 mg/l are defined as very toxic to aquatic organisms. Knowing how long a substance is likely to persist in the environment, and what is the potential for harm while it is there, then ena-

237

238

12  The Chemistry of Living Organisms

bles regulatory experts to make a judgement about the likely risks for a given production volume. The larger the actual or projected production volume, the more tests will be required to satisfy regulators that the material is acceptable. If, after clearance for manufacture at one tonnage level, the production volume rises, these additional tests must be carried out, and failure will result in enforced total withdrawal from the market. With environmental safety, just as was the case for safety in use, natural does not necessarily mean better than man‐made. For example, Amazonian natives know very well that some rainforest trees produce chemicals that are extremely toxic to fish, and they use them in fishing. We must also remember to keep things in perspective. The terpenoid fragrance ingredients produced from turpentine (see Chapter 14 for definitions and production routes) have been estimated to account for less than one tenth of a percent of the volatile organic chemicals (VOCs) produced by coniferous trees. Furthermore, some of these tree‐produced chemicals fail the ready biodegradability test. From time to time, other safety concerns arise, and one example is that of ­oestrogenic activity. This ability of substances to mimic the female hormone oestrogen possibly may cause problems such as male infertility and sexual and reproductive abnormalities. Many plant materials exhibit this property and are thus known as phytoestrogens. The best known are probably those present in soy milk. The plasticiser used in polyvinyl chloride (PVC) film and some other plastics is bis(diethylhexyl) phthalate. Some concerns exist that, at very high concentrations, this substance might cause oestrogenic effects. Exposure resulting from its use as a plasticiser will increase circulating oestrogen activity but only by one five‐thousandth the amount caused by eating the recommended daily amount (RDA) of five pieces of fruit. Nonetheless, the similarity of its chemical structural class to that of diethyl phthalate (DEP, which has been extensively tested and shown to be completely safe) has led to a campaign to have the latter discontinued as a perfumery solvent, and the industry has abandoned use of DEP as a solvent. As stated earlier, the fragrance industry strives to ensure product safety for ethical, legal, and commercial reasons. It therefore has a long history of testing and evaluating ingredients for safety in use and after use. Some companies arrange for their own testing of properties such as biodegradability, but extensive testing of ingredients is also carried out by the Research Institute for Fragrance Materials (RIFM), at the request of the International Fragrance Association (IFRA). The latter is a body funded by the fragrance industry, which sets out guidelines for the safe use of fragrance ingredients, both natural and synthetic. The limits on use imposed by IFRA are followed by all the major fragrance houses, and these invariably meet or exceed those imposed by government bodies. In most countries of the world, a company must have the government’s permission before manufacturing or using a new chemical entity. The European Union has now set a new trend through legislation known as Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) requiring that

 Review Questions

all existing substances (including natural extracts) are registered, tested, and approved by the European government before manufacture in or importation into any member state of the European Union. Since most fragrance houses and most of their larger customers are international in operation, this ruling affects the entire industry. An international company must satisfy the laws of every country in which it operates. The exact process of registration of chemical substances varies from country to country. As an example of such a process, we can look at the elaboration of an IFRA standard. Initially a dossier is prepared containing the relevant information about the material. This document will include physicochemical data such as melting/ freezing point, boiling point, density, vapour pressure, water solubility, partition coefficient (log P), flash point, auto‐ignition temperature, and stability at various pH levels. Test data will include both toxicological data (such as Ames mutagenicity test, skin irritation, eye irritation, and skin sensitisation) and environmental toxicological data (such as OECD biodegradability and aquatic toxicity). Based on this data, a risk assessment and a provisional material safety data sheet (MSDS) will be issued. All of this material will be reviewed by a committee known as the Research Institute for Fragrance Materials Expert Panel (REXPAN) that will draft a proposed standard. The exposure to the consumer can be estimated based on the amount of final product used per day, the level of fragrance in that product, the proposed level of the ingredient in the fragrance, the amount that is likely to be transferred to the consumer by this exposure, and hence the amount that will be retained on the skin. The ratio of this final level to parameters, such as NOAEL, will then show the safety margin in use. The draft standard will thus contain recommended safe use levels in the entire range of consumer products into which the ingredient is likely to be incorporated. Obviously, the levels may well vary from one product type to another. For instance, the fragrance in a skin cream will have much more contact with the skin of the user than will that of a laundry detergent. REXPAN will discuss the proposed standard with all the member associations, the client industry, and the IFRA scientific committee and will then draft a final wording based on the comments they receive, and this information will be communicated as a new standard to the industry by the IFRA secretariat. In the case of clearance by a government agency, the company proposing to manufacture or import a substance will prepare the dossier, and the decision will be made by the experts of the government agency.

Review Questions 1 To which classes of natural products do the chemicals shown in Figure 12.28 belong? 2 Figure 12.29 shows two cyclopentanone derivatives. Heptylcyclopentanone is a useful fragrance ingredient with a sweet fruity floral odour. It is prepared by hydrogenation of heptylidenecyclopentanone. If some starting material

239

240

12  The Chemistry of Living Organisms

OH H

HO

OH

O

OH OH

O OH H H

O OH

H O

OH

O O

H H

O

O

Fructosyl fructose

Tripalmitin

O

H N

H2N

O

O

N H

OH H N

O OH

O

Phenylalanylleucylthreonylalanine

Figure 12.28  Some natural substances. O

O

Heptylcyclopentanone

Heptylidenecyclopentanone

Figure 12.29  Two cyclopentanone derivatives.

(i.e. heptylidenecyclopentanone) from the hydrogenation reaction were left in the final product, it would probably affect the odour. Would it also affect its safety in use? 3 The structures of α‐ionone and α‐damascone are shown in Figure 12.30. Are the limitations on use of these two materials likely to differ? O

α−Ionone

O

α−Damascone

Figure 12.30  α‐Ionone and α‐damascone.

 Review Questions OH

R

OH

R

Compound A

Compound B

OH

R

R Compound C

O

OH

Compound D

Figure 12.31  Selected alcohols.

4 Figure 12.31 shows the structures of four alcohols. In each case, the fragment R is readily biodegradable. Which of the four alcohols would you expect to be readily biodegradable?

241

243

13 The Mechanism of Olfaction ­The Role of Olfaction in Biology Chemoreception is the process by which living organisms gain information about their environment by detecting the presence of chemicals in it. Smell (olfaction) and taste (gustation) are forms of chemoreception and are the oldest of our senses. The sense of smell gives organisms information about changes in the chemistry of their environment and therefore gives warning of dangers such as fire, bad food, and predators, and it also alerts to opportunities such as good food, potential mates, and social organisation. Taking all of this into consideration, the sense of smell must be time based, capable of quickly dealing with complex mixtures of molecules and also recognising previously unknown molecules. In order to achieve this, the mechanism of olfaction is extraordinarily complex as we will see in this chapter. The sense of smell is more important than sight for most mammalian species, and most mammals use about twice as many different types of smell receptors as humans do. Only a few primates (humans, gorillas, orangutans, chimpanzees, and rhesus macaques) and diurnal birds rely more on sight than other senses. Interestingly, those primates with a poorer sense of smell are the only mammals with colour vision, suggesting an evolutionary trade‐off of smell for colour. Nonetheless, smell is still important for humans and, of course, forms the basis of the fragrance industry. It is obvious that, for the industry to better serve its customers, it is desirable to understand as much as we can about the sense of smell. Smell is the oldest of our senses, and the basic chemistry employed by humans is probably similar to that which first appeared in primitive organisms since nature does not abandon systems that work but rather modifies, refines, and adapts them through the course of evolution. We can therefore learn something about our sense of smell by studying that of other animals. However, it is important to remember that other species sometimes use smell in a different way from us and discoveries about how their sense of smell works do not necessarily translate directly to human olfaction. For instance, many insects use pheromones as a means of communication and for control of social activity, but this does not mean that human behaviour can be controlled by volatile chemicals in the same

Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

244

13  The Mechanism of Olfaction

way. The rest of this chapter is a summary of what is currently understood about the chemistry of the sense of smell in humans.

­The Organs Used in Olfaction The most important organs used in the process of human olfaction are shown in Figure 13.1. Odorants enter the nose either from the front by sniffing (inhalation), which is called the ortho‐nasal route, or from the back by diffusion from the throat and/ or airways, and this is referred to as the retro‐nasal route. The retro‐nasal route is, of course, of major importance in tasting of food, whereas for perfumes, the ortho‐nasal route is the one we need to consider.

Glumerulus Olfactory bulb

Cribriform plate

Olfactory epithelium

Receptor cell

Cilia

Orbitofrontal cortex Thalamus Cribriform plate Nasal turbinates

Amygdala

Odorant molecules

Trigeminal nerve

Piriform cortex

Figure 13.1  Key organs involved in human olfaction. Source: Courtesy of Givaudan.

 ­The Organs Used in Olfactio

Our nasal cavity is divided into two parts that are separated by the nasal septum, a vertical plate that runs from the front to the back of the nose. The airflow through each of our nostrils is always different in that the airway on one side is always more restricted than that on the other because of swelling of the mucous tissues. This means that less air flows through that side of the nasal cavity and does so more slowly. After some time, the ‘noses’ swap over, and the one that had the faster flow becomes the slow one, and vice versa. This is a very clever device of nature. Because the airflows are different, the physical chemistry of adsorption of odorant molecules by the epithelium is different, and so the pattern of firing of receptors will be different. The signals from the two are not combined until after the level of the olfactory bulbs, and the brain knows which is the ‘fast’ nostril and which is the ‘slow’ at any given time. This adds considerably to the sensitivity of the nose, and the effect can even be reproduced by using two electronic noses operating with different airflows and comparing the results with those obtained using only one. Frogs have four noses, a wet nose and a dry nose on each side of their heads. When breathing air, they use the dry noses, and when submerged, the wet noses. The receptors in their wet noses more closely resemble those used by fish. Coming back to human olfaction, bones known as nasal turbinates are located in each nasal cavity. The role of these is to cause turbulence in the airflow and hence increase the efficiency of contact between inhaled air and the surfaces of the nasal cavity. The surface of the nasal cavity is suffused by nerve endings of the trigeminal nerve. This nerve is part of the body’s mechanism for detecting irritants, heat, and cold and so will respond to various molecules in the inhaled air. About 70% of odorous molecules also stimulate the nasal trigeminal nerve, so this is part of the total phenomenon of odour perception. The most important organ, as far as detection of odorous molecules is concerned, is the olfactory epithelium. This is a patch of greenish yellow tissue, several square centimetres in area and 100–200 μm thick. It is found on the roof of the nasal cavity and runs down a little onto the septum. The epithelium contains the receptor cells, and each of these possesses a number of hair‐like projections known as cilia. The cilia are 20–200 μm long and are bathed in a mucus layer that is 35 μm thick. The mucus flows backwards at a rate of 1–6 ml/min. The receptor proteins sit in the cell wall, in the cilia. Each receptor cell only produces one type of receptor protein, and the other end of the receptor cell is located in the olfactory bulb. Individual receptor cells fire spontaneously at a rate of 3–60 impulses/ second, and stimulation by an odorant increases this rate of firing. The mucus contains, among other things, proteins known as odour binding proteins (OBPs) and metabolic proteins such as oxidative enzymes known as cytochrome P450s and hydrolytic enzymes, and their functions are described below. The presence of the mucus is important, and it has been shown that, even in electronic noses, an artificial mucus improves the sensitivity of the instrument, presumably by increasing the level of adsorption of the odorant. The roof of the nasal cavity takes the form of a bony plate known as the cribriform plate. This plate contains channels that allow the receptor cells to extend through it. Vertically above the epithelium and on the other side of the cribriform plate is the olfactory bulb. We each have two olfactory bulbs: one on the

245

246

13  The Mechanism of Olfaction

right that receives signals from the right epithelium (and hence the right ‘nose’) and another on the left that receives its signals from the left ‘nose’. Odours learned using only one nostril are recognised when later presented to the other, so we thus know that information from the two bulbs is exchanged at a higher level in the brain. All of the nerve endings from a given type of receptor cell converge onto the same area of the bulb, and these centres are known as glomeruli. The glomeruli corresponding to receptor types with similar selectivities are located close to each other on the bulb. The signals leaving the bulbs go simultaneously to the brain regions known as the piriform cortex and the amygdala. Nerve pathways from these areas to the thalamus and orbitofrontal cortex are complex and flow in both directions as will be seen later.

­The Process of Olfaction Transport to the Receptors A molecule can only be detected by the olfactory receptors if it can come into physical contact with them. Since they are located high up in the roof of the nasal cavity and only exposed to gases/vapours that flow through the nose, this means that the molecule must be volatile enough to produce a sufficient concentration in the air entering the nose. In general, this means that it must have a molecular weight below 300 Da, which roughly equates to an empirical formula containing fewer than 20 carbon atoms, as larger molecules are too heavy to enter the vapour phase at ambient temperature and pressure. Molecules that are very polar will form strong non‐bonded interactions with the medium containing them, which will lower their ability to break free from the solid or liquid phase and enter the gaseous state. Thus, molecules with too low a log  P are unlikely to be volatile enough to reach the nose. Odorous molecules generally have log P values in the range 2–7. The nature of the medium containing the perfume molecule will affect its actual volatility because of non‐bonded interactions. For example, a molecule that is volatile enough to leave a paper blotter and move into the vapour space around it might interact strongly with a component in a shower gel. This will result in a reduction in the vapour pressure of that ingredient in the headspace over the gel relative to that which would be expected based on its performance on a perfumer’s blotter. Thus, the overall impression of a perfume containing it will be distorted on moving from the one medium to another. This is why perfumes should always be evaluated from the product which will finally contain them. Perfumers learn from experience how an ingredient will behave in different product bases and will adapt their formulae to suit the application. Physical chemical calculations of vapour pressure can, of course, aid the perfumer in doing this. After reaching the airspace of the nasal cavity, the odorous molecules must cross the aqueous layer of the mucus in order to reach the receptor cells. It is possible that they do this through simple diffusion. The mucus contains proteins known as OBPs. These proteins belong to the family known as lipocalins, which are proteins normally associated with removal of unwanted molecules from the

 ­The Process of Olfactio

body. Their role in olfaction could therefore be purely removal of excess odorant. Although it is possible that they might play a role in transporting odorant molecules (particularly the less water soluble ones) across the mucus to the receptors, any more active role in perception is unlikely since experiments using receptor proteins in cell culture indicate that the OBPs are not necessary for signal generation. Consequently, their role is probably to remove odorants from the epithelium and therefore continually refresh the population of odorants in the mucus, giving the time factor that is needed for survival. The metabolic enzymes include hydrolytic enzymes (that hydrolyse esters and peptides), and oxidative enzymes such as those belonging to the group of cytochrome P450s serve the same purpose, but one consequence of these is that the receptor proteins sometimes detect the products of metabolism as well as the original odorant. One well‐ established example is that of a woody odorant that is oxidised by a specific P450 enzyme to a molecule that smells of raspberry. This means that people lacking that enzyme smell wood, whereas those with the enzyme describe the smell as raspberry. The Receptor Event Olfactory receptor proteins belong to a family known as 7‐transmembrane G‐ protein‐coupled receptors or GPCRs. Members of this family also include the receptors for adrenalin (adrenergic receptors or adrenoceptors for short) and morphine (opioid receptors). The gene family coding for the olfactory receptors is the largest in the genome and contains information for synthesis of over 1000 different receptor proteins. The receptor proteins sit right across the cell wall, and the chain of their protein backbones crosses the cell wall seven times. In each of the transmembrane sections, the secondary structure is that of an α‐helix. One end of the protein chain (the one with a free amino group) sits outside the cell, and the other (the one with a free carboxylic acid group) is found inside the cell. There are three lops outside the cell and three inside. This basic structural outline is shown in Figure 13.2. This is the case for mammalian receptors; insect receptors have amino terminus of the sequence inside and the acid terminus outside the cell. NH3+

Extracellular medium

Cell wall

CO2– Intracellular medium

Figure 13.2  A seven‐transmembrane G‐protein‐coupled receptor.

247

248

13  The Mechanism of Olfaction

7

6

1

5 BP

2 3

4

Figure 13.3  View of an olfactory receptor protein from above the cell wall showing the location of the binding pocket.

In forming the tertiary structure of the receptor proteins, the seven transmembrane helices come together to form a column running right across the cell membrane. This column has a pocket at the core of the structure, closer to the extracellular surface, into which odorant molecules can be bound by non‐bonded interactions. Not surprisingly this region of the structure is known as the binding pocket. Examination of variability in DNA sequences (so‐called homology modelling) and computer model studies suggest that it is most likely to be formed between helices 3, 4, 5, and 6. This is shown in Figure 13.3 which is a schematic view, looking down onto the cell membrane from outside the cell. Experimental evidence in support of this concept of the odorant binding site can be found in the crystal structures (determined by X‐ray crystallography) of rhodopsin and the β‐adrenergic receptor. Rhodopsin is the pigment in the retina that detects light and hence gives us the sense of sight. In order to detect light, a molecule of vitamin A is chemically bound to rhodopsin since it is the vitamin A that actually senses light. Thus, when the structure of rhodopsin was determined, it contained a molecule of vitamin A in the binding site. In order to obtain a crystalline sample of the β‐adrenergic receptor, the crystallographers had to add a molecule that would bind to the active site and thus stabilise the protein structure. Such molecules are known as ligands. Rather than use the natural ligand, adrenaline, they used a synthetic molecule called carazolol, which was chosen to help makes the crystallisation easier. These results, together with those from numerous experiments combining computer modelling of the receptor proteins with observation (either direct or indirect) of receptor activation, enable us to be fairly confident that the event that activates the receptor protein is binding of an odorant molecule into the binding pocket through non‐bonded interactions such as hydrogen bonding and space‐filling hydrophobic interactions. The receptor protein is coupled to another protein called a G‐protein, which is really three smaller proteins linked together. When an odorant binds to the receptor protein, it changes its structure in such a way that the G‐protein breaks and splits into its three component parts. The mechanism by which the binding of an odorant releases the G‐protein (Golf in the case of olfaction) is known as the global toggle switch, and a simple schematic of its mode of action is shown in Figure 13.4. The process is rather like a scissor action: as the outer surface of the receptor protein closes around the odorant to bind it, the interior surface is forced open. One part of the G-protein binds to the gap and the other fragments break away from it. Once the G‐protein has broken up, one of the component parts then interacts with an enzyme in the cell interior. The enzyme can be either adenylyl cyclase or phospholipase C. These enzymes are converted from inactive to active forms by the action of the G‐­protein fragment and begin to catalyse chemistry in the cell.

 ­The Process of Olfactio Odorant Odorant binding pocket

Lipid membrane

Extracellular surface

Odorant binding

G-protein

Intracellular surface

Figure 13.4  The global toggle switch.

One of the reactions catalysed by adenylyl cyclase is to convert adenosine triphosphate (ATP) to cyclic adenosine monophosphate (cAMP), and similarly, phospholipase C produces inositol triphosphate (IP3). These two small molecules, cAMP and IP3, are what are called second messengers because they now carry the chemical signal on to yet another protein, an ion channel. Like the olfactory receptor proteins, the ion channel proteins sit right across the cell wall. As their name implies, they provide a channel that allows ions to cross through the non‐polar cell wall. The ion channels are not open all the time, but are closed at the start of the signal cycle, and ions cannot pass through. The second messengers act as gatekeepers and open the gate of the ion channel. Calcium ions now begin to flow into the receptor cell, which creates an electrical imbalance between the interior and exterior of the cell. This action all takes place in the cilia on the nasal side of the receptor cell, but, as a result, the whole cell becomes depolarised, which enables an electrical discharge to occur at the other end of the cell, in fact right in the olfactory bulb. The chemical message in the nose therefore becomes an electrical signal in the brain in the space of a few milliseconds. This sequence of events is illustrated schematically in Figure 13.5 using adenylyl cyclase and the ATP–cAMP conversion as the second messenger system. The Combinatorial Nature of Olfaction In 1973, Ernst Polak proposed that smell is a combinatorial sense, and this was confirmed experimentally in 1999 by Linda Buck and co‐workers who showed that each type of odorant can activate a range of olfactory receptor types and each type of receptor responds to a variety of odorants. Thus, instead of a simplistic situation in which activation of a given receptor type would be directly interpreted as indicative of a certain odour, the brain is presented with a pattern of receptor activation, which it compares with patterns it has seen previously. Identifying a pattern as matching one from the brain’s library of odour patterns then constitutes recognition of the odour. It is easy to see how this allows us to have a very wide repertoire of learned odours, indeed many more than the

249

250

13  The Mechanism of Olfaction Odorant

Ion channel Receptor protein Cell wall

G-protein Adenylyl cyclase

NH2 N

N

HO

P

O OH

P

O

OH P

O O P O HO

O

O HO

OH

Ca2+

N N

N

N

N

OH O O O

NH2 N

O OH

Cyclic adenosine monophosphate (cAMP)

Adenosine triphosphate (ATP)

Figure 13.5  The sequence of events in converting a chemical signal (odorant) in the nose to an electrical signal in the brain.

­ umber of receptors, since every combination of receptor signals represents a n different odour. But the complexity does not end there, as we shall see shortly. The genome contains codes for over 1000 types of olfactory receptors. Most mammals use 800–900 of these, but the five primate species mentioned used much smaller sets. Each human uses 350–400 of these, and statistically it is very unlikely that any two humans (other than identical twins) use exactly the same combination of receptor types. Thus, even at this basic level, the sense of smell is unique to each individual, that is, we each have our own olfactory view of the universe. Techniques are now available for cloning olfactory receptor proteins into cells, including human cells, in cell culture. Mechanisms for indicating when the receptor has been activated can also be cloned into the cell. This now allows researchers to screen receptors against libraries of odorants to determine which receptors respond to which odorants. There is a database, held at the Weizmann Institute in Israel, of all known human olfactory receptors, and it is known as the human olfactory receptor database or HORDE. A number of research groups have published results of screening human receptors against odorant libraries, and all have found that receptors vary widely in their selectivity. Figure 13.6 shows the sort of result that such researchers find. The map shown in Figure 13.6 is constructed for purely illustrative purposes and does not represent any real map. In it we see what might be the result of screening 10 receptors against 25 odorants. The receptors are labelled A–J along the top and the odorants 1–25 on the side. The shading in each square represents the strength of the response of a receptor to an odorant. For example, receptor A responds strongly to odorants 1 and 2. Receptor A is what is known as a broadly tuned receptor because it responds to 11 of the 25 odorants tested against it. Receptor J is a narrowly tuned receptor since it responds to only one of the odorants, albeit strongly. Compounds

 ­The Process of Olfactio A

B

C

D

E

F

G

H

I

J

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Figure 13.6  A typical receptor odorant screening map.

6, 12, and 22 have not given any response to these 10 receptors, and there must therefore be other receptors that do respond, assuming that these three compounds do elicit odours. Receptor I is known as an orphan receptor because, although its gene sequence is known and it can be cloned into cells, no odorants have been identified yet that are capable of activating it. Figure 13.6 shows how complex the interpretation of an odour signal is, especially when we remember that there are about 400 receptor types in the nose and an almost infinite variety of odorant molecules. The picture is even more complex because each receptor type can have a number of variants. For instance, there might be what is known as a single nucleotide polymorphism (SNP) in the genetic code for that protein. This means that different individuals might have one nucleotide difference in the total code for the protein. That difference will result in the synthesis of a receptor protein that differs by one amino acid in the sequence because one of the 3‐letter ‘words’ has been changed. Such a difference might alter the receptive range of the protein. There are definitely examples of this happening. Looking through HORDE reveals that in some cases up to 14 different variations exist for one

251

252

13  The Mechanism of Olfaction

receptor type. If the average were found to be 10 variants per gene, then we would each be using 350–400 receptor types from a pool of 4000. One very important point we learn from this section and the next is that the phenomenon that we call smell only exists in the higher parts of the brain. For example, smell does not exist at the level of the receptor cells. For a long time, chemists (and others) have been tempted to correlate molecular structure with odour and to try to predict the odours of novel molecules before synthesising them. This is purely a statistical exercise since there is no simple correlation between activation of a single receptor type and the ultimate percept of odour in the higher brain. In other words, we can guess the odour of the novel molecule by analogy with known odorants of similar structure, but we cannot predict it with a high degree of certainty. The Perception of Odour Figures 13.7–13.10 show a simple illustration of how the combinatorial nature of olfaction operates. In Figure 13.7 we see an odorant; we will call it the red odorant, approaching an array of six different receptor types. The molecules of the red odorant activate the receptors to different levels. They activate receptor types 1 and 2 to a moderate level. They activate receptor type 4 only weakly but have a strong effect on receptor type 6. They leave receptor types 3 and 5 ­unaffected. This generates a signal pattern in the olfactory bulb as shown schematically in the bar chart. The pattern then proceeds on through the higher brain and is recognised in the orbitofrontal cortex as corresponding to the odour, which it has learned to call ‘red’. Similarly, in Figure 13.8, we see the blue odour activating receptors 1, 5, and 6. This gives a different pattern in the olfactory bulb, and the higher brain recognises this as ‘blue’ by comparison with stored (i.e. learnt) Receptors

1 2

3 Odorant

4 3.5 3 2.5 2 1.5 1 0.5 0

1

2

3

4

5

Signal processing

4

5

6

Figure 13.7  Perception of one odorant.

Cognitive brain

6

 ­The Process of Olfactio Receptors

1

2

4 3.5 3 2.5 2 1.5 1 0.5 0

1

2

3

4

5

6

Signal processing 3 Odorant 4 Cognitive brain

5

6

Figure 13.8  Perception of a second odorant.

­ atterns. If a mixture of the two odorants is presented to the nose, we might p expect to see the picture shown in Figure 13.9 in which the pattern in the bulb is the simple sum of those of ‘red’ and ‘blue’, which might be recognised as ‘purple’. However, life is more complicated than that. Several research groups have reported antagonism of one odorant by another. In other words, the presence of one odorant prevents a receptor, which would normally recognise the second

Receptors

1

2

16 14 12 10 8 6 4 2 0

1

2

3

4

Odorant 3

4 Odorant

5

Signal processing

5

6

Figure 13.9  Perception of a mixture of two odorants.

Cognitive brain

6

253

254

13  The Mechanism of Olfaction Receptors

1

2

16 14 12 10 8 6 4 2 0

1

2

3

4

5

6

Signal processing

Odorant 3

4 Odorant

5

Cognitive brain

6

Figure 13.10  Perception of a mixture of two odorants with interactions at receptor level.

odorant, from doing so. Thus, our perfume formula of ‘red’ + ‘blue’ might actually behave more along the lines shown in Figure 13.10. Here we see that the molecules of the ‘blue’ odorant block receptor 2 so that it can no longer recognise the ‘red’ odorant. The signal that passes on to the brain is therefore a different signal; we will call it ‘purple 2’ that is distinct from both the ‘red’ and ‘blue’ signals and from the ‘purple’ sum that we might have expected by combining ‘red’ and ‘blue’. In these examples, we are considering only six types of receptors and two types of odorant molecule. In the human nose, there are 350–400 types of receptors and a potentially infinite variety of odorant molecules. It is clear therefore that the signal patterns leaving the olfactory bulb are very complex indeed and the ability of the brain to analyse and recognise them is truly awesome. The complexity of the olfactory process does not stop at the olfactory bulb. Figure 13.11 shows the main pathways that the signals follow through the brain up to the orbitofrontal cortex where the conscious awareness of odour exists. Signals from the olfactory bulb are sent to the piriform cortex and the amygdala. The latter is part of the limbic system that is associated with emotions and memory, so this probably accounts for the fact that the sense of smell has rapid and sometimes dramatic effects on memory and emotions. Signals that have been processed in the piriform cortex are passed on to the amygdala and the thalamus, and all three of these parts of the brain exchange signals with the orbitofrontal cortex, thus generating a complex network of interactions as shown in Figure 13.11. Odour signals are processed initially in regions of the brain that are dedicated to odour, and there is also a region dedicated to storage of learnt odour patterns. However, they are connected to the regions dealing with other senses such as sight and hearing, and the brain compares input from all the senses before making a judgement on what is being observed. When nerve signals from

 ­The Process of Olfactio

Orbitofrontal cortex

Thalamus

Piriform cortex Olfactory bulb

Amygdala

Receptor cells

Figure 13.11  Major neural pathways in olfactory perception.

two senses are congruent (i.e. both agree on the nature of the total stimulus), there is enormous (over 1000‐fold) amplification of the combined signal that goes on up to the higher parts of the brain. For example, someone smelling a rose perfume while looking at a picture of a rose is much more likely to correctly and confidently identify the smell than if they were exposed only to the smell without the visual clue. This also explains the inability of wine experts to correctly identify a white wine when a tasteless red dye has been added to it. The conflict between the visual stimulus (i.e. the red colour) and the odour stimulus (i.e. the bouquet of a white wine) leads the brain to make wrong conclusions about the olfactory signal. Figure 13.11 shows that five of the eight neural connections are two way and the red wine/white wine experiment is an example of this in action. The visual stimulus is received first as the subject sees the wine before tasting it. This then causes the higher brain centres to anticipate a red wine bouquet, and the feedback down the olfactory pathway is geared to the wrong terms of reference as a result. In this way both context and expectation affect the interpretation of the pattern of signals coming in from the olfactory bulb. Experience is also important in decoding the signal pattern since the more stored memories of odour patterns there are, the more readily and accurately the brain will be able to match an incoming pattern. Language is also important in odour recognition, and development of a good odour language is a vital part of a perfumer’s or evaluator’s training. The human brain seeks patterns and is exceptionally good at recognising them even if the material is incomplete or degraded. For example, a mixture of benzyl acetate, indole, and hexylcinnamaldehyde will give a distinct impression of jasmine, if the ratios of the three are within a certain range. The olfactory pattern elicited at the receptors by this mixture will clearly be different from that of real jasmine absolute, but the match is close enough for the brain to make the connection between the two odour impressions. This example also illustrates how each odour represents a separate image in the brain. When benzyl acetate mol-

255

256

13  The Mechanism of Olfaction

ecules contact the olfactory epithelium, the odour image in the brain is one that has a distinctly fruity character. Similarly, hexylcinnamaldehyde gives a fatty odour image, and indole a faecal one. Add the three together, and we get a fourth image rather than a sum of the three individual ones, and the nearest match is in the jasmine area. It takes considerable training to develop the skills to deconvolute the composite image into its component parts and to realise that the jasmine odour does contain fruity, fatty, and faecal components. This skill is an important part of a perfumer’s training. In summary, the process of odour perception begins when odorant molecules enter the nose and are recognised, through stereoelectronic (that is, the electron distribution in space) interactions, by the receptor proteins in the olfactory epithelium. The resultant nerve signals are mapped onto the olfactory bulb, and the output from this part of the brain is analysed by a variety of other brain structures before finally being interpreted as an odour in the cortex. Odour is not a molecular property; rather, odorants elicit an electrical stimulus in one apart of the brain that is used, along with other inputs, to construct a mental image higher in the brain. It is that mental image that is ‘odour’ and the image perceived by an individual are unique to that individual at that point in time. In terms of intensity, character, and hedonics, this interpretation will also be dependent on expectation, experience, and context. Odour perception is extremely complex, and it is important for those working in the industry to appreciate this and to understand the basic causes of this complexity. Albert Einstein said that, ‘The most beautiful thing we can experience is the mysterious. It is the source of all true art and science’. Despite the very significant recent advances in our understanding of it, olfaction still holds plenty of mystery and therefore provides a fertile ground for both artistic and scientific exploration. Deeper coverage of all aspects of olfactory perception can be found in the book Chemistry and the Sense of Smell, details of which are in the bibliography.

Review Questions 1 Why do we lose our sense of smell when suffering from a cold? 2 Why do we lose the sense of taste when we have a cold? 3 Why is it difficult to predict what a mixture of two odorants will smell like? 4 We each have a unique perception of the olfactory universe. How does this come about?

257

14 Natural Fragrance Ingredients Chapter  12 introduced the concepts of primary and secondary metabolites. This chapter is concerned with the latter category, that is, the secondary metab­ olites, into which the vast majority of natural fragrance chemicals fall. After covering nature’s reasons for producing odorous molecules, we consider how it does so and why this leads to families of fragrant molecules. Hence, we come to understand what is meant by the terms terpenoids, shikimates, and polyke­ tides. Organic chemicals are not stable in the environment and are degraded both by enzymic action and by simple chemical reactions. Some of the products of such degradation are odorous, and we shall also discuss the more important of these.

­Why Does Nature Produce Odorous Chemicals? Production of chemicals requires expenditure of energy and material and thus places a burden on the organism that does so. Therefore, unless there is a sur­ vival advantage, the genes responsible for production of secondary metabolites will be bred out. The benefits of producing odorous secondary metabolites are varied and include chemical communication and chemical defence. Substances used for chemical communication between the cells of one single organism are known as hormones, those used to communicate between different organisms are known as semiochemicals. Semiochemicals, which enable communication between two members of the same species, are known as pheromones, and those that carry messages from one species to another, as allelochemicals. The bulk of naturally derived fragrance derivatives fall into the categories of allelochemicals and defence chemicals. The flower oils are allelochemicals. Thus, such odorants as geraniol and linalool are made by plants in order to attract insects as pollina­ tors. The woody and balsamic notes are usually defence chemicals. They are pro­ duced in response to injury, and their role is to form a protective barrier, both physical and chemical, to prevent fungi and bacteria from entering the wound and damaging the plant. For example, myrrh is a secretion of the tree Commiphora abyssinica. When damaged, the tree produces a secretion that hardens (by air‐ catalysed polymerisation) into a glassy solid and seals the wound. However, as a further avenue of defence, the secretion contains a number of volatile chemicals Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

258

14  Natural Fragrance Ingredients

Figure 14.1  An antimicrobial sesquiterpenoid present in myrrh.

such as the sesquiterpenoid (see later for a definition of this term) shown in Figure 14.1, which has antibacterial and antifungal properties.

­ asic Principles of Biosynthesis: Enzymes B and Cofactors Biosynthesis is the word used to describe the building up of chemical substances in living organisms. It is also known as biogenesis. In synthesising chemicals, nature uses the same reactions that chemists do. The catalysts that it employs in doing so are enzymes, which have already been introduced in Chapter 12. These are globular proteins containing an active site that holds all the reacting species together and thus lowers the activation energy. In general, enzyme‐catalysed reactions are fast and selective. If the enzyme catalyses acid/base reactions such as ester hydrolysis, then the enzyme alone is probably all that is required in addi­ tion to the reagents. However, many enzymes need cofactors as reagents or energy providers. Cofactors are small molecules that play a vital role as reagents in the reaction and usually need to be recycled afterwards as, unlike enzymes, they are chemically changed by the reaction. Three cofactors are particularly important in the biosynthesis of fragrance components. Here we will give a brief introduction to these three species, and later in the chapter we will see many examples of their use in different biosynthetic reactions. Adenosine triphosphate, usually abbreviated to ATP, is shown in Figure 14.2. The structure comprises the nucleotide adenine attached to a ribose unit and three phosphate units attached to the ribose. The tri‐ester function is high in energy and readily loses one of the phosphate units to form adenosine diphos­ phate (ADP). During this process, an alcohol becomes phosphorylated (that is, esterified with phosphoric acid) as shown in the figure. The phosphate anion is a good leaving group, and so the phosphorylated alcohol becomes an electrophile and can be alkylated by nucleophiles as shown in the figure. Nicotinamide adenine dinucleotide phosphate, NADP for short, contains two nucleosides connected through the sugars by a diphosphate bridge as shown in Figure 14.3. One of the nucleotide units, derived from nicotinamide, contains a quaternary ammonium salt. This can be reduced to a dihydropyridine ring sys­ tem by addition of the equivalent of a hydride anion (H−). The reduced form is known as NADPH, the H indicating the reduction owing to the addition of a

­Basic Principles of Biosynthesis: Enzymes and Cofactor NH2 N

N

ATP

N

N

HO HO

O

O

O

P

P

P

O HO

O HO

O O HO

OH

RCH2OH

R

O

O P

OH

+

ADP (adenosine diphosphate)

OH

Nu.

Nu– HO R

Nu

OH P

O

O–

Figure 14.2  Adenosine triphosphate (ATP).

hydride anion. Figure 14.3 shows how the NADP/NADPH system can be used to oxidise an alcohol to a ketone or reduce a ketone to an alcohol. Nicotinamide adenine dinucleotide (NAD) is another cofactor that fulfils a very similar role but differs from NADPH by possessing only two phosphate groups in its structure, the two that hold the two nucleotides together. The other of these three key cofactors is known as coenzyme A or CoA and is shown in Figure 14.4. The crucial part of CoA, as far as biosynthetic reactions are concerned, is the thiol group at one end of the molecule. This cofactor can form thioesters with carboxylic acids as shown in the figure. The electron‐­withdrawing effect of the thiol group polarises the carbonyl group further and makes the car­ bonyl carbon atom more susceptible to attack by nucleophiles. It also has the effect of increasing the acidity of the hydrogen atoms next to the carbonyl carbon atom and therefore making it easier to form the corresponding carbanion. Both of these possible reactions are shown in the figure. In all of these cofactors, the polar groups of the nucleotides, the sugars, and the phosphates serve as centres for hydrogen bonding and hence for recognition by the enzymes they serve. In Figure 12.21, we saw NADPH about to reduce the anion of pyruvic acid to one of lactic acid. That figure shows clearly how the

259

260

14  Natural Fragrance Ingredients NH2 O

N

NADP

NH2

N

N N+

+ R

H

O O

OH

O

O

P

P

O

R′

O O

OH

HO HO

N

O

HO

OH

P

OH OH

O

NH2 O

NH2

N

NADPH

N

N N

+ R

O

+

HO

H+

O

P

P

O HO

O R′

O O

N

O O

OH O

HO

OH

P

OH OH

O

Figure 14.3  Nicotinamide adenine dinucleotide phosphate (NADP). NH2 O HS

N H

N H

N

N

O

HO OH O O P P O O O HO HO P

OH

Coenzyme A

N

N O

OH

O

O O H

OH

H H O

O – H H

S

H

CoA Base

Figure 14.4  Coenzyme A.

H H

S

O

CoA H Nu

H H

Nu

+

HSCoA

­General Pattern of Biosynthesis of Secondary Metabolite

cofactor and substrate are both recognised by the enzyme and held together in the geometric relationship required to enable the reaction to proceed.

­ eneral Pattern of Biosynthesis of Secondary G Metabolites Figure 14.5 shows the more important routes in the general overall scheme of biosynthesis of secondary metabolites in plants. Green plants convert carbon dioxide from the atmosphere, water, and sunlight into glucose through a process known as photosynthesis. Atmospheric carbon dioxide contains some of the radioactive isotope 14C, and thus radioactivity is introduced into all plant mate­ rial by this route as discussed in Chapter 6. The glucose molecule can be broken down by a process known as glycolysis (from the Greek for splitting of sweet material), and one of the fragments produced is phosphoenolpyruvate. Reaction of this with erythrose phosphate (another product of glycolysis) gives a substance known as shikimic acid. This acid is the key intermediate for the shikimate family of natural products, including a substantial number of fragrance components. Shikimic acid is also the precursor for coumarins, flavonoids, and lignins. Some coumarins are odorous, and some are undesirable components of essential oils, such as bergaptene, which is a phototoxin as explained in Chapter 12. Lignin, one of the principal structural materials of plants (cellulose being the other), is also derived from shikimic acid. If, instead of erythrose phosphate, the phosphoe­ nolpyruvate reacts with coenzyme A, the product, after loss of carbon dioxide from the intermediate, is acetyl coenzyme A, or acetyl CoA for short. Coupling of this through aldol‐type reactions, as already mentioned in Chapter 12, gives the polyketides and lipids. Acetyl CoA is also a key intermediate in the biosyn­ thesis of mevalonic acid, and this, in turn, is a key intermediate in the biosynthe­ sis of the terpenoids. This family of natural products provides by far the largest Carbon dioxide + Water Glucose Green plants + Sunlight Glycolysis O

OH

OP

Lignans coumarins flavonoids

–

CO2

Phosphoenolpyruvate

OH

HO OH

Coenzyme A

Polyketides lipids

Shikimic acid O

CoA Acetyl coenzyme A

O HO

OH OH

Mevalonic acid

Figure 14.5  General pattern of biosynthesis of secondary metabolites.

Terpenoids

261

262

14  Natural Fragrance Ingredients

number of natural fragrance ingredients. The next four sections of this chapter look at each of these pathways in more detail. The fragrance chemicals produced by plants fall into structural families, and relationships between different sub­ stances are clearly apparent. The reason for this order is the way in which plants put them together. Molecules that come off a biosynthetic ‘production line’ at different points still bear the hallmarks of that ‘production line’. By inspection of the molecular structure of a plant product, we can make a good guess as to how the plant put the molecule together. In fact, botanists use the knowledge of which chemicals a plant produces in order to help in classification of the plant, since the products of biosynthesis tell us something about which enzymes are active in the plant and hence about the genetic make‐up of the plant. This approach to clas­ sification is known as chemotaxonomy.

­Polyketide Biosynthesis In Figure 14.6, we see the basic steps of polyketide biosynthesis. One molecule of acetyl CoA reacts with a base to form the enolate anion and displaces the coenzyme from a second molecule of acetyl CoA. This reaction generates the CoA ester of a β‐keto acid, which contains two more carbons in its chain than did the starting acid. Of course, this new material also contains a CoA ester function, which can react with the enolate anion derived from a third acetyl CoA molecule. In this case, the product is a β, δ‐diketo acid containing four carbon atoms more than the starting material. Obviously, repetition of the pro­ cess will lead to the build‐up of longer chains with ketone functions on alternate carbon atoms, hence the name polyketides. Polyketide chains are reactive molecules because methylene groups sand­ wiched between carbonyl functions are readily deprotonated, and in a polyketide O

O

+ CoA

R

CoA

–

Condensation O

O

R

CoA O CoA

O R

O

O CoA

Figure 14.6  Polyketide biosynthesis.

­Lipid Biosynthesi O

O

– O

OH

O

–H2O OH

OH O

Aldol reaction

HO Enolise

O

O

O

OH

OH

Aldol O condensation

O

OH

O

Orsellinic acid

HO O OH

O

Methyl 3-methylorsellinate

Figure 14.7  Further steps in polyketide biosynthesis.

chain, electrophilic carbonyl groups are always within easy reach of the resultant enolate anion. Figure 14.7 shows a typical reaction of this type. The starting material is the tri‐keto‐octanoic acid formed from stepwise reaction of four acetyl CoA molecules. Deprotonation of the second carbon atom in the chain gives an anion, which is well placed to add to the electrophilic carbon atom sev­ enth in the chain. (It is a general rule in organic chemistry that reactions forming five‐ or six‐membered rings or involving six‐membered transition states are highly favoured.) Dehydration of the aldol reaction product gives the aldol con­ densation product. If the two ketone functions in this molecule tautomerise to give the enol forms, the result is an aromatic system, and this molecule is known as orsellinic acid. One of the most important perfume ingredients in this family of natural products is a derivative of orsellinic acid in which a methyl group is added to the ring and the acid function is esterified with methanol. This material, methyl 3‐methylorsellinate, is the most important component, in odour terms, of oakmoss and treemoss extracts. A number of the minor components of these extracts are sensitisers, and so the synthetic material, which is free of such unde­ sirable properties, is preferred on safety grounds. Members of the polyketide family of natural products usually retain evidence of oxygenation at alternate carbon atoms, which is how to recognise them as such. The evidence could be the presence of an alcohol, phenol, ketone, acid, or ester group, or it could be a double bond resulting from an aldol condensation.

­Lipid Biosynthesis If, instead of adding to another molecule of acetyl CoA, the β‐keto ester shown in Figure 14.6 suffers reduction of the ketone group by NADH, the resultant prod­ uct is the corresponding β‐hydroxy ester. This can be dehydrated to give the cor­ responding α,β‐unsaturated ester. Addition of ‘hydride’ to the β‐position followed by protonation of the α‐position then gives an ester with a carbon chain, which is two carbons longer than the original. This sequence of reactions is shown in

263

264

14  Natural Fragrance Ingredients O

O

NADH

OH

O

CoA

O

–H2O CoA

O

NADPH

CoA

CoA

Figure 14.8  Reduction of a polyketide to form a fatty acid.

Figure 14.8, and it is the basis of lipid biosynthesis. The whole reaction sequence can be repeated, adding another two carbon atoms to the chain, and further additions can and do occur, always with addition of two carbon atoms, hence the predominance of even‐numbered fatty acids in nature. The most common fatty acids are those with 16 or 18 carbon atoms in their chains. These materials and higher homologues have no odour. Perfume ingredients (and malodours) derived from lipids are therefore mostly degradation products, which will be described later. One important exception to this generalisation is the jasmine family of materi­ als. These are synthesised from a fatty acid known as arachidonic acid as shown in Figure 14.9. Arachidonic acid contains 20 carbon atoms and four double bonds. Enzyme‐controlled autoxidation of arachidonic induces both oxygena­ tion and cyclisation to form a five‐membered ring in the centre of the molecule. This step is important in the formation of some animal and plant hormones including the prostaglandins, which are of importance in the animal kingdom, and jasmonic acid, which is a plant growth hormone. Methyl jasmonate, the methyl ester of the latter, is an important contributor to the odour of jasmine flowers, as is jasmone, a closely related analogue, also produced in this way from arachidonic acid.

O

O H

O

OH

OH HO

O

O

O O

Arachidonic acid

O

O O O

OH

Jasmone

Methyl jasmonate O

O O HO

O

Figure 14.9  Arachidonic acid and the biosynthesis of jasmonic acid derivatives.

­The Shikimic Acid Pathwa

­The Shikimic Acid Pathway Figure 14.10 shows some of the key intermediates in the shikimic acid pathway. This biosynthetic pathway takes its name from shikimic acid, which is synthe­ sised in six steps from phosphoenolpyruvate and erythrose 4‐phosphate, as already described above. A further molecule of phosphoenolpyruvate is added to shikimic acid, and, after several further steps, prephenic acid is obtained. In these two precursors, namely, shikimic acid and prephenic acid, we can see the pattern of a six‐membered ring carrying a one‐ or three‐carbon side chain attached to carbon‐1 of the ring and oxygenation at carbons 3, 4, and/or 5 of the ring. These features characterise the members of the shikimic acid family. However, the oxygen atoms present in the final product are not those that were originally present in shikimic acid. The original oxygen atoms are all lost in the biosynthesis, and new oxygen atoms introduced at a later stage. The new oxy­ gen atoms are sometimes introduced at carbon‐2, but carbons 3, 4, and 5 are the more common sites of oxygenation. Phosphoenol pyruvate PO

OH

CO2H

CO2H

CO2H

O

OH

O HO

O OP

OH OH

OH

OH

Shikimic acid

Erythrose 4-phosphate

Prephenic acid

NH2

NH2

O

O Phenylalanine

OH

OH

Tyrosine

OH O

O

O

OH

OH

OH

HO

OH

p-Coumaric acid

Cinnamic acid

o-Coumaric acid

O

OH

O

O

O

O O

OH

OH

HO

HO Methylenecaffeic acid

HO

Ferulic acid

Figure 14.10  Key intermediates in the shikimic acid pathway.

Caffeic acid

265

266

14  Natural Fragrance Ingredients

Decarboxylation and dehydration of prephenic acid, followed by conversion of the ketone function to an amine, gives phenylalanine, one of the 21 essential amino acids. We now see a feature that recurs in many places in the shikimic acid pathway, namely, different potential routes from one intermediate to another. Which exact route operates in any given plant will depend on its genetic make‐ up, and it is possible that a mixture of routes will operate. At this point in the total pathway, two routes are possible from phenylalanine to p‐coumaric acid. The first involves elimination of ammonia to give cinnamic acid followed by hydroxylation in the para‐position to give p‐coumaric acid. The second route reverses this reaction sequence by first hydroxylating phenylalanine to give tyrosine, another essential amino acid, and then eliminating ammonia to give p‐ coumaric acid. Cinnamic acid can also be hydroxylated in the ortho‐position to give o‐coumaric acid. Hydroxylation of p‐coumaric acid at one of the meta‐­ positions gives caffeic acid, and methylation of this oxygen atom produces feru­ lic  acid. Oxidation of the methyl group and coupling to the adjacent phenolic hydroxyl group produces a five‐membered ring, and the resultant acid is known as methylenecaffeic acid. The role of these intermediates in producing odorant molecules will be discussed in the following paragraphs. Methyl anthranilate, 2‐phenylethanol, and indole are all products derived from prephenic acid by different routes. The relationship between indole and pheny­ lalanine is obvious from the figure. Indole is derived from degradation of pro­ teins and hence its faecal odour character. However, it is an important contributor to the odour of some floral odours and of jasmine in particular. These structures are shown in Figure 14.11. In Figure 14.12 we see the routes by which a variety of essential oil components are synthesised from cinnamic acid. Reduction of cinnamic acid gives cinnamal­ dehyde and cinnamyl alcohol. Cinnamaldehyde is important to the cinnamon CO2Me NH2 OH

Methyl anthranilate CO2H

O O

OH

OH

2-Phenylethanol

Prephenic acid CO2H NH2 Phenylalanine

N H Indole

Figure 14.11  Odorants from prephenic acid and phenylalanine.

­Terpenoid O

Cinnamic aldehyde

Cinnamic acid

O

HO

OH

O

OH

OH

p-Coumaric acid

o-Coumaric acid

Coumarin

Cinnamic alcohol

O

O

O

OH

O

OH

O

Anisaldehyde

O

O Estragole (methylchavicol)

Anethole

Figure 14.12  Odorants from cinnamic acid.

odour, and cinnamyl alcohol is found in a wide variety of plant oils. Esters of both cinnamic acid and cinnamyl alcohol are also widespread in nature. As stated above, hydroxylation of cinnamic acid can give either the ortho‐ or para‐hydroxy acids. The most important odorant derived from the former is coumarin, the compound responsible for the sweet smell of new mown hay and a key fragrance ingredient. p‐Coumaric acid is the precursor for anisaldehyde. Anisaldehyde has an odour reminiscent of hawthorn blossom. p‐Coumaric acid is also the precur­ sor for the important herbal component estragole (also known as methylchavi­ col) and anethole, a key component of anise, having an odour characteristic of that spice. Reduction of the side chain in ferulic acid gives the isomeric materials eugenol and isoeugenol as shown in Figure 14.13. Eugenol is the major component of oil of cloves and strongly resembles this oil in odour character. Both isomers are found in a variety of flower oils, and eugenol is a particularly important contributor to the odour of carnations. Cleavage of the side chain gives vanillin, the principal odour component of vanilla beans. By a similar set of reactions, methylenecaffeic acid gives safrole (the major component of sassafras oil) and heliotropin, which comprises about 90% of the oil of heliotrope and has an odour closely resembling that of the flower.

­Terpenoids The terpenoids constitute the largest family of natural fragrance ingredients. The first terpenoids to be characterised were obtained from turpentine, and it

267

268

14  Natural Fragrance Ingredients O

O O

O

OH

OH

O

HO

Methylenecaffeic acid

Ferulic acid

O

O

O

HO

HO Eugenol

O

O

O

Safrole

Isoeugenol

O

O

O

HO Vanillin

Heliotropin

Figure 14.13  Odorants from ferulic acid.

is therefore not surprising that that is how the term was derived. Terpenoids are defined as natural products that are built up of isoprene units. Isoprene is a five‐carbon olefin, 2‐methyl‐1,3‐butadiene, and its characteristic structural pattern can be seen in all terpenoid structures. Since the first terpenoids to be identified contained 10 carbon atoms (that is, two isoprene units), they were known as monoterpenoids, and terpenoid nomenclature still operates in mul­ tiples of two isoprene units as can be seen from the table in the adjacent text box. Isoprene itself is therefore classified as a hemiterpene. As far as perfum­ ery is concerned, the most important group of terpenoids is the monoterpe­ noids. Sesquiterpenoids are found in many essential oils and often play a very important part in their odours, but the percentage of sesquiterpenoids with an odour is much lower than that of monoterpenoids, virtually all of which have an odour. The higher terpenoids are generally too poorly volatile to pos­ sess odours. However, some degradation products of higher terpenoids are important perfumery ingredients, as we will see in the section on degradation products. In older literature, the term terpene is sometimes used interchangeably with terpenoid. It is more properly used to describe terpenoid hydrocarbons. In some instances, the term refers to monoterpenoid hydrocarbons, for exam­ple, when speaking of deterpenation of essential oils, what is usually meant is the removal, by distillation, of monoterpenoid hydrocarbons, such as limonene. (Sometimes nowadays, deterpenation or ‘folding’ of oils is carried out using processes other than distillation, solvent extraction, for instance.) When the  term is encountered, its exact meaning must be determined from the context.

­Terpenoid

Classification of terpenoids No. of carbon atoms

No. of isoprene units

Hemiterpenoids

5

1

Monoterpenoids

10

2

Sesquiterpenoids

15

3

Diterpenoids

20

4

Sesterterpenoids

25

5

Triterpenoids

30

6

Tetraterpenoids

40

8

Carotenoids

40

8

Polyisoprenoids

>40

>8

Steroids are a special subgroup of triterpenoids, though not all steroids have 30 carbon atoms. Isoprene

Head Head Tail

Tail

Figure 14.14  Isoprene and the head‐to‐tail linkage in terpenoids.

Figure 14.14 shows the structure of isoprene and also how two isoprene units join together in a head‐to‐tail fashion. If we imagine the isoprene molecule to resemble a tadpole or a fish with the first two carbon atoms and the pendant methyl group representing the ‘head’ and the other two carbons of the chain representing the ‘tail’, then the ‘head‐to‐tail’ fusion is easy to visualise. This link­ age is the most common between isoprene units. Tail‐to‐tail links are found in the central junction of two very special classes of terpenoids, the steroids and the carotenoids. These large molecules contain 30 and 40 carbons, respectively, in their starting skeletons. They are therefore too large to be volatile enough to pos­ sess an odour. In the perfumery industry we come across them mostly in the form of degradation products, so they will be discussed in the sections on degra­ dation products and malodours. In Figure 14.15 we see some typical terpenoids; linalool and limonene, two monoterpenoids; and caryophyllene, a sesquiterpenoid. The structures are shown in the normal way and then with the bonds joining the isoprene units shown as dotted lines. This illustration enables us to see how the units are linked together to give the final product structure. The third representation omits the bonds that were not part of the original chain and thus enables us to trace out the linear isoprene dimer or trimer as appropriate. The last representation shows

269

270

14  Natural Fragrance Ingredients HO

HO

HO

HO

Linalool

Limonene

Caryophyllene

Figure 14.15  Three terpenoids and the isoprene units they contain.

the isoprene units completely disconnected from each other. By working from the last set of structures back to the first, we can see how the isoprene units are first joined together in a chain and then cyclised by further bond formation between atoms in the chain. The three key building blocks of the terpenoid family are mevalonic acid, isopentenyl pyrophosphate, and prenyl pyrophosphate. As can be seen from Figure 14.16, three molecules of acetyl CoA condense to form mevalonic acid. A few species have a biosynthetic route to terpenoids, which does not involve mevalonic acid, but the route as shown in the figure is the predominant one and is the route used by the vast majority of terpenoid producing organisms, both plant and animal. Subsequent decarboxylation, dehydration, and phosphoryla­ tion of mevalonic acid gives isopentenyl pyrophosphate. This molecule can be isomerised enzymically to prenyl pyrophosphate and vice versa. Hydrolysis, elimination, or nucleophilic substitution of either of these pyrophosphates will produce hemiterpenoids. For example, hydrolysis of prenyl pyrophosphate will give prenol, elimination of pyrophosphoric acid from isopentenyl pyrophosphate will give isoprene, and nucleophilic attack by an acetate ion on prenyl pyrophos­ phate will give prenyl acetate. The mechanism of coupling of isoprene units to form linear terpenoids is shown in Figure 14.17. Isopentenyl pyrophosphate is deprotonated by base, and the resultant anion reacts as a nucleophile displacing pyrophosphate from pre­ nyl pyrophosphate to give geranyl pyrophosphate. This reaction can be used to produce monoterpenoids, or it can add another molecule of isopentenyl pyroph­ osphate to give farnesyl pyrophosphate  –  the precursor for all of the

­Terpenoid O 3

O

OH

CoA

S

Mevalonic acid HO

HO O O O

P

O

OH

HO O O

OH P

O

OH

P

O

OH P

OH

Isopentenyl pyrophosphate

Prenyl pyrophosphate AcO–

H2O

–H4P2O7 O

OH

O

Prenol

Isoprene

Prenyl acetate

Figure 14.16  Three key terpenoid building blocks and some hemiterpenoids.

–Base

H

Isopentenyl pyrophosphate O

O Prenyl pyrophosphate

O

P

O

O OH O P

O

OH O

P

OH

P

OH

OH OH –Base

H

Isopentenyl pyrophosphate O

P

O

O OH O O Geranyl pyrophosphate

O

P

O

P

OH

P

OH O

O Farnesyl pyrophosphate

O

OH

O

OH O

P

OH OH

Figure 14.17  The mechanism of coupling of isoprene units.

P

OH OH

271

272

14  Natural Fragrance Ingredients O OH O P P OH O O OH

+ OH

O

Geraniol

Citral

Limonene

O

+

Camphor

p-menthane

OH α-Terpineol α-Pinene

OH

Borneol

+

Camphane (bornane)

+

Pinane

β-Pinene

Figure 14.18  Formation of some typical monoterpenoids from geranyl pyrophosphate.

sesquiterpenoids. Further additions of isopentenyl pyrophosphate will, of course, generate even higher members of the terpenoid family. In Figure 14.18, we see how geranyl pyrophosphate is used to produce various monoterpenoids. The initial step involves cleavage of the phosphate ester link to give a positively charged species, the geranyl carbocation. In the figure, it is drawn as a free cation. However, in reality, the free carbocation is unlikely to form. What happens is that the carbon–oxygen bond is weakened by binding to an enzyme active site, which polarises the bond even more than it is naturally, and so makes the carbon atom behave as if it were a cation. Similarly, the inter­ mediate carbocations will not exist as free species but as enzyme stabilised, par­ tially charged entities. For simplicity we will consider the free carbocations to exist, but the reader should always bear in mind that these species are actually transient in nature and bound in some way to an enzyme active site. If the geranyl carbocation reacts with water, the latter will add as a nucleo­ phile, and, after loss of a proton, geraniol will be formed. Oxidation of the alcohol function of geraniol can then produce citral. Alternatively, the positive charge on carbon‐1 of the geranyl molecule can be trapped by the electrons at the other end of the molecule, which gives a cyclic carbocation that has the p‐menthane skeleton. (The term skeleton is used to describe the basic carbon framework of the molecule, and in the figures, italicised type is used for names

­Terpenoid

of skeleton types, whereas normal type is used for names of individual com­ pounds. Double bonds and positive charges are sometimes included in the figures with italicised names, but it must be pointed out that the name refers only to the saturated, uncharged carbon framework.) Since the electrons of the double bond have moved into the space between carbons 1 and 6, the positive charge is now located on what was carbon 7 of the original chain. Reaction of this carbocation with water produces α‐terpineol, whereas elimination of a proton produces limonene. Just as the initial geranyl carbocation was able to add across the ring, so this new carbocation can add to the double bond at the opposite side of the p‐­ menthane ring. Addition to one end gives the pinane skeleton, whereas addi­ tion to the other gives the camphane (also known as bornane) skeleton. The cyclisation to give pinane is more favoured in that it gives a tertiary cation in the product. However, it is less favoured because of the formation of a strained four‐membered ring. The enzyme used for the cyclisation will control which way the ring will close. When a carbocation is present at the position shown in the figure, the pinane skeleton can rearrange to the camphane by means of the Wagner–Meerwein rearrangement. This reaction, named after the two chem­ ists who discovered it, is an important feature of terpenoid chemistry. If the camphyl carbocation shown reacts with water, the product is isoborneol, and oxidation of this gives camphor. Some of the more significant acyclic monoterpenoids are shown in Figure 14.19. Geraniol, its (Z)‐isomer (see Chapter 2 for definition) nerol, and citronellol are known as the rose alcohols because of their importance in the essential oil of roses. However, they also occur in a very wide variety of other essential oils, especially flower oils. Linalool takes its name from linaloe oil and also occurs in a very wide variety of essential oils. Its odour character is softer and less characteristically rosy than the rose alcohols, making it a key ingredient of most floral perfumes. OH OH

OH OH

Geraniol

Nerol

O

Citronellol

Linalool

O O

O

O

O

Geranyl acetate

Linalyl acetate

Figure 14.19  Some acyclic monoterpenoids.

Citral

Citronellal

273

274

14  Natural Fragrance Ingredients

Geranyl acetate and linalyl acetate are also widespread in nature and important perfume ingredients. Citral is the substance responsible for the distinctive char­ acter of lemons. Since the lemon odour is associated with cleanliness in consum­ ers’ minds, the lemon note is highly desired in fragrances for cleaning products. However, citral itself cannot be used, because of its poor stability in such prod­ ucts. Citronellal is found in oils such as citronella and Eucalyptus citriodora to which it gives a distinct lemon/citrus character. Figure 14.20 shows some cyclic monoterpenoids. Limonene is a very common component of essential oils. It is best known as the major component in citrus oils, in which it is found as the dextrorotatory isomer, d‐limonene. l‐Menthol is the substance from which the p‐menthane (i.e. 1‐methyl‐4‐isopropylcyclohexane) skeleton takes its name. It occurs in various plants of the mint family. It is so not important as a fragrance ingredient in its own right, but rather as a flavouring and in oral care (e.g. toothpaste) and cosmetic preparations (e.g. shaving cream). In all of these applications, it is the cooling properties of l‐menthol rather than its odour that are required. l‐Menthol acts on the temperature sensing trigeminal nerves and gives the impression that the temperature being experienced is lower than it actually is. l‐Carvone is another mint component and has an odour very charac­ teristic of spearmint. It is often cited as an example of odour differences between enantiomers since its antipode, d‐carvone, has an odour close to those of dill and caraway. α‐Terpineol is present in many essential oils. It has an odour reminiscent of lilac although it is one flower oil in which it does not  occur. The isomeric pinenes are the major components of turpentine, the α‐­isomer being the predom­ inant one in most pine species. Both have a characteristic pine odour. Camphor is the major component of camphorwood oil but is also found in many other oils. For instance, it contributes a fresh top note to such flower oils as lavender and to herb oils such as sage and rosemary. Isobornyl acetate has a fruity and woody smell and is found in various essential oils such as sage. O

OH

OH d-Limonene

l-Menthol

l-Carvone

O α-Pinene

β-Pinene

Figure 14.20  Some cyclic monoterpenoids.

Camphor

α-Terpineol

O Isobornyl acetate

O

­Terpenoid

Just as the monoterpenoids are biosynthesised from geranyl pyrophosphate, so the sesquiterpenoids are synthesised from farnesyl pyrophosphate. Since farnesol has three double bonds, as opposed to geraniol’s two, the scope for intramole­cular reactions (i.e. reactions occurring within a single molecule rather than between two different molecules) is much greater, and so the diversity of sesquiterpenoids is much greater than that of monoterpenoids. Because of their higher molecular weight and hence lower volatility, many sesquiterpenoids are odourless. However, those with odours tend to be very significant components of the essential oils containing them. Some typical odorous sesquiterpenoids are shown in Figure 14.21, in the context of their biosynthetic pathways. As in Figure 14.18, italicised type is used to indicate structural classes, and normal type refers to names of distinct molecular entities. As with the biosynthesis of OH + +

α-Bisabolol Bisabolane

+

O

Campherenane

+

Cedrane α-Atlantone

+ OH

Germacrane

α-Santalane

Cedrol +

Khusane

Gualane

O Guaiol

+

α-Patchoulane

α-Santalol

OH

HO

OH

Zizanal

Patchouli alcohol

Figure 14.21  Formation of some typical sesquiterpenoids from farnesyl pyrophosphate.

275

276

14  Natural Fragrance Ingredients

monoterpenoids, the process begins with the cleavage of the bond to the pyro­ phosphate ester. If the farnesyl carbocation thus produced cyclises to the 6,7‐­double bond, the bisabolane skeleton is produced, and hydrolysis of it gives α‐bisabolol, which occurs in many oils. The anti‐inflammatory properties of chamomile oil are due to the α‐bisabolol it contains. Another odorous bisab­ olane derivative is α‐atlantone, which is an important contributor to the odour of Atlas cedarwood oil. Further intramolecular cyclisations and rearrangements of the bisabolane skeleton can lead to numerous different structural types, two of which are cedranes and campherenanes. Hydration of the cedrane carbo­ cation gives cedrol, a major component of Virginian and Chinese cedarwood oils. Another alternative is for the cedrane skeleton to rearrange further to give the khusane skeleton, which is of significance for the odour of vetiver oil. Vetiver oil is a complex mixture of sesquiterpenoids. The most abundant components contribute very little to the odour, and it is a handful of minor components, such as zizanal, which are responsible for the beautiful, rich aroma of vetiver. Several of the other key odour components of vetiver oil are very closely related to ziza­ nal and have the khusane skeleton. The campherenane skeleton can be rear­ ranged to the α‐­santalane skeleton, and subsequent enzymic oxidation produces α‐­santalol, the key odour component of East Indian sandalwood oil. If the farnesyl carbocation cyclises initially to the 10,11‐double bond, it produces the germacrane skeleton, and one possibility for this outcome is to rearrange to the guaiane skeleton. Hydration of this gives guaiol, the major component of guai­ acwood oil. The guaiane skeleton can rearrange to the α‐patchoulane skeleton, and, after further rearrangement and hydration, this leads to patchouli alcohol, the major component of patchouli oil. HO

P

O

P

OH

Geranyllinalyl pyrophosphate

O OH O O

O

OH P O OH O OH

Geranylgeranyl pyrophosphate

P

O

Phytoene

α-Carotene

Figure 14.22  Tail‐to‐tail coupling and the biosynthesis of carotenoids.

­Degradation Product

Most linkages between isoprene units occur in a head‐to‐tail fashion as described above. However, as mentioned earlier, the steroids (a subset of the triterpenoids) and the carotenoids (a subset of the tetraterpenoids) contain a tail‐to‐tail link. Figure 14.22 shows how geranylgeranyl pyrophosphate and gera­ nyllinalyl pyrophosphate come together via a tail‐to‐tail link to produce phy­ toene, the precursor for the carotenoids. The chain of phytoene can be cyclised to give six‐membered rings at one or both ends. The example shown in the fig­ ure is α‐carotene, one of the pigments responsible for the colour of carrots. α‐ Carotene is used by sighted animals as a storage system for the chemically more delicate vitamin A. Carotene can be stored in the body and then oxidised to give vitamin A when required. Vitamin A is the key part of the pigment used in the eye to detect light. The relevance of the carotenoid pigments to fragrance chem­ istry will be made apparent in the next section of this chapter.

­Degradation Products Once released into the environment, carbon‐based chemicals are subject to vari­ ous degradation mechanisms, principally involving oxidation, either autoxida­ tion or enzymic oxidation. The carotenoids are an example of this degradation. The long polyunsaturated chain between the two rings can be broken by either autoxidation or by enzyme action. The carotenoids are not volatile enough to possess odours, but as the molecule is broken down, some of the smaller frag­ ments produced are odorous. The two major classes of odorous carotenoid frag­ ments are the ionones and damascones. The more significant members of these families are shown in Figure 14.23. Comparison of the structures of the ionones and damascones in Figure 14.23 with that of α‐carotene in Figure 14.22 shows how these degradation products are related to their carotenoid precursors. In the both the ionones and damascones, the prefix α‐ indicates that the double bond in the ring runs away from the side chain, whereas the prefix β‐ indicates a ring

O

α-Ionone

O

β-Ionone

O

α-Damascone

O

β-Damascone

Figure 14.23  Ionones and damascones.

O

γ-Ionone O

Damascenone

277

278

14  Natural Fragrance Ingredients

double bond that joins the carbon carrying the single methyl group to that carry­ ing the side chain. If the double bond is exocyclic (i.e. connects a methylene group to the ring), then the isomer’s name takes the prefix γ‐. The difference between the ionones and damascones lies in the relative positions of the ketone function and double bond on the side chain. In the ionones, the double bond lies closer to the ring, whereas in the damascones the ketone group is next to the ring. The ionones possess woody and violet‐type odours, whereas the damascones are responsible for the fruity top notes of rose. In the ionone and damascone series, the α‐ and γ‐isomers tend to have more floral odours, whereas the fruity aspects are more dominant in the β‐isomers. Perhaps the best known degradation products in perfumery are those com­ prising ambergris. Ambergris is produced in the intestinal tract of the sperm whale (Physeter macrocephalus) in response to irritation of its gastrointestinal tract. One of the major components of ambergris, as produced by the whale, is the triterpene ambreine, the structure of which is shown in Figure 14.24. When ambergris is released into the sea, it is exposed to salt water, oxygen, and sun­ light, which in combination are responsible for a number of degradation reac­ tions that break the ambreine molecule into smaller fragments. All of these changes have been replicated in the laboratory. Three of the degradation prod­ ucts that contribute to the odour of ambergris are also shown in Figure 14.24. Dihydro‐γ‐ionone might look like a carotenoid degradation product, which it could be, but when found in ambergris, it is derived from the monocyclic end of the ambreine molecule. It contributes tobacco‐like notes to the overall odour profile of ambergris. Ambrinol is derived from dihydro‐γ‐ionone by cyclisation and is responsible for the earthy notes in ambergris. The most important of all the ambreine degradation products is the naphthofuran, which is known under various trade names such as Ambrofix, Ambrox, and Ambroxan. This material is largely responsible for the characteristic odour of ambergris.

OH H Ambreine

O

H

O

OH dihydro-γ-ionone

Ambrinol

Figure 14.24  Ambreine and some of its degradation products.

H Amberfix

­Malodour

HO

OH O Iripallidal O

OH

α-Irone

Figure 14.25  Formation of α‐irone.

Another group of compounds that, at first sight, might look like carotenoid degradation products is the irones. These are responsible for the characteris­ tic buttery, woody odour of iris rhizomes and hence orris, which is extracted from them. In this case the precursor is the triterpenoid iripallidal. The irones carry one methyl group more than do the corresponding ionones. The nomen­ clature system for ionones and irones is the same, and so the irone illustrated in Figure 14.25 is known as α‐irone. The double bonds in the chains of fatty acids serve as centres for oxidation and possible subsequent oxidative cleavage by both autoxidation and enzymic mech­ anisms. Cleavage of the chain produces molecules small enough to have odours, and the introduction of oxygen functions can introduce aldehyde, ketone, and hydroxyl groups. The hydroxyl groups can, of course, react with the acid func­ tion that is already present and therefore produce lactones. These reactions can produce fragrant components of essential oils, but they can also produce malo­ dours, as we will see in the next section.

­Malodours Some plants and animals produce foul odours as defence mechanisms. Anyone who has encountered an angry skunk will be very aware of this phenomenon. Some plants produce odours reminiscent of decaying flesh in order to lure flies into a trap. However, the malodours of most interest to the fragrance industry are those that are found in everyday human life and that we wish to eliminate or cover up. The main categories of malodours we come across in the industry are body odours (especially underarm malodour), kitchen malodour, bathroom malodour (faecal and urine), smoke malodour, pet malodour, mildew, and malodorous components of consumer goods. In some cases, low levels of malo­ dours contribute to a fragrance, for example, indole in jasmine. However, mostly they are considered undesirable, and one of the roles of perfume is to hide malodour. Consumer goods can have intrinsic malodours as a result of the manufactur­ ing process. The oldest example is soap, which has an unpleasant fatty odour unless rigorously purified. Two of the most difficult products to perfume in this respect are hair waving lotion and depilatory cream. An active ingredient in each

279

280

14  Natural Fragrance Ingredients

is thioglycolic acid (see Chapter 11 for details) and some degradation products of this acid possess intense malodours. In order to develop fragrances to cover the odour of the product base, it is always helpful and sometimes necessary to determine the nature of the substances responsible for the malodour. Obviously, this will vary from product to product, so each new consumer product presents a new challenge for the fragrance chemist. Tobacco smoke malodour is a very complex mixture of products of combus­ tion of the tobacco. Some of the more organoleptically significant compo­ nents include small molecules such as lower amines, acrolein (propenal), and acrylate esters (propenoates) and a range of heterocyclic materials including pyridines, pyrazines, and piperidines. The odour of mildew is characterised by geosmin, a fungal metabolite. The substances responsible for the other malo­ dour categories are mostly degradation products of lipids, proteins, terpe­ noids, and so on. They fall into four main categories, fatty acids, thiols, amines, and steroids. Lipid degradation can result in the formation of short chain fatty acids. Some of these are also produced by loss of the amino group from an amino acid. Butyric acid is produced in a number of ways, including from the spoiling of milk and milk products. It has a typical sour milk odour and is part of many kitchen malodours. Fatty acids containing between 4 and 10 carbon atoms are major components of sweaty malodours. One of the most important of these is 3‐methyl‐2‐hexenoic acid. This acid was first isolated from the sweat of schizo­ phrenics and so is sometimes referred to as schizophrenic acid. It was once thought to have an odour characteristic of sufferers of this condition, but it is now known that it is equally present in all human perspiration. It is not present in the free state in fresh sweat but rather is esterified to a pro­ tein. Fresh sweat is odourless. The odour develops as bacteria on the skin begin to digest the sweat components. As they do so, they hydrolyse the bond between the protein and the 3‐methyl‐2‐hexenoic acid, thus releasing this intensely malo­ dorous acid. Degradation of nitrogen‐ and sulfur‐containing protein fragments produces thiols and amines, and some of these have very intense and unpleasant malo­ dours. Hydrogen sulfide is also known as bad egg gas since it is responsible for the strong smell of rotten eggs. It is also present in various other kitchen malo­ dours. Trimethylamine is responsible for the characteristic malodour of stale fish. Methanethiol is a significant component of bad breath, and 3‐­mercapto‐3‐ methyl‐1‐butanol is responsible for the distinctive note of cat urine. As with 3‐ methyl‐2‐hexenoic acid, this material is present as a conjugate with a protein – hence the difficulty in eliminating the odour as the protein serves as a reservoir from which this extremely intense malodorant is slowly released into the atmosphere. Ammonia is present in urine, the body’s method of removing this toxic product of protein degradation. Indole and skatole are degradation products of the amino acid tryptophan and are major contributors to the odour of faeces. There are very few odorous steroids since the high molecular weight of this group of materials makes them relatively non‐volatile. However, one odorous

­Malodour SH

O

O OH OH

O Methyl acrylate smoke

3-Methyl-2-hexenoic acid stale human sweat

3-Mercapto-3-methyl-l-butanol cat urine

OH

O Androstenone male human sweat

Geosmin mildew

Figure 14.26  Some common malodour components and their sources.

steroid, which does present a malodour problem, is androstenone. This is a deg­ radation product of male sex hormones and so represents a part of the male axillary (underarm) malodour. Androstenone and some other representative malodour components are shown in Figure 14.26.

Review Questions 1 To which families of natural products do the following substances belong? O O

O OH

O

O

Ethyl everninate

O O

Deodarone

Dehydrocostus lactone

O O O

O O O Acorone

β-Asarone

Theaspirane

281

282

14  Natural Fragrance Ingredients

2 Are the following substances likely to be plant products? O O O

O Inonyl acetate

Verdyl acetate O

O Cervolide

O

O LiIial

283

15 Synthetic Fragrance Ingredients ­ hy the Industry Uses Synthetic Fragrance W Ingredients? With the large number of essential oils and other plant extracts that can be used in perfumery, one might ask, ‘Why make synthetic fragrance ingredients?’ However, synthetic fragrance ingredients have become the mainstay of the modern fragrance industry. The four main factors that have brought this trend about are shown in Figure 15.1. Essential oils are fine in relatively innocuous bases such as fine fragrance, but, as we saw in Chapter 11, modern perfumes are used in a wide range of consumer goods, and many of these products represent stability issues for fragrance ingredients. Many of the components of essential oils are delicate molecules and do not survive in consumer goods. For example, soaps and laundry detergents are alkaline, so acetate esters are prone to hydrolysis when incorporated in them. Replacement of the acetate function by a methyl ketone often gives a product with a similar odour to the ester but better stability at high pH. Similarly, hydrogenation of allylic alcohols, such as geraniol and linalool, gives saturated alcohols, which are more stable to oxidants such as bleach. Stability is not the only way in which performance of an ingredient can be improved. Others include compatibility with the product or its container and delivery from the product to the headspace (the air around the product). For a variety of reasons, essential oils are expensive to produce. The yield of oil in dried plant material is usually only 1% or 2% by weight, so a large volume must be harvested in order to produce a small amount of oil. Some plants can be harvested mechanically, but others, such as jasmine, must be hand‐picked. It takes about 7 000 000 jasmine flowers to produce a single kilogram of absolute, and so the labour cost is high, even in countries with cheap labour. If molecules identical Superior performance in use Cost Security of supply Safety/regulatory pressure Figure 15.1  Driving forces for the use of synthetic fragrance ingredients. Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

284

15  Synthetic Fragrance Ingredients

to those in the oil can be produced for a few dollars per kilogram, then the price competition between jasmine plant and chemical plant is very uneven. Plants such as patchouli and vetiver only grow in tropical climates and often only in limited geographical regions. Patchouli, for instance, is only produced in certain parts of Indonesia. This limited availability means that the harvest can be devastated by bad weather or natural disasters such as tsunamis, earthquakes, or volcanoes. Frankincense and myrrh are produced only in the Horn of Africa, a region that has been subject to wars and droughts throughout history. All of these factors make essential oil production a somewhat insecure business. By using chemical processes situated in several locations, the supply of ingredients is made much more secure. The fourth reason for the growth of synthetic ingredients, namely, safety, may come as a surprise to some. Contrary to public perception, the major issues with safety relate to essential oils and their components rather than synthetic ingredients. For example, of the 26 suspected ‘allergens’ listed in the Seventh Amendment to the European Cosmetics Directive, 18 are either plant extracts or common essential oil components. As mentioned already in Chapter  12, many oils (for example, fig leaf and sassafras) have been banned on safety grounds by the industry, and others are restricted or require detoxification before use (for example, bergamot). In most countries, the government has introduced legislation concerning safe use of chemicals (of both plant and man‐made origin, since there is no distinction between the two in toxicological terms). Clearly any responsible fragrance company will ensure that it operates within the law in each country where its products are sold. Few legally prohibited ingredients exist (the industry tends to act ahead of any such legislation and remove products before governments become involved). Much more common is the requirement for labelling of ingredients on products containing them. Such labelling requirements vary from country to country and from product to product. For instance, the approved level of an ingredient might be different for a ‘rinse‐off’ product such as shower gel and a ‘leave‐on’ product such as an underarm deodorant. Although the need to label does not prevent use of an ingredient, many consumer goods companies prefer not to have, in their products, ingredients that require hazard labels. An example of legal labelling requirements is shown in Figure 15.2. These labels are associated with citrus oils. Which labels are required will depend on the product and the country. For example, in Europe, the environmental hazard warning of a dead fish and dead tree would be required on a drum of orange oil but not on a bottle of fine fragrance that contained orange oil as one ingredient. However, the fragrance would have to carry the label ‘limonene’ as that is one of the 26 alleged ‘allergens’ covered by the Seventh Amendment. Over recent years, the growth in regulations has created an increasing need for safer synthetics to replace ‘natural’ ingredients.

­The Economics of Fragrance Ingredient Manufacture Fragrance ingredients are used by fragrance houses and also by those consumer goods companies (such as Procter & Gamble and Colgate‐Palmolive) that have internal perfumery departments. The large perfumery companies (such as

  ­The Economics of Fragrance Ingredient Manufactur

Flammable Harmful: may cause lung damage if swallowed Sensitiser

Limonene

Irritating to skin

Citral

Figure 15.2  Hazard labels associated with citrus oils.

Givaudan, Firmenich, and International Flavors and Fragrances [IFF]), some of the smaller ones, and a very few of the larger consumer goods companies (e.g. Kao) produce fragrance ingredients. These companies clearly produce ingredients in order to support their creative perfumery activities, and they also sell ingredients to each other. Some companies specialise in essential oil production and sell oils to fragrance companies but do not blend their own fragrances. Large chemical companies do produce fragrance ingredients, but this practice is usually as a consequence of their interests in other areas since the volumes used by the fragrance industry are tiny compared with those of the bulk chemical industry. For example, Badische Anilin und Soda Fabrik (BASF) and Dutch State Mines (DSM) both produce terpenoid vitamins and therefore use as intermediates such compounds as linalool and citral, which they also sell to the fragrance industry and use as intermediates for production of other higher volume fragrance ingredients. Similarly, Kuraray is primarily a synthetic rubber manufacturer, so they produce butadiene and isoprene as monomers and can use these to gain entry to the terpenoid market. It is only the highest volume fragrance ingredients, such as linalool or phenylethanol, which are of interest to these large chemical companies since their production plant tends to have large capacity and to be dedicated to individual processes. Smaller, fine chemical companies with versatile, smaller‐scale production equipment will be more interested in what, to the fragrance industry, represent the medium or lower volume ingredients. These fine chemical companies often earn the bulk of their income from the pharmaceutical industry, and so their cost base tends to put them at the top end of the price range of fragrance ingredients. A typical fragrance ingredient life cycle is shown in Figure 15.3 in the form of a graph of profitability over time. Initially, investment is necessary in the discovery of the new molecule, and the level of the investment increases as the material is evaluated and developed. Only a small percentage of the candidate novel

285

286

15  Synthetic Fragrance Ingredients

Profit/kg Captive use

Patent expiry

Speciality

Discovery

Commodity

Launch Time

Development

Figure 15.3  Typical ingredient life cycle.

­ olecules will be taken right through the development process, for reasons that m will be explained later. Successful candidates will be patented and launched, almost invariably as a captive ingredient, that is, one that is available only to the perfumers of the company that developed it. Development costs are high, especially the costs of convincing regulators that the compound is safe for use. These costs are reflected in a relatively high profit margin, but, in reality, the costs can only be recovered through the contribution of the new ingredient to the success of fragrances containing it. For this reason, it is very unlikely that a novel fragrance ingredient could be developed profitably by anyone other than a major fragrance producer. The volume that will be required is very much unknown at this point. Some ingredients remain as low tonnage products for their entire life, while others grow relatively rapidly to high volume ingredients. This uncertainty adds to the commercial risk that is run by the fragrance company. Patents are valid only for a period of 20 years, and so the patent on a new material will eventually expire. At this point, other fragrance companies or fine chemical companies may decide to produce the material (assuming that it has grown successfully) in competition with the company that developed it. This eventuality, of course, leads to a reduction in the profitability since price reductions are an inevitable consequence of such competition. In order to dissuade competitors or other companies from manufacturing the ingredient, a fragrance company will usually decide to release the captive ingredient for third‐party sales at some time during the life of the patent. What was a notional internal profit then becomes a real profit. However, the margin will already be falling because of the threat of imminent competition. After release for third‐party sales and while only a small number of producers remain, the product is usually referred to as a specialty. If the product is successful enough that it grows to a high volume (by fragrance ­industry

  ­The Economics of Fragrance Ingredient Manufactur

standards), it may become large enough to attract attention of larger chemical companies. Usually by this stage the volume consumed is relatively stable, and so such companies are able to make accurate predictions of return on investment. If so, they will be able to develop more cost‐efficient production methods based on their technological strength (or lower operating costs in the case of producers in developing countries) and advantages of scale. In these cases, the margin falls to a relatively low level, and only the most efficient producers will be able to remain in the market. At this point the ingredient has become a commodity. These phases of development, and hence product categories, are also illustrated in Figure 15.4. Commercially, the fragrance ingredient market is difficult because the volumes are in a similar order to those of pharmaceuticals but the prices are closer to those of the bulk chemical market. Consequently, fragrance ingredients often tend to ‘piggy back’ on other products such as intermediates or by‐ products from other industries. Apart from price, availability and security of supply are important factors in selecting feedstocks. Both plant and petrochemically derived raw materials are used. In many cases, the advantage of plant‐derived feedstocks is that their chemical structures are similar to those of fragrance ingredients. This distinction is important since fragrance ingredients often require a degree of structural complexity that is expensive to introduce from petrochemical feedstocks. All of these economic factors are in continual interplay, and the balance often moves from one point to another as outside circumstances change. Upon looking at some of the major production routes in the next part of this chapter and in Chapter 17, we will see examples of the various driving forces in action and the shifting of the economic balance over time. Captive ingredients

Specialities

Commodities

Unknown volume

Known, moderate volume

Known volume Low value

High price Medium price

High volume

Margin steady Margin falling

Margin steady

Investment justified by compound fragrance sales

Investment justified by calculated return

Produced by fragrance houses

Produced by fragrance houses and/or fine chemical companies

Figure 15.4  Categories of fragrance ingredients.

Investment justified by technological capability

Produced by fragrance houses and/or bulk chemical companies

287

288

15  Synthetic Fragrance Ingredients

­ roduction of Fragrance Ingredients from Polyketides P and Shikimates Vegetable oils are produced for many purposes, mostly as edible oils. This process makes them available as a ready source of lipid‐derived starting materials for the synthesis of fragrance ingredients. Hydrolysis of the vegetable oil releases the glycerol and fatty acids of which they are composed. Various fatty acids serve as useful raw materials for fragrance ingredients, the most important of which are castor, coconut, and rapeseed oils. Castor oil is rich in ricinoleic acid, which cleaves on pyrolysis to give heptanal and undecylenic acid. On base‐catalysed aldol condensation with benzaldehyde, heptanal gives amylcinnamic aldehyde, a floral jasmine ingredient. Similar condensation with cyclopentanone, followed by hydrogenation, gives heptylcyclopentanone, a floral, peachy material. Undecylenic acid is the starting material for a range of materials containing an 11‐carbon chain. For example, reduction to the aldehyde (which actually involves reduction to the alcohol and reoxidation to the aldehyde) gives undecylenic aldehyde. The corresponding saturated aldehyde is obtained by hydrogenation of undecylenic aldehyde, and trade names have to make clear which of the two any given products is. Treatment of undecylenic acid with a mineral acid such as sulfuric acid causes the double bond to isomerise and move along the chain. When it reaches the fourth carbon atom from the far end, it is trapped by the carboxylic acid function and forms the lactone, undecalactone. This material has a powerful and long‐lived coconut like odour. These examples of castor oil‐derived materials are shown in Figure 15.5. Coconut oil contains octanoic acid from which octanol and octanal can be obtained. One of the products produced from these is hexyl cinnamic aldehyde. This process is exactly analogous to production of amylcinnamic aldehyde from heptanal as described above, the difference in the two products simply being that hexyl cinnamic aldehyde has one more carbon in its side chain. Rapeseed oil is rich in erucic acid. Cleavage of the double bond in erucic acid by ozonolysis gives pelargonic acid and brassylic acid, a dicarboxylic acid, that is, one with an acid function at each end of the chain. Formation of the cyclic dilactone between brassylic acid and ethylene glycol gives ethylene brassylate, one of the highest volume macrocyclic musks. These products are shown in Figure 15.6. The major shikimate raw materials for fragrance ingredient production are clove and sassafras oils and turpentine and lignin (from wood). Clove oil is rich in eugenol, which is extracted from it for use per se as a fragrance ingredient and also as a starting material for methyl eugenol, isoeugenol, methylisoeugenol, and vanillin. Conversion of eugenol to vanillin involves ozonolysis of the double bond, and the same reaction, when applied to isosafrole, (prepared from safrole, the major component of sassafras oil) gives heliotropin, a powerful, sweet floral ingredient with an odour characteristic of the flower from which it takes its name. Turpentine is a major source of terpenoid materials, but it also is a source of anethole. Degradation of lignin, a component of the fibrous material of wood, produces vanillin, although this process is not a major route to the latter. These materials are shown in Figure 15.7.

  ­Production of Fragrance Ingredients from Polyketides and Shikimate HO

O

Ricinoleic acid

OH Pyrolysis Undecylenic acid

O

Heptanal

OH

+ O

O

O

Amylcinnamic aldehyde O

Undecylenic aldehyde

O Heptylcyclopentanone

O

Undecalactone

Figure 15.5  Fragrance ingredients from castor oil.

Erucic acid O Ozonolysis

OH

O OH Pelargonic acid

+

OH

HO O

O Brassylic acid

Ethylene brassylate

O O

Figure 15.6  Ethylene brassylate from rapeseed oil.

O O

289

290

15  Synthetic Fragrance Ingredients O

O

O

Eugenol

HO

Methyleugenol

O

Oxidation

Isoeugenol HO

O

O O

HO

Methylisoeugenol Vanillin

O

O

O

O

O

O Safrole

O Isosafrole

O

O

Heliotropin

Anethole

Figure 15.7  Some shikimate‐derived fragrance ingredients.

­Terpenoid Production Just as the terpenoids represent the largest group of natural fragrance ingredients, they are also the major group of synthetic fragrance ingredients. However, terpenoids are used not only as fragrance and flavour ingredients but also as food colours, vitamins, pharmaceuticals, active ingredients in cosmetics, resins, and so on. The major producers therefore tend to have interests other than fragrance. Some have product portfolios based on sulfate turpentine as a by‐product of the Kraft paper process. Others, such as BASF and DSM, are chemical companies engaged in vitamin production, and another, Kuraray, is a synthetic rubber company that, because of the monomers used for rubber manufacture, has access to the feedstocks and technology required for manufacture of fragrant terpenoids. The largest volume fragrant terpenoid produced is α‐terpineol of which about 30 000 tonnes are made per year, mostly for use as a disinfectant. About 20 000 tonnes/annum of l‐menthol are produced, and production of ingredients such as linalool, citral, and citronellol each runs at about 5000–10 000 tonnes/annum, which is similar to the total for the ionone family of ingredients. Borneol and isoborneol, carvone, and the synthetic sandalwood ingredient family all are in the 1000–2000 tonnes/ annum category. To put this into context with the bulk chemical industry,

  ­Terpenoid Productio

150 000 000 t of ethylene are produced annually, and production of ethylbenzene, phenol, and adipic acid are also all in the millions of tonnes per annum. For the fragrance industry, the most important groups of terpenoids are the rose alcohols and citral. Citral is not important as an ingredient in its own right but rather as an intermediate for the ionone family of ingredients. All five of these key materials are available from essential oils. Geraniol is obtained from palmarosa (and derived species such as dhanrosa and jamrosa) and geranium. Geranium is also a source of citronellol. Citronella provides geraniol, citronellol, and citronellal. Linalool can be obtained from ho leaf, rosewood, and linaloe oils. Litsea cubeba serves as a source of citral. However, some of these oils (e.g. rosewood and L. cubeba) are becoming rare because of overharvesting and so, in the interests of conservation, are no longer used commercially. The chemicals obtained from the essential oils are much more expensive than their synthetic counterparts, so the synthetics are the dominant materials used. The price and availability differences are greater for linalool than the other rose alcohols, and so, while maybe up to 10% of geraniol is essential oil derived, for linalool, the figure is about 1%. The principal synthetic routes to these materials and the most important conversions between them are outlined in Figure  15.8. The basic feedstocks are either turpentine or petrochemicals such as butadiene, acetone, and acetylene. Those companies that produce terpenoids from turpentine tend to enter the group of materials via geraniol, while the petrochemical routes mostly produce linalool initially. Citral is accessible from either of these by oxidation, the

Isobutylene acetone/acetylene

Pinenes

OH O

OH

Linalool Geraniol

Citral

OH Citronellol

O Citronellal

Figure 15.8  Key routes to citral and the rose alcohols.

Ionones Vitamins

291

292

15  Synthetic Fragrance Ingredients

­ xidation of linalool obviously requiring a concomitant allylic rearrangement o (i.e. with a shift of the oxygen from C3 to C1 and a movement of the double bond in the opposite direction). Aldol condensation of citral and subsequent cyclisation produces the ionone family of ingredients. Partial hydrogenation of geraniol gives citronellol, which can be oxidised to citronellal. Geraniol, linalool, and citronellol are used as ingredients in their own right and also as starting materials for esters, acetals, and other related ingredients. Citronellal is also important as a feedstock for hydroxycitronellal and l‐menthol. Some of these processes will be described in a little more detail in Chapter 17. Since the birth of the synthetic fragrance ingredients industry, process development chemists have striven to improve the efficiency of the processes used. Clearly, it is best to use the shortest route possible, the route with the lowest energy requirement, and the route with the smallest volume of by‐products or other waste material. All of these make economic sense, and the second two also lead to the most acceptable processes in terms of environmental impact. It is also desirable to use starting materials that are inexpensive, readily available from a secure supply and, preferably, from a renewable source. It is not always possible to achieve all of these goals simultaneously, and in such cases, the optimum compromise is sought. The history of citral production from petrochemical feedstocks will serve as an example of how development chemists strive for continual improvement. One useful tool for the process development chemist is the concept of atom efficiency. This parameter can be calculated for any proposed chemical route without even doing any laboratory work. The atom efficiency of a process is defined as the molecular weight of the desired product, expressed as a percentage of the sum of the atomic weights of everything that goes into the process. This assumes 100% conversion of starting materials to desired intermediate product at every stage of the process and so represents the best possible outcome for the route. In practice, this is never achieved. However, by calculating the atom efficiency of different proposed synthetic routes, the development chemist can easily see which one potentially offers the most attractive possibility. Citral was first synthesised by Barbier, Bouveault, and Tiemann in 1898. The route they used is shown in Figure 15.9. Their objective was to confirm the molecO

O Br

Br Br

O

Base

NaOH O

O

O

I

Zn

O O

O O

Ca(OCHO)2 D

O )2 2+

Ca

Figure 15.9  The Barbier–Bouveault–Tiemann route to citral.

Ca(OH)2

O

  ­Terpenoid Productio O O

O

Figure 15.10  The Arens–van Dorp concept for synthesis of citral.

ular structure proposed for citral, and this route served that purpose. However, it would not be an attractive manufacturing route. Even cursory inspection reveals that, per molecule of citral produced, two bromine atoms, one iodine atom, one zinc atom, one calcium atom, and several carbon and oxygen atoms are thrown away as waste in the process. A little arithmetic shows an atom efficiency of less than 20%. In other words, even if everything proceeds as well as theoretically possible, for every kilogram of citral produced, there will be more than 4 kg of waste. Clearly this is unacceptable in both economic and environmental grounds. In the 1940s, two Dutch chemists, Arens and van Dorp, proposed a synthesis of citral starting from two molecules of acetylene and two of acetone. The basic concept is shown in Figure 15.10. A molecule of citral is shown on the right of the figure, and, on the left, we see two molecules of acetylene and two of acetone orientated as they will be when incorporated into the citral molecule. The dotted lines show the new bonds that must be formed between these four building blocks. Arens and van Dorp realised the concept in 1948, and the exact route they used is shown in Figure  15.11. Addition of acetylene to acetone was effected using sodium amide as a base. Partial reduction of the acetylenic bond of the product, followed by treatment with phosphorus tribromide, gave prenyl bromide. The second acetone equivalent was actually used in the form of ethyl acetoacetate. This molecule can be considered an acetone molecule to which an ethoxycarbonyl function has been added. This second carbonyl function further activates the hydrogen atoms between itself and the carbonyl group of acetone and facilitated the base‐catalysed alkylation with prenyl bromide. Hydrolysis of the resultant keto ester and subsequent decarboxylation of the acid function gives methylheptenone. Instead of simply adding another molecule of acetylene, Arens and van Dorp used an acetylenic ether. A Grignard reagent was used to deprotonate this acetylenic molecule, and the resulting acetylenic Grignard reagent was added to methylheptenone. Hydrogenation under Lindlar conditions gave the corresponding olefinic material, which was in fact an enol ether. Hydrolysis of this and concomitant dehydration led to citral. The use of stoichiometric (i.e. one molecular equivalent of reagent per molecule of starting material) quantities of sodium amide, zinc/copper couple, and phosphorus tribromide, and the loss of several carbon and oxygen atoms, meant that the atom efficiency was still only about 20%. However, the basic principle was established, and this route is capable of development into an efficient manufacturing process. The discovery of the Carroll reaction enabled a major advance in the Arens–van Dorp approach. In the Carroll reaction, an allylic alcohol is heated with a β‐keto

293

294

15  Synthetic Fragrance Ingredients OH Na/NH3

O

OH

Zn/Cu H2O

PBr3 Br O

O

(1)

/Base

O (2) H3O+ (3) –CO2 O

O

O H2O

OH

HCI

H2

O

OH

Pd/BaSO4 O

Mg Br

Figure 15.11  The Arens–van Dorp synthesis of citral.

ester. The allylic alcohol displaces the original alcohol of the ester, and then a pericyclic reaction (see Chapter 7) occurs with loss of carbon dioxide to give a ketone as product, in this case, methylheptenone. Addition of acetylene gives linalool, which can be rearranged to citral. This synthesis is shown in Figure 15.12. The atom efficiency is now up to almost 50%, a huge improvement. OH O

OH

H2/cat.

Base

O

O O

O

H+

O OH KOH

Figure 15.12  The use of the Carroll reaction in the synthesis of citral.

  ­Terpenoid Productio

The major loss of atom efficiency in the above route stems from the loss of ethanol and carbon dioxide from the ethyl acetoacetate starting material. This issue was addressed with the discovery of the Claisen rearrangement. This rearrangement is another pericyclic reaction, and, in this case, an enol allyl ether rearranges to give a γ,δ‐unsaturated ketone. Heating methylbutenol with the methyl enol ether of acetone (methoxypropene) leads to exchange of the former with the methanol of the latter, and the resultant enol allyl ether rearranges to give methylheptenone. The remainder of the synthesis then proceeds as before. Using only a catalytic amount of base for the two acetylene additions brings the atom efficiency of this route up to nearly 55%. Successful completion of this route was reported in 1957. It became the first large‐scale commercial process for production of citral and is still in use today. Figure  15.13 shows the route using the Claisen rearrangement. Another of the prime routes to citral currently in use is shown in Figure 15.14. This route was developed by BASF chemists in the late 1970s and is a good example of the elegance that can be achieved through process development. Ene reaction of isobutene with formaldehyde (both inexpensive, readily available materials) gives isoprenol. Isomerisation of this over a palladium catalyst gives prenol. Alternatively, air oxidation using a silver catalyst, followed by isomerisation over palladium, gives prenal. Heating prenol and prenal together leads to formation of the allyl vinyl ether. This material undergoes a Claisen rearrangement, and then by rotating the product around the central carbon–carbon bond, it is set up to undergo a Cope rearrangement, another pericyclic reaction. The product of this reaction is citral. The last three reactions proceed sequentially in the same reactor, and so citral is reached in four steps (two of which run in parallel) using only heat and catalysts. The atom efficiency is about 80%, and the sole by‐product is water. Two molecules of water are produced per molecule of citral,

OH O

OH

H2/cat.

Base

O

O

H+

O OH Base

Figure 15.13  The use of the Claisen rearrangement in the synthesis of citral.

295

296

15  Synthetic Fragrance Ingredients

H

OH

Pd

O

∆ OH O

(1) Ag:SiO2/O2 (2) Pd

O

O

∆

O

O

Figure 15.14  Citral from isobutylene.

one from the oxidation reaction and one from the ether formation. If by‐products must be formed, then water is just about the best we could have. The other major source of raw materials for terpenoid synthesis is turpentine. Turpentine can be obtained by tapping pine, spruce, and fir trees, and this material is called gum turpentine. However, a much more important source is the cheaper, crude sulfate turpentine or CST, which is produced when timber is pulped to make paper in the Kraft process. The major components of CST are α‐pinene and β‐pinene, which can be separated from the rest of the CST by distillation. Both pinenes are used as raw materials for production of a wide range of perfume ingredients including the rose alcohols and hence citral, ionones, and vitamins. α‐Pinene is the major component of turpentine. It can be hydrogenated to give pinane, which serves as an intermediate to the rose alcohols and some other fragrance materials, as shown in Figure 15.15. Autoxidation of pinane produces a hydroperoxide, which can be reduced to give pinanol. On pyrolysis, the four‐ membered ring in pinanol breaks open by a pericyclic reaction to give linalool. Under the pyrolysis conditions, linalool is unstable and undergoes an ene reaction to produce materials called plinols. The pinanol pyrolysis is therefore run under conditions that minimise the formation of plinols. However, the small amounts of plinols formed are still sufficient to make the linalool unusable as such. The plinols are very difficult to separate from linalool, and so the linalool is isomerised to geraniol, and then the plinols can be removed. From the geraniol, it is then possible to prepare perfumery quality linalool and citral as an intermediate for the ionones and vitamins. Pyrolysis of pinane gives citronellene (also known as dihydromyrcene). Hydration of this material gives dihydromyrcenol, which has become a major ingredient in the relatively short time since its introduction to the palette. It is the ingredient responsible for the new masculine freshness made popular by fragrances such as Drakkar Noir and Cool Water. Under different conditions, citronellene undergoes acid‐catalysed cyclisation, rearrangement, and hydration to give cyclodemol, and this alcohol is used as a

  ­Terpenoid Productio

Pinanol

OH

OH OH

Linalool

Geraniol

Pinane

α-pinene

R″′ R″″ O OH

O

R

R′ R″

OH Dihydromyrcenol

O

Citronellene

Cyclodemol

Alicyclic musks

Figure 15.15  Fragrance ingredients from α‐pinene via pinane.

starting material for a recently discovered family of musk ingredients known as the alicyclic musks. These will be discussed in more detail below. Some of the other fragrance ingredients produced from α‐pinene are shown in Figure  15.16. Treatment of α‐pinene with dilute aqueous acid gives terpineol. Epoxidation of α‐pinene followed by Lewis acid‐catalysed rearrangement of the epoxide gives campholenic aldehyde. This aldehyde is the precursor for a family of ingredients with sandalwood odours. They are prepared by aldol condensation of campholenic aldehyde with another aldehyde or ketone and then reduction of the carbonyl function of the product to an alcohol. Acid‐catalysed rearrangement of α‐pinene gives camphene. Acid‐catalysed hydration of the latter gives isoborneol, which can be oxidised to camphor. When a mixture of camphene and guaiacol (2‐methoxyphenol) is treated with strong acid, the camphene rearranges to a mixture of isomeric monoterpenoid carbocations, and these products add to the phenol in a Friedel–Crafts‐type reaction. Hydrogenation of the resultant adducts leads to a complex mixture of cyclohexanols. In fact, 128 different isomers are found in a typical product mixture. Each contains a cyclohexane ring with one hydroxyl group (the other oxygen function is lost in the hydrogenation) and a monoterpenoid ring system attached at one of the other positions in the cyclohexane ring. Three of these possible structures elicit strong sandalwood odours. Separation would be impracticable commercially, so the mixture is used as such. Each producer has one or more grades of material available, and they are sold under many different trade names. Generically they are usually referred to as isobornyl cyclohexanols. Perhaps the simplest route to a fragrance ingredient from β‐pinene is the addition of formaldehyde in the presence of acetic acid to give the material known as nopyl acetate. When β‐pinene is pyrolysed, the four‐membered ring breaks open

297

298

15  Synthetic Fragrance Ingredients

R O

OH α-terpineol

Campholenic aldehyde

OH R′ Campholenic aldehyde family of sandalwood ingredients

OH Camphene α-pinene

O

Camphor

OH

Isobornylcyclohexanol family of sandalwood ingredients

Isoborneol

Figure 15.16  Some other products from α‐pinene.

exactly as we saw above with pinane and pinanol, but in this case, the product is myrcene, which is a particularly useful intermediate for fragrance ingredients production. Myrcene has a diene function, and it is therefore capable of undergoing the Diels–Alder reaction with a suitable dienophile. For example, with acrolein, the product is Myrac Aldehyde. Hydration of the double bond in the side chain of Myrac Aldehyde gives Lyral, a major muguet ingredient. However, a more efficient route to Lyral is to protect the diene function of myrcene by addition of sulfur dioxide, hydrate the isolated double bond, release the diene function again, and then carry out the Diels–Alder reaction. The woody ingredient Iso E Super is also prepared from myrcene by a Diels–Alder reaction though in this case, a second reaction, an acid‐catalysed cyclisation, also occurs. The diene function of myrcene cannot be hydrated directly, but indirect methods, for example, addition of hydrogen chloride and hydrolysis of the resultant hydrochloride, can be used to produce geraniol from myrcene. Thus β‐pinene can also be used as an entry to the rose alcohols, citral, ionones, and vitamins. Myrcene also serves as a starting material for l‐menthol. This route uses d‐citronellol as an intermediate and will be discussed in more detail later. All of these routes from β‐pinene are shown in Figure 15.17. Citrus oils are another source of raw materials for terpenoid synthesis. These oils are produced mainly as by‐products of the fruit juice industry. Whole fruits are pressed to obtain the juice, which also squeezes the oil of the glands in the peel. This oil is separated off, and since it is not required for the juice business, it

  ­Terpenoid Productio OH O

O

O

Myrac Aldehyde®

Iso E Super®

Lyral®

β-Pinene

OH Myrcene

O

Geraniol

O

O

OH

d-Citronellal

l-Menthol

Nopyl acetate

Figure 15.17  Some fragrance ingredients from β‐pinene.

is available as a by‐product. The major component of the citrus oils is d‐limonene. The chirality of this starting material is used to good effect in the synthesis of carvone. Addition of nitrosyl chloride (NOCl) to limonene occurs preferentially at the double bond in the ring to give the adduct shown in Figure 15.18. Nitrosyl chloride is not used per se, but the combination of a nitrite ester and hydrochloric acid behaves in the same way as nitrosyl chloride and is much easier to use. The adduct is a blue oil and exists in equilibrium with a dimeric form that is a colourless solid. Treatment of the adduct with a weak base results in dehydrochlorination and rearrangement to give the oxime of carvone (known as c­ arvoxime). CI

‘NOCI’

N

N

O

O

OH H3O+

–HCI

l-Carvone

d-Limonene

[O]

Citrus oils Valencene

Figure 15.18  Fragrance ingredients from citrus oil components.

Nootkatone

O

299

300

15  Synthetic Fragrance Ingredients

Carvone can then be produced by hydrolysis of the oxime. Readers will note from Figure  15.18 that the chiral centre is not involved in this reaction sequence. However, because of the selectivity of the regiochemistry of the reaction sequence, the chirality is maintained. To put this in another way, in the sequence of reactions that convert limonene oxide to carvoxime, the carbon atoms at that side of the ring are fixed in their positions relative to each other throughout the reaction sequence. This means that the absolute stereochemistry at the opposite end is maintained, and, since it is independent of the reactions and unaffected by them, the original (R)‐configuration of the chiral centre of the starting d‐limonene is conserved, giving l‐carvone as the product. Another product available from orange oil is the sesquiterpenoid hydrocarbon valencene. Oxidation of valencene produces nootkatone, and, as with limonene, the stereochemistry of the starting material ensures that only one stereoisomer of the product is obtained. This stereoisomer of nootkatone is the one responsible for part of the characteristic scent of grapefruit. Indian turpentine is rich in the sesquiterpenoid hydrocarbon longifolene. In the presence of acid, longifolene rearranges to give isolongifolene. Epoxidation of the latter with a peracid and subsequent acid‐catalysed rearrangement gives isolongifolanone. Prins reaction (acid‐catalysed addition of formaldehyde) of isolongifolene and acetylation of the resultant alcohol gives a mixture of isomeric acetates known as Amboryl Acetate. Both isolongifolanone and Amboryl Acetate have woody, amber odours. These reactions are shown in Figure 15.19. The major components of cedarwood oil are cedrol, cedrene, and thujopsene. These materials serve as starting materials for a variety of fragrance ingredients, some of which are shown in Figure 15.20. Cedrol is easily dehydrated to cedrene, and the mixture of the two can often be used in place of pure cedrene if the reaction conditions are acidic. Methylation of cedrol gives cedryl methyl ether, and esterification with acetic acid gives cedryl acetate. The latter can also be obtained by addition of acetic acid to cedrene. Epoxidation of cedrene gives cedrene oxide, and this material can be used to give Ambrocenide via its reaction with acetone in the presence of a Lewis acid. Probably the most significant derivative of cedarO O

Longifolene

Amboryl acetate®

H+

O

(1) H2CO/H+

(1) RCO3H (2) H+

Isolongifolanone

(2) Ac2O

Isolongifolene

Figure 15.19  Fragrance ingredients from longifolene.

+

O O

  ­Terpenoid Productio

O

O

O

O

O

O

Cedrene oxide

Cedryl methyl ether Cedryl acetate

Ambrocenide®

OH

Cedrene Cedrol

H

Thujopsene

O

O H

‘Acetyl cedrene’

Figure 15.20  Fragrance ingredients from cedarwood oil.

wood is acetylcedrene. This product is sold under a variety of trade names, including methyl cedryl ketone and Vertofix. To prepare it, cedarwood oil is treated with either acetyl chloride or acetic anhydride under strongly acidic conditions using either a Lewis or Brønsted acid. The major product of the reaction is acetylcedrene, but the most important is the acetylated product from the product of acid‐catalysed rearrangement of thujopsene. All of these cedarwood derivatives have what is described as woody/amber odours. The balance of woody and amber notes varies from one product to another and from one grade of a product to another. The most important amber ingredient is the naphthofuran known under such trade names as Ambrofix, Ambrox, and Ambroxan. This material was first identified in ambergris. A triterpenoid called ambreine is produced in the intestine of the sperm whale and excreted into the sea. Exposure to salt water, air, and sunlight causes a number of chemical reactions to occur, and the ambreine degrades to a complex mixture of chemicals known as ambergris. In odour terms, the most important of these chemicals is the naphthofuran. The scarcity of natural ambergris means that virtually all of the material used in commerce is synthetic. Some of it is produced from petrochemical precursors, but the largest source is from sclareol (a diterpenoid that occurs in clary sage and is isolated from the distillation residues after the essential oil has been removed from the herb). The chemistry involved in this process is shown in Figure  15.21. The oxidation of sclareol to sclareolide was originally carried out using heavy metal oxidants such as potassium permanganate or sodium dichromate, but nowadays it is done using a biotechnological approach involving microbial oxidation. Sclareolide is

301

302

15  Synthetic Fragrance Ingredients O OH OH H Sclareol

O

[O] H Sclareolide

(i) LiAlH4 or BH3

O

(ii) H+/–H2O H Ambrofix

Figure 15.21  Amber ingredients from clary sage.

used as a tobacco flavour as well as an intermediate in this synthesis. Reduction of the carbonyl function in sclareolide using either lithium aluminium hydride or borane gives the corresponding diol, and this product can be cyclised to the naphthofuran under acidic conditions.

­ roduction of Fragrance Ingredients P from Petrochemicals Crude mineral oil is separated into fractions by distillation. From the point of view of our industry, the most important are the gases such as ethane and the naphtha fraction, which contains hydrocarbons with 5–10 carbon atoms per molecule. The latter is also the fraction that is used to manufacture gasoline. Other processes carried out in oil refineries are cracking that reduces the carbon chain length of molecules; steam cracking, which increases the degree of unsaturation in molecules; and reforming, which increases the degree of branching of chains and cyclisation. The largest part of the volume produced is used as fuel. Other major uses, in decreasing order of volume, are polymers (polyolefins, synthetic rubber, polyesters, polyamides), solvents, lubricants, and surfactants. The fragrance industry’s use of petrochemicals is tiny compared with these others and to many other branches of the chemical industry. In the field of petrochemicals, therefore, even more than in the turpentine area, the fragrance industry tends to make tactical use of the feedstocks and intermediates that are produced to serve much larger industries. We have already seen how basic petrochemicals such as acetone, acetylene, and isobutene are used to produce terpenoid fragrance ingredients. In this section we will look at the other fragrance ingredients that are manufactured from petrochemicals. Perhaps the most basic of all petrochemical feedstocks is ethylene. The chemistry used with ethylene can also be applied to higher olefins, especially those with double bonds at the end of the chain. Two commercially important reactions of olefins are oligomerisation/polymerisation and hydroformylation. Oligomerisation means the joining together of a small number of monomers (single double bond units), whereas polymerisation means the joining of many olefinic units into a long chain. Because fragrance ingredients are small molecules, oligomerisation is much more important for us than is polymerisation. Hydroformylation is the addition of carbon monoxide to an olefin in the presence of hydrogen and a catalyst. Depending on the reaction conditions, the

  ­Production of Fragrance Ingredients from Petrochemical O O

R

R′

R′OH O R

OH [O]

Oligomerisation

CO/H2 R

R

OH – H2

R

O

Figure 15.22  Aliphatic fragrance ingredients from ethylene.

product will be an alcohol or an aldehyde, and of course, each of these chemicals is easily converted to the other. Oxidation of either leads to the corresponding acid, and from these, the esters can be prepared. Thus, the combination of oligomerisation and hydroformylation opens up routes from ethylene to a large range of aliphatic alcohols, aldehydes, acids, and esters as shown in Figure 15.22. The figure shows the most common pattern of reactions, that is, the hydroformylation gives an alcohol that can be dehydrogenated to the corresponding aldehyde or oxidised to the corresponding acid that can then be esterified. Butane and butenes can be prepared from ethylene but are also isolated directly from petroleum and from natural gas. Dehydrogenation of these materials gives butadiene, which is a key intermediate in petrochemical synthesis as shown in Figure 15.23. The main use of butadiene is in the production of polymers. Direct polymerisation is the main application, and one of the main uses is in synthetic rubber manufacture. Cyclic dimerisation gives cyclooctadiene, and cyclic trimerisation gives cyclododecatriene. These chemicals are then used as starting materials for polymers. For example, cyclododecatriene is the starting material for production of dodecane dicarboxylic acid, which is used in manufacture of Nylon 12. Both of these cyclic olefins are used for preparation of fragrance ingredients, for example, macrocyclic musks, as will be seen later. Isobutene, also known as isobutylene, is another important intermediate in the petrochemical industry. Addition of methanol to it gives methyl tertiary butyl ether (MTBE), which is used as an anti‐knock component of unleaded petrol (gasoline). Dimerisation of isobutene gives di‐isobutylene, which can be hydrogenated to give the trimethylpentane commonly known as iso‐octane. This material is the fuel standard on which the octane rating system is based. Hydroformylation of di‐isobutylene gives an alcohol commonly known as inonanol. The aldehyde derived from this alcohol as well as some of its esters is used as

303

304

15  Synthetic Fragrance Ingredients

Butane/butene Polymers especially synthetic rubber

Butadiene Cyclooctadiene

Cyclododecatriene

Figure 15.23  Butadiene as an intermediate.

fragrance ingredients. Addition of two molecules of formaldehyde to isobutene gives a dioxane, which can be cleaved to give isoprene. Isoprene is another feedstock for the polymer industry, and it is also a useful starting material for fragrance ingredients, especially in the terpenoid field. These reactions of isobutene are shown in Figure 15.24. Figure 15.25 shows some useful intermediates that can be prepared from acetone. Base‐catalysed aldol reaction of acetone is easy and gives diacetone alcohol. It can be reduced to the diol (known as hexylene glycol), and then the hydroxyl groups can be eliminated, either one at a time (the tertiary alcohol dehydrating first) or together in one pot under more vigorous conditions. The

MTBE

Isobutene 2CH2O

O Iso-octane

OH

Di-isobutylene

Inonanol

+ O

O Isoprene

Figure 15.24  Isobutylene as an intermediate.

CH2O

+

H2O

  ­Production of Fragrance Ingredients from Petrochemical O

OH

O Acetone

Hexylene glycol

Mesityl oxide

OH O

O

OH OH

Phorone Diacetone alcohol OH

O

Ligustral

O

Methylpentadiene

Methyl isobutyl ketone

Methyl isobutyl carbinol

O

Isophorone

Figure 15.25  Intermediates from acetone.

final product is methylpentadiene. Obviously, this material is a good reagent for the Diels–Alder reaction, and, for example, addition of acrolein (propenal) gives Ligustral, a useful green top note ingredient. Isophorone is a tantalising intermediate because of its structural relationship to the ionones and damascones, but no really efficient route from it to these valuable fragrance ingredients has yet been developed. Benzene can be isolated from the naphtha fractions of petroleum and is an important synthetic intermediate for our industry. Some of the more important processes using it are shown in Figure 15.26. One process of major importance is the styrene monomer/propylene oxide (or SMPO) process. It is illustrated, starting with benzene at the top right of the figure. First, ethylene is added to benzene in a Friedel–Crafts reaction to give ethylbenzene. Autoxidation of the latter gives the benzylic hydroperoxide, which is then reacted with propylene. The result gives a mixture of styrallyl alcohol (1‐phenylethanol) and propylene oxide. The latter has many applications including the manufacture of solvent and antifreeze agents. One trace by‐product is 2‐phenylethanol. It may be a trace by‐product as far as the petrochemical industry is concerned, but for the fragrance industry, the volume available is substantial. The major issue in isolating 2‐phenylethanol (known to the fragrance industry mostly as phenyl ethyl alcohol) from the major products of the SMPO process is that of achieving satisfactory organoleptic quality. One of the key problems is that, when 2‐phenylethanol is isolated (by distillation) from the SMPO product mixture, it contains minute traces of 2‐ethylphenol. This phenol has such an intense odour that even parts per million of it will render the 2‐phenylethanol unusable as a fragrance ingredient. Methods have been developed for efficient purification of 2‐phenylethanol obtained from the SMPO

305

306

15  Synthetic Fragrance Ingredients O

OH O2

O

O +

H Styrallyl alcohol

+

O/AlCl3 O

–H2O

OH

OH H2/cat.

[O] Ac2O

2-Phenylethanol O Trace

MeOH

H+

O

O

O

O Styrallyl acetate

PADMA

Phenylacetaldehyde

Figure 15.26  Benzene as feedstock.

process, and this process now represents the major source of this most important fragrance ingredient. Acetylation of styrallyl alcohol (e.g. using acetic anhydride as reagent) gives the fragrance ingredient styrallyl acetate, but the major use of styrallyl alcohol is in the manufacture of styrene, by dehydrating it. Styrene is, as everyone knows, important for its polymer, polystyrene. Epoxidation of styrene gives styrene oxide. Hydrogenation of styrene oxide gives 2‐phenylethanol. Another process for manufacture of 2‐phenylethanol is the addition of ethylene oxide to benzene. These two routes give a higher quality of 2‐phenylethanol, but the cost of the by‐product from the SMPO process is lower. Acid‐catalysed rearrangement of styrene oxide gives phenylacetaldehyde. This aldehyde is quite unstable and is usually supplied as a solution in 2‐phenylethanol, which stabilises it by formation of a hemiacetal. Another way of delivering phenylacetaldehyde is via its dimethyl acetal, phenylacetaldehyde dimethyl acetal or PADMA for short. The acetal has an odour reminiscent of the sharp green tones of the aldehyde and, in acidic conditions, will hydrolyse to release the aldehyde. A family of aldehydes exists, which for convenience, I will refer to as the muguet aldehydes. These important fragrance ingredients can be synthesised from benzene as shown in Figure 15.27. Friedel–Crafts‐type addition of acrolein diacetate (R′  =  H) or methacrolein diacetate (R′  =  CH3) to cumene (R  =  H) or tertiary butylbenzene (R = CH3) as appropriate gives (after hydrolysis of the intermediate enol acetate) either Cyclamen aldehyde, Bourgeonal, or Lilial. The latter two are also accessible from toluene by the route shown at the bottom of the figure. Acid‐ catalysed t‐butylation of toluene followed by oxidation of the methyl group gives t‐butylbenzaldehyde. Aldol condensation of t‐butylbenzaldehyde with acetaldehyde or propionaldehyde, followed by hydrogenation of the double bond, gives Bourgeonal or Lilial, respectively. This method is not so successful in the case of

  ­Production of Fragrance Ingredients from Petrochemical R′ O

+

R

O

O O

O

O

O

Bourgeonal

Cyclamen aldehyde

Lilial

O [O] H+ t-Butylbenzaldehyde

Figure 15.27  Manufacture of the muguet aldehydes.

Cyclamen aldehyde because of the difficulty of selectively oxidising the methyl group in the presence of the isopropyl group in p‐cymene (4‐isopropyltoluene). The manufacture of muguet aldehydes is illustrated in Figure 15.27. This account brings us to the chemistry of toluene, another starting material isolated from naphtha. In Figure 15.28, we see the two major routes used for the benzylic oxidation (that is, oxidation of the methyl group adjacent to the benzene ring) of toluene. Air oxidation is the cheaper and easier alternative, but it tends to give complete oxidation to the acid, benzoic acid. Even under conditions designed to give under‐oxidation, the aldehyde will only constitute a small percentage of the total product, and the amount of alcohol produced will be negligible. Benzoic acid is a high volume chemical, and so, as with the SMPO process, the small amount of by‐product produced will still be significant as far as the fragrance industry is concerned. The other way to effect the benzylic oxidation of toluene is by chlorination under free radical conditions. (Chlorination under acidic conditions leads to chlorination in the ring rather than on the methyl group.) The mono‐, di‐, and trichlorides are produced as a mixture. The balance between them depends on reaction conditions, in particular on the ratio of ­chlorine to toluene in the reactor. These three products are known as benzyl

307

308

15  Synthetic Fragrance Ingredients OAc

OH H2 O

NaOAc

H2O

Cl2/R* Cl O2

Cl

+ By-product

O

H2O

Cl

+

Cl

O H2O OH

Cl

Cl

Figure 15.28  Benzylic oxidation of toluene.

c­hloride, benzal chloride, and benzo trichloride, respectively. Hydrolysis of the chlorides gives the corresponding oxygenated materials. Hydrolysis of benzyl chloride gives benzyl alcohol; of benzal chloride, benzaldehyde; and of benzo trichloride, benzoic acid. Normally the chlorination is run at conditions to give mostly benzyl chloride with some benzal chloride as a by‐product. There is no sense in using chlorination to prepare the trichloride, as the acid is available much more cheaply from the air oxidation. Hydrolysis of benzyl chloride is problematic because of the formation of side products, so it is usually preferable to treat it with sodium acetate to form benzyl acetate and then obtain the alcohol by hydrolysis of the acetate. Besides, the acetate is the most important of the benzyl esters in perfumery, and so it is best to prepare it directly from the chloride. Some of the more important uses of these oxidised derivatives are shown in Figure 15.29. Shown at the top of the figure are the three oxidation products: benzyl alcohol, benzaldehyde, and benzoic acid. The alcohol is used to prepare a wide range of esters, the most significant of which is benzyl acetate, a major component of jasmine oil. Similarly, the acid is used to prepare a range of benzoate esters, many of which provide long‐lived fixative balsamic notes in fragrances. A number of products are the result of the Prins reaction of benzaldehyde. This reaction is the acid‐catalysed addition of an aldehyde to an olefin. In the two examples shown, a di‐olefin is used, and the initial product is therefore an unsaturated alcohol. This cyclises to produce a pyran derivative (a six‐­membered ring containing one oxygen atom). If the diene is methylpentadiene (see above for its synthesis from acetone), then the product is Pelargene, whereas isoprene gives Rosyrane. The latter is not so important in its own right but is an intermediate in the production, by hydrogenation, of the alcohol known as Mefrosol or

  ­Production of Fragrance Ingredients from Petrochemical O

OH

Benzyl alcohol

O

Toluene

Benzoic acid

O

R

O

O

R

O

Benzyl esters

Benzaldehyde

Benzoate esters RCH2CHO O

O Rosyrane

O Pelargene

OH

O Mefranal

R Cinnamic aldehydes

Mefrosol

OH R Cinnamic alcohols

Figure 15.29  Toluene as feedstock.

Phenoxanol. This material is an important ingredient in the rose family of odorants. Dehydrogenation of Mefrosol gives Mefranal. Aldol condensation of benzaldehyde gives cinnamic aldehyde derivatives. Using acetaldehyde as the other component in the aldol condensation gives cinnamaldehyde itself (R = H in the figure). Higher aldehydes are also important, particularly the condensation products of heptanal and octanal, known as amyl cinnamic aldehyde and hexyl cinnamic aldehydes, respectively. These two aldehydes are substantive and provide a good fatty background for jasmine‐type fragrances. Reduction of the aldehyde function provides the cinnamic alcohols. Only the lower members of this series, such as cinnamyl alcohol itself (R  =  H in Figure  15.29), have sufficient odour to make them useful in fragrances. Naphthalene is another hydrocarbon that serves as a starting point for synthesis of fragrance ingredients. It is best known as an alternative to camphor in mothballs and other products to repel the said insects. Friedel–Crafts acylation of naphthalene gives methyl naphthyl ketone. Ring oxidation to give β‐naphthol followed by alkylation of this phenol gives, according to the alkylating agent used, methyl β‐naphthyl ether, commonly known as yara yara, or ethyl β‐­naphthyl ether, commonly known as nerolin bromelia. More vigorous oxidation cleaves one of the rings of naphthalene to give phthalic acid. This acid is also obtained by air oxidation of o‐xylene (1,2‐dimethylbenzene). A variety of phthalate esters are

309

310

15  Synthetic Fragrance Ingredients

used in many applications. Some higher phthalates are used as plasticisers in polymers such as poly vinyl chloride (PVC). Adverse publicity concerning these higher phthalates has had an effect on our industry as a result of lack of understanding, with the result that diethyl phthalate (DEP), which once served as a perfumery solvent, has virtually disappeared from use in our industry. Using a reaction known as the Hoffmann degradation, anthranilic acid can be produced from phthalic acid. The methyl ester of this acid occurs in many essential oils to which it imparts a characteristic heavy sweet odour. It is important in perfumery because of its ability to form Schiff ’s bases with aldehydes and ketones (see Chapters 3 and 8). The reactions of naphthalene are shown in Figure 15.30. Phenol is a very important intermediate for fragrance ingredients. It is manufactured as shown in Figure 15.31. Friedel–Crafts addition of propene to benzene gives cumene, and autoxidation of this material gives cumene hydroperoxide.

O

O

O

Yara yara

Methyl naphthyl ketone

Nerolin bromelia

Naphthalene O

O

O

O

OH

O

OH

O

O

Schiff’s bases

NH2

O

DEP

Phthalic acid

Methyl anthranilate

Figure 15.30  Naphthalene as feedstock. Benzene

Phenol OH

O

+ Cumene Propene

Cumene hydroperoxide

OH

+ O

Acetone

Figure 15.31  Manufacture of phenol.

  ­Production of Fragrance Ingredients from Petrochemical

Cleavage of this hydroperoxide gives two valuable feedstocks, phenol and acetone. Etherification and alkylation of phenol can be used to prepare the spicy/herbal ingredients anethole and estragole (also known as methylchavicol). However, most of the anethole required by the industry is isolated from CST as mentioned earlier. Self‐etherification of phenol gives diphenyl oxide, which is used as a middle note in rose and geranium fragrances, particularly for fragrances for use in soaps and laundry detergents. Reaction of phenol with isobutylene under acidic conditions produces the phenol known as terbutol. This phenol is an intermediate in the production of some antioxidants, and hydrogenation of it gives the corresponding cyclohexanol. The acetate ester of the latter is used in quantity in the fragrance industry under various names, though the acronym PTBCHA (for p‐tertiary‐butylcyclohexyl acetate) tends to be the most common. Treatment of phenol with carbon dioxide under acidic conditions gives salicylic acid. Acetylation of the phenolic function gives acetylsalicylic acid, better known as aspirin, whereas esterification of the acid function gives a variety of fragrance ingredients. This illustration is another example of the fragrance industry’s reliance on the intermediates of a larger industry. Methyl salicylate is the major component of oil of wintergreen and has an odour very reminiscent of that oil. The higher esters of salicylic acid, such as hexyl salicylate, have odours that fall into the category of base notes, and for this author at least, hexyl salicylate is strongly reminiscent of the end notes of certain sun screen oils. The addition of carbon dioxide to phenol is quite regioselective in that the major product, by far, is the ortho‐substituted phenol. Addition of formaldehyde is less selective and gives a mixture of the ortho‐ and para‐hydroxymethylphenols. Oxidation of these materials to the corresponding aldehydes gives salicylaldehyde and p‐hydroxybenzaldehyde, respectively. Condensation of salicylaldehyde with acetic anhydride (and concomitant loss of one of the acetate units) gives coumarin, an important balsamic note with a hay‐like character. Methylation of the phenolic function of p‐hydroxybenzaldehyde gives anisaldehyde, which has a strong sweet odour reminiscent of hawthorn blossom. These reactions are all shown in Figure 15.32. Oxidation of the ring in phenol gives guaiacol, and this process is the starting point for yet another range of sweet and spicy fragrance ingredients as shown in Figure 15.33. Monomethylation of guaiacol followed by formylation (addition of a CO group) gives vanillin, the component responsible for most of the characteristic odour of vanilla. The analogous sequence using ethylation rather than methylation gives ethyl vanillin, which has an odour similar in character to that of vanillin but more intense. Vanillin is not very stable chemically and tends to cause discoloration in consumer products, so various analogues have been produced to try to replicate the odour but avoid the problems with colour. One such is the isobutyl ester of vanillin, known as Isobutavan, and another is Ultravanil, which is prepared by reduction of the aldehyde function of ethyl vanillin. Preparation of the cyclic methylene ether of guaiacol followed by formylation gives heliotropin, also known as piperonal. This aldehyde comprises about 98% of the oil of heliotrope flowers, from which it takes its name. When camphene is added to guaiacol under strongly acidic conditions, the terpenoid hydrocarbon

311

312

15  Synthetic Fragrance Ingredients OH O O

O

Estragole

Anethole O

OH

Terbutol

O

O

Ptbcha OH

OH

O

OH

Phenol

Salicylic acid

O

OH

O +

OH

Aspirin

OH

O

p-Hydroxybenzaldehyde

Salicylaldehyde

O

O O

O

O

R Salicylates

O Coumarin

Diphenyl oxide

O Anisaldehyde

Figure 15.32  Phenol as feedstock.

OH

OH

O OH

O Phenol

O

O

Guaiacol

O Heliotropin

O Helional OH

OH O

O

OH

O

OH O

O

O

C10H17 Terpenophenols

O Vanillin

O Isobutavan

O Ethylvanillin

Ultravanil

Figure 15.33  Ingredients from phenol via guaiacol.

undergoes various rearrangements, and some of the carbocations produced add to the phenol to give a mixture of terpenoid substituted phenols. Hydrogenation of this mixture causes one of the oxygen atoms to be lost by hydrogenolysis (i.e. loss of a group by addition of hydrogen across the bond linking it to the remainder of the molecule). Each of the possible reduction products can exist as a mixture of geometric isomers, and so the final product is a mixture of 128 different C10 substituted cyclohexanols, varying by the nature of the C10 species, the relative positions of attachment of the C10 unit and the hydroxyl function to the ring and the relative geometry of the two ring substituents. Three of these isomers

  ­Production of Fragrance Ingredients from Petrochemical

possess strong sandalwood odours. It would not be commercially worthwhile to separate the active isomers from the remainder, and so the whole reaction product mixture is used as such. The odour varies with different product compositions and is dependent on the exact conditions used. Thus, each manufacturer has one or more distinct products, which are sold under a variety of trade names. Generically, they are referred to as sandalwood terpenophenols or as isobornyl cyclohexanol. The latter is, of course, a misnomer as it suggests only a limited set of the isomers present in any of the products. Hydrogenation of phenol gives cyclohexanone, which is also manufactured by oxidation of cyclohexane. Oxidation of cyclohexanone under appropriate conditions results in cleavage of the ring and the formation of cyclohexane dicarboxylic acid, commonly known as adipic acid. Adipic acid is a primary intermediate in the manufacture of Nylon 6, and its high volume of production reflects this phenomenon. As far as the fragrance industry is concerned, the ready availability of adipic acid opens up possibilities for the synthesis of jasmine ingredients as shown in Figure  15.34. When the calcium or barium salts of adipic acid are heated strongly, a reaction related to the Claisen condensation (an aldol type of reaction) occurs to form a five‐membered ring. The remaining acid function is in the β‐position relative to the newly formed ketone group, and so it decarboxylates (loses carbon dioxide) at the high temperature involved, and the overall product of the reaction is therefore cyclopentanone. Aldol condensation of cyclopentanone with aldehydes produces 2‐alkylidenecyclopentanones, which can be hydrogenated to give 2‐alkylcyclopentanones. Those with five‐, six‐, and seven‐membered side chains are useful ingredients, available under various trade names. In the figure, the names Quintone, Jasmatone, and Heptone are used, respectively. As these names suggest, the six‐membered side chain gives OH

O

Phenol

Nylon

Cyclohexanone [O]

O HO

OH

Cyclohexane

O

O

O

Adipic acid O

γ-decalactone

Cyclopentanone

O

O

O

O

O

Quintone

O

Methyl dihydrojasmonate

Jasmatone

Figure 15.34  Adipic acid as feedstock.

Heptone

313

314

15  Synthetic Fragrance Ingredients

the strongest jasmine impression of the three. The material with the five‐­ membered side chain has a somewhat more ethereal character, while the seven‐ membered side chain tends towards a peachy note. Baeyer–Villiger oxidation (use of a peracid to convert a ketone to an ester) enables lactones to be produced from these cyclic ketones. The example shown in the figure shows the formation of γ‐decalactone from Quintone. The shorter chain lactones prepared in this way have buttery odours, whereas the longer ones are more coconut in character. Cyclopentanone is also the key intermediate for production of methyl dihydrojasmonate. Figure 15.35 shows, in more detail, some of the chemistry of the schemes outlined in Figure 15.34. The first step is the aldol condensation of cyclopentanone with pentanal to give 2‐pentylidenecyclopentanone. The condensations with hexanal and heptanal are exactly analogous, and hydrogenation of these 2‐alkylidenecyclopentanones gives the saturated analogues as shown in Figure 15.34. Treatment of 2‐pentylidenecyclopentanone with either an acidic catalyst or a platinum group metallic catalyst causes isomerisation of the double bond into the more stable endocyclic position to give 2‐pentylcyclopentenone. This substance is a good acceptor in the Michael reaction, so 1,4‐addition of the anion derived from diethyl malonate in the presence of base leads to formation of the Michael adduct as shown in the figure. Partial hydrolysis of this diester (that is, hydrolysis of only one of the ester functions) and subsequent decarboxylation of the carboxylic acid thus formed gives methyl dihydrojasmonate. This material is an analogue of methyl jasmonate, an organoleptically important component of O O Base

O H+ or Pd

+ O

2-Pentylidenecyclopentanone

2-Pentylcyclopentenone O O

Base

O O O

O

O O

O

Methyl dihydrojasmonate

Figure 15.35  Manufacture of methyl dihydrojasmonate.

O

O O

Michael adduct

  ­Production of Fragrance Ingredients from Petrochemical

jasmine oils and extracts, and was first introduced to the market under the trade name Hedione®. It has proved very valuable in many different types of fragrance composition, and its production volume has increased so dramatically that it is now one of the higher volume fragrance ingredients. Another cyclopentane derivative of importance as a starting material from fragrance ingredients is cyclopentadiene. The five‐membered ring of this molecule is rather strained, so its double bonds are more reactive than usual. They are capable of reacting individually as dienes or together as a dienophile in the Diels– Alder reaction. Thus, one molecule of cyclopentadiene can add to another in the Diels–Alder reaction to give a dimeric product known as dicyclopentadiene (DCPD). This reaction occurs spontaneously above about −40 °C, and so the dimer is the common form of the material. The monomer can be released by heating the dimer to 200 °C, so both species are available for further reactions. A  selection of the products available from them is shown in Figure  15.36. An obvious application is to use cyclopentadiene as the diene component in Diels– Alder reactions, and the examples shown are Chrysanthal (from cyclopentadiene and 2‐hexenal) and Herbanate (from cyclopentadiene and ethyl 4‐methyl‐2‐­ pentenoate). The double bond in the six‐membered ring of DCPD is strained and more reactive than usual for a 1,2‐disubstituted double bond. For example, in the presence of a strong acid, it will add to carboxylic acids to produce esters. A number of these materials are available under a variety of trade names. The examples shown in the figure are the acetate (Verdyl acetate), the propionate (Verdyl propionate), the isobutyrate (Gardocyclene), and the pivalate (Pivacyclene). These esters have mostly sweet, floral odours. Addition of methanol to DPCD gives an ether, known as Verdalia, which has a green odour. As its

O O Chrysanthal

O

Herbanate

Cyclopentadiene

O

O

O Dupical

Pivacyclene

O

Dicyclopentadiene

O O Fruitate

Gardocyclene O

O Verdalia

O

O Verdyl acetate

Figure 15.36  Cyclopentadiene as feedstock.

O

O Verdyl propionate

315

316

15  Synthetic Fragrance Ingredients

name suggests, Fruitate has a fruity odour, whereas the odour of Dupical is fresh floral. Both of these require more complex synthesis routes from DPCD. The musks form a large group of important ingredients that are best discussed as an odour family rather than on the feedstocks from which they are prepared. The evolution of the musk family is also discussed in Chapter 17 in the context of sustainability. There are four main groups of musks, and each of these will be described separately. The first synthetic musks to be commercialised were the nitro musks. They were discovered by chance when it was found that some intermediates in the synthesis of potential explosives related to trinitrotoluene (TNT) possessed musk odours. The nitro musks are prepared by nitration of the appropriate aromatic precursors using a mixture of concentrated nitric and sulfuric acids. In the late nineteenth and early twentieth centuries when no other synthetic musks had been discovered, no good routes to nature identical musks existed, the price of natural materials was very high, and the nitro musks became important fragrance ingredients. During the twentieth century a number of issues began to emerge with the nitro musks. Although they are inexpensive in raw materials terms, the processes used to manufacture them do entail some degree of hazard. Some nitro musks were found to be phototoxic, and all of them are relatively slow to biodegrade in the environment. A combination of these factors and the introduction of alternative musks resulted in a decline in nitro musk use, and today, they are essentially obsolete. Figure 15.37 shows some typical nitro musks. Musk ambrette is still mourned by perfumers who find its particular character very difficult to copy using other musks. In the second half of the twentieth century, the musk market was dominated by the polycyclic musks (PCMs). The materials in this group contain two or three rings fused together, hence their names. Three typical examples are shown in Figure 15.38, together with brief outlines of their syntheses. Fixolide (also known O

NO2

O2N

NO2

Musk xylene

O2N

NO2

O2N

Musk ketone

NO2

O2N

O NO2 Musk ambrette

Figure 15.37  Some nitro musks.

Moskene

  ­Production of Fragrance Ingredients from Petrochemical

O

+

Fixolide

+

O

OH Galaxolide

HO +

O

Traseolide

Figure 15.38  Some polycyclic musks.

as Tonalid) is prepared from p‐cymene, which is available from turpentine. Reaction with two molecules of neohexene (3,3‐dimethyl‐1‐butene) results in cycloaddition of one of the neohexene molecules to give the bicyclic hydrocarbon, which, on treatment with an acylating agent such as acetyl chloride/­aluminium chloride, gives Fixolide. In the cycloaddition stage, the second neohexene molecule serves to form a cation from p‐cymene and is lost as neohexane in the process. A similar cycloalkylation reaction between isoamylene (2‐methyl‐2‐butene) and cumene (isopropylbenzene) gives the bicyclic hydrocarbon from which Galaxolide is prepared. Acid‐catalysed addition of propylene oxide gives an alcohol, and reaction of this material with formaldehyde results in ring closure to form Galaxolide. Isoamylene also serves as a starting material for the synthesis of Traseolide. In this case, it is reacted with the alcohol shown in the figure and which is prepared by reduction of the ketone made by Friedel–Crafts isobutyrylation of toluene. Acetylation of the resultant hydrocarbon gives Traseolide. The PCMs are currently in decline because of relatively poor biodegradability and are being replaced by the macrocyclic musks and the recently discovered alicyclic musks. As discussed above, these latter musks are formed from the alcohol cyclodemol, and a typical example is Helvetolide, the first member of the family to be discovered. Its synthesis is shown in Figure  15.39. Addition of ­cyclodemol to isobutylene epoxide gives an alcohol that, when esterified with propionic acid, gives Helvetolide.

317

318

15  Synthetic Fragrance Ingredients H3C CH3

H3C O

H3C CH3

H3C CH3

H3C O

OH CH3 Cyclodemol

CH3

CH3 CH3

OH

O

CH3

CH3 O

O

CH3 CH3 Helvetolide

Figure 15.39  Synthesis of Helvetolide.

The natural musks such as muscone and ambrettolide are members of the macrocyclic musk group. This fact was established in the first part of the twentieth century by Leopold Ruzicka. The structures might look simple to produce, for example, by forming an ester from a long chain carboxylic acid, which has a hydroxyl group at the opposite end of the chain from the acid function; however, it is not straightforward. The problem is one of entropy, disorder. It is much more likely for the carboxylic acid to react with the hydroxyl group at the end of another molecule of the hydroxy acid than for the two ends of one chain to come together. This leads to polymerisation as one molecule reacts with another and the dimer then adds to a third molecule and so on. This problem was solved by Ruzicka who used very high dilution to prevent reaction between two different molecules and thus improve the chance of intramolecular esterification. Ruzicka was awarded the Nobel Prize in chemistry for this work. However, the high dilutions required to achieve this result in poor reactor utilisation in the factory and thus lead to high process costs. The English chemist, Carrothers, devised an alternative system in which a hydroxy acid is allowed to polymerise. The polymer is then depolymerised under vacuum so that, as monomeric lactone is produced, it distils out of the system. This process works fairly well, but material is lost by side reactions thus lowering the overall yield of the process. With the development of methods for cyclodimerisation and cyclotrimerisation of olefins, materials such as cyclooctene, cyclododecene, and cyclododecanone became available, and development of ring expansion methods has opened new, more efficient routes to macrocyclic musks. Figure 15.40 shows some macrocyclic musks. The top three are the key natural musks: muscone, which is found in the musk deer, ambrettolide from ambrette seeds, and civetone from the civet cat. Nowadays, synthetic versions (nature identical materials) are used instead of these natural sources. The lower part of the figure shows two of the highest production volume macrocyclic musks, cyclopentadecanolide and ethylene brassylate. The preparation of the latter is described earlier in this chapter. Two examples of the abovementioned ring expansion strategy for macrocyclic musk synthesis are shown in Figure 15.41. The first example starts from cyclododecanone, which is available from cyclododecatriene. The carbanion derived from 3‐hydroxy‐1‐butyne is added to the carbonyl group to give the acetylenic diol. This molecule is perfectly set up to undergo what is known as the Nazarov reaction. This reaction gives an unsaturated bicyclic ketone. The ring expansion is achieved using a reaction known as the Eschenmoser fragmentation, which produces an acetylenic ketone that is easily hydrogenated to give muscone. The

  ­Production of Fragrance Ingredients from Petrochemical O O

O

O Ambrettolide

Civetone

Muscone O O

O

O

O

O Cyclopentadecanolide

Ethylene brassylate

Figure 15.40  Some macrocyclic musks.

OH O

O

OH Nazarov reaction

O

O

Eschenmoser fragmentation

H2/cat. Muscone O [O] Metathesis Animusk

Figure 15.41  Synthesis of macrocyclic musks by ring expansion.

second synthesis uses a reaction known as olefin metathesis. When subjected to olefin metathesis, two cyclooctene molecules come together to form the 16‐ membered ring of cyclohexadecadiene. Oxidation of one of the double bonds produces a musk known as Animusk. Menthol, or more correctly l‐menthol, is an important flavour material. It is not used much in fragrances, but oral care products such as toothpaste and mouthwash depend heavily on it. The processes for manufacture of oral care flavours are more similar to those for manufacture of fragrances than to those used to manufacture other flavours. So oral care flavours are often supplied from

319

320

15  Synthetic Fragrance Ingredients

the fragrance divisions of F&F companies. There are a number of processes for production of l‐menthol, and, since these serve as an excellent example of the factors that must be considered in working towards sustainability, they will be described in Chapter 17.

­What Is Required of a Fragrance Ingredient? Once upon a time perhaps, the only requirement of a fragrance ingredient was that it had an attractive odour. Nowadays this is far from being the case. Figure  15.42 shows some of the more important requirements of modern fragrance ingredients. Odour is clearly a requirement, and for practical purposes we can break this down into character, intensity, tenacity, and radiance. Character is the odour type: musk, rose, jasmine, marine, and so on. Intensity is the strength that is perceived by the person smelling the material. (Not to be confused with detection threshold, as explained in Chapter 10.) Tenacity is the time the odour will last on a perfumer’s blotter, a piece of towelling, skin, hair, and so on. Radiance is the ability of an odour to ‘fill’ a space. It is similar to ‘bloom’ from soap (the ability of the odour to suffuse the air around the soap) and ‘trail’ (the ability of a fragrance to leave an odour trail in the wake of a wearer). Radiance, bloom, and trail are all combinations of physical and chemical properties relating to diffusion from substrate to air and sensory properties such as threshold and intensity. For all of these odour properties, no right or wrong answer exists; each ingredient has its

During manufacture

Raw materials

In environment

Plant

Process

Safety In use

Availability

Cost

Character Freedom to operate Intensity

Odour

Tenacity

Requirements of a fragrance ingredient In formulae

Radiance

Deodorancy

Microbial effects

Additional benefits

Insect repellency

Figure 15.42  Requirements of a fragrance ingredient.

Performance

In products

  ­What Is Required of a Fragrance Ingredient

own blend of the four facets, and all combinations are useful in perfumery. The perfumer can use an intense lemon note with low tenacity and radiance just as he/she can use a musk with a low intensity but long tenacity and high radiance. What the perfumer does want is good performance in formulae and in products. Performance in formulae refers to the way an ingredient works with other ingredients. The perfumer does not want an ingredient that either disappears into the background and makes no contribution to the overall odour or dominates an accord and provides a strident monotone instead of a harmonious blend. The only way to determine the outcome for anyone other than for an experienced perfumer is to work with an ingredient and discover how it performs and how best to use it. Performance in products is more predictable. A labile ester (i.e. one that is easily hydrolysed) will not survive in a laundry detergent, for example. In addition to performance predictions, a fragrance company will evaluate each ingredient in a wide range of product bases in order to provide good quality data on stability in products. In this context, usually it is organoleptic rather than chemical stability that is meant. The organoleptic stability depends on the chemical stability and also on the ability of the ingredient to be released from the product matrix into the air and hence delivered to the consumer’s nose. Overall organoleptic stability is shown schematically in Figure 15.43. The perfume is introduced into a product matrix that may contain other compounds that can degrade some of the fragrance ingredients or prevent them from release back into the atmosphere when in use. The perfumer must know how each ingredient will perform in each product type so that the fragrance released into the air when the product is used will give the desired fragrance effect. Nowadays customers also like perfumes to have properties in addition to their odours, known as additional benefits or secondary benefits. Deodorancy is perhaps the most obvious and certainly the most important. It is the ability of a

Headspace Fragrance ingredients in bottle or drum Degradation

Release

Product matrix containing various active components

Figure 15.43  From bottle to nose.

321

322

15  Synthetic Fragrance Ingredients

fragrance or fragrance ingredient to reduce a malodour. For example, a customer producing an underarm deodorant will appreciate a perfume that works with the base ingredients to suppress the ultimate perception of sweat malodour. Many essential oils and other fragrance ingredients are known to have some antimicrobial activity and can be useful in, for example, reducing spoilage of a product such as soap that might represent food to bacteria. Similarly, essential oils and other ingredients might possess some effects in repelling insects. The best known example is citronella oil, and it is used in candles to repel mosquitoes and other biting insects, a great asset when eating outdoors in summer. It is also important that an ingredient can be made available when the customer wants it and in the quantity that the customer requires. For this outcome, we need a robust chemical process that can operate at the anticipated scale and can easily be scaled up if necessary. The raw materials should be readily available and preferably from at least two different suppliers (to improve security of supply in case of unexpected problems with one of them). Obviously, the raw material and process costs must be such that the final cost of the ingredient is at a level that the performance can justify. Freedom to operate means that the process can be run without infringing any third party’s patents or any laws. Safety is of paramount importance. Three aspects of safety must be considered in this context: safety in the factory, safety in use, and safety after use. For ethical, legal, and financial reasons, it would be very undesirable for a company to endanger either its employees or equipment. Therefore, rigorous safety checks (called HAZOP studies  –  an acronym for hazard and operability) will be carried out before any process is run in a factory. Equally clearly, it would not be in the interests of any company to endanger its customers or consumers. Therefore, any ingredient will be rigorously evaluated for safety in use before being made commercially available. Of course, stringent legal requirements have also been created to ensure consumer safety. Finally, the consequences of releasing the perfume into the environment must be evaluated to ensure that no harm is done to the environment after use of the perfume.

­How Novel Fragrance Ingredients Are Designed? The reasons for making novel fragrance ingredients are much the same as those given above for using synthetic ingredients. Security of supply is less of an issue. The key drivers are therefore to find materials with performance in use that is superior to existing materials, ingredients that perform as well as existing ingredient but cost less, to replace ingredients that have been lost through safety or regulatory issues, and, of course, to seek continual improvements in safety in use and in the environment after use. The fragrance discovery chemist must bear in mind all of the requirements of ingredients that are described in the preceding section of this chapter. He may be working to a brief from perfumers such as ‘Find me a lemon scented molecule that is stable to hypochlorite bleach, repels flies and cockroaches and biodegrades rapidly in a sewage treatment plant’. Alternatively, he may select a target based on his own knowledge of what is required in the industry combined with

  ­How Novel Fragrance Ingredients Are Designed

some technical possibility that he has seen in the chemical literature or similar. The example of a brief cited here is something of a joke in the industry, since such a material is much sought after for application in fragrances for hard surface cleaners. The lemon scent is perceived as indicating cleanliness, the hard surface cleaner will contain hypochlorite bleach as an active ingredient, a lasting fragrance that repels flies and cockroaches will help keep the surface clean for longer, and everyone wants ingredients that biodegrade readily. However, the criteria of molecular structure for a lemon odour tend to lead to molecules that are not stable to hypochlorite. Equally, materials that are stable to hypochlorite are unlikely to biodegrade rapidly since bacterial enzymes tend to be less aggressive than hypochlorite. Similarly, the active insect repellents in citronella oil are not stable to hypochlorite bleach. To be able to balance all of the criteria, the discovery chemist must be well versed in synthetic organic chemistry. Initially he must provide material for perfumery evaluation, but, if his new molecule is selected for development, he should have some idea of potential commercially feasible synthesis routes. Therefore, he must have an awareness of the chemicals market (to know what raw materials are available), and he must have some understanding of process chemistry (so as not to give development chemists an impossible task). In addition, he must be capable of developing structure–activity relationships (SARs) for odour properties, performance properties, additional benefits, and safety. Whatever the background to the chemist’s project, he is, metaphorically speaking, starting with a blank page on which he must draw structures for molecules that might meet his objectives for a new ingredient. How does he go about this? Five major approaches to novel fragrance ingredient discovery are random screening, mechanistic understanding, copying leads, statistical design, and making what you can. Let us look at each of these in turn. Random screening has a very low chance of success. It would, theoretically, be possible to design more organic compounds than there are carbon atoms in the universe to make one molecule of each. The chance of picking a structure at random and finding it to be a better fragrance ingredient than those already on the palette is remote. However, some element of randomness in screening does give a chance of a significant breakthrough into a new class of ingredients. Designing new fragrance ingredients is somewhat like a lottery. There are two certain ways to win a lottery. One is to buy all the tickets, and the other is to buy only the winning ticket. High throughput screening of random samples is an approach along the lines of buying all the tickets. The problem with this approach in a lottery is that it will cost more than the prize. The same is true in fragrance ingredient discovery in that acquisition and evaluation of a sufficient number of random compounds until a winner is found will cost more than the ingredient is worth. At the other end of the scale, we could just make the winning ingredient. But how do we know which molecule to make? Accurate prediction of physical properties such as substantivity and stability in customers’ bases should be possible. As far as odour is concerned, if we were to understand the whole process of olfaction in total detail, then we might be able to make only the right molecule. However, the complexity of the process of olfaction (the combinatorial nature of receptor interaction and the nature of subsequent neuroprocessing) and ­variation

323

324

15  Synthetic Fragrance Ingredients

of odour perception from one individual to another make perfect odour prediction essentially impossible. Since these two approaches to ingredient design are unlikely to succeed, it is the other three that have been the mainstay of the industry and probably will be for the foreseeable future. Copying leads has led to many successful new ingredients. Initially, the leads came from nature and many still do. By identifying the molecules that are responsible for the wonderful scents of nature, we can imitate them by making analogues, which is how molecules such as Hedione (an analogue of methyl jasmonate from jasmine absolute) were discovered. Now that successful synthetic ingredients exist on the market, these also serve as leads, and often we see one fragrance company developing molecules that are similar to those that another company has developed and used successfully. This approach has a relatively high chance of technical success. The difficulties lie in finding good natural leads and gaps in competitors’ patents. Since the new molecules are similar to known ones, the progress in extending the palette is usually incremental. Statistical design of molecules improves the success rate over that of random screening. However, it is by nature an interpolative process, and therefore, as with lead optimisation, progress tends to be incremental. Nonetheless, this approach has been the major tool of the fragrance discovery chemist for some time now. The tool used is known as structure–activity relationships (SARs), or, if the activity is quantified, quantitative structure/activity relationships (QSARs). Many of the ingredient requirements shown in Figure  15.42 are amenable to treatment using SARs or QSARs. The basic principle behind (Q)SARs is the hypothesis that molecular structure determines all the properties of a chemical entity, and so by looking at a number of molecules, including some that have the property and some that do not, a statistical relationship can be found connecting certain features of molecular structure with the desired property. Any molecule possessing these features would then be expected to have a good chance of having the same activity. Examples of QSARs in physical chemistry would include the prediction of boiling points or log P values. The usefulness of calculated log P values has been described in earlier chapters. The more steps that are involved between the structure and the activity, the harder it is to find a good SAR and the less meaningful that SAR will be in mechanistic terms. For odour, very many steps stand between the structure and the activity. First is the delivery to the nasal air, then absorption into the mucus, transport across the mucus, interaction with the receptor, coding of receptor signals onto the olfactory bulb, and processing in the bulb and then a vast number of neural pathways that run sometimes in parallel and sometimes in series and interact with each other. This very complex route explains why structure–odour relationships are difficult to define, usually not highly accurate and never really tell us much about the mechanism of olfaction. The best structure–odour relationship is probably that which John Amoore proposed for molecules with a camphoraceous odour. This rule states that any hydrophobic molecule with an ellipsoidal shape, having a long axis of 0.95 nm and a short axis of 0.75 nm, will smell camphoraceous. This rule has few, if any, exceptions, but why it works is unknown. Other types of SAR models include spatial models with defined dimensions for critical features of molecular

  ­How Novel Fragrance Ingredients Are Designed

structure, models using properties of frontier orbitals, and olfactophores. The last are odour equivalents of pharmacophores. These models show the surface electronic properties of molecules. Some parts of the surface will be electrically neutral and therefore represent essentially simply physical bulk, while other parts will carry positive or negative electrostatic charges. The idea is that a protein that recognises a small molecule will do so through these spatial and electronic fields. Molecules with similar shapes, charge distributions, and polarisability would therefore be expected to be recognised by the same proteins. This model applies well to pharmaceutical research where a single protein is being targeted in the search for a new drug, but, in the fragrance industry, if we are trying to match a known molecule, then we must match its interactions with all 400 or so active human olfactory receptors. This task is daunting, but SARs have been used in the design of many successful fragrance ingredients such as Javanol, Rossitol, Belambre, and Azurone (Figure 15.44). The last and perhaps simplest approach is to make what you can. This method might sound either lazy or naive, but the general approach is to make molecules that we know can be made from readily available starting materials using easy chemistry (preferably chemistry that the company already uses routinely in its factories) and that SARs would suggest are likely to have useful properties. This approach offers the best chance of ensuring smooth transfer from laboratory to manufacturing and enables better cost prediction and hence better market evaluation. If it is not target directed using SARs, then it will have a low chance of finding a good molecule. The discovery chemist will make only a small sample of the material initially, and it will be used for the first tests in a screening process. The simplest test is odour character. It is done simply by smelling the material. Because of the subjectivity of odour, a panel, usually of experienced fragrance chemists and perfumers, should do this test. If the material passes one test, it will move on to another until the whole range of tests has been carried out and the material’s properties can be profiled against the requirements shown in Figure 15.42. Every novel molecule can fail at any of the tests, and so it is wise to carry out the OH OH Javanol Rossitol O O

O O

Belambre

O Azurone

Figure 15.44  Successful fragrance ingredients designed using structure–odour relationships.

325

326

15  Synthetic Fragrance Ingredients

c­ heapest tests first. This is why the odour properties are usually estimated first. These tests are cheap, can be carried out in minutes, and require only a small sample of the material. Performance tests require larger amounts (usually a few hundreds of grams) of material for incorporation into formulae and product bases and take considerably longer. Assessing feasibility of manufacture will require pilot‐scale batches perhaps up to hundreds of kilograms and could take anything from a few months to a few years. Consequently, the costs of this are much higher. Initial safety and environmental testing and registration will take about a year and will cost up to half a million dollars. Obviously, these very expensive phases of testing will be left to the end of the evaluation process. The total cost of bringing a new material from concept to production is likely to be between one half and one million dollars. The volume growth of novel ingredients is not fast as perfumers need to learn how to use them and then formulae containing them need to go through customers’ selection processes. If the return on R&D investment were to be calculated on the volume of material sold over the first years of the ingredient’s life while a patent on it is still in force, then it would probably never be economical to launch new ingredients. The real return is through the sales of the fragrance containing the ingredient, the ingredient itself contributing to the fragrance through improved performance. This fact is why novel fragrance ingredients are almost invariably launched only by companies with a significant volume of compounded fragrance sales. The initial investment is too high to be borne by smaller fragrance companies and is why novel ingredients come mostly from the larger fragrance companies. Being simultaneously aware of all of the factors outlined in Figure  15.42 requires the brain of an experienced fragrance chemist, and it is why the author believes that the assertion of Ernest Beaux (creator of Chanel 5) that ‘One has to rely on chemists to find new aroma chemicals creating new, original notes. In perfumery, the future lies primarily in the hands of chemists.’ is as true today as it was in 1921 when he created Chanel 5.

Review Questions 1 Why is turpentine a useful raw material for fragrance ingredient synthesis? 2 The most important component of saffron oil is safranal. Why might it be desirable to find a synthetic analogue for this material? O

Safranal

Review Questions  327 O

O

O tBuOCI

Base O

OH

+

Cl

O

Br

O

Na2CO3

CH2(CO2Me)2 H2/cat.

O

O

CO2Me MeLi OH

MeO2C (1) H+/H2O (2) H2/cat.

CrO3

O

O O Jasmone

O Methyl jasmonate

Figure 15.45  Büchi’s synthesis of jasmone and methyl jasmonate.

3 Figure 15.45 shows the route used by Professor George Büchi to synthesise both jasmone and methyl jasmonate. Cyclohexane‐1,3‐dione is alkylated with 1‐bromo‐2‐pentyne, and the product chlorinated with t‐butyl hypochlorite. Treatment with base then results in a ring contraction (known as the Favorski reaction). The product of this reaction can be converted to jasmone by hydrogenation under Lindlar conditions, addition of methyl lithium, and oxidation with chromium trioxide. Alternatively, Michael reaction with dimethyl malonate followed by hydrolysis of one of the ester groups, decarboxylation of the resultant acid, and Lindlar hydrogenation gives methyl jasmonate. This synthesis is very ingenious, but would not make a good commercial route. Why not? What analogues of the two products would you consider making in order to provide alternative products that would be easier to manufacture?

329

16 Chemical Information ­ ow New Chemical Information Is Generated H and Published? Research chemists work on projects and carry out experiments in order to test hypotheses, to develop syntheses, or purely to find out what happens under certain conditions. The results of their experiments may be of academic and/or commercial interest. If the discovery is of academic interest only, researchers will probably wish to publish the work in order to make the results available to others and, of course, to build their reputation as scientific researchers. If, however, the discovery is one that can be exploited commercially, researchers or their employers will probably wish to protect it through the means of a patent. Very often, the work leading to a patent is published in an academic journal after publication of the patent. In this chapter, we will see how information is made available and the ways in which it is made available to those who are interested. Patents Commercially valuable results will usually be published initially as a patent. A patent is essentially a contract between the researcher (or his employer) and the government of the country in which the patent is filed. In return for the researcher disclosing whatever discovery he has made, the government grants the patent holder the right to benefit exclusively from the findings for a fixed period (usually 20 years) in that country. A researcher working in a commercial organisation will almost certainly have signed an agreement with his employer that any patent rights arising from discoveries of the researcher will belong to the employer. The discoverer will be named on the patent as inventor and the employer as assignee (i.e. patent owner). As far as the fragrance industry is concerned, the three common types of patent claims are composition of matter (compound per se), application, and process. For a claim to be granted, it must be novel (i.e. not already known), non‐obvious (i.e. someone skilled in the art, that is, someone who understands the field of the invention, would not have been able to deduce it logically from information publicly available at the time of filing of the patent), and useful (i.e. it must have a practical application). When a new fragrance molecule is discovered, it might be the subject of a composition of matter patent Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

330

16  Chemical Information

claim. In order for such a claim to be granted, the molecule would have to be unknown at the time of filing, and it would have to be non‐obvious. For example, if an acetate and butyrate of an alcohol are both known, then it will be hard to convince a patent examiner that the propionate is not an obvious molecule to make. However, if the acetate and butyrate both smell of bananas and the propionate smells of roses, then that might well be considered to be surprising and hence non‐obvious. It must also be shown that the new molecule in question has a use, so patents on novel fragrance ingredients will contain examples of how the ingredient can be used to improve perfume formulae. Application patents include such examples as molecules that were known as such but not in the context of fragrance (of course, its use in fragrance would then have to be surprising) or molecules having specific uses, such as deodorant activity, which are not obvious. A process patent would cover, for example, a specific process for manufacture of a fragrance ingredient. Once again, the process would have to have some non‐obvious advantage over any known processes. If a patent is infringed, the patent holder can sue the infringer for damages. Obviously, he must be able to prove that his patent has been infringed. This corroboration is easiest with a composition of matter patent since possession of the substance without the permission of the patent holder constitutes infringement. With an application patent, it can be more difficult since the infringer might claim that he was using the substance for a purpose other than that claimed in the patent. When drafting any patent on fragrance or fragrance ingredients, it is important to cover all potential applications since, otherwise, a competitor could obstruct a patent by patenting one application missed out by the original inventor and his patent agent. The second person would not be able to exploit his patent, but the existence of it would present a possible problem for the first person’s customer and the safest route for the customer to avoid legal proceedings against him would be to avoid using the ingredient, process, or applications patented by both patent holders. For a process patent, the difficulty is that of knowing that one’s patent is being infringed. This summary of patents is only a very brief one, and many layers of subtleties and technicalities exist, which is why every fragrance company hires the skills of at least one good patent attorney. In addition to all the legal complexities of patents, there are some chemical issues of which we must be aware. For example, if someone discovered that the compound in Figure 16.1a had interesting fragrance properties and wished to patent it, he would probably start with a patent claim for a generalised structure around the key discovery. Such a generalised structure might look like the one in Figure 16.1b. The inventor might have made and reported only 10 of these structures, but the patent claim will cover all 65 356 possibilities. Chemical Abstracts will only record the 10 that were reported. Therefore, someone else carrying out a search for patented structures of this type will only find 10. If he were to make and commercialise one of the other 65 346, he would find himself infringing the patent. Equally, the more molecules covered in a general claim, the more likely it is that an inventor is covering a known compound and therefore risking invalidating his patent claim. If a discovery or the results of a research project are of interest to the scientific community or to the fragrance industry, then it can be published in a variety of ways. The work carried out by a student working for a higher degree will be

­How New Chemical Information Is Generated and Published O

O

O

O

R1 R2 R3

R4 R1 – R4 each can be C1 to C5 saturated alkyl radical (a)

(b)

Figure 16.1  Specific and generalised structural formulae.

published as the master’s or doctorate thesis, depending on the degree for which he has been working. Throughout the world, learned societies such as the American Chemical Society (USA), the Royal Society of Chemistry (UK), and the Gesellschaft Deutscher Chemiker (Germany) abound. One of the activities  of such societies is to publish journals that publish reports of scientific research.  So, for example, results of chemical research will be found in the Journal of the American Chemical Society (American Chemical Society), Organic and Biomolecular Chemistry (Royal Society of Chemistry), and Angewandte Chemie (Gesellschaft Deutscher Chemiker  –  in association with Wiley). Com­mercially produced journals, such as Tetrahedron, which is a journal for organic chemistry, also exist. Articles submitted to such journals are subjected to a process of peer review in which two or more experts in the field are asked to comment on whether or not the work is novel and of a standard that merits publication. Other journals, such as Perfumer & Flavorist magazine are directed at specific industries. These journals often contain a mixture of scientific publications (peer reviewed in the better journals) and articles about the industry. Reviews and Books Someone wishing to learn about a subject or to update their knowledge of a specific topic on which they are not an expert will probably not have time to find, read, and assimilate primary literature such as patents and journal articles. For this reason, experts in a given topic write books and reviews. Books are sometimes written for the complete novice (e.g. textbooks) and sometimes serve as longer reviews. Reviews are written essentially for those who are familiar to some extent with the subject. They will therefore be pitched at a higher level. They might be published as a book or appear in journals. Some journals such as Angewandte Chemie and Perfumer & Flavorist magazine contain reviews as well as original research publications, while journals, such as Chemical Reviews (published by the American Chemical Society), are devoted solely to reviews. Abstracts The volume of publications in the field of chemistry is currently so vast that no single person could manage to read all of them, even if they did nothing else. In order to keep up with developments in a field of interest, you can read

331

332

16  Chemical Information

only selected journals, but some material will inevitably be missed. Therefore, systems have been developed to enable people to find publications of interest to them. The first stage is through the provision of abstracts. Abstracting services scan all journals, patents, books, and other publications (e.g. conference proceedings) and write short summaries of the contents. These days many journals ask the author to write an abstract that can then be used by the abstracting services. The largest abstracting service in the field of chemistry is Chemical Abstracts. It is run by the American Chemical Society. Chemical Abstracts was originally published in hard copy containing a list of abstracts and indexes of article titles, authors, chemical formulae, and chemical abstracts (CAS) numbers. The CAS number is a unique identifier for a given chemical substance and is very useful for retrieving information about a given substance as will be seen later. The sheer volume of the hard copy of Chemical Abstracts makes manual searching very time consuming. Chemical Abstracts is now published also in electronic format, and searching using such tools as Sci­ Finder is fast and efficient. Patents are abstracted by services such as Chemical Abstracts, and specialist patent abstracting services such as Derwent also exist. Various patent offices, such as the US and European offices, also have abstract services and are mostly available online through their websites. These patent abstracting services also give information about the status of patents; for example, whether or not an application has been granted or whether a granted patent is still in force or has been abandoned by its owner. Some of the publications in the biosciences are also likely to be of interest to workers in the field of olfaction, so we should also search the relevant abstracting services in bioscience, for example, PubMed or Medline. Another type of abstract is the citation index. When an author cites previously published work, the citation is added to various citation indices, such as Science Citation Index. So, if we find one particular paper of interest, it might be worth checking in a citation index to see who is referring to it later since those later authors will probably be working on a very similar topic and their work is also likely to be of interest to us. Figure 16.2 summarises the general pattern of chemical information as described in the previous paragraphs. Abstracts

Research publications journals, patents

Reviews, books

Citation indices

Figure 16.2  Chemical information process.

­How to Find Chemical Information

­How to Find Chemical Information? If we want to find more about a given molecule or subject, where and how should we start? The first thing is to have access to one or more good libraries. Municipal libraries are unlikely to hold much in the way of specialist publications, but major libraries, such as the British Library or the Library of Congress, will do. University libraries are likely to hold a good range of journals, and most major fragrance companies maintain a specialist library. Electronic resources are increasingly important, and many journals and abstract services are available online – for a fee. Most learned societies have libraries that their members can use. For example, members of the Royal Society of Chemistry can use the RSC Library in London. To search for information on a specific molecule, the best approach is to use Chemical Abstracts, preferably through CAS online or SciFinder. If you know the CAS number for the molecule, it is very easy; just search for that number, and you will find almost every published reference to it. For example, a search for [78‐70‐6] will find a plethora of references to linalool. This number will find references to linalool for which the original source did not specify stereochemistry. If you want to find only the (S)‐(+)‐isomer, you should search for [126‐90‐9], whereas [126‐91‐0] will find only the (R)‐(−)‐isomer and [22564‐99‐4] the racemate. Another useful source of information on individual molecules is ChemSpider. This can be found by a web browser and once on the site, the user will find it easy to use. To search for a research topic, it is best to use a search engine such as SciFinder, though Google (or better, Google Scholar) will turn up a good number of references. If you set your search criteria too widely, you will find so many hits that it will be difficult or too time consuming to select those of real interest. It is best to choose a search engine that will allow you to refine a search, that is, to search the first hit list again using narrower criteria. If you have access to hard copy Chemical Abstracts, it can be searched by subject, chemical formula, author, or CAS number. Indices of these are compiled for each year, and also cumulative indices covering 10 year periods make the task physically easier. The indices will lead you to the abstracts, and then, from these, you can go to the original journal. It is important to remember that searches on abstracts will only retrieve information from the abstract, meaning that what you can find is determined by what the abstracter thinks is important in the original. Similarly, searches using keywords will only retrieve what the author thinks are the most important words or phrases in his paper. Browsing in original journals takes time, but it is the only way for the researcher to see everything that is there and to make up his own mind on its relevance or otherwise for his work. There is a list of recommended further reading in the bibliography that is intended to help explore in more detail some of the subjects covered in this book.

333

335

17 Towards a Sustainable Future ­What Is Sustainability? Sustainability is a topic that receives a great deal of attention in modern busi­ ness life. However, it is not always clear what is meant by sustainable. If it means that something can continue indefinitely, then it is a fiction. The earth we live on is not sustainable; at some point, the sun will explode and destroy the earth. Indeed, the universe itself is not sustainable in that sense. It began with the big bang and is currently expanding, but it will eventually cease to expand and col­ lapse back into the nothingness that was there before the big bang. Therefore, in reality the movement towards ‘sustainability’ is a movement towards longer‐ term survival rather than a guarantee of indefinite operation. For practical purposes in the modern fragrance industry, one commonly used definition of sustainability is along the lines of ‘satisfying the needs of the present generation without destroying the ability of future generations to satisfy theirs’. Of course, anything that is socially, politically, or scientifically unacceptable is likely to damage the immediate future of a company or destroy it completely. Equally, companies must operate profitably; otherwise they will fail as a result of insol­ vency. Therefore, both short‐ and long‐term issues must be considered when planning the future of a flavour and fragrance company. In this chapter, we will look at how the industry has progressed towards increased operational sustain­ ability. We will look at this in terms of commercial, product safety, natural ingredients, synthetic ingredients, and social and health factors and will also consider how important to keep up to date with technical information and cur­ rent thinking and social trends. Ethical issues are also important. Decent and sensible people will want to oper­ ate ethically, and those with duller consciences will still seek to behave ethically because of the penalties they will suffer for not doing so. Financial behaviour has always been a topic for ethical scrutiny, but nowadays large fragrance companies will examine not only their own practices but also those of all companies and individuals along their supply chain. A supplier of any goods or services that behaves in an unethical way will affect the standing of the company that uses those goods or services. This is particularly important with essential oils where there is often a long supply chain of growers, distillers, traders, and so on.

Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

336

17  Towards a Sustainable Future

Unwanted practices would include such things as adulteration of essential oils, exploitation of workers, and child labour. Adulteration of essential oils was a more serious issue in the past, but modern analytical chemistry has served the industry well in the fight against it. One hundred years ago, the dilution of an essential oil with a synthetic equivalent of a major component of it would have been difficult to detect, for instance, addi­ tion  of synthetic linalyl acetate to lavender oil in a proportion such as not to reduce the optical rotation beyond what might be expected on the basis of natu­ ral variation. However, nowadays with techniques such as GC‐MS and isotopic analysis, it would be very difficult to fool the analytical department of a good fragrance house. Of course, as fragrance companies develop methods to protect themselves, the crooks will show their ingenuity by seeking new methods of cheating. The fragrance industry must therefore always be on the alert for such activity. Exploitation of workers or child labour is difficult to prove since inspec­ tions usually have to be announced in advance, giving unethical employers the opportunity to hide their normal practices. There is also an issue of culture to be taken into account. Imagine the situation where a small essential oil grower operating in a poor country has the help of his/her child to harvest the crop. Use of child labour is unacceptable in a culture such as those in Europe or the United States, and so, the ethical multinational company would not wish to purchase the oil thus produced. However, that small farmer probably cannot afford to pay an adult for help and cannot manage to harvest the crop him/herself. By refusing to buy the product, the large company would therefore be guilty of preventing the farmer from earning a living. Similarly, working practices that are acceptable in one country might not be in another. Such ethical issues are therefore complex and must be handled sensitively, and imaginative solutions sought. Ernest Beaux, the perfumer who created Chanel 5, said that ‘One has to rely on chemists to find new aroma chemicals creating new, original notes. In perfum­ ery, the future lies primarily in the hands of chemists’. When he made that remark in the early decades of the twentieth century, he was thinking only of new fra­ grance ingredients that perfumers could use to achieve a competitive advan­ tage in novelty of odour over competitors. However, although chemists are still vital to the industry, their role is changing, and expertise in other subjects, espe­ cially botany, biochemistry, molecular biology, neuroscience, and psychology, is needed to complement that of the chemists. As outlined in Chapter 13, we now know that smell is a combinatorial sense with odorous molecules acting on an array of different receptor types, some broadly tuned others responding to a nar­ rower selection of odorant molecules, and that this is followed by a very complex pattern of interpretation by the brain. As Wilson and Stevenson (2006) write in their book Learning to Smell, ‘With a relatively few exceptions, neither odour physico‐chemical feature extraction at the receptor sheet, nor spatial maps of those features in the olfactory bulb, nor simple convergence of those features in cortical circuits are sufficient to account for the rich experience that is olfaction’. This means that the discovery chemists’ dream of being able to accurately, pre­ cisely, and consistently predict the odour of potential new molecules is a dream

­What Is Sustainability

that will never be realised. At the same time, odour novelty is now only one of many driving factors for research in the fragrance industry. Important factors driving change in the industry nowadays include safety, environment, resources, market trends, performance requirements, sustainability, and scientific dis­ covery. In addition to directed research programmes, opportunity, serendipity, and  external factors can contribute to new discoveries that move technology forward. The world demand for fragrance is growing with the growing population. Mineral oil and sulfate turpentine will inevitably rise in price as availability declines. Naturally sourced ingredients are currently substantially higher in price than identical chemicals made from mineral oil or sulfate turpentine and are unlikely to become substantially less expensive in the future. At the same time, the market trend is for less expensive products. Similarly, the consumer market is moving towards ‘natural’ products, but safety testing continues to find safety issues with natural chemicals. Increasingly stringent environmental con­ straints on providing ingredients, both natural and synthetic, are also growing. Altogether this presents a significant challenge for the industry and the chemists and biochemists working in it. Only a company that seeks inventive solutions and invests in research will be able to meet the needs of its customers and thus ensure survival. Many of the sub‐paragraphs below will show that the necessary research for the future will involve multidisciplinary teams. Commercial Feasibility The dependence of a company’s survival on commercial factors is so obvious that it can be overlooked when considering the whole picture. A fragrance company exists because it provides consumers with a product they want at a price they are prepared to pay. If the price is higher than the consumer will accept, then the company will have a short future. Most companies borrow money to finance developments, and, if they cannot afford the repayments, again they will fail in the short term. They also need to finance research since the industry as a whole is progressing, and companies that do not keep up with technological, safety, and legal developments will become uncompetitive and fail, unless they can find a niche where they can continue to operate, probably on a relatively small scale. All of the developments towards sustainability that we will consider below have cost implications, and so commercial factors must always be added into the survival equation. Safety in Use It is clearly not in the interests of any fragrance company to harm its customers, and all responsible modern fragrance companies go to considerable lengths to ensure that their products do not cause harm to those who use them. Of course, we must remember that it is impossible to prove that anything in life, in general, is totally safe. We can only prove that there are no known instances where harm has been done. We must also remember to distinguish between hazard and risk.

337

338

17  Towards a Sustainable Future

Cars are not safe as the figures on road traffic accidents will easily demonstrate. So, a definite hazard exists. Every time we cross a road, we are exposed to a hazard, that of being injured or killed by a moving car. However, we still cross roads because we, consciously or unconsciously, calculate the risk and carry out a quick risk/benefit analysis. We can, of course, reduce the risk by using a cross­ ing where the traffic is stopped by lights. We are all used to making risk/benefit analyses when crossing a road, but when it comes to various other hazards, humans generally are very good at identifying hazard but poor at calculating risk. Chapter  12 contains a discussion of hazards associated with fragrance ingredients and the methods that the industry uses to assess and reduce the risks involved, so this will not be repeated here. Once the industry identifies a risk as unacceptable, it will take appropriate action. For example, when it was discov­ ered that some ingredients had phototoxic properties, action was taken. In the case of bergaptene (Figure 17.1) in bergamot oil, the solution was to remove the bergaptene by distillation and use analytical controls to ensure its complete removal. With those nitro musks found to have the same problem, the solution was to withdraw them from the market. Problems such as mutagenicity, skin sensitisation, or phototoxicity are known, and so fragrance ingredients can be tested to identify any hazards. However, occasionally some previously unidentified problems are found. For example, the phenomenon of teratogenicity (the ability of substances to damage foetuses in utero) was unknown until thalidomide (Figure 17.1) was used as a pain‐killing drug. Toxicologists in the industry therefore keep up to date with developments in all related fields to ensure that the fragrance industry does not have any nasty surprises. It is also important to increase our understanding of the biochemistry of toxins, mutagens, allergens, phototoxins, teratogens, and other such unwanted effects because, by doing so, we can improve design of novel fragrance ingredi­ ents by ensuring that they do not elicit such unwanted reactions. O

O

O

N O

H N O O

O

O

O

Bergaptene

Thalidomide

Estragole

OH O Linalool

Citral

O

O Eugenol

Figure 17.1  Safety and natural chemicals.

Dihydroeugenol

­What Is Sustainability

One point worth mentioning is the difference in approach to product safety between flavours and fragrances. Flavours are treated legally as part of the food industry, and historical use plays a large part in determining ingredient safety. Fragrances come under chemicals and cosmetics regulations, and these are based on toxicological testing of pure chemical substances. We will take estragole as an example. In the flavour industry, estragole (Figure 17.1) is used in the form of various herbs such as tarragon (Artemisia dracunculus) and fennel (Foeniculum vulgare). Such herbs have been used as food flavours for millennia without any problem, and so their use is considered safe. However, when estragole is tested as a pure chemical, it is found to be a suspect mutagen, and so it is no longer used in fragrances. Is it really too dangerous to put on the skin yet safe enough to be eaten? Certainly, means of administration of substances can have an effect on adverse reactions, but this is not always the case. Perhaps in the future, a more universal system for estimation of risk will be developed spanning both flavours and fragrances. Chapters 12 and 15 mention the issue of labelling of perfume ingredients on the packaging of the final consumer product such as shower gels or cosmetic creams. The requirement for labelling does not mean a legal ban on the use of an ingredient, but consumers do not necessarily understand correctly the reasoning behind the label and do tend to avoid products labelled as containing substances identified by a scientific sounding name. Therefore, consumer goods companies sometimes prefer to avoid fragrances containing ingredients that require label­ ling. This also tends to place constraints on the fragrance company and the pal­ ette of ingredients it uses. When thinking of safety testing of fragrance ingredients, many people assume that testing on animals is involved. In the past this was certainly the case, but nowadays use of laboratory animals is questioned on an ethical basis. The industry is therefore moving towards other methods of evaluating safety. The use of the Ames test to determine mutagenicity and therefore give a possible pointer to carcinogenicity has already been mentioned. Probably the greatest issue for fragrance ingredients is that of skin sensitisation. In the past, laboratory animals and human volunteers were used. Some of the leading fragrance companies have jointly developed an in vitro (i.e. use of artificial systems in a laboratory vessel such as a Petri dish) method of testing chemicals for skin sensitisation potential. This is more ethical and much less expensive than use of laboratory animals. The industry is currently working to persuade government regulatory bodies that the new tests are good enough to be used in place of the tests that governments demand at present. Similar approaches to testing for other properties such as biodegradability and aquatic toxicity will doubtlessly be developed as well. Such research certainly requires multidisciplinary teams of chemists and biologists. Computer modelling will also play an increasingly important part in safety testing. Artificial intelligence systems are being developed as well as structure/ activity relationships, and the predictive accuracy of these will increase and, one hopes, will one day be accepted as sufficiently reliable to replace testing and allow more rapid and less expensive screening to help in the addition of new, safer ingredients to the perfumers’ palette.

339

340

17  Towards a Sustainable Future

­Natural Fragrance Ingredients There are two common misconceptions about natural fragrance ingredients. Many people, particularly those outside our industry, consider that renewable materials are sustainable and natural materials are safe. When considering sus­ tainability, we must remember the following two facts, and we will look at some examples of natural fragrance ingredients where each of these facts is of importance: Renewable does not necessarily mean sustainable. Natural does not necessarily mean safe. Linalool (Figure 17.1) can be extracted from rosewood oil (Dalbergia nigra) and citral (Figure 17.1) from Litsea cubeba. Demand for linalool and citral far exceeds the potential of the natural sources to meet the need, and, in any case, nowadays both species are endangered and no longer used for fragrance ingre­ dients. Both linalool and citral are readily available from other sources as shown in Chapter 15, and so this is not an issue. However, in the case of sandalwood, the key odour components are not so easily synthesised. Indian sandalwood was extracted from the tree Santalum album. This tree is a parasite and needs other trees to support it. For this and other reasons, S.  album cannot be grown in sustainable plantations. Sandal­ wood oil was highly sought after as a fragrance ingredient, and so the tree was over harvested to the point where it is now on the endangered species list and the Indian government has banned harvesting of it. Anticipating this develop­ ment, Givaudan came up with an interesting solution. Australian sandalwood, Santalum spicatum, is a tree that is a native of South‐Western Australia and produces a similar oil to that of S. album. Givaudan teamed up with govern­ ment, university researchers, and an NGO, ‘Circle of Wisdom’, to develop sus­ tainable plantations of Australian sandalwood and thus provide a source of the natural oil. This secured a sustainable source of sandalwood oil and provided local employment in plantations controlled by Australian Government, and the royalty paid on the oil is passed 100% to the Aboriginal community fund. Similarly, Givaudan works with locals to help ensure sustainable production of products such as agarwood and vanilla. In the case of tonka beans, sustainable production in the Amazon rainforest helps to deter destruction of the rainfor­ est for agricultural land. The English essential oil producer, R.C. Treatt, also works with local people such as in India where they provided wells to enable sustainable production of essential oils and other crops. The safety issues with bergamot and herbs such as tarragon are mentioned above, and they are not alone. Clove oil and substances produced from it are also restricted on safety grounds, and saffron is no longer used as a fragrance ingredient. One driving force for novel ingredients is the replacement of such natural chemicals by safer alternatives. The various methods used in the search for novel ingredients are discussed in Chapter 15. In the case of safer replace­ ments for natural ingredients, one obvious approach is to determine why the natural chemical produces an adverse reaction and then modify the structure to prevent that from happening. For example, the problem with clove oil is eugenol (Figure  17.1), the major component. The allyl side chain p‐ to the

­Synthetic Fragrance Ingredient

O Carene

Cineole

Figure 17.2  Carene and cineole.

­ henolic group makes the molecule susceptible to autoxidation, and this reac­ p tion results in formation of substances that are suspected skin sensitisers or mutagens or both. Therefore, a simple solution is to hydrogenate the side chain to give dihydroeugenol (Figure 17.1), which is much less susceptible to autoxi­ dation and still retains quite a clove odour character. Essential oils and plant extracts that are rich in one chemical component have potential as feedstocks for related ingredients. Since the oils and extracts are from plants, they are renewable and possibly sustainable if other factors per­ mit.  Chemicals that are readily available from inexpensive oils clearly have an economic advantage. Examples would include carene from certain turpentines and cineole from a wide range of Eucalyptus species (Figure  17.2). However, research to date has found few useful routes from either of these to anything of interest to the fragrance industry. In Chapter 15 we saw how extracts such as the pinenes, longifolene, limonene, and cedrol are used in this way. There could be other natural chemicals that would be available sustainably but not used to date. Chemists and botanists could work to identify candidates and ways to use them. In the section on biotechnology below, I will suggest additional ways in which researchers could seek methods by which feedstocks extracted from sustainable natural sources might be used.

­Synthetic Fragrance Ingredients In Chapter 15, the synthesis of citral is used as an example of the continual striv­ ing to improve production processes, and so that aspect of sustainability need not be discussed separately here. In the next four parts of this chapter, we will consider three other aspects of sustainability of synthetic ingredients and then look at menthol, which shows various routes operating in balance. Synthetic Fragrance Ingredients A: Use of By‐Products Any synthetic route to a fragrance ingredient that depends on a by‐product from another industry can only be used while that by‐product remains available at a competitive price. In Chapter  15 we saw how many terpenoid fragrance ingredients can be produced from the pinenes in turpentine. Gum turpentine obtained by tapping pine trees is more expensive than the petrochemical starting materials used in other routes to the same fragrance ingredients and is therefore not currently competitive. Sulfate turpentine is much less expensive than gum

341

342

17  Towards a Sustainable Future

turpentine because it is a by‐product of paper manufacture. However, produc­ tion of new paper from wood pulp is falling because of increased recycling of paper and use of computers in place of books and paper records. So, use of sulfate  turpentine is competitive at present and renewable but not necessarily sustainable since it depends on the market for new paper. In Chapter 15 we also saw how 2‐phenylethanol is produced as a by‐product from the manufacture of styrene and propylene oxide. This is the cheapest source of 2‐phenylethanol at present, but, should the market for either styrene or pro­ pylene oxide fail, then that route to 2‐phenylethanol would become uncompeti­ tive, and therefore it is inherently unsustainable. Similarly, the current route to the jasmine family, as described in Chapter 15, starts from adipic acid, a basic feedstock for nylon synthesis. Any decline in nylon production would therefore affect the sustainability of that route to jasmine odorants. Synthetic Fragrance Ingredients B: Environmental Impact Chapter 15 covered the environmental impact of chemical processes, but here it is the impact of the products themselves that we will discuss. To put the industry into context, the total volume of fragrance oil produced annually is about 300 000 tonnes, just enough to half‐fill a super tanker, whereas trees are estimated to release 100 000 000 tonnes of isoprene into the atmosphere each year. However, we do try to reduce total environmental load, especially of those products where biodegradation is slow. Perhaps the two most problematic odour classes in this respect are musks and woody odorants. Most of the woody odorants are natural or derived from natural precursors, but this does not nec­ essarily mean that they pass the Organisation for Economic Co‐operation and Development (OECD) ready biodegradability test; indeed a large proportion of woody ingredients do not. Perhaps this is an indication of the sensitivity of the test since trees have been producing prodigious quantities of odorous chemi­ cals since they first appeared on earth, but nature does take care of these even­ tually. There are two main approaches to reducing the environmental impact of fragrance ingredients. We can either develop new molecules that will biode­ grade more rapidly than those used currently, or we can reduce the volume used (environmental load) by developing new ingredients that are more intense and therefore can be used at lower volumes. In order to design ingredients that are readily biodegradable, we need to understand the mechanisms by which organic chemicals are degraded by micro­ organisms in the environment and, particularly, those present in sewage treat­ ment plants. One important mechanism of biodegradation is the β‐oxidation pathway. This is rather like the reverse of the route by which fatty acids are syn­ thesised. The first step of this pathway is to produce a carboxylic acid. This could be by hydrolysis of an ester or by oxidation of an alcohol, aldehyde, or methyl group. The carbon next but one, that is the one at the β‐position relative to the acid, is oxidised to a ketone, and this undergoes a retro‐aldol condensation to give acetate and an acid with two carbons less in the chain. The basic outline of the process is shown in Figure 17.3. Clearly, any substituent in the β‐position will prevent oxidation to the ketone and therefore block that route of degradation.

­Synthetic Fragrance Ingredient O R

R CH3

O OH

R

O O

R +

OH H 3C

etc.

OH O OH

Figure 17.3  The β‐oxidation pathway.

By understanding this and other biodegradation routes, the fragrance chemist can seek to design novel ingredients that will biodegrade more rapidly. Any fra­ grance ingredients or fragments of them that escape from the sewage treatment plant before complete degradation will possibly be assimilated by fish or insects such as daphnia. It is therefore also important to profile degradation in waste water systems and investigate the effects of any substances likely to reach aquatic organisms on those organisms. Once again, understanding the metabolic pro­ cesses of these creatures will help in the design of fragrance ingredients that will do no harm in the environment but rather be rapidly and safely degraded. The evolution of musk and sandalwood ingredients will serve to illustrate how reduction in odour threshold enables fragrance companies to reduce the environmental load by using low volumes of ingredients with low threshold to achieve the effect that would require a greater volume of an equivalent with a high threshold. The earliest musk ingredients to be used were derived from animals. Muscone was isolated from the anal glands of the musk deer. It is renewable, has a low odour threshold of 4.5 ng/l, and is readily biodegradable, but it is not sustainable because of animal welfare issues and its very high price. The discovery of nitro musks in the nineteenth century was therefore a welcome development at the time. One of the most successful of these was musk ketone, which was discov­ ered in 1894. Its odour threshold is 0.1 ng/l, considerably lower than that of muscone. However, some of the nitro musk family were found to be phototoxic, their synthesis is hazardous because of the explosive risk in aromatic nitration reactions, and all of them are slow to biodegrade in the environment. Musk ketone was one of the last to be withdrawn from use. In the 1920s, Leopold Ruzicka discovered the high dilution technique for synthesis of macrocyclic compounds. Consequently, musks such as cyclopentadecanolide became avail­ able. Cyclopentadecanolide itself became available in 1927. It has a threshold of 2.1 ng/l, which is lower than that of muscone but higher than that of musk ketone. However, the increase in volume of use would be outweighed by its more ready biodegradability. While the nitro musks were still accepted for use, they possessed an advantage in cost over macrocyclic musks because the high dilu­ tion technique is expensive in process costs. The discovery of the indane and tetralin (polycyclic) musks in the 1950s was another significant breakthrough because they are less expensive to manufacture than macrocyclic musks and have odour thresholds in the same region as those of nitro musks. A good exam­ ple is Galaxolide, launched in 1965 and with a threshold of 0.9 ng/l. However, the polycyclic musks are also relatively slow to biodegrade in water treatment plants,

343

344

17  Towards a Sustainable Future NO2

O

O

O2N Muscone

Musk ketone

O O

O

O

O

Cyclopentadecanolide

Galaxolide

Nirvanolide

OH

OH Radjanol

Javanol

Figure 17.4  Musk and sandalwood evolution.

and so the search for new, more sustainable musks still continues. Nirvanolide was launched in 2001. It has a threshold of 0.1 ng/l and is readily biodegradable. It can be made from readily available starting materials and so meets most of the criteria for sustainability. Lowering of process costs in manufacture will make Nirvanolide and similar musks sustainable options for the future. The synthesis of synthetic sandalwood ingredients from campholenic alde­ hyde was described in Chapter 15. The first member of the family was one now known under various names such as Radjanol. These ingredients are less expen­ sive than natural sandalwood oil, so they became important even before the oil from Indian sandalwood became unavailable. However, they are relatively slow to biodegrade, and so the development of analogues with lower odour thresholds was a target of research. This resulted in the development of Javanol that has a detection threshold 10 times lower than that of Radjanol (Figure 17.4). To achieve a desired level of sandalwood odour, the perfumer can use a fraction of the amount of Javanol that would be needed with Radjanol and therefore the envi­ ronmental load will be lower. Synthetic Fragrance Ingredients C: Biotechnology Chapter 15 includes an account of the preparation of the important amber ingre­ dient Ambrofix from sclareol, which is extracted from clary sage and is therefore renewable. The traditional route to Ambrofix from sclareol goes through sclare­ olide. As discussed in Chapter 15, this oxidation reaction originally used heavy metals and has now been replaced by a more sustainable oxidation based on bio­ technology. However, the reduction of the lactone function and cyclisation to Ambrofix is still a problem in chemical terms since a powerful reductant is

­Synthetic Fragrance Ingredient O OH

O

O

OH H Sclareol

H Sclareolide

H Ambrofix

Figure 17.5  Sclareol to Ambrofix.

required and potential loss of stereochemical integrity (a very important factor for Ambrofix) is a serious issue. Therefore, biotechnology might play a role in improv­ ing sustainability in the future, either by developing a suitable reductive system or by finding a route that does not involve over‐oxidation to sclareolide (Figure 17.5). Figure 15.18 showed how l‐carvone is made from d‐limonene and nootkatone from valencene. These processes are currently carried out using chemical processes that are not without issues and, like the synthesis of Ambrofix described above, need tight control of stereochemistry, so biotechnology could well play a role in improv­ ing the sustainability of both. An enzyme capable of inducing macrolactonisation of linear hydroxy acids to give macrocyclic musks would revolutionise musk synthesis and improve sustainability of that family of fragrance ingredients. In addition to improving the sustainability of known processes such as those described above, biotechnology might well provide totally novel approaches to producing sustainable fragrance ingredients. For example, chemistry has so far failed to provide a route from cineole to useful fragrance ingredients. There must be microorganisms on the floor of eucalyptus forest that degrade cineole. Might it be possible, using gene technology to modify degradative enzymes in these organisms and so interrupt degradation of cineole to carbon dioxide, arresting the process at a useful intermediate? Certainly, there is current research into using microorganisms to synthesise chemical feedstocks for the future using basic starting materials such as sugar. Synthetic Fragrance Ingredients D: Finding the Balance Sustainability came into prominence in the closing decades of the twentieth century. Nowadays it is a factor in the minds of everyone making decisions about the future of businesses. For any activity to be sustainable, it must take into consideration economic, social, and environmental factors, and it is always difficult to decide where the best balance lies between various possible courses of action. l‐Menthol serves as a good example to illustrate the issues involved and how no one solution can provide a complete answer. Usually a compromise must be reached, and the various routes to menthol show how different pro­ duction routes can operate simultaneously, each with advantages in some respects but none with a total clear long‐term advantage over all the others. Variations in technology, politics, weather, etc. can move the position of each route so that, at one time, one route may provide the best overall position but, at another time, a different route might be better. For example, in the case of l‐

345

346

17  Towards a Sustainable Future

menthol, a poor mint harvest would have a deleterious effect on the economic competitiveness of the natural material and thus shift the balance towards the synthetic routes. Originally l‐menthol was extracted from mint plants such as peppermint, Mentha piperita, and currently corn mint, Mentha arvensis, is a particularly important source. China was once the major producer of l‐menthol from M. arvensis, but it has now been overtaken by India. The oil is extracted from the plant by distillation, and the l‐menthol in it is separated from the other oil components by crystallisation. Chapter 2 explained that each asymmetric centre in a molecule gives rise to two different stereoisomers. Inspection of the structure of l‐menthol will quickly reveal that three of its carbon atoms carry four different substituents; in other words, the molecule contains three asymmetric centres. This means that there are 23 = 8 possible stereoisomers. Of these, l‐menthol is the one with the most powerful physiological cooling effect. When l‐menthol comes in con­ tact with the skin or mucus membranes, it produces an illusion of cold. This effect is sought after for cosmetic and flavour applications, and this is the rea­ son why l‐menthol is so important commercially. Any synthesis of menthol must therefore give l‐menthol as the product if it is to be commercially success­ ful, and thus stereochemistry is a key factor in all l‐menthol production routes. Introduction of homochirality (the presence of only one stereoisomeric form) into a chemical synthesis is not easy, and so natural feedstocks where only one chiral form is produced by enzymes are, in principle, attractive. Table 17.1 shows some homochiral feedstocks that have been used as starting materials for preparation of l‐menthol. However, it is clear from the table that some of these synthesis routes are rather long and in other cases, the feedstock is either too expensive or not available in sufficient quantity to meet demand. Consequently, some of these are only of academic interest, and none have gone beyond niche markets. At present, there are three synthetic routes to l‐menthol that operate in com­ petition to the natural product extracted from mint. We will now look at each of these in turn, in historical order of introduction to the market, and then compare all four main commercial routes in terms of sustainability. Table 17.1  Menthol from other natural feedstocks. Chemical

Oil

Steps

Carene

Indian turpentine

7

Citronellal

Citronella

2

Limonene

Citrus

5

Phellandrenes

Eucalyptus, pine, etc.

5

β‐Pinene

Pine, spruce, fir

3–7

Piperitone

Eucalyptus dives

3

Pulegone

Pennyroyal

2

­Synthetic Fragrance Ingredient Hydrogen/cat. Cat.

OH

OH

OH

m-Cresol Thymol

Mixture of 8 stereoisomers Fractional distillation

O O

R

O

+ O

Esterify + OH

l-Menthol

Stereospecific enzymic hydrolysis

OH

d-Menthol

Fractional crystallisation

l-Menthol OH

Figure 17.6  The Symrise route to l‐menthol.

The Symrise Route

Figure 17.6 shows the route used by Symrise to produce l‐menthol. It starts with m‐cresol and propylene, two readily available petrochemicals. These materials are reacted together in the presence of a suitable catalyst to produce thymol. Hydrogenation of thymol produces a mixture of all eight possible isomers of the cyclohexanol. The pairs of diastereomers (see Chapter 2 for an explanation of isomers) can be separated by fractional distillation. Thus, a racemic mixture of l‐menthol and its enantiomer d‐menthol can be obtained. This mixture has to be resolved, that is, the enantiomers have to be separated from each other. Originally this was done by fractional crystallisation, but modern methods use enzymes that discriminate between the two enantiomers, for example, by hydrolysing only one enantiomer of an ester of the racemate. The seven unwanted isomers can be recycled by adding them to the next batch of thymol, since the isomers are inter­ converted under the hydrogenation conditions. The Takasago Route

The route used by Takasago is shown in Figure  17.7. It starts from β‐pinene, which is pyrolysed to give myrcene. Addition of diethylamine to myrcene cata­ lysed by a strong base (lithium diethylamide, which is generated by adding butyl

347

348

17  Towards a Sustainable Future N Heat

N N Li

β-Pinene Myrcene Ru BINAP

Cat.

l-Isopulegol

O

N

d-Citronellal

Hydrogen/cat.

l-Menthol OH

Figure 17.7  The Takasago route to l‐menthol.

lithium to diethylamine, is shown in the figure) gives geranyl diethyl amine. Treatment of this material with a ruthenium complex moves the double bond from its original position to the one next to the nitrogen, that is, an allyl amine has been converted to an enamine. The catalyst is known as ruthenium BINAP because of the use of a binaphthyl ligand. The brilliant feature of this synthesis is that the catalyst used is a single enantiomer {Ru(S) ‐ BINAP2+ ClO4−}, and because of that the hydrogen atom, which is removed from the starting material, is replaced selectively from one side only with the result that the product consists of a single enantiomer – the enamine of d‐citronellal. This can be hydrolysed to d‐citronellal, and, when this is cyclised using an acid catalyst, the chirality of the one asymmetric carbon imposes a specific chirality on the new asymmetric cen­ tres that are being formed, and the product, after hydrogenation, is l‐menthol. Professor Noyori of Nagoya University in Japan was awarded the Nobel Prize for inventing and developing this catalyst. The BASF Route

Figure 15.14 shows how BASF manufactures citral in just four steps, two of which run in parallel. The route starts from two basic raw materials, isobutene and formaldehyde, and uses only heat and catalysts to achieve the result. In about

­Synthetic Fragrance Ingredient

O S-cat. E-Citral

Hydrogen/cat.

Acid cat. R-cat.

O

O

d-Citronellal

OH

l-Isopulegol

l-Menthol

Z-Citral

Figure 17.8  The BASF route to l‐menthol.

2010, BASF entered the menthol market using a route starting from citral. This route is shown in Figure 17.8. Citral exists as a mixture of two geometric isomers (see Chapter 2 for an expla­ nation of isomers), and a mixture of both is produced by the synthesis shown in Figure 15.14. Since these are geometric isomers rather than stereo­isomers, they have different physical properties, including boiling points. Fractional distilla­ tion can therefore be used to separate them. The ingenious feature of this next part of the synthetic route is the use of homochiral catalysts to selectively hydro­ genate the 2,3‐double bond of citral to give only one stereoisomer of the product. Using one isomer of the catalyst, E‐citral is hydrogenated to d‐citronellal, and use of the opposite enantiomer of the catalyst gives d‐citronellal from Z‐­citral. So, d‐citronellal is produced selectively from each of the citral isomers. The cata­ lyst is a rhodium‐based one. The ‘S‐cat’ shown in Figure 17.8 is Rh(CO)2(S,S)‐ chirophos, and the ‘R‐cat’ of Figure 17.8 is Rh(CO)2(R,R)‐chirophos. Let us now compare these four routes from the point of view of sustainability, at least as far as feedstocks, environment, and energy are concerned. Menthol Sustainability Feedstock  The Symrise and BASF processes currently depend on mineral oil

as the basic raw material. It is a finite resource, and so its price will rise steadily in real terms until the supply is exhausted. The percentage of mineral oil consumption used for fragrance ingredient production is tiny. If alternative sources of energy were to be developed, leading to a lower demand for petroleum as fuel, then the stocks of mineral oil would last much longer for use as chemical feedstocks, but the price would rise because of loss of scale advantages in production costs. Of course, research might lead to development of other sources of the basic feedstocks. For instance, fermentation technology might make them available. In that respect, the feedstocks for the BASF process would probably be easier targets than the m‐cresol needed for the Symrise process.

349

350

17  Towards a Sustainable Future

The Takasago process starts from turpentine, and, since it is a renewable resource, the first thought is that it is more sustainable. However, the turpentine used is crude sulfate turpentine (CST) rather than gum turpentine. The latter would prove far too expensive at present. CST is a by‐product of the paper industry, and so the fragrance industry is at the mercy of technical and economic factors in the paper industry. With the growth of electronic communication technology and recycling of paper, the need for paper production could decline, and that would result in a tightening of supply of CST and consequent increase in cost. Natural l‐menthol is, of course, renewable, but supply will only be able to meet demand as long as sufficient land is available for mint production. At present, India is the largest producer of l‐menthol, but India’s population is rising quickly, so for how long will India be able to devote land to mint instead of food production? Environment  As far as the environment is concerned, the use of mineral oil means removing something from the Earth’s crust that we cannot put back, and this is against one of the principles of sustainability. In effluent terms, the petrochemicals business is much cleaner than generally believed. The amount of waste produced per tonne of petrochemical is very small. So, once the oil is extracted from the earth, the processes taking it to the basic feedstocks of the BASF and Symrise routes to l‐menthol do less damage to the environment than forestry and farming. Gum turpentine can be produced with relatively little damage to the environ­ ment, other than those noted below concerning monoculture, if the turpen­tine is produced in plantations of one species of tree. However, CST requires the fell­ ing of trees and involves the Kraft process to make paper from the timber. The Kraft process has significant environmental issues associated with it in terms of pollution. Cultivation of mint plants and forestry plantations with a limited variety of species has environmental implications associated with species conservation, biodiversity, and the hazards of monoculture. If only one species is grown over a considerable area, then any disease that attacks the plants in that plantation will have disastrous consequences for the area. Monoculture is also disadvan­ tageous for wildlife, such as insects and birds, where a varied habitat is prefer­ able. Mint cannot be grown in the same ground indefinitely without replacing the nutrients that the plants require. This will be discussed again in the next paragraph. Energy  In energy terms, mineral oil requires relatively little energy to extract. Process chemists and engineers will always strive to reduce the energy needed for purification and chemical processing, but obviously some energy is required. In the Symrise l‐menthol process, energy is needed to produce the necessary pressures and temperatures for the alkylation, hydrogenation, and distillation processes, and the recycle loops in the resolution contribute significantly to this since the seven unwanted isomers are recycled through the process time and

­Pro‐fragrance

again. The BASF process fares a little better in energy terms, but energy is still needed for handling gases under pressure and for distillation. Energy is required for the Takasago process, not only for the chemical stages but also for the felling, transport, and processing (i.e. converting logs to pulp) of trees in order to produce CST. Surprisingly, the production of l‐menthol from mint also requires a great deal of energy. In addition to the energy for harvesting and distillation is the energy used in fertiliser production. Mint cannot be grown in the same field year after year unless fertiliser is added to replace the nutrients removed from the soil by the mint plants. Fertilisers require fixed nitrogen (i.e. nitrogen at the oxidation level of ammonia or nitrate), and it is made from atmospheric nitrogen gas using the Haber process. This process is very energy intensive. In the United States, the land used for mint production is scorched using gas burners in order to control weeds. Therefore, using modern agricul­ tural methods, the production of l‐menthol from mint herb is, surprisingly, actu­ ally the highest consumer of energy (in other words, has the largest carbon footprint) of the four approaches. The use of alternative agricultural methods is of questionable sustainability because of the increasing pressure on land use for food production. This pressure will inevitably mean that the most efficient farm­ ing methods possible will have to be used. Summary  It is evident that estimating the relative sustainability of each of the processes currently is very difficult and even more so to estimate the future position. The four routes to l‐menthol exist in economic balance at present and probably will for some time to come. Which one is the most cost efficient at any given time shifts continuously depending on numerous factors. The mint growers and chemical producers will undoubtedly continue to invest in research to improve the cost efficiency and sustainability of their individual processes, especially with respect to those aspects that are the weak points as far as sustainability is concerned. Determination of the ultimate sustainability of each, and hence predicting which is the most secure for long‐term supply, is far from straight­forward. This quandary might prompt the question of why make flavour and fragrance ingredients. The answer is that one of the pillars of sustainability is the social aspect. These materials are important in helping to provide the standard of life to which we have become accustomed or to which we aspire.

­Pro‐fragrances Prodrugs are well known in the pharmaceutical industry. These are substances that are not physiologically active in themselves but are converted into active drugs in the body. In earlier chapters we saw how Schiff ’s bases, compounds formed between an amine and an aldehyde or ketone, can serve to protect an aldehyde and release it during use. The limitation of this is that the only really acceptable amine is methyl anthranilate, and this has a strong odour of its own. Although Schiff ’s bases are more robust chemically than aldehydes, they are still not completely stable to acids, bases, and oxidants. However, the idea of using a

351

352

17  Towards a Sustainable Future O Light

O OH

R

OH

O O

+ O

ROH

O

R

Figure 17.9  Light‐induced release of odorants from a pro‐fragrance.

compound to deliver an odorant in use is an attractive one. One good example of a pro‐fragrance system uses the photochemical isomerisation of cinnamic acids. Ultraviolet light can convert the more stable trans‐form of a cinnamic acid to its less stable cis‐isomer. If an ester of a cinnamic acid is used and the benzene ring of the cinnamate carries an ortho‐hydroxy group, then the cis‐isomer formed in the reaction can undergo an intramolecular transesterification reaction to release the alcohol component of the starting cinnamate ester. The scheme is shown in Figure 17.9. Such a mechanism can be used to prevent a volatile alcohol from being lost by evaporation while a perfumed product is in opaque packaging before use. For example, in a laundry powder, the cardboard or plastic container will protect the starting ester from light, but, when deposited on fabric during washing and then exposed to sunlight when the clothing is worn, the reaction can proceed, and the volatile alcohol will be released. If the starting cinnamate ester carries no other substituent, then the other product of the reaction will be coumarin. Although the odour of coumarin is not so striking as that of methyl anthranilate, this still means that an accord of two odorants is released rather than a single volatile component. However, this is not necessarily a disadvantage, and Givaudan have developed a useful product of this type. Addition of a long alkyl chain into the coumarin system would make the molecule more fibre substantive if that property was required, for instance, for a laundry detergent product. It has been shown that metabolic enzymes in the nasal mucus can carry out chemical conversions on odorants rapidly and certainly within the space of one breath. One known example is the conversion of a woody odorant into one that smells of raspberry. The reaction is effected by an enzyme known as CYP2A13 that belongs to the P450 class of cytochrome enzymes. The starting odorant and the one produced from it in the nose are shown in Figure 17.10. This enzymic chemistry opens up another possibility for pro‐fragrances. For example, it should be possible to develop a pro‐fragrance molecule stable enough to survive a harsh consumer product such as laundry detergent or household cleaner but which is converted by nasal enzymes to an odorant that would not survive the storage and/or use conditions.

Air CYP2A13

HO

Figure 17.10  Metabolism of an odorant in the nose by known as CYP2A13.

­Social and Health Factor

Research into pro‐fragrance systems requires skilled and imaginative chem­ ists, and the fruits of this area of research will, no doubt, prove of value in the future in terms of delivery of delicate odorants. Furthermore, pro‐fragrance ingredients could also be a means of reducing environmental load by eliminating the need to use higher volumes of odorant in order to ensure that sufficient remains to provide the desired effect at the point of use.

­Social and Health Factors As mentioned in the preceding paragraph, sustainability in the fragrance indus­ try is not limited to design, production, and use of fragrance ingredients but also involves numerous social and health and well‐being aspects. This section of the chapter investigates how chemistry might be used to increase sustainability of the industry in those directions also. The discussion is divided into subsec­ tions covering understanding of olfaction, malodour management, and health/ well‐being. Understanding Olfaction The last few decades have seen enormous advances in our understanding of olfaction. In the 1970s, the chemist, Ernst Polak, concluded that smell is a combinatorial sense, and this has now been confirmed by Richard Axel and Linda Buck who also identified the family of genes responsible for the odour receptor proteins and showed conclusively that Polak’s conclusion was correct. Techniques such as those developed by Hiro Matsunami now allow the cloning of the odorant receptor proteins in cell culture and investigation of their responsive ranges. Again, this confirms Polak’s conclusion by showing that some receptors respond to a limited range of odorant molecules, while others are widely tuned and are activated by many odorants with diverse molecular structures. It has also been clearly demonstrated that the receptors are not tuned to odour but to stereoelectronic properties of the odorants. It has also been possible to demonstrate that one odorant might be capable of antagonis­ ing a receptor, that is, preventing it from responding to an odorant that would normally activate it in the absence of the antagonist. Meanwhile neuroscien­ tists have used techniques such as functional magnetic resonance imaging (fMRI) and optogenetics to study how odour signals are processed in the brain. We have established that the patterns of activation at the olfactory epithelium and the olfactory bulb (see Chapter 13 for an overview of the organs involved in smell perception) do contain a degree of information about the chemical structure of the odorant, but this is lost higher in the neuroprocessing. It has also been shown that anosmias are related to individual odorants rather than to an odour class and that anosmia to a single odorant when smelt by itself does not necessarily affect the ability to detect its presence in a fragrance for­ mula. The establishment of a database, known as HORDE, of odour receptor proteins has enabled analysis of the data it contains to show genetic effects on odour perception and to establish that variations between individuals is such

353

354

17  Towards a Sustainable Future

that we each have a unique perception of the universe as far as odour is concerned. However, there is still a great deal to learn. The largest published maps of odor­ ant/receptor response cover only a small fraction of receptors and a tiny fraction of odorants. We need to know much more before we can really begin to under­ stand the relationships between odorant structure and the percept that it elicits. The role of enzymes in the nose and how they affect perception has only just begun to be investigated. Current evidence suggests that odour binding proteins serve only to remove odorant molecules from the nose, but this has not been confirmed. The most complex part of the process of olfaction lies in the neural circuits where signals generated at the olfactory epithelium and processed and converted into the perception that we call odour. This science is still much in its infancy, and there is a vast amount to discover. Future discoveries in these areas will open up new paths for increasing sustain­ ability of the fragrance industry. One current market trend is towards personal­ ised perfumes. This dream might never be achievable, but advances in our understanding could well help a movement in that direction. For example, know­ ing an individual’s genotype might one day enable better choice of ingredients for a perfume tailored for that individual. The complexity of olfaction suggests that a totally personally engineered fragrance formulation is unlikely to be possible, but another way of approaching the issue would be to use computational meth­ ods and artificial intelligence to process the vast amount of data that exists concerning fragrance formulae and consumer perception. Indeed, Symrise have announced that they have developed just such a system. Similarly, analysis of data concerning ingredient use could also help in optimising formulae and even aid in design of new ingredients. Malodour Management Fresh human sweat is odourless, and it is bacterial action that generates the mal­ odorant molecules. Currently we try to disguise these odours by covering them with an odorant, often a more powerful one or one that has been identified by sensory scientists to be more efficient than others at doing so. Progress has been made recently in understanding bacterial biochemistry and identifying the enzymes involved. If volatile inhibitors of those enzymes could be found that are either odourless or have a pleasant odour, then inhibition of the enzymes in vivo would be an effective way to reduce malodour through perfume. Since odorant receptor proteins can be antagonised, discovery of a molecule capable of blocking reception of a malodorant molecule is a possible route to malodour control. A recent patent (WO2018138369) demonstrates that this is feasible. Of course, if the malodorant activates a wide range of receptors, it would be necessary to block them all, and that might have effects on perception of other, desired, odorants. Further research into agonists (activators) and antago­ nists (blockers) of odorant receptor proteins will be necessary to extend efforts in this direction.

­Social and Health Factor

Health and Well‐Being There has been a great deal of interest recently in relationships between perfume/ sense of smell and health and well‐being. The two main strands of thought are diagnosis and treatment. Loss of the sense of small in the older population has been studied extensively. Two diseases of old age are particularly important in this respect, namely, Alzheimer’s and Parkinson’s diseases. Smell loss could be an early symptom, particularly in the case of Parkinson’s disease since the olfactory bulb is one of the first brain areas to be attacked by it. Different forms of Parkinson’s affect odour perception differently, and chemists might well find a role in developing diagnostic tools for these and other forms of dementia. It is known that dogs can be trained to detect some types of cancer in patients. They do this through smell and presumably are either detecting specific marker chemicals or a specific pattern of normal components in human body odours. Either way there could be an opportunity for fragrance chemists to develop analytical tools to do the same. It is important to maintain a good sense of smell throughout life and to reduce or prevent loss in ageing. One obvious example is the ability of smell to give warning of dangers such as spoiled food or leaking gas. Much of what we call flavour is actually smell since there are only five basic tastes. Smell is therefore important in maintaining a healthy diet by encouraging consumption of a variety of nutritious food. This is particularly important in care of the elderly and those undergoing certain medical treatments such as cancer chemotherapy that distort the sense of smell. Better understanding of these issues will lead to improved treatment and the possibility of opportunities for fragrance technology. Modern life is stressful and related mental health issues such as depression are increasing. It has been known from time immemorial that perfume can play a role in combatting this. The writer of the Book of Proverbs knew it over 2000 years ago when he wrote ‘Perfume and incense bring joy to the heart’, and in the sixteenth century Michael de Montaigne wrote that ‘Physicians might, I believe, make greater use of scents than they do, for I have often noticed that they cause changes in me, and act on my spirits according to their qualities’. Modern research has confirmed such views and found that perfume promotes positive attitude, relaxation, alertness, and happiness, aids sleep, and relieves stress, depression, irritation, and apathy. On the other hand, loss of the sense of smell leads to depression and lowering of BMI. Not all fragrances and chemical components thereof have the same effect on stress and depression. For instance, it has been found that muguet and tuberose oils relieve depression, an effect that could be due either to physiological or psy­ chological processes. One particularly interesting and well‐researched finding is that frankincense relieves stress and induces a feeling of calm. Incensole acetate has been identified as the component responsible for this, and it has been shown to act on the TRPV3 receptor. This is the receptor that menthol activates in the skin and mucus membranes to give a perception of cold. However, it is also found in neurons of the brain in an area responsible for stress and anxiety. The action of incensole acetate is to block the stress signal and induce calm. Research into

355

356

17  Towards a Sustainable Future

O

O

O

Figure 17.11  Incensole acetate.

the effects of fragrances and their ingredients could well lead to treatments for mood disorders (Figure 17.11). Another feature of modern life is that the market is moving increasingly quickly. This, coupled with the idea of personalised fragrances, will give rise to the need for faster creation and maturation of perfumes. Once again chemists will be needed for the development of such technology.

­Information The importance of information technology was outlined in Chapter 16, and all that needs to be added is that, as the amount of information grows, so does the need to develop effective tools for analysing it.

­Conclusion The fragrance industry is highly competitive, and for any company to survive and grow, in other words to have a sustainable future, it needs to develop all of the following factors: ●● ●● ●● ●● ●● ●● ●● ●●

Better fragrance performance in products. Improved safety performance of fragrances. Improved environmental performance of fragrances. More efficient fragrance ingredient design. More efficient fragrance ingredient selection. Faster fragrance creation. More effective fragrance creation. Better total odour management.

Research is vital in all of these endeavours. The path from basic research to application will need interdisciplinary teams using their skills, opportunity, and serendipity and taking good notice of external factors in the industry and outside it. As noted above, Ernest Beaux’s basic assertion is still as true as in the 1920s, but the activities of fragrance chemists in the twenty‐first century will be more in terms of working in interdisciplinary teams with other scientists and with those involved in the wider issues of the business and modern life.

357

Answers to Review Questions ­Chapter 1 1) A mixture can be separated into its constituents by purely physical means, whereas the component elements of a chemical compound can only be released by chemical transformation. 2) A chemical element is composed of identical atoms, whereas a compound contains elements of two or more elements bonded together. 3) The atomic number of an element is the number of protons in its nucleus. 4) The atomic weight of an element is the sum of the number of protons and the number of neutrons in its nucleus. (Many elements have different isotopes, that is, nuclei in which the number of protons is the same but the number of neutrons varies. In such cases the atomic weight is usually expressed as the average atomic weight of a representative distribution of the isotopes.) 5) The valency of an element is determined by the number of unpaired electrons in its outer valence shell. 6) Caesium iodide is an ionic compound and therefore will be much more ­soluble in water than in liquid paraffin.

­Chapter 2 1) There are three and they have the following structures.

2) There are 19 and their structures are shown below. Taken with question 1, this shows how, as the number of carbon atoms in a molecule increases, the number of possible isomers rises very much faster. Symmetry means that there are not as many as some might expect. For instance, 1‐hexene is identical to 5‐hexene. The double bond adds E‐ and Z‐isomers, and in this case there is also one example of an asymmetric carbon and hence two enantiomers for one of the structures. Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

358

Answers to Review Questions

3) The chiral centres are indicated by arrows in the structures below. Citronellene (also known as dihydromyrcene) is prepared from a-pinene by hydrogenation and then pyrolysis. It is an important precusor for the synthesis of fragrance ingredients such as dihydromyrcenol. It contains one chiral carbon atom.

Limonene is the major component of citrus oils. It contains one chiral carbon atom

Hydrogenation of limonene produces p-menthane and in doing so, creates a second chiral centre in the molecule.

At first sight, it might seem that a-pinene has two asymmetric carbon atoms but they are connected together in a rigid bridge structure and so serve as a single asymmetric centre.

Answers to Review Questions

4)

a) 1,2‐Dimethylbenzene is better known as ortho‐xylene, a product from the pyrolysis of wood.

b) 2‐Methylbuta‐1,3‐diene is better known as isoprene. It is the basic structure behind the family of terpenoids. This is the most important family of natural products as far as the fragrance industry is concerned.

c) Hydrogenation of 2,2,4‐trimethylpent‐4‐ene produces iso‐octane, the standard for determination of fuel performance in internal combustion engines. The olefin is used as a starting material for production of some fragrance ingredients.

d) (E)‐2‐Methylene‐6,10,10‐trimethylbicyclo[7.2.0]undec‐5‐ene is better known as caryophyllene, the main hydrocarbon found in clove oil. Numbers have been added to the structure to help show how the name is derived. The bridgehead atoms are numbered 1 and 9. The longest bridge contains seven atoms and is numbered first, starting from the bridgehead and working towards the olefinic group ensuring that the latter is given the smallest possible number. Numbering then continues around the next largest ring. 8

7

9 11

6

5

10

4

1 2

3

e) Tricyclo[5.2.1.02,6]deca‐3,8‐diene is known as dicyclopentadiene. It is an important precursor for the family of fragrance ingredients known as cyclenes. To draw this structure from the name, it is best to start by drawing two dots to represent the bridgehead carbon atoms. The two longest bridges (i.e. the five and two carbon bridges) should then be added, and then the third bridge (one carbon) can be put in place. Next, the skeleton should be numbered, starting at a bridgehead and going round the largest bridge first. 10 7 8

9

5

6

4

1 2

3

359

360

Answers to Review Questions

Numbering continues across the second bridgehead and round the second largest bridge back to the first bridgehead and then onto the third bridge. Having done that, it is now easy to put the olefins and the fourth bridge into place.

­Chapter 3 1) The functional groups in the fragrance ingredients are as follows: ●●

●●

●● ●● ●●

●● ●● ●● ●● ●● ●● ●● ●● ●● ●●

●●

●● ●● ●●

Aldehyde C18® is a classical example of a misnomer. It is not an aldehyde but a lactone. In the days before spectroscopic analysis, deliberate misnomers of this type were common as a method of trying to maintain a degree of secrecy about the chemical composition of fragrance ingredients. Modern analytical methods make such tactics futile. Aurantion® contains an aromatic ring, an ester, an imine, and a tertiary alcohol. Boronal® is an aldehyde with two olefinic groups. Buccoxime® is an oxime. Exaltolide is a lactone, and, because of the size of the ring, it is referred to as a macrocyclic lactone. Frescile® is a nitrile. Gardamide® contains an aromatic ring and an amide group. Hedione® contains both an ester group and a ketone. Jasmacyclene® contains both ester and olefin functions. Jasmatone® is a ketone. Mayol® is a primary alcohol. Mefrosol® is a primary alcohol and it also contains an aromatic ring. Nerolin Bromelia® is an ether with two fused aromatic rings. Ortholate® is an ester. Precyclemone B® has a slightly misleading name. The –one ending hints at it being a ketone, but it is in fact an aldehyde, which also contains two double bonds. Rose oxide is an ether. Since the ether function is in a saturated six‐membered ring, it would also be described as a tetrahydropyran. Sandela® is a secondary alcohol. Traseolide® contains an aromatic ring and a ketone group. Vanillin, in addition to the aromatic ring, contains three functional groups: an aldehyde, an ether, and a phenol.

­Chapter 4 1) Leaving a perfume in a hot environment will cause the temperature of the perfume to rise. Perfume ingredients are not always totally unreactive, and chemical reactions can occur between them. By heating the perfume, any such reactions will be accelerated and therefore cause unnecessary damage to the perfume.

Answers to Review Questions

2) New York City is at sea level, whereas Mexico City is at high altitude. Consequently, the atmospheric pressure in Mexico City is lower than that in New York City. This will result in a reduction of the boiling point of the solvent in Mexico City relative to that in New York City, which means that a lower temperature will be required near Mexico City. 3) We actually perspire at the same rate at the same temperature. We appear to perspire more in humid conditions because the air in this case is closer to saturation with water vapour making it more difficult for the water in the perspiration to evaporate. 4) You would make up a sample of your perfume in his product and store it at 50 °C for the two weeks. Since the rate of any reaction causing incompatibility would approximately double for each 10 °C rise in temperature, the rate at 50 °C would be approximately eight times the rate at 20 °C (2 × 2 × 2 = 8). Therefore in two weeks at 50 °C, you will have approximately replicated the effect of 16 weeks at 20 °C.

­Chapter 5 1) In an aircraft hold, the temperature drops very low when at high altitude. If a perfume contains a solid ingredient, for example, coumarin, it is possible that its solubility limit will be exceeded at the low temperature causing it to crystallise out of the perfume. Theoretically, it should redissolve when warmed up again. However, in practice this is often a slow process, and crystals will be detectable in the oil. 2) Column chromatography would be a good technique. The dye will show up easily on a white stationary phase in a glass column making it easy to know which fractions to collect. Columns can be quite large enabling a large sample to be processed and therefore giving enough material to make analysis possible. 3) Nuclear magnetic resonance (NMR) is likely to be the most informative technique for this purpose. It will identify all of the carbon and hydrogen atoms in the molecule and give information about how they are attached to each other. Mass spectroscopy will (possibly) give the molecular weight and some information about fragments of the structure. Infrared (IR) will identify functional groups present, and ultraviolet will give some information about double bonds, if sufficient are present and conjugated to each other. Obviously, it is best to combine all of the techniques if possible. 4) Gas chromatography (GC) would be the best technique to use for this problem. Comparison with standard traces of lavender and rosemary oils would very quickly indicate which was present in the perfume. Since lavender oil is a complex mixture of components with a broad range of boiling points, these will probably distil right across the boiling range of all the other components in the perfume, making separation by distillation impossible. Such compounding mistakes cannot be corrected, and a batch of perfume will be lost unless the resultant composition can be used in a different application.

361

362

Answers to Review Questions

­Chapter 6 1) Density is usually a very accurate measurement, and linalool is a pure single substance, so this discrepancy should lead you to carry out another test. For example, the density of citronellol is 0.855 g/cc, and this is much closer to your measured value. The odours of linalool and citronellol are not entirely dissimilar, both being floral with a hint of citrus, and so it is possible that the drum has been mislabelled. 2) There are several techniques that could be used. A simple acid/base titration would be the cheapest, and it would enable you to quantify the amount of acid present. GC would also show up the presence of the acid and, by use of reference standards, would enable you to quantify the level of acid. IR and NMR would show the presence of acid by the appearance of spectral lines characteristic of the acid, but quantification would be more difficult with either of these techniques. 3) NMR would be the technique of choice since it will probably give you enough information to determine the structure without recourse to other methods. 4) Gas chromatography–mass spectrometer (GC‐MS) is the obvious first port of call in cases such as this. You know which odorous components you put into the fragrance, so, by comparing the GC trace of your fragrance against the GC of either the headspace of the soap bar or an extract from the soap bar, you can identify any volatile materials that did not originate from the fragrance. The MS will then, we hope, give you enough information to identify the offending materials. If the mass spectrum does not give a clear identification of the problem component, you will have to use preparative GC to separate enough of it to enable a structure determination by NMR.

­Chapter 7 1) Leaving a drum of perfume in the sun will mean that the contents will become warm. Perfume components are not inert, and chemistry between them is possible. Heating the perfume will provide energy and thus could help overcome the energy of activation and consequently enable reactions to proceed, which would not do so if the perfume were kept cold. 2) All three materials are esters and therefore could be hydrolysed. However, in the case of ethyl acetoacetate, the two hydrogen atoms attached to the carbon atom between the ketone and ester functions are relatively acidic because of the electron‐withdrawing properties of the two carbon–oxygen double bonds. The enolate anion is therefore easily formed and can undergo reactions with many other fragrance ingredients (and, of course, other substances present in the customer’s product.) 3) Although the components of a perfume are well below their boiling points when left at room temperature, entropy still comes into play. The molecules will evaporate because they will be more disordered in the gas phase

Answers to Review Questions

than they were in the liquid. Heat is required for this process (to overcome the latent heat of vaporisation), but it will be taken from the environment. 4) Here we are seeing the interplay of enthalpy and entropy. In order to get the reaction to proceed, we must overcome the energy of activation, hence the need to heat the starting materials. Enthalpy drives the reaction towards the formation of the adducts (or products). However, entropy favours the starting materials because two molecules (diene + dienophile) are a less ordered state than the single molecule of the product where all of the atoms are in fixed positions relative to each other. In the Gibbs equation, the entropy term is multiplied by the temperature. Therefore, as the temperature increases, the entropy term exerts a greater effect. In other words, at lower temperatures, enthalpy wins (the product is formed), whereas, at higher temperatures, entropy wins and the product breaks down to regenerate starting materials.

­Chapter 8 1) Laundry powder is alkaline with a pH of 10–11. Though it is a powder, it does contain some water, and so reactions that occur in aqueous base will take place, even when the powder is in a sealed box before use. Once in use, of course, there is plenty of water around, and the rate of loss of base‐sensitive ingredients will accelerate. Tetrahydrolinalool and acetaldehyde ethyl phenylethyl acetal will both be stable at this pH as they contain no functional groups that would be deprotonated by aqueous base and no functions that could be hydrolysed by it. Citral is an aldehyde and will therefore undergo aldol‐type reactions with itself or any other carbonyl group present in the product. Benzyl acetate is an ester and one which hydrolyses readily. Much of it will be hydrolysed in a product at high pH. 2) The main risk to alcohols in acidic products is that of elimination of water. The most stable of this group will be Mefrosol since protonation of its alcohol function would give a relatively unstable primary carbocation. Citronellol would be the next most stable. As with Mefrosol, its primary alcohol function will not dehydrate readily. However, it contains a double bond that could be hydrated giving an odourless diol. Overall, therefore, it will be slightly less stable than Mefrosol. The other two alcohols will be much less stable. Geraniol has an allylic alcohol function, that is, an alcohol with a double bond next to it. This double bond will stabilise the positive charge generated by protonation of the oxygen atom and therefore make elimination easier. Dihydromyrcenol has a tertiary alcohol, and so protonation would lead, after loss of water, to a relatively favoured tertiary carbocation. This will make dihydromyrcenol prone to elimination under acidic conditions. 3) Evolution of a vinegar smell suggests the formation of acetic acid. Since the product contains aqueous acid, it would be reasonable to think that it might cause hydrolysis of esters in the perfume. Putting these two pieces of information together, the first thing to do would be to check the perfume formula to see if it contains any acetate esters.

363

364

Answers to Review Questions

4) A vanilla smell and a brown colouration in soap suggests that vanillin in the perfume is undergoing aldol‐type reactions catalysed by the alkali present in the soap. The first thing to do would be to check the formulation for vanillin or related compounds such as ethylvanillin.

­Chapter 9 1) Rossitol® will be the more stable in hypochlorite bleach. It is a tertiary alcohol and is therefore not susceptible to oxidation. Mefrosol, on the other hand, could be oxidised to the corresponding acid by hypochlorite. 2) Lilial® could be oxidised to the corresponding acid by peroxy bleaches. 3) Chromic acid is a very strong acid and will dehydrate tertiary alcohols to olefins. The olefins are then oxidised by the chromic acid. 4) Tonalid® is a methyl ketone and hence will be susceptible to the chloroform reaction. Cyclopentadecanolide is an ester and can therefore be hydrolysed by the strongly basic conditions of a hypochlorite bleach. Galaxolide® would therefore stand the best chance of these three musks in a product containing hypochlorite bleach.

­Chapter 10 1) The idea of a fabric conditioner fragrance is to leave a fresh/clean odour on the cloth. Therefore base notes would be more appropriate as they will survive the rinse and drying cycles and remain on the cloth rather than evaporating off during the laundry cycle. 2) Feminine fragrances tend to have more floral notes than those intended for use by men. 3) If the ingredient has a low threshold, then probably less of it will be needed to produce the desired effect in a fragrance. This has advantages, in terms of difficulty of analysis and copying, cost advantages (especially if the ingredient is an expensive one), and environmental advantages in that the environmental impact will be less.

­Chapter 11 1) The first thing to do in this case is to look at your perfume formula and see if there is any indole or vanillin in it or indeed any accords, essential oils, and so on, which are likely to contain indole or vanillin. If you find any, then reformulate the fragrance without them, and test it in the customer’s soap base. 2) Colour formation between shower gel components and fragrances is not common. Some other factor is more likely to be the cause. The best thing in this case

Answers to Review Questions

is to open a discussion with the customer and try to establish as much as he is prepared to tell you about his base product and then work from there, preferably together with the customer, to try to establish what might be going on. 3) Linalool is a tertiary allylic alcohol and so will be prone to acid‐catalysed dehydration. Hydroxycitronellol is a tertiary alcohol and so will be prone to acid‐catalysed dehydration. If it dehydrates, the resultant product is prone to acid‐catalysed cyclisation. Its aldehyde function is also prone to other acid‐ catalysed reactions. Only Mefrosol is likely to survive the low pH of an antiperspirant. 4) The aldehyde group of citronellal can be oxidised by the bleaches in the powder and can also undergo aldol reactions catalysed by the strong bases present. Terpinyl acetate is an ester and is therefore prone to hydrolysis and other reactions catalysed by the base present. Frescile is a nitrile and is therefore relatively resistant to the basic and oxidising conditions as is the ether Anther®. These latter two compounds can be attacked, e.g. by oxidation of the benzene ring in Anther and by formation of the N‐oxide of Frescile, but the extent of reaction will probably be low enough to allow the bulk of the material to survive and provide sufficient odour to give the desired effect.

­Chapter 12 1) Fructosylfructose is a disaccharide based on two fructose units and is therefore a carbohydrate. Tripalmitin is a triglyceride containing three palmitic acid fragments esterified to glycerol and is therefore a lipid. Phenylalanylleucylthreonylalanine is a tetrapeptide composed of four amino acids and is therefore a protein. 2) Heptylidenecyclopentanone is an α,β‐unsaturated ketone and therefore a Michael acceptor. It is thus more likely to be a skin sensitizer. Consequently, suppliers of heptylcyclopentanone will ensure the absence of heptylidenecyclopentanone from their product. 3) Both α‐ionone and α‐damascone are α,β‐unsaturated ketones and are therefore potential sensitizers. However, Michael addition occurs at the β‐carbon atom of the α,β‐unsaturated ketone function, and, in the case of α‐ionone, this carbon atom is adjacent to the cyclohexene ring, and the ring carbons on either side of the point of attachment of the side chain both carry methyl groups. Indeed, one of them carries two methyl groups. This cluttering of the space around the potentially reactive carbon atom makes it difficult for reactive species to approach it, a phenomenon known to chemists as steric hindrance. The net effect is to reduce the Michael reactivity of α‐ionone relative to that of α‐damascone, and this, in turn, reduces its sensitization potential. The permitted levels of α‐ionone in products are therefore higher than those for α‐damascone. In practice this sensitization is not a problem since the intense odour strength of α‐damascone means that perfumers never use it at levels that would be a problem.

365

366

Answers to Review Questions

4) In all four cases, biodegradation is likely to start with oxidation of the alcohol group. With the exception of compound c, this reaction will result in formation of the corresponding carboxylic acids. The acid formed from compound a will easily undergo β‐oxidation to a point where the readily biodegradable fragment R is reached. The molecule will therefore be readily biodegradable overall. In the case of compound b, the β‐position is quaternary, meaning the carbon atom is bonded to four other carbon atoms. Oxidation is therefore not possible using the β‐oxidation pathway. This fact will slow down its biodegradation considerably, and, unless another route is open to the bacterial enzymes, the substance is likely to fail the ready biodegradability test. Compound c has two routes open to it. If oxidation first happens at the alcohol function, the initial product will be a ketone. The steps of the β‐oxidation pathway will then proceed but instead of releasing a molecule of acetic acid, the fragmentation will release a more highly oxygenated species and so, overall, we would expect molecule c to be readily biodegradable using this route. Alternatively, the terminal carbon atom could undergo ω‐oxidation to give an α‐hydroxy acid. This substance will then biodegrade along the β‐oxidation pathway. Molecule d has an oxygen atom in the β‐position and so falls into the same category as molecule b since this oxygen atom has no hydrogens attached and therefore none that can be removed by the bacterial enzymes in the first step of β‐oxidation.

­Chapter 13 1) When we have a cold, the mucus layer becomes thicker. This has the effect of reducing the airflow through the nose and therefore reducing the number of odorant molecules that pass into the nose. The increased thickness of the aqueous mucus also makes it more difficult for odorants (which are not very hydrophilic) to pass through it to reach the olfactory epithelium. 2) Most of what we normally refer to as taste is actually retronasal (i.e. approaching the nasal cavity from behind, in other words from the throat) smell and so, loss of the sense of smell as explained above, also causes loss of ‘taste’. 3) One odorant can affect the ability of the receptor proteins to recognise a second odorant. Thus one might block recognition of the other. Thus the signal pattern generated at the olfactory epithelium could be different from the simple sum of the patterns of the two individual odorants. Furthermore, even if just a simple sum of the two individual patterns, the pattern will be different from either of these and will be interpreted in the brain as a third pattern.

Answers to Review Questions

4) At the most basic level, the variation between individuals in terms of which 350–400 olfactory receptors is such that no two people (other than identical twins) are likely to use the same set of receptor types. Neural networks in the brain will also differ subtly from one individual to another. Furthermore, experience is also important in decoding the initial signal pattern, and, since everyone has a different experience of the world, each will have a unique set of odour references and processing systems based on them.

­Chapter 14 1) In the following figure the compounds of the question are shown just above representations of them in which some bonds have been removed or drawn as dotted lines, in order to illustrate the biogenetic origin: ●●

●●

●●

●●

●●

●●

Methyl everninate, which occurs in oakmoss and treemoss, is a polyketide. The lower drawing shows the original chain of four acetate units with carbonyl groups on every alternate carbon atom. Three of these units are still clearly evident in the final structure; the fourth is slightly disguised as it is the site of an aldol condensation in which the oxygen was eliminated as water. Deodarone is a component of Himalayan cedarwood oil. It is a sesquiterpenoid, and the lower drawing shows the three isoprene units as entities with solid lines. The dotted lines show the bonds joining the isoprene units together. Dehydrocostus lactone occurs, as its name suggests, in costus oil. It is also a sesquiterpenoid, and the lower figure is drawn in the same way as for deodarone. Acorone is found in calamus and is another sesquiterpenoid. Once again, the lower figure is drawn in the same way as for deodarone. Asarone is a shikimate as is evidenced by the three carbon chain attached to a benzene ring with hydroxylation sites in positions 3 and 4 relative to the side chain. It has an additional oxygenation site at position 6. This might lead one to suspect that it could be a polyketide, but the methoxy groups next to each other at positions 3 and 4 assure us that this is not the case. Theaspirane occurs in the oils of osmanthus and boronia. It is a carotenoid degradation product and hence a terpenoid. There are only 13 carbon atoms that might be misleading, but that is because of the position of cleavage of the original chain. Structural features such as the geminal dimethyl group and the close resemblance to the ionones make it clear that it is a terpenoid derivative. (Geminal refers to the fact that both methyl groups are attached to the same carbon atom.)

367

368

Answers to Review Questions

O O

O OH

O

O

O

Ethyl everninate O

O O Dehydro costus lactone

Deodarone O

OH O

OH

O

HO O

O O O O

O Acorone

O

β-asarone

Theaspirane

OH

O

OH

O

HO OH

2) None of these ingredients are of natural origin. If we consider all of the biosynthetic routs we have studied, there is no means in nature for introduction of a tertiary butyl group. This rules out inonyl acetate and Lilial as potential natural products. Nature does make macrocyclic ketones and lactones, but these come from the polyketide pathway, and it is not possible to insert an ether link into a polyketide chain using nature’s chemistry. Therefore, Cervolide is not a natural structure. The bridged ring system of verdyl acetate does resemble those of some terpenoids, but terpenoids will always have pendant methyl groups that verdyl acetate does not. O

O

O

O Inonyl acetate

Verdyl acetate O

O Cervolide

O

O LiIial

Answers to Review Questions

­Chapter 15 1) Turpentine is available from coniferous trees, which therefore, in principle, represents a renewable resource. Crude sulphate turpentine (CST) is available as a by‐product of paper manufacture, and so it is available at relatively low cost. The major components are the pinenes, and these allow easy access to structures, which already contain many of the features associated with fragrance ingredients, thus making subsequent synthetic routes easier. 2) Saffron oil is extracted from the stamens of a crocus species (Crocus sativus), and so harvesting is very expensive. The double bonds and aldehyde functions of safranal make it sensitive to acidic, basic, and oxidising media, and so its stability in functional products will be poor. Furthermore, safranal is an α,β,γ,δ‐unsaturated aldehyde; therefore it is a potential skin sensitizer or mutagen, and so there might also be safety reasons for seeking a replacement. 3) The main reasons why the syntheses shown in Figure 15.45 would not make good production routes are as follows. The route to either product contains six steps and so will mean high process costs (both operator time and reactor occupation). A number of the reagents used have handling hazards: t‐butyl hypochlorite (strong oxidant), methyl lithium (violent reaction with water), and chromium trioxide (strong oxidising agent). The effluent will be significant: bromide waste in the first step, t‐butanol waste in the second, chloride waste in the third lithium, and chromium waste in the jasmone route and carbon dioxide in the methyl jasmonate route. Both starting materials and many of the reagents are expensive: ●●

As far as analogues are concerned, the obvious thing to try is to use a saturated side chain since the presence of the cis‐double bond in the side chain complicates the syntheses. This approach allows the much shorter routes shown in Figure 15.35 to be used. These routes also employ reagents that are safer and easier to use and avoid most of the waste described above. By removing the methyl substituent and double bond from the ring of jasmone, it is possible to use readily available and inexpensive cyclopentanone as the starting material for synthesis of jasmone analogues such as Jasmatone (again, see Figure 15.35).

369

371

Glossary Absolute  the alcoholic extraction of a concrete. Accord  a pleasing and harmonious combination of perfume notes. Alicyclic  a molecule containing rings that are not aromatic rings. Aliphatic  a molecule with an open chain structure either linear or branched. Allelochemicals  substances that carry messages between individuals of different species. Analyte  a substance being subjected to chemical analysis. Anion  a negatively charged ion. Aromatic  a molecule containing a ring in which electrons are delocalised. Asymmetric carbon atom  one that carries four different substituents. Atom  the smallest unit of an element that retains the chemical properties of that element. Atomic number  the number of protons in the nucleus of an atom, which is characteristic of an element. Atomic weight  the weight of an atom relative to that of a hydrogen atom. It equals the sum of the number of protons and the number of neutrons in its nucleus. ATP  adenosine triphosphate, a cofactor for a number of enzymic reactions. Azeotrope  a mixture of two or more liquids that has a single boiling point and from which the component liquids cannot be separated by distillation. Base note  the least volatile part of a perfume. Biodegradability  the rate at which a material will be degraded in the environment. Biogenesis  the process by which a plant or animal manufactures a chemical. Biosynthesis  the process by which a plant or animal manufactures a chemical. Bloom  the ability of a perfume or perfume ingredient to fill space when delivered from a product matrix. Bond  a link or force between atoms that hold them together at a set distance. Carbohydrates  chemical substances composed of carbon, hydrogen, and oxygen, with the molecular ratio of hydrogen to oxygen in the empirical formula being 2 : 1. Carcinogens  materials that cause changes in cell chemistry that could become cancerous. Cation  a positively charged ion. Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

372

Glossary

Centre of asymmetry  a point in a molecule that imposes chirality on the structure, e.g. a carbon atom carrying four different groups. Chiral  an object that cannot be superimposed on its mirror image. In the case of carbon atoms, one that carries four different substituents. Chiral centre  see centre of asymmetry. Chord  a pleasing and harmonious combination of perfume notes. Chromatography  separation of components of a mixture using differences in the affinity of the components for two phases, one of which remains stationary and the other moves across it. Cis‐  two groups lying on the same side of a ring or double bond. Coenzyme A (CoA)  a cofactor for a number of enzymic reactions. Cofactors  small molecules that play a vital role as reagents in enzymic reactions. Colloid  a stable suspension of solid particles in a liquid. Compound  a chemical substance made up of two or more elements, in which the constituent elements cannot be separated by simple physical processes and the chemical and physical properties of which are different from those of the constituents. Concrete  material extracted from a plant using a hydrophobic solvent. Condensation  the process of changing from a gas to a liquid. Conformational isomers  isomers differing in the angle of rotation around the bonds between atoms in the molecule. Conformer  a conformational isomer. Critical pressure  the pressure at which all three phases of a substance exist simultaneously. Critical temperature  the temperature at which all three phases of a substance exist simultaneously. Crystallisation  the formation of a pure crystalline solid either by cooling a pure sample of the liquid form of the substance to its melting point or by cooling or concentrating a solution of the substance. Detection threshold  the lowest concentration at which an odorant can be detected by smell. Detergent  a substance that reduces the surface tension between two immiscible liquids or between a liquid and a solid surface. Deterpenation  removal of monoterpene hydrocarbons from an essential oil. Diastereomers  molecules that differ in having a different configuration at more than one chiral centre. Discord  a combination of perfume notes that produces a jarring effect. Distillation  purification of a liquid by converting it from the liquid phase to the vapour phase and then condensing it back to the liquid phase. Ecotoxin  a substance that causes damage to the environment when it is released into it. Electron  a negatively charged particle that is found in the outer portion of atoms and molecules and that takes part in chemical bonding. It has no mass. Electrostatic attraction  the attractive force between two particles with opposite electrical charges. Eluent  the material flowing out of a chromatography column.

Glossary

Empirical formula  a chemical formula that shows the relative proportions of elements in a compound. Emulsion  a mixture of two immiscible liquids in which one is dispersed as droplets in the other. Enantiomer  a molecule that cannot be superimposed on its mirror image. End note  the least volatile part of a perfume. Enfleurage  process of extracting hydrophobic components from a plant by contact with a bed of fat. Enzymes  globular proteins that serve as nature’s catalysts. Essential oil  the volatile oil obtained from plant material. Evaporation  the process of changing from a liquid to a gas. Expressed  an oil (usually citrus) that has been obtained from plant material by squeezing the plant material without heating it. Eye irritant  a substance that causes irritation when it comes into contact with the eye. Fats  hydrophobic substances composed of fatty acid esters of glycerol. Folded oil  an essential oil from which the monoterpenoid hydrocarbons have been removed. Fractional distillation  separation of two or more liquids by distillation through a high efficiency column so that components with different boiling points are collected separately. Freezing  the process of changing from a liquid to a solid. Fusion  the process of changing from a solid to a liquid. Geometrical isomers  molecules differing by the geometry across rings or double bonds. Glycolysis  the splitting of a sugar molecule to produce smaller building blocks for synthesis or as a step in burning to produce energy. Haptens  small molecules that are recognised as foreign substances by the immune system. Head note  the most volatile part of a perfume. Heart note  the central part of a perfume that is the part with medium volatility. Hormones  substances that carry messages from one part of an organism to another. Hydrocarbon  a compound containing only atoms of carbon and hydrogen Hydrogen bonding  the attractive force between a positively polarised hydrogen atom and a negatively polarised atom (usually oxygen, nitrogen, or sulfur) or a double bond or benzene ring. Hydrophilic  water loving. Hydrophobic  water hating. Hydrophobic bonding  the attraction between hydrophobic molecules in an aqueous environment resulting from a reduction in the total surface area exposed to water. Impact  the size of the impression that an odour makes on the person perceiving it. Intensity  the size of the impression that an odour makes on the person perceiving it.

373

374

Glossary

Internal standard  a reference material added to a sample to be analysed so that, by knowing the relative sensitivity of the test method to both the standard and the component being measured, an accurate quantification of the test substance can be determined. Ion  an atom or molecule that carries an electrical charge as the result of gain or loss of valence electrons. Isomers  molecules with the same empirical formula but differing in their molecular structures. Isotopes  atoms with the same number of protons but different numbers of neutrons in their nuclei. Thus, isotopes will have the same atomic number (hence the same chemical properties) but different atomic weights. Latent heat of fusion  the heat required to change a material from being a solid at its melting point to a liquid at the melting point. Latent heat of vaporisation  the heat required to change a material from being a liquid at its boiling point to a gas at the boiling point. LD50  the dose required to kill 50% of animals, usually expressed in terms of milligram of material per kilogram body weight of an animal of a given species. Lecithins  substances based on glycerol in which two of the alcohol groups are esterified with fatty acids and the third with a radical containing a phosphate group. Lipids  hydrophobic substances comprising glycerol esterified with hydrophobic radicals. Lipocalins  proteins that engulf small molecules, usually as part of a mechanism for removal of these from the organism, sometimes as part of a process of recognition. Lipophilic  oil loving. Lipophobic  oil hating. Melting  the process of changing from a solid to a liquid. Meta‐  a 1,3‐relationship of substituents on a benzene ring. Metal  a chemical substance that is malleable, ductile, and a good conductor of heat and electricity. Micelle  an emulsion in which the droplets are stabilised, e.g. by the presence of a detergent. Middle note  the central part of a perfume, that is, the part with medium volatility. Mixture  a combination of two or more chemical substances, the constituents of which can be separated by simple physical processes and in which the constituents retain their individual chemical and physical properties. Molecular recognition  the fitting together of two molecules or parts of molecules, in which weak non‐bonded interactions cooperate to form a stronger association. Molecular weight  the weight of a molecule relative to that of a hydrogen atom. Molecule  the smallest unit of a chemical compound that retains the chemical properties of that compound. Mutagen  a substance capable of modifying the genetic make‐up of an organism.

Glossary

NADP (H)  nicotinamide adenine dinucleotide phosphate, a cofactor involved in red‐ox reactions involving hydride transfer. Neutron  a neutral species found in the nucleus of an atom and having a mass of 1 Da. Non‐bonded interactions  attractive forces between molecules (or parts of molecules) that do not involve formal chemical bonds. Note  an individual odour character. Nucleic acids  chemical substances composed of purine and/or pyrimidine bases (nucleotides) that have been bonded to saccharide molecules (forming nucleosides) and then connected together into chains by means of phosphate ester links. The basis of genetic information. Nucleus  the positively charged core of an atom. Orbital  the volume of space occupied by an electron of an atom or molecule. Ortho‐  a 1,2‐relationship of substituents on a benzene ring. Para‐  a 1,4‐relationship of substituents on a benzene ring. Peptides  polymers made of α‐amino acids linked together through amide (peptide) bonds. Persistence  the ability of a perfume or perfume ingredient to last over time. Pheromones  substances that carry messages between individuals of the same species producing an inate response. Phospholipids  substances based on glycerol in which two of the alcohol groups are esterified with fatty acids and the third with a radical containing a phosphate group. Phototoxin  a substance that produces skin damage when exposed to sunlight. Polyketides  substances that are biosynthesised by linear coupling of acetate units. Pomade  the mixture of fat and extract resulting from the process of enfleurage. Positional isomers  molecules differing by the points of attachment of substituents, side chains, rings, or double bonds. Primary structure  the sequence of amino acids in a peptide chain. Proteins  polymers made of α‐amino acids linked together through amide (peptide) bonds. Proton  a positively charged species found in the nucleus of an atom and having a mass of 1 Da. Racemate  a mixture of equal parts of opposite enantiomers. Racemic mixture  see racemate. Racemic modification  see racemate. Radiance  the ability of a perfume or perfume ingredient to fill space. Receptor proteins  proteins whose role is to recognise specific substances and alert the cellular chemistry to the existence of these in cell’s environment. Recognition threshold  the lowest concentration at which an odorant can be identified by smell. Reduction  the removal of oxygen (or an equivalent) from a compound, addition of hydrogen to it or gain of an electron. Resolution  separation of a single enantiomer from a racemate. Rf  retention factor. The extent to which a substance moves on a chromatography plate or column relative to the movement of the solvent front in the same time.

375

376

Glossary

Rt  retention time. The time taken for a material to pass through a gas chromatography column. (The column length, diameter, coating material or packing, identity of carrier gas, speed of carrier gas flow, and oven temperature must all be defined when quoting Rt values since all of these parameters will affect Rt.) Saccharides  chemical substances composed of carbon, hydrogen, and oxygen, with the molecular ratio of hydrogen to oxygen in the empirical formula being 2 : 1. Saturated  a molecule containing as much hydrogen as can be added without breaking its basic carbon structure – in other words, a molecule containing no double or triple bonds. Saturated solution  a solution in which the liquid is incapable of dissolving any more of the dissolved substance in question. Secondary structure  the way in which peptide chains fold into sheets or helices in a given protein structure. Semiochemicals  substances that carry messages from one organism to another. Shikimates  substances that are biosynthesised using shikimic acid as an intermediate. Skin irritants  substances that irritate the skin on contact. Skin sensitizers  substances that are recognised by the immune system and cause an allergic reaction on repeated exposure. Solidification  the process of changing from either a gas or a liquid to a solid. Solute  a material that is dissolved in a liquid. Solution  a liquid containing a dissolved solid. Solvent  a liquid containing a dissolved solid, liquid, or gas that is not identical to the liquid itself. Spectrometry  the gaining of information about the molecular structure of a material by examining its ability to absorb electromagnetic radiation of various wavelengths (infrared, ultraviolet/visible, or radio frequency). (Mass spectrometry is a misnomer.) Spectroscopy  the gaining of information about the molecular structure of a material by examining its ability to absorb electromagnetic radiation of various wavelengths (infrared, ultraviolet/visible, or radio frequency). (Mass spectroscopy is a misnomer.) Steam distillation  distillation of material with concomitant distillation of water. Stereoisomers  molecules differing by the relative orientation in space of groups around an asymmetric centre. Structural formula  a drawing showing the relative positions of atoms in a molecule. Structural isomers  molecules differing by the points of attachment of substituents, side chains, rings, or double bonds or by different ring sizes, junctions, fusions, etc. Sublimation  converting a solid to vapour phase and then returning it to the solid phase. Used as a method of purifying a volatile solid from mixtures containing other solids that are not volatile.

Glossary

Super‐saturated solution  an unstable state in which a liquid contains more of another substance than it can normally support. Surfactant  a substance that reduces the surface tension between two immiscible liquids or between a liquid and a solid surface. Systematic name  a name that uses a defined set of rules (especially the IUPAC rules) to connect that specific name with a specific molecular structure. Tenacity  can mean persistence, but also can mean the ability of a perfume or perfume ingredient to stick to a surface such as cloth or hair. Terpeneless oil  an essential oil from which the monoterpenoid hydrocarbons have been removed. Terpenoids  substances that are biosynthesised using isopentenyl pyrophosphate and prenyl pyrophosphate as intermediates and whose structure is composed of isoprene units. Tertiary structure  the way in which the peptide chains, helices, and sheets of a given protein arrange themselves in space. Tincture  a solution/suspension prepared by extraction of a natural material with ethanol. Top note  the most volatile part of a perfume. Trail  the ability of a perfume to occupy the space through which a wearer of it has passed. Trans‐  two groups lying on opposite sides of a ring or double bond. Trivial name  a common name given to a compound (often based on the source from which it was first isolated) that does not necessarily give any information about its molecular structure. Valency  the combining power of an atom or group, in terms of hydrogen units. van der Waals attraction  the long‐range attractive forces between molecules that are not bonded to each other. Vaporisation  the process of changing from liquid to gas. Visualisation  process enabling colourless components of a mixture to be seen, after they have been separated by thin layer chromatography.

377

379

Bibliography Books Anonis, D.P. (1993). Flower Oils and Floral Compounds in Perfumery. Carol Stream, IL: Allured Publishing Corp. Arctander, S. (1960). Perfume and Flavour Materials of Natural Origin. Elizabeth, NJ: S Arctander. Arctander, S. (1969). Perfume and Flavour Chemicals, vol. 2. Elizabeth, NJ: S Arctander. Blakemore, C. and Jennett, S. (eds.) (2001). The Oxford Companion to the Body. Oxford University Press. Calkin, R.R. and Jellinek, J.S. (1994). Perfumery, Practice and Principles. Wiley. Curtis, T. and Williams, D.G. (2001). Introduction to Perfumery, 2e. Ellis Horwood. Edwards, M. (1996). Perfume Legends. HM Editions. Edwards, M. (1998). Perfume Legends: French Feminine Fragrances. La Quinta, CA, Crescent House Publishing. Edwards, M. (2004). Fragrances of the World. Carol Stream, IL: Allured Publishing Corp. Grob, R.L. and Barry, E.F. (eds.) (2004). Modern Practice of Gas Chromatography, 4e. Wiley. Jellinek, P. and Jellinek, J.S. (1997). The Psychological Basis of Perfumery, 4e. Blackie Academic and Professional. Kaiser, R. (2006). Meaningful Scents around the World. Wiley‐VCH. Mann, J., Davidson, R.S., Hobbs, J.B. et al. (1994). Natural Products: Their Chemistry and Biological Significance. Longman. Müller, P.M. and Lamparsky, D. (eds.) (1991). Perfumes, Art, Science and Technology. Elsevier. Ohloff, G. (1994). Scent and Fragrances. Springer. Sell, C.S. (2003). A Fragrant Introduction to Terpenoid Chemistry. Cambridge: Royal Society of Chemistry. Sell, C.S. (ed.) (2006). The Chemistry of Fragrances from Perfumer to Consumer, 2e. Cambridge: Royal Society of Chemistry. Sell, C.S. (2007). Terpenoids. In: Kirk‐Othmer Encyclopedia of Chemical Technology, vol. 22. Wiley. Sell, C.S. (2009). Olfaction. In: The Wiley Encyclopedia of Chemical Biology, vol. 3 (ed. T.P. Begley), 500–509. Wiley. Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

380

Bibliography

Sell, C.S. (2014). Chemistry and the Sense of Smell. Hoboken, NJ: Wiley. ISBN: 978‐0‐470‐55130‐1. Surburg, H. and Panten, J. (2006). Common Fragrance and Flavour Materials, 5e. Wiley‐VCH. van Toller, S. and Dodd, G.H. (eds.) (1988). Perfumery, the Psychology and Biology of Fragrance. Chapman and Hall. Williams, D.G. (1996). Chemistry of Essential Oils. Micelle Press. Williams, D.G. (2004). Perfumes of Yesterday. Weymouth: Micelle Press. Wilson, D.A. and Stevenson, R.J. (2006). Learning to Smell. The Johns Hopkins University Press.

Journal Articles Axel, R. (1995). The molecular logic of smell. Sci. Am. 273 (4): 154–159. Fráter, G., Bajgrowicz, J., and Kraft, P. (1998). Fragrance chemistry. Tetrahedron 54 (27): 7633–7703. Kraft, P., Bajgrowicz, J., Denis, C., and Fráter, G. (2001). Odds and trends: recent developments in the chemistry of odorants. Angew. Chem. Int. Ed. 39 (17): 2981–3010. Malnic, B., Hirono, J., Sato, T., and Buck, L.B. (1999). Combinatorial receptor codes for odours. Cell 96: 713. Rossiter, K.J. (1996). Structure–odour relationships. Chem. Rev. 96 (8): 3201–3240. Sell, C.S. (2006). On the unpredictability of odour. Angew. Chem. Int. Ed. 45 (38): 6254–6261.

381

Index a absolute alcohol  90 absolute zero of temperature  133, 135 absorption spectrum  113, 116, 117, 119 abstracts 331–332 acetal formation  159, 160 acetals  50, 158 formation of  159 acetonitrile 56 acetylenes  26, 166 acetylide anion  167 acid anhydrides  50–51 acid catalyzed addition to cyclopropanes  164 to olefins  163–164 acid chloride  50–51, 59, 63, 148, 167, 168 acids 111 and bases perfume ingredients stability  152 pH 150–152 strong and weak  149–150 in water  150 in consumer goods anhydrous aluminium chloride 199 calcium salts  198 limescale removal/prevention  198 pH control agents  198 zirconium chloride  199 activated aluminium chlorohydrate (AACH) 199

activated zirconium aluminium glycine (AZAG) complex  199 acyclic monoterpenoids  273 acylium ions  149 adenosine triphosphate (ATP)  249, 250, 258, 259 adipic acid  291, 313, 342 adulteration, of essential oils  336 air fresheners  192, 194, 209, 212 alcohols  40–41, 43 dehydration 162–163 aldehydes  45, 112, 158–159, 177–178, 182, 183 aromatic 45 perfumery 46 aldol condensation  155–158, 263, 288, 292, 297, 309, 313, 314 aldol reaction  155–158, 209, 236, 263, 304, 365 alicyclic aldehydes  45 alicyclic materials  32 alicyclic musks  297, 317 aliphatic fragrance ingredients, from ethylene 303 alkanes  18, 20–23, 125 alkenes  22–25, 31 alkyl ethoxylates (AEs)  77, 208 alkyl peroxy radical  182 alkynes 26 allelochemicals 257 allyl isothiocyanate  234 α,β‐unsaturated carbonyls  165 α,β‐unsaturated ketone  156, 157 α‐carbon 142

Fundamentals of Fragrance Chemistry, First Edition. Charles S. Sell. © 2019 Wiley-VCH Verlag GmbH & Co. KGaA. Published 2019 by Wiley-VCH Verlag GmbH & Co. KGaA

382

Index

α‐carotene 277 α‐helix protein structure  227 α‐hydroxy amine  54 α‐pinene  28–29, 86 α‐terpineol  43, 164 aluminium chloride  148, 149, 167, 199, 317 ambergris  94, 189, 278, 301 amboryl acetate  300 ambrofix  278, 301, 344, 345 Ames test  234, 339 amide formation  56 amide/peptide 55–56 amines  53, 79, 160, 201, 208, 280 amygdalin 220 sources of  220 structure of  220 amylcinnamic aldehyde (ACA)  157, 288 analysis chemical methods acids 111 aldehyde 112 base 111 COD 112 ester value  111–112 ketones 112 peroxide content  111 phenols 112 titration 110 water 112 physical methods boiling point  108 colour 109 density 108 flashpoint 109 melting point  108 optical rotation  109 refractive index  109 viscosity 109 spectroscopic methods  113–114 eugenol 127–131 GC‐MS 127 IR 118–120 NMR 120–124 quality control  131–132 UV 114–118

androstenone 281 aniline  149, 150, 208 Animusk 319 anionic surfacants  77 anions stabilisation, by hyperconjucation 143 anisaldehyde  45, 46, 267, 311 anisole (methoxybenzene)  169 antibacterial agents, in consumer goods 207–208 antimicrobial sesquiterpenoid  258 antioxidants  180, 184–185 antiperspirants  148, 152, 199 application patents  330 arachidonic acid  264 Arens–van Dorp synthesis, of citral 293–294 aromatic aldehydes  45 aromaticity  32, 62 aromatic molecules, electrophilic substitutions 168–170 Arquad and Hamburg Ester Quat (HEQ) 78 asymmetric carbon atoms  33, 34, 219 atom efficiency  292–295 atomic absorption  113 atomic number  4–8, 25, 35, 37, 61 atomic structure  7–8 atomic theory  1–7 atomic weight  4–8, 292 atoms 4 electronic structure  9–11 Aufbau Principle  10 autoxidation  171, 180 of fats  184 in perfume containing an aldehyde 182 of pinane  296 of 4‐tetrahydronapthalene  1–3, 181 azeotropes, formation of  90 azo‐dyes 206

b Baeyer–Villiger reaction  177 Barbier–Bouveault–Tiemann synthesis, of citral  292 bases 111

Index

in consumer goods  199–200 BASF route, to l‐menthol production 349 benzene  31–33, 43, 62, 69–71, 90, 95, 162, 163, 169, 181, 201, 217, 305, 310 benzoic acid  49, 307, 308 benzophenone  46, 118 benzylic oxidation, of toluene 307–308 bergaptene  234, 261, 338 beryllium fluoride (BeF2) 11 beryllium oxide (BeO)  11 β‐oxidation pathway  342–343 bilayers 82–84 biodegradability  211, 235, 237–239, 317, 339, 342, 343 biological oxygen demand (BOD)  112 biosynthesis of carotenoids  276 enzymes and cofactors  258–261 of fatty acids  223 of jasminic acid derivatives  264 polyketide 262–263 principles of  258–261 of secondary metabolites  261–262 Biradex N  68–69 crystal packing  70 one molecule of  69 structure of  69 bloom  193–194, 320 boiling point  21, 22, 29, 36, 72, 85, 86, 88–91, 100, 101, 108, 138, 216, 234, 239, 324, 349 borneol  103, 290 boron trifluoride  148, 167 Bourgeonal  45 brassylic acid  49, 288 Brønsted acids  147, 148, 167, 168, 301 Brønsted base  150 burette 110 “burning” of TNT vs. burning of coal 140 butene  23, 24 butenolides 50 butylated hydroxyanisole (BHA)  185, 203

butylated hydroxytoluene (BHT)  185, 203

c calcium carbonate (CaCO3)  136, 198, 199 calcium hydrogen carbonate/ Ca(HCO3)2 136 calcium hydroxide  148, 198 canonical structures  32, 162, 164 carazolol 248 carbanions  141, 142, 164, 180 carbocations  141, 162 formation 143 stabilisation, by conjucation  143 carbohydrates  78, 82, 218–221 carbon  3, 8, 10–12, 14, 15, 17–66, 69, 119, 121, 124, 126, 130, 132, 134, 140–142, 153, 155, 160, 162, 164–166, 177, 178, 180, 182, 191, 223, 236, 246, 288 carbon dating  8, 132 carbonic acid (H2CO3)  136, 151 carbon‐magnesium bond  166 carbon orbitals  15 carbon‐oxygen bond  40, 142, 144, 159, 272 carbonyl carbon  45–47, 54, 142, 153, 154, 156, 160, 164, 178, 182, 200, 259 carbonyl compounds  51, 62, 155, 156, 158, 164, 166, 203 carboxaldehyde 45 carboxylic acid chlorides  59 carboxylic acids  39, 47–50, 55, 59, 77, 94, 95, 137, 154, 169, 178, 199, 205, 225, 236, 247, 259, 288, 303, 313–315, 318, 342, 366 carboxylic ester  49, 61, 77 carene 341 carotenoids  269, 277–279 Carroll reaction, in citral synthesis  294 caryophyllene  28, 29, 103, 269 catalysts  5, 60, 113, 140, 167, 199, 213, 225, 228, 258, 295, 348, 349 cationic surfactants  78, 79 C–C double bond formation  23

383

384

Index

C–C σ bond  23 C–C triple bond formation  26 ceramide‐derived lipid  83 chalcones 68 Chanel No 5  45, 188–190, 192 character, odour  320 chelating agents, in consumer goods 205–206 chemical abstracts (CAS) numbers 332 chemical bonding  10, 12–16, 215 chemical information  333 chemical issues, patents  330 chemical oxygen demand (COD)  112, 178 chemical reactions  73, 112, 114, 133, 135–137, 141, 197, 228, 257, 301 chemical shift  121–124 chemoreception  215, 243 chemotaxonomy 262 chirality  33, 109, 217, 299, 300, 346, 348 chiral molecules  33 chlorine oxidation states of  174 oxoacids, structure of  175 chloroform reaction  122, 128, 178, 179, 201 chlorophyll 95 C6H5NH2 149 CHO (carbon, hydrogen, oxygen) compounds 39 chromatography 95 column 99 gas 100–104 HPLC 100 mobile phase  96 paper 98 principles of  96, 97 stationary phase  96 thin layer  98–99 types 96 visualisation technique  98 chromium, red‐ox cycle of  171–172 cineole 341 cinnamic acid  49

cis‐2‐butene 23 cis‐6‐dodecene 24 5‐cis‐undecatriene 24 citation index  332 citral  291, 340, 349 Arens–van Dorp concept  293–294 Barbier–Bouveault–Tiemann synthesis route  292 Carroll reaction  293–294 Claisen rearrangement  295 from isobutylene  296 citric acid  205 citronellol  43, 191, 231, 273, 290–292, 298 citronellonitrile 57 citronellyl nitrile  57 citrus oils  24, 34, 127, 274, 284, 285, 298, 299 Claisen rearrangement, in citral synthesis 295 clog P values  75 13 C NMR spectra  124 C=O bond  45 cocamido propyl betaine (CAPB)  79 coenzyme‐A (CoA)  259, 260 cofactors 258–261 colloids 84 column chromatography  96, 99, 100 common salt  5, 13, 67 concretes  2, 95 conformational isomers  17–19 conformers 19 constant boiling mixture  90 consumer goods acids in  198–199 antibacterial agents in  207–208 bases in  199–200 chelating agents in  205–206 cosmetics and toiletries  210 fine fragrance  209 household products  212–214 laundry powders  211–212 malodours in  279–281 nucleophiles in  200–201 personal wash category  210–211 photo‐active agents in  206–207 reactive ingredients in  208

Index

reductants in  202–204 surfactants in  204–205 coordination number, of a metal ion 205 13 C or carbon 13  8, 120, 123, 126, 130, 132 14 C or carbon 14  8, 132, 261 cosmic radiation  132 coupling constant  123 covalent bonds  12 formation of b/w two hydrogen atoms  14 in methane  15 critical point  72 critical pressure  72 critical temperature  72, 95 crude sulfate turpentine (CST)  296, 350 crystallisation  74, 85, 93–94, 248, 346, 347 cubic crystal lattice  67 cyanide  56, 57, 230 cyclic monoterpenoids  103, 273, 274 cyclobutane  27, 144 cyclohexane conformation  27 cyclopentadecanolide  318, 343 cyclopentadiene 315 cyclopentane  26, 27, 315 cyclopropane  27, 164 Cytochrome P450  5, 245, 247

d damascones  277, 278 Darzens reaction  61 Dean and Stark apparatus  90 decanoic acid  79 decanonitrile 57 defence chemicals  257 denaturing 228 density  41, 84, 108, 109, 239 deodorancy 321 deterpenation  92, 268 deuterium  8, 120 deuterochloroform (CDCl3)  122, 128 dextrorotatory (d‐)  34, 35, 274 D‐glyceraldehyde see dextrorotatory (D‐)

1,3‐diaminopropane 79 diamond crystal structure  70 dicyclopentadiene (DCPD)  315 Diels–Alder reaction  144, 145, 298, 305, 315 diethyl ether  44, 165 digitalis 231 dihydroeugenol 341 dihydromyrcenol  43, 296, 297 dihydroxyacetone (DHA) reaction, with perfume ingredients  208 dimethylbenzene 32 dimethyl sulfoxide (DMSO)  59 diols  41, 51, 159 1,3‐dioxan 61 1,3‐dioxolane 61 distillate  2, 85–88 distortionless enhancement by polarisation transfer (DEPT) 123 disulfides  59, 203, 204, 226–228 d‐limonene 34

e E‐and Z‐geometric isomers  35 eclipsed conformation  18 ecotoxicity  232, 235 Efetaal  51 electromagnetic radiation  107, 113, 206 electromagnetic spectrum  113, 114 2‐electron bonds  18 electronic structure atoms 9–11 transition metals  11 electrophiles  147, 152–154, 167–169, 233, 234 electrostatic forces  67, 215 elements  2–11, 17, 39, 111, 116, 165, 223 empyreumatic distillation  90 emulsions  79–81, 84, 204, 205 enamines  54–55, 160–161, 200, 201, 348 enantiomeric molecules  33 endothermic reactions  139 end point of the titration  110

385

386

Index

energy  9, 11, 32, 73, 91, 95, 113, 114, 116–120, 123, 128, 133–135, 138–140, 206, 207, 219, 236, 350–351 energy of activation  138–140, 181 enfleurage  2, 92, 94 enolate anion  142, 155, 156, 262, 263 enthalpy  134, 135, 138 entropy  134–135, 137, 139, 149, 230, 318 environmental impact, of fragrance ingredients 342–344 enzymes  140, 199, 211, 213, 228, 230, 245, 247, 258–261, 345–347, 352, 354 epoxide  61, 180, 297, 317 Eschenmoser fragmentation  318 essential oils  36, 44, 59, 90–92, 95, 100–102, 104, 108, 109, 111, 112, 127, 131–133, 160, 187, 190, 209, 233, 261, 268, 273–275, 283, 291, 301, 310, 322, 341 ester hydrolysis  137, 154–155, 162, 258 esterification  75, 94, 137, 138, 154–155, 162, 300, 311, 318 esters carboxylic 49 in perfumery  50 phosphate 49 sulfate 49 value 111–112 ethane  17–19, 177, 302 1,2‐ethanediol 52 ethanol  2, 41, 49, 73, 74, 90, 94, 95, 111, 121–123, 125, 126, 158, 194, 209, 295 ethanolamine  199, 200 ethanol‐insoluble materials  94 ethers  44–45, 58, 180, 183 ethoxylation 77 ethyl acetate formation  49, 90 ethyl alcohol  41, 93, 305 ethylene brassylate  49, 288, 289, 318 ethylenediaminetetra‐acetic acid (EDTA) 205 ethylene epoxide  61

ethylene glycol distearate (EGDS)  78 ethylene oxide  61, 77, 306 ethyne 26 eugenol 340 carbon DEPT spectrum  130 1 H NMR spectrum  128 infrared spectrum  128 mass spectrum  131 proton NMR spectrum  129 spectroscopic technique  127 ultraviolet spectrum  129 exothermic reaction  139 eye irritants  234

f farnesene 103 fatty acids  223–224 calcium salts of  76 fine fragrance  43, 90, 152, 163, 190, 194, 195, 209, 283, 284 first law of thermodynamics  133 Fixolide  316, 317 flame ionisation detector (FID) 101–102 flashpoint 109 florentine 91 fluorine  10, 11, 148 fractionating still  88, 89 fragment based method  235 fragrance companies commercial feasibility  337 ethical issue management  335–336 safety testing  237, 238 fragrance industry  29, 39, 59, 81, 93, 94, 96, 107, 113, 172, 175, 230, 232, 234, 235, 238, 243, 279, 283, 285, 286, 291, 302, 305, 307, 311, 313, 325, 329, 330, 335–338, 341, 350 fragrance ingredients  66 from α‐pinene 296–299 from β‐pinene 296–299 categories 287 from cedarwood oil  300–301 from citrus oil components  299 development costs  285 life cycle  285

Index

from longifolene  300 manufacture, economics of 284–287 from petrochemicals  302–320 raw materials for  288 requirements of  320–322 fragrance oils  109, 209, 342 fragrant aldehyde  55 frankincense  284, 355 Franklin acids and bases  149 free energy  135 free radical  22, 141, 176, 181 chain reactions  182 Friedel–Crafts acetylation of anisole 169 Friedel–Crafts acylation reaction  167, 168 Friedel–Crafts alkylation reaction 167 Friedel–Crafts reaction  149, 167–168, 305 fructose representation of  219 sucrose formation from  220, 221 functional groups  40 carbon‐nitrogen single bonds  65 carbon‐oxygen double bonds  63 carbon‐oxygen single bonds  63 divalent sulfur  64 hexavalent sulfur  64 other nitrogen  65 oxygen‐oxygen single bonds  64 tetravalent sulfur  64

g Galaxolide  317, 343 γ‐lactones 50 γ‐undecalactone 50 gas chromatography (GC)  96, 100 chromatogram, of lavender oil  103 detectors 104 GC‐sniffing 102 temperature ramping  101 gas chromatography with mass spectrometry (GC‐MS)  102, 124, 127 gas/liquid chromatography (GLC)  96

geometric isomers  22–25, 312, 349 geraniol  43, 231, 257, 272, 273, 283, 291, 292, 296, 298 geranium‐scented diphenyl ether  44 Gibbs equation  135 Gibbs’ free energy  135 global toggle switch  248, 249 globular proteins  228, 258 glucose aldehyde function of  220 representation of  219 sucrose formation from  220, 221 glycerol  75, 78, 159, 205, 224, 288 glycerol monostearate (GMS)  78, 205 glycidates 61 glycolysis 261 glycosides 220 Google Scholar  333 graphite  17, 70, 71 Grignard reaction  165–167 gum turpentine  296, 341, 350

h halogens  5, 104, 168 haptens  232, 233 HAZOP studies  322 height equivalent per theoretical plate (HETP) 88 Helional  55 Helvetolide  317, 318 hemiacetals 50–51 hemiketals 50–51 Herboxane  51 heteroatoms acid anhydrides  50 acid chloride  50 alcohols 40–43 aldehydes 45–46 esters 49–50 ethers 44–45 hydrogen bonding  39 ketones 46–47 peroxy compounds  52 phenols 43–44 heterocyclic compounds  60–66 hexane  21, 73, 94, 95 hexavalent sulfur  60, 64

387

388

Index

high performance liquid chromatography (HPLC)  96, 100 hormones  257, 264 human lymphocyte test  234 human olfaction, organs used in 244–246 human olfactory receptor database (HORDE) 250 hybridisation, of orbitals  11–12 hydrocarbons 15 alkynes 26 aromatic rings  31–33 cyclic structure  26–28 greek leters  30–31 polycyclic structures  28–29 rules for isomers  37 rules for nomenclature  36–37 stereoisomerism 33–36 stereoisomers 37 hydrodiffusion 91 hydroformylation  302, 303 hydrogen bonding  39–41, 216 in acetic acid  217 between acetic acid molecules  217 between alcohol molecules  216 in methanol  41 between methanol and benzene  217 in water  216 hydrogen chloride  147, 149–151, 174, 298 hydrogen fluoride  149 hydrogen molecule formation  17, 18 hydrophilic  16, 75, 77, 78, 81, 82, 218 hydrophobic bonding 217–218 water‐hating 75 hydroxide 153 hydroxylamine (NH2OH)  112, 161 hydroxyl (‐OH) group  41 hyperconjugation  142, 143 hyphenated technique  108 Hypo‐Lem  57

i i‐butane 20 imines (Schiff ’s bases)  54, 55 incensole acetate  355–356

indole  63, 210, 211, 255, 256, 266, 279, 280, 364 inductive effect  142, 153, 178 inert gas‐like electron shells  12 infrared (IR) spectroscopy  107, 118–120 inherently biodegradable substance 235 intensity, odour  193, 320 International Fragrance Association (IFRA) standard  238, 239 International Union of Pure and Applied Chemists (IUPAC)  24, 42 ion channels  227, 249 ionic bonds  12, 13 ionic compounds  13 ionic liquids  95 ionising radiation  114 ionones  277, 278, 296, 298, 305 ions and radicals formation  142 from methane  141 irones 279 iron, oxidation and reduction of  172 iso‐butane 20 iso‐butene/iso‐butylene 24 isobutylene  24, 144, 296, 303, 304, 311 isolongifolanone 300 isomeric butanes  19 isomeric butenes  23–24 isomeric 6‐dodecenes  24 isomerism  20, 33 isopentenyl pyrophosphate  270, 272 iso‐propyl,‐CH(CH3)2 22 isopropyl myristate (IPM)  78 isotopes 8

k Karanal  159 Karl Fischer method  113 Kelvin scale  135 ketals  50–52, 61, 62, 158–159, 178 ketones  46–47, 54, 112, 128, 142, 159, 169, 177, 178, 180, 183, 199–201, 203, 208, 233, 310, 368 kitchen malodours  279, 280

Index

l

m

lactone  49, 50, 177, 279, 314, 318 laevorotatory (l‐) 34 latent heat of fusion  73 latent heat of vaporisation  73, 91 laundry powders  133, 152, 154, 158, 170, 179, 192, 202, 211, 212, 352 Laureth‐2 77 lavandulyl acetate  103 l‐carvone 274 LD50  231, 232 lecithins  79, 83, 224 Lewis acids  147–149, 167, 300 Lewis base  148 Ligustral   45, 55, 144, 159, 305 limonene 34 enantiomers of  35 linalool  103, 131, 132, 144, 231, 273, 285, 291, 292, 294, 296, 297, 333, 340 linear alkyl‐benzene sulfonates (LAS) 77 lipases  199, 213 lipids 75 fatty acids  223 in living organisms  224 phospholipids  224, 225 sphingolipids  224, 225 triglycerides 224 lipocalins  226, 230, 246 lipophilic (oil‐loving)  75 liquid bleach  109, 201, 213 liquid splitter  89 lithium  4, 10, 11 lithium fluoride (LiF)  11 lithium oxide (Li2O) 11 l‐limonene 34 l‐menthol  274, 319 BASF route  349 from Mentha arvensis  346 structure 346 sustainability 345 Symrise route  347 Takasago route  347–348 l‐menthyl benzoate  94 log P  15, 16, 75, 191, 211, 233, 239, 246, 324 lutensols 78

machine dishwashing powders and gels 213 macrocyclic musks  28, 288, 317–319, 343, 345 magnesium cation  13 magnetic nuclei  120 malodour management  353, 354 malodours  39, 59, 171, 173, 174, 202, 264, 279–281, 322 mammalian cell wall  79, 83, 84, 226 mass spectrometry (MS)  102, 107, 114, 124–125, 127 basic elements  125 ethanol 126 fragmentation 126 ions fragment  125 matter  1–16, 67–84, 134 medium rings  28 Mefrosol  191, 308, 309 melting point  108 Mentha arvensis  346 menthol 35 sustainability energy 350–351 environmental implications  350 feedstocks 349–350 meso‐isomer 36 metallurgy 1 methane  14–15, 18 methanesulfonic acid  60 methane thiol  58 methanol  40, 41, 69, 216, 217, 263, 295, 315 (Z)‐3‐methoxy‐4‐methylhept‐3‐ene 25 methyl anthranilate  39, 54, 55, 114, 160, 161, 266, 351, 352 methyl carbanion  141 methyl dihydrojasmonate  194, 314 methyl ethyl 2‐hydroxyethyl salts  95 2‐methyl‐2‐(1´ethylprop‐1‐yl) propan‐1,3‐diol 159 methylheptenone  166, 293–295 3‐methyl‐2‐hexenoic acid  280 methyl iso‐amyl ketone  47 methyl jasmonate  49, 264, 324, 327 methyl ketones  47, 178, 179, 201, 283

389

390

Index

4(R)‐(+)‐1‐methyl‐4‐(1´‐methylethenyl) cyclohex‐1‐ene 24 2‐methylpropane 20 methyl radical  22, 25, 40, 126, 131, 141 methyl vinyl ketone  233 meting or fusion  73 mevalonic acid  261, 270 micelles  81, 212 Michael acceptors  233, 235 Michael reaction  164–165, 233, 314 molecular orbital  14, 114, 117 molecular recognition chirality 217 defined 215 electrostatic attraction  215 hydrogen bonding  216–217 hydrophobic bonding  217–218 van der Waals attraction  215 monosaccharides 219–221 monoterpenoid acetate  103 monoterpenoid alcohol  103 monoterpenoids  103, 268–270, 272–276 muguet aldehydes  306–307 musk ingredients  297, 343 musk ketone  57, 58, 343 mutagenic materials  234 myrcene  103, 298, 347

n n‐alkanes, boiling points  22 naphthalene 309–310 numbering systems  33 naphthofuran  278, 301, 302 natural chemicals, classes of  218 natural fragrance ingredients  340 estraction 95 naturally sourced ingredients  337 Nazarov reaction  318 n‐butane 20 neural pathways, in olfactory perception 255 neutralisation of sulfuric acid (H2SO4) 136 neutral surfacants  78 Newman projection  18

nicotinamide adenine dinucleotide phosphate (NADP)  229, 258, 260 Nirvanolide 343 nitriles  156, 157, 161–162 see also cyanide nitrobenzene  57, 58 nitro compounds  57 nitrogen detection  104 nitrogenous molecules  39 nitromusks  316, 343 n‐Octanol 75 non‐soap detergents (NSDs)  76, 211 non‐superimposability of asymmetric carbon atoms  34 isomer 33 nootkatone  127, 300, 345 normal butane/n‐butane 20 nor‐patchoulenol 43 novel fragrance ingredient discovery chemist’s role  322–323 copying leads  324 high thoroughput screening  323 odour character  325 performance tests  325 random screening  323 statistical design  324 n‐pentane 20 nuclear magnetic resonance (NMR)  108, 120–124 nucleic acids  218, 221–222 nucleophiles 152–154 in consumer goods  200–201 reaction, with perfume ingredients  200 nucleosides 221 nucleotides  221, 222, 251, 259 numerical prefixes  21, 24, 26

o ocimene 103 octenylsuccinic acid (OSA)  78 odorous molecules  17, 39, 221, 245, 246, 257, 336 odour binding proteins  226, 245–247, 354 odour perception  36, 245, 252–256, 324, 353, 355 odour properties  198, 320, 323, 326

Index

OECD test  235 olefins  23, 24 acid catalyzed addition to  163–164 metathesis 318 olfaction  41, 353 combinatorial nature  249–252 organs used in  244–246 receptor event  247–249 role in biology  243 transport to receptors  246–247 olfactory epithelium  245, 256, 353, 354 olfactory receptor proteins  225, 227, 247–250 oligomerisation  302, 303 oligosaccharides 219 opacifiers 207 optical brighteners  207 optical rotation  109, 336 orbitals hybridisation of  11–12 shape of  9–10 organic compounds  17, 101, 102, 165, 172, 178, 221 organic reactions  140–145 organoleptic analysis  198 organosulfur compounds, oxidation of 173 organs used, in olfaction  244 orsellinic acid  263 3‐oxapentane 44 1,3‐oxathiane 61 oxazolidine 62 oxidation and reduction (red‐ox) reactions 171 chromium 171–172 iron 171–172 oxidative decarbonylation, of aldehyde 183 oximes  54, 55, 112 oxirane 61 oxygenated organics, oxidation levels of 177–178 oxygen bleaches  177, 179

p palm kernel oil (PKO)  76 palm oil (PO)  76, 224

paper chromatography  96, 98 paraffins 18 para‐tertiary‐butylcyclohexanol (PTBCHol) 138 para‐tertiary‐butylcyclohexyl acetate (PTBCHA)  138, 155 patents  286, 322, 324, 326, 329, 330, 332 pearlisers 207 pentane  20, 26 peracetic acid  180, 202 perception of odour  252–256 perfume aldehydes and muguet notes  190 base notes  189–190 bloom 193–194 detection threshold  192 impact 192 ingredients 187 dihydroxyacetone reaction with 208 nucleophiles reaction with  200 middle/heart notes  189–190 molecules  75, 205, 208, 210, 246 muguet and sandalwood notes  190 notes, chords and discords  187 odour families  188 persistence and tenacity  191 physical and chemical factors  194 radiance 193–194 recognition threshold  192 requirement of  187 scent trail  194 stability  198, 210 top notes  188, 190 types of  192 perfumery nitriles  57 pericyclic (or electrocyclic) reactions 144 periodic table of elements  4, 5 Perkin triangle  87 peroxide  52, 59, 111, 176, 182, 233 peroxy compounds  52 peroxy species, in organic chemistry 176 Persil 202 personalised perfumes  354

391

392

Index

personal wash products  211 phase transition  72, 73 phenols  44, 112, 310–312 antiseptic 43 in essential oils  44 phenylacetaldehyde dimethyl acetal (PADMA)  51, 306 phenylethanol  43, 91, 191, 266 2‐phenylethanol  305, 342 pheromones  243, 257 phosphate esters  49, 221 phosphatidylcholine  83, 224 phospholipids  224, 225 phosphoric acid  49, 224, 258 photo‐active agents, in consumer goods 206 photosynthesis 261 phototoxic materials  234 pinane ring system  29 π‐electrons 32 of naphthalene  32 system of vanillin  158 π*‐orbitals 117 plant derived feedstocks, advantage of 287 platinum metals  5 polarising filters  33 polar solvents  13, 15 polycyclic musks (PCMs)  316, 317, 343 polyketide biosynthesis  262–263 polymerisation  257, 302, 303, 318 polypeptides  225, 227 polysaccharides  219, 221 pomade  94, 95 p‐orbitals 10 prenyl pyrophosphate  270 primary metabolites carbohydrates 218–221 defined 218 lipids 223–225 proteins 225–230 principle of chemical equilibrium 137–138 principle of microscopic reversibility 137–138 process patent  330

pro‐drugs 351 pro‐fragrance system  351 proteases/peptidases 199 proteins 226 chemical communication  226 globular 228 lipocalins 226 primary structure  227 secondary structure  227 tertiary structure  228–229 protium 8 protons  7, 8, 10, 12, 13, 128, 129, 141, 163 PTBCHA formation and loss  155 p‐toluenesulfonic acid  60

q quality control (QC)  107, 131–132 quantitative structure‐activity relationships (QSAR)  324 quaternary ammonium salts  53, 78, 79, 258 quats  53, 78

r racemate  36, 333, 347 racemic mixture  36, 94, 347 radiance  193, 194, 320 radioactive isotopes/radio‐isotopes  8, 104, 132, 261 radio frequencies  107, 120 (R)‐and (S)‐nomenclature 36 raw materials, for fragrance ingredients 288 reaction profiles  138–140 readily biodegradable substance  235 receptor event, olfaction  247 receptor odorant screening map  251 reductants in, consumer goods  202–204 reflux  88, 89 refractive index  100, 109, 131 refractometer 109 regular polygons  26, 27 Research Institute for Fragrance Materials (RIFM)  238 resinoid 95 resolution 36

Index

resonance 32 reviews and books  331 RIFM Expert Panel (REXPAN)  239 rose alcohols  273, 291, 296 rule of thumb  138, 197

s saccharides  218–219, 221 sandalwood ingredients  290, 343–344 sandalwood oil  276, 340, 344 Santalum album  340 saponification 75 saturated hydrocarbons  177 saturated solution  74 scented candles  212 scent trail  194 Schiff ’s bases  54–55, 160–161, 178, 200 schizophrenic acid  48, 280 SciFinder  332, 333 sclareolide  301, 302, 344 secondary metabolites  218 biosynthesis of  261–262 second law of thermodynamics  134 semiochemicals 257 sense of smell  2, 131, 215, 243, 244, 250, 254 separation and purification crystallisation 93–94 distillation 85–93 azeotropes, formation of  90 Dean and Stark apparatus  90 fractionating still  89 hydrodiffusion 91 under reduced pressure  87 steam still  91 natural fragrance ingredients estraction 95 sublimation 93 sesquiterpenoid hydrocarbons  103, 200 sesquiterpenoids  103, 268, 275–276 shikimate raw materials, for fragrance ingredient production  288 shikimic acid  261, 265–267 σ*‐orbital 117 silicones  78, 100

skew conformations  18 skin irritation  232, 234, 239 skin sensitisation  232–234, 239 sodium bisulfite (NaHSO3)  200, 202 sodium cation  13, 67 sodium chloride  13, 208 crystal  67, 68 crystal lattice, fragment  14 sodium cumenesulfonate (SCS)  77 sodium hydroxide (NaOH)  75–76, 136 sodium hypochlorite  109, 174, 175, 186, 201 sodium ion  67, 74 sodium lauryl ether sulfate (SLES)  77 sodium nonanoyloxybenzenesulfonate/ percarbonate bleach system 202 sodium or potassium  75, 111 sodium sulfate (Na2SO4) 136 soft electrophiles  153, 233, 234 solid shading  14, 34 solubility  43, 74–75, 85, 91, 93, 96, 164, 234, 239 solute 74 solvent  2, 13, 16, 69, 74 extraction 94 soot/carbon black  69 space‐filling model  15 spectrometers 113 basic components  116 spectroscopic techniques  114, 115, 127–131 spectrum, of red dye  116–117 sphingolipids  224, 225 sphingomyelin  83, 225 sp3 orbitals  11, 12, 14 staggered conformation  18, 28 standard temperature and pressure (STP) 21 states of matter bilayers 82–84 colloids 84 detergency 81–82 emulsions 79–81 gas 71 liquids 71 micelles 81

393

394

Index

states of matter (contd.) simple phase diagram  72 solids crystalline solid  68 cubic crystal lattice  67, 68 diamond crystal structure  70 graphite 70 soot/carbon black  69 surfactants, oil/water interface  76 stereoisomerism 33–36 stereoisomers  34–37, 346, 349 steroids  95, 269, 277 storage conditions, of perfumes  197 structure‐activity relationships (SAR) 324 structure‐odour relationship 324–325 s‐type orbital  9 styrene monomer/propylene oxide (SMPO) process  305 sub‐critical water  95 sublimation 93 sulfate esters  49, 59 sulfate turpentine  290, 296, 337, 341, 342 sulfonate  59, 77 sulfonic acids  59–60, 77, 169, 173 sulfonyl chlorides  59, 60 sulfoxides  59, 173 sulfur  3, 39, 58, 104 divalent sulfur compounds  58 hexavalent 60 molecule 39 oxidation states of  172 oxoacids, structure of  175 thio‐ether 59 thiols or mercaptans  58 sulfuric acid  49, 59, 60, 98, 136, 148, 149, 173, 230, 288, 316 super‐saturated solution  74 surface active agents  53, 75, 204 surfactants 75 anionic surfacants  77 cationic surfacants  79 in consumer goods  205 DEFI 77 LAS 77

neutral surfacants  78 sulfonic acids  77 two‐phase systems  73 sustainability, in fragrance industry definition 335 health and wellbeing and perfume/ sense of smell  355–356 malodour management  354 natural fragrance ingredients 340–341 social and health factors  353–356 synthetic ingredients  341–351 understanding olfaction  353–354 Symrise route, to l‐menthol production 347 synperonics 77 synthetic fragrance ingredients  283 biotechnology 345 environmental impact  342–344 factors for the use of  283 legal labelling requirements  284 l‐menthol 345 safety 284 secure supply using chemical process 284 stability issues  283 use of by‐products  341

t Takasago route, to l‐menthol 347–348 tartaric acid  36 tautomers  54, 155, 156, 160, 208 temperature ramping  101 tenacity, odour  320 terpeneless oil  92 terpenoids  95, 267, 290 building blocks  270, 271 defined 268 isoprene and head‐tail linkage in 269 isoprene units, mechanism of coupling  270, 271 terpineol  103, 297 tertiary‐butyl,‐C(CH3)3 22 tertiary‐butyl carbocation  143 tetraacetylethylenediamine (TAED) activators  179, 180, 202

Index

1,1,1,2‐tetrafluoroethane 95 tetrahedron  12, 14, 17, 53, 59, 331 tetrahydrofuran  61, 165 tetrahydropyran 61 tetrahydrothiophene 61 tetramethylsilane (TMS)  121 thermodynamics 133–135 thiazolidine 62 thin layer chromatography (TLC)  96, 98–99 thio‐ether  58, 59, 173, 174, 202 thioglycolic acid  200–204 thiols or mercaptans  58 thiomethanol 58 thionyl chloride (SOCl2) 50 thioterpineol 59 thyme 44 tincture 94 titration  110–113, 198, 221 tobacco smoke malodour  280 tocopherol  185, 203 toxicology 230–231 trail  194, 320 1,3‐trans  24 trans‐2‐butene 23 trans‐6‐dodecene 24 7‐transmembrane G‐protein coupled receptors 247 Traseolide  149 trickle down  195 triglycerides 224 trihydric alcohol glycerol  159 2,6,6‐trimethylbicyclo[3.1.1]hept‐2‐ ene 29 1,3,5‐trioxane  61, 159 triphenylmethane dyes  206 triple bonds  26, 28, 56 triple point  72 1,2,4‐trisubstituted aromatic ring  128 triterpene ambreine  278 tritium 8 trivial names  24, 32, 45, 46, 48, 57, 68 turpentine  28, 31, 86, 187, 238, 274, 288, 290, 291, 296, 300, 302, 326, 337, 341, 342, 350 2s orbital  9, 22

u ultimately biodegradable substance 235 Ultravanil  158 ultraviolet (UV)  100, 113–120, 206, 352 ultraviolet/visible (UV/Vis)  107, 113 unit cell, of crystal  67 unwanted acetal formation  160 unwanted aldol condensations  157

v valence  4, 9–11, 14, 19, 56, 148, 171, 300, 345 Van der Waals interactions  68–69 vanillin  158, 210, 288, 311 vapor pressure  73 vapour phase chromatography (VPC) 96 vapour splitting device  89 vegetable oils  15, 75, 199, 288 viscosity 109 visualisation 97–99 vitamin B12  5 volatility, perfume ingredients  191

w Wagner–Meerwein rearrangement 273 wanted acetal formation  160 wanted aldol condensations  157 water (H2O)  2, 3, 13, 15, 77, 112–113, 136 of cohobation  91 woody odorants  247, 342, 352 ω‐oxidation 236

x xylene 32

y ylang‐ylang  39, 190 Ysamber K  52

z Zwitterionic surfactants  79, 80

395

WILEY END USER LICENSE AGREEMENT Go to www.wiley.com/go/eula to access Wiley’s ebook EULA.