Characterization and Measurement of Flavor Compounds 9780841209442, 9780841211216, 0-8412-0944-8

Table of contents :
Title Page......Page 1
Copyright......Page 2
ACS Symposium Series......Page 3
FOREWORD......Page 4
PdftkEmptyString......Page 0
PREFACE......Page 5
1 Sensory Evaluation of Food Flavors......Page 7
History......Page 8
Consumer Testing......Page 9
Descriptive Analysis......Page 11
Statistical Analysis......Page 12
Literature Cited......Page 15
2 Substances That Modify the Perception of Sweetness......Page 17
Sweetness Inhibitors......Page 18
Mechanism of Action......Page 25
Acknowledgments......Page 29
Literature Cited......Page 30
Chemically-induced Oral Heat as Part of Flavor......Page 32
Psychophysical Characterization of Oral Irritation......Page 35
Sensory Evaluation of Pepper Heat......Page 39
CONCLUSIONS......Page 46
Literature Cited......Page 48
4 Analysis of Chiral Aroma Components in Trace Amounts......Page 49
Alcohols......Page 50
4- and 5-Hydroxyacid esters......Page 53
Investigation of chiral compounds formed during microbiological processes......Page 56
Determination of the enantiomeric composition of chiral aroma constituents in tropical fruits......Page 60
Literature Cited......Page 65
5 A New Analytical Method for Volatile Aldehydes......Page 67
Literature Review......Page 68
Experimental......Page 71
Sample preparations......Page 77
Results and Discussion......Page 78
Literature Cited......Page 83
6 The Use of High-Performance Liquid Chromatography in Flavor Studies......Page 85
Sweetness......Page 86
Bitterness......Page 89
HPLC Analysis of Flavors and Essential Oils......Page 91
HPLC as a Flavor Research Tool......Page 94
LC/HPLC......Page 97
HPLC/GLC/MS......Page 98
Literature Cited......Page 99
7 Capillary Gas Chromatographic Analysis of Volatile Flavor Compounds......Page 101
Sample Preparation......Page 102
Literature Cited......Page 114
8 High-Resolution Gas Chromatography-Fourier Transform IR Spectroscopy in Flavor Analysis Limits and Perspectives......Page 115
Literature Cited......Page 125
9 Tandem Mass Spectrometry Applied to the Characterization of Flavor Compounds......Page 127
Principles......Page 128
Types of Experiments......Page 131
Analytical Characteristics of MS/MS......Page 133
Applications to Flavor Compounds......Page 135
Applications......Page 137
Problems and Potentials of MS/MS......Page 141
Literature Cited......Page 143
10 Automated Analysis of Volatile Flavor Compounds......Page 144
Principle of Operation......Page 145
Experimental......Page 146
Results and Discussion......Page 148
Conclusion......Page 154
Literature Cited......Page 159
11 Supercritical Fluid Extraction in Flavor Applications......Page 160
Historical Perspective......Page 161
Operation of a Supercritical Fluid Extraction Process......Page 163
Applications of Supercritical Fluids to the Extraction and Characterization of Flavors......Page 170
Literature Cited......Page 180
Author Index......Page 182
A......Page 183
D......Page 184
G......Page 185
I......Page 186
O......Page 187
S......Page 188
Ζ......Page 189

Citation preview

289

ACS SYMPOSIUM SERIES

Characterization and Measurement of Flavor Compounds Donal U.S. Department of Agriculture

Cynthia J. Mussinan,

EDITOR

International Flavors and Fragrances

Developed fŕom a symposium sponsored by the Flavor Subdivision of the Division of Agricultural and Food Chemistry at the 188th Meeting of the American Chemical Society, Philadelphia, Pennsylvania, August 26-31, 1984

American Chemical Society, Washington, D.C. 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

Library of Congress Cataloging in Publication Data Characterization and measurement of flavor compounds. (ACS symposium series, ISSN 0097-6156; 289) "Developed from a symposium sponsored by the Flavor Subdivision of the Division of Agricultural and Food Chemistry at the 188th Meeting of the American Chemical Society, Philadelphia, Pennsylvania, August 26-31, 1984." Includes bibliographies and indexes 1. Flavor—Analysis—Congresses I. Bills, Donald D., 1932. II. Mussinan, Cynthia J., 1946.111. American Chemical Society. Division of Agricultural and Food Chemistry. Flavor Subdivision. IV. American Chemical Society. Meeting (188th: 1984: Philadelphia, Pa.) V. Series. TP372.5.C46 1985 ISBN 0-8412-0944-8

664

85-22913

Copyright © 1985 American Chemical Society All Rights Reserved. The appearance of the code at the bottom of the first page of each chapter in this volume indicates the copyright owner's consent that reprographic copies of the chapter may be made for personal or internal use or for the personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc., 27 Congress Street, Salem, MA 01970, for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to copying or transmission by any means—graphic or electronic—for any other purpose, such as for general distribution, for advertising or promotional purposes, for creating a new collective work, for resale, or for information storage and retrieval systems. The copying fee for each chapter is indicated in the code at the bottom of the first page of the chapter. The citation of trade names and/or names of manufacturers in this publication is not to be construed as an endorsement or as approval by ACS of the commercial products or services referenced herein; nor should the mere reference herein to any drawing, specification, chemical process, or other data be regarded as a license or as a conveyance of any right or permission, to the holder, reader, or any other person or corporation, to manufacture, reproduce, use, or sell any patented invention or copyrighted work that may in any way be related thereto. Registered names, trademarks, etc., used in this publication, even without specific indication thereof, are not to be considered unprotected by law. PRINTED IN THE UNITED STATES OF AMERICA

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

ACS Symposium Series M . Joan Comstock, Series Editor Advisory Board Robert Baker U.S. Geological Survey Martin L. Gorbaty Exxon Research and Engineering Co. Roland F. Hirsch U.S. Department of Energy Herbert D. Kaesz University of California—Los Angeles Rudolph J. Marcus Consultant, Computers and Chemistry Research Vincent D. McGinniss Battelle Columbus Laboratories

Robert Ory Virginia Polytechnic Institute and State University y Carnegie-Mellon University James C. Randall Phillips Petroleum Company Charles N. Satterfield Massachusetts Institute of Technology W. D. Shults Oak Ridge National Laboratory Charles S. Tuesday General Motors Research Laboratory

Donald E. Moreland USDA, Agricultural Research Service

Douglas B. Walters National Institute of Environmental Health

W. H. Norton J. T. Baker Chemical Company

C. Grant Willson IBM Research Department

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

FOREWORD The ACS S Y M P O S I U M S E R I E S was founded in 1974 to provide a medium for publishing symposia quickly in book form. The format of the Series parallels that of the continuing A D V A N C E S I N C H E M I S T R Y S E R I E S except that, in order to save time, the papers are not typeset but are reproduced as they are submitted by the authors in camera-ready form Papers are reviewed under the supervision of th Advisory Board and are selected to maintain the integrity of the symposia; however, verbatim reproductions of previously published papers are not accepted. Both reviews and reports of research are acceptable, because symposia may embrace both types of presentation.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

PREFACE B E C A U S E T H E P E R C E P T I O N O F F L A V O R involves human sensory organs, the characterization and measurement of flavor compounds require more than qualitative and quantitative chemistry. Both chemical and sensory studies are necessary to develop a full understanding of the role of a compound in food flavor. With the introduction of gas chromatography (GC) in the late 1950s, earl and sensory techniques by column when separating distillates or extracts of food. Not surprisingly, major components detected by the G C detector often had little or no perceptible odor, although some minor components were potently odorous. Familiarity with the odors of several hundred authentic compounds permitted these early workers to assign tentative identifications to many of the unknown compounds that eluted from their GC columns, and conclusive identifications by rigorous chemical or instrumental methods were often just necessary formalities. The use of the nose as an auxiliary GC detector is still of value, even though the sophistication of instrumental and sensory techniques for flavor compounds has increased greatly. In the broadest sense, flavor is a perceptual complex, an integration of information received by the brain from the five major human senses. The perception of the flavor of a food or beverage is influenced by appearance, textural or "mouth-feel" properties, sounds associated with mastication in the oral cavity, and sensations derived from the two chemical senses: taste and smell. Although taste and smell usually are most important to flavor and are the senses influenced by flavor compounds, the senses of sight, touch, and hearing must not be discounted. For example, American consumers expect potatoes to be white, but Europeans enjoy yellowish potatoes with a higher content of carotenoids; mashed potatoes should have a uniform, nongrainy texture; and potato chips must produce an audible crunch in the mouth. The purpose of this volume is not to explore the optical and structural characteristics of food, but one must bear in mind the modifying influence that the senses other than olfaction and gustation have on the perception of flavor. The number of compounds that may contribute to flavor is large. In general, those that influence the sense of smell must be volatile enough to reach the olfactory organ in the upper nasal cavity, whereas those that influence the sense of taste need not be volatile but must have at least a vii In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

degree of solubility in water. Volatility and solubility do not assure that a compound will be perceived as having an odor or taste. The human sensory apparatus is quite "blind" to many compounds and has variable sensitivity to those that elicit the sensation of flavor. Some compounds are organoleptically detectable and contribute to food flavor at the parts-per-billion level, whereas others must be present at concentrations exceeding a part per thousand to serve as flavor stimuli. The natural flavor of foods typically depends upon the presence of a large number of flavor compounds of several chemical classes, with varying volatilities and polarities, in both large and small concentrations, and in the proper proportions to each other to elicit the characteristic sensation of flavor for a given food. Thus, the characterization and measurement of flavor compounds provide a formidable and continuing challenge. The authors of this boo flavor research. Many improvement capabilities have accrued in recent years, and state-of-the-art instrumentation and instrumental techniques for flavor analyses compose a substantial portion of this volume. New methods for extracting, derivatizing, and otherwise manipulating flavor compounds are another important part of this book, as are the chapters that deal with sensory evaluation. As editors, we are grateful to the authors for their contributions and to our respective employers for their support of our effort. DONALD

D.

BILLS

U.S. Department of Agriculture Philadelphia, PA 19188 CYNTHIA

J.

MUSSINAN

International Flavors and Fragrances Union Beach, NJ 19118 June 24, 1985

viii In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

1 Sensory Evaluation of Food Flavors M . R. McDaniel Department of Food Science and Technology, Oregon State University, Corvallis, OR 97331-6602

Many changes have occurred in the sensory analysis of flavor in the past half century, beginning with the phasing out of the inappropriate term "Organoleptic," and the terms to describ Analysis" or "Sensory Science." In the food and flavor industries, bench scientists using unsophisticated methods have been replaced by highly educated specialists (M.S. and Ph.D.'s in Food Science and Psychophysics) directing large sensory groups. These specialists utilize highly trained judges and sophisticated test designs and analysis to solve sensory problems. Consumer testing in most industry applications now is done with consumers. Even with the continuing development of instrumentation to replace the human judge, sensory analysis continues to expand its contribution to flavor analysis.

Sensory evaluation (sensory science) i s a s c i e n t i f i c d i s c i p l i n e that concerns the presentation o f a stimulus ( i n t h i s case a flavor compound, a f l a v o r , or flavored product) to a subject and then evaluation o f the s u b j e c t s response. The response i s expressed as, or translated i n t o , a numerical form so that the data can be s t a t i s t i c a l l y analyzed. The sensory s c i e n t i s t then collaborates with the research or product development team to i n t e r p r e t the results and to reach decisions. Sensory s c i e n t i s t s stress that decisions, such as product formulation, are made by people, not by the r e s u l t s of a sensory t e s t , although such r e s u l t s may provide powerful guidance i n the decision-making process. Sensory science i s unique i n that i t requires human subjects. This i n i t s e l f creates challenges, some of which w i l l be discussed i n t h i s paper. The sensory s c i e n t i s t , often working as a part of a research team, also i s unique because t r a i n i n g i n a number o f f i e l d s i s necessary to the success o f the program. The t r a i n i n g of sensory s c i e n t i s t s has not proceeded as rapidly as has the appreciation of and need for sensory s c i e n t i s t s i n the flavor and 1

0097-6156/ 85/ 0289-0001 $06.00/ 0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

food i n d u s t r i e s . In the past, v i r t u a l l y anyone, regardless of t r a i n i n g and background, might be handed a sensory methods manual, and t o l d that they were i n charge of the sensory program. Fortunately this rarely takes place today. Sensory s c i e n t i s t s are s t i l l i n r e l a t i v e l y short supply, but there are dozens of h i g h l y trained men and women who are currently d i r e c t i n g sensory programs i n both flavor and fragrance companies and t h e i r c l i e n t s ' companies. A sensory s c i e n t i s t working i n the area of flavor would be expected to have a background i n food science and at l e a s t basic knowledge i n the areas of physiology, psychology and s t a t i s t i c s . History Forss (1) reviewed the r e l a t i o n s h i p between sensory analysis and flavor chemistry, and provided a discussion of sensory characteri z a t i o n of f l a v o r . Williams e t a l . (2) also addressed the problem of r e l a t i n g the many know perceived by the i n d i v i d u a chapter w i l l center more on reviewing advances i n sensory methodo logy as they have been adapted by the flavor and food industry. Moskowitz (3) i n h i s book "Product Testing and Sensory Evaluation of Foods" reviewed the h i s t o r y of sensory evaluation beginning with the study of psychophysics. Psychophysics i s the study of the r e l a t i o n s h i p between a p h y s i c a l stimulus and a subject's psychological response to that stimulus. Very e a r l y work was done i n Germany by E.M. Weber and G.T. Feckner over a hundred years ago. They were seeking quantitative laws o f human perception, s p e c i f i c a l l y how to measure our a b i l i t y to discriminate. The study of psychophysics advanced r a p i d l y over the following years and along with i t , so d i d testing methodology. As the psychologists developed and tested t h e i r methods, food s c i e n t i s t s and sensory analysts borrowed those methods and applied them to the study of food. Moskowitz {3) further observed that sensory analysis evolved greatly during World War I I . Much of t h i s work was done by the U.S. Army Quarter-Master Corps, l e d by Peryam and P i l g r i m , developers of the 9-point Hedonic scale. Also during t h i s time, Stevens (4), a psychologist at Harvard U n i v e r s i t y , was working on the psychophysics of hearing, and h i s work l e d to the development of a theory of scales of measurement, and ultimately to the development of the power law. The power law i s generally accepted to define the r e l a t i o n s h i p between psychol o g i c a l response and physical stimulus. The l o g of t h i s equation i s readily recognizable as the equation for a s t r a i g h t l i n e , therefore p l o t s of psychological response versus physical stimulus concentration on a log-log p l o t r e s u l t i n a s t r a i g h t l i n e . These p l o t s , depending upon which sensation i s being measured, be i t e l e c t r i c a l shock, brightness or sweetness, have d i f f e r e n t slopes, and the slope i s defined as the power of function. Food s c i e n t i s t s and psychophysicists have applied these laws to food evaluations and have obtained valuable information. Although knowledge of the power law has been a v a i l a b l e for many years, food s c i e n t i s t s have used t h i s information only r e cently to define how the various sensory q u a l i t i e s i n foods

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

1.

McDANIEL

Sensory Evaluation of Food Flavors

change, with changes i n concentration. I t i s s u r p r i s i n g that many food companies with only a few major products s t i l l do not know what the sensory properties of their products are or how they change with changes i n formulation. Scaling Methodology A new method c a l l e d magnitude estimation, a form of r a t i o s c a l i n g , also emerged from Stevens (4J work. To estimate the magnitude of some stimulus, p a n e l i s t s simply assign a number to r e f l e c t the stimulus i n t e n s i t y . P r i o r to the use of magnitude estimation, food s c i e n t i s t s and other psychologists used rather a r b i t r a r y category scales i n order to quantify perceptions. The scales could vary i n length, from a three point scale to perhaps a nine point scale or even l a r g e r , usually with each scale point representing some i n t e n s i t y and designated as small, moderate, l a r g e , extreme or by some other d e s c r i p t i v e the scale i t s e l f and th one p l o t t e d the r e l a t i o n s h i p between psychological response and stimulus i n t e n s i t y , one usually would not obtain a l i n e a r response. A log-log p l o t of stimulus i n t e n s i t y versus psychological response for magnitude estimation usuall y r e s u l t s i n a s t r a i g h t l i n e . Many studies have attempted to compare the r e s u l t s of magnitude estimation versus some of the more standard s c a l i n g techniques or semi-structure or unstructured l i n e techniques and d i f f e r e n t r e s u l t s have arisen from these various studies (5-8). Based on my experience, the more important consideration i s the degree of t r a i n i n g of the panel u t i l i z e d i n trained panel work rather than the type of scale that i s used. However, magnitude estimation results do tend to show more differences when the differences are very small than do other methods. Consumer Testing Perhaps the biggest advance i n consumer t e s t i n g i s the use of consumers instead of t r y i n g to obtain consumer data from people who do not use the product or who are too close to the product (a company's own workers). I t once was standard procedure to do in-house consumer t e s t i n g with company employees evaluating the products that produced t h e i r paychecks. This was not an unbiased sample and l e d to many expensive mistakes by industry. Most companies now use a marketing organization to do c e n t r a l l o c a t i o n testing or home placement t e s t i n g of their products. I h i s i s not to say that a l l in-house consumer t e s t i n g i s i n c o r r e c t . The concern i s that d e f i n i t e r i s k s are involved, and one must understand that misinformation may r e s u l t . Some very valuable d i r e c t i o n a l information can be gained i n the r i g h t circumstances by using in-house panels. V a l i d i t y of t h i s information can only be judged through experience with the i n d i v i d u a l i n d u s t r i e s and products involved. One of the most important things to learn i n sensory evaluat i o n i s that "experience" i s c r i t i c a l i n making methodology decisions. Years of sensory t e s t i n g on one product or one product

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3

4

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

l i n e y i e l d s valuable information that should be used by the researcher and the product developer i n making decisions regarding their current l i n e . However, using the same methodology j u s t because i t ' s been used for years and years i s not necessarily a wise decision. Some methodologies are incorrect or i n s e n s i t i v e , or the words used i n scales have l i t t l e r e l a t i v e meaning. One should not be a f r a i d to t e s t new methodology versus o l d methodology to see i f valuable information can be gained by switching to another method. The food industry also has learned that they must test consumers i n large numbers. However, unfortunately, there i s no magic "large" number, and the number of judgments that i s decided upon must be based, again, on the r i s k they are w i l l i n g to take. Another valuable lesson learned over the years i n t e s t i n g products i s that consumers are a l l very d i f f e r e n t , and these very d i f f e r e n t segments of the population, where a product market e x i s t s , must be changing the formulatio consumers of the product to see i f the change makes a difference to them. They aJlso may wish to test consumers of t h e i r product versus consumers of a competitor's product, but t h i s y i e l d s d i f f e r e n t information. I f the consumers of their products are c h i l d r e n , they must t e s t c h i l d r e n , and perhaps also the parent who purchases the product. I f the consumers of a product are on s p e c i a l d i e t s , people on s p e c i a l d i e t s should be the subjects of the t e s t . For example, a non-gluten bread may be highly appreciated by one on a gluten-free d i e t , but t o t a l l y unacceptable to those who have no need to r e s t r i c t gluten i n t h e i r d i e t s . Experience has shown that the order of sample presentation, when a number of samples are to be presented at one time, i s very important. The f i r s t product may strongly bias the evaluation of the product following i t , and one may find a s i g n i f i c a n t order e f f e c t with some products. The sample presentation order i n any test must be balanced or randomized. In analyzing data from such tests, one should consider the average score of one sample presented f i r s t versus i t s average score when presented second. Ihe f a c t that your competitor's product presented f i r s t somehow causes judges to score your product presented second lower, can be valuable information. What was i t about the competitor's product or about your product that caused t h i s difference? The amount of sample i s also c r i t i c a l . Very often i n a taste t e s t , people are given j u s t a "taste" and t h i s may not be enough for them to t r u l y evaluate the product. For example, when testing soup, where the flavor w i l l b u i l d up, mouthful a f t e r mouthful, i t may be necessary to have each p a n e l i s t consume an e n t i r e bowl of soup. The f i r s t impression gained from one or two b i t e s of a product may be t o t a l l y d i f f e r e n t than the f i n a l impression a f t e r consuming an e n t i r e serving. This can be true with many products, e s p e c i a l l y highly flavored products.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

M c DAN I EL

Sensory Evaluation of Food Flavors

Difference Testing Difference t e s t i n g has not changed greatly over the years. The t r i a n g l e t e s t and duo-trio t e s t remain popular and well-accepted, although much e f f o r t has been extended to prove one better than the other. The best advice i s to use the one that f i t s your test situation. When conducting one o f the pure difference t e s t s , the triangle or duo-trio t e s t , one must r e a l i z e that there are two d i s t i n c t types, one being a s i m i l a r i t y taste and the other being a difference test. In the case of the s i m i l a r i t y t e s t the experimental samples are a c t u a l l y d i f f e r e n t from the c o n t r o l . However, the purpose i s not to create a perceptible difference but to e f f e c t a cost reduction or to change suppliers of raw materials without changing the product i d e n t i t y . The actual goal of such a p r o j e c t i s to change the product without i n f l u encing consumer perceptio error of concern here i of concluding that the two samples are not d i f f e r e n t when they are d i f f e r e n t . In such t e s t i n g , i t i s desirable to keep the beta r i s k very low, but keeping both the alpha and beta r i s k s low i s d i f f i c u l t . When alpha i s low, beta tends to be high, and when beta i s low, alpha tends to be high. The only way to insure that both errors are small i s to t e s t a very large number of people. In a difference t e s t , rather than a s i m i l a r i t y t e s t , one would i n t e n t i o n a l l y make the samples d i f f e r e n t and then ascertain whether judges could detect the difference. Product D r i f t A major concern i n the food industry i s product d r i f t or subtle, step-wise changes that take place i n the product over time. Product d r i f t can occur when the o r i g i n a l ( f i r s t ) product i s tested and found not d i f f e r e n t from the new (second) product. The second product i s changed i n y e t another way, but i s found not to be d i f f e r e n t from the newest (third) product. However, the t h i r d product may be d i f f e r e n t from the f i r s t product. Hie best way to avoid product d r i f t i s to intimately know a product. This means knowing exactly what sensory c h a r a c t e r i s t i c s are present i n your product and a t what i n t e n s i t i e s . This type of sensory analysis requires the use of descriptive analysis techniques. Because d e s c r i p t i v e analysis involves using a trained panel, developing a set of descriptors, and r a t i n g their i n t e n s i t i e s , i t can be very expensive, but so can product drift. Descriptive Analysis One of the most e x c i t i n g developments i n sensory evaluation over the past decades has been the emergence and popularity of descriptive analysis or the use of highly trained panels to describe the sensory c h a r a c t e r i s t i c s of foods. This i s perhaps the most important development i n sensory evaluation methodology.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

6

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

Only when products are described i n d e t a i l and the i n t e n s i t y of descriptors rated can true product differences or d r i f t be noted. The type of d e s c r i p t i v e analysis chosen should be based on the v a r i e t y of products produced by the company. I f a company produces only one type of product, a more l i m i t e d or s p e c i f i c t r a i n i n g such as quantitative d e s c r i p t i v e analysis described by Stone et a l . (9) may s u f f i c e . I f a company has a large variety of products, flavor p r o f i l e t r a i n i n g established by the Arthur D. L i t t l e Company may be more e f f i c i e n t . Many laboratories combine the best of both methods and develop t h e i r own method. The key to success with e i t h e r method i s panel t r a i n i n g and the e s t a b l i s h ment of appropriate terminology. To i l l u s t r a t e d e s c r i p t i v e a n a l y s i s , I w i l l draw from both the wine and beer industry. Oregon State University's Sensory Science Laboratory, located i n the Department of Food Science and Technology, i s heavily involve p r i n c i p l e problems and solution and beer should be transferable to other products. Common wine d e s c r i p t o r s , such as s o f t , hard, f a t , are ambiguous. What do s o f t or hard mean when r e f e r r i n g to wine? The goal of d e s c r i p t i v e analysis i s to use precise terms, even r e f e r r i n g to s p e c i f i c chemical e n t i t i e s when possible. In the wine industry, objective sensory analysis must overcome the h i s t o r i c a l romance of wine. A c l a s s i c example i s the following description of a wine, " I t ' s a naive domestic burgundy without any breeding but I think y o u ' l l be amused by i t s presumption." Such a d e s c r i p t i o n obviously lacks meaningful sensory terms that convey an impression of the wine's aroma and taste. In work on Pinot Noir q u a l i t i e s i n our laboratory, a set of sensory descriptors were developed to aid i n evaluating the e f f e c t of several processing v a r i a b l e s , Henderson and McDaniel (10_) . A trained panel used the b a l l o t shown i n Figure 1 to describe wine produced by d i f f e r e n t malolactic cultures. There are several ways to display this type of i n f o r mation. The QDA method joins descriptor i n t e n s i t y points together to v i s u a l l y display difference. This works very n i c e l y for two to three comparisons. When one has more than two to three samples to compare, other types of s t a t i s t i c a l analyses and methods of displaying the r e s u l t s may be employed. This w i l l be covered i n a l a t e r section. S t a t i s t i c a l Analysis Sensory s c i e n t i s t s r e l y greatly on s t a t i s t i c a l analysis to aid i n i n t e r p r e t a t i o n of data. Ihey also continue to argue endlessly about what i s correct and i n c o r r e c t . Some of the questions that are posed include: which i s the r i g h t analysis; are the assumptions of the t e s t being v i o l a t e d ; i s the data good enough i n the f i r s t place to have s t a t i s t i c a l analyses applied to i t . Some of the most i n t e r e s t i n g advances i n sensory analysis i n the l a s t 20 years have been i n the area of s t a t i s t i c a l evaluation of the r e s u l t s . M u l t i v a r i a t e analysis i s an example of a new type of s t a t i s t i c a l analysis applied to food system. M u l t i v a r i a t e

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

1.

M c DAN I EL

1

Sensory Evaluation of Food Flavors

NAME SAMPLE # DATE Using the 9-point intensity scale shown below, rate each sample for all attributes listed. FOR AROMA ONLY 1 - none 2 - threshold 3 - slight 4 - slight to moderate Overall Intensity

5 - moderate 6 - moderate to large 7 - large 8 - large to extreme 9 - extreme 1st tier Frui ty

Citrus Berry

qrapefruit blackberry strawberry raspberry

Tree Fruit Dried Fruit

cherry strawberry jam raisin fig prune

Spicy

Spicy

black pepper cloves

Vegetative

Canned/cooked

Earthy Caramelized

Caramelized

honey buttery butterscotch

Chemical

Pungent

Ethanol

Sul fur Microbiological

lactic

Figure 1. B a l l o t used by trained panel f o r evaluation of wine.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

8

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

analysis i s used when observations are c o l l e c t e d on many d i f f e r e n t v a r i a b l e s , and i s p a r t i c u l a r l y h e l p f u l when one may be overwhelmed by the sheer bulk of the data that has been c o l l e c t e d . There are several types of multivariate a n a l y s i s , 1) p r i n c i p a l component analysis; 2) factor analysis; 3) c l u s t e r a n a l y s i s ; 4) discriminate a n a l y s i s . P r i n c i p l e component analysis and factor analysis are very s i m i l a r , a major difference being that factor analysis requires assumption of normality. Both methods take a l i s t of variables and reduce t h i s to a smaller number of factors (made up of these o r i g i n a l v a r i a b l e s ) . Such analyses have been applied in wine research (13, 14). An example would be i n the evaluation of wine where sensory and/or a n a l y t i c a l measurements have been taken on many samples. The goal would be to reduce the number of variables necessary to a c t u a l l y show meaningful differences between the wine samples. Factor analysis with vector loading i s h e l p f u l i n t h i s s i t u a t i o n . The loadings t e l l you how each variable f i t s under eac Cluster analysis allow together i n a multidimensional space. An analogy can be drawn to playing cards, which may c l u s t e r i n obvious ways by s u i t or card, or by other means inherent i n some game playing r u l e s . Cluster analysis i s generally a stepwise progression. An example, from hop variety research involving eight v a r i e t i e s and many samples of each, i s an appropriate subject for a p p l i c a t i o n of t h i s technique (15). The samples clustered near one side would be most s i m i l a r . Also, any two compounds whose concentration r a t i o s were r e l a t i v e l y constant i n a l l or most samples would have a high s i m i l a r i t y value and would c l u s t e r on the same side. With discriminate analysis one i s concerned about how observations d i f f e r and one sets the rules to d i s t i n g u i s h between populations. The r e s u l t i s a type of c l a s s i f i c a t i o n or sorting of observation into groups. For example, an unknown wine v a r i e t y may be c l a s s i f i e d among known v a r i e t i e s . Another example i s the a p p l i c a t i o n of response surface methodology (16_, 17) . The response i s some function of the design variables i n the t e s t and a l l of the variables are well c o n t r o l l e d and p r e c i s e l y measurable. In order to v i s u a l i z e response surface methodology, imagine that you are viewing a mountain on the horizon. This would be a single variable χ and i t s response, y showing the maximum. Imagine adding a second dimension, 2, coming s t r a i g h t out at you, to give the mountain dimension, and then s t a r t i n g at the very top of the mountain, taking s l i c e s of that mountain on a h o r i z o n t a l axis at equal response l i n e s . I f you then look down on that s l i c e of the mountain, you w i l l see c i r c l e s , the smallest inner c i r c l e equaling the l a r g e s t response. Response surface procedures are not used to understand the mechanism of the underlying system, but rather to determine what optimum operating conditions are or to determine a region of the t o t a l space of the factors i n which c e r t a i n operation s p e c i f i c a t i o n s are met.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

1. M e DANIEL

Sensory Evaluation of Food Flavors

Conclusion Almost everyone i s now u t i l i z i n g the computer for s t a t i s t i c a l analysis o f sensory data. Some laboratories also are using computers to gather the data as w e l l (18). A computerized sensory system would benefit most laboratories by freeing workers from laborious data entry and a n a l y s i s . A l s o , i t would allow for a more thorough analysis o f the data. I t should not replace inspection o f the raw data by the sensory s c i e n t i s t , but allow t h i s to occur more e a s i l y . The f i e l d of sensory evaluation has matured over the years. We have learned through expensive mistakes to rigorously control test s i t u a t i o n s to obtain v a l i d data and to analyze the data as thoroughly as possible to maximize understanding o f products. I believe the future o f sensory evaluation w i l l involve an expansion o f the use o f descriptive analysis i n many d i f f e r e n t s i t u a t i o n s , such as i development and researc competition i n the flavor industry, flavor companies are increasingly expanding their sensory work and sensory capa­ b i l i t i e s . This i s necessary, not only for the flavor company to understand the products they are producing but to be able to s a t i s f a c t o r i l y service t h e i r c l i e n t companies. Literature Cited 1. 2.

3. 4. 5. 6. 7. 8. 9. 10. 11.

Forss, D.A., in Flavor Research Recent Advances. (S.R. Tannenbaum and P. Watson, ed.). Marcel Dekker, Inc. 1981. p. 125 Williams, Α.Α.: Lea, A.G.H.; Timberlake, C . F . , in Flavor Quality Objective Methods. (R.A. Scanlan, ed.) ACS Symposium Series No. 51, Washington, D.C., American Chemical Society, 1977. p. 71. Moskowitz, H.R. Product Testing and Sensory Evaluation of Foods. Food and Nutrition Press, Inc. Westport, CT. 1983. Stevens, S.S. Science 1946, 103, 677-678. Moskowitz, H.R. and Sidel, J . L . J . Food Sci. 1971, 36, 677-680. McDaniel, M.R. and Sawyer, F.M. J . Food Sci. 1981, 46, 178-181. Giovanni, M.E. and Pangborn, R.M. J . Food Sci. 1983, 48, 1175-1182. Shand, P . J . ; Hawrysh, A . J . ; Hardin, R.T.; and Jeremiah, L . E . J . Food Sci. 1985, 50, 495-500. Stone, H.; Sidel, J.; Oliver, S.; Woolsey, H.; and Singleton, R.C. Food Technology 1974, 28(11), 24-34. Henderson, L.A. and McDaniel, M.R. Personal communication, 1985. Ennis, D.M.; Boelens, H.; Haring, H.; and Bowman, P. Food Technol. 1982, 36(11), 83-90.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9

10 12. 13. 14. 15. 16. 17. 18.

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

Johnson, R.A. and Wichern, D.W. Applied Multivariate Statistical Analysis. Prentice-Hall, Inc. Englewood Cliffs, New Jersey, 1982. Wu, L . S . ; Bargmann, R.E.; and Powers, J.J. J . Food Sci. 1977, 42, 944-952. Williams, A.A. J . Inst. Brew. 1982, 88, 43. Stenroos, L . E . and Siebert, K . J . ASBC Journal 1984, 34, 55. Henika, R.G. Food Technology 1982, 36(11), 96-101. Myers, R.H. Response Surface Methodology. Allyn and Bacon, Inc. Boston, 1984. Brady, P . L . ; Ketelsen, S.M.; and L . J . P . Ketelsen. Food Technology 1985, 39 (5), 82-88.

RECEIVED

July 29, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2 Substances That Modify the Perception of Sweetness Michael A. Adams Monell Chemical Senses Center, University of Pennsylvania, Philadelphia, PA 19104

Human taste response is modified by several plant-derived substances. The detergent sodium dodecyl sulfate, as well as triterpene saponins from the leaves of several plant species (most notably Gymnema sylvestre and Ziziphus jujuba) wil sensation in man; one hour for G. sylvestre and about fifteen minutes for Z. jujuba. The mechanism of action seems to be related, in part, to the surfactant properties of the materials. Structures of the modifiers and possible mechanisms of action are discussed. The number of substances known to modify sweet taste i s quite small. M i r a c u l i n , a substance from the f r u i t of Synsepalum dulcificum, has the a b i l i t y to make sour s t i m u l i taste sweet. The f r u i t s of t h i s West A f r i c a n shrub ("miracle f r u i t " ) have been used by natives to improve the f l a v o r of maize bread, sour palm wine, and beer ( 1_). The taste modifying e f f e c t i s of quite long duration, sometimes l a s t i n g for more than three hours. M i r a c u l i n i s a glycoprotein with a molecular weight of about 4U,000 (2). Not much i s known about i t s structure or i t s mode of action, but i t has been proposed (3_) that t h i s substance binds to the taste c e l l surface, where i t s e f f e c t i s then manifested. I t was further suggested that exposure of the taste receptors to acid contained i n a food or beverage causes a conformational change i n the membrane enfolding the receptor, which allows the sugar moieties bound to miraculin to stimulate the sweetness receptor. The speed of response of miraculin i s on the order of milliseconds, and t h i s i s taken as evidence that the sweetness receptor l i e s on the outside of a taste c e l l , since the speed of response i s apparently too fast to allow for transport i n t o the c e l l i n t e r i o r . Additional evidence that the e x t e r i o r surface of the taste receptor c e l l plasma membrane i s the l o c a t i o n of the sweet receptor i s provided by the action of the chemostimulatory proteins, monellin and thaumatin. Monellin occurs i n the f r u i t of the African serendipity berry (Dioscoreophyllum cumminsii), and thaumatin i s found i n the f r u i t of Thaumatococcus d a n i e l l i i , also 0097-6156/ 85/ 0289-0011 $06.00/ 0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

12

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

from A f r i c a . Both of these proteins have an intensely sweet t a s t e . Studies have shown that they are e s s e n t i a l l y carbohydratefree, i n contrast with miraculin. Monellin i s a single peptide chain with a molecular weight of about 11,000, and thaumatin, believed also to be a s i n g l e peptide chain, has a weight of approximately 21,000. Evidence has been presented i n d i c a t i n g that the t e r t i a r y structures of these proteins are c r i t i c a l to the sweet taste , and Cagan (5) has suggested that the size of these molecules renders i t u n l i k e l y that they are being transported i n s i d e the taste receptor c e l l . I t i s reasonable to conclude that t h e i r taste-stimulating e f f e c t s are most l i k e l y being manifested at the c e l l surface. The e f f e c t s of miraculin, monellin and thaumatin, taken together, provide evidence that the sweetness receptor s i t e i s located at the surface of the taste receptor c e l l , i n or i n close proximity t o , the plasma membrane. Sweetness I n h i b i t o r s There are only a few sweetness i n h i b i t o r s known. This a r t i c l e w i l l focus on three: sodium dodecyl s u l f a t e (SDS), the gymnemic acids (GA) and the z i z i p h i n s (ZJ). Each of these substances has the a b i l i t y to diminish or eliminate the a b i l i t y to recognize sweetness. The i n t e n s i t y and duration of the effect varies with the i n h i b i t o r . The best-known sweetness i n h i b i t o r i s sodium dodecyl sulfate (SDS), also known as sodium l a u r y l s u l f a t e . This substance i s a twelve carbon surfactant that i s quite commonly used as a detergent i n toothpaste. The observation i s often made that a f t e r brushing one's teeth, the taste of orange juice i s unusually b i t t e r . This has been ascribed to the presence of SDS i n the d e n t i f r i c e (6). The duration of the e f f e c t , as measured i n psychophysical experiments, i s very b r i e f - on the order of minutes, and the suppression effect i s not complete. A substance i s o l a t e d from the Indian shrub Gymnema s y l v e s t r e , has a profound a b i l i t y to reduce perceived sweetness of sugar s o l u t i o n s . The e f f e c t was noticed over a century ago when two B r i t i s h inhabitants of an Indian v i l l a g e found that, a f t e r chewing the leaves of G_^ s y l v e s t r e , the sweetness of t h e i r tea disappeared (7). The sweetness suppressing a c t i v i t y i s due to a mixture of several triterpene saponins which have c o l l e c t i v e l y been termed the gymnemic acids. For most people exposed to the e f f e c t s of GA, sweetness suppression i s complete and the e f f e c t l a s t s for about an hour. Much work was done i n the I960's by S t o c k l i n , Sinsheimer and t h e i r coworkers to i s o l a t e , p u r i f y , and elucidate the structures of these taste i n h i b i t o r s (8-11). Many substances without antisweet a c t i v i t y were i s o l a t e d from the leaves of G^. s y l v e s t r e , including hydrocarbons, stigmasterol, 3-amyrin, lupeol and phytol, among others. A crude antisweet extract was, however, produced by mineral acid p r e c i p i t a t i o n of an aqueous l e a f extract. This extract, when dried, accounts f o r about 10% of the l e a f material. Chloroform extraction was then used to remove some remaining i n a c t i v e material. Pfaffmann (12) determined that the mixture of taste modifying substances was g l y c o s i d i c i n nature, as glucose,

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2. A D A M S

Modifying the Perception of Sweetness

13

arabinose, and glucuronolactone were obtained upon hydrolysis. The hydrolysate was found to have no antisweet e f f e c t . Yackzan (7) postulated that the sweetness-inhibiting gymnemic acid must be a saponin, based on i t s foaming i n s o l u t i o n , a b i l i t y to produce red blood c e l l l y s i s , and i t s g l y c o s i d i c nature. Extraction of the crude l e a f f r a c t i o n with acetone removes a l l p h y s i o l o g i c a l l y active material. This acetone-soluble material was chromatographed on s i l i c a g e l by S t o c k l i n (9) to give the various c l o s e l y - r e l a t e d gymnemic acids. Further p u r i f i c a t i o n by Sinsheimer and Subba Rao led to the s t r u c t u r a l i d e n t i f i c a t i o n of gymnemagenin (I) and gymnestrogenin (II) as the aglycones of the gymnemic acids (Figure 1). The major taste active substance appears now to be gymnemagenin ( I ) , e s t e r i f i e d with glucuronic acid. This saponin i s c a l l e d gymnemic acid A, ( I I I , Figure 2 ) . The i d e n t i t y of the R groups i s presently unknown, although claims f o r formic, a c e t i c , i s o v a l e r i c and t i g l i c acids have been made UO, 13, 1M). No further s t r u c t u r a l work the l a s t several years Once a p u r i f i e d gymnemic acid became a v a i l a b l e , much psycho­ physical work was done to understand the nature of the sweetness i n h i b i t i o n e f f e c t . The work of Bartoshuk and co-workers i l l u s t r a t e s the course taken (15). The r e s u l t s of a t y p i c a l experiment are shown i n Figure 3. The sweetness of a sucrose s o l u t i o n was almost completely suppressed a f t e r holding a gymnemic acid solution i n the mouth f o r a few seconds. Further experiments were carried out to determine the e f f e c t of gymnemic acid on the other taste q u a l i t i e s (sour, b i t t e r and s a l t y ) . No e f f e c t of gymnemic acid on these tastes was observed. Early work with GA extracts had produced an apparent i n h i b i t i o n of b i t t e r taste, but t h i s effect was l a t e r a t t r i b u t e d to cross-adaptation to the taste of the crude l e a f extract, which was i t s e l f quite b i t t e r . Experiments with refined (and tasteless) extracts showed no bitterness suppression. A t h i r d group of sweetness i n h i b i t i n g substances has recently been i s o l a t e d from the leaves of the Middle Eastern t r e e , Ziziphus jujuba. This material, consisting of a mixture of 15-20 triterpene saponins, has a c t i v i t y s i m i l a r to that of the gymnemic acids, but the duration of the e f f e c t i s much shorter, on the order of 15 minutes f o r the average subject. The duration of suppression varies from subject to subject, and i s usually not complete. The i s o l a t i o n , p u r i f i c a t i o n , and s t r u c t u r a l characterization of these sweetness i n h i b i t o r s , which have been termed the z i z i p h i n s , i s under i n v e s t i g a t i o n i n our laboratory and some of our recent r e s u l t s are described below. Psychophysical and preliminary chemical studies were c a r r i e d out with 2^. jujuba (ZJ) extracts by Meiselman and co-workers (16). Their chromatographic investigations showed that the gymnemic acids were not present i n ΖJ l e a f extracts, but t h e i r observations of the physical properties of l e a f extract solutions d i d indicate that saponin-like substances were present. The e f f e c t of z i z i p h i n treatment of the tongue on sweet, sour, b i t t e r , and s a l t y tastes were examined. As with the gymnemic acids, ΖJ only reduced the i n t e n s i t y of sweet taste. A more detailed i n v e s t i g a t i o n of the z i z i p h i n s was carried out by Kennedy and Halpern (17). Included i n t h e i r work was a phylogenetic analysis of Z. jujuba and Gymnema sylvestre (Figure

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

14

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

II Figure 1. Structure of gymnemagenin (I) and gymnestrogenin ( I I ) .

Figure 2. Structure of gymnemic acid A, ( I I I ) , the most abundant gymnemic a c i d .

before

χ after G s >
eop eophyto

Subdivision

Angiosperme»

Class

Dichotyledoneoe

_ _ ____ : 1

Subclass

Archichlamydeae

Order

R h a m n o u :s

Family

Rhomnoceoe

A

Metochlomydeae

Gept

A

males

—

,

Asclepiodoceae piadai

Genus

Ziziphus

Gymnema ymnen

j\\ Species

sylvestre

jujubo /Ziziphus

jujuba]

/ Gymnema

sylvestre

Figure 4. Phylogenetic c l a s s i f i c a t i o n of Ziziphus jujuba and Gymnema s y l v e s t r e . Lineage. Since separation occurs e a r l y , a t the subclass l e v e l , these two species do not appear c l o s e l y r e l a t e d . S o l i d arrows indicate single pathways. Open arrows i n d i c a t e multiple pathways. Reproduced with permission from Ref. 17. Copyright 1980, ILR Press, Ltdo

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

J

ADAMS

Modifying the Perception of Sweetness

Ground, dried leaves of Ziziphus jujuba

MeOH-H 0 {2-\), 3x 2

residue

hexane ext.

organi

Et 0 ext. 2

X organic

organic 1. evap.MeOH 2. Et 0

aqueous BuOH ext.

aqueous

2

ppt.

supernatant

Crude antisweet fraction, ~ 4 % of leaf; C a . 1 5 - 2 0 components

Figure 5. Scheme f o r i s o l a t i o n of crude antisweet f r a c t i o n from the leaves of Ziziphus jujuba. Taste bioassays were performed at each step to a s c e r t a i n antisweet a c t i v i t y . An X i n d i c a t e s that no antisweet a c t i v i t y was detected.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

18

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

Crude antisweet fraction (4%

Flash chromatography (silica gel, solvent A )

of leaf; saponin containing)

ν

Y

DCC

Prep TLC

HPLC

(solvent A)

( reverse phase, C-18

(solvent A)

Solvent A = GHCI : MeOhh PrOH= H 0 3

2

Solvent Β = MeOH=H 0 2

Solvent C = C H C N = H 0 3

2

Figure 6. Techniques used f o r i s o l a t i o n and p u r i f i c a t i o n of antisweet substances from the crude antisweet l e a f f r a c t i o n of Ziziphus jujuba*

IV

v

Figure 7o Structure of e b e l i n lactone (IV) and jujubogenin (V).

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2.

ADAMS

Modifying the Perception of Sweetness

acid-catalyzed rearrangement. Thus, i t presently appears that a l l of our saponins are jujubogenin-based, d i f f e r i n g only i n the type, number, and linkage of sugar moieties. From our hydrolysis studies we have i d e n t i f i e d glucose, xylose, fucose, arabinose and rhamnose as the sugars of the saponins. The smallest saponin we have i s o l a t e d thus f a r i s jujubogeninglucose-rhamnose and our general approach to structure i d e n t i f i c a t i o n i n t h i s system can be i l l u s t r a t e d using t h i s molecule as an example. This substance was i s o l a t e d from the crude antisweet f r a c t i o n by s i l i c a g e l f l a s h chromatography. I t was p u r i f i e d by HPLC on a reverse-phase s i l i c a g e l column, followed by prep-scale s i l i c a g e l TLC. Acid hydrolysis gave an aglycone (shown to be ebelin lactone by NMR and mass spectrometry) and the sugars glucose and rhamnose. The i d e n t i t i e s of the sugars were v e r i f i e d by t h i n layer chromatography, gas chromatography, and by comparison with sugar standards. We have used desorption-chemica (DCI-MS) (21_) to obtain jujubogenin for t h i s molecule. DCI-MS i s an "in-beam technique i n which vaporization-ionization of r e l a t i v e l y n o n - v o l a t i l e molecules i s f a c i l i t a t e d . The sample was deposited, as an aqueous s o l u t i o n , on a deactivated rhenium wire filament (Figure 8). A f t e r evaporation of the solvent, the filament was inserted into the mass spectrometer's electron beam. The wire was then rapidly heated to effect k i n e t i c desorption of the substance. An i o n i z i n g gas (ammonia i n t h i s case) was used to f a c i l i t a t e ion production. The sample desorbed very quickly and cleanly from the wire as shown i n the reconstructed ion chromatogram i n Figure 9 ( i n s e t ) . The mass spectrum i s shown i n Figure 9. A molecular ion was not seen under the run conditions. An ion corresponding to the ammonia adduct of jujubogenin-glucose i s seen at m/z 652, which indicates that rhamnose i s the ultimate sugar; glucose i s bonded d i r e c t l y to jujubogenin. The ion for the ammonia adduct of jujubogenin appears at m/z U90. This simple experiment permitted assignment of the carbohydrate linkage order, however, i t does not say anything about the nature of carbohydrate linkages themselves. This information may be obtained by methylation a n a l y s i s or c i r c u l a r dichroism techniques, and these determinations are currently being pursued i n our laboratory. Mechanism of Action Sodium dodecyl s u l f a t e , the gymnemic acids and the z i z i p h i n s have a l l been termed "surface a c t i v e " taste modifiers because they a l l possess detergent-like properties. These molecules a l l have a polar and a non-polar end and they are capable of penetrating the phospholipid membranes that are believed to be components of sweetness receptors. Any speculation about the mechanism of a c t i o n of these substances must take i n t o account the experimental observations concerning miraculin, monellin, and thaumatin, which were presented at the beginning of t h i s a r t i c l e . Those observations suggested that transport of the modifier to the c e l l ' s i n t e r i o r was not occurring and the i n h i b i t i o n e f f e c t i s manifested at the surface of the c e l l .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

19

20

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

Figure 8 Preparing a sample f o r DCI mass spectrometry« Reproduced with permission from the Finnigan MAT Corporation. Copyright 1982, Finnigan MAT Corp. G

472

1ΘΘ.8-

jujubogenin

ebelin lactone 454

50.0 H

EL - H20 436

jujubogenin • NH4 490

634

*4* M/E

UjHillll^..fU.lfMiL 500

45Θ

550

652

600

650

Figure 9. DCI mass spectrum of jujubogenin-glucose-rhamnose. Conditions: ammonia i o n i z i n g gas at 0.25 t o r r ; filament heating rate 50 deg./sec; 0 3 sec/scan. Inset f i g u r e : reconstructed i o n chromatogram of jujubogenin-glucose-rhamnose. Reproduced w i t h o

permission

from R e f .

29.

C o p y r i g h t 1981,

Academic P r e s s ,

Inc.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2.

ADAMS

Modifying the Perception of Sweetness

The size of the molecules and the speed with which the modifying effect appears argue against a c t i o n taking place i n the i n t e r i o r of a taste c e l l . DeSimone's work on the physicochemical e f f e c t s of tastants on phospholipid membranes provides i n s i g h t i n t o possible mechanisms of modifier action. He has shown that GA and SDS are about equivalent i n surfactant strength and that d i l u t e solutions of both substances are capable of penetrating phospholipid monolayers (6^). Surfactants are known to affect c e l l function by d i s r u p t i n g the structure or s p a t i a l arrangement of membrane l i p o p r o t e i n s (22). Surface-active taste modifiers are postulated to exert t h e i r e f f e c t s by a l t e r i n g a membrane-related process that r e l a t e s i n some way to the transduction of a taste s i g n a l . This d i s r u p t i o n of c e l l s t r u c t u r a l proteins, or related phospholipid membranes, might be a t t r i b u t e d to a l t e r a t i o n s i n the surface pressure (and surface free energy) of a membrane through i n s e r t i o n of the taste modifier into the membrane (23). Additional insight i n h i b i t o r s i s provided b (2U). Two facts are known with c e r t a i n t y concerning molecules which produce taste e f f e c t s : they must be soluble i n s a l i v a and they must be able to occupy a receptor s i t e . To disrupt or i n h i b i t t a s t e , one or the other of these c h a r a c t e r i s t i c s can be i n t e r f e r e d with. Of the two, the l a t t e r i s the one most amenable to influence. From a k i n e t i c standpoint a stimulus molecule combines with a receptor to form a stimulus-receptor complex. This reaction occurs at a c e r t a i n rate. This complex may then break apart, g i v i n g back the receptor and the stimulus molecule; t h i s occurs at another rate. The formation of the stimulus-receptor complex eventually r e s u l t s i n the f i r i n g of a nerve and the production of a taste sensation. The rates of formation and breakdown of the complex are influenced by temperature. B i r c h has found that by increasing the temperature of tastant s o l u t i o n s , the perceived i n t e n s i t i e s of the solutions increase and the response-time p r o f i l e changes as w e l l (Figure 10). For a given temperature and sucrose concentration, taste sensation plateaus at a c e r t a i n time, suggesting that the sweetness receptors have become saturated ( i . e . that the rate of stimulus-receptor complex formation and breakdown are equal). A l i n e can be drawn on the graph from the beginning point of stimulus perception to the plateau point and the slope of a l i n e so drawn i s c a l l e d the magnitude estimation rate (MER). A plot of the r e c i p r o c a l s of MER values versus corresponding r e c i p r o c a l sucrose concentrations can be made; one obtains a Lineweaver-Burk type of p l o t (Figure 11). This plot indicates a low a f f i n i t y of sugar f o r receptor and a f f i n i t y values obtained by t h i s method agree generally with those of Cagan (25), which were determined by a r a d i o a c t i v e l y - l a b e l l e d sugar binding assay. In s i m i l a r fashion, a Lineweaver-Burk type plot for a gymnemic acid i n h i b i t e d tongue responding to sucrose solutions can be constructed. The appearance of t h i s p l o t i s quite d i f f e r e n t from that of Figure 12 and i s not c h a r a c t e r i s t i c f o r e i t h e r competetive or non-competitive i n h i b i t i o n . The r e s u l t s of the analysis seem to i n d i c a t e that i n h i b i t i o n of sweetness by gymnemic acid i s of a mixed k i n e t i c type which might r e s u l t from two i n h i b i t i o n mechanisms operating simultaneously. The i m p l i c a t i o n of t h i s i s that i n h i b i t i o n i s not immediately r e v e r s i b l e by increasing the

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

21

22

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

20° C 30° C 40° C

Figure 10. Time-intensity p l o t f o r sugar sweetness. Magnitude estimation (ME) Reproduced with permission from Ref. 29. 0

Copyright

1981,

Academic P r e s s , I n c .

F i g u r e 11. R e c i p r o c a l p l o t of MER versus c o n c e n t r a t i o n of sweet s t i m u l u s . Reproduced w i t h p e r m i s s i o n from Ref. 29. Copyright 1981, Academic P r e s s , I n c .

Figure 12. Structure of methyl 4,6-dichloro-4,6-dideoxy-a-Dgalactopyranoside (VI), D i C l - g a l .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2.

ADAMS

Modifying the Perception of Sweetness

concentration of sugar on the tongue. Thus, B i r c h postulates that sweetness suppression by gymnemic acid may be the r e s u l t of the i n h i b i t o r binding to the receptor for a period of time ( f o r up to an hour or so) during which time access of sugar molecules to the receptor i s blocked. Also, turnover of i n h i b i t o r i s much slower at the receptor than i s the turnover of sugar molecules. Therefore, a f t e r a receptor s i t e i s freed of bound gymnemic acid the sugar molecules must then compete with the proximate i n h i b i t o r for access and binding. A very powerful but s h o r t - l i v e d i n h i b i t i o n of sweet taste response i n the g e r b i l i s caused by methyl 4,6-dichloro-U,6dideoxy- α-D-galactopyranoside ( D i C l - g a l ) , VI, (Figure 12) a sugar d e r i v a t i v e synthesized and studied by Jakinovich (26). D i C l - g a l i s a potent sweetness i n h i b i t o r , but only when mixed with sucrose. I t has no long term i n h i b i t i n g e f f e c t on the sweet taste response. Also, pretreatment of the tongue with D i C l - g a l has no e f f e c t on subsequent response to sucros behaviors of SDS, GA an substance i s a competetive i n h i b i t o r , but experiments with a r t i f i c i a l sweetners suggest that the mechanism of i n h i b i t i o n i s possibly more complex than t h i s . Prospects for Future Research There i s a growing body of knowledge r e l a t i n g structure to function for sweet t a s t e . In s p i t e of t h i s , many questions remain unanswered with regard to the physical nature of the sweetness receptor, including the question of whether i t i s indeed a s i n g l e receptor at a s i n g l e s i t e . Once the structures of several d i f f e r e n t i n h i b i t o r s are known i n d e t a i l i t should be possible to use them as probes of the receptors, to a s c e r t a i n the physical l i m i t s and s t e r i c - e l e c t r o s t a t i c requirements of the system. By systematically varying the structure of i n h i b i t o r s and measuring the psychophysical e f f e c t s , s t r u c t u r e - a c t i v i t y c o r r e l a t i o n s can be constructed. These data, when taken together, may then allow construction of a s e l f - c o n s i s t e n t picture of the sweetness receptor. This, then, w i l l allow us to better understand the taste transduction process. F i n a l l y , i t i s i n t e r e s t i n g to speculate on what function sweetness i n h i b i t o r s such as the gymnemic acids and the z i z i p h i n s might perform for the plants i n which they occur. One thought i s that they might serve as feeding deterrents for predatory i n s e c t s . I t has been shown that gymnemic acids act as feeding deterrents to the Southern army worn (Prodenia eridana). The e f f e c t i s demonstrable with sugar-free d i e t s , thus the gymnemic acids do not appear to act on the sweet taste i n the i n s e c t , as they do i n humans (27,, 28). The sweetness i n h i b i t o r s may also function as defenses against mammalian herbivores: by d u l l i n g the sense of sweetness, the i n t r i n s i c b i t t e r n e s s of a plant might be accentuated, thus encouraging the browser to find t a s t i e r fare. Acknowledgments Much of the z i z i p h i n chemistry was carried out by Dr. Frank Koehn, and psychophysical studies were performed by Drs. C l a i r e Murphy and

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

23

24

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Carol Christensen. This work was p a r t i a l l y supported by NIH grant 5T32NS07176-05 (postdoctoral traineeship t o F.E.K.) and by the Ambrose Monell Foundation. Literature Cited 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.

Kurihara, K.; Beidler, L. M.; Science; 1968, 161, 1241-1243. Brower, J. N.; van der Wei, H.; Franke, Α.; Henning, G. J.; Nature (London); 1968, 220, 373-374. Kurihara, K.; Beidler, L. M.; Nature (London); 1969, 222, 1176-1179. van der Wei, H.; In "Olfaction and Taste, IV"; Wissenshaftliche Verlagsgesellschaft MBH, Stuttgart; 1972, 226-233. Cagan, R. H.; Science; 1973, 181, 32-35. DeSimone, J. Α.; Heck, G. L.; Bartoshuk, L. M.; Chem. Senses; 1980, 5, 317-330. Yackzan, K. S.; Ala. (15). Subba Rao, G.; Sinsheimer, J. E.; Chem. Commun.; 1968, 16811682. Stocklin, W.; Ag. and Fd. Chem.; 1969, 17, 704-708. Sinsheimer, J. E.; Subba Rao, G.; Mcllhenny, H. M.; J. Pharm. Sci.; 1970, 59, 622-628. Subba Rao, G.; Sinsheimer, J. E.; J. Pharm. Sci.; 1971, 60, 190-193. Pfaffmann, C.; In "Handbook of Physiology"; American Physiology Society, Washington, D. C., 1959; Section 1, Vol. 1, p. 507. Dateo, G. P.; Long, L.; Agric. Fd. Chem.; 1973, 21, 899-903. Stocklin, W.; Helv. Chim. Acta; 1967, 50, 491. Bartoshuk, L. M.; Dateo, G. P.; Vandenbelt, D. J.; Buttrick, R. L.; Long, L.; "Olfaction and Taste, III"; Pfaffmann, C., Ed.; Rockefeller University Press, New York; 1969, pp. 436444. Meiselman, H. L.; Halpern, B. P.; Dateo, G. P.; Physiol, and Behav.; 1976, 17, 313-317. Kennedy, L. M.; Halpern, B. P.; Chem. Senses; 1980, 5, 123147. Still, W. C.; Kahn, M.; Mitra, Α.; J. Org. Chem.; 1978, 43, 2923-2925. Ogihara, Y.; Inoue, O.; Otsuka, H.; Kawai, K.-I.; Tanimura, T.; Shibata, S.; J. Chromatogr.; 1976, 128, 218-223. Kawai, K.-I.; Akiyama, T.; Ogihara, Y.; Shibata, S.; Phytochem.; 1974, 13, 2829-2832. See: Cotter, R. J.; Analyt. Chem.; 1980, 52, 1589A-1606A, for review. Kagawa, Y.; In "Methods in Membrane Biology"; Korn, E. D., Ed.; Plenum Press, New York, 1974; Vol. 1, pp. 201-269. Quoted in (23), p. 215. DeSimone, J. Α.; In "Biochemistry of Taste and Olfaction"; Cagan, R. H. and Rare, M. R., Eds.; Academic Press, New York, 1981, pp. 213-229. Ray, Α.; Birch, G. G.; Life Sciences; 1981, 28, 2773-2781. Cagan, R. H.; Biochim. Biophys. Acta; 1971, 252, 199-206.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

2. ADAMS

26. 27. 28. 29.

Modifying the Perception of Sweetness

Jakinovich, W.; Science; 1983, 219, 408-410. Granich, M. S.; Halpern, B. P.; Eisner, T . ; J. Insect Physiol.; 1974, 20, 435-439. Harborne, J. B . ; "Introduction to Ecological Biochemistry"; Academic Press, New York, 1977; pp. 149-150. Birch, G. G . ; In "Biochemistry of Taste and Olfaction"; Cagan, R. H.; Kare, M. R . , E d s . ; Academic Press, New York, 1981; pp. 163-73.

RECEIVED August 14, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3 Sensory Responses to Oral Chemical Heat 1

2

Harry Lawless and Marianne Gillette 1

S. C. Johnson & Son, Inc., Racine, WI 53403 Research and Development, McCormick & Company, Inc., Hunt Valley, M D 21031

2

Two areas of research, psychophysics and sensory evaluation, have made recent contributions to the understanding of l sensation f heat derived fro peppers. Psychophysica observer's response compounds, focussing on such aspects as time-intensity functions, areas of oral stimulation, correlation with evoked salivary flow, interactions with basic tastes, and effects of sequential stimulation. Sensory evaluation of the heat level of ground red pepper has recently been advanced by the validation of a new method which solves many of the problems inherent in the previous Scoville procedure. The new method is based on anchored graphic rating by panels who are trained with physical reference standards. The procedure has shown excellent r e l i a b i l i t y , fine discriminations among samples, and high correlations with instrumental determinations of capsaicinoid content of pepper samples.

Chemically-induced Oral Heat as Part of Flavor Many v a r i e t i e s of red pepper, derived from plants of the genus Capsicum, are used i n d i f f e r e n t cuisines around the world f o r t h e i r sensory properties of o r a l chemical "heat", v o l a t i l e f l a v o r and color. Determination of the degree of heat i n a pepper sample has been a d i f f i c u l t problem f o r both sensory and instrumental analysts of f l a v o r . Furthermore, the l i t e r a t u r e concerning the sensory physiology and perceptual responses of the "common chemical sense" (as defined later) has lagged behind other areas of study of the chemical senses. The purpose of t h i s paper w i l l be to review recent developments i n two areas, the development of a standard method f o r sensory analysis of ground red pepper heat and the psychophysical characterization of observers' responses to o r a l chemical i r r i t a t i o n induced by spice-derived compounds. 0097-6156/ 85/ 0289-0026$06.00/ 0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3.

LAWLESS A N D GILLETTE

Sensory Responses to Oral Chemical Heat

P h y s i o l o g i c a l l y , the senses responsible for our perception of f l a v o r can be divided into three anatomical systems. In the o r a l cavity, the c l a s s i c a l gustatory pathways through the tongue and soft palate are responsible f o r our s e n s i t i v i t y to the four basic tastes, sweet, sour, s a l t y , and b i t t e r . In the nasal passages, the olfactory receptors provide s e n s i t i v i t y to a wide v a r i e t y of v o l a t i l e compounds, producing the sensations we normally assign to smell. In addition to these two systems, the trigeminal nerves i n both the o r a l and nasal c a v i t i e s provide s e n s i t i v i t y to thermal, t a c t i l e , i r r i t a t i o n and pain sensations(1). The trigeminal innervation i s also chemically s e n s i t i v e to compounds which are pungent, astringent or i r r i t a t i v e , and hence provide an important part of our appreciation of flavor as a whole. This paper i s concerned with the character of t h i s system, sometimes referred to as the common chemical sense. The results presented here focus on o r a l chemical "heat", as characterized by the sensations (warm to painful) e l i c i t e d by re from the more nasal sensatio naso - pharyngeal i r r i t a t i o n induced by such substances as ammonia or freshly ground horseradish. While many flavor compounds have i r r i t a t i v e or astringent properties (2), most have gustatory or o l f a c t o r y properties as well. However, several families of compounds, from three d i f f e r e n t spices, are potent stimuli of oral heat sensations, and i n t h e i r chemically pure forms are nearly devoid of side tastes and odors. Representative members of these families are shown i n Figure 1. Capsaicin i s an example of the heat p r i n c i p l e s derived from red pepper. Other structures i n the capsaicinoid family vary i n t h e i r heat l e v e l , depending upon the saturation of the double bond (a) and the side chain length (3). N-vanillyl-n-nonamide, "synthetic capsaicin", i s a r e a d i l y available and e a s i l y synthesized compound used as a chemical heat standard (4, see also below), i n which the branched side chain of capsaicin i s replaced with a saturated straight chain. Piperine i s derived from black pepper, and has three isomers depending upon the c i s - or trans- configuration of the two double bonds (b and c ) . Gingerol i s a pungent compound from ginger, with related compounds varying i n pungency depending upon dehydration at (d) and upon chain length. Two d i s c i p l i n e s have recently brought resources to bear upon the sensory characterization of these compounds. In sensory psychology, the techniques of psychophysics have been used to characterize responses to various o r a l chemical i r r i t a n t s . Also, the f i e l d of applied sensory evaluation has addressed the issue of determining the heat value of various unknown samples using human observers. This second area of work has largely been driven by a need to replace the widespread procedure known as the " S c o v i l l e determination" f o r measuring sensory heat with a more r e l i a b l e method. However, the orientations of the two d i s c i p l i n e s are d i s t i n c t l y d i f f e r e n t . A psychophysical study w i l l focus on determining c h a r a c t e r i s t i c s of the observer, tend to use naive untrained subjects, simple stimuli and ask f o r r e l a t i v e l y simple judgements (responses). The goal i n such studies i s to uncover fundamental attributes of sensory function, such as observer s e n s i t i v i t y , temporal and s p a t i a l properties of sensation, or dose-response curves (psychophysical functions). Sensory

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

27

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

CAPSAICIN

GINGEROL

PIPERINE

Figure 1. Structures of f l a v o r compounds inducing o r a l i r r i t a t i o n or "heat".

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3.

LAWLESS AND GILLETTE

Sensory Responses to Oral Chemical Heat

29

evaluation, on the other hand, i s oriented towards uncovering sensory attributes of the stimulus or product. This w i l l usually be done with trained, experienced or otherwise " c a l i b r a t e d " observers who are used i n order to elucidate the sensory properties of unknown or uncharacterized s t i m u l i . However, since the act of sensation i s an interaction of an observer with a stimulus, the two approaches often end up using s i m i l a r methods (e.g., scaling) to provide s i m i l a r information (e.g., observer response as a function of stimulus concentration). Because of t h i s p a r a l l e l orientation i n methods, and because of the concurrent advances recently made i n both areas, we have integrated results from the two f i e l d s i n t h i s paper. The f i r s t section w i l l focus on psychophysical characterization of o r a l chemical i r r i t a n t s . The f i n a l section w i l l discuss the development of a new sensory method f o r evaluation of ground red pepper heat. Psychophysical Characterizatio One important a t t r i b u t e of o r a l chemical i r r i t a t i o n i s the long time-course of the sensations e l i c i t e d , both i n onset of sensation after tasting the stimulus, and i n the lingering duration of the sensation a f t e r expectoration. In the results below, subjects were given emulsions of various spice-derived compounds (or mixtures) to taste. The sample was swirled vigorously around the mouth f o r 30 seconds, expectorated, and then various ratings were asked of the subject at fixed i n t e r v a l s . The method of magnitude estimation was used, i n which subjects assign numbers to sensations i n proportion to the sensation e l i c i t e d by reference s t i m u l i t r i e d e a r l i e r (usually NaCl s o l u t i o n s ) . Figures 2 and 3 show data f o r the time-course of sensations e l i c i t e d by various i r r i t a t i v e compounds (5). To a f i r s t approximation, these functions can be characterized by the following r e l a t i o n s h i p : S = k C

n

e-T/m

where η i s a power function exponent characterizing the growth of sensation magnitude (S) with concentration (C), and m i s a constant characterizing the rate of decay of sensation over time (T) (k i s a proportionality constant and e the base of the natural logarithm). These operating c h a r a c t e r i s t i c s , power function exponents and decay constants, can provide reference points f o r comparing compounds. In these studies, piperine tended to have higher exponents (faster growth of sensation with concentration) than n-vanillyl-n-nonamide. I t i s also worth noting that the majority of o l f a c t o r y and taste compounds have exponents less than or approaching 1.0, while p a i n f u l stimuli (e.g., e l e c t r i c shock) tend to have much higher exponents. These i r r i t a n t compounds f a l l more i n the range of flavor compounds than pain s t i m u l i (5), with exponents less than or about equal to 1.0. Another c h a r a c t e r i s t i c of i r r i t a t i v e stimulation of the trigeminal nerve i s the defensive reflexes (e.g., sneezing) invoked by the body to remove or d i l u t e the offending substance. In the case of o r a l chemical heat, the burning sensation from capsaicin invokes sweating, tearing, and copious s a l i v a t i o n . Salivary flow

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

τ

1

1

1

1

1 Ο

1 4.0

1

1

μ

~Τ

PPM

V ANILLYL NON AMIDE

0 1

2

3

4

Minutes a f t e r

5

6

7

8

10

Expectoration

Figure 2. Median perceived o r a l i r r i t a t i o n from four concentrations of v a n i l l y l nonamide, piperine and capsicum oleoresin over time. Reproduced with permission from Ref. 5, copyright 1984, IRL Press Limited.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

LAWLESS A N D GILLETTE

Sensory Responses to Oral Chemical Heat

Minutes

after

Expectoration

Figure 3. Median perceived o r a l i r r i t a t i o n from four concentrations of v a n i l l y l nonamide, piperine and ginger oleoresin over time. Reproduced with permission from Ref. copyright 1984, IRL Press Limited. In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

32

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

rate c l o s e l y p a r a l l e l s the subjective ratings of sensation i n t e n s i t y . Whole-mouth s a l i v a r y flow was c o l l e c t e d during the i n i t i a l two minutes of o r a l bum from the subjects who tasted the twelve compounds i n Figure 2. The c o r r e l a t i o n of mean s a l i v a r y flow rate with mean peak sensation i n t e n s i t y was .91. The "subjective" ratings, i n t h i s case, f i n d a source of v a l i d a t i o n since a high c o r r e l a t i o n exists with an "objectively" measured physiological response. In addition to the pattern of stimulation over time, these compounds also show s i m i l a r i t i e s and differences i n s p a t i a l patterns of stimulation. After a whole-mouth rinse, subjects were asked to report the areas of the mouth i n which they perceived the heat. Not s u r p r i s i n g l y , the number of o r a l areas that subjects reported systematically increased with concentration, and decreased over time as the burn faded i n i n t e n s i t y (5). The patterns of perceived stimulation across the mouth also d i f f e r e d among compounds. The red peppe showed pronounced anterio tongue). The ginger compounds, on the other hand, also showed some posterior stimulation, with many subjects reporting a b i t i n g sensation on the s o f t palate and throat. These data suggest that compounds may d i f f e r q u a l i t a t i v e l y i n the areas they most e f f i c i e n t l y stimulate. One further aspect i n which i r r i t a t i v e compounds d i f f e r i s t h e i r interaction with the tastes, sweet, sour, s a l t y and b i t t e r . A recent perceptual study sought to gain insights on the old adage that too much pepper makes i t hard to taste your food. In t h i s study, s t i m u l i representing the c l a s s i c a l four basic tastes were given a f t e r rinses with capsicum oleoresin (an extract of red pepper) or piperine. Psychophysical ratings of the perceived i n t e n s i t y of the tastants showed that under conditions of intense o r a l i r r i t a t i o n , there was some p a r t i a l i n h i b i t i o n of the taste sensations (6). In addition, the pattern of i n h i b i t i o n d i f f e r e d for the two i r r i t a n t s . The pungency of capsicum worked mostly against sour and b i t t e r tastes, and l e f t s a l t i n e s s i n t a c t , while the e f f e c t s of piperine were more broad, influencing a l l four taste q u a l i t i e s . These i n h i b i t o r y e f f e c t s of i r r i t a t i o n on taste are p a r a l l e l to s i m i l a r e f f e c t s of nasal i r r i t a t i o n on odors. For example, C O 2 , a potent nasal i r r i t a n t , w i l l p a r t i a l l y mask odors which are presented simultaneously (8). Whether o r a l bum can influence odor or aroma perception i s at t h i s time unknown. Since a series of several s t i m u l i are usually presented to subjects i n psychophysical studies, the opportunity arises to study e f f e c t s of sequential stimulation. These e f f e c t s can have p r a c t i c a l implications such as the number of s t i m u l i which may be sampled i n applied sensory evaluation without fatigue. During the time-intensity ratings which produced the data i n Figures 2 and 3, s t i m u l i were presented i n d i f f e r e n t orders, with the f i r s t and t h i r d concentrations i n the series of each compound presented either before or a f t e r the stimulus of next higher concentration. As shown i n Figure 4, s t i m u l i were usually judged to be much weaker when presented a f t e r a stronger compound. This suggests that there i s a sequential desensitizing e f f e c t during tasting and that experimenters should be careful concerning the number of s t i m u l i that may be reasonably given i n one s i t t i n g . This desensitization

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3.

LAWLESS AND GILLETTE

Sensory Responses to Oral Chemical Heat

p a r a l l e l s the e f f e c t s of capsaicin observed i n the pharmacological l i t e r a t u r e (8,9), where systemic or t o p i c a l administration renders animals i n s e n s i t i v e to chemical i r r i t a t i o n . Whether or not people became desensitized during long-term dietary intake of pepper i s unclear. One study examining thresholds f a i l e d to see a difference between c h i l i consumers and non-consumers (10). However, lack of e f f e c t s at threshold may not r e f l e c t above-threshold changes i n responsiveness. However, short-term d e s e n s i t i z a t i o n may depend upon the s p a t i a l c h a r a c t e r i s t i c s of stimulation. When stimuli are constantly refreshed, and stimulation i s limited to a small area, a pattern of ever-increasing sensation buildup i s observed. Figure 5 shows a flow chamber ("geschmackeslupe" a f t e r Hahn (11)) used to d e l i v e r constant controlled stimulation. When v a n i l l y l nonamide was flowed over a subject's tongue (2 ml/sec, 0.8 cm area, 8 subjects), the sensation continued to grow as shown i n Figure 6, about doubling during a psychophysical characterizatio that influence these sensations and t h e i r s p a t i a l and temporal interactions. 2

Sensory Evaluation of Pepper Heat Commercially, ground red peppers are purchased, sold, blended, and used based upon t h e i r sensory heat l e v e l s . Generally, the higher the heat, the higher the p r i c e . In order to produce a consistent product the heat l e v e l of capsicum products are monitored by sensory and chemical/instrumental methods (12-15). H i s t o r i c a l l y , the only sensory method f o r the assessment of heat i n red pepper has been the S c o v i l l e Heat Test (16, 17). While t h i s method o r i g i n a l l y f i l l e d the need f o r a means of measuring and expressing heat i n red pepper products, i t has become u n i v e r s a l l y c r i t i c i z e d f o r i t s lack of accuracy and p r e c i s i o n (4, 15, 18-20). S p e c i f i c problems noted with the S c o v i l l e Heat Test are: b u i l d up of heat, rapid taste fatigue and increased taste threshold as a r e s u l t of the 5 samples required f o r tasting, ethanol b i t e i n t e r f e r i n g with capsicum heat, lack of s t a t i s t i c a l v a l i d i t y , lack of reference standards, the 16 hour extraction time, the error of central tendency (tendency to pick the middle concentration of the series) and poor precision. The development of new instrumental methods to replace the troubled S c o v i l l e procedure necessitated the design of an improved sensory method f o r v a l i d a t i o n of instrumental precision. A new sensory method (4) was designed which offered the following procedural improvements over the S c o v i l l e Method: 1) a 20 minute aqueous extraction, (2) no ethanol used, 3) reference standard included i n each test, 4) trained panelists, 5) timed t a s t i n g , r i n s i n g and recess, 6) one d i l u t i o n f o r a l l samples, and 7) use of a graphic l i n e scale to score the heat sensation. To evaluate the aqueous extraction procedure, ground red pepper was extracted with spring or d i s t i l l e d water, at 20°C, 75°C, and 90°C, with or without 20-2000 mg/L Polysorbate - 80 or Polysorbate - 60. The aqueous extractions were compared to 5 hour ethanol extractions. Residues from a l l extractions were also evaluated f o r residual heat. A 20 minute simmering extraction of

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

33

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

30-r

47. 1 45--

40--

1

1 After

Weaker

After

Stronger

Stimulus Stimulus

Μ "ι e

0.978 ppm

0 * 5 ppm

1.3 ppm

2 . 0 ppm

Stimulus

Figure 4. Mean perceived intensity (at the peak of the time-intensity function) of v a n i l l y l nonamide i n d i f f e r e n t presentation orders.

I

/

Ο Figure 5. Geschmackeslupe f o r presenting flowing stimuli to o r a l surfaces. Scale = 1 cm. I - inflow port, 0 - outflow port, Τ - port f o r stimulation, approximately 1 cm i n diameter. Inflow and outflow tubes i n the b a r r e l are concentric.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985. m

Figure 6. Mean (+ 1 S.E. ) perceived intensity of 2 ppm v a n i l l y l nonamide during presentation through Geschmakeslupe, over seven minutes.

Time

^

36

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

ground red pepper i n 90°C spring or d i s t i l l e d water, with 200 ppm polysorbate -60 or -80 was found to optimize extraction of the sensory heat components. A panel of 12 highly experienced tasters was used to develop the new method. Pepper heat was rated on a 15 cm l i n e scale anchored at 0 (no heat), 1.25 cm (threshold heat), 5.0 cm ( s l i g h t heat), 10.0 cm (moderate heat), and 15 cm (strong heat). The panel selected standard concentrations of N-vanillyl-n-nonamide to be used f o r c a l i b r a t i o n and t r a i n i n g of future panelists. A reference control (0.44 ppm N-vanillyl-n-nonamide) was selected f o r use an an i n t e r n a l standard f o r " s l i g h t " during each sensory t e s t . Red l i g h t s were used to eliminate possible influences of v a r i a t i o n i n the color of the products. Two samples were evaluated per test, the known control and a test pepper extract. The control, coded "C" was served f i r s t , followed by an unknown test sample i d e n t i f i e d with a random double l e t t e r code. A l l samples were presented as 10 ml portions i n p l a s t i c medicin samples using the followin 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.

Cleanse palate before f i r s t sample (control) with unsalted cracker and 20°C spring water. Take entire f i r s t sample (control) i n mouth, hold f o r about 5 seconds, swallow slowly. Wait 30 seconds (timed). Rate f i r s t sample at " s l i g h t " on b a l l o t . Cleanse palate with unsalted cracker and 20°C spring water for 60 seconds (timed). Rinse with 20°C spring water immediately p r i o r to second sample. Take entire second sample (test sample) i n mouth f o r about 5 seconds, swallow slowly. Wait 30 seconds (timed). Rate second sample. Panel dismissed i f only one sample i s to be evaluated. I f two samples are to be evaluated: Wait 5.0 minutes. Clease palate well with water and crackers during t h i s time. Repeat steps 1 through 9 f o r the second set of samples.

Panelists placed a mark on the scale expressing t h e i r impression of the heat i n the test sample. Sensory Heat Ratings were obtained by measuring the distance i n cm from the "0" mark to the panelist's rating f o r each sample. The mean of a l l panelist's ratings f o r each sample represents i t s sensory heat rating. To evaluate the c o r r e l a t i o n of sensory responses with High Pressure Liquid Chromatography capsaicinoid quantitation, samples from 60 l o t s of ground red pepper were selected to represent the normal range of S c o v i l l e Heat Units found i n red pepper. These 60 peppers were analyzed instrumentally f o r the 3 major capsaicinoid analogs (nordihydrocapsaicin, capsaicin and dihydrocapsaicin), f o r 8 additional physical/chemical parameters (water a c t i v i t y , moisture, color, surface area and p a r t i c l e s i z e ) , and s e n s o r i a l l y by the S c o v i l l e heat tests. The 60 peppers were also assessed using the new sensory method f o r heat ratings. A l l possible single

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3.

LAWLESS AND GILLETTE

Sensory Responses to Oral Chemical Heat

37

and multiple regressions were performed i n order to determine the optimal instrumental alternative f o r the new sensory method, as well as to further substantiate the p r e c i s i o n of the new sensory method (4). Several very strong relationships (r > 0.90) were found between the instrumental and sensory measurements (Table 1). The strong c o r r e l a t i o n (r=0.94) between the t o t a l concentration of c a p s a i c i n o i d s and sensory heat ( T a b l e I ) , and the r e l a t i v e ease of measuring such, lead to the selection of t h i s chemical parameter as the most desirable predictor of the sensory heat. The p r e d i c t i v e strength of the equation r e l a t i n g percent capsaicinoids with sensory heat ratings was then tested using 20 red peppers. The mean average deviation between predicted and actual sensory heat ratings was less then 1 cm on the 15 cm scale (4). Thus, the equation "sensory heat rating = 31.26 (percent capsaicinoids) 0.21** p r e c i s e l y predicts sensory ratings f o r heat i n ground red pepper. Lesser p r e d i c t i v e relationship t r a d i t i o n a l l y determine heat ratings (r=0.40) or the percent capsaicinoids (r=0.48). Based upon the sensory heat values f o r i n d i v i d u a l capsaicinoids as determined by Todd, et. a l . (15), S c o v i l l e - type heat units were calculated f o r each red pepper based upon the capsaicinoid content of the peppers (13). Linear regression demonstrated the relationship between these calculated S c o v i l l e heat units and the new sensory heat ratings to be very good (r=0.94) (Figure 7). Therefore, both the HPLC and the new sensory method can provide output translated into S c o v i l l e units f o r universal understanding. The accuracy of the new sensory method was confirmed using a set of 15 " a r t i f i c i a l * * red peppers of known oleoresin capsicum content (Figure 8). The c o r r e l a t i o n of sensory heat rating with the percent oleocapsicum was 0.94. The p r e c i s i o n of the new method was demonstrated by collaborative study within 13 laboratories (22), and by r e p e a t e d t e s t i n g w i t h i n one l a b (4) ( T a b l e s I I and I I I ) . Furthermore, the new sensory method avoids several problems inherent i n the S c o v i l l e procedure. Heat b u i l d up, fatigue, and increased threshold are minimized by use of a standardized i n i t i a l sample, as well as timed r i n s i n g between samples. Ethanol b i t e i s avoided by use of an aqueous extraction. The panel data may be manipulated s t a t i s t i c a l l y due to the l i n e a r i t y of the scale and the number of panelists. Reference standards are included. Extraction time i s reduced from 16 hours to 20 minutes. Reproducibility of r e s u l t s has been demonstrated. The error of central tendency i s avoided by not having a "middle** sample. The new method i s more comparable to normal food usage as i t i s an aqueous rather than ethanol extraction. This method i s currently being used f o r routine laboratory analysis of red pepper heat. Results have been consistent and continue to correlate well with HPLC data. A s i m i l a r procedure has also been used f o r sensory evaluation of black pepper heat. The American Society f o r Testing and Materials (ASTM, Committee E-18) has conducted a collaborative study testing the new method i n comparison to the S c o v i l l e Method. ASTM E-18 i s currently preparing to document i t as a standardized test method. Also, a modification of the method i s being prepared f o r oleoresin capsicum and f o r low-heat capsicums.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

3

c

d

b

9

Scoville Heat U n i t s

1

30.38 (% Capsaicinoids) + 0.522 (Aw) -

Sensory =

5.5 11.1**

0.476 0.475

Sensory

78.3** 11.1**

214.9**

280.5**

1

9

f

d

c

b

31.26 (% Capsaicinoids) -

1.99 (Calculated Scoville) 45.54 (% Capsaicin) + 0.256

Sensory = Sensory = Sensory =

1.45 ( S H U ) + 0.303

1.61

- 0 . 1 4 9 (ΔΕ) + 8.39

Calculated Scoville = 0.683 (SHU) + 4684.67 Sensory = 1.45 (SHU) + 0.303

Sensory =

Sensory = 4 6 8 . 3 2 (% Nordihydrocapsaicin)

0.223

0.214

31.02 (% Capsaicinoids) + 10.272 (Moisture) -

Sensory =

2.12

3 4 . 6 6 (% Capsaicinoids) - 0.364 (b) + 7.07

Sensory =

151.1** 284.1**

34.18 (% Capsaicoinoids) - 0.216 (ΔΕ) + 10.24

Sensory =

188.8** 159.0**

Equation of line 212.6**

Calculated Scoville Heat Units

0.475

0.899

0.938 0.921

0.939

0.946 0.943

0.959 0.954

Total capsaicinoids as determined by HPLC (Hoffman et al., 1983). Change In total color, determined by Hunter Colorimeter, Model D25M-9. Color value, Hunter Colorimeter, Model D25M-9. Water activity determined by Beckman Water Activity Hygrometer. J Moisture determined by Azeotropic Distillation. Calculated using method described by Todd et al. (1977). Capsaicin as determined by HPLC (Hoffman et al., 1983). J Nordihydrocapsaicin as determined by HPLC (Hoffman et al., 1983). ASTA method 21.0; Scoville Heat Test. * * Statistically significant at 99% level of confidence.

1

f

% Nordihydrocapsaicin" a n d Δ Ε

% Capsaicin

Scoville Heat Units vs:

a

6

Calaulated Scoville Heat U n i t s

w

Sensory vs: % C a p s a i c i n o i d s a n d &E % Capsaicinoids a n d b % Capsaicinoids and a % Capsaicinoids a n d M o i s t u r e % Capsaicinoids

Variables

Table I . R e s u l t s o f R e g r e s s i o n A n a l y s e s on Sensory Heat Versus S e v e r a l A n a l y t i c a l Measurements of 40 Ground Red Peppers

LAWLESS A N D G I L L E T T E

Sensory Responses to Oral Chemical Heat

Figure 7. Sensory test heat ratings versus calculated S c o v i l l e Heat units ( i n thousands) f o r 60 red peppers. S c o v i l l e Heat units calculated based upon the method of Todd, e t . a l . (15). Reproduced with permission from Ref. 4, copyright 1984, I n s t i t u t e of Food Technologists.

PERCENT O L E O CAPSICUM

Figure 8. Sensory heat ratings versus concentration of oleoresin capsicum on paprika f o r a set of 15 a r t i f i c i a l red peppers. Reproduced with permission from Ref. 4, copyright 1984, I n s t i t u t e of Food Technologists.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

40

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Table I I .

Sensory Heat R a t i n g s f o r B l i n d Samples of Red P e p p e r

Duplicate

a

Pepper

A Β C D Ε F G H I J a

Sensory heat rating

Sensory heat rating

Session 1

Session 2

3.3 3.5 5.7 5.9 6.6 8.5 8.5 8.6

2.7 4.6 5.9 5.6 4.8 9.8 8.6' 8.8

Each sample evaluated at 2 different sessions by the same panel (n = 10). No pair of duplicate samples is different when analyzed by a parted t test.

CONCLUSIONS 1.

Sensations of o r a l chemical heat, as induced by pepper-derived compounds are amenable to psychophysical investigation and sensory evaluation by supra-threshold scaling techniques such as magnitude estimation and anchored graphic rating scales.

2.

Different heat-inducing compounds may be characterized by d i f f e r e n t pschophysical functions, perceived areas of o r a l stimulation, and interactions with taste sensations.

3.

Repeated sampling of intense o r a l heat stimuli may r e s u l t i n short-term desensitization. However, t h i s e f f e c t may depend upon the s p a t i a l and temporal parameters of stimulation, since constant stimulation of small areas of the o r a l epithelium leads to sensation growth, rather than desensitization.

4.

A new rating method f o r evaluation of the sensory heat of ground red pepper samples shows important advantages over the t r a d i t i o n a l S c o v i l l e method, e s p e c i a l l y i n the areas of accuracy, r e l i a b i l i t y and ease of administration.

5.

The new sensory method shows excellent c o r r e l a t i o n with instrumental determination of capsaicinoid content of red pepper samples, and can be converted to S c o v i l l e units f o r universal understanding.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

III-

4.0 ± 2 . 3 5.1 ± 3.1 9.4 ± 2 . 1 7.8 ± 2 . 4 9.1 ± 1.7

0.6 ± 0.6

(η = 10)

Panel Β

Lab #1

4.7 ± 2.3 3.4 ± 2.0 8.3 ± 1 . 7 10.0 ± 1 . 7 11.3±0.9

0.4 + 0.5

(η = 9)

Panel C

Study

4.8+ 1.3 6.0 ± 2.0 4.4 ± 1 . 7 9.7 ± 2.1 11.9 ± 1.7

0.5 + 0.8

(η = 9-11)

Panel D

Lab #2

6.1 ± 1.4 6.5 ± 1.4 7.5 ± 1.4 11.0 ± 1.4 12.0 ± 1.4

1.0+ 1.1

(η = 4-5)

4.5 5.3 7.6 9.7 11.4

0.6

(η = 5)

Χ

Lab

Inter-

1.1 1.2 1.9 1.2 1.4

0.23

(η = 5)

σ

Lab

Inter-

S e n s o r y Heat Method'

Panel Ε

Lab #3

on Red Pepper

Sensory heat ratings

R e s u l t s of C o l l a b o r a t i v e

0.22 0.22 0.25 0.12 0.12

0.38

(η = 5)

σ/Χ

Lab

Inter-

Means and standard deviations for 6 ground red peppers tested in 3 labs by 5 separate panels. Laboratory Means, Standard Deviations and Coefficients of Variations. Coefficient of Variation = σ / Χ ; an approximation-of method inter-laboratory precision.

0.6 + 0.6

3.0 ± 1 . 7 5.4 ± 2 . 1 8.4 ± 2 . 5 10.0 ± 1 . 4 12.6 ± 1 . 7

1

2 3 4 5 6

3

(η = 10)

Pepper

Panel A

Table

42

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Literature Cited 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

Silver, W. L.; Maruniak, J. Α.; Chem. Senses 1981, 6, 295-305. "Fennaroli's Handbook of Flavor Ingredients," Second Edition; Furia, T. H. and Bellanca, Ν., Ed.; Cleveland, 1975; Vol. II. Govindarajan, V. S. In "Food Taste Chemistry";Boudeau, J. C., Eds.; ACS SYMPOSIUM SERIES No. 115, American Chemical Society; Washington, D.C, 1979; p. 53. Gillette, M. H.; Appel, C. E.; Lego, M.C. Journal of Food Science 1984, 49. Lawless, H.; Chem. Senses 1984, 9, 143-155. Lawless, H.; Stevens, D. A.; Physiol. Behav. 1984, 32, 995-998. Cain, W. S.; Murphy, C. L.; Nature 1980, 284, 255-257. Jancso, N.; Bull Millard Fillmore HOSP. Buffalo 1960, 7, 53-57. Nagy, J. I.; "Handbook 15 Rozin, P.; Schiller, D.; Mot. Emot. 1980, 4, 77-101. Hahn, H.; "Beitrage fur Reizphysiologie", Heidelberg, 1949. Bajaj, K. J. AOAC. 1980. 63(6):1314. Hoffman, P.G.; Salb, M.C.; Galetto, W.G. J. Agr. Food Chem. 1983. 31(6:1326). Palacio, J. J. AOCA. 1977. 60(4):970 Todd, P.H.: Bensinger, M.G.: Biftu, T. J. Food Sci. 1977. 42(3):660. Scoville, W.L. J. Amer. Pharm. Assn. 1912. 1 : 453. American Spice Trade Association. 1968. "Official Analytical Methods" Method 21.0. Suzuki, J.I.; Tasign, F.; Morse, R.E. Food Technol. 1957. 11:100. Maga, J.M. Critical Rev. Food Sci Nutrition. 1975. July : 177. Govindajaran, V.S.; Narasimhan S.; Khanaraj, S.; JFS&T, India. 1977, 14(1):23. Rhyu, H.Y. J. Food Sci. 1978. 43(5): 1632 American Society for Testing and Materials. Committee E-18 on Sensory Evaluation, Subcommittee E-18, 03 on "Other Senses". Unpublished data.

RECEIVED August 5, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

4 Analysis of Chiral Aroma Components in Trace Amounts R. Tressl, K.-H. Engel, W. Albrecht, and H. Bille-Abdullah Forschungsinstitut für Chemisch-technische Analyse, Technische Universität Berlin, Seestrasse 13, D-1000 Berlin 65, Federal Republic of Germany

Methods for the capillary gas chromatographic separation of optical pounds after formation derivatives were developed. Analytical aspects of the GC-separation of diastereoisomeric esters and urethanes derived from chiral secondary alcohols, 2-, 3-, 4- and 5-hydroxyacid esters, and the corresponding and lactones were investigated. The methods were used to follow the formation of optically active compounds during microbiological processes, such as reduction of keto-precursors and asymmetric hydrolysis of racemic acetates on a micro-scale. The enantiomeric composition of chiral aroma constituents in tropical fruits, such as passion fruit, mango and pineapple, was determined and possible pathways for their biosynthesis were formulated. C a p i l l a r y column g a s c h r o m a t o g r a p h y h a s p r o v e d t o b e a s u i t a b l e t e c h n i q u e f o r t h e d e t e r m i n a t i o n o f t h e enant i o m e r i c c o m p o s i t i o n o f c h i r a l compounds i n t r a c e amounts The g a s c h r o m a t o g r a p h i c r e s o l u t i o n o f c h i r a l s u b s t a n c e s can be a c h i e v e d e i t h e r b y s e p a r a t i o n o f e n a n t i o m e r s o n a n o p t i c a l l y a c t i v e s t a t i o n a r y phase o r by formation o f d i a s t e r e o i s o m e r i c d e r i v a t i v e s and a n a l y s i s on a n o n - c h i r a l p h a s e (2). S e p a r a t i o n o f enantiomers on c h i r a l phases w i t h o u t d e r i v a t i z a t i o n s have been d e s c r i b e d o n l y f o r a few comp o u n d s , s u c h a s m o n o t e r p e n e s (3)· M o s t s e p a r a t i o n s require d e r i v a t i z a t i o n o f t h e enantiomers, e.g. c o n v e r s i o n i n t o N - c o n t a i n i n g d e r i v a t i v e s (A) , a n d a t p r e s e n t t h e number o f a v a i l a b l e t h e r m a l l y - s t a b l e c h i r a l s t a t i o n a r y phases i s l i m i t e d . D u r i n g o u r s t u d i e s o f t h e b i o g e n e s i s o f c h i r a l comp o u n d s , we d e v e l o p e d m i c r o - m e t h o d s f o r t h e a n a l y s i s o f 0097-6156/85/ 0289-0043S06.00/ 0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

44

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

n a t u r a l l y o c c u r i n g trace-components u s i n g n o n - c h i r a l phas e s . I n t h i s p a p e r we d e s c r i b e two methods f o r f o r m i n g d i a s t e r e o i s o m e r i c d e r i v a t i v e s s u i t a b l e f o r gas c h r o m a t o graphic i n v e s t i g a t i o n s of a broad spectrum of c h i r a l h y d r o x y compounds, t h e f o r m a t i o n o f d i a s t e r e o i s o m e r i c e s t e r s o f R-(+)-oC-methoxy- °C - t r i f l u o r o m e t h y l p h e n y l a c e t i c a c i d (R-(+)-MTPA) , and o f d i s t e r e o i s o m e r i c u r e t h a n e s o f R- ( + ) - p h e n y l e t h y l i s o c y a n a t e (R-(+)-PEIC). By u s i n g h i g h r e s o l u t i o n c a p i l l a r y gas c h r o m a t o g r a phy, t h e s c o p e o f use o f t h e s e r e a g e n t s , w h i c h w e r e i n i t i a l l y i n t r o d u c e d f o r N M R - a n a l y s i s (5)and f o r GC s e p a r a t i o n o f s e c o n d a r y a l c o h o l s (6), was e n l a r g e d c o n s i d e r a b l y (Z'Q) · e a p p l i c a t i o n o f t h e s e methods t o t h e a n a l y s i s o f o p t i c a l l y a c t i v e compounds f o r m e d d u r i n g m i c r o b i o l o g i c a l p r o c e s s e s and t o t h e d e t e r m i n a t i o n o f t h e e n a n t i o m e r i c c o m p o s i t i o n o f c h i r a l aroma c o n s t i t u e n t s i n some t r o p i c a l f r u i t s i s describe T n

A n a l y t i c a l aspects of the GC-separation of d i a s t e r e o i s o m e r i c R-(+)-MTPA and R - ( + ) - P E I C d e r i v a t i v e s Alcohols In Table I the data f o r the c a p i l l a r y GC-separation o f d i a s t e r e o i s o m e r i c R-(+)-MTPA e s t e r s and R - ( + ) - P E I C d e r i v a t i v e s o f some c h i r a l a l i p h a t i c a l c o h o l s a r e summar i z e d . For comparison, the data f o r the s e p a r a t i o n o f i s o p r o p y l u r e t h a n e d e r i v a t i v e s o f a l c o h o l e n a n t i o m e r s on an o p t i c a l l y a c t i v e s t a t i o n a r y p h a s e (9) a r e a l s o l i s t e d . I t i s o b v i o u s t h a t t h e s e p a r a t i o n s a r e s t r o n g l y dep e n d e n t on t h e s t r u c t u r e s o f t h e a l c o h o l s . The h i g h e s t s e p a r a t i o n f a c t o r s w i t h i n the s h o r t e s t time of a n a l y s i s were o b t a i n e d w i t h d i a s t e r e o i s o m e r i c u r e t h a n e d e r i v a t i v e s . The methods d e s c r i b e d a r e a l s o u s e f u l f o r t h e i n v e s t i g a t i o n of c h i r a l terpene a l c o h o l s . Figure 1 presents the s e p a r a t i o n of the e i g h t menthol stereoisomers. Neom e n t h o l ( 1 ) , n e o i s o m e n t h o l ( 2 ) , m e n t h o l (3) and i s o m e n t h o l (4) c a n be s e p a r a t e d and i s o l a t e d by p r e p a r a t i v e GC. By f o r m a t i o n o f R-(+)-MTPA e s t e r s , o n l y n e o m e n t h o l and m e n t h o l c o u l d be r e s o l v e d . I n c o n t r a s t , m e n t h o l s t e r e o i s o m e r s c o u l d be s e p a r a t e d a f t e r d e r i v a t i z a t i o n w i t h R - ( + ) - P E I C S c h u r i g and Weber (3) s e p a r a t e d t h e e n a n t i o mers o f m e n t h o l by c o m p l e x a t i o n c a p i l l a r y gas c h r o m a t o g r a p h y . B e n e c k e and Kônig (K>) i n v e s t i g a t e d i s o p r o p y l urethane d e r i v a t i v e s u s i n g a c h i r a l s t a t i o n a r y phase.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

180

170

170

170

Undecanol-2

Octanol-3

Octanol-4

1-0cten-3-ol

R

l

20.05

ll'll

18.05

20.80

^8.23

34.38

^2.45

]£·|£

6.65

R2

fc

t

Q

2

1

1

2

2

4

2

4

1.042 ~~

-

0

Q

Q

#

, e

K

1

U

1

1.021

1.018

T

1

1

1

(

(

c

)

)

Q

5

5

0

160

160

6

,

8

?

6

160

R

1

R2

4

5

2

2

6

7

1.055

5

9

8

8

0

1.018

ο

Q

0

Q

·

15.38 16.22

κ

1

1

K

1

1 .039

14.45

15.38

15.00

12.90

17.33

\\\*%

6.70

fc

t

R - ( + ) - P E I C

160

o

a

n

T

n

1

1

1

3

5

(

(

Q

5

5

1

c o l u m n DB 2 1 0 ; 30 m/0.33 mm i . d .

)

115

115

5

t

R

l

c

5.60

R2

(

)

Q

0

Q

1

2

1

7

2

6

column

32.50

1

1

K

21.55

27.70

65.85

48.70

1.015

\\\*S>

t

I P I C

(c) s e p a r a t i o n o f t h e i s o p r o p y l i s o c y a n a t e ( I P I C ) d e r i v a t i v e s o n a f u s e d s i l i c a XE-60-L-valine-(S)- p h e n y l e t h y l a m i d , 50 m/0.25 mm i . d .

(b) s e p a r a t i o n on a f u s e d s i l i c a

c

)

115

1

o

b

(a) s e p a r a t i o n o n a g l a s s c a p i l l a r y c o l u m n OV 1 0 1 ; 50 m/0.32 mm i . d .

170

Nonanol-2

)

170

c

0ctanol-2

o

170

(

Heptanol-2

T

R - ( + ) - M T P A

S e p a r a t i o n Data f o r t h e C a p i l l a r y GC-Separation o f C h i r a l Secondary A l c o h o l s

170

I.

Pentanol-2

Compound

Table

1#

°

16

46

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

"g

"

20min.

F i g u r e 1. C a p i l l a r y GC s e p a r a t i o n o f m e n t h o l i s o m e r s as R-(+)-MTPA d e r i v a t i v e s (CP Wax 57 CB, 50 m/0.32 mm i . d . , 180 °C) and R - ( + ) - P E I C d e r i v a t i v e s (F I : CP Wax 57 CB, 50 m/0.32 mm i . d . , 230 °C; F I I : DB 210, 30 m/0.33 mm i . d . , 180 ° C ) .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

4. TRESSL ET A L .

47

Chiral Aroma Components in Trace Amounts

2- a n d 3 - H y d r o x y a c i d

esters

Table I I presents the data f o r the GC-separation o f R-(+)-MTPA a n d R - ( + ) - P E I C d e r i v a t i v e s o f 2- a n d 3 - h y d r o xyacid esters. Whereas R-(+)-MTPA e s t e r s o f 2 - h y d r o x y a c i d s c a n b e r e s o l v e d o n p a c k e d columns (1J_) , t h e s e p a r a t i o n o f R - ( + ) MTPA d e r i v a t i v e s o f c h i r a l 3 - h y d r o x y a c i d s r e q u i r e s a n e f f i c i e n t c a p i l l a r y c o l u m n . The R - ( + ) - M T P A - e s t e r s p o s s e s s h i g h e r s e p a r a t i o n f a c t o r s t h a n t h e P E I C d e r i v a t i v e s , which depend o n t h e c h a i n l e n g t h o f t h e a c i d . The s e p a r a t i o n o f R-(+)-MTPA d e r i v a t i v e s o f 3 - h y d r o x y a c i d e s t e r s i s i n f l u e n c e d b y t h e s t a t i o n a r y p h a s e . The e f f i c i e n c y o f s e p a r a t i o n improves w i t h i n c r e a s i n g p o l a r i t y o f t h e phase, t h e h i g h e s t QC-value was o b t a i n e d w i t h a bonded t r i f l u o r o p r o p y l m e t h y l s i l i c o n e phase (1^2)

T a b l e I I . S e p a r a t i o n D a t a f o r t h e C a p i l l a r y GCS e p a r a t i o n o f R-(+)-MTPA a n d R-(+)-PEIC D e r i v a t i v e s o f Some 2- a n d 3 - H y d r o x y a c i d E s t e r s R-(+)-MTPA Compound T(°C)

(a)

I™ ^R2

R-( + )-PEIC T(°C) t

R2

c/3 c 70

D

z > z

δ

N

H

70

χ >

n

T A K E O K A ET A L .

7.

Capillary GC Analysis of Volatile Flavor Compounds

105

tures as low as 0°C. As shown i n Figure 8, the resistance to water and alcohol of t h i s p a r t i c u l a r bonded PEG has been exploited to achieve analysis of a l c o h o l i c beverages by d i r e c t s p l i t i n j e c t i o n ; the low temperature c a p a b i l i t y permitted a s t a r t i n g temperature of 35°C, and resulted i n baseline separation of acetaldehyde, methyl acetate, ethyl acetate and methanol before the appearance of the e t h y l alcohol peak. The shape of the l a t t e r peak i s quite good, and f a c i l i t a t e s (rather than i n t e r f e r e s with) the assessment of intermediate and higher b o i l i n g components. Such i n j e c t i o n s , however, do r e s u l t i n the deposition of wine s o l i d s and other non­ v o l a t i l e residues on the front of the column; i n time, t h i s i n v a r i a b l y leads to a loss of column e f f i c i e n c y . However, since these phases are non-extractable with water, the water-soluble

s 9

8

«·

5 min

»

Figure 8. Chromatogram of a German white wine (Mosel region) on a bonded polyethylene g l y c o l phase, capable of low tem­ perature operation. S p l i t i n j e c t i o n . Note that by s t a r t i n g at 35°C, acetaldehyde ( 1 ) , methyl acetate ( 2 ) , ethyl acetate (3), methanol (4), and e t h y l a l c o h o l (5) have a l l been cleanly resolved. Column, 30 m χ 0.25 mm DB-Wax*; 35°C f o r 5 min, 6°/min to 230°C, 5 min hold. Other peak assignments: (6) 1-propanol; (7) 2-methyl-l-propanol; (8) 2-ethyl-l-butanol; (9)3-methyl-l-butanol. * J and W S c i e n t i f i c , Inc.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

106

C H A R A C T E R I Z A T I O N A N D M E A S U R E M E N T OF FLAVOR C O M P O U N D S

residues can be removed by p e r i o d i c a l l y back-flushing the column with a few mL of d i s t i l l e d water at room temperature. Other developments i n column technology include the a v a i l a b i l i t y of columns with smaller and with larger diameters, coated with a wide range of stationary phase f i l m thicknesses (5, 6). Although these topics have always interested chromatographers, recent advances can be l a r g e l y a t t r i b u t e d to ( i n microbore) Schutjes and h i s colleagues (e.g. (7-9)), and ( i n megabore) to Ryder et a l . (10). Grob (11, 12) was probably the f i r s t to achieve p r a c t i c a l r e s u l t s with u l t r a - t h i c k f i l m columns, which were also considered by Jennings (13, 14), by Sandra et a l . (15) and by E t t r e (16, 17). Separation e f f i c i e n c y i n terms of the number of t h e o r e t i c a l plates per meter of column length varies inversely with column radius: better separation i s achieved on smaller diameter columns. Columns whose inner diameters are less than 100 urn, however, are extremely d i f f i c u l t t interfac with l inlet d detectors In a d d i t i o n , t h e i r capacitie overloaded, and t h e i r splitter present time i s the most p r a c t i c a l means of introducing a sample on these very small bore columns) can be capricious. Even the 100 um ID column suffers from these l i m i t a t i o n s ; s k i l l e d chromatographers have used them to good advantage, but at our present state-of-thea r t , many w i l l experience considerable f r u s t r a t i o n with these columns· The development of the megabore column may go down i n chromatographic h i s t o r y as having the most benefit f o r the most chromatographers. The vast majority of chromatographers have never experienced the t h r i l l of c a p i l l a r y chromatography. Some 75% of the chromatographic community continues to employ packed columns. Many such users accept the l i m i t a t i o n s of the packed column because they are more comfortable with the more f a m i l i a r , simpler apparatus, and are apprehensive about making any changes to the i n j e c t o r , wary of "complex" make-up gas adapters, suspicious about possible d i l u t i o n e f f e c t s , and unhappy with the facts that a d i f f e r e n t flame j e t may be required, and that they must develop a better understanding of the system. Even the word " c a p i l l a r y " triggers a negative response i n these users. Judged by our present standards of performance, the megabore i s not a c a p i l l a r y column; i t i s a large diameter open-tubular column. I t achieves i t s optimum l i n e a r gas v e l o c i t y at flow volumes of ca. 5-7 cc/min. These v e l o c i t i e s are too low to rapidly f l u s h the i n l e t that i s normally present on the packed column instrument, and as a consequence would lead to the slow introduction of a broad, d i l u t e band of sample, r e s u l t i n g i n poor separation. S i m i l a r l y , delivery of t h i s r e s t r i c t e d volume of c a r r i e r gas to the flame i o n i z a t i o n detector would disrupt the "normal" gas flow r a t i o s ; the detector would not be operating i n a plateau region, and s e n s i t i v i t y would be lower. A l t e r n a t i v e l y , by sacrificing approximately 50% of i t s p o t e n t i a l separating e f f i c i e n c y , the megabore can be operated at v e l o c i t i e s that would s a t i s f y the requirements of the packed column i n l e t and the detector (e.g. 30 cc/min). Operation under these conditions makes i t possible to substitute the megabore d i r e c t l y f o r the packed column without instrumental modification; even under these " l e s s -

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

7. TAKEOKA ET AL.

Capillary GC Analysis of Volatile Flavor Compounds

107

than-ideal" conditions, as compared to the packed column, the much more i n e r t megabore d e l i v e r s superior separation i n a shorter analysis time at increased s e n s i t i v i t y (Figure 9 ) . Obviously, quantitative r e l i a b i l i t y w i l l also be enhanced. Because i t can be d i r e c t l y substituted, the megabore column o f f e r s a r e a l i s t i c a l t e r n a t i v e to the majority of the chromatographic community who s t i l l use packed columns. I f increased r e s o l u t i o n i s ever required, the separation e f f i c i e n c y can be doubled by reducing the c a r r i e r flow toward i t s optimum value, and adding make-up gas at the detector.

Β

A - 4-CHLOtOPHENOL 15 BCtcr χ 0 . 5 3 mm MEGABORE COLUMN Β - DODBCAKE BOTH SYSTEMS:

U

20 c c / e l n HELIUM CARRIER NO MAKE UP CAS 150· C ISOTHERMAL FID CHART SPEED 3 c a / a l n

C - 1-DECTLAKIKE D - 1-UNDECANOL Β - TETIADECANB Ρ - ACENAPTHENE G - PENTADECANB

Figure 9. Results of the d i r e c t s u b s t i t u t i o n of a large-bore fused s i l i c a open tubular column. Solutes ( i n order of e l u t i o n ) : 4-chlorophenol, dodecane, 1-decylamine, 1-undecanol, tetradecane, acenaphthene, pentadecane. Top, 6 f t χ 1/8 i n OD s t a i n l e s s s t e e l column packed with Chromosorb W coated with OV 101; 20 mL/min helium c a r r i e r gas; 1.1 uL i n j e c t e d oncolumn. Bottom, same i n l e t , same detector (FID), 15 m χ 0.53 mm Megabore column coated with a 1.5 um f i l m thickness of bonded polymethylsiloxane (DB-1) substituted, and operated at the same 20 mL/min helium flow; 0.3 uL i n j e c t e d i n same manner. The time scale i s the same i n both chromatograms. Note t a i l i n g of the phenol and a l c o h o l , and complete disappearance of the amine on the packed column.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

CHARACTERIZATION A N DM E A S U R E M E N T O F FLAVOR

108

COMPOUNDS

Acknowledgments Portions of Figure 1 were reproduced with permission of J and W S c i e n t i f i c , Inc. The authors are also g r a t e f u l to R.R. Freeman, E. Guthrie, R. Lautamo, and L. Plotczyk f o r data on the Megabore c o l ­ umn, and f o r access to the chromatograms shown i n Figures 8 and 9.

Literature Cited 1.

Takeoka, G.; Jennings, W. J. Chromatogr. Sci. 1984, 22, 177-184. 2. Grob, Κ., Jr.; Mueller, R. J. Chromatogr. 1982, 244, 185-196. 3. Mehran, Mehrzad, Cooper, W. J. and Jennings, W. Paper No. 21, presented at ACS National Meeting, 28 April-03 May 1985, Miami Beach, FL. 4. Reeve, V.; Jeffries, J.; Weihs, D.; Jennings, W. Paper pre­ sented at Labcon West, 23-25 April 1985, San Mateo, CA. 5. Jennings, W. Paper 09 October 1984, St. 6. Jennings, W.; Takeoka, G. Paper presented at "Neue Ulmer Gespraeche", Symposium on Capillary Chromatographic Analysis of Drugs and Pharmaceuticals, 06-09 May 1985, Neu Ulm, Germany. Proceedings in press, Huethig Publishing Co. 7. Schutjes, C. P. M. Doctoral Thesis, Eindhoven University of Technology, The Netherlands, 1983. 8. Schutjes, C. P. M.; Vermeer, Ε. Α.; Cramers, C. A. Proc. 5th Int. Symp. on Capillary Chromatography, 1983, p. 29. 9. Schutjes, C. P. M.; Vermeer, Ε. Α.; Scherpenzeel, G. J.; Bally, R. W.; Cramers, C. A. J. Chromatogr. 1984, 289, 157-162. 10. Ryder, B. L.; Phillips, J.; Plotczyk, L. L.; Redstone, M. Paper No. 497, Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, 5-9 March, 1984, Atlantic City, ΝJ. 11. Grob, Κ., Jr.; Grob, K. Chromatographia 1977, 10, 250-255. 12. Grob, K.; Grob, G. J. High Res. Chromatogr. 1983, 6, 133-139. 13. Jennings, W. Paper No. 286, Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, 5-9 March, 1984, Atlantic City, NJ. 14. Jennings, W. J. Chromatogr. Sci. 1984, 22, 129-135. 15. Sandra, P.; Temmerman, I.; Verstappe, M. J. High Res. Chromatogr. 1983, 6, 501-504. 16. Ettre, L. S. Chromatographia 1983, 17, 553-559. 17. Coleman, P.; Ettre, L. S. J. High Resol. Chromatogr. 1985, 8, 112-118. RECEIVED August 21, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

8 High-Resolution Gas Chromatography-Fourier Transform IR Spectroscopy in Flavor Analysis Limits and Perspectives Heinz Idstein and Peter Schreier Lehrstuhl für Lebensmittelchemie, Universität Wü rzburg, Am Hubland, D-8700 Wü rzburg, Federal Republic of Germany The capability o coupled with Fourie (HRGC-FTIR) for the analysis of food flavors is demonstrated by selected examples (tropical fruits, endive, fermentation byproducts). A problem associated with HRGC-FTIR analysis in flavor research is its relative low sensitivity and dynamic range compared to mass spectrometry (MS). To overcome this problem, a dynamic compression ("DYCOM") system consisting of a linked HRGC-FTIR-MS combination, in which multidimensional packed-capillary gas chromatography (MDGC) is integrated, is proposed. With this system, operation with f u l l sensitivity of both spectroscopic methods is possible. The separation of v o l a t i l e trace components usually i s achieved with high-resolution gas chromatography (HRGC) using c a p i l l a r y columns (_1). However, with a complex mixture of substances of d i f f e r e n t chemical classes, as found among food flavors, one separation technique may not be s u f f i c i e n t to provide the maximum amount of information about the constituents, and repeated separations under a variety of chromatographic parameters may be necessary. More and improved information can be obtained by simultaneous introduction of the sample onto two columns of d i f f e r e n t p o l a r i t y and p a r a l l e l double detection (e.g., a nonspecific detector i n tandem with a selective detector) (2). Gas chromatography-mass spectrometry (HRGC-MS) occupies a special place among the a n a l y t i c a l techniques for investigations of v o l a t i l e flavor compounds, since i t provides maximum information from minimum sample material (.3). HRGC-MS provides both chromatographic (linear retention index) and s t r u c t u r e - s p e c i f i c information (MS-spectrum). Nevertheless, the method i s not capable of discriminating between d i f f e r e n t isomers. In these cases, one must obtain additional information with infrared or NMR spectroscopy. Early attempts to combine an IR instrument with an a n a l y t i c a l gas chromatograph were only partly successful; complications were related to the fact that components eluted from the GC column i n time i n t e r v a l s too short to permit matching the scale time of the IR 0097-6156/85/0289-0109$06.00/0 © 1985 American Chemical Society In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

110

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

equipment then available (4^). Recently, dispersive IR units have been modified to make them compatible with GC, and additional s e n s i t i v i t y was gained by applying Fourier transform (FT), leading to both better signal-to-noise r a t i o , and/or spectra i n a shorter time (5-7). The importance of obtaining IR information i n addition to mass spectra w i l l be demonstrated by an example taken from the work of our colleague Adam (8) (Figure 1). After photochemical transformation of compound ·_1 the main reaction product 2 (> 95 %) (Figure 1, top) was separated by preparative gas chromatography and then characterized by various spectroscopic methods. As to another reaction product (< 5 %) detectable by HRGC (Figure 1, bottom l e f t ; black square) and characterized by MS (Figure 1, bottom right) the hypothetical structures 3-8 can be formulated. Due to the low amount of the product, NMR spectroscopy could not be used for structure elucidation, but HRGC-FTIR analysis solved the problem. As outlined i n Figure 2 the vapor phase FTIR spectru exocyclic methylene grou except formula 3_ (Figure 1). In the area of f l a v o r s , we recently reported the r e s u l t s obtained from HRGC-FTIR studies of t r o p i c a l f r u i t s (10). During our HRGC-MS studies on cherimoya (Annona cherimolia, Mill.) volatiles, we obtained, among others, the mass spectrum shown i n Figure 3. At f i r s t glance, the spectrum suggests that the compound could be 2-pentenol, and the unkwown and 2-pentenol had s i m i l a r GC retention times. However, the "on-the-fly" IR spectrum of the same unknown (Figure 4j) showed absorption bands at 3082, 1755, 1650, 1177 and 895 cm i n d i c a t i n g an unsaturated (Z)-configurated ester. Comparison of the chromatographic and spectroscopic data with those of synthesized reference samples proved that the i d e n t i t y of the unknown compound was (Z)-2-pentenyl butanoate. Through support provided by various FTIR manufacturers,we were able to extend our HRGC-FTIR studies of v o l a t i l e s including guava fruit (Psidium guajava, L.) (11), fermentation byproducts (12) and vegetables (13). The following examples are drawn from these i n v e s t i gations. Figure 5 shows the FID trace from the HRGC-FTIR study of a polar s i l i c a gel f r a c t i o n of v o l a t i l e s (alcohols, hydroxy esters, lactones etc.) obtained from model fermentations using d i f f e r e n t yeast species and s t r a i n s . The numbers i n c i r c l e s indicate the r e g i s t r a t i o n of FTIR spectra. In some cases, minor constituents were detected by FTIR, since they were good IR absorbers. Some examples of vapor phase FTIR spectra taken from t h i s run are presented i n Figure 6, i . e . those of two major components, 2-methyl-l-propanol and 2-phenylethanol (top, l e f t and r i g h t , respectively) and of two compounds, ethyl l a c t a t e (bottom l e f t ) and Y-butyrolactone (bottom, r i g h t ) that were detectable i n much lower concentrations. Similar r e s u l t s were obtained i n our study of a mid-polar silica gel f r a c t i o n of endive v o l a t i l e s . Figure 7 shows the FID trace of t h i s f r a c t i o n and again the numbers i n c i r c l e s indicate where FTIR spectra could be recorded. In t h i s case, p r a c t i c a l l y a l l major compounds were detected (and i d e n t i f i e d ) by IR spectroscopy, again including minor

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

8.

IDSTEIN AND SCHREIER

HRGC-FTIR

in Flavor Analysis

6

111

7

8

91

I

Figure 1. Top: Major product (_2) ( >95 %) and hypothetical structures (3-8) ( ( c f . Figure 1).

50

100

150

m/e

Figure 3. Mass spectrum (70 eV) of unknown compound of cherimoya f r u i t v o l a t i l e s (explanation c f . text).

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

8.

HRGC-FTIR in Flavor Analysis

IDSTEIN AND SCHREIER

113

100

4000

300

cm"

Figure 4. Vapor phase FTIR spectrum of unknown compound from HRGCFTIR analysis of cherimoya f r u i t v o l a t i l e s . Same compound as i n Figure 3; explanation c f . text.

® ® M?) -

(2)

υ

®

10

20

30

mm

Figure 5. FID trace from the HRGC-FTIR study of s i l i c a gel f r a c t i o n I I I of fermentation byproducts obtained i n model fermentation studies with Saccharomyces cerevisiae (12). Column: 30 m χ 0.32 mm i . d . CP 57 CB (df = 0.2) (Chrompack). 50-130 C, 2°/min. 130-200°C, 5°/min.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

114

> Ε

10

20

30

min

Figure 6. FID trace from the HRGC-FTIR study of s i l i c a gel f r a c t i o n I I of endive v o l a t i l e s (13). Column: 30 m χ ο.32 mm i . d . CP 57 CB (df = 0.2) (Chrompack). 50-130° C, 2°/min. 130-200° C, 5°/min.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

Figure 7. Vapor phase FTIR spectra taken from HRGC-FTIR study of fermentation byproducts (12). Top: 2-methyl-l-propanol ( l e f t ) ; 2phenylethanol ( r i g h t ) . Bottom: ethyl lactate ( l e f t ) ; Y-butyrolactone ( r i g h t ) .

Î

ft

ft"

S'

Ci

%

το

m m

η !»

ζ/5

> α

m Ζ

Η

α

po

116

CHARACTERIZATION A N D M E A S U R E M E N T OF FLAVOR

COMPOUNDS

v o l a t i l e s i n some cases. Some vapor phase IR spectra from this run are selected as examples i n Figure 8. In t h i s Figure, the on-the-fly" spectra of (E)-2-hexen-l-ol (top, left), (Z)-3-hexen-l-ol (top, r i g h t ) , l-penten-3-ol (bottom, l e f t ) and phenylethanal (bottom, r i g h t ) are presented. The IR spectra and the other recorded FTIR data were discussed elswhere (12,13). Tt

As already reported for our recent study on t r o p i c a l f r u i t s , the p r i n c i p a l capability of HRGC-FTIR for the analysis of food flavors employing WCOT columns was confirmed; however, there are some l i m i t a t i o n s . F i r s t of a l l , the s e n s i t i v i t y of the method has permitted recording HRGC-FTIR spectra for some fractions of as l i t t l e as about 50 ng/peak. On the other hand, detection of strong bands exhibited by a molecule does not always enable component i d e n t i f i c a t i o n , especially i f one considers that a further l i m i t a t i o n to rapid and positive i d e n t i f i c a t i o n by HRGC-FTIR analysis i s s t i l l the lack of adequate IR vapor phase spectral bank such data (15,16). As t that on a routine basis the HRGC-FTIR method requires approximately 50 ng/peak, while the GC-MS technique requires less than 1 ng/peak i n order to record good interprétable IR- and MS spectra. This indicates that HRGC-FTIR i s more than one order of magnitude less sensitive than HRGC-MS for present state-of-the-art systems. This last-mentioned fact i s also the major l i m i t a t i o n i n coupling both IR and MS i n tandem, which i s p r i n c i p a l l y possible due to the non-destructive character of IR spectroscopy. Three important papers have been provided on t h i s topic (17-19), but i n a l l three publications the chromatographic procedures and coupling systems did not correspond to the state-of-the-art i n this f i e l d , i . e . SCOT columns with low separation capacity or even a j e t separator were used. Consequently, these combinations were only optimized for FTIR spectroscopy, which led to strong loss i n s e n s i t i v i t y of mass spectrometry i n comparison to HRGC-MS coupling systems. Whereas with a HRGC-FTIR-MS system problems of overloading arise i n HRGC and MS, i n FTIR spectroscopy one has to extend the detection l i m i t to a value common for MS. Our idea i s to solve t h i s problem by using packedc a p i l l a r y multidimensional gas chromatography (MDGC) with a suitable gas chromatographic apparatus. Figure 9 outlines our ideas for a linked HRGC-FTIR-MS system, which can obtain the highest sensitivity of both spectroscopic techniques. In t h i s scheme, i n which the loading capacity of the column (a), the detection l i m i t of FTIR (b) and that of MS (c) are outlined, four t y p i c a l examples for the concentration dynamics of peaks (1-4) separated by gas chromatography are represented. In the combination packed GC (PGC)-FTIR-MS (A) peak 1 i s eluted i n the optimal area for detection with both spectroscopic methods; peak 2 i s overloaded and may contain more than one component; peak 3 can only be detected by MS, and the amount of peak 4 i s even below the MS detection l i m i t . In comparison to system (A), using HRGC (B) the separation capacity can be improved, e.g., leading to separation of compounds 1 and 3 to l a / l b and 3a/3b, respectively, but the dynamics of concentrations are not influenced. With t r a d i t i o n a l MDGC (Β), i . e . using p a r t i a l "heart-

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

8.

IDSTEIN AND SCHREIER

HRGC-FTIR

in Flavor Analysis

117

Figure 8. Vapor phase FTIR spectra taken from HRGC-FTIR study of endive v o l a t i l e s (13). Top: (E)-2-Hexen-l-ol ( l e f t ) ; (Z)-3-Hexenl - o l ( r i g h t ) . Bottom: l-penten-3-ol ( l e f t ) ; phenylethanal ( r i g h t ) .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985. ι INTERUHRGC H FTIR

PQC

H MS

HRGC TRAD. MDGC

PQC

TRAP

OVCOM

MDGC

HRQC

FTIR

«DYCOM»

Figure 9. Proposal f o r a linked HRGC-FTIR-MS system with integrated multidimensional gas chromatography (MDGC) to be used with f u l l s e n s i t i v i t y of both spectroscopic techniques. A: Packed GC-FTIR-MS. B: T r a d i t i o n a l MDGC system with FTIR-MS. C: Proposed "DYCOM" system consisting of packed-capillary MDGC-HRGC-FTIR-MS. a: Loading capacity of column. b,c: Detection l i m i t s of FTIR and MS, respectively. Further explanations , c f . text.

H FTIR H MS

PGC

MS

en

D

Ο c ζ

•Ό

ο

n

70

Ο

H Ο •n -n r

m z

70 m

> C

m

D

Ο Ζ > ζ

το Ν

Η m

>

ο se

8.

IDSTEIN AND SCHREIER

HRGC-FTIR

119

in Flavor Analysis

cut", the amount of the overloaded compound 2 can be reduced i n such a way that with HRGC an improved separation w i l l be obtained (2a/2b). Nevertheless, the lack of spectroscopic detection for peaks 3a/3b (FTIR) and 4 (MS) s t i l l remains. Our proposal f o r a dynamic compression of concentrations dynamics as needed for flavor mixtures i s outlined i n system C. The dynamic compression ("DYCOM") works i n the following way: Peaks l a / l b and 2a/2b would be handled by the t r a d i t i o n a l MDGC technique, i . e . p a r t i a l "heart-cut" would be performed, but using the f u l l flow of the packed column onto the "DYCOM" trap, compounds 3a/3b and 4 would be enriched and could then be detected by FTIR (3a/3b) and MS (4), respectively. The i n d i v i d u a l elements of such a device are a l l commercially available, e.g., the first three elements of system C p r a c t i c a l l y correspond to the Sichromat-2 MDGC apparatus from Siemens (20). As peak triggering i s desirable, the computerization of the commercial system should be extended. Such a linked HRGC-FTIR-MS system using the f u l l s e n s i t i v i t y of both spectroscopic method are sure that i t w i l l b the near future. Acknowledgment s

We are indebted to the manufacturers Bruker, Bio-Rad Digilab, and Nicolet f o r the analysis of flavor mixtures by HRGC-FTIR. Special thank i s expressed to Mrs. E.M. Gotz-Schmidt and Mr. R. Hock for their contributions to sample preparation. Literature 1.

Cited

2.

Jennings, W.; Takeoka, G. In "Analysis of Volatiles"; Schreier, P., Ed.; W. de Gruyter: Berlin, New York, 1984; pp. 63-75. Schomburg, G.; Husmann, H.; Podmaniczky, L.; Weeke, F.; In "Analysis of Volatiles"; Schreier, P., Ed.; W. de Gruyter: Berlin, New York, 1984; pp. 121-150.

3.

Ten Noever

de Brauw,

W. de Gruyter: 4. 5. 6. 7. 8. 9.

10. 11.

12. 13.

Low,

M.J.D.;

Berlin, Freeman,

M.C.;

In "FLAVOUR

New York, S.K.;

J.

1981; Agric.

'81";

Schreier,

P.,

Ed.;

16,

525-

pp.253-286. Food Chem. 1968,

528. Ericksson, M.D.; Appl. Spectrosc. Rev. 1979, 15, 261-278. Griffiths, P.R.; De Haseth, J.A.; Azarraga, L.V.; Anal. Chem. 1983, 55, 1361 A. Herres, W.; In "Analysis of Volatiles"; Schreier, P., Ed.; W. de Gruyter: Berlin, New York, 1984; pp. 183-217. Adam, W.; Dörr, M.; Hill, K.H.; Peters, E.M.; Peters, K.; Von Schnering, H.G.; J. Amer. Chem. Soc., 1984, in press. Nyquist, R.A.; "The Interpretation of Vapor-Phase Infrared Spectra. Vol. 1. Group Frequency Data"; Sadtler/Heyden: Philadelphia, 1984. Herres, W.; Idstein, H.; Schreier, P.; HRC & CC 1983, 6, 590-594. Schreier, P.; Idstein, H.; Herres, W.; In "Analysis of Volatiles"; Schreier, P., Ed.; W. de Gruyter: Berlin, New York, 1984; pp. 293-306. Hock, R.; Dissertation, Univ. Würzburg, 1985. Götz-Schmidt, E.M.; Dissertation, Univ. Würzburg, in prep.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

120

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

14. 15.

Welti, D.; "Intra-Red Vapour Spectra"; Heyden: London, 1970. Nyquist, R.A.; "The Interpretation of Vapor-Phase Infrared Spectra. Vol. 2."; Sadtler/Heyden: Philadelphia, 1984. "FLATCO" (Flavours-Attractants-Contaminants) HRGC-FTIR-MS data bank; Schreier, P., Ed.; STN International, Columbus, Karlsruhe, Tokyo, in prep. Wilkins, C.L.; Giss, G.N.; Brissey, G.M.; Steiner, S.; Anal. Chem. 1981, 53, 113-117. Crawford, R.W.; Hirschfeld, T.; Sanborn, R.H.; Wong, C.M.; Anal. Chem. 1982, 54, 817-820. Wilkins, C.L.; Giss, G.N.; White, R.L.; Brissey.G.M.; Onyiriuka, E.C.; Anal. Chem. 1982, 54, 2260-2264. Oreans, M.; Müller, F.; Leonhard, D.; Heim, Α.; In "Analysis of Volatiles"; Schreier, P., Ed.; W. de Gruyter: Berlin, New York, 1984, pp. 171-182.

16.

17. 18. 19. 20.

RECEIVED August 26, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9 Tandem Mass Spectrometry Applied to the Characterization of Flavor Compounds Kenneth L . Busch and Kyle J. Kroha Department of Chemistry, Indiana University, Bloomington, IN 47405

Tandem mass spectrometry (MS/MS) i s a new analytical technique applied to problems in food and flavor analyses. Rapidit against chemical noise mixtures for functional groups are attributes of MS/MS that make it attractive for such problems. Sample collection and pretreatment differ from methods used in GC/MS. Correct choice of an ionization method i s paramount. Daughter ion MS/MS spectra are used for compound identification via comparison with those of authentic compounds, and parent and neutral loss spectra are useful in functional group analysis. Applications to direct analysis of volatiles emitted from fruits and to spice analyses are considered. Why use MS/MS a n a l y s i s o f v o l a t i l e components from food and f l a v o r components? Figure 1 provides the answer. The top trace i s the c a p i l l a r y column gas chromatographic p r o f i l e o f the concentrated v o l a t i l e s from a knockwurst sausage sample. The temperature program o f 55°C t o 180°£ a t 5° per minute establishes the time scale from beginning t o end o f run as 25 minutes. Coupled t o a mass spectrometer f o r i d e n t i f i c a t i o n , each o f the many compounds can be examined by the mass spectrometer f o r only a few seconds. The bottom s e r i e s o f figures i l l u s t r a t e s the d i r e c t MS/MS a n a l y s i s of the v o l a t i l e s from the sausage sample. A stream o f a i r i s swept over the sausage and c a r r i e d i n t o the source o f the mass spectrometer. Ions are formed from the v o l a t i l e constituents, and the f i r s t analyzer o f the instrument scans (5 s) t o provide a mass spectrum o f the mixture. A p a r t i c u l a r i o n i s selected from a l l o f those formed, excited by c o l l i s i o n , and i t s fragment ions analyzed by a second mass analyzer (5 s ) . The MS/MS spectrum thus obtained i s compared t o the spectrum o f the authentic compound (contained i n the laboratory l i b r a r y ) and the i d e n t i t y o f the compound established (10 s ) . The t o t a l time f o r a n a l y s i s by MS/MS i s under a minute, including the time required t o load the sausage i n t o the sample 0097-6156/85/ 0289-0121 $06.00/ 0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

122

boat. A p a r t i c u l a r i o n i c component present i n the mass spectrum can be i d e n t i f i e d i n only a few seconds. Since a l l ions are a v a i l a b l e continuously, the a c q u i s i t i o n o f data can be t a i l o r e d t o the i n t e n s i t y o f the s i g n a l . For strong s i g n a l s , a few seconds s u f f i c e s ; f o r weak s i g n a l s , the i n t e g r a t i o n time can be lengthened appropriately. Note t h a t the i o n chosen f o r the MS/MS experiment at m/z 163 i s only a minor component i n the mass spectrum. The analyst has the freedom t o examine any i o n formed i n the source i n any order, u n l i k e GC/MS, which allows examination o f the sample only i n a short "time window" established by the chromatography. To re-examine a compound i n GC/MS, the e n t i r e sample must be reintroduced t o the gas chromatograph. In MS/MS, the o r i g i n a l i o n i s simply again selected by the f i r s t mass analyzer. F i n a l l y , t h e time evolution o f a number o f compounds can be followed d i r e c t l y w i t h MS/MS, f o r example, as the sample i s heated. In GC/MS t h i s simple experiment generates a number o f samples, each o f which must be d i s c r e t e l y analyzed. Background. As an a n a l y t i c a l technique, tandem mass spectrometry i s j u s t entering i t s second decade o f development. The v a r i e t y o f reported a p p l i c a t i o n s b e l i e s i t s r e l a t i v e youth. Tandem mass spectrometry (MS/MS) grew out o f e a r l y work which used met e s t a b l e ion t r a n s i t i o n s i n order t o e s t a b l i s h i o n structures and i n t e r r e l a t i o n s h i p s . A f t e r extensive applications t o i o n s t r u c t u r a l studies, i t s usefulness i n d i r e c t complex mixture analysis became apparent w i t h the e a r l y work o f Cooks (1-3). I t s successes i n problem s o l v i n g are summarized i n a recent book edited by McLafferty (4). Now, w i t h several commercial instruments a v a i l a b l e , MS/MS i s being evaluated f o r a p p l i c a t i o n i n several new areas, i n c l u d i n g biochemical a n a l y s i s , forensic chemistry, and food and f l a v o r analyses. The p r i n c i p l e s o f MS/MS w i l l be summarized i n the f i r s t p a r t o f t h i s chapter. The second part o f the chapter w i l l deal w i t h the reported a p p l i c a t i o n s o f MS/MS t o f l a v o r analysis. Principles Ion Processing. As mentioned, MS/MS began w i t h the study o f meta stable ions (_5 ). Metastable t r a n s i t i o n s are observed from ions which undergo a d i s s o c i a t i o n while i n t r a n s i t through the instrument. The t r a n s i t i o n i s a chemical reaction c h a r a c t e r i s t i c of the nature o f the i o n . In MS/MS, the instrument i s modified so that the reactions occur more frequently and the masses o f the reacting i o n and the product i o n can be established. The approach t o MS/MS i s thus quite d i f f e r e n t from that f o r high r e s o l u t i o n mass spectrometry. There the exact mass measurement which provides the empirical formula o f the i o n i s a p h y s i c a l measurement. In metastable i o n studies, the focus i s on the nature o f the i n d i v i d u a l chemical reactions. Each metastable ion represents an i n d i v i d u a l l y defined t r a n s i t i o n f o r which masses and abundances o f both products and reactants can be s p e c i f i e d . The a n a l y t i c a l advantage t h a t accrues i s based on the greater information inherent i n a chemical rather than a p h y s i c a l

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

MS-MS Applied to Flavor Compound Characterization

BUSCH AND KROHA

approach. An i o n once formed i s not simply measured t o e s t a b l i s h i t s mass and i t s r e l a t i v e abundance, but rather i s processed i n experiments designed t o define i t s chemical r e a c t i v i t y . Independent sequential analyses. In order t o explain the b a s i s o f an MS/MS experiment consider the b a s i c r e a c t i o n sequence shown i n r e a c t i o n 1. M]+

+

>

N

m 2

+

+

1TV3

+ Ν

(1)

The goal i s t o e s t a b l i s h the masses o f both the products and reactants o f t h i s chemical r e a c t i o n . We thus require an analyzer t o e s t a b l i s h Μ^ , and a second analyzer t o e s t a b l i s h rrç*. The mass of the neutral i s defined by difference. The r e a c t i o n takes place between the two analyzers, aided by energy added t o the reactant i o n i n t h i s region (vide i n f r a ) . With the a d d i t i o n o f a source (S r e a c t i o n as *, the bloc established (Figure 2). This diagram a l s o explains the various experiments a v a i l a b l e i n MS/MS. The s a l i e n t points are: 1) there are two mass analyzers t o characterize the r e a c t i o n ; 2) the t r a n s i t i o n from reactant t o product occurs between the analyzers; 3) the analyzers operate independently. There must be a source o f energy i n order t o i n i t i a t e the r e a c t i o n between the analyzers beyond the inherent metastable i o n abundances. T y p i c a l l y the interanalyzer region i s f i t t e d w i t h a c o l l i s i o n c e l l which contains about a m i l l i t o r r of target gas Ν (often nitrogen or helium). The incoming i o n c o l l i d e s w i t h the target gas, transforming some f r a c t i o n o f the k i n e t i c energy of the i o n i n t o i n t e r n a l energy which then causes fragmentation. +

Resolution. In MS/MS, each o f the two independent mass analyzers can be operated a t a low r e s o l u t i o n while r e t a i n i n g a high o v e r a l l s e l e c t i v i t y . Since e x t r a c t i o n of the highest r e s o l u t i o n from a given analyzer requires a disproportionate e f f o r t , the s o l u t i o n o f demanding a n a l y t i c a l problems i s s i m p l i f i e d . By analogy, a chromatographer reduces the performance requirements o f a s i n g l e stage separation of a complex mixture by adding a simple sample p r e f r a c t i o n a t i o n . The same general p r i n c i p l e i s apparent i n MS/MS. Signal-to noise. Mass spectrometers are e x t r a o r d i n a r i l y s e n s i t i v e devices, having the a b i l i t y t o analyze nanogram amounts o f sample. MS/MS, as discussed above, deals w i t h the chemistry o f i o n i c reactions, and thus i t i s o f t e n chemical rather than e l e c t r o n i c noise t h a t establishes the l i m i t o f detection (6). Chemical noise i s produced by matrix constituents other than the sample, the reactions o f which may be i n s u f f i c i e n t l y resolved from the sample reaction o f i n t e r e s t . The r o l e o f the a n a l y t i c a l chemist i s t o design the MS/MS experiment t o provide the best p o s s i b l e d i s c r i m i n a t i o n against the chemical noise i n the system. The problem i s a s i g n i f i c a n t one; i n complex mixtures, the matrix constituents are present i n great excess, and t h e i r reactions are unknown. However, the use o f several stages o f independent mass a n a l y s i s can provide a very high signal-to-noise r a t i o . Table I summarizes the operation o f a generic species o f analyzers given an i n i t i a l mixture w i t h equal parts o f s i g n a l and noise. For t h i s example, each analyzer passes 50% o f the s i g n a l but only 10% o f the In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

123

124

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

SAUSAGE χ—

^STREAM

CONCENTRATE NJECT

QC/M 26i

fU

Mi

1M Itl

Jl

•ft

•a!

It*

M E C T ANALY88 MS/MS

10·

ΊΊ

20·

ΜΙΑ·

Figure 1. Comparison o f the time scales o f the procedures o f GC/MS and MS/MS a n a l y s i s o f v o l a t i l e f l a v o r compounds emitted from sausage samples (19).

Figure 2. Simple diagram o f an MS/MS instrument and three scanning modes based on changes i n mass between the parent and the daughter i o n . See t e x t f o r d e t a i l s .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

BUSCH AND KROHA

MS-MS Applied to Flavor Compound Characterization

noise. Even w i t h t h i s crude d i f f e r e n t i a t i o n , a f t e r two stages o f a n a l y s i s , the s i g n a l t o noise r a t i o has increased from u n i t y t o 25:1. With further a n a l y s i s , i t r i s e s r a p i d l y even as the t o t a l s i g n a l l e v e l decreases. I n MS/MS, the choice o f i o n i z a t i o n method, ion p o l a r i t y , and the mass analysis i t s e l f a l l contribute t o the d i f f e r e n t i a t i o n o f s i g n a l from noise, and the stepwise enhancement i s o f t e n several hundred t o one rather than 5:1 as shown i n t h i s example. Table I .

Signal t o Noise Enhancements w i t h M u l t i p l e Analyses

Number of stages 0 1 2 3

Intensities Signal Noise

Signal-to-noise ratio

100 250 125

10 1

25 125

Types o f Experiments The a n a l y s i s i n a mass spectrometer i s not based on the mass o f an ion, but rather on i t s mass t o charge r a t i o , m/z. Thus i f e i t h e r the mass o r the charge o f the i o n i s a l t e r e d i n an MS/MS experiment, the change can be followed by the second mass analysis. The experiments o f MS/MS can be subdivided i n t o those which involve changes i n mass o r changes i n charge. There e x i s t s a t h i r d category o f experiments which involve changes i n r e a c t i v i t y independent o f mass and charge, but these experiments w i l l not be discussed i n t h i s chapter. Changes i n mass. The most ccnrnon experiments i n MS/MS are based on changes i n mass. These are summarized i n Figure 2. Assume a complex mixture has been introduced i n t o the source, and that ions are formed corresponding t o each constituent o f the mixture. The f i r s t mass analyzer s e l e c t s ions o f a s p e c i f i e d mass which are passed i n t o the c o l l i s i o n region between the analyzers. Here the a d d i t i o n a l energy imparted by c o l l i s i o n causes the breakup o f t h i s parent i o n i n t o smaller fragment ions. The masses o f the fragment ions, termed daughter ions, i s established by scanning the second mass analyzer. The r e s u l t i n g spectrum i s c a l l e d a daughter i o n MS/MS spectrum, and consists o f a l l o f the fragment ions from a selected parent i o n (Figure 2a). Since the mass analyzers operate independently, i t a l s o i s possible t o set the second mass analyzer t o pass only daughter ions of a selected mass t o the detector. The f i r s t mass analyzer i s then scanned across the mass range. A s i g n a l a t the detector i s noted when the f i r s t mass analyzer passes a parent i o n that fragments t o the s p e c i f i e d daughter i o n . The spectrum that i s obtained i s c a l l e d a parent i o n MS/MS spectrum, and consists o f a l l the precursor ions o f a s p e c i f i e d daughter i o n (Figure 2b). I f both mass analyzers are scanned a t the same rate w i t h a

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

125

126

CHARACTERIZATION A N D M E A S U R E M E N T O F FLAVOR

COMPOUNDS

constant mass o f f s e t χ between them, then signals w i l l be observed at the detector whenever a parent i o n passing through the f i r s t mass analyzer produces a daughter i o n w i t h a mass χ daltons l e s s than the parent i o n . The spectrum obtained i s c a l l e d the constant neutral loss MS/MS spectrum, and consists o f a l l the parent ions i n the parent ion/daughter i o n p a i r s r e l a t e d by the l o s s o f a neutral of s p e c i f i e d mass (Figure 2c). The three MS/MS experiments described above provide d i f f e r e n t information i n complex mixture a n a l y s i s . The daughter i o n MS/MS spectrum i s often obtained when targeted compound analysis i s performed. The parent i o n selected corresponds t o the targeted component, and the daughter i o n spectrum obtained from the mixture i s compared t o that obtained f o r the authentic targeted compound introduced t o the source under the same conditions. In t h i s way, the presence o f the target can be established and o f t e n quantitated. Parent and constant neutral l o s s MS/MS spectra are more o f t e n used f o r i d e n t i f i c a t i o used f o r both targeted compoun unknown mixture. Experience often shows that there are c h a r a c t e r i s t i c daughter ions o r neutral losses that occur f o r s p e c i f i c functional groups. For instance, 149 as a daughter i o n i s c h a r a c t e r i s t i c i n the MS/MS spectra o f phthalates. A scan f o r parent ions o f the s p e c i f i e d daughter i o n 149 would thus pinpoint a l l o f the various phthalates present i n a mixture, regardless o f whether each was known t o be present o r not. S i m i l a r l y , the protonated molecular ions o f carboxylic acids t y p i c a l l y form daughter ions by l o s s o f carbon dioxide. A constant n e u t r a l l o s s MS/MS spectrum w i t h the o f f s e t s p e c i f i e d as 44 daltons (the weight o f the neutral fragment OO2) w i l l pinpoint parent ion/daughter i o n p a i r s that undergo a chemical reaction t y p i c a l o f carboxylic acids, again without p r i o r knowledge o f t h e i r presence. +

+

Changes i n charge. The c o l l i s i o n s used t o add energy t o ions and cause fragmentation a l s o may cause changes i n charge o f the i o n . Singly-charged ions can be oxidized t o doubly-charged ions i f the energy imparted by the c o l l i s i o n i s higher than the second i o n i z a t i o n p o t e n t i a l o f the molecule (charge s t r i p p i n g , Equation 2 ). Doubly-charged ions passed i n t o the c o l l i s i o n c e l l can be reduced t o the singly-charged i o n (Equation 3) i n a process known as charge exchange; the n e u t r a l involved i n the c o l l i s i o n acquires the balancing charge. F i n a l l y , negative ions can be converted i n t o the corresponding p o s i t i v e ions i n the oxidation process known as charge inversion (Equation 4 ) . Μ]+ M!

+ Ν 2+

+ Ν

M" + Ν

2

> M ] * + e" + Ν > M]* + N+ > M+ + 2e~ + Ν

(2) (3)

(4)

The energetic requirements f o r these reactions are d i f f e r e n t from those o f the reactions which involve changes i n mass. For the most part, these reactions are observed i n high energy c o l l i s i o n s . They have predominately been used f o r studies o f i o n structure, but have r e c e n t l y been used f o r complex mixture a n a l y s i s . This expansion i s

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

BUSCH AND KROHA

MS-MS Applied to Flavor Compound Characterization

based on the r e l a t i o n s h i p between i o n s t r u c t u r e and the f a c i l i t y o f charge changing reactions. Nitrogen-containing compounds f o r instance, form doubly-charged ions more r e a d i l y than other classes o f compounds. An experiment based on the r e a c t i o n of doubly charged ions can thus be s p e c i f i c f o r nitrogen-containing ions. #

A n a l y t i c a l C h a r a c t e r i s t i c s o f MS/MS MS/MS was f i r s t considered as a replacement f o r GC/MS. I t s true character as a complement t o t h a t method i s now r e a l i z e d , and the most demanding of a n a l y t i c a l problems often require the f u l l d i f f e r e n t i a t i n g power o f a GC/MS/MS combination. The choice between GC/MS o r MS/MS f o r a p a r t i c u l a r a p p l i c a t i o n must r e s t on r e l a t i v e merits o f s e n s i t i v i t y , s e l e c t i v i t y , and speed, each o f which w i l l now be b r i e f l y discussed. S e n s i t i v i t y . I t i s misleadin s e n s i t i v i t y f o r e i t h e r GC/M problem, the instrument, the experiment, the spectrum, and the operator. In general, f o r modern instruments, analyses o f compounds a t the nanogram l e v e l can be considered routine f o r e i t h e r technique. Lower l i m i t s o f detection are a v a i l a b l e w i t h s p e c i a l a t t e n t i o n t o the experiment. For example, i n GC/MS, selected i o n monitoring i s used t o increase s e n s i t i v i t y . In t h i s experiment, the mass analyzer no longer scans across the f u l l mass range, but rather integrates signal* i n a few mass windows corresponding t o ions o f i n t e r e s t . The e f f e c t i v e r e s o l u t i o n o f the chromatographic separation i s u s u a l l y increased. The selected i o n monitoring technique i s u s e f u l when compounds o f p a r t i c u l a r i n t e r e s t are known t o produce c h a r a c t e r i s t i c ions i n the source. Thus Harvey {!) demonstrated that the t r i m e t h y l s i l y l (IMS) d e r i v a t i v e s o f diphenylpropanoids form c h a r a c t e r i s t i c ions a t m/z 266 (Scheme 1). S e t t i n g the mass analyzer o f the spectrometer t o pass only m/z 266 pinpoints the e l u t i o n o f such compounds from the gas chromatographic column. S e n s i t i v i t y f o r selected i o n monitoring experiments t y p i c a l l y i s reported i n the low picogram l e v e l , although i n favorable cases, low femtogram s e n s i t i v i t y can be achieved. In MS/MS, the selected i o n monitoring experiment i s transformed i n t o selected r e a c t i o n monitoring. Both mass analyzers are set t o pass s p e c i f i e d parent ion/daughter i o n p a i r s . As i n selected i o n monitoring, there i s a time advantage as the mass analyzers are not scanned but rather integrate s i g n a l . There i s an added s p e c i f i c i t y over selected i o n monitoring i n t h a t both the reactant and the product are s p e c i f i e d . I f sample i n t r o d u c t i o n i s v i a the d i r e c t probe, the v a p o r i z a t i o n p r o f i l e provides a t h i r d parameter v i a which the compound can be i d e n t i f i e d . MS/MS has been used f o r the i d e n t i f i c a t i o n o f targeted compounds i n complex mixtures a t the nanogram l e v e l (8). A t lower l e v e l s , matrix constituents a f f e c t the p r e c i s i o n of the response, n e c e s s i t a t i n g e i t h e r higher r e s o l u t i o n measurements or sample cleanup. The l a t t e r route has been s u c c e s s f u l l y pursued under the guise o f GC/MS/MS down t o l e v e l s o f 10-100 pg i n pharmaceutical a p p l i c a t i o n s (£).

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

127

128

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

S p e c i f i c i t y . Yost (10) has compared the r e l a t i v e informing power of GC/MS and MS/MS. The informing power o f an a n a l y t i c a l procedure i s expressed mathematically i n terms o f the number o f " b i t s . " A c a p i l l a r y gas chromatography column o f 1 0 p l a t e s combined w i t h a quadrupole mass spectrometer w i t h u n i t mass r e s o l u t i o n up t o 1000 daltons provides an informing power o f 6.6 χ 1 0 b i t s . An MS/MS instrument comprised o f two sequential quadrupole mass analyzers o f the same performance provides an informing power o f 1.2 χ b i t s . Within the l i m i t o f the assumptions made, the informing powers are i d e n t i c a l . There are, however, various experimental parameters i n MS/MS which can be used as a d d i t i o n a l r e s o l u t i o n elements. These include energy resolved, pressure resolved, and angle resolved MS/MS experiments ( 4 ) . These parameters are balanced by v a r i a t i o n o f gas chromatographic conditions, such as the stationary phase chosen, the temperature program followed, and a d d i t i o n a l steps o f sample p u r i f i c a t i o n and pretreatment. I n summary, both GC/MS and a n a l y t i c a l problems, bot c h a r a c t e r i s t i c o f such combined methods ( 1L1 ). 5

6

Speed o f Analysis. The introductory example focussed on the speed of MS/MS a n a l y s i s o f v o l a t i l e compounds. However, there are several aspects o f a n a l y t i c a l speed o f i n t e r e s t i n MS/MS. The f i r s t , and that which has received the most a t t e n t i o n , i s the time required f o r a n a l y s i s . The analogy between GC/MS and MS/MS involves the comparison o f r e t e n t i o n times through the gas chromatograph w i t h i o n f l i g h t times through the f i r s t mass analyzer. The former occupies between 10* and 10^ s, and the l a t t e r on the order o f microseconds. I n the s p e c i f i c example o f targeted compound a n a l y s i s i n a complex mixture, MS/MS can o f f e r a s i g n i f i c a n t time advantage i n the examination o f a large number o f samples. G l i s h has shown t h a t the a n a l y s i s o f mixtures using a preset protocol o f selected reaction monitoring can occur a t near the r a t e o f sample introduction i n t o the source o f the instrument (12). The time advantage o f MS/MS a l s o i s exemplified by the a b i l i t y to s e l e c t from a mixture o f ions i n the source any parent i o n , i n any order, and t o return as necessary t o that parent i o n f o r p r e c i s e measurements. This independence o f access p e r s i s t s f o r the duration o f the sample residence time i n the source. This i s i n marked contrast t o the s i t u a t i o n i n GC/MS, where each sample i s a v a i l a b l e f o r examination by the mass spectrometer only during the r e t e n t i o n time window. To repeat the measurement, the e n t i r e sample must be r e i n j e c t e d . The source residence time f o r most samples introduced v i a the d i r e c t i n s e r t i o n probe i s on the order of a minute o r two, depending on the temperature o f the source and the r a t e o f heating o f the probe t i p i t s e l f . The f i n a l aspect o f speed t o be considered i s the information f l u x . In MS/MS, the sample may be a v a i l a b l e f o r minutes rather than the seconds corresponding t o the width o f a gas chromatographic peak. I n the absence o f a preset protocol, experimental decisions must be made i n r e a l time. For example, what parent i o n should be selected f o r a daughter i o n MS/MS spectrum? Does the spectrum o f parent ions from the source change w i t h probe temperature? What should the c o l l i s i o n energy and

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

BUSCH AND KROHA

MS-MS Applied to Flavor Compound Characterization

pressure be? The newest generation o f data systems can make such decisions automatically w i t h i n c e r t a i n preset l i m i t s ( 13 ). The r a t e a t which information must be obtained and processed i n these l a t t e r MS/MS experiments i s much higher than t h a t i n a GC/MS experiment. Applications t o Flavor Compounds The use o f gas chromatography/mass spectrometry i n food and f l a v o r analyses i s now well-established, and reviews are p l e n t i f u l (14-16). By contrast, the use o f MS/MS i n t h i s area i s l e s s widespread. In part t h i s has been due t o the longer a v a i l a b i l i t y o f commercial GC/MS instruments as opposed t o MS/MS instruments, but a l s o i n no small p a r t due t o the enormous success o f the GC/MS method i t s e l f . Food and f l a v o r analyses deal perforce with the i d e n t i f i c a t i o n and q u a n t i t a t i o n o f v o l a t i l e components o f complex mixtures. Gas chromatography able t o separate such compound o f the mass spectrometer allows the i d e n t i f i c a t i o n o f the eluted compound a t the very low l e v e l s found i n many food and f l a v o r mixtures. As i n GC/MS, the analyst using MS/MS must be concerned w i t h sample handling ( c o l l e c t i o n , treatment, and œntandnation) and sample a n a l y s i s ( i o n i z a t i o n method and mass measurement). Sample c o l l e c t i o n , treatment, and contamination. In GC/MS, sample treatment i s o f t e n extensive. In preparation f o r GC/MS a n a l y s i s o f nutmeg, Harvey (7) ground 100 mg o f nutmeg t o a f i n e powder and extracted f o r an hour w i t h e t h y l acetate. The f i l t e r e d e x t r a c t was frozen t o p r e c i p i t a t e t r i g l y c e r i d e s , f i l t e r e d again, and then d e r i v a t i z e d overnight w i t h a standard t r i m e t h y l s i l y l a t i o n reaction. By contrast, i n the MS/MS a n a l y s i s o f nutmeg by Davis (17), 10-50 mg o f ground nutmeg are loaded i n t o a glass c a p i l l a r y , introduced d i r e c t l y i n t o the source o f the mass spectrometer, and vaporized by a short heating program. A constant concern i n the a n a l y s i s o f f l a v o r components i s a l t e r a t i o n and contamination o f the sample. Losses o f v o l a t i l e components are a major problem. The extensive sample preparation involved i n GC/MS o f f e r s ample opportunity f o r transformations and losses because o f sample handling and exposure t o chemical d e r i v a t i z i n g reagents. In MS/MS, sample handling i s o f t e n reduced and the chances f o r outside contantination minimized. Sample carryover, a problem during e x t r a c t i o n procedures f o r GC/MS, i s not eliminated i n MS/MS, but evolves i n t o a problem o f source œntamination. This problem was severe i n some e a r l y MS/MS work, but now seems under c o n t r o l w i t h the use o f removable i o n volumes i n the source. The concentration and homogenization o f sample that occurs i n GC/MS pretreatment i s not a v a i l a b l e i n MS/MS. Sample inhomogeneities thus become o f much greater concern. Sample t o sample v a r i a t i o n i s already f a i r l y high i n samples o f n a t u r a l o r i g i n , and the extensive a p p l i c a t i o n o f MS/MS may require more c a r e f u l sampling procedures than c u r r e n t l y employed. For t r a c e analyses, a simple form o f sample pretreatment i s often employed i n MS/MS t o concentrate the sample and t o preserve the c l e a n l i n e s s o f the i o n i z a t i o n source.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

129

130

CHARACTERIZATION A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

Sample i o n i z a t i o n . Requirements f o r sample i o n i z a t i o n are much more severe i n MS/MS than i n GC/MS. For MS/MS, the i o n i z a t i o n method should create one i o n f o r each component, and the structure o f the i o n should be the same as that o f the neutral surrogate. Electron i o n i z a t i o n u s u a l l y does not f u l f i l l these requirements, since the ions formed often include those from rearrangement reactions, and the degree o f fragmentation i s excessive. Chemical i o n i z a t i o n provides the r e q u i s i t e s i n g l e i o n f o r each ccrnponent o f the matrix i n the form o f the quasimolecular i o n (M+H) . However, chemical i o n i z a t i o n i s s e n s i t i v e t o source parameters and matrix e f f e c t s , and these problems are exacerbated by the d i r e c t introduction o f a complex mixture i n t o the source. The e f f e c t s can be compensated t o some degree by the use o f an i s o t o p i c a l l y l a b e l l e d i n t e r n a l standard f o r q u a n t i t a t i v e work. In the analysis o f unknowns i n complex mixtures, the nature o f the source chemistry should be a constant concern. +

Sample D e r i v a t i z a t i o n . Th described by Harvey (7) i s designed t o increase the v o l a t i l i t y and s t a b i l i t y o f the components so that they can be separated i n the gas chromatograph. With d i r e c t probe introduction, MS/MS i s u s u a l l y able t o deal w i t h samples o f lower v o l a t i l i t y ; hence, d e r i v a t i z a t i o n i s not required. D i r e c t probe temperatures reach as high as 400° C, vaporizing many samples d i r e c t l y i n t o the vacuum o f the mass spectrometer source. D e r i v a t i z a t i o n i s used i n MS/MS for the sanewhat d i f f e r e n t purpose o f imparting a s p e c i f i c chemical r e a c t i v i t y t o the analyte. Consider the t r i m e t h y l s i l y l d e r i v a t i v e s often used t o increase v o l a t i l i t y and s t a b i l i t y . The electron i o n i z a t i o n mass spectra o f these d e r i v a t i v e s often contain fragment ions such as the t r i m e t h y l s i l y l c a t i o n i t s e l f IMS*, o r fragment ions due t o losses o f neutral species containing the t r i m e t h y l s i l y l moiety. The same fragmentation reactions are expected i n the MS/MS spectra o f these d e r i v a t i v e s . Treatment o f a mixture w i t h a s i l y l a t i n g reagent converts free hydroxyl groups t o t h e i r -0TMS d e r i v a t i v e s , and then a second l a b e l l e d s i l y l a t i n g reagent converts amino groups t o t h e i r -NH-d9lMS d e r i v a t i v e s . A parent i o n scan o f IMS* (at m/z 73) pinpoints a l l o f the precursor ions that contained a free hydroxyl group. A parent i o n scan o f dg-TMS* pinpoints the precursor ions w i t h a r e a c t i v e amino group. Ctammon ions i n the two MS/MS spectra represent molecules t h a t contain both r e a c t i v e groups. A d e r i v a t i z a t i o n scheme i n v o l v i n g a constant neutral loss MS/MS scan has been described by Zakett (18). Phenols and amines react w i t h a c e t y l c h l o r i d e t o form acylated d e r i v a t i v e s which commonly lose the neutral fragment ketene i n the MS/MS reaction. A neutral l o s s scan f o r the l o s s o f 42 daltons w i l l thus i n d i c a t e the molecular weights o f any compound which has undergone d e r i v a t i z a t i o n . The strategy was used s u c c e s s f u l l y i n the analysis o f these f u n c t i o n a l groups i n a synthetic f u e l sample (18). Applications t o f l a v o r compounds have not y e t been reported, but w i l l undoubtedly be extensively exploited considering the d i v e r s i t y o f d e r i v a t i z a t i o n chemistry already developed.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

BUSCH AND KROHA

MS-MS Applied to Flavor Compound Characterization

Applications Food aromas. Labows and Shushan (19) have reviewed the d i r e c t a n a l y s i s o f food aromas by a commercial MS/MS system using an atmospheric pressure i o n i z a t i o n source. The sample i n l e t i s a simple a l l - g l a s s device that c o l l e c t s v o l a t i l e components emitted by food materials and d i r e c t s them i n t o the mass spectrometer. Losses due t o sample preparation are minimized, as are absorption or decomposition problems associated w i t h chromatographic f r a c t i o n a t i o n . P r o f i l e s of aroma compounds obtainedfcyt h i s method are claimed t o be more accurate than those obtained using other a n a l y t i c a l methods. Because o f the high d i s c r i m i n a t i o n against chemical noise i n the MS/MS system, detection l i m i t s can be very low, reported as 0.5 ppb f o r ethyl butyrate, 0.8 ppb f o r l i n a l o o l , and 45 ppb f o r limonene. These l i m i t s were established w i t h daughter i o n MS/MS spectra. Figure 3 shows the spectrum o f authentic nootkaton as the parent ion) and the i o n a t the same mass emitted d i r e c t l y from a g r a p e f r u i t . The match between the two spectra confirm the presence of t h i s targeted compound i n the emitted v o l a t i l e s . I t a l s o has been i d e n t i f i e d i n the v o l a t i l e s frcm i n t a c t oranges. The experiment can be completed i n l e s s than a minute, without sample preparation. Note that the u n i t mass r e s o l u t i o n o f the t r i p l e quadrupole instrument allows an accurate assignment o f abundances f o r daughter ions o f adjacent mass i n t h i s MS/MS spectrum. A c l o s e examination o f the spectra show t h a t the match between the authentic and the target compound i s not p e r f e c t . E i t h e r the instrumental parameters were not constant, or there i s an a d d i t i o n a l component a t m/z 219 i n the v o l a t i l e s emitted by the g r a p e f r u i t . I t i s a t t h i s stage that a simple p r e f r a c t i o n a t i o n experiment, or an a l t e r n a t i v e i o n i z a t i o n method becomes necessary t o e s t a b l i s h the number o f components present a t t h i s mass. Other MS/MS experiments were used t o give information u s e f u l f o r functional group c h a r a c t e r i z a t i o n . Fragmentation t o m/z 18 (NH4" ") i s i n d i c a t i v e o f amines, and m/z 19 (H30 ) i s a t y p i c a l fragment i o n from alcohols. Thus a parent i o n scan f o r these daughter ions pinpoints compounds o f these groups i n the emitted v o l a t i l e s . Acetate esters produce daughter ions a t m/z 43 and m/z 61. A parent i o n scan f o r the l a t t e r produces the daughter i o n MS/MS spectrum shown i n Figure 4, which i s the sum of a l l the parent ions o f a l l of the acetate esters i n the v o l a t i l e s emitted from a banana. The base peak a t m/z 131 represents the parent i o n o f isoamylacetate, known t o have the c h a r a c t e r i s t i c banana odor. Figure 5 i s the neutral l o s s scan f o r l o s s o f 44 daltons, a f a m i l i a r loss from negative ions of c a r b o x y l i c acids. Thus the ions a t m/z 87, 89, and 121 are most l i k e l y from butanoic o r pyruvic a c i d , l a c t i c a c i d , and benzoic a c i d , respectively. The i d e n t i t y o f these ions are confirmed by examining the daughter i o n spectra of the authentic compounds and the peaks obtained i n the d i r e c t a n a l y s i s o f the sample, i n t h i s case a Teewurst sausage. D i r e c t analyses o f v o l a t i l e s has been suggested as a means o f screening food products that might otherwise pass a g r i c u l t u r a l borders. The s e n s i t i v i t y seems t o be s u f f i c i e n t l y high f o r t h i s purpose, and the MS/MS a n a l y s i s possesses the r e q u i s i t e speed and 4

+

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

131

132

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Scheme 1.

2 1 9 * MS/MS ORANGE RA

m/z RA

m/z NOOTKATONE 219* MS/MS

Figure 3. Daughter i o n MS/MS spectra o f suspected nootkatone emitted frcm g r a p e f r u i t and the authentic compound (19). PARENTS OF 6 1 *

m/z

Figure 4. Parent i o n MS/MS spectrum f o r acetate esters emitted from a banana, representing an example o f f u n c t i o n a l group screening by MS/MS ( 1 9 ) .

m/z

Figure 5. Neutral l o s s MS/MS spectrum f o r l o s s o f 44 daltons (carbon dioxide) which pinpoints (M-H)" molecular ions f o r carboxylic acids. L a c t i c a c i d and benzoic a c i d are i d e n t i f i e d a t 89"" and 121", r e s p e c t i v e l y , although these are not the l a r g e s t peaks i n the spectrum ( 1 9 ) .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

MS-MS Applied to Flavor Compound Characterization

BUSCH AND KROHA

s p e c i f i c i t y f o r r e a l time a n a l y s i s . A comprehensive study o f the interferences that might be expected i n such use wDuld be needed t o evaluate the s u i t a b i l i t y o f t h i s technique. Spice a n a l y s i s . Davis has studied the composition o f nutmeg using MS/MS (17). Nutmeg has been extensively studied because o f the large number o f psychoactive species alleged t o be present. This study i s noteworthy because o f the use o f both h i g h energy and low energy MS/MS t o acquire daughter i o n spectra f o r the various compounds contained w i t h i n the nutmeg, and the use o f programmed thermal desorption from the d i r e c t i n l e t probe o f the mass spectrometer i n order t o e f f e c t a crude d i s t i l l a t i o n o f the sample. Isobutane was used as the reagent gas i n order t o ensure t h a t most o f the constituents form protonated molecular ions (M+H) i n the mass spectrum w i t h a minimum o f fragmentation. Charge exchange was used as an a l t e r n a t i v e method of i o n i z a t i o n i n order t o form parent ion daughter i o n spectra o The isobutane chemical i o n i z a t i o n mass spectrum o f nutmeg obtained a t a probe temperature o f \SO°C d i f f e r s from that obtained a t 200°C. Higher mass v o l a t i l e s are not evaporated i n t o the source u n t i l the temperature o f the probe i s elevated t o the higher temperature. As w i t h many analyses o f t h i s type, the amount o f sample i s not a l i m i t i n g factor. The temperature can thus be held steady f o r several minutes a t a given value, allowing several independent MS/MS experiments t o be completed. At higher probe temperatures, thermal degradation o f the sample can become a problem. Comparison o f the daughter i o n MS/MS spectra o f authentic 4-allyl-2,6-dimethoxyphenol a t m/z 195 and the same mass i o n from the nutmeg sample i s presented i n Figure 6. The spectra are s u f f i c i e n t l y s i m i l a r that the presence o f t h i s compound i n nutmeg can be confirmed. Of p a r t i c u l a r value i s the sharp charge s t r i p p i n g peak a t 97.5 on the mass scale. This i s the product o f an o x i d a t i o n r e a c t i o n o f the s i n g l y charged 195 t o the doubly-charged 1 9 5 as a r e s u l t o f the h i g h energy c o l l i s i o n . This r e a c t i o n occurs frequently with nitrogen- and oxygen-containing compounds. Two points should be noted. F i r s t , the match, although close, i s not exact. This indicates that not a l l o f the i o n current a t t h i s mass i s due t o t h i s compound alone, as i n the case described above. Secondly, the width o f the peaks f o r the daughter ions are very wide, compromising both the assignment o f masses and the r e l a t i v e abundances. This i s a consequence o f the instrument used, which was a reverse-geometry sector instrument (20). Daughter i o n a n a l y s i s i s accomplished w i t h a k i n e t i c energy analyzer, which mirrors the k i n e t i c energy release observed as a consequence o f the fragmentation reaction. Although t h i s value can be used as a probe o f the mechanism o f the fragmentation i t s e l f , i t i s a disadvantage i n MS/MS work f o r these reasons. I t i s known that nutmeg contains diphenylpropanoids o f c y c l i c and a c y c l i c forms. The a c y c l i c form fragments t o c h a r a c t e r i s t i c daughter ions a t m/z 193, and the c y c l i c form t o daughter ions a t m/z 203 (Scheme 2). These daughter ions can be set as products i n a parent i o n scan t o examine the e n t i r e nutmeg mixture f o r parent ions o f these classes o f diphenylpropanoids. Figure 7 shows the +

+

+

2+

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

133

134

CHARACTERIZATION A N D M E A S U R E M E N T O F FLAVOR C O M P O U N D S

1 7

g

RA

NUTMEG 195+

MS/MS

165 154

LA RA

ι

m/z

4-ALLYL-2.5-DIMETHOXYPHENOL 195* 165 MS/MS 154

179

JL

m/z

Figure 6. Daughter i o n MS/MS spectrum o f the protonated molecular ion from 4 - a l l y l - 2 , 6-dimethoxyphenol as an authentic compared t o the spectrum obtained from an i o n o f the same mass formed d i r e c t l y from a nutmeg sample ( 17).

OCH. 203**

CYCLIC OCH

QCH,

a

OCH, 193*

1

ACYCLIC Scheme

193

PARENT ION MS/MS PARENTS OF 193 +

ACYCLIC

RII I 203

2.

357 371 I 401

PARENT ION MS/MS PARENTS OF 203 +

CYCLIC 327 357

3

7

!

221 _i

Figure 7 . Parent i o n scans f o r two isomeric forms o f diphenylpropanoids found i n nutmeg. The common ions a t m/z 355, 357, and 371 i n d i c a t e the presence o f both forms o f the compound a t those masses (17).

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

BUSCH AND KROHA

MS-MS Applied to Flavor Compound Characterization

r e s u l t of t h i s experiment. Parent ions at 357 , 371 , 387 , and 4 0 1 are indicated t o be a c y c l i c diphenylpropanoids. cyclic forms are i n d i c a t e d i n the parent i o n scan at 327*, 341 , 355 , 357+, 371 , and 375 . The common parent ions 355 , 357 , and 371 are c l e a r l y i n d i c a t e d as c o n s i s t i n g of both forms of diphenylpropanoid structures. The parent i o n at 355 i s thought t o be a dehydrodiphenylpropanoid d e r i v a t i v e o f m y r i s t i c i n , i d e n t i f i e d f o r the f i r s t time i n nutmeg. +

+

+

+

+

+

+

+

+

Problems and P o t e n t i a l s o f MS/MS Problems. The simultaneous i n t r o d u c t i o n of a l l the components o f a mixture i n t o the source o f the mass spectrometer c o n s t i t u t e s one o f the strengths o f MS/MS, but i s a l s o the cause of several problems. F i r s t i s the problem o f sample carryover and source c l e a n l i n e s s . One o f the tenets o f normal operation o f a mass spectrometer i s t o introduce as l i t t l e sampl s i z e s are often quite larg Newer instruments are designed f o r easier source cleaning or u t i l z e i o n volumes which are replaceable through the d i r e c t probe i n l e t . I t i s p o s s i b l e t o change the e n t i r e source volume w i t h each sample and thus minimize t h i s problem. Less e a s i l y ameliorated i s the phenomenon known as the "matrix e f f e c t " . I d e a l l y , the chemical i o n i z a t i o n source produces one i o n f o r each constituent o f the mixture, and the r e l a t i v e abundances o f the ions formed are i n proportion t o the amount o f constituent present. The "matrix e f f e c t " i s a term t h a t describes enhancement or suppression o f i o n s i g n a l f o r a s i n g l e component due t o the presence of the matrix. This i s an i n s i d i o u s problem because the matrix i s not characterized, and may change from sample t o sample. For targeted compound a n a l y s i s , the usual s o l u t i o n i s t o employ an i s o t o p i c a l l y l a b e l l e d i n t e r n a l standard that i s introduced i n t o the mixture as a whole. Quantitation o f the s i g n a l o f i n t e r e s t i s then derived from the r e l a t i v e abundances o f the ions corresponding t o the l a b e l l e d and unlabelled forms o f the analyte. Since the unlabel l e d / l a b e l l e d i o n p a i r p e r s i s t s i n many o f the daughter ions formed by c o l l i s i o n , several confirming r a t i o s can be obtained i n a s i n g l e experiment. The matrix e f f e c t a l s o may be evident i n chemical noise which p e r s i s t s i n the spectrum, and i s f a r greater i n analyses near the detection l i m i t . This was h i g h l i g h t e d i n the paper by Bursey (21) i n which a matrix e f f e c t i n the determination of polychlorinated organic compounds was found. D i r e c t probe MS/MS r e s u l t s were systematically high compared w i t h those from GC/MS o r GC/MS/MS. The r a p i d throughput p o s s i b l e i n an MS/MS screening protocol i s not obtained without concomitant r i s k . MS/MS i s an empirical method of a n a l y s i s . As i s evident from the examples presented, the i n t e r p r e t a t i o n of a daughter i o n MS/MS spectrum i s o f t e n based on the same c o r r e l a t i o n p r i n c i p l e s derived from e l e c t r o n and chemical i o n i z a t i o n mass spectrometry. More often, the comparison o f the spectrum obtained t o t h a t o f the authentic compound i s used f o r i d e n t i f i c a t i o n . This i s a fundamentally u n s a t i s f y i n g procedure. While e l e c t r o n and chemical i o n i z a t i o n spectra can be compared t o a s p e c t r a l l i b r a r y which has been compiled over the past t h i r t y years, no comparable l i b r a r y o f MS/MS spectra e x i s t s . Data systems may be used w i t h i n i n d i v i d u a l

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

135

136

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

laboratories t o create l i b r a r i e s o f s p e c t r a l data which are then searched i n the usual manner, but n a t i o n a l and i n t e r n a t i o n a l databases are nonexistent. Several obstacles must be overcome t o create such bases. F i r s t i s the d i v i s i o n o f MS/MS spectra i n t o those r e s u l t i n g from high energy c o l l i s i o n processes (as on sector mass spectrometers) and those obtained under low energy c o l l i s i o n conditions as p r e v a i l on the m u l t i p l e quadrupole instruments. The spectra thus obtained are often s i m i l a r , but perhaps not t o the p o i n t where a cross c o r r e l a t i o n can be drawn. Charge s t r i p p i n g and charge i n v e r s i o n are processes t h a t are confined l a r g e l y t o high energy MS/MS spectra, f o r example. The l e s s than u n i t mass r e s o l u t i o n o f the reverse geometry sector instruments f o r daughter ion MS/MS spectra WDuld be a problem i n reducing spectra i n t o a standard l i b r a r y form, as both the masses and the r e l a t i v e abundances o f the daughter ions are known w i t h a l i m i t e d c e r t a i n t y . F i n a l l y , the low energy daughter i o n MS/MS spectra are affected by instrument parameter the c o l l i s i o n gas pressure operation has been accepted. While these s p e c t r a l e f f e c t s are valuable i n e x t r a c t i n g a d d i t i o n a l information frcm the MS/MS spectrum, they do represent a s i g n i f i c a n t obstacle t o the standardization of MS/MS l i b r a r i e s . The most hopeful d i r e c t i o n might be i n advanced data systems w i t h memory s u f f i c i e n t t o accept a l l spectra obtained, and searching algorithms sophisticated enough to deal w i t h m u l t i p l e spectra o f a s i n g l e compound. As mentioned e a r l i e r , q u a n t i t a t i o n w i t h MS/MS i s often c a r r i e d out w i t h i n t e r n a l standards. Without standards, the accuracy and p r e c i s i o n o f q u a n t i t a t i o n i s reduced due t o matrix e f f e c t s . Rough estimates can be q u i c k l y obtained with MS/MS. For many problems, t h i s information i s more than s u f f i c i e n t . For instance, new drugs are o f t e n derived d i r e c t l y frcm p l a n t m a t e r i a l . One o f the f i r s t questions asked i s the r e l a t i v e concentration o f the desired material i n the various p l a n t p a r t s . MS/MS has been used t o provide the approximate amounts o f the targeted compound i n roots, stems, p e t a l s , o r flowers. The p l a n t t i s s u e w i t h the highest concentration o f compound i s then extracted. In food and f l a v o r analyses, the accuracy of the q u a n t i t a t i v e data required from MS/MS may be l i m i t e d by the v a r i a b i l i t y of the sample i t s e l f . In a n a l y s i s o f a large number o f samples, MS/MS provides a quick i n d i c a t i o n o f the amounts o f compounds o f i n t e r e s t . I f v a r i a b i l i t y f a l l s outside o f a preset tolerance, then only those samples are flagged f o r more exhaustive workup and a more rigorous q u a n t i t a t i v e a n a l y s i s . This a b i l i t y t o focus a n a l y t i c a l resources on samples o f i n t e r e s t i s a valuable property of the MS/MS experiment. P o t e n t i a l . The development o f MS/MS f o r analyses o f foods and f l a v o r s w i l l f o l l o w the same growth curve as i t has i n other a p p l i c a t i o n s . A t t h i s point, only the f i r s t part o f the growth curve i s evident. Several i n d u s t r i a l l a b o r a t o r i e s are beginning t o use MS/MS on a routine b a s i s , and commercial pressure w i l l d r i v e the exp>ansion o f the method. The speed o f the MS/MS a n a l y s i s i s a strong i n i t i a l advantage. I n the long term, i t i s l i k e l y t o be the f l e x i b i l i t y of MS/MS a n a l y s i s t h a t w i l l s u s t a i n i t s use i n these areas, and j u s t i f y the high i n i t i a l cost o f the instrument. I t i s

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

9.

BUSCH AND KROHA

MS-MS Applied to Flavor Compound Characterization

137

only a matter o f programming the experiment that d i f f e r e n t i a t e s the a n a l y s i s o f nutmeg from the a n a l y s i s o f acetate esters emitted as v o l a t i l e s frcm f r u i t . A l l these protocols can be developed and then stored w i t h i n the data system as standard methods o f a n a l y s i s , and then c a l l e d up as needed. The a b i l i t y o f MS/MS t o search f o r classes o f compounds i n a mixture w i l l be as valuable i n food and f l a v o r analyses as i t i s i n other complex mixture analyses, such as the pharmaceutical o r environmental f i e l d s . Flavors are complex mixtures, but o f t e n c o n s i s t o f groups o f chemically s i m i l a r compounds. I t i s p r e c i s e l y the i d e n t i f i c a t i o n o f these groups f o r which parent i o n and n e u t r a l l o s s MS/MS experiments are p a r t i c u l a r l y adept. This i s a c h a r a c t e r i s t i c that i s p a t e n t l y not a v a i l a b l e w i t h GC/MS, which has been the usual method o f analyses o f these mixtures. Literature Cited 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

19. 20. 21.

Kruger, T. L.; Litton, Anal. Chem. 1976, 48, 2113. Kbndrat, R. W.; McClusky, G. A.; Cooks, R. G. Anal. Chem. 1978, 50, 1222. Kruger, T. L.; Litton, J. F.; Cooks, R. G. Anal. Lett. 1976, 9, 533-542. McLafferty, F. W., Bd., "Tandem Mass Spectrometry"; Wiley: New York, 1983. Cooks, R. G., Beynon, J. H.; Caprioli, R. M.; Lester, G. R., "Metastable Ions"; Elsevier: Amsterdam, 1973. Kbndrat, R. W.; Cooks, R. G. Anal. Chem. 1978, 50, 1251A. Harvey, D. J. J. Chrom. 1975, 110, 91-102. Plattner, R. D.; Yates, S. G.; Porter, J. K. J. Agric. Food Chem. 1983, 31, 785-789. Richter, W. J.; Blum, W.; Schlunegger, U. P.; Senn, M. In "Tandem Mass Spectrometry"; McLafferty, F. W., Ed.; Wiley: New York, 1983. Fetterolf, D. D. y Yost, R. A. Int. J. Mass Spectrom. Ion Proc. 1984, 62, 33-50. Hirschfeld, T. A. Anal. Chem. 1980, 52, 297A. Glish, G. L.; Shaddock, V. M.; Harmon, K.; Cooks, R. G. Anal. Chem. 1980, 52, 165-167. Kirby, H.; Sokolow, S.; Steiner, S. In "Tandem Mass Spectrometry"; McLafferty, F. W., Ed.; Wiley: New York, 1983. Issenberg, P.; Kobayashi, A.; Mysliwy. T. J. J. Agric. Food Chem. 1969, 17, 1377-1386. Horman, I. Gazz. Chim. Italiana 1984, 114, 297-303. Horman, I. Bicmed. Mass Spectrom. 1981, 8, 384. Davis, D. V.; Cooks, R. G. J. Agric. Food Chem. 1982, 30 495-504. Zakett, D,; Cooks, R. G. In "New Appraoches in Coal Chemistry"; Blaustein, B. D.; Bockrath, B. C.; Friedman, S., Eds.; ACS SYMPOSIUM SERIES No. 169, American Chemical Society: Washington, D. C., 1981; pp. 267-288. Labows, J. N.; Shushan, B. Amer. Lab. 1983, 15(3), 56-61. Beynon, J. H.; Cooks, R. G.; Amy, J. W.; Baitinger, W. E.; Ridley, R. Y. Anal. Chem. 1973, 45, 1023A. Voyksner, R. D.; Hass, R. D.; Sovocool, G. W.; Bursey, M. M. Anal. Chem. 1983, 55, 744-749.

RECEIVED June 24, 1985 In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

10 Automated Analysis of Volatile Flavor Compounds Robert G. Westendorf Tekmar Company, Cincinnati, OH 45222-1856

Volatile organic compounds present in foods have a significant impact on flavor quality. The analysis of these compounds can be quite difficult, since th sampl i ofte t amenabl to direct GC injection present in very being important to flavor. The technique of dynamic headspace sampling was used for the isolation and concentration of volatiles prior to analysis by gas chromatography. Samples, which may be heated, are purged with an inert gas, sweeping any volatile compounds present out of the sample. The volatiles are trapped on Tenax, which is then thermally desorbed and backflushed to inject the sample into the GC. Using new instrumentation, this method was fully automated. Samples run include fruits, fruit products, edible oils, and oil-based foods. Detection limits in the low part-per-billion range were obtained with 2-8% reproducibility. The i n s t r u m e n t a l a n a l y s i s o f f l a v o r i n a food m a t e r i a l can be an e x t r e m e l y d i f f i c u l t t a s k . There a r e many f a c t o r s t h a t i n f l u e n c e f l a v o r . Of t h e s e , one o f the more i m p o r t a n t , y e t a l s o most d i f f i c u l t t o a n a l y z e , i s the p r o f i l e o f v o l a t i l e o r g a n i c compounds p r e s e n t . The d i f f i c u l t y a r i s e s from the f a c t t h a t t h e r e may be many v o l a t i l e s p r e s e n t a t v e r y low c o n c e n t r a t i o n s i n a c o m p l i c a t e d m a t r i x . A v a r i e t y o f methods have h i s t o r i c a l l y been used f o r t h i s a n a l y s i s . The m a j o r i t y o f these methods have u t i l i z e d gas chromatography (GC), d i f f e r i n g i n chromatographic parameters and sample p r e p a r a t i o n t e c h n i q u e s . Chromatographic systems have e v o l v e d tremendously i n r e c e n t y e a r s . Column t e c h n o l o g y has advanced t o a v e r y h i g h l e v e l o f s e p a r a t i o n power. Sample p r e p a r a t i o n t e c h n i q u e s , on the o t h e r hand, have not e v o l v e d as r a p i d l y as GC t e c h n o l o g y . A number o f d i f f e r e n t p r e p a r a t i o n t e c h n i q u e s have been used f o r the a n a l y s i s o f f l a v o r 0097-6156/85/0289-0138$06.00/0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

10.

WESTENDORF

Automated Analysis of Volatile Flavor Compounds

139

v o l a t i l e s , i n c l u d i n g s o l v e n t e x t r a c t i o n , steam d i s t i l l a t i o n , e q u i l i b r i u m headspace s a m p l i n g , and dynamic headspace s a m p l i n g . Of these methods, dynamic headspace sampling i s p r o b a b l y the l e a s t w e l l known, y e t i t has a number of advantages over o t h e r t e c h n i q u e s i n use. Methods u t i l i z i n g dynamic headspace t e c h n i q u e s were r e p o r t e d as e a r l y as 1960 (_1 ). These methods g e n e r a l l y u t i l i z e d c r y o g e n i c t r a p s , or c r y o g e n i c a l l y c o o l e d t r a p s c o n t a i n i n g column p a c k i n g materials (2) or molecular sieves ( 3 ) . Dynamic headspace a n a l y s i s (DHA), a l s o know as purge and t r a p a n a l y s i s , u t i l i z i n g new porous polymers as t r a p p i n g agents and i n c o r p o r a t i n g m u l t i - p o r t v a l v e s f o r f l o w s w i t c h i n g was i n t r o d u c e d f o r the a n a l y s i s of o r g a n i c c o n t a m i n a n t s i n water i n 1974 ( 4 ) . Various forms of DHA have been used f o r a v a r i e t y of food samples. F r u i t s and j u i c e s have been i n v e s t i g a t e d by a number of r e s e a r c h e r s . Schamp and D i r i n c k (_5) found over f o r t y compounds i n a study of s t r a w b e r r y v a r i e t i e s i n Golden D e l i c i o u s a p p l e s the GC/MS i d e n t i f i c a t i o n of over twenty d i f f e r e n t compounds i n s e v e r a l v a r i e t i e s of a p p l e s and a p p l e p r o d u c t s . A d d i t i o n a l work w i t h a p p l e s has been r e p o r t e d by Westendorf (]_) » T h i s same paper r e p o r t e d the a n a l y s e s of d a i r y p r o d u c t s , v e g e t a b l e o i l s , and a r t i f i c i a l flavors. E d i b l e o i l s have been e x t e n s i v e l y s t u d i e d s i n c e the p r e s e n c e of v o l a t i l e s was f i r s t r e c o g n i z e d as an i n d i c a t o r of o i l q u a l i t y ( 8 ) . C o n s i d e r a b l e work w i t h o i l s and o i l - b a s e d foods u s i n g manual p r o c e d u r e s has been r e p o r t e d by J a c k s o n e t a l . ( 9 , H ) ) , Dupuy e t a l . (JUL ,_12), and S e l k e (_13). I n 1983 R o b e r t s ( 1 4 ) f i r s t r e p o r t e d the use o f an automated DHA procedure f o r o i l v o l a t i l e s . P r i n c i p l e of O p e r a t i o n DHA i s based on the p a r t i t i o n i n g of v o l a t i l e compounds between a sample and the vapor phase above the sample a t a r a t e dependent on a v a r i e t y of f a c t o r s . These i n c l u d e the v o l a t i l i t y of the s u b j e c t compound, i t s s o l u b i l i t y i n the sample m a t r i x , homogeneity of the m a t r i x , t e m p e r a t u r e , and sample c o n t a i n e r configuration. I n e q u i l i b r i u m headspace a n a l y s i s , the sample i s s e a l e d i n a c l o s e d v e s s e l and the v o l a t i l e s a r e a l l o w e d t o e q u i l i b r a t e between the sample and vapor phase. An a l i q u o t of the vapor phase i s then i n j e c t e d i n t o a GC f o r a n a l y s i s . In DHA, the sample i s purged w i t h an i n e r t gas, sweeping the v o l a t i l e s out of the sample c o n t a i n e r . The purge gas i s t h e n passed through a s h o r t column c o n t a i n i n g a porous polymer a d s o r b e n t which s e l e c t i v e l y r e t a i n s the sample compounds w h i l e a l l o w i n g the purge gas and any water vapor t o pass t h r o u g h . By p u r g i n g i n t h i s manner, the e n t i r e o r g a n i c c o n t e n t s o f the vapor phase can be s u b j e c t e d t o GC a n a l y s i s , not j u s t an a l i q u o t . In a d d i t i o n , s i n c e the purge gas i s c o n t i n u a l l y b e i n g removed from the sample the c o n c e n t r a t i o n of o r g a n i c s i n the vapor above the sample remains e s s e n t i a l l y z e r o . T h i s s i g n i f i c a n t l y enhances r e c o v e r y by p r o m o t i n g f u r t h e r p a r t i t i o n i n g of the v o l a t i l e s i n t o the vapor s t a t e . A f t e r the purge s t e p i s c o m p l e t e d , the a d s o r b e n t column i s h e a t e d t o r e l e a s e the o r g a n i c s and b a c k f l u s h e d v i a a 6 - p o r t v a l v e t o sweep the sample t o the GC. When u s i n g c a p i l l a r y

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

140

columns, a c r y o g e n i c f o c u s i n g s t e p i s employed t o sharpen the i n j e c t i o n p r o f i l e ( 1_5). S e p a r a t i o n and d e t e c t i o n are c a r r i e d out i n the GC, n o r m a l l y under temperature programmed c o n d i t i o n s . A diagram o f the f l o w scheme i s i l l u s t r a t e d i n F i g u r e 1. G o a l s . The p r i m a r y g o a l o f t h i s work was t o adapt new i n s t r u m e n t a t i o n t o the f u l l y automated a n a l y s i s o f v o l a t i l e f l a v o r compounds i n foods w i t h o u t compromising any o t h e r aspect of a n a l y t i c a l c a p a b i l i t y . The g o a l s o f t h i s work, i n a d d i t i o n t o automation, i n c l u d e d : 1. Recover the maximum range o f compounds p o s s i b l e , from v e r y v o l a t i l e gases t o h i g h e r b o i l i n g , l e s s v o l a t i l e compounds. 2. A c h i e v e the maximum s e n s i t i v i t y p o s s i b l e , p r e f e r a b l y t o p a r t - p e r - b i 1 1 i o n (ppb) l e v e l s , s i n c e many compounds have important o r g a n o l e p t i c q u a l i t i e s a t low l e v e l s . 3. A c h i e v e the b e s t r e p r o d u c i b i l i t y p o s s i b l e . 4. Keep a r t i f a c t f o r m a t i o v o l a t i l e compounds foun sample. 5. E l i m i n a t e c r o s s - c o n t a m i n a t i o n between samples. Achievement o f g o a l s 1-5 was c o n s i d e r e d n e c e s s a r y f o r the g o a l , t o t a l automation, t o be o f any p r a c t i c a l v a l u e .

primary

I n s t r u m e n t a t i o n f o r Automated A n a l y s e s . A l l samples were run u s i n g c o m m e r c i a l l y a v a i l a b l e automated DHA equipment i n t e r f a c e d to a m i c r o p r o c e s s o r GC. The DHA apparatus c o n s i s t s o f t h r e e p a r t s : the b a s i c c o n c e n t r a t o r (TEKMAR Model 4000), i n c o r p o r a t i n g the purge system, s w i t c h i n g v a l v e , and adsorbent; a t e n - p o s i t i o n automatic sampler (TEKMAR Model 4200); and a c a p i l l a r y column c r y o g e n i c t r a p (TEKMAR Model 1000). Under normal c o n d i t i o n s t h e o p e r a t o r f i r s t s e t s a l l o p e r a t i n g c o n d i t i o n s , loads the samples, r a i s e s the sample h e a t e r s ( h i g h performance e l e c t r i c m a n t l e s ) , and p l a c e s the i n s t r u m e n t s i n a u t o m a t i c mode. The automatic sampler advances t o the f i r s t p o s i t i o n and s i g n a l s the c o n c e n t r a t o r t o s t a r t . The prepurge, p r e h e a t , and purge steps are performed by the c o n c e n t r a t o r , which then sends a ready s i g n a l t o the c a p i l l a r y c r y o t r a p . When the c r y o t r a p a l s o r e c e i v e s a ready s i g n a l from the GC, i t w i l l c o o l t o a p r e s e t temperature and s i g n a l the c o n c e n t r a t o r t o s t a r t d e s o r b i n g the sample. When d e s o r p t i o n i s complete the c r y o t r a p w i l l heatup t o i n j e c t the sample and s i m u l t a n e o u s l y output s i g n a l s t o s t a r t the temperature program o f the GC and t o s t a r t data a c q u i s i t i o n on an i n t e g r a t o r . The i n t e g r a t o r w i l l a l s o read the sample p o s i t i o n number from the a u t o m a t i c sampler v i a a BCD i n t e r f a c e . While t h e GC run c o n t i n u e s the c o n c e n t r a t o r w i l l r e c o n d i t i o n and then c o o l the adsorbent t r a p . When the t r a p has c o o l e d the automatic sampler w i l l a g a i n advance and the above c y c l e w i l l r e p e a t . Experimental M a t e r i a l s . Samples were o b t a i n e d from a v a r i e t y o f s o u r c e s . F r u i t s and j u i c e s were purchased l o c a l l y . A l l o i l s , peanut b u t t e r , and food s t a r c h samples were s u p p l i e d by food p r o c e s s o r s . F l a v o r compounds used i n the p r e p a r a t i o n o f standards were

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

WESTENDORF

Automated Analysis of Volatile Flavor Compounds

Injection Port

F i g u r e 1:

Gas f l o w scheme of dynamic headspace gas chromatography system.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

142

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

purchased from A l d r i c h (Milwaukee, WI) and used w i t h o u t f u r t h e r purification. C o n c e n t r a t o r C o n d i t i o n s . The c o n c e n t r a t o r system used was a TEKMAR Model 4000 w i t h a TEKMAR Model 4200 t e n p o s i t i o n a u t o m a t i c sampler. A TEKMAR Model 1000 C a p i l l a r y I n t e r f a c e was used f o r i n t e r f a c e t o the GC. The o p e r a t i n g c o n d i t i o n s were as f o l l o w s : sample s i z e s - peanuts, f r u i t s : 1.00g, peanut b u t t e r , food starch: lOOmg, o i l s : 0.5ml, j u i c e s : 5.0ml. Sample temperatures - f r u i t s , j u i c e s , food s t a r c h : ambient (23 C ) , peanuts, peanut butter: 100 C, o i l s : 150 C. Prepurge 3.5 min. a t 50ml/min. n i t r o g e n , preheat 3 min. f o r 100 samples, 5 min. f o r 150 samples, purge 10 min. a t 50ml/min. n i t r o g e n . The t r a p was 12" X 1/8" s t a i n l e s s s t e e l packed w i t h 24cm (approx. 150mg) Tenax TA (Chrompack, B r i d g e w a t e r , N J ) . The t r a p was p r e h e a t e d at 175°C and then desorbed a t 180°C f o r 4 min. The t r a p was baked a t 225 C f o r 10 min sampler mounts were c o n t i n u o u s l C a p i l l a r y I n t e r f a c e was c o o l e d w i t h l i q u i d n i t r o g e n f o r c r y o t r a p p i n g t h e desorbed sample and heated f o r 10 seconds f o r GC i n j e c t i o n . Gas Chromatograph C o n d i t i o n s . The GC was a VARIAN 6000 w i t h a flame i o n i z a t i o n d e t e c t o r . The i n j e c t i o n p o r t was heated a t 200°, the d e t e c t o r a t 250°. The d e t e c t o r range was 10-12 AFS, a t t e n u a t i o n 64 except as noted. The column was 25m X 0.32mm f u s e d s i l i c a DB5 w i t h a 1.0 m i c r o n f i l m t h i c k n e s s (bonded SE54, J&W S c i e n t i f i c , Rancho Cordova, CA). The c a r r i e r gas was hydrogen a t 47 cm/s. The column was temperature programmed from an i n i t i a l temperature o f 35 C, h e l d f o r 4 min., t o a f i n a l temperature o f 200 C a t 4°C/min. R e s u l t s and D i s c u s s i o n V o l a t i l i t y Range. The range o f f l a v o r compounds r e c o v e r e d depends on a v a r i e t y o f f a c t o r s . The two p r i m a r y f a c t o r s a r e t h e sample m a t r i x and temperature. The m a t r i x can a f f e c t r e c o v e r y i n two ways. The f i r s t i s the s o l u b i l i t y o f t h e f l a v o r compounds i n the m a t r i x . Compounds t h a t have a poor s o l u b i l i t y w i l l be purged more e f f i c i e n t l y than compounds o f h i g h s o l u b i l i t y . The second i s t h a t v o l a t i l e s may be p h y s i c a l l y bound i n the sample. Flavor compounds may be p r e s e n t p r i m a r i l y i n t h e i n t e r i o r o f a sample, as i s the case w i t h c o f f e e beans. I n t h i s case t h e mass t r a n s f e r r a t e o f the v o l a t i l e s through t h e m a t r i x becomes a l i m i t i n g f a c t o r . F o r samples o f t h i s t y p e , g r i n d i n g o r o t h e r w i s e homogenizing the sample can s i g n i f i c a n t l y i n c r e a s e the r e c o v e r y . However, c a r e must be taken t o a v o i d the p o s s i b l e l o s s o r adulteration of v o l a t i l e s during t h i s process. A cryogenic g r i n d i n g procedure developed f o r p l a s t i c s (16) has been s u c c e s s f u l l y a p p l i e d t o food m a t e r i a l s i n the a u t h o r ' s laboratory. I n c r e a s i n g the temperature o f a sample s e r v e s t o i n c r e a s e r e c o v e r i e s by i n c r e a s i n g the vapor p r e s s u r e o f the v o l a t i l e compounds. The e f f e c t o f i n c r e a s i n g temperature on a c o r n o i l sample i s i l l u s t r a t e d i n F i g u r e 2. Temperature i s g e n e r a l l y the

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

WESTENDORF

i g u r e 2:

Automated Analysis of Volatile Flavor Compounds

Recovery of c o r n o i l v o l a t i l e s w i t h i n c r e a s i temperature.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

144

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

p r i m a r y v a r i a b l e used i n o p t i m i z i n g r e c o v e r y . The r e c o v e r i e s o f a number o f r e p r e s e n t a t i v e compounds from c o r n o i l a r e l i s t e d i n Table I .

Table I . R e p r e s e n t a t i v e Compounds i n Corn O i l , 150 i t e n t i o n Time 1.03 min. 1.36 2.12 3.67 3.75 5.33 6.16 6.24 6.58 7.67 9.07 9.15 12.06 14.79 14.94 16.79 17.68 ND: n o t determined

Compound Butane Pentane Hexane Heptane 1-Butanal 1-Pentanal Octane 1-Hexanal 2-0ctene(a trans-2-Hexena Nonane 1-Heptanal Decane Undecane Amyl A l c o h o l ( a ) 1-Hexanol(a) l-Nonanol(a)

Recovery 847o

Reproduc: 10.57o

82 74 65 47 50 61 51

2.9 5.3 3.6 3.6 4.9 4.0 4.9

55 43 50 41 ND ND ND

3.4 3.4 1.9 6.0 4.9 8.3 2.8

a: t e n t a t i v e

identification

A p o t e n t i a l problem w i t h h e a t i n g some samples i s t h e presence o f l a r g e amounts o f water vapor. The Tenax adsorbent used i s h y d r o p h o b i c , and does n o t t r a p w a t e r . However, water vapor w i l l condense on a c o l d Tenax t r a p . Excess amounts o f water c a n i n t e r f e r e w i t h an a n a l y s i s i n two ways. I f the amount o f water p r e s e n t i s e x t r e m e l y h i g h , i t w i l l condense on t h e Tenax i n sufficent quantity to p h y s i c a l l y block a s i g n i f i c a n t portion of the a v a i l a b l e t r a p p i n g s u r f a c e . T h i s l e a d s t o reduced t r a p p i n g e f f i c i e n c y , d e g r a d i n g both s e n s i t i v i t y and r e p r o d u c i b i l i t y . F o r samples o f t h i s t y p e , t h e sample temperature must n o t be r a i s e d above 65 t o 95 , depending on the water c o n t e n t o f t h e sample. At 60 , however, even aqueous samples can be run w i t h o u t l o s i n g t r a p p i n g e f f i c i e n c y ( Γ7 ). A second way i n which water may i n t e r f e r e i s i n t h e GC s e p a r a t i o n and d e t e c t i o n . Water may degrade t h e column used, o r i n t e r f e r e i n t h e d e t e c t i o n p r o c e s s . Samples t h a t have i n t r o d u c e d s u f f i c i e n t water i n t o t h e column t o e x t i n q u i s h t h e flame i o n i z a t i o n d e t e c t o r have been encountered i n the a u t h o r ' s l a b o r a t o r y . S i n c e water i s n o t t r a p p e d on t h e Tenax, i t i s p o s s i b l e t o remove most o f i t by p a s s i n g d r y n i t r o g e n through t h e t r a p b e f o r e t h e d e s o r p t i o n s t e p (18). The v o l a t i l i t y range o f t h e compounds t h a t can be a n a l y z e d by dynamic headspace c o n c e n t r a t i o n extends from o r g a n i c compounds t h a t a r e gases a t room temperature up t o compounds c o n t a i n i n g about t e n t o t h i r t e e n carbon atoms p e r m o l e c u l e , depending on t h e number and n a t u r e o f any s i d e c h a i n s . Table I l i s t s a number o f r e p r e s e n t a t i v e compounds i n a c o r n o i l . Table I I l i s t s some

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

10.

WESTENDORF

Automated Analysis of Volatile Flavor Compounds

a d d i t i o n a l compounds commonly encountered i n v e g e t a b l e o i l samples. Compounds t y p i c a l l y found i n f r u i t samples i n c l u d e p r i m a r i l y e t h y l e s t e r s w i t h a number o f aldehydes and a l c o h o l s i n the t h r e e t o n i n e carbon range. Table I I I l i s t s the GC/MS i d e n t i f i c a t i o n s o f the compounds encountered i n an apple sample (from r e f . 6 ) .

Table I I : A d d i t i o n a l Compounds Found i n V e g e t a b l e Benzaldehyde cis-2-trans-4-Decadienal 2-Decenal 2-Heptanone 1-Nonanal Octenal

Benzyl A l c o h o l trans-2-trans-4-Decadienal 2,4-Heptadienal M e t h y l E t h y l Ketone 2-Nonanone 1-Pentanol

Oils

1-Butenal 2-Decanone Diacetyl 1-Hexanol Octadiene

Table I I I : GC/MC I d e n t i f i c a t i o n o f Apple V o l a t i l e s (from r e f . 6) Compound

Compound

N-Propyl B u t y r a t e Butanal N-Butyl Propionate Ethyl Acetate N-Amyl A c e t a t e 1-Butanol M e t h y l Caproate N-Propyl A c e t a t e Ethyl-2-Methyl Butyrate M e t h y l Butanoate N-Butyl-N-Butyrate 2-Methyl B u t a n o l Ethyl-N-Caproate N-Hexanal N-Hexyl A c e t a t e Ethyl-N-Butanoate I s o p r o p y l Hexanoate N-Butyl A c e t a t e 1-Hexanol 2-Hexanal 2-Methyl B u t y l A c e t a t e N-Hexyl-N-Butyrate 2-Methyl-2-Methyl-Propyl Butyrate

S e n s i t i v i t y . The s e n s i t i v i t y o b t a i n a b l e depends p r i m a r i l y on the e f f i c i e n c y w i t h which a compound i s r e c o v e r e d from the sample. As w i t h the v o l a t i l i t y range, r e c o v e r y g e n e r a l l y i s most e f f e c t i v e l y i n c r e a s e d by r a i s i n g the sample temperature. A d d i t i o n a l f a c t o r s a f f e c t i n g s e n s i t i v i t y i n c l u d e t r a p p i n g and d e s o r p t i o n e f f i c i e n c i e s , column r e s o l u t i o n , i n t e r f e r e n c e s , and detector s e n s i t i v i t y . F o r o i l s the lower l i m i t o f d e t e c t i o n f o r the m a j o r i t y of the compounds l i s t e d i n T a b l e s I and I I i s on the o r d e r of 1 t o 100 ppb. F o r o i l samples, nonane, which i s o f t e n added as an i n t e r n a l s t a n d a r d , i s d e t e c t a b l e t o l e s s than 5ppb. R e p r o d u c i b i l i t y . R e p r o d u c i b i l i t y i s a t h i r d f a c t o r t h a t depends p r i m a r i l y on r e c o v e r y . As a g e n e r a l r u l e , r e p r o d u c i b i l i t y improves as r e c o v e r y i n c r e a s e s . For the m a j o r i t y o f compounds f o r which the r e c o v e r y i s g r e a t e r than 40%, the r e p r o d u c i b i l i t y w i l l be on the o r d e r o f 2 - 8% r e l a t i v e s t a n d a r d d e v i a t i o n (RSD). T h i s number i s s t r o n g l y a f f e c t e d by column r e s o l u t i o n , however, s i n c e many foods samples tend t o g i v e c o m p l i c a t e d chromatograms. As the r e c o v e r y drops below 40% the r e p r o d u c i b i l i t y r a p i d l y d e t e r i o r a t e s . For v e r y v o l a t i l e compounds, the purge e f f i c i e n c y

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

146

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

i s g e n e r a l l y h i g h , but t r a p p i n g e f f i c i e n c y may become a f a c t o r . The r e p r o d u c i b i l i t y o b t a i n e d f o r pentane i n c o r n o i l i s 2.97 RSD, but the r e p r o d u c i b i l i t y f o r butane i s 10.57 RSD. T h i s s u g g e s t s t h a t butane i s not q u a n t i t a t i v e l y t r a p p e d by Tenax. A number of r e s e a r c h e r s are c u r r e n t l y e v a l u a t i n g new s o r b e n t m a t e r i a l s , which may be used a l o n e or i n c o m b i n a t i o n w i t h Tenax to improve the a n a l y s i s of compounds w i t h lower m o l e c u l a r w e i g h t s . When h e a t i n g samples, an a d d i t i o n a l f a c t o r c o n c e r n i n g r e p r o d u c i b i l i t y i s i n t r o d u c e d . Not o n l y must the h e a t e r c o n t r o l be p r e c i s e , but a time must a l s o be a l l o w e d f o r the sample temperature to e q u i l i b r a t e w i t h the h e a t e r b e f o r e s t a r t i n g to purge. Samples p l a c e d i n a h e a t e r at 150 C do not i n s t a n t l y r e a c h and e q u i l i b r a t e at 150 C . The r a t e at which the a c t u a l sample temperature r i s e s i s e x t r e m e l y d i f f i c u l t to r e p r o d u c e . If p u r g i n g i s begun b e f o r e the sample has e q u i l i b r a t e d , the r e s u l t i n g r e p r o d u c i b i l i t y may be a c c e p t a b l e , but can be improved. By a l l o w i n g a p r e h e a t tim to e q u i l i b r a t e b e f o r e p u r g i n g r e s u l t i n g from h e a t e r v a r i a t i o n s , d i f f e r i n g samples or sample s i z e s ( i . e . d i f f e r e n c e s i n heat c a p a c i t y ) , or g e o m e t r i c v a r i a t i o n s i n the sample h o l d e r ( e . g . s o l i d chunks w i l l have d i f f e r e n t amounts of s u r f a c e a r e a c o n t a c t i n g the w a l l s of the v e s s e l ) can be m i n i m i z e d . 0

o

A r t i f a c t Formation. The p o t e n t i a l f o r the f o r m a t i o n of a r t i f a c t s i s p r e s e n t i n v i r t u a l l y e v e r y a n a l y t i c a l method. An advantage of dynamic headspace a n a l y s i s i s t h a t no s o l v e n t i s u s e d , e l i m i n a t i n g the g r e a t e s t s i n g l e s o u r c e o f a r t i f a c t s . U n f o r t u n a t e l y , however, o t h e r mechanisms of a r t i f a c t f o r m a t i o n do exist. The i n s t r u m e n t a t i o n must be d e s i g n e d to m i n i m i z e any possible a r t i f a c t s . When r u n n i n g heated samples t h e r e are a number o f p o s s i b l e mechanisms f o r the f o r m a t i o n of a r t i f a c t s . F o r compounds t h a t are t h e r m a l l y l a b i l e the o n l y method o f p r e v e n t i n g t h e i r d e s t r u c t i o n i s to m a i n t a i n the sample temperature below any p o s s i b l e breakdown p o i n t . F o r samples s u b j e c t to o x i d a t i o n , an a d d i t i o n a l s t e p i s needed. The v o l a t i l e s p r e s e n t i n o i l s , f o r i n s t a n c e , are p r i m a r i l y formed through o x i d a t i o n r e a c t i o n s . At the temperatures n o r m a l l y used to run o i l s , the sample w i l l r a p i d l y r e a c t w i t h any t r a c e s o f oxygen p r e s e n t to form new compounds o r i n c r e a s e the c o n c e n t r a t i o n s of o x i d a t i o n p r o d u c t s already present. T h i s can be a g g r a v a t e d by the p r e h e a t i n g s t e p used to a c h i e v e temperature r e p r o d u c i b i l i t y , a l l o w i n g a g r e a t e r time f o r the hot sample to r e a c t w i t h any oxygen d i s s o l v e d i n the sample o r p r e s e n t i n the sample v e s s e l . To remove the oxygen a "prepurge" s t e p i s added i n which the sample i s purged f o r a s h o r t time b e f o r e heat i s a p p l i e d . Any oxygen p r e s e n t w i l l be removed from the sample and r e p l a c e d by n i t r o g e n . Note t h a t t h i s prepurge gas must a l s o be p a s s e d through the Tenax to t r a p any compounds t h a t may be purged at the low t e m p e r a t u r e . After p r e p u r g i n g , the sample h e a t e r i s t u r n e d on and the temperature a l l o w e d to r i s e . The e f f e c t s of the p r e p u r g i n g s t e p can be noted i n F i g u r e 3, which i l l u s t r a t e s a c o r n o i l sample run twice under i d e n t i c a l c o n d i t i o n s w i t h the e x c e p t i o n of the p r e p u r g e s t e p . For f u r t h e r c o n s i d e r a t i o n of i n t r o d u c i n g a r t i f a c t s , the p o s s i b i l i t y of c r o s s c o n t a m i n a t i o n between samples must f i r s t be examined. I t i s p o s s i b l e i n any i n s t r u m e n t a l method f o r p a r t of

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

10.

WESTENDORF

Automated Analysis of Volatile Flavor Compounds

jL

i

JjUiJLjJ.„^^ F i g u r e 3:

V a r i a n c e s i n c o r n o i l sample w i t h o u t ( t o p ) and w i t h (bottom) prepurge.

American Chemical Society Library 1155 16th St.,

N.W.

In Characterization andWashington, Measurement of D.C. Flavor 20036 Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

147

148

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

one sample t o " c a r r y o v e r " i n t o t h e next r u n , i n f l u e n c i n g t h e r e s u l t s o f the second run. With a c o n c e n t r a t o r system, the p o s s i b l e causes o f c a r r y o v e r a r e t h r e e f o l d : incomplete d e s o r p t i o n , c o n d e n s a t i o n , and a d s o r p t i o n . Any p o s s i b l e t r a c e s o f sample components remaining on t h e Tenax t r a p as a r e s u l t o f incomplete d e s o r p t i o n a r e g e n e r a l l y removed v i a a bake s t e p , i n which the t r a p i s heated t o a temperature above t h a t used f o r d e s o r p t i o n w h i l e p a s s i n g gas through the t r a p . Condensation and a d s o r p t i o n can occur i n the v a l v i n g and t r a n s f e r t u b i n g used. These can be e l i m i n a t e d by h e a t i n g the v a l v e and l i n e s . However, t h i s introduces the p o s s i b i l i t y o f a r t i f a c t formation v i a c a t a l y t i c decomposition on t h e h o t s u r f a c e s i n t h e i n s t r u m e n t . The e f f e c t o f temperature d i f f e r e n c e s i s i l l u s t r a t e d i n F i g u r e 4 on a soy o i l sample. There was no measurable change f o r any sample r u n between ambient and 100 C t e m p e r a t u r e s . However, above 100 C some r e a c t i o n s began t o o c c u r . For the m a j o r i t y of samples, 100 C was s u f f i c i e n samples c o n t a i n i n g l a r g might have c a r r y o v e r a , highe temperature f o r r a p i d cleanup.

A p p l i c a t i o n s . I n a d d i t i o n t o o i l s and a p p l e s , a number o f o t h e r samples were used t o e v a l u a t e t h e method. These i n c l u d e d a comparison o f an orange t o a food s t a r c h w i t h added orange o i l ( F i g u r e 5 ) . As i s t y p i c a l f o r many a r t i f i c i a l f l a v o r s o r f l a v o r - a d d e d samples, t h e p r i m a r y d i f f e r e n c e s occur i n t h e e a r l y , most v o l a t i l e , p o r t i o n o f t h e chromatogram. F r u i t samples c a n a l s o be e v a l u a t e d f o r v a r i e t a l d i f f e r e n c e s ( F i g u r e 6 ) , o r f o r e v a l u a t i o n of seasonal v a r i a t i o n s , r i p e n i n g s t u d i e s , o r d e t e c t i o n of s t o r a g e abuse. The e v o l u t i o n o f f l a v o r v o l a t i l e s through two d i f f e r e n t p r o c e s s e s f o r peanut p r o d u c t s i s i l l u s t r a t e d i n F i g u r e 7. The two chromatograms r e p r e s e n t p r o d u c t s made from t h e same l o t o f raw peanuts, which had v i r t u a l l y no v o l a t i l e s p r e s e n t p r i o r t o p r o c e s s i n g . There a r e a number o f a d d i t i o n a l sample types f o r which DHA i s c u r r e n t l y b e i n g used. These i n c l u d e d a i r y p r o d u c t s , such as m i l k and cheese, carbonated beverages, powdered d r i n k mixes, beer, wine, c o f f e e , meats, s p i c e s , g r a i n s , c e r e a l s , and v a r i o u s forms o f candy. S t u d i e s underway i n c l u d e e v a l u a t i o n of p r o c e s s i n g t e c h n i q u e s , s h e l f - l i f e , p a c k a g i n g , and q u a l i t y c o n t r o l o f p r o d u c t s and incoming m a t e r i a l s . Conclusion A method f o r t h e automated a n a l y s i s o f v o l a t i l e f l a v o r compounds i n foods i s d e s c r i b e d . V o l a t i l e compounds a r e removed from t h e sample and c o n c e n t r a t e d v i a t h e dynamic headspace t e c h n i q u e , w i t h subsequent s e p a r a t i o n and d e t e c t i o n by c a p i l l a r y column gas chromatography. With t h i s method, d e t e c t i o n l i m i t s o f low ppb l e v e l s a r e o b t a i n a b l e w i t h good r e p r o d u c i b i l i t y . T h i s method has e x p e r i e n c e d r a p i d growth i n r e c e n t y e a r s , and i s now i n r o u t i n e use i n a number o f l a b o r a t o r i e s .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

WESTENDORF

Automated Analysis of Volatile Flavor Compounds

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

150

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

F i g u r e 5: Comparison o f f l a v o r p r o f i l e s o f orange (bottom) and o r a n g e - f l a v o r e d food s t a r c h ( t o p ) .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

10.

WESTENDORF

Automated Analysis of Volatile Flavor Compounds

milium

Lit

1

ILL F i g u r e 6:

V a r i e t a l d i f f e r e n c e s between w h i t e ( t o p ) and r e d (bottom) grape j u i c e s .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

151

152

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

F i g u r e 7:

Comparison o f r o a s t e d peanuts ( t o p ) and peanut b u t t e r (bottom) made from same l o t o f raw peanuts.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

10.

WESTENDORF

Literature 1. 2. 3. 4. 5.

6.

7.

8. 9. 10. 11. 12. 13. 14.

15.

16. 17.

18.

Automated Analysis of Volatile Flavor Compounds

Cited

Nawar, W.W., and Fagerson, I.S.; Anal. Chem., 1960, 32, 1534 Hornstein, I., and Crowe, P.F.; Anal. Chem., 1962, 34, 1354 Morgan, M.E., and Day, E.A.; J. Dairy Sci., 1965, 48, 1382 Bellar, T.A.; Lichtenberg, J.J.; Jour.AWWA 1974,66,739. Schamp, N.; Dirinck, P.; in "Chemistry of Foods and Beverages: Recent Developments"; Charalambous, G.; Inglett, G., Eds.; Academic; New York, 1982; p25. Keenaghan, J.; Meyers, M.C.; "Analysis of Volatile Organics in Foods and Beverages by Headspace Concentration-GC/MS"; presented at the 34th Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy; Atlantic City, N.J.; March, 1984. Westendorf, R.G.; "Trace Analysis of Volatile Organic Compounds in Foods by Dynamic Headspace Gas Chromatography"; presented at the 35th Chemistry and Applied March, 1984. Scholz, R.G.; Ptak, L.R.; J.Am.Oil Chem. Soc. 1966,43,596. Jackson, H.W.; Giacherio, D.J.; J.Am.Oil Chem. Soc. 1977,54,458-460. Jackson, H.W.; J.Am.Oil Chem. Soc. 1981,58,227,231. Dupuy, H.P.; Fore, S.P.; Goldblatt, L.; J.Am.Oil Chem. Soc. 1971,48,876. Dupuy, H.P.; Fore, S.P.; Goldblatt, L.; J.Am.Oil Chem. Soc. 1973,50,340. Selke, E.; J. Am.Oil Chem. Soc. 1970,47,393 Roberts, J.; "Semiautomated Dynamic Headspace Analysis of Vegetable Oil Volatiles"; presented at the 74th Meeting of the American Oil Chemists Society; Chicago, IL; May, 1983. Capillary Column Use in Purge and Trap Gas Chromatography II, Use of the Model 1000 Capillary Interface"; Application Note B021684; Tekmar Company, Cincinnati, OH. "Plastic Sample Preparation"; Application Note B081882; Tekmar Company, Cincinnati, OH. Westendorf, R.G.; "Optimization of Parameters for Purge and Trap Gas Chromatography"; presented at the 32nd Pittsburgh Conference on Analytical Chemistry and applied Spectroscopy; Atlantic City, NJ; March, 1981. "Purge and Trap Analysis Using a Photoionization Detector: Removal of Water Interference"; Application Note B042281; Tekmar Company, Cincinnati, OH.

RECEIVED June 24, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11 Supercritical Fluid Extraction in Flavor Applications Val J . Krukonis Phasex Corporation, Nashua, ΝH 03060

Supercritical fluids are receiving increasing attention as extraction solvents because of their pressure-dependent solvent properties, often displaying the ability to extract selectively one (or a few) component(s) from mixture; the chemical, pharmaceutical, engaged in developing processes using powers. Because selective, "dial-in-a-dissolvingpower" action can often be conferred to supercritical fluids by virtue of the pressure applied on it, they are attractive for extracting flavors present in natural materials. Supercritical fluids, especially carbon dioxide which is an active solvent at room temperature, exhibit a frequent ability to fractionate flavor and aroma components present as complex mixtures of compounds in such materials as pepper, ginger, allspice, and other spices. An overview of research on supercritical fluid solubility phenomena and on process operation provides the framework to understand the extraction capabilities of supercritical fluids; three examples of ginger, allspice, and apple essence extractions illustrate these capabilities. By now, nearly every chemist has had some introduction to the subject of s u p e r c r i t i c a l extraction i n one form or another, and i t would seem that a f t e r scores of papers, newsreleases, and trade journal a r t i c l e s , only so much can be said about the background and early f i n d i n g s , the thermodynamic interactions between dissolved solutes and high pressure gases, the equations of state that can correlate and predict s o l u b i l i t y behavior, the many applications of the technology (some of which are i n f l a v o r s ) , the f u l l scale coffee and hops extraction plants now i n operation, etc. What, then, can a paper e n t i t l e d " S u p e r c r i t i c a l Fluids - Overview and S p e c i f i c Examples i n Flavors Applications" give that* s new? hopefully, a d i f f e r e n t development of the h i s t o r i c a l perspective 0097-6156/ 85/ 0289-Ό154$06.25/ 0 © 1985 American Chemical Society

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

K R υ ΚΟΝIS

155

Supercritical Fluid Extraction

and overview of s u p e r c r i t i c a l f l u i d phenomena, and some recently obtained data on flavor extraction and separation. Some of the h i s t o r i c a l perspective i s extracted (no pun intended) from a previous paper of the author (1) and i s expanded with a chronological development of s o l u b i l i t y phenomena based upon an additional compilation of recent work on naphthalenes u p e r c r i t i c a l solvent systems. The new data on f l a v o r extraction and f r a c t i o n a t i o n point out the most unique feature of super­ c r i t i c a l f l u i d solvents, v i z . , their often-demonstrated selective dissolving power properties, a s e l e c t i v i t y that i s achieved because the dissolving power of s u p e r c r i t i c a l f l u i d s i s pressure-dependent and can, therefore, be adjusted.

Historical Perspective The a b i l i t y of a s u p e r c r i t i c a l f l u i d to dissolve low vapor pressure materials was f i r s t reporte They described their s o l u b i l i t pressure glass c e l l s , and they observed that several inorganic s a l t s (e.g., cobalt chloride, potassium iodide, potassium bromide, f e r r i c chloride) could be dissolved or precipitated s o l e l y by changes i n pressure on ethanol above i t s c r i t i c a l point (T = 234 C). For example, increasing the pressure on the system caused the solutes to dissolve, and decreasing the pressure caused the dissolved materials to nucleate and p r e c i p i t a t e , i n the words of the authors, "as a snow." In an h i s t o r i c a l l y i n t e r e s t i n g aside, there was i n i t i a l controversy about this finding a f t e r i t was reported at a meeting of the Royal Society (London). Professor W. Ramsay, i n a subsequent paper delivered to the Royal Society (3) stated that based upon his reproduction of one of Hannay and Hogarth s experiments, he concluded that "the gentlemen have observed nothing unusual, but merely the ordinary phenomenon of s o l u b i l i t y of a s o l i d i n a hot l i q u i d . " Ramsay i n the same paper also took to task Dr. T. Andrews (the Andrews of carbon dioxide c r i t i c a l point phenomena fame) for "purposely abstaining from speculating on the nature of matter at the c r i t i c a l point, whether i t be l i q u i d or gaseous, or i n an intermediate condition, to which no name can be given." Ramsay went on to describe some other of his own experiments on c r i t i c a l point phenomena, and he concluded by saying "I am i n c l i n e d to think that carbonic anhydride, examined by Dr. Andrews, i s abnormal i n this respect, but of this I am by no means certain." As i s now well known, and as also must have been known to Andrews, carbon dioxide, the anhydride to which Ramsay r e f e r s , i s not abnormal i n i t s c r i t i c a l behavior. In s t i l l l a t e r presentations to the Royal Society (4_,5) , Hannay responded to Ramsay's charge, asking "permission to point out some errors into which Prof. Ramsay had f a l l e n " ; he did point out those errors, and he presented results of more experiments on dissolving solutes i n s u p e r c r i t i c a l f l u i d s which for posterity substantiated that the finding of the pressure-dependent dissolving power of a s u p e r c r i t i c a l f l u i d was indeed a new phenomenon. As might be concluded from the excerpted statements, references 2-5 provide i n t e r e s t i n g reading of developments during a period of time 1

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

156

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

when c r i t i c a l point phenomena were s t i l l incompletely understood or accepted. The pressure dependence of the dissolving power of a s u p e r c r i t i c a l f l u i d i s not limited to inorganic s a l t solutes, but i s a r e l a t i v e l y general phenomenon exhibited by a l l s o l i d and many l i q u i d solutes (as long as the solute i s not i n f i n i t e l y miscible with the solvent). After the f i r s t report by Hannay and Hogarth, a number of other papers reported on s o l u b i l i t y phenomena with a variety of s u p e r c r i t i c a l f l u i d solvents and organic s o l i d and l i q u i d solutes. Solvents included carbon dioxide, nitrous oxide, the l i g h t hydrocarbons, and solutes covered the gamut of organic compounds, v i z . , a l i p h a t i c s , aromatics, halogenated hydrocarbons, heteromolecules, t r i g l y c e r i d e s , and the l i k e . Booth and Bidwell i n 1949 presented an excellent review of nearly seventy-years of early research (6); several other comprehensive reviews cover l a t e r periods up to 1984 (]_,&_,9) . Supercritical flui attention i n the mid-to-lat development e f f o r t on activated carbon regeneration (10), alcohol-water separation (11) , chemotherapeutic drugs, flavor (12), and aroma extraction (13) , and on many other separations. Many of these papers introduced the phenomenon of the pressure-dependent dissolving power of a s u p e r c r i t i c a l f l u i d using naphthalene s o l u b i l i t y data as an example, since naphthalene s o l u b i l i t y has been studied more thoroughly than any other organic (or inorganic) compound. There i s a substantial body of quantitative information available, and even more importantly, naphthalene's s o l u b i l i t y behavior i s representative of the behavior of many other compounds in s u p e r c r i t i c a l solvents. Naphthalene s o l u b i l i t y models the s o l u b i l i t y of, for example, terpenes which have not been studied to as great an extent, but which are of commercial i n t e r e s t , or the behavior of t r i g l y c e r i d e s extracted from seeds with s u p e r c r i t i c a l carbon dioxide, or the s o l u b i l i t y of s t i l l more complex materials for which only scant data e x i s t . In 1948, a study of the s o l u b i l i t y and phase behavior of naphthalene dissolved i n s u p e r c r i t i c a l ethylene was reported by two workers from The Netherlands, Diepen and Scheffer (14) , and this now-classic paper was followed by two others from the same authors (15,16) and from others who reproduced the naphthalene s o l u b i l i t y data of Diepen and Scheffer and extended the studies to other s u p e r c r i t i c a l solvents. For example, Tsekhanskaya et a l measured s o l u b i l i t i e s of naphthalene i n ethylene and i n carbon dioxide i n 1962 (17 ,18) . Others are King and Robertson i n 1962, who studied the s o l u b i l i t y of naphthalene i n hydrogen and argon (19); Najour and King i n 1970, naphthalene i n s u p e r c r i t i c a l methane, ethylene, and carbon dioxide (20); McHugh and P a u l a i t i s , 1980, naphthalene i n carbon dioxide (21) , Kurnik and Reid, 1982, naphthalene i n carbon dioxide (22); Schmitt and Reid, 1984, naphthalene i n ethane, trifluoromethane, and chlorotrifluoromethane (23); and f i n a l l y , Krukonis et a l (24) , and McHugh et a l (25) , who studied the s o l u b i l i t y and phase behavior of naphthalene i n the solvent s u p e r c r i t i c a l xenon i n 1984. Thus, as might be concluded from the long l i s t of references c i t e d , naphthalene s o l u b i l i t y has been tested thoroughly, and a large amount of data exists on i t s

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

Supercritical Fluid Extraction

KRUKONIS

157

behavior over a wide range of pressure-temperature-gas conditions. Because of t h i s large data base, s u p e r c r i t i c a l f l u i d applications and process operation are quite frequently described i n terms of naphthalene s o l u b i l i t y , the concept being extended to other systems "by association"; some of the "association" w i l l be developed subsequently i n this section. Figure l i s a graph of naphthalene s o l u b i l i t y i n carbon dioxide at 45 C (15 C above the c r i t i c a l temperature of carbon dioxide) taken from Reference 17· As i s obvious from an examination of the data, the s o l u b i l i t y of naphthalene increases dramatically when the pressure i s increased beyond the c r i t i c a l pressure of 73 atm. The s o l u b i l i t y (given i n units of grams/liter i n the reference) approaches about 10% (w/w) at a pressure l e v e l of 200 atm. Figure 2 assembles more data on the s o l u b i l i t y of naphthalene in carbon dioxide from the previously mentioned works which was assembled by Modell (26 w i l l a i d i n the explanatio process operates. The curves i n the figure give isobars of s o l u b i l i t y as a function of temperature, and the dotted curve shows the s o l u b i l i t y of naphthalene i n l i q u i d carbon dioxide up to the c r i t i c a l temperature of 31 C and i n saturated vapor carbon dioxide. As one other example of a system f o r which a substantial amount of data has been obtained, Figure 3 shows the s o l u b i l i t y behavior of t r i g l y c e r i d e s i n s u p e r c r i t i c a l carbon dioxide (27 ,28) . The absolute values, the pressure, and temperature levels are d i f f e r e n t , but the c h a r a c t e r i s t i c "fan" of curves i s similar i n shape to those i n Figure 2, which p a r t i a l l y lends credence to the statement that the behavior of many s u p e r c r i t i c a l solvent-incompletely miscible solute binary systems i s similar. As an additional example, Figure 4 shows the s o l u b i l i t y behavior of a very d i f f e r e n t solute-solvent system, silica-water (29). Comparison of the s o l u b i l i t y levels with temperature and pressure levels again points out that the absolute values for a l l systems are not the same; however, the general shapes of the curves of a l l systems, i . e . , the "fans" referred to e a r l i e r , are the same. Many other s o l u b i l i t y diagrams can be constructed for other solute-solvent pairs; for example, phenanthrene (30), benzoic acid (31), anthracene (32) , phenol and chlorinated phenols (33) , biphenyl (21), represent just a small l i s t of solutes that have been studied for their s o l u b i l i t y and phase behavior i n methane, ethylene, ethane, and carbon dioxide. They a l l exhibit the s i m i l a r i t y of the data shown i n Figures 2, 3, and 4. Figure 2, showing the s o l u b i l i t y behavior of naphthalene i n s u p e r c r i t i c a l carbon dioxide, and Figures 3 and 4, which point out that the behavior i s quite general, w i l l be used to explain how a "generic" s u p e r c r i t i c a l f l u i d extraction process operates. Q

Operation of a Supercritical Fluid Extraction Process Because of the dissolving c h a r a c t e r i s t i c s shown i n Figures 1-4, i t i s possible to design i n d u s t r i a l processes to extract, purify, and fractionate materials based on changes i n pressure of a s u p e r c r i t i c a l f l u i d solvent, at high pressure effecting an

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

158

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

KRUKON1S

Figure 2.

Supercritical Fluid Extraction

S o l u b i l i t y of Naphthalene i n Carbon Dioxide.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

159

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

160

20

30

40

50

60

TEMPERATURE Figure 3.

e

70

βΟ

C

S o l u b i l i t y of Triglycerides i n Carbon Dioxide.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

Supercritical Fluid Extraction

KRUKONIS

160

240

320

400

480

560

TEMPERATURE Figure 4.

S o l u b i l i t y of S i l i c a i n Water.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

162

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

extraction and at a lower pressure effecting a separation of the dissolved material from the solvent which can then be recycled to the extractor for further extraction. The process can be either batch or continuous depending upon the nature of the feed and the nature of the extraction, i . e . , whether i t be a p u r i f i c a t i o n or fractionation, or an extraction from a reaction mass. A schematic diagram of a process that uses a s u p e r c r i t i c a l f l u i d as a pressure-dependent solvent to extract an organic substance i s given i n Figure 5a. Four basic elements of the process are shown, v i z . , an extraction vessel, a pressure reduction valve, a separator for c o l l e c t i n g the material dissolved i n the extractor, and a compressor for recompressing and recycling f l u i d . A n c i l l a r y pumps, valving, f a c i l i t i e s for f i l l i n g the vessel and for f l u i d make-up, heat exchangers, and similar equipment are omitted from the figure for c l a r i t y and ease of presentation. Figure 5b shows the extensive data on s o l u b i l i t y of naphthalene i n carbon dioxide as a function o i n Figure 2. Some process operating parameters are indicated on two s o l u b i l i t y isobars i n Figure 5b; E, represents conditions i n the extractor, e.g., 300 atm, 55°C, and the conditions which exist i n the separator, 90 atm, 43 C. The extractor vessel i s assumed to be f i l l e d with naphthalene i n admixture with another material, which for ease of discussion i s assumed to be insoluble i n carbon dioxide. Carbon dioxide at condition E. i s passed through the extraction vessel wherein i t extracts trie naphthalene from the insoluble material. The carbon dioxide-naphthalene solution leaving the extractor i s expanded to 90 atm through the pressure reduction valve and as indicated by the directed path i n Figure 5b. Because the equilibrium s o l u b i l i t y has been reduced from about 5% to 0.2% during the pressure reduction step, naphthalene precipitates from the solution. The naphthalene i s collected i n the separator, and the carbon dioxide leaving the separator i s recompressed and returned to the extractor. Recycling continues u n t i l a l l the naphthalene i s extracted. The directed l i n e segment E^-S, i n Figure 5b and i t s reverse represent approximately the c y c l i c process on the s o l u b i l i t y diagram. As an alternative to extraction and separation using pressure reduction ( i . e . , the path E^-S.), the process can operate at constant pressure with temperature changes i n a s u p e r c r i t i c a l f l u i d used to e f f e c t the extraction and separation steps. For example, and again s t a r t i n g at , the stream leaving the extractor could be passed through a heat exchanger (instead of a pressure reduction valve) and cooled to, for example, 20 C as indicated by the directed portion on the 300 atm isobar. Cooling the stream would result i n the p r e c i p i t a t i o n and c o l l e c t i o n of naphthalene i n the separator at conditions S^* The carbon dioxide leaving the separator could then be heated back to 55 C before being recycled to the extractor. This mode of operation, instead of requiring a high Δ Ρ compressor would employ a low Δ Ρ and less expensive blower to supply f r i c t i o n a l loss which i s small. Although not s p e c i f i c a l l y accented on the s o l u b i l i t y diagram, there are regions where the s o l u b i l i t y decreases with increasing temperature, a behavior not usually exhibited by l i q u i d solvents,

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

KRUKONIS

Supercritical Fluid Extraction

Figure 5. S u p e r c r i t i c a l F l u i d Extraction - a. b. Operating Paths.

163

Process Diagram;

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

164

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

and t h i s behavior can be used to advantage i n carrying out a s u p e r c r i t i c a l f l u i d extraction and separation. For example, one could carry out the extraction of naphthalene at 30 C and 80 atm and cause a 10-fold change i n dissolving power by heating the solution only 5 or 10 C. At the end of the extraction cycle, e s s e n t i a l l y a l l the naphthalene i s i n the separator and the insoluble material i s l e f t i n the extractor. The process can be considered as a naphthalene extraction or as an "insolubles" p u r i f i c a t i o n depending upon which i s the desired product.

Applications of Supercritical Fluids to the Extraction and Characterization of Flavors In addition to the a b i l i t y of carbon dioxide to act as a l i p o p h i l i c solvent, i t s dissolving power can be t a i l o r e d to carry out fractionations and r e l a t i v mixture, and t h i s uniqu characterization of f l a v o r s . Several representative examples that point out t h i s a b i l i t y are presented. Ginger. Figure 6 i s a chromatogram obtained by c a p i l l a r y column gas chromatography (GC) of an extract of ginger that was obtained by extracting shredded ginger root with methylene chloride i n a Soxhlet extractor. As can be seen from a cursory examination of the chromatogram, ginger extract i s a complex mixture of many components (no attempt was made to i d e n t i f y any of them for this presentation). Shredded ginger was also extracted with s u p e r c r i t i c a l carbon dioxide and the extract(s) analyzed by GC. Figure 7 i s a schematic diagram of the continuous flow laboratory apparatus used for the extraction t e s t s . The primary elements are a gas supply, compressor, extraction vessel, flow control and pressure reduction valve, gas flow meter, and t o t a l gas meter, plus such a n c i l l a r y equipment as heaters, temperature and pressure control, etc., which are not shown i n the schematic diagram for ease of discussion. The conduct of the extraction was as follows: An amount of ginger was introduced into the pressure vessel, the vessel sealed, and connected to the system. (The pressure vessels are t y p i c a l l y 2" nominal, SCh 160,316 st pipe, 1 to 4 f t long, 1/2 to 2 l i t e r s volume.) The temperature of the system was brought to the desired temperature, and carbon dioxide from a manifold and available at about 1200 p s i was fed to the suction side of a diaphram compressor. The compressed carbon dioxide at the desired pressure was passed through the g i n g e r - f i l l e d extractor, and soluble materials present i n the ginger were dissolved. The solution leaving the extractor was passed through the flow control valve and expanded to ambient pressure which caused the dissolved materials to nucleate and precipitate i n the c o l l e c t o r . The 1 atm gas passed subsequently through the flow meter and dry test meter. Gas volume data coupled with gravimetric determinations allow concentrations, y i e l d s , rates of extraction, and similar information to be calculated. Figure 8 i s a chromatogram of the extract obtained by

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

KRUKON1S

Figure 6. Ginger.

Supercritical Fluid Extraction

Gas Chromatogram - Methylene Chloride Extract of

Sample Collector Dry Test Meter

CO2

Supply

Figure 7. Laboratory Apparatus f o r S u p e r c r i t i c a l F l u i d Extraction.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

166

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

extracting ginger with s u p e r c r i t i c a l carbon dioxide at conditions of 5000 p s i and 50 C. For the q u a l i t a t i v e purpose of this discussion, the chromatograms shown i n Figure 6 and Figure 8 are e s s e n t i a l l y the same, which shows that 5000 psi carbon dioxide i s comparable i n i t s extraction c h a r a c t e r i s t i c s to methylene chloride. However, the physical appearance of the two extracts was s l i g h t l y d i f f e r e n t , v i z . , the carbon dioxide extract was a yellow-green, thin paste, whereas the methylene chloride extract was darker green. In order to i l l u s t r a t e the pressure dependent dissolving power of a s u p e r c r i t i c a l f l u i d , another charge of shredded ginger was extracted with carbon dioxide at two other pressure l e v e l s . The charge was f i r s t extracted at a low pressure, 1500 psi and 50 C. The extract that was collected was a pale yellow l i q u i d . The pressure was then raised to a medium pressure l e v e l of 3000 p s i ; and another volume of gas passed through the same charge of shredded ginger. The extrac i n color, but a thicker test carried out at 5000 p s i . Figure 9 i s a chromatogram of the 1500 p s i extract and Figure 10 the chromatogram of the 3000 psi extract; Figure 9 i s c l e a r l y different from the chromatogram of the 5000 p s i extract shown i n Figure 8 i n the number and r a t i o s of peaks. A cursory comparison of Figures 8, 9, and 10 reveals that carbon dioxide at 1500 p s i does not extract many of the longer retention components ( i . e . , those peaks to the right of the chromatograms)· A closer examination of the chromatograms shows that 1500 p s i carbon dioxide does extract some of the long retention time compounds, while i t does not extract a l l the low retention time components. The r e l a t i o n between the retention time on a chromatographic column and the s o l u b i l i t y i n carbon dioxide was not determined i n this study, and there probably i s no monotonie r e l a t i o n . Retention time may r e f l e c t more than say a vapor pressure phenomenon and can be a result of p o l a r i t y considerations. In a homologous series, for example, the retention time on a column i s related to solely vapor pressure, and analogously the s o l u b i l i t y i n a s u p e r c r i t i c a l f l u i d would exhibit a monotonie relationship, v i z . , the higher the vapor pressure (and therefore the lower the retention time), the higher the s o l u b i l i t y . However, across families of compounds, for example, organic acids, ketones, esters and aldehydes, there i s probably not a general c o r r e l a t i o n between retention time on a s p e c i f i c column packing and vapor pressure, and s i m i l a r l y there would not be a c o r r e l a t i o n between retention time and the s o l u b i l i t y i n a s u p e r c r i t i c a l f l u i d . Figure 9 does show, nevertheless, that some long retention time components are extracted while other low retention time materials are not; i t i s offered that the mixture characterized by the chromatogram i n Figure 9 would be d i f f i c u l t to achieve by d i s t i l l a t i o n or by l i q u i d solvent extraction, and one of the advantages of s u p e r c r i t i c a l f l u i d extraction l i e s i n the a b i l i t y to i s o l a t e or fractionate certain groups of compounds for subsequent evaluation or characterization.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

KRUKONIS

Supercritical Fluid Extraction

Figure 8. of Ginger.

Gas Chromatogram - Carbon Dioxide Extract (5000 psi)

Figure 9. of Ginger.

Gas Chromatogram - Carbon Dioxide Extract (1500 psi)

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

167

168

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Figure 10 shows the chromatogram of the 3000 p s i extract of ginger described e a r l i e r ; a comparison of Figures 9 and 10 and the high pressure extract whose chromatogram was given i n Figure 8 shows that the extract taken sequentially after f i r s t stripping the ginger at 1500 p s i was depleted of some of the low retention time components that were concentrated at 1500 p s i as shown i n Figure 9. The gravimetric analysis of a l l the extracts i s given i n Table I.

Table I.

Gravimetric Analysis of Ginger Extracts Amount of Ginger Charged

Test Soxhlet, methylene chloride

5.07 g

Amount of Extract Not determined

5000 p s i , carbon dioxide 1500 p s i , carbon dioxide 3000 p s i , carbon dioxide

32.44 g

Same charge as above

0.35 g

1.74 g

Piaento Berries ("All Spice"). A similar set of extraction tests was carried out with crushed pimento berries, and q u a l i t a t i v e l y similar results were obtained. Figure 11 i s a chromatogram of the methylene chloride extract of pimento berries, and no commentary on the complexity of the chromatogram i s needed. Figure 12 i s the high pressure carbon dioxide extract of pimento berries and the two chromatographs are similar i n appearance. Extraction conditions with the carbon dioxide were again 5000 p s i , 50 C. A second charge of crushed pimento berries was extracted sequentially at the same two pressure levels tested for the ginger, 1500 p s i and 3000 p s i , and the chromatograms of the two extracts are shown i n Figures 13 and 14, respectively. The low pressure extract was almost water-white, and the 3000 and 5000 p s i extracts were yellow to yellow-green pastes. A comparison of Figures 12, 13, and 14 allows the same q u a l i t a t i v e conclusions to be drawn as they were for the ginger extractions, v i z . , low pressure carbon dioxide i s different i n i t s extractive c a p a b i l i t i e s than i s high pressure carbon dioxide, and based upon the appearance of the chromatogram 5000 p s i carbon dioxide extracts the same components that methylene chloride does. A d d i t i o n a l l y , the sequential 3000 psi extract i s depleted i n some of the soluble components that were concentrated i n the 1500 p s i extract. For completeness, Table II gives the gravimetric analysis of the extracts of pimento b e r r i e s .

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

KRUKONIS

Figure 10. of Ginger.

Supercritical Fluid Extraction

Gas Chromatogram - Carbon Dioxide Extract (3000 psi)

Figure 11. Gas Chromatogram - Methylene Chloride Extract of Pimento Berries.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

170

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

-HFigure 12. Gas Chromatogram - Carbon Dioxide Extract (5000 psi) of Pimento Berries.

Figure 13. Gas Chromatogram - Carbon Dioxide Extract (1500 psi) of Pimento Berries.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

KRUKONIS

171

Supercritical Fluid Extraction

Table I I . Gravimetric Analysis of Pimento Berry Extract Test Soxhlet 5000 p s i 1500 p s i 3000 p s i

Amount of Pimento Berries 4.34 g 25.97 g 27.41 g Same charge as above

Amount of Extract Not determined 2.41 g 0.32 g 1.64 g

Extraction/Concentration of Flavors from Solutions. Carbon dioxide i s also an i d e a l solvent for concentrating essence components, and one example of t h i s c a p a b i l i t y i s presented i n t h i s subsection. Figure 15 i s a chromatogram of a synthetic apple essence consisting of eight C^-Cg esters, alcohols, and aldehydes dissolved i n water. Extraction was carried out i n the same laboratory apparatus shown i n Figure 7 with the differences that the vessel was f i t t e d with a check valve to prevent d i s t r i b u t o r sieve tray the carbon dioxide flowing through the charge of l i q u i d . An amount of s o l u t i o n (302 g i n t h i s test) was contacted with 800 g of carbon dioxide i n the batch-continuous mode described e a r l i e r , and the extract was c o l l e c t e d downstream of the expansion valve i n a trap maintained at dry i c e temperature to capture the v o l a t i l e components. A chromatogram of the extract i s given i n Figure 16. A quick comparison of Figures 15 and 16 does not provide q u a l i t a t i v e differences to be discerned because the peaks r i s e o f f - s c a l e i n both cases. The feed s o l u t i o n was composed of eight components, each present at about 0.02% ( f o r a " t o t a l " concentration of 0.19%). Each of the two peaks indicated by arrows i n both the s t a r t i n g essence and the extract a c t u a l l y consists of two components which were resolved i n the GC analysis (note the double, almost-superposed time p r i n t above each peak) but because of the scale selected they are not v i s u a l l y i d e n t i f i a b l e as separate peaks on the chromatogram. The integrator output from the GC analysis of the e x t r a c t , however, showed that the t o t a l concentration i n the extract was 12.4% (an average of about 1.5% per component)· Further inspection of Figures 15 and 16 shows that some trace components (market Τ i n both figures) are concentrated i n the e x t r a c t . These trace components were present i n the compounds used to prepare the synthetic apple essence and are related aldehydes, acids, etc. The chromatogram of the r a f f i n a t e , the depleted s o l u t i o n from the e x t r a c t i o n , i s given i n Figure 17, and comparison provides i n t e r e s t i n g q u a l i t a t i v e r e s u l t s . F i r s t , a l l the component concentrations have been reduced s u b s t a n t i a l l y (which i s not surprising because high carbon number l i p o p h i l i c materials are r e a d i l y extracted by carbon d i o x i d e ) . However, the peak heights (and, more q u a n t i t a t i v e l y , the integrator output not reproduced here) show that not a l l the components are extracted equally. A d d i t i o n a l l y , the two peaks marked by arrows can now be seen to contain two components each. As was found for the previously discussed ginger and pimento berries extractions, not a l l the apple essence components are extracted equally. Shultz et a l (34) have

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

172

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

III Figure 14· Gas Chromatogram - Carbon Dioxide Extract (3000 psi) of Pimento Berries.

Τ τ Figure 15. Gas Chromatogram - Synthetic Apple Essence.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

KRUKONIS

Supercritical Fluid Extraction

\ \

.1 II

Figure 16. Essence.

II

Gas Chromatogram - Carbon Dioxide Extract of Apple

Figure 17.

Gas Chromatogram - Depleted Apple Essence.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

174

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

shown, f o r example, that esters exhibit higher d i s t r i b u t i o n c o e f f i c i e n t s than aldehydes and carboxylic acids i n organic-watercarbon dioxide systems. In closing, this paper was not intended to represent an exhaustive process development e f f o r t i n flavors extraction from natural materials nor a development of the quantitative a n a l y t i c a l c a p a b i l i t i e s of s u p e r c r i t i c a l carbon dioxide. However, even though the examples and the conditions of extraction were somewhat a r b i t r a r y , they point out some of the interesting features of the pressure dependent dissolving power properties of s u p e r c r i t i c a l f l u i d s . They can be further refined by virtue of more narrow ranges and ratios of pressure and temperature to accomplish s t i l l more narrow separations. On the other hand, the h i s t o r i c a l perspective and operation of s u p e r c r i t i c a l extraction was intended to supply the background to serve as an a i d i n appreciating the motivation f o r the current a c t i v i t i e s , not j u s t i n for the process developmen food, pharmaceutical, and polymer industries.

Literature Cited

1. 2. 3. 4. 5. 6. 7.

8. 9. 10. 11. 12. 13. 14. 15. 16.

Krukonis, V. J., Branfman, A. R., and Broome, M. G. 87th Nat. Mtg.; AIChE: Boston, August 1979. Hannay, J. B., Hogarth, J. Proc. Roy. Soc. (London) 1879, 29, 324-6. Ramsay, W. Proc. Roy. Soc. (London) 1880, 30, 323-9. Hannay, J. B. and Hogarth, J. Proc. Roy. Soc. (London) 1880, 30, 178-88. Hannay, J. B. Proc. Roy. Soc. (London) 1880, 30, 484-9. Booth, H. S. and Bidwell, R. M. Chem. Rev. 1949, 44, 477-513. Irani, C. Α., and Funk, E. W. In CRC Handbook: Recent Developments in Separation Science; CRC Press, Boca Raton, Florida, Vol. III, Part A, 1977; pp. 171-9. Williams, D. F. Chem. Eng. Sci. 1981, 36, 1769-88. Paulaitis, M. E., Krukonis, V. J., Kurnik, R. T., and Reid, R. C. Rev. in Chem. Eng. 1983, 1, 179-250. Modell, M., deFilippi, R. P., and Krukonis, V. J. 87th Nat. Mtg.; AIChE: Boston, August 1979. Moses, J. M., Goklen, Κ. E., and deFilippi, R. P. 1982 Ann. Mtg.; AIChE: Los Angeles, November. Caragay, A. B. Perfume and Flavorist 1981, 6, 43-55. Schultz, E. G., and Randal, J. N. Food Tech 1970, 24, 94-8. Diepen, G. A. M. and Scheffer, F. E. C. J. Am. Chem. Soc. 1948, 70, 4081-5. Diepen, G. A. M. and Scheffer, F. E. C. J. Am. Chem. Soc. 1948, 70, 4085-9. Diepen, G. A. M. and Scheffer, F. E. C. J. Phys. Chem. 1953, 57, 575-8.

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

11.

17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30.

31. 32. 33. 34.

KRUKONIS

Supercritical Fluid Extraction

175

Tsekhanskaya, Yu. V., Iomtev, M. B. and Mushkina, Ε. V. Russ. J. Phys. Chem. 1964, 38, 1173-6. Tsekhanskaya, Yu. V., Iomtev, M. B. and Mushkina, Ε. V. Russ. J. Phys. Chem. 1962, 36, 1177-81. King, A. D., and Robertson, W. W. J. Chem. Phys. 1962, 37, 1453-5. Najour, G. C. and King, A. D., Jr., J. Chem. Phys. 1966, 45, 1915-21. McHugh, Μ. Α., and Paulaitis, M. E. J. Chem. Eng. Data 1980, 25, 326-9. Kurnik, R. T. and Reid, R. C. Fluid Phase Equilibria 1982, 8, 93-105. Schmitt, W. J. and Reid, R. C. 1984 Ann. Mtg.; AIChE: San Francisco, November. Krukonis, V. J., McHugh, Μ. Α., Seckner, A. J. J. Phys. Chem. 1984, 88, 2687-9. McHugh, Μ. Α., Seckner, Mtg.; AIChE: San Modell, Μ., deFilippi, R. P., Krukonis, V. J. and Robey, R. J. 87th AIChE Mtg., Boston, August 1979. Friedrich, J. P., List, G. R. and Spencer, G. F. 75th Am. Oil Chem. Soc. Mtg., Dallas, May 1984. Krukonis, V. J. 75th Am. Oil Chem. Soc. Mtg., Dallas, May 1984. Kennedy, G. C. Econ. Geol. 1950, 45, 629-36. Eisenbeiss, J. A Basic Study of the Solubility of Solids in Gases at High Pressures, Final Report, Contract No. DA18-108-AMC-244(A), Southwest Research Inst., San Antonio, August 6, 1984. Kurnik, R. T., Holla, S. J. and Reid, R. C. J. Chem. Eng. Data. 1981, 26, 47-51. Najour, G. C. and King, A. D., Jr. J. Chem. Phys. 1970, 52, 5206-11. VanLeer, R. Α., and Paulaitis, M. E. J. Chem. Eng. Data 1980, 25, 257-9. Shultz, W. G. and Randall, J. N. 1970, 24, 94-8.

RECEIVED September 9, 1985

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

Author Index Adams, Michael Α., 11 Albrecht, W., 43 Bille-Abdullah, H., 43 Busch, Kenneth L., 121 Ebeler, S., 95 Engel, K.-H., 43 G i l l e t t e , Marianne, 26 Hayashi, Tateki, 61 Idstein, Heinz, 109 Jennings, W., 95 Kr-oha, Kyle J . , 121 Krukonis, Val J . , 154 Lawless, Harry, 26

Schreier, Peter, 109 Shibamoto, Takayuki, 61 takeoka, G., 95 Tressl, R., 43 Westendorf, Robert G., 138

Subject Index Amino acids and peptides, HPLC analysis, 83 Analytical c h a r a c t e r i s t i c s , tandem mass spectrometry, 127-29 Analytical method, v o l a t i l e aldehydes, 61-77 Apple essence carbon dioxide extract, gas chromato gram, 17 3f depleted, gas chromatogram, 173f synthetic, gas chromatogram, 172f Apple v o l a t i l e s , gas chromatographic i d e n t i f i c a t i o n , 14 5t Aroma c h i r a l components i n trace amounts, analysis, 43-59 food, analysis by tandem mass spectrometry, 131-33 A r t i f a c t formation, automated analysis of v o l a t i l e flavor compounds, 146-48 Aspertame, HPLC analysis, 81,83 Automated analysis of v o l a t i l e flavor compounds, 138-52

A Acetate esters emitted from a banana, parent-ion tandem mass spectrum, 132f Acetoxyacid esters, hydrolysis, Candida u t i l i s , 51-54 2-Acetylthiazolidine, recovery e f f i c i e n c y , 72 Acid-catalyzed hydrolysis of saponins, 15,19 Alcohols c h i r a l , gas chromatographic separation, 44-46 o p t i c a l l y pure, formation, 51,52f Aldehydes and cysteamine, products and spec­ t r a l data, 66t standards, gas chromatographic analysis, 65-67 v o l a t i l e , a n a l y t i c a l method, 61-77 American Society f o r Testing and Materials, sensory evaluation of pepper heat, 37

177

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

Author Index Adams, Michael Α., 11 Albrecht, W., 43 Bille-Abdullah, H., 43 Busch, Kenneth L., 121 Ebeler, S., 95 Engel, K.-H., 43 G i l l e t t e , Marianne, 26 Hayashi, Tateki, 61 Idstein, Heinz, 109 Jennings, W., 95 Kr-oha, Kyle J . , 121 Krukonis, Val J . , 154 Lawless, Harry, 26

Schreier, Peter, 109 Shibamoto, Takayuki, 61 takeoka, G., 95 Tressl, R., 43 Westendorf, Robert G., 138

Subject Index Amino acids and peptides, HPLC analysis, 83 Analytical c h a r a c t e r i s t i c s , tandem mass spectrometry, 127-29 Analytical method, v o l a t i l e aldehydes, 61-77 Apple essence carbon dioxide extract, gas chromato gram, 17 3f depleted, gas chromatogram, 173f synthetic, gas chromatogram, 172f Apple v o l a t i l e s , gas chromatographic i d e n t i f i c a t i o n , 14 5t Aroma c h i r a l components i n trace amounts, analysis, 43-59 food, analysis by tandem mass spectrometry, 131-33 A r t i f a c t formation, automated analysis of v o l a t i l e flavor compounds, 146-48 Aspertame, HPLC analysis, 81,83 Automated analysis of v o l a t i l e flavor compounds, 138-52

A Acetate esters emitted from a banana, parent-ion tandem mass spectrum, 132f Acetoxyacid esters, hydrolysis, Candida u t i l i s , 51-54 2-Acetylthiazolidine, recovery e f f i c i e n c y , 72 Acid-catalyzed hydrolysis of saponins, 15,19 Alcohols c h i r a l , gas chromatographic separation, 44-46 o p t i c a l l y pure, formation, 51,52f Aldehydes and cysteamine, products and spec­ t r a l data, 66t standards, gas chromatographic analysis, 65-67 v o l a t i l e , a n a l y t i c a l method, 61-77 American Society f o r Testing and Materials, sensory evaluation of pepper heat, 37

177

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

178

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Β Banana, acetate ester emission, parention tandem mass spectrum, 1 32f Beer commercial, polyphenols, 90t HPLC analysis, 88,90t Bitterness, analysis, 83-84 Black pepper, chromatograms, 101f

C Calibration curve gas chromatographic, f o r t h i a z o l i d i n e , 69f preparation f o r formaldehyd analysis, 68 preparation f o r methyl glyoxa analysis, 69 Candida u t i l i s acetoxyacid ester hydrolysis, 51-54 2-heptyl acetate hydrolysis, 51 2-octyl acetate hydrolysis, 51 C a p i l l a r y gas chromatographic analysis, v o l a t i l e flavor compounds, 95-107 C a p i l l a r y gas chromatographic separation, o p t i c a l isomers of c h i r a l compounds, 43-59 Capsaicin effects of burning sensation, 29,32 HPLC analysis, 85 sensory responses, 27-33 Capsicum oleoresin, oral i r r i t a t i o n , 30f Carbon dioxide s o l u b i l i t y of naphthalene, 156-59 s o l u b i l i t y of t r i g l y c e r i d e s , 157,160f use as an extraction solvent, 96,98-104 Carbon dioxide extract of apple essence, gas chromatogram, 173f of ginger, gas chromatogram, I67f,l69f of pimento berries, gas chromatogram, 170f,172f Charge changes, tandem mass spectrometry, 126-27 Chemically induced oral heat as part of f l a v o r , 26-29 Cherimoya f r u i t v o l a t i l e s FTIR spectrum, 113f mass spectrum, 112f Chiral aroma components i n trace amounts, analysis, 43—59

Chiral compounds formed during microbiological processes, i n v e s t i g a t i o n , 50-54 Chloroform, quantity of formaldehyde i n commercial forms, 70 Chloroform extract of decaffeinated coffee, gas chromatograms, 73f of soy sauce, gas chromatograms, 75-76f Chromatogram(s) black pepper, 101f cinnamon essential o i l , 89f daisy, 104f Erigeron glaucus, 104f Eriophyllum staechadifolium, 103f Grindelia s t r i c t a venulosa, 102f gumplant, 102f

sunflower, 103f tarragon, 99f wine, 106f Cinnamon essential o i l , chromatogram, 89f Citrus limonoids, HPLC analysis, 84-85 C l i n i c a l alcohols, gas chromatographic separation, 44-46 Cluster analysis, sensory evaluation of food f l a v o r s , 8 Coffee formaldehyde content, 741 methyl glyoxal content, 74t Colorimetric analysis formaldehyde, 64 sugars, 80 Constant neutral loss tandem mass spectrum, description, 124f,126 Consumer t e s t i n g , sensory evaluation of food flavors, 3-4 Corn o i l automated analysis, I47f compounds, I44t recovery of v o l a t i l e s with increas­ ing temperature, I43f Cysteamine effect of quantity on t h i a z o l i d i n e recovery, 70 production of t h i a z o l i d i n e s , 62 D Daisy, chromatogram, 104f Daughter-ion tandem mass spectrum description, 124f,125 nootkatone emitted from grapefruit, 132f protonated molecular ion, 134f

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

INDEX

179

Decaffeinated coffee, chloroform extract, gas chromatograms, 73f Derivatization, sample, tandem mass spectrometry, 130 Descriptive analysis, sensory evalua­ tion of food f l a v o r s , 5-6,7f Desorption-chemical i o n i z a t i o n mass spectrometry jujubogenin-glucose-rhamnose, 20f sample preparation, 19,20f Dichloromethane, quantity of formal­ dehyde i n commercial forms, 70 Difference t e s t i n g , sensory evaluation of food flavors, 5 Diphenylpropanoids found i n nutmeg, parent-ion scans, 134f Discriminate analysis, sensory evalua­ tion of food flavors, 8 Duo-trio t e s t , sensory evaluatio food flavors, 5 Dynamic compression system, flavor analysis, 118-19 Dynamic headspace analysis, uses, 139 Dynamic headspace gas chromatographic system, gas-flow scheme, 14lf Dynamic headspace sampling, i s o l a t i o n of v o l a t i l e s prior to gas chromatographic analysis, 138-52

Ε Ebelin lactone, structure, 18f Endive v o l a t i l e s FTIR spectra, 11 7f s i l i c a gel f r a c t i o n , FID trace, 1l4f Erigeron glaucus , chromatogram, 104f Eriophyllum staechadifolium, chromatogram, 103f Extraction l i m i t a t i o n s , 96 s u p e r c r i t i c a l f l u i d , flavors applications, 154-74

F Factor analysis, sensory evaluation of food f l a v o r s , 8 Fermentation byproducts FTIR spectra, 11 5f s i l i c a gel f r a c t i o n , FID trace, 113f Flavanone glycosides, HPLC analysis, 84-85 Flavor(s) automated analysis, 138-52 c a p i l l a r y gas chromatographic analysis, 95-107

Flavor(s)—Continued chemically induced oral heat, 26-29 description, 95 extraction and concentration from solutions, 171-74 HPLC analysis, 79-92 HRGC-FTIR anlysis, 109-19 p r o f i l e s , orange and orange-flavored food starch, 1 50f sensory evaluation, 1-9 supercritical fluid extraction, 154-74 tandem mass spectrometric analysis, 121-37 Fluid extraction, s u p e r c r i t i c a l , flavors applications, 154-74 Food sensory evaluation of f l a v o r s , 1-9

analysis i n food samples, 72-77 c a l i b r a t i o n curve preparation, 68 i n chloroform, 70 i n coffee, 74t conventional a n a l y t i c a l methods, 63-64 i n dichloromethane, 70 effects, 61 forms, 62 i n soy sauce, 74t uses, 61 Fourier transform IR spectra cherimoya f r u i t v o l a t i l e s , 113f endive v o l a t i l e s , 117f fermentation byproducts, 115f Fourier transform IR spectroscopy-high resolution gas chromatography (FTIR-HRGC), f l a v o r s , 109-19 Free induction decay (FID) trace s i l i c a gel f r a c t i o n of endive v o l a t i l e s , 114f s i l i c a gel f r a c t i o n of fermentation byproducts, 113f F r u i t s , t r o p i c a l , enantiomeric com­ position of c h i r a l aroma constituents, 54-59 G Gas chromatogram(s) apple essence depleted, 1 73f synthetic, 172f carbon dioxide extracts apple essence, 173f ginger, I67f,l69f pimento berries , 170f,172f

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

180

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Gas chromatogram(s)—Continued chloroform extracts decaffeinated coffee, 73f soy sauce, 73f methylene chloride extracts ginger, I65f pimento berries, I69f Gas chromatography c a l i b r a t i o n curve for t h i a z o l i d i n e , 69f c a p i l l a r y , v o l a t i l e flavor compounds, 95-107 c h i r a l separations alcohols, 44-46 hydroxyacid esters, 47-50, lactones, 48-50, l i m i t a t i o n s , 95 and mass spectrometry, HPLC, 92 use with dynamic headspac i n analyzing v o l a t i l compounds, 138-52 v o l a t i l e aldehyde standards, 65-67 Gas-flow scheme, dynamic headspace gas chromatographic system, I4lf Gas-liquid chromatography, comparison to HPLC i n flavor studies, 79-80 Ginger, s u p e r c r i t i c a l f l u i d extraction of f l a v o r s , 164-68 Gingerol, sensory responses, 27-33 D-Glucose, decomposition products, 62 Grape j u i c e s , v a r i e t a l differences between white and red types, 151f Grapefruit, nootkatone emission, daughter-ion tandem mass spectrum, 1 32f Gravimetric analysis ginger extracts, I68t pimento berry extracts, 1711 Grindelia s t r i c t a venulosa, chromatogram, 102f Gumplant, chromatogram, 102f Gymnema sylvestre effects on the taste i n t e n s i t y of sucrose, 14f phylogenetic c l a s s i f i c a t i o n , I6f Gymnemagenin, structure, 14f Gymnemic acids mechanism of action i n taste modification, 19,21-23 structure, I4f sweetness-suppressing a c t i v i t y , 12-20 Gymnestrogenin, structure, 14f H Headspace sampling with an on-column i n j e c t o r , gas chromatography, 96,97f

Heat oral chemical, sensory responses, 26-41 pepper, sensory evaluation, 33,36-40 2-Heptyl acetate, hydrolysis, Candida u t i l i s , 51 High-performance l i q u i d chromatography (HPLC ) amino acids and peptides, 83-84 beer, 88,90t capsaicinoids, 85 c i t r u s limonoids, 84-85 comparison to gas-liquid chromatography i n flavor studies, 79-80 flavanone glycosides, 84-85 flavors, 85-91 hop b i t t e r acids, 83 l i c o r i c e extracts, 85

soy f l o u r , 92 sugars, 81 tea, 91 ter penes, 91-92 use i n flavor studies, 79-92 use with gas chromatography and mass spectrometry, 92 v a n i l l a extract, 85-87 v a n i l l y l nonamide, 85 High-resolution gas chromatographyFourier transform IR spectroscopy (HRGC-FTIR), f l a v o r s , 109-19 Hop b i t t e r acids, HPLC analysis, 83 Hydrolysis acetoxyacid esters, 51-54 2-heptyl acetate, 51 2-octyl acetate, 51 saponins, 15,19 Hydroxyacid esters , c h i r a l , gas chromatographic separation, 47-50

I Independent sequential analyses, tandem mass spectrometry, 123,124f I n h i b i t i o n of sweetness, 12-23 Instrumentation for automated analyses of v o l a t i l e flavor compounds, 140 Ion processing, tandem mass spectrometry, 122-23 Ionization, sample, tandem mass spectrometry, 130 I r r i t a t i o n , o r a l , psychophysical characterization, 29-35

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

181

INDEX

J Jujubogenin, structure, 18f Jujubogenin-glucose-rhamnose, desorption-chemical ionization mass spectrum, 20f L Lactones, c h i r a l , gas chromatographic separation, 48-50 Lavender and peppermint o i l s , chromatogram, 89f Licorice extracts, HPLC analysis, 85 L i z a r d t a i l , chromatogram, 103f

Methylene chlorine extract—Continued of pimento berries, gas chromatogram, I69f Microbiological processes, investigation of c h i r a l compounds formed, 50-54 M i r a c u l i n , taste-modifying e f f e c t , 11 Molar r a t i o , effect on methyl glyoxal and cysteamine reaction, 67 Monellin, taste-stimulating e f f e c t , 11-12 Multidimensional gas chromatography, packed c a p i l l a r y , use i n flavor analysis, 116-19 Multivariate analysis, sensory evaluation of food f l a v o r s , 6,8 Mushroom blanching water, HPLC analysis, 88,91

M Magnitude estimation rate versus sweet stimulus concentration, 21,22f sensory evaluation of food flavors, 3 sensory responses to oral chemical heat, 29 Mango, enantiomeric composition of c h i r a l aroma constituents, 54-59 Mass changes, tandem mass spectrometry, 125-26 Mass spectrometry cherimoya f r u i t v o l a t i l e s , 112f and gas chromatography, HPLC, 92 tandem, characterization of flavor compounds, 121-37 Mechanism of action, substances that modify the perception of sweetness, 19 Megabore column, description, 105-6 Menthol isomers, gas chromatographic separation, 46f R-(+)-gî- Met ho xy-Q?- t r i f 1 uor omet hyl phenylacetic acid derivatives, gas chromatographic separation, 44-50 Methyl glyoxal analysis i n food samples, 72-77 a n a l y t i c a l methods, 65 c a l i b r a t i o n curve preparation, 69 characteristics, 64 i n coffee, 74t and cysteamine reaction effect of molar r a t i o , 67 effect of pH, 68 foods containing, 61 i n soy sauce, 74t Methylene chloride extract of ginger, gas chromatogram, l65f

Naphthalene, s o l u b i l i t y and phase behavior, 156-59 Nasal passages, sensory perceptions, 27 Neutral loss tandem mass spectrum, negative ions of carboxylic acids, 132f Nonsugar sweeteners, HPLC analysis, 81,83 Nootkatone emitted from grapefruit, daughter-ion tandem mass spectrum, 132f Nutmeg, diphenylpropanoids, parent-ion s cans, 13 4f 0

2-0ctyl acetate, hydrolysis, Candida u t i l i s , 51 O i l s , HPLC analysis, 85-88,89f Optical isomers of c h i r a l compounds, c a p i l l a r y gas chromatographic separation, 43-59 Oral cavity, sensory perceptions, 27 Oral chemical heat, sensory responses, 26-41 Oral i r r i t a t i o n capsicum oleoresin, 30f ginger oleoresin, 31f piperine, 30-31f psychophysical characterization, 29-3 5 v a n i l l y l nonamide, 30-31f,34f Orange and orange-flavored food starch, flavor p r o f i l e s , 150f

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

182

CHARACTERIZATION AND MEASUREMENT OF FLAVOR COMPOUNDS

Ρ Packed-capillary multidimensional gas chromatography, use i n flavor a n a l y s i s , 116-19 Parent-ion scans, diphenylpropanoids found i n nutmeg, 134f Parent-ion tandem mass spectrum acetate esters emitted from a banana, 1 32f description, 124f,125 Passion f r u i t , enantiomeric composi­ tion of c h i r a l aroma constituents, 54-59 Peanut butter, comparison to roasted peanuts, 1 52f Pepper heat, sensory evaluation, 33,36-40 Peppermint and lavender o i l s chromatogram, 89f Perception of sweetness, substance that modify, 11-23 pH, effect on methyl glyoxal and cysteamine reaction, 68 R-(+)-Phenylethylisocyanate deriva­ tives , gas chromatographic separation, 44-50 Pimento berries, s u p e r c r i t i c a l f l u i d extraction of f l a v o r s , 168-71 Pineapple, enantiomeric composition of c h i r a l aroma constituents, 54-59 Piperine oral i r r i t a t i o n , 30-31f sensory responses, 27-33 Polyethylene glycol type phases, use i n gas chromatography, 98,105,106f Polyphenols i n commercial beers, 90t Postcolumn reaction systems, sugar detection, 81 Potato chips, sugar chromatogram, 82f P r i n c i p a l component analysis, sensory evaluation of food f l a v o r s , 8 Product d r i f t i n the food industry, 5 Protonated molecular i o n , daughter-ion tandem mass spectrum, 134f Psychophysical characterization of oral i r r i t a t i o n , 29-35 Psychophysics, d e f i n i t i o n , 2 Pulsed amperometric detector f o r sugars, 81 R Red pepper heat, sensory evaluation, 33,36-40 Regression analyses, sensory heat versus a n a l y t i c a l measurements of red peppers, 38t

Reproducibility, automated analysis of v o l a t i l e flavor compounds, 145-46 Resolution, tandem mass spectrometry, 123 Response surface methodology, sensory evaluation of food f l a v o r s , 8 Roasted peanuts, comparison to peanut butter, 152f S Saccharin, HPLC analysis, 81,83 Salivary flow, effect of oral chemical heat, 29,32 Sample c o l l e c t i o n , treatment, and contamination, tandem mass spectrometry, 129

hydrolysis, 15,19 Sausage sample, tandem mass spec­ trometric analysis of v o l a t i l e flavor compounds, 121-22,124f Scaling methodology, sensory evalua­ tion of food f l a v o r s , 3 S c o v i l l e determination red pepper heat, 33 sensory heat, 27 Sensation magnitude, function charac­ t e r i z i n g growth, 29 Sensitivity automated analysis of v o l a t i l e flavor compounds, 145 tandem mass spectrometry, 127 Sensory evaluation food f l a v o r s , 1-9 pepper heat, 33,36-41 Sensory responses to oral chemical heat, 26-41 Sequential analyses, independent, tandem mass spectrometry, 123,124f Signal-to-noise r a t i o , tandem mass spectrometry, 123,125 S i l i c a , s o l u b i l i t y i n water, 157,l6lf S i l i c a gel f r a c t i o n of endive v o l a t i l e s , FID trace, 1l4f of fermentation byproducts, FID trace, 113f Sodium dodecyl s u l f a t e mechanism of action i n taste modification, 19,21-23 sweetness-suppressing a c t i v i t y , 12-20 Solutions, extraction and concen­ t r a t i o n of f l a v o r s , 171-74 Solvent, effect on t h i a z o l i d i n e recovery, 69-70

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.

INDEX

183

Soy f l o u r , HPLC analysis, 92 Soy o i l sample, automated analysis, I49f Soy sauce chloroform extract, gas chromatograms, 75-76f formaldehyde content, 74t methyl glyoxal content, 74t S p e c i f i c i t y , tandem mass spectrometry, 128 Speed of analysis, tandem mass spectrometry, 128-29 Spice analysis, use of tandem mass spectrometry, 133-35 S t a t i s t i c a l analysis, sensory evalua­ tion of food flavors, 6,8 Sucrose, taste i n t e n s i t y , effects of Gymnema sylvestre, 14f Sugars chromatogram, potato chips colorimetric analysis, 8 HPLC analysis, 81 sweetness, time-intensity p l o t , 22f Sunflower, chromatogram, 103f S u p e r c r i t i c a l f l u i d extraction flavors applications, 154-74 h i s t o r i c a l perspective, 155-57 laboratory analysis, l65f process operation, 157,162-64 Sweet stimulus concentration, mag­ nitude estimation r a t e , 21,22f Sweeteners, nonsugar, HPLC analysis, 81,83 Sweetness analysis, 80-83 i n h i b i t o r s , 12-23 perception, substances that modify, 11-23

Thiazolidines—Continued recovery effect of cysteamine quantity, 70 effect of pH, 68f effect of solvent, 69-70 Trace amounts, c h i r a l aroma components, 43-59 Triangle t e s t , sensory evaluation of food f l a v o r s , 5 Trigeminal nerves, sensory perceptions, 27 Triglycerides, s o l u b i l i t y i n carbon dioxide, 157,l60f Tropical f r u i t s , enantiomeric composi­ tion of c h i r a l aroma constituents, 54-59

V a n i l l y l nonamide HPLC analysis, 85 oral i r r i t a t i o n , 30-31f,34f sensory responses, 27-33 Vegetable o i l s , compounds, I45t V o l a t i l e aldehydes a n a l y t i c a l method, 61-77 gas chromatographic analysis of standards, 65-67 V o l a t i l e flavor compounds automated analysis, 138-52 c a p i l l a r y gas chromatographic analysis, 95-107 V o l a t i l i t y range, automated analysis of v o l a t i l e flavor compounds, 142-45

W Τ

Tandem mass spectrometry a n a l y t i c a l c h a r a c t e r i s t i c s , 127-29 background, 122 characterization of flavor compounds, 121-37 problems and potentials, 135-37 Tarragon, chromatograms, 99f Tea, HPLC analysis, 91 Terpenes, HPLC analysis, 91 Thaumatin, taste-stimulating e f f e c t , 11-12 Thiazolidines c a l i b r a t i o n curve, 69f gas chromatogram, 66f proposed formation mechanism, 62

Water, s o l u b i l i t y of s i l i c a , 157,l6lf Wine chromatogram, 104f evaluation b a l l o t , 7f Ζ

Ziziphins mechanism of action i n taste modification, 19,21-23 sweet nés s - s uppr es s i ng a c t i v i t y , 12-20 Ziziphus jujuba i s o l a t i o n and p u r i f i c a t i o n of a n t i sweet substances, 17-l8f phylogenetic c l a s s i f i c a t i o n , 16f

In Characterization and Measurement of Flavor Compounds; Bills, D., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1985.