Photochemistry Volume 9 : a review of the literature published between July 1976 and 1977

Citation preview

A Specialist Periodical Report

Photochemistry Volume 9

A Review of the Literature published between July 1976 and June 1977

Senior Reporter D. Bryce-Smith, Department of Chemisfry, University of Reading Reporters N. S. Allen, Universify of Salford M. D. Archer, The Royal lnsfifution, London H. A. J. Carless, Birkbeck College, University of London R. B. Cundall, University of Salford R. Devonshire, Universify of Shefield A. Gilbert, University of Reading W. M. Horspool, University of Dundee M. Wyn-Jones, University of Salford J. M. Kelly, University of Dublin J. F. McKellar, Universify of Salford D. Phillips, Universify of Soufhampton S. T. Reid, University of Kent

The Chemical Society Burlington House, London, wiv OBN

British Library Cataloguing in Publication Data

Photochemistry. Vol. 9. - (Chemical Society. Specialist periodical reports). 1. Photochemistry I. Bryce-Smith, Derek 11. Series 541’.35 QD708.2 73-17909 ISBN 0-85186-085-0 ISSN 0556-3860

Copyright @ 1978 The Chemical Society All Rights Reserved No part of this book may be reproduced or transmitted in any form or by any means -graphic, electronic, including photocopying, recording, taping or information storage and retrieval systems - without written permission from The Chemical Society

Organic formulae composed by Wright’s Symbolset method

PRINTED IN GREAT BRlTAIN BY JOHN WRIGHT AND SONS LTD., AT T H E STONEBRIDGE PRESS, BRISTOL BS4 5NU

Contents Introduction and Review of the Year By D. Bryce-Smith

xiii

Part I Physical Aspects of Photochemistry Chapter 1 Spectroscopic and Theoretical Aspects By R. Devonshire

3

1 Introduction

3

2 Calculations of Electronically Excited States Atomic Systems Molecular Systems

3 7 8

3 Absorption Spectroscopy Thermal Lensing and Photoacoustic Spectroscopy Single-photon Absorption Spectra &-T, Absorption Spectra Sl-S, Absorption Spectra Tl-Tn Absorption Spectra Two-photon Absorption Spectra Multi-photon Absorption Processes Linear Dichroism Circular Dichroism Stark and Related Effects

19 19 22 24 25 25 26 30 31 33 36

4 Intramolecular Excited-state Decay Processes Theories of Radiative Decay Processes Radiative Lifetimes from Absorption Spectra Theories of Radiationless Decay Processes Tunnel-effect Theory Intersystem Crossing Internal Conversion Vibrational Relaxation The Singlet State Resonance Fluorescence Fluorescence An ti-Stokes Fluorescence Fluorescence Quantum Yields Fluorescence Polarization

38 38 39 40 41 43 44 45 46 46 47 49 51 53

-\Y

iv

/

Circularly Polarized Fluorescence The Triplet State Phosphorescence Optically Detected Magnetic Resonance (ODMR) Electron Spin Resonance (e.s.r.) Single Vibronic Level Studies

Contents 54 55 57 59 62 62

5 Intermolecular Excited-state Decay Processes Electronic Energy Transfer Vibrational Energy Transfer

67 68 70

6 Photochemical Excited-state Decay Processes Theories of Photochemistry Chemically Induced Dynamic Magnetic Polarization Magnetic Field Effects Photoionization Photodetachment Electron Transfer Photoisomerization Proton Transfer in the Excited State Photofragmentation

71 71 72

7 Laser-induced Photochemistry

86 86 87 88 88

Laser-induced Dissociation Laser-induced Photoionization Laser-induced Chemical Reactions Laser Isotope Separation 0ther Is0t ope Separation Techniques Laser-induced Effects

Chapter 2 Photophysical Processes in Condensed Phases By R. B. Cundall and M. Wyn-Jones 1 Introduction

73 78 79 80 80 81 83

90 90 92

92

92 2 Excited Singlet-state Processes Singlet Quenching by Energy Transfer and Exciplex 109 Formation Electron Donor-Acceptor Complexes and Related Charge109 transfer Phenomena Fluorescence Quencing by Inorganic Species 113 113 Heavy-atom Quenching Excimer Formation and Decay 114 3 Triplet-state Processes 116 Exciplexes, Triplet Quenching, and Energy-transfer Processes 125

,y

Contents

V

E.s.r. and Microwave Studies, and Related Triplet-state Topics

4 Physical Aspects of Some Photochemical Studies Photo-oxidation Photolysis Photoisomerization Photochromism Chemiluminescence Triboluminescence Chapter 3 Gas-phase Photoprocesses By D. Phillips

131 133 133 134 135 136 137 138 1 40

1 Alkanes and Alkenes

140

2 Aromatic Molecules

142

3 Carbonyls and other Oxygen-containing Compounds

153

4 Radical Reactions

159

5 Sulphur-containing Compounds

161

6 Nitrogen Compounds

162

7 Halogenated Compounds

166

8 Atom Reactions

171

9 Miscellaneous

174

10 1.r. Laser-induced Reactions, and Isotope Separation

176

11 Atmospheric Reactions Extraterrestrial Phenomena Upper Terrestrial Atmosphere Halocarbons and O3 Tropospheric Reactions and Photochemical Smog Detection of Atmospheric Constituents and Pollutants H Atom, H,, and H 2 0 Reactions 0 Atoms, O,, and O3 HO, Reactions Nitrogen, NO,, and HNO, Reactions CO and CO, Reactions SOa Reactions

178 178 178 179 181 182 183 184 186 188 190 191

Contents

vi

Part I/ Photochemistry of Inorganic and Organomefallic Compounds By J. M. Kelly 1 Photochemistry of Transition-metal Complexes Titanium Vanadium Chromium Molybdenum and Tungsten Manganese Iron Ruthenium Cobalt Rhodium and Iridium Nickel Platinum Copper Silver and Gold Zinc Mercury Lanthanides Act inides

195 197 198 198 201 201 202 204 208 210 212 212 213 213 214 214 214 216

2 Transition-metal Organometallics and Low-oxidation-state Compounds Titanium and Zirconium Chromium, Molybdenum, and Tungsten Manganese and Rhenium Iron and Ruthenium Cobalt, Rhodium, and Iridium Nickel Copper Mercury Thorium

217 218 218 224 225 230 235 235 236 236

3 Main-group Elements Anions Barium Boron Silicon Germanium and Tin Nitrogen, Phosphorus, Arsenic, Antimony, and Bismuth Oxygen and Sulphur Selenium and Tellurium Halogens and the Noble Gases

239 239 240 240 242 242 244 244 245 246

vii

Contents

Part //I Organic Aspects of Photochemistry Chapter 1 Photolysis of Carbonyl Compounds By W. M. Horspool

25 1

1 Introduction

25 1

2 Norrish Type I Reactions

253

3 Norrish Type 11 Reactions

263

4 Rearrangement Reactions

269

5 Oxetan Formation

273

6 Fragmentation Reactions

276

Chapter 2 Enone Cycloadditions and Rearrangements: Photoreactions of Cyclohexadienones and Quinones By W. M. Horspool

279

1 Cycloaddition Reactions Intramolecular Intermolecular Dimerization

279 279 283 288

2 Enone Rearrangements

290

3 Photoreactions of Thymines etc.

308

4 Photochemistry of Dienones Linearly Conjugated Dienones Cross-conjugated Dienones

312 312 315

5 1J-, 1,3-, and l&Diketones

320

Chapter 3 Photochemistry of Olefins, Acetylenes, and Related Compounds By W. M. Horspool 1 Reactions of Alkenes Addition Reactions Halogeno-olefins Group Migration Reactions cis-trans-Isomerization

336

336 336 339 339 340

...

Contents

Vlll

2 Reactions involving Cyclopropane Rings

341

3 Reactions of Dienes

358

4 Reactions of Trienes and Higher Polyenes

364

+

5 [2 21 Intramolecular Reactions

372

6 Dimerization and Intermolecular Cycloaddition Reactions

377

7 Miscellaneous Reactions

382

Chapter 4 Photochemistry of Aromatic Compounds By A. Gilbert

396

1 Introduction

396

2 Isomerization Reactions

396

3 Addition Reactions

402

4 Substitution Reactions

414

5 Intramolecular Cyclization Reactions

426

6 Dimerization Reactions

442

7 Lateral-nuclear Rearrangements

447

Chapter 5 Photo-reduction and -oxidation By H. A. J. Carless

449

1 Conversion of C=O into C-OH

449

2 Reduction of Nitrogen-containing Compounds

456

3 Miscellaneous Reductions

459

4 Singlet Oxygen

462

5 Oxidation of Aliphatic and Alicyclic Unsaturated Systems

463

6 Oxidation of Aromatic Compounds

47 1

7 Oxidation of Nitrogen-containing Compounds

473

8 Miscellaneous Oxidations

48 1

ix

Contents

Chapter 6 Photoreactions of Compounds containing Heteroatoms other than Oxygen By S. T. Reid 1 Nitrogen-containing Compounds

4 84

Rearrangement Addition Miscellaneous Reactions

484 484 507 51 1

2 Sulphur-containing Compounds

516

3 Compounds containing other Heteroatoms

522

Chapter 7 Photoelimination By S. T. Reid

526

1 Photodecomposition of Azo-compounds

526

2 Elimination of Nitrogen from Diazo-compounds

533

3 Elimination of Nitrogen from Azides

538

4 Photodecomposition of other Compounds having N-N Bonds

543

5 Photoelimination of Carbon Dioxide

544

6 Fragmentation of Organosulphur Compounds

546

7 Miscellaneous Decomposition and Elimination Reactions

551

Part / V Polymer Photochemistry By N. S. Allen and J. F. McKellar 1 Introduction

559

2 Photopolymerization Photoinitiation of Addition Polymerization Photocondensation Polymerization Phot ografting Photochemical Cross-linking

559 559 563 563 564

3 Optical and Luminescence Properties of Polymers

565

Contents

X

4 Photochemical Processes in Polymeric Materials Synthetic Polymers Polyolefins Polystyrenes and Related Polymers Poly(viny1 halides) Poly(viny1 alcohol) Pol yacrylates Polyamides Polyesters Rubbers Natural Fibres Miscellaneous Polymers Reactions of Singlet Oxygen Photodegradable Polymers Photosensitive Polymers Photoactive Additives

572 572 572 573 575 575 576 576 577 577

5 Photostabilization Processes in Polymeric Materials

582

6 Photochemistry of Pigments and Dyes

585

7 Appendix: Review of Patent Literature Photopolymerizable Systems Table A1 : Prodegradants and U.V. Sensitizers Table A2: Photostabilizers Table A3 : Optical Brighteners

588 588 590 592 598

577

579 580 58 1 58 1 58 1

Part V Photochemical Aspects of Solar Energy Conversion By M. D. Archer 1 Photochemistry

Valence Isomerization and Energy Storage Photochemical Decomposition of Water 2 Photoelectrochemistry at Semiconductor Electrodes

Photoelectrolysis of Water Titanium Dioxide and Titanate Anodes 0ther Electrodes for Pho t oanodic Oxygen Evolution p-Type Semiconductors for Photocathodic Evolution of Hydrogen from Water Regenerative Phot oelectrochemical Cells A

3

Photoelectrochemistry at Insulator Electrodes

603 603 605 606 606 607 609

611 612 615

Contents

xi 4 Photoelectrochemistry at Metal Electrodes Photogalvanic Cells and Effects Pigment-covered Metal Electrodes

615 615 616

5 Photochemistry in Vesicles, Liposomes, and Membranes

616

6 Photosynthesis

617

7 Photovoltaic Cells Inorganic Homojunct ion Cells Silicon 0ther Materials Inorganic Heterojunction Cells Metal-Semiconductor Cells Metal-Insulator-Semiconductor Cells Organic Semiconductors

618 618 61 8 618 619 619 620 622

Author Index

623

Introduction and Review of the Year

For Volume 9 of ‘Photochemistry’ we are pleased to welcome five new Reporters. Drs. Allen and McKellar have contributed the review of polymer photochemistry, and Professor Cundall and Dr. Wyn-Jones have taken over responsibility for coveringphotophysical processes in condensed phases. Dr. Devonshire has reviewed the spectroscopic and theoretical aspects of photochemistry over the two-year period July 1975-June 1977 inclusive. The chemical aspects of photobiology and developments in instrumentation and techniques will in future be reviewed on a biennial basis, so developments during the present year will be covered in Volume 10. There has been solid progress in most areas, with a good proportion of notable reports. Among developments in the physical photochemistry of small molecules, the observation, by Goddard and Csizmadia, of an excimer-like state of HP formed by the rectangular approach of two hydrogen molecules is of particular interest. Molecular fluorine has been rather neglected by spectroscopists (one understands why), but the recent use of fluorine in electron-beam pumped lasers has stimulated Hay and Cartwright to make some extensive calculations on excited states of this molecule. Likewise, the use of argon and other rare gases in lasers is stimulating studies of the corresponding diatomic species. A theoretical study of HCN by Shwenzer et al. has led to the surprising conclusion that the energies of several of the excited states appear to be higher than those of the corresponding states in HNC. Attention is drawn to some interesting ab initio improved-virtual-orbital (IVO) calculations on excited states of SF, (Hay). Ridley and Zerner have very successfully applied a modified INDO technique to calculate triplet state energies of benzene, pyridine, and the diazines, and Ha’s ab initio SCF calculations on s-tetrazine indicate that the lowest excited state is, after all, a triplet, and not the singlet which had been predicted from a localized exciton model. Gordon has provided further confirmation that excited CH, has a tendency to distort from tetrahedral symmetry. Brith-Lindler and Allen have revealed a new electronic transition in ethylene, and van der Werf has presented an important paper on decay processes in biacetyl. Harrigan and Hirota have described some very informative studies of total decay processes in the 3 m n * states of benzaldehyde and related molecules, based on microwave-induced delayed phosphorescence (MIDP) and phosphorescence-microwave double resonance (PMDR) techniques. The rapid development of laser photochemistry noted in previous Volumes continues apace, notably in application to isotope separation. 1.r. laser excitation of tetramethyl-1,Zdioxetan sensitized by CHsF produces luminescence in the visible region, and quantitative formation of acetone (Turro et aZ.). Ambartzumian

xiv

Photochemistry

and Letokhov have presented a very timely review of specific ‘thermal’ reactions induced by i.r. radiation from COa lasers: further new examples involving boron compounds have been described this year. Another application of lasers is in the ‘thermal lens’ technique, which seems likely to facilitate the measurement of extremely low extinction coefficients and has applications in observing S-T, vibrational overtone, and multi-photon absorptions, and detecting short-lived transient species. This technique involves a laser-produced transverse temperature profile in a liquid. It has been used to measure the visible absorption spectra of benzene and other aromatic molecules. Thus the lA,, --f lBzuabsorption in liquid benzene which normally appears over the 225-270 nm band appears over 465-540 nm in the two-photon absorption spectrum : see Long, Swafford, and Albrecht, and Twarowski, Klyger ; and compare Wunsch et al. A most interesting report by Zewail, Orlowski, and Jones describes the use of laser radiation to form true eigenstates and the singlet Born-Oppenheimer state. New picosecond and sub-picosecond laser techniques are permitting the direct measurement of vibrational relaxation times of polyatomic species, even large dye molecules, in dilute solution (Ippen et al., Ricard and Ducuing, Popovic and Menzel, inter alia). Romanovskis and Smits have proposed an explanation for the photo-orientation effect which can be induced by resonant plane-polarized light. Freed has presented a general theory of collision-induced electronic relaxation in ‘small’ and ‘intermediate case’ molecules. In several important papers, Formosinho has employed tunnel-effect theory to provide new insights into hydrogen-abstraction by excited carbonyl compounds, UOz+, and other species. The theory has been confirmed experimentally by Abbott and Phillips through measurement of the rate constants for hydrogen abstraction by TI benzophenone from tertiary amines, although some modification of parameters may be needed for other hydrogen donors. Kimbel and Mandel have presented and applied an important new theory of two-level atom resonance fluorescence: see also the related interesting study of atom recoil effects by Lam and Berman. Razi Naqvi and Wild have reported the first example of intermolecular triplet -f doublet energy transfer: see also Hodgson et al. and Topp. Attention is drawn to an important article by Harshbarger and Robin on applications of opto-acoustic methods in the determination of absolute yields oflumin escence and other processes which produce an increase in translational energy. Gosele and co-workers have presented a new theory of long-range energy transfer under conditions where both donor and acceptor can diffuse, and they have also provided some interesting extensions of the original Forster treatment. Salem has developed a theory of photochemical reactions which envisages the primary product as largely a singlet or triplet biradical or as one of the two possible zwitterions corresponding to biradicals, and he has introduced the new term ‘topicity’ which refers to the number and nature of reactive sites generated in a primary photochemical process. It is well known that in principle, and sometimes also in practice, applied magnetic fields can influence any process involving a change in the multiplicity of an intermediate. Such effects have been noted, for example, on the tendency for primary ‘cage’ recombination in photolysis of dibenzoyl peroxide (Tanimoto

Introduction and Review of the Year

xv et al.), the photodimerization of acenaphthylene (Ichimura and Watanabe), and in 13Cisotope enrichment in photolysis of dibenzyl ketone (Buchachenko et ai.): see also Haberkorn’s theoretical discussion of such effects. There seems to be a growing interest in this area. Until recently, 1,4-bonding in the benzene ring to produce the Dewar-isomer has been the only clear example of a non-dissociative reaction from an upper excited singlet state. de Mayo and his co-workers have now described stereospecific intermolecular addition reactions of adamantanethione which appear to occur from the 7r*S2 state. The chemistry differs from that previously reported for the nr* triplet state. Gotthardt and Nieberl have reported that S2 and Tl xanthone react with 1,2-dimethoxyethylenes in a non-stereospecific manner. Some interesting examples of energy transfer from T2 states have been reported by Ladwig and Liu. Benzene is used as energy carrier from the T. donor to the dicyclopentadiene acceptor. Complementary detailed reviews of benzene photochemistry have been presented by Cundall, Robinson, and Pereira and by Bryce-Smith and Gilbert; the former deals mainly with photophysical aspects whereas the latter is largely, though not wholly, concerned with the chemistry. van der Waals and his colleagues have presented two important papers on Tl benzene. They conclude that the structure can be greatly influenced by the crystal field. Thus the distortion from hexagonal symmetry is quinoid in a borazole host and antiquinoid in a C8D6host. Durnick and Kalantar have described phosphorescence polarization measurements which appear inconsistent with the conventional assignment of 7’, benzene as SBl,; but it would probably be premature to abandon this assignment for the alternative 3B2u. Birks has presented a useful discussion of discrepancies between theoretical and observed radiative lifetimes of excited aromatic molecules: benzene still seems to give more trouble in this respect than do the higher aromatic species. Noyes et ai. have shown that no significant photoisomerization of benzene to fulvene and benzvalene occurs at 266.8 nm, in contrast with the well known isomerization at 253.7nm. This observation supports previous proposals of other workers that meta-bonding in S1benzene occurs from an upper vibrational state. Controversial proposals of reversible intersystem crossing in benzene and naphthalene are critically assessed by David Phillips in Part I, Chapter 3. Howard and Schlag’s very high-resolution study of excitation of naphthalene at selected levels has enabled them to evaluate the effects of rotational excitation on rate constants for decay of the lBSustate. Schroder, Neusser, and Schlag have also described an important T-T single vibronic level study of this hydrocarbon. Turro et al. have obtained prismane, Dewar-benzene, and benzvalene indirectly via the S, state of (1); but the Tl state gives the diazocine (2) through a most

0 N-N

Photochemistry

xvi

unusual preference for fission of C-C rather than C-N. Electron affinities may provide a better guide than ionization potentials for predicting the course of photoadditions of ethylenes to the benzene ring (Mirbach, Mirbach, and Saus). Fluorescence studies are tending increasingly to concentrate on upper excited states these days, and two-photon excitation (TPE) is proving a popular technique. There is increasing use of the synchrotron as a light source for fluorescence decay measurements (see, e.g., Lopez-Delgado et al.). Hara et al. have described some interesting pressure-dependent shifts of luminescence both to longer and to shorter wavelengths, and Uchida and Tomura have observed that increase of pressure decreases emission decay times for crystalline naphthalene and pyrene. Luminescence procedures are being increasingly applied in biological systems, for example in amino-acid analysers (Lund et a/.). Morgan et al. have suggested a possible relationship between carcinogenicity of pol ycyclic aromatic hydrocarbons and the energy of their S , states. Further evidence has appeared that ‘dual phosphorescence’ (e.g. from xanthone in low temperature glasses) results from the existence of two distinct ground-state conformers, each having a characteristic triplet state (Pownall and Mantulin). Yang et al. have shown that medium-ring cycloalkanones exhibit unusually high phosphorescence quantum yields at 77 K, with a maximum (0.44)at Clo. Inoue et al. have described examples of intramolecular y-H transfer to vibrationally excited m* triplet states of alkenes in the gas phase which are analogous to Norrish Type I1 reactions of carbonyl compounds. Cohen et al. have observed that traces of an aliphatic (but not aromatic) thiol can promote the photoreduction of benzophenone by 2-butylamine, possibly by catalysing proton transfer from the amine radical-cation. Photoadditions of quadricyclane to anthracene and acridine have provided the first examples of [47~ 20 201 processes (Sasaki et d.).A termolecular exciplex appears to be involved in the photoreaction of anthracene with NN-dimethylaniline (Yang et al., Saltiel et al.). Few photoreactions of tetracyanoethylene are known, so the remarkable cyanation of a tertiary amine described by Ohashi et al. is of special interest. A considerable amount of evidence has been accumulating that the photolysis of many organic halides can produce both free radicals and carbonium ions, the latter probably being formed from the radicals by electron transfer within the solvent cage. Thus Kropp and his co-workers have shown that photolysis of vinyl iodides can provide a useful source of vinyl cations, and Mueller, Parlar, and Korte have employed the irradiation of aryl and vinyl halides in CDBODas the basis of a procedure for selective replacement of the halogen by deuterium in high yields: see also related papers by Charlton and Williams, and Gokhale et al. Bunnett et al. have reported the photoreaction of aryl and vinyl halides with enolate ions in liquid ammonia. Ausuabel and Wiznen have proposed the following unusual mechanism for photoisomerization of 1,2-dichloroethylenes.

+

+

Introduction and Review of the Year

xvii

Sasse and his co-workers have provided further interesting examples of photodimerization of naphthalenes, and have described a particularly remarkable product from 2-napthoic acid anhydride. Cohen et aE. have established the longsuspected fact that photodimerization of anthracene proceeds by way of an excimer : see also the related studies by Bouas-Laurant et al. Joussot-Dubien, Salem, and their co-workers have detected a short-lived transient formed by laser irradiation of 1-phenylcyclohexene which may be the highly strained transisomer. However, a previous report by White et al. of the photochemical production of a mono-trans-isomer of 1,3,4,6-tetraphenylcyclo-octatetraenehas been withdrawn. Nakazaki et al. have isolated the remarkable doubly transbridged ethylene ( 3 ) following xylene-sensitized irradiation of the cis-isomer. Among many interesting developments in cyclopropane photochemistry during the year, one may note the ingenious application by Spielmann et al. of the reaction ethylene + cyclopropane -+ cyclopentane as a key step in a

(3)

synthesis of the tris-homobenzene (4). Masamune et al. have reported an efficient new photochemical route to cyclobutadiene. On the other hand, Maier and Reisenauer have shown that irradiation of ( 5 ) in an argon matrix at 7 K gives the n-complex (6) rather than free cyclobutadiene; but at room temperature it gives the adduct (7), presumably by addition of cyclobutadiene to (5). Chambers and Maslakiewicz have provided evidence for the generation of a short-lived azacyclobutadiene by irradiation of a fluorinated pyridazine derivative.

\ - I

(5)

(7)

Attention is drawn to the interesting double-irradiation procedure employed by Griffin et al. in their study of oxygen ylides. Irradiation of formaldehyde in aqueous alkali gives pentaerythritol, 2-hydroxymethylglycerol, and carbohydrates (Shigemasa et al.). Wigfield et al. have observed that the reduction of ketones by NaBH, in diglyme can be accelerated by irradiation at 254nm. Wang has described an interesting photochemical peptide synthesis on a polymer support bearing pendant phenacyl groups. Binkley has observed that alcohols may be oxidized to aldehydes or ketones by irradiation of their pyruvic esters. A study of the n -+ n* excitation of polyene aldehydes by Langlet indicates that the excitation is localized on the carbonyl group, in agreement with the spectral properties. The dienone (8) is believed to be an intermediate in the photoisomerization of benzyl phenyl ether to p-benzylphenol (Bausch et al.).

Photochemistry

xviii

+

Sasaki et al. have reported that the mode of addition of norbornadiene to 21 type with phenanthraquinone and o-quinones tends to be of [2 acenaphthenequinone (+ oxetans), but of [4 21 type with the stronger electronacceptor o-chloranil (-+a dioxene). It will be interesting to know whether these exemplify a more general effect. Saito, Yazaki, and Matsuura have described the first example of a ‘walk rearrangement’ in which aziridine nitrogen undergoes migration. Irradiation of epoxydiphenylsuccinimide gives the unusually stable carbonyl ylide (9). Hata’s observation that intersystem crossing in the photoisomerization of isoquinoline N-oxide to isocarbostyril is promoted by application of a strong magnetic field provides yet another example of the phenomenon noted earlier in this review. The observation of Polo and Chow that toxic nitrosamines may be efficiently photodegraded in acid solution might well have some practical applications. Okazaki et al. have employed a photochemical procedure to provide the first example of a stable o-thioquinonemethide. Kobayashi et al. have described a novel photorearrangement of aromatic azoxysulphones to arenediazonium sulphonates. CIDNP studies with a 1-norbornyl derivative have indicated that the photoelimination of nitrogen from azoalkanes can involve successive rather than concerted fission of the two C-N bonds (Green, Dubay, and Porter); but results obtained by Greiner, Schneider, and Rau suggest that this conclusion may not apply generally, at least not in cyclic species. Barton and his co-workers have obtained an unusualIy stable steroidal dithiet by photoelimination of ethylene from the corresponding dithiin. Freerksen et al. have described a new photochemical procedure for functionalizing the C-18 angular methyl group in steroids (+ CH,CN). Interest in organophosphorus and organosilicon photochemistry seems to be increasing. Two groups have obtained the silaethylene Me,Si= CHMe by photolysis of Me,SiCHN,: it is stable below 45 K (Chapman et al., Chedekel et al.: cf. Ando et al.). Ishikawa et al. have described the following unusual reactions.

+

+

SiMe,

SiMe,OMe

+

Me O S iMe,OM e

Introduction and Review of the Year

xix

Photo-oxidation is being increasingly used as a synthetic technique. Among mechanistic developments, Davidson and Trethewey have shown that the common use of amines or /I-carotene as probes for lo, in dye-sensitized oxidations is liable to produce ambiguous results because of the ability of these compounds to quench S1Rose Bengal and methylene blue. They must therefore be used at very low concentrations. The same authors have also proposed a comprehensive rate expression for dye-sensitized photo-oxidations. Although most amines quench lo2‘dimol’ emission, Deneke and Krinsky have observed that cyclic tertiary diamines can enhance this. Kim et al. and Evans and Tucker have generated lo, directly from *02by laser excitation at high pressures. Dioxetans and endo-peroxides are attracting a markedly increasing level of interest (see Part 111, Chapter 5). Among other interesting reports, Wynberg et al. have prepared an optically active dioxetan from the corresponding optically active alkene and have reported that the chemiluminescent thermal decomposition involves the emission of circularly polarized light. The perennial question of perepoxide intermediates in dye-sensitized photooxidation reactions continues to be controversial. Thus Jefford and Borschung have been unable to reproduce the previous findings of Schaap and Faler (solvent pinacolone -+ t-butyl acetate) which had been taken as strong evidence for a perepoxide intermediate in the photo-oxidation of bisadamantylidene. It seems remarkable that the products now appear to be dependent on the nature of the sensitizer used. We may see some further interest in Schenck’s earlier proposal of a dye-oxygen complex : this might well undergo peroxide-like reactions. Such an idea does indeed seem to be suggested by a report (Shimizu and Bartlett) that alkenes can be converted into epoxides, often in greater than 90% yield, by benzil- or biacetyl-sensitized oxygenation in benzene solution. Apart from the obvious synthetic application, there is the intriguing possibility that a sensitizer-oxygen complex could be involved, as in the following scheme.

Dewar and Thiel’s calculations on endo-peroxidation of butadiene by lo2suggest a two-step pathway involving a perepoxide intermediate. Attention is drawn to the preparatively useful general synthesis of a-diketones via photo-oxygenation of enamino-ketones which has been described by Wasserman and Ives. Schaap, Burns, and Zaklika have reported a new chemiluminescent reaction based on silica-gel-catalysed isomerization of an endo-peroxide to a dioxetan. Binkley has reported that the photolysis of pyruvate esters provides a useful general and mild method for the selective oxidation of alcohols to carbonyl compounds, often in almost quantitative yields. A useful feature is that the acetaldehyde and carbon monoxide by-products are volatile and therefore easily

xx

Photochemistry

removed. Jones, Edwards, and Parr have described an interesting method for photo-oxidation of alkanes to alkenes which involves catalysis by cupric pivalate and benzophenone. Kwart et al. have observed that the lifetime of singlet oxygen can be extended by reversible formation of an adduct with phenyl ally1 sulphide: this again is relevant to the question of sensitizer-oxygen complexes in photooxygenation reactions. Davidson has noted that the methylene-blue-sensitized decarboxylation of or-ketocarboxylic acids in the presence of oxygen previously reported by Jefford et al., and thought to involve singlet oxygen, occurs even more efficiently in the absence of oxygen. The photochemical benzylic hydroxylation described by Libman and Berman employs nitrobenzene as the oxidant, and has already been applied practically in the steroid field. Among developments in inorganic photochemistry, van Quickenburne and Ceulemans, and Burdett have separately proposed important new alternatives to Adamson's rules for predicting the course of photosolvolysis reactions of octahedral metal complexes. Viscosity seems to be a less important factor than previously believed in determining quantum yields for photoaquation reactions in mixed solvents (Wong and Kirk; CJ Liu and Zink). Attention is drawn to a particularly interesting laser-excitation study of the photoaquation of cobalt complexes described by Langford and Vuik. Straws and Ford have made the observation, unsuspected in the light of previous work, that photoaquation of rhodium complexes can occur with inversion of configuration. The 2E state of [Cr(bipy),I3+, which is an intermediate in photosubstitution reactions, has been shown to react as a strongly oxidizing species (Balzani and co-workers). Katz and Gafney have reported that the photolysis of [Cr(NH3)5N3]2+proceeds uia a nitrene complex [Cr(NH,),NI2+ rather than the redox process previously suggested : compare the observations of Inoue, Endicott, and Ferraudi on the corresponding rhodium complex, and the interesting work on pentacyano-complexes by Miskowski, Nobinger, and Hammond. Betts and Buchanon have made the interesting observation that the carbon dioxide evolved on irradiation of potassium ferrioxalate is isotopically enriched in 12Cto an extent dependent on wavelength. Photochemical disproportionation of Cut to Cuo and Cu" has been reported (Loginov and Shagisultanova: Volger and Kern). [Ru(bipy)J2+ and related complexes continue to attract considerable attention, but last year's report of the ability of a complex of this type in a monolayer assembly to catalyse the photodissociation of water has not been confirmed in studies using more highly purified material. Some as yet unidentified impurity appears to be necessary. Various flash spectroscopic quenching studies have confirmed that the reduction potential for [Ru(bipy),12+/[Ru(bipy),]+ is close to the thermodynamic limit of about 0.8 V. The latter complex reduces oxygen to give the superoxide ion. A small anodic photocurrent is produced at an SnO, semiconductor electrode on irradiation of [Ru(bipy),lz+ solutions (Gleria and Memming), and a much larger cathodic photocurrent is produced using this species in acid solution in the presence of a mild oxidant such as O z , methylviologen, or Fe3+ (Kobayashi et al., J. Phillips et al.). Peterson and Demas, and Giordano et al. have used ruthenium complexes to provide the first direct evidence for acid-base equilibria in excited states of metal complexes.

Introduction and Review of the Year

xxi

Hydrogen is produced on irradiation of Ti3+ solutions in the presence of catalytic amounts of Cu+, from [Mo,(SO,),]~- in sulphuric acid solution (Stevenson and Davis), and also from irradiation of aqueous ethanolic solutions containing V2+ (Koryakin et al.). The evolution of hydrogen on irradiation of Eu2+in acid solutions occurs with a quantum yield which varies with the square root of the hydrogen ion concentration (Davis et al., inter alia). Jacobs et al. have made the interesting observation that the photoreduction of Ag+ in zeolites leads to the evolution of oxygen, and that hydrogen is released on heating to 600 “C. Workers in the field of metal carbonyls should note the three important papers by Poliakoff on absolute quantum yields for the reaction Mo(CO), Mo(CO), CO in argon and methane matrices, and selective photoisomerization of pentaco-ordinate chromium and tungsten carbonyl-thiocarbonyl species. Little has been known on the photochemistry of metal cluster compounds, so the reports on iron and copper cluster species merit attention (Bock and Wrighton, and Jarvis, Pearce, and Lappert respectively). Burdett and co-workers have reported the interesting phenomenon that irradiation of Fe(CO), with plane-polarized light in CO matrices at 20 K induces reorientation of the molecules such that absorption of light having the particular polarization employed is reduced. Crichton and Rest have reported examples of the photo-fixation of molecular nitrogen. Thus irradiation of Co(CO),(NO) in a nitrogen matrix gives Co(CO),(NO)(N,) and probably Co(CO)(NO)(N,),. Attention is drawn to an interesting paper by Grossweiner and Baugher on the behaviour of photoproduced solvated electrons, for example from I- -+ I * + e-aq. The recombination lifetime is about lo4 times greater than that predicted from the Noyes theory, probably owing partly to the large initial displacement from the radical co-product. There is a marked growth of interest in the preparation of active catalysts for hydrogenation, polymerization, olefin dismutation, and silane addition reactions, etc., and many further examples have been described this year (see Part 11). Ramsey and Anjo have trapped the ‘boryne’ species RB (R = 1-naphthyl) which appears to be formed on irradiation of tri-1-naphthylboron. In the field of polymer photochemistry, new procedures for radical photografting and cross-linking may be possible using polymers rendered photochemically active by incorporation of vanadium chelate residues (Aliwi and Bamford). There is considerable technological interest in photochemical crosslinking, and various workers have described the production of hybrid polymers by various photografting procedures. Charge-transfer photopolymerization, e.g. by tetrahydrofuran-Br, and pyromellitic anhydride, has been attracting increased attent ion. Oxygen-induced excimer fluorescence (‘oxiplex’ emission) has been observed from polystyrene doped with aromatic compounds (Kenner and Khan). Exciplex emission in polymer matrices can differ significantly from that in fluid solution (Martic et al.). Luminescence spectroscopy is being widely used to identify minor light-absorbing groups in polymers which may be involved in the

+

--f

xxii

Photochemistry

photodegradation. Conjugated enone and dienone groups are often found, notably in polypropylene (see Allen and McKellar, inter aliu). On the other hand, it has been found that the photostability of polypropylene can be markedly improved by heating in an inert atmosphere in order to decompose hydroperoxide groups (Chakraborty and Scott; cf. Carlsson et al.), Yet Allen, Bullen, and McKellar have expressed some doubt concerning the role of hydroperoxides in the degradation of polyethylene. Nicholls and Pailthorpe have obtained further evidence to indicate that the photo-yellowing of wool results from the reaction of tryptophanyl residues with singlet oxygen. Some progress has been made in understanding the various complex mechanisms by which photostabilizers operate in polymers, although many features still remain uncertain (see Carlsson, inter alia). New techniques continue to be developed, for example the grafting of o-hydroxybenzophenone stabilizers onto polymer chains (Birchill and Pinker ton). Among many aspects of atmospheric and extraterrestrial photochemistry, the much-debated question of ozone destruction in the upper atmosphere by halocarbons and other products of human activities still occupies a prominent position. Dr. Archer’s review of solar energy aspects contains much of general photochemical interest, especially for those photochemists whose horizons extend significantly beyond their laboratory walls. Talbert et al. have re-defined the conditions which must be met for a photochemical solar energy storage system to be more advantageous in use than a conventional solar thermal system. Encouraging improvements in the performance of thin film, amorphous, and polycrystalline silicon, indium phosphide, and other photovoltaic cells continue to be reported. Metal-insulator-semiconductor (MIS) cells are also looking promising, but single-crystal cells based on GaAlAs still seem to have some of the highest efficiencies (up to ca. 22%). ‘Regenerative photoelectrochemical cells’ can show conversion efficiencies up to ca. 10% at 633 nm (Ellis et d). These cells contain a redox couple in solution, one electrode reversible towards this couple and the other a semiconductor such as n-CdTe (usually the anode). The solute consumed at one electrode is regenerated at the other. These cells are attracting great attention, and further improvements are to be expected. As mentioned above, the non-reproducibility of the photodecomposition of water using a ruthenium-bipyridyl complex in a monolayer assembly has unfortunately been confirmed (Sprintschnick and co-workers, Valenty and Gaines). Another approach to the photoelectrolysis of water discussed in two papers by Butler involves the use of cells having p - and n-type semiconductors for the cathode and anode respectively. Osa and Fujihira have described cells having electrodes such as TiO, and SnO, ingeniously modified by covalent chemical bonding of a dye molecule such as rhodamine B. D. Bryce-Smith

Part I PHYSICAL ASPECTS OF PHOTOCHEMISTRY

1 Spectroscopic and Theoretical Aspects BY R. DEVONSHIRE

1 Introduction This Chapter covers the two-year period July 1975 to June 1977 inclusive. The format is broadly similar to that followed by David Phillips in previous years. My aim has been to make papers relatively easy to find, given that Specialist Periodical Reports do not contain a subject index. The selection of papers for the areas covered in this section is of necessity a very limited one; a simple listing of relevant papers would more than fill the available space. The end result is somewhere between being a telephone directory and a personal scrapbook. A number of areas have developed significantly since this section last appeared and coverage of Fhese is reasonably complete, uiz. optoacoustic and related studies, single vibronic level studies, especially of the simple dicarbonyls, magnetic field effects, and laser photochemistry in general. No attempt has been made to report on the extensive literature on the photolysis of atomic or crystalline systems. My overriding impression having completed this report of current work in a number of areas in photochemistry is of the subject’s tremendous vitality at the present time. 2 Calculations of Electronically Excited States The papers reported in this section are concerned with developments in the methods of calculation of spectroscopic and other related excited-state properties of atoms and molecules of interest in photochemistry. Although the coverage of calculations on particular systems is concentrated in this section, a number of papers containing extensive calculations appear in later sections. An inequality formulation of the well known Hund’s rules has been investigated numerica1ly.l Restricted Hartree-Fock SCF energies of various states of atoms and ions of the He, C, N, and 0 isoelectronic series were used to examine the behaviour of AE, A, and A, the differences in total energy, electronelectron repulsion, and electron-nuclear repulsion respectively, in going from a singlet to a triplet state. AE was found to be positive in all cases because A was always positive and large enough to more than compensate for occasional negative values for A found, in particular, for the neutral atom cases. Viewed as a two-step process, the change from singlet to triplet involves firstly a conversion with frozen orbitals to an approximate triplet state which will certainly be at lower energy because of the smaller V,e, followed by a J. P. Colpa, A. J. Thakkar, V. H. Smith, jun., and P. Randle, Mol. Phys., 1975, 29, 1861.

3

4

Photochemistry

contraction of the wavefunction which results both in an increase in K e , thus making it possible for an overall change in Ge, A(Ge), that is negative, and in a large decrease in Gn which makes A always positive. A recent SCF theory of excited states has been used to calculate hole states of free atoms.2 In the method each shell is specified by a subset of orbitals and an occupation number. A suitable choice of starting orbital exponents is found from the equivalent core (EC) approximation. The results of calculations on various hole states of Ne, Na, Mg, and S are in excellent agreement with experimental measurements, and the method should prove to be particularly valuable in the interpretation of Auger and shake-up processes as it is successful for any number of open shells. The use of the one-Hamiltonian and sequential orthogonalization methods in the calculation of particle and hole states has been discussed3 and a number of SCF methods used in the calculation of excited states are compared in a paper which extends recent restricted Hartree-Fock calculations in high-spin open shells to singly excited ~ t a t e s . The ~ concept of localization in molecular excited states is critically examined in an interesting paper which proposes the use of localized molecular orbitals to describe delocalization phenomena.S It is demonstrated how such a change from the usual discussion in terms of delocalized molecular orbitals can give a valuable insight into configuration interaction processes. An algorithm has been suggested to solve the divergence problem of the quadratically convergent SCF method,B and the CND0/2 method with configuration interaction has been successfully applied to the problem of excitedstate geometries in a study of the lowest excited singlet state of a number of small molecule^.^ An algorithm for the calculation of electronic structure and spectra using the CND0/2 CI method has also been published.8 The energies of the alternant molecular orbitals, AMO’s, for any alternant homonuclear molecule having a singlet ground state have been shown to be related to the conventional Hartree-Fock MO energies by a simple expression involving the correlation corre~tion.~A combined A M 0 and generatorco-ordinate method has been used to study excitations in the electron-gas.10 Ab initio SCF-LCAO calculations of a number of simple molecules, mainly diatomic species, were found to be sufficiently accurate to give the general character of the electron distribution.ll A theoretical approach which combines the ideas of the valence-electron and Thomas-Fermi-Dirac theories differs from earlier related treatments in that the core potential is completely nonempirical and is the same for all valence levels.12 The effects of many-electron D. Firsht and R. McWeeny, Mol. Phys., 1976, 32, 1637. E. R. Davidson and L. Z. Stenkamp, Internat. J. Quantum Chem., Symposia, 1976,10, 21. J. W. Caldwell and M. S. Gordon, Chem. Phys. Letters, 1976, 43, 493. J. Langlet and J.-P. Malrieu, in ‘Localisation and Delocalisation in Quantum Chemistry’, Vol. 11, ed. 0. Chalvet, R. Daudel, and S. Diner, Reidel, Dordrecht, Netherlands, 1976, p. 15. I. C. Chang, Chem. Phys. Letters, 1975, 36, 611. P. Zahradnik and J. Leska, Chem. Phys. Letters, 1976, 41, 293. 8 P. V. Zhukov and V. A. Gubanov, Zhur. strukt. Khim., 1975,16,326. N. N. Tyutyulkov, Internat. J. Quantum Chem., 1975, 9, 683. l o B. Laslowski, P. van Leuven, and 1. L. Calais, Physica, 1975, 80A, 561. l1 A. V. Nuikkanen, V. I. Perevozchikov, and L. A. Gribov, Zhur. priklad. Spektroskopii, 1975, a a

22, 1082. l2

J. Goodisman, Internat. J. Quantum Chem., 1976, 10, 341,

Spectroscopic and Theoretical Aspects 5 rearrangements in excited states on the expressions for the electronic excitation energy have been discussed, and model calculations of transition energies for Be indicate that relaxation corrections are substantial for core excitations, but are small for valence tran~iti0ns.l~ The use of Green’s functions in the calculation of energy levels from spectral properties is featured in a general review of the application of Green’s functions in quantum mechanic^.^^ Perturbation treatments based on Green’s functions have been used in the calculation of excitation energies in v-electron systems l6 and excited state lifetimes in atomic systems.16 The calculation of excited states using the natural transition orbital (NTO)method has been proposed for systems where the ground state is given by a correlated wa~efuncti0n.l~Ab initio calculations of energies using floating spherical Gaussian orbitals (FSGO’s) have been reviewed l8 and their special features well illustrated in calculations for a number of simple molecules using a variety of basis sets of fixed and floating SGO’s.lS The results are compared with those obtained using STO’s and GTO’s. A new multiconfiguration method for the calculation of excited states of atoms and molecules has been described.20 The method is stable and avoids some of the common problems of root flipping and poor convergence by shifting the desired SCI root below the other roots. Test calculations on the 2Sstates of Li and the lS states of Be gave results very close to the full CI energies. A different approach is presented by the so-called generalized Brillouin theorem (GBT) multiconfiguration method which is suitable for the calculation of (highly) excited states.21 Very good agreement with experimental values for excited states of Li is found in test calculations using the method. Test calculations on valence-electron excited states of C, H20, and CH20 using the GBT-MC method have also been carried out.22 Optimization procedures in the multiconfiguration SCF method have been Projected-unrestricted Hartree-Fock theory has been applied to the calculation of electronic spectra for symmetric m01ecules.~~The results of these PUHF calculations for a number of simple molecules using the all-valence-electron (INDO) approximation compare favourably with results obtained using CI methods, and significant improvements are expected from ab initio PUHF calculations. The effects of electron correlation on the excited-state energies and oscillator strengths of atoms and the methods of calculating them have been reviewed.25 In the recently developed self-correlated field method, correlation effects are concentrated on electron pairs associated with the same space orbital but with Is l4 l6 la l7 l8

2o

aa 2a

P. W. Deutsch and T. C. Collins, Internat. J. Quantum Chem., 1975, 9, 213. S. M. Blinder, Znternat. Rev. Sci., Phys. C k m . , Ser. Two, 1975, 1, 1. V. J. Ihaya, S. Narita, K. Yamaguchi, and M. Nakayama, Prog. Theor. Phys., 1976,55, 1685. H. M. M. Mansour and K. Higgins, Nuom Cfmento SOC.ital. Fis. A , 1976,36, 196. K. Deguchi, K. Kanno, and H. Miyagawa, Chem. Phys. Letters, 1976,39, 169. M. Afazl and J. Ahmad, Pakistan J. Sci. Znd. Res., 1975, 17, 113. R. M. Archibald, D. R. Armstrong, and P. 0.Perkins, Rev. Roumuine Chim., 1975,20, 1371. F. Grein and A. Banerjee, internat. J. Quantum Chem., Symposia, 1975, 9, 147. W. H. E. Schwarz and T. C. Chang, internat. J. Quantum Chem., Symposia, 1976,10,91. T . C. Chang and W. H. E. Schwm, Theor. Chim. Acta, 1977,44,45. S . Polezzo, Theor. Chim. Acta, 1975, 40,245. J. C. Schug, B. H. Lengsfield, and D. A. Brewer, Internat. J. Quantum Chem., \977,11, 591. A. W. Weiss, Ado. At. Mol. Phys., 1973, 9, 1.

6

Photochemistry

opposite spin functions.26 When applied to the 2S and 2P ground and excited states respectively of three-electron atomic systems, which should reveal the effects of the outer electron on the inner pair, it is found that the inner-outer correlation effects are probably more important for the ground state than for the excited state. In neutral and singly ionized first-row atoms there are several states belonging to the 2s2pm configurations which are not the lowest for their symmetry. Calculations of oscillator strengths for transitions involving these states using restricted Hartree-Fock (RHF) methods have not been successful unless charge wavefunctions are used for both the initial and final states. Wavefunctions which do not suffer variational collapse to the lower states of the same symmetry for these non-lowest 2s2pmstates have recently been calculated using a non-closed shell many-electron theory (NCMET) 27 and successfully applied to the calculation of oscillator strengths involving O(1) and O(I1) transitionsY2* N(1) and N(I1) transition^,^^ and C(1) and F(II) transition^.^^ Correlation effects have been reviewedYs1discussed for the excited states of a ten-electron and treated in the spin-density-functional f o r m a l i ~ r n . ~ ~ A classification scheme for the doubly excited states of a two-electron atom has been given 34 and triply excited states of a three-electron atomic system have been discussed within a formalism whereby many-electron systems are described by a large number of product functions of different configuration^.^^ The calculation of atomic electric dipole oscillator strengths is of considerable current interest and this results, in part, from their relevance to studies in astrophysics and plasma physics. There have been a number of developments in this area, including the formulation of a reasonably complete non-relativistic manybody theory which gives accurate results for single and multiple excitations to discrete and autoionizing states in atoms. The first-order theory of oscillator strengths (FOTOS) 36 is able to predict the electron correlations which are important to the oscillator strength and has been successfully applied to oneand two-electron excitations in Li(I), N(I), and N(T1) 36 and to a number of other An important step in the theory is the derivation of ‘first-order symmetries’ by finding out the type of angular distortions which are allowed for initial and final states owing to the symmetry of the dipole operator. Taking the authors’ example,38the first-order symmetry distortion of the Be ground state whose Fermi-sea (a set of spin orbitals 39) configurations are 1s22s2, ls22p2yields

as 28

31

sa 33 34

36 3t3 s7

38

R. Lissillour and C. R. Guerillot, Internat. J . Quantum Chem., 1975, 9, 627. W. L. Luken and 0. Sinanoglu, Phys. Rev. ( A ) , 1976, 13, 1293. W. L. Luken and 0. Sinanoglu, J . Chem. Phys., 1976,64, 1495. W. L. Luken and 0. Sinanoglu, J. Chem. Phys., 1976, 64,3141. W. L. Luken and 0. Sinanoglu, J. Chern. Phys., 1976,64,4680. U. Fano and C. D. Lin, At. Phys., 1975, 5, 47. A. N. Ivanova and N. V. Rabin’kina, Optika i Spektroskopiya, 1975, 38, 1049. 0. Gunnarsson and B. I. Lundqvist, Phys. Rev. (B), 1976, 13,4274. S. I. Nikitin and V. N. Ostrovsky, J. Pliys. (B), 1976, 9, 3141. M. Ahmed and L. Lipsky, Phys. Rev. (A), 1975, 12, 1176. C. A. Nicolaides and D. R. Beck, Chem. Phys. Letters, 1975, 36, 79. (a)C. A. Nicolaides and D. R. Beck, J. Chem. Phys., 1976, 66, 1982; (b) D. R. Beck and C. A. Nicolaides, J. Quant. Spectroscopy Radiative Transfer, 1976, 16, 297; (c) D. R. Beck and C. A. Nicolaides, Canad. J. Phys., 1976, 54, 689. D. R. Beck and C. A. Nicolaides, Internat. J. Quant. Chem., Symposia, 1976, 10, 119. D. R. Beck and C. A, Nicolaides, Internat. J. Quant. Chem., Symposia,1974,8, 17.

Spectroscopic and Theoretical Aspects

7

for the Yoexcited states the following two symmetry ‘configurations’: (sssp), (sspd). For the Be ls22s2p IPO excited state the corresponding first-order symmetry configurations are: (sspp), (ssss), (sssd). The third one of these is discarded because it does not yield an overall lS symmetry. The calculation proceeds by calculating the radial function for each of the allowed symmetries. The theory also has semiquantitative value in that it can predict those electron correlations which are important in photoelectron spectroscopy, in the photoionization of noble gases, and on molecular transition probabilitie~.~~ Many-body perturbation approaches to the calculation of transition probabilities have been reviewed.40 The oscillator strengths of a number of doubly ionized elements of astrophysical interest have been calculated 41 as have those for the 2s3p-2s3d transition in C , N, and 0 using a model Hamiltonian which includes an effective potential for an optical electron in a many-electron atom and in a positive The calculation of oscillator strengths using a model potential method has been compared with the quantum defect method.43 The important and much debated question of which of the three formally equivalent expressions for the non-relativistic oscillator strengths, written in the length, velocity, or acceleration forms respectively, is the more appropriate has been clarified in a paper which identifies conditions for which good agreement is found between all three expression^.^^ This is illustrated by caIcuIations on the 1s22s22p2Po-ls22s2paaD transition in C(I1) and O(1V). Atomic Systems.-Of the numerous papers concerned with the calculation of excited-state energies and oscillator strengths in particular atomic systems, only a few are reported on here. A generalized oscillator strength equation has been applied to the case of initially excited H-like The FOTOS method was used to calculate the oscillator strength of the Li ls22s 2S-2Po transition in a paper which is concerned with highly excited states of atoms which interact with Rydberg and continuum states of the same ~ y r n m e t r y .A ~ ~comparison of non-relativistic and relativistic oscillator strengths for transitions in singly ionized Si, Ge, Sn, and Pb shows the importance of spin-orbit interactions of the optical electron.47 The Green’s function method has been used to calculate mean excitation energies of ground- and excited-state hydrogen-like atoms and a number of papers deal with aspects of the excited states of helium, namely the behaviour of interelectronic angular distribution f u n ~ t i o n s CI , ~ ~wavefunctions for the 3s stateYs0 solutions to the Schrodinger equationYK1 and model calculations relevant to studies of non-adiabatic core polarizations and penetration corrections in alkali-like atoms.62 A technique to calculate approximate Brueckner or H. P. Kelly, At. Inner-shell Processes, 1975, 1, 431. E. Bremont, J. Quant. Spectroscopy Radiative Transfer, 1976, 16, 137. I. I. Gutman, Optika i Spektroskopiya, 1975, 39, 1001. 43 I. V. Avitova and L. I. Podlubryi, Optika i Spektroskopiya, 1975. 38, 1059. 44 C. A. Nicolaides and D. R. Beck, Chem. Phys. Letters, 1975, 35, 202. 46 M. Matsuzawa, K. Omidvar, and M. Inokouti, J. Phys. (B), 1976,9,2173 46 C. A. Nicolaides and D. R. Beck, J. Chem. Phys., 1977, 66, 1982. 47 J. Migdalek, J. Quant. Spectroscopy Radiative Transfer, 1976, 16, 265. 4a I. Shimamura, J. Phys. SOC. Japan, 1976, 40, 239. 4g K. E. Banyatd and D. J. Ellis, J. Phys. (B), 1975, 8,2311. 6o F. W. Birss and P. Senn, J. Chem. Phys., 1975, 63, 1276. a I. L. Hawk and D. L. Hardcastle, J . Compur. Phys., 1976, 21, 197. 6a R. H. Young and W. J. Deal, Znternat. J. Quantum Chem., 1975, 9, 835.

40 41

Photochemistry natural orbitals for systems with a single valence electron has been used to calculate the hyperfine interaction of some excited states of alkali atoms,6Sand non-relativistic Hartree-Fock solutions for excited states of In+ in the form of analytical expansions in STO’s have been obtained?* Molecular Systems.-A many-body perturbation theory suitable for the calculation of molecular excited states has been used to calculate the two lowest singlet and triplet states of Zg+, Xu+,nu,and n, symmetries for molecular hydrogen with very good accuracy.s5 The method uses Brandow expansion diagrams, which can be used for degenerate states and which are a development from the linked Goldstone perturbation diagrams used in calculations of energy levels in atomic systems. SCF-CI calculations of the B”Z,+ and D’n, states of the hydrogen molecule were carried out to give the values of the electronic wavefunctions at the crossing point of their potential curves.66 The values were required in a study of collision-induced predissociation. Calculations on the uncharacterized third l&,+ state of H2show that its Born-Oppenheimer potential curve has a double minimum.57 The inner (Rydberg-like) minimum occurs at about 1.99 bohr and the deeper outer (‘ionic’) minimum at 3.30 bohr. The state has been assigned the spectroscopic term symbol G + K1Zg+(lsa3da+ 2p7r2): this accounts for spectral features previously assigned to separate G and K electronic states. Non-exponential wavefunctions in elliptical co-ordinates for a number of electronic states of the molecular hydrogen positive ion, Ha*, have been described.s8 A calculation of the 2Zu+first excited state of H2+ has been made using a modified form of the MS-MA perturbation theory which improves its short-range p e r f o r m a n ~ e . ~An ~ excimer-like state has been identified in a study of the ground and low-lying excited states of H4 formed by the rectangular approach of two H2 Ab initio SCF-MO-CI calculations were carried out and one of the excited states of H4 could be correlated with ground (X’Zu+) and first excited singlet (B’Z,+) states of the H2molecule. The results are illustrated in Figure 1. When the results of calculations on the ground and excited states of ozone using an SCF Xa scattered wave method were compared with those from more sophisticated methods which take explicit account of electron correlation, they were found to give an adequate description of the lower excited states, but not the higher states which are more ionic in character.6f Comparatively little theoretical or experimental work has been carried out on the electronic spectroscopy of molecular fluorine, F2. The recent application of fluorine in electron-beam pumped lasers has naturally led to increased interest in its spectroscopic properties. A preliminary report of some extensive CI calculations on excited states of F2and Fa+as a function of internuclear distance 8

6s 64

58 67 68 tio

60

I. Lindgren, J. Lindgren, and A. M. Martensson, 2. Phys. (A), 1976,279, 113. R. Bauer, K. Differt, and L. Schwan, J. Phys. (Paris), Colloq., 1976,7, 177. U. Kaldor, J. Chem. Phys., 1975,63,2199. A. G. Ritchie, Internut. J. Quantum Chem., 1976, 10, 1071. R. M. Glover and F. Weinhold, J. Chem. Phys., 1977,66,303. I. Tamassy-Lentei, Actu Phys. Acud. Sci. Hung., (a) 1976,40, 111; (b)ibid., 1976,40,777. M. Battezzati and V. Magnasco, J.C.S. Faraduy ZI, 1976,72, 508. J. D. Goddard and I. G. Csizmadia, Chem. Phys. Letters, 1976, 43, 73. R. P. Messmer and D. R. Salahub, J. Chem. Phys., 1976,65,779.

9

Spectroscopic and Theoretical Aspects

discusses the results for the lower excited states of F2.S2A number of avoided curve-crossings between valence, Rydberg, and ionic states of the same symmetry were found. The B + X transition electric dipole moment of molecular bromine, 81Br2,has been calculated from measurements of absorption to selected vibrationrotation levels in the sllOu+ excited The value obtained, I & 1, = 0.12D2, gives a value of 20 ps for the radiative lifetime. Assignments of a number of transitions in the B -+ X systems of la71,and lznI2 have been made following the calculation of rotational distortion constants for the B3110U+and X1&+ states of lZ7I2from the Rydberg-Klein-Rees potential^.^^

-

'

I

' \ ' I

z20bq;:~mt&

-

r

-f

L

'

'

'H,'

r ( H - H l 5 2 . 2 0 bohr

-1.6 -

1

I

2D

1

I

30

I

I

I

4.0 5.0 R (bohr)

I

I ¶I I 60 ''00

Figure 1 CI results for the rectangular approach of two H, molecules ( R = 2.20 bohr) to the optimum square H4 (Reproduced by permission from Chem. Phys. Letters, 1976, 43, 73)

Ab initio calculations of the excited states and potential energy surfaces of the Li,H molecule have been carried out and this will help in the interpretation of the results of molecular-beam experiments on the reaction.sa H

+ Li,

Li,H

LiH

+ Li

All the low-lying states of Li,H were found to involve substantial charge transfer. The Brandow formulation of degenerate MBPT mentioned above in connection with H2 has been used to calculate the ground and seven lowest excited states of boron monohydride, BH, and the accuracy achieved suggests it could be applied to larger molecules.sa Ab initio calculations of the ground and excited states of NeH+, NeH, and linear asymmetric NeH,+ indicate that there is no pseudocrossing or crossing of the investigated potential energy curves for the excited states of NeH+, effects which are important for the corresponding argon rn01ecules.~~ Spectral assignments of transitions in the simple alcohols and ethers have resulted from improved virtual orbital (IVO) calculations on water, methanol, and dimethyl ether.@ The lowest-energy transitions in each of the molecules correspond to excitations out of an oxygen lone-pair orbital to an essentially 63 64

eK 66

J. P. Hay and D. C. Cartwright, Chern. Phys. Letters, 1976, 41, 80. F. Zaraga, N. S. Nogar, and C. B. Moore, J. Mol. Spectroscopy, 1976, 63, 564. K. K. Yee, J.C.S. Faraday II, 1976,72,2113. W. B. England and N. H. Sabelli, J. Chem. Phys., 1975,63,4596. P. S. Stem and U. Kaldor, J. Chem. Phys., 1976, & 2002. I, K. Vasudevan, Mol. Phys., 1975, 30,437. W. R. Wadt and W. A. Goddard, tert. Chem. Phys., 1976,18, 1.

10

Photochemistry

Rydberg-type orbital. The vibrational spectrum of the electronically excited 8)of HaOhas been analysed using a Green’s function method and state the ground (2A”) and lowest excited (2A’) states of the hydroperoxyl radical, Hoe, have been calculated in an all-valence-electron CI The results agree well with experimental measurements and a radiative lifetime for the 2A’ excited state of lo-, s is predicted. Extensive ab initio CI calculations of vertical excitation energies to a number of excited states of the C,, hydronium ion, HsO+, have been r e p ~ r t e d .In ~ ~order of increasing energy the four lowest vertically excited states of H,O+ are predicted to be ,A1, lA1, ,E, and lE. The recent extensive CI calculations on water have proved to be valuable in understanding the spectroscopy of alcohols and ethers, and similar work on the sulphur analogue H2S would naturally be of interest in future studies of mercaptans and dialkyl sulphides. Such calculations have recently been reported and show the dominance of Rydberg transitions for the low-lying excited states of H2S.72973 The first absorption band is assigned to a strong Rydberg 2b1 -+4s transition overlying a much more valence-like 2bl -+ 3dxz (xz in-plane) system, in support of earlier suggestions concerning this broad 40 000-60 000 cm-l 73 A comparison of the ab initio and experimental vertical excitation energies of H2S is shown in Table 1.

(c+-

Table 1 Comparison of ab initio and experimental vertical excitation energies of H,S Orbital excitation 26, + 4s

2bl

+ 3b2

Upper states 1’ B,

26, + 4p 2b1 -f 3d

1’A2 2IA,, 2!!& 2 3 3’&, 11B2, 3’A,, 3IA2

5a, -+ 4s

4 x 1

9

a

a

Vertical excitation energyleV Calculated Expt. 6.2 1 32)5.0 -f 7.4 6.6 1 7.64, 7.79, 7.97 7.85, 8.02, 8.18 8.52, 8.56, -- 8.59, - 8.66, 8.81

-

8.13, 8.94 9.33

(8.93)

Allowed transitions are underlined. From the DECI calculations using the ground-state orbitals. The value in parentheses is assumed, but has not been observed.

Combination of lifetime measurements for a number of levels (v’,J’) in the

B1nustate of Na2 with Franck-Condon factors and r centroids calculated from

spectroscopic data has shown that the squared transition moment, I R(r) 12, increases by 30% as the internuclear separation is increased from 0.25 nm to 0.611m.’~ Such data will provide a critical test of the wavefunctions used to describe the states involved in the transition. Model potential calculations of interaction potentials in a number of electronic states of the Li2+molecule show 68 70

71 72

’*

S. Trovato, A. Millefiori, and A. Raudino, J. Mol. Structure, 1976, 33, 133. R. J. Buenker and S. D. Peyerimhoff, Chem. Phys. Letters, 1976,37, 208. R. C . Raffenetti, H. J. T. Preston, and J. J. Kaufman, Chem. Phys. Letters, 1977, 46, 513. M. F. Guest and W. R. Rodwell, Mol. Phys., 1976, 32, 1075. S. Shih, S. D. Peyerimhoff, and L. J. Buenker, Chem. Phys., 1976, 17, 391. W. Demtroeder, W. Stetzenbach, M. Stock, and J. Witt, J. Mol. Spectroscopy, 1976, 61, 382.

Spectroscopic and Theoretical Aspects

11

that a high degree of accuracy can be achieved for intermediate and large separations as is shown by comparisons with the results of extensive ab initio calculation^.^^ The results, together with those for the Na2+molecule, have been used to calculate electronic transition moments and oscillator strengths for transitions from the ground 2&,+(ns) state to the excited states 2&+(ns), 211u, and 2&,+(np), in the two molecule^.^^ The potential energy curves for several excited electronic states of Liz+ have been calculated using a frozen-core method and six of them are found to be binding, viz. four 2Cg+states and two 2nu Ab initio equations-of-motion methods have been used to obtain vertical excitation energies for valence states arising from the 1'IIuS2IIu1 multiplet in the C 0 2 molecule 78 and ab initio calculations, unlike previous semi-empirical treatments, predict the ground electronic state of the C0,- radical anion to be a 2B1 state and to have a CZz, symmetry structure, in agreement with the interpretations of i.r. and e.s.r. measurement^.^^ The singlet-triplet absorption spectra of the NO2- and NOs- ions were calculated using the CNDO/CI method and found to agree well with experimental measurements, particularly for the NO,- ion.*O Very little is known about the excited electronic states of the diazenyl radical, HN,, unlike the situation for its widely studied isoelectronic partner, the formyl radical, HCO. Ab initio calculations on HNa show that it is only weakly bound and has an equilibrium bond angle of 124°.81 The vertical spectrum is predicted to show an isolated low-energy 211+ 2A' transition at about 2.0 eV and a number of other valence transitions in the region 5.5-8.0 eV. No evidence for a low-lying triplet state was found in an ab initio generalized valence bond (GVB) CI study of the vertical excited states of ammonia, NH,. The presence of such a state was suggested by threshold electron-impact experiments. The eight lowest excited states have essentially Rydberg-type characteristics; in particular the singlet-triplet energy splittings have the small values expected for such states.82 A theoretical study of the electronic excited states in HCN has unexpectedly shown that several of the states in HNC have lower energy than their equivalent states in HCN.83 The potential energy surfaces of the lowest Al symmetry excited state of H,S84 and of the three internal co-ordinates of the ground, 2A", and first excited, 2A', states of the HSO radical 85 have been investigated using ub initio methods. A review of ab initio calculations of potential energy surfaces for ground and excited states has been published.86 The excited electronic states of SF6 have been investigated by an a6 initio improved virtual 76 'O

77 78

7e

82

8a

Bs

86

C. Bottcher and A. Delgano, Chem. Phys. Letters, 1975, 36, 137. K. Kirby-Docken, C. J. Cerjaxi, and A. Dalgarno, Chem. Phys. Letters, 1976, 40, 205. W. Mueller and M. Jungen, Chem. Phys. Letters, 1976, 40. 199. W. England, D. Yeager, and A. C. Wahl, J. Chern. Phys., 1977,66,2344. S. P. So,J.C.S. Faraday II, 1976.72, 646. B. F. Minaev, Optika i Spektroskopiya, 1976, 41, 752. K. Vasudevan, S. D. Peyerimhoff, and R. J. Buenker, J. Mol. Structure, 1975, 29, 285. R. Rianda, R. P. Frueholz, and W. A. Goddard, tert., Chem. Phys., 1977, 19, 131. G. M. Schwenzer, H. F. Schaefer, tert., and C. F. Bender, J. Chern. Phys., 1975, 63, 569. F. Flouquet, Chem. Phys., 1976, 13, 257. A. B. Sannigrahi, K. H. Thunemann, S. D. Peyerimhoff, and R. J. Buenker, Chem. Phys., 1977, 20, 25. S. D. Peyerimhoff and R. J. Buencker, in 'The New World of Quantum Chemistry', ed. B. Pullman and R. Parr, Reidel, Dordrecht, Netherlands, 1976, p. 213. 2

12

Photochemistry

orbital (IVO) method using an extended Gaussian basis with which it is possible to describe excited states which involve diffuse orbitals.87 In the IVO method the energy of a state arising from an excitation from a $ k orbital to a #z orbital is obtained by constructing the Hamiltonian for orbital #z with the appropriate ion core obtained by removing orbital $k. The error in the resulting excitation energies is expected to be similar to the error in the ionization potential calculated using the same ion core and is therefore corrected for by the difference between the Hartree-Fock and experimental ionization potentials. The results show an ordering of the excited-state orbitals with increasing energy as follows : 6alU(u*),7a1,(4s), 6tlu(4p), 2t2,(3d), 4eU(3d),8a1,(5s), and 7t1,(5p). (The bracketed symbols are the Rydberg designations, i.e. the lowest orbital is a valence antibonding orbital and the rest are diffuse Rydberg orbitals.) The lowest-energy transition in SF, is therefore the dipole-forbidden state arising from the Itl, + 6ar, excitation and is predicted to occur just below 10 eV. In recent years interest in the ultraviolet spectroscopy of rare-gas diatomic molecules has greatly increased because of their use in laser systems. The broad excimer emissions resulting from transitions between bound singlet excited states and repulsive ground-state potentials are a source of tunable ultraviolet laser radiation.88 The lowest excimer state of Ar2*, the 3&,+ state, shows up clearly in the results of scattering experiments of metastable Ar* atoms by ground-state Ar atoms. The results also indicated the presence of another state, probably the 3C0+ state, which has a small barrier and/or a shallow well. Ab initio calculations have supported these assignments: the features of the scattering experiment, notably those which involve the 3Cu+state, were almost exactly reproduced by calculations on the theoretical potential curves.s9 Spectroscopic data for the electronic levels of the homonuclear diatomic molecules of Group V elements are discussed in a study of the electronic states of Bi2.00 A new ground state, XI, is proposed for Biz, with symmetry O,+,to explain a number of experimental observations. The effect of the level of CI on the calculated ordering of excited states, excited-state geometries, and bond energies has been examined in calculations on the low-lying excited states of methylene fluoride, CH2F2, using the INDO rnethod.Ol In the case of vertical excitation energies and bond energies, N-CI calculations, which include all configurations of a given symmetry with coefficient 20.1 in the first three states of that symmetry, are found to be adequate. This is illustrated for vertical excited state ordering in Figure 2. In an interesting study the effect on their vertical electronic spectra of substituting a F atom for a H atom in the molecules trans-di-imide (N2H2)and diazirine (N,CH,) to give N2F2and N2CF2, respectively, is calculated using ab initio SCF-CI methods. The results are summarized in Figure 3 and show that the ‘perfluoro effect’ causes a reordering of states for trans-di-imide with the It3(77 -+T*) transitions

89

s1

P. J. Hay, J. Amer. Chem. SOC.,1977, 99, 1013. For a review see C. W. Werner and E. V. George in ‘Principles of Laser Plasmas’, ed. G. Bekefi, Wiley, Chichester, 1976, p. 421. R. P. Saxon and B. Liu, J. Chem. Phys., 1976,64, 3291. G. Gerber and H. P. Broida, J. Chem. Phys., 1976, 64, 3423. M. S. Gordon, Chem. Phys. Letters, 1976,37, 593.

Spectroscopic and Theoretical Aspects

13

red-shifted and the remainder blue-shifted and for diazirine the relative ordering remains the same: the l(w + m * ) transition is blue-shifted and the remaining ones red-shifted.a2 Ab initio calculations on formamide using the restricted Roothan 15

-

14

-

13

-

W aJ

3A 20;

-lB

2

12

U

w

a

10 L

vo 341 N-CI s-CI Figure 2 Vertical states of CH,F2 (Reproduced by permission from Chem. Phys. Letters, 1976,37,593) 12.0

1

W

I

21)

Figure 3 Comparison of calculated energy levels for electronically excited states [3J(n+,7r*)]and PJ(7r,m*)] in trans-di-imide, trans-dipuoride, diazirine, and diflaorodiazirine (Reproduced by permission from Chem. Phys., 1976, 15, 103) Oa K. Vasudevan and W.E. Kammer, Chem. Phys., 1976,15, 103.

14

Photochemistry

open-shell method have found a significant dependence upon the molecular geometry of the energy difference between the lowest two triplet states and their conformational preference^.^^ In calculations which again illustrate the power of modern ab initiu CI methods to predict with considerable accuracy complex molecular spectra, the spectrum of acetone has been studied and compared with the results of a previous similar study of formaldehyde. The lowest transitions are assigned to ls3(n,37*) (the weak dipole-forbidden J+-F A 1 band) and a single 3(7r,7r*)-type transition. The remaining classified bands +- F A 1 ) are all Rydberg-type transition^.^^ MO calculations on nitromethane and some of its derivatives show a shift of electronic charge away from the N atom and towards the central C atom on excitation from the ground state to the lowest triplet The singlet and triplet energy levels of the dicarbonyl compounds, trans-glyoxal and trans-biacetyl, molecules of some importance in recent single vibronic level (SVL) studies, have been investigated by SCF-CNDO/S calculations.~s The open-shell SCF calculations which use a second-order perturbation treatment of the electronic correlation suitable for the higher triplet states locate the lowest 3Bg state relative to the lowest excited state, the 3A, state, in agreement with recent measurements on biacetyl. Ab initiu SCF calculations using a double-5 basis set of contracted Gaussian functions for both the cis- and trans-forms of glyoxal have also appeared.07 An extremely large singlet-triplet splitting for the 1-B, state in trans-glyoxal is found which places state between the 3A, and 3B, states. As the authors point out, this large the splitting may be an artifact of the basis set when applied to 7~-7r* excitations such as B,. The results are summarized in Figure 4. Vertical excitation energies to the valence and Rydberg excited states of formic acid below 11.O eV have been investigated by ab initiu frozen-core single-excitation configuration mixing (CM) calculation^.^^ The method is apparently quite accurate for Rydberg states as correlation and reorganization energies tend to cancel with those of the ion, but it may underestimate valence-state excitation energies. A number of transitions to valence T* MO's are found amongst the predominantly Rydbergtype transitions, no-ns, np, and nd, found in this molecule. The excited states of diformamide, (HCO),NH, and N-methyldiformamide, (HCO),NMe, have been examined using CNDO-CI calculation^.^^ The results of ab initio STO 4G calculations on the excited states of methane are qualitatively similar to those from earlier INDO studies: the tendency of electronically excited methane to distort from tetrahedral symmetry is again clearly demonstrated.loOThe lowest adiabatic state is 'B, in C,,symmetry which is some lOeV lower than lE(C3,,). The lBl(C,,) state is found to be unstable to dissociation to lBl methylene and lAl H2whereas the INDO results also predicted it to be the lowest adiabatic state, but to be stable. The vertical excited states of

(&c,fi

Ds

C. N. Baird and H. B. Kathpal, Chem. Phys. Letters, 1976, 43, 315.

94

B. Hess, P. J. Bruna, R. J. Buenker, and S. D. Peyerimhoff, Chem. Phys.,

96

1976, 18, 267.

V. I. Slovetskii, T. A. Chenchik, and I. A. Abronin, Izvest. Akad. Nauk S.S.S.R., Ser. khim., 1977, 289.

J. M. Leclercq, C. Mijoule, and P. Yvan, J. Chem. Phys., 1976, 64, 1464. O7 C. E. Dykstra and H. F. Schaefer, tert., J. Amer. Chem. Suc., 1976, 98, 401. D. Demoulin, Chem. Phys., 1976, 17, 471. 9B J. Maranon and 0. M. Sorarrain, 2. Nuturforsch., 1977, 32a, 103. loo M. S. Gordon, Chem. Phys. Letters, 1976, 44, 507. IM

15

Spectroscopic and Theoretical Aspects

propane have been examined in an INDO study and predictions concerning the photodissociation of alkanes are made on the assumption that excited valence states are the precursors to such processes.1o1 The orbitally forbidden 2E'+- 2Ae" transition in the methyl radical which has not, as yet, been observed experimentally is predicted to occur in the ultraviolet by an ab initio SCF study which assumes the ground state to be planar.lo2 Cis -

Tfons

F -226.0

I 3

- 226.1 h -,226.2

P

------\,

&:::--

k-

-

b

------

0

C

w

I-A?

,226.3

-

226.4

-

'

Icm-'

I

I-Bz

_____-Figure 4 The excited states of glyoxal. Energies correspond to vertical excitation from the cis or trans ground state. Triplet states are shown with solid lines and singlet states with broken lines. Symmetry state designations are given for each singlet-triplet pair of states. Correlation lines between cis and trans states connect states which must correlate on the basis of symmetry, but do not represent any internal rotation potentials (Reproduced by permission from J. Amer. Chem. SOC.,1976,98, 401)

The t rimethylenemethane biradical cont hues to attract considerable experimental and theoretical work. An ab initio GVB-CI calculation o f the ground and excited states of the molecule in both planar and twisted geometries has again shown that the ground state of the planar form is a triplet, 3Aa' in Dghsymmetry, and additionally that the lowest singlet state of the planar form, lE', will twist to a more stable bisected form.lo3 A semi-empirical CI study of the wavefunctions for the electrons in the lE' state shows again the preference for an Iol Io2 loS

P. M. Saatzer, R. D. Koob, and M. S. Gordon, J. Amer. Chem. Soc., 1975,97,5054. Y. Ishikawa and P. C. Binning, jun., Chem. Phys. Letters, 1976,40,342. J. H.Davis and W. A. Goddard, tert., J. Amer. Chem. SOC.,1976,98,303.

16

Photochemistry

orthogonal geometry in singlet trimethylenemethane.lo4 Good experimental evidence that the ground state is indeed a triplet has been obtained in an e.s.r. study over a range of temperature^.^^^ Semi-empirical calculations and qualitative MO arguments have been used to discuss the effect of substituents on the ground states of substituted trimethylenemethanes.lo6 A new ab initio study of the excited states of ketene CHzCO has been p~b1ished.l~'Three basis sets of contracted Gaussian functions were used to calculate the vertical excitation energies of 18 excited electronic states. The results show that for the five lowest states (FA1, 3A2,lAz, 3A1,2'A1), reliable CI studies would result from using the manageable double-t: basis and need not involve polarization and Rydberg functions. Although an experimental result, it is appropriate to mention here that the methylene CH,(3B3 ++ CH2(lA1)energy splitting, a valuable datum in theoretical calculations on this species, has been found from ketene photolysis experiments to be 8.3 k 1 kcal mol-1.108 A large-scale configuration-interaction (CI) calculation of the triplet-singlet splitting in CH2 which uses a double-zeta plus polarization basis for CH2 gives a value of 15.4 kcal/m01e.~~~ Recent experimental and theoretical work on the excited states of ethylene has been reviewed resulting in a table showing the assignments of virtually all the lower-energy states.l1° Ab initio frozen-core calculations of a number of excited-state geometries in ethylene show that Rydberg states are linear while valence ...(l 7 ~ ~ ) ~ ( l n ,states *) may be cis and trans bent or assume a gauche, non-planar, geometry if the in-plane or out-of-plane component, respectively, of the 1n,* MO is excited.lll An important ab initio study of the n-electron states of trans-l,3-butadiene has been pub1ished.ll2 Five of the excited states, 13&, 13A,, 2lA,, 33Bu,and 16A,, had a valence-like charge distribution and could be correlated with the valence N and T states of ethylene. They may be regarded in terms of a 'two-vinyl' model as either NT (the first two) or TT (doubly excited). All of the low-lying singlet excited states, except for 2'4, involved diffuse excited orbitals. Ab initio calculations of the low-lying nn* excited states of trans-l,3,5-hexatriene have led to the suggestion that the singlet-state ordering is 2lA, < llBu, opposite to that calculated for butadiene but the same as that found in the longer polyenes where a weak electric dipole-forbidden transition occurs at lower energies than the allowed t r a n ~ i t i 0 n . l An ~ ~ interesting study of the n-* transition in polyenic aldehydes compares the results of calculations which treat the excitation in terms of delocalized molecular orbitals, as is generally done, or in terms of localized molecular orbitals, i.e. the excitonic model of excited The n-v* excitation is shown to be localized in the W. T. Borden, J. Amer. Chem. SOC.,1976,98, 2695. R. J. Baseman, D. W. Pratt, M. Chow, and P. Dowd, J. Amer. Chem. SOC.,1976, 98, 5726. lo6 B. K. Carpenter, R. D. Little, and J. A. Berson, J. Amer. Chem. SOC.,1976, 98, 5723. l o 7 C. E. Dykstra and H. F. Schaefer, J. Amer. Chem. Soc., 1976,98,2689. lo* J. W. Simons and R. Curry, Chem. Phys. Letters, 1976, 38, 171. lo* A. H. Pakiari and N. C. Handy, Theor. Chim. Acta, 1975,40, 17. R. S. Mulliken, J. Chem. Phys., 1977, 66, 2448. ll1D. Demoulin, Chem. Phys., 1975, 11, 329. lla R. P. Hosteny, T. H. Dunning, jun., R. R. Gilman, and A. Pipano, J. Chem. Phys., 1975,62,

lo4

lo6

4764. 113 114

D. A. Luippold, Chem. Phys. Letters, 1976,43, 55. J. Langlet, Theor. Chim. Acta, 1975, 38, 199.

Spectroscopic and Theoretical Aspects

17

mcpo region thus explaining the reason for the constancy of the n-n* excitaton

energy in polyenic aldehydes in all but the few with a low number of double bonds. The successful application of the semi-empirical PPP model in the interpretation of the MCD signs and polarizations of nn* transitions in conjugated hydrocarbons has been discussed.116 The T,-So oscillator strengths and radiative lifetimes of the triplet nw* and vn* states of benzaldehyde have been calculated using semi-empirical methods in a study which discusses the strong correlation which exists between the nn* and m* triplet energy splitting, the lifetimes, and ZFS parameters in aromatic carbonyl compounds.ll6 An interesting paper has examined the problem posed by the observation of dual phosphorescence in lowtemperature glassy solutions of phenyl alkyl ketones.l17 CNDO calculations strongly support the idea that the dual emission arises because the most stable conformations in the emitting n -+ w* triplet state and the ground state are different (trans and gauche forms respectively). A modified INDO technique has been very successful in calculating the excited triplet-state energies of benzene, pyridine, and the diazines.ll* The major problem associated with the application of this technique to triplet states after its earlier success in the calculation of singlet excited states concerns the choice of excited configurations for the configuration interaction. In general, it is found to be more sensitive to the inclusion of higher lying configurations. The four lowest triplet states of benzene are calculated in good agreement with experiment. Pyridine is found to have a 3Bl(n-7r*) triplet nearly degenerate with the lowest-lying 3A1(n--n-*)triplet which agrees with recent experimental work. Ab initb calculations of the oscillator strengths for the allowed transitions and Franck-Condon excitation energies of transitions from the ground state to the ly3A1(4bl -+ 5b1), lg3A1(3a, + 4a2), 13Ba(3a2-+ 5b1), and 1,3B,(4b1-+ 4a,) excited electronic states of carbazole show good agreement with the ordering of states found experimentally and although all too large the calculated transition energies have an excellent linear relationship with experimental values.ll@ The transition energies from the ground to low-lying singlet and triplet states in a number of pyrylium derivatives have been calculated by SCF-PPP-CI methods lZo and the electronic structures of singlet excited states of benzoquinolines calculated by a semi-empirical SCF-T-MO-CI method have been correlated with experimental dipole-moment and acidityconstant measurements.121 The electronically excited states of the DNA bases adenine, thymine, guanine, and cytosine have been studied using the self-consistent renormalized random phase approximation (RPA) 122 and the radiative phosphorescent lifetime of caffeine has been calculated using the CNDO/S method 123 following its successful nS lla 117 118

J. Michl, Znternat. J. Quantum Chem., Symposia, 1976, 10, 107. C. Mijoule and P. Yvan, Chem. Phys. Letters, 1976, 43, 524. J. Langlet and P. Gacoin, Theor. Chim. Actu, 1976,42,293. J. E. Ridley and M. C. Zerner, Theor. Chim. Acta, 1976, 42, 223. L. E. Nitzsche, C. Chabalowski, and R. E. Christofferson, J. Amer. Chem SOC., 1976, 98, 4794.

lZo lZ1 laa 12a

M. D. Gheorghiu and A. T. Balaban, Rev. Roumaine Chim., 1976,21, 1513.

H.Yamaguchi, T. Ikeda, and H. Mametsuka, Bull. Chem. SOC.Japan.

H. Ito and Y. J. Ihaya, Bull. Chem. SOC.Japan, 1976,49, 3466. G. Lancelot. Mol. Phys., 1975, 29, 1099.

1975,48, 1118.

18

Photochemistry

application in the cases of azulene and formaldehyde. Studies of the excited states of the phenoxyl radical lZ4and the ethidium cation 126 have been published. The use of the RPA in the calculation of the triplet spectra for a number of conjugated molecules has been critically examined and it is shown how the problem of triplet instability, i.e. imaginary triplet levels, can be overcome.lZ6An ab initio SCF-CI calculation of the excited states of s-tetrazine has been published and clearly establishes that the lowest excited state in this molecule is a triplet state and not a singlet state as had been predicted by a localized exciton model. This surprising prediction offered an explanation for the very strong fluorescence properties of the molecule, but the present calculations show that this probably occurs because there is no TT* triplet state between the two lowest excited states, 1B3u(nn*)and 3 B 3 u ( n ~ * )The . phosphorescence has recently been observed for the first time and was found to have a very short lifetime. The energy levels and their assignments for this interesting molecule are given in Table 2.12'

Table 2 A summary of calculated transition energies of s-tetrazine and comparison with available experimental values

Electronic state and orbital promotion 'A,, (ground state) 3B3u(n -+ n*) 1B3u(n --f n*) 3B2u(n -+ n-*) 3B1u( n + n*) 3A, (n + n*) l A , (n + n*) lB3, (n -+ n*)

Transition energies calculated (eV)

lBzu(n-+ n*) 3Beu(n+ n*) lBl0 (n -+ n*) l A , ( n -+ lBzu (.rr + n 1

n*i

lB2,, ( n + n*)

3B1u(n + n*) 3B1, (n T*) 3Au (n + n*) sBsu (n + n*) -+

lBSu(n + n*) 3Bzg(n + n*) 1B3g(n n*> 1BIU(n + n*) --f

Experimentally observed transitions Band Range origin

Oscillator strength (a.u.) Calc. Expt.

2.04 2.80 4.05 4.44 4.66 4.18 5.10 5.99 6.05 6.24 6.46 6.5 1 6.59 6.60 1.76 1.86 8.14 8.42 8.71 10.39 10.45

1.66-2.07 2.22-2.10

1.69 2.25

0.011

5.01

0.023 0.053

3.88 (?) 4.43-5.40

0.004

0.001

0.098

0.017 0.296

The unusual fluorescence properties of 1,l'-binaphthyl have been investigated calculation and extended CI which determines the energies of a number of the lowest-lying singlet states as a function of the dihedral by an SCF-LCAO-MO lZ4 l*6 lZ6

lZ7

V. A. Kuzmin and I. V. Khudyakov, Doklady Akad. Nauk S.S.S.R., 1976,227, 1394. A. Waleh, B. Hudson, and G. Loew, Biopolymers, 1976, 15, 1637. 0. Matsuoka and H. Ito, Theor. Chim. Acta, 1975, 39, 111. T. Ha, Mol. Phys., 1975, 29, 1829.

Spectroscopic and Theoretical Aspects

19

angle.lZ8 The state responsible for the strongly red-shifted fluorescence in fluid solutions achieves stabilization in the more coplanar geometry that it has compared with the ground state by having a large amount of charge resonance character. A closely related study of the excimer of l-methylnaphthalene shows again the importance of charge resonance states in systems where there are two interacting identical moieties: a description in terms of a simple exciton model is clearly 3 Absorption Spectroscopy Thermal Lensing and Photoacoustic Spectroscopy.-The application of the thermal lens technique to a number of problems in absorption spectroscopy is an interesting and significant development.130 The technique is closely related to the optoacoustic technique for measuring luminescence quantum yields reported on in Section 4 of this Chapter. In the technique the non-radiative decay of energy absorbed from a laser beam (TEMoo mode) by a liquid results in a transverse temperature profile being formed which, because most liquids have a negative temperature coefficient of the refractive index, is effectively a divergent lens. The focal length of the thermal lens will depend on the absorbed power, and a small pinhole-sized detector area at the centre of the beam beyond the sample is able to monitor sensitively changes in the absorbed power. Measurements of extinction coefficientsas low as dm3mo1-1 cm-l (or equivalently low optical densities) will be possible using the technique.130 In addition to measurements of vibrational overtone bands and spin-forbidden electronic transitions (see below) the technique can readily be applied to the detection of photoproduced transient species and multiphoton absorption processes. The first reported measurementsusing a newly developed dual-beam synchronous thermal lensing instrument which uses two CW lasers, one of which is tunable, to probe for absorptions and the second, of fixed frequency, to monitor the lensing, are in a paper showing the singlet-triplet absorption spectrum of anthracene (overlapping in part with a vibrational overtone band) in ethyl iodide in the range 15 800-17 400 cm-l and the strong absorption band shown in Figure 5 , in liquid benzene in the same spectral region which corresponds to the fifth overtone of the C-H stretching vibrafion.ls0 The value of the technique is nicely illustrated in a subsequent paper from the same authors who report the visible region thermal lensing absorption spectra of [lH,]- and [2H,]-benzene, methyl-substituted benzenes, naphthalene, and anthracene.lsl The absorption arising from highly excited C-H stretching vibrations is clearly seen in eachcase and the excited vibrational state involved is found to be the most anharmonic of the many states at the same quantum level, i.e. the state which is furthest removed from the strict selection rules of the harmonic oscillator. Such studies are of particular relevance to theories of radiationless transitions in, for example, aromatic hydrocarbons where the states reached here by absorption are the very states implicated as

M. F. M. Post, J. W. Eweg, J. Langelaar, J. D. W. van Voorst, and G. Ter Maten, Chem. Phys., 1976, 14, 165. lZB M.F. M. Post, J. Langelaar, and J. D. W. van Voorst, Chem. Phys., 1976, 15,445. l a o M. E. Long, R. L. Swafford, and A. C. Albrecht, Science, 1976, 191, 183. 131 R. L. Swaffrod, M. E. Long, and A. C. Albrecht, J. Chern. Phys., 1976, 65, 179. 12*

20 Photochemistry accepting modes in the radiationless process, and they are also relevant to theories of multiphoton excitation of molecules by infrared lasers reported on elsewhere. They are also relevant to theories of intramolecular vibrational redistribution, since the absorption profiles are narrow, implying that the levels reached are long-lived. This would be unexpected if a fast redistribution of energy occurred.

f

- 2.0

\

I\ I \

F

* -

‘0

-

n

’E

- 1.5 ‘a V

E

T

\

X U

U

4

- 1.0 y

E!

T

i

I7400

I7000

I6600

16200

X Y

w

- 0.5

I5800

E (ern-')

Figure 5 Absorption of neat benzene obtained with the single-beam apparatus, ( 0 )[lHS]benzene and (0) [2H,]benzene (Reproduced by permission from J. Chem. Phys., 1976, 65, 179)

The technique has been developed further by introducing a pulsed probing laser to increase greatly the laser power and, therefore, the probability of observing multiphoton a b ~ o r p t i o n s133 . ~ ~The ~ ~ time development of the focal length of the thermal lens and the resultant intensity changes at a detector in the centre of the far-field beam were analysed theoreti~a1ly.l~~ The analysis shows that observation of two-photon absorption using the technique is perfectly feasible and indeed, with reasonable values of the various parameters involved, its effects may be more pronounced than for one-photon absorptions. The following paper describes the application of the method to the measurement of the two-photon absorption spectrum of liquid benzene, chosen because the fluorescence-detected two-photon spectrum of benzene vapour is known already, making direct comparisons p0ssib1e.l~~The spectrum was measured over the range 360-540 nm and showed three principal features. The lowest-energy of these corresponds to the vibronically allowed lBZu-+ lAl, transition in the region 465-540 nm, the next, a weak band at 425 nm, corresponds to the vibronically lga

133

A. J. Twarowski and D. S. Kliger, Chem. Phys., 1977, 20,253. A. J. Twarowski and D. S. Kliger, Chem. Phys., 1977, 20, 259

Spectroscopic and Theoretical Aspects

21

induced lBlu -+ lAls transition, and the highest is a very strong band peaking at wavelengths below 360 nm which is assigned to the lEeU+- lA1, transition. Interestingly the strong bandat 391 nm seen in the vapour phase work is not observed and this supports its assignment as a transition to a Rydberg state which is markedly broadened and weakened in condensed phases. The potential of time-resolved photoacoustic spectroscopy, particularly in studies of opaque samples, has been demonstrated in a recent investigation of

-

Figure 6 The 4; vibronically induced band of the x1A2 - XIAl system of H&S. The peak intensity of this band is 1.5 times that of the 0; band (Reproduced by permission from Chem. Phys., 1977, 22, 199)

Rose Bengal dye.lg4 The sample was excited simultaneously by two pulsed dye lasers operating at 474 nm and with pulse widths of 6 ns. It was observed that a 30% larger photoacoustic signal was produced when the two beams overlapped on the sample than when they did not. A previous picosecond flash photolysis investigation of the dye had shown that the S, t S1absorption of 474 nm is stronger than the S1+- Soabsorption at the same wavelength. This observation, and the fact that S1fluoresces strongly whereas S, does not, together provide a convincing explanation of the signal enhancement. Time-resolved nanosecond photoacoustic spectroscopy would be possible on introducing a variable time delay between the two pulses. Optoacoustic measurements of forbidden transitions in thioformaldehyde, ~ ~ vibronically H2CS, and disulphur monoxide, S 2 0 , have been ~ e p 0 r t e d . l The induced 4: band of the L1A2-z1A1 system in H2CS obtained in this work is shown in Figure 6. Calibration of the apparatus, which used chopped radiation from a tunable CW dye laser, with measurements on a known transitionin lS4 la6

M. G. Rockley and J. P. Devlin, Appl. Phys. LPtters, 1977, 31, 24. R. N. Dixon, D. A. Haner, and C. R. Webster, Chem. Phys., 1977, 22, 199.

22

Photochemistry

thiophosgene, ClzCS, indicated that oscillator strengths as low as f z should be detectable in a multi-pass cell operated at atmospheric pressure. Optoacoustic measurements of visible vibrational overtone bands in CHo and NH3 lS6and infrared transitions of H 2 0 have also been Single-photon Absorption Spectra.-A classification scheme for the excited states of a family of non-alternant hydrocarbons has been proposed.lS8 Classification schemes for the excited states of benzenoid hydrocarbons are extensively used, particularly those of Clar and Platt, but the wide variety of types of non-alternant hydrocarbons has limited the development of such useful generalizations for this group of molecules. A study of the low-lying electronic excited states of molecules derived from the two parent pericondensed tricyclics acenaphthylene (1) and pleiadiene (2) showed that they could be classified together and related in some respects to states of naphthalene. Three low-lying transitions labelled K , L, and M were characterized experimentally in a total of

eleven hydrocarbons and their energies were well reproduced by semi-empirical PPP calculations and also with the qualitative predictions of a simple MO theory, i.e. PMO. The agreement between the experimental and calculated (PPP) K , L, and M transition energies is illustrated in Figure 7. A detailed study of the low-lying excited states of a number of bridged [14]annulenes with anthracene perimeter has been p~b1ished.l~~ Absorption, polarized emission, linear dichroism, and magnetic circular dichroism measurements are used to assign seven electronic transitions below 35 000 cm-l. Four of these transitions can be correlated with those of anthracene in agreement with a theoretical model of Platt and Moffitt, and the other three appear to arise from a hyperconjugative effect of the saturated bridge, i.e. 'u-n mixing'. A many-body theoretical approach to the calculation of vibrational structure in molecular electronic spectra has been deve10ped.l~~The theory provides a new approach to the calculation of Franck-Condon factors for polyatomic molecules and is applied with some success to the optical absorption and electron impact spectra of N, and CO and to the photoelectron spectra of H,CO and DZCO.l4l A study of the fluorescence excitation and excitation polarization spectra of indole and tryptophan in propylene glycol results in a clear resolution of their excitation spectra into 'La and lLb bands.142 This is illustrated for the case of G. Stella, 5. Gelfand, and W. H. Smith, Chem. Phys. Letters, 1976, 39, 146. M. S. Shumate, R. T. Menzies, J. S. Margolis, and L. G. Rosengren, Appl. Optics, 1976, 15, 2480. lSB J. Michl and J. F. Miller, J. Amer. Chem. SOC.,1976, 98, 4550. laB J. Kolc, J. Michl, and E. Vogel, J. Amer. Chem. Soc., 1976, 98, 3935. 140 L. S . Cederbaum and W. Domcke, J. Chem. Phys., 1976, 64, 603. W. Domcke and L. S. Cederbaum, J . Chem. Phys., 1976,64,612. B. Valeur and G. Weber, Photochem. and Photobiol., 1977,25, 441. 13'

Spectroscopic and Theoretical Aspects

23

H n

(bl

Figure 7 (a) Experimental energies of transitions K, L, and M in hydrocarbons of acenaphthylene and pleiadiene families (0-0 band); (b) calculated (PPP) energies of transitions K, L, and M in hydrocarbons of acenapthylene and pleiadiene families (vertical transition) (Reproduced by permission from J. Amer. Chem. SOC,1976,98, 4550)

indole in Figure 8. The simplicity of the method employed in the paper results from the additive property of the emission anisotropy, r, a property not shared by the degree of polarization, p, which is still widely used in polarization studies. An interesting paper investigates the large variations in the absorption bandwidths of radical anions of the aromatic hydrocarbons by making model calculations of a one-electron bound-free transition in a cylindrical square-well

24

Photochemistry

p0tentia1.l~~The relative broadness of the benzene anion absorption band compared with that of the naphthalene anion may be attributed, at least in part, to the different angular momentum quantum number changes in the two cases as these result in different centrifugal barrier heights and therefore different lifetimes. The theory of light absorption by ions in solution has been examined 144 and the near-infrared absorption spectra of organic peroxy radicals 145 and the ultraviolet absorption spectrum of t-butyl radicals 146 have both been reported. The important theoretical problems posed by the broad absorption band of the solvated electron continue to attract attention. The effect of cavity distortions, higher excited states, and two-electron cavity species on the absorption band shape have been calculated, and in each case no clear relationship to the observed

Figure 8 Resolution of the excitation spectrum of indole (Reproduced by permission from Photochem. and Photobiol., 1977, 25, 441)

band is f 0 ~ n d . l ~A' particle in a spherical box model has been used to generate the solvated electron absorption spectrum148 and in an interesting paper it is shown how the application of the method of moments to the spectrum gives the characteristic parameters of any given model p0tentia1.l~~ Spectroscopic evidence which suggests the importance of both short- and long-range interaction terms in models for the solvated electron spectrum is discussed in a paper which demonstrates the clear correlations between halide ion and solvated electron absorption A valuable discussion of Rydberg states maxima in a wide range of in atomic and molecular spectroscopy has been p~b1ished.l~~ S,-T, Absorption Spectra. The lowest triplet state of acenapthenequinone crystals was characterized as 3 B l ( n ~ *in) a Zeeman study and the So-T, transition was found to borrow intensity from the z-polarized lAl(.zr~*)-lAl transition T. Watanabe, T. Shida, and S. Iwata, Chem. Phys., 1976, 13, 65. R. R. Dogonadze, E. M. Itskovitch, A. M. Kuznetsov, and M. A. Vorotyntsev, J. Phys. Chem., 1975,79,2827. 146 H. E. Hunziker and H. R. Wendt, J. Chem. Phys., 1976, 64, 3488. lP* D. A. Parkes and C. P. Quinn, Chem. Phys. Letters, 1975, 33, 483. I P 7 N. R. Kestner and J. Logan, J. Phys. Chem., 1975,79,2815. l P 8 V. Mazzacurati and G. Signorelli, Lettere Nuooo Cimento, 1975, 12, 347. l P 9 M. G. Debacker, 5. N. Decarpigny, and M. Lannoo, J. Phys. Chem., 1977, 81, 159. 160 M. F. Fox and E. Hayon, J.C.S. Faraday I , 1976, 72, 1990. lS1 R. S. Mulliken, Accounts Chem. Res., 1976, 9, 7. lP3 lP4

Spectroscopic and Theoretical Aspects 25 through the y-component in the spin-orbit i n t e r a ~ t i 0 n . l ~Oxygen-induced ~ singlet-triplet absorption spectra of a series of trans-acyclic azo-compounds showed that the lowest triplet state in these compounds is 55.5 k 1.5 kcal mol-1 above the ground The results compared well with values obtained from trapped electron and differential electron scattering spectroscopy and from triplet energy transfer studies. A detailed study has been made of the polarized phosphorescence excitation and emission spectra of the simple a,F-unsaturated ketone 9-methyl-5(10)octalin-l,6-dione and several related steroidal en one^.'^^ The results indicated that the lowest triplet state, 3(77,77*), of the simple ketone is planar, or nearly so, in direct contrast with earlier emission studies and calculations which had indicated a highly non-planar 3(7r,~*)state. The low-resolution results of the earlier emission studies are shown to be misleading and the stabilizing effect of rotation about the C=C bond, which is shown in the previous calculations and led to the prediction of a twisted excited state geometry, is presumably more than compensated by the stabilizing effect of increasing double bond character in the C-C bond. The results pose some interesting problems since it is known that these molecules photoisomerize via the triplet state. Three low-lying triplet states of ketene have been identified in the threshold electron impact excitation spectrum at 3.8 eV (3A2),5.0 eV (3Al), and 5.8 eV (3B,). A frozen-core ab initiu calculation was very successful in predicting both the singlet- and triplet-state energies observed in the spectrum.lK5 The triplet states of the amide functional group have been investigated using the trapped electron (TE) method.15s The TE spectra of formamide and N,N-dimethylformamide were obtained and for comparison those of formaldehyde, acetyldehyde, and acetone also. Optical absorption data and the results from ab initio calculations are also used in assigning the lowest singlettriplet transitions in formamide and its methyl derivative to bands at 5.30eV and 5.00 eV respectively. S1-S, Absorption Spectra. The excited singlet-singlet absorption spectrum of rhodamine-6G has been measured and placed on an absolute scale by completely bleaching the solution with a laser ~u1se.l~'Triplet formation is assumed to be negligible and thus the excited-state concentration during bleaching is that of the ground-state molecules prior to excitation. The S1-S, absorption spectrum of 3,3'-bipyrenyl has features in common with that of the excimer of pyrene and which are not found in the S,-S, absorption spectrum of pyrene i t ~ e 1 f . lThese ~~ observations are analogous to those seen previously in equivalent naphthalene systems. Tl-T, Absorption Spectra. The effect of pressures up to 25 kbar on the energies of three triplet states of anthracene (3L,, 3Ka, has been measured. Comparison with earlier measurements of the pressure dependence of the lowest singlet state, M. Sano, T. Narisawa, and Y . J. Ihaya, Bull. Chem. SOC.Japan, 1975,48, 3469. J. Metcalfe, S. Chervinsky, and I. Oref, Chem. Phys. Letters, 1976, 42, 190. 15' C. R. Jones and D. R. Kearns, J. Amer. Chem. SOC.,1977,99, 344. l K 5 J. Vogt, M. Jungen, and J. L. Beauchamp, Chem. Phys. Letters, 1976, 40, 500. x6 R. H. Staley, L. B. Harding, W. A. Goddard, tert., and J. L. Beauchamp, Chem. Phys. 16a

u3

15'

lS8

Letters, 1975, 36, 589. G. Dolan and C. R. Goldschmidt, Chem. Phys. Letters, 1976, 39, 320. M. F. M. Post, J. Langelaar, and J. D. W. van Voorst, Chem. Phys. Letters, 1976, 42, 133.

Photochemistry

26

lLU,shows that the lLUand 3L, states cross in energy between 19 and 28 k b a ~ . l ~ ~ The photophysics of the triplet states of the diphenyl polyenes 1,6-diphenyl1,3,5-hexatriene (DPH) and l,S-diphenyl-l,3,5,7-octatetraene(DPO) have been investigated : this study includes measurement of the triplet-triplet absorption spectra using triplet energy sensitizers.16oThe spectra show very large extinction coefficients ( E ~ ->~ lo5 dm3mol-1 cm-l) similar in value to those of the closely related biologically important polyenes, e.g. /3-carotene. The build-up of the triplet-triplet absorption of acridine following picosecond laser excitation yielded a rate constant value of 8 x 1Olos-l.lS1 The rate is suggested to result from population of T1(m*) by internal conversion from T,(rrn*)formed by intersystem crossing from Sl(nn*) which lies just below the S,(.rrm*) state 'reached directly by absorption. Two-photon Absorption Spectra.-There has been a considerable growth of interest in multiphoton-particle interactions since the last report. This arises mainly because of their relevance to methods of laser isotope separation which are reported on later in this Chapter. This section concentrates mainly on twophoton excitations in which there are no selective photodissociation or photoionization processes, but which are of spectroscopic interest because of their possible avoidance of Doppler broadening, thus revealing new layers of spectroscopic detail, and because the symmetry rules which apply differ from those in conventional one-phot on excitations. Recent theoretical and experimental results relating to Doppler-free twophoton absorption spectroscopy have been discussed 162 and reviewed.ls3 The dependence of the emission anisotropy on the sum of the two frequencies involved in a symmetry-forbidden two-photon transition has been investigated and the possibility of determining the symmetry of the active non-totally symmetric vibrations from the structure seen in the dependence has been discussed.lG4 The power-broadening linewidths and level shifts for two-photon absorption by a rigid-rotor model have been calculated using a semi-classical theory 166 and the relationship between two different theoretical approaches to optical twophoton absorption has been discussed in a joint paper by their originators.lGs The time dependence 16' and structure (three-level atom case) 16* of two-photon resonance fluorescence spectra have been discussed. It has been claimed that two-photon cross-sections commonly used in the literature are dependent on the mechanism of the m e a ~ u r e m e n t .This ~ ~ ~ arises because the number of field modes involved in the process changes as the relative energy of the two photons is changed. A new cross-section is proposed which has dimensions of S p h o t o r 2molecule-' and which is a molecular constant. R. W. Shaw and M. Nicol, Chem. Phys. Letters, 1976, 39, 108. R. Bensasson, E. J. Land, J. Lafferty, R. S. Sinclair, and T. G. Truscott, Chem. Phys. Letters,

lbe

le0

1976, 41, 333.

Y. Hirata and I. Tanaka, Chem. Phys. Letters, 1976, 41, 336. B. Cagnac, Ann. Phys., 1975, 9, 223. N. Bloembergen and M. D. Levenson, Topics Appf. Phys., 1976, 13, 315. lE4 V. A. Gaisenok and A. M. Sarzhevskii, Optika i Spektroskopiya, 1975, 39, 514. leb M. D. Burrows and W. R. Saizman, Phys. Rev. (A), 1977, 15, 1636. lee D. Grischkowsky and R. G. Brewer, Phys. Reu. (A), 1977,15, 1789. lE7 E. Courtens and A. Szoke, Phys. Rev. ( A ) , 1977, 15, 1588. ls8 B. Sobelewska, O p f . Comm., 1977, 20, 378. lEe 0. Kafri, Chem. Phys. Letters, 1975, 36, 624. lE1 lsa

Spectroscopic and Theoretical Aspects

27

A timely paper develops expressions for the two-photon absorption crosssections in linearly and circularly polarized light for rotating diatomic molecules, systems for which experimental data are now a~ai1able.l~~ The authors again emphasize the particular value of polarization measurements in assigning state symmetries and point out a possible difficulty in using two-photon excitation spectra (emission-detected absorption spectra) as opposed to two-photon absorption spectra, when making accurate comparisons of theory and experiment. This difficulty can arise if additional photons interact with the emitting state being monitored. It has been pointed out that polarization effects in twophoton spectroscopy depend critically on the rotational branch being observed, and that care must be taken to avoid m i s a s ~ i g n m e n t s172 .~~~~ Coherent two-photon excitation 173 and two-photon-induced coherence 174 have been studied. In the former case linewidths narrower than the Fourier-transform limit of a single pulse of the excitation pulse train were observed for the Na 3 s - 5 s transition. One of the processes which can be carefully investigated now that Doppler-free spectroscopy is possible is collisional broadening of lineshapes. The theory of these collisional processes has been discussed and it is pointed out how they can be used to distinguish between two-quantum and stepwise contributions to the overall two-photon excitation rate.176 There have been a number of studies of the two-photon spectroscopy of atoms. The fine structure splitting of the Na 4F state was measured as 229 k 4 MHz 176 and a polarization rotation effect observed for the Na two-photon 3S-5S transition forms the basis of a very fast optical Highly excited S and D states of K were investigated experimentally and theoretically 178 and, in what is believed to be the first reported case, two-photon amplification has been observed, in coherently inverted K atoms.170 Hyperfine splittings and isotope shifts of several levels lE0and excitation of energetic D states (n = 11 ic 32) 181 in atomic Rb have been measured. Selective excitation of n 2D levels in Cs up to n = 19 has been achieved,ls2 and saturation of two-photon transitions has been observed in CslE3and Th.lE4 The isotope shift in the 1S-2s transition in atomic hydrogen and deuterium was measured to be 670.933 f 0.056 CHz.l*s The 1s Lamb shifts were measured to be 8.20 k 0.10 GHz(H) and 8.25 & 0.11 GHz (D). .The Ne 3s, J = 2 to 4d'(5/2),J = 3 transition has also been observed lE6by two-photon spectroscopy. R. G. Bray and R. M. Hochstrasser, Mol. Phys., 1976, 31, 1199. L. Wunsch, H. J. Neusser, and E. W. Schlag, Chem. Phys. Letters, 1976, 38, 216. R. Lombardi, D. M. Friedrich, and W. M. McClain, Chem. Phys. Letters, 1976, 38,213 179 R. Teets, J. Eckstein, and T. W. Hansch, Phys. Rev. Letters, 1977, 38, 760. 17' M. Matsuoka, Opt. Comm., 1975, 15,84. 17c, P. R. Berman, Phys. Rev. (A), 1976, 13, 2191. 17e P. F. Liao and 5. E. Bjorkholm, Phys. Rev. Letters, 1976, 36, 1543. 177 P. F. Liao and G. C. Bjorklund, Phys. Rev. Letters, 1976, 36, 584. 178 C. D. Harper, S. E. Wheatley, and M. D. Levenson, J. Opt. SOC. Amer., 1977, 67, 579. 179 N. Tanno, Y. Adachi, K. Yokoto, and H. Inaba, Opt. Comm., 1976,18, 111. 180 D. E. Roberts and E. N. Fortson, Opt. Comm., 1975,14,332. Y. Kato and B. P. Stoicheff, J. Opt. SOC.Amer., 1976, 66,490. lea S. M. Curry, C. B. Collins, M. Y. Mirza, D. Popescu, and I. Popescu, Opt. Comm., 1976, 16,

171

251. J. F. Ward and A. V. Smith, Phys. Rev. Letters, 1975, 35,653. la* C. C. Wang and L. I. Davis, jun., Phys. Rev. Letters, 1975, 35, 650. S. A. Lee, R. Wallenstein, and T. W. Hansch, Phys. Rev. Letters, 1975, 35, 1262. F. Biraben, E. Giacobino, and G. Grynberg, Phys. Rev. (A), 1975, 12, 2444.

28

Photochemistry

Studies of the E state in molecular iodine, which can be populated by twophoton absorption, have been extended to include measurements of the E + B transition radiative lifetime.ls7 This was found to be short (27 ns), indicating a strongly allowed transition, and strongly supports a 31108+ assignment for the E state. A more detailed investigation with two independently tunable dye lasers has located five electronic states of T2 in the region 40 00045 000 cm-l above the ground state and the so-called E fluorescence, seen in mild electrical discharges in 12, is seen to result from the emission from three of the five states to the B excited state.lss The A2Z+(v = 0) and (v = 0, 1,2) lQostate of NO has been studied and high-resolution Doppler-free two-photon resonances in 14NH3have been measured.lQ1~ lQ2 Two-photon absorption studies of a number of xanthenes, cyanines, and acridines have shown that the two-photon cross-sections in these molecules do not obey a general rule that had been proposed several years This work provoked further discussion in the 1 i t e r a t ~ r e .lQ6 l ~ ~ Fluorescence ~ spectra, corresponding to emissions from upper excited singlet states, have been observed for three xanthene dyes in a study which involved consecutive absorption of two photons, i.e. where two-photon absorption proceeds via one-photonproduced real intermediate states which in this study are the levels of the first excited singlet state.lae Apart from the thermal blooming experiments mentioned previously, there have been several other studies of the electronic spectroscopy of benzene. The symmetry- and parity-forbidden Sl + So (1B2u-1Al,)two-photon transition in benzene, C6H6 and C6D6,vapour has been measured at 2 cm-’ resolution, which is sufficient to show moderate rotational structure on the vibrationally induced bands.lQ7 The two-photon rotational band contours of the vibronic transitions were calculated and found to be in good agreement with the experiments which show the strong polarization changes in the rotational bands. Calculated and experimental results for the strongest totally symmetric TPE band bzu 14; in [lH,]benzene are shown in Figure 9. Progressions were observed where the intensity was greater for more highly excited quanta which shows the importance of anharmonic mixing in the excited state. The analogy between two-photon absorption and Raman scattering processes was demonstrated in an experiment on gas-phase benzene in which two lasers operating at different frequencies and polarizations were used for excitation.1es An extensive and detailed investigation of the two-photon fluorescence excitation spectrum of crystalline naphthalene at 77 K and naphthalene in durene at 1.6 K has been made with a single laser beam and at 1 cm-l resolution.1g9 N

D. L. Rousseau, J. Mol. Spectroscopy, 1975, 58, 481. M. D. Danyluk and G. W. King. Chem. Phys. Letters, 1976, 44, 440. J. A. Gelbwachs, P. F. Jones, and 5. E. Wessel, App. Phys. Letters, 1975, 27, 551. lg0 H. Zacharias, 5. B. Halpern, and K. H. Welge, Chem. Phys. Letters, 1976, 43, 41. lS1 W. K. Bischel, P. J. Kelly, and C. K. Rhodes, Phys. Reo. (A), 1976, 13, 1829. lg2 W. K. Bischel, R. R. Jacobs, and C. K. Rhodes, Phys. Rev. (A), 1976, 14, 1294. lQ9 B. Foucault and 5. H. Hermann, Opt. Comm., 1975,15,412. l S 4 F. P. Schaefer and W. Schmidt, Opt. Comm., 1976, 17, 11. Ig6 J. Ducuing, B. Foucault, and 5. P. Hermann, Opt. Comm., 1976, 17, 267. lee G. C. Orner and M. R. Topp, Chem. Phys. Letters. 1975, 36, 295. lQ7 L. Wunsch, F. Metz, H. J. Neusser, and E. W. Schlag, J. Chem. Phys., 1977, 66, 386. lg8 W. Hampf, H. J. Neusser. and E. W. Schlag, Chem. Phys. Letters, 1977, 46, 406. 19* R. M. Hochstrasser and H. N. Sung, J. Chem. Phys., 1977, 66, 3276.

Spectroscopic and Theoretical Aspects a)

l

iL;q

polar ir cd

29

11 pclarized

11

polarized

vo = I570 cm-l Figure 9 Comparison of the computer plot (a) with the experimental spectra (b) of the strongest totally symmetric TPE band bzu 14; in [lH,]benzene. In the calculated profiles the rotationaI branches are separately plotted. Qo and Q2 correspond to the isotropic and anisotropic contributions to the totally symmetric TP transitions as discussed in the text. The wavenumbers in (a) and (b) are drawn on the same scale. The band in circularly polarized light is enlarged by a factor of 10 in (a) and about 9 in (6) (Reproduced by permission from J. Chem. Phys., 1976, 66, 386)

30

Photochemistry

Amongst the many observations is the appearance of the one-photon o--O band levels in the two-photon spectrum. Their polarization behaviour suggests that they are magneto-electric dipole two-photon transitions. Single-photon excited single-vibronic-level (SVL) studies in the naphthalene S, state are complicated by the intrusion of S, levels at even moderate excess energies. In two-photon studies this problem is absent as the S, +- So transition is forbidden whereas in the single-photon case it is strongly allowed. Advantage has been taken of this fact in a study of SVL lifetimes in the S, state of naphthalene which indicates that the intersystem crossing to triplet states is dominated by FranckCondon factors and not by vibronic inductions and promoting modes.200 A study of anthracene solutions investigates the interference of singlet-triplet transitions on two-photon absorption measurements. Such transitions frequently occur in the same spectral region and it is suggested that only fluorescencedetected methods of measuring two-photon absorption cross-sections are suitable in these situations.201 A detailed study of the two-photon excitation spectra of [lHlo]- and [2H,o]-biphenylneat and mixed crystals at 77 and 1.8 K includes a theoretical description of the anisotropy of such spectra in crystals which reveals the importance of birefringence effects, neglected in previous theoretical treatments. 202 Phosphorescence-detected two-photon absorption spectra of pyrazine ,03 and triphenylene 204 have been measured. Phosphorescence detection has the advantage over its fluorescence counterpart that scattered laser radiation can be discriminated against because of the long phosphorescence lifetime. The band origin of the S2+- So transition in triphenylene obscured in the one-photon spectrum was observed at 33 200 cm-l. The discussion of two-photon allowed and vibronically induced transitions in the paper on pyrazine is particularly recommended. Table 3 shows the two-photon one-electron excitation pathways out of the l A , ground state of pyrazine to the lower energy excited g states. In an interesting note, a strong two-photon absorption in solutions of transstilbene is The band occurs between the one-photon A and B bands and supports the idea that low-lying electronic excited states exist for stilbene and longer polyenes having the same symmetry as the ground state. The twophoton absorption spectrum of 3,4-benzopyrene in the region 530-750 nm has been measured.20s Multiphoton Absorption Processes.-A number of papers have dealt with the more general aspects of the theory of multiphoton-particle interactions concerning especially the interpretation of infrared laser experiments reported on later in this chapter. They include papers on coherent multiphoton propagation effects such as two-photon self-induced the effects of photon loo 201 20% 80s

2.4

306 a06

20'

U. Boesl, H. J. Neusser, and E. W. Schlag, Chem.Phys. Letters, 1976, 42, 16. J. Kleinschmidt and M. Schubert, Exp. Tech. Phys., 1977, 25, 9. R. M. Hochstrasser and H. N. Sung, J. Chem. Phys., 1977, 66, 3265. P. Esherick, P. Zinsli, and M. A. El-Sayed, Chem. Phys., 1975,10,415. R. J. M. Anderson, G. R. Holtom, and W. M. McClain, J . Chem. Phys., 1977, 66, 3832. L. Singer, Z. Baram, A. Ron, and S . Kimel, Chern. Phys. Letters, 1977, 47, 372. J. Krasinski, W. Majewski, and M. Glodz, Opt. Comm., 1975, 14, 187. J. Katriel and S. Speiser, Chern. Phys., 1976, 12 ,291.

31 Table 3 One-electron excitation pathways in pyrazine for two-photon excitation from the ground state, 1A,, to the lower energy excited g states Spectroscopic and Theoretical Aspects

Intermediate states

Final states 3A,

a

a The low-lying doubly excited (nn*)a state 2lA, has been omitted. The only allowed excitation route to this state is successive out of plane (n-m) transitions which are each of weak intensity.

statistics,208 and solutions of the Schrodinger equations 209 for the basic mu1t iphot on-two-level atom problem. The possibilities for ul tra-high-resolut ion studies using multiphoton resonances in two-level systems have been explored,210 and a variational method for calculating cross-sections for various multiphoton processes has been proposed and applied successfully to a number of processes in the H atom.211 A review of multiphoton processes in atoms has appeared.,12 A simple procedure for calculating steady-state vibrational level populations in multiphoton absorption processes has been given which should prove useful in studies of laser-catalysed reactions despite its rather restricted Calculations of the probabilities of two- and three-photon transitions between the vibrational-rotational levels of diatomic and polyatomic molecules indicate that at the intensities required for three-photon processes dissociation and gas breakdown processes would be dominant.214 A model potential, analogous to the Fues potential, has been suggested for use in the calculation of multiphoton processes in diatomic molecules,21sand the results of quantum theoretical and semiclassical treatments of high-intensity light absorption have been critically compared.21s Linear Dichroism.-The applications of ultraviolet linear dichroism 217 and the linear dichroism of biological chromophores 218 have both been reviewed. The V. V. % l oV. L. zo8

21a 218

z14 216

zle 217

Ernst, 2. Phys. (B), 1976,23, 103. Emst, 2. Phys. (B), 1976, 23, 113. P. Chebotaev, Kvantovaya Electron. (Moscow), 1976, 3, 694. N. Labzovskii and N. S . Yakovleva, Optika i Spektroskopiya, 1975, 39, 1009. P. Lambropoulos, Adv. At. Mol. Phys., 1976, 12, 87. J. T. Hougen, J. Chem. Phys., 1976, 65, 1035. V. E. Merchant and R. N. Isenor, IEE J. Quantum Electron., 1976, 12, 603. N. L. Manakov and V. D. Ovsyannikov, Phys. Letters (A), 1976, 58, 385. J. M.Schurr, Internat. J. Quantum Chem., 1976, 10, 359. A. Yogew and Y Mazur, Method. Chim., 1974, lA, 430. J, Hofrichter and W. A. Eaton, Ann. Rev. Biophys. Bioeng., 1976, 5, 511.

Photochemistry

32

method of moments has been used in a study of linear dichroism associated with Jahn-Teller effects in the optical bands of activated crystals219and it has been noted that the measurable quantities, A,-Al, the dichroism, and A , the unpolarized absorption, strictly define the quantities A , and A , which are usually measured separately.z20 An explanation, based on a classical model, has been proposed for the photo-orientation effect, i.e. an induced anisotropy in an ensemble of anisotropically absorbing molecules when exposed to resonant plane-polarized light, wherein a light-induced torque tends to orient an oscillator perpendicularly to the electric field of the incident light .z21 A modified circular dichroism spectrometer has been used in measurements of the linear dichroism of small molecules which are partially orientated by a previously oriented polymer matrix.zzz A number of idealized situations for different relative orientations of the transition moment and the orientation axis are analysed, and the technique is applied to the near-ultraviolet bands of CS2, NOz, and SO,. A method, thought to be more general and reliable, for calculating the direction of the electronic transition moment of a planar molecule in a stretched polymer film has been proposedzz3and applied to naphthalene and acridine-cation in poly(viny1alcohol) films.224The relationship between electronic transition moments and the symmetry of electronic wavefunctions has been discussed with particular reference to the case of linear M O I ~ C U ~ ~ S . ~ ~ ~ A particularly interesting paper describes an investigation of picosecond timeresolved photoinduced dichroism in solutions of rhodamine dyes.2z6An intense polarized excitation pulse partially bleaches the dye, resulting in an anisotropic distribution of ground-state molecules which is monitored with a weak collinear probing beam polarized perpendicularly to the excitation pulse. The results are discussed in terms of a linear oscillator model of electric dipole transitions which predicts a value of 1 for the molecular dichroism, D, [D, = (al, - al)/ (a, + al) where CT! and a, are the absorption cross-sections parallel and perpendicular to the molecular axis, respectively]. The value of D, cannot be clearly obtained from studies of molecules in stretched films and the best evidence for the values of D, for a number of molecules, particularly dyes, comes indirectly from careful fluorescence polarization studies but here the interpretation of the results necessarily includes some assumptions concerning the relative orientations of the absorbing and emitting oscillators. The laser technique involves only the oscillator corresponding to absorption which in this case is the So-& transition in Rhodamine 6G and in Rhodamine B. Values of D , in the range 0.90 k 0.03 to 0.95 If: 0.03 were found for these dyes in a number of solvents and the deviations from the value predicted by the linear oscillator model were attributed to differently orientated transition moments for the transitions from various 219

Yu. E. Perlin, L. S. Kharchenko, and B. S. Tsukerblat, Zzvest. Akad. Nauk S.S.S.R., Ser. Fiz., 1976, 40, 1770.

220

221 222

223 224 226

226

A. Davidson and B. Norden, Chem. Scripta, 1975, 8, 95. T. Romanovskis and 0. Smits, Nature, 1975, 258, 137. B. Norden, Chem. Scripta, 1975, 7 , 167. K. R. Popov, Optika i Spectroscopiya, 1975, 39, 509. K. R. Popov, Optika i Spektroskopiya, 1975, 39, 656. L. E. Brus, J. Mol. Spectroscopy, 1977, 64, 376. A. Penzkofer and W. Falkenstein, Chern. Phys. Letters, 1976, 44, 547.

Spectroscopic and Theoretical Aspects 33 thermally populated vibrational levels of So to vibrational Franck-Condon levels in Sl. The use of linear dichroism measurements to aid in the assignment of electronic absorptions bands is well illustrated in important papers on the electronic states of bridged [14]annulenes with anthracene perirneferl3O and a series on the electronic states of a c e ~ ~ a p h t h y l e npleiadiene e,~~~ (cyclohepta [delnaphthalene),22* the relationship between their respective absorption and a general classification of the electronic states of these two non-alternant hydrocarbons and their close derivatives mentioned elsewhere in this Chapter. A similarly detailed study of fluoranthene and five isomeric aminofluoranthenes emphasizes the value of the stretched-film method for measuring linear dichroism when dealing with a series of closely related The polarized absorption spectra of aceheptylene, 3,5-dimethylaceheptylene, and 3,5,8,10-tetramethylaceheptylenehave been measured in stretched polyethylene sheets at 77 K. The results of PPP-CI calculations of the transition energies, oscillator strengths, and polarization directions agreed well with the experimental observations for the first ten The two lowest-energy transitions of pyridine l-oxide at 30 000 cm-1 and 36 000 cm-l were found to be in-plane polarized, perpendicular, and parallel to the long molecular axis respectively, in a study where orientation of the molecules was achieved by using an electrically aligned nematic liquid crystal mixture as the supporting medium.232 Circular Dichroism.-Papers dealing with the diagnostic use of circular dichroism (CD), e.g. conformational studies in biochemistry, are not covered here unless they serve to illustrate some general point. There has been further discussion in the literature of the measurement of the circular dichroism of fluorescent chromophores by monitoring the fluorescence intensity when the sample is excited alternately with left and right circularly polarized light, i.e. fluorescence-detected circular dichroism (FDCD). Although originally proposed as a means of resolving overlapping CD bands when one component is fluorescent there are a number of other interesting uses for the method, e.g. where an optically inactive fluorescent molecule monitors the CD of a non-fluorescent optically active molecule. This example and others have been analysed to give the relationship between observables and the molecular 234 The important effects of photoselection on FDCD measurements have been The fluorescence-detected magnetic circular dichroism (FDMCD) of both tryptophan and dye-labelled Fe"' cytochrome c was shown to be the same as that found by conventional transmission-detected magnetic circular dichroism (MCD).23e 227

2a8

a2D 230 231

232 233 234

a35

a36

E. W. Thulstrup and J. Michl, J. Amer. Chem. SOC., 1976, 98, 4533. J. Kolc and J. Michl, J . Amer. Chem. SOC.,1976, 98, 4540. J. Michl, J . Amer. Chem. SOC.,1976, 98, 4546. E. W. Thulstrup, M. Nepras, V. Dvorak, and J. Michl, J. Mol. Spectroscopy, 1976, 59, 265. P. Bischof, R. Gleiter, K. Hafner, M. Kobayashi, and J. Spanget-Larsen, Ber. Bunsengesellschaft phys. Chem., 1976, 80, 532. R. D. Peacock and B. Samori, J.C.S. Furuday II, 1975,71, 1909. I. Tinoco, jun. and D . H. Turner, J. Amer. Chem. SOC.,1976, 98, 6453. T. G. White, Y. Pao, and M. M. Tang, J. Amer. Chem. SOC.,1975, 97,4751. B. Ehrenburg and I. Z. Steinberg, J. Amer. Chem. SOC.,1976, 98, 1293. J. C. Sutherland and H. Low,Proc. Nut. Acud. Sci. U.S.A., 1976,73, 276.

34

Photochemistry

A linear response polarizability theory of the configurationally induced optical rotation in helical biopolymers has been formulated and applied to single~ ~ ~ fundamental aspects stranded poly-(A) having a DNA c o n f i g ~ r a t i o n . Some of rotational strength calculations have been examined238and a theory of the induced CD of achiral molecules in the presence of chiral molecules has been extended to include random orientations of the interacting Symmetry rules have been formulated which determine those states of an achiral complex which are expected to exhibit dispersion-induced CD (DICD) as a result of dispersive coupling with chiral species.24oA study of vibronic contributions to DICD suggests that it may be a useful method in giving symmetry assignments to

‘ a ) A€

t

0-1,

I

S isomers

R isomers

--

S isomers

geometry

02,

geometry

torsional amplitude -q,

Dz

geometry

torsional amplitude t q ,

A

f

I

D2

D2

gcqmetry

I 02,

geometry

t

torsional omplitude -9,

R isomers

DZ

geometry b

torsional amplitude +q,

Figure 10 Schematic representations of the variation of electronic energies with the torsional co-ordinate q1 for a D Z d molecule; ( a ) in the absence of chiral neighbours. Vibrational probability functions are indicated and the exact cancellation of vibronic contributions to the C D spectrum is depicted; ( b ) in the presence of chiral neighbours. The uncancelled vibronic contributions to the C D spectrum are shown (Reproduced by permission from Chem. Phys. Letters, 1976, 41, 225)

electronic and vibrational A new type of induced CD has been proposed for achiral molecules of DSd symmetry, e.g. s ~ i r e n e s .It~ arises ~ ~ from vibronic effects as illustrated in Figure 10, following the lifting of orbital degeneracy in the excited state of such a molecule by a chiral environment. A theoretical analysis of two-photon absorption by an optically active molecule shows that differential absorption, as between left and right circularly polarized 237

2s* 241 24a

H. Ito, T. Eri, and Y . 5. Ihaya, Chem. Phys. Letters, 1976, 39, 150. F. W. King, J.C.S. Faruday IZ, 1976, 7 2 , 225. S. F. Mason, Chem. Phys. Letters, 1975, 32, 201. P. E. Schipper, Znorg. Chim. Actu, 1975, 14, 161. P. E. Schipper, Chem. Phys., 1976, 12, 15. D. P. Craig and P. J. Stiles, Chem. Phys. Letters, 1976, 41, 225.

Spectroscopic and Theoretical Aspects

35 light, i.e. two-photon CD is to be expected.243Two-photon CD arising from a pair of achiral chromophores dissymmetrically situated with respect to each other has also been The vibrational CD of the O-H and C*-H stretching vibrations in 2,2,2-trifluoro-l-phenylethanol e46 and of the C-H stretching vibration in [%H,]tartaric acid246have been reported. The effect has also been observed in overtone and combination bands of other chiral m01ecules.~~~ A free-electron theory of the optical activity characteristic of a potential supported by an arbitrary space curve has been developed and applied to a number of model A study of the near-ultraviolet optical activity of dissymmetric pyridyl compounds has involved calculation of the optical rotatory strengths by treating the substituent groups, which contain an asymmetric carbon atom, as perturbations of the pyridyl chromophore (a one-electron The study suggests that the third singlet-singlet transition (second n + n* transition) of this chromophore contributes very little to the near-ultraviolet optical activity of these compounds, and that because of the closeness of the three lowest singlet states, pseudo-Jahn-Teller-type vibronic interactions may have profound effects on their CD spectra. The use of the important fluorescent probe molecule 1,8-anilinonaphthalenesulphonate (1,8-ANS) as a CD probe in complexes with proteins has been discussed following semi-empirical MO calculations on N-phenylnaphthylamines.260 Only small contributions of inherent dissymmetry to the rotational strengths of the long-wavelength band were found and therefore protein-dyecoupled oscillator interactions may be dominant in the induced CD bands observed for such systems.261 There is an increasing number of reports of CD spectral measurements being extended to the vacuum ultraviolet, including studies of collagen, gelatin, and p ~ l y ( p r o l i n e ) I poly(L-pro1ine)I ,~~~ and II,2s3 p-forming alkyl o l i g ~ p e p t i d e sand ,~~~ N-acetylglucosamine, glucuronic acid, and hyaluronic An MCD investigation of ethylene in the near vacuum-ultraviolet has revealed a new electronic transition, tentatively assigned to a 7~ -+ 3p, transition, beneath the N --f V transition.266 An interesting study of trapped electrons in frozen polar media by MCD gives strong evidence that in aqueous systems the solvation 243

244

24b 246

E. A. Power, J. Chem. Phys., 1975, 63, 1348. D. L. Andrews, Chem. Phys., 1976, 16, 419. L. A. Nafie, J. C. Cheng, and P. S. Stephens, J. Amer. Chem. SOC., 1975,97, 3842. H. Sugeta, C. Marcott, T. R. Faulkner, J. Overend, and A. Moscowitz, Chem. Phys. Letters, 1976, 40, 397.

247

248 z48

2bo 261 26a

253 264

2s6

T. A. Keiderling and P. J. Stephens, Chem. Phys. Letters, 1976, 41, 46. N. L. Balazs, T. R. Brocki, and I. Tobias, Chem. Phys., 1976, 13, 141. C. Yeh and F. S. Richardson, J.C.S. Faraday ZZ, 1976,72, 331. J. C. Smith and R. W. Woody, J. Phys. Chem., 1976, 80, 1094. See, for example, W. E. Muller, Z . physiol. Chem., 1976, 357, 1487. D. D. Jenness, C. Sprecher, and W. C. Johnson, jun., Biopolymers, 1976, 15, 513. M. A. Young and E. S. Pysh, J. Amer. Chem. SOC.,1975, 97, 5100. J. S. Balcerski, E. S. Pysh, G. M. Bonora, and C. Toniolo, J. Amer. Chem. SOC.,1976, 98, 3470. L. A. Buffington and E. S. Pysh, J. Amer. Chem. SOC.,1977,99, 1730. M. Brith-Lindner and S. D. Allen, Chem. Phys. Letters, 1977, 47, 32.

Photochemistry

36

environment of the trapped electron is highly symmetrical whereas in alcohol A related MCD-e.s.r. study of trapped glasses it is of much lower electrons in y-irradiated poly(viny1 alcohol) films indicates a glass-like trapping site arising from absorbed water.268 An MCD investigation through the Soret bands of ferricytochrome c and deoxyhaemoglobin showed opposite signs for the two As the former involves a low spin (S = *) ferric haem and the latter a high spin (S = 2) ferrous haem, dual spin-orbit coupling mechanisms are invoked. The first, dominant in deoxyhaemoglobin, is the normal spin-angular momentum interaction and the second, dominant in ferricytochrome c, is due to the interaction of the metal ion spin with the orbital motion of the porphyrin rr-electrons. MCD studies of dicycl~hepta[cd,gh]pentalene,~~~ the triphenylcarbenium ion,261and protoporphyrin derivatives have been reported.262 Stark and Related Effects.-A valuable review article on the dipole moments and polarizabilities of molecules in excited electronic states has appeared which discusses both the theory and results of gas-phase Stark measurements and condensed-phase electrochromism The theory of electrochromism, which is the electric field dependent change in absorption and emission coefficients, has been discussed with particular emphasis on solvent contributions to condensed-phase electrochromi~m.~~~ The wider applications of electrochromism studies have been reviewed 265 and their use in biological membrane studies has been discussed.2s6 An instrument has been developed for the measurement of medium-resolution gas-phase electrochromism in the region 300.00165.0 nm and in the first instance used to study the electrochromism spectrum of PhNHz in the region of the B2 +- Al (m*) transition at 293.84 nm.267The dipole moment change which was obtained compared favourably with a value obtained from high-resolution Stark effect studies. Electric field quenching of the exciplex fluorescence observed in a poly(N-vinylcarbazole) film doped with dimethyl terephthalate has been attributed to a field-assisted thermal dissociation of the exciplex into free carriers.266 Highintensity radiation has been found to quench the fluorescence from a number of complex organic molecules in the vapour phase and in solution and this has been explained in terms of a linear Stark effect.26sThe influence of an electric field on the co-operative transitions of linear biopolymers has been investigated with particular emphasis on the helix-random coil t r a n ~ f o r m a t i o n . ~ ~ ~ 267 258 260

2Go

261 2fi2

263

T. Takui, K. Itoh, Y. Waka, and H. Kawakami, Chem. Phys. Letters, 1975, 35,461. T. Takui, Y. Waka, H. Kawakami, and K. Itoh, Chem. Phys. Letters, 1975, 35, 465. M. A. Livshits, A. M. Arutyunyan, and Yu. A. Sharonov, J. Chem. Phys., 1976, 64, 1276. V. Kratochvil, J. Kolc, and 5. Michl, J. Mol. Spectroscopy, 1975, 57, 436. H. P. J. M. Dekkers and E. C. M. Kielman-Van Luyt, Mol. Phys., 1976, 31, 1001. T. Shimizu and T. Nozawa, Bioinorg. Chem., 1976, 6, 77. W.Liptay, in ‘Excited States’, Vol. 1, ed. E. C. Limy Academic Press, New York, 1974, p. 129.

2F4

266

267

270

W. Liptay, Ber. Bunsengesellscharftphys. Chem., 1976, 80,207. H. Labhart, Ber. Bunsengesellschaftphys. Chem., 1976, 80, 240. W. K. Cheng, Biomembranes, 1975, 7 , 56. G. C. Causley, J. D. Scott, and B. R. Russell, Rev. Sci. Instrum., 1976, 48, 264. M. Yokoyama, Y. Endo, and H. Mikawa, Chem. Phys. Letters, 1975, 34, 597. V. L. Bogdanov and B. D. Fainberg, Optika i Spektroskopiya, 1976, 41, 799. G. Schwarz, Biopolymers, 1975, 14, 1173.

Spectroscopic and Theoretical Aspects 37 Although the numerous papers on Stark effects in atomic spectroscopy are not covered here, attention is drawn to a review of Stark effects on linewidths and positions of non-hydrogen-like atoms which critically evaluates and collects data from a large number of published The electric dipole polarizabilities of the first excited singlet and triplet states of a number of conjugated hydrocarbons have been calculated using a modified Rayleigh-Schrodinger perturbation theory with an SCF-LCAO The results are compared with previous more sophisticated configuration interaction perturbation calculations. The contribution of intramolecular charge-transfer to the hyperpolarizabilities of nitroanilines and its relation to the first excited state energy, oscillator strength, and dipole moment of these molecules has been treated t h e ~ r e t i c a l l y . ~ ~ ~ The excited-state dipole moments of picrates of pyridine, 2-methylpyridine, 3-methylpyridineYand 4-methylpyridine 274 and of azulene and 3,5-dimethylcyclopenta[ef]heptalene 275 have been determined from measurements of the frequency of the solute absorption and emission spectra in a number of different solvents. A number of papers deal with Stark effects in molecular crystals and a few are selected here including an interesting paper on an ionic crystal potassium chlorochromate, KCrO,Cl, wherein the orientational degeneracy is removed in the applied field to give a so-called pseudo-Stark effect. The pseudo-Stark splittings are similar in magnitude to those found in molecular Stark effects have been reported on the luminescence spectra of a number of benzaldehyde the electric field-induced transition to the lowest triplet state of 9,lO-anthraquinone which leads to its assignment as 3B1g,278 the phosphorescence of p-benzoquinone which establishes that there is a double minimum potential in the first triplet on the electronic origins of the lBZu+- l A , absorption system of tetracene and pentacene,280and on the spectra of azulene and 1,3-diazaazulene in host crystals which show clearly the vibrational sublevel dependence of the dipole moment in the excited state.281 The use of Stark spectra to study the vibronic coupling between nearly degenerate electronic states has been discussed with particular reference to the n-n* spectra of quinones.282A study of the Stark spectra of phthalocyanines in solid solutions in poly(methy1 methacrylate) indicates that the dipole moments in the ground and excited state are both 271

27d

275 27e

278

27B 280

281

288

R. N. Konjevic and J. R. Roberts, J. Phys. and Chem. Ref. Data, 1976, 5, 209. M. Trsic, J. E. Sanhueza, and L. L. Espinoza, Infernat.J. Quantum Chem., 1976, 10, 429. J. L. Oudar and D. S. Chemla, J. Chem. Phys., 1977, 66, 2664. 0. Sedoyama, M. Higashi, Y. Akimoto, and H. Yamaguchi, Bull. Chem. SOC.Japan, 1977, 50, 299. H. Yamaguchi and T. Ikeda, Bull. Chem. Sac. Japan, 1976,49, 1762. J. H. Hoeg, C. J. Ballhausen, and E. I. Solomon, Mol. Phys., 1976, 32, 807. Y. Udagawa and D. M. Hanson, Chem. Phys. Letters, 1977, 45, 228 (and refs. therein). J. P. Galaup, J. Megel, and H. P. Trommsdorff, Chem. Phys. Letters, 1976, 41, 397. Y. Muyagi, M. Koyanagi, and Y. Kanda, Chem. Phys. Letters, 1976, 40, 98. 5. H. Meyiing, W. H. Hesselink, and D. A. Wiersma, Chem. Phys., 1976, 17, 353. R. Clark and G . J. Small, J. Chem. Phys., 1977, 66, 1779. H. P. Trommsdorff and J. P. Galaup, Ber. Bunsengesellschaftphys. Chem., 1976,80,229. L. M. Blinov, N. A. Kirichenko, and N. V. Duninin, Zhur. priklad. Spektroskopii, 1976, 25, 548.

38

Photochemistry 4 Intramolecular Excited-state Decay Processes

Theories of Radiative Decay Processes.-An elementary theoretical account of radiative transition phenomena in molecules has been given which includes linewidths and level shifts.284The treatment is based upon a particular formulation of Schrodinger theory and yields the Fermi golden rule for transition rates, the Rabi-Bloch-Siegert absorption profile, and the Weisskopf-Wigner approximation for spontaneous decay under appropriate conditions. A semi-classical theoretical treatment has successfully accounted for the coherent and directional properties of stimulated and spontaneous emission processes, thereby avoiding the usual development in terms of quantum electrodynami~s,~~~ and a previous study of the thermal quenching of luminescence which used a single-configurational co-ordinate has been extended by replacing the Condon-approximation overlap integrals with linear and derivative The concepts used in the description of elementary processes associated with luminescence have been discussed.287 A fundamental relationship between absorption and emission spectra derived by Stepanov from Kirchoff’s law of black body radiation has been re-examined in an interesting analysis which casts doubts on the ‘warm fluorescence’ interpretation of the anomalous temperatures frequently observed when using the Stepanov method in condensed media.2s8 The linear dependence of F(v) on v predicted by the Stepanov relationship F(v) 2 In [ W(v)/y3k(v)] = - (hv/kT)

+ D(T)

(2)

[where W ( Y )and k(v) are the luminescent power and absorption coefficient, respectively, of the molecule at frequency v, and D is a function independent of v] is observed in a number of systems, but the corresponding temperature is frequently found to differ from the ambient temperature T. Values above T and values below T have both been observed. The analysis shows that the above relationship holds only when the ground and excited states are at the same equilibrium temperature and that when they are not equal, distinct curvature should be apparent in the Stepanov plots. As such curvature has not yet been reported in cases of anomalous temperatures, the alternative explanation in terms of inhomogeneous broadening is favoured and this has important implications for investigations of, for example, resonant energy transfer between molecules in solution where it will result in overestimates of transfer rates when using calculated absorption-emission overlap integrals. A discussion of the radiative properties of molecules in vibrational-rotational levels near to the dissociation limit, i.e. so called ‘long-range molecules’, shows that their electronic radiative lifetimes rapidly approach those of the separated atoms whilst their vibrational lifetimes will be at least as long as several years! 2Hg Using what may prove to be a generally useful procedure the singlet-triplet 284 286

28e

287 288

P. W. Langhoff and W. H. Heffner, Internat. J . Quantum Chem. Symposia, 1975, 9, 461. A. V. Durrant, Amer. J . Phys., 1976, 44, 630. C. W. Struck and W. H. Fonger. J. Luminescence, 1976, 14, 253. Y . Toyazawa, J . Lumin., 1976, 12/13, 13. R. L. van Metter and R. S. Knox, Chem. Phys., 1976, 12, 333. W. C. Stwalley and W. T. Zemke, Internat. J . Quantum Chem., Symposia, 1976, 10, 223.

Spectroscopic and Theoretical Aspects

39

transition intensities of naphthalene and anthracene have been calculated.2D0 The treatment is based on INDO wavefunctions and on a semi-empirical spinorbit Hamiltonian. The relationship of molecular geometry in the excited states of conjugated molecules to their luminescence properties has been explored in a series of papers. SCF-CI-SC calculations of excited-state geometries have resulted in a classification system based on the following three categories: (i) strong conjugation over the whole molecule or on the periphery, (ii) long main conjugated framework linked to shorter conjugated fragments, and (iii) two or more slightly conjugated fragments linked to each 292 The classification is useful in the interpretation and prediction of the fluorescence ability of conjugated m ~ l e c u l e s2*4 .~~~~ Radiative Lifetimes from Absorption Spectra. The derivation of the Strickler and Berg equation which relates the radiative lifetime of molecules to the integrated absorption spectrum for the corresponding transition contains an approximation which is not widely recognized.2g6The factor J, defined as

where is the Franck-Condon overlap integral for upper and lower vibronic states U,,and Lt respectively, and ga and pL are the degeneracy factor and level density respectively of the lower state, is assumed to be 1 in the Strickler and Berg derivation. The limit, G, on the integral corresponds to the frequency of the U,,-+ L emission process to the zero-point level of L. The levels of L(1 > f ) above the emitting level of U are not included in the integral, and the closure procedure which sets J = 1 may be a poor approximation in many applications, leading to a large overestimate of the time-dependent total spontaneous transition probability. This approximation may lead to the large differences found between the calculated and experimental values of the radiative lifetimes for molecules such as NO2, SO2, and CS,. The refractive index, n, correction to fluorescence measurements arising from the change in collection geometry is well known, but it is less clear what effect the refractive index has on the actual radiative parameters, TRF and ~ R ( 7 =~1 /~k ~ ~of) the excited state. This question has been investigated by measuring the fluorescent lifetimes and the integrated absorption areas of isopentane solutions of 9, lediphenylanthracene over a wide temperature range.296 A correction in the range r2-r3 to T K F values was found and a number of 7 R p extrapolations to different environments which have appeared in the literature were reinterpreted. Values of the ratio of the experimental radiative lifetime, 7 R p , obtained from T p and $F measurements, to the radiative lifetime calculated from the absorption 2so

lB6

G. L. Bendazzoli, G. Orlandi, and P. Palmieri, Internut. J. Quunt. Chem., 1976, 10, 659. F. Fratev and A. Tadzher, J. Mof. Sfructure, 1975, 27, 185. F. Fratev, J . Mol. Structure, 1976, 30, 217. F. Fratev, 2. Nuturforsch., 1975, 30a, 1691. F. Fratev, W. Monev, 0. E. Polansky, S. Stojanov, and N. Tyutyulkov, 2. Nuturforsch., 1977,32a, 178. S . Lipsky, J. Chem. Phys., 1976, 65, 3799. J. Olmsted, tert., Chem. Phyx. Letters, 1976, 38, 287.

F

Photochemistry

40

~ close , to 1 , but several systems are known which show spectra, T ~ are~ often large deviations from 1 even after trivial reasons, e.g. overlapping absorption bands, are accounted for. A discussion of the anomalies observed in several systems, e.g. benzene, trans-stilbene, and retinol polyenes, has been published.2Q7 It has been proposed that the temperature and solvent dependence of TRY for the retinols and diphenylpolyenes arises from state mixing introduced by solvent perturbation~.~~~

Theories of Radiationless Decay Processes.-A general theory of collisioninduced electronic relaxation phenomena in small and intermediate case molecules has been Attention is focused on the collision-induced intersystem crossing process because of the availability of extensive experimental data for molecules such as glyoxal. The 'mixed-state' model appears as a limiting case of the general theory which is also able to demonstrate the general importance of rotational sublevels in these processes. A new approach to the problem of calculating Franck-Condon factors in polyatomic molecules has been proposed which has some analogy with the treatment of radiative transitions using the Franck-Condon principle.300 Calculations of the electronic or promoting part of both the rate of the non-radiative transitions, 'A2 --f 'Al and 'A2 3Az, and the radiative lifetimes of the states, 3A2(n+,T * ) , 'A2@ m*), lBl(n ++ o*), and lAl(rr ++ m*), in formaldehyde have been published.301 Simple molecular orbitals were used in the calculations. For the symmetry-forbidden radiative transition it is shown that the Born-Oppenheimer correction is of comparable magnitude to the second-order vibronic coupling. A brief account of theoretical work on radiationless transitions for the period 1971-1 973 has appeared.302 The use of lasers with well-defined coherence properties to excite molecules has produced evidence to suggest that true eigenstates of molecules, which may have both singlet and triplet character, may be prepared, even in large molecules, if the laser correlation time is relatively long and the molecular relaxation is slow. The singlet Born-Oppenheimer state, which has a much shorter lifetime than the true eigenstates, is produced on excitation with lasers with a much shorter correlation time.303 This interesting paper will certainly stimulate considerable discussion and further experimental work. A study of solventinduced effects on radiationless transitions distinguishes between two basic types, a so-called medium-induced change where the solute-solvent interaction is the perturbation and an effect arising from a change in the energy gap involved in the process.3o4 An expression for the non-radiative decay rate of an electronically excited atom or molecule or of a vibrationally excited molecule which interacts with the phonons of a crystalline host matrix has been calculated using a Green's function --f

f-)

2g7 288

299

304

J. B. Birks, Z. phys. Chem. (Frankfurt), 1976, 101,91. G. Hug and R. S . Becker, J. Chem. Phys., 1976, 65, 55. K. F. Freed, J. Chem. Phys., 1976,64, 1604. F. Metz, Chem. Phys., 1976, 18, 385. S. H. Lin, Proc. Roy. Soc., 1976, A352, 57. B. R. Henry and W. Siebrand, in 'Organic Molecular Photophysics', Vol. 2, ed. J. B. Birks, Wiley, Chichester, 1975, p. 303. A. H.Zewail, T. E. Orlowski, and K. E. Jones, Proc. Nut. Acad. Sci. U.S.A., 1977, 74, 1310. S. H. Lin, S. T. Lee, Y. H. Yoon, and H. Eyring, Proc. Nat. Acad. Sci. U.S.A., 1976,73,2533.

Spectroscopic and Theoretical Aspects

41

method.305 Application of the theory to the thermal quenching of anthracene fluorescence in different solvents gives satisfactory agreement with experimental results when the S, -+ transition is considered alone, no additional transition needing to be An earlier investigation of the effect of temperature on radiationless transitions has been extended by investigating the temperature dependence of multiphonon relaxation rate processes over a wide temperature range.3o7 The theoretical results are correlated with measurements on rare-earth ion impurities in crystalline lattice structures where non-radiative relaxation between the adjacent close levels generally involves several phonons. It is shown that Kiel’s equation, commonly used to interpret such data, represents a limiting case of the present treatment. Tunnel-eflect Theory. A recent development in the treatment of radiationless transitions since this was last reported in Volume 7 has been the successful application of the tunnel-effect theory308 to a number of important photochemical problems, namely the quenching of excited aromatic molecules by paramagnetic species,3oBradiationless processes in aromatic excimers and e x c i p l e x e ~ ,and ~ ~ ~ excited-state hydrogen-abstraction r e a c t i o n ~ . ~ These ~~-~~~ papers follow earlier applications of the theory to non-radiative processes in aromatic compounds 308 and to the cis-trans olefin 314 and benzene 316 photoisomerizations. The approach is of course empirical, but its value in the interpretation of a wide variety of phenomena has been clearly demonstrated. It is, perhaps, worth noting the strong current interest in the use of the quantum mechanical tunnelling concept to interpret the ion-electron recombination processes which occur in condensed-phase systems following photolysis and radioly~is.~~~ The tunnel-effect theory of radiationless transitions calculates non-adiabatic Franck-Condon factors between two electronic states by regarding the radiationless transition between them as a quantum mechanical tunnelling through the potential energy barrier formed by the initial- and final-state potential energy curves represented in Figure 11. Standard tunnelling expressions yield the following result after making suitable approximations : kNR= r exp {- (27~/h)[2p(D - E,)# A x ] } (4) where k,, the rate constant for conversion between the two states is related to v the frequency of vibration in the initial electronic state, p is the reduced mass of the oscillator, D is the crossing point of the two energy curves above the initial so8

910

sla 114

316

316

V. Yakhot, Chem. Phys., 1976,14,441. V. Yakhot, Chern. Phys. Letters, 1976, 40, 304. D. Knittel, H. Raizdadeh, H. P. Lin, and S. H. Lin, J.C.S. Furuday ZI, 1977, 73, 120. S. J. Formosinho, J.C.S. Faraday ZI, 1974, 70, 605. S. J. Formosinho, Mol. Photochem., 1976, 7 , 13. S. J. Formosinho, Mol. Photochem., 1976, 7, 41. S. J. Formosinho, J.C.S. Faraday ZZ, 1976, 72, 1313. S. J. Formosinho, J.C.S. Faraday 11, 1976,72, 1332. H. D. Burrows and S. J. Formosinho, J.C.S. Furaday II, 1977, 73, 201. S. J. Formosinho, in ‘Excited States of Biological Molecules’, ed. J. B. Birks, Wiley, Chichester, 1975, p. 555. S. J. Formosinho, Mol. Photochem., 1974, 6,409. For example see the review by M. J. Pilling and S. A. Rice in ‘Progress in Reaction Kinetics’, ed. R. B. Cundall and K. R. Jennings, in press.

42

Photochemistry

electronic state, i.e. the barrier height, E, is the vibrational energy in the initial electronic state and Ax is the distance between the potential energy curves of the two electronic states at the energy of the initial state, i.e. the barrier width.

I

0

I

I

0.08

I

0.04

I

0

- 0.04 I

x/nm Figure 11 Potential energy curves of C=O* and CH oscillators in reactants and C--0 and OH oscillators in products (Reproduced from J.C.S. Faraday 11, 1976, 72, 1313)

The electronic part of the transition is accounted for by a transmission factor, or forbidden factor, which acts as a multiplier to the expression for the vibrational part of the transition. For internal conversion this factor is unity except in cases where geometrical considerations may have to be included. For an intersystem crossing process, the forbidden factor is treated as an empirical parameter. To obtain numerical values for kNR,the characteristics of the appropriate oscillators are required, i.e. their frequencies, dissociation energies, and anharmonicities. The important barrier width parameter, Ax, may be found from a construction of the two potential energy surfaces, for which it is necessary to know the energy difference between their two minima and the displacement of the minima along an appropriate co-ordinate. For a polyatomic molecule this co-ordinate corresponds to that of the two-dimensional space of the particular vibrational mode which is wholly or partly responsible for the radiationless transition. The crossing point D is sensitive to the shapes and displacement of the potential energy curves. For the small displacements appropriate to intramolecular processes, the crossings may occur on either the attractive or the repulsive side of the potential energy curves. Both D and Ax must be considered when determining whether crossing is more favourable on the contraction of a bond or on the distension of a bond, i.e. on the repulsive or attractive sides, respectively, of the potentials. An interesting situation arises for large displacements when the repulsive part of the final-state potential energy surface crosses on the attractive

Spectroscopic and Theoretical Aspects 43 side of the initial-state potential energy surface. In such a case the rate will increase as the energy gap increases because D decreases, This is a case of an inverse energy gap rate law dependence not normally found for electronic relaxation processes of polyatomic molecules (see below). The application of these simple ideas has been quantitatively successful when applied to a variety of problems. The dominant role of the CH stretching mode (high v, low p) in the radiationless processes of aromatic compounds was clearly demonstrated in a paper which also discusses the effects of vibronic excitation, If the very rapid increase in the radiationless solvent, and temperature on kNR.a08 decay rate of benzene excited into its S1state with wavelengths below 240 nm is interpreted as a rapid change in Ax with vibrational energy this implies a crossing with a state with a low-slope potential energy surface and a continuum or quasi-continuum of levels, i.e. Channel 3 is a radiationless transition to a weakly bonding state.*16 Interesting conclusions were also reached in a study of exciplexes and excimers which revealed the importance of an internal conversion contrary to earlier suggestions. A study of the quenching of process Sl3 So,81o the excited states of aromatic molecules by paramagnetic species which regards the rate-determining step as the non-radiative transition of a quenching exciplex reveals that in the oxygen quenching of aromatic molecules, and the quenching of naphthalene by transition metal ions, the quenching is dominated by the C-H stretching vibrations. For the NO quenching of triplet states the C-H modes are again dominant for triplet energies below 14 500 cm-l but above this NO vibrations are In a stimulating paper the tunnel-effect theory is applied to the hydrogen abstraction reactions of excited carbonyl compounds, and the result is a unifying treatment with many new insights into the molecular factors which are important in these processes.s11 The theoretical approach is to regard the reaction as the non-radiative transition between an initial state which comprises the in-phase stretching motions of two oscillators, >C=O and H-C-, and a final state which is made up of two coupled oscillators, )&Oand -0-H. It is an example of the large-displacement case mentioned above. The approach in this paper was extended to hydrogen abstraction reactions by the excited uranyl (UOaa+)ion *lS and by thioketones, quinones, aza-aromatics, olefins, and azobenzenes.818 Intersystem Crossing. A new expression for the intersystem crossing rate, Sl-+ T,, in aromatic hydrocarbons has been developed using the Green's function technique.817 This represents an extension of earlier treatments because of the inclusion of cross-terms between the spin-orbit coupling and the nonadiabatically induced spin-orbit coupling which are found to make a significant contribution to the rate constant under certain conditions. The mechanism leads to the classification of promoting modes according to the perturbation involved. The first kind are governed by spin-orbit coupling matrix elements and the second kind by the nuclear momentum operator. Applying the results of an analysis of a three-electronic-state model to the benzene Sl(lB,,,) + T intersystem crossings indicates that the C-H out-of-plane vibration ol0 acts as a dominant promoting mode for both the direct, Sl3 T,, and the indirect, S1+ Tg-+ TI, *17

Y. Fujimura, N. Shimakura, and T. Nakajima, J. Chem. Phys., 1977,66,3530. 3

44

Photochemistry

mechanisms. Applied to the intersystem crossing processes in benzophenone and 9,lO-diazaphenanthrene, molecules where the El-Sayed rule is not obeyed, the rate constants can be explained in terms of indirect and mixed decay processes.31s The El-Sayed rule states that when both the lowest excited singlet and triplet states are of: the nn* (or mr*)type, the presence of a triplet TT* (or nn*) state which also lies below the singlet state will result in a significantly higher intersystem crossing rate than in a system where such a state is not present. The model used in this investigation involved four electronic states, the initial singlet and the final and two lowest triplet states, T,, T,, and T3 respectively, An analysis of the matrix elements for T,(nn*) -+ So radiationless transitions in aromatic molecules having non7bonding electrons has led to the general conclusions that the intersystem crossing induced by non-adiabatic coupling may not be much smaller than the one governed by adiabatic coupling for molecules with relatively small nrr*-rrn* electronic gaps, and that, symmetry apart, mechanisms involving vibronic coupling in the triplet manifold are much more important than those involving vibronic coupling in the singlet manifold.31n A number of nitrogen heterocyclic and aromatic carbonyl compounds are discussed in the paper. Marked effects on the nn* phosphorescence spectra of pyrazine following methyl substitution are regarded as good evidence for the so-called proximity effect on the radiationless transition rate.320 The 3nn*-snn* energy gap decreases on methylation, leading to an increasing interaction between the two states. An investigation of the luminescence from 9,lO-phenanthroline indicates that the failure to observe phosphorescence from this molecule in hydrocarbon solvents is due to a very rapid Tl -+ So process due again to the 37rn*,3n7r* states being very close in energy.321 The phosphorescence emission is seen in hydroxylic solvents because the hydrogen bonding shifts the energy levels so as to reduce their interaction. Studies of intersystem crossing processes in benzene and perdeuteriated benzene v a p o u r ~low-temperature ,~~~ n-alkane solutions of tetracene, perdeuteriobenzene, 1,2-benzanthracene, and 1,2,3,4-dibenzanthra~ene,~~~ 1,2,4,5-tetrachlorobenzene in a durene host benzoylacetone and its EDA and in aromatic amines 327 have been published. Internal Conversion. A study of gaseous CF3N0 suggests that this molecule is suitable for testing theories of internal conversion.328 Large variations in the collision-free fluorescence lifetimes as a function of excess energy in the first excited state were observed and there are two components in the fluorescence decay. A very fast loss channel, which is probably photodissociation, competes 3269

318 319 320

sal 82z

823 a24

s2s *28 337

328

N. Simakura, Y. Fujimura, and T. Nakajima, Chem. Phys., 1977, 19, 155. N. Kanamaru and E. C. Lim, J. Chem. Phys., 1976, 65,4055. S. L. Madej, S. Okajima, and E. C. Lim, J. Chem. Phys., 1976, 65, 1219. C. T. Lin and J. A. Stikeleather, Chem. Phvs. Leftem, 1976. 38. 561. S. J. Formosinho and A. M. Da Silva, J.C.S. Faruduy ZZ, 1976,72, 2044. R. H. Clarke and H. A. Frank, J. Chem. Phys., 1976,65, 39. W. M. Pitts and M.A..El-Sayed, Chem. Phys., 1977, 19,289. M. Mercantonatos, Inorg. Chim. Acta, 1977, 21,75. H. J. Haink and J. R. Huber, Chem. Phys. Letters, 1976,44, 117. H. J. Haink and J. R. Huber, J. Mol. Spectroscopy, 1976, 60,31. K. G. Spears and L. Hoffland, J. Chern. Phys., 1977,66,1755.

Spectroscopic and Theoretical Aspects

45

with internal conversion at large excess energies. The fluorescence quantum yields of several nucleic acid components in ethylene glycol-H,O solution at 160 K are considerably higher (factors of 2-10) in the 0-0 region than they are near the first absorption maximum.32eThis effect is attributed to a very rapid internal conversion process which is able to compete with vibrational relaxation. The strongly temperature-dependent fluorescence properties of isoquinoline in solution may be accounted for by a thermally activated depopulation of S1(m*)via a higher state which is possibly the S2(nrr*) ~tafe.~3O A detailed study of the internal conversion processes of p-xylene in solution has been made by monitoring S3, S2, and S, popultttions by their fluorescence emi$sions S, -+ So, S2 -+ So,and S1-+ The inductive effect of polar substituents on absorption band positions and intensities is not very marked and attempts to correlate any changes with Hammett or Taft constants are not generally-successful. However, a surprisingly. good correlation is found between logk and Ca* for the substituted benzenes where k is the S, non-radiative decay constant and C;a* the sum of the six substituent Taft constants ;a*.332 A simple model predicts good correlations also between a* and the activation energy of non-radiative S1decay and with the enetgy of the 0-0 phosphorescence band. Vibrational Relaxation. Picosecond laser techniques have made it possible to measure directly the vibrational relaxation of polyatomic molecules even when they are in dilute solutions. Using an elegant three-step method, the vibrational relaxation time in the ground electronic state of rhodamine 6G was measured to ~ ~ first step, a powerful be 4 ps in ethanol solution and 3 ps in l - b ~ t a n o l .In~ the pulse populates excited vibrational levels in S1which rapidly decay to the lowest vibrational level of S,. A second ‘preparing’ pulse stimulates emission from the lowest vibrational level of S1which is formed rapidly following the first pulse to a level in the vibrational manifold of So. A delayed third pulse probes the subsequent absorption out of the repleted lowest vibrational level of So. The technique has a number of advantages over previous methods which used spontaneous anti-Stokes Raman scattering to monitor the relaxation process. Unfortunately this scattering process has a very small cross-section. Another technique which has been developed recently measures the vibrational relaxation in a ground electronic state by monitoring the fluorescence from Sl as a function of the time delay between two picosecond pulses of light.834 The first of these is an infrared pulse which selectively excites a vibrational mode in So and this is followed by a second pulse which has just sufficient energy to populate S1from the excited ground-state vibrational level. The technique has also been used to investigate the relaxation time of the first electronic state of azulene, S,, using S2 fluorescence to monitor the relaxation.335 (This study showed that even faster techniques were necessary to observe this particular s2n 330

ssa 3s4 3s5

R. W. Wilson, J. P. Morgan, and P. R. Callis, Chkm. Phys. Letters, 1975, 36, 618. R. J. Huber, M. Mahaney, and J. V. Morris, Chem. Phys.,1976, 16, 329. T. A. Gregory and S. Lipsky, J. Chem. Phys., 1976,65,296. L. A. King, J.C.S., Perkin II, 1976, 14, 1725. D. Ricard and J. Ducuing, A . Chem. Phys., 1975, 62, 3616. A. Laubereau, A. Seilmeier, and W, Kaiser, Chem. Phys. Letters, 1975, 36, 232. J. P. Heritage and A, Penzkofer, Chem. Phys. Letters, 1976, 44, 76.

46

Photochemistry

relaxation and this has recently been achieved using the sub-picosecond pulses from a mode-locked CW dye laser: see ref. 369.) The same research group have reported measurements of vibrational relaxation in rhodamine 6G and rhodamine B 336 and in ethanol and methyl iodide.337 Vibrational relaxation in gas-phase rhodamine B is discussed in a paper which attempts to give a unified account of recent results on electronic and vibrational The Singlet State.-Resonance Fluorescence. There is considerable current interest in theoretical and experimental aspects of the problem of the interaction of light with resonant atomic systems. These studies are relevant both to atomic isotope separation, which is reported on elsewhere in this Chapter, and to theoretical physics where they provide good tests of quantum electrodynamical theory. A new theory of two-level atom resonance fluorescence within the framework of quantum electrodynamics (QED) has been published which avoids many of the important assumptions common to other theoretical The theory reveals, amongst other things, that certain steady-state properties of the fluorescence will be more valuable in testing QED than transient effects. The authors applied similar methods in a study of resonance fluorescence from a twoThe bandwidth of level atom excited by a laser beam of finite excitation is found to affect significantly the fluorescence spectrum. Another paper explores the Markovian and non-Markovian behaviour of two-level atom fluorescence.s41 A non-exponential fluorescence decay law is evidence for the non-Markovian character of the Bloch equations. An analysis of the problem of a three-level atom in an intense monochromatic field shows that in this case the fluorescence spectrum contains five lines and not three as in the case of the two-level atom.342 An atom suffers a recoil when it absorbs or emits a photon and this results in frequency shifting in spectral lineshapes. The effect of recoil on the resonance fluorescence spectrum of a two-level system excited with monochromatic radiation has been examined in the limit of a weak external field, i.e. + I E I 4 lo6 V m-1.343 The excitation of atomic fluorescence in alkali-metal vapours with radiation whose frequency is as much as 2000cm-l from the resonance frequency can be explained by assuming absorption takes place during pair collisions.844Theoretical treatments of resonance fluorescence in intense coherent resonant fields,34s scattered from collimated atomic beams,346for atoms with Zeeman degeneracies of the lower and upper higher-order intensity 336

337 338

33D 340 841

342

343 344

*M

346 347

A. Penzkofer, W. Falkenstein, and W. Kaiser, Chem. Phys. Letters, 1976,44, 82. K. Spanner, A. Laubereau, and W. Kaiser, Chem. Phys. Letters, 1976,44,88. A. Tramer and C. Tric, Ber. Bunsengesellschaftphys. Chem., 1977,81, 209. H.J. Kimble and L. Mandel, Phys. Rev. (A), 1976, 13, 2123. H. J. Kimble and L. Mandel, Phys. Rev. (A), 1977, 15, 689. K. Wodkiewicz and J. H. Eberly, Ann. Phys. (New York), 1976, 101, 574. B. Sobolewska, Opt. Comm., 1976, 19, 185. J. F. Lam and P. R. Berman, Phys. Rev. (A), 1976, 14, 1683. A. M.Bonch-Bruevich, S. G. Przhibel'skii, A. A. Federov, and V. V. Khromov, Zhur. eksp. reor. Fiz., 1976,71, 1733. R. Bonifacio and L. A. Lugiato, Opt. Comm., 1976, 19, 172. B. Renaud, R. M. Whitley, and C. R. Stroud, jun., J. Phys. (B), 1977, 10, 19. C. Cohen-Tannoudji and S. Reynaud, J. Phys. (Paris), Letters, 1977,38, L173.

Spectroscopic and Theoreticai Aspects

47

correlation function time f a c t o r i ~ a t i o n ,and ~ ~ ~from J = 4 transitions s49 have been published. It has been proposed that knowledge of the effects of magnetic fields on the polarization of resonance fluorescence can be applied to problems of astrophysical i n t e r e ~ t . ~ ~ * A number of studies have dealt with aspects of resonance fluorescence under conditions where there is appreciable reabsorption, or imprisonment, of the radiation.s61, The characteristics of the super-radiance from laser-pumped HF was found to be in poor agreement with a recent theoretical treatment.s63 Quantum beats have been observed in the superfluorescencefrom the 7P-7s transition in atomic Cs which occurs at 3 pm.ss4 The beats are observed in both an applied field and in zero field and the frequencies correspond not only to initial-level splittings but also to combinations of initial- and final-level splittings illustrating the fundamental difference of the observed beats from single-atom quantum beats. Fluorescence. The following papers are selected from a large number on the fluorescence properties of complex polyatomic molecules. The fluorescence spectrum of pyrene in a monocrystalline matrix of n-heptane at 8 K has been measured,s6 a further example of the potential spectroscopic value of this particular matrix. The polarized spectra led to vibrational assignments in good agreement with those found for pyrene in biphenyl and fluorene single crystals. The near-infrared electrochemiluminescence from solutions of 9,lO-diphenylanthracene has been attributed to &-c fluorescence.m6 The Marcus theory of electron transfer reactions predicts a high rate of formation of second triplet states, T,(k = 10s-lO1o 1 mol-f s-l), from the reaction of the positive and negative aromatic ions. Fluorescence lifetimes and quantum yields have been measured for the [4]- to [9]-*s7and [lo]- to [14]-helicenes and the bridged compound 3,15ethan0heptahelicene.=~ The fluorescence from these molecules is from the lLb states which are not seen in absorption. A regular trend in the value of the fluorescence parameters is observed for helicenes whose number of rings, N, is odd and a different regular trend if N is even. The sharp well-resolved emission spectra of the benzyl radical and several deuteriated derivativeshave been analysed followingtheir production by photolysis of the parent toluenes in polycrystalline methylcyclohexane at 77 K.36g The radiative lifetime of about 1 ps for the benzyl radical indicates a small oscillator strength for the first electronic Fluorescence from the diphenyl 1148

a49

as 0 861 a61 SbS

8b4 S66

866

ab7 368

S69

ado

G. S. Agarwal, Phys. Rev. (A), 1977,15,814. D.Polder and M. F. H. Schuurmans, Phys. Rev. (A), 1976, 14, 1468. A. A. Ruzmaikin, Asfron. Zhur., 1976,53, 550. M. J. Boxall, C. J. Chapman, and R. P. Wayne, J. Phofochem., 1975,4,281. (a) L. F. Phillips, J. Phofochem., 1975,4,407;(b)ibid., 1976,5,241 ;(c) ibid., 1976, 5, 277. R. Friedberg and B. Coffey, Phys. Reo. (A), 1976,13, 1645. Q. H.F. Vrehen, H. M. J. Hikspoors, and H. M. Gibbs, Phys. Rev. Letters, 1977, 38,764. Vo Dinh Tuan, U. P. Wild, M. Lamotte, and A. M. Merle, Chem.Phys. Letters, 1976, 39, 118. N. Periasamy and K. S. V. Santhanam, Chem. Phys. Letters, 1976,39,265. M. Sapir and E. Vander Donckt, Chem. Phys. Leffers,1975,36, 108. J. B. Birks, D. J. S. Birch, E. Cordemans, and E. Vander Donckt, Chem. Phys. Lefters, 1976, 43, 33. V. J. Momson and J. D. Laposa, Spectrochtrn. A d a , 1976,32A,1207. T. Okamura and 1. Tanaka, J. Phys. Chem., 1975,79,2728.

48 Photochemistry ketyl radical has been observed followifig laser excitation of solutions of benzophenone in hydrogen-containing The ketyl radical is formed by H-atom abstraction from the solvent by triplet benzophenone. If this is rapid, as in 2-methyltetrahydrofuran, the radical is itself excited during the laser pulse and radical fluorescence is seen. If radical fluorescence continues to build up after the laser pulse, as it does in hydrocarbon solvents, another mechanism operates to give excited ketyl radicals. It is proposed that the mechanism involves electronic energy transfer from the benzophenone triplet to the radical groundstate doublet, the first report of such a process. Two other investigations of this system have been r e p ~ r t e d363 .~~~~ Styrene and its simple derivatives have weak, structured, first absorption bands, lLb (lB2-+ lAl in C,,symmetry), and much stronger, almost featureless, second absorption bands, lLU (lAl lAl in CZvsymmetry). Excitation of styrene vapour under ‘colhion-free’ conditions in the lLb band results in single-exponential fluorescence decay but excitation in the lL, band results in a non-exponential decay. Resolution of the complex decay into two exponential decays gave a fast component whose lifetime and spectrum were clearly related to the ‘ L b emission seen on direct’excitation to that state and whose appearance is evidence of a lL, -+ ’Lb internal conversion process. The longer-lived emission (45 ns for trans-l-phenylpropene whose fast component has T z 2ns) has a spectrum which is to the high-energy side of the spectrum from the short-lived emission and clearly does not originate from the lLb state. It has been proposed364and subsequently confirmed 305 that the long-lived emission arises from a ‘twisted’ excited singlet state reached from the lL, singlet state by a change in molecular geometry, probably a rotation about the olefinic double bond. The lifetime of the weak fluorescence of s-tetrazine was found to be 450 -t55 ps and this gives a value of the fluorescence quantum yield of 1.1 x when combined with a radiative lifetime of 4.0 x lO-’s derived from the absorption Intersystem crossing in this molecule is negligible and the dominant decay channel, with a quantum yield approaching unity, is photochemical. A detailed analysis of the laser-excited fluorescence from isoquinoline vapour excited in the 3 10 nm absorption system has demonstrated that this transition arises from the near-resonance vibronic coupling between two electronic states; lA”(nn*),1A’(~7r*).3s7 Non-exponential fluorescence decay has been observed in dilute solutions of poly(p-meth~xyphenylacetylene).~~~ When viewed as a sum of exponential decays, the decay is consistent with a model in which backbone rotation maintains a distribution of conjugated sequence lengths in dynamic flux. --f

361

362 3f13 3f14

K. Razi Naqvi and U. P. Wild, Chem. Phys. Letters, 1976, 41, 570. M. R. Topp, Chem. Phys. Letters, 1976, 39,423. B. W. Hodgson, J. P. Keene, E. J. Land, and A. J. Swallow, J . Chem. Phys., 1975,63, 3671. R. P. Steer, M. D. Swords, P. M. Crosby, D. Phillips, and K. Salisbury, Chem. Phys. Letters, 1976, 43, 461.

366

386

387

368

K. P. Ghiggino, K. Hara, G. R. Grant, D. Phillips, K. Salisbury, R. P. Steer, and M. D. Swords, J.C.S., Perkin II, 1978, 88. R. M. Hochstrasser, D. S. King, and A. C. Nelson, Chern. Phys. Lerters, 1976, 42, 8. G. Gischer and A. E. W. Knight, Chem. Phys., 1976,17, 327. A. M. North, D. A. Ross, J. E. Guillet, and T. L. Nemzek, J. Chern. Res. (S), 1977, 49.

Spectroscopic and Theoretical Aspects

49

The first dear measurement of the lifetime of the first singlet state of azulene has recently been reported,8@gPrevious attempts to measure the lifetime had indicated a value of a fm. picoseconds. The method involved the division of sub-picosecond dye laser pulses which reach the sample after introducing a variable delay between them. The-ifirst pulse populates S1and the second pulse populates S2 from S, and iesulbs in tan easily detectable S,-So fluorescence. An S1lifetime of 1.9 k 0.2 ps. was determined in all four of the following solvents: cyclohexane, benzene, Wo&obaaene, and chlorobenzene, f.e. .no clear heavyatom effect. Picosecond laser tehniques have been used to measure flaorescence lifetimes in 3,3’-diethyloxadicarbocyanine iodide (DODCI),S70cryptocyanine, and 3,3’-diethyl-2,2’-thidricarbocyanheiodide (DTTC),373and the I aggregates of pseudois~cyanine.~~~ An interesting study of the photophysicat properties of fluorinated.acetones in the vapour phase has been p ~ b l i s h d .The ~ ~ radiative, ~ kR, and non-radiative, ~ N R ,decay constants for the first excited singlet states of these molecules are given in Table 4. The depehdence of k~ on the ground-state promoting mode Table 4 Rate constants for radiative and non-radiative decay Compound 7/ns 2.4 A 2.3 -t- 1 MFA 2.2 -4 1 DFA TrFA 4.0 15.0 TeFA PFA 32.5 84 a HFA

@F

0.0012 0.01

-0.001 Q.009 0.004 0.0075 0.0185

k R / 1 s-1 ~ kNR/107 S-1 (k&ss/l~f‘ s-1 5.0 42 1 .o 4.3 43 1 .o 4.5 45 0.9 2.3 25.0 0.6 2.7 6.6 1.1 2.3 3.1 0.8 2.2 1.2 0.6

--

~R/(~R)sB

5.0 4.3 5.0 3.8 2.5 2.9 3.7

frequency vm was stronger than the linear form predicted by a Herzberg-Teller coupling scheme, and approximated to the cubic dependence expected from a Born-Oppenheimer coupling scheme. Anti-Stokes Fluorescence. A detaj&l analysis of the Stokes and anti-Stokes fluorescence spectra of 1,€2-beozoperylpein n-heptane solution from 90 to +90 “C has been p~blished.4~4The flumqcence spectrum is resolved into three types of bands. The normal or Stokes fluorescence comprises five bands c o n e spmding to the Sq -+ S:* Si,Si, S:, $; transitions. Three bands at higher energy corliespond to the hdt fluorescence bands, St, Sq, S: -+ S:, and finally an mti-Kasha (emissions that do not originate from the lowest electronic excited state of a given multiplicity) fluorescence band corresponding to the S! + S: transition. (S% corresponds to a singlet electronic state SE in a vibrational level tr.) It is proposed that the So1 -+ S r (rn # 0), Sf-+ S: (n # 0), and S! -+ So0 transitions are induced by vibronic coupling to allowed S r -+ Sr,

-

870 871

an 373 374

E. P. Ippen, C. V. Shank, and R. L. Woerner, Chem. Phys. Letters, 1977, 46,20. J. C. Mialocq, A. W. Boyd, J. Jaraudias, and J. Sutton, Chem. Phys. Letters, 1976, 37, 236. H. Tashiro and T. Yajima, Chem. Phys. Letters, 1976, 42, 553. F. Fink, E. Klose, K. Teuchner, and S. Daehne, Chem. Phys. Letters, 1977, 45, 548. J. Metcalfe and D. Phillips, J.C.S. Furuday fZ, 1976, 72, 1574. J. B. Birks, C. E. Easterly, and L. G . Christophorou, J. Chem. Phys., 1977,66,4231.

Photochemistry S $ 3 Sg, and Sk -+ St transitions, respectively. The Sa fluorescence quantum efficiency is found to be 0.005 k 0.001, and its lifetime 21 2 4 ps. A method which involves polarized absorption and emission measurements has been proposed to resolve the &-So emission from overlapping S2-So, Sl-So emissions.s7b A weak emission from biphenylene has been attributed to S2(lLb)+ ,!?#A) The emission was dependent on the excitation wavelength, indicating that part of the emission is from unrelaxed states. The S2lifetime is estimated to be 4 ps. An investigation of the Sa anti-Kasha emission from 3,4-benzpyrene supports the statement that in the case of intermediate strong coupling the relative S8 emission yield of large isolated molecules is proportional to the ratio of the density of states in the Sa and Slmanifolds.377 The S2-So fluorescences of xanthione and thioxanthione compete with a rapid temperature-dependent Sa -+ Sl internal conversion process.378 S2-S0 fluorescences from [18]annulene and monofluoro[ 18lannulene have been observed in a 3-methylpentane glass at 4 K S 7 O This is the first reported case of strong Sa-So emission from an alternant hydrocarbon. The large Sa-Si energy gap of 9000 cm-l and the large oscillator strength observed for the S2-So transition both suggest the possibility of observing &-So fluorescence and possibly both the S2-S, and Sl-So emissions. Both the emissions are seen in the monofluoro[18]annulene but only the S2-So emission could be found in [18]annulene. The weak Sa-SI emission of azulene in a Shpolskii matrix has been reported.s80 The quantum yield of this emission in a methylcyclohexane glass at 77 K is only 4 x assuming an S,-So quantum yield of 0.03,and indicates a vibronically induced transition. Fluorescence from the first two excited singlet states of fluoranthene has been observed.S81 Within each of the emission bands considerable variations in the emission lifetimes were found, and this was attributed to emission from vibrationally unrelaxed excited states. Short-wavelength emissions from several polymethine dye molecules have been associated with S2-So emissions and not with the emissions from photodegradation Similarly, short-wavelength fluorescence emissions from upper excited singlet states have been observed for tetracene a83 and t h i o p h o ~ g e n e . ~ ~ ~ When molecules are promoted to excited states having an excess of vibrational energy there is competition between an intramolecular energy redistribution and electronic relaxation. If the amount of excess vibrational energy is small, electronic relaxation is the more rapid of the two processes and the effects of single vibronic levels can be observed. As the excess vibrational energy is increased, the vibrational redistribution is expected to become more rapid and 50

N

376

370

a77

37* s79

380

381 38a

384

L. Marguiles and A. Yogev, Chem. Phys. Letters, 1976, 37,291. H. Shizuka, T. Ogiwara, S. Cho, and T. Morita, Chem. Phys. Letters, 1976,42, 311. P. A. M. Van den Bogaardt, R. P. H. Rettschnick, and J. D. W. van Voorst, Chem. Phys. Letters, 1976, 41,270. M. Mahaney and R. J. Huber, Chem. Phys., 1975,9,371. U. P. Wild, H. J. Griesser, Vo Dinh Tuan, and J. F. M. Oth, Chem. Phys. Letters, 1976,41, 450. G. D. Gillespie and E. C. Lim,J. Chem. Phys., 1976,65,4314. D. L. Philen and R. M. Hedges, Chem. Phys. Letters, 1976,43, 358. 0. V. Przhonskaya and E. A. Tikhonov, Izuest. Akad. Nauk S.S.S.R., Ser. $z., 1975, 39, 2275. B. Nickel and G. Roden, Ber. Bunsengesellschaft phys. Chem., 1977, 81, 281. Y . Oka, A. R. Knight, and R. P. Steer, J. Chem. Phys., 1975,63,2414.

Spectroscopic and Theoretical Aspects

51

to result in a random distribution of vibrational level populations thus obscuring the effects of single vibronic levels. Evidence for such a competition in aromatic molecules has been found in two investigations which studied the excitation energy dependence of fluorescence lifetimes and quantum yields in dilute vapours of fluorene and /3-naphthylamine,88sand tetracene and p e n t a ~ e n e .The ~ ~ ~results for [lH12]-and [aHla]-tetraceneare shown in Figure 12. The initial increase in the radiationless decay rate is thought to be due to the excitation of a good accepting mode, the C-C stretching vibration, for S, + So internal conversion. The

I

I

I

30

40

50

Ecxc (IO'cm'')

Figure 12 Total decay rate (l/~g)of fluorescence of ( 0 ) [lH,,]-te~raceneand ( 0 ) [2H14]-tetracenevapours as a function of excitation energy (Reproduced by permission from Chem. Phys. Letters, 1976, 37, 403)

dependence tails off at higher excess energies because the vibrational redistribution is faster than electronic relaxation. The effect of different excitation energies on the relative amount of S1*hot fluorescence in the total Slstate fluorescence emission of 3,4-benzpyrene vapour can be explained in terms of a randomly distributed excess vibrational energy.387 A brief review of hot luminescence from impurity centres and other systems has appeared.888 A vibrationally unrelaxed emission has been observed following selective excitation of higher vibrational levels in matrix-isolated difluoromethylene38g and sodium nitrite.8B0 The reverse of the above processes, a vibrationally (or multiphonon) assisted electronic excitation, has been observed for triply ionized rare-earth ions in a number of Fluorescence Quantum Yields. The measurement of absolute quantum yields for luminescence by optoacoustic methods is an important development, as is the application of reIated methods to the study of weak absorption spectra. In these C.Huang, J. C. Hsieh, and E. C. Lim, Chem.Phys. Letters, 1976,37, 349.

a*8 a89

S . Okajima and E. C. Lim, Chern. Phys. Letters, 1976,37,403. P. A. M. Van den Bogaardt, R. P. H. Rettschnick, and J. I). W. van Voorst, Chem. Phys. Letters, 1976,43. 194. K. Rebane and P. Saari, J . Luminescence, 1976,12/13,23. V. E. Bondybey, J. MoI. Spectroscopy, 1976, 63, 164. J. Aaviksoo, A. Anialg, P. Saari, and T. Soovik, Fir. Tverd. Tela (Leningrad), 1977,19, 826. F. Auzel, J. Luminescence, 1976, 12/13, 715.

52

Photochemistry

methods all non-radiative decay processes which eventually lead to an increase in translational energy can be measured acoustically.3B2The measurements are relatively simple and accurate and involve the comparison of signals from the luminescent system under investigation and from a non-emitting, equally strongly absorbing system. Clearly the nature of the reference is sharply defined, i.e. it does not emit luminescence, in contrast to the standard in the complementary conventional studies of luminescence quantum yields where it is in principle as indeterminable as that of the sample itself. A useful discussion of the related calorimetric techniques has been published which includes descriptions of a number of calorimeter^.^^^ The fluorescence of vibrationally relaxed benzene vapour is used as a reference in vapour-phase studies of fluorescence quantum yields. An optoacoustic determination of the absolute fluorescence quantum yield, using an oxygenquenching technique, gave a value &B = 0.16 f 0.02, in good agreement with the previously widely accepted value of 0.18 k 0.04.394In another determination a value of t$F = 0.19 k 0.02 at 254 nm was obtained using an interesting variation of the technique in which the benzene itself acted as an internal The technique, which in principle could be extended to any system where the fluorescence quantum yield varies over the absorption spectrum, involves the measurement of both the normal fluorescence excitation spectrum and the photoacoustic excitation spectrum. The complementary nature of these measurements (in the absence of any photochemistry) enables a unique scaling of the two spectra to be made which leads to values for the absolute quantum yield over the region of overlap of the two spectra. The errors in quantum yields calculated in this way will be very large in systems where the wavelength variation of the quantum yield is small and' in the limit when #*(A) is a constant the technique cannot be used at all, but it clearly still has wide potential. The important application of optoacoustic methods to similar measurements in liquid solutions has been 3Q7 A simple analysis of the concentration quenching effect which was previously regarded as a limitation to fluorescent dye-labelling analytical techniques has shown that it has the effect of lengthening bleaching lifetimes, and that the integrated fluorescence emission obtained on complete bleaching does not depend on the fluorescence quantum efficiency.398Application of a technique based on this principle has resulted in the detection of single reagent rn01ecules.~~~ Expressions have been derived for the luminescence quantum yields of primary light absorbers and their nearest similar molecule neighbours in isotropic solutions when long-range energy transfer occurs.4oo Methods used to determine luminescence efficiencies have been reviewed 401 and a discussion of possible reference systems for use in the determination of 398

398

ag4 396 396 3g7

8Ba 398 400 401

W. R. Harshbarger and M. B. Robin, Accounts Chem. Res., 1973, 6, 329. J. B. Callis, J. Res. Nat. Bur. Stand., Sect. A , 1976, 80, 413. L. M. Hall, T. F. Hunter, and M. G. Stock, Chem. Phys. Letters, 1976, 44, 145. M. G . Rockley, Chem. Phys. Letters, 1977, 50, 427. W. Lahmann and H. J. Ludewig, Chem. Phys. Letters, 1977, 45, 177. I. 0. Starobogatov, Optika i Spektroskopiya, 1977, 42, 304. T. Hirschfeld, Appl. Opt., 1976, 15, 3135. T. Hirschfeld, Appl. Opt., 1976,15,2965. C. Bojarski, Acta Phys. Pol. (A), 1977, 51, 589. A. Bril and A. W. De Jager-Veenis, J. Res. Nat. Bur. Stand., Sect. A , 1976, 80, 401

Spectrascopic and Theoretical Aspects

53

fluorescence quantum yields in concentrated solutions has been published.40a The influence of intramolecular movements on the fluorescence quantum yield of a molecule has been discussed and the solvent viscosity dependence of the #F of methinecyanine dyes is regarded as an example of such an effect.403A photon-photoion coincidence method has been used to determine directly the quantum yields and lifetimes of molecular ion fluorescence from a number of simple species.4o4 Fluorescence Polarization. The majority of fluorescence polarization measurements are made in studies of the rotational behaviour of molecules, particularly in biological systems. This section is restricted to more fundamental aspects of fiuorescence polarization. If it is desirable to remove the effects of polarization of fluorescence from a particular measurement, e.g. quantum yield measurements, it is known that if for vertically polarized excitation and right-angle viewing the emission polarizer is set at 54.75' (or 125.25') to the vertical this elimination is in fact achieved. It has now been shown that this arrangement is just as effective for all viewing angles and, furthermore, that for horizontally polarized excitation, viewing at 45" (or 135") with the emission polarizer at the same angle as above also removes the polarization Combining the two conditions removes the necessity of an excitation polarizer altogether. A general expression for the concentration depolarization of luminescence caused by the formation of non-luminescent dimers has been derived.406 The concentration depolarization of fluorescence in mixtures of two dyes whose absorption spectra do not overlap, trypaflavine and rhodamine B, was found to be in good agreement with theoretical A study of acridine orange cations dissolved in PVA films showed that the fluorescence and probably the phmphorescence conoentration depolarimtions resulted from singlet-singlet energy transfer in the concentration range 10-s-10-4 mol 1-1 408 A model caiculation of the change in the components of the fluorescence polarization resulting from orientation of a chromophore which is part of a rigid molecule has been applied in a study of the electric field orientation of the 2-hydroxy-4,4'-diamidinostilbene-DNA complex.4oeThe transfer of anisotropy from linearly polarized Cd atoms to Cs atoms has been demonstrated in a study of sensitized fluorescence in a mixture of their v a p o u r ~ and , ~ ~a~ study of the magnetic depolarization of the 552.3 nm fluorescence from MgI produced by electron impact excitation gave a value for the lifetime of the 3s4d lD2level of 41 ns.411 A study of sodium (32P4, 32P4)fluorescence polarization and intensity roa 409 404 406

406

407 408 '08

410

(11

J. B. Birks, J. Res. Nat. Bur. Stand., Sect. A , 1976, 80, 389. G. Calzaferri and H. Gugger, Helo. Chim. Acta, 1976, 59, 1969. J. H. D. Eland, M. Devoret, and S. Leach, Chem. Phys. Letters, 1976, 43, 97. K. D. Mielenz, E. D. Cehelnik, and R. L. McKenzie, J. Chem. Phys., 1976, 64, 370. C. Bojarski and R. Bujko, Acta Phys. Chem., 1976, 22, 25. H. Cherek, Bull. Acad. Pol. Sci., Sdr. Sci. Math., Astron., Phys., 1976, 24, 135. T. Komiyama and Y. Mori, Bull. Chem. SOC.Japan, 1976, 49, 864. G. Weill and J. Sturm, Biopolymers, 1975, 14, 2537. E. K. Kraulinya, A. P. Bryukhovetsku, and L. I. Kartasheva, Optika i Spektroskopiya, 1976, 41, 903. U. Teppner, L. H. Goebel, G. Von Oppen, and K. H. Martens, Asrron. Astrophys., 1976,52, 381.

54

Photochemistry

using pulsed dye laser excitation has been published412as have the polarized emission spectra of the pyrene 413 and anthracene and methyl colanthin excimem414 A remarkably high emissionan isotropy is found for room-temperature solutions of Michler’s ketone [4,4’-bis(dimethylamino)benzophenone] in ethanol.41s This is attributed to an unusually short fluorescent lifetime (less than 15 ps) and not to impurities or aggregation. The degree of fluorescence polarization of benzene, mesitylene, triphenylene, and triphenylbenzene in glassy solutions at 203 K rises sharply above the value of 1/7 predicted for molecules of their symmetry when excited at the red-edge of their absorption The values fall to approximately 1/7 or lower as A decreases. This is interpreted as evidence for a lower symmetry configuration of the molecules when they are excited in the region of the 0-0 band and agrees with the interpretation of some recent MIDP studies on benzene. The polarized fluorescence and phosphorescence emissions from benzenethiol and thioanisole have revealed some structure in the emission polarization, particularly for the pho~phorescence.~~~ An interesting photoselection study of dinucleotides and polynucleotides gives results which are in poor agreement with predictions of both vibronic exciton theory and Forster An excitation wavelength-dependent fluorescence polarization has been discovered in a study of the rotational motion of l-naphthylamine.41Q Circularly Polarized Fluorescence. A quantum electrodynamical theory of circularly polarized emission (CPE) and magnetic circularly polarized emission (MCPE) has been developed.420 The effects of photoselection and Brownian rotational motion were accounted for in the final expressions for the emission transition probabilities. CPE has been observed from solutions of achiral molecules in optically active 422 The fluorescein-a-phenylethylamine system gave relatively large emission anisotropy factors (gem)which were found to vary across the emission band and may be indicative of the different vibronic coupling mechanisms involved in the CPE and total emission processes.421No solvent-induced circular dichroism (CD) was observed in this system, indicating that the solute-solvent interactions must be significantly different in the ground and emitting states. These observations demonstrate their potential value in the study of solutesolvent interactions. Studies of the circularly polarized resonance fluorescence of gas-phase species following excitation with circularly polarized light can lead to an identification of rotational features in a spectrum and give detailed information about the transfer of angular momentum in collisional p r o c e ~ s e s424 . ~ ~The ~~ V. Kroop and W. Behmenburg, 2. Naturforsch., 1976, 31a, 707. A. S. Ghosh, D. Gupta, and S. Basu, J. Photochem., 1975, 4, 227. 414 D. Gupta and S . Basu, J. Photochem., 1976, 6 , 145. 416 W. Liptay and H. J. Schumann, Chem. Phys. Letters, 1976,39, 427. 410 E. Leroy and H. Lami, Chem. Phys. Letters, 1976,41, 373. 417 P. G. Russell, J. Phys. Chem., 1975, 79, 1347. 418 R. W. Wilson and P. R. Callis, J. Phys. Chem., 1976, 80, 2280. I1@B. Valeur and G. Weber, Chem. Phys. Letters, 1977,45, 140. 4 2 0 J. P. Riehl and F. S. Richardson, J. Chem. Phys., 1976,65,1011. 421 H. G. Brittain and F. S. Richardson, J. Phys. Chem., 1976, 80, 2590. 42a H. G. Brittain and F. S. Richardson, J. Amer. Chem. SOC., 1977,99, 65. 423 A. J. McCaffery, S. R. Jeyes, M. D. Rowe, and H. Kato, Ber. Bunsengesellschaftphys. Chem., 1977, 81,225. 424 H. Kato, R. Clark, and A. J. McCaffery, Mol. Phys., 1976, 31, 943. 41a

Spectroscopic and Theoretical Aspects

55

matrix element for the thallium 62P4-72Pt M1 transition has been measured experimentally by observing the circular polarization of the fluorescence which results from interference with the Stark-induced 62Pi-72P4 El A high-resolution study of fluorescence spectra following monochromatic excitation shows three components at high excitation intensities, just as predicted theoretically for excitation with circularly polarized light.426Studies of the total emission (TE) and CPE spectra of the optically active rare-earth chelate system, TMS (3-trifluoroacety~-~-camphorato)europ~um(xxx), which is an n.m.r. lanthanide shift reagent,427and of a number of europium(xx1) and terbium(xxx) complexes with carboxylic acids have been published. It has been pointed out that the coupling of nuclear spin and rotational angular momentum will reduce the polarization ratio of resonance fluorescence.42e This applies both to circularly and linearly polarized fluorescence. The MCPE, MCD, and U.V. spectra of meso-tetraphenylporphine and its dihydrochloride have been The fluorescence MCPE could be interpreted in terms of the theoretical treatment for MCD with an appropriate interchange of the initial and final states. The Triplet State.-The

roles of triplet states in photochemical electron transfer reactions 4*1 and in the photodegradation of dye molecules in dye lasers 432 have been discussed. The frequency of the purely electronic transition So-q as well as the Sl-Tl energy splitting have been determined for anthraquinone, Ph2C0, and 4-phenylbenzophenone from a study of their thermally activated delayed fluorescence.433 The triplet quantum yields #T for 17 aromatic compounds in cyclohexane or benzene room-temperature solutions have been determined using a general method based on relative actinometry with a standard compound of known +21.434 The triplet-triplet extinction coefficients, cPT, were found from energy transfer measurements. Values of eT-T for anthracene and 9-bromoanthracene have been determined by a ground-state depletion method.436 Studies of single vibronic levels in the triplet manifold of large polyatomic molecules are not easily made because of the long lifetimes of triplet states. These states are populated from the singlet, S1,manifold by an intersystem crossing process, Sl-T,,, which is in competition with the radiative and non-radiative internal conversion processes, Sl-So. A stochastic model for triplet yields which takes into account these competing processes has been published.436 The model includes the vibrational relaxation mechanisms for S1and T’ and the dependence 426

426 427 420

4ao 430 481 48a 488

484 4a6

486

S . Chu, E. D. Commins, and R. Conti, Phys. Letters (A), 1977, 60,96. W. Hartig, W. Rasmussen, and R. Schieder, 2. Phys. (A), 1976, 278,205. H. G. Brittain and F. S. Richardson, J . Amer. Chem. SOC.,1976,98, 5858. C. K. Luk and F. S. Richardson, J. Amer. Chem. SOC.,1975,97,6666. P. A. Madden, Chem. Phys. Letters, 1975,35, 521. J. C. Sutherland, G. D. Cimino, and J. T. Lowe, in ‘Excited States of Biological Molecules’, ed. J. B. Birks, Wiley, Chichester, 1975, p. 28. A. K. Chibisov, High Energy Chem., 1976,10, 1. E. H. Halpern, Proc. SOC.Photo-opt. Znstr. Eng., 1976, 82, 20. V. A. Tolkachev, Optika i Spektroskopiya 1975,38, 397. B. Amand and R. Bensasson, Chem. Phys. Letters, 1975,34,44. M . B. Ledger and G. A. Salmon, J.C.S. Faraday ZI, 1976,72, 883. K. H. Fung and K. F. Freed, Chem. Phys., 1976,14,13.

56

Photochemistry

of the radiative and non-radiative decay rates on the vibrational energy. The model gives qualitative agreement with triplet yield data for naphthalene using available data and the discussion of these results serves to illustrate the use and limitations of this approach. The triplet lifetime of carbonyl in poly(viny1 phenyl ketone) has been measured using a picosecond mode-locked train of pulses from a ruby laser. This intermediate is involved in the primary step of Norrish Type I1 p h ~ t o d e g r a d a t i o n . ~ ~ ~ The nature of the excited triplet states in the photolysis of 0-acyloximes has been investigated,438and the lifetimes of the second excited triplet states, T,, of naphthalene and 1-chloronaphthalene have been determined using an energytransfer method which is unusual in that it involves the use of an additional molecule as an excitation carrier: for example, the T, state of naphthalene was found to have a lifetime of 12 k 2 p s using endo-dicyclopentadiene as the acceptor and benzene as the carrier.439 The triplet states of the nitronaphthalene^,^^, anth hone,^^^ p y r i d a ~ i n e ,and ~ ~ ~bilirubin 443 have been investigated. The S,-S, fluorescence of anthracene and 9-methyl- and 9-phenyl-anthracenes following Tl-Tnexcitation has been studied and it is shown that the T, (n > 2) -+S2 intersystem crossing is much higher in the substituted The temperature dependence of the T,-T' fluorescence and the T,-S, phosphorescence of 9-bromo- and 9,lO-dibromo-anthracene indicates the presence of a thermally activated S,-T, intersystem crossing process.445 The triplet energies of the three quenchers azulene, /%carotene, and ferrocene have been determined using a graded series of triplet sensitizers to be 39 kcal mol-1 and, in the ranges, 21-25 kcal-l mol-l and 3 8 4 1 kcal-l mol-l respectively.446 Variable-angle electron-impact spectroscopy has detected low-lying triplet states in furan (&,-TI, 3.99 eV, S,-T,, 5.22 eV), thiophen (S,-T,, 3.75 eV, S,-T,, 4.62 eV), and pyrrole (S,-T,, 4.21 eV).447 Evidence has been obtained which attributes the lack of phosphorescence of styrenes to both an inefficient intersystem crossing process and an efficient non-radiative relaxation from The results of oxygen-enhanced measurements of the S,-T, absorption spectra of several styrenes studied in this investigation are shown in Table 5 , which also shows the marked constancy of the S,-T, energy splitting in these molecules (except for the severely twisted cis-t-butylstyrene) making possible the location of triplet states of the phenylalkenes from measurements of their S,-S, absorption spectra. J. Faure, J.-P. Fouassier, and D.-J. Lougnot, J . Photochem., 1976, 5, 13. M. Yoshida and H. Sakuragi, Chem. Letters, 1975, 11, 1125. 439 C. C. Ludwig and R. S. H. Liu, J. Amer. Chem. Soc., 1976, 98, 8093. p 4 0 C. Capellos and K. Suryanarayanan, Internat. J. Chem. Kinet., 1976, 8, 541. 441 A. Garner and F. Wilkinson, J.C.S. Faraday II, 1976, 72, 1010. 442 C. J. Marzzacco, Bull. Chem. SOC. Japan, 1977, 50, 771. E. J. Land, Photochem. and Photobiol., 1976, 24, 475. S. Kobayashi, K. Kikuchi, and H. Kokubun, Chem. Phys. Letters, 1976, 42, 494. 446 G. D. Gillispie, and E. C. Lim, J. Chem. Phys., 1976, 65, 2022. p46 W. G. Herkstroeter, J. Amer. Chem. SOC.,1975, 97, 4161. u7 W. M. Flicker, 0. A. Mosher, and A. Kuppermann, Chem. Phys. Letters, 1976, 38, 489. Oo8 P. M. Crosby, J. M. Dyke, J. Metcalfe, A. J. Rest, K. Salisbury, 5. R. Sodeau, J.C.S., Perkin ZZ, 1977, 182. 437

438

Spectroscopic and Theoretical Aspects 57 The correlation of electrochemical redox potential data with the energies of electronic excited states has been examined for series of monocyclic aza-aromatics, polycyclic aza-aromatics and their N-oxides, condensed aromatic hydrocarbons, and benzene derivati~es.4~~ Triplet energies for members of the first three series were predicted from the linear free-energy relationships that were discovered. Intersystem crossing rates for a~ridine,~~O p h e n a ~ i n e ,and ~ ~ ~ anthrone and fluorenone 462 have been measured directly using picosecond laser techniques. A laser-excitation e.s.r. technique with which it is possible to measure directly the intersystem crossing rate of molecules with high fluorescence and low triplet yields has been applied to rhodamhe, fluorescein, and acridine

Table 5 So-Tl Oxygen-enhanced absorption spectra of some styrenes and 3-phenylpropene Compound Styrene

trans-1-Phenyipropene cis-l-Phenylpropene trans-1-Phenyl-3,3-dimethylbut-l -ene cis- 1-Phenyl-3,3-dimethyl but- l e n e 2-Phenylpropene 1-Phenylcyclopentene 3-Phen y lpropene (I

&-TI &9et/nm hlaxlnm 463.0 390.0 (463.0) 480.0 400.0 (478.0) 435.0 385.0 460.0 395.0 435.0 390.0 460.0 385.0 480.0 410.0 355.0 326.0

239

189

234 228 243 243 229

191.5 186 213.5 193.5 187

Measured in perfluoromethylcyclohexane.

Phosphorescence. An INDO/S method has been used to calculate the radiative triplet lifetimes of benzene and azaben~enes.4~~ The authors discuss the mechanisms for benzene phosphorescence in some detail and conclude that first-order vibronic coupling between the triplet manifolds of the *Bluand sElu states leads to the almost purely out-of-plane polarized phosphorescence, intensity being borrowed from the SElu-lAle intercombination. This conclusion agrees with earlier Zeeman and MIDP studies. Recent polarized phosphorescence studies, however, have led to different conclusions 466 (see below). The vibronic bandshape of the 2E, -+4A20 phoshorescence in 3dS complexes has been calculated.6* This doubly forbidden process occurs as a result of interactions of extended 3d3 orbitals of the central ion with the surrounding ligands. Oscillator strengths and lifetimes of the phosphorescence of a number of carbonyls have been calculated.467 460 461

dm

dm *&

467

R. 0. Loutfy and R. 0. Loutfy, Canad. J. Chem., 1976, 54, 1454. Y . Hirata and I. Tanaka, Chem. Phys. Letters, 1976, 41, 336. Y . Hirata and I. Tanaka, Chem. Phys. Letters, 1976, 43, 568. T. Kobayashi and S. Nagakura, Chem. Phys. Letters, 1976, 43, 429. M. Yamashita, A. Kuniyasu, and H. Kashiwagi, J. Chem. Phys., 1977, 66, 986. H. Ito and Y. J. Ihaya, Bull. Chem. SOC.Japan, 1976, 49, 940. T. J. Durnick and A. H.Kalantar, Chem. Phys., 1977,20,347. J. Kupka, Chem. Phys. Letters, 1976, 41, 114. P. Yvan, Mol. Phys., 1976,31,451; M. A. Winnik, S. N. Basu, C. K. Lee, and D. S. Saunders, J. Amer. Chem. SOC.,1976, 98, 2928.

Photochemistry

58

The conformations of n-alkyl esters of benzophenone-4-carboxylicacid are found to be similar in three different non-polar s01vents.~~*The evidence is based upon the insensitivity of the ratio of intramolecular quenching (found from variable chain length studies) to bimolecular quenching (from added quenchers) rate constants in the different solvents. The radiative decay constant of phosphorescence in these systems is found to be independent of chain length and The activity of out-of-plane vibrations of the aldehyde group of p-chlorobenzaldehyde is found to be very sensitive to the matrix.460 It is suggested that this results from concerted matrix and pseudo-Jahn-Teller perturbations which act together in acetophenone to give very strong activity and in opposition in p-xylene to give very weak activity. The aldehyde H-wag mode is, however, prominent in both solvents and is thought to gain activity through HerzbergTeller and intramolecular distortion effects. A detailed analysis of 9,lO-anthraquinone phosphorescence has been Vibronic activity results from the two low-lying triplet states, 3Au(n77*) and 3Blg(n7r*),interacting. Studies of the effect of temperature on phosphore~cence,~~~ the chemiluminescence of phosphorus under reduced pressure and oxygen-deficient and the phosphorescence of bilirubin 464 have appeared. A study of partial deuteriation on the phosphorescence lifetime of NNN’N’tetramethyl-p-phenylenediamine (TMPD) has shown that ring and methyl positions contribute independently and differently to 7p.466Ring substitution is 6-8 times more effective in reducing T~ than methyl substitution. The noticeable effect of deuteriation of methyl protons on T~ contrasts with the negligible effect observed for a similar substitution in toluene. Deuteriation of aromatic hydrocarbons has clearly been shown to affect the triplet radiative decay constants, kpT.466The ratio kpTH/kpTDwas found to be 1.20 k 0.07 for naphthalene, 1.39 k 0.06 for phenanthrene, and 0.98 k 0.04 for chrysene. It had almost become folklore that kpTHz kpTD z 0.03 s-l. A study of deuterium effects on the phosphorescence lifetime of phenanthrene has shown that the lifetime increases by 1% for deuteriation at the 2-positio11, about 6% for the 1-, 3-, and 4-positions and by 34% for the 9-positi0n.~~~ A very small positional variation of monodeuterium substitution on the T~ of benzonitrile has been The vibrational structure in the phosphorescence of p-chlorobenzaldehyde (PCB) is different in a slowly cooled [lH,,]methylcyclohexane (MCH) matrix from that in slowly-cooled [2H14]MCH.46gThe results indicate a high degree of distortability in the 37r7r* state of PCB which the authors attribute to vibronic interactions between closely spaced 37r7r*-3n77* states. M. A. Winnik and A. Lemire, Chem. Phys. Letters, 1977, 46, 283. 0. S. Khalil, L. Goodman, and S. H. W. Hankin, Chem. Phys. Letters, 1976,39, 221. 0. S. Khalil and L. Goodman, J. Phys. Chem., 1976, 80, 2170. 462 T. Iwao, Y.Gondo, and Y . Kanda, Mem. Fac. Sci., Kyushu Univ. Ser. C,1976, 10, 13. l d 3 R. J. Vanzee and A. U. Khan, J. Phys. Chem., 1976,80,2240. 464 D. J. W. Barber and J. T. Richards, Chem. Phys. Letters, 1977,46, 130. 46s N. Yoshida and N. Ebara, Bull. Chem. SOC. Japan, 1975,48, 709. J. B. Birks, T. D. S. Hamilton, and J. Najbar, Chem. Phys. Letters, 1976,39, 445. 16’ J. C. Miller, K. U. Breakstone, J. S. Meek, and S. J. Strickler, J. Amer. Chem. SOC., 1977,99,

lS9 460

1142.

J. D. Laposa, R. A. Nalepa, and G. L. LeBel, Mol. Photochem., 1976,7,465. 0 . S . Khalil and L. Goodman, J. Chem. Phys., 1976,66,4061.

lB8

Spectroscopic and Theoretical Aspects

59

In contrast with pyridine, 4-hydroxypyridine phosphoresces strongly in an EPA glass at 77 K. A study of the phosphorescence polarization in this system has appeared.470 A careful study of the polarization of phosphorescence from benzene in a 3-methylpentane glass leads466to the conclusion that radiative decay proceeds via the Asg spin sublevel in disagreement with theory and reliable microwave measurements. A significant out-of-plane polarized component of bzg intensity is found. A number of reasons for this anomalous and intriguing result are discussed. Polarized phosphorescence studies of the chlorophyll-like molecules rhodoporphyrin XV 471 and octaethylporphine 472 have been published. A concentration effect on the depolarization of phosphorescence has been reported 478 and a systematic study of the effect of the concentration and heavyatom depolarization of naphthalene phosphorescence has been carried out."* The metal phthalocyanines exhibit a wide range of luminescent properties which can be conveniently classified into three broad categories depending on the nature of the metal: light-metal, e.g. Mg; intermediate-metal, e.g. Pd; and heavymetal, e.g. Pt. The steady increase in the spin-orbit coupling strength, Z, in these categories is alone unable to account for the observed trends in luminescence behaviour, but in combination with an additional perturbation, a hypothetical crystal field splitting 2A, a model with considerable qualitative success is achieved.*76The observed variations in the intersystem crossing rate and zerofield splitting parameters can be related to the three situations when 2A %- Z, 2A 2, and 2A 4 2 as shown in Figure 13. Spectrally different emissions from apparently simple systems are often sensitive indicators for the presence of conformers, energy transfer, etc. Calculations indicate the presence of two stable isomers of propiophenone in the ground state, iirst n 3 n* state, and first n -+ T* singlet and triplet The dual phosphorescence of this molecule may be related to the conformational change in the n -+ n* state. Optically Detected Magnetic Resonance (0DMR). Optically detected magnetic resonance techniques continue to give detailed information about triplet substates and insights into the mechanisms of intersystem crossing processes. The triplet-state zero-field splittings (ZFS) and intersystem crossing (ISC) rate constants for bacteriochlorophyll and preparations from purple photosynthetic bacteria have been measured by zero-field ODMR.477~ 478 The importance of such measurements is nicely illustrated by subsequent papers from the same group where they are able to determine the relative orientations of the bacteriochlorophyll pair in the reaction centre of photosynthetic bacteria if, as is believed, N

473

S. Hotchandani and A. C. Testa, Spectrochim. Acta, 1976, 32A, 1659. A. T. Gradyusko and K. N. Solov'ov, Optika i Spektroskopiya, 1976,41, 57. K. N. Solov'ov, A. T. Gradyusko, M. P. Tsvirko, and V. N. Knyukshto, J. Luminescence,

47a

Y.Gondo, M. Hirai, T. Iwao, T. Kakibaya, T. Kuroi, H. Nagatomo, and Y . Kanda, Chem.

470

471

1976, 14, 365. 474 47b 476 477

Letters, 1975, 5,463. J. Friedrich, G. Weinzierl, and F. Doerr, 2.Nuturforsch., 1976, 31a, 748. T. H. Huang, K. E. Rieckhoff, K M. Voigt, and E. R. Menzel, Chem. Phys., 1977, 91, 25. J. Langlet and P. Gacoin, Theor. Chim. A d a , 1976,42, 293. R. H. Clarke, R. E. Connors, J. R. Norris, and M. C. Thurnauer, J. Amer. Chem. SOC., 1975, 97, 7178.

47*

R. H. Clarke and R. E. Connors, Chem. Phys. Letters, 1976,42, 69.

60

Photochemistry

the triplet signals arise from a reaction centre.dimer.479* 480 The authors stress the importance of using both ZFS and ISC data in the determination of the origin of triplet ODMR signals. ODMR studies of model compounds for phe~phytins,~~' and the pyridinium 484 have appeared. c ~ u m a r i nand , ~ ~p-chloroaniline ~

AcB+B3 Y

t.

-(-$

+2A)

I D2h

2A>>z

2A- 2

2A character and can give fluorescence. This is summarized in Figure 14.

fl

Figure 14 Schematic representation of the manifold of quasi-stationary states. The fluorescent states, which contain singlet character due to resonant mixing, are contained by the shaded areas, the non-fluorescent states by the non-shaded areas (black holes). The vertical arrows represent the various vibrational relaxation processes. Note the significant decrease of kT and k-, the latter only existing for the shaded areas Reproduced by permission from Chern. Phys., 1976, 16, 125)

As mentioned earlier for glyoxal, the values of p T (and ps) calculated from the Haarhoff formula using known and estimated vibration frequencies are significantly smaller than those found experimentally. The ratio pT/ps, however, shows good agreement between calculated and experimental values. The paradox is removed by considering significant coupling of the vibrational and rotational motions. As might be expected, methylglyoxal exhibits behaviour intermediate between that of its close analogues, glyoxal and biacetyl. Studies at relatively low excess excitation energies 619-521 (up to 1115 cm-l) were extended 522 to values of 3000 cm-l to observe behaviour characteristic of the ‘overlap region’ seen in the biacetyl studies. The expected fall in the ratio p T / p s was not found and may 619 520

621 622

R. R. R. R.

L. Opila, R. A. Coveleskie, and J. T. Yardley, J. Chem. Phys., 1975, 63, 593. A. Coveleskie and J. T. Yardley, Chem. Phys., 1975, 9, 275. A. Coveleskie and J. T. Yardley, Chem. Phys., 1976, 13, 441. van der Werf, E. Schutten, and 5. Kommandeur, Chem. Phys., 1976, 16, 151.

Spectroscopic and Theoretical Aspects

65

occur at even higher ex~essenergies, but in other respects an analysis in terms of ‘black hole’ and overlap regions was successful. The different behaviour of the three analogues can be completely understood on the basis of the differences in the level densities 622,andthis is illustrated in Figure 15 which shows the number of interacting triplet states N versus the excitation energy.

.I

\ . \2\

l

o

lo2

o/ I I biocetyl I

.I. /

116.3

I

I.

ii I

1

glyoxol

./I / /’

me-glyoxol

1.11.5

t / I.

./

I

/

22000

E x c i t a t i o n energy (cm-’1 2LOOO

Figure 15 Experimental N values for: glyoxal (a), methylglyoxal (M), and biacetyl (0) respectively. The dotted curves give the best fit of ksT[fpT (calc)]/2c, yielding the enhancement factors f. The solid lines give the computer calcuiated values for pT/ps (Reproduced by permission from Chem. Phys., 1976, 16, 151)

Azabenzenes provide another dass of molecules in which non-statistical radiationless transitions are expected. The S1singlet state and the lowest triplet states are dose in energy, and the vibrational degrees of freedom are limited in these molecules. Therefme the Sl-T ISC should exhibit ‘intermediate case’, or possibly even ‘small molecule case’, radiationless transitions. An interesting review of published data on the six azabenzenes includes new measurements of the decay processes from the S, zera-point levels (6l level in the case of s - t r i a ~ i n e ) . ~ ~ ~ Collision-induced triplet formation was clearly demonstrated for pyrazine, pyrimidine, and s-triazine. The dual fluorescences expected of molecules with b23

A. E. W. Knight and C. S. Farmenter, Chem. Phys., 1976, 15, 85.

66

Photochemistry

intermediate-case coupling were also observed in a separate study of pyrimidine v a p o ~ r .The ~ ~ problem ~ of differences in the calculated and experimental level densities mentioned previously has been discussed with particular reference to SVL fluorescence from pyrazine vapour.626 A pressure-dependent level density of the lowest triplet state is proposed to explain pyrazine fluorescence characteristics at low and high pressures: the general point is also made that any theoretical treatments must include all the electronic states and their associated vibrational manifolds. An extensive analysis of the fluorescence from selected vibronic levels of the Ill1(&) state of pyrimidine has been published.626 Strong anharmonic coupling in the excited state results in marked deviations from the predicted FranckCondon intensities. The forbidden 1B2u-1Al,(Sl-So)radiative transition of benzene is the classic example of a vibronically induced radiative transition and as such provides a crucial test of theories of such transitions. Within the first 2500 cm-l of the S1 manifold, benzene exhibits a number of truly discrete levels before, at higher energies, the density of states increases to the point where the width of a given vibronic level exceeds the level separation. New measurements of relative band intensities in progressions found in single vibronic level (SVL) fluorescence 527 and of radiative transition probabilities of single vibronic levels 628 have been published. The relative band intensities could be reproduced by first-order Herzberg-Teller theory only after the introduction of normal mode mixing.627 There are Fermi resonances of levels with h&b of 121 cm-l and 223 cm-l in the excited and ground states respectively which despite modest coupling energies (ca. 10 cm-l) produce large changes in the calculated radiative transition probabilities. The Franck-Condon intensities of members of the u1 progressions can be reasonably predicted by normal Herzberg-Teller theory. Combining these band intensity measurements with new lifetime and quantum yield data gives values for the radiative transition probabilities of the single vibronic Herzberg-Teller theory was unable to account for the relative single-channel radiative transition probabilities of these levels. Detailed studies of the SVL photophysics of naphthaIene vapour under collision-free conditions have been p ~ b l i s h e d . ~It~is~ not -~~ possible ~ to separate out cleanly SVL effects because some overlapping with neighbouring sequence levels is always present. The allowed purely electronic Sl-So transition is weak and numerous vibronically induced origins can be seen, most of which are stronger than the allowed origin. The SVL fluorescence spectra are discussed in terms of vibronic coupling In part they are considered as illustrating an interference effect between the Herzberg-Teller and Born-Oppenheimer transition moments. Large variations in the intersystem crossing (ISC) rate for the four lowest SVL's are indicated by fluorescence quantum yield measure524

626 626

628

519

K. Uchida, I. Yamazaki, and H. Baba, Chem. Phys. Letters, 1976, 38, 133. G. Fischer and R. Naaman, Chem. Phys., 1977, 19, 377. A. E. W. Knight, C. M. Lawburgh, and C. S. Parmenter, J. Chem. Phys., 1975, 63,4336. C. S. Parmenter, K. Y . Tang, and W. R. Ware, Chem. Phys., 1976, 17, 359. W. R. Ware, A. M. Garcia, C. S. Parmenter, M. D. Schuh, and K. Y . Tang, Chem. Phys., 1976, 17, 377. M. Stockburger, H. Gattermann, and W. Klusmann, J. Chem. Phys., 1975, 63, 4519.

Spectroscopic and Theoretical Aspects 67 ments.6a0 At higher excess energies a second non-radiative decay channel dominates which is not ISC and which may be internal conversion to the ground state. Naphthalene phosphorescence is found to have a lifetime of 4.2 ms which is significantly longer than previously reported values.631 The fluorescence of pyrene vapour under collision-free conditions clearly shows spectral components from both the S1and Szstates. The effect of cyclohexane and oxygen on the fluorescence can be explained by a straightforward kinetic analysis if it is assumed that internal conversion between the S2 and S1 states is reversible and very fast compared with other radiative and non-radiative processes from these states and that oxygen quenches both the S1and Szemissions whereas cyclohexane causes only vibrational relaxation.6a8 The S2 fluorescence consists mainly of a slow component which originates from an S2* state in equilibrium with an S1*state. The lifetimes of the S1and S2 emissions are very nearly equal. The Sa fluorescence quantum yields for several SVL’s of thiophosgene have been measured.633 The Szquantum yields are unusually high for this molecule, being in the range 0.20-1.0. An investigation of the photodecomposition of cyclobutanone from SVL’s showed considerable variations in the nature of the products obtained from different levels and the onset for predissociation was found to be only 700 cm-l, which is significantly lower than previous The fluorescence decay times of vibrational levels of several electronic states of nitric oxide [A2Z+(u = 0, 1,2,3), PII (u = 5), C2n(u = 0, l), and D2Z+ (0 = 0,1,2,3)] have been Attention was focused on the C2ll (u = 0) level as its energy coincides with the first dissociation limit of NO to ground state N and 0 atoms. Non-exponential decay behaviour was observed for this level. A detailed analysis of the SVL fluorescence of chromyl chloride shows that the 580 nm band system previously designated ‘system 11’ comprises at least two electronic Recently an interesting study of NO2 fluorescence lifetimes following excitation to the perturbed 2B2 state near 600nm has been published.s37 At certain excitation wavelengths, the decay of the total fluorescence is highly nonexponential even at Torr. Absorption in more weakly absorbing regions gives long-lived nearly exponential decays. The results are interpreted in terms of a single excited electronic state (*BJ variably coupled vibronically to upper vibrational levels of the ground state. 5 Intermolecular Excited-state Decay Processes Papers containing material relevant to this section are comprehensively reported on in later Chapters of this Volume; the few selected here are mainly concerned with the theoretical aspects of such processes. 6ao

6a1

bax baa

m4

b87

M. Stockburger, H. Gattermann, and W. Klusmann, J. Chem. Phys., 1975, 63, 4529. H. Gatterrnann and M. Stockburger, J. Chem. Phys., 1975, 63, 4541. K. Chihara and H. Baba, Bull. Chem. SOC.Japan, 1975,48, 3093. T. Oka, A. R. Knight, and R. P. Steer, J. Chern. Phys., 1977,66,699. K. Y. Tang and E. K. C. Lee, J. Phys. Chem., 1976,80, 1833. 0. Benoist D’Azy and R. Lopez-Delgado, Chem. Phys., 1975,9, 327. R. N. Dixon and C. R. Webster, J. Mot. Spectroscopy, 1976, 62, 271. V. M. Donnelly and F. Kaufman, J . Chem. Phys., 1977,66,4100.

Photochemistry

68

Electronic Energy Transfer,-There have been some interesting developments in the theory of long-range energy transfer when both the donor and acceptor molecules are able to diffuse. A new treatment of this problem has appeared 638 and has been shown to give good agreement with experimental results for the quenching of aromatic hydrocarbon fluorescence by Wurster's Blue significantly better agreement than that given by a number of earlier treatments. The theory is developed from a statistical treatment of pair probability densities, an approach which has been successfully applied to diffusion-controlled bimolecular reactions, and leads to a number of simple approximate analytical expressions for luminescence quenching. The concentration, CD*, of excited donors after a &excitation is given by

where T is the mean lifetime of excited donor molecules for the case CA = 0, and where the function @(t)takes the following form for the case where diffusion is dominant:

aD(t)= 4 r D r ~ ~ [+1 r m / ( ~ D t ) * ]

(6)

and the following form for the case where resonant energy transfer is dominant: QR(t) = 4rDr*[1

+ r*/(vDf)f]

(7)

with r* = 0.724(a/D)f. D is the sum of the diffusion constants for the donor and the acceptor, 01 is a measure of the intrinsic energy transfer efficiency, and r~ is the collision distance. In a subsequent paper 640 the authors examined in more detail the cases where both diffusion and energy transfer were of importance and discovered that a general treatment could be given in terms of an effective radius, re^, given by

where r~ z (1 /0.93)r*, Zo = (rF/rAD)2, and where f(&) [ z o.43Zoa/lY*(Zo)/Zt(Zo)] contains the modified Bessel functions K* and I&. As shown in Figure 16, to a good approximation r e z r m for Z , < 1 and r d z r~ for 2, > 1. The theory gives very good agreement with experimental measurements from a pyrene(D)perylene(A) The effect of the large changes in the diffusion constants of donors and acceptors in going from the liquid to the solid phase on the kinetics of longrange energy transfer by exchange interaction has been i n v e ~ t i g a t e d . ~ The ~~ discussion is in terms of a time-dependent radial distribution of acceptors around the donors and is analogous to an earlier treatment of quantum mechanical electron tunnelling in rigid matrices. Some interesting extensions of the original Forster treatment of long-range energy transfer have been made64awhich may find application in studies of 638 639 640

641

64a

U. Gosele, M. Hauser, U. K. A. Klein, and F. Frey, Chern. Phys. Letrers, 1975, 34, 519. L. R. Faulkner, Chem. Phys. Letters, 1976, 43, 552. U. K. A. Klein, R. Frey, M. Hauser, and U. Gosele, Chem. Phys. Letters, 1976, 41, 139. M. J. Pilling and S. A. Rice, J.C.S. Faradby 11, 1976, 72, 792. M. Hauser, U. K. A. Klein, and U. Goseie, Z . phys. Chem. (Frankfurt), 1976,101,255,

69 micellar systems. The cases for transfer to statistically distributed acceptors in one, two, and three dimensions and for single donor-acceptor pairs restricted to a range of possible geGmetric arrangements, e.g. the pair is confined within a sphere, were treated. Fluorescence quenching by exciton-exciton interactions has been discussed for the case where the exciton motion is controlled by anisotropic diffusion.64S A classical theory of energy transfer from an electric dipole to an absorbing medium has been extended to magnetic dipole and electric quadrupole In Kuhn’s famous work on energy transfer across fatty-acid monolayers, the fatty acid was treated as a weakly absorbing medium. The present work shows Spectroscopic and Theoretical Aspects

I 0

1.o

2.0

rF -

*

r~~

Figure 16 Ratio of the efectiue interaction radius reff and collision radius r m as a function of the ratio rF/rm, - - - approximation for rFlrm < 1, ---approximation firrp/rm > 1 (Reproduced by permission from Chem. Phys. Letters, 1976, 41, 139) - a

that the effect of the optical anisotropy of the fatty-acid layers on the fluorescent lifetimes and the energy transfer of emitters embedded in them is relatively small. In a .memorial lecture to Theodor Forster, Sir George Porter has discussed the problem of energy transfer in The serious discrepancies that can appear between quenching data obtained by steady-state and dynamic Stern-Yolmer-type studies on the same system have been clearly demonstrated in a study of the oxygen quenching of liquid toluene Such discrepancies arise mainly from one or more of the following three causes: (i) formation of a ground-state complex between the quencher and the fluorophore, (ii) time-dependent concentration gradient effects which result in k,, the quenching rate constant, being apparently time dependent, (iii) excited-state complex formation. In the present study the effects of both (i) and (ii) were observed. Studies of electronic energy transfer in condensed phases and in atomic vapours are commonplace. Only a limited amount of work has, however, been carried 643 b44

646

U. Gosele, Chem. Phys. Letters, 1976, 43, 61. R. R. Chance, A. Prock, and R. Silbey, J. Clrsm. Phys., 1976, 65,2527. G. Porter, Naturwiss., 1976, 63, 207. C. Lewis and W. R. Ware, J.C.S. Faraday ZI, 1976,72, 1851.

70 Photochemistry out on such phenomena in molecular vapours. This situation will probably change following the recent publication of two fascinating studies of electronic energy transfer between single vibronic levels in a benzene-aniline gas-phase mixture647and in a ben~ene-[~H,]benzenegas-phase mixture.648 In the first of these studies benzene was selectively excited to one of the 6l, 6111, or 6llZ levels of the lBzustate. The quenching of the resulting benzene fluorescence by the addition of aniline was matched by a corresponding enhancement of the aniline fluorescence. All of the measurements were carried out at pressures low enough to make the vibrational relaxation of excited vibronic levels comparable with that due to both energy transfer and radiative decay. The main conclusions of the investigation are that electronic energy transfer is the dominant, or possibly the only, mechanism of fluorescence quenching in the benzene(D)-aniline(A) system and its rate is independent of the initial benzene vibronic level involved in the transfer. It was also found that the transfer was accompanied by a significant amount of energy appearing as translational energy and/or vibrational energy of the ground-state donor molecule. The most important conclusion is that electronic energy transfer between gas-phase aromatic molecules probably involves short-range interactions, and that a theoretical treatment based on a Forster model, where higher terms in the multipole expansion of the Coulomb interaction are neglected, would therefore be inadequate. These conclusions are consistent with earlier theoretical work and with the results of the benzene(A)[2H,]benzene(D) i n v e s t i g a t i ~ n .In ~ ~this ~ latter investigation, a strong dependence of the transfer efficiency on the final state of the acceptor reached from a given initial donor vibronic level was detected. The transfer to the Oo level of the lBzu state of benzene from the initially pumped Oo level of [2H,]benzene (AE = -203 cm-I) was found to be more important than transfer to the nearresonant 16l level ( A E = + 37 cm-l). The feasibility of using electronic energy transfer as an efficient pumping mechanism for a dye vapour laser has been investigated for the potentially favourable pyrene(D)-perylene(A) It was concluded that it is highly unlikely that such a mechanism can form the basis of a successful organic vapour laser owing principally to the discovery that, as in the cases mentioned above, short-range interactions are involved in the energy transfer process which would mean that triplet energy transfer to the acceptor may be just as efficient as singlet energy transfer. Thus triplet absorption could well be as troublesome as it is in existing liquid dye lasers. Vibrational Energy Transfer.-The use of laser-induced infrared fluorescence techniques has already proved to be extremely valuable in detailed state-to-state kinetic studies of chemical reactions and energy-transfer processes. In the main such studies are carried out in the gas phase, but recently there have been a number of interesting reports of similar studies in the solid phase. A recent study of matrix-isolated carbon monoxide under conditions where radiationless vibrational relaxation is very slow has revealed a very efficient long-range dipole-dipole interaction which results in a relatively fast vibrational energy 647 64* 64*

C. Lardeux and A. Tramer, Chern. Phys., 1976, 18, 363. C. S. Parmenter, B. Setzer, and K. Y. Tang, J. Chem. Phys., 1977, 66, 1317. C. T. Ryan and T. K. Gustafson, Chem. Phys. Letters, 1976, 44,241.

Spectroscopic and Theoretical Aspects

71

transfer process.6s0s661 Excitation of the first vibrational level of 12Cf60with the frequency-doubled output of a CO, laser results in fluorescence emission from upper vibrational levels of laC160 and also of 13C160and 12C180which are present in natural abundance. A plausible explanation for this and related observations involves two energy transfer processes : firstly a resonance energy transfer process which results in a rapid diffusion, or migration, of vibrational energy between 14C1s0molecules, and secondly a single-phonon-assisted energy transfer process between the vibrational levels of all of the isotopic species present. The phonon energy can compensate for a range of vibrational energy level differences which result either from the effects of vibrational anharmonicity or from isotope effects. These observations could lead to a number of practical applications. The authors suggest that such processes could form the basis of an isotope separation method and possibly of a molecular solid-state laser. 6 Photochemical Excited-state Decay Processes Theories of Photochemistry.-An electronic theory of photochemical reactions has been discussed in which emphasis is placed on the essentially biradical character of the primary product. Four electronic states with very different characteristics can be visualized for such a biradical species. In cases where there are two clearly identifiable radical sites, two states arise when one electron is associated with each site, namely, the triplet and singlet biradical states, sD and lD, and the other two states, the zwitterionic states 2, and Z,, arise from the cases where both electrons are associated with either of the two sites labelled 1 and 2. A number of photochemical reactions are discussed in terms of these states and a number of other features important to the theory, i.e. surface crossings and avoided surface crossings, and a property called topicity which is defined as the total number and nature of available radical sites generated in the primary process. A comprehensive article on the dynamics of primary photochemical processes has been published 654 and an interesting discussion of the dynamics of electron excitation in the wider context of chemical reactions.666 A number of reviews of the use of potential energy surfaces in theories of photochemical reactions have appeared.666-SS8The rate constants for a generalized three-component photochemical reaction system have been expressed analytically6ss and the application of an expression for the one- and two-centre contribution to the excitation energy in PPP processes has been discussed.6so Photochemical primary processes in solution are discussed briefly within a quantum mechanical framework in a recent paper.661 66a1

H. Dubost and R. Charneau, in ‘Molecular Spectroscopy of Dense Phases - Proceedings of the 12th European Congress on Molecular Spectroscopy’, Elsevier, Amsterdam, 1976, p. 719. H. Dubost and R. Charneau, Chem.Phys., 1976,12,407. m L. Salem, Israel J. Chem., 1976, 14, 89. ua L. Salem, Science, 1976, 191, 822. 664 S. A. Rice in ‘Excited Statcs’, Vol. 2, ed. E. G. Lim, Academic Press, London, 1975, p. 111. 651 U. Fano, Phys. Todzy, 1976,29, 32. 668 J. J. Kaufman, Ado. Chem. Phys., 1975,28, 113. c67 A. Devaquet, Pure Appl. Chem., 1975,41,455. 668 J. Michl, Pure Appl. Chem., 1975,41, 507. wv G. K. Heidt, J. Photochem., 1976, 6, 97. A. Mehlhorn and J. Stumpe, 2. Chem., 1976,16,36. 6*1 H. Labhart, Chimia, 1977,31, 89. m0 661

12

Photochemistry

The reactivity of electronically excited atomic species produced by collisions 663 and the results of experimental measurein the gas phase has been reviewed ments of the reactivity of vibrationally excited molecules have been reviewed and compared with theoretical predictions.664Limitations on the parameters involved in any process of isotope enrichment based on selective vibrational excitation have been discussed using a simple kinetic Aspects of the photoreactions of organic molecules s66 and excited radicals 667 have been reviewed and attention has been drawn to similarities between photochemical, radical, and catalysed reactions.668 Ab initiu calculations have been carried out for two feasible forbidden photochemical reactions, i.e. feasible in terms of state symmetry but forbidden in terms of orbital symmetry, of formaldehyde.6se They indicate that the initial energy gap between intersecting configurations is the essential factor in determining whether or not there is an energy barrier to reaction. A theory of photochemical processes based on intramolecular energy transfer following the initial excitation of a single vibronic level has been developed and its principal features illustrated with data for the photodecomposition and radiationless decay of formaldehyde.670 Theoretical studies of the spin-forbidden reaction in which triplet methylene is produced by dissociation of singlet d i a ~ o m e t h a n e ,of ~ ~the ~ photochemical disrotatory closure of butadiene to cyclobutene 672 and of the photodimerization of arenaphthylene 673 have been published. Some fallacies concerning singlet and triplet mechanisms in photochemistry have been 662s

Chemically Induced Dynamic Magnetic Polarization.-An account of the current theoretical views of chemically induced dynamic electron (nuclear) polarization, CIDE(N)P, has been given; this contains a summary of recent The initial polarization in CIDEP is now attributed to a triplet mechanism in which the three triplet sublevels are formed at different rates from the photoexcited singlet state. The resulting polarization is transferred to radicals produced from the triplet state if this process occurs with spin conservation and within the longitudinal relaxation time of the triplet. Further polarization effects occur in CIDEP following radical formation, viz. the emissive-absorptive polarization. This arises from the same radical pair mechanism that results in CIDNP. The theory of these effects in CIDEP has been reviewed.676Experiments illustrating L62 66s

664 666

D. Husain, Ber. Bunsengesellschaftphys. Chem., 1977, 81, 168. D. L. King and D. W. Setser, Ann. Rev. Phys. Chem., 1976, 27, 407. J. Wolfrum, Ber. Bunsengesellschaftphys. Chem., 1977, 81, 114. K. Bergman, S. R. Leone, R. G. MacDonald, and C. B. Moore, Israel J . Chem., 1975, 14, 105.

667

670 671

H. D. Scharf and J. Fleischhauer, Method. Chim., 1974, lB, 650. A. I. Bogatyreva and A. L. Buchachenko, Uspekhi Khim., 1975,44,2171. R. Zahradnik and S. Beran, J. Catalysis, 1976, 44, 107. D. Grimbert and L. Salem, Chem. Phys. Letters, 1976, 43, 435. G. D. Gillespie and E. C. Lim, J. Phys. Chem., 1976, 80, 2166. S. N. Datta, C. D. Duncan, H. 0. Pamuk, and C. Trindle, J. Phys. Chem., 1977, 81, 923.

672

673

67Q 676

676

D. Grimbert, G. Segal, and A. Devaquet, J. Amer. Chem. SOC.,1975,97, 6629. K. A. Muszkat and S. Sharafi-Ozeri, Chem. Phys. Letters, 1976, 38, 346. J. B. Birks, Photochem. and Photobiol., 1976, 24, 287. J. K. S. Wan and A. J. Elliot, Accounts Chem. Res., 1977, 10, 161. P. W. Atkins and G. T. Evans, Ado. Chem. Phys., 1976,35, 1.

Spectroscopic and Theoretical Aspects

73

the simultaneous operation of the two polarization mechanisms have been rep~rted.~~~-~~~ Studies of CIDEP in the triplet ground state of diphenylmethylene,681in the photolysis of tetrafluoro-p-benzoquinone,6a2 and of maleic anhydride 683 and in chloroplasts at room temperature 684 have been published. It has been found that under certain limited conditions the Overhauser effect which is an electron and nuclear cross-relaxation effect, can be seen in CIDNP,677* in addition to the dominant effects of the radical pair mechanism. One of the studies477involved measuring the 18F n.m.r. signals from the photolysis of benzene solutions of tetrafluoro-l,4-benzoquinone (FQ) and tetrafluoro-l,4-hydroquinone(FQH,). Triplet-excited FQ abstracts a hydrogen atom from FQHa to give a pair of identical FQH radicals, a situation for which the radicaI pair theory predicts no net n.m.r. polarization. An emissive 18Fn.m.r. signal was observed. When chloroform solutions of FQ were photolysed the emission from 18Fwas dependent on the relative orientation of the plane of polarization of the incident light and the applied magnetic field, a result which cannot be explained by the radical pair theory but which is consistent with a mechanism in which the net electron polarization of initially polarized radicals is transferred to the nuclear spin states. lH CIDNP studies have been reported of photochemically induced hydrogen abstraction by aromatic nitro 688 of photodecarboxylation 6Bo and photorearrangement processes and of naphthalene-pyridinium e x c i ~ l e x e s .13C ~ ~ ~CIDNP studies of the photolysis of t-butyl h y d r ~ p e r o x i d e , ~ ~ ~ of benzoyl peroxideK8*and di-t-butyl ketone,6Qsand of a number of aldehydes and ketones69shave been made. Photolysis of dialkyl azo compounds has been investigated using nitrogen-15 CIDNP.687 The feasibility of constructing masers which use chemical or photochemical pumping of upper Zeeman energy levels has been 6a6p6a6

68*9

Magnetic Field Effects.-The influence of applied magnetic fields on photophysical and photochemical processes is the subject of an increasing number of K. Y. Choo and J. K. S. Wan, J. Amer. Chem. SOC.,1975,97,7127. B. B. Adeleke and J. K. S. Wan, J.C.S. Faraday I, 1976,72, 1799. 67B P. B. Ayscough, G. Lambert, and A. J. Elliot, J.C.S. Faraday I, 1976, 72, 1770. 680 K. A. McLauchlan and R. C. Sealy, Chem. Phys. Letters, 1976, 39, 310. 681 D. C. Doetschman, B. J. Botter, and J. Schmidt, Chem. Phys. Letters, 1976, 38, 18. 682 H. M. Vyas and J. K. S. Wan, Canad. J. Chem., 1976,54,979. 683 P. Wuensche and J. Hellebrand, 2. phys. Chem. (Leipzig), 1977, 258, 203. 684 R. Blankenship and A. McGuire, Proc. Nut. Acad. Sci. U.S.A., 1975,72, 4943. 686 H. M. Vyas and J. K. S. Wan, Chem. Phys. Letters, 1975,34,470. 686 F. J. Adrian, H. M. Vyas, and J. K. S. Wan, J. Chem. Phys., 1976, 65, 1454. m7 K. A. Muszkat and M. Weinstein, J.C.S. Perkin II, 1976, 1072. K. A. Muszkat and M. Weinstein, 2. phys. Chem. (Frankfurt), 1976, 101, 105. 6 * 9 I. A. Den Hollander and J. P. M. Van der Pleig, Tetrahedron, 1976, 32, 2433. m 0 P. R. Bowers, K. A. McLauchlan, and R. C . Sealy, J.C.S. Perkin ZI, 1976, 915. R. Bausch and H. P. Schuchmann, J.C.S. Chem. Comm., 1976,418. 682 J. Bargon and G. P. Gardini, Tetrahedron Letters, 1976, 2993. 6Bs S. A. Sojka, C. F. Poranski, jun., and W. B. Moniz, J. Amer. Chem. Soc., 1975,97, 5953. 684 C. F. Poranski, jun., W. B. Moniz, and S. A. Sojka, J. Amer. Chem. SOC.,1975, 97, 4275. 6g6 W. B. Moniz, C. F. Poranski, jun., and S. A. Sojka, J. Org. Chem., 1975, 40, 2946. 696 R. Benn and H. Dreeskamp, 2.phys. Chem. (Frankfurt), 1976, 101, 11. 607 J. G. Green, G. R. Dubay, and N. A. Porter, J. Amer. Chem. SOC., 1977,99, 1264. 608 V. L. Berdinskii, A. L. Buchachenko, and A. D. Pershin, Teor. i eksp. Khim., 1976, 12, 666. w7 m8

74

Photochemistry

investigations. After a long history of scattered and often erroneous reports, the area has developed to a point where it is possible to foresee important applications in the control of photochemical reaction pathways and possibly to isotope separation. A brief article 6QQ and a review of recent experiments 6oo are recommended introductory reading material. Magnetic fields can potentially influence any process where a change in the mu1tiplicity of an intermediate is involved. The field modifies the spin-rephasing processes which lead to the interconversion of spin multiplets. An important example of this effect is the modifkation by the field of the ratio of cage to bulk recombination products following homolytic fission of a bond in a molecule, represented by M-N, to give two doublet radicals, 2M and 2N. Spin conservation in this process would result in a singlet state of M N leading to a singlet radical pair 1{2M--2N). In the condensed phase there is a high probability of the pair colliding and recombining before diffusion leads to the loss of all correlation. Recombination, however, will only take place if the radical pair has overall singlet spin. If during separation a rephasing had occurred to produce a triplet radical pair, 3{2M-*2N), the radicals would diffuse apart again without recombining. Clearly anything that modifies the rephasing process will modify the relative number of singlet to triplet encounters and hence the relative yield of cage to bulk recombination. Theory identifies two interactions which can induce spin-rephasing. The first is the Zeeman interaction which, if the radicals are different, induces S-To crossing because of small differences in the g-values of the radicals in the pair. The second interaction is the hyperfine interaction of the electrons with the magnetic moments of the nuclei in the radicals. In the absence of an applied field the hyperfine interaction induces the three crossings S-T,, Th1 with essentially equal probability. The rephasing rate arising from the Zeeman interaction will simply increase as the applied field increases and would lead to an increase in the ratio of bulk to cage product if it, alone, operated. The hyperfine interaction gives the opposite behaviour. The applied field lifts the degeneracy of the three triplet substates until only one channel for rephasing is left open, the S-T, channel. The interaction finds it increasingly difficult to induce S-T+lrephasing as the S (or TOFT*,energy separation increases and the ratio of bulk to cage products would decrease if this mechanism operated alone. The Zeeman interaction will be the dominant rephasing interaction in high fields. The effect of an external magnetic field on the singlet sensitized photolysis of dibenzoyl peroxide (B,O), has been reported.601The yield of the geminate product B,OPh decreased by 8% and the yield of free-radical products increased by 2% in the presence of a field of 4.3T (1 T = 10 kG). The decrease in the B30Ph yield was proportional to the square root of the applied field strength, a result which agrees with predictions of the radical-pair theory of CIDNP. 69n

Boo 601

P. W. Atkins, Chem. Brit., 1977, 12, 214. P. W. Atkins and T. P. Lambert, Ann. Reports (A), 1975,72, 67. Y. Tanimoto, H. Hayashi, S. Nagakura, H. Sakuragi, and K. Tokumaru, Chem. Phys. Letters, 1976, 41, 267.

Spectroscopic and Theoretical Aspects 75 An important observation of isotope enrichment has been reported.602 The recombination of PhCH,CO* and PhCH2- radicals produced in the photolysis of (PhCH,),CO led to an enrichment of (PhCH,),CO in 13C, mainly at the carbonyl group. When photolysed in the presence of a magnetic field the enrichment decreases with increasing field strength. Reports have also appeared of magnetic field effects on the singlet- and triplet-sensitized photolysis of di benzoyl peroxide,so3the photochemical reaction of isoquinoline N - o ~ i d e , ~ ~ ~ and the photodimerization of acenaphthylene 605 to give cyclobutane-type dimers. In the latter study the field was found to increase the cis-trans ratio by 5-10%, the cis-product being produced via the singlet, the trans-product via the triplet. A situation closely related to that of radical pairs outlined above arises in the recombination of radical ion pairs produced in the radiolysis of liquid hydrocarbon solutions. The initially produced holes and electrons rapidly transfer to solute molecules having excited states below those of the solvent, resulting in geminate ion-pairs. Similar situations can be produced photochemically. In non-polar solvents the recombination is predominantly geminate and the energy release is large enough to produce excited states. The relative singlet to triplet yields resulting from such recombinations are sensitive to applied magnetic fields. The theory of this effect has been considered in some detail by Brocklehurst.606 The effect of the applied field is again to modify the spinrephasing of the initially singlet radical ion pair via, predominantly, the isotropic hyperfine coupling of the magnetic nuclei and the electrons. Several predictions of the theory have been verified experimentally. A marked isotope effect was observed in the scintillation pulse shapes from [lH,,]terphenyl and [2H,,]terphenyl in decalin, confirming the role of electron nuclear hyperfine interactions in the solute radical ions.6o7The results are illustrated in Figure 17. Similar isotope effects were found for continuously y-irradiated solutions of naphthalene and [2H,]naphthalene.608The fluorescence intensity is enhanced in a magnetic field by a smaller amount for the perdeuteriated compound than for its perprotonated parent molecule. Solvent effects on the radical ion recombination times also give the predicted behaviour. Large enhancements (40-50%) of the fluorescence from squalane solutions of fluorene following pulse radiolysis were observed for fields of 0.3 T.sou Triplet production during geminate recombination of radical ion pairs has been observed directly.610 The yield of such triplets was reduced by 80% from 6 oa

608

804 608 606

607 6 08

A. L. Buchachenko, E. M. Galimov, V. V. Ershov, G. A. Nikiforov, and A. D. Pershin, Doklady Akad. Nauk S.S.S.R., 1976, 228, 379. H. Sakuragi, M. Sakuragi, T. Mishima, S. Watanabe, M. Hasegawa, and K. Tokumaru, Chem. Letters, 1975, 3, 231. N. Hata, Chem. Letters, 1976, 6, 547. K. Ichimura and S. Watanabe, Sen'l Kobunshi Zairyo, 1975, 108, 29. B. Brocklehurst, J.C.S. Faraday 11, 1976, 7 2 , 1869. B. Brocklehurst, Chem. Phys. Letters, 1976, 44, 245. R. S . Dixon, F. P. Sargent, V. J. Lopata, and E. M. Gardy, Chem. Phys. Letters, 1977, 47, 108.

609

610

F. P. Sargent, B. Brocklehurst, R. S. Dixon, E. M. Gardy, V. J. Lopata, and A. Singh, J. Phys. Chem., 1977, 81,815. K. Shulten, H. Staerk, A. Weller, H.-J. Werner, and B. Nickel, Z . phys. Chem. (Fmnkfurt), 1976,101, 371. 4

Photochemistry

76

its zero-field value on application of a field of 0.05T in general agreement with theoretical predictions.606The system studied was pyrene (Py) and 3,5-dimethoxyNN-dimethylaniline ( D M D M A ) in methanol. Photoinduced electron transfer results in the formation of singlet-solvated radical ion pairs 1{2Py--*2 D M D M A + ) . These may recombine to give ground-state products although this process may

o

#

25-

,*

-.

o

0

"

'\

I

'

II

\

'

I

2'i 10:' I

I

I

2.0 -

r

I

I

I

I

I

I

15 -

(bl

I

I

I

I

0

1.0

0

/

I

I I

I

I

50

100

150

t Ins

Figure 17 Experimental values of R, the fluorescence intensity ratio - field on (0.16 Tesla)/ field of- versus time after the peak of the scintillation pulse; (O), 0.005 mol dm-3 terphenyl in decalin; (a),0.005 mol dm-3 [2H14]terphenyl in decalin; ( x ), 0.01 mol dm-3 terphenyl in benzene; (b) theoretical values of the ratio of the amounts of singlet terphenyl; - - -, [2H14]terphenyl character in the ion-pair wavefunction; -, (Reproduced by permission from Chem. Phys. Letters, 1976, 44, 245)

have low probability because the large energy difference between the ground and radical ion pair states makes the transition Franck-Condon forbidden. If spinrephasing occurs before the pair finally diffuse into the bulk then, when reencounter occurs, a triplet radical ion pair 3{2Py--.-2 D M D M A + } can give a triplet excited-state product because of the nearness of the triplet and radical ion pair states. In this way 3Py* and l D M D M A are produced in such geminate recombinations. From the discussion of neutral radical pair recombination given earlier, it is clear that inhibition of the rephasing, via hyperfine coupling, by an applied field will reduce the geminate triplet yield, in agreement withthe

Spectroscopic and Theoretical Aspects

77

experimental observations. Similar results were obtained in a laser photolysis following an study of solutions of pyrene and NN-diethylaniline in A key observation prior to the above reports earlier discussion of the was that of Mataga and Nagashimas13 who found triplet formation within one nanosecond of the formation of excited single anthracene by laser photolysis of solutions of anthracene and NN-diethylaniline in acetonitrile. A recent theoretical treatment of the effect of magnetic fields on radical recombination reactions 614 extends earlier work by including the effects of exchange interaction between the unpaired spins and the effects of diffusion. The general theory is based on the stochastic Liouville equation for the density matrix of the system. An investigation of the monomer and excimer delayed fluorescence of pyrene solution has shown similar effects of an applied magnetic field on both types of emission.615 The result suggests a common spin-selective step for monomer excited-state and excimer formation. A more detailed study found a marked temperature dependence of the field effect.61e The field dependence of the two delayed emissions was found to be identical in the high-temperature region but to diverge at low temperatures. This change was found to be a general feature of a number of different aromatic-solvent systems. The results can be explained equally well, at least quantitatively, by mechanisms which do or do not include a common spin-selective step. New and extensive data on the above effect have been publisheds1' and associated theoretical treatments have been given.61s,619 The theories are based on a generalized Liouville equation for the triplet pair density matrix. There have been several reports of the quenching of fluorescence of molecules in the gas phase by an applied magnetic field. The effect has been found for glyoxa1,620-622 carbon disulphide,622and nitrogen In the work on NO2 the effect resulted from a field-induced increase in the collisional quenching cross-section. It has been proposed 624 that these quenching observations can be explained by two types of mechanism: a direct and an indirect mechanism. The first enhances intramolecular radiationless decay by adding a Zeeman term to the matrix element of the perturbation and the second by modifying the initial or final states of the decay channel. An external magnetic field has been found to produce a large increase in the quantum yield of fluorescence from excited complexes of rubrene with 0xyge1-1.~~~ 612

613 614 615

616

620

621 622

623 624

626

M. E. Michel-Beyerle, R. Haberkorn, W. Bube, E. Steffens, H. Schroder, H. F. Neusser, E. W. Schlag, and H. Seidlitz, Chem. Phys., 1976, 17, 139. R. Haberkorn and M. E. Michel-Beyerle, 2.Naturforsch., 1976, 31a, 499. N. Mataga and N. Nakashima, Spectroscopy Letters, 1975, 8, 275. R. Haberkorn, Chem. Phys., 1977, 19, 165. H. van Willigen, Chem. Phys. Letters, 1975, 33, 540. D. A. Capitanio and H. van Willigen, Chem. Phys. Letters, 1976, 40, 160. J. Spichtig, H. Bulska, and H. Labhart, Chem. Phys., 1976, 15, 279. K. Lendi, P. Gerber, and H. Labhart, Chem. Phys., 1976, 18, 449. K. Lendi, P. Gerber, and H. Labhart, Chem. Phys., 1977, 20, 145. A. Matsuzaki and S. Nagakura, Z. phys. Chem. (Frankfurt), 1976, 101, 283. A. Matsuzaki and S. Nagakura, Chem. Phys. Letters, 1976, 37, 204. A. Matsuzaki and S. Nagakura, J. Luminescence, 1976, 12/13, 787. S. Butler and D. H. Levy, J. Chem. Phys., 1977, 66, 3538. P. W. Atkins and P. R. Stannard, Chem. Phys. Letters, 1977, 47, 113. B. M. Rumyantsev, V. I. Lesin, and E. L. Frankevich, Khim. oysok. Energii, 1977, 11, 132.

78

Photochemistry

An applied field can produce similar effects to those of microwaves under certain conditions. This has been demonstrated in a study of low temperature alkane solutions of Zn chlorophyll-b.62e The role of electron nuclear hyperfine interactions in the mechanism of triplet formation in photosynthesis has been Photoionization.-A review of experimental techniques used to measure the total and partial cross-sections of the primary processes in photoionization has been published.628The cross-sections of each of the three primary processes, which are (i) direct ionization, (ii) direct dissociative ionization, and (iii) direct multiple ionization, can now be measured very accurately in many different atomic and molecular systems. Generally speaking, the theoretical prediction of these crosssections for atoms is reasonably successful, but in the case of molecules much less so. A new orbital theory of molecular photoionization has been published 629 which retains the ease of interpretation of other orbital theories. The theory involves a pair of Dyson orbitals; one of these orbitals represents the outgoing electron and the other the remaining hole, and in the very high energy limit the final expression that is obtained reduces to the Born and sudden approximation forms. An investigation of processes in which there is a major redistribution of the electrons remaining in a molecule following the photoejection of an electron has shown that as a result of such changes the experimental cross-sections may be unusually small and/or have an angular distribution which is characteristic of excited orbitals of the parent m01ecule.~~~ Reviews of photoionization 631 and the Fano-Mies theory of photoionization resonances 632 have appeared. Photoionization cross-sections of the noble gases have been calculated using a multiconfiguration Hartree-Fock method 633 and the results were comparable with values obtained using the random-phase approximation with exchange and the R-matrix. A many-body theoretical approach has been used to treat the photoionization 634 and double photoionization 635 of helium; R-matrix theory has been applied to the photoionization of neon and the acceleration, length, and velocity forms of the dipole operator have been used to calculate subshell photoionization cross-sections in argon 637 and the random-phase approximation with exchange method has been used to study the effect of electron shell rearrangements following photoionization of xenon.638 The multiphoton ionization of xenon atoms has been investigated using tunable linearly and circularly polarized light.s3g In the experiments using linearly polarized light 626

627 628

62s

630 631 632 633 634

635

636 637

638 639

R. H. Clarke, R. E. Connors, and J. Keegan, J. Chern. Phys., 1977, 66, 358. R . Haberkorn and M. E. Michel-Beyerle, F.E.B.S. Letters, 1977, 7 5 , 5. J. A. R. Samson, Phys. Reports, 1976, 28C, 303. B. T. Pickup, Chem. Phys., 1977, 19, 193. J. Simons, J. Chem. Phys., 1975, 63, 3179. C. Bottcher, Daresbury Synchrotron Radiat. Lect. Note Ser., 1974, 1. L. Torap, Daresbury Synchrotron Radiat. Lect. Note Ser., 1975, 3. J. R. Swanson and L. Armstrong, jun., Phys. Reo., A , 1977, 15, 661. M. Chatterji and B. Talukdar, Phys. Letters ( A ) , 1975, 55, 143. B. Talukdar and M. Chatterji, Phys. Reu. ( A ) , 1975, 11, 2214. P. G. Burke and K. T. Taylor, J. Phys. ( B ) , 1975, 8, 2620. M. S. Pindzola and H. P. Kelly, Phys. Reu. (A), 1975, 12, 1419. M. Ya. Amusia, V. K. Ivanov, and L. V. Chernysheva, Phys. Letters ( A ) , 1976, 59, 191. D. T. Alimov and N. B. Delone, Zhur. eksp. teor. Fiz., 1976, 70, 29.

Spectroscopic and Theoretical Aspects

79

resonances were observed corresponding to bound electron states. They were not observed when using circularly polarized light. Calculations of photoionization cross-sections in lithium, sodium, and in 642 in carbon 644 and oxygen atoms,g44 in nitrogen,64Sin c h l ~ r i n e647 , ~and ~ ~ in ~ the lanthanum atom 648 have been published. The probability for autoionization of a number of doubly excited states of the bivalent atoms, beryllium, magnesium, and zinc has been Molecular photoionization cross-sections have been calculated for H2, CO, H20, and C2H4 using an effective plane-wave method,660for H2 using the AMOLCAO approximation,6s1 and for CPHI in the second Born approximation.662 Total and partial photoionization cross-sections and angular distributions for H2, N2,CO, and O2 have been calculatedgS3and rotational states of molecular ions have been directly observed in photoionization processes.6s4 The role of molecular ‘Rydberg’ states in the photoelectron ejection from organic molecules in polar and non-polar solvents has been discussed in a series of papers on the photophysics of solutions of the tetra-aminoethylenes.g6S-668 In recent years there has been considerable interest in photoejection from an aromatic analogue of these molecules, namely, N’N’NNtetramethyl-p-phenylenediamine(TMPD) when dissolved in low-temperature alkane glasses. The photoionization cross-sections of positive ions of oxygen atom 669 and of atoms of thelnoble gases 660 have been calculated. 643s

Photodetachment.-The orthogonilized plane wave and augmented plane wave methods for calculating photodetachment cross-sections have been compared,661 and an account of multichannel photodetachment theory has been published.662 A collection of data on binding energies in atomic negative ions has also appeared.663 A multichannel J-matrix formulation of close-coupling theory has 640 641 642 Oa3

644 645 640

648

650

M. Aymar, E. Luc-Koenig, and F. C. Farnoux, J. Phys. (B), 1976,9, 1279. M. J. Jamieson, Chem. Phys. Letters, 1976, 42,441. P. Tiwari, M. A. Hashim, and S. P. Ojha, Canad. J. Phys., 1975, 53, 1524. S. L. Carter and H. P. Kelly, J. Phys. (B), 1976,9, 1887. K. T. Taylor and P. G. Burke, J. Phys. (B), 1976, 9, L353. M. Le Dourneuf, L. Vo Ky, and A. Hibbert, J. Phys. (B), 1976,9, L359. A. F. Starace and L. Armstrong, jun., Phys. Reo. (A), 1976, 13, 1850. N. A. Cherepkov and L. V. Chernysheva, Phys. Letters (A), 1977, 60,103. M. Ya. Amusia and S. I. Sheftel, Phys. Letters ( A ) , 1976, 55, 469. R. Srivastava and D. K. Rai, J . Chim. phys., 1977, 74, 17. P. R. Hilton, S. Nordholm, and N. S. Hush, Chem. Phys., 1976, 15, 345. R. A. Pandey, D. N. Tripathi, and K. N. Upadhya, Proc. Nut. Acad. Sci., India, Part A , 1974,40,29.

6ba 663 664

OS7 668

OE0 6e1

063

A. P. Markin, Optika i Spektroskopiya, 1976,40, 662. F. Hirota, J. Electron Spectroscopy Related Phenomena, 1976, 9, 149. W. B. Peatman, Chem. Phys. Letters, 1975,36, 495. Y. Nakato and H. Tsubomura, J. Phys. Chem., 1975,79,2135. Y. Nakato, A. Nakane, and H. Tsubomura, Bull. Chem. SOC.Japan, 1976,49,428. Y. Nakato and H. Tsubomura, J. Luminescence, 1976, 12/13, 845. Y. Nakato, J. Amer. Chem. SOC.,1976, 98, 7203. D. W. Missavage, S. T. Manson, and G. R. Daum, Phys. Rev. ( A ) , 1977,15, 1001. A. Msezane, R. F. Reilman, S. T. Manson, J. R. Swanson, and L. Armstrong, jun., Phys. Rev. (A), 1977, 15, 668. M. Mohraz and L. L. Lohr, jun., Internat. J. Quantum Chem., 1976, 10, 811. C. M. Lee, Phys. Rev. ( A ) , 1975,11, 1692. H. Hotop and W. C. Lineberger, J. Phys. and Chem. Ref. Data, 1975, 4, 539.

80

Photochemistry

been used to calculate the one- and two-electron photodetachment cross-sections of the hydride ion 664 and photodetachment measurements on the negative ion of tetracyanoethylene, C2(CN)(, have been used, together with structural data for the molecule and its negative ion, to obtain a value of 2.3 Ih 0.3 eV for the electron affinity of C2(CN)4.665 A table of absolute electron affinities for a number of other electron-accepting organic molecules was prepared using the vaIue for C2(CN)4as a reference. Electron Transfer.-An interesting discussion of electron transfer reactions in solution is illustrated by ab initiu calculations of the potential surfaces for a model system (NH,, cyanoethylene, 2H20).666 The model system was chosen because the charge-transfer, CT, and locally excited, LE, states have different spatial symmetries and therefore surface crossing occurs. An analysis of several configurations of the four component molecules shows that as well as surface crossing brought about by the close approach of the donor and acceptor molecules, crossing can be induced by moving the solvent molecules when the donor and acceptor are at a fixed distance. The latter surface crossing is termed a solventassisted surface crossing. Two recent publications have explored the effect of harmonic 667 and anharmonic 668 intramolecular quantum modes on free-energy relationships for exothermic electron transfer reactions. The transition probabilities of such reactions involve both the low-frequency medium modes and high-frequency intramolecular reactant modes. A number of general effects of these latter modes were predicted, including oscillatory free-energy relationships at low temperatures and high frequencies, an effect which is analogous to the vibrational structure in optical transitions. Photoisomerhation.-The interesting and important photochemical problems arising from photoisomerization reactions continue to attract considerable theoretical and experimental research effort. A semiclassical trajectory approach to photoisomerization processes has been described.669 The theory of photochemical isomerization in polyenes has been discussed 314 and careful measurements of the isomerization yields of benzene vapour irradiated at wavelengths below 266.8 nm have been reported.671 The photoisomerizations of styrene 672 and the cis-trans isomerizations of stilbene 673-677 and stilbene derivatives 678 have also been the subject of recent publications. A theoretical study of the photochemical 6701

664

665

670

672

873 874 676 676

677 678

J. T. Broad and W. P. Reinhardt, Phys. Rev. ( A ) , 1976, 14, 2159. L. E. Lyons and L. D. Palmer, Austral. J. Chem., 1976, 29, 1919. G . Ramunni and L. Salem, Z . phys. Chem. (Frankfurt), 1976, 101, 123. J. Ulstrup and J. Jortner, J . Chem. Phys., 1975, 63, 4358. N. C. Sondergaard, J. Ulstrup, and J. Jortner, Chem. Phys., 1976, 17, 417. A. Warshel and M. Karplus, Chem. Phys Letters, 1975, 32, 1 1 . 5. N. Kushik and S. A. Rice, J . Chem. Phys., 1976, 64, 1612. S. A. Lee, J. M. White, and W. A. Noyes, jun., J. Chem. Phys., 1976, 65, 2805. M. C. Bruni, F. Momicchioli, I. Beraldi, and J. Langlet, Chem. Phys. Letters, 1975, 36, 484. J. B. Birks, Chem. Phys. Letters, 1976, 38, 437. D. J. S. Birch and J. B. Birks, Chem. Phys. Letters, 1976, 38, 432. R. Benson and D. F. Williams, J . Phys. Chem., 1977, 81, 215. J. Saltiel, D. W. L. Chang, E. D. Megarity, A. D. Rousseau, P. T. Shannon, B. Thomas, and A. K . Uriarte, Pure Appl. Chem., 1975, 41, 559. E. Heumann, W. Triebel, R. Uhlmann, and B. Wilhelmi, Chem. Phys. Letters, 1977, 45, 425. R. Matsushima, T. Kishimoto, and M. Suzuki, Bull. Chem. SOC.Japan, 1975, 48, 3028.

Spectroscopic and Theoretical Aspects 81 isomerization of oximes has appeared 670 and Monte Carlo calculations have been carried out for the photoisomerization reaction of 1,2-diphenylcy~lopropane.~~~ The direct and sensitized cis-trans photoisomerizations of cyclo-octene have been investigated and ground- and excited-state potential energy curves for the molecule have been proposed to account for the results.681 Studies of photoisomerization in cyanines,682, 683 in a z ~ r n e t h i n e685 ,~~ and ~ ~merocyanine 687 and in freebase porphyrin have been reported. Proton Transfer in the Excited State.-A review of the acid-base properties of electronically excited states of organic molecules has been published which contains extensive tables of experimental values for pK(So), pK(S1), and

Figure 18 A cubic form symbolizing the chemical species and thermodynamic quantities in a system involving one-electron transfer, one-proton transfer, and electronic excitation (Reproduced from J.C.’S. Faraday 1,1977, 73, 11)

pK(Tl).68DAn interesting discussion has appeared of the thermodynamic and extrathermodynamic (e.g. Hammett-type substituent constants) relationships between proton transfer, electron transfer, and electronic These relationships may be combined to give a three-dimensional ‘cycle’ referred to as the generalized Forster cycle. The scheme is shown in Figure 18. Each of the A. Dargelos, D. Liotard, and M. Chaillet, Theor. Chim. Acta, 1975, 38, 79. E. W. Valyocsik, Mol. Photochem., 1976, 7 , 131. m1 Y.Inoue, S. Takamuku, and H. Sakurai, J. Phys. Chem., 1977,81, 7. w* D. J. Lougnot and J. P. Fouassier, Compt. rend., 1976, 282, C, 265. 683 J. P. Fouassier, D. J. Lougnot, and J. Faure, J. Chim. phys., 1977, 74, 23. 08* W. G. Herkstroeter, J. Amer. Chem. SOC.,1976, 98, 330. 686 M. I. Knyazhanskii and M.B. Stryukov, Khim. vysok. Energii, 1976, 10, 98. 686 V. A. Kuz’min, A. M. Vinogradov, Ya. N. Malkin, M. A. Al’perovich, and I. I. Levkoev, Dolkady Akad. Nauk S.S.S.R., 1976, 227, 380. m7 V. A. Kuz’min, A. M. Vinogradov, N. N. Romanov, M. A. Al’perovich, I. I. Levkoev, and F. S. Babichev, Izvest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 1864. 688 S. Voker and J. H. van der Waals, Mol. Phys., 1976, 32, 1703. J. F. Ireland and P. A. H. Wyatt, Ado. Phys. Org. Chem., 1976, 12, 131. m0 Z. R. Grabowski and W. Rubaszewska, J.C.S. Faraday I, 1977, 73, 11. 67a

E8O

82

Photochemistry

corners of the figure represents a chemical species, each edge a process connecting two species and each face a thermodynamic cycle, a.g. the protolytic Forster cycle for the oxidized species is the nearest face in the plane of the paper. An examination of data for a number of different classes of organic molecules suggests a general common trend between changes in pK on excitation (ApK*) and on reduction (ApK,). The limitations of the original Forster cycle have again been An important factor in determining the acid-base properties of molecules in their ground electronic states is the charge of the atom that is protonated. The extension of the use of such static indices to excited electronic states has been discussed in a paper which examines the effect of the number of configuration interactions used in calculations of suitable parameters, including the charge, for a number of organic The order of excited-state pK’s was shown to follow, in general, the order of the derivative w of the energy E of the state with respect to the parameter &, in the PPP method, i.e. w = dE/dW,,.

Figure 19 Simplified carbon and extreme models of thiazine

The calculation of the parameter w for a number of aromatic compounds has led to the formulation of a number of general rules concerning the ordering of pK(S,,), pK(S1), and pK(T,) values in these mo1ecules.6g3 In the theoretical discussion of the trends in these values two new models were introduced. The basic form in the equilibrium was represented by a ‘carbon model’ in which the atom to be protonated is replaced by a carbon atom and the corresponding acidic form by an ‘extreme model’ in which a .rr-electron pair is completely localized on the same centre. This is illustrated for the case of thiazine in Figure 19. The study of acid-base equilibria in excited states of conjugated molecules using the method of group density analysis has been A correlation between the ability of a number of heterocyclic aromatic compounds to deactivate the triplet states of 2-naphthol and 1-anthrol by a hydrogen-transfer reaction and their reduction potentials suggests that an electron transfer from the triplet to the quencher is the primary process in such reactions.695 In an interesting paper many different important features of the photochemistry of rhodopsin have been accommodated in a new model in which the primary step is a photoinduced tautomerization of rhodopsin by H-transfer from the C-18 methyl group to the imidazole N to give a hexaeneamine imidazole structure for prelumirhodop~in.~~~ Trends observed in the Am values of a 601 69a

684 696

696

Z. R. Grabowski and A. Grabowska, 2. phys. Chem. (Frankfurt), 1976, 101, 197. J. C. Rayez, Z . phys. Chem. (Frankfurt), 1976, 101,429. J. C. Rayez, A. Chalvet, J. Joussot-Dubien, and J. Hoarau, Z . phys. Chem., (Frankfurt), 1976, 103, 1. R. Constaciel and 0. Chalvet, ref. 5, p. 373. S. Yamamoto, K. Kikuchi, and H. Kokubun, Bull. Chem. SOC.Japan, 1976,49,2950. K. Van der Meer, J. J. C. Mulder, and J. Lugtenburg, Photochem. and Photobiol., 1976, 24, 363.

Spectroscopic and Theoretical Aspects

83 number of intermediates in rhodopsin photochemistry were well reproduced by calculations on the structures proposed in this model.

Photofragmentation.-A unified theoretical scheme for the description of direct photodissociation and predissociation of polyatomic molecules has been published.sa7 The paper establishes an important connection between photodissociation processes and the general framework of scattering calculations. In the case of energy-resolved measurements the concept of a half-collision picture of dissociative dynamics introduced by earlier workers and which emphasizes the interfragment dynamics aspect of the problem was shown to be a meaningful one. It can be represented as a superposition of full collision processes on the dissociative potential surface. For the special case of a linear triatomic molecule, analytical quantum mechanical expressions for the crosssections for photon absorption, photon scattering, and photofragmentation into different vibrational channels are presented. The interfragment repulsion is assumed to be exponential and the diatomic fragment to be characterized by a harmonic potential. Numerical solutions for the molecules HCN, CICN, BrCN, and ICN are compared with recent experimental measurements on these molecules with the aim of assessing the role of intercontinuum coupling in determining the vibrational energy distribution of the photofragments. The results suggest that, at least in these molecules, it is the Franck-Condon vibrational overlap factor that is important for the vibrational partitioning, a conclusion reached by other workers. A more detailed discussion of the case of linear triatomics has been published which includes treatments of intercontinuum intrafragment coupling and of rotational It is pointed out that to complete the picture of photofragmentation presented in the paper spectroscopic data of the predissociating states and information about the form of the repulsive potential are both required. In the case of direct photodissociation, information about the repulsive potential can be obtained from an analysis of experimental absorption lineshape functions. A large isotope effect on the photofragmentation probability of HCN and DCN 1.45 eV above the threshold for production of CN(B2Z) has been predicted t h e o r e t i ~ a l l y . ~ ~ ~ The role of angular momentum conservation in photofragmentation reactions has only recently been discussed in detail. An appreciation of this role should give a valuable additional insight into such reactions (see below). Using a simple model in which the bond that breaks is assumed to be independent of the other bonds in the molecule, the special case of a photofragmentation reaction where the prepared state is non-linear but the fragmentation takes place from a linear state has been The photofragmentation of monochloroacetylene is thought to involve an intramolecular energy transfer between two electronic states whose geometries are the same as those in the model. The treatment shows that the rate of fragmentation depends sensitively on the precise partitioning between rotational and translational kinetic energy in the product. (07 e88

'0°

S. Mukamel and J. Jortner, J . Chem. Phys., 1976, 65, 3735. 0.Atabek, J. A. Beswick, R. Lefebvre, S. Mukamel, and J. Jortner, J. Chem.Phys., 1976,65,4035 0. Atabek, J. A. Beswick, R. Lefebvre, S. Mukamel, and J. Jortner, Chem. Phys. Letters, 1977, 45, 21 1. D. Florida and S. A. Rice, Chem. Phys. Lelters, 1975, 33, 207.

84

Photochemistry

A general quantum mechanical theory of the dissociation processes, photodissociation, predissociation, unimolecular decomposition, and collisionally induced dissociation of polyatomic molecules, has been In previous theoretical treatments the quasidiatomic assumption was made in which the reaction co-ordinate is taken to be a normal mode of vibration in the initial electronic state of the polyatomic molecule and is regarded as a pure bond vibration. The distribution of vibrational energy in the fragments is altered only by the interaction that occurs between the fragments during their recoil. The new model avoids these assumptions by taking full account of the proper normal modes of the initial state of the polyatomic molecules and of the final state of the dissociation fragments. The model is essentially a Franck-Condon description of dissociation processes in terms of multidimensional bound-continuum matrix elements. It is shown that the Franck-Condon rearrangement effects often play the dominant role in determining the product energy distributions compared with the effects of forces between the recoiling fragments; a similar conclusion The new model can was reached in an earlier treatment of a simple predict the occurrence of vibrational population inversions in the fragments using only information on the relative positions of the effective oscillator. The decay of superexcited states of molecules, i.e. states whose energy is greater than the first ionization potential, is an interesting problem. Unlike the situation in the threshold energy region where these states are usually vibrationally excited Rydberg states which stabilize by ionization, the states in the higherenergy region may decay by both dissociation and ionization. The theoretical calculation of the branching ratio for predissociation and preionization has been given using both a simple classical probabilistic model 703 and a quantum mechanical treatment which uses the Green function In the latter treatment the internal state distribution of the residual molecular ion following preionization either directly from the superexcited state or via the dissociative superexcited state is given in terms of Franck-Condon factors. A treatment of the dissociative ionization cross-sections of diatomic molecules has been given.'05 The transition probabilities and absorption spectra associated with the photodissociation of N20(X1C+)to give N2(lE,+) and O(lS) have been calculated using a method in which there is no artificial separation between the photoexcitation and dissociation The vibrational state distribution of the N2 was calculated for different initial vibrational levels of the N,O and as a function of the photon energy and this made it possible to identify the conditions for which a population inversion would exist in these fragment vibrational levels. The use of correlation diagrams to understand the role of the initial electronic state in determining the nature of the fragment species in photodissociation reactions has been illustrated in studies of H20,707 H20+,'08HCOf and COH+,709 701 702

703 704

705 '08 707 '08

709

Y . B. Band and K. F. Freed, J. Chem. Phys., 1976, 63, 3382. M. J. Berry, Chern. Phys. Letters, 1974, 27, 73. H. Nakamura, Chem. Phys. Letters, 1975, 33, 151. H. Nakamura, Chem. Phys., 1975, 1 0 , 2 7 1 . M. Parikh, Phys. Rev. ( A ) , 1975, 12, 1872. M. Shapiro, Chem. Phys. Letters, 1977, 46, 442. S. Tsurubuchi, Chem. Phys., 1975, 10, 335. R. A. Rouse, J. Chem. Phys., 1976, 64, 1244. P. J. Bruna, S. D. Peyerimhoff, and R. J. Buenker, Chem. Phys., 1975, 10, 323.

85

Spectroscopic and Theoretical Aspects

formaldehyde,710and of the monoxide and carbide of platinum.711 Two studies of the electronic states of methylene produced in photodissociation reactions have been published.712 The importance of studies of laser-induced fluorescence has already been mentioned in the section on vibrational energy transfer. A brief review of the application of this technique to the study of product-state distributions has appeared.713 The trapping efficiency of photofragment ions formed by polarized and unpolarized radiation in an ion cyclotron resonance (ICR) cell has been analysed theoretically and it is shown how the kinetic energy and anisotropic angular distribution of fragment ions can be estimated using this ICR technique.'14 A number of very interesting studies of photodissociation processes have been published in which either the laser-induced fluorescence of cyanogen free radicals CN(X2Z+)or the fluorescence of excited cyanogen radicals CN(A211 and B2Z+) produced in the photodissociation of various cyanogen derivatives has been used to monitor the details of the energy partitioning. A study of CN(A2n -+ X 2 X + ) fluorescence produced in the photodissociation of HCN with Xe line (147 and 129 nm) excitation showed that the vibrational level populations in the CN(A21T)state showed a steady decrease as u' increased.'15 A study of the rotational distributions of CN(X2C+) radicals produced in the flash photodissociation of ClCN revealed a bimodal distribution peaking at very high rotational levels, K' z 66, and which may be a result of non-linearity in the photoexcited state of the molecule.71s A careful high-resolution study of the CN(B2X+ -+ X2E+) fluorescence produced on photodissociation of BrCN with noble-gas resonance radiation has revealed a population inversion of vibrational levels in CN(B2X+)for 123.6 nm excitation and the parallel formation It is shown that the monotonic decrease previously of the longer-lived A 2 n reported for the vibrational distribution is a result of a collisionally induced intersystem crossing process between (A),>lo and (B)w>olevels. Electronic branching in the photodissociation of ICN to give CN radicals in both the A and B states has previously been Recent studies of this system show that excitation of ICN in the A state continuum at h > 220nm produces CN(X2Z+) with little vibrational excitation 720 but with disequilibrated rotational distribution^.^^^ At higher photon energies, h > 145 nm, a population inversion is observed in the X state of CN as a result of a near-resonant collisioninduced energy transfer from u' = 0, 1 levels of the A state to u = 4,5 levels of Time-dependent laser-induced fluorescence measurements indicate the X that the quantum yields of the A and B states are comparable at the higher 71g9

'lo 711 712

713 714

716

?16 718 719

720

R. L. Jaffe and K. Morokuma, J. Chem. Phys., 1976, 64,4881. 0. N. Singh and B. P. Asthana, Acra Phys. Pol. ( A ) , 1977, 51, 575. (a)V. P. Zebransky and R. W. Carr, jun., J . Amer. Chem. SOC.,1976, 98, 1130; (6) T. H. Richardson and J. W. Simons, Chem. Phys. Letters, 1976, 41, 168. F. Engelke, Ber. Bunsengesellschaftphys. Chem., 1977, 81, 135. R. Orth, R. C. Dunbar, and M. Riggin, Chem. Phys., 1977, 19, 279. E. N. Tereshchenko and N. Ya. Dodonova, Optika i Spektroskopiya, 1976, 41, 495. M. J. Sabety-Dzvonik and R. Codey, J. Chem. Phys., 1976, 64,4794. M. N. R. Ashfold and J. P. Simons, Chem. Phys. Letters, 1977, 47, 65. C. K. Luk and R. Bersohn, J. Chem. Phys., 1973,58,2153. M. J. Sabety-Dzvonik and R. J. Codey, J. Chem. Phys., 1977, 66, 125. A. P. Baronavski and J. R. McDonald, Chem. Phys. Letrers, 1977, 45, 172.

86

Photochemistry

photon energy. A table of results collected from previous investigations of the ICN system highlights the conflicting nature of many of these A recent theoretical treatment of the A-system photodissociation 721 is able to reproduce the results of an experimental investigation in which the presence of both the A and X states of CN is indicated 722 and is therefore in disagreement with the two investigations of ICN mentioned above. A very interesting study of the rotational energy disposal in the monochromatic vacuum-ultraviolet photodissociation of the cyanogen halides under collision-free conditions shows that the level of rotational excitation in CN(B),,, increases in the order ICN < BrCN < ClCN, i.e. as the mass of the departing halogen atom decreases.723 It is suggested that the high level of excitation for ClCN may result from a strong repulsion on the final potential surface or from some non-linearity in the predissociating state of the molecule. The results are discussed within the framework of a simple Franck-Condon model in which rotational excitation is a consequence of a bending motion of a linear intermediate state. The model is able to account for the rotational energy partitioning observed in a study of (CN), photodissociation by 150 nm radiation.724 This latter study shows that (CN), predissociates to give equal amounts of the A and X states of CN, which indicates that there is a single primary process at this wavelength. Studies of the photofragmentation of methylaminoacetonitrile and acetoitri rile,^,^ of iodine bromide, for which a semiclassical theory of predissociation is developed,726of alkyl 727 and aryl halides,72sand of formaldehyde 729 have also been published.

7 Laser-induced Photochemistry The general area of laser-induced photochemistry continues to develop rapidly, particularly in the field of isotope separation. In many respects the theoretical description of these processes is incomplete, but the increasing number of systematic experimental investigations of particular systems that are appearing in the literature should soon help this situation. Laser-induced Dissociation.-Theoretical treatments of multiphoton molecular dissociation attempt to relate molecular parameters to the following observables: the threshold powers for the onset of photofragmentation and for the onset of saturation effects, the order of the multiphoton process below saturation, and the dependence of the fragmentation probability on the laser 731 In the intensity region above the onset of saturation effects it is predicted that is proportional to the laser intensity to the 3/2 power over a broad intensity range.732 A theory of the dissociation of polyatomic molecuIes by intense infrared laser radiation has been given in which vibrational heating of the whole 7a1 72a 723

724 725 726

727 728 729

730 731 73a

U. Halavee and M. Shapiro, Chem. Phys., 1977, 21, 105. J. H. Ling and K. R. Wilson, J. Chem. Phys., 1975, 63, 101. M. N. R. Ashfold and J. P. Simons, J.C.S., Faraday II, 1977, 73, 858. R. J. Codey, M. J. Sabety-Dzvonik, and W. M. Jackson, J. Chem. Phys., 1977, 66,2145. I. P. Vinogradov and F. E. Vilesov, Optika i Spektroskopiya, 1976, 40, 311. M. S. Child, Mol. Phys., 1976, 32, 1495. S. 5. Riley and K. R. Wilson, Faraday Discuss. Chem. SOC.,1972, No. 53, 132. M. Kawasaki, S. J. Lee, and R. Bersohn, J. Chem. Phys., 1977, 66, 2647. P. L. Houston and C. B. Moore, J. Chem. Phys., 1976, 65,757. S. Mukamel and J. Jortner, Chem. Phys. Letters, 1976, 40,150. S. Mukamel and J. Jortner, J. Chem. Phys., 1976, 65, 5204. S. Speiser and J. Jortner, Chem. Phys. Letters, 1976, 44, 399.

Spectroscopic and Theoretical Aspects

87

molecule follows the excitation of a particular infrared-active mode which is in near-resonance with the applied field.733 Conditions have been identified for which it is possible to neglect the effects of coherence in the multistep photodissociation of molecules and the multistep resonant photoionization of atoms and thus use simple rate equations to describe these processes.734The mechanism of the photodissociation of polyatomic molecules by intense infrared radiation .~~~~ has been discussed 736-757 as also has the case of diatomic m o l e c u l e ~739 Laser-induced photodissociation studies have been reported for the molecular hydrogen-D+ for ortho-iodine in ortho-para i o d i n e r n i x t ~ r e s , ~ for ~ l -boron ~~~ 745 for 748 for ethane,748, 749 for ammonia,75o and for various alkyl halides.751 There have been a number of investigations of the infrared laser-induced photodissociation of sulphur hexafluoride. In a study of the effect of the pulse duration on the dissociation probability it was found that the latter increased as the duration of the pulses, which had a fixed energy density, was decreased but not nearly as rapidly as had been predicted t h e ~ r e t i c a l l y . ~In~ ~another study of SF6 it was found that the fractional dissociation depended strongly on the total integrated laser Other studies have investigated the effect of pressure in the range 0.01-0.3 Torr on the dissociation rate 754 and have determined the threshold for the collisionless dissociation.755The threshold was found to be a laser pulse energy effect with the same value for SF6, SiFp, and CFBC12. 7479

Laser-induced Photoionization.- Mu1t iphot on ionization probabilities have been measured using pulses from single and multimode lasers;756a review of the 733 734 736

736

D. P. Hodgkinson and J. S. Briggs, Chem. Phys. Letters, 1976, 43, 451. J. R. Ackerhalt and J. H. Eberly, Phys. Rev. (A), 1976, 14, 1705. R. V. Ambartsumyan, Yu. A. Gorokhov, V. S. Letokbov, G. N. Makarov, and A. A. Puretskii, Pis’ma Zhur. eksp. teor. Fiz., 1975, 22, 374. V. M. Akulin, S. S. Alimpiev, N. V. Karlov, and L. A. Shelepin, Zhur. eksp. teor. Fiz., 1975, 69, 836.

737

R. V. Ambartsumyan, N. V. Chekalin, V. S. Letokhov, and E. A. Ryabov, Chem. Phys. Letters, 1975, 36, 301.

738

73* 740

741

74a 743 744 746

746

747 748

U. Devi and M. Mohan, Phys. Letters (A), 1975, 53, 421. L. C. M. Miranda, Phys. Letters (A), 1976, 56, 374. N. P. F. B. Van Asselt, I. G. Maas, and J. Los, Chem. Phys., 1975, 11, 253. V. S. Letokhov and V. A. Semchishen, Doklady Akad. Nauk S.S.R., 1975,222, 1071. V. S. Letokhov and V. A. Semchishen, Spectroscopy Letters, 1975, 8, 263. V. I. Balykin, V. S. Letokhov, V. I. Mishin, and V. A. Semchishen, .Chem.Phys., 1976,17, 111. V. N. Burimov, V. S. Letokhov, and E. A. Ryabov, J. Photochem., 1976,5,49. R. V. Ambartsumyan, V. S. Dolzhikov, V. S. Letokhov, E. A. Ryabov, and N. V. Chekalin, Zhur. eksp. teor. Fiz., 1975, 69, 72. N. V. Chekalin, V. S. Dolzhikov, V. S. Letokhov, V. N. Lokhman, and A. N. Shibanov, Appl. Phys., 1977, 12, 191. J. Tardieu de Maleissye, F. Lempereur, and C. Marsal, Chem. Phys. Letters, 1976, 42, 472. J. Tardieu de Maleissye, F. Lemperreur, and C. Marsal, Ber. Bunsengesellschaftphys. Chem., 1977, 81, 235.

J. Tardieu de Maleissye, F. Lempereur, C. Marsal, and R. I. Ben-Aim, Chem. Phys. Letters, 1976, 42, 46. 760 761 763

763 764

766 766

J. D. Campbell, G. Hancock, and J. B. Halpern, Opt. Comm., 1976, 17, 38. W. Braun and W. Tsang, Chem. Phys. Letters, 1976,44, 354. P. Kolodner, C. Winterfeld, and E. Yablonovitch, Opt. Comm., 1977, 20, 119. J. D. Campbell, G. Hancock, and K. H. Welge, Chem. Phys. Letters, 1976, 43, 581. J. Dupre, P. Pinson, J. Dupre-Maquaire, C. Meyer, and P. Barchewitz, Compt. rend., 1976, 283, C, 311. M. C. Gower and K. W. BiIlman, Appl. Phys. Letters, 1977, 30, 514. T. U. Arslanbekov, N. B. Delone, A. V. Masalov, S. S. Todirashku, and A. G. Fainshtein, Zhur. eksp. teor. Fiz., 1977,72, 907.

88

Photochemistry

multiphoton ionization of atoms has been published 757 and the selectivity of the two-step photoionization of atoms has been The inadequacy of the momentum-translation approximation has been discussed with reference to the multiphoton ionization of atomic hydrogen 759 as have the differential crosssections for such processes.76o Multiphoton ionization experiments have shown polarization effects in a study of alkali singly and doubly charged ions have been produced in atomic A strong two-photon resonance effect has been studied in strontium v a p o ~ r as , ~have ~ ~ high-lying Rydberg and valence states in atomic uranium.764 The two-step (sequential) photoionization of formaldehyde has been and a multiphoton ionization technique has been used in measurements on benzene.786 Laser-induced Chemical Reactions.-A number of reviews of laser-induced reactions have a p p e a ~ e d . ~ A ~ ~theoretical - ~ ~ ~ treatment of the intermediatecoupling case has been published which gives an approximate method for calculating relative reaction rates of laser-catalysed reactions,770and a treatment of laser-induced rate processes in polyatomic gases 771 has also been published which differs substantially from a previously mentioned discussion of the ,~~~ Studies of laser-induced reactions of boron c o i n p o ~ n d s methanol-bromine mixtures,773excited ethylene and excited carbonyl sulphide (with atomic oxygen),774 excited ethylene,775 chlorofluorohydrocarbons,776nitric tetrafluoroh y d r a ~ i n e , ~ ~complexes ~ - ~ ~ O in the liquid phase,781and uranium hexafluoride 782 have been reported. There has also been an interesting report of the laserinduced isomerization of hexafluorocyclobutane to the less thermodynamically stable isomer h e x a f l u ~ r o b u t a d i e n e . ~ ~ ~ Laser Isotope Separation.-Several reviews of isotopically selective photochemistry have been p ~ b l i s h e d . ~A~ ~number - ~ ~ ~ of specific isotope separation procedures have been described,788and the mechanism of selective dissociation of polyatomic molecules by intense CO, laser pulses has been 767

J. S. Bakos, A h . Electron. Electron Phys., 1974, 36, 57.

758

N. V. Karlov, Yu. B. Konev, and A. M. Prokhorov, Kuantouaya Elektron. (Moscow), 1975,

759

750

761 762 763

764

765

766

767 758 769 770

771 772

773

774

775

2, 2453. A. Rachman and G. Laplanche, J . Phys. (B), 1975, 8, 826. Y. Gontier, N. K. Rahman, and M. Trahin, J . Phys. (B), 1975, 8, L179. G . A. Delone and N. L. Manakov, Zhur. eksp. teor. Fiz., 1976, 70, 1234. 1. S. Aleksakhin and I. P. Zapesochnyi, Ukrain. f i z . Zhur, 1976, 21, 1383. P. P. Sorokin, J. J. Wynne, J. A. Armstrong, and R. T. Hodgson, Ann. New York Acad. Sci., 1976, 267, 30. R. W. Solarz, J. A. Paisner, and L. R. Carlson, Opt. Comm., 1976, 18, 29. S. V. Andreev, V. S. Antonov, I. N. Knyazev, and V. S. Letokhov, Chem. Phys. Letters, 1977, 45, 166. P. M. Johnson, J. Chem. Phys., 1976, 64, 4143. A. M. Ronn, Spectroscopy Letters, 1975, 8, 303. F. Klein, F. M. Lussier, and J. I. Steinfeld, Spectroscopy Letrers, 1975, 8, 247. A, K. Petrov and V. P. Chebotaev, Khim. Plazmy, 1975, 2, 267. S. S. Bauer and K.-R. Chien, Chem. Phys. Letters, 1977, 45, 529. M. F. Goodman, J. Stone, and D. A. Dows, J . Chem. Phys., 1976, 65, 5052. H. R. Bachmann, H. Noeth, R. Rinck and K. L. Kompa, Chem. Phys. Letters, 1975, 33, 261. C. Willis, R. A. Back, R. Corkum, R. D. McAlpine, and F. K. McClusky, Chem. Phys. Letter.7, 1976, 38, 336. R. G. Manning, W. Braun, and M. J. Kurylo, J . Chem. Phys., 1976, 65, 2609. J. Bastiaens, D. De Keuster, and X. De Hemptinne, Bull. SOC.chirn. Belges, 1976, 85, 833.

Spectroscopic and Theoretical Aspects

89

as has the use of tunable dye lasers 790 and two-frequency infrared laser fields 791 in selective excitation studies. Laser-induced radiative preassociation and isotope separation has been described 792 and a unimolecular decomposition scheme has been used to calculate isotope enrichments following infrared laser excitation.793 A large isotope effect has been found for a low-temperature reaction 794 and an electric field isotope separation of the ions formed by infrared laser excitation of polyatomic molecules has been described.795New photochemical schemes for isotope separation have appeared 797 and a general discussion of laser isotope separation has been Reports of isotope separation experiments on boron,8oo carbon,801-ao4 n i t r ~ g e n , ~ ~ ~ - ~ ~8os ~ and chlorine have 798s

D. F. Dever and E. Grunwald, J. Amer. Chem. SOC.,1976, 98, 5055. J. C. Stephenson and S. M. Freund, J. Chem. Phys., 1976, 65, 1893. 778 V. I. Ksenzenko, A. V. Pankratov, and V. M. Shabarshin, Trudy Mosk. Inst. Tonkoi Khim. Tekhnol., 1975, 5, 53. 77s V. V. Gorlevskii, A. N. Oraevskii, A. V. Pankratov, A. N. Skachkov, and V. M. Shabarshin, Khim. vysok. Energii, 1976, 10, 443. 7 8 0 A. N. Oraevskii, A. V. Pankratov, A. N. Skachkov, and V. M. Shabarshin, Khim. vysok. Energii, 1977, 11, 152. 781 N. V. Karlov, N. P. Stupin, and L. A. Shelepin, Pis’ma Zhur. tekh. Fiz., 1976, 2, 828. 78a A. M. Ronn and B. I. Earl, Chem. Phys. Letters, 1977, 45, 556. 783 A. Yogev and R. M. J. Benmair, Chem. Phys. Letters, 1977,46,290. 784 V. S. Letokhov and C. B. Moore, Kvantovaya Electron. (Moscow), 1976, 2, 248,485. 786 V. S. Letokhov, Springer Ser. Opt. Sci., 1976, 3, 122. 786 R. V. Ambartsumyan and V. S. Letokhov, Accounts Chem. Res., 1977, 10, 61. m7 R. V. Ambartsumyan and V. S. Letokhov, Laser Focus, 1975, 11, 48. 788 R. C. Stern and B. B. Snavely, Ann. New York Acad. Sci., 1976, 267, 71. 78s R. V. Ambartsumyan and V. S . Letokhov, Comments At. Mol. Phys., 1976, 6, 13. 790 V. S . Letokhov, Spectroscopy Letters, 1975, 8, 697. 7s1 R. V. Ambartsumyan, Yu. A. Gorokhov, N. P. Furzikov, V. S. Letokhov, G. N. Makarov, and A. A. Puretskii, Pis’ma Zhur. eksp. teor. Fiz., 1976, 23, 217. 7uZ C. Schmidt, Chem. Phys. Letters, 1976, 37, 574. 7s3 K. C. Kim and J. M. McAfee, Chem. Phys. Letters, 1977,45235. 7 @ 4 A. McNeish, M. Poliakoff, K. P. Smith, and J. J. Turner, J.C.S. Chem. Comm., 1976, 21, 776 777

859.

V. M. Akulin, S. S. Alimniev, N. V. Karlov, N. A. Karpov, Yu. N. Petrov, and A. M. Prokhorov, Pis’ma Zhur. eksp. teor. Fiz., 1975, 22, 100. 796 C. T. Lin, Spectroscopy Letters, 1975, 8, 901. 797 I. Glatt and A. Yogev, J. Amer. Chem. SOC.,1976, 98, 7087. 798 X. De Hemptinne, Rev. Quest. Sci., 1976, 147, 315. 79e J. B. Marling, Chem. Phys. Letrers, 1975, 34, 84. S. M. Freund and J. 3. Ritter, Chem. Phys. Letters, 1975, 32, 255. J. H.Clarke, Y. Haas, P. L. Houston, and C. B. Moore, Chem. Phys. Letters, 1975, 35, 82. m 2 J. B. Marling, Opt. Comm., 1976, 18, 36. J. J. Ritter and S. M. Freund, J.C.S. Chem. Comm., 1976, 20, 811. R. R. Karl, jun. and K. K. Innes, Chem. Phys. Letters, 1975, 36, 275. 8os J. C. D. Brand, J. L. Hardwick, and K. E. Teo, Cunad. J. Phys., 1976, 54, 1069. H. Noguchi, Y. Izawa, and C. Yamanaka, Technol. Reports Osaka Univ., 1976, 26, 151. J. Dupre, P. Pinson, J. Dupre-Maquaire, C. Meyer, and P. Barchewitz, Compt. rend., 1976, 282,B, 357. R. V. Ambartsumyan, Yu. A. Gorokhov, V. S. Letokhov, G. N. Makarov, and A. A. Puretskii, Zhur. eksp. teor. Fiz., 1976, 71, 440. 809 D. R. Keefer, 5. E. Allen, jun., and W. B. Person, Chem. Phys. Letrers, 1976, 43, 394. P. Fettweis and M. Neve de Mevergnies, Appl. Phys., 1977, 12, 219. W. Fuss and T. P. Cotter, Appl. Phys., 1977, 12, 265. A. K. Petrov, Yu. N. Samsonov, A. V. Baklanov, V. V. Vizhin, and A. M. Orishich, Izvest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 2148. J. Stone, M. F. Goodman, and D. A. Rows, J. Chem. Phys., 1976, 65, 5062. G. Hancock, J. D. Campbell, and K. H. Welge, Opt. Comm., 1976, 16, 177. A. Yogev and R. M. J. Benmair, J. Amer. Chem. SOC.,1975, 97, 4430. 7D6

Photochemistry

90

appeared. Studies of the heavier elements uranium,816-s1s osmium,s1B samarium,820and europium have also been reported. Isotopically selective excitation of fluorescence in copper(1) iodide has been observed.8zz Other Isotope Separation Techniques. An isotope separation scheme has been proposed in which resonant microwave radiation can preferentially lead to photochemical reaction of the triplet state of molecules containing particular A scheme based on photoinduced changes in the electric and magnetic properties of molecules and atoms has been described 824 and the photocatalysed carbon isotope exchange between phosgene and carbon monoxide was found to result in enrichment factors greater than those expected t h e o r e t i ~ a l l y . ~ Large ~~ isotope effects in the photodissociation of polyatomic molecules have been discussedazs and an interesting scheme for producing heavy water has been described which involves an isotopically selective photodesorption process.827 Laser-induced Effects.-A theoretical model for the fluorescence excited in gases by continuous infrared excitation has been givensz8 and the kinetics of the threshold luminescence produced by the resonant COz laser excitation of gaseous sulphur hexafluoride and boron trichloride have been discussed.82BCO, laserexcited infrared emission in low pressure gases 830 and HF laser-excited visible luminescence in NH3, MeOH, and COZa3lhave been observed, and the effect of non-exciting radiation on the fluorescence of organic molecules in the vapour phase has been Laser-induced intermolecular electronic energy transfer has been discussed and observed in sulphur methyl fluoride-nitrogen dioxide and in a helium glow discharge.83s The enhancement of luminescent yields by pulsed laser excitation 837 and dissociative excitation transfer from sulphur hexafluoride following pulsed infrared excitation 838 have both been reported. *17

820

821

822

823 824

82b 826 827 828

829

830 831

L. J. Radziemski, jun., S. Gerstenkorn, and P. Luc, Opt. Comm., 1975, 15, 273. T. Kasuya and K. Shimoda, Kagaku, Kogaku, 1975,39, 353. T. E. Eriksen, Kem. Tidskr., 1976, 88,44. R. V. Ambartsumyan, Yu. A. Gorokhov, G. N. Makarov, A. A. Puretskii, and N. P. Furzikov, Chem. Phys. Letters, 1977, 45, 231. N. V. Karlov, B. B. Krynetskii, V. A. Mishin, A. M. Prokhorov, A. D. Savel’ev, and V. V. Smirnov, Kvantovaya Elektron. (Moscow), 1976, 3, 2486. N. V. Karlov, B. B. Krynetskii, V. A. Mishin, A. M. Prokhorov, A. D. Savel’ev, and V. V. Smirnov, Pis’ma Zhur. tekh. Fiz., 1976, 2, 961. C. R. Wu and D. A. Dows, J. Mol. Spectroscopy, 1975, 58, 384. T. Akaheh and A. M. El-Sayed, J. Phys. Chem., 1976, 80,2710. P. L. Kelley, N. M. Kroll, and C. K. Rhodes, Opt. Comm., 1976, 16, 172. Z. B. Vukmirovic and S. V. Ribnikar, J. Chem. Phys., 1977, 66, 7. Y. B. Band and K. F. Freed, J. Chem. Phys., 1975, 63,4479. T. Gangwer and M. K. Goldstein, ERDA Energy Res. Abstr., 1977, 2, Abst. No. 6720. R. R. Parenti, J. F. Whitney, and J. 0. Artman, J. Chem. Phys., 1975, 62, 3955. A. N. Oraevskii, A. V. Pankratov, A. N. Skachkov, and G. V. Shmerling, Kvantovaya Electron. (Moscow), 1975, 2, 1725. R. T. Bailey and F. R. Cruickshank, J. Phys. Chem., 1976, 80, 1596. B. K. Deka, P. E. Dyer, and D. J. James, Phys. Letters (A), 1977, 61, 30. V. L. Bogdanov, B. S. Neporent, and V. P. Klochkov, Izvest. Akad. Nauk S.S.S.R., Ser. jiz., 1975,39, 2295.

833 834

836 836 837

838

J. Jortner and A. Ben-Reuven, Chem. Phys. Letters, 1976, 41, 401. B. L. Earl, A. M. Ronn, and G. W. Flynn, Chem. Phys., 1975, 9, 307. S. M. Lee and A. M. Ronn, Spectroscopy Letters, 1975, 8, 915. A. Catherinot, B. Dubreul, A. Bouchoule, and P. Davy, Phys. Letters (A), 1976, 56, 469. E. F. Wyner, J. A. Sousa, and J. F. Roach, Spectroscopy Letters, 1975, 8, 419. B. J. Orr and M. V. Keentok, Chem. Phys. Letters, 1976, 41, 68.

Spectroscopic and Theoretical Aspects

91

I would like to thank Dr. B. Brocklehurst for helpful discussions about his own work and that of others on magnetic field effects and also to thank the numerous people who sent reprints of their work. I am indebted to Dr. D. Phillips for all the encouragement he has given me during the transition period in the writing of this Chapter; it was very much appreciated.

2 Photophysical Processes in Condensed Phases BY R. B. CUNDALL AND M. WYN-JONES

1 Introduction For the sake of continuity we have essentially, followed the successful format developed and presented by previous Reporters in earlier volumes. Restrictions in space have permitted only short comments on most papers. Where overlap occurs with the chapter on Physical Aspects of Photochemistry, readers should consult that chapter for further details. Some papers have been included in which means other than optical excitation have been used to produce excited states. 2 Excited Singlet-state Processes Birks has discussed the anomalies in radiative lifetimes of aromatic molecules in solution, in which measured values do not appear to agree with the theoretical values calculated from absorption measurements.2 Such anomalies are expressed by a factor R, which is the ratio of theoretical to experimental lifetime, and the examples discussed include benzene, trans-stilbene, diphenyl-polyenes, retinolpolyenes, biphenyl, and fluoranthene. Forster, in his early calculations of radiative lifetimes, assumed mirror symmetry between fluorescence and absorption, but Birks points out that such estimates are significantly altered in the case where mirror symmetry is only approximate. For example, in the case of biphenyl and fluoranthene, the weak So S1absorption is partially obscured by the adjacent strongly absorbing So -+ S2absorption, and it is only on correct evaluation of the extinction coefficient for the relevant So S, transition that agreement between theory and experiment is achieved. The experimental results for benzene, however, do not allow easy interpretation of the S, rate parameters, where fluorescence and intersystem crossing, as well as internal conversion (Channel 3), play a part in determining the overall behaviour. Fischer and Naaman3 have proposed a three-state model for the sharp and large fluctuations in line widths in electronic spectra of benzene. The Channel 3 radiationless transition has been explained on the basis of the energy dependence of the line broadening in the lBzu state. These predictions, however, appear to be at variance with experimental results on values of the triplet yields. Vacuum-u.v. excitation of benzene in rare-gas and nitrogen matrices at liquid-helium temperatures * reveals phosphorescence and fluorescence. Shifts in emissions are observed from one matrix to another, and they depend upon the vibronic bands in the case of fluorescence. A review on the --f

--f

1 1

a

J. B. Birks, 2. phys. Chem. (Frankfurt), 1976, 101, 91. S . Lipsky, J. Chem. Phys., 1976, 65, 3799. G. Fischer and R. Naaman, Chem. Phys. Letters, 1976, 42, 581. L. Hellner and C. Vermeil, J. Mol. Spectroscopy, 1976, 62, 313.

92

Photophysical Processes in Condensed Phases

93

state of the art in benzene photophysics up to the middle of 1975 has recently a ~ p e a r e d .In ~ order to obtain reliable experimental values for rate constants, a knowledge of quantum yields for fluorescence and phosphorescence is required. Birks gives a very useful and critical assessment of the use of various compounds as fluorescent standards in the estimation of quantum yields. Quinine bisulphate in sulphuric acid appears to be widely accepted as a fluorescent standard, as originally suggested by Melhuish, although some authors choose to use 0.1NH,SO, as solvent instead of 1N-H2S04. This variation requires care, since there is some experimental evidence to suggest that the values of the quantum yield increase on increasing the concentration of sulphuric acid in the aqueous solvent. Birks has also indicated pitfalls in the elucidation of the photochemical process schemes, and points out the advantages of S-T perturbation studies, especially using xenon, for this purpose. El-Bayoumi8 has discussed recent views on the various relaxation processes of excited molecules. In particular, he considers intramolecular geometric relaxation, intramolecular excimer formation, solventcage relaxation, and proton transfer, all of which occur on the nanosecond timescale. A detailed review of the theoretical and various aspects of lifetimes of atoms, molecules, and ions has also recently a ~ p e a r e d .Although ~ dealing with data on gas-phase systems, the discussion of new techniques makes it valuable for all those interested in measurement of lifetimes. The use of fluorescence techniques in liquid l1 as well as providing a powerful means for the examination of t.1.c. plates,l2Pl3has provided increased sensitivity and rapidity of analysis of a wide range of materials. Guilbault,13in a timely review on a rapidly expanding field, has given an account of the recent uses of fluorescence in the assay of substances directly on solid surfaces such as paper chromatograms, thin-layer chromatography (t.1.c.) plates, potassium bromide disks, and electrophoresis strips and silicone-rubber pads. The use of fluorescent techniques in molecular biology appears to be an ever more rapidly expanding field. The use of fluorescent probes in membrane studies 1 4 and the photochemistry of nucleic acids l5 are areas of considerable and topical interest to molecular biologists. Froehlich l6 has written a review paper which gives much insight into some luminescence assays available in clinical and biological analysis, there the assay requirements are for rapid, sensitive techniques and where the use of microsamples is an obvious necessity. These and other aspects of luminescence are reviewed in two recent twin-volume works edited by Wehry 170 and by Chen and Ede1h0ch.l~~ 6

7

10

11 l2

l3 15

17

R. B. Cundall, D. A. Robinson and L. C. Pereira, Adv. Photochem., 1977, 10, 174. J. B. Birks, J. Res. Nat. Bur. Standards, 1976, BOA, 389. J, B. Birks, Photochem. and Photobiol., 1976, 24, 287. M. A. El-Bayoumi, J. Phys. Chem., 1976, 80,2259. R. E. Imhof and F. H. Read, Reports Progr. Physics, 1977, 40,1. W. Slavin, A. T. Rhys Williams, and R. F. Adams, J. Chromatog., 1977, 134, 121. E. Johnson, A. Abu-Shumays, and S. R. Abbott, J. Chromarog., 1977, 134, 107. J. C. Young, J. Chromatog., 1977, 130, 392. G. G. Guilbault, Photochern. and Photobiol., 1977, 25, 403. G. Strauss, Photochem. and Photobiol., 1976, 24, 141. G. Lober and L. Kittler, Photochem. and Photobiol., 1977, 25, 215. P. Froehlich, Appl. Spectroscopy Rev., 1976, 12, 83. (a) ‘Modern Fluorescence Spectroscopy,’ ed. E. L. Wehry, Plenum Press, New York, 1976 vols. 1 and 2; (b) ‘Biochemical Fluorescence Concepts,’ ed. R. F. Chen and H. Edelhod, Marcel Dekker Inc., New York, 1975, Vol. 1 and 1976, Vol. 2.

94

Photochemistry

Anthracene still continues to provide interest to photocheniists, as evidenced by the appearance of several publications on various aspects of the luminescence of anthra~ene.l~-~* Yakhot l8, l9 provides a quantitative interpretation of the known fluorescence quenching data of anthracene in various solvents. The decrease of quantum yield at high temperature has been explained in terms of the S, T2 transition only, rather than the previously held view of the S, -+T3 transition. Bale et aI.20have measured the lifetime of anthracene emission at low temperature, and found it to be less than 1.9 ns at 5.4 K. This value is certainly lower than previously reported values, but Bale et al. account for this in terms of a super-radiant fluorescence arising from free excitons. Decay times of the fluorescence of naphthalene crystals doped with small amounts of anthracene as functions of both temperature and guest concentration have been studied by Kohler.21 The results are shown in Figure 1. Kohler et aI. have proposed a simple exciton hopping model to explain the experimental results, as an alternative to the long-range resonance-transfer theory. Kobayashi et aZ.22have obtained data on triplet-singlet intersystem crossing from an excited triplet state T, (n 3 2) to the S, state for meso-substituted derivatives of anthracene by means of a double excitation method which utilizes two excitation pulses of light fired at appropriate time intervals; a technique that was developed to investigate the mechanisms of the biphotonic excitation of fluorescence. Intramolecular quenching properties of pyridinium and N-methylimidazolium cations linked to an anthracene chromophore both directly and via a methylene bridge have been The pyridinium ions provide the stronger electron-accepting pro pert ies. Papers have appeared on the photophysics and photokinetics of the anthraceneNN-diethylaniline exciplex,24 anthracene-NN-dimethyIaniline (DMA) exci~~ plex,29-32and on anthracene-trans,trans-2,4-hexadiene e ~ c i p l e x . Nishimura et al. have studied the build-up of the concentration of the anthracene triplet state by means of nanosecond laser photolysis and measurements of transient T, -+ TI absorption processes: the triplet state of anthracene is produced from the relaxed exciplex fluorescent The lifetime of the anthracene-DMA exciplex is shortened as the concentration of DMA is increased, the possibility being that a triplex is f ~ r m e d ,which ~ ~ , ~arises ~ from interaction of a second --f

l8 l9 2o 21 2a 2s

24 26

2e

27

V. Yakhot, Chem. Phys., 1976, 14,441. V. Yakhot, Chem. Phys. Letters, 1976, 40, 304. A. G. Bale, N. J. Bridge, and D. B. Smith, Chem. Phys. Letters, 1976, 42, 166. M. Kohler, D. Schmid, and H. C. Wolf, J. Luminescence, 1976, 14, 41. S. Kobayashi, K. Kikuchi, and H. Kokubun, Chem. Phys. Letters, 1976, 42, 494. G. M. Blackburn, G. Lockwood, and V. Solan, J.C.S. Perkin ZZ, 1976, 1452. T. Nishimura, N . Nakashima, and N . Mataga, Chem. Phys. Letters, 1977, 46, 334. Ch. Jung and K. H. Heckner, Chem. Phys., 1977, 21, 227. D. Gupta and S. Basu, J. Photochem., 1976, 6, 145. D. :J. Mitchell, ;G. B. Schuster, and H. G. Drickamer, J. Amer. Chem. SOC.,1977, 99, 1145.

2e

ae

K. K. Rohatgi-Mukherjee and A. K. Gupta, Chem. Phys. Letters, 1977, 46, 368. 5. Saltiel, D. E. Townsend, B. D. Watson, P. Shannon, and S. L. Finson, J. Amer. Chem. Soc., 1977, 99, 884.

ao 81

aa

Man-Him Hui and W. R. Ware, J. Amer. Chem. SOC.,1976,98,4712. Man-Him Hui and W. R. Ware, J. Amer. Chem. SOC.,1976, 98,4718. N. C. Yang, D. M. Shold, and B. Kim, J. Amer. Chem. SOC., 1976,98, 6587. B. D. Watson, D. E. Townsend, and J. Saltiel, Chem. Phys. Letters, 1976, 43, 295.

Pho tophysical Processes in Condensed Phases

95

5z

!?

z

i W

K

50

100

150

200

TIME [ns) 1.0

-

0.75

e 5

0.5

Figure 1 Decay times of the fluorescence of (a) a pure naphthalene crystal at 4.2 K, mol of anthracene per mol of naphthalene (b) a naphthalene crystal doped with 8 x mol of anthracene per mol of at 300 K, (c) a naphthalene crystal doped with 8 x naphthalene at 4.2 K (Reproduced by permission from J. Luminescence, 1976, 14, 41)

96

Photochemistry

quencher molecule with the exciplex. Transient effects on the exciplex photokinetics associated with diffusion play an important part in a number of quenching ~ y s t e m s 31 . ~A ~ ~weak transient fluorescence (@* z 0.0036 and T x 27 ns) has mol I-l) and trans,transbeen detected from solutions of anthracene (6.7 x 2,4-hexadiene (1.76 moll-l) by employing single-photon counting. The emission has been tentatively assigned to the anthracene-l,3-diene exciplex. Hochstrasser and Nelson 34 have investigated the energy transfer between 1,2-benzanthracene molecules in highly viscous solvent by observing the polarized absorption of the S,state, using picosecond laser pulses for excitation. The results show that a

WAVELENGTH (nm) Figure 2 Fluorescence of anthracene (0.2 and its excimer in CHCI, at 32 "C. (Reproduced by permission from J. Chern. Phys., 1976, 65, 3375)

concentration-dependent depolarization of the spectra is dominant at concentrations higher than 3.5 x lo-, moll-l: the effect is indicative of energy transfer between 1,2-benzanthracene molecules. The influence of substituents on the internal and heavy-atom-induced intersystem crossing for six 9,lO-substituted anthracenes in heptane has been described by Jung and Heckner.26 Excimer emission from anthracene26 has been postulated to be above 450 nm, after measurements of the depolarization factors above and below 450 nm had been made. McVey et aL3!jhave reported the first direct observation of anthracene dimer emission and its characterization under ordinary experimental conditions. The spectrum is shown in Figure 2. The excimer of the anthracene molecule has a peak at 540 k 15 nm, a quantum yield of 0.010, and a lifetime of 1.45 ns. The effect of pressure on the fluorescence of 9-carbonyl-substituted anthracene 27 and the results of anthracenesulphonate-sensitized photo-oxidation of iodide ion28 have also been presented. Emission spectra from the higher excited states of tetracene excited by stepwise two-photon excitation have been presented by Nickel and Roden.36 The latter paper is particularly interesting in that emissions from the upper excited singlet states of aromatic hydrocarbons excited by this method have not apparently been reported hitherto. s1 s6

a6

R. M. Hochstrasser and A. C. Nelson, Optics Comm., 1976, 18, 361. J. K. McVey, D. M. Shold, and N. C. Yang, J. Chem. Phys., 1976, 65, 3375. B. Nickel and G. Roden, Ber. Bunsengesellschaft phys. Chem., 1977, 81, 281.

Photophysfcal Processes in Condensed Phases

97

Azulene still continues to retain its unique interest to spectroscopists, even after 20 years of detailed study, owing primarily to the readily observed S, -+ So fluorescence, and also because of the rapid relaxation from the S1state. Gillispie and LimS7have provided resolved spectra and values of quantum yield for the S, + So fluorescence of azulene. They quote a value of 4 x s-l for the radiative probability of this transition for azulene in a methylcyclohexane glass at 77 K, based on an assumed quantum yield of the S2-+ Sofluorescence of 0.03. Measurements of relaxation times from the first excited state of azulene have also been r e p ~ r t e d . ~Heritage ~ - ~ ~ and Penzkofer 3s found the relaxation time of azulene to be less than 2 p s when excited at different wavelengths, and Ippen et aZ.*O quote a similar lifetime of 1.9 ps in four different solvents; Wirth et aL41 have postulated a value of less than 1 ps. These values are much less than previously cited relaxation times of ca. 8 ps, and possibly reflect the improvements in instrumental set-up and resolution capabilities for this type of experiment. Marinari and Saltiel 42 have described a technique for the separation of radiative and non-radiative energy transfer in a trans-stilbene-azulene system. Front-face emissions from a cell containing high donor and acceptor concentrations are used, and the distortion of the donor emission spectrum due to radiative energy transfer is measured. This technique is considered useful in systems where, due to either short donor lifetimes or small spectral overlaps, energy transfer occurs only at high acceptor concentrations. The dipole moments of the two lower-lying singlet states of azulene and 3,5-dimethylcyclopenta[eflheptalene have been determined,43and Holzworth et al.44have studied the S3 -+So fluorescence of azuleno[5,6,7-~d]phenalene. Interest in the observation and characterization of the emission from the second excited singlet state is further shown in the case of [18]annulene 46 and of monofiuoro[l8]annulene. The observation of highly resolved spectra in Shpolski matrices and short (subnanosecond) lifetimes, as described in these papers, is especially noteworthy. Absorption and polarized emission studies of bridged [14]annulenes with an anthracene perimeter are also presented in a paper by Kolo et a1.“ The photophysical processes in stilbenes have been investigated theoretically and experimentally. Bartocci and Mazzucato4* have reported on the effect of excitation energy on the emission properties of trans-azastilbenes. The authors discuss an anomaly in the fluorescence excitation behaviour of 3-styrylpyridines (StP’s) and 3,3’-dipyridylethylenes (DPE’s) which differ from the other isomers of (StP) and (DPE). For trans-stilbene and 2- and 4-StP’s the excitation spectrum is a replica of the absorption spectrum. For 3-StP there was a distinct difference in the spectral profiles, similar behaviour being observed in acetonitrile and in 469

G. D. Gillispie and E. C. LimyJ . Chem. Phys., 1976, 65,4314. W. Kaiser, Optics Commun.,1976, 18, 197. 39 J. P. Heritage and A. Penzkofer, Chem. Phys. Letters, 1976, 44, 76. 4 0 E. P. Ippen, C. V. Shank, and R. L. Woerner, Chem. Phys. Letters, 1977,46, 20. 41 P. Wirth, S. Schneider, and F. Don, Chem. Phys. Letters, 1976,42,482. 4a A. Marinari and J. Saltiel, Mol. Photochem., 1976, 7 , 225. 48 H. Yamaguchi, T. Ikeda, and H. Mametsuka, Bull. Chem. SOC. Japan, 1976,49, 1762. I4 A. R. Holzwarth, K. Razi Naqvi, and U. P. Wild, Chem. Phys. Letters, 1977, 46, 473. 4s U. P. Wild, H. J. Griesser, and V. D. Tuan, Chem. Phys. Letters, 1976,41, 450. (6 U. P. Wild, Chimiu (Switz.), 1976, 30, 382. 47 J. Kolo, J. Michl, and E. Vogel, J. Amer. Chem. Soc., 1967, 98, 3935. 48 G. Bartocci and U. Mazzucato, Chem. Phys. Letters, 1977, 47, 541.

98

Photochemistry

water-ethanol Sol50 (v/v) at pH values above 7, where 3-StP is in the neutral form (pKB = 4.8). In acidic medium at pH < 3, where 3-StP is protonated, the excitation spectrum was coincident with the absorption found for stilbene and other isomeric StP's. The difference in spectral shape is regarded as being due to an allowed transition to an upper excited singlet state. Table 1 presents the fluorescence quantum yields of trans-stilbene and 3-StP in n-hexane and of neutral and protonated trans-3-StP and 3,3'-DPE in water-ethanol at various excitation wavelengths. Absorption and emission spectra of arylethylenes in the dissolved and crystalline state have been Information relating to transitions of even parity in stilbene solutions in the temperature range -40 to 20 "C has been obtained by using a probe beam method of time resolved spectroscopy 6o and by polarized dependent two photon excitation (TPE) at room temperature.61 Further examples of emission from the Sa excited states of molecules have been given for xanthione 62 and f l u ~ r a n t h e n e . ~ Transient ~ absorption spectra were recorded at times subsequent to excitation, and the spontaneous radiative lifetime of xanthione S2 fluorescence was given as 12 k 3 ps. Two distinct absorption bands were seen, the early time band, centred at 620 nm, being attributed to S2absorption whereas at longer times a band attributed to TIabsorption appears, centred at 540 nm. Similarly, fluoranthene has an S, fluorescence band centred at 450 nm, with a typical lifetime of 58 ns, whereas S2 emission occurs in the region 370-395 nm, with a typical lifetime of 143 ns. Dual fluorescence in some N-monosubstituted salicylamides64 in alcoholic solvents has been reported, whereas single fluorescence is observed in other solvents. The duality of fluorescence was ascribed to two different molecular species, one with and the other without intramolecular proton transfer in the lowest excited singlet state. The use of two-photon absorption in toluene,66 1,6-diphenylhexatriene, 1,8-diphenylo~tatetraene,~~ and naphthalene has given further information relating to the higher excited states. Lin and Topp6* used consecutive twophoton laser excitation to populate the higher singlet states of the xanthene dyes rhodamine 6G and rhodamine B. Three prominent emission bands were seen in the range 240-400 nm, the overall shape of the spectra being dependent on excitation wavelength and polarization. Rubrene was also studied. The use of two-photon absorption in the lifetime measurement 69 of triphenylmethane and indigo dyes 6o in ethanol give values of 5 and 17 ps respectively. The lifetimes of the excited states of 27 cyanine dyess1 in solution have been studied, under 49

so 61 6a

s5 66

57 6* 68

Ch. Goedicke and H. Stegemeyer, 2. phys. Chem. (Frankfurt), 1976, 101, 181. E. Heumann, W. Triebel, R. Uhlmann, and B. Wiehelmi, Chem. Phys. Letters, 1977, 45, 425. R. 5. M. Anderson, G. R. Holtom, and W. M. McClain, J. Chem. Phys., 1977, 66, 3832. R. W. Anderson, R. M. Hochstrasser, and H. J. Pownall, Chem. Phys. Letters, 1976,43,224. D. L. Philen and R. M. Hedges, Chem. Phys. Letters, 1976,43, 358. M. Kondo, Bull. Chem. SOC.Japan, 1976,49,2679. G. Beck and 5. K. Thomas, J.C.S. Faraday I, 1976,72,2610. G. R. Holtom and W. M. McClain, Chem. Phys. Letters, 1976, 44, 436. R. M. Hochstrasser and H. N. Sung, J. Chem. Phys., 1977,66,3276. H.-B. Lin and M. R. Topp, Chem. Phys. Letters, 1977, 47, 442. P. Wirth, S. Schneider, and F. Don, Chem. Phys. Letters, 1976, 42, 482. P. Wirth, S. Schneider, and F. Don, Optics Comm., 1977, 20, 155. L. D. Derkacheva, V. A. Petukhov, and E. G. Treneva, Optics and Spectroscopy, 1976,41,574.

-

Table 1 Fluorescence quantum yields of trans-stilbene and 3-StP in n-hexane and of neutral and protonated trans-3-StP and 3,3'DPE in water-ethanol at various values of A, n-Hexane

I

u n m 260 270 280 290 300 310 315

stilbene 0.038 0.038 0.036 0.035 0.035 0.036 0.035

3-StP 0.076 0.073 0.064 0.052 0.054 0.065 0.066

Water-ethanol 50/50 (v/v) r

3-StP (PH 8) 0.037 0.037 0.030 0.025 0.020 0.030 0.031

3-StPH+ (PH 2). 0.21 0.21 0.21 0.19 0.19 0.19 0.19

Water-ethanol 50/50 (v/v) A

3,3'-DPE (PH 8) 0.106 0.100 0.079 0.070 0.074 0.095 0.108

3'

2

$ 3

\

3,3'-DPEHi+ (PH 0) 0.024 0.023 0.022 0.023 0.023 0.022 0.024

2

2 E

n b

100

Photochemistry

picosecond laser pulses. Lifetimes ranged from 0.3 to 3.5 ns, and were also determined from spectroscopic data. Variations in lifetimes have been analysed in relation to the chemical structure of the dye. The evaluation of non-random errors 62 and methods of analysis of fluorescence decay data 63 are also of interest to photochemists. Single-photon counting 64 has been used to measure the fluorescent lifetimes of N-methylpiperidine (NMP) from subnanosecond to 25 ns. Such a compound provides simple second-order self-quenching kinetics together with single exponential decays, and a method of data treatment has been described. With highly concentrated NMP solutions, lifetimes of 0.30, 0.50, and 0.70 ns could be distinguished, and a lower limit of 0.2 ns was placed on the measuring system. The fluorescence spectra, quantum yields, and lifetimes of the higher helicenes have been r e p ~ r t e d ,these ~ ~ results being presented in Table 2. The lifetimes of tryptophan and tryptophan residues in bovine serum albumin 66 and of aminopyridines67 in different solvent systems have been reported, the fluorescent rate constant for the latter being smaller in nonpolar solvents due to vibronic mixing of n,r* and r,m* states than in polar and hydrogen-bonding solvents. Apart from a dependence of lifetimes of tryptophan residues on pH, there is a marked influence of the dielectric nature of the environment and excitation wavelength. The effect of glycerol on the lifetime of the excited N-arylaminonaphthalene singlet 68 is particularly interesting, in view of the extensive use of this type of molecule as a fluorescent probe for proteins and biological membranes. Fluorescence and phosphorescence spectra of several aromatic amines have been presented, and it has been concluded that the conjugation between nitrogen and aryl groups in the excited state is restricted. A steric explanation has been given for the lack of the expected photochemical ring-closure of 1- and 2-anilinonaphthalene to c a r b a ~ o l e s . ~The ~ relaxation of erythrosin 70 following picosecond excitation has been examined. Measurements of decay time and quantum yields as functions of temperature have been made for poly( 1-vinylnaphthalene) and its copolymers with methyl m e t h a ~ r y l a t e . ~Activation ~ energies and frequency factors have been reported and compared with reported values for methylnaphthalene and a,a’-dinaphthylpropane. Excimer dissociation has been found to be unimportant in polymer systems. Fink et aZ.72have reported on the decay time of I aggregates of pseudoisocyanine, using an optically switched Kerr cell, the time resolution being quoted as less than 5 ps. Decay times in the 25 ps range have been given for the dye 62

63

65

67

68

139 70

71 i2

J. Eisenfeld and D. J. Mishelevich, J. Chem. Phys., 1976, 65, 3384. A. R. Watkins, Mol. Photochem., 1976, 7 , 171. D. K. Wong and A. M. Halpern, Photochem. and Photobiol., 1976, 24, 609. J. B. Birks, D. J. S. Birch, E. Cordemans, and E. Vander Donckt, Chem. Phys. Letters, 1976, 43, 3 3 . W. X. Balcavage and T. Alvager, Mol. Photochem., 1976, 7 , 309. S. Babiak and A. C . Testa, J. Phys. Chem., 1976,80, 1882. R. P. De Toma, J. H. Easter, and L. Brand, J. Amer. Chem. SOC.,1976, 98, 5001. F. Fratev, 0. E. Polansky, and M. Zander, Z. Nuturforsch., 1976, 31b, 987. H. Al-Obaidi and G. A. Oldershaw, J. Photochem., 1976, 6, 153. C. D. M. Piens and G. Geuskens, European Polymer J., 1976, 12, 621. F. Fink, E. Klose, K. Teuchner, and S . Dahne, Chem. Phys. Letters, 1977, 45, 548.

Table 2 Fluorescence parameters of the helicenes Moiecuie Phenant hrene [4]Helicene [51Helicene [6]Helicene [7]Helicene

0.12 0.18 0.04 0.041 0.021

i71.4

0.019

10.2

[8]Helicene [BIHelicene [lOIHelicene [l 1JHelicene [12]Helicene [13lHelicene [14JHeIicene

0.014 0.014 0.0088 0.0088 0.0061 0.0097 0.0065

10.0 9.6 8.23 7.24 8.03 8.55 8.71

f

dns 63

@f = fluorescence quantum yield,

35.5

25.5 14.5 13.8

Tf

kIs1107 S-1 1.# 2.31 3.76 6.61 7.09

k1crn-l 28 830 26 200 24 900 24 100 22 500

Fllo/cm-l 1250

AP1crn-l 3250

1100 1200 lo00

3050 2600 2600

1.86

9.62

21 500

1200

2500

1.40 1.46 1.07 1.22 0.76 1.13 0.75

9.86 10.27 12.04 13.69 12.38 11.58 11.41

22 050 21 400 20 650 20 420 19 600 19 450 19 600

lo00 950 800 750 700 650 1050

2300 2600 2750 2700 3000 3350 3600

kfll0" s - ~ 1.90 5.07 1.57 2.83 1.52

= fluorescence lifetime,

kf = radiative rate constant, and k ~ = s radiationless rate constant. A = 3,15-ethano-heptahelicene.

-

S'

102

Photochemistry

NN'-diethylpseudoisocyanine chloride. The use of photon counting to study the isotope effect on the dissociation rate of carbazole in the excited state has been reported by Kishi et aZ.73 Direct evidence has been presented for rotational diffusion in fluorescein dye d e ri v ativ e ~.~~ Hochstrasser et aZ.75have reported the lifetime of sym-tetrazene in benzene solution at 300 K, using a Duguay-Hansen light gate, and report a new S1absorption spectrum that peaks at 473 nm, the lifetime being given as 450 ps. Decay times of tetracene crystals excited by synchrotron radiation have also been The use of the synchrotron as a light source for the measurement of the decay of fluorescence is likely to increase in the next few years. Several papers have appeared on various aspects of the luminescent properties and the lasing efficiency of several types of dye^,^'-^^ on photoinduced dichroism of rhodamine dyes,84 and on lifetime measurements using pulsed laser excita86 Popovic and Menzel 87 have described the use of a simplified phaset i 0 n . ~~1 shift lifetime-measuring instrument, and Heidt ** similarly used a phase-shift fluorometer as described by Hauser and Heidt (1975) to measure the lifetimes of some dyes. The phase-shift fluorometer apparently enables measurements of lifetimes of dyes with low quantum yields to be made with less difficulty than by the single-photon-counting pulse method. Numerous papers have appeared on the influence of physical and environmental effects on luminescence. These effects include p r e s s ~ r e , ~S ~O- ~I ~V ~ ~ ~ lo7 S , ~ ~ - ~ 73

T. Kishi, J. Tanaka, and T. Kouyama, Chem. Phys. Letters, 1977, 46, 383.

G. R. Fleming, J. M. Morris, and G. W. Robinson Chem. Phy.s, 1976, 17, 91. R. M. Hochstrasser, D. S. King, and A. C. Nelson, Chem. Phys. Letters, 1976, 42, 8 . 76 R. Lopez-Delgado, J. A. Miehe, and B. Sipp, Optics Comm., 1976, 19, 79. 77 M. W. Geiger, N. J. Turro, and W. H. Waddell, Photochem. and Photobiol., 1977, 25, 15. 78 J. P. Fouassier, D. J. Lougnot, and J. Faure, Optics Comm., 1976, 18, 263. 79 W. G. Herkstroeter, J. Amer. Chem. SOC.,1976, 98, 6210. U. K. A. Klein and F. W. Hafner, Chem. Phys. Letters, 1976, 43, 141. *l D. Basting, D. Ouw, and F. P. Schafer, Optics Comm., 1976, 18, 260. 82 K. H. Drexhage, J. Res. Nut. Bur. Standards 1976, 80A, 421. 83 W. Majewski and J. Krasinski, Optics Comm., 1976, 18, 255. 84 A. Penzkofer and W. Falkenstein, Chem. Phys. Letters, 1976, 44, 547. 85 H. Tashiro and T. Yajima, Chem. Phys. Letters, 1976, 42, 553. 86 J. Knof, F. J. Theiss, and J. Weber, Optics Comm., 1976, 17, 264. 87 Z. D. Popovic and E. R. Menzel, Chem. Phys. Letters, 1977,45, 537. G. K. Heidt, Appl. Spectroscopy, 1976, 30, 553. 89 K. Hara, G. B. Schuster, and H. G. Drickamer, Chem. Phys. Letters, 1977, 47, 462. O 0 D. K. Killinger, C. C. Wang, and M. Hanabusa, Phys. Rev. (A), 1976,13, 2145. Dl C. J. Mastrangelo and H. W. Offen, Chem. Phys. Letters, 1977, 46, 588. ga K. Uchida and M. Tomura, J. Luminescence, 1976, 14, 349. g3 R. P. Detoma and L. Brand, Chem. Phys. Letters, 1977,47, 231. g4 J. Herbick and A. Grabowska, Chem. Phys. Letters, 1977, 46, 372. g5 A. Nakajima, J. Mol. Spectroscopy, 1976, 61, 467. g6 A. Nakajima, J. Luminescence, 1976, 11, 429. 97 T. Hayashi, N. Mataga, Y. Sakata, and S. Misumi, Chem. Phys. Letters 1976, 41, 325. H. E. Lessing and M. Reichert, Chem. Phys. Letters, 1977, 46, 111. g9 M. L. Pandya and M. K. Machwe, Chem. Phys. Letters, 1976,44,561. l o o M. Sun and P.-S. Song, Photochem. and Photobiol., 1977, 25, 3. lol L. A. King, J.C.S. PerkinII, 1976, 1725. loa K. N. Solovev, V. N. Knyukshto, M. P. Tsuirko, and A. T. Gradyushko, Optics and Spectroscopy, 1976, 41, 569. D. Knittel, H. Raizdadeh, H. P. Lin, and S. H. Lin, J.C.S. Faraday ZZ, 1977, 73, 120. lo4 S. L. Madej, S. Okajima, and E. C. Lim, J. Chem. Phys., 1976, 65, 1219. lo5 Y. Kusumoto, V. Gondo, and Y. Kanda, Bull. Chem. SOC.Japan, 1976,49,2706. lo6 T. Azumi, K. Itoh, and H. Shiraishi, J. Chem. Phys., 1976, 65, 2550. lo7 G. Hug and R. S. Becker, J. Chem. Phys., 1976,65, 55. 74

75

Photophysical Processes in Condensed Phases

103

polar substituent effect,lo1 deuteration,102~108t e m p e r a t ~ r e , ~ ~ magnetic ~-~~~ field,l13,11* p ~ l a r i z a t i o n , ~and ~ ~absorption -~~~ and self-absorption quenching.124 Although all of these papers present significant advances in both theoretical and empirical knowledge, limitations of space permit only the following brief allusions. Hara et aZ.89studied the influence of external pressure on solid 2-adamantanone, and discovered three regions of pressure influence. In the region 0-10 kbar, a red shift in emission was seen, whereas in the critical 10-15 kbar region a dramatic shift of 500 cm-l to the blue was observed. Further increase of pressure resulted in an approximately linear blue shift in the emission maximum. The reversible discontinuity was attributed to a phase change within the adamantanone crystal. In contrast, however, the absorption and emission spectra of aromatic crystals shift towards lower energy under high pressure. Uchida and Tomura O 2 have reported emission decay times for naphthalene and pyrene crystals up to 1.5 kbar, and found the decay times to be shortened by 13 and 7%, respectively. The influence of isomeric solvents on the emission intensities of pyrene is presented in a paper by Nakajima.s5 The results show the importance of solvent polarity on band intensification in polar solvents, while dispersion interaction contributes mainly to emission intensity in non-polar solvents (see Figure 3). The kinetics of fluorescence solvatochromism are further elucidated by Lessing and ReichertsS in a theoretical paper which includes a consideration of the co-operative effects of light polarization and rotational diffusion. Nanosecond fluorescence spectroscopy has also been used to study the interaction of 2-anilinonaphthalene with polar solvent molecules.s3 The effects of solvent and temperature on the natural radiative lifetimes of substituted polyenes are discussed in a paper by Hug and Becker,lo7who used a simplified coupling model. They postulate that much of the dependence of T~ on the solvent and the temperature of retinals and diphenyl-polyenes is possibly due to solvent perturbations which cause state mixing, and that a decrease of T, with temperature arises because of an increase in solvent perturbation. Knittel et aZ.lo3have presented a theoretical treatment of the effect of temperature on radiationless transitions, with application to systems containing rare-earth ions. The temperature effect on the edge excitation red-shift of S-anilinonaphthalene-l-sulphonic acid has also been studied by Azumi et aZ.lo8 From Figure 4, L. C. Pereira, I. C. Ferreira, and M. P. F. Thomaz, Chem. Phys. Letters, 1976, 43, 157. 1976, 16, 329. J. Herbich and A. Grabowska, Chem. Phys. Letters, 1977, 46, 372. M. F. Anton and W. R. Moomaw, J. Chem. Phys., 1977,66, 1808. W. R. Moomaw and M. F. Anton, J. Phys. Chem., 1976,80,2243. J. Spichtig, H. Bulska, and H. Labhart, Chem. Phys., 1976, 15,279. K.Lendi, P. Gerber, and H. Labhart, Chem. Phys., 1976,18,449. B. R. Jennings and P. J. Ridler, Chem. Phys. Letters, 1977, 45, 550. M. L. Pandya and M. K. Machive, Chem. Phys. Letters, 1976, 44, 93. W.S. Struve, Chem. Phys. Letters, 1977, 46, 15. H. Shizuka, T. Ogiwara, S. Cho, and T. Morita, Chem. Phys. Letters, 1976, 42, 311. E. Leroy and H. Lami, Chem. Phys. Letters, 1976, 41, 373. R. C. Dorrance, T. F. Hunter, and J. Philip, J.C.S. Furuday 11, 1977, 73, 89. I. Tinoco, B. Ehrenberg, and I. Z. Steinberg, J. Chem. Phys., 1977,66, 916. I. Tinoco and D. H. Turner, J . Amer. Chem. Soc., 1976, 98, 6453. H. G. Brittain and F. S. Richardson, J. Phys. Chem., 1976, 80, 2590. G. Henderson, J. Chem. Educ., 1977, 54, 57

lo8

loB 111

114

117

118

l20 121

lza

124

J. R. Huber, M. Mahaney, and J. V. Morris, Chem. Phys.,

104

Photochemistry

it appears that the fluorescence resulting from excitation at the centre of the absorption band is shifted to the red, whereas fluorescence resulting from edge excitation exhibits little or no shifts. for isoquinoline lo9in ethanol increases The quantum yield of fluorescence (Of) from 0.012 at 295 K to 0.61 at 77 K, with a parallel increase of T~ from 0.25 to 9 ns. An analysis of the data shows that a spin-allowed internal-conversion

I

b

c

In C c

zP .501

a 0

22

24

26

28

Wavenumber ( l o 3 cm-')

..

Figure 3 Fluorescence spectra of pyrene (a) in cyclohexane (. .) and in N-methylformamide (-), and (b) in cis-dichloroethylene (-) and in trans-dichloroethylene (- - - -) (Reproduced by permission from J. MoZ Speciroscopy, 1976, 61, 467).

process is responsible for the observed temperature effect. Emission studies of quinoxaline,llo quinoline, and isoquinoline,lll as well as excited-state proton transfer in quinoline-isoquinoline systems,11zhave been reported. Two papers by Labhart and his co-workers deal with the influence of magnetic field on delayed fluorescence of aromatic hydrocarbons in ~ o l u t i o n114 . ~ ~The ~~ first paper describes the influence of magnetic field and the temperature, viscosity, and polarity of solvent on triplet-triplet annihilation (TTA); the second paper concerns the theoretical aspects of these results. The influence of an electric field115 on the intensity and polarization of fluorescence of ethidium bromide (EB) bound to DNA has also been investigated: such measurements can give insight into the nature of the stacking of

Photophysical Processes in Condensed Phases

105

I

I

/I

430

E .420 z

0 !-

; I 410 V LL 0

400

E 390 W z W

383 U

3

370

K

7

360

440

450

460

470

WAVELENGTH OF EMISSION MAXIMUM/MI Temperature dependence of fluorescence edge excitation red shvt for 8-anilinonaphthalene-1-sulphonicacid in propan-1 -01 (Reproduced by permission from J. Chem. Phys., 1976, 65, 2551)

Figure 4

E M I S S I O N WAVELENGTH Im

r1

Figure 5 Plot of ( A ) polarization versus emission wavelength, ( B ) relative intensity versus emission wavelength (Reproduced by permission from Chem. Phys. Letters, 1976, 44, 93)

106 Photochemistry groups within the DNA helix. The possible use of polarization measurements in distinguishing two closely emitting species has been demonstrated, using rhodamine as an example (see Figure 5).11* Picosecond lifetime and depolarization studies of emission from l(nn*) transazobenzene indicate that fluorescence contributes substantially to the emission decay pattern of this molecule. Biphenylene was found to exhibit abnormal behaviour, in that fluorescence was seen from the S,(lL,) state in cyclohexane, ethanol, and acetonitrile whereas rapid deactivation via S1 occurs in other Tinoco 121* has presented a detailed treatise on fluorescencedetected c.d. in both solution and rigid media. Henderson 124 discusses the influence of the phenomenon of self-absorption quenching on fluorescence intensities in an elementary but lucid fashion. The paper by Limlo4and his collaborators on substituent and temperature effects in emission from some N-heterocyclic molecules is a noteworthy contribution to this complex field of photophysics where interactions between nw* and n7r* states are so important. Emissions from biacetyl in water and alcohol solutions have been attributed to a hemihydrate and a hemiketal, re~pective1y.l~~ Acid-base and excited-state properties of 3,6-dimethoxy-9-phenylxanthen-9-ol have been examined in some detai1.lZ6 A reappraisal of the Forster cycle used in the calculation of the thermodynamics of the acid-base properties of excited states has been made by Grabowski and G r a b o w ~ k a . ' ~ ~ Papers have appeared on excited-state proton transfer in N-formylkynurenine derivatives,128on the effect of proton donors on the fluorescence characteristics of a ~ r i d o n e and , ~ ~ on ~ proton quenching of fluorescence of N-acetyl-L-tryptophanamide.130 Multiple fluorescenceshave also been reported for the protonated sulphonate~.~~~ form of N-alkyl-2-N-arylamino-6-naphthalene Nitrogen-laser excitation of benzophenone solutions generates not only the familiar luminescence spectrum but also a yellow emission, centred at 560nm, which is considered to arise from emission from the diphenylketyl radical. Studies on substituted naphthalenes,133 2-naphthylamine,13* tetrazine,135 and arylazomethines 136 have also been reported. The absorption of high-energy radiation produces large initial yields of 13* has discussed the production radical-cations and electrons. Brocklehurst 13'9

Y.J. Lee and J. G. Burr, Chem. Phys. Letters, 1976,43, 146. P. A. H.Wyatt and Z. M. Zochowski, Z . phys. Chem. (Frankfurt), 1976, 101, 143. la7 Z. R. Grabowski and A. Grabowska, Z . phys. Chem. (Frankfurt), 1976,101, 197.

126

126

lZ8

M. P. Pileni, P. Waliant, and R. Santus, Chem. Phys. Letters, 1976, 42, 89.

lae

H.Kokubun, 2.phys. Chem. (Frankfurt), 1976, 101, 137. R. F. Evans, C. A. Ghiron, R. R. Kuntz, and W. A. Volkert, Chem. Phys. Letters, 1976,42,

130

615. H. Dodiuk and E. M. Kosower, J. Amer. Chem. SOC.,1977,99, 859. 132 K. Razi Naqvi and U. P. Wild, Chem. Phys. Letfers, 1976, 41, 570. 133 R. D. Singh and S. N. Singh, Indian J. Pure Appl. Phys., 1976, 14, 583. 134 V. L. Bogdanov, V. P. Klochkov, and A. M. Makushenko, Optics and Spectroscopy, 1976, 41, 341. R. M. Hochstrasser and D. S. King, J. Amer. Chem. SOC.,1976, 98, 5443. lS6 N . A. Vasilenko, R. N. Nurmuckhametov, and I. L. Belaits, Russ. J. Phys. Chem., 1976, 50, 1399. lS7 B. Brocklehurst, J.C.S. Faraday ZZ, 1976, 72, 1869. 138 B. Brocklehurst, Faraduy Discuss., 1977, 63, 96.

Is1

Photophysical Processes in Condensed Phases

107

of excited states through geminate recombination of radical-ions, and developed the theory of changes of spin correlation i n the presence of a magnetic field. Double-resonance techniques and the low-field Zeeman PMDR method have been used to elucidate the mechanism of the SI-+ Tlnon-radiative process in d~ra1dehyde.l~~ The use of excitation sources other than light in producing excited states has also continued to receive some attention. Thus 01 and fl particles have been used to excite the scintillator bPBD140 in a methylcyclohexane glass at 77 K. The luminescence decay (observed was in the range 1-500 ns) comprising two components, a ‘fast’ one, attributed to excitation energy transfer from solvent to solute molecules ( 5 x 10-lo s) and a ‘slow’ component attributed to electron-cation recombination fluorescence and observed over a large time scale (1-500 ns). The use of luminescence studies in the biological field continues to provide an increasing number of publications. Thomas and Leonard 141 have reviewed the application of fluorescent heterocycles to the study of chemical and biological systems. Although not covering the whole range of heterocycles available, they nevertheless give a detailed account of the properties and uses of coumarins, pyrrolinones derived from fluorescamine, and analogues of nucleic acid bases. Fluorescamine provides a rapid, sensitive reagent for the detection of amines, but the hydrolysis of fluorescamine normally competes with the production of the fluorophore. A study of the effect of mixed aqueous solvents containing dimethyl sulphoxide on the fluorescence intensity of some fluorescamine derivatives has been The use of fluorescent tags as probes has been critically Blue dextran has assessed 143 and has been applied to amino-acid ana1y~ers.l~~ also been used as a probe for the nicotinamide-adenine dinucleotide domain in Another class of fluorescent nitrogen heterocycles includes the porphyrins 146-148 and c h l ~ r o p h y l l1509 .~~ 180~ ~ The parameters of the triplet state of chlorophyll and protochlorophyll 149 have been studied by determining the time dependence of fluorescence intensity. A study of the kinetics of triplet population allows lifetimes of triplet-state sublevels to be determined. Rate constants for quenching of chlorophyll-a singlets and triplets by a series of manganese complexes have also been measured in fluid The use of porphyrin in the study of photo-oxidation of methionine is described in the paper by Cauzzo et all4’ IndolelS1 is one of the most widely studied fluorescent heterocycles. Its excitation spectrum has been resolved by polarization measurements into the %, and lLb excitation bands (see Figure 6). 13g

140 141 lila 14s 144 145 146

14’ li18

160

161

A. Campion and M. A. El-Sayed, J. Phys. Chem., 1976, 80, 2201. F. Kieffer, C. Lapersonne-Meyer, and J. Klein, Chem. Phys. Letters, 1976, 40, 492. R. W. Thomas and N. J. Leonard, Heterocycles, 1976, 5, 839. P. M. Froehlich and T. D. Cunningham, Analyt. Chim. Acta, 1976, 84, 427. T. Hirschfeld, Appl. Optics, 1976, 15, 3135. E. Lund, J. Thomsen, and K. Brunfeldt, J. Chromatog., 1977, 130, 51. E. Stellwagen, Accounts Chem. Res., 1977, 10, 92. S. Volker and J. H. van der Waals, Mol. Phys., 1976,32, 1703. G. Cauzzo, G. Gennari, G. Jori, and 5. D. Spike, Photochem. and Photobiof., 1977, 25, 389. G. W. Canters, G. Jansen, M. Noort, and J. H. van der Waals, J . Phys. Chem., 1976, 80, 2253. R. Avarmaa, Chem. Phys. Letters, 1977, 46, 279. R. G. Brown, A. Harriman, and G . Porter, J.C.S. Faraday ZZ, 1977,73, 113. B. Valeur and G. Weber, Photochem. and Photobiol., 1977, 25, 441. 5

108

Photochemistry

Interactions of 7-azaindole with alcohol and water, as functions of pH, temperature, and solvent d e ~ t e r a t i o n ,and ~ ~ ~the photophysical properties of isoindole 163 and its benzo[f]- and dibenzo[e,g]-derivatives have also been studied. Luminescence studies of solid naphthalene haemoprotein~,~~~ ochratoxins 156 (which may involve intramolecular transfer of energy between chromophores), and tetracyclines 16’ have also been reported.

t

k

A/m Figure 6 Corrected excitation spectrum (- - - -) and excitation polarization spectrum of indole in propylene glycol at - 58 “C (Reproduced by permission from Photochem. and Photobiol., 1977, 25, 442)

Continued interest has been shown in DNA 158-161 and the nucleosides,ls2 owing partly to the relevance to cancer research. Of particular interest to photochemists is the paper by Morgan et aZ.ls3which attempts to establish a correlation between the carcinogenic activities of polycyclic aromatic hydrocarbons with their singlet, triplet energies and the phosphorescence lifetime. They have concluded that there is some correlation with the energy of the singlet excited state but not with the phosphorescent lifetimes. Interest has continued in the photochemistry of p o l y e n e ~ ,including ~~~ the visual pigment retinals, and their Schiff bases. Becker et aZ.165 P. Avouris, L. Yang, and M. A. El-Bayoumi, Photochem. and Photobiol., 1976,24,211. W. Rettig and J. Wirz, Helu. Chirn. Acta, 1976, 59, 1054. 154 R. Kopelman, J. Phys. Chem., 1976, 80, 2191. 155 F. Adar, M. Gouterman, and S. Aronowitz, J . Phys. Chem., 1976, 80, 2184. 156 A. M. Gillespie and G. M. Schenk, Analyt. Letters, 1977, 10, 161. lS7 H. Poiger and Ch. Schlatter, Analyst, 1976, 101, 808. 158 S. Sasson, S. Y . Wang, and M. Ehrlich, Photochem. and Photobiol., 1977, 25, 11. 159 W. Hauswirth and S. Y . Wang, Photochem. and Photobiol., 1977, 25, 161. 160 M. H. Patrick and J. M. Snow, Photochem. and Photobiol., 1977,25, 373. lE1 V. G. Tatake, T. S. Desai, and P. V. Sane, Photochem. and Photobiol., 1976, 24, 463. 162 H. L. Chung and J. Zemlicka, J. Heterocyclic Chem., 1977, 14, 135. 163 D. D. Morgan, D. Warshawsky, and T. Atkinson, Photochem. and Photobiol., 1977, 25, 31. 16* R. L. Christensen and B. E. Kohler, J. Phys. Chem., 1976, 80, 2197. IE5 R. S. Becker, G. Hug, P. K. Das, A. M. Schaffer, T. Takemura, N. Yamamoto, and W. Waddell, J. Phys. Chem., 1976, 80, 2265. 152

153

109

Photophysical Processes in Condensed Phases

have provided firm evidence to show that the lowest excited state of retinals in alkane solvents is principally of l(nn*) character, and they give the lifetime for a retinal Schiff base as 3 ns at 77 K. Singlet Quenching by Energy Transfer and Exciplex Formation.-Electron DonorAcceptor Complexes and Related Charge-transfer Phenomena. The fluorescence quantum yields of p-xylene 168 excited at 184.9 nm in iso-octane have been determined for the S3-+ So(Q3), S2 -+ So(a2),and S1-+ So (0,) transitions as functions of the concentration of p-xylene (c,) and of the concentration of the added quenchers CClp, CHCI3, and cyclo-C,F,,. In dilute solutions of p-xylene, it was concluded that the S3state either radiates (kr = 3 x lo9 s-l) or undergoes internal conversion to S, (k = 2.5 x 1014s-l). No other processes are significant. The S, state either radiates (kr = 1.3 x lo8 s-l), or undergoes internal conversion to S, (k = 1.0 x lof3s-l), or autoionises: all other decay processes are again negligible. In the absence of a quencher, the electron returns to the parent positive ion and regenerates S2,but not a lower state. In the presence of quencher, the electron is scavenged, and it no longer contributes to the eventual production of emission. Studies of the quenching of the red emission of aromatic thioketonesls9 by both ferrocene and anthracene have enabled the red emission to be assigned to phosphorescence rather than to fluorescence. A quenching rate constant ka = (5.8 & 0.4) x loDm-ls-l was obtained for both quenchers. Quenching of aromatic hydrocarbon singlets 170 and aryl ketone triplets by alkyl disulphides reveals that charge-transfer-stabilized exciplex formation is unimportant, and that endothermic singlet-singlet energy transfer is more efficient than predicted from the Arrhenius equation. Rate constants for singlet quenching of aromatic hydrocarbons, as determined by single-photon counting, are shown in Table 3. Table 3 Singlet quenching data for di-t-butyl disulphide Aromatic hydrocarbon (molarity ) Naphthalene (4 x Triphenylene (1 x lob2)

kq SoZvent

GH6 C6H6

MeCN Phenanthrene (3 x

C6H6

MeCN Chrysene (2.7 x

C6H,

MeCN Pyrene (1.2 x

C6H6

MeCN

Fluoranthene (2.8 x

C6H6

TO/U

106 38.6 37.0 57.4 57.5 44.4 44.3 304 282 47.8

x

1 rno1-I s-l 27 k 2 1.6 k 0.1 2.7 & 0.2 1.2 -L 0.1 2.2 0.1 0.31 k 0.02 0.74 & 0.1 0.14 i-0.01 0.28 k 0.01 0.17 k 0.01

*

Measurements of the quenching of benzene fluorescence that is excited by ionizing radiation might be expected to differ from those of u.v.-excited systems. T. Takemura, P. K. Das, G. Hug, and R. S . Becker, J. Amer. Chem. SOC.,1976, 98, 7099. A. N. Kriebel and A. C. Albrecht, J. Chem. Phys., 1976, 65,4575. lea T. A. Gregory and S. Lipsky, J. Chem. Phys., 1976,65, 296. 16s M. Mahaney and J. R. Huber, J. Photochem., 1976, 5, 333. 170 W. L. Wallace, R. P. Van Duyne, and F. D. Lewis, J. Amer. Chew. SOC.,1976, 98, 5319. lee 167

Photochemistry

110

Variations in decay measurements under different conditions may give timecorrelated information about events following excitation. West 171used pulsed proton irradiation to study benzene quenching, and postulated an intra-track quenching by transient species as a possible explanation. Further studies of the quenching of benzene by acetone, 2-butanone, and 3-pentanone in cyclohexane 172 and of trifluoromethylbenzene 173 by heterocyclic compounds reveal a diffusionlimited process to be involved. The transfer probability is not only related to the donor-acceptor separation but also to the angle between donor and acceptor moments. Fluorescence depolarization measurements can therefore be used to estimate the efficiency of energy transfer between donor and acceptor. Such measurements have been carried out in polymers, using styrene-phenylacetylene copolymer as the donor and perylene as the acceptor (see Table 4).174 The application of Forster's theory Table 4

Depolarization of styrene-phenylacetylene copolymer (donor) and perylene (acceptor), suspended in poly(methy1 methacrylate)

Acceptor concentration1 mol 1-1 0

0.55 x 1.46 x 3.65 x 7-30 x 18.30 x

10-3 10-3 10-3 10-3

Eficiency of transfer from donor to acceptor 0.40 0.59 0.73 0.81 0.99

Depolarization (1) 6.8 10.3 16.0 22.7 25.0

Depolarization (2) 3.2 3.4 4.7 5.8 10.5

Depolarization (3) 1.8 1.7 1.6 1.5 1.9 1.9

to the polymer system yields a value for the critical distance of 7.8 nm. Efficiency of transfer increases, together with depolarization from the acceptor, in accord with the predictions of Forster's theory. The Forster-Ore theory of concentration 176 and polarization depolarization has also been revised. Depolarization measurements 17' have been carried out for Acridine Orange cations in poly(viny1 alcohol) films. The results show that singlet-singlet energy transfer is responsible for depolarization of fluorescence, as well as of phosphorescence. Quenching results on excited singlet and triplet states of ketones by di-t-butyl nitroxide 17* and also on the quenching of fluorescence of 9-aminoacridine bound to polynucleotides 179 have been presented. In an attempt to understand the light-gathering capabilities of photosynthetic systems, fluorescence yields and lifetimes of chlorophyll a in a lipid vesicle and liposomes were determined.lsO Concentration quenching was observed in all systems, the extent depending on the lipid used. 1759

171

172 173 174

176

17' 17&

17s 180

M. L. West, J. Phys. Chem., 1977, 81, 377. W. Augustyniak, E. Wiechowicz, and J. Wojtczak, J. Photochem., 1976, 6, 229. M. E. Sime, D. Phillips, and Kh. Al-Ani, Mol. Phorochem., 1976, 7 , 149. J. R. MacCallum and L. Rudkin. Nature, 1977, 266, 338. T. Komiyama and Y . Mori, Bull. Chem. SOC.Japan, 1976, 49, 864. C. Bojarski and R. Bujko, Acra Phys. Chem., 1976, 22, 25. T. Komiyama and Y . Mori, Chem. Letters, 1976, 1081. D . S. Weiss, J. Photochem., 1976, 6, 301. Y . Kubota and Y . Fujisaki, Bull. Chem. SOC.Japan, 1977, 50,297. G. S. Beddard, S. E. Carlin, and G. Porter, Chem. Phys. Letters, 1976, 43, 27.

Photophysical Processes in Condensed Phases

111

The photophysics of the interaction of a-cyanonaphthalene in its first excited singlet state with 1,2-dimethyIcyclopentene as a quencher has been reported over the temperature range 0 to -40 "C in hexane. Analysis of the decay kinetics reveals four rate constants associated with exciplex formation and decay.lB1 The geometrical requirements which have to be satisfied for fluorescent intramolecular exciplex' formation and fluorescence quenching have been discussed,lB2and a simple frontier M.O. analysis of data for donor quenching of exciplexes has been developed.1s3 Such analysis allows order-of-magnitude predictions of quenching rates ( k ~to) be made. Observation of exciplex formation by radical-ion reactions in polar solvents has stimulated further work on the role of aromatic hydrocarbon interceptors in energy transfer from exciplexes.lB4 Measurements of luminescence were made to obtain the exciplex chemiluminescence efficiency for the reaction between tri-p-tolylamine cation radical and the anion of 1,4-dicyanobenzene: they give an indication of the role of exciplex formation in radical-ion annihilations occurring in polar media. Exciplex formation in 1,3-photoaddition reactions of anisole with olefins has been proposed,lB5and the modes of decay of exciplexes formed by the donor molecule 1,12-benzperyIenewith a series of dimethylaniline derivatives have been described.ls6 5-Methoxyindole lB7has been shown to be non-exciplex forming with solvent in polar solvents, in contrast with indole itself and its methyl-substituted derivative. This adds weight to the proposal that there are two limiting classes of exciplexes, namely, charge-transfer and dipole-dipole-stabilized. Investigations of the role of donor-acceptor complexes in charge-transfer reactions have also been made.188-1Q2Geometry and solvent polarity effects on cyclophanes lQ1reveal an enhanced radiationless decay with increasing solvent polarity. Further investigations of intramolecular electron donor-acceptor complexes have been carried O U ~ and , ~ also ~ on ~ sensitized ~ biphotonic radicalanion formation from solute-solvent donor-acceptor complexes.188 An interesting and instructive, if selective, review of energy-transfer processes has recently appeared.lQ3 A discussion of the major mechanisms of electronic energy transfer between organic molecules is given, with particular emphasis on the photo-physical and -chemical properties of polymers. Numerous other papers have appeared in the literature on charge-transfer reactions and their influence on quenching of excited states. Most of these are D. V. O'Connor and W. R. Ware, J. Amer. Chem. SOC.,1976, 98,4706. R. S. Davidson and K. R. Trethewey, J.C.S. Chem. Comm., 1976, 827. la3 D. Creed,R. A. Caldwell, H. Ohta, and D. C. DeMarco, J. Amer. Chem. Sac., 1977, 99, 277. lE4 H. Tachikawa and L. R. Faulkner, J. Amer. Chem. SOC.,1976, 98, 5569. lS6 R. Srinivasan and J. A. Ors, Chem. Phys. Letters, 1976, 42, 506. la6 A. R. Watkins, Chem. Phys. Letters, 1976, 43, 299. M. V. Hershberger and R. W. Lumry, Photochem. and Photobiol., 1976, 23, 391. lS8 K. Kimura and Y . Achiba, Chem. Phys. Letters, 1977, 46, 585. lE8 M. Yoshida, H. Tatemitsu, Y. Sakata, and S. Misumi, J.C.S. Chem. Comm., 1976, 587. l g o H. Dodiuk and E. M. Kosower, J . Phys. Chem., 1977, 81, 50. ls1 J. H. Borkent, J. W. Verhoeven, and Th. J. De Boer, Chem. Phys. Letters, 1976, 42, 50. lS2 S. Tazuke and K. Sato, J. Phys. Chem., 1976, 80, 1727. lB3 N. J. Turro, Pure Appl. Chem., 1977, 49, 405. lS1 ls2

112

Photochemistry

only of rather specialized interest as far as this Report is concerned but have been mentioned for ~ o m p l e t e n e s s . ~ ~ ~ - ~ ~ ~ An improved expression for the effective interaction radius has been derived for energy transfer accompanied by diffusion in the system pyrene-perylene in a 1 : 1 paraffin oil-1-methylnaphthalene the experimental data providing support for the theory. Results have been presented which show that the triplet spin in aza-aromatic molecules205is created in a plane through the axis of the lone-pair orbital on the nitrogen and the normal to the molecule. In molecules where this nrr plane does not coincide with the zero-field spin planes, it is then possible for the spin levels to share a common channel in the intersystemcrossing process. Nanosecond laser studies have also been carried out on the benzene-tetracyanobenzene molecular complex,2o6and on 1,2,4,S-tetracyanobenzene207 with aromatic donors, the results showing a marked dependence on the amounts of acetonitrile present in the solvent. The relative quantum yield and the lifetime of the fluorescent state were found to decrease with increasing amount of acetonitrile, effects attributable to a decrease in the radiative transition probability and an increase in non-radiative probabilities. The influence of a N-methyl substituent on the fluorescence, phosphorescence, and intersystem-crossing rate constants of aromatic amines, diphenylamine, acridane, and carbazole 208 in solution has been investigated. The strong enhancement of singlet-triplet intersystemcrossing efficiency between acridane and methylacridane has been attributed to the change in molecular geometry upon methylation. Picosecond laser techniques have been used to study the geometrical requirements and the influence of solvent polarity on the excited-state charge-transfer process in anthryl-(CH,)3-NN-dimethylaniline.20g The role of the ‘Rydberg’ state in charge-transfer-type interactions has been studied through absorption, fluorescence, and photocurrents for tetra-aminoethylenes 210 in aromatic hydrocarbons. It has been postulated that mixing between charge-transfer and ‘Rydberg’ configurations still remains significant even at relatively long distances between donor and acceptor, and thus shows the possibility of long-range intermolecular interactions. A practical application of energy transfer is found in the development and improvement of dye lasers.211 Such effects can provide ways of extending the M. Nepras, M. Titz, V. Zverina, and M. Matrka, Coff.Czech. Chem. Comm., 1976,41,2489. Y . Kubota, Chem. Letters, 1977, 311. lu6 J. Kaminski and A. Kawski, Z . Nuturforsch., 1977, 32a, 140. lS7 A. Ueno, T. Osa, and F. Toda, Polymer Letters Edn., 1976, 14, 521. lg8 D. Holten, M. Gouterman, and W. W. Parson, Photochem. and Photobiol., 1976, 23, 415. l@@ E. Vozary and L. Szalay, Acta Phys. Chem., 1976, 22, 17. R. Reisfeld, S. Nathanson, and E. Greenberg, J. Phys. Chem., 1976, 80, 2538. 201 T. Okada, T. Fujita, and N. Mataga, Z . phys. Chem. (Frankfurt), 1976, 101, 57. 2 0 2 R. J . Hurtubise, Analyt. Chem., 1976, 48, 2092. 203 U. Landman, A. Ledwith, D. G. Marsh, and D. J. Williams, Macromolecules, 1976, 9, 833. 2 0 4 U. K. A. Klein, R. Frey, M. Hauser, and U. Gosele, Chem. Phys. Letters, 1976, 41, 139. 205 R. A. Schadee, J. Schmidt, and J. H. van der Waals, Chem. Phys. Letters, 1976, 41, 435. 2 0 8 B. B. Craig and M. A. J. Rodgers, J.C.S. Furuduy ZI, 1976, 72, 1259. 2 0 7 B. B. Craig, M. A. J. Rodgers, and B. Wood, J.C.S. Furuduy IZ, 1976, 73, 349. 208 H. 5. Haink and J. R. Huber, Chem. Phys. Letters, 1976, 44, 117. 2 o e K. Gnadig and K. B. Eisenthal, Chem. Phys. Letters, 1977, 46, ,339. Y. Nakato, J. Amer. Chem. SOC.,1976, 98, 7203. 211 E. Weiss and S. Speiser, Chem. Phys. Letters, 1976, 40,220. lB4

lg6

Photophysical Processes in Condensed Phases

113 wavelength regions in which lasing is achieved by the use of suitably selected dye mixtures; a prediction verified by preliminary experiments with anthraceneperylene mixtures. Fluorescence Quenching by Inorganic Species. The influence of alkali-metal halides on the fluorescence and photoisomerization of 3-styrylpyridine, its 4-methoxy-derivative, and 1,2-di(3-pyridyl)ethylene, all in their protonated form, has been studied.212With the bromide and iodide anions the isomerization is less quenched than luminescence for 3-styrylpyridine, whereas it is enhanced for the other two compounds. This result was attributed to the formation of a charge-transfer complex followed by the induction of olefin intersystem crossing by a heavy-atom effect. The reversible photoreduction of thionine 213 by Fell still retains interest, especially in view of possibilities for solar energy conversion. Fluorescence lifetimes and the quantum yield of thionine at pH 2.5 have been measured as a fraction of added Fe'I quencher, using a picosecond flash technique. Similar results were found in media containing chloride and sulphate, the results obeying the Stern-Volmer relation. The overall second-order quenching rate constants were given as 2.9 x 10l2cm3mol-1 s-l with chloride and 3.5 x 10l2cm3mo1-1 s-l with sulphate. The principal reactions in acid (pH = 2.5) aqueous solutions are shown in equations (1)-(3). Interestingly, previous measurements have been TH+

hv

*TH+

3THi+ + Fe"

*THf

3TH+ ___+

TH;I+

H+

+ Fell1

(1)

3THi+

(2)

(3)

confined to studies of the triplet state, although the singlet state is slightly fluorescent. The use of iodide ions in quenching the fluorescence of pyrene that is solubilized within cetyltrimethylammonium bromide micelles has also been The existing model for quenching, which requires pyrene to diffuse from within the hydrocarbon core, is not acceptable : the water-lipid interface evidently penetrates deep into the micellar body, so as to be close to the included pyrene molecule. Energy transfer from singlet excited levels of organic molecules to Cri1I complexes in liquid and solid solutions has been studied.216 Energy transfer takes place in anthracene by an inductive-resonance diffusional mechanism, and evidence has been given that dyes of the xanthene and acridine series transfer energy by intramolecular diffusion and inductive resonance. Heavy-atom Quenching. Of continued interest in organic photochemistry is the quenching of excited states by compounds not able to accept energy by classical energy transfer. It is well known that the presence of heavy atoms, either bonded or in the environment, can strongly enhance the rates of molecular spin-forbidden transitions such as singlet-triplet intersystem crossing. 21a

213

G. Bartocci, U. Mazzucato, and P. Bortolus, J . Photochem., 1976, 6, 309. M. D. Archer, M. I. C. Ferreira, G . Porter, and C. 5. Tredwell. Nouueau J. de Chimie, 1976, 1, 9.

214 215

M. A. J. Rodgers, M. E. da Silva, and E. Wheeler, Chem. Phys. Letters, 1976, 43, 587. E. B. Sveshnikova and S. P. Naumov, Optics and Spectroscopy, 1976, 41, 131.

114

Photochemistry

The use o f such heavy atoms in enhancing and observing room-temperature phosphorescence has been described (see Figure 7).216 Fluorescence quenching studies of aromatic hydrocarbons by triphenyl derivatives of Group V elements and by triethylamine in both non-polar and polar solvents have been described. The evidence available for quenching tends to rule out simple electron or charge transfer, and infers that there is also some heavy-atom effect involving Group V elements. A search for evidence of exciplex formation in the quenching of azulene S2fluorescence did not prove fruitful.217

I

fluorercence

300

phosphorescence

400

500

WAVELEMGTH (nm)

Figure 7 Emission spectrum of 2-naphthol on filter paper with 0.5 mol 1-1 sodium iodide added (Reproduced by permission from J. Chem. Educ., 1976, 53, 654)

Decay routes in the singlet quenching of naphthalene by chloroacetonitrile have been measured,218and rate constants for the fluorescence decay of dyes in gelatin and adsorbed onto AgCl and AgBr microcrystals have also been Large decreases in both fluorescence yields and lifetimes are found in going from a gelatin environment to the surface of either halide, the bromide being much more effective in this respect than the chloride. Excimer Formation and D e c a y . Several papers have appeared on the luminescence and kinetics of dimer emission. Conformational differences between singlet and triplet excimers of naphthalene have been investigated,230and also the influence of the crystal structure of dianthronylidenethane 221 on the formation of the excimer. Studies of the magnetic field dependence 222 of monomer and excimer delayed fluorescence 216

217

218 219

220 221

J. L. McHale and P. G . Seybold, J. Chem. Educ., 1976, 53, 654. H. D. Burrows, S. J. Formosinho, A. M. da Silva, and S. E. Corlin, J. Photochem., 1976, 6, 317. F. H. Quina, Z. Hamlet, and F. A. Carroll, J. Amer. Chern. Soc., 1977, 99, 2240. A. A. Muenter, J. Phys. Chem., 1976, 80, 2178. P. C. Subudhi and E. C. Lim, Chem. Phys. Letters, 1976, 44, 479. H.-D. Becker, B. Karlsson, and A.-M. Pilotti, Chem. Phys. Letters, 1976, 44, 589. D. A. Capitanio and H. Van Willigen, Chem. Phys. Letters, 1976, 40, 160.

Photophysical Processes in Condensed Phases

115

reveal a monotonic decrease in intensity with increasing field strength at room temperature. Models for the triplet-triplet annihilation process have been put forward to explain the results. Emission spectra of pyrene over the temperature range 130-260 K and concentrations of to 2 x mol dm-3 have been The association energy of the pyrene excimer has been re-evaluated, using a method independent of experimental set-up response functions. The absorption spectra of the excimer have been published and compared with the S1+ S, absorption of the analogous 3,3’-bip~reny1.~~~ Excimer emission data have been presented for 9-cyanoanthracene 1,2-di(g-anthr~l)ethane,~~~ and anthracenophanes and 1,2-dianthryIethane~.~~’ Excimer fluorescence of biphenyl, p-terphenyl, trans-stilbene, diphenylacetylene, and perfluoronaphthalene arising from thermoluminescence of y-irradiated glasses 228 has been reported: these emissions have not as yet been observed following p ho t oexcitat ion. Perylene exhibits concentration quenching without any excimer emission. Excimer emission from perylene, however, can be observed in a rigid polymer matrix,229the quenching of fluorescence being interpreted as due to Forster inductive resonance to a trap site in which there is a neighbour sufficiently close to form an excited complex. Papers have appeared on excimer formation in polyesters having pendant 1-naphthylmethyl 230 and anthryl groups,231 and on substituted aw-diphenylall-trans-polyenes in the solid and organized monolayer assemblies.232 Sensitization of excimer fluorescence in polymer matrices can occur through singlet molecular oxygen.233 In particular, pyrene, 2,3-dimethylnaphthalene, and fluorene in polystyrene fluff emit entirely excimer emission under sensitization, in direct contrast with what is observed with photoexcited emission of the samples. It has long been known that absorption of near-u.v. light by DNA induces biologically deleterious effects such as dimerization of pyrimidine bases, chain breaks, and formation of DNA-protein cross-links. Pyrimidine dimers are extremely stable, and this therefore contributes to their adverse biological effects. Sutherland 234 has discussed the role of photoreactivating enzyme in monomerizing ‘cyclobutyl’ pyrimidine dimers, and postulates that electronic energy transfer occurs in the photoreactivation process. Further studies of the formation of pyrimidine dimer and the reaction with N’-formylkynurenine derivatives have 223

a24 225 226 227

228 228

230

231 232

233

M. E. Abu-Zeid, J. R. Lopez, P. Martinez, J. C. Acevedo, and R. Groff, Chem. Phys. Letters, 1977, 46, 558. M. F. M. Post, J. Langelaar, and J. D. W. Van Voorst, Chem. Phys. Letters, 1976, 42, 133. R. M. Macfarlane and M. R. Philpott, Chem. Phys. Letters, 1976, 41, 33. J. Ferguson, M. Morita, and M. Puza, Chem. Phys. Letters, 1976, 42, 288. T. Hayashi, N. Mataga, Y . Sakata, S. Misumi, M. Morita, and J. Tanaka, J. Amer. Chem. SOC.,1976, 98, 5910. M. Al-Jarrah, B. Brocklehurt, and M. Evans, J.C.S. Furaduy IZ, 1976, 72, 1921. J. A. Ferreira and G. Porter, J.C.S. Faraduy II, 1976,73, 340. S. Tazuke and F. Banba, Macromolecules, 1976, 9, 451. S. Tazuke and F. Banba, J. Polymer Sci. ,1976, 14, 2463. F. H. Quina, D. Mobius, F. A. Carroll, F. R. Hopf, and D. G . Whitten, 2. phys. Chem. (Frankfurt), 1976, 101, 151. R. D. Kenner and A. U. Khan, Chem. Phys. Letters, 1977, 45, 76. J. C. Sutherland, Photochem. and Photobiol., 1977, 25, 435,

116

Photochemistry

been published.235Other reports have described intramolecular excimer formation from dibenzyl ether and d i b e n ~ y l a m i n e and , ~ ~ ~aw-diarylalkane~,~~~ and excimer emission from diethyl p-phenylenedia~rylate.~~~ 3 Triplet-state Processes Although benzene has served as a prototype .rr-electron system for many years, there are still certain fundamental questions unanswered about the nature and location of the lowest and second lowest triplet states. Full characterization of the lowest triplet states, whose assignment is either 3Bluor 3B2u,is still required, and excitation is both orbitally and spin forbidden. Evidence is available to show that there is strong correlation between the zero-field splitting, the decay rates, and the energy difference between the lowest lying triplet states. The 3B1ustate is normally studied in solution or crystals, and the molecular distortions observed prevent an easy interpretation of the extrinsic and intrinsic effects. Extensive work on Tl benzene has been carried out, using [lH,]benzene in [2H6]benzene. Vergragt and van der Waals 239 have measured the zero-field splitting parameters of C6H6 and C6D6 in a borazole crystal host, using microwave-induced delayed phosphorescence at 1.2 K. The value of E is found to be three times larger for C6H6 in a borazole host than for C6H6 in C6D6. This result indicates that the distortion in borazole is much larger than in other hosts studied so far. It has been concluded that the distortion from hexagonal symmetry of benzene is quinoid in a borazole host, whereas it is antiquinoid in a C6D, host, and the structure of the triplet state of benzene is to a large extent determined by the crystal field. The lifetime of the triplet state of benzene in borazole was given as 9.12 k (0.02) s, which is slightly longer than in C6D6 as host (8.7 s). In a further paper, Christensen and van der Waals 240 describe investigations of the crystalfield effects of [lH,]benzene doped with [2H6]benzene. Such a system contains shallow traps which emit at low temperature. The phosphorescence spectrum of the doped crystal has been found to be almost identical with that of [lM,]benzene in [2H6]benzene.It is found that the X-traps have zero-field anisotropy factors, I E I, that are almost one order of magnitude smaller, whereas the splitting of 9: is almost a factor of ten larger than the corresponding values seen for C & &in C6D6. It has been suggested that rationalization of these apparently divergent results requires more general forms of the crystal-field potentials than those employed at present. Phosphorescence polarization 241 measurements on C&, and C6D6 in 3-methylpentane glass at 77 K have been carried out. The assignment of 3Bluto the lowest triplet state of benzene requires that phosphorescence polarization of the weak b,, bands be entirely in-plane. The emission was found to have 15-20% out-of-planepolarized content, and is therefore in conflict with the conventional 3Bluassignment for the lowest triplet state of benzene. This result, together with the in-plane character found for the e,, region, requires radiative decay via the A 2 , 236 236

237

P. Waliant, R. Santus, and M. Charlier, Photochem. and Photobiol., 1976, 24, 13. Y.-C. Wang and H. Morawetz, J. Anier. Chem. SOC.,1976, 98, 3611. K. Zachariasse and W. Kuhnle, 2. phys. chem. (Frankfurt),1976, 101, 267.

Photophysical Processes in Condensed Phases

117

spin sublevel, in disagreement with theory and with microwave work. These results, however, must be considered in the light of possible solvent perturbations and reorient ational depolarizat ion. Phosphorescence lifetimes and polarization measurements have also been made 242 for C6H6and C,D, in a range of 3-methylpentane-2-methylbutaneglasses at 77 K. The rotation times are independent of isomer, and the rotation time in 3-methylpentane is > 200 s. The phosphorescence lifetimes ranged from 5.03 to 1.86 s with an increase from 10 to 60% (by volume) of isopentane in the solvent. Phonon-assisted transfer of triplet excitation 243 energy between isotopic traps in three-component (host, trap, and supertrap) benzene isotopic mixed crystals demonstrates the dependence of off-resonance trapsupertrap excitation energy transfer on the lattice Franck-Condon factor. The efficiency goes through a maximum at a trapsupertrap energy separation of 50 cm-l, and decreases rapidly on either side of this value. Continued interest has been shown in the 'dual phosphorescence' from 2,4,5-trimethylbenzaldehyde in crystalline d ~ r e n e Previous . ~ ~ ~ evidence supports dual phosphorescence from a thermally populated upper 3nn* state and a lower (predominantly 3n7r*)state. Additional measurements have been carried out to reveal the nature of the second lowest triplet state over a temperature range 10-1 60 K. Two phosphorescence bands with different temperature dependences have been observed and identified with emission from the lowest 3nn* states of two different conformers that are formed uiu the rotation of the aldehyde group in the excited triplet state. Zero-field e.p.r. transitions and spin sublevel intersystem-crossing rates for the lowest triplet states of tetracene, perdeuteriotetracene, 1,2-benzanthracene, and 1,2,3,4-dibenzanthracenehave been obtained in n-alkane solvents at 2 K by optical detection of magnetic resonance, using an argon-ion laser as the fluorescence excitation source.24s Dynamic measurements of the build-up and decay of the microwave-induced fluorescence were used to determine the depopulating rate constants and the relative populating rates for each of the spin sublevels. These results are presented in Table 5. The role of the T2 state in intersystem crossing in 9- and 9,lO-substituted anthracenes has been postulated to account for the decrease in fluorescence efficiency with temperature. Direct evidence for the participation of the T, state is inherent in the observation of the T2-+ T, fluorescence of 9-bromoanthracene and 9,lO-dibromoanthracene in heptane The T, So phosphorescence of these molecules has also been recorded, thus permitting accurate determination of the electronic energies. Estimates of the quantum yields for 9,lO-DBA are given as aP(Tl So) = 6 x and Q(T2 +Tl) = 8 x lo-' and the T2 -+ TI internal-conversion rate is lo1' s-l. --f

--f

238 239 240 241

242

243 244 246

F. Nakanishi, H. Nakanishi, and M. Hasegawa, Nippon Kugaku Kuishi, 1976, 1578. P. J. Vergragt and J. H. van der Waals, Chem. Phys. Letters, 1976, 42, 193. R. L. Christensen and J. H. van der Waals, Chem. Phys. Letters, 1977, 45, 221. T. 5. Durnick and A. H. Kalantar, Chem. Phys., 1977, 20, 347. T. J. Durnick and A. H. Kalantar, J . Chem. Phys., 1977, 66, 1914. S. D. Colson, R. E. Turner, and V. Vaida, J. Chem. Phys., 1977, 66, 2187. W. Moehle and M. Vala, Chem. Phys. Letters, 1976, 41, 149. R. H. Clarke and H. A. Frank, J. Chem. Phys., 1976,65,39, G. D. Gillispie and E. C. Lim, J. Chem. Phys., 1976,65,2022.

Table 5 Rate constants for triplet state spin sublevel population ( P J and decay (ki)in polycyclic hydrocarbons k,/s-l P, : P, : P, a Molecule Sohen t k,/s-l k,/s-l 0.7 rt 0.1 0.4 -t 0.1 0.1 i. 0.05 1.0 : 0.2 : 0.02 Naphthalene [2H,]Naphthalene [2H,o]Anthracene [2H,o]Anthracene [2H,,]Tetracene [2H,2]Tetracene 1,3-Benzanthracene 1,2 ;3,4-Dibenzant hracene Pyrene 3,4-Benzpyrene 3,4-Benzpyrene

n-Heptane n-Heptane n-Nonane n-Nonane n-Heptane n-Octane Fluorene n-Octane n-Heptane

36.1 rf: 3.6 11.2 k 0.9

382 2 44 222 f 31 4.32 & 0.56 2.3 f 0.5

2.5 L- 0.5 3.61 f 0.18 4.32 & 0.22

19.9 f 1.4 6.5 f 0.9 604 f 55 435 -t 47 3.65 5 0.48 0.9 f 0.3 2.8 f 0.5 13.91 & 0.42 12.2 f 1.9

5.08 f 0.35 1.9 f 0.9 124 k 11 51.5 k 7.8 0.49 k 0.06 0.5 5 0.1 0.6 f 0.2 9.2 & 1.1 5.06 -t 0.74

1.0 : 0.5 : 0.03 1.0 : 0.7 : 0.1 0.63 : 1.0 : 0.20 0.51 : 1.0 : 0.12 1.0 : 0.85 : 0.11 1.0 : 0.38 : 0.21 1.0 : 0.15 : 0.3 1.0 : 0.6 : 0.8 1.0 : 0.8 : 0.2

a Axis designations: x , y refer to in-plane molecular axis, z to out-of-plane molecular axis. For molecules with symmetry lower than CZu, the x axis is taken as the top zero-field spin sublevel, y the middle zero-field level, and z the lowest energy zero-field level. H. Six1 and M. Schwoerer, 2. Naturforsch., 1970, 25a, 1383. R. H. Clarke and J. M. Hayes, Chem. Phys. Letters, 1974, 27, 556.

Photophysical Processes in Condensed Phases

119

Staerk 247 has described a 'triplet' spectrophotometer ; absorption and emission data, spectral as well as time-resolved, are obtained under identical conditions. Such a system will alleviate the need to rely on data from different sources in the evaluation of T-T annihilation parameters. Data are presented for 1,2-benzanthracene in n-hexane. Energy can be transferred in molecular crystals by singlet and triplet Frenkel excitons. Decay of the singlet excitons by fission, as it is called, can produce two triplet excitons, which in turn can take part in bimolecular annihilation, known as exciton fusion. This can result in the formation of one singlet exciton and lead to the occurrence of delayed fluorescence. Measurements of the decay of triplet excitons in crystalline tetracene248provide a rate constant /3t z 5 x lo3 s-l and a bimolecular annihilation rate constant ytt z 2 x 10-l1 cm3s-l. These values differ from previously reported ones by two orders of magnitude, but are close to those reported for anthracene and pyrene. These anomalies are accounted for by the different experimental approaches, whereby ytt is measured directly through the dependence of the quantum yield on intensity, whereas other determinations have involved arguments based on equilibria between fission and fusion processes. Aladekomo et aZ.,249in another paper on tetracene, try to . ~ ~ the~ diffusion remove the anomaly in the results of Arden et ~ 1 by measuring coefficient Dab. Direct measurements have hitherto been made for anthracene and naphthalene, such measurements being difficult to carry out on tetracene because of the weak delayed fluorescence. Observations of the influence that a magnetic field has on the mutual annihilation of triplet excitons have been made with 9,lO-dichloroanthracene and in platinum phthalocyanine single Photodetrapping in 9,lO-dichloroanthracene has been shown to be due to energy transfer from triplet exciton to trapped positive holes. Measurements of phosphorescence decay on platinum phthalocyanine under magnetic fields of up to 100 kG by Q-switched ruby laser excitation provide an annihilation rate constant of the order of lo-'* cm3s-1.262 The influence of a n electric field on the spin alignment of triplet traps of the polar 4,4'-dichlorobenzophenone neat crystal has been investigated.253 This changes the phosphorescence intensity and the phosphorescence microwave double resonance (PMDR) signals, indicating a field dependence of the structure and dynamics of triplet excitons, which in turn leads to a change in the relative pumping rates to the spin levels of both shallow and deep traps. The field is effective in coupling the exciton band anisotropically to the spin levels of the shallow traps. Several papers have appeared on the influence of solvent and of environment on triplet s t a t e ~ . ~ ~Highly * - ~ ~resolved ~ phosphorescence spectra arising from the 947

24B 260

2s1 262

263 264 256

Ls6 2s7 268

26B

H. Staerk, J . Luminesence, 1976, 11, 413. W. Arden, M. Kotani, and L. M. Peter, Phys. Status Solidi (b), 1976, 75, 621. J. B. Aladekomo, S. Arnold, and M. Pope, Phys. Status Solidi (b), 1977, 80, 333. W. Arden, M. Kotani, and L. M. Peter, Phys. Status Solidi (b), 1976, 75, 621. M. Kotani, Chem. Phys. Letters, 1976, 43, 205. K. Kaneto, K. Yoshino, and Y. Inuishi, Chem. Phys. Letters, 1976, 40,505. S. J. Sheng and M. A. El-Sayed, Chem. Phys. Letters, 1977, 45, 6. 0. S. Khalil and L. Goodman, J . Phys. Chem., 1976, 80, 2170. M. A. Winnik and A. Lemire, Chem. Phys. Letters, 1977, 46, 283. H. 5. Pownall and W. W. Mantulin, Mol. Phys., 1976, 31, 1393. D. J. W. Barber and J. T. Richards, Chem. Phys. Letters. 1977, 46, 130. G . Moller and A. M. Nishimura, J . Phys. Chem., 1977, 81, 147. E. M. Schulman, J. Chem. Educ., 1976, 53, 522.

120

Photochemistry

3B19(n7r*)-f lAg electric dipole-forbidden transition in 9,lO-anthraquinone 254 (AQ) and in [2H,]AQ at 4.2 K in n-alkanes are found to be sensitive to the unitcell characteristics of the n-alkane. In n-hexane and n-heptane, experimental analysis shows that the molecule retains its inversion centre, and the transition is vibronically induced by &type vibrations. Analysis of the weak bands allows assignment of an odd-quantum progression, a bl, skeletal deformation. This has been interpreted as evidence for potential surface distortion of the 3 B 1 g state arising from 3Au(n.r*)-3B1,(n~*)vibrational-electronic interaction. Phosphorescence quantum yields and emissive lifetimes were obtained for a series of Independent values of alkyl esters of benzophenone-4-carboxylic 185 f 15 s-l were obtained for the radiative rate constants. This value was independent of the chain length and of the solvent. This latter observation is somewhat unexpected, as it implies that the solvents examined shift the relative energies of the coupled pairs of singlet and triplet states similarly. Analysis shows the differential solvation in polar solvents of the relaxed and FranckCondon ground states of the ketone.256 The phosphorescence of xanthone and [180Jxanthone in ethanol-ether-isopentane (EPA) and in 3-methylpentane (3-MP) has been reported for the temperature range 12-125 K.25s Dual phosphorescence was observed, and attributed to the existence of two different conformations of the ground state, each of which has a characteristic triplet state. Quantum yields of xanthone phosphorescence in 3-MP and EPA are 0.65 2 0.07 and 0.85 k 0.09, respectively. The phosphorescence lifetimes are 115 5 5 nis in 3-MP and EPA, respectively. An anomalous blue shift has been reported 257 for bilirubin IX-ain solvents of increasing dielectric constant, although it displays a high extinction coefficient that is typical of a n +- n* transition, which would normally exhibit a red shift. This transition has been interpreted in terms of the presence of an amide group on the terminal pyrrole rings, and evidence has been presented of an emission at 77 K that is attributable to phosphorescence from an excited triplet state of energy 230 kJ mo1-l. The effects of solvent and substitution on the phosphorescence properties 258 of dimethylnaphthalene (DMN), quinoline, hydroxyquinolines (HQ), indene, indole, azaindole, benzimidazole, and azabenzimidazole have been reported. The properties include rate constants for decay of phosphorescence from the lowest triplet state and zero-field splitting parameters arising from the anisotropic spin-spin interaction of the triplet electrons. Observation of phosphorescence in organic materials requires low temperatures and the rigorous exclusion of oxygen. This imposes restrictions on the ease with which it can be demonstrated. However, when ionic organic materials are adsorbed onto surfaces having free hydroxy-groups and the support is rigorously dried, intense visible phosphorescence (of long lifetime) can be seen.259 This is demonstrated in Figure 8. The effect of deuteration260in slowing the radiationless decay of the triplet state of several free porphyrin bases has been found to depend critically on the position of the deuterium. Deuteration of either C-H bonds or N-H bonds produces significant effects on the radiationless decay rate constants. The 260

R. P. Burgner and A. M. Ponte Goncalves, Chem. Phys. Letters, 1977, 46, 275.

Photophysical Processes in Condensed Phases

121

effect of deuteration of the imino positions was found to be much larger than that of deuteration of the ring. Phosphorescence lifetimes have been given for phenanthrene, the five monodeuteriophenanthrenes, and perdeuteriophenanthrene in EPA and 3-methylpentane at 76 K.2e1 A significant variation in lifetime of up to 34% is noted with the position of deuteration: effects of deuteration on the phosphorescence lifetimes of benzonitrile and phenyl isocyanide in several solvents at 77 K have also been reported.262 The observed increases in the phosphorescence lifetimes upon deuteration have been attributed to a solventindependent decrease of the radiationless TI + Sorate constant.

WAVELENGTH,

nm

Figure 8 Comparison of emission and excitation spectra of sodium naphthalate absorbed on paper and frozen in solution (Reproduced by permission from J. Chem. Educ., 1976,53, 523)

The effect of molecular weight on the delayed fluorescence, phosphorescence spectra, and decay curve measurements at 77 K have been given for samples of poly(2-vinylnaphthalene) over the range of molecular weights 15 600-505 The results show an increase in intensity of delayed fluorescence relative to phosphorescence up to the sample of solution-polymerized material (mol. wt. = 100 OOO) of highest molecular weight, with a levelling off for the bulk polymers of higher molecular weight. Arguments have been put forward to indicate a triplet exciton migration distance of up to 700 chromophore units, the shortest triplet exciton capture time being approximately 0.3 s. The use of flash photolysis in the study of triplet-state spectra of rhodamine 6G, NN’-diethylrhodamine, and rhodamine B 264 and of 6,6’-di-n-hexyloxythioindigo, 5,5’-diethylselenoindigo,and four 5,5’-dialkylthioindigo dyes in EPA glass matrices have been The effect of aggregation on the radiative z61 262

*63 26p

266

J. C. Miller, K. U. Breakstone, J. S. Meek, and S. J. Strickler, J . Amer. Chern. Soc., 1977, 99 1142. J. D. Laposa, R. A. Nalepa, and G. L. Le Be), Mol. Photochem., 1976, 7 , 465. N. F. Pasch and S. E. Webber, Chem. Phys., 1976, 16, 361. V. E. Korobov, V. V. Shubin, and A. K. Chibisov, Chem. Phys. Letters, 1977, 45, 498. D. Schulte-Frohlinde, H. Herrmann, and G. M. Wyman, 2. phys. Chern. (Frankfurr), 1976, 101, 115.

122

Photochemistry

(T, -+ So) and non-radiative (7" -+ So and S, -+TI)transitions in xanthene dyes, eosin, and erythrosin dyes has been investigated as a function of dye concentration.266 The triplet decay time ( T ~ )decreases, while the relativeyields of phosphorescence to fluorescence increase with increasing concentration. The photochemistry of L-histidine in aqueous solutions has been studied at 254 and 185 nm.267Under an inert atmosphere, the photolysis quantum yield @ depends on the histidine concentration, the degradation originating from the triplet state. In the presence of oxygen, the photodegradation is more rapid, and essentially consists in an oxidizing cleavage of the imidazole ring. Andrews et ~ 1 have. described ~ ~ a~ simple algorithm for the analysis of tripletstate reaction-rate data from flash photolysis where an initial pseudo-first-order triplet reaction is overlaid by a second-order decay of products. This procedure is illustrated in the case of the photoreduction of fluorenone triplet by 1,4diazabicyclo-octane, a reaction which yields a pair of radical ions. The rise time of the T, +- 7 ' absorption of anthrone in benzene was found to be 70 ps, while that of fluorenone changed from 140 ps in cyclohexane to 12 ns in acetone.269 This effect was attributed to a change in character of the lowest excited singlet state of fluorenone from nrr* in acetone to nr* in cyclohexane. The lowest singlet-singlet transition in the three isomeric pyridine aldehydes has been assigned n + n* character, since only phosphorescence is detectable at 77 K.270 Pyridinium phosphorescence 271 begins to build up at 334 nm in acid ethylene glyco1,rand has a lifetime of 3.5 s and a quantum yield of 0.04. The non-phosphorescence of pyridine has been discussed in terms of a pseudo-JahnTeller mixing of the 3nn* and 3nn* states. Solvent and protonation effects on absorption and phosphorescence spectra of 2-, 3-, and 4-benzoylpyridines have been investigated.272 Triplet-triplet energy transfer from benzoylpyridines to biacetyl has been studied in benzene and aqueous neutral and acidic solutions, and the behaviour has been compared with that of benzophenone. Triplet yields and lifetimes of the three isomeric benzoylpyridines were found to be somewhat Table 6

Triplet lifetimes and yields of benzophenone (Bp) and benzoylpyridines (BP) in benzene solution Compound

BP 2-BP 3-BP 4-BP

1O%/s

5.3 0.6 0.54 k 0.07 1.3 zk 0.1 0.41 k 0.04

@f

1 0.7 k 0.1 0.9, k 0.1 0.7 & 0.1

lower than for benzophenone (see Table 6 ) . The position of the heteroatom in a benzoylpyridine greatly affects the sensitizing power in aqueous solution, the 2-isomer being inactive: in an acidic aqueous medium there is a decrease in triplet lifetime of 3-benzoylpyridine, the 2- and 4-isomers being inefficient sensitizers. 266 267 268

269

270

271 272

N. B. Joshi and D. D. Pant, J. Luminescence, 1976, 14, 1. C . Hasselmann and G. Laustriat, Ex. du J. Chinz. phys., 1976, 73, 767. L. 5. Andrews, J. M. Levy, and H. Linschitz, J. Photochem., 1976, 6, 355. T. Kobayashi and S . Nagakura, Chem. Phys. Letters, 1976, 43, 429. M. R. Padhye and C . J. Jahagirdar, Indian J . Chem., 1975, 13, 1300. A. G. Motten and A. L. Kwiram, Chem. Phys. Letters, 1977, 45, 217. G. Favaro, J.C.S. Perkin II, 1976, 133.

Photophysical Processes in Condensed Phases

123

+

Self-quenching data (3B* B + 2B; kBq)in benzene and carbon tetrachloride have been presented for a series of para,pava’-disubstitutedb e n z ~ p h e n o n e s . ~ ~ ~ Hammett plots of k,, with oP+ reveal a discontinuity, this being attributed to n-type exciplexes operating over regions of the Hammett plots where negative p values are observed, whereas r-type exciplexes operate over the regions where positive p values are seen. In the intermediate regions, mixed mechanisms operate. The low-temperature emission has been attributed to excitation of a ground-state complex. A kinetic study of the reaction between photoexcited benzophenone and a series of n-alkanes has also been carried out in dilute carbon tetrachloride s ~ I u t i o n s . ~This ~ * system has been analysed as a model for the behaviour in polymer systems. It is widely known that phenazine276is very weakly fluorescent, but phosphoresces strongly. This implies very efficient intersystem crossing to the triplet state. The build-up of triplet-triplet absorption for phenazine has been measured by picosecond spectroscopy. The measured rate constants for the build-up of the Tl states are 7 x 1Olo and 5 x 1O1O s-l in iso-octane and in methanol, respectively. The fast build-up of the Tl state is due to the strong spin-orbit coupling between Sl(nn*) and Tl(m*). Triplet-triplet absorption spectra, quantum efficiencies of intersystem crossing, triplet extinction coefficients, and triplet energy levels of 1,6-diphenylhexa-l,3,5in ethanol and benzene have been triene and l,S-diphenylocta-1,3,5,7-tctraene Quantum efficiencies of intersystem crossing were found to be less than 0.03 for both molecules in both solvents, thereby showing that the excited singlet is efficiently deactivated by internal conversion and fluorescence emission. Quantum yields, lifetimes, and the polarization of phosphorescence of 4- and 5-aminopyrimidines have also been measured in solvents of different polarity at 90 K.277 Phosphorescence data have been presented for alicyclic and acyclic ketones,278 the medium-sized cycloalkanones exhibiting unusually high quantum yields of phosphorescence at 77 K. These results are presented in Table 7. Fluorescence and phosphorescence spectra of isomeric trifluorotoluidines have been and also the phosphorescence spectra of coronene, chrysene, and 1,2-benzpyrene in 1-bromobutane at 4.2 K under laser excitation of the +- So transition.280 The phosphorescence assumes a line structure which is absent under monochromatic S1+ So excitation, and is due to the effect of the excess energy of the S1 -+So transition causing environmental inhomogeneities. The lowest triplet state of single crystals of 9,lO-anthraquinone 281 has been assigned as 3BIg,located at 22 153 cm-l. Observations of the Stark effects on the a78

a74 275 276

276

277 278

27D 280

281

M. W. Wolf, R. E. Brown, and L. A. Singer, J. Amer. Chem. Soc., 1977, 99, 526. M. A. Winnik and E. Shum, Macromolecules, 1976, 9, 875. Y. Hirata and 1. Tanaka, Chem. Phys. Letters, 1976,43, 568. Y. Hirata and I. Tanaka, Chem. Phys. Letters, 1976, 43, 568. R. Bensasson, E. J. Land, J. Lafferty, R. S. Sinclair, and T. G. Truscott, Chem. Phys. Letters, 1976, 41, 333. J. Smagowicz and K. L. Wierzchowski, J. Luminescence, 1976,14,9. N . 4 . C. Yang, D. M. Shold, and C. V. Neywick, J.C.S. Chem. Comm., 1976, 727. M. R. Padhye and S. A. Agnihotry, Indian J. Chem., 1976; 14A, 1. E. I. Alshits, R. I. Personov, and B. M. Kharlamov. Optics and Spectroscopy, 1976, 41,474. J. A. Galaup, 5. Megel, and H. P. Trommsdorff, Chem. Phys. Letters, 1976, 41, 397.

124

Photochemistry

Table 7 Measured values of quantum yields (ap) and lifetimes (7& with calculated values for rate constants of phosphorescence ( k p ) and of non-radiative

Ketone a Cyclopentanone Cyclohexanone Cycloheptanone Cyclo-octanone Cyclononanone Cyclodecanone Cycloundecanone Cyclododecanone Cyclotridecanone Cyclopentadecanone Acetone

Inm 460 480 440 435 430 435 440 445 440 440 460

Pentan-2-one Pentan-3-one Hexan-Zone Hexan-3-one Hept an-4-one Octan-Zone Octan-4-one Adamatan-Zone 2,2,4,4-Tetramethylpentan-3-one

460 460 460 460 460 450 460 I

460

rims 1.2 0.9 2.7 3.7 4.3 4.3 5.0 3.1 4.7 4.4 0.6, 0.6 0 0.4,' 0.33 j 0.8 1.1, 1.260 0.7 1.o 1.o 1.o 1.o

-

6.7, 8.60

kP a

0.1 1 0.064 0.33 0.42 0.41 0.44 0.36 0.37 0.3 1 0.30 0.042, 0.043,n 0.03 0.096 0.10 0.093 0.079 0.070 0.13 0.067 < 0.005 0.29, 0.89 j

92 68 120 110 95 100 72 120 66 67 71

k-t 740 1500 250 200 140 130 f

130 200 150 160 1600

120 91 130 79 70 130 67

1100 820 1300 920 930 870 930

43

110

-

-

a All values are for phosphorescence only. f 5 nm. f O . l ms. Based on the phosphorescence of benzophenone, (DD = 0.84, as the secondary standard, s. L. Murov, Ph.D. thesis, f Calculated from k-t = 7t-l University of Chicago, 1967. Calculated from k, = (D,,,-l. D. S. McClure, J. Chem. Phys., 1949, 17,905. E. H. Gilmore, G. E. Gibson, and D. S. k,. McClure, ibid., 1952, 20, 829; 1955, 23, 399. R. F. Borkman and D. R. Kearns, ibid., 1966, 44, 945. 3 M. O'Sullivan and A. C. Testa, J. Amer. Chem. SOC.,1970, 92, 258.

phosphorescence spectra of p-benzoquinone 282 in naphthalene crystals at temperatures between 1.7 and 30 K reveal a bond splitting of 24 cm-l, attributed to a double-minimum potential in the Tl state. The zero-field splitting of the triplet state of magnesium porphin283 solvated in ethanol is given by D = 0.035 cm-l and I El = 0.010 cm-l. Decay rates for the upper two spin components are found to be ca. 2Os-l, while that of the lower_-componentis ca. 2 s-l. The S1-+ T' non-radiative process in duraldehyde 284 has been analysed by the low-field Zeeman PMDR method. The results show that the majority of molecules undergo a direct S,(nn*) +T*(nm*) transition, but 10-30% of the molecules follow an indirect Sl(nn*) -+T,(nn*) -+T,(nn*) pathway. This latter transition has been attributed to the pseudo-Jahn-Teller forces which mix the T,(nn*) and Tl(mn*) states that are separated by only 0.05 eV. The two-photon phosphorescence excitation spectrum of triphenylene 285 in PMMA matrix at 77 K has been reported for the spectral region 32000282

283 284

285

Y. Miyagi, M. Koyanagi, and Y. Kanda, Chem. Phys. Letters, 1976,40,98. G. Jansen and J. H. van der Waals, Chem. Phys. Letters, 1976, 43, 413. A. Campion and M. A. El-Sayed, J. Phys. Chem., 1976, 80,2201. L. Singer, Z. Baram, A. Ron, and S. Kimel, Cheni. Phys. Letters, 1977,47, 372.

Photophysical Processes in Condensed Phases

125

36 000 cm-l. The band origin of the So -+ S2 transition not easily seen with onephoton absorption is seen at 33 200 cm-l. Experiments have been made on phosphorescence in poly(methy1 methacrylate) glasses to resolve previously reported anomalies in media which would be expected to be convenient for use in measuring phosphorescence at room temperature.28s The role of the triplet state in photosynthesis has been reviewed by Norris 287 and by McIntosh and Bolton.2R8E.s.r. signals which exhibit the chemically induced dynamic electron polarization phenomenon (CIDEP) are observed from free radicals associated with the PS I1 of chloroplasts and whole algae, This observation provides additional evidence for the possible role of the triplet states as intermediates in the mechanism of photochemical electron transfer in photosynthesis. Zero-field splitting parameters have been reported for 8-methoxypsoralen289which distinguish it from the less active derivatives and from the parent coumarin. A method has been developed for studying trapping of triplet excitons by observation of the dependence on temperature.290 Self-reversal of the exciton phosphorescence of 1,4-dibrornonaphthalene has been A computational analysis has been made of the conformation of the triplet eximer of naphthalene which shows it to be different from that of the singlet.292 A consequence of this conformation is breakdown of the 0 - 7 ~ separation which leads to enhancement of the TI So transition. The nature of the Tl(mr*) + So radiationless transitions in aromatic molecules having nonbonding electrons 293 has been examined within the framework of the pure spin adiabatic Born-Oppenheimer bases set. A heuristic model for intersystem crossing and zero-field triplet splittings has been established for metal phthalocyanines.2 9 4 --f

Exciplexes, Triplet Quenching, and Energy-transfer Processes.-The transfer of triplet excitation energy between isotopic traps in benzene isotopic mixed crystals 296 has been investigated using both steady and pulsed excitation sources. Application of the excitation-tunnelling transfer mechanism predicts the correct dependence of the transfer rate on trap depths. I n the first of two 298 hydrogen-abstraction reactions of electronically papers by Formosinho excited carbonyl compounds are treated as radiationless transitions, using the tunnel-effect theory. This theory is then applied to the triplet states of thioketones, quinones am-aromatic compounds, olefins, and azobenzenes. Good agreement with experiment was found and in general it has been concluded that the TT* states are more reactive than m*states. Rate constants for the quenching of 295p

2979

286

287 288

W. Windhager, S. Schneider, and F. Dorr, J. Photochem., 1976, 6, 69. J. R. Norris, Phorochem. and Photobiol., 1976, 23, 449. A. R. McIntosh and J. R. Bolton, Nature, 1976, 263,443. T . A. Moore, A. B. Montgomery, and A. L. Kwiram, Phorochem. and Photobioi., 1976, 24, 83.

290

2gs 204

296 296 297 298

D. D. Dlott and M. D. Fayer, Chem. Phys. Letters, 1976,41, 305. R. M. Macfarlane, U. Konzelmann, and D. M. Burland, J . Chem. Phys., 1976, 65, 1022. A. K. Chandra and E. C. Lim, Chem. Phys. Letters, 1977, 45, 79. N. Kanamaru and E. C. Lim, J. Chem. Phys., 1976,65,4055. T.-S. Huang, K. E. Rieckhoff, E.-M. Voigt, and E. R. Menzel, Chem. Phys., 1977, 19, 25. S. D. Colson, F. B. Tudron, R. E. Turner, and V. Vaida, J. Phys. Chem., 1976, 80, 2196. F. B. Tudron and S. D. Colson, J. Chem. Phys., 1976,65,4184. S . J. Formosinho, J.C.S. Faraday IZ, 1976, 72, 1313. S. J. Formosinho, J.C.S. Faraday ZZ, 1976, 72, 1332.

126

Photochemistry

phosphorescence of 3n7r* benzophenone in benzene solution and in perfluorinated solvents have been obtained in a variety of substrates, principally a m i n e ~ . ~ ~ ~ The abstraction of hydrogen from suitable donors by 3nn* benzophenone can be categorized into two general types, namely: (a) direct hydrogen transfer from the donor molecules to the photoexcited benzophenone :

and (b) charge-transfer pathways as with amines:

The above reaction schemes do not allow easy prediction of reaction pathways to be made, and a variety of rationalizations have been suggested. The results of hydrogen-abstraction reactions by 3nn* species have been presented and compared with those calculated using Formosinho’s The results are summarized in Table 8. Excellent agreement persists between observed and

Table 8 Second-order rate constants for hydrogen-abstraction reactions of ( 3 n ~ *benzophenone ) as observed and predicted by the author of ref. 299 ka/dm3 mol-1 s-l determined by other workers a 2.29 x lo9 2.69 x lo9

a

Substrate Triethylamine NN-Dimethylaniline n-Hexylamine Isopropylamine Cyclopentane Di-t-butyl sulphide t-Butylamine

kcz/dm3mol-l s-l (observed) 2.42 x 109 3.4s x 109 4.18 x lo8 2.95 x lo8 5.07 x lo6 5.74 x 107 6.8 x lo8

k/dm3 mol-l s-l (predicted) 2.51 x 109 3.16 x 109 8.91 x 1 0 7 1.41 x lo8 3.16 x 106 4.47 x 107 No predicted value is given

k/ka 1.04 0.9 1 0.21 0.48 0.62 0.78

See S. J. Formosinho, J.C.S. Faraday II, 1976, 72, 1313.

predicted rate constants for the substrates NN-dimethylaniline and triethylamine. Reasonable agreement is seen with di-t-butyl sulphide and cyclopentane, but a significant difference is seen for the two substrates n-hexylamine and isopropylamine. An explanation of these differences is sought in the parameters which define the rate of tunnelling. Energy transfer from the second triplet state of naphthalene, l-chloronaphthalene, and pyrene has been The second triplet state of naphthalene lies slightly below the S1 level. In order to increase the trapping efficiency of the very short-lived upper states, a third compound is introduced, at high concentration, which traps the Tzstate of the donor and which transfers the energy, via similar molecules, to the acceptor. The energy carrier was benzene, 300

G . D. Abbott and D. Phillips, Mol. Photochem., 1977, 7,289. C. C. Ladwig and R. S. H. Liu, J . Amer. Chem. SOC., 1976, 98, 8093.

PhotophysicaI Processes in Condensed Phases

127

the acceptor being endo-bicyclopentadiene. The average lifetimes of the & state of naphthalene and l-chloronaphthalene were 12 f 2 and 13 f 2 ps, respectively. Evidence has also been given for delayed fluorescence in naphthalene and phenanthrene arising from electron-cation r e ~ o r n b i n a t i o n .The ~ ~ ~existence of the triplet-triplet annihilation-delayed fluorescence of polyacenes in ethanol and EPA glasses has been known for several years. The possibility that this emission might originate from a two-photon photoionization followed by an electroncation recombination-delayed fluorescence (RDF) has been considered. RDF emission has been observed in anthracene single crystals, and of late the thermoluminescence emission of phenanthrene crystals that have been excited by U.V. light at 77 K has been investigated and attributed to an electron-cation recombination event. The intensities of delayed fluorescence are proportional to the square of the phosphorescence intensities and also to the square of the intensity of the excitation light; this shows the biphotonic nature of the delayed fluorescence emission. A study of the quenching of triplet sensitizers by a series of electron-acceptor salts of N-heteroaromatic compounds has been carried I n systems where both transfer of triplet excitation energy and electron transfer are energetically favourable, both processes occur competitively, and at the anticipated rates. The rate constant for triplet energy transfer from phenanthrene to stilbene shows non-Arrhenius dependence when studied in 2-methylpentane-2,4-diol over the temperature range 190-350 K 3 0 3 The results can be described by an equation of the form k = AT exp [ - B / ( T - To)],where B = (850 k 100) K and To = 158 K. Triplet energy transfer from poly(riboadeny1ic acid) to decarboxyThe formation of lo2from photoformylkynurenine has been excited trans-4-nitrostilbenes confirms previous findings on the transcis isomerization mechanism.so6 The triplet diffusion coefficient ( D ) of anthraceneSo6has been measured as a function of solvent and concentration (see Table 9). At low concentrations, D (in methylcyclohexane) = 2.67 ( 2 0.18) x cm2s-l and D (in cyclocme s-l. The effect of temperature (120octane) = 1.42 ( f 0.14) x 293 K) on the deactivation of triplet anthracene 307 by transition-metal ions, and of triplet phenanthrene by lanthanide ions, has been investigated in methanolwater (9 : 1) mixtures. The small activation energies observed are explained in terms of the shielding effects of the ligand molecule. At low temperatures, all the reactions become diffusion-controlled. Evidence for a triplet-singlet resonance energy transfer process involving the carbonyl-based photoactive chromophoric species in commercial polypropylene has been 9,lO-Dibromoanthracene was used as an efficient triplet quencher, and the critical transfer distance was found to be ca. 15 nm. Transfer of triplet excitation energy in vinyl M. Ewald, T. T. Quach, and G. Durocher, Canud. J. Spectroscopy, 1976,21, 175. A. R. Gutierrez, T. J. Meyer. and D. G. Whitten, Mol. Photochem., 1976, 7 , 349. F. S. Dainton, M. S. Henry, M. J. Pilling, and P. C. Spencer, J.C.S. Faruduy I , 1977,73,243. M. P. Pileni, P. Walrant, and R. Santus, Chem. Phys. Letters, 1976, 41, 12. 3 0 6 H. 5. Kuhn, R. Straatmann, and D. Schulte-Frohlinde, J.C.S. Chem. Comm., 1976, 824. 30u R. D. Burkhart, J. Phys. Chem., 1977, 81, 370. 3 ~ 7E. J. Marshall, N. A. Philipson, and M. J. Pilling, J.C.S. Furuduy ZI, 1976,72, 830. J. Homer and J. F. McKellar, Chem. undZnd., 1977, 158. 301 303

128

Photochemistry

polymers with pendant carbazolyl groups is evidenced by delayed fluorescence, phosphorescence, and (in some cases) excimer phosphorescence emission.3o9 Several papers have appeared on the role of the triplet states in energy-transfer reactions in organic dye p h o t o ~ h e m i s f r y . ~ ~ ~ - ~ ~ ~ Table 9 Lifetimes and diffusion coeficients of triplet anthracene in two diferent solvents and at different concentrations, at 25 "C [Anthracene]/

mol 1-1 x lo5

Solttent

Methylcyclohexane Met h ylcyclohexane Methycyclohexane Met hylcyclohexane Cyclo-octane Cyclo-octane Cyclo-octane a Average deviations in T values were about 2 %. 10 %. 0.99 2.0 4.9 12 4 0.94 1.89 4.72

r/ms a

D/cm2s-1 x lo6

8.19 3.22 5.75 3.15 4.8 1 3.76 1 .s2

2.67 1.52 1.24 < 1.0 1.42 1.33 0.9

Average deviations in D values were about

Although the reactivities of the nr* and w r * ketone triplets are now reasonably well defined for hydrogen-abstraction reactions, less is known of these states in charge-transfer reactions. Of particular interest is the structure of the nonluminescent exciplex that is presumably formed in the quenching 19-Naphthyl y-dimethylaminopropyl ketone undergoes inefficient (0d O.Ol), unquenchable type I1 photoelimination to 2-acetonaphthone in protic and aprotic solvents. The efficiency of the ketone as a photosensitizer indicates that it forms s) triplets in good yield (@isc = 0.76-0.88). The triplet, in long-lived (7 3 protic solvents, yields acetonaphthone (@ = 0.17 in methanol), and it has been postulated that the mr* triplet undergoes charge-transfer interactions with the amino-group in all solvents: only protonation of the exciplex catalyses its rearrangement into a diradical. The lack of rearrangement in a protic solvent implies a rather restricted cyclic geometry of the exciplex. Oxciplex emission from naphthalenes in polymer matrices, assigned to the process shown in equation (6), has been described.317 The assignment of the emission to an oxciplex

rather than to sensitized excimer emission has been made on the basis of the similarity of the emission to oxygen-induced absorption, and from energetic, kinetic, and intensity considerations. The role of the heavy-atom-induced singlet-triplet intersystem crossing as a path for decay in charge-transfer exciplexes has been reported for quaternary 309

310 311 312

313

314

s15 316

317

A. Itaya, K. Okamoto, and S. Kusabayashi, Bull. Chem. SOC.Japan, 1976, 49, 2037. E. Vogelmann and H. E. A. Kramer, Photochem. and Photobiol., 1976, 23, 383.

P. D. Wildes, N. N. Lichtin, and M. Z . Hoffman, Photochem. and Photobiol., 1977, 25, 21. M. Yamashita, A. Kuniyasu, and H. Kashiwagi, J . Chem. Phys., 1977, 66, 986. E. Vogelmann, S. Schreiner, W. Rauscher, and H. E. A. Kramer, Z . phys. Chem. (Frankfurt), 1976, 101, 321. A. D. Kirsch and G . M. Wyman, J. Phys. Chem., 1977, 81, 413. A. K. Chibisov, J. Photochem., 1976, 6 , 199. P. 5. Wagner and D. A. Ersfeld, J. Amer. Chem. Soc., 1976, 98,4515. R. D. Kenner and A. U. Khan, Chem. Phys. Letters, 1977,45,340.

Photophysical Processes in Condensed Phases

129

salts of 1,2-dipyridyIethylene~.~~* With the quenchers studied, charge-transfer exciplex formation is evident ; with the halogen-free quenchers, the exciplex decays rapidly, without isomerization. For most of the halogen-containing quenchers, the ‘heavy-atom effect’ involves quenching of the localized singlet to the exciplex followed by quantitative decay to the localized olefin triplet. This mechanism operates when the quencher oxidation potential is greater than 1.38 V, such that the exciplex lies energetically between the excited singlet and triplet states of the olefin; but when the exciplex lies lower in energy than the triplet, no isomerization is observed. Further reports of the heavy-atom effects and their exploitation in a variety of fields have been made.31s~320 Numerous reports have appeared on the use of flash photolysis and the similar technique of radiolysis in studying triplet absorption and kinetics. Quinoid compounds as a group are of considerable practical importance because they include a range of anthraquinone dyes and isoprenoid benzoquinones involved in photosynthesis and oxidative phosphorylation. Thus flash spectroscopy of 32a duroquinone in solution leads to the following principal IQO

hv ____,

kr 3Q

3Q

> ‘Qo

6H

(‘Qo is the ground state of duroquinone, 3Q the first triplet state, the semiquinone radical, QH2 the durohydroquinone, and SH the solvent). When the solution is flashed several times, the presence of QH, introduces a further reaction for the decay of 3Q, namely equation (1 1). Amouyal and Bensasson 3w investigated the transient species involved in the duroquinone and durohydroquinone reaction and showed the presence of 3Q and b H , together with the anticipated isosbestic points, in ethanol and cyclohexane. In water, 3Q and were observed, isosbestic points again being present. The formation of the triplet exciplex “Q,QHZ) as an intermediate in reaction (11) has also been suggested. ElectronIn watertransfer studies on Tl duroquinone have also been carried ethanol mixtures, the triplet abstracts electrons from a variety of substrates, forming durosemiquinone and oxidized donor, the rate constants for these redox processes being close to the diffusion-controlled limit (109-1010 m-l s-l). The duroquinone triplet also abstracts electrons from the solvents ethanol and acetone at a relatively slow rate whereas in n-hexane the triplet disappears predominantly uia a bimolecular triplet-triplet annihilation process. The effect of 8wsubstitution on flavin triplet and semiquinone properties has also been investigated.323 A

6

318 310 s20

s21 32a

s25

A. R. Gutierrez and D. G. Whitten, J . Amer. Chem. SOC.,1976, 98, 6233. T. Vo-Dinh, E. L. Yen, and J. D. Winefordner, Anafyt. Chem., 1976, 48, 1186. H. E. Lessing, A. Von Jena, and M. Reichert, Chem. Phys. Letters, 1976, 42, 218. E. Amouyal and R. Bensasson, J.C.S. Furuduy I, 1976, 72, 1274. R. Scheerer and M. Gratzel, J. Amer. Chem. SOC., 1977, 99, 865. D. E. Edmondson, F. Rizzuto, and G. Tollin, Photochem. and Photobiol., 1977, 25, 445.

130

Photochemistry

further paper dealing with the triplet excited states of a series of lY4-disubstituted anthraquinones 324 containing NHz, NHMe, NHAr, and OH groups in benzene gives triplet lifetimes of ca. 3 ps, the rapid decay being attributed to reversible hydrogen abstraction across the intramolecular hydrogen bond. Flash photolysis studies have also been reported for the triplet states of 5 - n i t r o q ~ i n o l i n e ~ ~ ~ ~ p-nitroaniline 326 (which, with that of 2-N-dimethylamin0pyridine~~~~ is probably a charge-transfer state), 9-fl~orenone,~~* and methyl 2-naphthyl ketone.329 Bifl~orenylidene,~~~ thionine triplet 3TH+ and its protonated form 3THz2+ (which provides evidence for the triplet exciplex formation),331rhoda mine R6G in ethanol s01utions,~~~ and tetramethylbenzidine 333 in organic solvents and aqueous micellar solutions have all been studied by flash photolysis. Nanosecond laser photolysis of indole 334 in cyclohexane produced transient absorption spectra attributed to indole triplets and indolyl radicals, the triplet lifetime of 16 ps being observed at low triplet concentrations. At higher concentrations, triplet-triplet annihilation takes place with a rate constant of 7 x lo9 s-l m-l. A complementary study of l-methylindole implied that the radical arose by dissociation of the N-H bond. The quenching properties of l-methylindole triplet 335 by various compounds have also been studied by pulse radiolysis and laser flash photolysis, charge-transfer interactions playing a predominant part in determining the quenching rate constants. The effect of quenching by oxygen on the triplet of l-methylindole, as studied by laser photolysis and pulse radiolysis, provides a rate constant k~ = 1.4 x 1Olodm3mol-1 s-l. This value is higher than that normally observed, and is attributed to the presence of low-lying triplet charge-transfer states which enhance the quenching efficiency.336 Inorganic ions have also been used to quench the triplet states of aromatic hydrocarbons and carbonyl Typical triplet-triplet energy transfer was demonstrated for NOz-, using flash and laser photolysis. Charge transfer is also an important mechanism for the very efficient quenching observed with many anions. Although benzophenone has been extensively used as a triplet sensitizer, comparatively little is known of xanthone, which has a high intersystemcrossing yield and a triplet energy of 310 kJ mol-l. A study of the laser photolysis of xanthone 339 reveals a peak around 600 nm, attributable to absorption by the first excited triplet state, and a triplet decay time of 92 ns in benzene. The 324

326 328 327 328

329 330

331

33a 333 334 336

336

s37 338

330

E. J. Land, E. McAlpine, R. S. Sinclair, and T. G. Truscott, J.C.S. Faraday Z, 1976,72, 2091. A. Cu and A. C. Testa, J. Photochem., 1976, 6 , 277. 1.Wolleben and A. C. Testa, J. Phys. Chem., 1977, 81, 429. J. Wolleben and A. C. Testa, Mol. Photochem., 1976, 7 , 277. A. Nakajima, Mol. Photochem., 1976, 7 , 251. D. I. Schuster and M. D. Goldstein, Mol. Photochem., 1976, 7 , 209. A. V. Sinoy and E. Van der Donckt, J.C.S. Faraday Z, 1976, 72, 2312. U. Steiner, G. Winter, and H. E. A. Kramer, Z . Nuturforsch., 1976, 31a, 1019. A. Dunne and M. F. Quinn, J.C.S. Faraduy Z, 1976, 72,2289. S. A. Alkaitis and M. Gratzel, J. Amer. Chem. SOC.,1976, 98, 3549. C. Pernot and L. Lindqvist, J. Photochem., 1976, 6 , 215. F. Wilkinson and A. Garner, J.C.S. Faraduy ZI, 1977, 73, 222. A. Garner and F. Wilkinson, Chem. Phys. Letters, 1977, 45, 432. A. Treinin and E. Hayson, J. Amer. Chem. SOC.,1976, 98, 3884. A. Garner and F. Wilkinson, J.C.S. Faraduy ZZ, 1976, 72, 1010. S. Yamamoto, K. Kikuchi, and H. Kokubun, Bull. Chem. SOC.Japan, 1976,49,2950.

Photophysical Processes in Condensed Phases

131

quenching constants of triplet xanthone by naphthalene, oxygen, and 3-methylindole are 9.5, 5.6, and 8.4 x lo9 dm3mol-I s-l, respectively. The hydrogen-bonding interaction of triplet 2-naphthol and l-anthrol with aromatic N-heterocycles 338 and the photocleavage reaction of naphthalene perifused c y ~ l o b u t a n e s which , ~ ~ ~ occur from upper triplet states, have been studied by flash photolysis. The triplet-triplet extinction coefficients of anthracene and 9-bromoanthracene have been determined by flash photolysis, using the method of Hadley and Keller.341 The respective values of E, are 53 000 k 2000, 85 700 & 3200, and 66 500 rt 3250 dm3mol-1 cm7. The use of pulse radiolysis and laser flash photolysis in the elucidation of problems directly applicable to medicine is demonstrated in a paper on the triplet excited state of The triplet state has an absorption maximum at 500 nm, a lifetime of 9 ps, an extinction coefficient of 8800 1 mol-lcm-l, an energy level of 150 kJ mol-l, and a rate of quenching by oxygen of 8.2 x lo8 mol-1 s-l. The significance of these data in the bilirubin photodestruction is discussed. In the study of exciplexes, pulse laser photolysis techniques have provided a time response better than those used hitherto. The system pyreneNN-diethylaniline has been studied in acetonitrile-toluene mixed solvents of varying composition.343 In the first few nanoseconds, the amine induces the pyrene*(S,) to undergo efficient intersystem crossing to produce pyrene*(T,). Subsequent changes were found to depend on the solvent composition. In the presence of MeCN, further pyrene*(Tl) is produced, possibly from the dissociation of a triplet exciplex having a neutral lifetime of 47 ns. The formation of triplet is discussed in terms of geminate ion separation and recombination. Dilute solutions of all-trans-retinal 334 in n-hexane have been studied, on the decapisosecond time-scale. A 8 ps pulse is used to excite the sample, the absorption spectra of the resulting transients being investigated over the 0-200 ps time-scale. Preliminary results reveal two transients : one corresponds to the triplet absorption with a rate constant of 34 ps (k 5), whereas the other is of lower energy than the T-T absorption, and appears instantaneously and decays away in ca. 20 ps. E.s.r. and Microwave Studies, and Related Triplet-state Topics.-In addition to the conventional optical techniques available for the study of triplet-state kinetics, continued interest has been shown in e.s.r., e.p.r., electron spin echo, and phos phorescence-microwave studies. The e.s.r. absorption spectrum of [2Hlo]pyrenein benzophenone 346 at - 20 "C has been reported. A slightly different procedure is outlined whereby the crystal is rotated about the laboratory-fixed crystal frame a, b, and c. Values of D and E were obtained which are the same as those previously reported for 12Hlo]pyrene in fluorene. 340 341 342

343 s44

K. Honda, A. Yabe, and H. Tanaka, Bull. Chem. SOC.Japan, 1976, 49, 2384. M. B. Ledger and G. A. Salmon, J.C.S. Fnroday II, 1976, 72, 883. E. 5. Land, Photochem. and Photobiol., 1976, 24, 475. I. P. Bell and M. A. J. Rodgers, Chem. Phys. Letters, 1976, 44, 249. R. M. Hochstrasser, D. L. Narva, and A. C. Nelson, Chem. Phys. Letrers, 1976, 43, 15. L. J. Noe and L. F. Wojdac, J. Phys. Chem., 1976, 80, 2519.

132

Photochemistry

The variation of the e.s.r. spectra of some rhodamine dyes as a function of concentration in methanol-water solutions at 90 K has been investigated.346 The spectra have been assigned to monomers and dimers, and information relating to the orientation of the molecules within the aggregates has been presented. The possible involvement of singlet oxygen in the photo-yellowing of l i g ~ ~ has in ~ prompted ~ ~ an investigation of the dependence of excitation of the triplet states on wavelength for a number of model lignin compounds in the region 300-400nm by the e.s.r. technique. It was shown that the triplet states lead either to phenoxyl radicals or to singlet oxygen, both of which have been postulated to play a role in the photo-yellowing of lignin. E.s.r. analysis has also been applied to triplet characterization of the bis(to1uene p-sulphonate) of hexa-2,4diyne-1,6-diol 348 and to aromatic compounds that have been orientated by stretching poly(viny1 alcohol) films that contain them.349 Both e . ~ . r . ~ ~and O optical detection of magnetic resonance (ODMR) 351 techniques have been applied to chlorophylls in order to obtain information relating mechanisms of intersystem crossing with the possible involvement of the excited triplet state in the initial photophysical processes of photosynthesis. Optically induced electron spin polarization 352 (OEP) results have also been presented for the triplet state of chlorophyll-a and -b in an organic glass, and for tetraphenyl-porphyrin and -chlorin free bases in a solid polymer at 200 K. The results show a dominant steady-state population in the top zero-field level associated with an arbitrary in-plane molecular axis. Cullick and Gerkin 353 have reported on the observation and characterization by e.p.r. of previously unreported single discontinuities in the zero-field splittings of triplet [2Hlo]phenanthrene in biphenyl and of triplet [2H8]naphthalene in biphenyl around a small temperature range near 39 K. Within this temperature range, single e.p.r. absorption signals characteristic of these systems above 39 K are converted into sets of four prominent e.p.r. absorption signals, which then persist to lower temperatures. The e.p.r. spectra of triplet exciton pairs in tetracene and charge-transfer crystals have also been Electron spin echo 356 and ODMR techniques have been applied to the analysis of the 'mini-exciton' in the naphthalene pair in a Cl,D8 host. An analysis using the phase memory time''7 as well as the shift in resonance frequency enables the energy separation between the two Davydov states, together with their zero-field splittings and lifetimes, to be calculated. Electron spin echo and [2HIO]-diphenylcarbene358 experiments of the ground triplet state of [lH,,]347 a48

350

351 352 353

354 356

356

H. Schmidt, J. Phys. Chem., 1976, 80, 2959. G. J. Smith, I. J. Miller, and W. H. Melhuish, Austral. J. Chem., 1976, 29, 2073. G . C. Stevens and D. Bloor, Chem. Phys. Letters, 1976, 40, 37. 5. Higuchi, T. Ito, M. Yagi, M. Minagawa, and M. Bunden, Chem. Phys. Letters, 1977, 46, 477. R. H. Clarke, R. E. Connors, T. J. Schaafsma, J. F. Klubenker, and R. J. Platenkamp, J. Amerr.Chem. SOC., 1976, 98, 3674. R. H. Clarke, R. E. Connors, H. A. Frank, and J. C. Hoch, Chem. Phys. Letters, 1977,45,523. J. F. Kleibeuber, R. J. Platenkamp, and T. J. Schaafsma, Chem. Phys. Letters, 1976, 41, 557. A. S. Cullick and R. E. Garkin, Chem. Phys. Letters, 1976, 42, 589. E. L. Frankevich, A. I. Pristupa, and V. I. Lesin, Chem. Phys. Letters, 1977, 47, 304. B. J. Botter, C. J. Nonhot, J. Schmidt, and J. H. Van der Waals, Chem. Phys. Letters, 1976 43, 210. C. Cheng, T. Lin, and D. J. Sloop, Chem. Phys. Letters, 1976,44, 576.

Photophysical Processes in Condensed Phases

133

embedded in benzophenone crystals at low temperatures (40-60 K and 4.2 K) show pronounced modulations due to proton and deuteron hyperfine interactions. A review of the electron spin echo experiments on the photochemically induced spin orientation in triplet diphenylcarbene has been given by D ~ e t s c h m a n , ~ ~ ' together with an interpretation of the highly preferential photochemical population of one of the triplet sublevels. The phosphorescence microwave photoexcitation spectrum of 1,4-dibromonaphthalene may be presented by monitoring the x-trap phosphorescence at 201 64 cm-l and modulating the amplitude of the 2942 MHz zero-field spin resonance.3s8 The spin alignment of the emitting Tl state has been found to depend upon the nature of the pumping state. Phosphorescence-microwave double resonance studies have been used to locate and characterize the electronic states of a naphthalene-tetrachlorophthalic anhydride charge-transfer The phosphorescent triplet state of p-chloroaniline 360 in a p-xylene host (n-trap and x-trap system) has been studied, using phosphorescence spectra, ODMR, and absorption spectra. The radiative properties of the individual TI spin states have been analysed, using the emission spectra for each spin state obtained by the microwave technique. Triplet states of keten have been investigated by trapped electron spectros c o ~ y and , ~ ~the~ triplet rr -+ n* transitions in thiophen, furan, and pyrrole by low-energy electron impact Evidence for the simultaneous contributions of the radical pair and the triplet mechanisms in the chemically induced dynamic electron polarization (CIDEP) of the photoreduction reactions of substituted benzoquinone, naphthoquinone, and anthraquinone by 2,6-di-tbutylphenol has been 4 Physical Aspects of Some Photochemical Studies

Photo-oxidation.-The role of singlet oxygen (lo,) in the sensitized photooxidations of organic molecules is of very general interest at present, and a short review of recent developments in the chemistry of singlet molecular oxygen has been given by R ~ s e n t h a l .Khan,36s ~~~ in a further interesting paper, details the discovery, generation, detection, and quenching of singlet oxygen, together with a discussion of its possible involvement in the photocarcinogenic activity of polycyclic hydrocarbons. It has now become increasingly apparent that many important reactions of singlet oxygen depend upon the existence of two lowlying excited singlet states, lAe and l&+. These two states have long lifetimes and favourable energy dispositions (lh, -+ 3Cu-, T = 45 min, voo = 7882.39 cm-l; lCU+-+ 31;g-,T = 7.1s, voo = 13 120.9080 cm-l), and give singlet oxygen its ss7

ss8

sSD 361 36a

s03 Se4

se6

D. C. Doetschman, J. Phys. Chem., 1976, 80,2167. P. N. Prasad, A. I. Attia, and A. H. Francis, Chem. Phys. Letters, 1977, 46, 125. P. Avouris and S. Sheng, Chem. Phys. Letters, 1977, 46, 295. N. Nishi and M. Kinoshita, Bull. Chem. SOC.Japan, 1976, 49, 1221. J. Vogt, M. Jungen, and J. L. Beauchamp, Chem. Phys. Letters, 1976, 40, 500. E. H. Van Veen, Chem. Phys. Letters, 1976, 41, 535. B. B. Adeleke and J. K. S. Wan, J.C.S. Faraday I, 1976, 72, 1799. I. Rosenthal, Photochem. and Photobiol., 1976, 24, 641. A. U. Khan, J. Phys. Chem., 1976,80,2219.

134

Photochemistry

unique pattern of reactivity. A few papers have been selected to illustrate the types of investigation being made and the progress achieved. Previous reports on the reactions of singlet oxygen with oxazolesSBghave shown that oxygen uptake leads to triamides, most probably through iminoanhydrides. Evidence has been presented to show that oxidation of 2-phenylcyclohexeno-4,5-oxazole in methanol with singlet oxygen leads to an N-aroylimino-anhydride. Photosensitized oxygenation reactions of 2,3-diphenylind0le,~~~ models for polymers,36s and 2-phenylnorbornene 360 have also been effected. The photosensitized formation and reaction of singlet oxygen with diphenylisobenzofuran solubilized in sodium dodecylsulphate micelles has been Oxygenated aqueous solutions of methylene blue were irradiated with red light, and the results indicate the formation of singlet oxygen through energy transfer from the triplet of ethylene blue. The singlet oxygen then diffuses, and penetrates to the interior of the micelles, where it reacts with the furan. An analysis of the kinetics of dye photosensitization, taking account of solvent isotope effects, sensitizer-substrate triplet-triplet energy-transfer processes, and quenching of the excited singlet state of the sensitizer by the substrate, has also been Quantitative measurements of the quenching of the excited singlet states of methylene blue and Rose Bengal by anions of /%carotene and 2,5-diphenylisobenzofuranare also given in this paper. Irradiation of 3,4,5-triphenyl-4-oxazolin-2-one(1) 372 in benzene, under nitrogen, with or without iodine, gave benzanilide and phenanthroxazolinone (2). In the presence of oxygen, however, (1) was photochemically converted into benzanilide, benzoic acid, and benzamidobenzophenones. This photooxidation reaction may involve an initial attack of singlet oxygen on (l), forming a dioxetan, followed by ring cleavage, to yield NN-dibenzoylaniline and the final products. The photoreaction of pyrazine derivatives under both nitrogen and oxygen 373 and the direct and eosin-sensitized photo-oxidation of glaucine 374 have been investigated. Tertiary amines have been extensively used as specific quenchers of lo2, and have been shown to inhibit formation of products in several systems, as well as dimol emission at 634nm in the gas phase. A recent report 376 inhibiting shows that stimulation of lo, dimol emission occurs in the presence of certain cyclic tertiary diamines in aqueous solution. Photo1ysis.-The photolysis of saturated compounds is receiving increasing attention, and results have been given for the photolysis of saturated chain acetals s66

387

368 QBg

37O

371 s72

973

374

37b

H. H. Wasserman and G. R. Lenz, Heterocycles, 1976, 5, 40. M. Nakagawa, N. Ohyoshi, and T. Hino, Heterocycles, 1976, 4, 1275. P. Bortolus, S. Dellonte, G . Beggiato, and W. Corio, European Polymer J., 1977, 13, 185. C. W. Jefford, A. F. Boschung, and C. G. Rimbanet, Helo. Chirn. Acta, 1976, 58,2542. A. A. Gorman, G. Lovering, and M. A. J. Rodgers, Photochem. and Photobiol., 1976,24,339. R. S. Davidson and K. R. Trethewey, J.C.S. ferkin It, 1977, 169. 0. Tsuge, K. Oe, and Y. Ueyama, Chem. Letters, 1976, 425. K. Yamada, K. Katsuura, H. Kasimura, and H. Iida, Bull. Chem. Soc. Japan, 1976, 49, 2805. L. Castedo, R. Suau, and A. Mourino, Anales. de Quim., 1977, 73, 290. C. F. Deneke and N. I. Krinsky, J . Arner. Chem. Soc., 1976, 98, 3041.

Photophysical Processes in Condensed Phases

135

which absorb in the 200nm region owing to what is assumed to be an n -+ O* transition. In the photolysis of liquid formaldehyde dimethyl acetal, detailed measurements of quantum yields of the products have been made, and a reaction scheme has been Anderson and Hochstrasser have described the development and application of a picosecond laser system to the study of the dynamics and the cage recombination of the radicals produced from tetraphenylhydrazine (TPH) in fluid The results published include both spectral and kinetic solution at 300 K.377 studies, and TPH has been shown to dissociate into radicals with a wellcharacterized spectrum in a time less than - 2 ps, while in times ranging from 0 up to about 2 ns, cage recombination of the radicals does apparently occur. The following further papers are of photophysical interest : flash photolysis studies of the reactive intermediate in the photochemistry of cyclohepten-2benzene-fumar~nitrile,~~~ and eosin and its complex with l y s ~ z y m e An .~~~ e.s.r. study of 1-hydropyridine-2,5-dicarboxylateanion radical in basic aqueous solutions has also appeared.361 Photoisomerization.-The following photoisomerization processes are of photophysical interest. The cis-trans photoisomerization of stilbene was originally attributed to population of the triplet states, although later theories support the view of rotation of the central bond to form a common twisted singlet. Although the mechanism of this photoisomerization is still not completely established, still less is known of the corresponding mechanism for 1,3-diphenyIpropene~.~*~, 583 Inactive and 14C-labelled substrates have been used to identify the reaction mechanisms. With the inactive substrate, quantum yields for cis-trans photoisomerization of 1,3-diphenylpropenes are shown in Table 10. Experiments Table 10 Quantum yields for the cis-trans photoisomerization of 1,3-diphenylpropenes Initial isomer cis degassed trans degassed trans undegassed

@Dcis+trans

0.071 4 0.006 0.065 k 0.004 0.066 k 0.007

* *

@trans+ota

0.01 0.10 & 0.01 0.10 0.01 0.11

with the active substrate showed the photoisomerization to be substantially unaffected by oxygen. The results show that the photolysis has no appreciable effect on the position of the label, and provide evidence against participation of triplets in the isomerization process. Results on sensitized i s ~ m e r i z a t i o n ~ ~ ~ provide stronger evidence against triplet participation where isomerization through the triplet state is accompanied by a change in the position of the label, whereas this change is not observed by direct isomerization. It has therefore been concluded that direct photolysis of 1,3-diphenylpropenes occurs by rotation of the double bond in one of the olefin singlet states. 376

a77 *78 s70 s80

s81

s83

H. Schuchmann and C. von Sonntag, J.C.S. Perkin ZZ, 1976, 1408. R. W. Anderson and R. M. Hochstrasser, J. Phys. Chem., 1976, 80,2155. R. Bonneau, P. F. de Violet, and J. Joussot-Dubien, Nouueau J. de Chimie, 1976, 1, 31. P. Schuler and H. Heusinger, Photochem. and Phorobiol., 1976, 24, 307. G. J. Fisher, C. Lewis, and D. Madill, Photochem. and Photobiol., 1976, 24, 223. H. Zeldes and R. Livingston, J. Magn. Resonance, 1977, 25, 67. 3. M. Figuera and M. T. Serrano, J.C.S. Faraday Z, 1976, 7 2 , 1534. J. M. Figura and M. T. Serrano, J.C.S. Faraday Z, 1976, 72, 2265.

136

Photochemistry

Photochromism.-Several papers have been presented by Korenstein et 02. relating to the photochromism and thermochromism of bianthrone and of bianthrone analogues (3-7).384-391 A detailed experimental account of the reversible photochemistry of bithioxanthene, over a wide temperature range, reveals the photoformation of two unstable isomers; E, absorbing at shorter wavelengths, and P, absorbing at longer wavelengths. Both isomers revert thermally to the

JqJJJ \:!z

X

co

z

0 NMe NMe NMe CH2 CHOH

11

Y H H H Me H H H

Z

H H H H Br H H

parent compound. The isomer E is believed to be produced through the triplet manifold, while P is formed directly from the excited singlet state. Bianthrone and its derivatives undergo reversible and irreversible reactions. At high temperature, those derivatives not substituted at the 1- and 1’-positions undergo irreversible photoreactions which compete with formation of the isomer.38KThe effects of temperature, external spin-orbit perturbation, and structure on the photoisomerization of dixanthylidenes have been 388 The reversible photochemistry of 10,lO’-dimethylbiacridaneand 10,10’-dihydrobianthrylidene,3Qo including the internal and external heavy-atom effects, and the structures of the short-wavelength isomers of bianthrone have recently been reported. Heterocyclic spiropyrans are photochromic, as illustrated in Scheme 1, and are found to have two parts: a ‘left’ heterocyclic part and a ‘right’ benzopyran a model compound or 2H-chromene part. 6-Nitro-8-methoxy-2H-chr0mene,~~~ for the ‘right’ part of spiropyrans, has provided information relating to the absorption, the luminescence, and the relative energies of the lowest nr* and

a

B hv (u.v.)

kT

O

R b

R

‘

’

-

0 5 i

R

R

Scheme 1 584

386 386

387 388

38g

3s0 391 392

R. Korenstein, K. A. Muszkat, and E. Fischer, J. Photochem., 1976, 5, 345. R. Korenstein, K. A. Muszkat, and E. Fischer, Helo. Chim. Acfa, 197% 59, 1826. R. Korenstein, K. A. Muszkat, and E. Fischer, J. Photochern., 1976, 5, 447. R. Korenstein, K. A. Muszkat, M. A. Slifkin, and E. Fischer, J.C.S. Perkin ZI, 1976, 438. R. Korenstein and K. A. Muszkat, ‘Environmental Effects on Molecular Structure and Properties,’ D. Riedel, Dordrecht, 1976, 561. R. Korenstein, G . Seger, K. A. Muszkat, and E. Fischer, J.C.S. Perkin II, 1977, 557. R. Korenstein, K. A. Muszkat, and E. Fischer, J.C.S. Perkin II, 1977, 727. R. Korenstein, K. A. Muszkat, and G. Seger, J.C.S. Perkin 11, 1976, 1536. P. Appriou, J. Brelivet, C. Trebaul, and R. Guglielmetti, J. Photochem., 1976, 6, 47.

Photophysical Processes in Condensed Phases

137

m* singlet and triplet states linked with the photochromic behaviour of the spiropyrans. Thermolysis of dioxetans 393 produced by the reaction of singlet oxygen with a series of l-methylene-4,4-diphenylcyclohexa-2,5-dienesproduces a range of products and a ketone derived from fission of the exocyclic double bond. This fission involves rearrangement of the excited state. The efficiency of generation of the excited state is ca. 16.6 k 3.2%,and is independent of the structure of the methy1 ketone by-product. Electrochromism involves a change in the absorption curve upon application of an electric field, and Labhart and his co-workers have discussed the possibilities of achieving large light modulation amplitudes by this effect. The requirements for achieving high electrochromic effects are considered, together with possible applications to dye lasers and colour

Chemi1uminescence.-An interesting review of very recent work on bioluminescence and chemiluminescence has been presented by Hastings and Wilson.39s The role of the dioxetans and dioxetanones in bioluminescence and chemiluminescence is discussed. To date, over 40 dioxetans have been isolated. All decompose at relatively low temperatures ( c 150 "C) into two carbonyl fragments, one of which has a good probability of being in the excited state (see Scheme 2). The electronically excited species can give rise to chemiluminescence 0-0 R1+R4 R2 R3 Dioxet an

+

h, R1

R2

R3

R4

Scheme 2

either directly, if the carbonyl product is a good emitter, or indirectly, by transfer of excitation energy. Another interesting property is that the excited carbonyl products are generated predominantly in the triplet rather than the singlet state. The review also covers the role of singlet molecular oxygen and electron-transfer processes in chemiluminescence. 1,2-Dioxetan has been produced by irradiation of 10,l O'-dimethyl-9,9'-biacridylidene (DBA) in several oxygen-saturated solvents Although it is stable at -78 "C in the presence of zinc tetraphenylp~rphin.~~~ at - 78 "C, 1,2-dioxetan decomposes into two molecules of N-methylacridone (NMA) at higher temperatures (71 z 1 min at 0 "C), with the emission of light. The chemi-excitation quantum yield for formation of INMA* is 0.036 k 0.018 for the photo-oxygenation (Aex > 550nm) or (thermal) oxidation of DBA by triphenyl phosphite ozonide. The quantum yield for formation of 3NMA has also been given, as 0.04 k 0.01. Chemiluminescence associated with cleavage of cyclopropenes by singlet oxygen Interest in this was aroused by the possibility that the has been 303 s94

396 398 sg7

H. E. Zimrnerrnan, G. E. Keck, and J. L. Pfledener, J. Amer. Chern. Suc., 1976, 98, 5574. J. K. Fischer, D. M. von Bruning, and H. Labhart, Appl. Optics, 1976, 15, 2812. J. W. Hastings and T. Wilson, Photochem. and Photobiol., 1976, 23, 461. K. Lee and L. A. Singer, J. Org. Chern., 1976, 41, 2685. G . W. Griffin, I. R. Politzer, K. Ishikawa, N. J. Turro, and M.-F. Chow, TetrahedronLetters, 1977, 1287.

138

Photochemistry

collapse of dioxetans derived from strained alkenes might be accompanied by chemiluminescence; dioxetans were indeed shown to be produced as intermediates, as shown in Scheme 3. Ph X '02

>

A

X

0-0

(8) a: X = H b ; X = Ph

P h l P h

= H'

0

0

1

A X = %

Ph

Scheme 3

A report on the inhibition and enhancement of the dimol chemiluminescence has recently appeared.398 The lo2dimol emission occurs of singlet oxygen (l&) at 634 and 703 nm, the emission of light being due to simultaneous transitions of two lo2molecules, both in their lAUstate, as shown:

The emissions at 634 and 703 nm correspond to the ground (0,0) and the first (0, 1) vibrational states respectively. Certain cyclic diamines were found to increase the dimol lo2emission, whereas other amines were inhibitory. The increase in light emission was seen for both 1,4-diazabicycl0[2,2,2]octane and NN'-dimethylpiperazine, but not, however, with the acyclic analogues. Results on chemiluminescent Schiff bases reveal that these compounds are generally chemiluminescent on oxidation, with reasonable efficiency.39g The Schiff bases derived from isobutyraldehyde have been found to be more chemiluminescent than those obtained from 2-phenylpropionaldehyde. It has been suggested that the presumed dioxetan intermediate decomposes to form exclusively triplet nn* excited states, and that the singlet emission observed is the result of energy transfer from the kinetic product. A further paper 400 discusses the stability of the amino-dioxetan, and indicates that this might be interpreted differently in view of the behaviour of a peroxide derived from tetrahydrocarbazole.

Triboluminescence (TBL).-Triboluniinescence is a type of luminescence arising from scraping, grinding, or breaking a crystal. Early research work concentrated on the discovery of new triboluminescent materials, but of late, attempts have been made to correlate TBL with the properties of the crystal; and it appears 398 39s 400

C. F. Deneke and N. I. Krinsky, Photochem. and Photobiol., 1977,25, 299. F. McCapra and A. Burford, J.C.S. Chem. Comm., 1976, 607. F. McCapra, C. Chang, and A. Burford, J.C.S. Chem. Comm., 1976,608.

Photophysical Processes in Condensed Phases

139

that TBL is intimately linked to the elastic properties of the crystal. Chandra 401 has presented results on the rise and decay kinetics of TBL in sucrose crystals. TBL emission does not appear instantaneously when the force is applied to the crystal but appears some time later, the delay being dependent on the magnitude of the applied force. The decay time of TBL from sucrose crystals was 4.2 x 10-3s, irrespective of the mode of crushing. It is also interesting to note that the area below the TBL decay-time curve increases with the length of the delay between impact and the observation of TBL. This is related to an increase of stored potential energy available for fracture, and hence TBL. ‘01

B. P. Chandra, Indian J. Pure Appl. Phys., 1976, 14,874.

6

3 Gas-phase Photoprocesses BY D. PHILLIPS

1 Alkanes and Alkenes In reaction (l), in which the C+ ion has kinetic energies of between 0.5 and 9.0 eV, the resulting CH+ is produced by a direct mechanism in the a"311 state.l The double photoionization cross-section of C(3P) has been measured.2 The fluorescence decay time of the C2(d"311g) state has been obtained by laser fluorescence excitation and laser pyrolysis studies. In the former study, the decay time was found to be 120 k 10 ns independent of vibrational level v' = 0-5. The fluorescence transition monitored is between the c?(~~I, -+ a 3rIn,) levels, and use of C , resonance fluorescenceas a monitoring technique in transient flame studies has been investigated.6 C t + H,

-

CH++ H

(1)

The Hg 6(3P,)-sensitized decomposition of ethane at high light intensities is said to occur via the mechanism shown in reactions (2) to (5)6 with radical reactions such as (6) and (7) and many others occurring at high intensities. It

+ C&Je H* + C2H6

Hg(3P,)

2C2H6* 2CzH5.

____+

____+

H*

+ CaHS* + Hg(6'So)

(2)

H2

+ C2H6.

(3)

n-C,Hlo CzH6

(4)

+ C,H*

(5)

should be noted that in many mercury atom sensitizations, reaction (2) is not a single-step process as written, HgH being formed initially. The rate constant

-

H*+CzH,*

CH,*+C2H&*

-

2CH3*

(6)

__I__,

C3Hs

(7)

for reaction (8) was measured as 1.5 x 1014cm3m~lecule-~ s-l at 300 Torr ethane and 35 "C. An initial step exactly analogous to (2) is proposed in the

C. A. Jones, K. L. Wendell, J. J. Kaufman, and W. S. Koski, J. Chem. Phys., 1976, 65, 2345. S. L. Carter and H. P. Kelly, J. Phys. (B), 1976, 9, 1887. T. Tatarczyk, E. H. Fink, and K. H. Becker, Chem. Phys. Letters, 1976, 40, 126. S. Leach and M. Velghe, J. Quant. Spectroscopy Radiative Transfer, 1976, 16, 861. D. G. Jones and J. C. Mackie, Combustion and Flame, 1976, 27, 143. J.-T. Cheng, Y . 4 . Lee, and C.-T. Yeh, J. Phys. Chem., 1977, 81, 687.

140

Gas-phase Photoprocesses

141

cadmium(3Pl)-sensitized photolysis of ethane at 533, 553, and 575 K, pressures ranging from 3 to 40T0rr.~ Reaction (4) produces 'hot' n-butane from which subsequent reactions can occur. Reactions occurring in the photo-oxidation of methane * and higher alkanes * will be discussed in Sections 5 and 11 of this chapter. The i.r. laser photodecompositions of ethane,lo cyclopropane,ll cycloCdFs," and ethylene 13-16 have been reported. The threshold energies for the competing reactions of triplet propene (9) and (10) are as shown,le but whereas process (10) is of most importance upon electron impact at 4.4eV, at which only the lowest triplet state at 4.4 eV is populated, process (9) is predominant in the mercury-photosensitized reaction at 4.9eV. The authors conclude that Hg(3P,) must populate the T2 state of propene, although since this lies at 6.1 eV, a non-vertical transition must be invoIved. C6H6 CsH6

____+

C3H,*

+ H-

+ CH3*

C2H3*

A H o = 3.8eV AH' = 4.0eV

(9) (10)

In the 147-nm photolysis of cis-but-2-ene and pent-l-ene, the new fragmentation processes (11) and (12) have been observed, with quantum yields as ~ h 0 w n . I ~

The order of efficiency for scavenging of methyl radicals was as follows: HI > NO2 > NO > O2 > H,S > SOa. But NO2, SO2, and H2S were also found to initiate cistrans isomerization in but-2-ene, and HI and SO2 and H2S suffered from the experimental disadvantage of giving iodine and sulphur deposits respectively on the reaction vessel windows. The vacuum-u.v. photolysis of other organic compounds has been reported.17" An analogue of the type I1 process in ketones has been found in an intramolecular H-abstraction via a six-membered transition state in Hg(3Pl)-sensitized reactions of hex-l-ene and cis-oct-Zene vapour.l8 Subsequent reactions of the diradicals so formed give rise to stable products (Scheme 1).

*

E. J. McAlduff and Y. H. Yuan, J. Photochem., 1976, 5, 297. R. A. Cox, R. G. Dement, P. M. Holt, and J. A. Kerr, J.C.S. Faraday I , 1976, 72, 2044. K. R. Darnall, W. P. L. Carter, A. M. Winer, A. C. Lloyd, and J. N. Pitts, jun., J. Phys. Chem., 1976, 80, 1948; W. P. L. Carter, K. R. Darnall, A. C. Lloyd, A. M. Winer, and J. N. Pitts, jun., Chem. Phys. Letters, 1976, 42, 22.

lo

J. Tardieu de Maleissye, F. Lernpereur, C. Marsal, and R. I. Ben-Aim, Chem. Phys. Letters, 1976, 42, 46.

M. L. Lesiecki and W. A. Guillory, J. Chem. Phys., 1977, 66,4317. la J. M. Preses, R. E. Watson, jun., and G. W. Glynn, Chem. Phys. Letters, 1977, 46, 69. 0. N. Avatkov, V. N. Bagratashvili, I. N. Knyazev, Yu. R. Kolomiisky, V. S. Letokhov, V. V. Lobko, and E. A. Ryabov, Kvantovaia Elektronika, 1977,4, 741. l4 S . L. Chin, Canad. J. Chem., 1976, 54, 2341. l6 J. Tardieu de Maleissye, F. Lempereur, and C. Marsal, Chem. Phys. Letters, 1976, 42, 472. R. Derai and J. Danon, Chem. Phys. Letters, 1977, 45, 134. l7 G. J. Collin and K. Bukka, J . Photochem., 1977, 6, 381. 17d G. J. Collin, J. Chim. phys., 1977, 74, 302. Y. Inoue, S. Takamuku, and H. Sakurai, Canad. J. Chem., 1976,54,3117.

l1

~II

Photochemistry

142

a 0

1,4 shift

0

0

1,5 shift

Scheme 1

Two-photon excitation spectroscopy tends to support the proposal that in diphenylhexatriene and diphenyloctatetraene, a one-photon forbidden lA, state is almost isoenergetic with the lBUallowed state.lg The photodissociation of C4_H8+ions has been studied.20 Emission from radical cations of diacetylene (A211u-+ X2nn,), triacetylene -+ T211u), and tetra-acetylene (J211u -+ 8211,) has been observed, and lifetimes of the 2 states have been measured.21

(x2n,

2 Aromatic Molecules Several studies on the two-photon fluorescence excitation spectrum of benzene vapour are now in agreement that the promoting mode for the transition is the ~ 1 vibration 4 of bzu~ y m m e t r y , ~ and ~ -this ~ ~ has been confirmed by observation of the fluorescence spectrum in which the main progression seen is the 10n1401.26 Relative band intensities have been measured in thirteen strong progressions found in seven single vibronic level spectra from benzene,26and it was shown that the Herzberg-Teller theory could not, in either first order or second order, account for the observed intensities. However, the introduction of Fermi resonances shown in Figure 1 involving levels separated by 121 cm-1 in the upper state and 223 cm-I in the lower brought experiment into coincidence. These Ievels involve the v6 promoting mode and vl totally symmetric ring lo

eo 2a 23

24 26 2e

G. R. Holtom and W. M. McClain, Chem. Phys. Letters, 1976, 44, 436. M. Riggin, R. Orth, and R. C. Dunbar, J. Chem. Phys., 1976, 65, 3365. M. Allan, E. Kloster-Jensen, and J. P. Maier, Chem. Phys., 1976, 17, 11. W. Hampf, H. J. Neusser, and E. W. Schlag, Chem. Phys. Letters, 1977, 46, 406. L. Wunsch, F. Metz, H. J. Neusser, and E. W. Schlag, J. Chem. Phys., 1977, 66, 386. J. R. Lombardi, R. Wallenstein, T. W. Hansch, and D. M. Friedrich, J. Chern. Phys., 1976, 65, 2357. A. E. W. Knight and C. S . Parmenter, Chem. Phys. Letters, 1976,43, 399. C. S. Parmenter, K. Y. Tang, and W. R. Ware, Chem. Phys., 1976, 17, 359.

143

Gas-phase Photoprocesses

breathing mode, and despite the rather large energy differences, and modest coupling matrix elements of about 10 cm-l, clearly cause marked deviations from harmonic radiative transition probabilities. This study has been amplified by decay time meas~rements.~~ Some of the relevant data are shown in Table 1,

6'I '

-.

L

6'i'

60It

c A€=ZZkq+

62 6,,,lo

i

t

6'

O0

00

'4g

'82

u

Figure 1 A schematic of some of the Lvg-vl' level mixings in both the lBZUand ground states of benzene

from which the value of the radiative rate constant k, has been obtained in the normal way and compared with values calculated on the Herzberg-Teller model. In Table l,f6 is the fraction of total intensity due to the Av, = k 1 progressions. The basis for the computed values O f kr6 shown in the final column was as follows.27 The general form of the first-order HT radiative rate constant kr from an upper state ub to the lower f a is:

where Qi is a promoting mode, 8 is the vibrational eigenfunction (harmonic) of the state ub or fa as required, and Mi contains electronic terms which are Qi dependent. The degenerate mode v 6 is treated as two independent promoting modes. Thus for radiative transitions induced only by the modes k, becomes k 6

Cx

V%-tzo M62

I b b-k

(22)

Alternatively, empirically

=

1

IF 9dv/s IF dv

(23)

If more than one progression is involved, each must be summed, e.g., for the ln61 origin, both the l,n601 and l,f1621 progressions must be integrated. If calculations are done using Franck-Condon factors, individual progressions are surnmed and the result finally weighted by the appropriate (6b 1 Q6 1 6a). Since in (20) and (21) the frequency ratios are essentially unity, kmas a function of vibronic level can be computed, with results given in Table 1. Several pertinent comments about these results are made by the authors. Within experimental error, kr6(6')/kr6(@) agrees well with theory. Mode sixteen does not influence the ratio, but one or two quanta of mode one in the pumped state produce a large departure of the experimental result from that predicted, as does excitation of two quanta of mode six. No refinement of Herzberg-Teller theory attempted by these authors could explain these deviations entirely satisfactorily, and perhaps further modelling is called for. An exactly solvable model for vibronic coupling with a non-totally symmetric harmonic mode has been proposed.28 The kr data discussed above require accurate measurement of fluorescence quantum yields, and methods of measurement have been reviewed recently.2B Anomalies in radiative lifetime measurements in other systems have also been discussed.so The non-radiative fates of benzene lBzuin the vapour phase are still the subject of much debate. It has been shown that for excitation at 266.8 nm to produce the zero-point level of the upper state, the fluorescence quantum yield is independent of pressure in the range 1-20 Torr, but above this decreases to around 0.18, 29

90

A. R. Gregory, W. H. Henneker, W. Siebrand, and M. Z. Zgierski, J. Chem. Phys., 1976, 65, 2071. J. B. Birks, J. Res. Nat. Bur. Stand., Sect. A, 1976,80,389; A. Bril and A. W. de Jager-Veenis, ibid., p. 401 ; J. N. Demas and B. H. Blumenthal, ibid., p. 409. J. B. Birks, 2.phys. Chem. (Frankfurt), 1976, 101,91.

Photochemistry the same as for excitation at 253 nm.31 This high-pressure quantum yield is in substantia1 agreement with that obtained from the spectrophone technique of 0.16 k 0.02.32 For 266.8 nm excitation the combined quantum yields of fluorescence and intersystem crossing to the triplet manifold measured by the isomerization of but-2-ene and phosphorescence of biacetyl methods are 0.97,31 whereas at shorter wavelength a quantum deficit is noted, which is attributed by these authors to the formation of the benzvalene isomer, although this is not directly observable. (Note that the direct concerted transformation of lBau benzene into benzvalene is not symmetry-allowed, whereas meta bonding can occur with retention of symmetry). By contrast, measurements using the but-2-ene method, which show triplet-state formation for 235.7 nm excitation becoming zero in the limit of low pressure, have persuaded other authors to propose a model involving reversible intersystem crossing.33 The authors first consider the alternative possibility that the vibrationally excited triplet state first formed upon intersystem crossing (24) has a lifetime sufficiently short that in the limiting low pressures used, triplet quenchers such as but-2-ene (25) cannot effectively compete with the unimolecular decay (26). Under such conditions the apparent triplet yield would approach zero, as is observed. This possibility is discounted, but the analysis may be based upon faulty premises. Firstly, it is stated that the relaxed triplet state has a decay time which is independent of pressure of other species at constant pressure of benzene. This certainly is not the case for the added gas pentane, however, since a quite strong dependence of r upon concentration of this species was noted in the experiments using the spectrophone technique.34 Thus the lifetime assumed for the relaxed triplet under the conditions of the experiments may be too long in the but-Zene sensitization work. Moreover, the authors use an empirical tunnel theory approach to calculate the increase in T, -+ So intersystem crossing rate with excess energy, in order to evaluate the lifetime of the hot Tl state initially formed upon S1-+ Tlintersystem crossing. This theory predicts a maximum increase of a factor of approximately 150, giving the rate constant for (25) of 3 x lo5s-l. The authors state that in order for inefficient quenching through (26) and/or (27) and (28) to give rise to apparent 146

'A, 3A,

3A,

+Q

+Q 3Ao + Q

3A,

3Ao

---+

3A,

(24)

____f

A,

(25)

-

Energy transfer

(26)

3A0

(27)

Energy transfer

(28)

A,

(29)

reductions in measured triplet yield through the isomerization of the but-2-ene, the rate constant for (25) would have to be in excess of 2 x lo7s-l. In fact, measurements using the spectrophone technique are consistent with this rate constant being even larger, namely 2 x lo8 s-1.32 If this is correct, and there dl y2

sa 8Q

S. A. Lee, J. M. White, and W. A. Noyes, jun., J. Chem. Phys., 1976, 65,2805. L. M. Hall, T. F. Hunter, and M. G. Stock, Chem. Phys. Letters, 1976, 44, 145. S. J. Formosinho and A, M. da Silva, J.C.S. Faraday II, 1976, 72, 2044. L. M. Hall, T. F. Hunter, and K. S. Kristjansson, Chem. Phys. Letters, 1976, 43, 404.

Gas-phase Photoprocesses 147 is no reason to doubt the value, then at pressures in the 0.1-20 Torr range the effects of vibrational relaxation in TI will certainly be observable, and could account for the observed dependence on pressure without invoking reversible S1.n-f intersystem crossing. The system would then be similar to that in naphthalene (uide infra) in which again conclusive evidence has been given which does not support a reversible intersystem-crossing mechanism. The spectrophone experiments have produced some results of further interest, in that the decay of hot triplets to ground state, step (25), appears not to be a

1 B2

"

E+ *3

kt

-3

%

I

t i

Figure 2 LeveIs used to account for decay of tripIet (3B1u)benzene. Best fit to experimental data was obtained with ks = 2 x lo8 s-l, k-s = 5 x lo7s-l, and kr = 3 x 107 s-1

(Adapted from Chem. Phys. Letters, 1976, 44, 145)

direct process, but involves an intermediate lower-lying triplet state, of unidentified character (but may be non-planar), as shown in Figure 2. The observed kinetics are in accord with this scheme, although there is no other evidence for the existence of the low-lying state. Rate constants required to fit the experimental data are shown in Figure 2. Two recent studies have probed the mechanism of electronic energy transfer in benzenes. In the first,36transfer from the zero-point level of lBZU perdeuteriobenzene to the lBZUelectronic state of benzene itself was investigated. Possible routes are shown in Figure 3, and it was found that despite the near-resonance of the 16' state of lBzubenzene with the initial state 0 ' of CsDs, the C6Hs zero-point level was predominantly populated, with cross-sections for processes I and I1 (Figure 3) of 0.6 and 0.1-4.2 of hard-sphere respectively. Among possible routes for populating the zero-point level of benzene by collision of the initial perdeuteriobenzene with thermal levels of benzene, the hv = 0 route (30) is C6D6(1B2u, 3s

0')

+ C6H6(1A1c7~00)

-

C I @ ~ ( ' A , ~00) , -!- CsH6('&,, 0 ')

C. S. Pmenter, B. Setzer, and K. Y.Tang, J. G e m . Phys.,

1977,66, 1317.

(30)

148

Photochemistry

dominant, and this has gas-kinetic efficiency. In a similar rate constants for the quenching of selected single vibronic levels of lBzu benzene by aniline were measured as given in Table 2. In both studies the conclusion was reached that the Forster long-range induceddipole interaction could not account for the observed efficiency of electronic energy transfer. A review of inter- and intra-molecular electronic and vibrational relaxation including discussion on benzene and larger molecules has appeared,37 and a theoretical paper has proposed that electronic relaxation can be used as a probe for the suggested intramolecular vibrational relaxation process.38 With regard to intramolecular vibrational relaxation, it must be said that the success 38800 -

38600I

E

0

Y

)r

CT

4' -

..

162

n c

(dg)

16'

38400-

L

a,

J

c

O0

w

38m-

-

38000

38066 cm-'

c6

H6

c6 D6

Figure 3 Vibronic levels in the lBzU(Sl) states of C6H6 and c&. The zero-point levels Oo are separated by 203 cm-l, and the v16 fundamental of S, C6H6lies 34 cm-l above the Oo C6D6 level being pumped (Reproduced by permission from J. Chern. Phys., 1977, 66, 1317)

Table 2 36

Quenching of 1B2u benzene by aniline State 6l 6l1' 6'12

Rate constant a/l mol-l s-l 4.3 x 1011 4.6 x loll 4.4 x 1011

Measurements made from decay times. Values from yield measurements are a factor of 1.4 higher. @

36 s7 38

C. Lardeux and A. Tramer, Chem. Phys., 1976,18, 363. A. Tramer and C. Tric, Ber. Bunsengesellshaft Phys. Chem., 1977, 81, 209. K. F. Freed, Chem. Phys. Letters, 1976, 42, 600.

Gas-phase Photoprocesses

149

of the local-mode approach in explaining the vibrational overtone spectra in benzene and related molecules suggests that redistribution of vibrational energy is a slow process.39 The absorption spectrum of benzene and perdeuteriobenzene in the 340 nm region corresponding to the So -+ T, transition has been Electronimpact studies on a variety of molecules including benzene, toluene, m-xylene, acetylene, CN, SO2,N2,and CO have revealed relatively long-lived triplet states,*l but in olefins, for example, no metastable triplets were detected. The authors draw a correlation between Iarge geometry changes between triplet and ground states and absence of metastable triplets. A comparative study of different types of electrical discharges for use in exciting benzene vapour has been made,42and intramolecular 1,2-methyl shifts in the y-radiolysis of xylenes and trimethylbenzenes in the vapour phase have been A van der Graff generator has been used to study the higher excited states of toluene with sub-nanosecond time r e ~ o l u t i o n . ~ ~ A further note has appeared in which ns-time-resolved emission spectroscopy is used to study the dual emission of styrene and related compounds excited to A weak higher-energy long-lived component of the total the S2(lL,) fluorescence was attributed to emission from a state in which some twisting about the olefinic double bond had occurred. This was confirmed by studies on 2-phenylnorbornene (1) in which such twisting motion is entirely suppressed, and in which the second component of fluorescence was absent.

A high-resolution two-photon excitation spectrum of naphthalene has been and radiative lifetimes obtained for two-photon excitated levels in the first excited By comparing results with one-photon excitation, which produces vibronic levels of different parity and ~ y m m e t r y it , ~was ~ shown that the measured fluorescence decay time decreases exponentially with increase in excess energy irrespective of symmetry or parity of the initially prepared state. Thus there is no strong dependence upon promoting mode frequency for the non-radiative decay process, and it is concluded that non-radiative decay probability is dominated by Franck-Condon factors rather than electronic 88 40

41 4a 43

44 46 46 47

46

R. L. Swoffard, M. E. Long, and A. C. Albrecht, J. Chem. Phys., 1976, 65, 179. J. Cariou, J. Lotrian, and A. Johannin-Gilles, J. Quant. Spectroscopy Radiative Transfer 1976, 16, 843. J. C. Hemminger, B. G. Wicke, and W. Klemperer, J . Chem. Phys., 1976, 65, 2798. M. Bigwood, A. Delaby, and S. Boue, Nouveau J . de Chimie, 1977, 1, 25. G. Perez, J . Phys. Chem., 1976, 80, 2983. G. Beck, J. T. Richards, and J. K. Thomas, Chem. Phys. Letters, 1976, 40, 300. K. P. Ghiggino, D. Phillips, K. Salisbury, and M. D. Swords, J . Photochem., 1977, 7, 141, V. Boesl, H. J. Neusser, and E. W. Schlag, Chem. Phys., 1976, 15, 167. V. Boesl, H. J. Neusser, and E. W. Schlag, Chem. Phys. Lerrers, 1976, 42, 16. W. E. Howard and E. W. Schlag, Chem. Phys., 1976, 17,123.

150

000-z

000.9

I

'

mo*z

oo'r4

0004

0.1 X AlISN3Q 7V311d0 005.1

009.1

009.

00s.

ooz.

008-

00s.

007.

3AIlVIOVMNON

oo+z

-01 X 31VcI

OOt*&

'9

009-

0-L X AlISN3Q 1V311d0 004 1 000.1

000.1

01 X 31Vcl 3 A I l V l O V t l

002-1

'9-

w >

EOL

Photochemistry

000 - 0

000.0

ooo*o

000.0

.G $ R v

Gas-phase Photoprocesses

151

matrix elements. An extremely high-resolution study on the excitation of naphthaIene seIected levels which are free of sequence congestion has permitted evaluation of the effects of rotational excitation upon radiative and non-radiative rate constants for the decay of lBaun a ~ h t h a l e n e .Results ~~ for the 8(b,,) vibrational band are shown in Figure 4, from which it can be seen that k, peaks at high J values, with k, decreasing in the same region. Ro-vibronic coupling would not be expected to produce such pronounced effects. However, the authors conclude that the use of the normal relationship (31) to give values of kr can be kr = @PhF

(31)

totally erroneous if ensemble averaging occurs, as in most studies. A theory was developed for non-radiative decay via spin-orbit coupling which predicts two variations in knr due to rotational excitation, one a dynamic rotational overlap effect, and secondly an effect of energy-gap branching which causes a change in vibrational overlap factors. The study strongly suggests that even higher resolution experiments will have to be carried out in the future, perhaps using the elegant supersonic jet method of rotational and vibrational cooling to permit selective e x c i t a t i ~ n . ~ ~ Further impressive results on triplet-state formation in naphthalene following single vibronic level excitation in the triplet state have been obtained using a tandem dual dye-laser arrangement.6o The results provide an explanation of the requirement for a relatively high pressure of buffer gas in conventional flash photolysis experiments in order to see triplet-triplet absorption in naphthalene,61 since the absorption profile of the initially formed, vibrationally excited triplet state is very narrow, but grows by collisional relaxation into the more familiar profile of absorption by the relaxed Tl state. Since the formation of ‘hot’ triplet naphthalene is still about 84% of the loss of singlet naphthalene under isolated molecule conditions, there is no need to invoke reversible mechanisms for the intersystem process as was done previously.61 These very important new results thus place naphthalene firmly in the statistical limit as far as the Sl.y\r)Tl intersystem-crossing process is concerned. Spectral effects of the breakdown of the Born-Oppenheimer approximation in molecules such as naphthalene 62 and experimental evidence for the preparation of molecular eigenstates and BornOppenheimer states by coherent light have been described.63 Biacetyl-sensitized annihilation-delayed fluorescence of anthracene vapour has been r e p ~ r t e d . ~The ~ fluorescence decay times of various deuteriated phenanthrenes, using 25 cm-l bandpass excitation, have been measured (Table 3) in the vapour phase.66 The results are of interest in that an inverse deuterium isotope effect is observed. This is best explained by a resonance mechanism66 for the S1-+ Tl intersystem-crossing process involving a higher triplet state, 11

R. E. Smalley, L. Wharton, and D. H. Levy, Accounts Chem. Res., 1977, 10, 139. H. Schroder, H. J. Neusser, and E. W. Schlag, Chem. Phys. Letfers, 1977, 48, 12. S. J. Formosinho, G. Porter, and M. A. West, Chem. Phys. Letters, 1970,6,7; Proc. Roy. SOC.,

6a

1973, A333,289. S. J. Strickler, J. Phys. Chem., 1976, 80, 2149.

49 6o

63 64 66

66

A. H. Zewail, T. E. Orlowski, and K. E. Jones, Proc. Nat. h a d . Sci. U.S.A., 1977, 74, 1310. V. T. Pavlova, Doklady Akad. Nauk Belorusskoi S.S.R., 1977, 21, 17. J. J. O’Brien, B. R. Henry, and B. K. Selinger, Chem. Phys. Letters, 1977, 46, 271. V. Lawetz, G. Orlandi, and W. Siebrand, J. Chem. Phys., 1972,56,4058.

152

Photochemistry

Table 3 Fluorescence decay times of deuteriated phenanthrenes Compound Phenanthrene [2,4-2H,]Phenant hrene [1 ,3-2H,]Phenanthrene [9,1O-2H,]Phenanthrene

55

Decay timelns 76.2 k 3 63.8 & 1 63.1 k 1 58.2 rt: 1.5

which must in this case have Al symmetry in point group CZu.There is such a level predicted at 27 120 cm-l, just below the S1state. Investigations of electronic energy transfer from pyrene to perylene have been stimulated by potential use as pumping mechanisms in vapour-phase dye lasers. The efficiency of energy transfer $ET as measured by equation (32), where CDp is

the quantum yield of fluorescence of perylene vapour, of value 0.07, 7 is the energy-conversion efficiency of around 20%, T~ is the pyrene radiative lifetime of -700ns, rt is the measured total decay time of the mixture, and k, is the net quenching rate constant, was found to be 3&40% under the conditions of the e~periment.~' This is inadequate for laser use. Stepwise two-photon excitation of fluorescence in tetracene vapour (33) and (34) is seen to produce fluorescence from upper singlet states (35).58 So

+ hv

S1i- hv

_I___,

S 1

(33)

S,

(34)

+ hv

(wheren > 1) (35) The prediction that, in the intermediate strong-coupling limit, the relative fluorescence yield from the S2 state compared with S1 is given for isolated molecules by the ratio of density of states (36) has been confirmed for S,

So

3,4-benzp~rene.~~ The S1* emission can be interpreted as due to that from an ensemble in which excess energy is distributed randomly over available vibrational For levels higher than the u' = 0 level of S1azulene, the decay time has been measured to be sub-picosecond,60whereas in solution the decay of the Boltzmann distribution of S1vibrational levels is 1.9 P S , the ~ ~ decay being electronic relaxation through internal conversion. The picosecond relaxat ion of acridine has also been investigated .6 67 68

5Sa

6B

6o

62

C. T. Ryan and T. K. Gustaf'son, Chem. Phys. Letters, 1976, 44,241. B. Nickel and G . Roden, Ber. Bunsengesellschufr Phys. Chem., 1977, 81, 281. P. A. M. van den Bogaardt, R. D. H. Rettschnick, and J. D. W. van Voorst, Chem. Phys. Letters, 1976, 41, 270. P. A. M. van den Bogaardt, R. P. H. Rettschnick, and J. D. W. van Voorst, Chem. Phys. Letters, 1976, 43, 194. P. Wirth, S. Schneider, and F. Dorr, Chem. Phys. Letters, 1976, 42, 482. E. P. Ippen, C. V. Shank, and R. L. Woerner, Chem. Phys. Letters, 1977,46,20. V. Sundstrom, P. M. Rentzepis, and E. C. Lim, J. Chem. Phys., 1977,66,4287.

Gas-phase Photoprocesses

153

The dye-laser-induced photodissociation of C7H8+ions in gaseous the angular distribution of ionic fragments from the photolysis of molecular ions,6a the sequential two-photon dissociation of the cyanobenzene cation,6s and fluorescence quantum yields and decay times for a variety of molecular ions including C6FB+,CsFsH+, and 1,2,4-C3F,H+g6 have been reported. In the last study, the fluorescence quantum yields of the hexa- and penta-fluorobenzene cations in the x a n d states respectively were unity, with decay times of 59 k 10 and 50 k 10 ns respectively. The high quantum yields in contrast with those of the neutral species undoubtedly reflect the absence of non-radiative decay routes in the ion through nearby states of differing multiplicity which are available to the neutrals. 3 Carbonyls and other Oxygen-containing Compounds

The photochemistry of formaldehyde and other prototype carbonyl compounds has continued to attract attention. The sequential two-step photoionization of formaldehyde by a nitrogen laser through steps (37) and (38) at 0.9 Torr has been

in~estigated.~~ By following the ion current as a function of time, the decay properties of the lAz state could be monitored. These appeared to be nonexponential and in the 20-30 ns region. These results are not in agreement with those from a recent single vibronic level study.68 Results from this for the 4l excited level are given in Table 4, together with those from an earlier study.@ Table 4 Decay characteristics of the 4l level of formaldehyde and deuteriated analogues 60 Compound H,CO H2C0 HDCO HDCO D,CO DzCO a

Excess energylcm-l 124.6 124.6 96.5 96.5 68.5 68.5

@F

0.035

-

0.048

-

1.00 -

q/ns 82 282" 141 290" 450 463 a

rr/s 2.3

-

2.9

-

4.5 -

Tnr/ns 85

45 x 103 148

Results from ref. 69.

It can be seen that except for D&O, a serious difference in lifetime measurements exists. The results show a very dramatic deuterium isotope effect, which is reflected in both the radiative lifetime Tr, but more dramatically in the non. isotope effects on Tr can be interpreted in terms radiative decay rate, T ~ Deuterium

I. R. Eyler, J. Amer. Chem. SOC.,1976, 98, 6831. R. Orth, R. C. Dunbar, and M. Riggin, Chern. Phys., 1977, 19, 279.

E6

aE E7

T. E. Orlowskii, B. S. Freiser, and J. L. Beachamp, Chem. Phys., 1976, 16, 439. J. H. D. Eland, M. Devoret, and S. Leach, Chem. Phys. Letters, 1976, 43, 97.

S. V. Andreyev, V. S. Antonov, I. N. Knyazev, and V. S. Letokhov, Chem. Phys. Letters, 1977, 45, 166. R. G. Miller and E. K. C. Lee, Chern. Phys. Letters, 1976, 41, 52. E. S. Yeung and C. B. Moore, J. Chem. Phys., 1973,58, 3988.

154

Photochemistry

of deviations due to the Herzberg-Teller approach, in which a linear dependence of kr upon the frequency of the promoting mode is expected (HT), or by deviations from the Born-Oppenheimer approximation (non-BO) where a cubic dependence of k, upon promoting-mode frequency is to be expected. Relative values of k, for the deuteriated formaldehydes are given in Table 5 compared Table 5 Relative radiative rate constants for deuteriated formaldehydes, 4' state 68 Experiment a1

HT

non-BO

2.0 i-0.5

1.82

6.03

kr(D,CO )

with values computed from the various models. It is evident that in this case, in contrast with the case of fluorinated acetones studied recently,70 the H T dependence seems to be appropriate, as was predicted theoretically for formaldehyde re~ently.~'For D,CO, the lA2 state was self-quenched, with a rate constant of 1.9 k 3 x 1O1O 1mol-1 s-l. A further study has shown that for the nearly degenerate 51 and 1141 vibronic states of A1A2 formaldehyde, the radiative rate constant for the former is 1.8 times greater than that for the 1141 Evidently the asymmetric C-H stretching vibration v5 enhances the radiative transition probability, a result unexpected on the basis of the vibronic activity in the absorption spectrum. The large deuterium isotope effect upon the nonradiative decay of formaldehyde has been interpreted as due to influence upon internal conversion to the ground That some non-radiative decay must occur prior to decomposition is evident, since in absorption the C=O stretching vibration is principally excited, yet decomposition must involve C-H bond rupture (39) or CO elimination (40) involving different normal modes.73 The quantum yields of reaction (39) have been remeasured for a variety of single vibronic levels of formaldehyde in the isolated-molecule limit,74 and results ranged from ~ 0 . 1for the 2l4l level to a maximum of 0.37 for the 2341 level. H2C0

hv

H*

+ *HCO

H 2 +CO

(39) (40)

These results are much lower than those measured earlier,7s which means that (40) may have substantial yields. This result is encouraging to those interested in using this reaction for laser isotope separation, since (39) is a complicating reaction owing to the formation of reactive hydrogen atoms. 70 71 72

73 74

76

J. Metcalfe and D. Phillips, J.C.S. Faraday 11, 1976, 72, 1574. S. H. Lin, Proc. Roy. SOC.,1976, A352, 57. K. Y. Tang and E. K. C. Lee, Chem. Phys. Letters, 1976, 43, 232. G. D. Gillespie and E. C. Lim, J. Phys. Chern., 1976, 80,2166. R. S. Lewis, K. Y. Tang, and E. K. C. Lee, J. Chem. Phys., 1976, 65,2910. R. D. McQuigg and J. G. Calvert, J. Amer. Chern. Soc., 1969, 91, 1590.

155

Gas-phase Pho toprocesses

A photochemical study of the rotational state dependence of #A2 formaldehyde decay using laser excitation has appeared In another study, it was found that the CO product from (40) appears at a rate 100 times lower than the which suggests a long-lived intermediate between decay of lA2 f~rmaldehyde,~~ the formation of products and loss of the state. Whether this could conceivably be the ground state or possibly an isomeric form of formaldehyde remains to be established. Although the fraction of product energy available carried as vibrational energy in the CO increases with excitation energy, the maximum is remarkably small, being only 4.5%.77 Calculations have been carried out on the fine structure and radiative decay of the 3Az(n7r*) state of f ~ r m a l d e h y d e and , ~ ~ the photolysis of this molecule in the presence of nitric oxide has been studied from a mechanistic point of ~ i e w . 7 ~ Energy transfer in the l A , state of glyoxal has been further investigated,80 with studies on [2H2]glyoxal(‘Au) and [I ,2-2H,]glyoxal (‘A”). Strong isotope effects were again observed on the collision-free lifetimes, owing to effects on the non-radiative decay constant. However, rate constants for the collisional depopulation of any vibronic level, although strongly dependent on the energy of the state, increasing sharply with increase in energy, are however very strikingly independent of isotope. Further studies on methylglyoxal 81 and biacetyl 8 2 confirm the presence of dual emission, with a microsecond and nanosecond component. The molecules belong in the intermediate coupling limit, and among the data given by the authors are the radiative and non-radiative relaxation rates from singlet and triplet levels, and the cross-section for quenching of the ‘slow’ fluorescence as a function of energy. Strong evidence was found for the participation of rotational states in the intramolecular relaxation. A real distinction has to be made between singlet levels which are isolated (low energy), where very large fluorescence cross-sections are found (termed by the authors a ‘black-hole’ region), and where singlet level widths overlap (at higher energies), where at least two effective collisions were required in biacetyl to obtain a thermalized triplet. The mean energy removed for effective collision was 2200 cm-l. The size of the ‘black-hole’ region decreases in going from glyoxal to biacetyl, as summarized in Figure 5. As observed previously, the fluorescence intensity and decay time of glyoxal can be reduced by approximately 15% by a magnetic field of 10 kG,83due to enhancement of the lA, -+ SA, intersystem crossing. Two mechanisms for this process which have different field dependencies have been The resonance fluorescence of an atom in a Plstate under the influence of a rotating magnetic field has been described.86 The photochemistry of propynal vapour has been probed using phosphorescence excitation spectroscopy.8s The conclusions reached in this study were

-

K. Y. Tang, P. W. Fairchild, and E. K. C. Lee, Chem. Phys., 1977,21, 3303. P. L. Houston and C. B. Moore, J. Chem. Phys., 1976, 65, 757. S. R. Langhoff and E. R. Davidson, J. Chem. Phys., 1976, 64, 4699. K. Tadasa, N. Imai, and T. Inaba, Bull. Chem. SOC.Japan, 1976,49, 1758. P. F. Zittel and W. C. Lineberger, J. Chem. Phys., 1977, 66,2972. R. van der Werf, E. Schutten, and J. Kommandeur, Chem. Phys., 1976,16, 151. R. van tier Werf and J. Kommandeur, Chern. Phys., 1976, 16, 125. a A. Matsuzaki and S. Nagakura, 2. phys. Chem. (Frankfurt), 1976, 101,283. 8‘ P. W. Atkins and P. R. Stannard, Chem. Phys. Letters, 1977, 47, 113. 86 W. Hartmann, J . Phys. (B), 1977, 10, 803. 8% J. R. Huber and D. Kumar, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 216. 76

77

156

Photochemistry

that for low excess energies in the initially pumped S1level, intersystem crossing was the main fate, but at the 2l excited level, at an excess energy of 2952 cm-l, predissociation predominates. Above this level, dissociation is reduced in importance again, and between 3000-5000 cm-l, internal conversion becomes

I

I

/ I

0

/I 0

tb I 0

o/ I

1 bracetyl f=

6.3

/ /

I

n

*I’

*/ /

/

/. /.

I

glyoxol

f:8.2

/

/ * /

-

Excitatlon energy (cm-1) I

2 L 000

I

1

26000

Figure 5 Experimental values of n, the number of triplet levels coupled to the singlet level, for rhe compounds shown. Dashed lines give the best J;t of the rate constant for singlet to triplet conversion, kST, from which an enhancement factor, f, can be calculated as shown. At high energies n should be given by the ratio of triplet to singlet level densities p ~ I p s and , solid lines show this ratio as calculated by computer (Reproduced by permission from Chem. Phys., 1976, 16, 151)

the main fate. Above 6000 cm-lS direct photodecomposition occurs. Phosphorbut no escence in the 3n?r* state of 2-furaldehyde has also been inn* emission. The decomposition of the molecule was studied, the quantum yield of CO formation being pressure- and energy-dependent. Thus at limiting low pressures, QCo approached 2, but decreased to zero with increase in pressure of parent compound or added CO,. Hg(3P,) sensitization gave the same products as direct photolysis at 253.7 nm but at 313 nm and 366 nm, direct photolysis produced CO with a quantum yield of only -0.01. a7

A. Gandini, P. A. Hackett, and R. A. Back, Canud. J. Chem., 1976,54, 3089, 3095.

157 I n the photo-oxidation of acetaldehyde, the ratio of rate constants for reactions (41) and (42) had the value kd1/kd2 = 186 Torr for M = N2.88For an overhead Gas-phase Photoprocesses

3MeCH0

+ O2

-

complex

MeCHO

___I+

Me-

+ CO + -HOz

(41)

+ O2 + M

sun in the U.S.A., the rate of H e 0 and HO, formation in the primary oxidation of acetaldehyde can thus be calculated to be 2.8 x and 8.7 x s-l respectively, with an overall rate coefficient for free radical production of s-l. These results have an important bearing on the chemistry of 2.3 x polluted urban atmospheres. The quantum yield of phosphorescence of benzaldehyde at 2 mTorr pressure has been shown to decrease rapidly upon excitation into the S1(m*)band compared with the Sl(nn-*) e x c i t a t i ~ nin , ~agreement ~ with an earlier study.@O Luminescence in the acetone photochemical system @ land a consideration of secondary reactions in the flash photolysis of this molecule in the vapour phase 92 have been reported. HF elimination in the photochemistry of hexafluoroacetone has been d e s ~ r i b e d . ~ ~ The quenching of phosphorescence of biacetyl excited at 404.7 nm by alcohols and iodides has second-order rate constants varying from 6 x lo6 for ethanol to 3 x lo6 for methylene iodide.Q4In the case of alcohols, hydrogen abstraction is the probable mechanism. In the photolysis of selected vibronic levels of cyclobutanone, the ratio of C3 products to C2products, C3/C2,is as large as 7.0 at 343.7 nm, compared with 2.0 at 326.3 nm.95 These new results indicate that the Sl TI intersystem-crossingyield is at least as large as 0.88 at long wavelengths of excitation, and the onset of S1 So internal conversion leading to predissociation may be at 700cm-l above the zero-point level rather than the 1200 cm-l estimated previously. A generalized valence bond description of the low-lying states of keten has been p r e ~ e n t e d . The ~ ~ photochemistry of epoxides in the vapour phase can be explained by the mechanism shown in Scheme Zg7Acetone is chiefly formed in this system by secondary reactions involving hydrogen abstraction. A novel technique in which i.r. laser excitation of tetramethyl-1,Zdioxetan gives rise to visible luminescence has been described, which is an example of i.r.

-

02

0%

8p

05 98

97

-

J. Weaver, J. Meagher, and J. Heicklen, J. Photochem., 1976, 6, 111. T. Itoh, T. Takemura, and H. Baba, Chem. Phys. Letters, 1976, 40, 481. J. Metcalfe, R. G. Brown, and D. Phillips, J.C.S. Faraday ZZ, 1975, 71, 409. S. N. Tong and S. Y . Ho, J. Chinese Chem. SOC. (Formosa), 1976, 23, 203. P. Ahlfors, T. Kauppinen, A. Maki, M.-L. Pohjonen, and J. Koskikallio, Acta Chem. Scad., 1976, A30, 740. J. E. Bassett and E. Whittle, Internat. J. Chem. Kinetics, 1976,8, 859; G . C. Pimentel, F. M. G. Tablas, J.'Hartmann, and E. Whittle, ibid., p. 877. R. W. Carr, jun. and M. P. Ramirez, J. Pholochem., 1977, 6 , 431. K. Y. Tang and E. K. C. Lee, J . Phys. Chem., 1976,80, 1833. L. B. Harding and W. A. Goddard, J. Amer. Chem. SOC.,1976, 98, 6093.

D. R. Paulson, A. S. Murray, D. Bennett, E. Mills, jun., V. 0. Terry, and S. D. Lopez, J. Org. Chem., 1977, 42, 1252.

158

Photochemistry

u"'

w

Scheme 2

laser electronic excitation.08 The overall process is represented by equation (43) and gives rise to quantitative acetone formation. The reaction is sensitized by methyl 0-0

I

t

I

I

Me-C-C-Me

-

0

II

hv(9.6 pm)

(MeF)

2

Me Me (2)

/

C

\

+ hv,,,nm

(43)

Me Me (3)

-

fluoride, which acts on the absorber of i.r. radiation through the steps (44)-(50). Further scissions of dioxetans have been re~0rted.O~ MeF

+ hvi.,.

MeFt MeFt (2)t

+ (2)

+ (2)

+ MeF

(2)t

+ (2)

(2)t land (2)(t)l (3)*

___j

MeFt MeF

+ (2)t

(44) (v-v

+ MeF(t) (2)(t) + MeF(t) (2)(t)

2(2)(t)

+ (3) + hv,,,

I

transfer) (v-t

transfer)

(45) (46) (47)

(48)

(3)*

(49)

(3)

(50)

The photochemistry of 1,1,1-trifluoro-3-bromoacetonehas been described.lO0 The energy dependence of the non-radiative relaxation of camphorquinone vapour has been investigated.lol It was found that k , increased exponentially with excess energy, as has been found for other carbonyl and the increase is consistent with the opening of a new internal-conversion channel. Computed Franck-Condon factors, using a displacement of 0.11 A and an upper-state C - 0 vibrational frequency of 1400 crn-l, give reasonable agreement with measured rates of internal conversion.1o1 98 Bg

100

101

W. E. Farneth, G. Glynn, R. Slater, and N. J. Turro, J. Amer. Chem. SOC.,1976, 98, 7877. H. E. Zimmerman, G. E. Keck, and J. L. Pflederer, J. Amer. Chem. Soc., 1976, 98, 5574. J. R. Majer, J. C. Robb, and Z. Y . Al-Saigh, J.C.S. Furuduy I, 1976, 72, 1697. P. Avouris, W. D. Hopewell, and M. A. El-Sayed, J . Chem. Phys., 1977, 66, 1376.

Gas-phase Photoprocesses

159

4 Radical Reactions Because the field is covered in other titles of the Specialist Periodical Reports series, scant details of these reactions will be given here. The recombination rates of Me radicals and ethyl, isopropyl, and t-butyl radicals have been determined using a phase-modulation method,lo2and that for Me radicals has been calculated in the low-pressure limit.lo3 Rate parameters for the reactions of Me with nitric oxide,lo4molecular and atomic oxygen,1o5and addition reactions to ethylene, acetylene, and benzene lo6and fluoroethylenes lo7have been reported. The photolyses of azoethane and hexafluoroazoethane at 366nm and one atmosphere pressure have been investigated.lo8 The oxidation of methyl radicals at room temperature loo and the spectroscopic detection of MeO- l1O, ll1 have been reported. In the latter study, the excitation threshold for the formation of MeO- from the corresponding nitrite was found to be 6.02 eV, compared with 5.90 and 5.82 eV, respectively, for ethyloxy and isopropyloxy radicals. The zero-zero band in MeOa was found to be at 305 nm, and fluorescence decay times were 3, 1, and 0.3 ps for the three radicals respectively.lll The oxidation of methane proceeds through reactions (51)--(55).8 Reaction (56) was found to be OH

+ CH,

+ 02(+M) MeO, + NO

Me

Me0 + 0 2 HOa+NO

-

H,O

+ Me

Me02(+ M) Me0

+ NO2

(51) (52)

(53)

H C H O + HO,

(54)

OH+NO,

(55)

unimportant, k68/k53 being only -0.05; k63 was found to be 1.2 x 10-la cms molecule-l s-l. In contrast with this system, the analogous reaction does occur MeO,

+ NO2

____+

products

(56)

in alkoxy radicals such as the n-butyloxy radical, but yields of butyl nitrites are much larger than expected, and thus it is necessary to involve reactions (57) and (58) to account for thisD Thus a distinction is to be drawn between long- and ROa

+ NO

ROaNO

-

ROZNO

(57)

RON02

(5 8)

short-chain alkanes in regard to these reactions. Alkoxy radicals having &hydrogens undergo 1$hydrogen shifts, e.g. (59).9 D. A. Parkes, D. M. Paul, and C. P. Quinn, J.C.S. Furaduy I, 1976,72, 1935; D. A. Parkes and C. P. Quinn, ibid., p. 1952. lo3 H. E. van den Bergh, Chem. Phys. Letters, 1976, 43, 201. l o 4 M. J. Pilling, J. A. Robertson, and G. J. Rogers, Znternat. J. Chem. Kinetics, 1976, 8, 883. l o 5 N. Washida and K. D. Bayes, Znternat. J. Chem. Kinetics, 1976, 8, 777. lo6 P. M. Holt and J. A. Kerr, Internat. J. Chem. Kinetics, 1977, 9, 185. lo' H. C. Low, J. M. Tedder, and J. C. Walton, J.C.S. Furaduy I, 1976, 72, 1707. lo* G. 0. Pritchard, F. M. Servedio, and P. E. Marchant, Internut. J. Chem. Kinetics, 1976,8,959. log D. A. Parkes, Znternat. J. Chem. Kinetics, 1977, 9, 451. no H. E. Radford and D. K. Russell, J. Chem. Phys., 1977, 66,2222. ll1 K. Ohbayashi, H. Akimoto, and I. Tanaka, J. Phys. Chem., 1977, 81,798. loa

1 60 0-

1 CH,CH,CH,CHCH,

-

Photochemistry OH

I tH,CH,CH,CHCH,

(59)

Rate constants have been given for the reactions of trifluoromethyl radicals with 0, and F2,112and with ethylene and fluorine-substituted propy1enes;ll3 for hydrogen-abstraction reactions of ethyl radicals with methylfluor~siIanes;~~~ ~ with 02, H,, and D,; and for chlorinefor reactions of isopropyl radicals l l 5 116 atom abstractions by cyclohexyl r a d i ~ a 1 s . l ~Reactions ~ of chlorine atoms with formaldehyde in C1,-0,-H,CO mixtures are represented by reactions (60)(65).11' At 23 "c,k62/k63 Was 5 f- 1 and k64/k63 Was 20.19.

-

+ + H,CO ---+ HCO + 0, HCO + 0, HCO + .c 1 2

hv366

C1

0 2

HCO,

2c1'

HCO

+ HCI

HC03

+ HO, C02 + OH

CO

HCOOH

There have been further kinetic studies to suggest that the CH2(,B1) and CH2(lA1) energy gap is around 8 kcal mol-l. In the first, the figure quoted was 8.7 i- 0.8 kcal m ~ l - l ,whereas ~ ~ ~ the second gives a value of 7.5 kcal mol-1.120 In the latter study the ratio of CH2(1Al)/CH2(3B1)was found to become constant at high pressures of inert gases, independent of the source of methylene, from either direct photolysis of diazirine or triplet-sensitized photolysis. This strongly suggests interconversion, from which simple kinetic arguments give the energy gap quoted. These results are in good agreement with results of numerical calculations which give a best fit of 8.0 kcal mo1-1,121 but the authors of this paper totally demolish the usefulness of their results by stating in a note added in proof that their theory could accommodate an energy gap as large as 19.5 kcal mo1-l. This value has been obtained recently from p.e.s.,12, and appears to be in marked contrast with the results obtained kinetically. A collision-complex model to account for the CH2(lA1)+ CH2(3B1)interconversion has been It has been shown that the vibrational energy distribution of CH,(lA1) produced on photolysis of CH,B, at 435.8 and 366 nm is reasonably broad, whereas that produced on photolysis of keten at 334 or 313 nm is very narrow.124 These V. S. Arutyunov, S. I. Gorshenev, and A. M. Chaikin, Reaction Kinetics and Catalysis Letters, 1977, 6 , 161. 113 H.C. Low, 5. M. Tedder, and J. C. Walton, J.C.S. Faraday I, 1976,72, 1300. 114 T.N.Bell, T. Yokota, and A. G. Sherwood, Canad. J . Chem., 1976, 5 4 , 2 3 5 9 . lls L. Szirovicza and F. Marta, Internat. J. Chem. Kinetics, 1976, 8, 897. 116 R. R. Baldwin, C. J. Cleugh, and R. W. Walker, J.C.S. Furaday I, 1976,72, 1715. 11' M.G . Katz, L. A. Rajbenbach, and G . Baruch, Internat. J. Chem. Kinetics, 1977, 9, 5 5 . T. L. Osif and J. Heicklen, J. Phys. Chem., 1976, 80, 1526. l l D H.M. Frey and G . J. Kennedy, J.C.S. Faraday I, 1977,73, 164. lZo F. Lahmani, J. Phys. Chem., 1976, 80, 2623. lz1 K. C. Kulander and J. S. Dahler, J. Phys. Chem., 1976, 80, 2881. 122 P. F. Zittel, G. B. Ellison, S. V. O'Neil, E. Herbst, W. C. Lineberger, and W. P. Reinhardt, J . Amer. Chem. Soc., 1976, 98, 3731. 123 K. C. Kulander and J. S. Dahler, Chem. Phys. Letters, 1976, 41, 125. 124 T. H. Richardson and J. W. Simons, Chem. Phys. Letters ,1976 .41, 168.

Gas-phase Photoprocesses

161

facts can explain some of the differences in results obtained recently by different groups of workers. A theoretical study of the direct production of CH2(3B1)from singlet diazamethane shows this to be fea~ib1e.l~~ There is evidence for 3CFa formation in the reaction of O(3P) with C2F4.12' Reactions of CH2(lA1)with c y c l o a l k e n e ~ methane,128 ,~~~ and but-2-enes,ll9 and of CH2(3B1)with 0, 129 and C02,1291130 have been reported. The rate constants for reactions (66) and (67) 3CH,+ CO, 3CH,

-

H C H O + CO

+ CH,N2

CH3*

+ -CHN2

(66)

(67)

were given as 3.9 k 1.9 x 10-14 and 1.0 x cm3mo1-1 s-l respe~tive1y.l~~ In the reactions of lCHa with hydrocarbons, the rate constant for energy randomization has been shown to be 4 x 10l2s-l.131 The rate of reaction (68) CH#AJ

+ Hz

CH,

(68)

has been calculated.132 The radiative lifetime of the CH,(blBl) state has been given as 1.9 k 0.15 ps.133 5 Sulphur-containing Compounds Fluorescence at 370 nm has been monitored from the B3&- state of S2,and the decay time given as 45 k 0.6 ns.134 The cross-section for self-quenching of this state was 81.3 k 4.7 A,. Laser excitation of the same S, state has been The use of the Strickler-Berg relationship to compute radiative lifetimes from absorption spectra has been questioned for small molecules such as CS2, SO,, NO2, and ND3.136Fluorescence from CS, vapour excited by a nitrogen laser at 337.1 nm has dual exponential decay, with components of 2.9 ps and 17 ps, attributed to emission from the v," = 1 and v," = 2 levels of the l A , state.137 Photodissociation of CS2 below 160 nm produces CS in the li3n state in high yields between 125 and 140 nm, although the quantum yield of S(lS) production is dO.1. The radiative lifetime of CS(CSII) is deduced to be 16 k 3 ns. Rate coefficients for the quenching of this species by CS,, NO, CO, O,, H2, COz,N2,Ar, and He are compared with similar values for CO(ii311). The reaction of He(2SS) with CS, produces strong emission from the CS+(AeII-+ zzZ+)and CS(AIII -+ x%+) with weaker emission observed from CS,+(A"erI, -+ CS2+(B2Z,+-+ XzrI,), and CS(G3n -+ 8lP) systems. The CS,+ band

z2rIg),

S. N. Datta, C. D. Duncan, H. 0. Pamuk, and C. Trindle, J. Phys. Chem., 1977, 81, 923. D. S. Y. Hsu and M. C. Lin, Chem. Phys., 1977,21, 235. l*' T. L. Rose, A. E. Haas, T. R. Powers, and J. M. Whitney, J. Phys. Chem., 1976, 80, 1653. 128 K. Shibuya, K. Obi, and I. Tanaka, Bull. Chem. SOC. Japan, 1976, 49, 2178. la9 D. S. Y. Hsu and M. C. Lin, Internat. J. Chem. Kinetics, 1977, 9, 507. lSo A. H. Laufer and A. M. Bass, Chem. Phys. Letters, 1977, 46, 151. 131 A. N. KO, B. S. Rabinovitch, and K. J. Chao, J . Chem. Phys., 1977, 66, 1374. lsa C. W. Bauschlicher, jun., K. Haber, H. F. Schaefer, and C. F. Bender, J. Amer. Chem. SOC., 1977, 99, 3610. lSa G. R. Mohlmann and F. J. de Heer, Chem. Phys. Letters, 1976, 43, 236. lS4 T. H. McGee and R. E. Weston, jun., Chem. Phys. Letters, 1977, 47, 352. lS6 S. R. Leone and K. G . Kosnik, Appl. Phys. Letters, 1977, 30, 346. S. Lipsky, J. Chem. Phys., 1976, 65, 3799. lS1 S. J. Silvers and M. R. McKeever, Chem. Phys., 1976, 18, 333. 137u G . Black, R. L. Sharpless, and T. G. Slanger, J. Chem. Phys., 1977,66,2113. 138 J. A. Coxon, P. J. Marcoux, and D. W. Setser, Chem. Phys., 1976, 17, 403. lz6

las

1 62

Photochemistry

systems were also observed in other 66 as was emission from the COS+ excited species.66 The reaction of CS with oxygen atoms produces vibrationally excited CO (69), populations of vibrational levels of which were proposed using tuned laser

-

cs + 0 cot + s (69) Quantum yields of fluorescence from the second excited singlet state of thiophosgene vapour have been determined, and results show that quantum yields for low-lying levels are close to unity.141 The results dictate a reassignment of the origin of the So -+S,system to 36 007 cm-l. For higher levels, the dramatic reduction in fluorescence quantum yield is most likely caused by predissociation. Photolysis of thiophosgene below 127 nm produces CS(2111).142The quantum yield of chlorine atom production is unity for 235.7 nm excitation over the pressure range 0.4-80 Torr, indicating reaction (70) as the predominant

-

C1,CS + hv ClCS + c1 (70) dissociation At longer wavelengths the quantum yield of C1 atom production declines to 0.38 at 435.8 nm, in the limit of low pressure. The pressure dependence permits an estimation of the lifetime of the lA2 state of 55 ns at 435.8 nm excitation, and around 6 ns at 366 nm. The results suggest a new method of C1 isotope enrichment.

6 Nitrogen Compounds Fluorescence from the fi and 2 states of N2+and N,O+ respectively has unit fluorescence quantum yield.66 The vacuum-u.v. photolysis of NH3 produces NH ( f 3 Z - ) . This has been studied by resonance fluorescence, permitting evaluation of the rate constant for reaction (71) to be 4.7 rt 1.2 x 10-l1 cm3mol-1 s-1. NH

+ NO

-

products

(71)

For the $TI state of NH, the v’ = 0 level has a lifetime of around 400ns, independent of J , whereas in the v’ = 1 level it varies from 400 ns for J = 5 to 41 ns for J = 41.1439144 For the state, ~ ( v ’= 0) is -400 ns and T(U’ = 1) is -50 ns. These results imply that very little NH is formed by the two-body radiative recombination of atoms in interstellar space, in contrast with the case with CH and OH. The rate constants for combination of NH(hlC+) with 0 and N have been given as 1.78 ? 0.9 x 10-l’ and 3.38 ? 0.07 x 10-l1 cm3mol-1 s-l for NH respectively, with values for the corresponding state This result is in contrast with another of ND an order of magnitude 10wer.l~~ study where virtually no difference was found between rate constants for reactions of NH(6lZ+) and ND(blX+) with various substrates despite a very strong deuterium isotope effect in the substrate i t ~ e 1 f . lThus ~ ~ Table 6 gives some

cln

139

W.J. Balfour, Canad. J. Phys., 1976, 54, 1969.

140

R. W. Dibble and J. R. Bowen, J. Chem. Phys., 1976, 65, 2028.

H.Okabe, J. Chem. Phys., 1977, 66,2058. 142 T. Oka, A. R. Knight, and R. P. Steer, J. Chem. Phys., 1977, 66, 699. 143 I. Hansen, K.Hoinghaus, C. Zetsch, and F. Stuhl, Chem. Phys. Letters, 1976, 42, 370. 144 W.Hayden Smith, J. Brzozowski, and P. Erman, J. Chem. Phys., 1976, 64,4628. u6 B. Gelernt, S. V. Filseth, and T. Carrington, J. Chem. Phys., 1976, 65, 4940. 141

146

C. Zetsch and F. Stuhl, J. Chem. Phys., 1977, 66, 3107.

163

Gas-phase Photoprocesses

of the data. For collision partners such as helium, rate constants are four orders of magnitude 10wer.I~~ The lifetimes of the dlC+state of NH and ND have been given as 46 k 5 and 62 & 5 ns respectively.148 Table 6 146

Rate constants for reactions of NH(blC+) and ND(@X+) NH

Substrate

ND 6" k x 1014/cm3mol-l s-l

6

Ic x lOl4/crn3mol-l s-l 86 1.7

H2 D2 C2H4

81

1.8 16 1.1

14 1.3

c2D4

The addition reaction of ground-state NH, with propylene has a rate constant given by k = 2.9 x lo8 exp[-4.3 k 0.2 (kcal)/RT] I mol-1 s-1.149 The radicalradical reaction of NH, is termolecular between 0-20 Torr, with efficiencies of third bodies in the ratio 4.0 : 1.0 : 0.4 for NH,, N2, and Ar respectively, and third-order rate constant for nitrogen as third body given by 2.5 f 1.3 x 10l2l2 moP2s-l.160 Reaction (72) represents the main fate of NH2. The z2B1--f x 2 A , absorption spectrum of NH2151 and microwave double resonance in NH2(x2Al)162 have been discussed. NH2+NH,

-

NH,+NH

(72)

In the photolysis of ammonia, NH2D, NHD2, and ND, at 214.4, 213.9, 206.2, and 185 nm, the quantum yield of decomposition was the same for all isomers, and presumably unity,lS3although the rate of decomposition was a factor of two greater in the NH, than ND3 at short wavelengths, and much larger at longer wavelengths. The quantum yield of D2 formation from ND3 was ~ 0 . 0 0 3and 0,009 at 213.9 nm and 185.0 nm respectively. At short wavelengths, NHz(2A,) is produced. The system is of little importance as a means of hydrogen isotope separation. Very weak fluorescence with @F 1.5 x has been observed in NH3 excited at 213.9 and 214.4 nm.ls4 A computer simulation of the photo-oxidation of ammonia has shown that a long-standing mechanism given in reactions (73) to (75) suffices,lS5and there exists no firm proof that a new mechanism proposed recently156is necessary to explain existing data.

-

NH, NH, NH, 147

118

14p 150

151 158

153

lS4 155

156

+ hv

+ O2 + NO

-

+H NO + H,O N, + H,O NH,

(73) (74)

(75)

C. Zetsch and F. Stuhl, Ber. Bunsengesellschnft Phys. Chem., 1976, 80, 1348, 1354. D. K. Hsu and W. H. Smith, J. Chem. Phys., 1977, 66, 1835. R. Lesclaux, J. C. Soulignac, and P. Van Khe, Chern. Phys. Letters, 1976, 43, 520. P. Van Khe, J. C. Soulignac, and R. Lesclaux, J. Phys. Chem., 1977, 81, 210. J. W. C. Johns, D. A. Ramsay, and S. C. Cross, Canad. J . Phys., 1976, 54. 1804. G. W. Hills and R. F. Curl, jun., J. Chem. Phys., 1977, 66, 1507. S. Koda and R. A. Back, Canad. J. Chem., 1977, 55, 1380; R. A. Back and S. Koda, ibid., p. 1387. P. A. Hackett, R. A. Back, and S. Koda, J. Chem. Phys., 1976, 65, 5103. S. Z . Levine and J. G. Calvert, Chem. Phys. Letters, 1977, 46, 81. R. K. M. Jayanty, R. Simonaitis, and J. Heicklen, J. Phys. Chem., 1976, 80, 433.

164

Photochemistry

Tom) A study of the Hg-NH3 exciplex emission at low pressures (~0.03 shows that the emission band at 300 nm is due to a complex between ammonia and the Hg(6,PO) state, as thought previ0us1y.l~~The quantum yield of this emission is around 0.1, the main fate of the complex being non-radiative. The kinetics of excimer lasers using NF, have been and dissociation of NH, by a pulsed C 0 2 laser has been 0 b s e r ~ e d . l ~ ~ ~ Photofragmentation studies continue to rely on cyano-compounds as a fruitful source of experimental data and calculation. The fluorescence of CN(J22n -+ x 2 C + ) in the photolysis of HCN159 and isotope effects in photo161including HCN and DCN, have been fragmentation of linear triatomics,160reported. In the photolysis at 160 nm of C4N2,CN(X2Z+)is produced with the relative populations in the u” = 0, u” = 1 , and d’ = 2 vibrational levels of 0.76 : 0.17 : 0.07, corresponding to a statistical distribution of energy in the predissociating state.ls2 The corresponding photolysis of C2Na at 160 nm again produces 95% of the CN(X) radical with populated u” = 0 and U” = 1 levels.1s3 There is some evidence to show that in this case equal amounts of CN(J21’It)and CN(T2C+).are formed. Energy partitioning in the CN 3 state following i.r. laser-induced photofragmentation has been studied,ls4and an electronic-transition CN laser proposed.ls5 Energy partitioning following photolysis of ICN,166-168a C1CN,ls9and BrCN 170 has been discussed. Visible fluorescence from the vacuumU.V. photolysis of NO-containing compounds such as alkyl nitrites, CF,NO, and ButNO is produced in process (77), following (76).171 CF,NO is a blue gas RON0

+

hv147

NO(Z2E+)

-

+ NO(J2Z+) + hv

ROO

(76)

NO(f211)

(77)

absorbing in the 52&720 nm region, the transition being due to an n -+ n* excitation, in which the NO stretching (vl), CN stretching (vJ, and C-NO bending (v7) vibrations feature pr0minent1y.l~~For selective level excitation, the fluorescence decay time varies smoothly with excess energy, whereas the excitation spectrum has discontinuities at around 627 nm. The non-radiative 15’ 158

lse 160

161

lBZ lR3 16*

le5

166

lb7 lb8

16B 170

I7l 172

A. B. Callear and C. G. Freeman, Chem. Phys. Letters, 1977, 45, 204. W. Chow, M. Stuke, and F. P. Schafer, Appl. Phys., 1977, 13, 1. 5. D. Campbell, G. Hancock, J. B. Halpern, and K. H. Welge, Chem. Phys. Letters, 1976, 44,404. E. N. Tereshchenko and N. Ya. Dodonova, Optika i Spektroskopiya, 1976, 41,495. 0. Atabek, J. A. Beswick, R. Lefebvre, S. Mukamel, and J. Jortner, Chem. Phys. Letters, 1977, 45, 21 1. 0. Atabek, J. A. Beswick, R. Lefebvre, S. Mukamel, and J. Jortner, J. Chem. Phys., 1976, 65, 4035. M. J. Sabety-Dzvonik, R. J. Cody, and W. M. Jackson, Chem. Phys. Letters, 1976, 44, 131. R. J. Cody, M. J. Sabety-Dzvonik, and W. M. Jackson, J. Chem. Phys., 1977, 66, 2145. M, L. Lesiecki and W. A. Guillory, J. Chem. Phys., 1977, 66, 4239. C. R. Quick, C. Wittig, and J. B. Laudenslager, Optics Commun., 1976, 18, 268. A. P. Baronavski and J. R. McDonald, Chem. Phys. Letters, 1977, 45, 172. M. D. Morse, K. F. Freed, and Y. B. Band, Chem. Phys. Letters, 1976, 44, 125. M. J. Sabety-Dzvonik and R. J. Cody, J . Chem. Phys., 1977, 66, 125. U. Halavee and M. Shapiro, Chem. Phys., 1977,21, 105. M. J. Sabety-Dzvonik and R. J. Cody, J. Chem. Phys., 1976, 64,4794. M. N. R. Ashfold and J. P. Simons, Chem. Phys. Letters, 1977, 47, 65. C . A. F. Johnson, V. Freestone, and J. Giovanacci, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 218. K. G. Spears and L. Hornand, J . Chem. Phys., 1977, 66, 1755.

165

Gas-phase Photoprocesses

decay from lower levels is principally internal conversion, and the authors argue that the new channel causing the discontinuity is the onset of direct dissociation. The fluorescence excitation spectrum of the HeNO, van der Waals complex has been observed using the elegant supersonic jet The spectroscopic linewidth suggests a lifetime of the complex of lo-" s, and it is clear that the fluorescence observed is that from excited NO2 produced upon dissociation of the excited complex. There is good evidence to show that azoisopropane isomerizes uia a common state as is the case in simple olefins, with an equal probability of returning to the cis or trans ground The photolysis of ethyleneimineat 147 nm is described by reactions (78) and (79),with quantum yields shown, although (80) may also GH,NH

hvlrr hh47

>

CzH4

CH3*

H*

+ NH + [H,CN]

= 0.38

(78)

@ = 0.47

(79)

+ C,H,N

(80)

occur.175 The absorption spectrum of methyldi-imide vapour has maxima at 360, 208, and 170 nm corresponding to n -+ n*, n -+ 3s Rydberg, and n -+ 3p Rydberg transitions respectively.17g When photolysed in the long-wavelength band, reaction (81) is the principal fate, although in the presence of ethylene NZH,

hv

NZ

+ HZ

(81)

C2H6is the product rather than hydrogen. It is proposed that decomposition occurs from high vibrational levels of the ground state formed by internal conversion.177 The photolysis of alkyl (But, Pri, and Et) nitrites at 254 nm proceeds as in Scheme 3.178 For But, 01 = 0.87, and 01 = 0.45 for Et. A molecular-beam study of the photodissociation of ethyl nitrite has also been carried CH3R'R'CONO

+ IZV

___f

(1 - a)CH3R1R2C0 + NO aCH3R1R2COf

J

CH,.

Nol

+ R1R2C0

R1,R2 = H or Me

CH3N0

Scheme 3 R. E. Smalley, L. Wharton, and D. H. Levy, J. Chem. Phys., 1977, 66, 2750. L. D. Fogel and C. Steel, J. Amer. Chem. SOC.,1976, 98, 4859. 176 A. A. Scala and D. Salomon, J. Chem. Phys., 1976, 65, 4455. 1 7 ~S. K. Vidyarthi, C. Willis, and R. A. Back, Canad. J. Chem., 1977, 55, 1396. C. Willis, R. A. Back, and J. M. Parsons, J. Photochem., 1977, 6, 253. 17* M. I. Christie and P. M. Hetherington, J. Photochem., 1977, 6, 285. 1 7 O A. F. Tuck, J.C.S. Faraday 11, 1977,73,689.

173

166

Photochemistry

For 1,4-diazabicycl0[2,2,2]octane (DABCO) excited in a glow discharge, DABCO emission is observed at low voltages, whereas at high voltages emission in the CN(B2X Z2Z)system is also seen.180 The nature of vibronic intensity borrowing in the ‘A, -+ 1B3utransition of pyrazine has been considered,181 and non-exponential decay in single vibronic level fluorescence from this system has been discussed in detail.la2 A study on the ps relaxation of acridine was mentioned earlier.62 --f

7 Halogenated Compounds To shorten this and succeeding Sections, ground-state phenomena will not be discussed, but merely listed. A review of fluorine atom (2p) reactions of F with H, and hydrocarbons,ls4 aldehydes,ls5 HCI and CI2,lR4 DF(d’ = 1),la6 and the recombination of F atoms lS7have been reported. reactions with H,, D2,188$ methane,190-193 methyl Chlorine atom (“p) chloride and fluoride,lgl ethane,lgl 03,1239lg2,lg4H2 029 lS3HNO lg3HO29 lg3 and acetylene lg5and NO lg6have been studied. Bromine atom (2Pi) recombination lg7 and reactions with hydrogen,lg8 Br2,1gaaHCI and DCl (u” = l ) , l g 9 9 2oo and 1,l-difluoroethylene 201 have been described. Recombination of I(2Pd) 203 and the reaction of this species with HF(u” = 2) 204 are reported. The pressure dependence of atomic fluorine laser transition intensities has been investigated,,05 and oscillator strengths for the CI(4s -+ 3p5)transitions reported.20s lg29

39

2021

F. Lahmani and R. Srinivasan, Chem. Phys. Letters, 1976, 42, 1 1 1. G. Fischer, S. Jakobson, and B. Scharf, Chem. Phys., 1976, 16, 237. lE2 G. Fischer and R. Naaman, Chem. Phys., 1977, 19, 377. lS3 W. E. Jones and E. G. Skolnik, Chem. Rev., 1976, 7 6 , 563. 184 N. F. Chebotaryov, G. V. Pukhalskaya, and S. Ya. Pshezhetsky, Kuantovaia Elektronika, 1977, 4, 872. lE5 D. J. Bogan, D. W. Setser, and J. P. Sung, J . Phys. Cliem., 1977, 81, 888; D. J. Smith, D. W. Setser, K. C. Kim, and D. J. Bogan, ibid., p. 898. lS6 H. K. Shin, Chem. Phys. Letters, 1977, 46, 260. 137 C. J. Ultee, Chem. Phys. Letters, 1977, 46, 366. 198 D. J. Stevens and L. D. Spicer, J . Chem. Phys., 1976, 64, 4798. lEQ R. A. Cox and R. G. Derwent, J.C.S. Farnday I , 1977, 73, 272. l B 0 R. G. Manning and M. J. Kurylo, J . Phys. Chem., 1977, 81, 291. lD1 D. A. Whytock, J. EL Lee, J. V. Michael, W. A. Payne, and L. J. Stief, J . Chem. Phys., 1977, 66, 2690. lg2 R. Watson, G. Machado, S. Fischer, and D. D. Davis, J. Chem. Phys., 1976, 65, 2126. lg3 M.T. Leu and N. B. DeMore, Chem. Phys. Letters, 1976, 41, 121. lQ4 M. A. A. Clyne and W. S. Nip, J.C.S. Faraduy II, 1976, 7 2 , 838. lQ6 F. S. C . Lee and F. S. Rowland, J . Phys. Chem., 1977, 21, 684. lQ8 H. Hippler and J. Troe, Internal. J . Chem. Kinetics, 1976, 8, 501. lQ7 D. T. Chang and G. Burns, Canad. J . Chem., 1976, 54, 1535. lQ8 I. H. Zimmerman, J.-M. Yuan, and T. F. George, J. Chem. Phys., 1977, 66, 2638. 1a8a F. Zaraga, S. R. Leone, and C. B. Moore, Chem. Phys. Letters, 1976, 42, 275. lg9 R. D. H. Brown, I. W. M. Smith, and S . W. J. van der Merwe, Chem. Phys., 1976, 15, 143. I. W. M. Smith, Chem. Phys., 1977, 20, 437. 201 J. M. Pickard and A. S . Rogers, J . Amer. Chem. SOC.,1976, 98, 6115. ao2 R. E. Antrim, G. Burns, and J. K. K. Ip, Canad. J. Chem., 1977, 55, 749. 203 W. H. Beck and J. C. Mackie, Chem. Phys. Letters, 1976, 44, 444. 204 J. Wanner, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 129. 206 L. 0. Hocker and T. B. Phi, A p p f . Phys. Letters, 1976, 29, 493. 206 M. A. A. Clyne and W. S . Nip, J.C.S. Faraday II, 1977, 73, 161.

lE0

Gas-phase Photoprocesses

167

Electron photodetachment cross-sections for C1- and Br- have been The photoexcitation of Br, to the D3II0,,+state provides a source of Br(2Pa) atoms. Reactions of these with propylene have been investigated, and under the conditions of these experiments have been shown to have shorter radical chains than in other systems.208The quenchings of Br(,Pa) by H,, D,, and HD have the rate constants 1.8 x 1O-l1, 1.8 x 10-l2,and 3.6 x 10-l1cm3molecule-ls-l respectively.209 Electronic-to-vibrational energy transfer from Br(2P&to HCN and the subsequent deactivation of HCN(OO1) have been studied.210 Deactivation of I(2Pi) by a variety of substrates including Br,,211 C2F,J,212 hydrogen halides,213and others 214 has been studied. Selected rate constants are given in Table 7. The radiative lifetime of I(2Pi) has been given as 108 ms.212r216

Table 7 Rate constants in cm3molecule-l s-l for deactivation of I(,P) Substrate HF HCl HCl HBr HBr HI HI Br2 C2FJ H2O D2O D C1 DBr DI

k 3.1 6.5 1.5 1.6 1.3 5.0 5.2 6.0 1.8 8.4 1.8 4.3 G 4.9 5.0

x 10-l2 x 10-l6

x x x x x x x

x x x x x

10-14 10-13 10-13 10-14 10-14 10-ll 10-17 10-13 10-14 10-I6 10-14 10-14

Ref. 213 213 214 213 214 21 3 214 21 1 212 214 214 214 214 214

Absolute rate constants for competing reactions of CnF2n+land I(,Pi) in the gas phase have been given,21sand a study has been reported of the production of I(2Pi)by collisional release in molecular iodine.217 A photo-initiated F2/H2and F2/D2laser with high electrical efficiency has been described.218 The quenching of C12(~3110,+) by either Cl, or Ar has a rate constant of 3.2 x 1013cm3mol-1 s-l, quenching being due to collision-induced predisso~iation.~~~ The two-photon excitation spectrum of I, 220 and laser-excited fluorescence have been discussed. A from the fi311yu+-+ xlC,+ system of 12'12 and 12912221 207 208

209

210 21a

21s 21r

z16 217 218

221

A. Mandi, Phys. Reu. (A), 1976, 14, 345. K. B. McAfee, jun., R. M. Lum, and R. S. Hozack, J. Chem. Phys., 1976, 64, 5073. J. R. Wiesenfeld and G . L. Wolk, J. Chem. Phys., 1976, 65, 1506. A. Hariri, A. B. Petersen, and C. Wittig, J. Chem. Phys., 1976, 65, 1872. P. L. Houston, Chem. Phys. Letters, 1977, 47, 137. F. J. Comes and S. Pionteck, Chem. Phys. Letters, 1976, 42, 558. A. T. Pritt, jun. and R. D. Coombe, J. Chem. Phys., 1976, 64, 2096. R. J. Donovan, C. Fotakis, and M. F. Golde, J.C.S. Faraday 11, 1976, 72, 2055. F. J. Comes and S. Pionteck, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 219. G. A. Skorobogatov, V. G. Seleznev, and 0. N. Slesar, Doklady Akad. Nauk S.S.S.R. Ser. khim., 1976, 231, 1407. D. H. Burde and R. A. McFarlane, Chem. Phys., 1976, 16,295. D. B. Nichols, R. B. Hall, and J. Doyle McClure, J. Appl. Phys., 1976, 47, 4026. R. E. Huie, N. J. T. Long, and B. A. Thrush, Chem. Phys. Letters, 1976, 44, 608. M. D. Danyluk and G. W. King, Chem. Phys. Letters, 1976, 44, 440. K. K. Yee, J.C.S. Faraday ZI, 1976, 72, 2113.

168

Photochemistry

high-resolution study on the hyperfine components of I, rotational levels has confirmed theoretical predictions that these lead to non-exponential decay of fluorescence.222The lifetime of the 43 -+ 2 band in emission from the BSIIou+ state has been given as 2.13 ps, different from the single rotational level [P(16) in the 42 + 2 band] of 1.82 ps.223The kinetics of selective excitation of 0-1, by lasers 224 and thermally dissociating I, in laser applications 225 have been described. An U.V. laser based upon the excitation of the iodine 6lZ;,+ state, in which collisional deactivation to the lasing I,31T,, level occurs, has been proposed.226-228 In the presence of oxygen, the fi state produces IO.,,' The lifetime of the b state is 15.5 k 0.5 ns,,,* and rate constants for deactivation to the lasing 31T2, level range from 1.88 x 10-l2 ~ m ~ m o l e c u l e - ~ for s - ~ Ne to 0.96 x 10-lo cm3 molecule-l s-l for C0,.228 The supersonic beam technique has been used to observe the photodissociation of the van der Waals complex 12He.229In this study it was found that excitation of level u in I, causes fluorescence exclusively from the u - 1 level; thus coupling between the I, stretch and van der Waals stretch must be weak. The B31J,+ states of BrC1,230IF, ICl, and IBr231 have been observed in emission. The quenching of BrC1(3110+)by BrCl, Cl,, He, and air had rate to 1.6 x 10-l2 cm3molecule-l constants ranging from 1.2 x Predissociation in IBr as a case of intermediate coupling has been u' x211i,v ' ' ) , , ~ ~ Emission spectra lifetimes and cross-sections for HCl+ and spectra of the ClCN+, BrCN+, ICNf B21T -+ Z2lJ and J2iC+-+ fTl systems 234 have been recorded. Exciplex lasers based upon XeF and related compounds 236 have prompted much research on such species, including papers on XeF,236-242KfF,240,241 and XeC1.244 Rate constants for reaction (82) were found to be 7.5 x 10-lo, 6.4 x 10-lo, and 2 x 10-lo cm3molecule-l s-l for A = Xe(",), Kr(3P,), and AI~~P,,,)

(x2X+,

XeF, 22z

223 2a4 z26

226

z27 228

22B 23u 231

232

233 z34 235 236

237

238 239

240 241 242

243 244

+ APP,,,)

-

XeF*

+ AF

--f

(82)

M. Broyer, J. Vigue, and J.-C. Lehmann, J. Chem. Phys., 1976,64,4793. K. Sakurai, G. Tareb, and H. P. Broida, Chem. Phys. Letters, 1976, 41, 39. V. I. Balykin, V. S. Letokhov, V. I. Mishin, and V. A. Semchishen, Chem. Phys., 1976,17, 111. D. R. Gray, H. J. Baker, and T. A. King, J. Phys. ( D ) , 1977, 10, 169. A. B. Callear and M. P. Metcalfe, Chem. Phys., 1977, 20, 233. A. B. Callear and M. P. Metcalfe, Chem. Phys. Letters, 1976, 43, 197. A. B. Callear, P. Erman, and J. Kurepa, Chem. Phys. Letters, 1976, 44, 599. M. S. Sim, R. E. Smalley, L. Wharton, and D. H. Levy, J. Chem. Phys., 1976, 65, 1216. 5. 5. Wright, W. S. Spates, and S. J. Davis, J. Chem. Phys., 1977, 66, 1566. M. A. A. Clyne and I. S. McDermid, J.C.S. Fnraday 11, 1976, 72, 2242, 2252. M. S. Child, Mol. Phys., 1976, 32, 1495. G . R. Mohlmann, K. K. Bhatani, and F. J. de Heer, Chem. Phys., 1977, 21, 127. M. Allan and J. P. Maier, Chem. Phys. Letters, 1976, 41, 231. H. Hutchinson, La Recherche, 1977, 8, 370. R. Burnham and N. W. Harris, J. Chem. Phys., 1977, 66, 2742. J. Tellinghuisen, G . C. Tisone, J. M. Hoffman, and A. K. Hays, J. Chem. Phys., 1976,64,4796. M. Rokni, J. H. Jacob, J. A. Mangano, and R. Brochu, Appl. Phys. Letters, 1977, 30, 458. J. G. Eden and S. K. Searles, Appl. Phys. Letters, 1977, 30, 287. J. G . Eden and S. K. Searles, Appl. Phys. Letters, 1976, 29, 356. J. E. Velazco, J. H. Kolts, and D. W. Setser, J. Chem. Phys., 1976, 65, 3468. J. E. Velazco, J. H. Kolts, D. W. Setser, and J. A. Coxon, Chem. Phys. Letters, 1977, 46, 99. Yu. A. Kudryavtsev and N. P. Kuzmina, Appl. Phys., 1977, 13, 107. R. Shuker, Appl. Phys. Letters, 1976, 29, 785.

Gas-phase Photoprocesses 169 A laboratory study of the absorption of C10,245$ 246 and cross-beam chemiluminescence studies on alkaline earth-C10, reactions 247 have been reported. Isotope separation using the photolysis of phosgene-CO mixtures has been discussed (vide infra).e48 Vibrational energy transfer in CD3F,249and energy distribution in CD3Cl, CD3H unimolecular decompositions 250 have been investigated. The photolysis of CCll at 213.9, 163.3, and 142.0nm has two wavelength-dependent primary (83) and (84), with quantum yields shown. The CCl, formed in (84)

does not react with CC14, but inserts into HCl and combines with other radicals. At 313 nm, the absorption cross-section is small (d 3.7 x cm2molecule-') and thus photolysis of this species in the troposphere is unimportant compared with competing processes. The photolysis of ethyl chloride at 147 nm can be summarized by Scheme 4, where (*) and (**) denote different electronic C2H5CI

+

C2H5 CI

4+

C2H5Cl

hv + C,H,Cl *

** C1

L

Y

-

C2H3Clt + H2

.1

i

C,H,f-+ HCl

C,H,

+ HC1

.1

CZH, + H2 Scheme 4

The photolysis of 2-chloropropane at 147 nm and 123.6-116.5 nm produces HCI plus propylene in a hot state from which allene, propyne, acetylene, and ethylene can be formed ~ u b s e q u e n t l y . ~ ~ ~ 1.r. multiphoton laser-induced reactions in methyl halides,254 and alkyl halides,266including CC1F3 and CC13F,256 have been studied. The photolysis of CC1,Br in the presence of CH2C12, CH2F2, CH,FCl, and CHFCI, has been described.e57The sequential two-photon photodissociation of methylene iodide proceeds by excitation first to a repulsive B1 state, forming CH21 and I, with 80-90% of available energy carried in the ro-vibrational levels of the CHJ P. Rigaud, B. Leroy, G. le Bras, G. Poulet, J. L. .Jourdain, and J. Combourieu, Chem. Phys. Letters, 1977, 46, 161. J. A. Coxon, J . Photochem., 1977, 6, 439. 247 F. Engelke, R. K. Sander, and R. N. Zare, J. Chem. Phys., 1976,65, 1146. 24* Z. B. Vukmirovic and S. V. Ribnikar, J. Chem. Phys., 1977, 66, 7. 54B L. A. Gamss, B. H. Kohn, A. M. Ronn, and G. W. Flynn, Chem. Phys. Letters, 1976,41,413. 2so J. D. McDonald and R. A. Marcus, J. Chem. Phys., 1976, 65,2180. 261 R. E. Rebbert and P. J. Ausloos, J . Photochem., 1977, 6, 265. 262 T. Ichimura, A. W. Kirk, G. Kramer, and E. Tschuikow-Roux, J . Photochem., 1976, 6, 77. 46a P. Gagnon and J. A. Herman, Canad. J. Chem., 1976, 54, 3470. 264 B. L. Earl and A. M. Ronn, Chem. Phys. Letters, 1976, 41, 29. .mb W. Braun and W. Tsang, Chem. Phys. Letters, 1976, 44, 354. 268 D. F. Dever and E. Grunwald, J. Amer. Chem. SOC.,1976, 98, 5055. 257 D. E. Copp and J. M. Tedder, J.C.S. Faraday I, 1976,72, 1177. 246

170

Photochemistry

fragment, followed by absorption of a second photon by CHJ to give methylene and iodine Two excited states have again been invoked in the photolysis of CH3CC13,259 reactions (85) and (87). Reactions (85) and (86) occur chiefly at excitation wavelengths > 220 nm, (87) for h < 220 nm. CH,CCI,* CH,CCl,**

-

CH,CC12 + HCI CH,CCI2 CH3CC1,

+ CI. + Cl*

The photochemical fluorination of fluoroform,260and the photobromination of bromobutane with alBr 261 have been described. Photolysis of RI at 254 nm and 155-175 nm, where R = C3F7, CF3, or iso-C3F7, has been discussed with reference to the iodine photochemical laser.262 Absorption cross-sections of CF2C12,263-26s CFC13,263and other halogenated methanes 266 of stratospheric importance have been measured, and the reactions of CFCl, with atmospheric-like ions Two different excited states have again been invoked in the photolysis of l,l-dichloroethylene,26a one leading to C2H2and C,HCl in the ratio of 0.3 : 1 as products, the other dissociating to give CH,Cl radicals. The photochemical isomerization of 1,2-dichIoroethylene by U.V. radiation 269 and i.r. multiphoton excitation 270 have been described. The trimerization of tetrachloroethylene sensitized by BC13271is a further example of a specific i.r. laser reaction. The efficiency of intramolecular vibrational energy redistribution in chloroacetylene as a model system has been In some elegant molecular beam work, the translational energy in the products of photodissociation of 1,5-di-iodonaphthalene, 4,4’-di-iodobiphenyl, 4,4’dibromobiphenyl, and 9-bromofluorene for 265-321 nm excitation were found to be independent of wavelength of excitation.273 This important study shows that the electronic energy absorbed by the aromatic moiety, but not the vibrational energy, is transferred to the C-X bond. Lifetimes of excited states were estimated to be 6 2 ps. 2659

268

2e1

263

264

286 266

287 268 26s

270

271 272

273

P. M. Kroger, P. C. Demou, and S. J. Riley, J. Chem. Phys., 1976, 65, 1823. T. S. Yuan and M. J. 5. Wijnen, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 310. V. I. Vedeneev, M. A. Teitel’boim, and A. A. Shoikhet, Izvest Akad. Nauk S.S.S.R. Ser. khim., 1976, 25, 1850. D. D. Tanner, E. V. Blackburn, Y . Kosugi, and T. C. S. RUO,J. Amer. Chem. SOC., 1977, 99, 2714. L. G. Karpov, A. M. Pravilov, and F. I. Vilesov, Kvantovaia Elektronika, 1977, 4, 889. C. C. Chou, W. S. Smith, H. V. Ruiz, K. Moe, G . Crescentini, M. J. Molina, and F. S. Rowland, J . Phys. Chem., 1977, 81, 286. K. M. Monahan, R. L. Russell, and W. C. Walker, J. Chem. Phys., 1976, 64, 5309. R. G . Green and R. P. Wayne, J. Photochem., 1977, 6, 371, 375. C. Hubrich, C. Zetzsch, and F. Stuhl, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 437. R. G . Hirsch, Atmos. Environment, 1976, 10, 703. R. Ausubel and M. H. J. Wijnen, J. Photochem., 1976, 5, 233. R. Ausubel and M. H. J. Wijnen, 2. phys. Chem. (Frankfurt), 1976,100, 175. R. V. Ambartzumian, N . V. Chekalin, V. S. Doljikov, V. S. Letokhov, and V. N. Lokhman, J. Photochem., 1976, 6 , 55. 13. R. Bachmann, R. Rink, H. Noth, and K. L. Kompa, Chem. Phys. Letters, 1977, 45, 169. W. L. Hase and C. S. Sloane, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 207. M. Kawasaki, S. J. Lee, and R. Bersohn, J. Chem. Phys., 1977, 66, 2647.

171 The laser-induced fluorescence of HgBr (B -+ 2)has been studied, and the decay rate of the B state found to be 4.3 x lo7s.274 Gas-phase Photoprocesses

8 Atom Reactions A review of lifetime measurements in atoms, molecules, and ions275has been written. Papers concerned with noble-gas atom processes have reported lineshapes in two-photon absorptions in atomic v a p o ~ r sreactions , ~ ~ ~ of metastable Ar with N2,277-27g Ar metastables with Kr, Xe,2sol281 and 0, C1 281 atoms, He(2lS) with Ne,280Kr metastables in krypton-nitrogen afterglows,2s2cross-sections between excited states of Kr and Xe,283Ar(3P2,0) reactions with chlorine atoms,206 fluorescence from laser-excited rare-gas r n e t a ~ t a b l e s ,excitation ~~~ transfer in n = 4 levels of He following dye-laser pumping of the 4lO level,2s6photoionization of Ne,286s287 He, Ar, and Kr,287and vacuum-u.v. fluorescence in Xe.288 Rotational predissociation in HeH+,289photodissociation of the dimer ions Ar2+,Ne2+, and (C02)2+,2s0 photoionization of van der Waals molecules Ar, and Kr2,291and the kinetics of emission from XeO 292 have also been discussed. The quenching of an afterglow of magnesium (") atoms 293 and the broadening of the Mg 285.2 nm line by inert gases 294 have been described. Near-resonant electronic energy transfer between Zn(lP,) 295 and Cd(l4) 2D6 and N0(Z2II) has been studied quantitatively, as has electronic energy transfer from Cd(53P,) to POPOP v a p ~ u r , ~ ~ ~ A model has been proposed for the mercury-photosensitized hydrogenabstraction reaction with alkanes in which electronic energy is transferred to the Laser magnetic resonance of the Hg(3P,) atom,2ssreactions of Hg(3Po) 274

277

278

278

281 282

283 284

28b 28E

287 288

288 290

291

292 293 29p

2B6 2D7 298 288

N. Djeu and C. Mazza, Chem. Phys. Letters, 1977, 46, 172. R. E. Imhof and F. H. Read, Rep. Progr. Physics, 1977,40, 1. J. E. Bjorkholm and P. F. Liao, Phys. Rev. (A), 1976, 14, 751. A. N. Schweid, M. A. D. Fluendy, and E. E. Muschlitz, jun., Chem. Phys. Letters, 1976, 42, 103. R. A. Sanders, A. N. Schweid, M. Weiss, and E. E. Muschlitz, jun., J. Chem. Phys., 1976, 65, 2700. J. Krenos and J. B. Bruno, J. Chem. Phys., 1976, 65, 5617. D. W. Martin, T. Fukuyama, R. W. Gregor, R. M. Jordan, and P. E. Siska, J. Chem. Pliys., 1976, 65, 3720. D. L. King, L. G. Piper, and D. W. Setser, J.C.S. Faruday 11, 1977, 73, 177. C. J. Tracy and H. J. Oskam, J. Chem. Phys., 1976, 65, 1666. R. Damaschini, J. Cahen, J. Brochard, and R. Vetter, J . Phys. (B), 1976, 9, L211. V. E. Bondybev and T. A. Miller, J. Chem. Phys., 1977, 66, 3337. M. J. Shaw and M. J. Webster, J. Phys. (B), 1976, 9, 2839. G. Baravian, J. Godart, and G. Sultan, Phys. Rev. (A), 1976, 14, 761. J. B. West and G. V. Marr, Proc. Roy. SOC.,1976, A349, 397. R. Brodmann, G. Zimmerer, and U. Hahn, Chem. Phys. Lerters, 1977,41, 160. R. Locht, J. G. Maas, N. P. F. B. van Assett, and J. Los, Chem. Phys., 1976, 15, 52. M. L. Vestal and G. H. Manclaire, Chem. Phys. Letters, 1976, 43, 499. C. Y. Ng, D. J. Trevor, B. H. Mahan, and Y. T. Lee, J. Chem. Phys., 1977,66,446. V. Ya. Aleksandrov, V. Ya. Vinogradov, V. V. Lugovsky, and I. V. Podmoshensky, Optika i Spektroskopiya, 1976, 41, 390. G. Taieb and H. P. Broida, J. Chem. Phys., 1976, 65, 2914. G. V. Zhuvikin, N. P. Penkin, and L. N. Shabanova, Optika i Spektroskopiya, 1977,42,236. W. H. Breckenridge, R. P. Blickensderfer, and J. FitzPatrick, J. Phys. Chem., 1976, 80, 1963. W. H. Breckenridge and J. FitzPatrick, J . Phys. Chem., 1976, 80, 1955. D. J. Ehrlich and J. Wilson, Optics Commun., 1977, 20, 314. G. C. Marconi, G. Orlandi, and G. Poggi, Chem. Phys. Letters, 1976, 40, 8 8 . J. W. C. Johns, A. R. W. McKellar, and M. Riggin, J. Chem. Phys., 1977, 66, 3962. 7

Photochemistry

172

and inert gases 300 and fluorescence from these,301and a laser fluorescence study of Hg(,PO) quenching rates ,029 ,03 have been reported. For the reactions below, the rate constants measured were kg8 = (7.1 k 1.4) x 10-l2 cm3molecule-l s-l and (1.55 k 0.16) x 1 O - , O cm-s s-l at 293 K 3 0 2 Quenching of Hg(,P,) by C2H4, 02,NO, N 2 0 , H2, CO, C02, NH,, CH4, and Xe had rate constants in units of 10-l6 cm3molecule-1 s-l of 3.4 x lo6,2.6 x lo6, 1.95 x lo6, 5.1 x lo6, 6.2 x lo4, 4.15 x lo3, 3.1 x lo3, 21.7, and 2.25 respectively.303 Further aspects of the (Hg),* excimers formed in (89) have been Hid3PI) + N, W 3 P 0 ) + N2 + Hg

-

Hid3PO)+ N2 Hg2C1,)

+ N2

(88)

(89)

The potential of the thallium-mercury mixed excimer as a laser system of high average power at 459 and 656 nm has been Sodium atom reactions have been reported with the following substrates : a r g ~ n , ~ O hydrogen ~ - ~ ~ l atoms,311 C0,312 other excited Na N20,,14 and other simple 316 Tunnelling rates for excited Na in a static electric field,317 and spin relaxation of Rb atoms in collision with rare gases have also been described. Atom reactions of the above kind are often accompanied by the appearance of chemiluminescence, and product distributions can also be probed by laserinduced fluorescence. Such studies have been carried out on the following N20,3239 324 CH,Br, CH,Br2, systems: Barium plus Cl2,,l912,320CC14,321S2C12,322 CHBr,, and CBr4,325C 0 2 and C0,326Sr (Ye)in CO/N20 alkaline300 301 302 303 304

305

306 307

308 300

310

311 312 313

314

318 317

318 31B 320

321 322 323 324 325

326 827

N. P. Penkin, T. P. Red’ko, and N. A. Krykov, Optika i Spektroskopiya, 1977, 42, 224. J. R. Woodworth, J. Chem. Phys., 1977, 66,754. L. F. Phillips, J.C.S. Furaday ZZ, 1976, 72,2082. L. F. Phillips, J.C.S. Faraday ZZ, 1977, 73,97. J. Skonieczny, Acta Phys. Polon. (A), 1976, 50, 117. J. Skonieczny, Acta Phys. Polon. (A), 1976, 49, 807. M. Stock, R. E. Drullinger, and M. M. Hessel, Chem. Phys. Letters, 1977, 45, 592. D. Drummond and L. A. Schlie, J. Chem. Phys., 1976, 65,3454. R. Duren, H. 0. Hoppe, and H. Pauly, Phys. Rev. Lerters, 1976, 37,743. R. E. Smalley, D. A. Auerbach, P. S . H. Fitch, D. H. Levy, and L. Wharton, J. Chem. Phys., 1977, 66, 3778. B. S. Ault, D. E. Tevault, and L. Andrews, J. Chem. Phys., 1977, 66, 1383. T. R. Proctor and W. C. Stwalley, J. Chem. Phys., 1977, 66,2063. D. S. Y. Hsu and M. C. Lin, Chem. Phys. Letters, 1976, 42, 78. M. Allegrini, G. Alzetta, A. Kopystynska, L. Moi, and G. Orriols, Optics Commun., 1976, 19, 96. R. C. Benson, J . Chem. Phys., 1977, 66,3879. J. R. Barker and R. E. Weston, jun., J. Chem. Phys., 1976, 65, 1427; S.-M. Lin and R. E. Weston, jun., ibid., p. 1443. I. V. Hertel, H. Hoffmann, and K. A. Rost, Chern. Phys. Letters, 1977, 47, 163. M.G. Littman, M. L. Zimmerman, and D. Kleppner, Phys. Rev. Letters, 1976, 37,486. F. A. Franz and C. Volk, Phys. Rev. (A), 1976, 14, 1711. C. A. Mims and J. H. Brophy, J. Chem. Phys., 1977, 66, 1378. C. R. Dickson, J. B. Kinney, and R. N. Zare, Chem. Phys., 1976, 15, 5 3 ; P. J. Dagdigian, H. W. Cruse, and R. N. Zare, ibid., p. 249. W. Schmidt, A. Siegel, and A. Schultz, Chem. Phys., 1976, 16, 161. F. Engelke and R. N. Zare, Chern. Phys., 1977, 19, 327. W. Felder, R. K. Gould, and A. Fontijn, J . Chem. Phys., 1977, 66,3256. M.A. Revelli, B. G. Wicke, and D. 0. Harris, J. Chem. Phys., 1977, 66,732. M. Rommel and A. Schultz, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 139. J. B. West and H. M. Poland, J. Chem. Phys., 1977, 66,2139. B. J. Jansen, Tj. Hollander, and H. Van Helvoort, J. Quantum Spectroscopy Radiative Transfer, 1977, 17, 193.

173

Gas-phase Pho toprocesses

earth atoms with BrCN,828 Sn plus Nz0,320A1 plus 02330 and halogens,331 Pb atoms plus 03,332 Cs with Fz,333 phosphorus flames 334, 336 and B with alkali General descriptions of these methods 338 have been given. Two-photon third harmonic generation in calcium vapour has been achieved,339 and lifetimes for the Z3P, states of Si and S,340the 8s level of the 26 099 and 29 973 cm-l levels of Er11,34z and Yb' and Yb'l 343 have been measured. The and the excitation of the 5d 2Dtlevel of Cs in collisions with rare-gas depolarization of the C S ( ~ P ,state ) by the same species34s has been studied. Phosphorescence from P(*&) has been observed by atomic resonance 3379

Recent studies347on the deactivation of Bi(2D3) and Bi(2Dt) by a variety of species have produced results at variance with earlier The point is illustrated in Table 8 with selected data. No explanation for the differences was Table 8 Quenching of Bi(zDt) and Bi(2D*) k0[Bi(24)1/ cm3molecule-1 s-1

Substrate H 2

DZ

co

Ref. 347 7.2 x 10-16

Ref. 348 7.9 x 10-14

x

1.1 x 10-14 7.6 x 10-14

a s

~ 1 . 4x

10-16

Ref. 347 1.0 x 10-l1 1.1 x 10-13 4.7 x 10-13

Ref. 348 5.6 x 10-l2 2.4 x 10-13 5.4 x 10-13

360 atoms with established. Reactions of ground-state Bi(4S+)340 and Sb(4S~) and SF6 have been studied, with kinetic data ethylene in the presence of He, Nz, and 3Pz states occur at given. The absorptions of the Ge atom in the 249.8, 259.3, and 275.5 nm, and thus the reactions of three individual levels can be studied by using atomic absorption at these wavelengths as a 828 s28

s30 831 832

333 s34 s35 3s6

337

338 s39 s40

34a s43

s47 34* 84g

860

s61

L. Pasternack and P. J. Dagdigian, J. Chem. Phys., 1976, 65, 1320. A. Yokozeki and M. Menzinger, Chem. Phys., 1977, 20,9. D. M. Lindsay and J. L. Gole, J. Chem. Phys., 1977, 66, 3886. S. Rosenwaks, J. Chem. Phys., 1976, 65, 3668. M. J. Kurylo, W. Braun, S. Abramowitz, and M. Krauss, J. Res. Nat. Bur. Stand., Sect. A , 1976,80, 167. L. H. Hall, J. Chem. Phys., 1977,66,2435. R. J. Vanzee and A. U. Khan, Chem. Phys. Letters, 1976, 41, 180. R. J. Vanzee and A. U. Khan, J. Chem. Phys., 1976, 65, 1764. V. C. Sridharan, D. L. McFadden, and P. Davidovits, J . Chem. Phys., 1976, 65, 5373. D. R. Preuss and J. L. Gole, J. Chem. Phys., 1977, 66, 2994; J. L. Gole and D. R. Preuss, ibid., p. 3000. F. Engelke, Eer. Eunsengesellschafi Phys. Chem., 1977, 81, 135. A. I. Ferguson and E. G. Arthurs, Phys. Letters (A), 1976, 58, 298. S. L. Varghese, C. L. Cooke, and B. Carnutte, Phys. Rev. (A), 1976, 14, 1729. G. Alessandretti, F. Chiarini, G. Gorini, and F. Petrucci, Optics Commun., 1977, 20, 289. B. Engman, J. 0. Stoner, jun., I. Martinson, and N. E. Cerne, Phys. Scripta, 1976, 13, 363. F. H. K. Rambow and L. D. Schearer, Phys. Reo. (A), 1976, 14,738. R. T. M. Su, J. W. Bevan, and R. F. Curl, jun., Chem. Phys. Letters, 1976, 43, 162. J. Guiry and L. Krause, Phys. Rev. (A), 1976, 14, 2034. D. Husain and P. E. Norris, J.C.S. Faraday ZZ, 1977, 73, 415. D. W. Trainor, J. Chem. Phys., 1977, 66, 3094. M. J. Bevan and D. Husain, J. Phys. Chem., 1976, 80, 217. D. Husain and N. K. H. Siater, J . Photochem., 1977, 6, 325. D. Husain and N. K. H. Slater, J. Photochem., 1977, 7 , 59. M. A. Chowdhury and D. Husain, J. Photochem., 1977, 7 , 41.

174

Photochemistry

The reactivity of excited atoms of this type,352and the dynamics of collisions between vibrationally excited diatomics and atoms have been reviewed.3s3* 3s4 Laser action in atomic copper,355-3s8atomic iridium,359and manganese 360 has been reported. Numerical studies on radiation trapping of argon 106.7 nm radiation have been discussed 9 Miscellaneous

The following diatomic species have been studied in either emission or from the point of view of photodissociation: Liz, in which 7 for the A%,+ state was 18 ns,362Na 29 363- 364 Rb,3655366 K 2,367 Ca2,36* NaK,36gNa halides,370HgBr,274MgH and MgD,371PO,372NeI,373ZnH, ZnD, CdH, and CdD,374LaS,37s AgBi,376 C10,377Ti0,378CaH,379ScO and YO,380SiF,381SrH and SrD,38230Se 383 and TlI.384 The laser-excited fluorescence of B02385 and the chemiluminescence of AI(Me), in active oxygen and nitrogen 386 have also been discussed. Fluorescence has been observed from UF6 in the vapour phase, with quantum yield of for excitation between 400 and 460 nm.3*7 The radiative lifetime 29

N

D. Husain, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 168. 353 M. Quack and J. Troe, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 160. s4 I. W. M. Smith, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 126. 365 M. A. Alaev, A. I. Baranov, N. M. Vereshchagin, I. N. Gnedin, Yu. P. Zherebtsov, V. F. Moskatenko, and Yu. M. Tsukanov, Kvantovaia Elektronika, 1976, 3, 1134. 366 A. A. Isaev, M. A. Kazaryan, G. Yu. Lemmerman, G. G. Petrash, and A. N. Trofimov, Kvantovaia Elektronika, 1976, 3, 1800. 357 J. R. McNeil and G. J. Collins, I.E.E.E. J. Quantum Electronics, 1976, 12, 371. s6a A. I. Fyodorov, V. P. Sergeenko, V. F. Tarasenko, and V. S. Sedoi, Izuest. Vyssh. Uch. Zav. Fiz. S.S.S.R., 1977, 135. 369 R. Burnham, Appl. Phys. Letters, 1977, 30, 132. 360 A. A. Isaev, M. A. Kazaryan, G. G. Petrash, and V. M. Cherezov, Kvantovaia Elektronika, 362

1976, 3, 1802.

L. F. Phillips, J. Photochem., 1976, 5, 241, 277. 362 P. H. Wine and L. A. Melton, Chem. Phys. Letters, 1977, 45, 509. s6s H. Itoh, H. Uchiki, and M. Matsuoka, Optics Commun., 1976, 18, 271. s64 R. H. Callender, J. I. Gersten, R. W. Leigh, and J. L. Long, Phys. Rev. (A), 1976, 14, 1672. 366 A. C. Tam and W. Hopper, J. Chem. Phys., 1976, 64,4337. 366 D. L. Feldman and R. N. Zare, Chem. Phys., 1976, 15,415. 967 S. Lemont, R. Giniger, and G. W. Flynn, J. Chem. Phys., 1977, 66, 4509. 36a K. Sakurai and H. P. Broida, J. Chem. Phys., 1976, 65, 1138. 369 M. Allegrini, L. Moi, and E. Arimondo, Chem. Phys. Letters, 1977, 45, 245. 370 N. Furuta, E. Yoshimura, Y. Nemoto, H. Haraguchi, and K. Fuwa, Chem. Letters, 1976,539. 371 W. J. Balfour and H. M. Cartwright, Canad. J. Phys., 1976, 54, 1898. 372 H. Haraguchi, W. K. Fowler, D. J. Johnson, and J. D. Winefordner, Spectrochim. Acta, 1976, 32, 1539. 373 P. Martin and J. Campos, J. Phys. (B), 1977, 10, 1265. 374 5. Dufayard and 0. Nedebec, J. de Physique, 1977, 38, 449. 375 R. W. Jones and J. L. Gole, Chem. Phys., 1977, 20, 311. 376 J. Lochet, J. Phys. (B), 1977, 10, 277. 377 J. A. Coxon, J. Photochem., 1976, 5, 337. 378 C. Linton and H. P. Broida, J. Mol. Spectroscopy, 1977, 64, 382, 389. 37g L. E. Berg, L. Klynning, and H. Martin, Optics Commun., 1976, 17, 320. w0 C. L. Chalek and J. L. Gole, J. Chem. Phys., 1976, 65, 2845. 381 S. J. Davis and S. G. Hadley, Phys. Rev. (A), 1976, 14, 1146. 383 M. Aslam Khan, M. Rafi, and S . J. A. Hussaine, J . Phys. (B), 1976, 9, 1953. 383 G. Gouedard and J. C. Lehmann, J. Phys. (B), 1976, 9, 2113; D. J. Greenwood and R. F. Barrow, ibid., p. 2123. 884 M. Kawasaki, H. Litvak, and R. Bersohn, J. Chem. Phys., 1977, 66, 1434. D. K. Russell, M. Kroll, and R. A. Beaudet, J. Chem. Phys., 1977, 66, 1999. s86 D. Golomb and J. H. Brown, Combustion and Flame, 1976, 27, 383. s87 A. Andreoni and H. Bucher, Chem. Phys. Letters, 1976, 40, 237; P. Benneti, R. Cubeddu, C. A. Sacchi, 0. Svelto, and F. Zaraga, ibid., p. 240.

Gas-phase Photoprocesses

175

at zero pressure was estimated to be 500 ns, with a self-quenching rate constant of 5.7 x 10-l2 cm3molecule-l s-l. Laser-induced dielectric breakdown of VF6 using the C 0 2 laser has been In the direct 147 nm photolysis of dimethylsilane, ten retrievable products plus a polymer are formed, and eleven primary processes are said to be necessary to account for these results.38gThe main processes involve molecular elimination, and although 147 nm excitation corresponds predominantly to an Si-H excitation, Si-C bond scission is the main result. Intermediates such as SiH2, SiHCH3, Si(CH,), are proposed to insert into Si-H bonds. The mercurysensitized photolysis of pentamethyldisilane and symmetrical tetramethyldisilane has, also been The photolysis of GeH31 at wavelengths longer than 200 nm produces I(2Pt) and I(2P+)in the ratio 1.3.3a1 The rate constant for deactivation of I(2P+)by GeH31 is large (6.9 f 1.0 x 10-l2cm3molecule-l s-l), which precludes the use of this material as the basis of the iodine photochemical laser. Single vibronic level fluorescence of Cr02C12, using a tunable cw laser as excitation source, has been A novel double-excitation method has been used to study vibrational depopulation kinetics in ground-state molecules, using electronic excitation of fluorescence as a probe.3Q8An ultra-short i.r. pulse is used to excite to some specified vibrational level of the ground state, and the relaxation of this is probed by a delayed pulse which excites to a fluorescent electronically excited state. Using this technique, which has wide applicability, the lifetime of an overtone at 5950 cm-l in the substituted coumarin (4) was found to be 4 f 1 ps at a vapour pressure of 1 Torr.

The ps fluorescence spectroscopy has also been carried out in a very novel The method is fashion using a cw laser excitation and radio essentially a phase-shift technique which, however, uses a null-type measurement. It has been used with success on Rhodamine B and Rhodamine 6G. The photodecomposition of POPOP type dyes (used in dye lasers) in the vapour phase has been investigated.3s5 388 380 391 39a

383 384

A. M. Ronn and B. L. Earl, Chem. Phys. Letters, 1977, 45, 556. A. G. Alexander and 0. P. Strausz, J. Phys. Chem., 1976, 80, 2531. I. N. Jung and W. P. Weber, J. Organometallic Chem., 1976, 114, 257. R. 5. Donovan, C. Fotakis, and H. M. Gillespie, J . Photochem., 1977, 6, 193. R. N. Dixon and C. R. Webster, J. Mol. Spectroscopy, 1976, 62, 271. J. P. Maier, A. Seilmeier, A. Laubereau, and W. Kaiser, Chem. Phys. Letters, 1977, 46, 527. Z. D. Popovic and E. R. Menzel, Chem. Phys. Letters, 1977, 45, 537. P. F. Liao, P. W. Smith, and P. J. Maloney, Optics Commun., 1976, 17, 219.

176

Photochemistry

10 1.r. Laser-induced Reactions, and Isotope Separation This burgeoning field can only be briefly reviewed here. The goal of efficient isotope separation has prompted much effort in this area, and several reviews have a p p e a ~ e d . ~ ~The * - ~multiphoton ~~ i.r. selective photoexcitation of species is economically attractive, and because of the availability of appropriate output wavelengths from the C 0 2 laser, SF6 has often been used as a model substance for such photodecomposition. Further work on SF, has appeared,406-411and some progress towards a theoretical understanding of the processes involved has been made.412-41sLaser-induced dielectric breakdown in the hugely commercially important UF, has been and reaction (90) found to be of importance. 2UFo

+ H2

-

2UF5

+ 2HF

(90)

Laser-induced isotope enrichment in rare-gas matrices 420 and narrow-line Doppler-free two-photon excitations which might have importance for selective excitation 421 have been discussed. Among specific isotope separations investigated, that of deu terium/hydrogen seems not to have made much progress. Thus ND3 photolysis163 has been rejected for this purpose, and formaldehyde photolysis suffers from complications due to H-atom formation in a primary process.74 Carbon isotope separation has been attempted by the photocatalysed exchange between phosgene and CO, reaction (91).24* Theoretically, the reaction should

12coc1,+ 13co 3~ 387 398

39g

4n0 4n1

4n2 4n3 404

4n5 4n6 407

408 409

410

412

413 414

415

m 417 41*

419

42n

421

-

13coc1, + 12co

(91)

R. V. Ambartsumyan and V. S. Letokhov, Accounts Chem. Res., 1977, 10, 61. R. N. Zare, ScientiJic American, 1977, 236, 86. R. V. Ambartsumyan, Yu. A. Gorokhov, S. L. Grigorovich, V. S. Letokhov, G. N. Makarov, Yu. A. Malinin, A. A. Puretsky, E. P. Filippov, and N. P. Furzikov, Kvantovaia Elektronika, 1977, 4, 171. N. G. Basov, I. S. Datskevich, V. S. Zuer, L. D. Mikheev, A. V. Startsev, and A. P. Shigorin, Kvantovaia Elektronika, 1977, 4, 638. V. S. Letokhov, Physics Today, 1977, 30, 23. 5. Tardieu de Maleissye, La Recherche, 1977, 8, 376. C. T. Lin, Spectroscopy Letters, 1976, 9, 615. G. Richards, Nature, 1977, 266, 303. J. Wolfrum, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 114.. J. D. Campbell, G. Hancock, and K. H. Welge, Chem. Phys. Letters, 1976, 43, 581. B. J. Orr and M. V. Keentock, Chem. Phys. Letters, 1976, 41, 68. J. Stone, M. F. Goodman, and D. A. Daws, Chem. Phys. Letters, 1976, 44, 411. D. R. Keefer, J. E. Allen, jun., and W. B. Person, Chem. Phys. Letters, 1976, 43, 394. D. S. Frankel, jun., J. Chem. Phys., 1976, 65, 1696. R. V. Ambartsumian, Y.A. Gorokhov, V. S. Letokhov, G. N. Makarov, and A. A. Puretsky, Zhur. eksp. teor. Fiz., 1976, 71, 440. M. J. Coggiola, P. A. Schulz, Y. T. Lee, and Y. R. Shen, Phys. Rev. Letters, 1977, 38, 17. C. D. Cantrell and H. W. Galbraith, Optics Commun., 1976, 18, 513; R. V. Ambartsumian, N. P. Furzikov, Y. A. Gorokhov, V. S . Letokhov, G. N. Makarov, and A. A. Puretzky, ibid., p. 517. J. Stone, M. F. Goodman, and D. A. Dows, J. Chem. Phys., 1976,65, 5062. M. F. Goodman, J. Stone, and D. A. DOWS,J. Chem. Phys., 1976, 65, 5052. Yu. I. Ponomaryov and M. R. Rasovsky, Optika i Spektroskopiya, 1976, 41, 545. K. C. Kim and J. M. McAfee, Chem. Phys. Letters, 1977, 45, 235. S. H. Bauer and K.-R. Chien, Chem. Phys. Letters, 1977, 45, 529. S. Speiser and J. Jortner, Chem. Phys. Letters, 1976, 44, 399. S. Mukamel and J. Jortner, Chem. Phys. Letters, 1976, 40, 150. B. Dellinger, D. S. King, R. M. Hochstrasser, and A. B. Smith, J. Amer. Chem. Soc., 1977, 99, 3197. J. C. Gamson, T. H. Einwohner, and J. Wong, Phys. Rev. (A), 1976, 14, 731.

177

Gas-phase Photoprocesses

have separation factors of 1.04 at 60 "C and 1.07 at - 35 "C, but upon Hg 253.7 nm irradiation, the 13Cseparation ratio was 1.092 at 60 "C and 1.17 at - 35 "C. The differences between theoretical and experimental results are accounted for by differences in the 12COC1, and 13COC12predissociation spectra. Several studies on boron isotope separation, using catalytic 422 methods, trimerization of tetrachloroethylene sensitized by BC13,271the exchange reaction of BCI3 with B(Me)2,423and the laser-augmented decomposition of D3BPF3424 have appeared. The last one used 930-950cm-l radiation, and showed that 4 x lo4 photons were required in absorption per molecule decomposed. The measured activation energy was 3.5 f 1 kcal mo1-1 compared with 29.3 kcal mol-1 for the purely thermal reaction, indicating the differencesthat excitation of a specific vibrational mode impose upon any system. Selective excitation of 0-1, 224 and the isotopically selective Br + Br, atomexchange reaction lgaahave been mentioned earlier. The C0,-laser-induced reaction of PF, + SO, has been The i.r. laser-induced isomerization (92) has been suggested as a model reaction for isotope Thus for the labelled compounds shown in (93), the b

G+yj D

h

D

(93)

ratio (5)/(6)imtd/(5)/(6)inf~*~~was > 1 for all irradiations. Note that only (6) absorbs the output from the CO, laser. If a non-degenerate system were used, chemical trapping of a specifically enriched species should be possible. Other electrocyclic reactions induced by multiphoton i.r. absorption have been investigated.427 Other i.r. laser-driven photodecompositions include hydrocarbons 428 and specifically ethane,lo cyclopropane,ll methyl halides,254alkyl halides,266CCIFs and CC13F,256ethylene,13-16 cycbC4~&12and cis- and trans-l,2-dichloroethylene.270In the last study, at a power density of > 5 x lo8W ern-,, three 4aa 428

424 426

426

4a7 428

C. T. Lin, T. D. Z. Atvars, and F. B. T. Pessine, J . Appl. Phys., 1977,48, 1720. F. Bachmann, H. Noth, R. Rinck, W. Fuss, and K. L. Kampa, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 313. K.-R. Chien and S. H. Bauer, J . Phys. Chem., 1976, 80, 1405. A. V. Eletskii, V. D. Klimov, and V. A. Legasov, High Energy Chemistry, 1976,10, 110. I. Glatt and A. Yogev, J. Amer. Chem. SOC.,1976, 98, 7087. A. Yogev and R. M. J. Benmair, Chem. Phys. Letters, 1977,46, 290. J. Tardieu de Maleissye, F. Lempereur, and C. Marsal, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 235.

178

Photochemistry

processes were observable : (i) selective (e.g. trans -+ cis) isomerization, (ii) selective molecular dissociation to give unexcited fragments ultimately yielding acetylene, (iii) formation of CH* and C,* electronically excited species. Conditions of laser power and wavelength could be chosen to optimize any of these three possibilities. The laser-enhanced reactions between 03(001) and NO 429 and NO(v = 1) and O3430s 431 have also been studied. 11 Atmospheric Reactions This section has of necessity been condensed even further in this Volume.

Extraterrestrial Phenomena.-Chemistry in interstellar the far-u.v. interstellar radiation field,433 the photochemistry of excited species,434 ion reactions with atomic oxygen and nitrogen of astrophysical the the chemistry of silicon in interstellar photosynthetic origins of H, and the formation of N H in interstellar regions, or rather the absence of formation by two-body radiative recombination between N and H have been reported. A review of the atmospheres of outer planets has appeared.438 The abundance of ethane and acetylene in the atmosphere of and the photochemistry and PH3441has been discussed. The in the Jovian atmosphere of NH34401441 photochemistry of N, in the Martian atmosphere 442 and the 0, dayglow emission from Mars and the abundance of Martian ozone443have been described. A series of reports by many authors has thoroughly reviewed the Martian A photochemical model for the escape of H, in an atmosphere with low O2 levels has been Upper Terrestrial Atmosphere.-The chemical response of the Earth’s middle atmosphere to changes in the U.V. solar flux 447 and photochemical deviations of thermospheric 0 and 0, densities from diffusive equilibrium448have been investigated. The photochemistry of doubly charged 0 ions in the thermosphere 4469

429

430

431 4S2 433

434 436 436

437 438

43g 44O

44l 442

443

444

445 445

447 448

S. M. Freund and J. C. Stephenson, Chem. Phys. Letters, 1976, 41, 157. J. C. Stephenson and S. M. Freund, J . Chem. Phys., 1976, 65, 1893. J. C. Stephenson and S. M. Freund, J . Chem. Phys., 1976, 65,4303. M. Taube, Chimiu, 1977, 31, 131. R. C. Henry, Astrophys. J., 1977, 212, 947. M. N. Vlasov, J. Atmos. Terrestrial Phys., 1976, 38, 807. F. C. Fehsenfeld, Astrophys. J., 1976, 209, 638. E. Greenbaum, Science, 1977, 196, 879. J. L. Turner and A. Dalgarno, Astrophys. J., 1977, 213, 386. G . E. Hunt, Ado. Physics, 1976, 25, 455. A. Tokunaga, R. F. Knacke, and T. Owen, Astrophys. J., 1976,209, 294. S. S. Prasad and L. A. Capone, J. Geophys. Res., 1976, 81, 5596. D. F. Strobel, Astrophys. J., 1977, 214, L97. Y . L. Yung, D. F. Strobel, T. Y . Kong, and M. B. McElroy, Icarus, 1977,30,26. J. F. Noxon, W. A. Traub, N . P. Carieton, and P. Connes, Astrophys. J., 1976, 207, 1025. Science, 1976, 193, 759-812. G . Visconti, J. Atmos. Sci., 1977, 34, 193. J. E. Frederick, Planetary Space Sci., 1977, 25, 1. J. E. Frederick, J. Geophys. Res., 1976, 81, 3179. E. Oran and D. Strobel, J. Geophys. Res., 1977, 82, 696.

Gas-phase Photoprocesses

179

and twilight emissions from odd N in the thermosphere449were also reported. Auroral NO concentrations have been The stability A review has appeared on the chemistry of the and possible destruction of the stratospheric 0, layer continue to exercise the talents of many groups.451aThe analysis of 0, world wide for possible trends is an area452crucial to the understanding of possible effects of man-made interferences with Nature, as is the diurnal U.V. photodissociation rate of stratospheric species such as 02,03.463 The destruction of O3 by Man's activities 454-455a such as nuclear weapons testing,466stratospheric flight,467and the use of fertilizers 458, 459 has been considered. The most recent cause for concern, halocarbons, are discussed in the next section. Modelling of 0, concentration^,^^^^ 460a the a correlation between relationship between O3 in W. Europe and O3 concentrations and Aitken nuclei the distribution of 0 atoms in the Earth's lower the sensitivity of O3 and NO2 stratospheric concentrations to surface temperature and atmospheric and the effects of nearby supernovae explosions on atmospheric O3 concentrations 465 have been commented upon. The altitude distribution of 0, and the OH Meinel emissions in the nightglow 466 and the fate of OH(v" G 9) in the stratosphere 467 have been discussed. Balmer and series fluorescence from photodissociation fragments of H2 and 02,468 resonance fluorescence from molecular H, 469 has been recorded. Halocarbons and 03.-The possibility of the production of chlorine atoms in the stratosphere by photochemical means,47owith the subsequent loss in ozone and competing chlorine-loss processes, has produced much kinetic data in the past year, and possible alternative tropospheric loss mechanisms for halocarbons, M. R. Torr, D. G. Torr, W. B. Hanson, J. H. Hoffmann, J. C. G. Walker, and A. 0. Nier, J. Geophys. Res., 1977, 82, 1008. 449a D. F. Strobel, E. S. Oran, and P. D. Feldman, J. Geophys. Res., 1976, 81, 3745. 4 5 0 E. Hyman, D. J. Strickland, P. S. Julienne, and D. F. Strobel, J. Geophys. Rev., 1976, 81,4765. 461 B. A. Thrush, Endeavour, 1977, 1, 3. 4510 R. S. Scorer, Atmos. Environment, 1977, 11, 277; C. Norman and C. Sherwell, Nature, 1976, 263, 268. 45a W. J. Hill, P. N . Sheldon, and J. J. Tiede, Geophys. Res. Letters, 1977, 4, 21. R. D. Rundel, J. Atmos. Sci., 1977, 34, 639. 464 P. Fabian, Naturwiss., 1976, 63, 273. 455 M. A. A. Clyne, Nature, 1976, 263, 723. S. Zanni, Rivista di Meterologia Aeronautica, 1976, 36, 319. 456 A. D. Christie, J. Geophys. Res., 1976, 81, 2583. 467 A. Ferrari and T. Tirabussi, Rivista di Meterologia Aeronautica, 1976, 36, 323. 468 H. S. Johnston, Science, 1977, 195, 1280. 469 H. S. Johnston, J. Geophys. Res., 1977, 82, 1767. 4 B 0 R. S. Harwood and J. A. Pyle, Quart. J. Roy. Meteorological Soc., 1977, 103, 319. 460a A. 5. Krueger and R. A. Miruner, J. Geophys. Res., 1976, 81, 4477. 461 R. Guicherit and H. van Dop, A m o s . Environment, 1977, 11, 145. 46a J. Podzimek, Pure Appl. Geophysics, 1976, 114, 925. 46a J. A. Logan and M. B. McElroy, Planetary Space Sci.,1977, 25, 117. *04 V. Ramanathan, L. B. Callis, and R. E. Boughnev, J. Atmos. Sci., 1976, 33, 1092. 466 R. C. Whitten, J. Cuzzi, W. J. Borucki, and J. H. Wolfe, Nature, 1976, 263, 398. 466 T. Watanabe, Y. Morioka, and M. Nakamura, Rep. Ionosphere Space Research Japan, 1976, 30, 41 ;T. Watanabe and T. Tohmatsu, ibid., p. 47. 467 G. E. Streit, G . Z. Whitten, and H. S. Johnston, Geophys. Res. Letters, 1976, 3, 521. 46* L. C. Lee and D. L. Judge, Phys. Rev. (A), 1976,14, 1094. 489 E. N. Alexandrov, S. M. Temchin, and L. P. Shishatskaya, Doklady Akad. Nauk S.S.S.R., 1977,233, 392. 470 K. Ya. Kondratiev and D. V. Pozdnyakov, Fiz. Atmosfery i Okeana, 1976, 12, 683. 4~

Photochemistry

180

including Freons, have also been considered. Halocarbon concentration measurements have been made in Alaska,471the British Isles,472 Tokyo,473a and and the predicted future levels of CCI,F (Fluorocarbon 11) for various hypothetical tropospheric removal rates Halocarbon behaviour has been further considered,476in particular the atmospheric diffusion and reactions of CHFCI, and CHF2C1,477 the reactions of Fluorocarbon 11 with atmospheric-like ions,267$ 478, 479 a comparison of the stratospheric reactivities of Fluorocarbons 11, 12,21, and 2Z,480and possible removal processes for F r e o n ~ , ~ ~ ~ 483 and OH radical^.^^^-^^^ The including reactions with O(l0) atoms latter reactions would result in reduced tropospheric OH concentrations, and there is some evidence for Clearly the absorption cross-sections for the halocarbons is of importance in determining their rate of stratospheric decomposition and there have been many including the temperature dependence.263 recent measurements of The photodecomposition of tropospheric CCI, was shown to be a negligible fate of this species.251 Production of C1 atoms in the stratosphere initiates reaction (94), and the 26Ks

Cl

+ 0,

48p9

-

C10

+ O2

(94)

importance of this reaction has prompted a reinvestigation of the Cl,/O, s y ~ t e r n489 , ~ with ~ ~ ~the possible roles of ClO*, ClNO,, and HOCl The absorption cross-section of C10 is thus of importance here, and has been reinvestigated in 246 Rate constants for important C1 atom The rate constant for the and C10 radical reactions are given in Table 9.4Q2-495 490p

l'4 472

47s

473a 474 476 476

478

479

480

481

48a

483

484 48s

486

488

490

402 493

4D5

4919

3779

E. Robinson, R. A. Rasmussen, J. Krasnec, D. Pierotti, and M. Jakubovic, Atmos. Enoironment, 1977, 11, 215. F. J. Sandalls and D. B. Hatton, Atmos. Environment, 1977, 11, 321. M. De Bortoli and E. Pecchio, Atmos. Enoironmenr, 1976, 10, 921. T. Ohta, M. Morita, and I. Mizoguchi, Atmos. Environment, 1976, 10, 557. P. W. Krey, R. J. Lagomarsino, and L. E. Tooknel, J. Geophys. Res., 1977, 82, 1753. F. S. Rowland and M. J. Molina, J . Phys. Chem., 1976, 80, 2049. D. H. Pack, J. E. Lovelock, G. Cotton, and C. Curthoys, Atmos. Environment, 1977,11, 329. C. Seigneur, H. Caram, and R. W. Carr, jun., Atmos. Environment, 1977, 11, 205. F. C. Fehsenfeld, D. L. Albritton, G. E. Streit, J. Davidson, C. 5. Howard, E. E. Ferguson, and R. G. Hirsch, Atmos. Enoironment, 1977, 11, 283. F. C. Fehsenfeld and D. L. Albritton, Geophys. Res. Letters, 1977, 4, 61; M. J. Campbell, ibid., p. 64. D. E. Robbins and R. S. Stolarski, Geophys. Res. Letters, 1976, 3, 603. R. J. Gelinas, D. K. Hall, and R. G. Nelson, Nature, 1977, 266, 229. R. Atkinson, G. M. Breuer, J. N. Pitts, jun., and H. L. Sandoval, J. Geophys. Res., 1976, 81, 5765. H. M. Gillespie, J. Ganaway, and R. J. Donovan, J. Photochem., 1977, 7, 29. C. J. Howard and K. M. Evenson, J. Chem. Phys., 1976, 64,4303. D. D. Davis, G. Machado, B. Conaway, Y.Oh, and R. Watson, J. Chem. Phys., 1976,65,1268. R. T. Watson, G. Machado, B. Conaway, S. Wagner, and D. D. Davis, J. Phys. Chem., 1977, 81, 256. H. B. Singh, Geophys. Res. Letters, 1977, 4, 101. R. D. Rundel and R. S. Stolarski, J. Geophys. Res., 1976, 81, 5759. S. S, Prasad, Planetary Space Sci., 1976, 24, 1187. J. A. Coxon and D. A. Ramsay, Canad. J. Phys., 1976,54, 1034; J. A. Coxon, W. E. Jones, and E. G. Skolnik, ibid., p. 1043. M. Mandelman and R. W. Nicholls, J. Quant. Spectroscopy Radiative Transfer, 1977, 17, 483; D. M. Cooper, ibid., p. 543. C. Park, J. Phys. Chem., 1977, 81, 499. M. S. Zahniser and F. Kaufman, J. Chem. Phys., 1977, 66, 3673. M. A. A. Clyne and W. S. Nip, J.C.S. Faraday I, 1976, 72, 2211. M. T. Leu, C. L. Lin, and W. B. De More, J. Phys. Chem., 1977, 81, 190.

181

Gas-phase Photoprocesses

Table 9 Rate constants for reactions of CI(,P) and C10 of stratospheric importance Radical CI

Substrate

c1

03

0 3

c1 c1

0 3 0 3

Cl

CH4

Cl

CH4

c1

CH*

Cl

CHI

c1

c10

CH* 0

CIO

0

c10

NO

k/cm3molecule-l s-l (log k = - 10.286 - 181/T) 1.35 x 10-l1 1.3 x 10-l1 3.1 x 10-l1 x exp (- 576/RT) 7.93 x 30-l2 x exp (-2530/RT) a 18.4 x 10-12 x exp (- 1545/T) 6.51 x lo-', x exp (- 1229/T) 7.94 x 10-12 x exp (-2437/RT) 1.2 x 10-13 3.38 x 10-l1 x exp (75/T) 10.7 x 10-l' x exp (-224/T) 1.13 x 10-l'

-

Ref. 194

1100 295 218-350

492

2 18-322

190

299-500

191

200-299

191

218-401

192

295

193 493

193 192

a

x exp(200/T) C10( M) NO2 2.66 x x exp (1140/T) (M = He) C10( M) NO2 1.15 x (M = Ar) Activation energy in cal mol-l. In units of cm6molecule-a s-l.

+ +

TemperaturelK

I

298

494 493 495 495

reaction of C1 atoms with methane over the temperature range 200-500 K does not obey a singleArr henius expression, an observation which could account for differing results obtained by different groups.191 The reaction between chlorine atoms and acetylene and the possible stratospheric significance of this have been commented upon,196and the reaction of O(l0) with C1, has been investigated.Ig6 The effect of H,O vapour on C10, and NO, reactions with 03,497 C10, + ClNO, reaction^,^^^ and reactions of chlorine nitrate4Q9have been discussed. NOCl is said to be less of a chlorine atom sink than previously thought, since measured absorption cross-sections for this species are greater than measured previously.6ooThe U.V. absorption of C10N0601and the chemical degradation of C10N02 in the stratosphere602have also been discussed. The high-resolution i.r. absorption spectrum of chlorine nitrate has been Tropospheric Reactions and Photochemical Smog.-A survey of kinetic data for atmospheric reactions has appeared,604and further studies on the kinetics of the 4B6

4s7

I. S. Fletcher and D. Husain, Ber. Bunsengesellschaft Phys. Chem., 1976, 80, 982. S. C. Lin, T. M. Donahue, R. J. Cicerone, and W. L. Chameides, J. Geophys. Res., 1976, 81, 3111. J. W. Birks, B. Schoemaker, T. J. Leck, R. A. Borders, and L. J. Hart, J. Chem. Phys., 1977, 66,4591.

4B9

601

F. S. Rowland, J. E. Spencer, and M. J. Molina, J. Phys. Chem., 1976, 80, 2711,2713. A. J. Illies and G. A. Takacs, J. Photochem., 1976, 6 , 35. L. T. Molina and M. J. Molina, Geophys. Res. Letters, 1977, 4, 83. D. D. Davis, G. Smith, G . Tesi, and J. Spencer, Geophys. Res. Letters, 1977, 4, 7. R. A. Graham, E. C. Tuazon, A. M. Winer, J. N. Pitts, jun., L. T. Molina, L. Beaman, and M . J. Molina, Geophys. Res. Letters, 1977, 4, 3. L. G. Anderson, Rev. Geophys. Space Phys., 1976,14, 151.

182

Photochemistry

photochemistry of the urban troposphere have been A gas cell for such studies606 and a smog-chamber study of reductions of NO2 concentration due to night-time reactions 507 have been the subjects of recent reports. A correlation has been drawn between emission and photochemistry of formaldehyde in urban areas.608 The effect of latitude on the potential for smog formation,509seasonal variations in tropospheric ozone,51oand ozone measurements in Antarctica,611 Los A n g e l e ~ and , ~ ~ St. ~ Louis 516 have been London,612,513 Britain reported. Models for air-pollutant transport have been developed,617,618 and the effect of CO on the atmospheric CO/OH/CH, cycle has been Problems relating to sulphur dioxide emission and aerosol formation have also been aired.520--623 Detection of Atmospheric Constituents and Pollutants.--LIDAR methods have been developed to measure ozone,524 N02,525HCl, CH4 and N20,526and A laboratory study of the potential of resonance Raman The possibility of using scattering for a LIDAR system has been carried Doppler-free two-photon spectroscopy for monitoring stratospheric gases has been a n a l y ~ e dand , ~ ~the ~ use of atomic and molecular fluorescence in this regard 5 3 2 a Mesospheric 0, can be measured from ground-based has also been mm-wave and stratospheric O3 and aerosols from orbiting photoelectric Rocket-borne 535 and balloon-borne 536 chemiluminescence detection of NO in the upper atmosphere has been achieved. T. E. Graedel, L. A. Farrow, and T. A. Weber, Atmos. Environment, 1976, 10, 1095. A. E. Ledford, jun. and W. Braun, Rev. Sci. I., 1977,48, 537. 607 R. M. Kamens, H. E. Jeffries, D. L. Fox, and L. Alexander, Atmos. Environment, 1977,11,225. W. S . Cleveland, T. E. Graedel, and B. Kleiner, Atmos. Environment, 1977, 11, 357. H. Nieboer, W. P. L. Carter, A. C. Lloyd, and J. N. Pitts, jun., Atmos. Environment, 1976,10,731. 610 P. Fabian and P. G. Pruchniewicz, J. Geophys. Res., 1977, 82, 2063. 611 S. J. Oltmans and W. J. Komhyr, J. Geophys. Res., 1976, 81, 5359. D. J. Ball, Nature, 1976, 263, 580. 613 H.N. M. Stewart, E. J. Sullivan, and M. L. Williams, Nature, 1976, 263, 582. 614 P. D. Moore, Nature, 1976, 263, 546. m J. T. Peterson and K. L. Demerjian, Atrnos. Environment, 1976, 10, 459. 618 W. H. White, Science, 1976, 194, 187. 617 D. L. Ermak, Atmos. Environment, 1977, 11, 231. 618 W. L. Chameides and D. H. Stedman, J. Geophys. Res., 1977, 82, 1787. 618 N. D. See, Science, 1977, 195, 673. 6 2 0 S. L. Heisler and S. K. Friedlander, Atmos. Environment, 1977, 11, 157. 6z1 A. B. Harker, L. W. Richards, and W. E. Clark, Atmos. Environment, 1977, 11, 87. 62a W. H. White, Nature, 1976, 264,735. m3 C.S. Kiang and P. Middleton, Ceophys. Res. Letters, 1977, 4, 17. 6 2 4 0. K. Kostko, N. D. Smirnov, and V. V. Fadeev, Kvantovaia Elektronika, 1976, 3, 2392. 626 T. Tsuji, H. Kimura, Y. Higuchi, and K. Goto, Japan J. Appl. Phys., 1976, 15, 1743. E. R. Murray, J. E. Vanderlaan, and J. G. Hawley, Appl. Optics, 1976, 15, 3140. 637 Y. Iwasaka, A. Mita, and K. Isono, Rep. Ionosphere Space Res. Japan, 1976, 30, 51. D. A. Ross and W. L. Kuriger, Appl. Optics, 1976, 15, 2958. 629 R. Dos Santos and W. H. Stevenson, Appl. Phys. Letters, 1977, 30,236. 630 J. G. Hochenbleicher, W. Kiefer, and J. Brandmuller, Appl. Spectroscopy, 1976, 30, 528. 631 J. Gelbwachs, Appl. Optics, 1976, 15, 2654. 632 K. Schofield, J. Quant. Spectroscopy Radiative Transfer, 1977, 17, 13. me C. L. Fincher, A. W. Tucker, M. Birnbaum, R. J. Paur, and W. A. McClenny, Appl. Optics, 606

633 634 635

638

1977, 16, 1359. H. Penfield, M. M. Litvak, C. A. Gottlieb, and A. E. Lilley, J. Geophys. Res., 1976, 81, 6115. G. L. Matloff, J. Astronautical Sci., 1976, 24, 365. C. J. Mason and J. J. Horwath, Geophys. Res. Letters, 1976, 3, 391. J. W. Drummond, J. M. Rosen, and D. J. Hofmann, Nature, 1977, 265, 319.

183

Gas-phase Photoprocesses

Trace amounts of atmospheric constituents have been detected using highresolution i.r. 563 O2 abundances in the thermosphere have been made from ion measurements,641and OH measurements 642 made by methods There including resonance absorption of sunlight 543 and laser have been various measurements of O3 concentrations 546-647 and SO2 c o n ~ e n t r a t i o n s649 .~~~~ 511-516e

H Atom, Ha, and H 2 0 Reactions.-Rate parameters have been reported for the reactions of hydrogen atoms with the following substrates : H atoms,550HF,551-553 DF,654C1,553Br,65396 5 5 and I,653rare gases and H2,656 02,557 NO,658NO 2, 659-664 039 65g NOCl,661C12,562~ 566-567 HCI and C1CF3,668N F2, 669 PH3, 570 C2H4, 571 MeCD,,572 C. M. Bradford, F. H. Murcray, J. W. Van Allen, 5. N. Brooks, D. G. Murcray, and A. Goldman, Geophys. Res. Letters, 1976, 3, 387. m 8 J. M. McAfee, E. R. Stephens, D. R. Fitz, and J. N. Pitts, jun., J. Quant. Spectroscopy Radiative Transfer, 1976, 16, 829. m9 R. J. Nordstrom, 5. H. Shaw, W. R. Skinner, W. H. Chan, J. G. Calvert, and W. M. Uselman, Appl. Spectroscopy, 1977, 31, 224. 640 W. J. Williams, J. J. Kosters, A. Goldman, and D. G. Murcray, Geophys. Res. Letters, 1976, 3, 383. M. Oppenheimer, A. Dalgarno, and H. C. Brinton, J. Geophys. Res., 1976, 81, 4678. 642 D. Perner, D. H. Ehhalt, H. W. Patz, V. Platt, E. P. Roth, and A. Volz, Geophys. Res. Letters, 1976, 3, 466. 643 C. R. Burnett, Geophys. Res. Letters, 1976, 3, 319. M. Hanabusa, C. C. Wang, S. Japar, D. K. Killinger, and W. Fisher, J. Chem. Phys., 1977, 66, 21 18. R. W. L. Thomas and A. C. Holland, Appl. Optics, 1977, 16, 613. 646 J. S . Randhawa, J. Photochem., 1976, 6, 147. 647 R. E. Basher, Quart. J. Roy. Meteorological SOC.,1976, 102, 667. 648 E. Sagawa and T. Itoh, Geophys. Res. Letters, 1977, 4, 29. 649 W. Jaeschke, R. Schmitt, and H. W. Georgii, Geophys. Res. Letters, 1976, 3, 517. 660 K. P. Lynch, T. C. Schwab, and J. V. Michael, Znternat. J. Chem. Kinetics, 1976, 8, 651. 6 6 1 J. F. Bott and R. F. Heidner, J. Chem. Phys., 1977, 66, 2878. R. F. Heidner and J. F. Bott, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 128. 663 M. D. Pattengill, J. C. Polanyi, and J. L. Schreiber, J.C.S. Faraduy IZ, 1976, 72, 897. 664 J. F. Bott, J. Chem. Phys., 1976, 65, 1976. stis H. Endo and G. P. Glass, J. Phys. Chem., 1976, 80, 1519. 666 F. Roussel, P. Pradel, and G. Spiess, Phys. Reu. (A), 1977, 15, 451. 657 J. R. Hislop and R. P. Wayne, J.C.S. Faruday ZI, 1977, 73, 506. 658 K. Oka, D. L. Singleton, and R. J. Cvetanovic, J. Chem. Phys., 1977, 66, 713. 669 M. A. A. Clyne and P. B. Monkhouse, J.C.S. Faraday ZZ, 1977, 73, 298. 660 J. E. Spencer and G. P. Glass, Chem. Phys., 1976, 15, 35. m1 H. G. Wagner, U. Welzbacher, and R. Zellner, Ber. Bunsengesellschaft Phys. Chem., 1976, 80, 1023. 6e2 P. P. Bemand and M. A. A. Clyne, J.C.S. Faraduy ZZ, 1977, 73, 394. c63 H. Haberland, W. von Lucadou, and P. Rohwer, Ber. Bunsengesellschaft Phys. Chem., 1977, 81, 150. m4 J. A. Silver, W. L. Dimpfl, J. H. Brophy, and J. L. Kinsey, J. Chem. Phys., 1976, 65, 1811. 666 D. G. Truhlar, J. A. Merrick, and 5. W. Duff, J. Amer. Chem. SOC.,1976, 98, 6771. 666 P. F. Ambridge, J. N. Bradley, and D. A. Whytock, J.C.S. Faraday I, 1976,72, 1157. 667 H. G. Wagner, U. Welzbacher, and R. Zellner, Ber. Bunsengesellschaft Phys. Chem., 1976, 80, 902. 668 P. F. Ambridge, J. N. Bradley, and D. A. Whytock, J.C.S. Faraday Z, 1976, 72, 2143; J. N. Bradley, D. A. Whytock, and T. A. Zaleski, ibid., p. 2284. J. I. Steinfeld, Ber. Bunsengesellschaft Phys. Chem., 1972, 81, 221. 670 J. H. Lee, J. V. Michael, W. A. Payne, D. A. Whytock, and L. J. Stief, J. Chem. Phys., 1976, 65, 3280. 671 G. Pratt and I. Veltman, J.C.S. Faraday I, 1976, 72, 1733. 672 G. D. Beverly and R. M. Martin, J. Phys. Chem., 1976, 80, 2063.

637

184

Photochemistry

p r ~ p y n e ,acetone ~ ~ ~ 574 and t h i ~ t a n e ,and ~ ~ ~MeONO and methyl hydroperoxide.576 Selected rate constants are given in Table 10. An H 2 0 i.r. laser pumped directly by electronic to vibrational energy transfer has been The reaction of H atoms with oxygen was shown to be a two-stage process, with HO, as an intermediate, this reacting with further hydrogen atoms to produce 02(1Cg+).657The efficiency of 02(1Zg+) formation relative to total H H02 reaction cross-section was 2.8 x

+

Table 10 Selected rate constant data for H atom reactions Subsirate CI2 c 1 2

NO2 NO2 0 3

NO(

k/cm3 molecule-l s-l k = -9.851 -250/T 2.2 x 10-1’ 1.1 x 10-10 log10 k = -9.319- 174/T loglo k = - 10.005 - 224/T loglo

+ M)

In units of cme

4.33 x

10-32 a

Temperai urel K

Ref.

300-750 300 298 298-653 298-638 298

562 562 562 559

559 558

s-l.

0 Atoms, 02,and 03.-Oscillator strengths for a variety of 0 atom transitions in the far U.V. have been given.578 Rate parameters for the reactions of O ( 3 P ) with the following substrates have been reported : O-,K7DO,( + 02),580 C0,581 NO( + M),582-584 N20and C02,585 HCl,492HBr,586C10N02,587OCS,588CS2,K89p 590 CS,589H2S,591n - b ~ t a n e ,ethylene,588 ~~~ olefins 592 including c y c l ~ p e n t a d i e n e , ~ ~ ~ tetrafluoroethylene,126methyla~etylene,~~~ MeCN and CF3CN,695a ~ e t o n e596 ,~~~~ and phosphorus t r i f l ~ o r i d e .Selected ~ ~ ~ rate constant data are given in Table 1 1 . Rate-constant data for the reactions of O(l0) with halocarbons 26K$ 598 has been reviewed in ref. 265, and will not be repeated here. Reaction of this species 4839

673 574 675 6i6

577 bi8

5iQ

680

681 682 58s

684

686 586

687

588 689

5a0

5Q1

6g2

6a3 6a4

6Q6

bae 6Q7

D. A. Whytock, W. A. Payne, and L. J. Stief, J. Chem. Phys., 1976, 65, 191. P. F. Ambridge, J. N. Bradley, and D. A. Whytock, J.C.S. Furaduy I , 1976, 72, 1870. 0. Horie, K. Kawamata, K. Onuki, and A. Amano, Chem. Letters, 1976, 753. G. K. Moortgat, F. Slemr, and P. Warneck, Internut. J. Chem. Kinetics, 1977,9,249; F. Slemr and P. Warneck, ibid., p. 267. A. B. Peterson, L. W. Braverman, and C. Wittig, J. Appl. Phys., 1977, 48, 230. M. A. A. Clyne and L. G. Piper, J.C.S. Faraday II, 1976, 72, 2178. H. C. Carlson and R. M. Harper, J. Geophys. Res., 1977, 82, 1144. L. G. Hogan and D. S. Burch, J. Chem. Phys., 1976, 65, 894. J. D. Kelley and R. L. Thommarson, J. Chem. Phys., 1977, 66, 1953. J. V. Michael, W. A. Payne, and D. A. Whytock, J. Chem. Phys., 1976, 65, 4830. R. Atkinson, R. A. Perry, and J. N. Pitts, jun., Chem. Phys. Letters, 1977, 47, 197. D. A. Whytock, J. V. Michael, and W. A. Payne, Chem. Phys. Letters, 1976, 42, 466. J. R. Wiesenfeld, Chem. Phys. Letters, 1977, 45, 384. J. E. Spencer and G. P. Glass, Internat. J. Chem. Kinetics, 1977, 9, 97. L. T. Molina, J. E. Spencer, and M. J. Molina, Chem. Phys. Letters, 1977, 45, 158. R. G. Manning, W. Braun, and M. J. Kurylo, J. Chem. Phys., 1976, 65, 2609. J. D. Kelley, Chem. Phys. Letters, 1976, 41, 7. R. E. Graham and D. Gutman, J. Phys. Chem., 1977, 81, 207. D. A. Whytock, R. B. Timmons, J. H. Lee, J. V. Michael, W. A. Payne, and L. J. Stief, J . Chem. Phys., 1976, 65, 2052. D. L. Singleton and R. J. Cvetanovic, J. Amer. Chem. Soc., 1976, 98, 6812. K. Nakamura and S. Koda, Internut. J. Chem. Kinetics, 1977, 9, 67. M. E. Umstead, R. G. Shortridge, and M. C. Lin, Chem. Phys., 1977, 20, 271. R. J. Bonanno, R. B. Tirnmons, L. J. Stief, and R. B. Klemm, J. Chem. Phys., 1977, 66, 92. 5. H. Lee and R. B. Timmons, Internat. J . Chem. Kinetics, 1977, 9, 133. I. B. Goldberg and H. R. Crowe, J . Phys. Chem., 1976, 80, 2407. I. S. Fletcher and D. Husain, J. Phys. Chem., 1976, 80, 1837.

185

Gas-phase Photoprocesses

with HCl,6nDNH3, N202, C2H6, C3H8, and CMe,,600 O,, N,, CO,, 03,and H20,601and CCl2O, CFCIO, and CF20s02have also been studied. The role of O(lS) in the lower and the rate constants for the deactivation of O(%) by H 2 0604 and N 2 0605 have been reported. Table 11 Selected rate constants for O(3P)atom reactions Substrate

HCI CIONO, NO(+ M) (M = He) NO(+ M) (M = Ar) NO(+ M) (M = N,) NO

n-Butane n-Butane a

k/cm3 rnolec~le-~ s-l 2 x 10-14 3.4 x ~ X P(- 840/T) 10.8 x exp (1040/RT) 14.6 x exp (940/RT) a, 15.5 x exp (1160/RT)a* 7 x 10-32 a 2.5 x exp (-4170/RT) 2.2 x 10-14

In units of cm6 molecule-l s-l.

Temperature/K 1100

Ref. 492 587 582 583 584 583 583 583

217-500 298439 217-500 298 298439 298

Activation energy in cal mol-'.

The photodecomposition of 0, in the Schumann-Runge continuum 606a and the efficiency of O(l0) atom production in the dissociative recombination of 02+ have been 02+ reactions are reported with N, NO+, and H20,e08 and cross-sections for photodetachment in 0 2 / C 0 2 / H 2 mixtures 0 given.6onA new nightglow band system in 0, has been reported.610 The formation of excited species in the quenching of o,PA,) 611 and electronic to vibrational energy transfer in the O,(lA,)/HF system 612 have been discussed. The quenching of O,(&lC,+) by 02,613, 614 H 616 N 2, 613 HBr,g16and other diatomics 618 has been studied quantitatively. The possible role of O,(l&-) in the photo-oxidation of benzene vapour 619 has been discussed. 606s

29

617s

M. C. Addison, R. J. Donovan, and H. M. Gillespie, Chem. Phys. Letters, 1976, 44, 602. I. S. Fletcher and D. Husain, Cunad. J. Chem., 1976, 54, 1765. 601 G . E. Streit, C. J. Howard, A. L. Schmeltekopf, J. A. Davidson, and H. I. Schiff, J. Chem. Phys., 1976, 65, 4761. 6 0 2 R. K. M. Jayanty, R. Simonaitis, and J. Heicklen, J. Photochem., 1976, 5 , 217. 603 T. G. Slanger and G. Black, Planetary Space Sci.,1977, 25, 79. 604 F. W. Bingham, A. Wayne Johnson, and J. K. Rice, J. Chem. Phys., 1976, 65, 1663. 606 T. G. Slanger and G. Black, J. Chem. Phys., 1976, 65, 2025. (06 D. C. Cartwright, N. A. Fiamengo, W. Williams, and S. Trajmar, J. Phys. (B), 1976, 9, L409. P. S. Julienne, J. Mol. Spectroscopy, 1976, 63, 60. L. L. Cogger, L. S. Smith, and R. M. Harper, Planetary Space Sci., 1977, 25, 105. 8 0 8 F. C. Fehsenfeld, Planetary Space Sci., 1977, 25, 195. 609 R. A. Beyer and J. A. Vanderhoff, J. Chem. Phys., 1976, 65,2313. P. C. Wraight, Nature, 1976, 263, 310. R. G. 0. Thomas and B. A. Thrush, Ber. Biinsengesellschaft Phys. Chem., 1977, 81, 177. S. Madronich, J. R. Wiesenfeld, and G. J. Wolga, Chem. Phys. Letters, 1977, 46, 267. as L. R. Martin, R. B. Cohen, and J. F. Schutz, Chem. Phys. Letters, 1976, 41, 394. S. A. Lawton, S. E. Novick, H. P. Broida, and A. V. Phelps, J. Chem. Phys., 1977, 66, 1381. M. Braithwaite, E. A. Ogryzlo, J. A. Davidson, and H. I. Schiff, J.C.S. Furaduy ZZ, 1976, 12, 2075. 616 M. Braithwaite, E. A. Ogryzlo, J. A. Davidson, and H. I. Schiff, Chem. Phys. Letters, 1976, 42, 158. M. Braithwaite, J. A. Davidson, and E. A. Ogryzlo, Ber. Bunsengesellschaff Phys. Chem., 1977, 81, 179. M. Braithwaite, J. A. Davidson, and E. A. Ogryzlo, J. Chem. Phys., 1976, 65, 771. T. F. Hunter, D. Rumbles, and M. G. Stock, Chem. Phys. Letters, 1977, 45, 145.

6oo

186

Photochemistry

Reactions between 0, and the following substrates have been reported : N0,429-431O(3P),620formaldehyde,621 cyanoacetylene, CS2, and thiophen,622 c i ~ - b u t - 2 - e n eother , ~ ~ ~ o l e f i n ~ and , ~ ~ tetrafluoroethylene.626 ~ The flash photolysis of 03,626 O3 spectroscopy with a CO, waveguide laser,627and MO calculations on Q 3 6 2 8 have also been discussed.

HO, Reactions.-Hydroxyl radical reactions with a wide variety of substrates have been studied, owing to the importance of the OH oxidation chains in polluted urban atmospheres. Selected rate-constant data for reactions of OH with the following substrates are given in Table 12: H2,629C0,629* 632 NO,629 NO,( + ~),633,634 HNO29 629, 635 N 207630, 636 CH 629, 6 3 l C2H6 9 631, 484 chlorinated ~' c p and C, and brominated me thane^,^^^, 485 F r e ~ n s n, ~- b~~~t a n e , ~CF3H,631 alkanes,638 halogen-substituted e t h a n e ~ , ~ ~ ~ 639-641 halogenated ~~* isopropyl alcohol, ethy1enes,'j4l c6 and C, a l k e n e ~ , aromatic diethyl, and di-n-propyl ethers,643ketones, chloroethanes and r n o n ~ t e r p e n e s , ~ ~ ~ and MeSH and MeNH2.645 In the study at 1300 K, relative rate constants for the OH reactions with H,, CO, CHI, CF3H, C2H4, and C2H6 were in the ratio 0.59, 0.18, 1.00, 0.19, 2.33, and 2,.88 respectively.631 In the reaction of OH with CzH4,the ratio of hydrogenabstraction product to addition product was found to be ~ 3 . 5 The . ~ rate ~ ~of reaction of OH with CO, in the kinetically uncommon but practically useful units of p.p.m.-l min-l, has been found to be 439 at 700 Torr, but only 203 at lOOTorr, implying pressure dependence.632 The rate is a factor of two larger 47

620

621 622 623 624

625 626

627

629

630 631

632 633

634

635

638

637 G38 G30 e40

641 642

643

644

645

G. A. West, R. E. Weston, jun., and G. W. Flynn, Chem. Phys. Letters, 1976, 42, 488. S. Braslavsky and J. Heicklen, Internat. J. Chem. Kinetics, 1976, 8, 801. S. Toby, F. S. Toby, and B. A. Kaduk, J. Photochem., 1977, 6,297. H. Niki, P. D. Maker, C. M. Savage, and L. P. Breitenbach, Chem. Phys. Letters, 1977,45,327. S . M. Japar, C. H. Wu, and H. Niki, J . Phys. Chem., 1976, 80, 2057. F. S. Toby and S. Toby, J. Phys. Chem., 1976, 80,2313. V. V. Timofeev, B. M. Popov, M. P. Povich, Y. N. Zhitnev, and Y. V. Filippov, 2hur.fiz. Khim., 1976, 50, 2144. R. T. Menzies, Appl. Optics, 1976, 15, 2597. R. P. Messmer and D. R. Salahub, J. Chem. Phys., 1976, 65, 779. R. A. Cox, R. G. Derwent, and P. M. Holt, J.C.S. Faraday Z, 1976, 72, 2031. R. Atkinson, R. A. Perry, and J. N. Pitts, jun., Chem. Phys. Letters, 1976, 44, 204. J. N. Bradley, W. D. Capey, R. W. Fair, and D. K. Pritchard, Internat. J. Chem. Kinetics, 1976, 8, 549. W. H. Chan, W. M. Uselman, J. G. Calvert, and J. H. Shaw, Chem. Phys. Letters, 1977, 45, 240. K. Erler, D. Field, R. Zellner, and I. W. M. Smith, Ber. Bunsengesellschuft Phys. Chem., 1977, 81, 22. C. Anastasi and I. W. M. Smith, J.C.S. Faraday IZ, 1976, 72, 1459. R. A. Fifer, J. Phys. Chem., 1976, 80, 2717. H. W. Biermann, C. Zetzsch, and F. Stuhl, Ber. Bunsengesellschaft Phys. Chem., 1976, 80, 909. R. A. Perry, R. Atkinson, and J. N. Pitts, jun., J. Chem. Phys., 1976, 64, 5414. K. R. Darnall, A. M. Winer, A. C. Lloyd, and J. N. Pitts, jun., Chem. Phys. Letters, 1976, 44,415. R. Atkinson, R. A. Perry, and J. N. Pitts, jun., J. Chem. Phys., 1977, 66, 1197. J. F. hleagher and J. Heicklen, J. Phys. Chem., 1976, 80, 1645. C. J. Howard,J. Chem. Phys., 1976, 65, 4771. R. A. Perry, R. Atkinson, and J. N. Pitts, jun., J. Phys. Chem., 1977, 81, 296. A. C. Lloyd, K. R. Darnall, A. M. Winer, and J. N. Pitts, jun., Chem. Phys. Letters, 1976, 42, 205. A. M. Winer, A. C. Lloyd, K. R. Darnall, and J. N. Pitts, jun.,J. Phys. Chem., 1976,80, 1635. R. Atkinson, R. A. Perry, and J. N. Pitts, jun., J. Chem. Phys., 1977, 66, 1578.

187

Gas-phase Photoprocesses

Table 12 Selected rate constants for OH radicaI reactions Substrate

co co NO2 (M) NO NZO HONO HONO n-Butane CH4 2,2,3-Trirnethylbutane CH,C1 CH2C1F CHClF, CF2CICFaCI Cyclohexene 1-Heptene Wl* a-Pinene t9-Pinene a-Limonene Methyl ethyl ketone Trichloroethane Isopropanol Diethyl ether

+

c 6 b

1,3,5-Trirnethylbenzene a

k/cm3 molecule-l s-l 1.54 x 10-13 2.7 x 10-13 3.8 x 10-17 1.2 x 10-l' (M = air, 760Torr) < 2 x 10-1s 6.6 x 10-l2 2.6 x 10-l2 2.72 x 10-l2 7.6 x 10-l6 3.83 x 10-l2 1.84 x 10-l2 exp (-2181lRT) a 2.84 x 10-l2 exp (- 1259/T) 9.25 x 10-13 exp (- 1575/T) 3 x 10-18 7.5 x 10-11 3.7 x 10-11 1 7 x 10-13 5.8 x 6.8 x 10-l1 1.5 x 10-10 3.3 x 10-12 4.5 x 10-12 7.2 x 10-l1 9.3 x 10-11 1.2 x 10-12 6.24 x 10-l'

TemperaturelK

Ref.

299 296 298 296 299 296 1000-1 400 297420 296 305 245-375

630 629 636 629 630 629 635 637 629 638 485 486 486 486 638 638 641 644 644 644 644 644 643 643 642 642

-

305 305

-

305 305 296-473 296-473

Activation energy in cal mol-'.

than previously accepted values. It is to be noted in Table 10 that halocarbons with abstractable hydrogens have much higher rates of reaction with OH than those which are completely halogenated (e.g. CF,ClCF,Cl). This implies that all but the latter variety should have very short tropospheric lifetimes. The intensity and pressure dependence of resonance fluorescence of OH induced by a tunable U.V. a potential OH U.V. laser,847OH measurements in a photochemical reactor using laser-induced the characterization of OH in some photochemical and some reactions of OH(v" - 1) 660 have been discussed in recent reports. Reactions between HO, and OH, and 0,'j6land between HO, and NO, to give pernitric acidsS2 (95) have been studied by the powerful techniques of laser magnetic resonance spectroscopy and Fourier transform spectroscopy respectively. HO,+NO, 646

647 g48

649

661 6sa

-

HOON02

(95)

D. K. Killinger, C. C. Wang, and M. Hanabusa, Phys. Rev. (A), 1976,13,2145. C. H. Chen and M. G. Payne, Optics Commun., 1976, 18, 476. C. H. Wu, C. C. Wang, S. M. Japar, L. I. Davis, M. Hanabusa, D. Killinger, H. Niki, and B. Weinstock, Internat. J . Chem. Kinetics, 1976, 8, 765. N. Jacob, I. Balakrishnan, and M. P. Reddy, J. Phys. Chem., 1977, 81, 17. J. E. Spencer and G. P. Glass,Internat. J. Chern. Kinetics, 1977, 9, 111. J. P. Burrows, G. W. Harris, and B. A. Thrush, Nature, 1977, 267, 233. H. Niki, P. D. Maker, C. M. Savage, and L. P. Breitenbach, Chem. Phys. Letters, 1977,45, 564.

188

Photochemistry

Nitrogen, NO,, and HNO, Reactions-It has been suggested that atomic nitrogen can be used as a probe of physical conditions in the interstellar medium.e63 The vibrational distribution of NO produced in reactions (96) and (97) has been N(45)

+

0 2

N(45) + O(3P)

-

->

NO+O

(96)

NO(Z4hC-)

(97)

p r ~ b e d and , ~ it~was ~ ~found ~ ~ that ~ CO, was twice, N 2 0 four times, and H 2 0 six times as effective as N, in removing N0(b4C-, u' = 2 or 3).655The quenching of N(2D) by a variety of species has been studied, that by N,O producing NO(B211,) (u' = O)."', 657 The yield of N(2D) atoms produced in the dissociative recombination of NO,658 fluorescence from N2+,N20+,669 and the reactions of 02+ with N and NO+ and NO2+ with 0 608 have been the subjects of recent reports. The collision-induced predissociation of N2+in the Z2Zg+,and states has also been described.6ao states of N2,661 laser spectroscopic studies on There have been reports on 5Zc,+ N2(B311g,u' = 13),662and the observation of this electronic state in the Lewisthe collisional deactivation of N2(E3C,+)to the Rayleigh afterglow of N2,663 lifetime measurements on N, lines below 500 and collisional quenching of 667 The deactivation rate of the state was found to be 1.2 x cm3molecule-' s-1.664 The values of the deactivation rate constants for N2(J3ZU+)by N, N,, and a Pyrex glass surface were 3.5 x s-1.667 lo-'' cm3atom-l s-l, 4.5 x 10-17 cm3molecule-l s-l, and 1.8 x The autoionization structure of NO near the threshold has been recorded,6s8 pressure-broadened linewidths in NO have been investigated,669and the atmospheric formation of NO from N2(J3Z,*) has been d i s c u ~ s e d .The ~ ~ ~effect of NO photolysis on NO, mixing ratios has been reported,671and NO and N20i.r. lasers have been proposed in which pumping is by direct electronic to vibrational energy u' = 0, 1, or 2 In the two-photon excitation of NO to the x2Z+, levels, lifetimes were found which varied from 672 174 -t 5 ns for the band-head of the d = 2 state to 220 ns for the K' = 12 level of u' = 0. Electronic quenching rate constants were also dependent on the level, varying from 1.38 x 10-lo cm3

x2rIu

c311,

N2(23&+).6669

653 654

666 657 668 65g

660

651

683 664 665 666 667 688

66g 670 671 67a

M. A. Dopita, D. J. Mason, and W. D. Robb, Astrophys. J., 1976, 267, 102. M. E. Whitson, jun., L. A. Darnton, and R. J. McNeal, Chem. Phys. Letters, 1976, 41, 552. I. M. Campbell and R. S. Mason, J . Photochem., 1976, 5 , 383. T. G. Slanger and G. Black, J. Chem. Phys., 1976, 64,4442. G. Black, R. L. Sharpless, and T. G. Slanger, J. Photochem., 1976, 5, 435. D. Kley, G. M. Lawrence, and E. J. Stone, J . Chem. Phys., 1977, 66,4157. A. L. Osherovich and V. N. Gorshkov, Optika i Spektroskopiya, 1976,41, 158. T. F. Moran, J. B. Wilcox, and L. E. Abbey, J . Chem. Phys., 1976,65,4540. M. Krauss and D . B. Neumann, MoZ. Phys., 1976, 32, 101. K. H. Becker, H. Engels, and T. Tatarczyk, Z . Naturforsch., 1976, 31a, 673. A. Y . M. Ung, J . Chem. Phys., 1976, 65, 2987. D. J. Burns, D. E. Golden, and D . W. Galliardt, J . Chem. Phvs., 1976, 65, 2616. J. A. Kernahan, E. H. Pinnington, and K. E. Donnelly, Phys. Letters (A), 1976, 57, 323. L. S. Polak, D. I. Slovetskii, and R. D. Todesaite, High Energy Chemistry, 1976, 10, 53. P. H. Vidaud, R. P. Wayne, M. Yaron, and Avon Engel, J.C.S. Faraday ZZ, 1976, 72, 1185. C. Y. Ng, B. H. Mahan, and Y. T. Lee, J . Chem. Phys., 1976, 65, 1956. G. D. T. Tejwani, B. M. Golden, and E. S. Yeung, J. Chem. Phys., 1976, 65, 5110. W. Swider, Geophys. Res. Letters, 1976, 3, 335. W. H. Duewer, D. J. Wuebbles, and J. S. Chang, Nature, 1977, 265, 523. H. Zacharias, J. B. Halpern, and K. H. Welge, Chem. Phys. Letters, 1976, 43, 41.

Gas-phase Photoprocesses

189

m~lecule-~ s-l for the K’ = 3 level of IJ’= 1 up to 2.02 x cms molecule-l s-l for the K’ = 3 levels of u‘ = 0. Quenching rates of NO(B211, v’ = 9) by He, H,, CO, CF4, N2,and C02 relative to quenching by ground-state NO have the values 0.12, 0.13, 0.13, 0.14, Absolute values for the v’ = 0 level range from 0.17, and 0.18 for He and Ar up to 3.2 x 10-locm3molec~le-~ s-l for ethylene.s67 2.9 x The ratios are, however, very different from those in the study on the IJ’= 9 level, being (relative to that for NO) 2.1 x 7 x lo-,, 0.21, 4.3 x and 5 x for He, Ha, CO, N,, and CO, respectively for the u’ = 0 level. The rate constant for reaction (98) has been remeasured as ks8 = 2.34 x 10-l2exp( - 1450/T) cm3m~lecule-~ s-1.674 Reactions between iodine atoms and NO

+ 0,

___+

NO,

+ 0,

(98)

nitric oxide and NO2 have been studied q u a n t i t a t i ~ e l y .Various ~ ~ ~ aspects of the electronic absorption spectra of NO, have been discussed, including the extinction coefficients for NO, and N204,676 ab initio potential-energy surfaces for several and electronic states of the @B, +- T2A1 and B2B, +- Z2A1 predissociation in the zero-point level of the 249.1 nm band The photodecomposition in this region (250-213.9 nm) gives the products shown in equation (99).a80 N 0 , ( ~ 2 B z+ ) hv

-

O(l0)

+ NO(On)

(99)

Laser fluorescence spectroscopy of N02(2B1,K > 0) 681 has been reported. The quenching of N02(2B,)fluorescence by a magnetic field is due to an increase in the collisional cross-section caused by the field, but is a strong and erratic function of the excitation wavelength.682This is consistent with a model where local perturbations caused by neighbouring states vary strongly from level to level. Magnetically induced changes in the absorption coefficient were also observed, and were attributed to the tuning in and out of resonance with the exciting light of energy levels that have energy differences. Fluorescence lifetime measurements have produced evidence of a perturbed 2Bz state of NOz near 600 nm.683In the laser-stimulated reaction of NO2* with COYit was shown that one in twenty collisions of the excited NO2* with NO, deactivated the excited it has been shown that In contrast with results in an earlier over the total pressure range from 100 to 760Torr, NO,/N, mixtures give a constant quantum yield of fluorescence, implying normal Stern-Volmer behaviour 673

674 t375 676

677 678

67B 680

681

T. Hikida, S. Nakajima, T. Ichimura, and Y. Mori, J. Chem. Phys., 1976, 65, 1317. J. W. Birks, B. Shoemaker, T. J. Leck, and D. M. Hinton, J. Chem. Phys., 1976, 65, 5181. H. Vandenbergh, N. Benoitguyot, and J. Troe, Internat. J. Chem. Kinetics, 1977, 9, 223. A. M. Bass, A. E. Ledford, jun., and A. H. Laufer, J. Res. Nat. Bur. $tand., Sect. A, 1976, 80, 143. C. F. Jackels and E. R. Davidson, J. Chem. Phys., 1976, 65, 2941. G . D. Gillespie and A. U. Khan, J. Chem. Phys., 1976, 65, 1624. K.-E. J. Hallin and A. J. Merer, Cunad. J. Phys., 1976, 54, 1157. W. M. Uselman and E. K. C. Lee, J. Chem. Phys., 1976, 65, 1948. G . I. Senum and S. E. Schwartz, J. Mol. Spectroscopy, 1977, 64, 75. S. Butler and D. H. Levy,J. Chem. Phys., 1977,66, 3538. V. M. Donnelly and F. Kaufman, J. Chern. Phys., 1977, 66,4100. I. P. Herman, R. P. Mariella, jun., and A. Jaran, J. Chem. Phys., 1976, 65, 3792. S. Braslavsky and J. Heicklen, J. Photochem., 1972/3, 1, 203.

1 90

Photochemistry

for NO, fluorescence.s8s This is an important result if fluorescence is to be used for remote sensing of NO2. The photodissociation of the N 2 0 cation has been and exact collinear calculations have been carried out on the photodissociation reaction Sources and sinks of atmospheric N20 have been d i ~ c ~ ~ ~ e d . ~ ~ ~ N,O(f'Z+)

hv

N,('C,+)

+ O(1S)

Predissociation in the HNO Al -+ lA" system has been The rate constant for reaction (101) has been measured as 1.55 x 10-17 cm3 molecule-l s-l.sgl HN02

+ HN03

-

2N02

+ H20

(101)

The photolysis of nitrous acid has been used as a source of hydroxy radicals, which were then treated with acetaldehyde [reaction (102)], with a rate constant d 2 x 10-l1 cm3molecule-l s-1.6g2 OH

+ MeCHO

MeCO

+ H20

(102)

The absorption cross-section for gaseous nitric acid has been found to be five-six times greater than previously accepted values, and the quantum yield of OH radical formation was found to be 0.92 k 0.16 for 365 nm irradiation.693 Computed rates of HONO loss by reaction with OH radicals are some three orders of magnitude lower than the loss rate through photolysis. HONO is formed in the atmosphere from OH NO, but since it is efficiently photolysed, it is not an effective sink for radicals. NOCl is also less of a chlorine atom sink than thought previously since the absorption cross-sectionsare greater than previous measurements had indicated.500 The rate of reaction of C10 with NO has been measured as 1.13 x 10-l1 e ~ p ( 2 0 0 / T ) .Models ~ ~ ~ of the upper atmosphere which include reactions 602 predict large HCI/C10N02 ratios, as found by experiof chlorine nitrate mental measurements. Further aspects of the CIO,, NO, cycle and the ozone stratospheric layer have been d i s c ~ s s e d . ~ 587 ~~~

+

4999

497s

CO and CO, Reactions.-Radiative lifetimes of the CO+(x2nt)

and the C 0 2 + ( 4 and CO2+(B) states,6s photodissociation and photodetachment of photodetachment in relaxed molecular negative ions in CO2/O,/H20 C03-,696 and spectroscopic studies on the 2 state of C02+,697 have been reported. e8e e87

M. Birnbaum, C. L. Fincher, and A. W. Tucker, J. Photochem., 1977, 6, 237. R. G. Orth and R. C. Dunbar, J. Chem. Phys., 1977, 66, 1616. M. Shapiro, Chem. Phys. Letters, 1977, 46, 442. M. B. McElroy, 3. W. Elkins, S. C. Wofsy, and Y . L. Yung, Rev. Geophys. Space Phys., 1976, 14, 143.

eDO

eol egg 694 Oo5

P. A. Freedman, Chem. Phys. Letters, 1976, 44, 605. E. W. Kaiser and C. H. Wu, J. Phys. Chem., 1977, 81, 187. R. A. Cox, R. G . Derwent, P. M. Holt, and J. A. Kerr, J.C.S. Faraday I , 1976, 72, 2061. R. A. Cox and R. G . Dement, J. Photochern., 1976, 6, 23. G. R. Mohlmann and F. J. de Heer, Chem. Phys. Letters, 1976, 43, 170. P. C. Cosby, 5. H. Ling, J. R. Peterson, and J. T. Moseley, J. Chem. Phys., 1976, 65,

5267. S. P. Hong, S. B. Woo, and E. M. Helmy, Phys. Rev. (A), 1977, 15, 1563. eo7 W. Sim and M. Haugh, J. Chem. Phys., 1976,65, 1616. e9e

191

Gas-phase Pho toprocesses

The importance of anthropogenic CO emissions on the CO/OH/CN cycle has been Observation of the CO(xln --f zlE+)sgs and(d"Af --f ~ 1 C + ) 6 gsystems B in the vacuum-u.v. photolysis of CO, has been made, and theoretical considerations of this dissociation have been SO, Reactions.-Detailed single-vibronic level fluorescence spectra of SO, in the 315-308 and 307-293 nm regions reveal that the upper state is not linear, but has an OSO angle similar to that in the ground The rate constant SO2

+ hv

313 nrn

313 lllll

3S02 ---+ SSO,

+ allene A

+ allene A +M 3SO, + SO, %O, + M so,* A

SO,*

+ allene

____+

____+

SO,*

+M 'so2

lS0,

+ allene

+ so, ISO, + M

lS0,

so2**

+ allene SO2** + NO

SO,**

where M is C02, NO, or H,O

_____+

___+

___+

____+

SO,*

3S02 SO,

+ hv,

A

+ allene 2CO + C2H4 SO,

not CO or C,H, not CO or C,H4

2S0, SO,

+M

IS02

SO, SO,**

+ allene

+ allene SO,** + M SO, + M

SO,

so2 C,H, SO,

4- hVp

+ CO + allene

2s0, SO,

+M

so, CO

not CO or C,H,

Scheme 5 688

700 701

E. Phillips, L. C. Lee, and D. L. Judge, J. Chem. Phys., 1977, 66, 3688. E. Phillips, L. C. Lee, and D. L. Judge, J. Chem. Phys., 1976, 65, 3118. Y. B. Band and K. F. Freed, J. Chem. Phys., 1976, 64,4329. R. J. Shaw, J. E. Kent, and M. F. O'Dwyer, Chem. Phys., 1976,18, 155, 165.

192

Photochemistry SO2

+ OH + (M)

____+

HSO,

(103)

of reaction (103) at room temperature is 1.6 x cm6mo1ecule-2s-1 with N2 as the third body, and this reaction is almost certainly the cause of the stratospheric sulphate layer, and is an important reaction in the conversion of 703 which has been commented upon.7o3o tropospheric SO, in its SBl state has been used as a sensitizer for the isomerization and other reactions of a variety of olefins, including 1,2-difluoro- and -dichloroeth~lenes,~O* pentene,'05 and 1,3-~entadienes.~O~ The reaction between S02(3B1)and cis-1,Zdifluoroethylene has a rate constant of 1.72 x 1O1O 1 mol-1 s-l, and produces a stationary state cis : trans products ratio of unity, in contrast with the case in 1,2-dichloroethylenein which this ratio is 1.8, and 1,3-pentadienes, in which the trans : cis ratio is 1.2. The photolysis of SO, in the presence of acetylene707and allene708has been thoroughly investigated, and in the latter case, reactions could be explained on the basis of Scheme 5. SO2* is formed at a constant fraction of absorbed intensity and does not produce products directly, but can be collisionally deactivated to produce SO,**, which does. Reference 708 also contains a considerable amount of tabulated rate constant information from the literature, to which the reader is directed. The photochemical addition of SO2 to arylcyclopropanes 709 and the spectra of SO, (and SO and S,) from SF6 afterglows 710 have been discussed. S02,7029

A. W. Castleman and I. N. Tang, J. Photochem., 1977, 6, 349. R. Atkinson, R. A. Perry, and J. N. Pitts, jun., J. Chem. Phys., 1976, 65, 306. 70wA. J. Alkezweeny and D. C. Powell, Atmos. Enoironment, 1977, 11, 179. 7 0 p F. B. Wampler, Internat. J. Chem. Kinetics, 1976, 8, 511, 519; F. B. Wampler and J. W. Bottenheim, ibid., p. 585. ?06 F. B. Wampler, Internat. J. Chem. Kinetics, 1976, 8, 935, 945. ? 0 6 F. B. Wampler, J. Photochem., 1977, 6, 183. '07 N. Kelly, J. F. Meagher, and J. Heicklen, J. Photochem., 1977, 6, 157. 708 K. Partymiller, J. F. Meagher, and J. Heicklen, J. Photochem.. 1977, 6, 405. ? O D D. E. Applequist and L. F. McKenzie, J . Org. Chem., 1977, 42, 1251. 710 D. Kley and H. P. Broida, J. Photochem., 1977, 6, 241.

?Oa

703

Part II PHOTOCHEMISTRY OF INORGANIC AND ORGANOMETALLIC COMPOUNDS By J. M. KELLY

Photochemistry of Inorganic and Organometallic Compounds

195

1 Photochemistry of Transition-metal Complexes Several reviews and the proceedings of a symposium have dealt with general aspects of work in this area. A comprehensive survey of the radiation chemistry of metal ions in aqueous solution also contains much of interest for the inorganic photochemist.6 New theoretical alternatives to Adamson's rules for predicting the course of photosolvolysis reactions of octahedral metal complexes have been considered in two recent publication^.^^ Both groups of authors have used angular overlap methods to calculate from spectroscopic data the excited-state metal-ligand bond energies, taking account of both c- and n-bonding. In the more detailed study,8 excellent agreement with experiment is reported for the prediction of the labilized ligand of some 29 CrI" and Co"' complexes (including [Cr(en),F,]+ and [Cr(en),FCl]+, which are exceptions to Adamson's rule). In the case of these Cr-F complexes an important feature appears to be that the Adamson rules are related to ( I * - I ) rather than to I * (I* and I being the bond energies in the excited state and ground state respectively). Thus, although calculations do indeed show that in the lowest excited state of [Cr(en),FCl]+ more of the excitation energy is concentrated in the Cr-F bond than in the Cr-Cl bond, the greater strength of the Cr-F bond in the ground state remains dominant, causing I*(Cr-F) to be greater than I*(Cr-Cl) and therefore the Cr-F bond less prone to labilization. As the vast majority of photosubstitution reactions of transition-metal complexes have been conducted in aqueous solution, little is known about the course of these transformations in organic media. Raising the bulk viscosity of the solvent might be expected to cause an increase in the rate of geminate recombination within the solvent cage of the fragments produced upon photodissociation of the metal-ligand bond. It had previously been supposed that this was the origin of the apparent decrease in the quantum yield for the photoaquation of [CO(CN)8]3- in water-alcohol mixed solvents.* However, consideration of a larger number of mixed solvents leads to the conclusion that any correlation with viscosity is largely f o r t u i t o ~ s .(In ~ some cases errors might have been introduced through negrecting the formation of [Co(CN),(~oIvent)]~-.) The same authors have determined the variation of the efficiency of CNSphotosubstitution in trans-[Cr(NH,),(NCS),]- (1) and in trans-[Cr(en),(NCS)F]+ (2) in various mixed solvent^.^ They have found that although there is a marked effect with complex ion (1) (Figure l), no similar dependence is observed for (2). This contrasting behaviour probably stems from the need for water molecules in the solvation shell of the negatively charged (1) to rotate into a favourable a 9

ti

9

M. S. Wrighton, Topics Current Chem., 1976, 65, 37. €3. B. Abrahamson, A. B. Ellis, D. S. Ginley, D. L. Morse, M. A. Schroeder,and M. S. Wrighton, Mol. Photochem., 1976, 7 , 263. J. F. Endicott, Suruey Progr. Chem., 1976,7,41. 'Inorganic Compounds with Unusual Properties', Ado. Chem. Series No. 150, ed. R. B. King, Amer. Chem. SOC., Washington, 1976. G . V. Buxton and R. M. Sellers, Coordination Chem. Rev., 1977, 22, 195. L. G. Vanquickenborne and A. Ceulemans, J . Amer. Chem. Soc., 1977, 99,2208. J. K. Burdett, Chem. Phys. Letters, 1977, 47, 43. F. Scandola, M. A. Scandola, and C. Bartocci, J. Amer. Chem. Soc., 1975, 97, 4757. C. F. C. Wong and A. D. Kirk, Canad.J. Chem., 1976,54,3794.

196

Photochemistry

orientation for the substitution reaction, whereas with the cation (2) the nucleophilic oxygen atom of the water is already suitably positioned. The importance of solvation in determining the properties of co-ordination compound excited states is further emphasized by Conti and Forster’s study of [Cr(CN)J3- phosphorescence lifetime in both 80% glycerol-water and aqueous solutions.1° In both solvents the lifetime of the 2E state is found to depend on the wavelength of excitation, the effect being most pronounced near the longwavelength edge of the (4T2+- 4A2) absorption band. This phenomenon is 1.0

0-8

0.6

0.4

0.2

t

0 1.0

0.u

0.6

0.4

0.2

Figure 1 Relative quantum yields for photosolvation of thiocyanate ion from trans[Cr(NH,),(NCS),]- as a function of aqueous mixed solvent composition. 0acetonitrile, 0 acetone, 0 ethanol, A ethylene glycol, and glycerol Reproduced by permission from Canad. J. Chem., 1976, 54, 397)

attributed to the presence of various long-lived solvates. Clearly if such solvates are sufficiently long-lived to influence the photophysical properties of the 2E state (T > 0.1 p),they must also affect the behaviour of the very short-lived 4T, state of the Cr“’ complex and, by extension, those of other complex ions, The effect of viscosity on charge-transfer photoreactions has been the subject of several earlier investigations. Many of these studies have been carried out in mixed solvents, such as glycerol-water, where other effects such as change in polarity or preferential solvation may also be important.ll Liu and Zink have lo

l1

C. Conti and L. S. Forster, J. Amer. Chem. SOC.,1977, 99, 613. J. F. Endicott in ‘Concepts of Inorganic Photochemistry’, ed. A. W. Adamson and P. D. Fleischauer, Wiley, New York and London, 1975, Chapter 3.

Photochemistry of Inorganic and Organometallic Compounds

197

eliminated this problem by using tris(dibenzyldithiocarbamato)iron(rrr) as a probe for viscosity effects in mixtures of benzene and long-chain hydrocarbons.12 Decomposition of the LMCT excited state in the presence of halogenated hydrocarbons (CHCI,, CC14,or CBr,) as scavengers proceeds as shown in equation (1). [Fe(dtc),]

ik

>

[Fe(dtc),X] + R-dtc

The dependence of the quantum yield of the reaction on the solvent viscosity has been analysed in terms of the Noyes treatment for radical pairs. Quenching and sensitization of the excited states of transition-metal complexes may involve a number of physical and chemical processes, and the factors determining which is predominant have not yet been thoroughly e1~cidated.l~ The deactivation of the triplet states of several organic molecules in aqueous solution [ET varying from 24 800 cm-l (xanthone) to 12 900 cm-l (protonated haematoporphyrin)] by [PdCl4I2-, [PtCI4l2-, and [Ni(CN),I2- proceeds by triplet-triplet energy transfer.14 In cases where the energy transfer is exothermic, the quenching is more efficient than it normally is for octahedral complexes, indicating that the ease of orbital overlap with these square-planar complexes facilitates the exchange mechanism for energy transfer. An exchange mechanism also appears to operate for the quenching of the singlet states of quinine and other organic bases by metal aquo-ions, although with Ag+ and Fe2+some type of charge-transfer process is imp1i~ated.l~ However, a resonance-energy-transfer mechanism has been suggested for the quenching of anthracene and rhodamine B fluorescence by [Cr(NCS),IS- and [Cr(NH3),(NCS),]-.ls Other related publications consider the quenching of eosine fluorescence l7 and duroquinone triplet state18 by various metal aquo-ions and the effect of heavy-metal ions on the excited-st ate interaction of the boric acid-benzoyl acet one complex.l9 Except for transition-metal organometallics and low oxidation state compounds, which are considered in Section 2, the photochemistry of compounds of each transition element, including metalloporphyrins, will now be treated systematically. Titanium.-Photo-oxidation of the solvent appears to be the primary process upon irradiation of ajl-unsaturated ketones 2o and nitriles 21 in methanolic solutions containing Tic&. Hydrogen is evolved upon irradiation of acidic solutions of Ti3+in the presence of catalytic amounts of Cut salts.aa The reaction appears to involve initial photo-oxidation of Cut to Cutl [equation (2)], no hydrogen being released in the absence of copper salts. l9 1’ l6 lo

l7

ao 21 22

P.-H. Liu and J. I. Zink, J . Amer. Chem. SOC.,1977,99,2155. V. Balzani, L. Moggi, F. Bolletta, and M. F. Manfrin, in ref. 4, p. 160. K.. C. Marshall and F. Wilkinson, Z . phys. Chem. (Frankfurt), 1976, 101, 67. J. A. Kemlo and T. M.Shepherd, Chem. Phys. Letters, 1977, 47, 158. E. B. Sveshnikova and S. P. Naumov, Optika i Spektroskopiya, 1976, 41, 225. T. A. Sakhverdov and Z. H. Turaeva, Optika i Spektroskopiya, 1976, 41, 1082. A. P. Darmanyan, I. V. Khudyakov, and V. A. Kuz’min, Izuest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 1772. M. Marcantonatos, Inorg. Chim. Acta, 1976, 16, 17; 1976, 19, 109; 1977, 21, 65, 75. T. Sat0 and S. Yoshiie, Chem. Letters, 1976, 415. T. Sato, G. Izumi, and T. Imamura, J.C.S. Perkin I, 1976, 788. K. L. Stevenson and D. D. Davis, Inorg. Nuclear Chem. Letters, 1976, 12, 905.

198 CU' + H+ Cull

+ Ti"'

hv

Photochemistry

+ 4H2

(2)

Cu' + TiIV

(3)

CU"

Vanadium.-Photoreduction of VOC13 in alcoholic solution proceeds in stages to give eventually V2+.23In the case of reduction of V3+ to V2+, acetaldehyde, the final oxidation product, is presumably formed by disproportionation of the ethoxy-radicals formed in the initial process (4).24 The quantum yield for V2+ formation is 0.23 at 280nm. Irradiation of the V2+ solution so formed causes = 0.009). evolution of hydrogen (Oaa0= 0.019;

Irradiation of Vv-doped silica gel in the presence of methane and oxygen leads to the reduction of Vv to VIV and the oxidation of methane.25 Aliwi and Bamford have studied the photolysis of 19 derivatives of type [VOQ2(0R)] (Q = 8-quinolyloxo-, R = alkyl) by a combination of spectroscopic, tracer, and spin-trapping experiments.26 These investigations indicate that the sole photochemical process is (5). The quantum yields for the various [VOQ,(OR)]

hv

VOQ2

+ *OR

(5)

derivatives lie in the range 0.7 x 10-3-9.2 x and increase with increasing electron-withdrawing power of the R-group. Photoactive polymers have been synthesized by attaching the VOQ2 chromophore to copolymers of methyl methacrylate and 2-hydroxyethyl methacrylate or of styrene and p-vinylbenzyl 28 The radicals formed on photolysis of the first type of copolymer cause grafting and cross-linking of the polymer and consequent gelation. The active initiator of the photoinduced polymerization of 2-methylpropl-ene2Dor of butadiene30 containing VC14 is the olefin radical-cation. This species is produced upon excitation of the corresponding olefin-VCI, chargetransfer complex.

Chromium.-Klaning has carried out experiments on the photoinduced oxidation of propan-2-01 using flash photolysis, quantum yield measurements, e.s.r. at low temperatures, and ~ p i n - t r a p p i n g . ~These ~ confirm earlier conclusions that the principal primary process is photolysis of the chromate ester [reaction (7)]. 23 24

25

ae 27 28

29

31

A. I. Kryukov, S. Y . Kuchmii, I. S. Shchegoleva, and A. V. Korzhak, Teor. i e k s p . Khim., 1977, 13, 135. B. V. Koryakin, T. S. Dzhabiev, and A. E. Shilov, Doklady Akad. Nauk S.S.S.R., 1976, 229, 128. S. A. Surin, A. D. Shuklov, B. N. Shelimov, and V. B. Kazanski, Khim. vysok. Energii, 1977, 11, 147. S. M. Aliwi and C. H. Bamford, J.C.S. Faraday I , 1977,73, 776. S. M. Aliwi and C. H. Bamford, Polymer, 1977, 18, 375, 381. C . H. Bamford, Polymer, 1976, 17, 321. L. Toman and M. Marek, Makromol. Chem., 1976, 177, 3325. J. Pilar, L. Toman, and M. Marek, J . Polymer Sci., Polymer Chem., 1976, 14, 2399. U. K. Klaning, J.C.S. Faraday Z, 1977, 73, 434.

Photochemistry of Inorganic and Organometallic Compounds

199

The CrIV species so formed normally disproportionates to Cr"I and CrV, or reacts with Crv to give Cr"' and CrV1. Related publications consider the photoreduction of dichromate by d i ~ l s , ~ by, poly(viny1 and on sensitization by methylene 35 [Me,CHOCrO,]-

hv ___3

Me,CO

+ CrIV + Hf

(7)

Irradiation of tran~-[Cr(en),(NH,)Cl]~+in its lowest quartet LF band (4E,) produces almost exclusively (100 k 3%) ci~-[Cr(en),(H,O)Cl]~+.~~ This pronounced stereochemical preference lends further support to the theory that Cr"I photosubstitutions proceed via a concentrated associative pathway in the quartet state. On excitation to higher LF states (4B2,4Eb)some equatorial labilization is also observed, indicating that under these conditions more than one reactive excited state is involved, probably the 4Bzand 4E,thermally equilibrated (thexi) states. The photoanation of [Cr(DMS0)J3+ (DMSO = dimethyl sulphoxide) by N3- in DMSO solution has a markedly higher quantum yield than that by SCN-, suggesting that the photosubstitution reactions of this complex under these conditions take place by an associative mechani~rn.~'This conclusion is further strengthened by the observation that the exchange of DMSO upon irradiation of the complex in deuteriated solvent is very inefficient. Hydroxide ion has in the past been used in mechanistic studies as a quencher of the 2Estate of Cr"' complexes. An investigation of the quenching of both the phosphorescence and thiocyanate phot osubstit u tion in [Cr(en>,(NCS),]+ reveals that OH- behaves differently from other quenchers, which act uia energy Thus whereas experiments with these other physical quenchers show that 20% of the yield of CNS- is 'unquenchable', results with OH- suggest a value of 55%. This larger amount of CNS- release must arise by OH- inducing CNS- formation from the doublet state of the complex, possibly uia a sevenco-ordinate intermediate. The photochemistry and photophysics of [Cr(bipy),13+(bipy = 2,2'-bipyridyl) have been the subjects of several recent paper^.^^-^^ The metal-centred (,E) state of the complex, formed on excitation with unit efficienc~,~~ may be monitored in fluid solution either by its emission or by conventional flash photolysis. It has been demonstrated that this state is highly oxidizing, efficiently accepting an electron from [Ru(bipy),12+ or Fe2+.40 Quenching studies indicate that the photosubstitution reactions, to give [Cr(bipy)2(H20)2]3+, [Cr(bipy)2(OH),]+, 01 [Cr(bipy),(OH)(H,O)l2+, depending on pH, proceed through the 2E state, although whether directly from this state or by reaction of the 4T2species formed sa s3 84

35

88

s7 s8

40

M. Miteva, P. Bonchev, and A. Malinovski, Doklady bolgarsk. Akad. Nauk, 1976,29, 81. V. Rek and M. Bravar, Nafta (Zagreb), 1976, 27, 267. M. Sasaki, S. Kikuchi, and K. Honda, Nippon Kagaku Kaishi, 1976, 886 (Chem. Abs., 1976, 85, 70 636). M. Sasaki, N. Sakai, S. Kikuchi, and K. Honda, Nippon Kagaku Knishi, 1976, 895 (Chem. Abs., 1976, 85, 70 637). C. F. C. Wong and A. D. Kirk, Inorg. Chem., 1976,15, 1519. C. H. Langford and J. P. K. Tong,J.C.S. Chem. Comm., 1977, 138. D. Sandrini, M. T. Gandolfi, A. Juris, and V. Balzani, J . Amer. Chem. Soc., 1977,99, 4523. F. Bolletta, M. Maestri, and V. Balzani, J. Phys. Chem., 1976, 80,2499. R. Ballardini, G. Varani, F. Scandola, and V. Balzani, J . Amer. Chem. Soc., 1976, 98, 7432.

Photochemistry

200

by back ISC is not yet apparent.41 The previously reported42luminescence of the 4T2 state on CW laser excitation of DMSO solutions of [Cr(bipy)J3+ has now been reassigned to some artefact, possibly some strongly emitting byproduct .41 The redox decomposition of [Cr(NH3),BrI2+ induced by light with wavelengths shorter than 360 nm is one of the few examples of this type of reaction established for CrI" Evidence includes the scavenging of Cr2+by [Co(NH,),FI2+ and the observation of Br2'- on flash photolysis in the presence of bromide ion. Although the observed quantum yields for Cr2+ are modest (0.02 d 0 2 5 4 d 0.04), the authors point out that recombination within the solvent cage may be very important, and that the primary redox yield may be an order of magnitude higher. A re-examination of the photolysis of [Cr(NH&N3)I2+ at 3 13 nm indicates that the primary photochemical reaction is formation of a chromium-nitrene complex [equation (8)] and not redox decomposition as previously

The reaction probably originates in a low-lying azide-centred excited state, as has previously been proposed for the analogous rhodium complex. In a recent study with the optically active complex tris-[( +)-3-acetylcamphorato]chromium(~~~) (3), it has been shown that the photoisomerization Me

H '-0 '

R (3)

(4) a ; 6;

R R

= =

Me Pri

proceeds via a square-pyramidal intermediate formed upon cleavage of one of the Cr-0 With the tris-@-diketonato)CrlI1complexes (4a) and (4b), the ratios ~)tsans-eie/~DcPs-ttana are markedly different C0.20 for (4a); Q 0.05 for (4b)].46 This difference has been attributed to the influence of steric factors on the rate of formation of products from the five-co-ordinate species formed on photolysis. Irradiation of [Cr(acac),] in the presence of (27)-alanine, (S)-valine, and (S)-phenylalanine is a convenient route to derivatives of the type [Cr(acac),L] .47

4s

44 46 46

47

M. Maestri, F. Bolletta, L. Moggi, V. Balzani, M. S. Henry, and M. Z. Hoffman, J.C.S. Chem. Comm., 1977, 491. N. A. P. Kane-Maguire, J. Conway, and C. H. Langford, J.C.S. Chem. Comm., 1974, 801. R. Sriram and J. F. Endicott, J.C.S. Chem. Comm., 1976, 683. M. Katz and H. D. Gafney, J. Amer. Chem. SOC.,1976, 98, 7458. S. S. Minor and G . W. Everett, jun., Znorg. Chem., 1976, 15, 1526. S. S. Minor and G . W. Everett, jun., Znorg. Chim. Acta, 1976, 20, L51. S. S. Minor, G . Witte, and G. W. Everett, jun., Znorg. Chem., 1976, 15, 2052.

Photochemistry of lnorganic and Organometallic Compounds

201 Energy-transfer experiments using CrllI complexes as both donors and acceptors have allowed the determination of their quantum yields for ISC in fluid solution at room temperature :39 trans-[Cr(en),(NCS),l+, 0.4; [Cr(CN)J3-, 0.5 ; [Cr(bipy)3]3+,1.O; [Cr(phen),13+, 0.2 ; [Cr(en),13++,0.7. Measurements of the phosphorescence lifetimes of various CrXI1complexes, [Cr(en),13+, [Cr(NH&(NCS)I2+, cis- and trans-[Cr(en),(NCS),]+, during quenching by [Cr(CN),J3reveal that the intermolecular energy transfer is r e v e r ~ i b l e . For ~ ~ this reason the system is not governed by Stern-Volmer kinetics, and apparent SternVolmer constants are quite misleading. Photophysical processes in Cr"' complexes have been discussed in a recent short review.4D Studies of the rise-time of the phosphorescence of various C P complexes confirm that the (4T2 + %E)ISC is a fast process (k > lo8 S - ~ ) . ~ O # 61 Other reports have dealt with the luminescence of [Cr(en)3]3+,6a[Cr(NH,),(OH)]2+,63[Cr(CN)6]3-y64 and cis-[Cr(en),X,]+ (X = F, C1, or Br).66 Molybdenum and Tungsten.-Photolysis of the metal-metal-bonded species [Mo2(S0JJ4- in sulphuric acid solution leads to production of hydrogen (@264

= 0.17) :as

2H+

+ 2[M0s(SOJ4l4-

hv

Ha

+ 2[MO,(S04)413-

(9)

Recent experiments, including isolation of the salt Ag,[W(CN),(H,O)], have confirmed that [W(CN),I4- undergoes efficient photoaquation in acid solution :a7 [W(CN)J4-

+ H,O

[W(CN)7(H20)]3-

+ CN-

(10)

Although at low pH photoreduction of [MO(CN),]~-, to give [Mo(CN)J4-, appears to proceed cleanlyYs8in alkaline solution secondary photochemical and thermal reactions cause complication^.^^ These include the formation of MoV1 by reaction of cyanide radical with the starting material.sD Other publications have considered the photochromism of alkylammonium molybdates 6o and the reduction of MoVXto MoV by laser radiation.61 Manganese.-Detailed knowledge of the excited-state interaction of chlorophyll and manganese complexes will be necessary for an understanding of the role of manganese in photosystem I1 of the chloroplast. In aqueous ethanol, both N. A. P. Kane-Maguire, C. G. Toney, B. Swiger, A. W. Adamson, and R. E. Wright, Inorg. Chim. A d a , 1977,22, L11. L.S. Forster, in ref. 4, p. 172. F.Castelli and L. S. Forster, J. Phys. Chem., 1977,81,403. O1 D. C. Stuart and A. D. Kirk, Rev. Sci. Znstr., 1977,48, 186. 6 8 G. L. Hilmes, H. G. Brittain, and F. S. Richardson, Inorg. Chem., 1977, 16, 528. 5s S. Decurtins, H. U. Gudel, and K. Neuenschwander, Inorg. Chem., 1977, 16,796. C. D. Flint and D. J. D. Palacio, J.C.S. Furaday II, 1977,73,649. I6C. D. Flint, A. P. Matthews, and P. J. OGrady, J.C.S. Faraday II, 1977,73,655. I6D.K. Erwin, G. L. Geoffroy, H. B. Gray, G. S. Hammond, E. I. Solomon, W. C.Trogler, and A. A. Zagars, J. Amer. Chem. SOC.,1977,99,3620. H. Mohan, J. Inorg. Nuclear Chem., 1976, 38, 1303. R. P. Mitra, B. K. Sharma, S. K. Gupta, and H. Mohan, Indian J. Chem., Sect. A , 1976,14, 457. 69 A. Marchaj, A. Plesinski, and Z. Stasicka, Roczniki Chem., 1976, 50, 1239. oo T.Yamase and T. Ikawa, Bull. Chem. SOC.Japan, 1977,50,746. dl B. V. Sarkisov, N. N. Milyaeva, V. A. Shkoda-Ul'yanov, and B. V. Anikeev, Ukrain. khim. Zhur., 1976, 42,936.

48

Photochemistry

202

excited singlet and triplet states of chlorophyll-a are quenched by manganese complexes, the efficiency decreasing in the order MnIV > Mnrrl > Mn11.62 With Mn" compounds, the quenching efficiency is dependent on the co-ordinated ligand (being most rapid with [Mn(phen)J2+}, and it is proposed that the deactivation process proceeds by enhanced radiationless conversion following complex formation [equation (1 l)]. Chl*

+Q

-

-

(Chl****Q)

Chl 3. Q

(1 1)

Irradiation (300 < X < 580 nm) of aqueous solutions of manganese(1v) sulphate leads to the evolution of oxygen and reduction of the complex to Mn111.63 Other authors have investigated the photoreduction of Mnrr1 in K2[Mn-2-a-hydroxyethylisochlorine,] acetate and have found, contrary to earlier work with related systems, that there is no evidence for concurrent hydroxyl radical formation.64 Other recent publications have considered the photodecomposition of Mn" complexes of both amine N-oxides 66 and edta,66 the Mn"-catalysed photooxidation of sulphite ions in and the luminescence68 and triboluminescence 6B of Mn" complexes.

Iron.-As previously noted for other alcohols bearing on a-hydrogen, two distinct processes are observed on photolysis of Fe3+ in acidic aqueous ethylene glycol solution.70 Thus hydroxyl radicals produced by reaction (12) react with the glycol to yield acetaldehyde, while oxidation by the CT state of the substrate in the second co-ordination sphere leads primarily to formaldehyde. With tertiary alcohols, where no a-hydrogen atom is present, direct reaction with the CT excited state does not occur.71 Fe3+ + H,O .OH

+ HOCH,CH,OH

hv

---+

+ *OH + H+ HOCHCH,OH + H,O Fe2+

(1 2) (13)

Methanol is oxidized to formaldehyde on aerobic irradiation of solutions of -tetra-azacyclo[Fe(tim)(MeOH)(OMe)]z+ (tim = 2,3,9,10-tetramethy1-1,4,8,11 tetradeca-1 ,3,8,10-tetraene).72 The primary process probably involves homolysis of the Fe-OMe bond. Other related studies include the use of benzene as a scavenger for *OH formed on irradiation of Fe3+,73and the photo-oxidation of alcohols,74tri~hloroethylene,~~ and poly(methy1 methacrylate) 76 by FeCl,. ea

e3 64 66 G8

' 6 68 69

70

1' 72 73

74 75

R. G. Brown, A. Harriman, and G . Porter, J.C.S. Faraday IZ, 1977,73, 113. T. S. Dzhabiev and V. Y. Shafirovich, Reaction Kinetics and Catalysis Letters, 1976, 4, 419. G. Vierke and M. Mueller, Z . Naturforsch., 1976, 31b, 816. L. C. Nathan, J. Cullen, and R. 0. Ragsdale, Inorg. Nuclear Chem. Letters, 1976, 12, 137. H. B. Lockhart and R. V. Blakeley, Etiviron. Letters, 1975, 9, 19. S. Lunak and J. Vepiek-SiSka, Reaction Kinetics and Catalysis Letters, 1976, 5, 157. D. Oelkrug and W. Kempny, Ber. Bunsengesellschaftphys. Chem., 1976, 80, 436. G. E. Hardy and J. I. Zink, Inorg. Chem., 1976, 15, 3061. J. H. Carey, E. G . Cosgrove, and B. G . Oliver, Canad. J . Chem., 1977, 55, 625. J . H. Carey, B. G . Oliver, and C. H. Langford, Canad. J. Chem., 1977, 55, 1207. D. W. Reichgott and N. J. Rose, J . Amer. Chem. Soc., 1977, 99, 1813. N. Jacob, I. Balakrishnan, and M. P. Reddy, J . Phys. Chem., 1977, 81, 17. V. I. Stenberg, S. P. Singh, N. K. Narain, and S. S. Parmar, J . Org. Chem., 1977, 42, 171. Z. A. Tkachenko, S. Y. Kuchmii, A. V. Korzhak, and A. I. Kryukov, Ukrain. khim. Zhiir., 1976, 42, 539.

78

N. I. Zaitseva, T. V. Pokholok, G . B. Pariiskii, and D. Y. Toptygin, Doklady Akad. Nauk S.S.S.R., 1976, 229, 906.

Photochemistry of Inorganic and Organonzetallic Compounds

203

It has been found that the carbon dioxide evolved on photodecomposition of potassium ferrioxalate is isotopically enriched in 12C,the extent of fractionation depending on wavelength 77 [(12C/13C),,,1,/(12C/13C)ref = 1.050 at 520 nm and 1.001 at 366 nm]. The results are interpreted in terms of two reactive intermediates. The photolysis of (Hphen),[Fe(~x),],~~ difficulties with the ferrioxalate a c t i n ~ r n e t e r ,and ~ ~ the solid-state photolysis of K3[Fe(ox),],3H20 8o have been the subjects of recent reports. Photolysis of the Fe"' complex of pyridine-2-carboxylicacid gives a surprisingly high yield of 2,2'-bipyridyl, possibly because the pyridine radical, formed after decarboxylation, remains co-ordinated to the iron atom.81 The formation of their Fe1I1 complexes and subsequent photo-oxidation cause the aerobic decomposition of edta 66 and sodium sulphite.82 Excitation of the LMCT state of Fe(S2CNRz)3(R = Et or CH,Ph) causes homolytic cleavage of the Fe-S bond, with, in the presence of halogenated hydrocarbons, the resultant formation of Fe(S2CNR2)2X.1as 83, 84 The quantum yield for the overall reaction is dependent on the C-X bond strength of the halogenated hydrocarbon, being markedly larger for CBr, than for CH,Br2 and for CHC13 than for PhCl. As noted in previous Reports, the nature of the primary photoprocesses of [Fe(CN),N0I2- is a matter of debate. Recent comparative e.s.r. and absorption spectroscopic investigations of its pliotolysis products in DMF, dimethylacetamide, or methanol solutions at both 77 K and room temperature demonstrate that the reaction course is also markedly medium-dependent.86 Other workers have studied the flash photolysis of [Fe(CN),I3- in the presence of NSand SCN-.66 Intramolecular electron transfer from the iron to the cobalt centre in the pyrazine derivative (5a) is induced by excitation of the Fe-heterocycle MLCT band at 620 nm (CS = 0.9).87 In the analogous complex (5b) the quantum yield is dramatically lower (0.02), probably owing to the lower oxidizing power of the

( 5 ) a; I, = (NH,),

b; L

'' 7D *O

83 84

86

87

= en

R. H. Betts and W. D. Buchannon, Canad. J. Chem., 1976,54, 2577. P. Thomas, M. Benedix, and H. Hennig, Z . Chem., 1977, 17, 114. W. D. Bowman and J. N. Demas, J. Phys. Chem., 1976,80,2434. E. L. Simmons, J . Phys. Chem., 1976,80, 1592. T. Kirnura, J. Kamimura, K. Takada, and A. Sugimori, Chem. Letfers, 1976, 237. S . Lunak and J. Vepfek-SiSka, Coll. Czech. Chem. Comm., 1976,41,3495. D. P. Schwendiman and J. I. Zink, J. Amer. Chem. Sac., 1976,98,4439. G. L. Miessler, G. Stuk, T. P. Smith, K. W. Given, M. C. Palazzotto, and L. H. Pignolet, Inorg. Chern., 1976, 15, 1982. R. B. Zhutkovskii and K. I. Zamaraev, Khim. oysok. Energii, 1976,10, 150. D. J. Kenney and H. M. Lin, Bull. Inst. Chem., Acad. Sinica, 1976, 23, 53. D. A. Piering and J. M. Malin, J. Amer. Chem. SOC.,1976, 98, 6045. 8

204

Photochemistry

Co(en), moiety.87Intermolecular electron transfer from other pentacyanoferrate(I1) complexes to the [ R ~ ( b i p y ) ~ ]excited ~+ state is responsible for quenching of its emission.88 Further studies have been carried out to assess the potential of thionine-Fe" 8 Q and iodine-Fe" systems as solar energy converters.Qo The efficiency of photodissociation of the Fe-L bonds in carbonyl, nitrosyl, isocyanide, and oxygen complexes of haemoglobin and myoglobin upon excitation of the porphyrin may be rationalized in terms of a one-electron MO theory, when account is taken of the d,(Fe)-n(L) character of the bond.B1 The rebinding of carbon monoxide following laser flash photolysis of carbonylhaem protein complexes has been monitored at temperatures between 5 and 340 K. The results for carbonylmyoglobin reveal that in the recombination reaction it is necessary to surmount (or tunnel through) four separate activation barriers 93 Horse-radish peroxidase caused by the solvent, the globin, and the haem.B2~ becomes photosensitive when it is converted from its high-spin paramagnetic form into a low-spin diamagnetic species on complexation with cyanide.Q4 Ruthenium.-The photosensitization and luminescence properties of [Ru(bipy),12+have been the subject of many publications in the past few years. As in other areas of inorganic photochemistry, however, there is a dichotomy between luminescence studies at low temperature (77 K and below) and those on the photophysical and photochemical processes under ambient conditions. From low-temperature data it has been suggested that the luminescent MLCT (d-n*) state of [Ru(bipy),12+consists of a manifold of three states separated by ca. 60 cm-l, and therefore thermally equilibrated except at extremely low t e m p e r a t ~ r e s96. ~Two ~ ~ groups of authors have recently reported on the variation of the luminescence lifetime of the complex in a variety of solvents as a function Q8 In both studies non-Arrhenius behaviour was of temperature (up to 373 noted, indicating that at higher temperatures (T > 320 K) some nonluminescent state plays a major role in the radiationless deactivation of the emitting excited state. Raising the temperature also causes the disappearance of the solvent isotope effect on the emission lifetime (at 298 K, TH,O = 0.58 ,us, TD$J = 1.02 p: at 363 K, TH,O = 0.10 ps, TD,O = 0.10 ps). From these and other data, Van Houten and Watts argue that a LF excited state lying 3600 cm-l above the emitting MLCT level is responsible for its efficient deactivation at those higher temperatures where its population is possible. Observation of photoinduced substitution of [Ru(bipy)J2+ at 368 K corroborates this h y p o t h e ~ i s . ~Other ~ authors have recorded the luminescence spectra of K).971

O1 92

93 94 96

96

*'

98

H. E. Toma and C. Creutz, Ingor. Chem., 1977, 16, 545. T. Sakata, Y. Suda, J. Tanaka, and H. Tsubomura, J. Phys. Chem., 1977,81, 537. T. Ohta, S. Asakura, M. Yamaguchi, N. Kamiya, N. Gotoh, and T. Otagawa, Internat. J. Hydrogen Energy, 1976, 1, 113. T. Kitagawa, Y. Kyogoku, T. Iizuka, and M. I. Saito, J. Amer. Chem. Soc., 1976, 98, 5169. N. Alberding, R. H. Austin, S. S. Chan, L. Eisenstein, H. Frauenfelder, I. C. Gunsalus, and T. M. Nordlund, J. Chem. Phys., 1976, 65,4701. L. Eisenstein, Internat. J. Quantum Chem., Quantum Biol. Symp., 1976, 3, 21. Y.-J. Kang and J. D. Spikes, Biochem. Biophys. Res. Comm., 1977, 74, 1160. G. A. Crosby, in ref. 4, p. 149. K. W. Hipps and G. A. Crosby, J . Amer. Chem. SOC.,1975,97,7042. J. Van Houten and R. J. Watts, J . Amer. Chem. SOC., 1976, 98, 4853. S. R. Allsopp, A. Cox, S. H. Jenkins, T. J. Kemp, and S. M. Tunstall, Chem. Phys. Letters, 1976, 43, 135.

Photochemistry of Inorganic and Organometallic Compounds

205

[R~(bipy).(phen)~-,]~+,~~ and of [RuL312+ [L = 2-(2'-pyridy1)quinoline or 2,2'-biq~inoline].~~~ Previous work has shown that * [Ru(bipy),12+ may undergo intermolecular deactivation by energy transfer or by either oxidative or reductive electron transfer [equations (14)-(16)]. This is further illustrated in a recent study of the

+Q * [ R u ( b i ~ y ) ~ l+ ~ +Q *[Ru(bipy),12+

+

*[R~(bipy)~]~+Q

-

+ [Ru(bipy),I3+ + Q[Ru(bipy),]+ + Q+

[ R ~ ( b i p y ) ~ ] ~ +*Q

(14)

(1 5 ) (16)

luminescence lifetime and intensity of [ R ~ ( b i p y ) ~ ]and ~ + [Ru(phen)J2+ during quenching by some 20 metal Diffusional energy and electron transfer are the principal deactivation routes, although some static quenching due to ion-pairing occurs with anionic complexes. No evidence could be found for heavy-atom or paramagnetic quenching. The neutral complex [R~(bipy),(CN)~], being free of the problems of ion-pairing or electrostatic repulsion, offers some advantages over [Ru(bipy)J2+ as a sensitizer.lo2 Nevertheless, static quenching occurs with Cu2+, Ni2+, and Co2+, where the co-ordinated cyanide acts as a bridging ligand. Complexes of the type [RuLSl2+with Me-, Ph-, CI-, Br-, or NO,-substituted phenanthrene or bipyridine ligands all possess lowest-lying MLCT states which may be quenched by Fe3+, Eu3+, or Cr3+.lo3 In the first two cases quenching is by oxidative electron transfer (15), whereas for Cr3+ energy transfer takes place. Other authors have used C P complexes as energy-transfer acceptors to estimate the quantum yield for intersystem crossing in [Ru(bipy),(CN),] (a = 1.O) and [Ru(phen)J2+ (@ = 0.65).38 For this purpose @=C for [Ru(bipy)J2+ was taken to be 1.0, as is now generally accepted. Competitive electron- and energytransfer processes are responsible for the quenching of * [Ru(bipy),12+by various 1,2-dipyridylethylene~.~~~ Thus with 1,2-bis-(3-pyridyl)ethylene the rate constant for energy transfer (14) is 6 x lo* dm3mol-1 s-l while that for oxidative electron transfer (15) is 7 x lo6dms mol-1 s-l. Flash photolysis is a useful tool for studying electron-transfer reactions not only of excited states but also of reactive ground-state compounds. Thus from the decay kinetics of the corresponding [RuLgl3+,formed from *[Ru(bipy),12+ or * [ R ~ ( p h e n ) ~ and ] ~ + Fe3+,the rate constants for processes (17) and (18) have been deduced (1.8 x lo9 and 1.2 x lo0 dm3mol-1 s-l).lo6 A drop in the rate constant for electron transfer between *[Ru(bipy)J2+ or *[Ru(4,4'-diMebipy)J2+

+ [Fe(phen)J2+ [Ru(phen),13+ + [Ru(bipy)J2+ [Ru(bipy),13+

-

+ [Ru(phen),12+ + [Ru(bipy),13+

[ R ~ ( b i p y ) ~ ] ~ +[Fe(phen),13+

(1 7) (1 8)

G. A. Crosby and W. H. Elfring, J. Phys. Chern., 1976, 80,2206. D. M. Klassen, Inorg. Chem., 1976, 15, 3166. lol J. N. Demas and J. W. Addington, J. Amer. Chem. SOC.,1976, 98, 5800. loa J. N. Demas, J. W. Addington, S. H. Peterson, and E. W. Harris, J . Phys. Chem., 1977, 81, 1039. 103 C. T. Lin, W. Boettcher, M. Chou, C. Creutz, and N. Sutin, J. Amer. Chem. Soc., 1976,98, 6536. lo( A. R. Gutierrez, T. J. Meyer, and D. G. Whitten, Mol. Photochem., 1976, 7 , 349. lo5 R. C. Young, F. R. Keene, and T. J. Meyer, J . Amer. Chem. SOC.,1977,99, 2468.

OD

loo

Photochemistry

206

and other [M(bipy),12+ complexes on increasing the free energy for electron transfer may be confirmation of the 'inverted region' predicted by Marcus theory.lo6 Flash spectroscopy has provided direct evidence for the reductive quenching (16) of *[Ru(bipy),12+ by Eu2+, [Ru(NH3)J2+, or [Fe(CN)6]4-,107or by various organic donor molecules, including phenothiazines.lo8, lo9 The eeciency of quenching of [Ru(bipy),12+ luminescence by [Fe(CN),LI3- complexes parallels their reduction potentials, again indicating a reductive mechanism.88 All these quenching experiments confirm that the reduction potential for *[Ru(bipy),12+/ [Ru(bipy),]+ is close to the thermodynamic limit of approximately 0.8 V. It has also been observed that the Ru' complex reacts rapidly with oxygen to give the superoxide ion [equation (19)].10*plo9

-

+

[R~(bipy)~I+ + O2 lR~(bipy),]~+ 0,(19) It has been known for some time that [Ru(bipy),12+ is an efficient (@ = 0.85) singlet oxygen sensitizer, and this property has now been utilized in the construction of a laser actinometer.l1° Similar activity is also reported for various substituted phenanthroline complexes of ruthenium(r1) and osmium(Ir).lll The precise nature of the complex { [Ru(bipy),] 02},+ is not clear, some authors favouring an exciplex which may subsequently form singlet oxygen or undergo deactivationll' whereas others prefer a species with a high degree of charge transfer { [Ru(bipy)J3+ ...02-).l12 The small anodic photocurrent produced on irradiation of [Ru(bipy),12+ at a SnO, electrode has been attributed to direct electron transfer from the excited state to the conduction band of the semiconductor [equation (20)].l13 Much 0..

*[Ru(bipy),12+

anode

, [Ru(bipy),13+ + e-

(20)

larger cathodic currents are observed on irradiation of acidic solutions of [Ru(bipy),12+ in the presence of mild oxidants (X) such as oxygen, methylviologen, or Fe3+; the proposed mechanism is that shown by reactions (21) and (22).ll4, 115 From the variation of the magnitude of the photocurrent with Fe3+ *[Ru(bipy),I2+ [Ru(bipy),I3+

+X + e-

cathode,

[Ru(bipy),I3+ + X[Ru(bipy),12+

(21) (22j

concentration the lifetime of the excited Ru" complex has been estimated and found to be in good agreement with that determined directly in luminescence Emission from * [Ru(bipy)J2+ is induced during cathodic polarization of S i c or GaP electrodes in solutions of [Ru(bipy),13+, indicating that electron transfer from the semiconductor to the Ru"' complex occurs [i.e. the reverse of reaction (2O)].ll6 106

C.Creutz and N. Sutin, J . Amer. Chem. Soc., 1977, 99, 241.

C. Creutz and N. Sutin, J. Amer. Chem. SOC., 1976, 98, 6384. M. Maestri and M. Gratzel, Ber. Bunsengesellschaft phys. Chem., 1977, 81, 504. l o g C.P. Anderson, D . J. Salmon, T. J. Meyer, and R. C. Young, J . Amer. Chem. SOC.,1977,99, 1980. J. N. Demas, R. P. McBride, and E. W. Harris, J. Phys. Chem., 1976, 80, 2248. n1 J. N. Demas, E. W. Harris, and R. P. McBride, J. Amer. Chem. SOC.,1977, 99, 3547. l l a J. S. Winterle, D. S. Kliger, and G . S. Hainmond, J . Amer. Chem. Sac., 1976, 98, 3719. 113 M.Gleria and R. Memming, Z . phys. Chem. (Frankfurt), 1975, 98, 303. 114 S. 0. Kobayashi, N. Furuta, and 0. Simamura, Chem. Letters, 1976, 503. 115 J. Phillips, C.H. Langford, and J. A. Koningstein, J.C.S. Chem. Comm., 1977, 425. l T 6 M. Gleria and R. Memming, 2. phys. Chem. (Frankfurt), 1976, 101, 171. lo7

lo8

Photochemistry of Inorganic and Organometallic Compounds

207 The exciting observation by Whitten and co-workers of oxygen and hydrogen evolution on irradiation of assemblies of monolayers of (6) has provided a

,:%

Ly----N\ / O=t

0

stimulus for many groups. However, it has now been demonstrated that the behaviour of such monolayers is dependent on the particular sample used and on its purity.lla, lle Highly purified samples of (6) do not catalyse the photodecomposition of water, but are themselves destroyed by undergoing ester hydrolysis. It is likely that the reactive species in the active monolayer assembly is some trap centre, possibly another Ru" complex. Derivatives of the type [ R ~ I ~ ( b i p y ) ~ have L ~ ] ~provided + the first direct evidence 121 With for acid-base equilibria in the excited states of metal 117

G. Sprintschnik, H.W. Sprintschnik, P. P. Kirsch, and D. G . Whitten, J. Amer. Chem. SOC.,

11*

G. Sprintschnik, H.W. Sprintschnik, P. P. Kirsch, and D. G. Whitten, J. Amer. Chem. SOC.,

1976,98,2337. 1977,99,4947. S. J. Valenty and G . L. Gaines, J. Amer. Chem. Soc., 1977, 99, 1285. lZo S. H.Peterson and J. N. Demas, J. Amer. Chem. SOC.,1976, 98, 7880. 119

P. J. Giordano, C. R. Bock, M. S. Wrighton, L. V. Interrante, and R. F. X. Williams, J . Amer. Chem. SOC., 1977,99, 3187.

lZ1

Photochemistry

208

[Ru(bipy),(CN),] (D) in strongly acid solution ([Hf] = 2.5 mol dm-s) emission is only from *D even though the protonated forms DH+ and DHA2+ predominate in the ground state (27% and 65% respectively).120 These observations are consistent with rapid excited-state protonation-deprotonation equilibria involving *D, *DH+, and *DH2+in which *D is favoured. With complex (7) emission is observed from both the protonated and deprotonated forms [reaction (23)].121

(74

(7 b)

However, the ground-state and excited-state complexes have quite different pK, values. From the pH dependence, the pK, for the excited state was calculated to be 8.50, indicating that the excited state of (7) is a weaker acid than the ground state (pKB = 5.5). Thus at pH = 3.5, on excitation of the deprotonated form (7b), emission originates mainly from the excited state of the protonated species (7a). This behaviour is in agreement with the expected increased negative charge on the ligand in the MLCT excited state. Monolayers of the dioctadecyl ester of [Ru(CO)(py)(mesoporphyrin IX)] exhibit some remarkable properties.122 Thus, in contrast to the behaviour of the parent complex [Ru(CO)(py)(mesoporphyrin IX)], the co-ordinatively unsaturated species (8) formed on photoinduced CO expulsion may be readily isolated and characterized [reaction (24)]. It reacts with oxygen or nitrogen to form stable derivatives, which on further irradiation in the presence of CO re-form the original carbonyl compound [reactions (25) and (26)].

-

hv

[ R ~ ( C O ) ( P Y ) ( P ~ ~ P ~ Y ~ ~ ~ )[Ru(~y)(porphyrin)l I + CO [ R u ( p ~ ) ( ~ o r ~ h ~+ r i nY2 )I

(24)

(8)

[ R U O T , ) ( P Y ) ( P ~ ~ P ~ Y ~ ~ ~ ) (25) I

hv

[Ru(Y,)(PY)(PorPhY~n)I+ co [Ru(Co>(PY>(PorPhYrin>l + y, (26) In [Ru(NO)X,]~- (X = Cl, Br, or I)123and [R~(bipy),(NO)Cl]~+,l~~ substitution of NO by solvent is induced by light, although it is not clear at present whether the initial step is expulsion of NO or NO+. The RuIV complex [Ru(S,CNE~~)~CI] has been isolated after photolysis of [Ru(S2CNEt2),] in chloroform or dichloromethane.126 Cobalt.-Excitation of ligand-field states of Co”’ ammine complexes normally leads to low-efficiency photoaquation. The reactive species is probably a thermally equilibriated (thexi) singlet state.126 The ligand labilized may be F. R. Hopf and D. G. Whitten, J. Amer. Chem. SOC.,1976,98, 7422. A. B. Nikol’skii, A. M. Popov, and I. V. Vasilevskii, Koord. Khim., 1976, 2, 671. la4 R. W. Callahan and T. J. Meyer, Inorg. Chem., 1977, 16, 574. la6 K. W. Given, B. M. Mattson, and L. H. Pignolet, Znorg. Chem., 1976,15, 3152. lZ6 A. W. Adamson, in ref. 4, p. 129. lZa

123

Photochemistry of Inorganic and Organometaliic Compounds 209 predicted by rules similar to those introduced by Adamson for Cr"' complexes. Thus with tran~-[Co(en),(NH,)Cl]~+,ammonia aquation (@ = 1.5 x lo-,) is more important than chloride aquation (@ = 3.0 x although unlike the corresponding CrlI1 system the reaction proceeds with retention of stereochemistry [reactions (27) and (28)].12' The low quantum yields are attributed hv

trans-[Co(en),(NH,) C1I2+ ----+ rr~ns-[Co(en),(NH,)Cl]~+

+ NH, truns-[Co(en),(NH,)(H,0)I3+ + C1trans-[Co(en),( H,O) C1I2+

(27) (28)

to efficient radiationless deactivation. With ~is-[Co(en),(NH,)Cl]~+the overall reactions are low-efficiency chloride aquation (0= 3.1 x and ammonia aquation (0= 2.1 x although it is suggested that labilization of one of the ethylenediamine Co-N bonds followed by either recombination or displacement of an adjacent ligand is an important reaction pathway. Direct population of the triplet states of [Co(NH3)J3+ or of [CO(NH,)~C~]~+ by laser excitation (647 nm) causes photoaquation with lower quantum yields than those found for irradiation into the spin-allowed ligand-field bands (514 nrn).l2* Interestingly the relative amounts of chloride versus ammonia aquation are inverted (1 : 3 at 514 nm, 1 : 0.11 at 647 nm). Thus, as in the thermal reactions, direct population of the triplet state induces mainly chloride replacement, whereas excitation of the singlet states produces predominantly the anti-thermal product. On the other hand, the bimolecular reaction of [Co(enI,l3+ with [Fe(CN),I4- is more efficient upon direct triplet excitation, and it has been proposed that this reaction proceeds from a ST state formed by intersystem crossing.128 In an analogous fashion to [Co(NH3),(N02)12+,[Co(NH3),SCNI2+undergoes both linkage isomerization and redox decomposition, the ratio of the two processes being essentially independent of ~ave1ength.l~~ (In contrast, [Co(NH&,NCSI2+ does not isomerize.) It is probable that the isomerization, redox decomposition, and (low-efficiency) photoaquation proceed through the secondary radical pair [equations (29)-(3 l)], whereas re-formation of the [CO(NH,)$CN]~+ takes place either by direct deactivation of the LMCT state or by collapse of the solvent-caged radical pair. [CO(NH,),]2+, *SCN [cO(NH,)6]2+,*SCN [CO(NH3)612+,*SCN

-

[Co(NH&,NCS12+ Co2+

+ 5NH3 + *SCN

[Co(NHJ,I3+,NCS-

(29) (30)

(31)

Two Russian groups 131 have reported some observations on intermediates formed on U.V. photolysis of [(NH,),CO(O,CR)]~+ (R = alkyl). In aerated solution, complexes of the type [CO(NH&,O~CO(NH~)~]~+ could be identified,13* 13*9

la'

lZ8 12@

130 131

R. A. Pribush, R. E. Wright, and A. W. Adamson, J . Amer. Chem. SOC.,1977, 99, 2495. C. H. Langford and C. P. J. Vuik, J . Amer. Chem. SOC.,1976, 98, 5409. M. Orhanovic and N. Sutin, Inorg. Chem., 1977, 16, 550. L. N. Neokladnova, V. A. Repina, and N. I. Zotov, Zhur. neorg. Khim.,1976,21,149. A. L. Poznyak, V. V. Pansevich, and S. I. Arzhankov, Koord. Khim., 1976,2,472.

210

Photochemistry

whereas species formed at 77 K were assigned to [Co(NH3),RI2f on the basis of characteristic bands in the U.V. absorption These latter complexes are assumed to be formed by decomposition of the radical pair [equations (32) and (33)]. Other authors describe the photoaquation and photoredox reactions [(NHs)5Co(02CR)]2+

[(NH,),CO]~+,*CO,R

hv

[(NHJ,CO]~+,*CO,R [(NH,),CoRI2+

+ CO,

(32)

(33)

of [ C O ( ~ ~ ) ~ ( S , O ~ )thermal ,]-,~~~ gravimetric data for both anation and redox and the photoproducts after solid-state photolysis of [CO(NH,),(H,O)]X,,~~~ polymerization of acrylamide by [CO(NH~)~]~+-SCNBy use of electrochemicalmonitoring it has beenc onfirmed that the benzophenone triplet state sensitizes the photoreduction of [Co(NH3)J3+uia energy Pulse radiolysis studies with Co"' ammine and macrocyclic complexes provide useful information about the reactivity of the corresponding Co" complexes in solution .136 -ls9 For example, [Co(NH3),C1]+ decomposes to [CO(NH,)~]~+ within 2 ps, whereas the remaining ammonia ligands are lost in a stepwise fashion with half-lives of 10, 65, and 540 ps.136 The reactivity of [Co(acac),] produced photochemically has been compared with that formed from 6oCo recoil species in [Co(a~ac),].l~~ Irradiation of [CO(CN),(SO,),]~- in the presence of labelled cyanide ion produces a mixture of cis- and trans-[Co(CN),(*CN)(SO3)l4-, in contrast with the thermal reaction which proceeds stereospecifically to give the trans-isomer as the sole product.141 This scrambling in the photochemical reaction is taken as evidence for the creation of an excited state of [CO(CN),(SO,)]~-upon photodissociation of the starting material. Phosphorescence spectra of [CO(CN),]~-,'~~ and of [CO(CN)~(SO~)]~and [CO(CN),(NO,)]~-,~~~ have been recorded at low temperatures.

Rhodium and Iridium.-Previous work has suggested that, in contrast with Cr"' complexes, the photoaquation of Rh"' compounds proceeds with retention of configuration. However, most experiments have been conducted either with trans-[RhL,X,]+ complexes or with cis-complexes, such as [Rh(cyclam)X,]+ (cyclam = 1,4,8,1l-tetra-azacyclotetradecane), which contain a ligand that resists isomerization. Strauss and Ford 144 have now shown that photoaquation of cis-[Rh(NH3),C12]+leads to the trans-product (0= 0.33) [equation (34)]. 134

K. P. Balashev, I. P. Serova, and G . A. Shagisultanova, Koord. Khim., 1977, 3, 82. G. d'Ascenzo, U. B. Ceipidor, and A. Marino, Thermochim. Acta, 1976, 16, 69. F. Takemura and K. Sakaguchi, Nippon Kagakrr Kaishi, 1976, 1470 (Chem. Abs., 1976, 85,

135

13. S. Hall, K. F. Dahnke, S. S. Fratoni, and S. P. Perone, J . Phys. Chem., 1977, 81, 866.

133

160 653).

ne-J. Lilie, N. Shinohara, and M. G. Simic, J . Amer. Chem. SOC.,1976, 98, 6516. 13*

M. G. Simic, M. Z. Hoffman, and N. V. Brezniak, J . Amer. Chem. Soc., 1977, 99, 2166. A. M. Tait, M. Z.Hoffman, and E. Hayon, Internat. J. Radiation Phys. and Chem., 1976, 8, 691.

139

141 142 143

144

J. F. Endicott, J. Lilie, J. M. Kuszaj, B. S. Ramazwamy, W. G . Schmonsees, M. G . Simic, M. D. Glick, and D. P. Rillema, J . Amer. Cliem. SOC.,1977, 99, 429. Y . Nishi and T. Tominaga, Radiochem. Radioanalyt. Letters, 1976, 24, 249. K. F. Miller and R. A. D. Wentworth, Inorg. Chem., 1976, 15, 1467. K. W. Hipps, G. A. Merrell, and G. A. Crosby, J . Phys. Chem., 1976,80, 2232. B. Loeb and F. Zuloaga, J. Phys. Chem., 1977, 81, 59. D. Strauss and P. C. Ford, J.C.S.Chem. Comm., 1977, 194.

-

21 1

Photochemistry of Inorganic and Organometallic Compounds ci~-[Rh(NH3)4Cl,]+

+ HZO

hu

+

tr~ns-[Rh(NH3)4Cl(H,O)]~+ C1-

(34)

This important result, coupled with those of earlier studies, is most readily rationalized by assuming that the light-induced cleavage of the Rh-CI bond is followed by isomerization of the five-co-ordinate fragment, so that identical products are formed from both cis- and trans-complexes. Other authors have made further observations on the photoaquation of [Rh(NH3)6X]2+(X = C1, Br, I, or OH).145 Excitation of the ligand-field states of Ir"' ammine complexes induces photoaquation [equation (35)] with quantum yields which are very similar to those found previously for the analogous Rh"' complexes (Table l).14s This quantitative Table 1 Comparison of quantum yields for photosubstitution reactions (35) upon ligand-field excitation of Rh"' and 1r"I complexes 146 Rh"' I P Excitation wavelength/nm 313 254 313 313

Complex [M(NH3)S13+

[M(NHs)s(HzO>13+ [M(NH3)6C1I2+

0 0.09 0.082 0.42 0.13

Excitation wuvelength/nm a) 313 0.075 254 0.07 313 0.43 280-380 0.12-0.16

similarity, although possibly fortuitous, is quite surprising, as the ligand-field excited-state lifetimes are much shorter for iridium compounds than for the analogous rhodium species, and further the thermal substitution rates are smaller for the heavier element. These moderately efficient photoaquation reactions ]~+ contrast with those observed for CO"' complexes (@ for [ C O ( N H ~ ) ~= 2 x 10-4).

+

-

[M111(NH3)5L]n+ H,O {L = NH,, H,O, or C1-)

hv

+

[M(NH3)5(H20)]3+ L(n-3)+ (35)

The production of metal nitrene complexes on irradiation ( A > 214 nm) of [Rh(NH3)6N3]2+in acidic solution has been reinvestigated.14' When account is taken of the photolysis of [Rh(NH3)6(NH2C1)]3+ it can be demonstrated that process (36) is the only significant mode of reaction; no evidence was found for

+ H+ [Rh(NH,),NHJ3+ + HCl [Rh(NH3),N,lZ+

hv

[Rh(NH,),NHl3+

+ N,

[Rh(NH3)5(NHzCl)]3+

(36) (3 7)

redox decomposition such as is observed for [CO(NH,),(N,)]~+. A qualitative treatment of the energetics of the system indicates that decomposition to the nitrene complex should be energetically favourable for both the cobalt and rhodium compounds, and it is suggested that d,--p, bonding in the excited state of the azide complex promotes production of nitrene products by inducing mixing of the LMCT and ligand-centred excited states. The nature of the state responsible for metal nitrene production, however, is put in doubt by recent experiments las 146

G. A. Shagisultanova and V. V. Yasinetskii, Zhur.$z. Khim., 1976, 50, 91. A. W. Zanella, M. Talebinasab-Sarvari,and P. C. Ford, Znorg. Chem., 1976, 15, 1980. T. Inoue, J. F. Endicott, and G. J. Ferraudi, Znorg. Chem., 1976, 15, 3098.

Photochemistry

212

with [M(CN),N3I3- (M = Rh or Ir).14* Irradiation ( A = 313 nm) in absorption bands assigned to azide ligand-centred excited states led not to nitrene formation but to azide aquation (38) (a = 0.42 for Rh, 0.60 for Ir). Although this effect

-

+

[M(CN)5NJ3- + HZO [M(CN)5(H,0)]2- N,(38) may be due to weaker back-bonding in the nitrene due to the winteraction of the cyanide ligands, the contrast with the reactions of the penta-ammine complexes is nevertheless surprising. A Rh"' superoxo-complex is formed upon irradiation as shown in equation hv

(39).'49

hv

[Rh(en>2(No2)21+ + 0, ---+ [Rh(en)2(No,)(02)l++ NO2 (39) Recent publications have considered the luminescence of bis-chelated derivatives of Ir ''I, including [Ir(bipy),(H,O)( bipy)]3+,150* 151and of trans- [R hBr2(4-Mepy),]+ and trans-[RhX,(py),]+ (X = C1 or Br).14, Nickel.-Pulse laser irradiation of [Ni(dpp)Br,] [dpp = 1,3-bis(diplienylphosphino)propane] in dichloromethane perturbs the equilibrium between its square-planar and tetrahedral forms.152 In acetonitrile, monitoring of the absorbance and conductivity following the pulse reveals that ionization is also occurring in this more polar solvent [equation (40)]. sp-[Ni(dpp)Br,]

-

7 tet-[Ni(dpp)Br,] ~[Ni(dpp)Br]+,Br~[Ni(dpp)Br]++ Br-

(40) Other publications have considered the photoreduction of Ni2+ in methanol s01utions,~~~ the photoisomerization of the azo-substituent of a Nil' Schiff-base and the quenching of organic singlet and triplet states by [Ni(4-Mepy),CI2] and [Ni(4-Mepy)a(NCS)2].155

Platinum.-Transient Pt"' complexes have been formed both by the reaction of pulse-radiolytically generated hydroxyl radicals with Pt" complexes and by flash photolysis of Pt'" species.15s It is proposed that the first transient observed on photolysis of [Pt(en),Cl2I2+is the distorted octahedral trans-[Pt(en),C1(H20)l2+. Photoredox decomposition of [PtN4X2I2+(N4 = 4NH3, 4MeNH,, 4EtNH,, 2en, or 2 x 1,3-diaminopropane; X = C1, Br, or I) has been reported.ls7 Photosolvation (0= 0.1) of cis-[Pt(l-naphthylamine),CI,] in aqueous DMF solution occurs upon irradiation into the intra-ligand band (Amx = 293 nm) [reaction (41)].158 No reaction is found after direct excitation of the LF states. V. M. Miskowski, G. L. Nobinger, and G. S. Hammond, Inorg. Chem., 1976, 15, 2904. R. D. Gillard, J. D. Pedrosa de Jesus, and L. R. H. Tipping, J.C.S. Chem. Comm., 1977, 58. 160 R. J. Watts, B. Griffith, and J. S. Harrington, in ref. 4, p. 201. 161 R. J. Watts, J. S. Harrington, and J. Van Houten, J. Amer. Chem. Sac., 1977, 99, 2179. 15a L. Campbell and 5. 5. McGarvey, J.C.S. Chem. Comm., 1976, 749. 163 E. F. Abdrashitov and L. A. Tikhomirov, Khim. vysok. Energii, 1976, 10, 401. D. P. Fisher, V. Piermattie, and 5. C. Dabrowiak, J. Amer. Chem. Soc., 1977, 99, 2811. lSb A. Guarino, G. Occhiucci, E. Possagno, and R. Bassanelli, J . Photochem., 1976,5, 415. lb6 J. C. Brodovitch, D. K. Storer, W. L. Waltz, and R. L. Eager, Internat. J. Radiation Phys. and Chem., 1976, 8, 465. lb7 A. V. Loginov and G. A. Shagisultanova, Zhur. priklad. Spektroskopii, 1976,49, 2353. 16* A. Vogler and A. Kern, Angew. Chem., 1976, 88, 686; Angew. Chem. Internat. Edn., 1976, 15, 625. 148

ld8

-

Photochemistry of Inorganic and Organometallic Compounds

cis-[Pt(l-naphthylamine),Cl,]

+ solvent

21 3

hV

cis-[Pt (1-napht hy lamine)(S)Cl,]

+ 1-naph t hy lamine

(4 1)

This labilization effect is attributed to the reduced basicity of the amine in the excited state. The photoisomerization of [Pt(PEt,),CI,] has been re-examined in greater detai1.lS9 In degassed solution the quantum yields (@)trans-eis = 1.0, @cis-trans 0.09) are essentially independent of wavelength. The transcis interconversion is partially quenched by oxygen or piperylene; however, 13% remains unquenchable even at higher quencher concentrations. Intense emission from [pt(OP(OH),},(P(OH),}~] in aqueous solution at room temperature 160 and activation-energybarriers for intersystem crossing in platinum phthalocyanine have been the subjects of recent articles. Copper.-Suspensions of cuprous chloride in water are photochromic, turning green in the presence of U.V. light.f62~1e3 This is due to the formation of fine particles of copper of grain size 3 nm [equation (42)]. 2CuCl

A

cuo

+ cu2+ + 2c1-

(42)

Excitation of the MLCT band of [Cu(2,9-diMephen),]+ in the presence of ColI1 complexes leads to oxidation of the copper complex and production of Co2+,probably by electron transfer from the CT excited state to the acceptor complex.164 Other recent papers have considered the photoredox reactions of Cull carboxylates at 77 K,166the photopolymerization of acrylamide by bis(g1utamato)the use copper(II),166the use of [Cu(acac),] as a light stabilizer for polyamide~,~*~ of Cu2+and benzophenone as a catalyst system for the photodehydrogenation of cyclohexane,ls8 and the photochemistry of $?-unsaturated ketones 16@ and isoxazoles in the presence of Cua+. Silver and Gold.--Photoreduction of Ag+ has been monitored in aqueous solution and in alcohols.171 As has previously been observed for photoreactions of the CT state of Fe3+,this reaction is most efficient with alcohols having an a-hydrogen, such as ethanol or propan-2-01, but inefficient for t-butyl alcohol. Photoreduction of silver ions in zeolites (ZO- is the zeolite lattice) gives concomitant oxygen evolution [equation (43)].17a On heating to 873 K, hydrogen is released. S. H. Goh and C. Y. Mok, J. Inorg. Nuclear Chem., 1977, 39, 531. R. P. Sperline, M. K. Dickson, and D. M. Roundhill, J.C.S. Chem. Cornm., 1977, 62. l S 1 T.-H. Huang, K. E. Rieckhoff, and E. M. Voigt, Canad. J. Phys., 1976, 54, 633. la* B. Carlsson and G. Wettermark, J. Photochem., 1976, 5, 321,421. 188 B. Carlsson, C. Leygraf, and G. Hultquist, J. Photochem., 1977, 7 , 51. 164 D.R. McMillin, M. T. Buckner, and B. T. Ahn, Inorg. Chem., 1977, 16, 943. 166 V. K. Rumas and A. L. Poznyak, Doklady Akad. Nauk Belarus. S.S.R., 1976,20, 1106. l a o P. Natarajan, K.Chandrasekaran, and M. Santappa, J. Polymer Sci.,Polymer Letters, 1976, lS9

u0

14,455. 167

A. L. Margolin, I. A. Kabanova, L. M. Postnikov, and V. Y. Shlyapintok, Vysokomol.

Soedineniya, Ser. B, 1976, 18, 378. G.H.Jones, D. W. Edwards, and D. Parr, J.C.S. Chem. Cornm., 1976,969. la@ T. Sato, K.Tamura, K. Maruyama, 0. Ogawa, and T. Imamura, J.C.S. Perkin I, 1976, 779. 170 T. Sato, K. Yamamoto, K. Fukui, K. Saito, K. Hayakawa, and S. Yoshiie, J.C.S. Perkin I, 1976, 783. 171 H. Hada, Y.Yonezawa, A. Yoshida, and A. Kurakake, J. Phys. Chem., 1976,80,2728. 178 P. A. Jacobs, J. B. Uytterhoeven, and H. K. Beyer, J.C.S. Chem. Cornm., 1977, 128. lea

Photochemistry

214 2Ag+

+ 2 2 0 - + H,O

2Ag0

+ 22021. +

(43)

Formation of a thin layer of colloidal silver is the first stage in the Iightinduced decomposition of silver azide.17, The photochemical redox reactions of [AuCI,]- and carboxylic acids have been investigated.17, Zinc.-A pronounced enhancement of the phosphorescence intensity and diminution of the fluorescence of bis-(N-salicylidenealkylamine)zinc(II) complexes compared with the monosubstituted derivatives is attributed to enhanced ISC through interaction of the two ligands with each NN’-(4,4’-Sulphonyldipheny1ene)dimaleimide photopolymerizes in the presence of ZnBr2.17s Mercury.-The observation that photolysis of various amino-acids in the presence of HgClz gives methylmercury derivatives may be environmentally i ~ p 0 r t a n t . l ’ ~ Lanthanides.-The general interest in photochemical decomposition of water has prompted several groups of workers to examine the photolysis of Eu2+ in acid solution in more detai1.178-1so The quantum yield for hydrogen evolution in perchloric acid solutions (8.7 x 10-3-3.74 mol drn-,) varies with the square root of the hydrogen-ion c o n c e n t r a t i ~ n .This ~ ~ ~ dependence is that predicted by the Noyes theory for reaction of H+ with the geminate radical pair [EuOH,HI2+ formed from the excited state. 2Eu2+ + 2H+

hv

2Eu3+

+ H,

(44)

In another recent study it has been demonstrated that in strong solutions the perchlorate ion may not always be inert.179 Thus at 77 K in perchloric acid (9 mol dm-3) glasses, photolysis of Eu2+produces C103, which was identified by its u.v.-visible absorption spectrum [reaction (45)]. No reaction was observed at perchloric acid concentrations of 0.2 mol drn-,. Eu2+

+ HCIOd

hv

Eu3+

+ CIO, + OH-

(45)

An absorption band observed in the U.V. spectrum (Amax = 225 nm) after photolysis of Eu2+ in 8 mol dm-3 hydrochloric acid glasses at 77 K has been assigned to hydrated hydrogen atoms,lso although it is markedly different from that previously reported, but similar to that of the HOz*radical (Amx = 225 nm, Emax = 1200).le1 Investigation of the influence of thiocyanate ion on the intensity and lifetime of the emission from both the 5D1 and 6Do levels of [Eu(CF,COCHCOCF,),] shows that the quenching occurs by interaction with the metal-centred state and A. V. Dubovitskii, E. V. Prokhorin, and G . B. Manelis, Khim. uysok. Energii, 1976,10, 59. B. S. Maritz and R. Van Eldik, J . Znorg. Nuclear Chem., 1976, 38, 2124. 176 T. Ohno, S. Kato, A. Takeuchi, and S. Yamada, Bull. Chem. SOC. Japan, 1977, 50, 6. 17* D. C. Phillips, D. H. Davies, and J. A. Jackson, Makromol. Chem., 1976, 177, 3049. 177 K. Hayashi, S. Kawai, T. Ohno, and Y. Maki, J.C.S. Chem. Comm., 1977, 158. 178 D. D. Davis, K. I. Stevenson, and G. K. King, Inorg. Cltem., 1977, 16, 670. 170 V. P. Kazakov, R. G. Bulgakov, and M. M. Konoplya, Khim. vysok. Energii, 1976, 10, 181. lB0 V. V. Korolev and N. M. Bazhin, Chem. Phys. Letters, 1976, 43, 469. 181 P. Pagsberg, H. Christensen, J. Rabani, G . Nilsson, J. Fenger, and S. 0. Nielsen, J. Phys. Chem., 1969,73, 1029. 178

174

Photochemistry of Inorganic and Organometallic Compounds

215

not with the l(n-m*) ligand state as previously postulated.ls2 Other authors have also observed quenching of the 6Do level of Eu3+ by CNS- caused by complex formation in water, methanol, or acetone Energy transfer between [Tb(thd),] and [Eu(thd),] (thd = ButCOCHCOBut) has been monitored in a variety of Lifetime and intensity data indicate that there is no significant contribution from static quenching due to formation of [Tb(thd),]-[Eu(thd),] dimers. Other authors have reviewed work on energy transfer between cationic donors and rare-earth ions la6and discussed the sensitization of EuS+ luminescence in AMP or ATP complexes (ATP = adenosine 5'-triphosphate) at room temperature.la8 The circularly polarized emission (CPE) from Eu3+ and Tbs+ in L-malic acid complexes has been examined in D,O and H 2 0 solutions as a function of pH.ls7 Other systems for which CPE data have recently been published are tris(3-trifluoro-d-camphorato)europium(111),~~~ tris-(p-diketonato)europium(Irr) complexes in optically active s01vents,~~~ and Tb"' in substituted bovine cardiac troponin C.lDo The observed pressure-induced variation of the quantum yield of luminescence is attributed to a lowering of the triplet from [Eu(PhCOCHCOMe),]-[Hpip]+ energy level (by ca. 1500 cm-1 at 55 kbar).lQ1 Luminescence from filter-paper adsorbates of lanthanide-IS-diketone complexes has been described.lQ2 The effect of solvent deuteriation on the rate of radiationless decay has been examined for Sms+in water, DMSO, and acetonitrile,lQ3and for Dys+ in DMSO, acetone, and a c e t ~ n i t r i l e .The ~ ~ ~lower rate in deuteriated solvents is predicted by the modified energy-gap theory of radiationless fransitions,lQSbut the role of anharmonicity is invoked to explain the effect for deuteriated a c e t ~ n i t r i l e . ~ ~ ~ Luminescence data, particularly lifetime determinations, have been used to establish the nature of the primary solvation shell for lanthanides in aqueous propan-1-01 ~ 0 1 u t i o n s ,as ~ ~probes ~ for the metal environment in lanthanide edta and enzyme complexes,1Beand in polymer^.^^^^ lD8 Other publications describe the emission characteristics of rare-earth complexes of 1,8-naphA. J. Twarowski and D. S. Kliger, Chem. Phys. Letters, 1976, 41, 329. V. P. Gruzdev and V. L. Ermolaev, Optika i Spektroskopiya, 1977, 42, 781. H. G. Brittain and F. S . Richardson, J.C.S. Faraday II, 1977,73, 545. R. Reisfeld, Structure and Bonding, 1976, 30, 65. R. Reisfeld, S. Nathanson, and E. Greenberg, J. Phys. Chem., 1976, 80, 2538. H. G. Brittain and F. S . Richardson, Inorg. Chem., 1976, 15, 1507. H. G . Brittain and F. S. Richardson, J. Amer. Chem. SOC.,1976,98, 5858. H. G. Brittain and F. S. Richardson, J. Amer. Chem. SOC., 1977, 99, 65. lgO H. G. Brittain, F. S. Richardson, R. B. Martin, L. D. Burtnick, and C. M. Kay, Biochem. Biophys. Res. Comm., 1976, 68, 1013. lgl V. A. Voloshin, A. I. Savutskii, and A. I. Kas'yanov, Optika i Spektroskopi-va, 1976,40,607. lQ8 S . B. Meshkova, L. I. Kononenko, A. A. Kucher, and N. S. Poluektov, Doklady Akad. Nauk S.S.S.R., 1976,227, 384. laa G . Stein and E. Wuerzberg, 2.phys. Chem. (Frankfurt), 1976,101, 163. lo4 V. L. Ermolaev, E. B. Sveshnikova, and V. S. Tachin, Optika i Spektroskopiya, 1976,41,343. ln6A. I. Krutous and I. M. Batyaev, Koord. Khim., 1976,2, 1041.

lS2

lE3

lB*

lQ8

W. D e W. Horrocks. G. F. Schmidt, D. R. Sudnick, C. Kittrell, and R. A. Bernheim, J . Amer. Chem. SOC., 1977, 99, 2378. S. L. Davydova, V. A. Barabanov, L. G. Koreneva, and V. F. Zolin, Izoest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 2673. V. F. Zolin, L. G. Koreneva, V. A. Barabanov, and S. L. Davydova, Koord. Khim., 1976,2, 695.

216

Photochemistry

thyridines lQ9 and the luminescence properties of lanthanide ions in POC1,-MCI, mixtures.200,201 The luminescence properties of a large number of rare-earth porph yrin complexes have been investigated.202 The emission characteristics of these compounds are governed by the relative energies of the porphyrin singlet and triplet excited states and those of the metal. Thus when the S1and TI levels of the porphyrin are below those of the metal, fluorescence and phosphorescence are normally observed, although the fluorescence is weak because of enhanced ISC due to spin-orbit coupling and/or the paramagnetism of the metal. However, with Nd and Yb complexes, where the metal-centred excited states are below the porphyrin S1and TI levels, no porphyrin emission, only that characteristic of the metal, is observed. Intermolecular transfer of energy to ytterbiumaetioporphyrin from porphyrins having their triplet levels lower in energy than that of the acceptor has been monitored by flash photolysis and emission In this case energy transfer, presumably via an exchange mechanism, directly populates the emitting 2F5/2level of the ytterbium. Complexes of Y 'I1, Lu"', and ThIV of meso-tetraphenylporphyrin are noteworthy in exhibiting fluorescence from the second excited singlet In the case of the Th" compound this emission is actually more intense than that from the lowest singlet state. Actinides.-The tunnel-effect theory of radiationless transitions has been applied to hydrogen-abstraction processes by the excited state of the uranyl ion *[U02]2+.205Good agreement between the theoretical and experimental values was found for the rate constants of quenching of *[UO2l2+by various alcohols and ethers. It has also been predicted that reversible hydrogen abstraction is the principal route for the deactivation of *[U02]2+in aqueous solution at room temperature. Chibisov and co-workers have published a full account of their flash photolysis studies of the photochemical reactions of [U02]2+.206The recording of the absorption spectra of the transient *[UO2I2+and Uv species has allowed them to obtain quantitative data on the physical (46) and chemical (47) deactivation routes for the excited states in the presence of quenchers such as alcohols, amines, and organic acids.

+

*Uv* Q *Uvl + Q

---+

Uv' Uv

+ *Q + Q'+

(46) (47)

Electron transfer (47) is responsible for the quenching of uranyl ion luminescence by halide ions.2o7,208 Thus, on flash photolysis of UOZ2+in the leg

201

202

203 204

206

206

207 208

M. Ngsee, H. W. Latz, and D. G . Hendricker, in ref. 4, p. 214. J. Chrysochoos and P. Tokousbalides, J. Luminescence, 1976, 14, 325. B. Jezowska-Trzebiatowska, W. Ryba-Romanowski, Z. Mazurak, and K. Bukietynska, Chem. Phys. Letters, 1976, 43, 417. K. N. Solov'ev, M. P. Tsvirko, and T. F. Kachura, Optika i Spektroskopiya, 1976,40, 684. M. P. Tsvirko, K. N. Sokol'ev, and V. V. Sapunov, Optika i Spektroskopiya, 1976, 40,843. L. A. Martarano, C.-P. Wong, W. De W. Horrocks, and A. M. Ponte-Goncalves, J. Phys. Chem., 1976,80,2389. H. D. Burrows and S. J. Formosinho, J.C.S. Faraday ZZy 1977,73, 201. G . I. Sergeeva, A. K. Chibisov, L. V. Levshin, and A. V. Karyakin, J. Photochem., 1976, 5, 253. H. D. Burrows and J. D. Pedrosa de Jesus, J. Photochem., 1976,5,265. Y . Yokoyama, M. Moriyasu, and S. Ikeda, J . Znorg. Nuclear C'hem., 1976, 38, 1329.

Photochemistry of Inorganic and Organometallic Compounds

217

presence of I-, CNS-, or Br-, the species 12*-, (CNS),'-, and Br,*- have been The light-induced exchange reaction between labelled Uvl and UXv species in 8 mol dm-3 hydrochloric acid proceeds via initial production of Uv as shown in equation (48).,09

*U"'+ U'"

__I,

2uv

(48)

Although uranium(v) complexes normally disproportionate in aqueous solutions, photolysis of [UO,Cl,(py),] in dry ethanol provides a convenient route to UOC13.210 More extended irradiation gives [U(OEt),]. It is postulated that the first stage in the reaction is photochemical production of U02CI. A detailed study of the photolysis of uranyl oxalate has been presented.211 The photoactive species has been identified as [UO,(oxH),]. Related topics recently reported are the decomposition of uranyl formate in the solid state2', and the photoreduction of [U02]2+by citrate ions213 and by tri-n-butyl phosphate.214 Other publications discuss the possible application of [UO2lZ+in solar energy conversion,215the use of uranyl nitrate as a sensitizer for preparation of cellulose-styrene graft and the photochemical formation of furan derivatives by irradiation of $-unsaturated ketones in methanol containing U02C12.217 The luminescence intensity of [UO2I2+ increases with H+ concentration.218 The effect is attributed to changes in the co-ordination sphere of the ion which lead to reduced efficiency of deactivation. Other recent papers have described the polarized luminescence of uranyl corn pound^.^^^^ 220 U.V. irradiation increases the extent of disproportionation of plutonium(1v) in acidic 3Pu4++ 2H20

2Pu3+ + [PuO2I2++ 4H+

(49)

2 Transition-metal Organometallics and Low-oxidation-state Compounds Recent reviews have considered the photochemistry of transition-metal organometallics and its biological implications,222possible applications of organo209 210

212

213 214 216

216 a17

m 219

220

221 222

Y. Kato and H. Fukutomi, J. Inorg. Nuclear Chem., 1976,38, 1323. S. Sostero, 0. Traverso, C. Bartocci, P. di Bernardo, L. Magon, and V. Carassiti, Inorg. Chim. Acta, 1976, 19, 229. A. G. Brits, R.Van Eldik, and J. A. Van den Berg, Z. phys. Chem. (Frankfurt), 1976,102,203. B. Claude1 and 5. P. Puaux, Compr. rend., 1976, 282, C, 571. A. Ohyoshi, Asahi Garasu Kogyo Gijutsu Shoreikai Kenkyu Hokoku, 1975, 26, 57 (Chem. Abs., 1976, 85, 134 106). C. K. Rofer-Depoorter and G. L. Depoorter, J. Inorg. Nuclear Chem., 1977,39, 631. C. K. Joergensen, Naturwiss., 1977, 64, 37. N. P. Davis, J. L.Garnett, and R. Urquhart, J. Polymer Sci., Polymer Letters, 1976, 14, 537. T. Sato, 0. Ito, and M. Miyahara, Chem. Letters, 1976, 295. V. P. Kazakov, R. G. Bulgakov, Y. E.Nikitin, and G. L. Sharunov, Optika i Spektroskopiya, 1976,41, 1091. V. E. Karasev, in 'Inorganic and Nuclear Chemistry - Herbert H. Hyman Memorial Volume', ed. J. J. Katz, Pergamon, Oxford, 1976, p. 199. L. V. Volod'ko, A. 1. Komyak, M. R. Posledovich, and S. E. Sleptsov, Izvest. Akad. Nauk S.S.S.R., Ser.fiz., 1975, 39, 2235. H. A. Friedman, L. M. Toth, and 5. T. Bell, J. Inorg. Nuclear Chem., 1977, 39, 123. E. A. Koerner von Gustorf, L. H. G. Leenders, I. Fischler, and R. N. Perutz, Adu. Znorg, Chem. Radiochem., 1976, 19, 65.

218

Photochemistry

metallics to photo-imaging co-ordination compounds as photocatalysts,224and photoactivated catalytic hydrogenation with transition-metal

As in previous Reports, a selection of recent examples of photochemical substitution reactions of metal carbonyls is presented in Table 3 (p. 237). Titanium and Zirconium.-Earlier work has shown that photolysis of [Cp,TiMe,] (Cp = q6-cyclopentadienyl) in the presence of diphenylacetylene yields the metallocycle (9). It has now been demonstrated that under similar conditions Ph

Ph (9)

the mono-insertion product [Cp,TiMe(CPh= CPhMe)] is also produced (in 18% yield).226With C,F,C=CC,F, no metallocyclic species is formed, and the mono-insertion product [CPzTiMeiC(C,F,)=C(C,F,)Me)] is the only organometallic product. Irradiation of [Cp,ZrMe,], [Cp,ZrPh,],227 or [Cp,Zr(CPh= CMe,),] in solutions of chlorinated hydrocarbons induces cleavage of the Zr-C bond, and leads eventually to [Cp,ZrCI,]. Similarly, dissociation of the Ti-Ge is the primary photochemical process upon irradiation of [Cp2Ti(GePh3),] and related The methylation of co-ordinated carbon monoxide [reaction (50)] presumably proceeds by reaction of photochemically produced [(q5-C,Me5),Zr(CO)] with hydrogen. 230 [(q5-C5Me,),Zr(CO),]

+ 2H2

hv

[(q5-C5Me5)Zr(OMe)H]

+ CO

(50)

Chromium, Molybdenum, and Tungsten.-Our knowledge of the properties of co-ordinatively unsaturated compounds formed on photolysis of metal carbonyls owes much to matrix-isolation studies with the Group VI metal hexacarbonyls at low temperatures. Absolute quantum yields for such reactions are difficult to determine under these conditions, but in a recent paper the relative quantum efficiencies for reactions (51) have been It was found that PhCH2CI CH2C12.250From changes in the i.r. spectra it has been deduced that photolysis of [CpM(CO),], (M = Mo or W) in methyltetrahydrofuran glasses at 30 or 80 K also leads to rupture of the metal-metal bond.251 The isocyanide complexes [M(CNR),] [M = Cr, Mo, or W, R = Ph or iPh (2,6-di-isopropylphenyl)] are formal analogues of the hexacarbonyls, and like them undergo photosubstitution in the presence of pyridine, albeit with lower quantum yields [reaction (62)].252 However, the sensitivity of the quantum yields for the substitution of the molybdenum and tungsten compounds to the = 0.011; for [W{CN(iPh)},] @ = 0.0003) bulk of the ligand (for [W(CNPh),] suggests to the authors that the reactio‘n has associative character. Another interesting feature of these complexes is that they are all strongly luminescent at 77 K, and the molybdenum and tungsten species also emit at room temperature in fluid solution. These observations are consistent with the lowest excited state being MLCT in character. Further evidence for the dipolar nature of this state is the efficient oxidation of the complexes in chloroform solution [reactions (63) and (64)]. In both cases the initial process is probably electron transfer (65). 250

262

R. M. Laine and P. C. Ford, Inorg. Chem., 1977, 16, 388. D. M. Allen, A. Cox, T. J. Kemp, and Q. Sultana, Inorg. Chim. Acta, 1977, 21, 191. K. R. Mann, H. B. Gray, and G. S. Hammond, J. Amer. Chem. SOC.,1977,99, 306.

Photolysis of the metallocyclobutane derivative (19) yields ethylene as the principal product, in contrast with the thermal It is postulated that the primary photochemical process is production of the q3-cyclopentadienyl derivative (20), which appears to be in equilibrium with the olefin-carbene complex (21) (Scheme 2). This latter reaction is similar to the essential step previously proposed in catalytic olefin dismutation.

(19)

k

Scheme 2

Other authors have described the cis to trans isomerization of [M(CO),(carbene),] c o r n p l e ~ e s , ~ ~the ~ - ~preparation ~* of an active catalyst for olefin dismutation by photolysis of a mixture of [W(CO),{C(OEt)Ph}] and TiC11,267 the polymerization of phenylacetylene induced by photodecomposition of [(mesitylene)M~(CO)~],~~~ and the production of a photocatalyst by irradiation of [Mo(CO),] and amberlite.26s Mechanistic studies confirm that the light-induced interaction of [Mo(N,),(dppe),] [dppe = 1,2-bis(diphenylphosphino)ethane] is initiated by photoexpulsion of dinitrogen from the complex.260s 868

264

m

268

269 260

281

M. Ephritikhine and M. L. H. Green, J.C.S. Chem. Comm., 1977,926. K. Oefele, E. ROOS, and M. Herberhold, 2.Naturforsch., 1976, 31b, 1070. M. F. Lappert, P. L. Pye, and G. M. McLaughlin, J.C.S. Dalton, 1977, 1272. M. F. Lappert and P. L. Pye, J.C.S. Dalton, 1977, 1283. Y . Chauvin, D. Commereuc, and D. Cruypelinck, Mukromol. Chem., 1976,177,2637. L. M. Svanidze, E. A. Mushina, A. M. Sladkov, N. I. Sirotkin, A. K. Artemov, I. R. Gol’ding, T. G . Samedova, G. N. Bondarenko, and B. E. Davydov, Vysokomol. Soedineniya, Ser. B, 1977, 19, 51. S. Ivanov, R. Boeva, and S. Tanielyan, Reaction Kinetics and Catalysis Letters, 1976, 5, 297. J. Chatt, A. A. Diamantis, G. A. Heath, N. E. Hooper, and G . J. Leigh, J.C.S. Dalton, 1977, 688. J. Chatt, R. A. Head, G. J. Leigh, and C. J. Pickett, J.C.S. Chern. Comm., 1977, 299.

224

Photochemistry

Manganese and Rhenium-There is now general agreement that the dominant decomposition process for photoexcited [Mn,(CO),,] is cleavage of the metalmetal bond [reaction (66)], and this is confirmed by recent report^.^^^-^^* The recombinatioii process (67), monitored by flash photolysis, is rapid in both

cyclohexane (k = 4 x lo9dms mol-1 s-l) and THF (k = 2 x logdms mol-1 s-1).262 However, small amounts of a long-lived photoproduct are also formed, and it is suggested that it is secondary photolysis of these species which leads to the permanent photodecomposition previously observed. The radical Mn(CO), may be scavenged by the spin-trapping agents nitrosodurene 263 or 2,4,6-trit-butylnitro~obenzene,~~~ and the product [Mn(CO),{fi(O)Ar)] has been identified by e.s.r. Other spin-trapped paramagnetic species are produced under different experimental conditions. For example, if the concentration of spin-trap is is observed, this deficient, a signal attributed to [{Ar(0)I%}(OC)4MnMn(CO)6]+ possibly indicating a reaction of type (68).a63 E.s.r. investigations confirm that

in basic solvents the disproportionation of the Mn(CO), radical produces Mn" [Mn,(CO),,]- has been identified by e.s.r. following 60Coy-ray irradiation of [Mn,(CO),,I at 77 K.265 Photoreaction of [Mn,(CO),,] and hydrogen gives [HMn(CO),], whereas with [Re,(CO),,], depending on the excitation wavelength, [HRe(CO),], [H,Re,(CO),], or [H3Re3(C0)14]is produced.266 These and other substitution experiments substantiate previous proposals about the lability of the Group VII metal pentacarbonyls. A detailed kinetic study of the polymerization of tetrafluoroethylene photoinitiated by [Mny(C0),,] reveals that the propagating species is Mn(CO),CF2CFz*.28~ 267 Irradiation of mixtures of [Mn,(CO),,] with cyclohexadienes or 260 butadiene yields small amounts of substitution 1.r. analytical data on the products of photolysis of [M(CO),X] (M = Mn or Re, X = Cl or Br) in methyltetrahydrofuran glasses at 77 K demonstrate that CO is expelled from the manganese complexes but not from their rhenium analogues.251 The results of a thorough investigation of the photosubstitution reactions of [CpM(CO),L] (M = Mn or Re, L = amine) have been communicated.270The study shows that the only reaction observed is amine substitution (69). This 262

263 264 286 266

267

a68 26e

J. L. Hughey, C. P. Anderson, and T. J. Meyer, J. Organometallic Chem., 1977, 125, C49. A. Hudson, M. F. Lappert, and B. K. Nicholson, J.C.S. Dalton, 1977, 551. A. S. Huffadine, B. M. Peake, B. H. Robinson, 5. Simpson, and P. A. Dawson, J. Organometallic Chem., 1976, 121, 391. S. W. Bratt and M. C. R. Symons, J.C.S. Dalton, 1977, 1314. B. H. Byers and T. L. Brown, J. Amer. Chern. SOC.,1977, 99,2527. C. H. Bamford and S. U. Mullik, Polymer, 1976, 17, 225. M. Zoeller, H. E. Sasse, and M. L. Ziegler, 2. anorg. Chem., 1976, 425, 257. M. Zoeller and M. L. Ziegler, 2. anorg. Chem., 1976, 425,265. P. J. Giordano and M. S. Wrighton, Znorg. Chem., 1977, 16, 160.

-

Photochemistry of Inorganic and Organometallic Compounds [CpM(CO),L] + Y

7tV

[CpM(CO)2Y]

225

+L

(69)

behaviour is consistent with the observation that trisubstituted products [Cp,MnL,] are only formed when the entering ligand has r-bonding characteristics similar to those of carbon monoxide. The quantum yields for reaction (69) are dependent on the nature of the metal and the ligand L but independent of both the nature and concentration of the reactant Y, as expected for a photodissociative mechanism. Although the quantum yield for substitution of [CPM~(CO)~L] is high (0.14-0.40) for all derivatives, those for [CpRe(CO),L] complexes depend markedly on the electron-accepting ability of the ligand (e.g. for L = 4-methylpyridine, 0 = 0.29; for L = 4-acetylpyridine, 0 < This variation with the nature of L is attributed to the presence of a low-lying MLCT excited state. As in the analogous case of [W(CO)gL],240 these MLCT excited states appear to be substitutionally unreactive. The species responsible for reaction (69) is probably a triplet LF state, even though the MLCT band is lowest in the absorption spectra for most of the derivatives considered. In the

_________-

ground state

---- ------

high reactivity

intermediate reactivity

no reactivity

(L = piperidine, THF, CO,

(L=3- Br - py

(L= 4 - acetyl- py 4 - benroyl py, etc.)

4

3,5-diCl

- py)

-

- Me - PY, PY,

etc.1

Figure 2 Relaxed excited-state energies and photoreactivities of [CpRe(CO),L] derivatives (Adapted by permission from Inorg. Chem., 1977, 16, 160)

case of the rhenium derivatives the marked dependence of quantum yield on L is presumably indicative of a changeover in the nature of the lowest-lying thermally equilibrated excited state (Figure 2). Photochemical synthesis of optical isomers of [(Me(CO,Me)Cp)Mn(CO)(PPh,)(P(OMe),)] 271 and the preparation of bridged complex of the type

r-----l

[Cp(CO),Mn-AsMe,-M] been reported.

[M = Mn(CO),, Re(CO),, or Co(C0)J

873

have

Iron and Ruthenium.-The absence of solvated electrons following flash photolysis of either acetylferrocene (22a) or benzoylferrocene (22b) 274 in aqueous or a71 F. Le Moigne, R. Dabard, and M. Le Plouzennec, J. Organometullic Chern., 1976,122,365. B7a

U. Richter and H. Vahrenkamp, J. Chem. Res. (9, 1977, 156. E. K. Heaney and S. R. Logan, Inorg. Chim. Actu, 1977,22, L3. E. K. Ijeaney, S. R. Logan, and J. A. Powell, J.C.S. Furuday I, 1977,73, 699.

Photochemistry

226 O‘.J-.’ C O

i

R

Fe

I

(22) a; R = Me

b; R = Ph

alcoholic solutions demonstrates that, contrary to what had been previously photoionization is not an important reaction for these compounds [reaction (70)]. It is suggested that the photoprocess leading to decomposition involves intramolecular charge transfer. [CpFe(CpCOR)]

1LV

--x+

+

[CpFe(CpCOR)]+ e-

(70)

Other reports on the photochemistry of ferrocene and derivatives describe the quenching of the singlet states of naphthalene, phenanthrene, triphenylene, and tetracene by f e ~ r o c e n ethe , ~ ~polymerization ~ of epichlorohydrin initiated by photolysis of ferrocene-TiCl, mixtures,277and the ferrocene-catalysed photocondensation of cyclohexyl isocyanate and butan-l-01.~~~ At present, little is known about the photochemistry of metal cluster compounds. Therefore the appearance of a recent report on [CpFe(CO)], is of special It has been shown that the compound is inert towards photosubstitution but that in the presence of halogenocarbons efficient photo-oxidation occurs [reaction (71)]. As in the analogous reaction with ferrocene, charge transfer to the halocarbon in a CTTS excited state appears to be responsible for transformation (71). [CpFe(CO)],

+ RX

hv

[CpFe(CO)],+ + X-

+ R-

(71)

Irradiation of [Fe(CO),] in carbon monoxide matrices at 20 K with planepolarized light induces partial orientation of the sample.28oThus excitation in the ( l A ; - lE’) band of the [Fe(CO),], with consequent dissociation and recombination (72), results in movements of the molecules so that they are less [Fe(CO),]

hv

[Fe(CO),]

+ CO

-

[Fe(CO),]

(72)

likely to absorb light of the particular polarization used. In another study, photochemical reactions of 13CO-enriched [Fe(CO),] in methane matrices have been initiated by i.r. radiation.281 Excitation with a CO laser, tunable in the region 1900-2000 cm-l, causes production of the methane complex [reaction (73)]. Interestingly, the laser radiation not only causes selective excitation of 276

278

280

0. Traverso, R. Rossi, S. Sostero, and V. Carassiti, MoZ. Photochem., 1973, 5 , 457. E. Penigault, A. M. Braun, and J. Faure, Compt. rend., 1976, 283, C, 655. K. Kaeriyama, J. Polymer Sci.,Polymer Chem., 1976, 14, 1547. S. P. McManus, H. S. Bruner, H. D. Coble, and M. Ortiz, J. Org. Chem., 1977, 42, 1428. C. R. Bock and M. S. Wrighton, Inorg. Chem., 1977, 16, 1309. J. K. Burdett, J. M. Grzybowski, M. Poliakoff, and J. J. Turner, J . Amer. Chem. Soc., 1976, 98, 5728. A. McNeish, M. Poliakoff, K. P. Smith, and J. J. Turner, J.C.S. Chem. Comm., 1976, 859.

-

Photochemistry of Inorganic and Organometallic Compoimcls [Fe(CO),l

+ CH,

hv (i. 1.1

[Fe(CO),(CH,)I

221 (73)

molecules with differing degrees of isotopic substitution but also of molecules with an identical degree of substitution but differing stereochemistry. A recent publication on the photolysis of matrix-isolated [Fe(CO),(NO),] at 20 K reports that photoexpulsion of CO to give the planar [Fe(CO)(N0)2]is the predominant process.282 Photolysis of [CpFe(CO),X] (X = CI, Br, or I) in methyltetrahydrofuran matrices at 77-80 K also causes CO Whereas the photochemical synthesis of [Fe,(CO),] from [Fe(CO),] is well established and may be conducted at room temperature, the analogous routes to [Os,(CO),] and [Ru,(CO),] require irradiation at low temperatures (233 K) in heptane [Fe(CO),] photocatalyses the reaction of trialkylsilanes with olefins under mild conditions.284 It is proposed that the active catalytic species is [(olefin)Fe(H)(SiRd(C0)31. Photolysis of [Fe(CO),], methyl acrylate, and 2,3-dimethylbutadiene yields compound (23).285 In further experiments it could be shown that the reaction proceeds via the photoexcitation of either [(q2-CH,=CHC0,Me)Fe(CO),] or

(23)

HyQ+i3

(24) R = H or Me

OC-RU 1 \ CF3

oc co

the [(q4-diene)Fe(CO),] complex, leading to [(q2-CH2=CHCOzMe)(q2-diene)Fe(CO),] as a common intermediate. 1 : 1 Adducts of type (24) are formed upon photolysis of various [(1,3-diene)Fe(CO),] complexes with hexafluorobut-2-yne, whereas the 1 : 2 derivative (25) is produced from [(1,3-cyclohexadiene)282

283 284 286

0. Crichton and A. J. Rest, J.C.S. Dalton, 1977, 656. J . R. Moss and W. A. G. Graham, J.C.S. Dalton, 1977, 95. M. A. Schroeder and M. S. Wrighton, J. Organometallic Chem., 1977, 128, 345. F. W. Grevels, U. Feldhoff, J. Leitich, and C. Krueger, J. Organometallic Chem., 1976, 118, 79.

228

Photochemistry

R u ( C O ) , ] . ~ ~No ~ clear distinction could be made between photoinduced CO expulsion and alternative photoformation of the [(v2-diene)M(CO),] complexes as the initial step in these reactions. A related publication gives details of the structure determination of a photoproduct (26) formed from [(1,3-cyclohexadiene)Fe(CO),] and hexaflu~ropropene.~~~ Other authors have presented a full paper on the formation of (27) from dimethyldiacetylene and [Fe(C0>,],288described the photoinduced reactions of

[Fe(CO),] with dicyclopropylacetylene,28Qand reviewed the [Fe(CO),]-catalysed formation of cyclic ketones.290 U.V. irradiation of [CpFe(CO),], with hexafluorobut-2-yne gives a ferrocyclohexadienone complex (28), the first example of such a system to be structurally By contrast [CpRu(CO),], reacts with the same acetylene to give the mononuclear complexes (29) and (3Q).248

0, CF3

oc’ oc’

RI A f H

oc CF,

CF3

(29)

Photolysis of ether solutions of [CpFe(CO),(B1,Hl3)], a compound where the borane fragment is bonded to the metal by an Fe-B single bond, yields the icosahedral metallocarbaborane [CpFe(+BloHloCOEt,) 1, apparently by CO insertion into the boron cage.291 Substitution of [Fe(CO),] by triferrocenylphosphine only occurs if the mixture is irradiated under reflux.aB2Other novel substitution reactions of [Fe(CO),] are considered in Table 3 (see p. 237).

287 28*

2g0

281 293

M. Bottrill, R. Davies, R. Goddard, M. Green, R. P. Hughes, B. Lewis, and P. Woodward, J.C.S. Dalton, 1977, 1252. R. Goddard and P. Woodward, J.C.S. Dalton, 1977, 1181. F. R. Young, D. H. O’Brien, R. C. Pettersen, R. A. Levenson, and D. L. Von Minden, J. Organometallic Chem., 1976, 114, 157. R. Victor, V. Usieli, and S. Sarel, J. Organometallic Chem., 1977, 129, 387. E. Weissberger and P. Laszlo, Accounts Chem. Res., 1976,9,209. R. V. Schultz, F. Sato, and L. J. Todd, J. Organometallic Chem., 1977, 125, 115. G. P. Sollott, D. L. Daughdrill, and W. R. Peterson, J. Organometallic Chem., 1977,133, 347.

Photochemistry of Inorganic and Organometallic Compounds

229

Recent communications give further details on the photoproduction of (3 1) from (32),293describe the photoreaction of the cationic complex [{CpFe(CO),},(H,C=C=C=CH,)]2+,294 and report on the lack of isotope effects in the photochemical format ion of [(butadiene)Fe(CO),]. 296

c~F~(co), (31)

-

A recent example of photochemical orthometallation is the solid-state reaction (74).296

[FeH,(dPPe),l,CsH,

hv

[FeH,(ceH4PPhCH,CHzP~hZ)(dPPe)lYceH6 (74)

Photoisomerization of [Ru(CO)CI,(PMe,Ph),Y] [equation (75) ;L = PMe,Ph, Y = P(OMe), or PPh(OMe),] is probably initiated by dissociation of either a

(33)

(34)

phosphine or carbon monoxide.297However, a quite different type of mechanism has been proposed for the isomerization of several analogous trimethylphosphineiron derivatives such as (35).2gs In solution, this rearrangement is accelerated by excess carbon monoxide but retarded by iodine. These observations may be rationalized by the mechanism given in Scheme 3, which is supported by isolation

(37) Sckme 3 ass 2B4 286

A. N. Nesmeyanov, N. E. Kolobova, 1. B. Zlotina, B. V. Lokshin, I. F. Leshcheva, G. K. Znobina, and K. N. Anisimov, Izuest. Akud. Nauk S.S.S.R., Ser. khim., 1976, 1124. T. E. Bauch, H. Konowitz, and W. P. Giering, J. Organometallic Chem., 1976, 114, C15. R. Noyori and K. Yokoyama, BUN. Chem. SOC.Japan, 1976,49, 1723. T. Ikariya and A. Yamamoto, J. Organometallic Chem., 1976, 118, 65. C. F. J. Barnard, J. A. Daniels, J. Jeffrey, and R. J. Mawby, J.C.S. Dalton, 1976, 1861. M. Pankowski and M. Bigorgne, J. Orguuornetallic Chem., 1977,125 :231.

Photochemistry

230

of (37) from the solid-state photolysis of (35). Photoisomerization of [RuCIH(CO),(PPh,),] has also been described.lBO Irradiation of [RuCIH(CO)(PPh,),] leads to CO release and formation of the active hydrogenation catalyst [RuClH(PPh,),] ;2aa the usefulness of this method is restricted by the thermal reaction (76). This latter reaction is not reversed on [RuCIH(CO)(PPh,),]

+ CO

-

[RuClH(CO),(PPh,),]

+ PPh,

(76)

irradiation, isomerization of the dicarbonyl complex being the only process observed. Hydrogen is evolved upon excitation of the dihydride complex [RuHdCO)(PPh&I. Cobalt, Rhodium, and Iridium.-The only photochemical reaction observed ([14]aneN, = formula (38)) on irradiation of [C0([14]aneN~)(OH,)Me]~+

is homolytic cleavage of the Co--(methyl) bond [reaction (77)].800 The quantum yield for the reaction (0.30) is essentially independent of wavelength (250 < h < 540 nm), the threshold energy therefore being less than 18 000 cm-l [Co([l4]ane Na)(OH2)Me12+

hv

+ Me- + HzO

[Co([l4]ane Na)I2+

(77)

(217 kJ mol-l). The reaction is induced upon irradiation into absorption bands much lower in energy than any associated with charge transfer. This behaviour is in marked contrast with the redox reactions of other Co"' complexes such as acidopenta-ammines, where redox decomposition is principally associated with LMCT bands and where the quantum yield is often wavelength dependent. The dependence of the homolysis energy (AHB)on the covalency of the Co--(methyl) bond has been attributed to the lower difference in ligand-field stabilization energy (ALFSE) compared with that of other Co-X complexes (see Scheme 4).300,

301

It has long been suspected that the initial step on photolysis of methylcobalamin (39; R1 = Me) is also homolytic cleavage of the Co--(methyl) bond (78), and this has now been verified by flash p h o t o l y ~ i s .In ~ ~the ~ presence of oxygen, the Co"-cobalamin (B& so formed decays, following second-order kinetics, presumably due to process (80). 208 300

302

G. L. Geoffroy and M. G. Bradley, Znorg. Chem., 1977,16, 744. C. Y . M o k and J. F. Endicott, J. Amer. Chem. SOC.,1977, 99, 1276. J. F. Endicott, Inorg. Chern., 1977, 16, 494. J. F. Endicott and G. J. Ferraudi, J. Amer. Chem. Suc., 1977, 99, 243.

-

Photochemistry of Inorganic and Organometallic Compounds Co3+(g)

+5

W

+ x-(g)

1

LFSE(II1)

IP(Co)

+ Uo

lHs

AHB

co2+(g)

+ AHs

23 1

+ w g ) + x(g)

1 lHs

+ U,'

LFSE (11)

AHB

CO"'Lg( x-)(g)

Coll'b(X-)(aq)

+ EA(X)

~Co1'L,,*X>(g)

(Co"L,,*X}(aq)

Scheme 4

A review of recent developments in B12chemistry contains some discussion on photochemical reactions.808 Other publications have considered the photodecomposition reactions of phenacylcobalamin,304and of some sixteen analogues of coenzyme B12 (39) with the ligand R composed of various base and sugar moieties closely related to those of the natural 5'-deoxyadeno~yl.~~~ Me-B12 Me

Bier

+ 0,

+ OOOMe

R'

hv

BIzr

+ Me*

(78)

MeOO-

(79)

B,,,-OOMe

(80)

I R2

R3 R2-

R ofH..o / /

Me H,OH

Me:

306

H (39)

G. N. Schrauzer, Angew. Chem., 1976, 88, 465; Angew. Chem. Znternat. Edn., 1976, 15, 417. K . L. Brown, M. M. L. Chu, and L. L. Ingraham, Biochemistry, 1976,15, 1402. T. Toraya, K. Ushio, S. Fukui, and H. P. C. Hogenkamp, J. Biol. Chem., 1977,252,963.

Photochemistry

232

It has been found that acidic solutions of alkylaquocobaloximes (40) [R = Pr", Pri, n-hexyl, or Ph(CH&; L = HzO] are much more susceptible to photodecomposition than are neutral s o l ~ t i o n s . Further, ~~~ at pH = 7, the alkyl group is transformed into the corresponding 1-alkene, whereas in acidic solutions the products are those derived from free radicals. This pH dependence of the reactions has been attributed to protonation of a reactive LMCT state in acidic solution [reactions (8 1) and (82)]. The photoreactions of dihydroxy-

alkylcobaloximes have been studied as model systems for the diol dehydration reactions catalysed in vivo by coenzyme B13.307 The homolysis of the Co-C bond in [Co(CN),CH,PhI3- after excitation to the LMCT state leads to overall reaction (83) in degassed solution (@a13 = 0.13), or to reaction (84) in aerated In the latter case, the intermediacy of a peroxy-complex [Co(CN),O,CH,PhI3- is probable. 2[Co(CN)5CH,PhI3[Co(CN)5CH2PhI3-

+ 0,+ HzO

+ PhCHZCHzPh (83) [CO(CN)~OH~]~+ PhCHO + OH-

hv

2[C0(CN)5l3-

hv

(84)

Controversy still surrounds the photoinsertion reactions of oxygen into the Co-C bonds of alkylcobaloximes (85). A comparison with the thermally induced insertion reaction has led Bied-Charreton and Gaudemer to propose [(dmg),CoLRl

+ 0,

hv

C(dmg),Co(L)(OOR)I

(85)

that the primary photoprocess is not rupture of the Co-C bond but rather a direct bimolecular reaction of oxygen with the activated ~ ~ b a l o ~ i r nIne . ~ ~ ~ another study it has been shown that the quantum yield for decomposition of [(dmg),Co(py)Me] and related complexes is greater in aerated than degassed solutions.310 Oxygen insertion into the Co-C bonds of alkyl cobalt porphyrins has also been reported.311

30Q 310 311

B. T. Golding, T. J. Kemp, P. J. Sellars, and E. Nocchi, J.C.S. Dalton, 1977, 1266. B. T. Golding, C. S. Sell, and P. J. Sellars, J.C.S. Chem. Comm., 1976, 773. A. Vogler and R. Hirschmann, 2. Naturforsch., 1976, 31b, 1082. C. Bied-Charreton and A. Gaudemer, J. Organometallic Chem., 1977,124,299. G . Roewer, C. Kraetzschmar, and G. Kempe, Z . Chem., 1976, 16, 67. M. Perree-Fauvet, A. Gaudemer, P. Boucly, and J. Devynck, J. Organometallic Chem., 1976, 120, 439.

Photochemistry of Inorganic and Organometallic Compounds

233

By studying the i.r. spectra of [Co,(CO),] before and after photolysis, it has been demonstrated that [Co,(CO),] exists in three isomeric forms in lowtemperature argon and hexane matrices.31a The only photoproduct observed is [CO,(CO),].~~~ Upon irradiation into bands associated with a non-bridged isomer of [Co,(CO),], where cleavage of the Co-Co bond might be expected, no evidence could be obtained for Co(CO), fragments. This may indicate that recombination is very efficient. Carbon monoxide expulsion is the primary process upon photolysis of [Co(CO),(NO)] in argon or methane matrices at 20 K314 Photosubstitution occurs in nitrogen matrices, yielding [Co(CO),(NO)(N,)] and probably [Co(CO)(NO)(N,),I* The limiting value of the quantum yield for photochemical insertion of SnCI, into the Co-Co bond of [(Bun3)PCo(CO),], is 1.0 for 365 nm radiation and ca. 6.0 for 546 nm radiation.,lS This surprising observation is attributed to the initiation of a chain reaction at 546 nm, with [Co(CO),(PBun3)] as the chain carrier, but a stoicheiometric reaction at 365 nm. This is the more remarkable as excitation at 365 nm is in a band assigned to a a-u* transition of the Co-Co bond. Substitution of CO in [CI,SnCo(CO),] by PBun3,PPh3, or AsPh, is catalysed by light.,ls The initial step appears to be homolytic cleavage of the Co-Sn bond in [Cl,SnCo(CO),] or its base complex. The co-ordinatively unsaturated species [CpCo(CO)] has been identified upon low-temperature (195 K) photolysis of [CpCO(CO)2].317This species may then dimerize to form [Cp,Co,(CO),] or react with the starting material to give [Cp,Co,(CO),]. In the presence of substituted acetylenes at 195 K, irradiation of [CpCo(CO),] yields [CpCo(CO)(RC=CR)] (R = Ph or Me).,', On warming the mixture with diphenylacetylene, [Cp,Co,(CO)(PhC=CPh)], [Cp,Co,(CO)(PhC=CPh)], and (41) are formed. 0

Molecular hydrogen is released when [IrCIH2(PPh3),] ant, mer- or fac[IrH,(PPh),] are irradiated with U.V. or visible light [e.g. reaction (86)].,ls The elimination of hydrogen proceeds in a concerted fashion. Thus when a mixture of [IrClH,(PPh,),] and [IrCID,(PPh,),] was photolysed, no HD was produced. [IrC1H2(PPh3),] sl* a18

als a16

t17

s18 319

hv

[IrCl(PPh,),]

+ H,

R. L. Sweany and T. L. Brown, Inorg. Chem., 1977,16,415. R. L. Sweany and T. L. Brown, Inorg. Chem., 1977, 16,421. 0. Crichton and A. J. Rest, J.C.S. Dalton, 1977, 536. P. F. Barrett, A. Fox, and R. E. March, Canad. J. Chem., 1977, 55, 2279. M. Absi-Halabi and T. L. Brown, J. Amer. Chem. SOC.,1977, 99, 2982. W.4. Lee and H. H. Brintzinger, J. Organometallic Chem., 1977, 127, 87. W.4. Lee and H. H. Brintzinger, J. Organometallic Chem., 1977,127,93. G. L. Geoffroy and R. Pierantozzi, J. Amer. Chem. SOC.,1976, 98, 8054.

(86)

Photochemistry

234

On the basis of simple MO theory (Figure 3), it appears that population of a o,a-ya orbital is responsible for the labilization of hydrogen. Strohmeier and co-workers have shown that the activity of [IrCI(CO)(PPh,),] and related complexes as catalysts for homogeneous hydrogenation is enhanced by U.V. r a d i a t i ~ n320-324 . ~ ~ ~ ~The effectiveness of different catalysts of type [IrCI(CO)(PR,),] (R = Ph, Pri, OPh, or cyclohexyl) for the photoactivated The activity hydrogenation of various activated olefins has been

\

\

\

\ \* /..HO

L

Q

0

*2

cis - t Ir H2L,l L

Figure 3 Molecular orbital energy level diagram for six-co-ordinate dihydride complexes of iridium (Reproduced by permission from J. Amer. Chem. Soc., 1976,98, 8054)

is dependent on the nature of both the substrate and the phosphine ligand. With ethyl acrylate, for example, the most active catalyst is [IrCI(CO)(PPri,),].321 With certain substrates, e.g. cyclohexa-l,3-diene, the hydrogenation is truly photocatalytic, the reaction proceeding at the same rate after the light source is switched off, and a study of the dependence of the reaction rate on the light intensity has been carried With [IrCl(CO)(PPh,),] in the absence of solvent this photocatalytic hydrogenation is exceptionally efficient, possessing a turn-over number of about 100 OO0.324 Photolysis of [RhCI(CO)(PPh,),] in aerated solution leads to oxidation of its ligands to carbon dioxide and triphenylphosphine The reaction is inhibited by excess phosphine present in solution, supporting the assignment of phosphine dissociation (87) as the primary process. a20 321 928

324 325

W. Strohmeier and L. Weigelt, J. Organometallic Chem., 1977, 129, C47. W. Strohmeier and L. Weigelt, J. Organometallic Chem., 1977, 125, C40. W. Strohmeier and L. Weigelt, J . Organometallic Chem., 1977, 133, C43. W. Strohmeier, H. Steigerwald, and L. Weigelt, J. Organometallic Chem., 1977, 129, 243. W. Strohmeier and H. Steigerwald, J. Organometallic Chem., 1977, 125, C37. G. L. Geoffroy, D. A. Denton, M. E. Keeney, and R. R. Bucks, Inorg. Chern., 1976,15,2382.

-

Photochemistry of Inorganic and Organometallic Compounds [RhCI(CO)(PPh,),]

hv

[RhCI(CO)(PP$)]

235

+ PPh3

(87)

Nickel.-A full paper on the matrix photolysis of CpNi(N0) has been published.32s In carbon monoxide matrices, short periods of irradiation lead to substitution of NO, yielding [CpNi(CO)] and [CpNi(CO),], whereas Ni(C0)4 is formed on longer exposure. In non-reactive matrices (argon, methane, or nitrogen), photolysis produces an isomeric species, possibly [CpNi]+NO- or more probably one of the type [CpNi(NO*)], in which the nitrosyl group behaves as a one- or two-electron ligand. The Nio complex [Ni(NO)Cl(dppe)], upon irradiation, reacts with oxygen to form the Nil1species [Ni(NO&Cl(dppe)].327The initial step appears to be an attack of oxygen on an (MLCT ?) excited state of the starting material to give a peroxynitrate complex [Ni(NO$Cl(dppe)]. This latter species reacts with [Ni(NO)Cl(dppe)] to yield (42), which has been characterized by e.s.r., and which decomposes to the final product.

0 0

(dppe)ClNi-N

//

'N-NiCl(dppe)

\

/

0-0

(42) Copper.-In the presence of copper(r) chloride, norbornadiene effectively = 0 . 3 4 . 4 in CHC13).328The photoactive isomerizes to quadricyclene (aal3 species is a 1 : 1 norbornadiene-CuC1 complex. It is also noteworthy that no dimeric species are formed in this case, in contrast with what is found with other organometallic catalysts such as [Cr(CO),]. If analogous systems can be developed in which visible light is active in inducing isomerization, these might have potential for solar energy storage.

Photolysis of 7-methylenenorcarane (43) in the presence of CU' salts leads to both fragmentation (giving cyclohexene and acetylene) and to rearrangement [the main products being (44) and (45)].329 The most probable mechanism is one in which photoexcitation causes rearrangement of a Cur-olefin r-complex to a-complexes of types (46) and (47) : complex (46) then fragments to give acetylene and cyclohexene, whereas (47) forms (44), (43, and other minor products. The photodecomposition of the copper cluster compound [Cu(CH,SiMe,)], has been described.330 0. Crichton and A. J. Rest, J.C.S. Dalton, 1977, 986. m7 R. Ugo, S. Bhaduri, B. F. G. Johnson, A. Khair, A. Pickard, and Y. Benn-Taarit, J.C.S. Chem. Comm., 1976,694. *28 D. P. Schwendiman and C. Kutal, Inorg. Chem., 1977, 16, 719. 82p R. G. Salomon and M. F. Salomon, J. Amer. Chem. SOC.,1976,98, 7454. 8 3 0 J. A. J. Jarvis, R. Pearce, and M. F. Lappert, J.C.S. Dalron, 1977, 999. 886

9

Photochemistry

236

Mercury.-The results of CIDNP investigations on the photodecomposition of dialkylmercury (R1HgR2) compounds have been ~ ~ m m ~ n i ~ a332t e In d.~~~~ deuteriated benzene or toluene, decomposition to give R1R2, RlH, R2H, or olefins proceeds through a triplet precursor, either directly [equation (S9)] or perhaps by initial formation of a radical pair (R1Hg*,R2*)followed by rapid decomposition of the RHg- species. In chlorinated solvents, reaction involves both triplet (89) and singlet processes [e.g. reaction (90)].331

R1HgR2 R1HgR2

+ CCI,

hv

+

R1**ReT Hg

_c__,{R1HgR2,CC1,}

-

S R1**CC13

+ R2HgCl (90)

Other papers dealing with reactions of related mercury derivatives describe the photodecomposition of B U ~ C H ~ H ~ B of U(SiMe2Ph)2Hg,334 ~,~~~ and of [ ( B U ~ C H , ) , S ~ ] , Hand ~ , ~ ~the ~ polymerization of styrene induced by SiMe, radicals formed on photodissociation of (SiMe3)2Hg.33e Thorium.-On photolysis Cp,ThPri forms Cp,Th and [approximately equal quantities of propane and ~ r o p e n e . ~ This ~ ' behaviour contrasts with thermal decomposition, where intramolecular hydrogen abstraction occurs [reaction (91)]. Evidence has been presented to show that the photochemical reaction does not proceed by homolytic cleavage of the Th-C bond, but rather by /?-elimination of the olefin [reaction (92)]. 2Cp,ThPri Cp,Th-CHMe, 831

s89

s34

A

hv

[Cp,Th(C,H,)], Cp3ThH

+ 2cH8

+ H,C=CHMe

(91) (92)

F. J. J. de Kanter, Org. Magn. Resonance, 1976, 8, 129. R. Benn, Chem. Phys., 1976, 15, 369. W. A. Nugent and J. K. Kochi, J. Organometailic Chem., 1977, 124, 327. E. N. Gladyshev, L. 0. Yuntila, G. A. Razuvaev, N. S. Vyazankin, and V. S. Sokolov, Izuest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 204. 0.A. Kruglaya, B. V. Fedot'ev, I. B. Fedot'eva, and N. S. Vyazankin, Zhur. obshchei Khim., 1976,46, 1517.

838

-

H. Ikeda, Y. Miura, and M. Kinoshita, Makromol. Chem., 1976, 177,2647. D. G. Kauna, T. J. Marks, and W. A. Wachter, J. Amer. Chem. SOC., 1977,99, 3877.

Photochemistry of Inorganic and Organometallic Compounds

237

Table 3 Photochemical substitution reactions of metal carbonyl compounds Reactant L Ph2P(CH,) ,PPhZ n = 2, 3, or 4 [Ph*P(CH2)2PPhCH212 Ph2P(CHa),PPh, n = 1,2, or 4 GH4RX R = H, Me, or Ph, X = C1 or Br MeN(PF,), MeN(PF3,

Products cis-[V(CO)1L]-

Ref. 338

[V(CO),L]- and polymeric [W(CO)4h&12cis-[CpV(CO),L]

338 338 339, 340 341 342 343 344 345 346 347 348 349

350

D. Rehder, L. Dahlenburg, and I. Miiller, J. Organometallic Chem., 1976, 122, 53. M. Schneider and E. Weiss, J. Organometallic Chem., 1976, 121, 345. s40 U. Franke and E. Weiss, J. Organometallic Chem., 1976, 121, 355. 841 R. B. King and K.-N. Chm, Inorg. Chim. Acta, 1977,23,L19. sra R. B. King and J. Gimeno, J.C.S. Chem. Comm., 1977, 142. 848 L. Weber, 2.Naturforsch., 1976, 31b, 780. s44 J. Schmidt and D. Rehder, Chem. Letters, 1976, 9,933. 846 W. A. Schenk, J. Organornetallic Chem., 1976, 117, C97. 141 I. S. Butler, N. J. Coville, and D. Cozak, J. Organometallic Chem., 1977, 133, 59. 847 F. Y.Petillon and D. W. A. Sharp, J. Fluorine Chem., 1976,8, 323. 848 A. M. Rosan and J. W. Faller, Synth. React. Znorg. Metal-org. Chem., 1976, 6, 357. * 4 ~ R.B. King, M. G. Newton, J. Gimeno, and M. Chang, Znorg. Chim. Acta, 1977,23,L35. aw A. N. Nesmeyanov, V. V. Krivykh, P. V. Petrovskii, and M. I. Rybinskaya, Doklady Akad. Nauk S.S.S.R., 1976,231, 110. P. Jutzi and W. Steiner, Chem. Ber., 1976,109, 3473.

Photochemistry

238

Table 3 (cont.) Substrate

Reactant L

Products

ReJ

351

352 353 353 346

354

352 352 355

356

357

362

353 354

365

3s6

P. M. Treichel and H. J. Mueh, Znorg. Chim. Acta, 1977, 22, 265. M. K. Chaudhuri, A. Haas, and A. Wensky, J. Organometallic Chem., 1976, 116, 323. P. Jutzi and A. Karl, J. Organometallic Chem., 1977, 128, 57. R. Aumann, H. Woermann, and C. Krueger, Angew. Chem., 1976, 88, 640; Angew. Chem. Internat. Edn., 1976, 15, 609. R. Aumann and H. Averbeck, Angew. Chem., 1976, 88, 641; Angew. Chem. Internat. Edn., 1976, 15, 610.

367

368

M. S. Brookhart, G. W. Koszalka, G. 0. Nelson, G. Scholes, and R. A. Watson, J. Amer. Chem. Soc., 1976,98, 8155. M. G. Newton, R. B. King, M. Chang, and J. Gimeno, J. Amer. Chem. SOC.,1977,99,2802.

Photochemistry of Inorganic and Organometallic Compounds Table 3 (cont.) Substrate

Reactant L

Products [Fe(CO),L] and rFe(CO),Ll

239 Ref.

359

360

SbPh,

361

362

3 Main-group Elements Reports on the photochemistry of anions in solution are dealt with in the next section: those on other compounds are considered under the particular element concerned. Anions.-It is well established that the photolysis of inorganic anions, such as iodide, causes ionization [reaction (93)]. However, it has previously been I-

hv

1-

+ qq-

(93)

observed in scavenging experiments that the decay profile of the solvated electrons so produced differs markedly from that of those formed upon radiolysis of water. Recently the reactions of the solvated electron have been monitored in the submicrosecond region following 265 nm laser photolysis of iodide, [Fe(CN),]-, tryptophan, or t y r ~ s i n e .Either ~ ~ ~ recombination of the electron with its radical co-product or bimolecular reactions of the electron with a scavenger may account for its decay. The experimentally determined decay profiles have been shown to correlate well with those determined theoretically, although the recombination lifetime (ca. s) is about lo4 times longer than that predicted by the Noyes theory. This behaviour is apparently caused both by the large initial displacement ( 1 nm) of the electron from its radical co-product after photodissociation and by the energy barrier created by the hydration of the electron. The effect of temperature on the CTTS transition of iodide ion in supercooled water or glassforming aqueous solutions 364 and correlation of ,,v for the CTTS transition in N

saa

T. Boschi, P. Vogel, and R. Roulet, J. Orgunometallic Chem., 1977, 133, C36. T. Boschi, P. Narbel, P. Vogel, and R. Roulet, J. Orgunometullic Chem., 1977,133, C39. D. J. Cane, E. J. Forbes, and T. A. Hamor, J. Orgunometullic Chem., 1976, 117, C101. E. G. Bryan, A. L. Burrows, B. F. G. Johnson, J. Lewis, and G . M. Schiavon, J. Orguno-

*us

L. I. Grossweiner and J. F. Baugher, J . Phys. Chem., 1977, 81, 93.

864

A. Barkatt and C. A. Angell, J. Phys. Chem., 1977,81, 114.

8sB 861

metallic Chem., 1977, 129, C19.

Photochemistry

240

chloride, bromide, and iodide ions with that of the solvated electron in some 43 solvents, including acetone, ethers, and alcohols,S66have been reported. Bromide and iodide ions enhance the photoisomerization yields for protonated trans-4’-methoxy-3-styrylpyridine and diprotonated trans-l,2-bis-(3-pyridyl)ethylene, while quenching their The effect is attributed to complex formation with consequent heavy-atom induction of ISC to the reactive triplet states of the olefin derivatives. Examples of iodide quenching of fluorescence of aliphatic of pyrene in aqueous m i ~ e l l e s and , ~ ~ of ~ flavins in aqueous solution36ghave been described. The decay of the triplet state of triphenylene in ethanol solutions at 77 K is accelerated by potassium iodide, the effect on the radiative transition being greater than that on the radiationless dea~tivation.3~0 Photo-oxidation of cyanide to cyanate at titanium dioxide surfaces has been The redox reaction appears to involve oxidation of cyanide ions by valence-band holes of the semiconductor [reaction (94)]. CN-

+ 20H- + 2p+

-

CNO-

+ H20

(94)

Oxidation of carbonate ions by the triplet state of duroquinone in micellar solutions has been monitored by laser p h o t ~ l y s i s .It~ ~has ~ been demonstrated that singlet oxygen is involved in the photo-oxidation of iodide ion by various 374 Hydrogen abstraction from the hydroanthracene sulphonate derivatives.373* carbon by an excited state of the ion appears to be responsible for the photoreduction of nitrite ions in the presence of alkanes.376 The luminescence of nitrite 376, 377 and of selenite 378 ions has been studied in aqueous glassy media at temperatures from 4.2 to 200 K. Barium.-Flash photolysis of the barium salt of triphenylethylene causes electron ejection, allowing the study of the kinetics of the equilibrium shown in reaction

-

(99.379

TPE+ Ba2+TPE2Ba2+TPE- + TPE’(95) Boron,-Photodecomposition of tri-l-naphthylboron in non-polar solvents appears to proceed via two quite distinct pathways [reactions (96) and (97)].3**

hv

(naph),B

s67 a*8

SE8 s70

371

373 374 s76

w8 a77

378

37D 380

+ (naph),

(96)

M. F. Fox and E. Hayon, J.C.S. Faraday I, 1976,72,1990. G. Bartocci, U. Mazzucato, and P. Bortolus, J. Photochem., 1977, 6, 309. I. V. Sokolova amd L. V. Orlovskaya, Zhur. priklad. Spektroskopii, 1977, 26, 167. M. A. J. Rogers and M. E. Da Silva e Wheeler, Chern. Phys. Letters, 1976, 43, 587. S. Bergstrom and B. Holmstrom, Chem. Scripta, 1976, 9, 193. J. Najbar and A. Chodkowska, J. Luminescence, 1976,11,215. S . N. Frank and A. J. Bard, J. Amer. Chem. SOC.,1977,99, 303. R. Scheerer and M. Graetzel, Ber. Bunsengesellschaft Phys. Chem., 1976, 80,979. K. K. Rohatgi-Mukherjee and A. K. Gupta, Indian J. Chem., Sect. A , 1976, 14, 723. K. K. Rohatgi-Mukherjee and A. K. Gupta, Chem. Phys. Letters, 1977,46, 368. G. L. Petriconi and H. M. Papee, Water, Air, Soil Pollution, 1976, 5, 403. M. U. Belyi, Y . T. Kononenko, and I. Y . Kushnirenko, Ukrain. $2. Zhur., 1977, 22, 306. Y. T. Kononenko and I. Y . Kushnirenko, Ukrain.f;z. Zhur., 1976,21, 1599. I. Y. Kushnirenko and S. G . Nedel’ko, Zhur. priklad. Spektroskopii, 1976, 25, 812. B. De Groof, G. Levin, and M. Szwarc, J. Amer. Chem. SOC.,1977,99,474. B. G. Ramsey and D. M. Anjo, J. Amer. Chem. SOC., 1977,99, 3182.

3 ~ 3 366

(naph)B: (48)

24 1 The boryne species (48) has been trapped by cyclohexene, giving (49). In protic solvents irradiation of tribenzylboron, or its ammonia complex, i~ducesheterolytic cleavage of a B-C bond.881 The resulting benzyl anion protonates to give the major product, toluene.

Photochemistry of Inorganic and Organometallic Compounds

Mercury-photosensitized reactions of N-mono-, -di-, or -tri-methylborazine in the presence of hydrogen give exclusively C- C-bonded diborazinyl derivatives. The products arise by coupling of radicals formed upon hydrogen atom abstraction from the methyl group of the borazine Photodecomposition of carbonyl sulphide in closo-carbaboranes provides a convenient synthetic route to boron-substituted mercapto~arbaboranes.~~~ The reaction involves initial generation of sulphur atoms in their lD excited state and subsequent insertion into a B-H bond [e.g. reaction (98)]. On further irradiation these mercaptans may dehydrodimerize to form boron-bridged disulphides [reaction (99)].

’s 4-

2,4-GB6H7

-

3-SH-2,4-C&H, (GB,H,S),

2GB6&SH

+ Ha

(98) (99)

The study of specific reactions induced by i.r. radiation from carbon dioxide lasers continues to be a rapidly developing area, and is the subject of a recent review article.S84Excitation of diborane by CO, laser radiation in a vibrational mode associated with a rocking vibration of a terminal BH2group causes reactions quite different from those observed on thermolysis. Thus, in the presence of hydrogen sulphide, HB(SH), and HSBaHs are formed,s86and with isobutylene alkylated diboranes are the products.886 Upon similar i.r. excitation, boron trichloride sensitizes the ‘trimerization’ of tetrachloroethylene to hexachlorobenzene in ca. 88% yield.387The C0,-laser-induced disproportionation of boron trichloride and trimethylboron 888 and the laser-augmented decomposition (100) 388 have been examined in greater detail. 2HaB-PFs

hv (i.r.1

+

+ 2PF3

(1W

880

Vo Van Chung, K. Inagaki, M. Tokuda, and M. Itoh, Chem. Letters, 1976,209. G . A. mine and R F. Porter, Inorg. Chem., 1977, 16, 11. J. S. Plotkin and L. G. Sneddon, J. Amer. Chem. Soc., 1977,99, 3011. R. V. Ambartzumian and V. S. Letokhov, Accounts Chem. Res., 1977, 10, 61. H. R. Bachmann, F. Bachmann, K. L. Kompa, H. Noth, and R. Rinck, Chem. Ber., 1976,

s86

Y. A. Adamova, A. A. Bukharov, A. V. Pankratov, and A. N. Skachkov, Zhur. neorg.

881 889

88s 884

109, 3331. 887

88D

Khim., 1976,21,938. H. R. Bachmann, R. Rinck, N. Noth, and K. L. Kompa, Chem. Phys. Letters, 1977,45, 169. F. Bachmann, H. Noth, R. Rinck, W. FUSS,and K. L. Kompa, Ber. Bungengesellschaft phys. Chem., 1977,81, 313. K. R. Chien and S. H. Bauer, J. Phys. Chem., 1976,80, 1405.

242

Photochemistry

Silicon.-The organic photochemistry of silicon derivatives is considered in Part 111, Chapter 6 . Previous studies have provided much indirect evidence for the participation of compounds containing silicon-carbon double bonds as reactive intermediates. By use of matrix-isolation techniques, two groups of workers have now spectroscopically characterized such a s p e c i e ~ . 391 ~ ~ ~Photolysis 1 of trimethylsilyldiazomethane (50) in argon matrices produced a photostationary mixture containing trimethylsilyldiazirine (5 1) (Scheme 5 ) . On extended irradiation Me

A Y Me

RN2

Me,Si' SiMe,

Me, Si -C,

H

Me,Si=CHMe N

Me,Si-C' I'N H

+A +

(53) Me

xSiMe,

(52)

1

Me,Si

y

Me (54)

Scheme 5

(A > 300nm), the silaethylene (52) was produced. On warming to room temperature, this unsaturated species dimerized, giving (53) and (54). Longerwavelength irradiation (?.> 360 nm) of the diazirine-diazomethane mixture yielded the carbene Me3SiCH, which could be characterized by e.s.r. as a linear ground-state triplet. At the low temperatures used in these experiments this carbene species did not appear to isomerize thermally to the silaethylene (52), although such a rearrangement probably takes place at room Photolysis of the silacyclopropene (55) in the presence of [PdCI,(PEt,),] gives (56).a93

(55)

(56)

Germanium and Tin.-A full report on the synthesis of stable germaniumand tin-centred radicals of type *M[CH(SiMe3)2]3,*M[N(SiMe,),],, and *M[N(SiMe,)ButI3 (M = Ge or Sn), formed by photolysis of the bivalent 3D0 391

s82

393

0. L. Chapman, C. C. Chang, J. Kolc, M. E. Jung, J. A. Lowe, T. 5. Barton, and M. L. Tumey, J. Amer. Chem. SOC., 1976, 98, 7844. M. R. Chedekel, M. Skoglund, R. L. Kreeger, and H. Shechter, J. Amer. Chem. Soc., 1976,98, 7846. W. Ando, A. Sekiguchi, and T. Migita, Chem. Letters, 1976, 779. M. Ishikawa, T. Fuchikami, and M. Kumada, J.C.S. Chem. Comm., 1977, 352.

243 alkyls or amides [reaction (lol)], has been published.3B4An alternative route to the tervalent alkyl complexes M[CH(SiMe3)J3 is shown in equation (102).3B5 The metal-centred radical *Sn(CH,CMe,Ph),, which is sufficiently stabfe to be observed by e.s.r. in solution at room temperature, has been prepared inter alia by reaction (103).3Bs

Photochemistry of Inorganic and Organometallic Compounds

(PhMe&CHd,Sn-Sn(CH,CM%Ph),

hv

2*Sn(CHzCMeaPh), (103)

Production of mSnBun3 by photochemical cleavage of the Sn-Sn bond in (Bun,Sn), has allowed the determination of the rate of reaction of this tincentred radical with dialkyl-selenium or -tellurium Photodesulphurization of (57) by (Bu*,Sn), is a useful method for preparing tetrathiafulvalenes (5 8).398

-

Dibutylstannylene, formed by photodissociation of polymeric dibutyltin, inserts into the carbon-halogen bond of alkyl halides [e.g. reaction (104)].39B Bu,Sn

+ EtBr

Bu,EtSnBr

(104)

The light-induced addition of GeMe,H to fluoroethylene~,~~~ the photoinitiated polymerization of methyl methacrylate in the presence of organotin the photodecomposition of dibutyltin dilaurate and other PVC stabilizers,40aand the U.V. decomposition of (Bu,Sn),O 403 have all been described recently.

8B1 308

Ioo

Iol ‘Oa *03

A. Hudson, M. F. Lappert, and P. W. Lednor, J.C.S. Dalton, 1976,2369. M.J. S. Gynane and M. F. Lappert, J. Organometallic Chem., 1976,114,C4. H. U. Buschhaus, hL Lehnig, and W. P. Neumann, J.C.S. Chem. Comm., 1977,129. J. C. Scaiano, P. Schmid, and K. U. Ingold, J. Organometallic Chem., 1976,121,C4. Y.Ueno, A. Nakayama, and M. Okawara, J. Amer. Chem. SOC.,1976,98,7440. S. Kozima, K. Kobayashi, and M. Kawanisi, Bull. Chem. SOC.Japan, 1976,49,2837. K. D. R. Winton and J. M. Tedder, in ref. 219, p. 29. S. F. Zhil’tsov, V. N. Kashaeva, G. I. Anikanova, Y. A. Kaplin, and L. F. Kudryavtsev, Izv&st. Vyssh. Uchebn. Zaved. Khim. i khim. Tekhnol., 1976,19,633. Y. Oki, F. Mori, and M. Koyama, Kobunshi Ronbunshu, 1977,34,43(Chem. Abs., 1977,86, 122 184). D.Kloetzer and U. Thust, Chem. Tech. (Leipzig), 1976,28, 614.

244

Photochemistry

The photoreduction of K2Sn2(ox),in aqueous solution to give Sn2+has been monitored both spectrophotometrically and by radiochemical determination of 14CO 404 2.

Nitrogen, Phosphorus, Arsenic, Antimony, and Bismuth.-A comparison of the photodissociation reactions of MPh3 (M = N, P, As, Sb, or Bi) has been made at both 77 K and 300 K, using e.s.r. and spectrophotometric With the aid of the spin-trap reagent PhCH=N(0)But, quantum yields for the production of phenyl radicals from the arsenic (0.05), antimony (O.l), and bismuth (0.15) derivatives could be estimated, whereas with triphenylamine and triphenylphosphine the radical concentration was too low to be detected. Quenching of the fluorescence of aromatic compounds by MPh, (M = N, P, Sb, or As) involves a deactivation of the excited state by both charge-transfer and heavy-atom induced ISC processes.4o6 Tertiary phosphines are readily photo-oxidized to the corresponding phosphine oxide [reaction (105)].407For PPh,, PEtPh,, PEt,Ph, PMe,Ph, and PMePh, the quantum yields (at 254 nm) range from 3.3 to 5.4, indicating that a chain reaction mechanism is operative. 2PR3

+ O2

hv

2R3P0

(105)

Persistent radicals of the type PR2 have been prepared by irradiation of [(Me3Si),CHI2MCl or [(Me,Si),N],MCl (M = P or As) in the presence of electron-rich olefins such as (59).408

(59)

Photolysis of the ylide Me,&-%P(O)Ph, gives singlet pho~phinylnitrene.~~~ Attack of triplet benzophenone upon tetraphenyl- or tetraethyl-diphosphine causes cleavage of the P-P bond as indicated in equation (106).*1° Ph2COT

+ R2PPR2

-

Ph2k0PR2

+ -PR,

(106)

Oxygen and Sulphur.-Autoradioluminescence of solutions of uranyl ion in 57% aqueous perchloric acid under intrinsic 238Ua-radiation has been assigned to emission from the 3B1state of A report on the thermoluminescence of polycrystalline ice after U.V. irradiation has been published.412 404 406

406

407

408

40Q

410

*11

E. L. J. Breet and R. Van Eldik, Znorg. Chim. Acta, 1977, 21, 89, 95. S. G. Smirnov, A. N. Rodionov, K. L. Rogozhin, 0. P. Syutkina, E. M. Panov, D. N. Shigorin, and K. A. Kocheshkov, Zzvest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 335. H. D. Burrows, S. J. Formosinho, A. M. Da Silva, and S. E. Carlin, J. Photochem., 1977, 6, 317. G. L. Geoffroy, D. A. Denton, and C. W. Eigenbrot, Znorg. Chem., 1976, 15, 2310. M. J. S. Gynane, A. Hudson, M. F. Lappert, P. P. Power, and H. Goldwhite, J.C.S. Chem. Comm., 1976, 623. E. Kameyama, S. Inokuma, and T. Kuwamura, Bull. Chem. SOC.Japan, 1976,49, 1439. R. Okazaki, K. Tamura, Y. Hirabayashi, and N. Inamoto, J.C.S. Perkin Z, 1976, 1924. G. L. Sharipov, V. P. Kazakov, and R. G. Bulgakov, Zhur. priklad. Spektroskopii, 1977,26 289.

J. R. Pettit and P. Duval, Solid State Comm., 1976, 19, 475.

Photochemistry of Inorganic and Organometallic Compounds

245

Recent publications have considered the photolysis of hydrogen peroxide in methanol, using e.s.r.-monitored flash photolysis 413 and the photo-oxidation of sulphur by hydrogen A mixture of addition products is formed following the photodissociation of OF, in the presence of FCONSF2.415 Multiphoton dissociation of SF, by carbon dioxide laser radiation has been 416-421 investigated by a number of research Selenium and Tellurium.-Two groups have presented their results on the photodeselenation (107) of benzyl diselenides in the presence of tertiary p h o ~ p h i n e s . ~ ~The ~ , *primary ~~ process appears to be cleavage of the Se-Se bond, followed by a chain reaction [steps (108)--(112)], and this is confirmed by PhCH,SeSeCH,Ph

+ PPh,

PhCH,SeSeCH,Ph PhCH,Se*

+ PPh3

PhCH,SePPh, PhCH,=

+ PhCH,SeSeCH,Ph 2PhCH2*

hV

hv

____+

_____+

____+

PhCH,SeCH,Ph

+ Ph3PSe

(107)

2PhCH2Se-

(108)

PhCH,SePPh,

(109)

PhCH,.

+ SePPh,

PhCH,SeCH,Ph PhCH2CH,Ph

+ *SeCH,Ph

(1 10) (1 11) (1 12)

the observation of quantum yields greater than unity at phosphine concentrations greater than 0.05 mol dm-3.422 Photolysis of diethyl and dibenzyl ditellurides also results in demetallation [reaction (1 1 3)].424 However, in this case the reaction appears to involve rupture of the Te-C bond rather than the Te-Te linkage. RTeTeR

hv

RTeR

+ Te

(113)

The photochemical and photopnysical properties of bis(benzoylmethy1)tellurium dichloride have been At 77 K, emission is observed from the lowest-lying triplet state, which appears to be a T-T* species, although with appreciable n-n* character. At room temperature, this excited state reacts either by a process similar to the Norrish Type I1 reaction [equation (114); J. Hellebrand and P. Wuensche, Faserforsch. Textiltech., 1976, 27, 589. I. N. Barshchevskii, Zhur.$z. K'him., 1976, 50, 1176. I. Stahl, R. Mews, and 0. Glemser, Chenr. Ber., 1977, 110,2398. *lo J. Dupre, P. Pinson, J. Dupre-Maquaire, C. Meyer, and P. Barchewitz, Compt. rend., 1976, 283, C, 311. B. J. Orr and M. V. Keentok, Chem. Phys. Letters, 1976, 41, 68. H. N. Rutt, J. P h p . (D), 1977, 10, 869. a9 V. D. Klimov, V. A. Kuz'menko, and V. A. Legasov, Zhur. neorg. Khim., 1976,21,2100. 420 A. K. Petrov, Y. N. Samsonov, A. V. Baklanov, V. V. Vizhin, and A. Orishich, Izvest. Akad. Nauk S.S.S.R. Ser. khim., 1976,2148. 421 P. Kolodner, C. Winterfeld, and E. Yablonovitch, Opt. Cornrn., 1977, 20, 119. d m J. Y. C. Chu and D. G. Marsh, J. Org. Chem., 1976,41, 3204. D. H. Brown, R. J. Cross, and D. Millington, J.C.S. Dalton, 1977, 159. 424 D. H. Brown, R. J. Cross, and D. Millington, J. Organometnllic Chem., 1977, 125, 219. ras D. G. Marsh, J. Y. C. Chu, J. W. Lewicki, and J. L. Weaver, J. Amer. Chem. SOC.,1976,98,

413

'14

8432.

246

Photochemistry

the ylide could not be isolated] or by cleavage of the Te-C (1 1511.

bond [equation

(PhCOCH,),TeCI, (PhCOCH,),TeCl,

hv

hv

PHCOCH, PhCOkH,

+ PhC(O)CH+eCl, + PhCOCH,?eCI,

(1 14)

(1 15)

Upon photolysis in an argon or nitrogen matrix at 8 K, the selenadiazole (60) decomposes to yield ethynylselenol, selenoketen, and acetylene.42s On further irradiation, the selenoketen partially isomerizes to ethynylselenol. Photolysis of (61) provides a convenient route to the novel heterocyclic compound (62).427 In the presence of oxygen, irradiation of (63) gives (64).428

0

Ph-C

Ph-C

II 0

Ph (63)

(44) Halogens and the Noble Gases.-Chlorine difluoride has been identified by i.r. and Raman spectroscopy, following its preparation by irradiation of chlorine monofluoride and fluorine in low-temperature nitrogen The results of a detailed investigation of the emission from chlorine in noble-gas matrices have been communicated.430 The photochemical synthesis of krypton difluoride from its elements has been examined q ~ a n t i t a t i v e l y . ~ ~The l - ~highest ~~ quantum yield (0.01 3) was recorded d26 427

428

430

431

432

433

J. Laureni, A. Krantz, and R. A. Hajdu, J. Amer. Chem. SOC.,1976, 98, 7872. B. Pakzad, K. Praefcke, and H. Simon, Angew. Chem., 1977,89,329; Angew. Chem. Internat. Edn., 1977, 16, 319. E. Luppold, W. Winter, and E. Mueller, Chem. Ber., 1976, 109, 3886. E. S. Prochaska and L. Andrews, Inorg. Chem., 1977, 16, 339. V. E. Bondybey and C. Fletcher, J. Chem. Phys., 1976, 64, 3615. A. A. Artyukhov, V. A. Legasov, G. N. Makeev, B. M. Smirnov, and B. B. Chaivanov, Khim. vysok. Energii, 1976, 10, 512. A. A. Artyukhov, V. A. Legasov, G. N. Makeev, L. A. Palkina, B. M. Smirnov, and B. B. Chaivanov, Khim. uysok. Energii, 1977, 11, 88. A, A. Artyukhov, V. A. Legasov, G . N. Makeev, and B. B. Chaivanov, Khim. vysok. Energii, 1977, 11, 89.

Photochemistry of Inorganic and Organometallic Compounds

247

for 313 nm irradiation of solid krypton in liquid fluorine at 77 K.433 Monohalides of the noble gases (KrF, XeF, XeCI, and XeBr) have been prepared by photolysis of the elements at 20 K, and their emission spectra have been 436 Photolysis of argon-xenon-ozone mixed matrices at 22 K yields xenon monoxide, which was identified by its U.V. absorption 434 436

B. S. Ault and L. Andrews, J. Chem. Phys., 1976, 65, 4192. B. S. Ault, L. Andrews, D. W. Green, and G. T. Reedy, J. Chem. Phys., 1977, 66,2786. B. S. Ault and L. Andrews, Chem. Phys. Letters, 1976, 43, 350.

Part III ORGANIC ASPECTS OF PHOTOCHEMISTRY

1 Phot o Iy s i s of Carbony I Co mpo und s BY W. M. HORSPOOL

1 Introduction A subjective review of the 1974 literature has appeared.l Other reviews have dealt with aspects of the photochemistry of natural products,2 with organic molecules in the triplet state,* and with newer methods for the determination of intersystem crossing efficiencie~.~ Intersystem crossing efficiencies for propanal, heptanal , and dodecanal have been measured as 0.64, 0.18, and 0.19 respectively.6 A flash photochemical study of methyl naphthyl ketone has shown that the triplet lifetime of the ketone is concentration dependent.8 Spectroscopic studies have been carried out on the aminoketones (l), (2), and (3).’ The supposed self-quenching of 1,l,l-trifluoroacetone phosphorescence is due to impurities.8 The phosphorescence of aryl ketones is quenched by alkyl disulphides.@4-Amino- and 4-hydroxybenzophenone are inefficient in sensitizing the phosphorescence of biacetyl owing to the ability of the ketones to quench the phosphorescence of biacetyl at diffusion controlled rates.1° Triplet energy transfer in the ketones (4) and ( 5 ) has been investigated,ll and the deactivation of benzophenone triplets via exciplex formation has been reported.l* The photochemistry and photophysical properties of Michler’s ketone have been studied in considerable detail.15 The photochemical reactions of cycloalkanones have been analysed14 in terms of ‘Salem Diagrams’.16 The hydrogen-abstracting ability of ketones such as benzophenone, acetophenone, and acetone in their triplet states has been known for many years and continues to be exploited, as in the investigation of the photochemical reactions 1

a 8 4 6

D. R. Brewer, I. E. Kochevar, G. B. Schuster, J. E. Shields, and N. 5. Turro, Mol. Photochem., 1976,7, 85 (Chem. Abs., 1976,85,20 189). K. Nakanishi and H. Sato, Nat. Prod. Chem., 1975,2, 523 (Chem. Abs., 1977,86,70 962). A. Gilbert, Znt. Rev. Sci.,Org. Chem. Ser. Two, 1975, 10, 277 (Chem. Abs., 1977,86,71045). G. S. Hammond, US.NTZS, Ad. Rep. AD-A0 18549, 1975 (Chem. Abs., 1976,84,150014). P. Lebourgeois, R. Arnaud, and J. Lemake, J. Chim. Phys., 1976,73, 135 (Chem. Abs., 1977 86, 16 005).

6

7

* 9

10

11

la 14 16

D. I. Schuster and M. D. Goldstein, Mol. Photochem., 1976,7, 209. A. M. Halpern and A. L. Lyons, jun., J. Amer. Chem. SOC.,1976,98, 3242. A. Gandini and P. A. Hackett, J. Photochem., 1976, 6, 75 (Chern. Abs., 1977, 86, 71 191). W. L. Wallace, R. P. Van Duyne, and F. D. Lewis, J. Amer. Chem. Suc., 1976,98,5319. G. Favaro, J. Photochem., 1976,6, 139 (Chem. Abs., 1977,86, 71 198). P. J. Wagner, D. A. Ersfeld, and B. J. Scheve, Exciplex. Pruc. Meet., 1975, 359 (Chem. Abs., 1976, 85,4861). M. W. Wolf, R. E. Brown, and L. A. Singer, J. Amer. Chem. Soc., 1977,99,526. D. I. Scbuster, M. D. Goldstein, and P. Bane, J. Amer. Chern. SOC.,1977,99, 187. N. J. Turro, W. E. Farneth, and A. Devaquet, J. Amer. Chem. SOC.,1976, 98, 7425. L. Salem, W. G. Dauben, and N. J. Turro, J. Chim. Phys., 1973,70,694; L. Salem, J. Amer. Chem. SOC.,1974,96,3486; W. G. Dauben, L. Salem, and N. J. Turro, Accounts Chem. Res., 1975, 8, 41; A. Devaquet, Topics Current. Chem., 1975 $4, 1.

25 1

Photochemistry

252 Me

0

(4)

of these ketones with 2-phenyl-l,3-dio~an.~~ Pentaerythritol, 2-hydroxymethylglycerol, and a low yield of carbohydrate molecules are produced when formaldehyde is irradiated in aqueous solution in the presence of an inorganic base.17 The addition of the radical (6) (obtained by the photolysis of di-t-butyl peroxide in acetic acid) to ethylene affords butyric acid and other carboxylic acids.le CHaCOaH (6)

Both the direct and sensitized (xanthone, benzophenone, or acetophenone) irradiation of the acetylquinoline (7a) results in the reduction of the acetyl function to ethyl yielding (8).1° The sensitized path to (8) involved rapid reduction of the acetyl group to a hydroxy-ethyl substituent. This compound (9), which was isolated only in small amounts, rapidly undergoes loss of hydroxyl upon secondary irradiation. The loss of hydroxy and alkoxy groups from such sites has been demonstrated in the ferrocenyl series.2o The direct irradiation of (7a) also yielded

M c1 c o\~ p M e O C 6 H , ,

MeoQ&-Me*c6H4 CI \ (7) a ; R = Ac

b;R

=

(8)

H CHOHMe

M CeI o\~ p - M e O C 6 H 4 (9) 1e l7

M. Suzuki, T. Inai, and R. Matsushima, Bull. Chem. SOC.Japan, 1976,49, 1585. Y. Shigemasa, Y. Matsuda, C. Sakazawa, and T. Matsuura, Bull. Chem. SOC.Japan, 1977, 50,222.

Y.Suhara, Yukuguku, 1976.25, 75 (Chem. Abs., 1976, 85, 32 375). 19

G. A. Epling, N. K. N. Ayengar, and E. F. McCarthy, TetrahedronLetters, 1977, 517. C. Baker and W. M. Horspool, Chem. Cornrn., 1971,615.

Photolysis of Carbonyl Compounds

253 0

-c=o

(J

II

-0 C

(10)

(J"""""" (11)

(7b) together with the ethylquinoline (8). While product (7b) arose from the singlet state a definite mechanism for its formation has not been established.le A study of the photochemical reactions of aromatic aldehydes (e.g. benzaldehyde) in benzene with added t-butylamine has shown that the radical (10) formed by the abstraction of the aldehydic hydrogen from a ground-state molecule of aldehyde by an excited-state molecule is trapped as the amide (1 1).21 Wigfield et a1.22have reported that cyclohexanones can be successfully reduced by sodium tetrahydroborate in diglyme when the solutions are irradiated at 254nm. In the absence of irradiation no reduction was detectable. Ph2C(OH)CH&MeaOH (12)

*CH,CMe,OH (13)

For many years t-butanol has been used as a sdvent for photochemical reactions due to its reluctance to take part in photochemical reactions. However, in recent years there have been reports 2s that this non-reactivity is to be questioned. Pratt and his co-workers 24 have found that when benzophenone is irradiated in t-butanol the diol (12) is produced in accordance with the proposal that benzophenone abstracts a hydrogen atom from a methyl group to yield the radical (13). This is contrary to the suggestion, made by Stille and his group,2s that hydrogen was abstracted from the hydroxyl function. 2 Norrish Type I Reactions

The photochemical decarbonylation of a series of fluorinated ketones (14) has been described.2s The reaction, in all cases, involves a Norrish Type I C-C fission affording radicals from which the ultimate products are obtained. A study of the photochemistry of the aldehydes (15) in inert solvents has been published. The reactions of the aldehydes arise either by Norrish Type I fission from the triplet state or by intermolecular hydrogen abstraction from both the Sl and TI states. No evidence for Norrish Type I1 intramolecular hydrogen abstraction 28 or-Fission is also encountered following U.V. irradiation of was ketone (16).29 The reaction affords a substituted benzyl radical (17) and a ketene 21

22

z8

*'

*6 26

*8

R. S. Davidson, J. Edwards, and S. K. Warburton, J.C.S. Perkin I, 1976, 1511. D. C. Wigfield, S. Feiner, and F. W. Gowland, Tetrahedron Letters, 1976, 3377. W. Lowowski and T. W. Mattingly, J . Amer. Chem. SOC.,1965, 87, 1947. D. W. Alexander, A. C. Pratt, and A. E. Tipping, Tetrahedron Letters, 1976,2893. E. W. Kummerle, T. A. Rettig, and J. K. Stille, J. Urg. Chem., 1975, 40,3665. V. I. Saloutin, Y . N. Agishiev, and N. A. Ryabinin, Zhur. org. Khim., 1975, 11,2623 (Chem. A h . , 1976, 84, 157 959). C. W. Funke and H. Cerfontain, Tetrahedron Letters, 1973,487. C . W. Funke, J. L. M. de Boer, J. A. J. Geenevasen, and H. Cerfontain, J.C.S. Perkin II, 1976, 1083.

H. Iida, S. Aoyagi, and C. Kibayashi, Heterocycles, 1976,4,697 (Chem. Abs., 1976,85,21007).

Photochemistry

254

(14) n

=

2, 4, or 6

(15) n

=

1, 2, 3, or 4

p

Me0 Me0

(16)

(18). The ketene is trapped as the corresponding acid while the benzyl radical yields 1,2-di-(2',3'-dimethoxyphenyI)ethane. The photochemical reactions of the hydroperoxy ketones (19) probably involve acyl radicals in a radical chain mechani~rn.~~ The activation energy for the photochemical cleavage of butan-2one triplets in the gas phase has been measured as 12.71 kcal mo1-1 (53.25 kJ m01-1).31 The Norrish Type I fission of phenacyl systems has been used as a method for the formation of protected peptides using a solid-state polymer-support system. In the reaction sequence the phenacyl unit is anchored to a resin and the peptide is constructed by replacement of the bromine in an ester-forming reaction.32 Other amino acids are then added in the desired sequence. Finally, irradiation at 350nm cleaves the constructed peptide from the phenacyl system (Scheme 1). A series of bridgehead phenyl ketones, some of which are shown below, has been synthesized and their photochemical reactivity The ketones (20) and (21) undergo photochemical a-cleavage in benzene solution to afford products derived from the cycloalkyl and the PhCO radicals. Ketones (22a and b) are almost inert to photolysis in benzene but they do undergo photochemical reduction and Norrish Type I cleavage in propan-2-01. Low temperature (- 70 "C)irradiation of the spirocyclobutanones (23) and (24) in methanol yielded the ring-expanded products (25) and (26) re~pectively.3~ so 81 s2 s3 34

Y . Sawaki and Y. Ogata, J. Amer. Chem. SOC.,1976, 98, 7324. E. Abuin and E. A. Lissi, J. Photochem., 1976, 5, 65 (Chem. Abs., 1976, 84, 142 987). S . 4 . Wang, J . Org. Chem., 1976, 41, 3258. H.-G. Heine, W. Hartmann, F. D. Lewis, and R. T. Lauterbach, J. Org. Chern., 1976,41,1907. K. Kimura, M. Takamura, S . Koshibe, M. Juro, Y.Fukuda, and Y . Odaira, Bull. Chem. SOC. Japan, 1976,49, 741.

Photolysis of Carbonyl Compounh

255

Me resin-@O&mr

1 resin O

Me

OGly-Boc

I Z-Lys(Z)-Phe-Phe-Giy-OH

f

iv

resin

I

C O - , ,

0

Me Co- cI H

I

Z-Lys(Z)-Phe-Phe-Gly-0 Reagents: i, Boc-Gly-0- Cs+; ii, Boc-Phe-OH; iii, Z-Lys(Z)-OH; iv, hv Scheme 1

4 c=o I

(22) a; n b;n

=

2

=

1

Ring expansion of this type yielding an intermediate carbene, subsequently trapped by solvent, is common in cyclobutanones.s6 Ring expansion was also encountered in the photochemical reactions of the cyclobutandiones (27).32 Ring expansion, although most common in cyclobutanones, also arises in other systems. Thus in (28d and e) photochemical ring expansion is efficient and yields the acetals (29).36 In this instance the substitution pattern stabilizes the biradical intermediate (30) more than (31). But this latter type of intermediate dominates the photochemical ring-opening reactions of furanones (28a-c) For recent reviews see D. R. Morton and N. J. Turro, Adu. Photochem., 1974, 9, 197; P. Yates and R. 0. Loutfy, Accounts Chem.!Res., 1975, 8, 209. P. Yates, A. K. Verma, and J. C. L. Tam, J.C.S. Chern. Comm.,[1976, 933.

Photochemistry

256

where Yates and his co-workers 33 have shown that esters (32) are formed from a ketene produced by disproportionation of the biradical (3 1). An intermediate case between the two routes is found with furanone (28f) which affords a small (2%) amount of a ketal together with the ester (32f).

0

0 CO,R

R3

R410,&I

R5

R2

(32) a-c

(31)

(30)

R5 and f : R

=

R2 Me or Et

Direct irradiation of the azetidone (33) in methanol gave the ester [(34), 56x1 and the amide [(35), These products are the result of photochemical fragmentation of the starting material (33) induced by initial Norrish Type I cleavage. The formation of the ester is good evidence for the fragmentation following the path shown in Scheme 2 whereby a methylene ketene (36), an elusive class of compound, is produced. The ester (34) isolated is a secondary product since the Me Me

+

N. '

Me

N

Me

K

+ Me

Me Me

K

(33)

Me

Mc FCH-C-NHMetMeNH, Me I1 0

Me Me

+ )(

(3s)

0

(36)

Me

K CH II

o/'C

OMe

( 37)

3 N0 H M e

?OMe0

(35)

(34) Scheme 2

37

p. H. Mazzocchi, M. W. Bowen, and J. Kachinsky, J.C.S. Chem. Comm., 1977, 53.

Photolysis of Carbonyl Compounds

257

primary product is the ester (37) which undergoes photochemical deconjugation to afford (34). The minor product (35) is also formed by a deconjugation path involving the amide (38). This amide (38) is formed by the trapping of (36) by methylamine, a product of hydrolysis.

Irradiation of the steroidal ketone (39) in dry dioxan affords the epimerized product (40).58 A similar epimerization is encountered in the photochemical interconversion of ketones (41a) and (41b).39 The process in this case is dependent upon the nature of the substituent at C-3. When this is H as in (41c), the ketoaldehyde (42) is formed in high yield (at low photochemical conversions in benzene). It is argued that the presence of the substituent at C-3 prevents the 0

m

Me R (41) a ; R b; R

= =

cis-Me trcilis-Me

c;R=H

1

0

.

b; R

=

R

=

C;

Me, Y = H, X = Br Ph, Y = Otosyl, X = H

easy transfer of hydrogen either from the benzylic position or from the hydroxy function in the intermediate biradical (43). These reactions occur from the triplet state of the molecules (41). However, the singtet state is involved in the photochemistry of (44).39 These molecules undergo elimination of HX when irradiated and yield the isocoumarins (45). The exact nature of the intermediate involved in this transformation is not known aIthough an ionic mechanism has been discounted. The ketone (46) undergoes Norrish Type I fission when irradiated to yield an intermediate ketene intramolecularly trapped as the lactone (47).40 Another lactone (48) was also isolated as were the dehydration products of both (47) and 38

R. Jacquesy and H. L. Ung, Tetrahedron, 1976, 32, 1375.

*s

N. K. Hamer, J.C.S. Chem. Comm., 1977,239.

40

G. Ruecker and E. Dyck, Arch. Pharrn., 1976,309, 638 (Chem. A h . , 1977,86,43 832).

258

Photochemistry OH

OH

OH

(49)

(50)

(48). Intramolecular trapping is also involved in the photochemical conversion of the ketone (49) into the hemiketal (50).41 However, in this instance disproportionation of the biradical formed by Norrish Type I fission yields an aldehyde which undergoes intramolecular hemiketalization. Substituted cycloalkanones have been further studied. Thus Weiss et aZ.42 have reported that 2-methylcyclo-octanone and 2-methylcycloheptanone yield a-fission products (51) and (52) and (53) and (54) respectively as the major products. The full range of products formed in the photolyses is shown in Scheme 3. The cyclobutanol products (56) and (57) and (58) and (59) are formed by a second photochemical step involving Norrish Type I1 reactivity of the major products (51) and (52) and (53) and (54). In the case of the cycloheptanone a ketene is also OH

+

42

+

A. F. Thomas and M. Ozainne, Helo. Chini. A d a , 1976, 59, 1243. D. S. Weiss, P. M. Kochanek, and J. J. Lipka, Tetrahedron Letters, 1977, 1261.

Photolysis of Carbonyl Compoundr 259 produced which is trapped as the ester (55). On the basis of the stereochemical requirements within the transition state for disproportionation of the biradical(60) it is argued that conformation (61) will be preferred for C8 rings and above. Thus the alkenal ratio (51) : (52) is 1.57. For the C , system the preferred transition state (62) is reflected in the ratio (0.45) of the olefins (53) and (54).4a Several groups

of workers have studied the photochemistry of cyclododecanone (63a) where the photochemistry is dominated by hydrogen-abstraction processes from both the S, and Tl states. Weiss and Kochanek43 have examined the influence of substituents upon the reaction pathways in a study of the photochemistry of the methyl analogue (63b). The photochemistry of this compound is, surprisingly, dominated by a-cleavage leading to the products shown in Scheme 4. The yields

+ (63) a ; R b;R C; R

= = =

H Me CD3

(65) O (C,H,, 85%) (cyclohexane, 77%)

(C6H67 5%) (cyclohexane, 6%)

Me (C6Hf3,

lo%)

(cyclohexane, 17%)

Scheme 4

are shown in parentheses. The a-cleavage process leading to (64) and (65) is quenched by piperylene indicating that process to arise from the triplet state. The rate constant is in fact similar to that for 2-methylcyclohexanone. Not all of the or-cleavage reaction is quenchable to the same extent, and it seems likely that the formation of (64) and (65) has some singlet component. The disproportionation pathway for the formation of (64) and (65) was verified by the photolysis of the deuteriated material (63c). The ketone (66a) undergoes photochemical cleavage in methanol (with added sodium hydrogen carbonate) to afford the aldehydes (67a) and (68a) via Norrish a

D. S. Weiss and P. M. Kochanek, Tetrahedron Letters, 1977, 763.

260

Photochemistry

Type I reactions.44 In this case both modes of fission compete, yielding the two aldehydes in a ratio of 45 : 55. trans-Verbanone (66b) also yields two aldehydes (67b) and (68b) in a ratio of 40 : 60. These results illustrate the influence of the methyl group on the ratio of cyclobutane to cyclobutene products. Irradiation

(66) a ; R b;R

= =

CH,OAc CH,

(68)

(67)

of the adducts (69) and (70) in aqueous dioxan led to the isolation of the acids (71a and b) which were formed by Norrish Type I fission of the ring containing a carbonyl Similar fission was encountered in the irradiation of (72) in

(71) a ; R b; R

(72)

ti =

=

=

CH,CH=CI-ICH, (CH,),

1 or 2

An earlier publication 46 reported in note form the photochemical results of the cleavage of certain bicyclo-octanones. A full account of this work *' dealing specifically with the ketones (73) has now been published. The results reported 44 46

48 d7

P. D. Hobbs and P. D. Magnus, J. Amer. Chem. SOC.,1976, 98,4594. Y.Tobe, H. Omura, A. Kunai, K. Kimura, and Y . Odaira, Bull. Chem. SOC.Japan, 1977,50, 319. W. C. Agosta and S. Wolff, J. Amer. Chem. SOC.,1975,97,456. W. C. Agosta and S. W O EJ. Amer. Chem. SOC.,1976, 98,4182.

Photolysis of Carbony1 Compounds

261

R1 R2 R3 But Me H Me R' H Me H H H Me Me

R2 (73)

H Me H H H H H H H Me H

RI; H H H H H H

K7

R4

R5

H H H Me OMe H H H H H

H H H H H H But Me Me0 H

H

H

H H H H H H H H H H H H H H H H H H H H H H M e H H H M e

show that Scheme 5 is operative in the formation of products and that conformational relaxation of the six-membered ring is important in the determination of the ratio of products (ketene : aldehyde) formed. A further publication has given particular consideration to the conformational problems associated with

CH,CHO

O

R

R

H Scheme 5

the six-membered ring biradical (74) formed when the ketone (75a) undergoes Norrish Type I photochemical fission.** This conformational inversion can be followed by suitable deuterium labelling and Agosta and Wolff 48 have sought to evaluate this in a study of the two ketones (75b and c). In each case the

&Rl

(74)

R2

(75) a; R1 = R2 = H b; R1 = H, R2 = D C; R1 = D, R2 = H

aldehyde was isolated and was shown to be a mixture of the two deuteriated compounds (76) and (77), as in Scheme 6. The selectivity shown by the reaction is thought to arise from stereoelectronic control where there is interaction between the bond being broken within the biradical and the singly-filled adjacent p-orbital. W.C. Agosta and S. Wolff, J. Amer. Chem. SOC.,1976, 98, 4316.

262

Photochemistry 2%

90%

nrr*-Excitation of the gibberellin (78) in ethyl acetate results in the formation of the aldehyde (79) by Norrish Type I cleavage.49 The aldehyde is itself photoreactive and yields the oxetan (80)by [2 21 addition. Other multicyclic systems also undergo a-cleavage as in benzene-sensitized irradiation of the tricyclononanones (81a and b) via the intermediate aldehydes (82a and b): these were not isolated since they undergo ready photochemical [2 21 cycloaddition to afford

+

+

bR1

@

&Rl

0

( X I ) a ; R1 = R2 = Me b ; R1 = H, R2 = Me c ; R1 = H, R2 = ally1

(82)

(83)

(84)

R R

But

a; R = Ph

b; R

=

4-MeC6%

R Ph

(85)

(86)

the oxetans (83a and b).60 An oxetan (83c) was also isolated from the irradiation of (Sic), but this product was accompanied by (84) which is thought to be formed by Norrish Type I1 hydrogen abstraction from the allylic site. The resultant biradical then ring closes and undergoes dehydration to afford the isolated product. Pyrex-filtered irradiation of the adducts (85) in benzene solution affords a high yield of the ketene (86).61 dB

61

G. Adam and T. V. Sung, Tetrahedron Letters. 1976, 3989. P. hlargaretha, Helv. Chim. Acta, 1976, 59, 2902. F. Toda and K. Tanaka, J.C.S. Chem. Comm., 1976, 1010.

Photolysis of Carbonyl Compounds

263

3 Norrish Type I1 Reactions Alkyl ketones (4-methylpentan-Zone, heptan-2-one, 5-methylheptan-3-oneYand 5-methylhexan-2-one) were irradiated in several solvents in a study of solvent effects on the Norrish Type I1 elimination reaction.62 The effect of solvent is almost entirely due to the quenching of the excited state by the solvent and not because of a change in solvent polarity. Arrhenius parameters have been evaluated for the intramolecular photochemical abstraction of y-hydrogens in several alkyi ketones.63 The fate of the biradical produced by such an abstraction is reverse transfer, bonding to- yield a cyclobutanol, or fission to yield an enol. Thus the enol of acetone was obtained by U.V. irradiation of acetonepropan-2-01 mixtures at -70 0C.64The lifetime of the resultant enol was relatively long and there was no appreciable decay in 5000 s. Enols (87)can also be produced by the photochemical Norrish Type I1 reactions of the ketones (88) at - 70 oC.66

aRi2

r

R3

R4

R4

(87)

(88) R1

R1 R1 R1

R2

a; R1 =

b; R1 = c ; R2 = d ; R1 = e ; R1 = f; R1 =

&3R R' \ R'

R3

= R4 = Me, R3 = H = Me, R2 = R3 = R4 = H = R2 = H, R3 = Me, R3 = = Me, R2 = R3 = H,R4 = E

CH,C(Me), Ph

RZ

= R3 = H, 11 = 1 Me, R2 = R3 = H,12 = Me, R1 = R3 = H, 11 = R3 = Me, R2 = H, 12 = R2 = Me, R3 = H, 11 = R2 = R3 = Me,11 = 1

1 1 1 1

g;n h; II i ; IZ j; n k;n 1; ?I

= = = = = =

2 2 2 2 2

2

(89)

The influence of ring substitution upon the photochemical reactivity of the ketones (89a-f) has been studied in considerable detaiLss The results, which are shown in Table 1, illustrate the fact that 3-methyl substitution stabilizes the m* state of the benzoyl group more so than 4-methyl substitution. As the energy difference between the nm* and the m* states increases the rate of tripIet state hydrogen abstraction decreases. Similar studies were carried out for substituted valerophenones (89g-1).66 Other arylated ketones have also been studied. Thus Table I

Quantum yields and rate constants from photoreactions of (89).66 Ketone %I@Cyclobutanol kq7IM-l k~/108sec-l (894 (89b) (89d (894 (89d (890

aa 6*

6d

0.35 0.14 0.27 0.13 0.13 0.015

0.03 0.012 0.019 0.008 0.010

660 1140 3600 3100 6050

0.004

2650

7.6 2.6 1 .o 1.1 0.5 0.06

M. V. Encina and E. A. Lissi, J. Photochem., 1976,5, 287 (Chem. Abs., 1976,85, 93 466). M. V. Encina and E. A. Lissi, J. Photochem., 1977,6, 173 (Chern. Abs., 1977, 86, 105 486). A. Heme and H. Fischer, Helv. Chim. Acta, 1975, 58, 1598. A. Heme and H. Fischer, Angew. Chem. Internat. Edn., 1976, 15, 435. P. J. Wagner, M. J. Thomas, and E. Harris, J . Amer. Chem. SOC.,1976, 98, 7675.

264

Photochemistry

the direct irradiation of the ketone (90) in benzene or acetonitrile solution affords 2-acetonaphthone with a quantum yield of 0.01. The reaction of the ketone cannot be quenched and it is likely that it arises from the singlet nr* state. However, Wagner and Ersfeld67 point out that the Norrish Type I1 reaction could also arise from the triplet nn* state in competition with deactivation of this state to the lowest triplet (m*)state. The intersystem crossing in the molecule is 76% as efficient as that in benzophenone (a = 1.0). When the irradiation is carried out in methanol the quantum yield for the elimination reaction rises to 0.17,0.16 of which is quenchable by piperylene. The authors 67 suggest that while intramolecular exciplex formation occurs in all solvents only in protic solvents does the exciplex rearrange to a 1,4-biradical which permits the fragmentation reaction to take place. The Norrish Type I1 elimination reactions of the azetidones (91) lead, by the elimination of acetophenone, to the formation of the thioxoazetidones (92).68 The irradiations were carried out in acetonitrile solution

using Pyrex filtered light. Elimination is also encountered in the irradiation of the keto-ether (93) which yields the deformylated product (94).60 Hydrogen abstraction following photochemical excitation of the flavone (95) affords the biradical (96) whose fate is dimerization and loss of hydrogen to give (97).s0 Excitation to the Sz (m*)state of the thiones (98) leads to Norrish Type I1 reactivity.61*6 2 67 6*

6o 62

P. J. Wagner and D. A. Ersfeld, J. Amer. Chem. SOC., 1976, 98, 4515. A. Brandt, L. Bassignani, and L. Re, Tetrahedron Letters, 1976, 3975. T. G. Fourie, D. Ferreira, and D. G. ROUX,J.C.S. Perkin I, 1977, 125. R. Nakashima, K. Okamoto, and T. Matsuura, Bull. Chem. SOC.Japan, 1976,49, 3355. P. de Mayo and R. Suau, J. Amer. Chem. SOC., 1974,96, 6807. A. Couture, K. Ho, M. Hoshino, P. de Mayo, R. Suau, and W. R. Ware, J. Amer. Chem. SOC., 1976,98, 6218.

Photolysis of Carbonyl Compounds

(93)

(98) R

=

CH,CH,Me, (CH,),Me, CH,CHMe,, (CH,)Me, (CH&Ph, CH,CH,

Ring closure within a 1,6biradical is also used in the synthesis of multicyclic systems, e.g. in the quantitative photochemical conversion of the ketone (99) into the tricyclohexanol (100). The authors63claim this as the first example of the tricycl0(2,2,0,O~~~)hexane ring system. Irradiation of the xanthone (101) leads to

the photoenol (102) by a Norrish Type XI hydrogen-abstraction process. The The results of a flash photoenol could be trapped with oxygen or metric study of the photoenolization of several acetophenones have been published.6s 64

E. C. Alexander and J. Uliana, J. Amer. Chem. SOC.,1976, 98, 4324. W.A. Ayer and D. R. Taylor, Canad.J. Chem., 1976,54, 1703. D. M. Findlay and M. F. Tchir, J.C.S. Furaduy Truns. I, 1976, 1096.

Photochemistry

266

(101)

( 102)

A study of the Norrish Type I1 cleavage reaction of the esters (103) in cyclohexane has been carried out.66 The quantum yield for the formation of acid (Scheme 7) is dependent upon the nature of the substituent on the ester side-

a ; (1)

(103) a ; X b;X c; X d;X

= = = =

x

=

Ni* W

g;X h;X

= =

CH,NMe, SMe Scheme 7

Me Ph OMe OEt e ; X = NMe,

f;

=

b ; (1) = C ; (1) = d ; (I) = e ; (1) = r; (I) = g ; (1) = h; (I) =

0.003 0.008 0.023 0.009 0.15 0.1

0.018 0.08

chain. Reasonable efficiency is found when the substituent contains an aminoor thio-group. This efficiency is thought to result from the operation of an electron-transfer mechanism. The use of Norrish Type I1 elimination reactions for synthetic purposes has seen two further applications in the past year. Thus U.V. irradiation of pyruvic esters provides a method for the oxidation of alcohols to aldehydes and ketones.s7 A typical example of this is shown in Scheme 8. The carbohydrate esters (104a)

0

0

Scheme 8

undergo Norrish Type I1 elimination to afford the methylene derivative (105). In competition with this process Norrish Type I fission affords the saturated derivative (104b).68 A major product of the photolysis of the adamantyl ester (106) has been identified as the hydroxyketone (107).69970 This product is thought to arise from the 66

67 68

6s lo

J. D. Coyle and D. H. Kingston, J.C.S. Perkin ZZ, 1976, 1475. R. W. Binkley, Synth. Commun.,1976, 6,281 (Chem. Abs., 1976, 85, 142 105). R. W. Binkley and J. L. Meinzer, J. Carbohydrate, Nucleosides, Nucleotides, 1975, 2, 465 (Chem. Abs., 1977,86, 5695). N. A. Marron and J. E. Gano, J. Amer. Chem. Soc., 1976,98,4653. J. E. Gano and L. Eizenberg, J. Amer. Chem. Soc., 1973,95,972.

Photolysis of Carbonyl Compounds

267

CH,XR

(104) a; R = Me, I%, CH,Ph, CH,CH,Ph, Ph,CH or Ph,MeC; X = OCO

b;RX

=

(105)

H

path outlined in Scheme 9 whereby irradiation of the ester in methanol (254 nm) affords the biradical (108) which either fragments to yield adamantene (109) or ring closes to the unstable oxetanol (110). This material, which was not isolated, undergoes ring opening to afford the observed product (107).

Mazzocchi and Bowen 72 have examined the photochemistry of a number of alkyl amides and have come to the conclusion, in agreement with other earlier that the Norrish Type I1 process yielding (1 11) and (112) from amide 71p

(114) a ;

R'

b; R' 71

73

= =

R2 = Me Ph, R2 = H

(115) a ; R = H b;R-R = (CH=CH),

P. H. Mazzocchi and M. J. Bowen,J. Org. Chem., 1976,41, 1279. P. H. Mazzocchi and M. J. Bowen, J. Org. Chem., 1975,40,2689. C. H. Nicholls and P. A. Leermakers, J. Org. Chern., 1970,35,2754. 10

Photochemistry

268

(113) is extremely inefficient. The reaction of the amides is dominated by the Norrish Type I fission and this induces radical reactions with the solvent. Low efficiency is also observed in the Norrish Type I1 reactions of the benzamides (1 14),73an observation consistent with the fact that the luminescence characteristics of the benzamides (114) indicates the lowest excited state to be TT* in ~haracter,'~However, when a charge transfer mechanism is involved, as with (1 15), the efficiency of the elimination reaction is greatly enhanced 74 (see ref. 57). 1,SHydrogen transfer reactions also arise in medium-sized ring systems such A more as cyclododecanone which yield the products shown in Scheme

(D

=

0.004

Q) = 0.043((Dtri1, = 0.041 ; (bsing = 0.002) Scheme 10

recent study has provided quantitative information and has shown that the compounds produced are all primary p h o t o p r o d u ~ t s .The ~ ~ quantum yields for the production of the various products are shown under the appropriate structure. Norrish Type I1 photochemical reactivity has been described for the hydroxyketone (116) which yields (117a) and (117b) probably by the route shown in Scheme 11.77

(1 17b)

Scheme 11 74

75

76

77

J. D. Coyle and D. H. Kingston, Tetrahedron Letters, 1976, 4535.

B. Camerino and B. Patelli, Experientiu, 1964, 20, 260; T. Mori, K. Matsui, and H. Nozaki, Tetrahedron Letters, 1970, 1175, Bull. Chem. SOC.Japan, 1971, 44, 3440; K. H. Schulte-Elte, B. Willhalm, A. F. Thomas, M. Stoll, and G. Ohlaff, Helv. Chim. Acta, 1971, 54, 1759. P. J. Burchill, A. G. Kelso, and A. J. Power, Ausrrul. J. Chem., 1976, 29, 2477. A. Marchesini and U. Pagnoni, Gazzerta, 1976, 106, 663 (Chem. Abs., 1977, 86, 88 773).

Photolysis of Carbonyl Compounds 269 4 Rearrangement Reactions A full account of the photochemical Favorskii reaction of chlorocyclobutanones (1 18) has been '@ A typical example of this is the photochemical ring contraction of (1 18a) which affords the products shown in Scheme 12. The ringcontraction reaction is dependent on the stereochemistry of the C1-substituent.

R1 = Me, R'

R'

=

= C0,Me CO,Mc, R? = M e

Scheme 12

A previous study of the lactam (119) showed that it undergoes ring expansion in methanol to afford the lactam (120).s0 A further investigation of the reaction has suggested that the ring expansion involves the formation of a zwitterionic intermediate (121).81 The invo€vement of such an intermediate is indicated by the dependence of the reaction upon the concentration of methanol. The ringexpansion reaction only takes place with lactams fused to the bicyclo(2,2,1) ring /

NR

NR

(119) a ; R = H b ; R = Me

'@

8o

G. Jones, jun., and L. P. McDonnell, J.C.S. Chern. Comm., 1976, 18. G. Jones, jun., and L. P. McDonneU, J. Amer. Chem. SOC.,1976,98, 6203. H. L. Ammon, P. H. Mazzocchi, W. J. Kopecky, jun., H. J. Tamburin, and P. H. Watts, jun., J. Amer. Chem. SOC.,1973,95, 1968, P. H. Mazzocchi, T. Kalchak, and H. J. Tamburin, J. Org. Chem., 1976, 41, 2808.

Photochemistry

270

P11

(127) Scheme 13

system, Thus the rearrangement of (122) takes place to afford (123), but the lactams [e.g. (124) and (125)] do not yield ring-expanded products although other products were formed.81 A full account 8 2 of photochemical transformations involving enols has been published following earlier preliminary r e p o r t ~84. ~ Thus ~ ~ the 3-chromanone (126) photoisomerizes into the isomeric compound (127) via the enol intermediates shown in Scheme 13. Other compounds such as (128) also photorearrange by way of enols (Scheme 14).84a98 5 An enol is also thought to be involved in photo-cis-trans-isomerizationof the flavones (1 29).86

82

83 84 86 88

A. Padwa, A. Au, G . A. Lee, and W. Owens, J . Amer. Chem. SOC.,1976,98, 3555, A. Padwa, and G. A. Lee, J. Amer. Chem. SOC., 1974,96, 1634. (a) A. Padwa, and A. Au, J. Amer. Chem. SOC.,1974,96, 1633; (b) 1975, 97,242. A. Padwa and A. Au, J. Amer. Chem. SOC.,1976, 98, 5581. T. G. Fourie, D. Ferreira, and D. G. ROUX, J.C.S. Perkin I, 1977, 125.

Photolysis of Carbonyl Compounds

27 1

&OMc \

d\

0

I

M

c

0

0

(131)

( 130)

(132) a ; R' b ; R'

I

=

Ph, K' H , R'

= €3 = Ph

(133) 3 : R' h ; R'

-=

=

PI), H , R'

H == 1'11

Irradiation (high pressure Hg lamp) of the flavone (130) in benzene solution affords the hydroxychalcone [(131), 14%].60 Pete and his co-workers have continued their investigation of the photolytic reactions of keto-epoxides and have examined the transformations of the isomers (132) and (133).87 Irradiation was carried out using a Pyrex filter and benzene solutions and gave the products shown in Scheme 15. Subsequent investigations

Scheme 15

showed that (134) was the principal product and that secondary irradiation of this compound gave all the others formed, Separate irradiation of (134) gave an additional product (135) which is formed by a-fission in the dicarbonyl compound (134) and rebonding within the biradical (136). In competition with this mode of

a7

J. Muzart and J. P. Pete, Bull. SOC.chim. France, 1976, 1953.

272 (134)

’I”

f

a!

o-:..,

Photochemistry (137)

+

I’IlCIIO

H l’hC=O

Scheme 16a

a-fission, the alternative pathway shown in Scheme 16a leads to the olefin (137) when the radical pair diffuse apart from the solvent cage in which they are initially formed. The other two products (137) and (139) arise as a result of y-hydrogen abstraction. Thus (137) and (139) are formed as shown in Scheme 16b where

(138)

Scheme 16b

the two processes are shown. In other keto-epoxides Muzart and Pete have studied epimerization and have proposed that this probably occurs by a carbonyl ylid path involving the fission of the C-C bond of the oxiran ring. However, the epimerization path is always minor even when a phenyl group is attached to the oxiran ring, e.g. as in (140). Under conditions of acetone sensitization, the oxiran (140) affords the adduct (141). Studies on the epimerization reactions of (142) have also been carried out by the same Ring opening of the ketoepoxide (143) to the 1,3-diketone (144) has also been A variant on H

88

J. Muzart and J. P. Pete, Tetrahedron Letters, 1977, 303. J. Muzart and J. P. Pete, Tetrahedron Letters, 1977, 307.

Photolysis of Carbonyl Compounds

273

this reaction has been reported for the photochemical rearrangement of the ketone (145) to the ketoaldehyde (146).B0 Presumably the reaction involves fission of a C - 0 bond in the oxiran ring to afford the biradical (147) which, perhaps because of the driving force of aromatization, follows the path shown in Scheme 17.

( I 47)

(145)

Scheme 17 5 Oxetan Formation The selectivity of the photochemical addition of benzophenone to the methylenenorbornene (148) has been studied and compared with the photochemical additions of the same ketone to methylenenorbornane (149) and norbornene.Ol The products from the addition reaction are the two oxetans (150a) and (150b) and there is no trace of a product derived from the endocyclic double bond. The

( 149)

( I 50) a ; R' b ; R'

= =

Ph, R' = 1-1 H, R2 = Ph

R1 = Ph, R2 = H b ; R 1 = H, R2 = Ph

(151) a ;

P. S. Venkataramani, N. K. Saxena, R. Srinivasan, and J. Ors, J. Org. Chem., 1976,41,2784. A. A. Gorman, R. L. Leyland, C. T. Parekh, and M. A. J. Rodgers, Tetrahedron Letters, 1976, 1391.

Photochemistry

274

(153)

(154)

methylenenorbornane (149) affords (151a) and (151b) while norbornene yields (152), a product which has been previously A kinetic study compared the quenching efficiencies of the three olefins with respect to the ketones, but there is little evidence to suggest that the additional double bond in the methylenenorbornene (148) plays any major part. The authors 91 suggest that the most likely reason for the specificity of the reaction of (148) is preferential formation of a biradical complex at the exocyclic site where steric factors will be at a minimum. Irradiation of the benzophenone derivative (153) in norbornene yields the oxetan (154) showing that the triplet-excited carbonyl group was reactive.93 Under conditions where no cycloaddition could take place, and in the presence of a hydrogen-donating solvent, hydrogen abstraction by the benzoyl carbonyl group dominates the reactions. Rivas and Bolivar O4 have reported the photochemical addition of the ketones (155) to l-benzoylpyrrole affording the 2 : 1 adducts (156). Oxetans (157) are also produced when the imidazoles (158) are irradiated in the presence of benzophenone.06 l-Alkylated imidazoles gave products of the addition of ketyl radicals

(155)a;X b;X C; X

=

Y = CII N , Y = CH Cli, Y = N

(1 57) R

=

Ac, PhCO, or CO -

= =

COPh

83

D. Scharf and F. Korte, Tetrahedron Letters, 1963, 821 ; D. R. Arnold, R. L. Himman, and A. H. Glick, Tetrahedron Letters, 1964, 1425. Y . Tsujimoto, Y . Shigemitsu, T. Miyamoto, and Y . Odaira, BUN. Chem. Soc. Japan, 1976,

84

49, 1445. C. Rivas and R. A. Bolivar, J. Heterocyclic Chem., 1976, 13, 1037.

OL

T. Nakano, C. Rivas, C. Perez, and J. M. Larrauri, J. HeterocycZic Chem., 1976, 13, 173.

82

Photolysis of Carbonyi Compounds

275

(160) a ; R b;R

=

Ac

=

H

to the alkyl group. The adduct (159) is formed when benzophenone is added photochemically to the azaindole (160a). The acetyl group appears to be important for the success of the reaction since the irradiation of benzophenone in the presence of (16Ob) failed to yield a A full account of the photochemical addition of propanal to cyclohexa-1,3diene has been 98 The products formed are shown in Scheme 18. H

ECHO

+

H cyclohexadiene dimers

H +

Scheme 18

The dimers of cyclohexadiene are thought to be formed by the direct excitation of the diene yielding a triplet state which subsequently attacks and adds to a ground-state diene molecule, but the oxetans are formed via the SIstate of the propanal. Indeed a singlet exciplex mechanism has been suggested and substantiated by a kinetic analysis of the reaction. A detailed study of the photochemical formation of oxetans (e.g. Scheme 19) from acetaldehyde, pivalaldehyde, 0 11

R-CH

+IR-/

h",3,0nm+]R~o

+ [-RJ

z -

a; 6 b; 5 c; 8 d; 6

a ; 54 b ; 53 c; 53 d ; 54

a ; 12 b; 14 c; 19 d ; 13

R

R

=

a, Me; b, But; c, c-C,H,; or d, c-C,H,

c; 17 d ; 23

Scheme 19 96

97 9a

T. Nakano and M. Santana, J. Heterocyclic Chem., 1976, 13, 585. T. Kubota, K. Shima, S. Toki, and H. Sakurai, Chem. Comm., 1969, 1462. K. Shima, T. Kubota, and H. Sakurai, Bull. Chem. SOC.Japan, 1976,49,2567.

Photochemistry

276

cyclopropane- and cyclobutane-carbaldehydes with dienes and alkenes has shown that the reaction occurs from the S1state of the aldehyde.sg Indeed the oxetan-forming reaction is predominant with intermolecular hydrogen-abstraction reactions minimally in competition, A detailed kinetic study reveals that exciplex formation does occur but that only 10%of exciplex species forms oxetan and the predominant reaction of the exciplex is decay to the ground state. The regiospecificity of the reaction follows the pattern established for other systems whereby the direction of addition is dictated by the relative stabilities of the possible 1,4-biradical intermediates. The oxetans are also formed by addition to the more substituted double bond of the dienes, again an observation which has literature precedence. One problem encountered is associated with the ease of ring opening of cyclopropyl radicals. Such species would be generated if a 1,4-biradical is intermediate in the formation of oxetans. However no evidence for such a ring-opening process was detected. The authorsQBargue that the absence of ring opening must be due to the fact that oxetan formation must be los s-l) in agreement with the proposal that a singlet 1,6biradical faster (k is involved. A copolymer containing oxetan linkages has been prepared by the U.V. irradiation of aromatic diketones in the presence of tetramethylallene. N

6 Fragmentation Reactions Martini lol has described the photochemical decarbonylation of the fluorinated ketonic ethers [e.g. (161)] yielding the ethers [e.g. (162)l. Irradiation of the spirocompounds (163), in benzene gives the cyclic ketones (164) via decarbonylation and loss of nitrogen.lo2

(163)

=

1,2, or 3 (1 64)

The hydroxy-ester (165) undergoes photochemical decarboxylation to afford (166).60 Photodecarboxylation of the acid (1 67) in acetonitrile has been reported.lo3 An e.s.r. and CIDNP study of the sensitized decarboxylation of C. W. Funke and H. Cerfontain, J.C.S. Perkin ZI, 1976, 1902. D. J. Andrews and W. J. Feast, J. Polymer Sci.,Polymer Chem. Edn., 1976, 14, 319 (Chem. Abs., 1976, 84, 122 388). Io1 T. Martini, Tetrahedron Letters, 1976, 1865. l o a H. Meier and D. Daniil, Chem.-Zrg., 1976,100, 89 (Chem. A h . , 1976,84, 164 692). l o 3 N. Suzuki, Y.Fujita, T. Yamabayashi, Y. Deguchi, and Y . Izawa, J.C.S. Perkin Z, 1976,1901. Bg

loo

277

Photolysis of Carbony1 Compounds

0

@ 0

( 169)

carboxylic acids has been carried A novel example of decarboxylation has been reported in the photolysis of the diester (168) which gives a high yield of the paracyclophane (169).loS A study of the photochemical reactions of the pyrethroid (170a) has been carried out.lo8 In hexane only three products (170b), (17Oc), and (171) are produced, but in methanol the reaction is very complex yielding products of isomerization, debromination, and products from the fission of C-0 bonds via radicals (172) and (173).lo6 Benzophenone sensitization is not very efficient, but yields an increased amount of the isomerized product (170b). The cyclobutanones (174) and (175) undergo photochemical conversion into a ketene (176) which is trapped as the ester (177) when the irradiation is carried out in methan01.l~' The fragmentation of the cyclobutanone to the vinylketene presumably involves

(170) a ; R1 = H,

Ra = Br2C=CH b; R1 = Br,C=CH, RZ = H c; R1 = N, R2 = BrCH=CH

(171)

P. R. Bowers, K. A. McLauchlan, and R. C. Sealy, J.C.S. Perkin ZI, 1976, 915. M. L. Kaplan and E. A. Truesdale, Tetrahedron Letters, 1976, 3665. loo L. 0. RUZO,R. L. Holmstead, and J. E. Casida, Tetrahedron Letters, 1976, 3045. lo' R. D. Miller, D. L. Dolce, and V. Y . Merritt, J. Org. Chem., 1976, 41, 1221. lo'

lo6

Photochemistry

278

(174)

q) (1 75)

0

0

..-c=c=0

0

..- ...&

o:, ( 177)

0 ( 179)

the fission of bond ‘a’ in (174) and (175). The cyclobutanone derivative (178), however, does not follow this fragmentation path and instead undergoes decarbonylation to (179) and loss of ketene to yield cyclohexa-1,4-diene.lo7 Irradiation of benzophenone in the presence of tetraphenyldiphosphine results in fission of the P-P bond probably as a result of Schenck-type attack of benzophenone on the phosphine. After sulphurization the products are tetraphenylethylene, thiobenzophenone, and phosphinic acids, the formation of which is in accord with the route suggested above.lo8 lo*

R. Okazaki, K. Tamura, Y. Hirabayashi, and N. Inamoto, J.C.S. Perkin I, 1976, 1924.

2 Enone Cycloadditions and Rearrangements : Photoreactions of Cyclo hexadieno nes and Qu inones BY W. M. HORSPOOL

1 Cycbaddition Reactions Intramolecular.-A biradical pathway is thought to be involved in the photochemically induced cyclizations of the dienones (1) yielding (2).l The conditions for the reactions are critical since many side-products can be formed. Thus whereas cyclization to (2) can predominate in hydrocarbon solvents, the use of methanol can give a methyl ester via a keten, e.g. ester (3) from (lc) according to Scheme 1. Tamura ef aLahave previously demonstrated that the photocyclization R1 R2 * R 4 0

R2A R 4

R3

R1 = C1, R2 = R3 = R4 = H b ; R1 = H,R3 = OAC, R3 = R4 = H

(1) a; C;

R'

d ; R'

RZ = = R2 =

=

H, R3 = CI, R4 = H RJ = H, R4 = Cl

(2) a; 13% b ; 8% c ; 1% d ; 52%

MeOH

hv

___j

G O M e

c1 (3) Scheme 1

+

of cyclohexenone (4a) gives the crossed [2 21 product (5).2 In an extension of their studies,8 they have examined the photochemistry of closely related compounds bearing terminal methyl groups. In this case the reaction path does not follow that originally outlined, but diverges to yield the tetrahydrofuran derivative f6a) from (4b) and the pyrrolidine (6b) from (44. The reaction is essentially an ene-reaction and probably involves initial interaction between the 1,$-diene to yield a biradical (7) which does not ring-close as in the previous examples. Instead, as a result of the steric factors introduced by the methyl groups, a hydrogen-abstraction path is followed. This was fully demonstrated for the deuteriated compound (4d) which yielded (6c). a

F. T. Bond, C-Y. €30, and 0.McConnell, J. Org. Chem., 1976,41, 1416. Y . Tamura, H. Ishibashi, M. Himi, M. Ikeda, J. Org. Chem., 1975,40,2702. Y. Tamura, H. Ishibashi, and M. Ikeda, J. Org. Chern., 1976, 41, 1277.

279

Photochemistry

280

H, X = 0 or NMe Me,X = 0 c;R M e , X = NMe d ; R = CD,, X = 0

(4) a; R b;R

(6) a; X b; X C; X

&/ ""X

=

= =

= = =

0, R1 = H, R2 = Me NMe, R' = H, R2 = Me 0,R1 = D, R2 = CD,

(7)

The allyl-substituted cyclohexenones (8) are all photochemically reactive and can be converted by irradiation (366 nm) in cyclohexane or acetonitrile into the The quantum yields for the processes are given tricyclononanones (9) and

(8) R = H, Me, or Pri (9) R = H, 100% (0 = 0.19) (10) R R = Me, 79% (CD = 0.14) R R = Pri, 67% (0= 0.20) R

= = =

H, Me, 21% ((D = 0.14) Pri, 33% (0 = 0.14)

in parentheses. The influence of solvent polarity on the mode of addition was minimaL4 In the cyclopentenone series, Agosta and his co-workers have shown that a benzene solution of (11) upon irradiation through a Uranium glass filter (A > 340 nm) afforded the tricyclic compound (12). Closely related to this type of intramolecular [2 +2] addition is the synthesis of cage compounds. Thus the cage ketones (13) are readily obtained by photolysis of the adducts (14).6 The cage ketones (1 3) are themselves photochemically labile and undergo decarbonylation to afford (15) which can be formed by the photochemical [2 21 addition of the dienes (16). A description of the synthesis of deuteriated cubanes has been published.' The key step in the reaction sequence is the photochemical ring closure of the dienone (17) to the bishomocubane (18). A study of the cage formation of (19) from (20) by photolysis in the presence of rhodium(1) catalysts has been reported.8

+

W. Frost1 and P. Margaretha, Helv. Chim. Acta, 1976,59,2244. S . Wolff, S. Ayral-Kaloustian, and W. C. Agosta, J. Org. Chem., 1976, 41, 2947. T. Tezuka, Y. Yamashita, and T. Mukai, J. Amer. Chern. SOC.,1976, 98, 6051. E. W. Della and H. K. Patney, Austral. J. Chem., 1976, 29, 2469. G. Jones and B. R. Ramachandran, J. Photochem., 1976,5, 341.

28 1

Enone Cycloadditions and Rearrangements

9 (11)

Me

Photochemistry

282

+

A few years ago Becker reported the photochemical [2 21 intramolecular cycloaddition of cyclohexa-2,4-dienones to afford cage compounds. Yamamura e l aLl0 have now reported the analogous reaction in the photochemical conversion of the neolignan (21a) into (22a). A similar cycloaddition occurs when

R

M e 0 OMe

Me

Me0 ‘(21) a ; R = allyl

b; R

=

M e 0 OMe

0

Prn

oH

OMe

OMe

the reduced compound (21b) is converted into (22b). Both of the above conversions took place when the light employed was filtered through potassium chromate solution. When a Pyrex filter was used for the irradiation, compound (21b) was converted into two oxetans (23) and (24). This reaction presumably follows the initial cyclization process [to (22b)l and then involves one or two Norrish Type I1 hydrogen abstractions and cyclization to afford the oxetans. A head-to-tail cage compound (25) is produced by irradiation of the naphthalenone

qo allyl

0

& allyl

0

lo

H.-D. Becker, Annalen, 1973, 1675. S. Yamamura, Y. Terada, Y.-P. Chen, M. Hong, H.-Y. Hsu, K. Sasaki, and Y.Hirata, Bull. Chem. SOC.Japan, 1976,49, 1940.

Enone Cycloadditions and Rearrangements

28 3

dimer (26).11 No evidence for head-to-head addition was found. Head-to-head cycloaddition is, however, encountered in the Pyrex-filtered irradiation of a methylene chloride solution of the dienone (27) which yields the [2 21 adduct (28) (50%).12

+

Another mode of [2 + 21 cycloaddition is found in enone (29) which affords the oxetan (30) by irradiation (Pyrex filter) in benzene s01ution.l~ Oxetans (31) are also formed from the irradiation of the acylnorbornenes (32) in benzene s01ution.~* Details of the photochemistry of the aldehydes (33) have been reported l6 in supplementation of the results previously reported in note form.lS

R', ,R1

0

(33) R'

=

R2 = H

R1 = H, R2 = Me

Intermolecular.-Several reviews have been published during the year. Otsuki et a1.l' have published a short review of photochemistry and Pete has reported l1

Y. Nakamura, J. Zsindely, and H. Schmid, Heterocycles, 1976,5,427 (Chem. Abs., 1977,86, 89 650).

la

l9 l4 lo

l7

W. Oppolzer, R. Achini, E. Pfenninger, and H. P. Weber, Helv. Chim. Acta, 1976, 59, 1186. N. F. Feiner, G. D. Abrams, and P. Yates, Canad. J. Chem., 1976, 54, 3955. R. R. Sauers, R. Bierenbaum, R. L. Johnson, J. A. Thich, J. Potenza, and H. J. Schigar, J. Org. Chem., 1976, 41,2943. B. Guiard, B. Furth, and J. Kossanyi, Bull. SOC.chim. France, 1976, 1552. J. Kossanyi, B. Guiard, and B. Furth, Bull. SOC.chim. France, 1974, 305. T. Otsuki, A.Takuwa, and K. Maruyama, Yuki GosefKagaku KyokaiShi, 1976,34,734 (Chem. A h . , 1977, 86, 88 461). J. P. Pete, C. R . Congres Nut. SOC.Sauantes Sect. Sci., 1975, 95, 87 (Chem. Abs., 1976, 85, 62 230).

Photochemistry

284

on some aspects of the photochemical addition of olefins to enones. A review has been published on the synthetic utility of bicyclo[3,2,0]heptanones. Many of these compounds have been synthesized by photochemical addition of olefins to cyclopentenones.lD Margaretha 2o has reviewed the photochemical cycloadditions of cyclic oxa-enones. An account of the deactivation encountered in photochemical cycloaddition reactions in the solid phase has been published.21 An examination of the efficient photochemical addition reactions of several styrenes (34) to the enols of the two 1,3-diketones (35) has provided a novel route to 1,5-diketones (36).22 The additions of the styrenes to the enol are stereo- and regio-specific, e.g. the addition of the E-propene (34d) yields the threo-ketone (37) (Scheme 2). The [2 21 photocycloaddition of the methyl carboxylate (38)

+

I'll W

R

to cyclopentene is reported to yield the adduct (39)(67%).23 Photochemical addition of the same carboxylate (38) to isophorone (40) and to ethoxy vinyl ether yielded (41) and (42) respectively. 19 20

21

22

23

S. M. Ali, T. V. Lee, and S. M. Roberts', Synthesis, 1977, 155. P. Margaretha, Chimiu (Switz.), 1975,29, 203 (Chem. Abs., 1975, 83, 177 511). G. N. Gerasimov, 0. B. Mikova, E. B. Kotin, and A. D. Abkin, V. sb., XI Mendeleeusk. S'ezd Obshch. i Prikl. Khimii. Ref. Dokl. i Soobshch., 1974, 151 (Chem. Abs., 1976, 84, 180 742). P.-F. Casals, J. Ferard, and R. Ropert, Tetrahedron Letters, 1976, 3077. P. A. Wender and J. C. Lechleiter, J. Amer. Chem. SOC.,1977, 99, 267.

Enone Cycloadditions and Rearraagements

285

High yields of the cycloadducts (43a) (74"A and (43b) (80%) were obtained when the enone (44)was irradiated in cyclohexa-l,4-diene and cyclo-octa-ly5-diene re~pectively.~~ The resultant adducts (43) were irradiated in the presence of the enone (44) which yielded the 2 : 1 adducts (45a) (25%) and (45b) (31%). These

(43) a ; 11 = 1 b;tt ='2

(44) 0 H

& R

0 =

H or Me

(46)

(47)

'

,OSiMe,

dSiMe3 (48)

?

OSiMe A 3

n

-I--.

,

(49)

adducts were converted into propellane derivatives. The synthetic usefulness of [2 + 21 cycloaddition reactions to enones has been exploited many times. A further example of this reaction, with specific application to the hydroazulene skeleton, has been reported following the photochemical [2 + 21 addition of cyclopentenones (46) to the silyloxy-cyclopentene (47).25 This affords the two isomers (48) and (49) (total yield ca. 70%). In cycloaddition reactions of this type there is always the question of regio-selectivity. Thus does the olefin add solely to the C=C or is there some contribution from an oxetan-forming process? Thi and Margaretha26have examined this point with respect to the enones (50) and have shown that with (50a), on irradiation at 366 nm in acetonitrile, oxetan formation predominates.26 When cyclohexane is used as solvent oxetan formation is the exclusive reaction path. These results are summarized in Table 1 which also includes results from other cyclopentenones. The results clearly indicate the profound effect which fluoro substituents can have upon the cycloaddition reactions. A kinetic analysis of the process was carried out.26 Irradiation of benzene

B5

e*

Y. Tobe, H. Omura, A. Kunai, K. Kimura, and Y. Odaira, Bull. Chern. SOC.Japan, 1977, 50, 319. D. Tremont, P. De Clercq, D. De Keukeleire, and M. Vandewalle, Synthesis, 1977, 46. G. V. Thi and P. Margaretha, Helv. Chim. Acta, 1976, 59, 2236.

Photochemistry

286

(50) a ; R1 = F, R2 = Me (51) a ; R' = F, R2 = Me (52) a ; R1 = F, R = Me b; R' = H, R2 = Me b ; R1 = H, R2 = Me b ; R1 = H, R2 = Me C ; R' = R? = H C ; R' = RZ = H C; R' .= R2 = H

Yields ofproduct from the irradiation of (50) with 2,3-dimethyZbut-2-ene 26

Table 1

Enone

Producrs (%)

Solvent

(504

C6H12

(504

MeCN

(50b) (50b) (50c) (50c)

C6H12

MeCN

(51) 100

(52)

75 53 29

25 47

-

C6H12

-

MeCN

I

71 100 100

solutions of the furanone (53) in the presence of methyl acrylate and vinyl acetate gave the adducts (54). Cycloaddition to cyclopentene gave (55).27 The photoaddition of 3-methylbut-3-en-l-yne to cyclohex-2-enone (via its triplet state) affords six ethynylbicyclo[4,2,0]octanones where head-to-tail

6

Me

0

0

i3Me

(53)

@ R

= C0,Me or OAc, R2 = R3 = H ( 5 5 ) R = a-or 13-H b; R' = R3 = H, R2 = C0,Me or OAc c; R1 = R2 = H, R3 = C0,Me or OAc

(54) a ; R'

adducts predominate.28 The photoaddition of aldehydes (e.g. butyraldehyde) to cyclohex-2-enone results in the formation of 3-butyrylcyclohexanone (56).2a Various other enones and aldehyde pairs also react in this fashion to yield

T. Ogino, T. Kubota, and K. Manaka, Chem. Letters, 1976, 323 (Chem. Abs., 1976, 85, 20 982). as 29

A. K. Margaryan, E. P. Serebryakov, and V. F. Kucherov, Izvest. Akad. Nauk S.S.S.R., Ser. khim., 1976, 840 (Chem. Abs., 1976, 85, 77 116). D. R. Hicks, R. C. Anderson, and B. Fraser-Reid, Synth. Comm., 1976, 6,417 (Chem. Abs., 1976, 85, 176 346.)

Enone Cycloadditions and Rearrangements

287 1,4-diketones. Dimedone enol acetate (57) undergoes photoaddition to cyclopentene to afford the adducts (58) and (59).30 Wiesner 31 has rationalized the photochemical cycloaddition reactions of olefins to enones in terms of a tetrahedral /3-carbon of the enone with an electron-rich orbital in a pseudo-axial orientation. Cargill et have carried out a test of this hypothesis using the enone (60). This, in its excited state, according to the Wiesner model, should have the conformation shown in (61). The [2 21

+

0

photoaddition of ethylene to (60) yields (62) exclusively. The formation of this product (whose structure was determined by X-ray) clearly lends weighty support to the Wiesner proposals. Another example of this selectivity has been reported by Ziegler and Kloek33who found that the photochemical addition of allene to the enone (63) afforded the [2 + 21 adduct (64) as the major product. This was

w

Me

*$.

C0,Me

accompanied by a minor addition product whose constitution was not fully investigated. The results obtained by Jones and KearnsY3* from a study of the electronic structure of the enones (65)-(67), has suggested that the excited state geometry is almost the same as that of the ground state and that there is little distortion of the molecule. Furthermore, the triplet state of these enones indicates that it is m* in type. Clearly, further work has to be carried out to rationalize, in

sa

M. Umehara, T. Oda, Y. Ikebe, and S. Hishida, Bull. Chem. SOC.Japan, 1976, 49, 1075. K. Wiesner, Tetrahedron, 1975, 31, 1655. R. L. Cargill, T. A. Bryson, L. M. Krueger, J. V. Kempf, T. C. McKenzie, and J. Bordner.

84

J. Org. Chem., 1976,41, 4096. F. E . Ziegler and J. A. Kloek, Tetrahedron, 1977,33, 373. C. R. Jones and D. R. Kearns, J. Amer. Chem. SOC.,1977,99, 344.

81

Photochemistry

288

OH

(66) R = H or Ac

(65)

(67)

electronic terms, why molecules such as (60) and (63) behave in such a selective fashion in cycloaddition reactions. Photoadditions of ethylene and 2,3-dimethylbut-2-ene to the gibberellin (68) have been reported35 to afford the [2 21 adducts (69). The structure of the adduct (70) has been determined by X-ray d i f f r a ~ t i o n . ~ ~

+

K

(69) R

=

R

=

H, l a , 2%and Ip, 2/3 Me, ICY, 23 and lp, 2 p

The irradiation (Pyrex filter) of solutions of the cyclic enones (71) in hexane with added isoprene and cyclopentadiene results in the formation of the DielsAlder adducts (72) and (73).37 The formation of the adducts provides reasonable evidence for the formation of reactive trans-enones (74) following photochemical excitation.

Dimerization.-A review of the photochemical dimerization of cinnamic acid and its derivatives in the crystalline state has been published.38 Whitten and co-workers 39 have studied the solid-state dimerization of the octadecyl esters of cinnamic and p-chlorocinnamic acids. The esters of cinnamic acid crystallized in three forms, one of which was photochemically stable. The other two crystalline modifications gave the dimers (75a) and (76a) respectively. The ester of the 35

35

37 38

3D

B. Voigt and G. Adam, Tetrahedron, 1976, 32, 1581. L. Kutschabsky, G. Reck, B. Voigt, and G. Adam, Tetrahedron, 1976,32, 2021. H. Shinozaki, S. Arai, and M. Tada, Bull. Chem. SOC.Japan, 1976, 49, 821. F. Nakanishi, Kobunshi, 1977,26, 188 (Chem. Abs., 1977,86, 121 803). J. Bolt, F. H. Quina, and D. G . Whitten, Terrahedron Letters, 1976, 2595.

Enone Cycloadditions and Rearrangements

(75) a ; K'

=

b ; R'

=

289

octadecyl, R2 = H octadecyl, R = Cl

p-chlorocinnamic acid gave only two crystalline types which gave (75b) and (76b) upon photo-dimerization. Solution-phase dimerization also occurs with cinnamic acid derivatives. Thus the cinnamates (77) undergo photochemical dimerization in solution to afford mainly the dimers (cis-78) and (trans-79), the

(78) Ar

=

p-NO,C,H,

relative proportions of which are dependent upon solvent and concentration. Isomerization of the cinnamate occurs in competition with the dimerization. A quantitative study of the reaction indicates that the trans-dimer (79) arises from the m r * triplet state while both the singlet and the triplet state contribute to the formation of the cis-dimer (78). High solvent polarity also influences the amount of cis-dimer while low polarity increases the amount of the transdimer.40 The reductive dimerization of chalcone has been reported.41

Irradiation of the enol ester (80) in the crystalline state affords the cyclobutane dimer (81).42 When the enol ester is irradiated in solution, a mixture of the transand cis-isomers results. A dimeric product was obtained when (82) was irradiated in chloroform solution. 40

41 43

T. Ishigami, T. Murata, and T. Endo, Bull. Chem. SOC.Japan, 1976, 49, 3578. J. K. Sugden, Synth. Comm., 1976, 6,93 (Chem. Abs., 1976,84, 164 341). E. Wachsen, R. Matusch, D. Krampitz, and K. Hartke, Annalen, 1976, 2137.

290

Photochemistry

The triplet state of the pentenolide (83) produced by irradiation at 254nm affords the three anti-dimers (84a), (84b), (85).43 The pentenolide also undergoes photochemical addition to cyclopentene, affording the two products (86).43 The 0

9 Me

(83)

(84) a; R1 = H, R2 = Me b; R1 = Me, R2 = H

(86) R1 = Me, R2 = H

RL= H, R2 = Me influence of ring size upon the photochemistry of the esters (87) has been investigated. Irradiation of (87a) in pentane through a Vycor filter led to the production of two dimers, the major of which was identified as (88).44 Irradiation of (87b) under identical conditions gave no dimers, although the compound was photochemically reactive and was converted into the deconjugated compound (89) in

70% yield. Reduction and hydroxylation of the ethylenic bond also occurred but only to a small extent. It is clear that the deconjugation reaction arises by a Norrish Type I1 process which, owing to the enhanced flexibility of the eightmembered ring, takes place in preference to the cis-trans isomerization about the double bond which occurs in the cycloheptene case. 2 Enone Rearrangements Direct irradiation (350 nm) of the chalcones (90) in methanol solution afforded the cis-isomers (91).46 The Z-isomer of the enone (92a) can be photochemicaIly converted into the E-isomer. This compound (92b) then undergoes thermal dimerization to a cyclobutane d e r i v a t i ~ e . The ~ ~ direct irradiation of the enol esters (93) results in their isomerization into (94) by Z-E isomerization around p3 44

46 46

K. Ohga and T. Matsuo, Bull. Chem. SOC.Japan, 1976,49, 1590. T. W. Gibson, S. Majeti, and B. L. Barnet, Tetrahedron Letters, 1976, 4801. D. Ferriera and D. G. ROUX, J.C.S. Perkin I, 1977, 134. Y. Maki and M. Sako, Chem. and Pharm. Bull. (Japan), 1976, 24, 2250 (Chem. Abs., 1977, 86, 105 433).

Enone Cycloadditions and Rearrangements M

e

O

Me0

u

O

R

2

291 MeO M %/

ORL

\

/ \

(90) R1 = CH,OMc, R2 = Me R' = Me, R' = CH,OMe

OR2 (91)

Me

a : & R L

R2

( G O Ph C 0 2 M e

OC0,Me (92) a; R1 = CO,Me, R2 = H (93) i z = 1, 2, 3, or 4 b; RL = H, R2 = C0,Me

(94)

fhe enol double bond.47 A study of the photochemically induced isomerization of pro t onated trans-crotonaldehyde, trans-crotonic acid, trans-t iglaldehyde, trans-pent-3-en-2-one, and trans-3-methylpent-3-en-2-one at low temperatures has been r e p ~ r t e d . ~ ~ trans-cis Isomerization results when the enone (95a) is irradiated (313 nm) in benzene solution. There is no evidence for hydrogen-abstraction reactions with

w

R2

(95) a; R1 = H, R2 = Me b ; R 1 = Me, R2 = H

(97)

the However, irradiation of the ynone (96) under the same conditions afforded the cyclobutanol (97) and the fragmentation product (98), both of which arise by way of a Norrish Type I1 process. The triplet energy of the ynone (96) was estimated to be 73.1 kcal mol-1 and the excited state was shown to be nn* in type. There is therefore a considerable similarity between the ynone and valerophenone which has a similar triplet energy and undergoes similar fragmentation and cyclobutanol formation.40 Hydrogen abstraction is fairly

dB

K. Hartke and E. Wachsen, Annalen, 1976,730. R. F. Childs, E. F. Lund, A. G. Marshall, W. J. Morrisey, and C. V. Rogerson, J . Amer. Chem. SOC.,1976,98, 5924. P. S. Engel, M. E. Schroeder, and M. A. Schexnayder, J. Amer. Chem. SOC.,1976,98, 2683.

Photochemistry

292

common in enones. Earlier in this Chapter, enone (87b) has been reported to undergo deconjugation to (89) by a Norrish Type I1 process.44 Other deconjunction reactions, in competition with trans-cis isomerization, occur with the enones (99).50 Photochemical enolization also falls into this reaction type and some further examples have been reported during the year. A full account of the generation of the enols (100) from the photolysis of l-acetylcyclo-octene (101) R'

(100) a; R1 = Me, R' = OH b ; R1 = OH, R2 = Me

0

(101)

62 The multiplicity of the excited state is unknown at the has been present time. However, it is thought that the excitation affords a state which undergoes cis-trans isomerization to afford the trans-isomer (102). I n this form, secondary excitation of the nn* band permits hydrogen abstraction (a Norrish Type I1 process), affording a biradical which yields the enols. Surprisingly cis-trans isomerization was not reported for (87b),44possibly owing to analytical difficulties. Several groups of workers have reported from time to time the apparent photochemical stability of some unsaturated ketones such as mesityl A reinvestigation of several of these ketones (103)-(105) has shown that

1

they do undergo photochemical reaction and form enols (from the singlet state) when irradiated in This reaction was demonstrated by deuteriation studies and also by low-temperature i.r. examination which showed the presence of an enol group. A 1,Shydrogen transfer from carbon to oxygen in the singlet excited state of 2-propionyl-6-methyl-2,3-dihydro-4H-pyran (1 06) affords stereoselective cyclization to the dehydro-brevicomin (1 07)? Hasegawa et aZ.66have studied the photochemical activity of the N,N-dialkyl unsaturated amides (108) in benzene. The photochemical reactivity is dependent 6o

61 62 b3

b4 65

.s6

A. Deflandre, A. Lheureux, A. Rioual, and J. Lemaire, Cunud. J. Chem., 1976, 54, 2127. R. Noyori, H. Inoue, and M. Kato, J. Amer. Chem. SOC.,1970, 92, 6699. R. Noyori, H. Inoue, and M. Kato, Bull. Chem. SOC. Japan, 1976, 49, 3673. N. C. Yang and M. 5. Jorgensen, Tetrahedron Letters, 1964, 1203; P. J. Wagner and G . S. Hammond, J . Amer. Chem. SOC.,1966, 88, 1245; L. F. Friedrich and G. B. Schuster, ibid., 1969, 91, 7204; D. R. Morton and N. J. Turro, ibid., 1973, 95, 3947. M. Tada and K. Miura, Bull. Chem. SOC.Japan, 1976,49, 713. P. Chaquin, J. P. Morizur, and J . Kossanyi, J. Amer. Chem. Soc., 1977,99,903. T. Hasegawa, M. Watabe, H. Aoyama, and Y. Omote, Tetrahedron, 1977, 33,485.

Enone Cycloadditions and Rearrangements

293

(106)

(107)

to a large extent upon the substitution at the nitrogen atom. Thus the amides (108a-f) are photochemically inert, whereas amides (1OSg-i) afford the azetidones (109). These products are formed by the abstraction of a hydrogen from one of the N-alkyl groups by the F-carbon of the enone moiety. Hydrogen abstraction R2*N

R32

R" (108) a; R1 = b; R1 = c ; R1 = d ; R1 = e; R1 = f f R1 =

R2 = H, R3 = Et Me, R2 = H, R3 = Et H, R2 = Me, R3 = Et Ph, R2 = H, R3 = Et R2 = H, R3 = Me H, R2 = R3 = Me

g; R' = h; R1 = i ; R1 = j; R1 = k; R1 =

R2 = H, RJ = PhCH2 Me, R2 = H, R3 = CH2Ph H, R2 = Me, R3 = CH2Ph R2 = H, R3 = Pri H, R2 = Me, R3 = Pri

RICH2-*- --H 0B C H 2 P h (1 10)

(109)g; 70% h; 30% i ; 84%

by a carbon atom in a cyclic enone system is not The resultant biradical cyclizes to afford the product. Although alternative routes were considered, the weight of evidence favours the route shown in Scheme 3. Another reaction path manifests itself with the amide (108i) which also afforded the dealkylated product (1 10) by an unknown mechanism. However, dealkylation is also encountered with the amides (108j, k) and this is rationalized in Scheme 4.

Scheme 4 67

A. B. Smith and W. C. Agosta, J. Amer. Chem. SOC.,1974,96,3289.

294

Photochemistry

The importance of ground-state conformation upon the outcome of the reaction was demonstrated using the amides (1111, m). The amide (1111) gave only the azetidone (1 121) in low yield (lo%), while the amide (1 1lm) gave the azetidone (1 12m) (7.5%) and the dealkylation product (1 13) (19.5%). The authors s6 reason that the enhanced yield of products in the case of (112m) is the result of the enhanced population of the conformer (1 14), i.e. as shown in (1 1lm). Hydrogen abstraction by the a-carbon of an enone is encountered in the irradiation (Pyrex filter) of the enone (115) in benzene solution which affords (116) as the major

product by way of biradical (117).68 A minor product of unknown composition was also isolated. Both products are formed from the triplet state of the enone, a fact which is substantiated by the quenching of the formation of both when piperylene is used as a quencher. The enone (115) undergoes photochemical reduction to (118) when it is irradiated in propan-2-01, Cyclization to (1 19) is encountered in the irradiation of the chalcone (120) in EtOH-H,PO,. An ionic mechanism is likely.6s

The study of the effect of metal salts (in this publication UOzClawas used) on the photochemical reactions of organic compounds has continued.60 The compounds (121), (122), and (123) were all irradiated in the presence of the salt in methanol solution and were converted into the furan derivatives (124), (125), and (126) respectively. 68

69

6o

B. Gioia, A. Marchesini, G . D. Andretti, G. Bocelli, and P. Sgarabotto, J.C.S. Perkin I, 1977,410. S. Matsuura, M. Iinuma, T. Ito, M. Kuroiwa, and N. Higashi, Yukugaku Zusshi, 1976, 96, 393 (Chem. Abs., 1976,85, 5455). T. Sato, 0. Ito, and M. Miyahara, Chem. Letters, 1976, 295 (Chem. A h . , 1976, 85, 5435).

Enone Cycloadditions and Rearrangements

295

0

R

d

(123) R

=

R H or Mc

The methylene ketones (127) undergo photochemical (Pyrex filter) conversion into the intermediate allenylketens (128) as a result of Norrish Type I cleavage.61 When the reaction of (127a) is carried out in benzene solution, intramolecular cyclization affords the allenyl-Iactone (129); but when methanol is used as the solvent the isomeric lactone (130) is formed by a path not involving isomerization

-y-

&.oOH

Pp ;h *11‘ P

(127) a; X b;X

x

Ph+.

Ph

= =

H Br Ph Ph

L& Ph

Ph

P1Ac

Po Ph Ph

of (129). The butenolide (130) is thus thought to be formed from the ion (131) which, in methanol, is protonated upon the allenic carbon atom. The presence of a bromine atom in the cyclobutenone (127b) introduces additional problems through the tendency for photodissociation of the C-Br bond. Thus irradiation of (127b) in benzene affords (132) and (133) probably as shown in Scheme 5 , whereas in methanol only (132) is formed. Toda et aLs2have also reported the irradiation of the related methylenecyclobutene (134) in the presence of benzophenone or oxygen. This treatment yielded the allenyl ketone (135) in 31% yield. The route to the formation of (135) in the presence of benzophenone (or 0,)is thought to involve photochemical addition of the benzophenone (or 0,) to the allenyl keten (136) which is formed when the cyclobutenone ring F. Toda and E. Todo, Bull. Chem, SOC.Japan, 1916,49,2503. F. Toda, Y. Todo, and E. Todo, Bull. Chem. SOC.Japan, 1976, 49,2645.

Photochemistry

296

(133)

Scheme 5

opens photochemically. The intermediate (1 37) formed in the addition reaction undergoes loss of diphenylketen (or CO, in the case of oxygen addition) to afford the isolated product. A keten intermediate (138) is also involved in the photochemical rearrangement of (139) into (140). The route to product is Ph

Ph,CH (134)

OMe Ph,CH

Ph PI1

OMe

Ph2CH

(136)

(135)

OMe

Ph,CH

OMc

(137)

thought to involve the path shown in Scheme 6, but the presence of a keten could not be shown either chemically or spectr~scopically.~~ It is proposed that the keten path could still operate, but that the keten is produced with sufficient vibrational energy to transform rapidly into the final product. a-Fission is encountered in irradiation (254 nm) of the pyrazolone (141) in ethanol to give the

Scheme 6

azo-compound (142).64 The reaction is thought to involve fission of the pyrazolone to yield a biradical (143) which after ring closure to (144) breaks down to yield a benzenediazonium salt. This adds to starting material to yield the observed product. 63

B4

A. Padwa, A. Ku, and E. Sato, Tetrahedron Letters, 1976, 2409. K.Tsutsumi, I. Takagishi, H. Shizuka, and K. Matsui, J.C.S. Chem. Comm., 1976, 685.

Enone Cycloadditions and Rearrangements

297

N=NPh

'fiR

(141)

R

=

OfiR

O q , I