摘要:

Chemical Society Reviews Vol 6 No 3 1977 Page JOHN JEYES LECTURE Chemicals which Control Plant Growth 261By R. L.Wain Enaminones By J. V. Greenhill 277 Ion-Molecule Reactions in the Evolution of Simple Organic Molecules in Interstellar Clouds and Planetary Atmospheres By W. T.Huntress, Jr. 295 KELVIN LECTURE Across the Living Barrier By David E. Fenton 325 Metal-ion-promoted Reactions of O~~O-SulphUr Compounds By D. P. N. Satchel1 345 Corrigendum 372 The Chemical Society London Chemical Society Reviews Chemical Society Reviews appears quarterly and comprises approximately 25 articles (ca. 500 pp) per annum. It is intended that each review article shall be of interest to chemists in general, and not merely to those with a specialist interest in the subject under review. The articles range over the whole of chemistry and its interfaces with other disciplines.Although the majority of articles are intended to be specially commissioned, the Society is always prepared to consider offers of articles for publication. In such cases a short synopsis, rather than the completed article, should be sub-mitted to The Editor, Books and Reviews Section, The Chemical Society, Burlington House, Piccadilly, London, W1V OBN. Members of the Chemical Society may subscribe to Chemical Society Reviews at €5.00 per annum; they should place their orders on their Annual Subscription renewal forms in the usual way. Non-members may order Chemical Society Reviews for €14.00 ($30) per annum (remittance with order) from: The Publications Sales Officer, The Chemical Society, Blackhorse Road, Letchworth, Herts., SG6 lHN, England. 0 Copyright reserved by The Chemical Society 1977 Published by The Chemical Society, Burlington House, London, W1V OBN Printed in England by Eyre & Spottiswoode Ltd, Thanet Press, Margatc

ISSN:0306-0012    

DOI:10.1039/CS97706FP007

出版商:RSC    

年代:1977

数据来源: RSC

摘要:

JOHN JEYES LECTURE* Chemicals which Control Plant Growth By R. L. Wain UNIVERSITY OF LONDON, WYE COLLEGE, ASHFORD, KENT, TN25 5AH 1 Introduction It is indeed an honour for me to have been chosen to receive the first John Jeyes Medal and Lecture Award. In doing so I wish to pay tribute to an outstanding man. His lively research interests and business acumen led him to important achievements whose impact on society are still very evident today. John Jeyes, born in 1817, was the second son of Northampton parents. He developed interests in botany and horticulture as a boy and these interests stayed with him throughout his life. His name has been associated with the breeding of the ‘Jeyes Conquerer’ pea-though some would have it that the seed actually came from an Egyptian tomb !At any rate, this quite famous variety was developed by John Jeyes.When he came to London in 1859 he found that conditions amongst the poor were appalling: hygiene was non-existent and disease was prevalent everywhere. Aware of Lister’s classical work on phenol, Jeyes began to study coal tar and the tar acids in particular. In 1877, one hundred years ago, he took out his fist British patent on disinfectants. Not long afterwards he set up a company, ‘Jeyes Sanitary Compounds Co. Ltd.’, to handle what was to become the famous Jeyes Fluid. So John Jeyes was a pioneer in the development of chemicals with biological activity and he was also interested in plants. Perhaps this is the reason why the Council of the Chemical Society decided that this, the first John Jeyes Lecture, should be concerned with chemistry and with plants.Now plants, directly or indirectly, provide all the food of man and animals so a knowledge of how they grow is of first-rate importance. Let us consider what happens when a seed germinates. It sends down a root to collect water and nutrients from the soil and produces a shoot which when it appears above the ground can take in carbon dioxide from the air and absorb energy from the sun. Thus provided with energy and an adequate supply of food and water, it has the wherewithal to grow. By a process of cell division new cells are produced in the tips of shoot and root and when these tiny cells enlarge- mainly by taking in water-the tissues increase in size and growth results.Then there must also be something in seeds which will ensure that the plants which grow from them will be true to type, with the same form, size, and structure as *Based on the John Jeyes lecture first given at the Chemical Society Annual Congress in London on 31st March 1977. Chemicals which Control Plant Growth all other plants of the same kind. These controlling influences in the seed we call genetic. They are dependent on hereditary factors which, carried in the nucleus of cells, come down to the plant from its parents. Quite a lot is known about plant growth in relation to its nutrition and more and more is being discovered about the chemistry of its genetic make-up, but there is another aspect of plant growth which is no less important.Why should all the various growth processes occur just when and how they do? Why should plants always bend towards the light? Why should roots appear at the base of a cutting and why should a ripe apple fall from the tree? We now know that such processes are controlled by a complex of growth hormones and inhibitors, all of which are extremely potent chemicals produced by the plant itself. No account of this fascinating chapter of botanical research would be complete without reference to Charles Darwin who in 1880, presenting his findings in a book ‘The Power of Movements in Plants’, came to the conclusion that some ‘influence’ was operating from the tip of a shoot which made the plant respond to 1ight.l This ‘influence’ of Darwin’s is now known to be exerted by growth hor- mones which, synthesized in the growing tip of a shoot, diffuse downwards and promote growth by causing the cells to enlarge.Two types of cell elongation hormone are now recognized-the auxins and the gibberellins. The other funda- mental process which is concerned in plant growth is cell division by which new cells are produced. Naturally occurring compounds which can influence this process are called cytokinins. In addition, the gaseous hydrocarbon ethylene and two endogenous hormone inhibitors abscisic acid and xanthoxin operate in the complex of compounds which control the growth and development of plants. Hormones and hormone inhibitors are also important in determining the plant’s response to environmental factors such as light, temperature, and various stress conditions such as drought and waterlogging.These then, are the chemicals which control plant growth about which I shall be talking in this address-with a ‘fluidity’ which I hope, would have met with the approval of John Jeyes himself! 2 The Auxins Went in 1926 demonstrated that oat seedlings contain a diffusable substance which would promote their growth; this was the first clear indication that a growth hormone occurs in plants.2 The discovery eight years later by K0g13 that indole-3-acetic acid (IAA) (1) is capable of promoting the elongation of plant cells mcH2c02HH (1) Indole-3-acetic acid C. Darwin, ‘The Power of Movement in Plants’, John Murray, 1880.a F. W. Went, Proc. kon. ned. akad. Wetenschap., 1926, 30, 10. F. KogI, A. J. Haagen-Smit, and H. Erxleben, Z. physiol. Chern., 1934, 228, 90. Wain focused attention on this compound which is now recognized to be the most important of the class of growth hormones known as the auxins. It is worth recording that IAA had been found in human urine over 50 years before4 though it was not synthesized5 until 1904. IAA is widely distributed in plants where it appears to be synthesized from tryptophan by a series of enzyme-controlled reactions.6 The amounts of endogenous IAA are extremely small and are controlled at the physiological levels required for normal growth by the capacity of the plant to biosynthesize the compound, to destroy it by the action of an IAA oxidase,7 and to conjugate the molecule with such compounds as amino-acids,s thereby removing its activity.The discovery of IAA as an auxin led chemists to examine the growth- regulating activity of compounds with similar structures. As a result, a wide range of active compounds has become available. These include arylacetic acids [e.g. a-naphthylacetic (2) and 2,3,6-trichlorophenylaceticacids], aryloxy-acids [e.g. 2,4-dichlorophenoxyacetic acid (2,4-D) (3) and 2-methyl-4-chlorophen- oxyacetic acid (MCPA) (4)], and certain benzoic acids [e.g.2,3,6-trichlorobenzoic acid (5)J Certain 2,6-disubstituted phenols [e.g. (6)] are also a~tive.~ CH,CO,H OCH,CO,H Cl Cl (2) a-Naphthylacetic acid (3) 2,4-D (4) MCPA CO,H OH Unlike the natural auxin IAA, these synthetic growth regulators are not readily inactivated within plant tissues and some of them are therefore much more physiologically active than IAA itself.Commercial uses for these synthetic growth substances were soon found and include promoting the rooting of cuttings, J. P. Neaki and N. Sieber, J. prakt. Chem., 1882, 26, 333. A. Ellinger, Ber. deut. Chem. Gesellschaft, 1904, 37, 1801. F. Wightman, Canad. J. Bot., 1962,40, 689. 'Y. W. Tang and J. Bonner, J. Arch. Biochem., 1947, 13, 11. * W. Andrea and N. E. Good, Plant Physiol., 1955, 30, 380. R. L. Wain and H. F. Taylor, Narure, 1965, 207, 167. Chemicals which Control Plant Growth (Plate l), reducing fruit fall, setting fruit in the absence of pollination (Plate 2), and selective weed control (Plates 1-7 are between pp.274 and 275). Much research has been carried out on the relationships between chemical structure and capacity to induce growth effects in plants. In the phenoxyacetic acids chlorine substituted in the ring affects activity. Thus, the low activity of phenoxyacetic acid increases in the order 2 >3 >4 in the monochloro- derivatives. In the dichlorophenoxyacetic acids, the 2,3-, 2,4-, and 3,4-isomers are all active but the 2,6-and 3,5-derivatives are not.10 Studies on the effect of introducing alkyl groupings into the side-chain of aryloxyacetic acids (7)-(10) have shown that the molecule must have at least one hydrogen atom on the carbon adjacent to the carboxy-group for activity to be shown.llJ2 ArO -C -CO, t --&--I I I IH H Me (7)Acetic (8) Propionic (9) Butyric (10) Isobutyric derivative derivative derivative derivative (active) (active) (active) (inactive) When the molecule possesses an asymmetric carbon atom, as with the above propionic and butyric derivatives, high activity is shown by one enantiomorph and the other is inactive.13 It is of interest here to note that steric considerations also apply with certain cinnamic acids where the cis-isomer is extremely active as a growth substance and the trans-isomer is inactive.14 The literature on the mode of action of growth substances and on chemical structure in relation to plant growth-regulating activity is extensive and has been reviewed.15~16 A.Effects of Auxins on Enzyme Activities.-It would seem logical to expect that physiologically active compounds which are capable of controlling plant growth and development must exert influences on enzyme systems within the plant. The effect of auxins on enzyme activity in inulin-containing plant tubers has recently been investigated in our 1ab0ratory.l~ Discs cut from the roots of chicory and Jerusalem artichoke were held in the surface of auxin solutions for varying lo R. L. Wain and F. Wightman, Ann. Appl. Biol., 1963, 40,244. l1 D. J. Osborne and R. L. Wain, Science, 1951, 114, 92. l9 C. H. Fawcett, R. L. Wain, and F. Wightman, Ann. Appl. Biol.,1955, 43, 342. l8 M. S. Smith and R. L.Wain, Proc. Roy. SOC.,1951, B139, 118. l4 E. N. Ugochukwu and R. L. Wain, Ann. Appl. Biol., 1968, 61, 121. R. L. Wain and C. H. Fawcett, ‘Plant Physiology’, Vol. VA, Academic Press, New York, 1969. l6 J. L. Garraway and R. L. Wain, in ‘Drug Design’, Vol. VII, Academic Press, New York, 1976. l7 R. L. Wain, P. P. Rutherford, E. W. Watson, and C. M. Griffiths, Nature, 1965, 203, 504. Wain periods at 25 "C.Considerable increases in the size and weight of the discs occurred (see Table). The effect was shown by a wide range of substances which Table Water uptake and invertase activities of'chicory and Jerusalem artichoke discs after 3 days' treatment at 25 "C Water uptake, Units* of Tissue Treatment increase as of' invertase activity initial wt.x 106 Chicory Fresh tissue -46.3 Water 35.2 52.5 10-5~2,4-~ 280 3392 Jerusalem artichoke Fresh tissue -0 Water 24 0 10-5~2,4-~ 96 324 * 1 Unit represents 2 pmol hexose liberated per min at 25 "Cper mg initial dry weight. were active in standard tests for plant growth-regulating activity. Furthermore, structural requirements which have been established for auxin activity were found to operate in the water uptake response.l* In further studies it was shown that treatment of discs of Jerusalem artichoke with 2,4-D led to a breakdown of fructosans and the production of soluble reducing sugars within the tissues and also to an increase in the rate of respiration. It would therefore seem that the resulting changes of osmotic pressure and the provision of energy arising from the liberated sugars together promote the observed water uptake.Since the changes in carbohydrate status within the tissues are initiated by the growth-substance treatment, studies have been made on the enzyme systems which might be inv01ved.l~ Hydrolytic enzymes were extracted from discs of chicory and Jerusalem artichoke which had been treated with 2,4-D and also from the controls, and the extracts were fractionated on cellulose columns. The invertase activity is shown in the Table. The increases in invertase activity observed in both tissues following treatment with 2,4-D could not be associated with an increase in enzyme protein, but this finding does not rule out the possibility that a small proportion of new enzyme with very high specific activity is produced.It is also possible that the growth- substance treatment caused the enzymes to become released from a bound form on the cell wall; alternatively, the activity of enzymes already present could have become greatly increased by the action of the growth substance. B. Selective Weed Control.-Synthetic auxins such as 2,4-D and MCPA, when applied at low rates per acre to certain plant species, produce such drastic growth effects that the plants outgrow themselves and cannot survive; other species, P. P. Rutherford, C. M. Griffiths, and R. L. Wain, Ann. Appl. Biol., 1966, 58, 467. A. E. Flood, P. P. Rutherford, and E. W. Watson, Nature, 1967, 212, 1049. Chemicals which Control Plant Growth however, including cereals and grasses, are able to withstand these low doses and remain unharmed.Used in this way these chemicals are applied annually to millions of acres of crops to remove weed competition and they have made a tremendous contribution to world food production. Even so, their range of use is limited; for example, they are of no value for controlling weeds in clover, lucerne, and other legumes because these crops are just as susceptible to them as the weeds. Some years ago, our basic studies on the oxidation of w-substituted fatty acids within plant tissues led to a new type of herbicidal selectivity. The principle involved is logical; the susceptible plant when treated with a compound which is inactiveper se converts it enzymically into a herbicide and this ‘lethal synthesis’ operating within the cells leads to destruction of the plant.An example of how selectivity can be achieved by this approach is provided by certain y-phenoxy- butyric and ephenoxycaproic acids.20p21 When applied to susceptible species these compounds become converted into their coenzyme A derivatives which in presence of an appropriate b-oxidase enzyme system undergo /%oxidation of the side-chain (Scheme 1). In this way the highly herbicidal acetic acid derivative is produced, a molecule of acetylcoenzyme A being lost at each stage. OCH*CH 2CH ,C H,C H2C0,H OC H ,CH ,C H ,CO*H OCH,COtH 8-oxidation of 8-oxidationOMe-QMe -QMe CoA derivative of CoA derivative CI CI CI ~-(2-Methyl-4-y(2-Methyl-4-2-Methyl% ch1orophenoxy)-ch1orophenoxy)-chlorophenoxy-caproic acid butyric acid (MCPB) acetic acid (MCPA) (herbicidal)Scheme 1 Certain legume crops possess a fl-oxidase enzyme system which is not adapted to operate with these synthetic substrates.Degradation therefore does not occur and the crop plants remain unharmed although many weed species are destroyed. Some 20 homologous series of phenoxy-acids were used in the physiological and biochemical investigations which established that the herbicidal selectivity of these butyric and caproic acids depends upon differential lethal ~ynthesis.~~-24 The two compounds chosen for commercial development were y-(2,4-dichloro- phenoxy)butyric acid (2,4-DB) and y-(2-methyl-4-chlorophenoxy)butyricacid soR.L.Wain, Ann. Appl. Biol., 1955,42, 151. R. L. Wain, J. Agric. Food Chem., 1955,3, 128. 2a R. L. Wain and F. Wightman, Proc. Roy. SOC.,1954, B142, 525. 23 C.H. Fawcett, R. M. Pascal, M. B. Pybus, H. F. Taylor, R. L. Wain, and F. Wightman, Proc. Roy. SOC.,1959, B150, 95. 21 C. H. Fawcett, R. L. Wain, and F. Wightman, Proc. Roy. Soc., 1960, B152, 231. Wain (MCPB). Both of them are widely used in agriculture for selective weed control in certain legume crops and in cereals undersown with cl0ver.~59~~ Another example of lethal synthesis which has formed part of our research2’ is the conversion of aryl- and aryloxy-ethylamines into the corresponding alde- hydes. This only occurs when the appropriate amine oxidase systems are present within the treated plant.The aldehyde so formed then becomes oxidized to the corresponding acid (Scheme 2) which, as with 2,4-D, may be strongly herbicidal. OCH 2CH ,NHp OCH,CHO OCHBCOpH CI c1 c1 24 2,4-Dichloro- 2,4-Dichloro-2,4-D (herbicidal) phenoxy)-ethylamine phenoxy-acetaldehyde(2,4-D ethylamine) Scheme 2 That the amine oxidation is dependent on the presence of an amine oxidase has been demonstrated by isolating the enzyme from pea seedlings and effecting the conversion in vitro. Furthermore, the herbicidal effects produced by treating plants with 2,4-D ethylamine do not appear if the plants are pretreated with the amine oxidase inhibitor 2-hydroxyethylhydrazine.27 With some phenoxy-acid herbicides, selectivity can arise not from differential lethal synthesis but because they are more readily degraded to inactive products in one species than in another.An example of this is mecoprop, 2-(2-methyl-4- ch1orophenoxy)propionic acid, fist prepared in the reviewer’s laboratory in 1953.28 The racemic form of this substance was shown to be as active as 2-methyl- 4chlorophenoxyacetic acid (MCPA) in four different tests for plant growth- regulating activity. and Leafe30 demonstrated that mecoprop also controls two weeds, chickweed (Stellaria media) and cleavers (Gallium aparine), which are not well controlled by MCPA. Evidence obtained from radioactive tracer studies indicates that the resistance of cleavers to MCPA is due to a detoxification within the tissues whereby both carbon atoms of the side-chain are lost.31 The steric effect of the a-methyl group in mecoprop, however, protects against this degrad- ation, enabling the compound to accumulate in the tissues and produce its R.L. Wain, Agriculture, 1957, 63, 575. 26 ‘Weed Control Handbook’, ed. J. Fryer and R. Makepeace, Blackwell Publications, Oxford, 1970. 27 R. J. Nash, T. A. Smith, and R. L. Wain, Ann. Appl. Biol., 1968, 61, 481. 2E C. H. Fawcett, D. J. Osborne, R. L. Wain, and R. D. Walker, Ann. Appl. Biol., 1953, 40, 232. G. B. Lush, Proceedings of the IIIrd British Weed Control Conference, Blackpool, 1956, British Weed Control Council, London, p. 625. 30 E. L. Leafe, ref. 29, p. 633. 31 E. L. Leafe, Nature, 1962, 193,485.Chemicals which Control Plant Growth physiological effects. MCPA is also subject to other types of chemical modifica- tion and indeed most herbicides are metabolized at varying rates in plants. De- carboxylation, hydrolysis, dealkylation, hydroxylation of ring, conjugation, and ring cleavage are all known to occur. When breakdown occurs to a greater extent in one plant species than another, we have a basis for herbicidal selectivity. 3 The Gibberellins The second group of endogenous plant growth hormones are the gibberellins. This area of research began with the discovery by Kurasawa in 1926 that the cell- free sterile filtrate of a fungus Gibberella fujikuroi, which causes pale spindly growth of rice, produced marked growth stimulation when applied to seedlings of rice and grasses.32 Gibberellins occur in all higher plants and some 52 of them have now been characterized; they differ only slightly in structure from the well known gibberellic acid GA3 (11) which is produced commercially from fungal cultures.OH HO H Me -CH* CO,H (1 1) Gibberellic acid GA, Like auxins, gibberellins promote stem extension and fruit growth. They can also stimulate flowering in some plants and overcome dormancy of certain seeds. Gibberellins promote fruit setting and the growth of seedless grapes and improve skin quality and delay the ripening of citrus fruits. GA3 is also used in the brewing industry to stimulate the production in barley of the enzyme a-amylase, which plays a key role in the breakdown of starch during malting.Recent research has led to the discovery of a number of synthetic compounds which appear to act by inhibiting gibberellin biosynthesis. A plant so treated must therefore depend mainly on its auxin for extension growth and so it will be retarded. Not all growth retardants work in this way but by their use it is possible to obtain dwarfed plants with shortened internodes. Such plants are sturdy and healthy with deep green leaves and in some cases they have been found to have greater pest and disease resistance. Most of the many growth retardants synthesized in our laboratory are ammonium, phosphonium, or sulphonium salts;33 4-chlorobenzyltri-n-butyl-so E. Kurosawa, J. Nut. Hist. SOC.Formosa, 1926, 16, 213.33 B. E. A. Knight, H. F. Taylor, and R. L. Wain, Ann. Appl. Biol., 1969, 63, 21 I. Wain ammonium bromide, for example, is very effective for dwarfing important legume crops such as French bean (Phaseolus vulgaris) and soya bean (Glycine max), as well as certain ornamentals, when sprayed on to the young plants at 500-1500 p.p.m. (Plate 3). The dwarfed plants which result from these treatments take up less space so that more plants per acre can be grown. Whether or not this leads to higher yields per acre is now under investigation. Some of our other compounds, e.g. N-methyl-N-chloromethylpyrollidiniumbromide, are highly effective in dwarfing wheat and oats,34 as is the commercial product chlormequat (2-chloroethyltrimethylammoniumchloride). Dwarf wheat varieties can also be obtained by breeding; this has been achieved at the Plant Breeding Institute at Cambridge, and Borlaug’s now famous dwarf high-yielding wheat varieties in which the Japanese Norin 10 dwarfing gene was incorporated into Mexican wheats are playing an important part in the ‘Green Revolution’.Recent work in the reviewer’s laboratory has indicated that the dwarfing effect which operates in such genetic dwarfs may arise from the synthesis of hormone-inhibitory substances within the plant. The compounds responsible, which would appear to be ‘natural’ growth retardants, are now being intensively studied. A well known commercial synthetic growth retardant is NN-dimethylamino- succinamic acid, which, among other uses, has been shown to reduce extension growth and to promote flowering in apple trees. 4 Cytokinins As already stated, auxins and gibberellins are hormones which promote cell enlargement.The other fundamental process which determines growth is cell division by which new cells are produced. The hormones which can influence this process in plants are known as the cytokinins. The first compound possessing cytokinin activity to be discovered was 6-furfuryladenine (kinetin), isolated in 1955 from autoclaved herring sperm DNA.~~s~~ However, it was not until 1964 that the first naturally occurring cytokinin, zeatin, was reported. This was found by Letham37 in developing maize kernels and like kinetin it is a 6-substituted adenine derivative [6-(4-hydroxy-3-methylbut-2-enyl)aminopurine(12)].3*Zeatin and its riboside occur widely in plants. Cytokinins can delay senescence.A detached radish leaf, for example, spot- treated with a cytokinin, remains green in the treated area when the rest of the leaf becomes yellow.39 Furthermore, it has been shown that a 14C-labelled amino-acid applied to the non-treated area of the leaf migrates to the site of cytokinin 34 V. K. Chamberlain, K. Chamberlain, and R. L. Wain, Ann. Appl. Biol., 1976, 82, 589. a6 C. 0. Miller, F. Skoog, M. H. von Saltza, and F. M. Strong, J. Amer. Cherrr. SOC., 1955, 77, 1392. s6 C. 0. Miller, F. Skoog, F. S. Okumura, M. H. von Saltza, and F. M. Strong, J. Amer. Chem. SOC.,1955,77, 2662. 37 D.S. Letham, Life Sci.,1963, 2, 569. D. S. Letham, J. S. Shannon, and I. R. McDonald, Proc. Chem. SOC.,1964, 231. s8 A. E. Richmond and A. Lang, Science, 1957,125, 650. Chemicals which Control Plant Growrlz MeC -CH,OH II NH-CH,-CH I (12) Zeatin treatment ;40 this indicates that the chemical is promoting normal metabolism at the expense of the rest of the senescing leaf. Cytokinins therefore have potential for promoting the life of fresh vegetables such as lettuces and cabbages. They can also be used for promoting the life of cut flowers and overcoming the dormancy of certain seeds. Cytokinins appear to be synthesized in the roots of plants. Many 6-substituted adenine derivatives have been tested for cytokinin ac- ti~ity.~lA useful test is to use explants of tobacco pith tissue on sterile agar medium.The control agar is provided with all the necessary nutrients, mineral elements, and growth substances except cytokinins ; since cell division cannot take place no growth OCCUTS. When the agar contains cytokinin, however, the cells do divide and then enlarge, thereby leading to the growth of undifferentiated callus tissue. In recent research carried out in the reviewer’s laboratory, cytokinin activity has been found in a range of 6-substituted oxypurines. When examined in the tobacco pith test some of them [e.g. 6-benzyloxypurine (13)] not only promote callus growth but this then undergoes morphological differentiation leading to the production of intact tobacco plants42 (Plate 4).OCH,PhI (1 3) 6-Benzyloxypurine As we have seen, natural cytokinins operate with auxins and gibberellins in the hormonal complex which controls growth and development. In an attempt to determine whether, in the growth of the wheat coleoptile, these operate together or in sequence, a study was made of the changes in the protein (and enzyme) pattern at different stages of gro~th.~3 Protein fractions were taken from coleop- tiles at all growth stages and antiserum was raised from the combined fractions in 40 K. Mothes, ‘RCgulateurs Naturels de la Croissance VCgdtale’, Paris, Edition du C.N.R.S., 1964, p. 131. *l K. Rothwell and S. T. C. Wright, Proc. Roy. SOC.,1967,B167, 202. 42 E. J. Wilcox and R. L. Wain, Ann. Appl. Biol., 1976, 84, 403.43 S. T. C. Wright, Symp. SOC.Exp. Biol., 1963, 17, 18. Wain the rabbit. Using a technique involving the diffusion, in an agar gel, of protein extracts of coleoptiles of different ages and their antibodies, it was shown that qualitative changes in protein occurred during growth and cellular differentiation of the wheat coleoptile. It was logical therefore to expect the responses of the coleoptiles to gibberellin, kinetin, and IAA to vary at different stages of growth. This was found to be the case; there were clear indications that these substances exert their growth effects by acting in a well defined sequence.44 The first early phase of development was one of cell enlargement (approximately 0-30 h after sowing) influenced mainly by GA; the second (approximately 30-60 h after sowing) in which many of the cells were undergoing cell division was influenced mainly by kinetin.The third phase was a final period of cell enlargement (approxi- mately 60-120 h after sowing) which was promoted mainly by IAA. It has since been postulated that these three classes of growth regulator may operate in a similar sequential manner in the growth of fruit.45 Furthermore, it has been shown that tobacco pith tissue grown in a nutrient medium under sterile conditions also appears to have a sequential requirement for GA, kinetin, and IAA.46 5 Ethylene At this stage, mention should be made of another compound which can exert profound physiological effects on plants. This is the simple unsaturated, gaseous hydrocarbon ethylene.Tt is evolved by certain plants and especially by ripening fruits, and it can promote abscision of both fruit and leaves. Much work has been carried out on the mode of action of ethylene, the specific effects of which must be related to its unsaturated character and its small molecular size, since higher olefins are much less active and the corresponding paraffin, ethane, is inactive. A compound which breaks down to yield ethylene following application to plant tissues has interesting commercial uses. This compound is ethephon, 2-chloroethylphosphonic acid. One striking effect which it produces is to stimul- ate greatly the flow of rubber latex when applied to rubber trees on a band of smoothed bark below the tapping cut.It is also used for accelerating maturity and early ripening of tomatoes and citrus fruits and for inducing uniform flowering in pineapples. 6 Hormone Inhibitors The three main types of hormone so far discussed are concerned with the promotion of growth. It has long been suspected, however, that inhibitory compounds might operate in growth control processes as, for example, when, owing to unfavourable environmental conditions, growth ceases altogether. In 1965 a naturally occurring hormone inhibitor was isolated from cotton fruits and identified as 3-methyl-5-( l-hydroxy-4-oxo-2,6,6-trimethylcyclohex-2-en-1-yl)-44 S. T. C. Wright, Nature, 1961, 190, 699. 45 J. Van Overbeek, Proc. Campbell Soup Co. PI. Sci.Symp., 1962, 38.J. P. Nitsch, ‘Biochemistry and Physiology of Plant Growth Substances’, ed. F. Wightmanand G. Setterfield, Runge Press, Ottawa, 1968, p. 563. 271 Chemicals which Control Plant Growth cis,trans-penta-2,4-dienoicacid (14).47Shortly afterwards the same compound was found in sycamore leaves and in lupin pods. The synthesis of this inhibitor, which has been given the name abscisic acid, was achieved in Cornforth’s lab~ratory.~~ (14) Abscisic acid Abscisic acid has since been found in a wide range of plant species, and physio- logical studies have revealed that it can inhibit the activity of auxins, gibberellins, and cytokinins. Not only this, but we have also shown that abscisic acid operates in defending plants against the effects of physiological stress.49950 For example, when water is withheld from a tomato plant the wilting plant responds by producing up to fifty times the normal level of abscisic acid in its leaves (Plate 5).The effect of this is two-fold. Firstly, the resulting inhibition of growth hormone activity stops the plant from growing and energy is thereby conserved; secondly, a closure of the leaf stomata is induced51 and water loss by transpiration is cut down. By two mechanisms therefore the build up of endogenous abscisic acid provides the plant with a better chance of survival during the drought period. In our laboratory we have compared the capacity of a Mexican drought- resistant maize variety (‘Latente’) with two others which are not drought resis- tant.52 It was found that when subjected to standard water stress conditions ‘Latente’ produced much more abscisic acid than the other two varieties, again indicating the important role played by this inhibitor in plants subjected to drought.A similar rapid increase in abscisic acid levels has also been shown to occur when plant roots are waterlogged (Figure, opposite p. 275).53 Abscisic acid therefore provides a defence mechanism against physiological stress. In addition to the free acid, abscisyl p-D-glucopyranoside has been isolated from plant tissues.54 A feature of the abscisic acid molecule is its resemblance to Vitamin A, a substance which is produced in the animal liver from certain carotenoid pigments supplied in the food.Although carotenoids are present with chlorophyll in all 67 K. Okhuma, F. T. Addicott, 0. E. Smith, and W. E. Thiessen, Tetrahedron Letters, 1965, 29, 2529. 48 J. W. Cornforth, B. V. Millborrow, and G. Ryback, Nature, 1965, 206, 715. ‘*S. T. C. Wright, Planta, 1969, 86, 10. so S. T. C. Wright and R. W. P. Hiron, Nature, 1969, 224, 719. I1R. J. Jones and T. A. Mansfield, J. Exp. Bot., 1970, 21, 714. sa A. LarquB-Saavedra and R. L. Wain, Nature, 1974, 251, 716. Ba R. W. P. Hiron and S. T. C. Wright, J. Exp. Bot., 1973, 24, 769. K. Koshimizu, M. Iniu, M. Fukui, and T. Mitsui, Agric. and Biol. Chem.(Japan), 1968,32, 789. Wain green leaves, and therefore occur abundantly throughout the plant kingdom, their physiological function within the plant has never been properly elucidated. However, the similarity between Vitamin A and abscisic acid led to speculations in our laboratory55 on whether carotenoids might serve as precursors of abscisic acid within the plant.Since plants in the dark grow taller than those in the light, it follows that light inhibits growth. Light therefore might be a factor which could promote the conversion of carotenoid into the inhibitor abscisic acid. This reasoning agreed with one of our earlier findings56 that a growth inhibitor is produced in dwarf pea plants when they are exposed to white light. To test the above hypothesis, equal volumes of an acetone solution of nettle leaf carotenoids were applied uniformly to two fiIter papers and the solvent was evaporated.One of these papers was kept in darkness and the other was exposed to the light for 1 hour; they were then placed in Petri dishes, moistened with water, and each was sown with cress seeds. The dishes were then placed in an incubator. After 60hours in the dark it was found that all the seeds on the paper which had been kept in the dark had germinated whereas no germination had occurred on the paper which had been exposed to the light (Plate 6). Thus, simple exposure of the mixed carotenoids to light had led to the formation of a potent seed-germination inhibitor.55 However, further investigation showed that the inhibitor was a neutral substance; it was therefore not abscisic acid. Examination of a range of pure carotenoids by the above procedure revealed that the precursors of the inhibitor were certain xanthophylls of which the most important is the epoxide violaxanthin (1 5).57 Large quantities of this pigment were extracted from orange peel.When photo-oxidized three main products (16)-(18) were identified. Of these, one showed growth inhibitory activity fully comparable with that of abscisic acid.58 It was a mixture of the 2-cis-4-trans- and 2-trans-4- trans-isomers of 5-( 1,2-epoxy-4-hydroxy-2,6,6-trimethyl-l-cyclohexyl)-3-methy1-pentadienal. These geometric isomers were separated and most of the inhibitory activity was found to reside in the 2-cis-4-trans-isomer. In view of its formation by the oxidation of certain xanthophyll pigments this new inhibitor has been given the name xanthoxin.59 That the inhibitor is formed within the intact plant has also been established; pea seedlings, for example, exposed to short intervals of light contain some seven times more xanthoxin than similar plants held in the dark.60 Thus, in xanthoxin we have another naturally occurring inhibitor which oper- ates in the chemical control of plant growth.Its formation from xanthophyll epoxides offers for the first time an explanation of why plants grow taller in the dark. Not only this, but the bending of plants towards the light may depend at 66 H. F. Taylor and T. A. Smith, Nature, 1967, 215,1513. 66 G.M. Simpson and R. L. Wain, J. Exp. Bot., 1961, 12,207. 67 H.F. Taylor and R. S. Burden, Proc. Roy. SOC.,1972, B180,317. 6* H.F.Taylor and R. S. Burden, Phytochernistry, 1970, 9,2217. 68 H. F. Taylor and R. S. Burden, Nature, 1970, 227,302. 'a R.S. Burden, R. D. Firn, R. W. P. Hiron, H. F. Taylor, and S. T. C. Wright, Nature, 1971, 234, 95. Chemicals which Control Plant Growth least in part upon the production of the inhibitor on that side of the plant which is exposed to the light. (15) Violaxanthin HOw (16) Inactive (I 7) Inactive HOWHOHOwHo 2-cis-4-trans-Xanthoxin 2-frans-4-rrans-Xanthoxin 7 Promotion of Root Growth So much then for the hormones and hormone inhibitors which, in the light of present knowledge, control the growth of plants. What about that vital part of the plant which is not normally seen-the roots growing in the soil ? Research on hormones in relation to root growth has so far not been extensive and, indeed, it is only recently that indole-3-acetic acid has been unequivocally shown to operate in roots.In some of our recent experiments we have demonstrated that root growth is restricted on exposure to light, that the light receptor area is the root cap, and that with some species only a short flash of light is enough to produce an inhibitory effect.61-63 An interesting development in this area of research came unexpectedly when we were examining the biological activity of 3,5-di-iodo-4-hydroxybenzoicacid 81 H. Wilkins, R. S. Burden, and R. L. Wain, Ann. Appl. Biol., 1974, 78, 337. 6a H. Wilkins and R. L. Wain, Planta, 1975, 123, 217. 63 H. Wilkins, A.LarquC-Saavedra, and R. L. Wain, ‘Plant Growth Substances’, Hirokawa Publ. Co., Tokyo, 1973, p. 1231. Plate 1 Effect of indole-3-acetic acid in promoting rooting. Top row : Bean cuttings rooting in water through action of auxin moving down stem from the growing tip. Bottom row: Similar cuttingsplaced in solution containing 50p.p.m. indole-3-acetic acid Plate 2 Setting tomatoes. Top : Untreated bottom truss ojoutdoor tomuto plunt. Bottom : Similar plant oj' which the bottom flower. truss had been sprayed with ~(2-naphthoxy)- propionic acid at 180p.p.m. in water. s3Plate 3 Eflect of spraying the Wye retardant 3-chlorobenzyltributylamrnonium bromide at 10 5, lo-', and 10 rnol 1-l on to French bean plants Plate 4 Cytokinin activity of 6-benzyloxypurine.Tobacco pith segments on left fail to grow on nutrient medium containing no cytokinin. In flask on right containing the oxypurine, growth has occurred and morphological differentiation has also taken place with the production of young plants Plate 5 Tomato plant on left supplied with adequate water has only one unit of abscisic acid in its leaves. In response to water stress the leaves of the wilting plant on right have built up 52 times this amount of the inhibitor Plate 6 Production of u seed germination inhibitor on exposure of carotenoid pigments ‘to light. Carotenoid-treated paper- in did1 on left which had been held in the dark permits 100% germinution of cress seeds whereas a similar paper k>hich hod pwioiislv been exposed to light allows no germination.Top dish shows germination on untreated papeI’ Plate 7 Efect of DIHB in prornoting root growth of bean seedlings (top row) and pea seedlings (bottom row). Plants on right show growrh in loose soil: those in middle, plmts in compacred soil;those on leJi, plants in compacred soil treated with 10 mol 1 DtHB g200 0) 100sai a .I 50 7 P Q) d)=o I I I I 0 24 48 72 96 Hours water logged Figure Levels of abscisic acid (ABA) in dwwrf bean seedlings: 0 =soi/waterlogged; 0=soil not waterlogged Wain (DIHB) (19), a compound closely related to ioxynil (20), a selective herbicide discovered in the reviewer’s laboratory.64 When cress or rice seedlings were grown in a normal culture solution, roots exposed to light were found to grow to only OH OH CO,H CN (19) DIHB (20) Ioxynil one-third the length of those growing in the dark.However, this inhibitory effect was completely removed when DIHB was present in the solution at 10-5 moll-1. This effect can be spectacular; with seedlings of cress (Lepidium sativum), for example, roots standing in a 10-5 mol 1-1 solution of DIHB grow three times longer than those not receiving the treatment.65 Thus, DIHB removes a constraint on root growth which is imposed by exposure to light.66 Unfortunately, roots in the soil are in the dark so the above beneficial response from DIHB treatment cannot be exploited. However, other constraints on the growth of roots can operate in the field, such as mechanical impedence.When this occurs, DIHB treatment has been shown to have a beneficial effect on root growth.67 This is illustrated for compacted soils in Plate 7. The agricultural implications of these findings and studies on the mode of action of DIHB are now being intensively examined. In this Review I have endeavoured to show that research on hormones and inhibitors is not only providing a better understanding of plant growth but that it is leading to agricultural developments and increased food production. The biological properties of these unique organic molecules would surely have interested John Jeyes had he been with us here today. At any rate, I would Iike to think so. 64 R. L.Wain, Nature, 1963, 200, 28. 65 R. L. Wain, H. F. Taylor, P. Intarakosit, and T. G. D. Shannon, Nature, 1968, 217, 870. 66 R. L. Wain, P. Intarakosit, and H. F. Taylor, Mededel. Rijk. Land. Gent, 1968, 33, 1341. 67 S. M. Wilkins, H. Wilkins, and R. L. Wain, Nature, 1976, 259, 329. 3 275

ISSN:0306-0012    

DOI:10.1039/CS9770600261

出版商:RSC    

年代:1977

数据来源: RSC

摘要:

Enaminones By J. V. Greenhill SCHOOL OF PHARMACY, UNIVERSITY OF BRADFORD, BRADFORD, WEST YORKSHIRE BD7 IDP 1 Introduction The term enaminone is used to indicate any compound containing the conjugated system N-C=C-C=O. It may be a mono-enamine of a lY3-diketone (vinylog- ous amide) or of a 3-keto-ester (vinylogous urethane). Designations sometimes used, such as enamino ketone or p-amino-a$-unsaturated ketone, are misleading in that the compounds rarely show the physical or chemical properties normally associated with ketones. This review covers the chemical and (briefly) physical properties of enaminones and attempts to demonstrate their potential uses in synthetic and medicinal chemistry. 2 Preparation The most used general method for the synthesis of enaminones involves reaction between ammonia or a primary or secondary amine and a 1,3-diketone or 3-keto-ester1 (Scheme 1).This occasionally fails, for example with the very weak 6,. --6,.X (4) Scheme 1 'K. Dixon and J. V. Greenhill, J.C.S. Perkin Ii,1974, 164. Enaminones bases 0-and p-nitroaniline.2 Conversion of the diketone into a vinylogous acid halide or vinylogous ester followed by reaction with the base often gives a kinetically favoured route3 [e.g.(2) R = Me -+ (4) R1=H, R2=p-N02C&-]. The addition of a base to an acetylenic ester or ketone (when available) provides an excellent method of preparati~n.~ The reactions of enamines with acid chlorides are sometimes empl~yed,~ as in the standard preparation of 2-propionyl cyclohexanone (Scheme 5).Similarly, the ring closure [(S) +(6)] is a general method for polycyclic compounds. With nitriles (7) ring closure is catalysed by magnesium perchlorate to give the amidine (8) which gives the enaminone (9) on basic hydrolysis.6 0 (5) N I c10; RIR R = Me, Et,N(CH,),-, or Et,N(CH,)3-, n = 1, 2, 3,4. t-Butylamine reacts with dimedone in refluxing xylene to give, in addition to the expected enaminone, the dienamine-dione (11). This red compound(hmaxHao286 nm, E 14 200; AmaxHao 410 nrn, E 26 100) has coplanar rings.' The same compound is obtained from~the trione (10). For confirmation, the alternative a J. V. Greenhill, J.C.S. Perkin I, 1976, 2207. K. Dixon and J. V. Greenhill, J.C.S.Perkin I, 1976, 2211. E. Winterfeldt and J. M. Welke, Cheni. Ber., 1968, 101, 2381 ; K. Bowden, E. A. Braude, E. R. H. Jones, and B. C. L. Weedon, J. Chem. Soc., 1946, 45. S. Hunig, E. Benzing, and E. Lucke, Chem. Ber., 1957, 90, 2833. A. I. Meyers, A. H. Reine, J. C. Sircar, K. B. Rao, S. Singh, H. Weidmann, and M. Fitzpatrick, J. Heterocyclic Chem., 1968, 5, 151.'J. V. Greenhill, J. Chem. SOC.(C), 1970, 1002. Greenhill structure (12) was prepared from the methoxy derivative of (10). The planes of the rings in this colorless 2-substituted enaminone were at an angle to each other* (AmaxHaO 290 nm, E 17 800). To avoid dimerization in this example, the diketone was converted into the vinylogous bromide (c.5 3; X = Br) before reaction with t-butylamine to give [c.J (4) R1 = H, R2= But].3 Tautomerism It is well established that enaminones of all types exist predominantly in the carbonyl form (13). Undoubtedly this is stabilized by the contribution of the mesomer (15). Dipole moments have been reportedg for two cyclohexanedione derivatives and in both cases are over 6 D. N.m.r. studieslO have confirmed the contribution of zwitterionic forms. The importance of the carbonyl tautomer for compounds derived from acetyl- acetone or dimedone and primary amines is clearly shown by analysis of the -NH-CHz-spin-spin splitting. For example, compound (16; R = H) shows 5.3 Hz, (16; R = Ph) JABJAB 6.8 Hz, and compound (17; R = Ph)JAB 5.3 Hz, all in deuteriochloroform.ll In the case of compound (17; R = H) derived from [15N]methylamine, the 15N-H~ spin coupling is 94.3 Hz with a superimposed M.Ramli, Ph.D. Thesis, University of Bradford, 1973. D. Pitea and G. Favini, J.C.S. Perkin ZI, 1972, 142. loE. J. Cone, R. H. Gamer, and A. W. Hayes, J. Org. Chem., 1972, 37, 4436. l1 G. 0. Dudek and R. H. Holm, J. Anrer. Chem. Soc., 1962,84,2691. Enaminones 4.9 Hz coupling from the methyl group.12 From pKa values of a series of 3-amino- cyclohex-Zenones and their 0-and N-methyl derivatives13 it has been shown that the carbonyl form is favoured over the enol form by a factor of the order of 108. A recent suggestion14 that benzylamine and dimedone react together in hexane to give first the enol-imine tautomer is probably mistaken.It seems likely, from the data provided, that the compounds studied were amine salts of dimedone. 4 U.V.,I.r., and N.M.R. Spectra Four conformers of the enaminone system are possible (Scheme 2). For U.V. spectra to be compared they must be measured in solvents of similar polarity. Most are reported in water or alcohol which stabilize the charge separated form of the excited state.13 In consequence, enaminones absorb at shorter wavelengths in non-polar solvents.13J5 trans-s-trans cis-s-bans frans-s-cis cis-s-cis Scheme 2 Fixed trans-s-transcompoundsleJ7 generally absorb in the range 285-305 nm with molecular extinction constants of 25 OOO to 35 OOO 1 mol-1 cm-1. cis-s-cis Systems1J8 show longer wavelength absorption (300-320 nm) but, more important, lower E values (10 OOO to 20 OOO I mol-1 cm-1).cis-s-transCompounds could not be distinguished from cis-s-cis forms on these data.15 The difficulty of designing enaminones with fixed trans-s-cis configurations has presumably prevented any reliable data for this type becoming available so far. The ranges quoted are for polar solvents and do not include compounds with aromatic ligands. Primary enaminones absorb at the lower ends and tertiary enaminones towards the higher ends of the wavelength ranges. Most of the acyclic compounds in this review are drawn in the s-cis form for convenience, but this is probably the major conformer only when stabilised by hydrogen bonding. Enaminones derived from cyclohexane-l,3-dionel8Jgshow two or three very strong i.r.absorption bands in the range 1540-1610 cm-1. For acyclic com- pound~1~~~0the range is approximately 1540-1660 cm-1. There are no other la G. 0. Dudek and E. P. Dudek, J. Amer. Chem. SOC., 1964, 86,4283. l3 J. V. Greenhill, J. Chem. SOC.(B), 1969, 299. l4 E. J. Kikta and J. F. Bieron, Org. Magnetic Resonance, 1976, 8, 192. l6 C. Kashima, M. Yamamoto, and N. Sugiyama, J. Chem. SOC. (C), 1970, 11 1. l8 K. Ramalingham, M. Balasubramanian, and V. Baliah, Indian J. Chem., 1972, 10, 62. l7 J. V. Greenhill, J. Chem. SOC.(C), 1971, 2699. D. L. Ostercamp, J. Org. Chem., 1970, 35, 1632. J. Dabrowski and K. Kamienska-Trela, Spectrochim. Acta, 1966, 22, 211 and references cited therein. ao N. J. Leonard and J. A. Adamcik, J.Amer. Chem. Soc., 1959, 81, 593. Greenhill bands in the double bond stretching region. Because of the strong mesomeric interactions these bands have been assigned to the whole of the C=C-C=O system rather than the separate units.lg A recent, careful, examination of con- formational effects treats the v(C=C) and v(C=O) vibrations as out-of-phase and in-phase coupled modes.21 Proton chemical shifts in enaminones generally show the expected values. The only difficulty lies in assigning the signals for protons adjacent to the carbonyl group and the double bond. For compounds derived from dimedone the higher field signal usually at about r 7.7 to 7.9 is taken by most authors to represent the C-6 methylene group while another signal at about r 7.5-7.7, which is sometimes broadened by allylic coupling to the vinyl C-H, represents the C-4 methylene group.22 For acetylacetone derivatives of type (16; R = alkyl or arylalkyl) the methyl signals appear at T 8.0 to 8.3 andr 7.9 to 8.1 in CDC13.l1 The unequivocal assignment of these signals awaits further research.Only a few examples of 13C n.m.r. spectra have so far appeared, but the tech- nique seems to be insensitive to changes in configuration and conformation.23 On the other hand, 14N n.m.r. signals fall in a range close to that of amides, but slightly shifted towards the region characteristic of amines.24 5 Protonation The pKa values of four 3-alkylaminocyclohex-2-enonesrange13 from 2.96 to 3.10. Similarly for three tricyclic cis-s-transenaminones [e.g.(25)] pKa's of 2.82 to 2.98 are rep0rted.~5 Nevertheless, many vinylogous amides give stable salts with strong acids.20p26 Invariably the system protonates on oxygen (18).In the U.V. spectra this is shown by a hypsochromic shift of 10-18 nm, although a few aromatic enaminones2' show shifts as small as 4 nm. C-Protonation (19) would effectively remove the U.V. absorption and N-protonation (20) would give much larger hypsochromic shifts. To be sure such weak bases are fully protonated, the spectra must be measured in at least 0.1 M mineral acid.27 Little or no shift is seen for solutions of salts in water or ethanol although this did not prevent some authors from reaching correct conclusions.26p2* I.r.20 (one or two strong bands near 1600 cm-l) and n.m.r.studies29 have confirmed the predominance of 0-protonation. In the one case studied of protonation of a vinylogous urethane, the strong U.V. absorption at 240 and 295 nm disappeared in acid soIution,30 clearly showing C-protonation. z1 D. Smith and P. J. Taylor, Spectrochim. Acta, 1976, 32A,1477. 22 C. Kashima, H. Aoyama, Y. Yamamoto, and T. Nishio, J.C.S. Perkin ZI, 1975, 665. 23 G. R. Bedford and P. J. Taylor, Org. Magnetic Resonance, 1977, 9, 49. 24 J. Dabrowski, A. Skup, and M. Sonelski, Org. Magnetic Resonance, 1969, 1, 341. 25 A. I. Meyers, A. H. Reine, and R. Gault, J. Org. Chem., 1969, 34, 698. 28 G. H. Alt and A. J. Speziale, J. Org. Chem., 1965, 30, 1407. 27 J. V. Greenhill, J.C.S. Perkin I, 1976, 2207.W. Sobotka, W. N. Beverung, G. G. Munoz, J. C. Sircar, and A. J. Meyers, J. Org. Chem., 1965, 30, 3667. p9 H. E. A. Kramer, Annulen., 1966, 696, 15. OoJ. C. Powers, J, Qrg. Chem,, 1965, 30, 2534. Enaminones 6 Alkylation The alkylation of vinylogous urethanes which, after hydrolysis, gave a-alkylated p-keto esters usually offers no advantage over the direct alkylation of the p-keto-e~ter.~lWhen the very reactive propargyl bromide is involved, however, the use of an enaminone prevents the dialkylation which occurs when ethyl acetoacetate is the ~ubstrate.3~ Vinylogous amides derived from butane-l,3-dione and pentane-2,4-dione react with methyl iodide to give C-methyl derivatives.33-35 4-Pyrrolidinyl-3-penten-2-one (21) gives a C-methyl derivative (22) contaminated with some O-methyl salt.Ethyl iodide, on the other hand, gives the O-ethyl salt (23) in 31 % pure yield.20 Trans-s-trans Enaminones derived from cyclohexane-l,3-diones give high yields of O-alkylated salts, even when methyl iodide is used.13J5s25 N-Methylation has been achieved by preliminary de-protonation with sodium hydride followed by treatment with methyl iodide. The identity of the product has been confirmed by independent synthesis13 (Scheme 3). This methylation of (24; R = C6H11) R = -CH,CH,Ph, -CBHIl Scheme 3 31 W. M. Lauer and G. W. Jones, J. Amer. Chem. SOC.,1937, 59,232. 32 G. Eglington and M. C. Whiting, J. Chem. Soc., 1953, 3052. a3 A. Combes and C. Combes, Bull. SOC.chim.France, 1892, 7, 778. 34 N.K. Kochetkov, Bull. Akad. Sci. U.S.S.R.,1953, 833. 35 N. K. Kochetkov, M. G. Ivanova, and A. N. Nesmeyanov, Bull. Akad. Sci. U.S.S.R., 1956, 687. GreenhiII also gave a trace of 2-methyl derivative, detected by t.l.~.3~ Although potentially useful, only a few examples of N-alkylation of enaminone anions have been reported, and further developments may require the use of different basic catalysts. The cis-s-transenaminone (25) reacts with methyl iodide25 in non-polar solvents to give the C-methyl derivative (26), but in alcoholic solvent to give the 0-methyl compound (27). In aprotic polar solvents (acetonitrile etc.) the ratio of (26):(27) increases with reaction time. It was shown that nucleophilfc attack by the iodide ion can reverse the reaction (25) -(27) and allow the C-methyl derivative (26) to ac~urnulate.~~ A new technique for regiospecific alkylation at the y position of a tertiary enaminone involves de-protonation with butyl-lithium or lithium di-isopropyl- amide, followed by treatment with an alkyl halide37 (Scheme 4). Compound (28) R = Me, Ph, -0Me R1= Me, Prn, Bz,-CH,-CH=CH2, -CH2-COOEt Scheme 4 gives either the a' (29) or the y (30) derivative, depending on whether base or enaminone is in excess.38 These methods may have useful synthetic applications; for example a recently reported ring closure [(31) -+ (32)] may have value in alkaloid ~ynthesis.3~ Quaternization of enaminones has not been achieved and only simple chloro- vinyl ketones react with tertiary amines to give quaternary ammonium saIts.35 G.V. Kondrat'eva, V. I. Gunar, L. F. Ovechkina, S. I. Zav'yalov, and A. I. Krotov, Bull. Akad. Sci. U.S.S.R. 1967, 609. 37 M. Yoshimoto, N. Ishida, and T. Hiraoka, Tetrahedron Letters, 1973, 39; T. A. Bryson and R. B. Gammill, Tetrahedron Letters, 1974, 3663. J. E. Telschow and W. Rensch, J. Org. Chem., 1975, 40, 862. Enaminones (31) (32) 7 Acylation Acetylation or benzoylation of primary or secondary enaminones gives generally N-acyl derivatives.16J8 From simple acyclic secondary enaminones mixtures of N- and a-C-acetylated derivatives are obtained in the presence of triethylamine; with pyridine traces of rather unusual 0,N-diacetyl derivatives are reported, but these compounds have not yet been well chara~terised.3~ An enamine from cyclohexanone (33) gives an enaminone with one equivalent of an acid chloride (34).With excess acid chloride 0-acylation must occur because mild hydrolysis gives an enol ester5 (Scheme 5). Scheme 5 39 L. Kozerski, Tetrahedron, 1976, 32, 1299. Greenhill The teritary enaminone (35) reacts with acetic anhydride or acetyl chloride to give the C-acetyl derivative (37) in good yield. Pivalyl chloride and several substituted benzoyl chlorides also give C-acyl derivatives, but trichloracetyl chloride gives a chloriminium salt (38). It was suggested that initial 0-acetylation gave the unstable salt (36), but normally attack by the chloride ion at the carbonyl carbon (1) reformed the original reactants and the more slowly formed C-acetyl derivative (37) accumulated.When the alternative attack by chloride ion at ring carbon (2) could release a trichloracetate ion (better leaving group) the chlori- minium salt (38) was f~rmed.~O Compound (38) is also formed when tosyl chloride, picryl chloride or, best, phosphorus pentachloride are used. These all give anions which are good leaving groups. CI (37) (35) (38) Acylation of a series of acyclic tertiary enaminones with dichloroketen gave cc derivatives from vinylogous urethanes and y derivatives from vinylogous amides41 (Scheme 6). COCHCI R = Me,Ph -R COCHCI, Scheme 6 40 G. H. Alt and A. J. Speziale, J. Org. Chem., 1964, 29, 798. 41 M.Yoshimoto,T. Hiraoka, and Y. Kishida, Chem. andPharm. Bull. (Japan), 1970,18,2469. Enaminones 8 Reactions with Grignard Reagents The Grignard reactions on acyclic enaminones, which have been reported, all give nitrogen-free products through 1,4 addition42143 (Scheme 7). The cis-s-trans 0 0 R = Ph, Me; R’ = Ph, Me,Et Scheme 7 compound (39) does give the enamine alcohol (40) in high yield, but with methyl magnesium iodide only 10% of the 1,4 addition product (41) is obtained together with unreacted starting material and a mixture of unidentified compounds. Preliminary conversion of the enaminone (39) to its 0-acetyl iminium salt (42) allows preparation of the amino ketone (41) in a useful 60 to 65% ~ield.4~ 0 0 ($0CI-Organometallic reactions on acyclic or cis-s-trans enaminones clearly have synthetic potential, but several attempts in this laboratory to employ trans-s-trans E.Benary, Chem. Ber., 1931, 64,2543. ra N. K. Kochetkov, Bull. Akad. Sci. U.S.S.R.Div. Chem. Sci., 1954, 37. 44 A. J. Meyers and S. Singh, Tetrahedron Letters, 1967, 5319. Greenhill enaminones derived from cyclohexane-l,3-diones have failed, good recovery of starting material always being made.45 9 Reactions with Aldehydes and Ketones The anion of the acyclic enaminone (43) generated at -60°C reacts with benzaldehyde to give the y substituted product37 (44).However, the same ena- minone reacts with p-nitrobenzaldehyde under catalysis by toluene-p-sulphonic acid, sodium ethoxide, triethylamine or 1,4-diazabicyclo-octane at reflux temperatures to give a similary substituted product41 (45).One example has been reported where an enaminone anion reacts with a ketone to give a tertiary alcohoP7 [(46) -(47)]. OH P-N0,C.H ,CHO OH 0 0 The derivative of o-phenylenediamine(48)reacts readily with aldehydes to give the hexahydrodibenzodiazpinones (49)in a Mannich-type reaction.46 Dimedone derivatives (50) react with formaldehyde under neutral conditions to give the methylenebisenaminones (51) which on heating with acid rapidly 46 K. Dixon, Ph.D. Thesis, University of Bradford, 1976. ‘O S. Miyano and N. Abe, Chcm. and Pharm. Bull. (Japan), 1972, 20, 1588. Enaminones change to the hexahydroacridinediones (52).Either the original enaminones or the derivatives (51) react with aqueous acidic formaldehyde at room temperature to give the spiranes (53), again via an intramolecular Mannich-type reaction. Acetaldehyde under the same conditions converts the enaminone (50; R = H) into a tetra hydro benzoxazinone (54) .17 IH (54) 0 0(7-Jp-m-N O NH NH -aIIR II R RR Robinson annelation of 2-methylcyclohexane-1,3-dionewith pent-Zen-4-one was examined as the first stage of a total synthesis of calarene. Only a low yield of unwanted trans-isomer was obtained. Conversion of the dione to its enaminone before annelation gave an improved yield (27%) of a 1 : 1 mixture of isomers (Scheme 8) from which the cis-isomer was separated and the synthesis com- pleted.47 The reaction involved Michael addition of the vinyl ketone to the 2-position of the enaminone followed by hydrolysis of the immonium inter- mediate and aldol condensation.38 10 Reactions with other Electrophiles Treatment with N-bromosuccinimide, N-chlorosuccinimide or 1 equivalent of molecular bromine generally gives the a substituted enaminone.48~49 Dimedone has been used for N-protection in peptide synthesis and removed as 2,2-dibromo- 47 R.M. Coates ane J. E. Shaw, J. Amer. Chem. SOC.,1970, 92, 5657. 48 T. Tokumitau and T. Hayashi, Nippon Kagaku Kaishi, 1973, 11, 2152. 49 I. Jirkovsky, Canad. J. Chem., 1974, 52, 55. 288 Greenhill Calarene Scheme 8 dimedone by treatment with excess bromine.50 Primary and secondary enami- nones react readily with isocyanates at elevated temperature to give vinylogous ureas by ct sub~titution.~gTertiary acyclic compounds react even at room temperature to give 1:2 adducts in which both the ct and y positions are s~bstituted.~' Acyclic enaminones give, with dibenzoylperoxide, a-benzoyl derivatives which are converted to oxazoles in refluxing acetic acid52 (Scheme 9).R RCO Ph OCOPh R = Me,-OEt Scheme 9 11 Reduction Selective reduction of the enaminone system could provide a source of p-amino-ketones. In many examples investigated, however, the standard techniques fail to give any reaction and under more forcing conditions either nitr0gen~3,~~ or oxygen,54is split out of the molecule. On hydrogenation over platinum, the simpie enaminone (55) gives only butanone.It is suggested that hydrogenolysis removes the nitrogen atom first and then the olefin bond is saturated since Mannich bases give amino-alcohols on h~drogenation.~3 The closely related compound (56) also gives the neutral ketone (57) when hydrogenated over palladium. Over rhodium 1o B. Halpern and L. B. Jones, Nature, 1964, 202, 592. .sl 0. Tsuge and A. Inaba, Bull. Chem. SOC.Japan, 1973,46,286; 2221. 5* H. 5. Jakobsen, E. H. Larsen, P. Madsen, and S. 0.Lawesson, Arkiv. Kemi., 1965,24, 519. 53 J. C. Martin, K. R. Barton, P. G. Gott, and R. H. Meen, J. Org. Chem., 1966, 31, 943. 64 J. V. Greenhill, M. Ramli, and T. Tomassini, J.C.S. Perkin I, 1975, 558. Enaminones or ruthenium, however, the saturated amino-alcohol(58) results and with lithium aluminium hydride the amino-ketone (59) is obtained.53 3-Aminocyclohex-2-enone is hydrogenated over Raney nickel at 70 "C to a mixture of the 3-aminocyclohexanols. The method gives a product considerably richer in the trans-isomer than the previously used procedure, reduction of 3-acetamidophenol.54 2-Acetyl cyclopentanone (60) reacts with ammonia to give only one product (61) which is unaffected by metal hydride reducing agents.Hydrogenation gives an amino-alcohol which can be converted to an amide and oxidised to the amido- ketone (62). The appearance of the methyl signal as a doublet in the n.m.r. spectrum confirms the structure of the enaminone (61). 2-Acetyl cyclohexanone (63) on the other hand, reacts with ammonia at the ring carbonyl group to give (64) so a similar series of reactions gives an amido-ketone (65) with a singlet methyl signa1.54 The resistance of the enaminone system to reduction is illustrated by the reaction of the vinylogous urethane (66) with excess lithium aluminium hydride or lithium borohydride.The non-conjugated ester group is reduced and the al- cohol spontaneously ring closes to the furanone (67). Acid hydrolysis54 gives tetronic acid (68). Although trans-s-trans enaminones have not been reduced with metal hydrides, the cis-s-trans compound (39)44 and some acyclic enaminoneP are reduced to saturated amino-ketones. 12 Heterocyclic Syntheses Simple chlorovinyl ketones and their derived enaminones react together at room 66 G.N. Walker, J. Org. Chem ,1962, 27, 4227. Greenhill ....H, -NHCOR + 3 steps 0 0 COOMe ButNHaCOOMe ButNH R, HOlii, temperature to give pyridine derivatives in good yield. Tertiary enaminones prepared from these chlorovinyl ketones react with cyanoacetamide to give 6-alkyl-3-cyanopyrid-2-0nes~~(Scheme 10). pi+ R CN I CHICONH, _____f R CNMesR n R = Me, Et, Prn, But, nCsH,, Scheme 10 4 291 Enaminones Cyclodehydration of the enaminones (69) with polyphosphoric acid gives the expected tetrahydropenanthridines(70). When a different substituted aniline hydrochloride and zinc chloride are used a mixture of tetrahydroacridines is obtained (Scheme 11).The original arylamine moiety was retained afrer re- arrangement (71) or replaced by the reacting arylamine (72).66 A similar rearrange- ment must be involved in the reaction of a 2-aminomethylene cycloalkanone with a 1,3-dione57 (Scheme 12). R = Me,-OEt Scheme 12 4Aminouracil reacts with a series of enaminones to give fused heterocycles (e.g. Scheme 13). Transamination precedes ring closure.58 3-Aminocyclohex-2-enone (73) reacts with propynal to give the ketone (74) which can be reduced by the Huang-Minlon method to 5,6,7,8-tetrahydro- quinoline.69 With methyl propiolate the same enaminone gives the 5-oxotetra- B. D. Tilak, H. Berde, V. N. Gogte, and T. Ravindranathan, Indian J. Chem., 1970, 8, 1. G. Bouchon, K-H.Spohn, and E. Breitmaier, Chenz. Ber., 1973, 106, 1736. 6* W. Remp and H. Junek, Monatash., 1973, 104, 1101 and references cited therein. *' C. Ruangsiyanand, J. Rimek, and F. Zymalkowski, Chew. Ber., 1970, 103,2403. Greenhill 0 OR Scheme 13 hydroquinolone (75) which is an intermediate in the preparation of 4-azaster-oids.60 Acyclic tertiary enaminones react with hydroxylamine to give high yields of 5-alkyl isoxazoles which are a source of /3-keto-nitriles (Scheme 14). The parent chlorovinyl ketones give mixtures of 3-alkyl and 5-alkyl is0xazoles.~3 R = Me,Prn; R’ = Me,Et Scheme 14 M.A. T.Sluyter, U. K. Pandit, W. N. Speckamp, and H. 0.Hiusman, Tetrahedron Letters, 1966, 87. Enaminones Irradiation of allylamine derivatives of dimedone gives bicyclo [2,1,1]hexanes61 (Scheme 15).0 R R = Me,Ph, -CH2-CH=CH, Scheme 15 13 Conclusion The reactions of enaminones vary somewhat with conformation and nitrogen substitution, but the physical and chemical properties are sufficiently similar for the group to represent a class of organic compounds in its own right. With five positions vulnerable to electrophilic attack and two to nucleophilic attack, the enaminone system shows interesting and sometimes complicated reactivity. Enaminones are already established as synthetic intermediates, particularly in heterocyclic chemistry, but, for many of the reactions mentioned in this review, only a few examples have been reported. Although it is not possible to comment on the generality of these, their wider use in synthetic schemes clearly merits investigation.In the future, they may be useful as protected amines. As pro-drugs their potential is considerable. An enaminone derivative of a physiologically active amine may well show improved transport across biological membranes and allow a high concentration of the amine to be released close to the site of action. Such derivatives have the added advantage, always attractive to medicinal chemists, that they may have physiological activity in their own right. Recent years have seen a rapid rise of interest in this area, and we can look forward to a continuing expansion of the literature of enaminone chemistry. 61 Y. Tamura, H. Ishibashi, M. Hirai, Y.Kita, and M. Ikeda, J. Org. Chem. 1975,40, 2702.

ISSN:0306-0012    

DOI:10.1039/CS9770600277

出版商:RSC    

年代:1977

数据来源: RSC

摘要:

Ion-Molecule Reactions in the Evolution of Simple Organic Molecules in Interstellar Clouds and Planetary Atmospheres By W. T. Huntress, Jr. JET PROPULSION LABORATORY, PASADENA, CALIFORNIA 91 103, U.S.A. 1 Introduction There is a general appreciation among scientists for the fundamental physics involved in the structure of the universe as we perceive it through the media of astronomy. The word astrophysics has been coined by this appreciation, and the subject is fostered by the application of the newest developments in modern physics to the science of astronomy. Neutron stars, black holes, and quasars are just a few examples of the marriage of modern physics and astronomy. The new understanding of the universe as a dynamic and evolving entity, alive with the exotic fires of an only dimly understood relativistic particle physics, is a per-ception which was made possible by modern physics and which in turn provides impetus to the progress of both physics and astronomy.The ancient and revered science of astronomy has undergone a rebirth in recent decades in the wake of these advances. The science of chemistry is now beginning to make its mark upon astronomy and space science in the same way that physics has done in the past 20 years. Chemistry, in the sense of nuclear reactions in stars and of the spectroscopic detection of atomic and molecular species in the atmospheres of stars and the planets, has been a part of astronomy for almost the entire century. It is only since the advent of the space programme, of radio astronomy, and of air pollution, that chemistry has been considered in astrophysics and space science in the sense of chemical reactions in these environments.Through the technique of radio astronomy, organic molecules have been discovered in the depths of interstellar clouds. The work being done to explain the origin of these molecules1 is providing a new science of astrochemistry. The space programme has opened new spectro- scopic eyes on the atmospheres of the other planets in our solar system and has been communicating back new data on the chemical constituents of the atmos- pheres of Mercury, Venus, Mars, and Jupiter. Work is now needed on the chemistry of these atmospheres in order to explain their composition.2 In turn, recent concern over chemical pollution of our own atmosphere has stimulated much new effort into photochemical reactions in the Earth’s atmosphere.3 Reviewed by W.D. Watson in ‘Atomic and Molecular Physics and the Interstellar Matter’, North-Holland, Amsterdam, 1975, and by A. Dalgarno and J. H. Black, Reports Progr. Phys., 1976, 39, 573. a J. S. Lewis, Ann. Rev. Phys. Chem., 1973, 24, 339. Department of the Environment, CentraI Unit on Environmental Pollution, ‘ChIoro-fluorocarbons and their Effect on Stratospheric Ozone’ (Pollution Paper No. 5 H.M. Stationery Office, 1976). Ion-Molecule Reactions in the Evolution of Simple Organic Molecules Ionic processes play a very significant part in the chemistry of interstellar clouds and planetary atmospheres.In this review we will examine the role of ion-molecule reactions in the chemical processes which result in the evolution of organic molecules in extraterrestrial environments. The approach will be to start with the ion chemistry of the most tenuous of astrophysical environments and work up to the most dense. The most tenuous are the thin interstellar clouds dispersed throughout the galaxy. The densities in these clouds range from 10 to lo9 molecules ~m-~ depending on their mass and evolutionary history. Working up through astronomical condensations of increasing density, we will follow the important contribution made by ion chemistry to the chemical evolution of both inorganic and organic compounds in these clouds from atomic to diatomic to polyatomic species.These chemical compounds eventually find their way as frozen volatiles in comets and are present as the gaseous envelopes around planetary objects when the interstellar cloud collapses through the stellar nebula stage to form a sun and a planetary system. From this point we will examine the importance of ion-molecde reactions for the synthesis of organic molecules in planetary atmospheres of reducing composition. Jupiter is an example of such a planet in our own solar system, and the primitive atmosphere of the Earth was most likely reducing in nature also.4 These same ionic reactions may also be responsible for the formation of some of the intermediate organic compounds which have been observed in laboratory studies of the prebiotic, abiogenic synthesis of complex organic molecules, those compounds required to initiate and sustain chemical life on our own and other planets in the universe.5 2 Ion-Molecule Reactions Ion-molecule reactions are defined as the formation of new product species as the result of a collision between an ion and a neutral atom or molecule.Ions are formed in extraterrestrial environments by the impact of energetic particles or photons on neutral species. In interstellar clouds, atoms and molecules can be ionized by both the galactic ultraviolet (u.v.) flux from starlight and by cosmic rays and X-rays.l In planetary atmospheres, solar photoionization in the tenuous part of the upper atmosphere forms an ionosphere about the planet.6 In more dense regions of planetary atmospheres, where clouds can form from condensable species and where precipitation can occur, ions are formed in electrical discharges in thunderstorms.’ Cosmic rays can also produce a low level of ionization in planetary atmospheres above the heights at which the pressure is of the order of several atmospheres.4 W. T. Huntress, Jr., J. Chem. Educ., 1976, 53, 204. 5 C. Ponnamperuma and N. W. Gabel, Space Life Sci.,1968, 1, 64. a M. J. McEwan and L. F. Phillips, ‘Chemistry of the Atmosphere’, Arnold, London, 1975. 7 A. Bar-Nun and A. Shaviv, Icarus, 1975, 24, 192. EL. A. Capone, R. C. Whitten, S. S. Prasad, and J. Dubach, Astrophys. J., 1977, 33, 495. L. A. Capone, R. C. Whitten, J.Dubach, S. S. Prasad, and W. T. Huntress, Jr., Icarus, 1976, 28, 367. Huntress The types of ion-molecule reaction which will be important are a function of the density. For pressures much below 1 atmosphere, bimolecular ion-molecule reactions are generally the most important. Reaction (1) is a very simple example H2+ + H2 4H3+ + H (1) of a bimolecular ion-molecule reaction important in dense interstellar clouds and in the atmospheres of the Jovian planets where Hz is the dominant molecular species. At the very low densities (-10-106 molecules cm-3) of interstellar clouds, the time between collisions is very long, from 3 years to 20 minutes, and C++ H2 + CHa+ + hu (2) radiative association reactions become important. Reaction (2) is a prime ex- ample of a radiative association reaction important in interstellar clouds.1 In this process, a photon (hv) is emitted rather than a heavy neutral particle as in reaction (1).The C+ion is otherwise unreactive towards molecular hydrogen. Bimolecular reactions can and most often do occur at nearly every collision between an ion and a molecule when the process is thermodynamically allowed. On the other hand, the probability for emission of a photon to stabilize the collision complex is very small during the short time the ion and neutral species are in close contact. Radiative association reactions are therefore very much slower than bimolecular reactions. Radiative association reactions have not yet been confirmed in laboratory experiments, and the existence of the radiative association process is based primarily on theory.These reactions are very much akin to collisional association reactions. An appropriate example is the reaction between methyl cations and water [reaction (3)], where the excited intermediate complex formed during the collision between CH3+ and H2O can either dissociate back into the reactants (no reaction) or can emit an infrared photon to produce protonated methanol (radiative association), or where before dissociation the complex can undergo another collision with a third spectator species, Ha in this case, which stabilizes the complex to form protonated methanol. The collisional association reaction (3b) becomes important only at high pressures (greater than about 10-3 atm) because of the requirement for a third-body collision, but collisional association reactions are also clearly related to the radiative association reaction because the rates for both processes depend on the lifetimem of the intermediate complex.Collisional association reactions are commonly observed at high pressures in the Ion-Molecule Reactions in the Evolution of Simple Organic Molecules lab~ratory,~from which the dissociative lifetime Td can be deduced. From Tdand from Tr,the radiative lifetime of the excited intermediate ion, the rate for the radiative process can be estimated. The lifetime of the complex towards dissoci- ation, Td,is temperature dependent since at lower encounter velocities the excit- ation energy in the complex is smaller and therefore the lifetime is longer.The larger74at low temperatures in interstellar clouds may allow radiative association reactions to become important even in competition with bimolecular reaction channels for the complex. Smith and Adamsg have observed the collisional association reaction (3b) and have proposed that the radiative analogue (3a) may be important for methanol synthesis in interstellar clouds (p E 106 molecules cm-3, T = 10-30 K). Likewise, the collisional process (3b) may be important for methanol synthesis in the lower atmosphere of Jupiter (p z 1020 molecules cm-3, T = 300 K). Both of these reactions become important because the bimolecular reaction between CH3+ ions and H20, reaction (4), to produce CH3+ + H2O [CH30H2+]"+ CH20H+ + H2 (4) protonated formaldehyde does not occur.1o Reaction (4) is one of the few thermodynamically allowed biomolecuiar ion-molecule reactions which do not occur at the collision rate.More often, thermodynamically allowed ion-molecule reactions occur with nearly unit efficiency: almost every collision results in a chemical reaction. Even conden- sation reactions requiring rearrangement in an intermediate complex are often observed to occur at the collision rate.lO Collisions between ions and molecules are in general 1~lOOOtimes more rapid than collisions between neutral collision partners. Consequently, this results in much larger rate constants for ion-molecule reactions than for reactions between neutral species.Both the larger collision frequencies and the large chemical reaction efficiency in ion- molecule collisions are the result of the charge on the ion.11 The ion induces a dipolar field in the polarizable neutral collision partner. This field forms a potential between ion and neutral which is much more attractive than the simple van der Waals attraction between neutral species. The ion and neutral are attracted over very much larger distances than are neutral species. The collision cross-section for ions with neutrals is therefore much larger than the hard-sphere cross-section of molecular dimensions (-few A2) appropriate for neutral-neutral collisions. Ion-molecule collision cross-sections are of the order of a few hundred As.Because of the form of the ion-induced dipole potential, the collisional rate constant is also independent of ion velocity, and consequently of kinetic temperature. The reason that reaction efficiency in ion-molecule collisions is high is also due to the ion-induced dipole field. From infinite separation, the energy gained in bringing ion and neutral together at molecular dimensions along this ' D. Smith and N. Adams, Astrophys. J., 1977, in press. loW. T. Huntress, Jr., Astrophys. J., Supp. Ser., 1977, in press. l1 E. W. McDaniel, 'Collision Phenomena in Ionized Gases', Wiley, New York, 1964. Huntress potential is of the order of tens of kcal mol-1.11 This is sufficient energy to over- come most kinetic barriers to reaction, so that most ion-molecule reactions have little if any activation energy.Their rate constants can therefore be essentially independent of temperature. This fact is extremely important in astrophysical applications.1 The temperature in the interior of interstellar clouds can be as low as 3-10 K, at which temperature most chemical reactions among neutral reactants are infinitely slow, excepting a few radical-radical reactions. Ion- molecule reactions can be extremely fast even at these low temperatures. Another feature of the ion-induced dipole potential is that it allows for spiralling and multiple impact encounters,ll so that the time the ion and molecule remain in intimate contact can be much longer than in neutral-neutral collisions. This longer time facilitates the formation of longer-lived intermediate complexes, which are important for bimolecular as well as associative ion-molecule reactions.3 Diffuse Interstellar Clouds The conditions in diffuse interstellar clouds have been reviewed recently both by Watson and by Dalgarno and B1ack.l These are regions in interstellar space where the density of gas increases above the normal -0.1 molecules cm-3 pervading the distances between stars. Diffuse interstellar clouds have densities of the order 10-103 molecules cm-3 and are optically thin. Stars are visible through them, and some of the species which are present in these clouds are identified by their absorptions in the spectra of background stars. In many cases these regions are illuminated by radiation from nearby stars, which resonantly excites the atoms present in them.The luminous regions of the Great Nebula in Orion pictured in Plate 1* are examples of diffuse interstellar clouds. The tem- peratures in diffuse clouds can be as high as several hundred degrees (see Table 1). The majority of the material in diffuse clouds is atomic, and they contain only a very small proportion of molecular species other than Hz. In diffuse clouds n(H2) < n(H) in general. Table 2 gives the cosmic abundances of elements as they are found in the universe. The composition given in Table 2 represents the average composition of interstellar and stellar material as it has been compiled from spectral observations of the sun and the stars, from particles in the solar wind, and from meteoritic abundances.12 Hydrogen and helium dominate the composition of the universe.Jn diffuse clouds, atoms are ionized by cosmic rays, by X-rays, and by the U.V. flux from the collective starlight in the galaxy. The galactic U.V. flux has a cut-off above 13.6 eV. Hydrogen atoms in the space near stars are ionized by U.V. light with energies above 13.6eV so that the light which reaches interstellar clouds only contains photons with energies of less than 13.6 eV. The galactic U.V. flux through diffuse interstellar clouds therefore is not capable of ionizing atoms with ioniza- tion potentials greater than 13.6 eV, such as H, He, 0,N, Ne, or Ar atoms. These latter atoms are ionized by cosmic rays and X-rays.The atoms with ioniza- tion potentials less than 13.6 eV, such as C, S, C1, Si, and metal atoms, are * Plates 1-3 are between pp. 306 and 307. l2 A. G. W. Cameron, Space Sci. Rev., 1973, 14, 383; ibid., 1973, 15, 121. lon-Molecule Reactions in the Evolution of Simple Organic Molecules Table 1 Typical conditions in interstellar clouds Difuse Dense Density 10-103 cm-3 lO3-lO6 cm-3 Ion mean free time 3 years-10 days 10 days-20 min Temperature 100-30 K 20-10 K* n(Hzlln(H)Major ion Fractional ionization 0.05-102 Cf 10-4-10-5 102-106 S+, Mg+, metal ions 10-5-10-7 Chemical equilibration time 150-10 My 10-4.1 My *For clouds with embedded proto-stars, or with hot stars nearby, the temperature can be much higher.The Orion molecular cloud with p Z 10&-10' molecules cm-* has a temperature T M 70-90 K. Table 2 Cosmic elemental abundances and ionization potentials Element Relative abundance Ionization potentiallev H 1.oo 13.6 He 0.063 24.6 0 5.9 x 10-4 13.6 C 3.5 x 10-4 11.3 N 8.5 x 10-5 14.6 Ne 7.6 x 10-5 21.6 Si 3.5 x 10-5 8.2 Mg 3.0 x 10-5 7.7 S 1.6 x 10-5 10.4 Ar 5.5 x 10-6 15.8 Fe 3.2 x 10-6 7.9 Al 2.5 x 6.0 Ca 2.1 x 10-6 6.1 c1 2.0 x 10-6 13.0 Na 1.5 x 5.2 F 1.1 x 10-6 17.4 ionized by the galactic U.V. flux and by cosmic rays and X-rays. The ionizing galacticU.V. fluis of the order of 106-107 times higher than the ionizing cosmic ray and X-ray flux, so that in diffuse clouds U.V. ionization of C atoms is the most important ionization mechanism.The recombination of C+ions with elec- trons is a radiative process, and is consequently very slow. Practically all of the carbon atoms are ionized in diffuse clouds and n(C+) z n(e-) z no(C) where no(C) is the cosmic abundance of carbon. All S,CI, Si, and metal atoms are ionized by the galactic U.V. as well. H+ and He+ ions are produced by cosmic and X-ray ionization of H and He, but these atoms remain mostly neutral and the number of Hf and He+ ions is of the order of, or less than, the number of C+ ions. In diffuse clouds, where the species are mostly atomic, the oniy processes which can occur in gas-phase collisions between atomic ions and atoms, other than simple scattering, are charge transfer and radiative association.Among the more 300 Huntress abundant atoms H, He, C,N, and 0 and their ions, the only charge-transfer reaction which occurs at thermal energies is the reaction H+ + o of + H (5) Reaction (5) is an accidentally resonant charge-transfer reaction. The ionization potentials of H and 0 atoms are almost exactly the same: IP(0) = 13.62 and IP(H) = 13.60 eV. Charge-transfer reactions are unusual among ion-molecule reactions in the sense that they involve transfer of a very light particle, the electron, at potentially very large distances (>100A) so that they therefore exhibit strong quantum mechanical effects.13 Charge-transfer reactions among the first row of atoms are generally not observed even though thermodynamically allowed.For example, the IP of He is greater than that of 0,but the He+-O charge-transfer reaction does not occur. Charge-transfer reactions require that there be a state available in the product ion which is resonant in energy with the recombination energy of the reactant ion, and furthermore that the probability be large for the transition from the ground state of the neutral reactant to the resonant state of the product This condition is met by the ground states of H, 0+,and 0 in the forward and reverse reactions (5). The forward reaction (5) only is very slightly endothermic (AH = +0.5 kcal mol-1, corresponding to about 250 K) and therefore probably has a strong temperature dependence. The charge-transfer reaction (5) does not contribute in any large way to the synthesis of diatomic ions or molecules in diffuse clouds other than to enhance the O+ ion density.14 The important step required to start the fires of chemistry burning is the synthesis of diatomic species. Once this is accomplished, fast bimolecular ion-molecule reactions become possible and new species can be synthesized.In order to synthesize diatomic species, however, radiative processes and heterogeneous reactions at interstellar grain surfaces must be invoked.Heterogeneous recombination of H atoms on grain surfaces is the generally accepted mechanism for synthesis of interstellar H2:l grainH + H+H2 Molecular hydrogen is non-polar and has a very low boiling point. It is easily evaporated from grain surfaces even at low interstellar temperatures.Any other molecular species, such as CH, even if formed on a grain surface, is difficult to evaporate back into the gas phase at interstellar temperatures. Reaction (6) is required in order to explain the presence of interstellar H2; gas-phase processes do not work. Gas-phase reactions are probably responsible, however, for the synthesis of most of the other molecules observed in interstellar clouds. J. B. Laudenslager, W. T. Huntress, Jr., and M. T. Bowers, J. Chem. Phys., 1974, 61,4600. l4 S. S. Prasad and W. T. Huntress, Jr., submitted to Asrrophys. J. Ion-Molecule Reactions in the Evolution of Simple Organic Molecules The gas-phase mechanisms acting to produce diatomic species in diffuse clouds are formation of CH+ by the radiative association of Cf and H and for- mation of OH by associative detachment of 0-to H or H-to 0, Cf + H + CHf + hv (7) 0-+ H -+ OH + e-(8) H-+ 0 -+ OH + e-(9) Reactions (8) and (9) owe their origin to the formation of 0-and H-by the radiative association of electrons with 0 and H atoms: e-+ 0 -+ 0-+ hv (10) e-+ H -+ H-+ hv (1 1) There are no laboratory data on any of the reactions (7)-(11), but there are theoretical calculations for the rate constants of the rate-determining radiative association steps (7), (lo), and (ll).15J6 If either or both of the detachment reactions (8) and (9) are fast, then OH formation can be just as important as CH+ formation via (7) in diffuse c10uds.l~ Generally, reactions of negative ions are not important in interstellar clouds because the abundance of negative ions is far less than the abundance of positive ions.16 Negative ions are formed only by pair processes (AB -+ A+ + B-) in ionization events, or by electron attachment reactions.The cross-sections for these processes are very small compared with positive-ion formation and scattering.ll Reactions (8) and (9) are important as an initiating process only in diffuse c10uds.l~ Radiative association as in reaction (7) occurs by a transition between energy levels during a collision which leaves the diatomic molecule or ion in a state below the asymptotic dissociation limit along which the reactants have approached each other. This situation is illustrated in Figure 1.The probability for this transition, and the consequent emission of a photon during collision, is largest for transitions between electronic energy levels where the lifetime of the excited state towards radiation is the shortest (71-2 10-6-10-s s). If the lifetime of the collision complex is of the order of several vibrations (Td 2 s) then the probability for emission per collision is of the order of P = Td/Tr % 10-7-10-5. Vibrational transitions from the dissociation limit to lower states take too long (7r z 10-3-10 s) for these processes to be important for synthesis of diatomic species. For more complex intermediates such as in reaction (3), 7d can be quite long owing to the number of available modes into which the collisional excitation can be channeled.Vibrational transitions can become effective under these circumstances and radiative rate constants relatively large. For diatomic species, however, the only potential candidates for radiative association are those which have low-lying electronic states correlated with the reactants in their ground states, as in Figure l5 E. Herbst, J. G. Schubert, and P. R. Certain, Astrophys. J., 1977, in press.leA. Dalgarno and R. A. McCray, Astrophys. J., 1973, 181, 95. Huntress 4 3 2 1 UJeV 0 -1 -2 -3 -4 0 1 2 3 4 RIA I Figure 1 Illustration of radiative processes forming CH+ during collisions of Cf and H, or Hf and C. The former is more important because of the large C+/C ratio in difluse interstellar clouds.An allowed radiative transition from the low-lying A‘n state to the X’Z+ state can result in CH+formation in a small fraction of C+-H collisions 1. Only reactants in their ground states need be considered in interstellar clouds, since there is sufficient time available between collisions for even the longest lived of excited states to decay radiatively to the ground state. The temperature is also so low in interstellar clouds that any repulsive character in the potential curve for the low-lying excited state will prevent the reactants from approaching to internuclear distances sufficiently close for a radiative transition to take place. Given these restrictions and the relative abundances of atoms and atomic ions in diffuse clouds, the ionic radiative association reaction 303 Ion-Molecule Reactions in the Evolution of Simple Organic Molecules C+ + H-, CH+ + hv (1 2) appears to be the most important in these regions.1 Having now available three diatomic species, Hz, CH+,and OH, we can proceed to examine what bimolecular ion-molecule reactions will occur involving these species and what new molecules or ions will be formed.The rate of formation of Hz is by far the greatest among these three so that much more Hz will be available than CH+ or OH. Beginning with Hz, and examining all of the potential reactions of H+, He+, C+, N+, and O+ions with Hz, we find that H+, He+, and C+do not react rapidly with H2. Both O+and N+react rapidly with Hz to initiate a chain of reactions by successive hydrogen-atom abstractions from Hz to form Haof and NHs+ ions [reactions (13-(18)].Aside from H and He, the Hz molecule is gener-Of + Hz+ OH+ + H (1 3) kH20+ + H 12 + H N+ + Hz+ NH+ + H (1 4) kNHz+ + H (17) kNH3+ + H ally the next most abundant species in diffuse clouds. Even for densities as low as 10 moIecules cm-3, the fractional abundance of Hz is greater than that of 0 or N atoms14 (carbon is mainly ionic). The OH+, NH+, HzO+, and NHz+ ions do not react with H or He, so that reactions with Hz are the most important. The react ion NHs+ + Hz -P NH4+ + H (19) is thermodynamically allowed, but experiments have shown it to be inordinately slow.17 The abstraction chain (16)-(18) therefore does not proceed to complete H atom saturation as it does in the oxygen analogue, reactions (13)-(15).The OH radicals formed by reactions (8) and (9) probably react with the most abundant ion, C+,to produce CO+.The CO+ ion reacts further with Hz to produce HCO+, the ion which has been identified as the source of the ‘X-ogen’ line observed in interstellar clouds [reactions (20) and (21)].ls The CO+ion may also C+ + OH-, CO+ + H (20) kHCO+ + H react with H atoms by charge transfer to form neutral CO: W. T. Huntress, Jr. and V. G. Anicich, Astrophys. J., 1976, 208, 237. R.C.Woods, T. A. Dixon, R. J. Saykalley, and P. G. Szanto, Phys. Rev. Letters, 1975, 35, 1269. Huntress CO+ + H + H+ + CO (22) Carbon monoxide has been observed in abundance in diffuse interstellar clouds (Table 3).Table 3 Atomic and molecular abundances in c-Ophiuchii Abundances Atomlmolecule Observed C+ 2.7 x 10-4 2.3 x 10-4 N 9.5 x 10-5 5.8 x 10-5 C 6.9 x 7.6 x 10-6 co 2.0 x 10-6 2.0 x 10-6 OH 9.8 x 9.3 x 10-8 CH 6.5 x 10-8 7.1 x 10-8 CH+ 1.8 x 10-8 4.0 x 10-10 CN 1.7 x 1.1 x 10-8 H2O <4.5 x 10-9 2.0 x 10-9 c2 <1.0 x 10-8 2.6 x NH < 1.4 x 10-7 2.6 x 10-11 CH2 1.8 x 10-7 CH2+ 1.4 x 10-9 CH3+ 7.2 x 10-9 C2H 1.8 x 10-8 HCO+ 9.3 x 10-11 HzCO 1.0 x 10-10 HCN 8.0 x 10-10 NH2 3.5 x 10-14 NH3 1.1 x 10-14 Black and Dalgarno model (1H (1.w 332 0.79 0 5.6 x 10-4 0.79 4.6 x 10-4 The CH+ion formed by reaction (7)does not react rapidly with H or He, but does react with H2 [reactions (23) and (24)].The reaction to form CH4+ from CH+ + H2 3 CHz+ + H (23)kCH3+ + H CH3+ and H2 is endothermic, so that the chain terminates at the CH3+ ion. An alternative mechanism to forming hydrocarbon ions is via radiative association of C+with H2:l C+ + H2 + CH2+ + hv (25) A large rate constant has been theoretically predicted15 for reaction (25), but has Ion-Molecule Reactions in the Evolution of Simple Organic Molecules not yet been confirmed in the lab~ratory.~JS If as large a fraction as 10-6 of the collisions between C+ and H2 result in CH2+ formation, then reaction (25) rather than (7) is the more important mechanism for initiating hydrocarbon ion for- mation. A large rate constant for reaction (25) may also help to explain the large CH+/CH ratio observed in clouds where n(H2) > ~z(H)~O:for large fractional abundances of H2, CH+ is rapidly destroyed by reaction (23), and CH is produced from CH+ by reactions (23) and (24) followed by dissociative recombination of CH2+ and CH3+ ions with electrons as in reaction (26b) below.The additional quantities of CH+ required to match the observed CH+/CH ratio might be formed by reaction (25) followed by U.V. photodissociation of CH2+ ions [or CH3+ ions formed viu reactions (25) and (24)] to give CH+.The present status of the interstellar CH+/CH ratio problem has been reviewed recently by Dalgarno.21 The major loss mechanism for ions, and the major mechanism for forming neutral species from ions in diffuse clouds, is dissociative recombination with electrons [reactions (26)-(29)]. Once the density is sufficiently high for the CH3+ + e-+ CH2 + H (26a) CH + 2H +OH+H (28b) HCO++e--+CO+H (29) formation of diatomic species initiated by reactions (6)-(9), the successive processes of ionization, reaction, and recombination proceed rapidly to syn- thesize new polyatomic species.Dalgarno and Blackf have reviewed more extensively the chemistry of diffuse interstellar clouds and have constructed a model of a typical diffuse cloud, 5-Ophiuchii, based on this chemistry.22 Table 3 shows the molecules which have been observed in 5-Ophiuchii and the results of Dalgarno and Black’s model. The agreement is very good for those molecules which have been observed and for those for which observational upper limits are available. The model also predicts densities for molecular species for which observations have not yet been made.The fractional density of molecular species is very small in diffuse clouds because of the presence of the strong U.V. radiation field, which dissociates neutral as well as ionic molecular species. In more dense clouds, where the interior of the cloud is shielded from the galactic l9 F. C. Fehsenfeld, D. B. Dunkin, and E. E. Ferguson, Astrophys. J., 1974, 188, 43. 2o J. H. Black, A. Dalgarno, and M. Oppenheimer, Astrophys. J., 1975, 199, 633. *l A. Dalgarno, in ‘Atomic Processes and Applications’, ed. Burke and Moiseiwitsch, North Holland, Amsterdam, 1976.aa J. H. Black and A. Dalgarno, Astrophys. J. Suppl. Ser., 1977, 34, in press. Plate 1 Great Nebula in Orion. Luminous regions are thin, high-temperature regions. Darker condensations of denser material are visible against the luminous background (Copyright by the California Institute of Technology and Carnegie Institution of Washington. Reproduced by permission from the Hale Observatories) Plate 2 Results of an electrical discharge in a methane-ammonia gas mixture (Reproduced by permission from Zcarus, 1969, 10, 386) Plate 3 Imaging photopolarimeter colour photograph of Jupiter taken by the Pioneer 11 spacecraft from a distance of los km on 2 December 1974 I I I I’= 1011 COSMIC RAYS 1 -1012 -IIII 11 I$ 103 104 105 n(e ), electrons cm J Figure 4 Simplified representation of the ionic processes occurring at various altitudes in the atmosphere of Jupiter or Saturn.Solar photoionization and photodissociation occur in the upper atmosphere. Cosmic ray and electrical discharge ionization cause a second ionization peak at lower altitudes Huntress U.V. radiation, the fractional concentration of molecular species increases considerably. 4 Dense Interstellar Clouds For densities much larger than several hundred molecules ~m-~, most of the hydrogen in interstellar clouds is in molecular form and the cloud becomes increasingly shelf-shielded from the galactic U.V. flux. Cosmic ray ionization becomes the most important ionization mechanism.In the course of conden- sation through the diffuse stage a significant fraction of C+ has been converted into CO via reactions such as (20)--(22) and (29) so that n(C0) z lO-5-lO-* n(H2). Carbon monoxide is the next most abundant molecular species after H2. Other molecules are present in quantities of about 10-6-10-1°n(H2). Table 4 lists some of the molecules which have been observed by radio astronomers in clouds with densities of 103-106 molecules cm-3, New molecules are being Table 4 Molecular species observed in dense clouds. Estimated fractional abundances given by (x) = loz Diatomic Tetra-atomic Octa-atomic €32 (0) NH3 CH~CZCCN (-10) OH (-7) H2CO HCOOCH3 (-10) CH (-8) HzCS co (-4) HNCO Nona-atomic CN cs (-8) (-7) Penta-atomic CH~CHZOH CH30CH3 (-10) (-10) NS (-8) CHzNH so (-7) HCOOH SiO (-7) HCECCN SiS (-7) Hexa-atomic Triatomic CH30H HCO (-8) NHzCHO HCO+ HCN (-7) (-6) Hepta-atomic HNC (-6) CH3CHO N2Hf (-7) CH3NH2 C2H (-7) H2C=CHCN H2S so2 (-8) (-7) CH~CECH HCrC-C=CCN (-10) ocs (-8) added to the list constantly as observations proceed.Complex organic molecules are observed, some containing as many as nine atoms, and no doubt even more complex organic molecules will be discovered. Just over a decade ago it was beyond suggestion that such highly evolved organic molecules could ever exist in such cold, thin, isolated regions of interstellar space. It is becoming increasingly Ion-Molecule Reactions in the Evolution of Simple Organic Molecules obvious that ion-molecule reactions are responsible for the synthesis of a large number of them.The C+ ion remains a cornerstone in the formation of organic species in dense clouds, even though they are present in much smaller fractional abundance than in diffuse clouds.' The ionization rate is lower in dense clouds owing to the lack of U.V. ionization. Carbon is mainly in the form of atomic carbon and CO in dense clouds. However, the C+ ion is still one of the most abundant ionic species present. It is produced by the reaction He+ + CO -+ C+ + 0 + He (30) The He+ ions are produced by cosmic ray ionization of He. Since n(He)/n(Hz) FZ 0.15, reaction (30) represents a large fraction of the total ionization rate.The importance of reaction (30) results from the fact that the He+ ion does not react rapidly with H2. The reaction He+ + HZ+ H+ + H + He (31) occurs in less than 0.01 % of the collisions with Hz at 300 K,23 and is even slower at interstellar temperatures.17 Ionization of H2 by cosmic rays accounts for the major proportion of the ionization in dense clouds. For every H2 molecule ionized, an H3+ ion is produced via reaction (1). H3+ ions are rapidly lost via dissociative recombination with electrons, whereas C+ ions are lost only very slowly by radiative recombination. For this reason C+ ions are an important ionic species in dense as well as diffuse clouds. The function of H3+ ions in dense cIouds is to produce protonated species via reactions of the type (32).lJo For 0 atoms this reaction initiates the abstrac- H3+ + X -+ XH+ + Hz (32) tion chain to produce OH and HzO.For C atoms this reaction initiates the abstraction chain to produce the CH3+ ion and CH and CH2 via reaction (26).For N atoms, reaction (32) is not thermodynamically allowed, and an alternative reaction channel (33) may initiate the abstraction chain terminating in NH3+ via H3+ + N + NH2+ + H (33) reaction (18). The NH3+ ion does not abstract an H atom rapidly from H2 at low tempera- tures. Abstraction is possible, however, from OH and H20:10J7 NH3+ + OH -+ NH4+ + 0 (34) NH3+ + H2O -,NH4+ + OH (35) Ammonia can then be synthesized by the dissociative recombination of NH4+ ions with electrons : NH4+ + e--+NH3 + H (36) as R.Johnsen and M.A.Biondi, Icarus, 1974, 23, 139. Huntress Alternatively, NH3 may be synthesized directly from NH3+ by charge transfer to a metal atom such as Mg, Si, Na, Al, or Fe: NH3+ + Mg -+ Mg+ + NH3 (37) Because OH, H20, and the metals are so much less abundant than H2, the NH3/H20 ratio is expected to be small, of the order of 10-3, even though the cosmic N/O ratio is about 0.5. This low NH3/H20 ratio has been confirmed in the Orion nebula,24 which provides support for the gas-phase ion-molecule reaction theory for interstellar molecular synthesis. The temperature must be of the order of 200 K or more in order for reaction (19) to be sufficiently fast for the NH3/H20 ratio to approach the cosmic N/O value.Hydrogen cyanide is another commonly observed molecule in dense clouds. One method of synthesis is shown in reactions (38) and (39). However, in the C+ + NH3 + HKN+ + H (38) HKN+ + e--+ HCN + H (or CNH + H) (39a) -+ CN + 2H (39b) absence of effective NH3 synthesis, more likely reactions are14 and Cf + NH2 + HCN+ + H (40) followed by CH3+ + N -+ HCN+ + H2 HCNf + H2 + HzCN+ + He (41) (42) and reaction (39). The H2CN+ ion most likely has the structure HCNH+ so that the observation of both HCN and HNC in interstellar clouds with near equal densities fits nicely with the theory of production by ion-molecule reactions. The HCO+ ion is isoelectronic to HCN and is produced by a number of reactions (43)-(46)1J4 in addition to reactions (20) and (21).Reaction (46) C+ + H2O + HCO+ + H (43) H3+ + CO -+ HCO+ + H2 (44) CH3+ + 0 -+ HCO+ + H2 (45) CH + 0 + HCO+ + e-(46) is a chemi-ionization reaction between neutral radicals, and may be fast even at interstellar temperatures. The carbon-oxygen family of observed interstellar molecules includes CO, CH20, and CH30H. Carbon monoxide is synthesized via dissociative recombination of HCOf ions with electrons [reaction (29)]. Methanol is most likely synthesized via radiative association of CH3+ ions with H2O [reaction (3a)],9 followed by dissociative recombination. Ion-molecule I' J. W. Waters, J. J. Gustincic, J. J. Kakar, T. B. H. Kuiper, H. K. Roscoe, P. N. Swanson, A. R. Kerr, and P. Thaddeus, Astrophys.J., 1977, in press. Ion-Molecule Reactions in the Evolution of Simpie Organic Molecules schemes to produce formaldehyde have been difficult to formulate, and present theorye5 favours the neutral reaction CH3 + 0 + CHzO + H (47) Neutral radical-radical recombination reactions can have very small activation energies and may possibly occur at interstellar temperatures. The difficulty with reaction (47) is the problem of synthesizing CH3 by a low-probability process, i.e, radiative recombination of CH3+with electrons in competition with reaction (26). A more likely process would seem to be the radical-radical reaction (48) where CHz + OH -+ CH2O + H (48) CH2 is derived from reaction (26a) and OH from reaction (28b).A possible scheme for formaldehyde production involving ion-molecule reactions is CH3+ + OH + CH20+ + Hz (49) CH20’ + Mg + Mg+ + CH20 (50) A variant of this scheme suggested by Dalgarno et aL25 uses reaction (51) CH3+ + 0 + CHzO+ + H (51) but recent laboratory work26JO indicates that the product of the CH3+-0 reaction is most likely HCO+ + HZand not CHzOf + H. Radiative association of HCO+ and HZ has also been suggested [reactions (52) and (53)],27 but HCO+ + HB-+ CHz0H.t + hv (52) CHzO + H (53) laboratory work19 indicates that the structure of intermediate complex, in collisional association reactions at least, is probably a cluster ion HCO+.Hz and not protonated formaldehyde. The ethynyl radical observed in dense clouds is readily explained from ion chemistry in the hydrocarbon family.C-C bonds are synthesized by the reactions14 Cf + CH + Cz+ + H (54) CH3+ + C + C2H+ + Hz followed by C2+ + H2 -+ CzH+ + H CzH+ + Hz --+ C2Hz+ + H (58) a5 A. Dalgarno, M. Oppenheimer, and J. H. Black, Nature Phys. Sci.,1973, 245, 100. 26 F. C. Fehsenfeld, Astrophys. J., 1976, 209, 638. 27 E. Herbst and W. Klemperer, Astrophys. J., 1973, 185, 505. 310 1: Huntress The abstraction chain from H2 stops at CzHz+. Dissociative recombination with electrons produces C2 and C2H from C2H2+: C2H2+ + e--+C2H + H (59d + C2 + 2H (59b) Gas-phase chemical reaction schemes have been propo~ed~~~~~*~ for the synthesis of a few of the other species listed in Table 4, such as SO, SN, CS, OCS, H20, HzCS, N2H+, SiO, and SiS, but no schemes have yet been put forward for the synthesis of most of the other more complex species listed in Table 4.Smith and Adamsg have suggested that radiative association of CH3+ ions with HzO, NH3, CH20, CH30H, and CH3NH2 can lead to production of the protonated ions of methanol, methylamine, acetaldehyde, ethanol (or dimethyl ether), and dimethylamine, respectively. The parent neutrals could then be formed from the protonated ions by dissociative recombination with electrons. Except for the reaction with H20, fast bimolecular reactions occur between CH3+ and the other neutrals listed by Smith and Adams. The importance of radiative association versos bimolecular reaction depends on the temperature dependence of the life- time of the intermediate complex.Methyl cations do not undergo bimolecular reactions with CO or HCN, and radiative association in these collisions can lead to synthesis of ketene and acetonitrile as in reactions (60)-(63). A number of CH3+ + CO -+ CH3CO+ + hv (60) kCH2CO + H (61) CH3+ + HCN + CH3CNH+ + h~ (62) 1:CH3CN + H (63) such reactions can be envisioned to synthesize many of the species given in Table 4. For an astrochemist, the prospect of constructing mechanisms for the synthesis of interstellar molecules and testing them both in the laboratory and against observations is an exciting one. An interdisciplinary approach is required since hints can sometimes be gleaned from observations, and in the course of constructing synthetic mechanisms from laboratory work or theoretical work the astrochemist can often suggest confirming observations to the radio astronomer.5 Planetary Atmospheres A. Origin and Evolution of Planetary Atmospheres.-As the more massive of dense interstellar clouds proceed on a course of self-gravitating collapse, both obser- vation and theory suggest that the central densities and temperature can get sufficiently high to initiate nuclear reaction, resulting in star formation. There do appear to be infrared-emitting objects located at the central point of some dense 28 J. L. Turner and A. Dalgarno, Astrophys. J., 1977, 213, 386. 311 Ion-Molecule Reactions in the Evolution of Simple Organic Molecules clouds, such as Orion A.These objects may be proto-stars. At densities n > 106 molecules ~m-~, and at elevated temperatures of the order of several hundred degrees, reactions among neutral species become ever more important. Gradually, as the proto-star becomes a new sun and the initiating nebula builds planets around the central sun, all of the gaseous material in the nebula is converted into stable, reduced forms. Because of the overwhelming relative abundance of hydrogen, carbon is converted almost entirely into CH4, nitrogen into NH3, and oxygen into H2O. It is curious to note that in the highly reducing environment of dense interstellar clouds a large percentage of the carbon is in an oxidized form, CO, but that at the higher temperatures and densities of proto-stellar nebulae, CH4 is probably the major reservoir of carbon.This difference may be due to the lower temperature and much higher fractional ionization in interstellar clouds compared with proto-solar nebulae and planetary atmospheres. In the above scenario, large massive planets will develop atmospheres contain- ing molecular hydrogen with about 15% helium and very much smaller per- centages of CH4, NH3, and H2O according to the cosmic abundances of C, N, and 0.Jupiter is a primary example of such a body in our own solar system. Jupiter is a gas giant composed mostly of Hz and He.29Methane and ammonia have been detected in Jupiter’s atmosphere in roughly their equivalent cosmic abundances. Whatever rocky core Jupiter may possess is small compared with its own deep atmosphere.The planet is almost entirely gaseous because of the very low cosmic abundance of the refractory, solid-forming elements such as Fey Mg, Al, and Si. The radius of a Jovian rocky core could be no more than one-fifth of the total radius of the planet. The pressure at the bottom of the Jovian atmosphere is of the order of 108 atmospheres, and the temperature is of the order of 20 OOO K. Under these conditions, hydrogen is completely dissociated and probably exists in metallic form. Saturn is very similar to Jupiter, and Uranus and Neptune are both somewhat smaller examples of the same class of gas giant. Together, these four planets are referred to as the Jovian family of planets. The relative sizes of the planets of our solar system are shown in Figure 2, and Table 5 gives what is known of their atmospheric composition.Table 5 Atmospheric Compositions Venus Earth Mars Jupiter Major species c02 N2 02 c02 H2 He Minor species H2O 02 Ar CO2 N2 Ar CH4 NH3 co €320 H2O H2O 03 03 02 C2H2 HCI co co C2H6 HF NzO 03 co PH3 ** A source book on Jupiter, the major member of the Jovian family of planets, is edited byT. Gehrels: ‘Jupiter’, University of Arizona Press, Tucson, 1976. Huntress c-n0W z E W I--CL z) d W70 > ws il.I Ion-Molecule Reactions in the Evolution of Simple Organic Molecules The smaller terrestrial planets, Mercury, Venus, Earth, and Mars, are examples of planets which were too small in their formative stages to accrete massive atmospheres from the surrounding solar nebula.The dividing line between gas giant and terrestrial-type planet is defined by whether or not hydrogen atoms can escape from the upper atmosphere by thermal excitation. The gas giants are so massive that H atoms cannot escape from their atmospheres in times approaching that of the age of the solar system. The gas giants are also distant from the sun and their atmospheres are cold. The terrestrial planets are closer to the sun, and their atmospheres can get much warmer. The terrestrial planets are also quite small. Hydrogen atoms take only several times 104 years to escape from the upper atmosphere of the Earth.Thus, through the continuing process of loss of hydrogen to space, the atmospheres of the terrestrial planets have become more oxidizing in nat~re.~O-33 Furthermore, any original atmospheres the terrestrial planets may have accreted in their formative stages must have subsequently been completely lost very early in their history. The present atmospheres of the terrestrial planets can only be of secondary origin; that is they can only have been formed by outgassing of the interior. The evidence for loss of the primordial atmosphere of the Earth comes from the observation that the so-called ‘rare’ gases, Ne, Ar, Kr, and Xe, are in fact rare in the Earth’s atmo~phere.~~ The same situation holds true for Mars’s atmosphere.Neon and the natural isotopes of Ar, Kr, and Xe are simply not present in the atmospheres of Earth and Mars in their cosmic proportions; they are depleted by factors of the order of lo7.Helium readily escapes from the atmosphere of Earth, but not the heavier noble gases. Where did all these gases go? The geological record shows that the Earth’s interior is strongly degassed owing to tectonic activity. Therefore, the missing quantities cannot be trapped in the interior. The only remaining explanation is that the entire primordial atmos- pheres of Earth and Mars (Venus and Mercury probably as well), if indeed there were any primordial atmospheres, must have been entirely lost at some early catastrophic epoch in their formative stages.Jupiter, on the other hand, has clearly retained all of its initial atmosphere since hydrogen and helium are present in roughly their cosmic proportions. Jupiter at least among the gas giants is a fossil relic of the nebula which condensed to form the sun. Jupiter’s atmospheric composition has probably not changed since the time of formation of the planets. Some clues as to the composition of the primordial atmosphere outgassed from the primitive terrestrial planets can be gained from theoretical studies and by examining the composition of the gases evolved by heating meteorite^.^' Certain classes of meteorite, such as the chondrites, are believed to be relics left over from a swarm of particles out of which the solid bodies of the terrestrial planets were formed.Table 6 shows the composition of the gas released from a F. P. Fanale, Chem. Geol., 1971, 8, 79. 31 F. P. Fanale, Zcarus, 1971, 15, 279. 32 A. J. Meadows, Planet. Space Sci., 1973, 21, 1467. 33 L. Margolis and J. E. Lovelock, Zcarus, 1974, 21, 471. Huntress Table 6 Primordial Rock Gas Composition Released above 1050 K Normal chondrite Earth basalt Equilibrium theory c02 64 50 55 Ha 26 40 36 co 4 3 6 N2 2 2 3 CH4 4 2 - H2O - 1000 1500 normal chondritic meteorite and from a primitive Earth basalt (primitive un- modified volcanic rock) when heated to melting. The third column in Table 6 shows what would be expected on the basis of chemical equilibrium between rock melt and gas at 1050 K.A large quantity of hydrogen is present in addition to C02 and H20, and on release of this hot gas from the interior of the Earth to the surface the gas cools rapidly where the chemical equilibrium (64) shifts strongly to methane formation, especially with the condensation of H2O to form oceans. The initial, secondary atmosphere of the Earth was therefore probably reducing in nature and contained carbon largely as CH4. Methane would then have been slowly oxidized back to C02 in the Earth’s primitive atmosphere by thermal processes, by solar photolysis, and by other energetic mechanisms as the equilibrium is shifted back to C02 by the rapid escape of H2 from the atmosphere: CH4 + 2H20 + C02 + 4H2 t (65) The difference between Mars and Earth is that Mars, being a much smaller planet, lost its reducing atmosphere much more rapidly than did the Earth31 and did not (probably because of the loss of a reducing atmosphere) develop life on its surface.The present O2/N2 atmosphere is due to the action of biology on the Earth’s surface and in the 0ceans,3~9~3 and the Earth’s primitive C02 atmosphere has been completely dissolved and deposited by the Earth’s oceans. The difference between Earth and Venus is that Venus, being much closer to the sun, accreted out of higher-temperature material that contained very little water of crystallization.2 Hence, very little H2O was outgassed from Venus’s interior. Therefore, ocean formation was not possible and the atmosphere remained forever as C02.If all of the C02 presently dissolved in the Earth’s oceans and deposited in the Earth’s sedimentary formations (carbonates) were to be released into the atmosphere, the Earth would have almost the exact equivalent of Venus’s atmosphere; almost pure C02 at a surface pressure of 90 atm!30*31 For the abiogenic synthesis of organic compounds, a reducing environment is req~ired.~The oxidizing atmospheres of the present-day terrestrial planets are incapable of such synthesis. The primitive methane atmosphere of the Earth probably was, and the present-day atmospheres of the gas giants almost certainly are, sites where the abiological synthesis of organic compounds from simple hydrides can and did take place. The classic experiment was conducted by Stanley Ion-Molecule Reactions in the Evolution of Simple Organic Molecules Miller34 in the early 1950's.Miller showed that simple amino-acids were produced in electrical discharges through mixtures of gaseous CH4, NH3, and H2O. These are exactly the same starting materials available in the atmospheres of the gas giants, and probably in the Earth's primitive atmosphere as well. Plate 2 shows the results of an electrical discharge in a mixture of methane and ammonia. A red tarry substance condenses on the wall of the reaction vesseI, which on hydrolysis yields amino-acids. The intermediate products have been identified as HCN, a-aminonitriles, CH20, and other aldehyde~~J4 Curiously enough, similar molecules are also observed in dense interstellar clouds! The coloration of the material in Plate 2 is tantalizingly close to the coloration of the cloud bands in the Jovian atmosphere shown in Plate 3.It is tempting to ascribe the Jovian colorations to the same processes which produce the colours in the experiments illustrated in Plate 2. Gas-phase production HCN, HCHO, a-Amino nitriles, CHd, NHJ, HzO ionizing radiation Aldehydes, Unsaturated hydrocarbons + Amino-acids Solution HCHO, Aldehydes Sugars Figure 3 Generalized scheme for the abiogenic synthesis of biologically important com- pounds, starting with the simple hydrides in the gas phase The general scheme of prebiotic, or abiogenic, synthesis of organic compounds is illustrated in Figure 3. From simple hydrides in the gas phase, energetic processes such as solar u.v., cosmic rays (ionizing radiation) and discharges produce simple organic precursors such as HCN, CH20, or-aminonitriles, aldehydes, and unsaturated hydrocarbons. These species are captured in aqueous solutions (oceans in the case of the primitive Earth, water droplets in a massive planet-wide cloud deck in the case of Jupiter) where they undergo hydrolysis and condensation to form amino-acids, purines, and pyrimidine bases and sugars.*'S. L. Miller, Science, 1953, 117, 528. Huntress These latter are the basic building blocks for formation of peptides, nucleosides, nucleotides, and various other complex organic material required for the initiation and sustenance of organic chemical life.The theory of chemical evolution leading towards the origin of life has been discussed in detail in a number of review books35 and articles.5 In the following sections, an examination is made of what is presently known about the potential contributions that ionic processes make to the overall scheme of gas-phase organic synthesis in both the prebiotic synthesis experiments and in their natural counterpart in the atmos- pheres of the Jovian planets. B. Ionospheric Chemistry.-The ionosphere is that upper region of the atmosphere where the density is very low (S1014 molecules cm-3 or S10-5 atm) and the species present are ionized by the solar U.V. flux. In general, photochemical products are present also, formed by solar photodissociation of atmospheric species below the ionosphere and having diffused up to more tenuous ionospheric heights.In the case of the gas giants,2129 atomic hydrogen is the most abundant photochemical product present in the ionosphere. Some methane is also present in the lower part of the ionospheres of the gas giants, but no NH3 or H20. These latter gases are condensed out of the atmosphere into cloud decks at much lower altitudes. Figure 4* provides a simplified representation of the processes occurring at various levels in the Jovian atmosphere. The chemistry of the ionospheres of the Jovian planets is then the ion chemistry in a mixture principally of H, Hz, and He with a small percentage of methane near the lower boundary.36 Without NHs or HzO, the ionospheric chemistry is not effective in producing organic species except for minor amounts of hydrocarbons.The ionospheres of the outer planets are warmer (N80-1 50 K) and denser than interstellar clouds so that the reaction of He+ ions with Hz [reaction (31)], albeit slow, is an efficient He+ ion loss mechanism. Solar ionization of H, He, and H2 then results in H+ and H3+ being the most important ions, and H3+ is rapidly lost by recom- bination with electrons to produce H atoms. At lower altitudes where CH4 is present, reactions (66) and (67) convert hydrogenic ions into hydrocarbon ions, H+ + CH4 + CH3+ + Hz (66a) + CH4f + H (66b) H3+ + CH4+ CH5+ + Hz (67) followed by reactions (68) and (69). Most of the ionization of CH4 is accomplished CH3+ + CH4 CzHs+ + Hz (68) CH4+ + CH4 --* CH5+ + CH3 (69) * Opposite p.307 36 A source book on these processes is edited by S. L. Fox: ‘The Origin of PrebiologicalSystems’, Academic Press, New York, 1965. 36 W. T. Huntress, Jr., Adv. Atom. Mol. Phys., 1974, 10, 295. Ion-Molecule Reactions in the Evolution of Simple Organic Molecules by reactions (66) and (67) since hydrogen is so abundant, but solar photoioniz- ation of CH4 and its photochemical products does contribute. Neither CH5+ nor C2H5' reacts with H2 or CH4, but at the lowest altitudes, where the pressures are highest, three-body collisional association reactions, e.g. (70)-(74), can become H2 CH3+ + H2 -+ CHsf H2 CH5+ + CH4 --+ CzHg+ H,C&5' + HZ -+ C2H7' H,+ CH4 -+ C3H9.+ (73) CzH7' H2+ CH4 -+ C3Hii' (74) important.These and higher-order clustering reactions have been used in model calculations of the chemistry in the ionospheres of the Jovian planets.8 These reactions serve to produce hydrocarbons by subsequent dissociative recombi- nation with electrons, such as reactions (75) and (76). Ethane and acetylene have CZH5' + e-+ C2H4 + H (75a) 4C2Hz + H + H (75W C2H7' + e--+ C2H6 + H (76) been identified with the atmosphere of Jupiter.29 At lower altitudes the atmosphere is shielded from solar ionizing photons. Ionization can still occur at these lower Ievels owing to the penetration of cosmic rays. Model calculations of the atmospheres of Saturn, Uranus, and Neptune show that ionization by cosmic rays peaks near the 1 atm level and can produce number densities of ions and electrons ( N lo4~m-~)nearly equal to that produced by solar photoionization hundreds of kilometres higher in the ionosphere.8 The base of the water cloud in the atmosphere of Jupiter is near the 1 atm level.Thus, cosmic ray ionization can provide the activation energy necessary to initiate ionic reactions leading to the synthesis of new organic species in the regions of the atmosphere within and below the cloud layers in the Jovian pl mets where NH3 and H2O exist. C. Ion Chemistry in the Lower Atmosphere of the Jovian Planets.-In addition to cosmic ray ionization, discharges in the turbulent cloudy regions in the lower atmospheres of the Jovian planets during thunderstorms can also contribute to primary ionization.An electrical discharge was the energy source used in the classic Miller experiment,34 and electrical discharges remain one of the primary energy sources advocated for the prebiotic synthesis of organic compounds in the atmosphere of the primitive Earth.5J5 Earth-based and spacecraft photographs Huntress of Jupiter’s cloud bands show the atmosphere to be extremely turbulent.29 Swirling vortices are clearly evident, and convection occurs on a large scale with giant plumes reaching high into the atmosphere. Under these conditions there is every reason to believe that thunderstorm activity can be pre~alent.3~ Laboratory experiments clearly show the formation of the organic compounds necessary to begin and sustain life during electrical discharges under the conditions present at these levels in the Jovian atmo~phere.~*~5 Since ion formation is a necessary part of electrical discharges, ion chemistry in discharges may play an important role in the synthesis of organic compounds both in the laboratory experiments and in the present atmospheres of the Jovian planets.At pressures appropriate to the cloudy regions in the Jovian atmosphere, 0.1 atm and above, three-body collisional association reactions become very important. Even if a bimolecular ion-molecule reaction is possible, the colli- sional association reaction can often dominate. Methyl cations are among the most important of ionic species produced by ionization of H2, CH4, NH3, and H2O since they are formed in almost half of the methane ionization events, and since they are a very good condensation reagent.10 There are no bimolecular reactions between CH3+ ions and Hs, so that the ion is stable in H2 and can react with the trace species present.At high pressures, collisional association of CH3+ ions with H2 can form CH5+ in a hydrogen atmosphere. However, the rate constant of the cH3f-H~ association reaction is a hundred times smaller than, for example, the rate constant for the CH3+-H20 rea~tion.~ This relationship probably also holds true for association of CH3+ ions with any polar molecule in relation to Ha, so that even in a hydrogen-dominated atmosphere association reactions of CH3+ ions can still be important.Also, the structure of the CH5+ ion involves a three-centre bond so that the ion looks very much like a CH3+-H2 complex (1). It is possible, therefore, that in collisional association reactions the CH5+ ion may react very much like a CH3+ ion via attendant H2 elimination, although this premise remains to be examined in the laboratory. In the discussion to follow we will emphasize the importance of association reactions at atmos- pheric pressures for organic synthesis. Attention was first called to this point by Hiraoka and Kebarle38 with regard to alcohol and organic acid formation. Here we expand on this notion to illustrate further the utility of the mechanism for abiogenic synthesis of organic species by ionic reactions at high pressures.It is important to emphasize that at this point most of the mechanisms suggested are speculative, 37 A. Bar-Nun, Icarus, 1975, 24, 86. K. Hiraoka and P. Kebarle, J. Anter. Chem. Soc., 1977,99, 360,366. Ion-Molecule Reactions in the Evolution of Simple Organic Molecules An excellent example of competition between bimolecular reaction and association reaction is found in the case of CH3+ ions reacting with NH3 [reaction (76)]. At low pressures (< atm) only the CHZNHZ+ ion product is observed + H*CHpNH2+ CHINHI+ CH,' + [CHINHIf]* from this reactionlo (another reaction channel forming NH4+ ion is also observed but is unimportant for the discussion). Recombination of CHZNHZ+ ions with electrons can then lead to methanimine or HCN formation: Methanimine is not a particularly stable species and can decompose readily to HCN.[At low interstellar temperatures, however, methanimine is clearly stable since the molecule has been observed in these regions by radio astronomers. The reaction sequence (76a)-(77) may account for its presence.] The reaction sequence (76a)-(78) could be responsible for HCN formation in the Jovian lower atmosphere, but at these high pressures the collisional association reaction (76b) can also become important. The atmospheric pressure at which reaction (76b) competes effectively with (76a) can be calculated by equating the bimolecular and three-body rates, or [Hz] 2 K~~/K76b.Using K76s z cm3 s-l, K76b z 10-26 cm6s-1, and Loschmidt's number (at 300 K), we find [Hz] 2 0.003 atm. so that the collisional association reaction can compete even at fairly low pressures on the atmospheric scale.Reaction (76b) has been observed at pressures as low as 0.2 Torr in helium by Smith and ad am^.^ Dissociative recombination of the product ion with electrons can then yield methylamine: CH3NH3+ + e--+CH~NHZ+ H (79) Methanol can likewise be produced by the collisional association of CH3+ and H2O followed by recombination with electrons : H* CH3+ + HzO 3 CH30Hz+ kCH3OH + H The proton affinity of ammonia is very high and, since ammonia is abundant in Huntress the Jovian atmosphere, proton transfer to NH3 is a viable alternative to recom- bination for formation of neutral from protonated species.For example, CH30H2+ + NH3 4NH4+ + CH30H (81) Collisional association in the Jovian atmosphere leads to synthesis of the same products as radiative association in interstellar clouds. Like HCN, formaldehyde is an important precursor for synthesis of higher-order organic compounds in prebiotic synthesis schemes. Formaldehyde may be synthesized in discharges by (3a) followed by dissociative recombination of C&OH2+ with electrons, CH30H2+ + e-+ CH2O + H2 + H (8W by the sequence of reactions (82), (53), (83),lO CH3+ + CH30H -+ CH20H+ + CH4 (82) CH20H+ + e--+ CH2O + e-(53) + NH3 + CH2O + NH4+ (83) or by the neutral reactions (47)and (48). Carbon monoxide could likewise be produced from formaldehyde by the series of neutral reactions H + CHzO + HCO + H2 (84) H + HCO 4 CO + H2 (85) Carbon monoxide has been detected in the Jovian atmosphere39 at a mixing ratio CO/H2 z 10-9.The CO molecule cannot exist in equilibrium at low temperatures (< 1100 K) in the highly reducing environment of the Jovian atmosphere, and must be produced by non-equilibrium processes such as those just described. In addition to HCN, CH20, CH3NH2, and CH30H, other more complex organic species can be synthesized by methyl cation association reactions in the lower atmospheres of the Jovian planets. Nitriles can be synthesized by conden- sation with HCN [reactions (86)-(88)]. Aldehydes can be synthesized by HI CH3+ + HCN 4CH3CNH+ (86) +CH3CN e- + H (87) +CH3CN + NH4+ NHs (88) condensation with CO [reactions (89)--(92)] or by condensation with CH2O H,CH3+ + CO 4 CH3CO+ ?@ R.Beer, Astrophys.J., 1975, 200, L167. Ion-Molecule Reactions in the Evolution of Simple Organic Molecules e-+CH3CHO + H (91) NHs [reaction (93)], followed by reaction (97) or (92). The methyl cation does not HI CH3+ + CH2O -+ CH3CH=OH+ (93) exhibit bimolecular reactions with HCN or CO so that the high-pressure associ- ation reactions are the only available pathways for these reactions in the Jovian atmosphere. Likewise in interstellar clouds, radiative association of CH3f with HCN and CO is possible to yield nitriles and aldehydes. There is a bimolecular reaction between CH3+ and CH20, CH3+ + CH2O + HCO+ + CH4 (94) but at high pressures in the Jovian atmosphere reaction (93) might possibly contribute.Aldehyde formation via radiative association of CH3+ and CH20 in interstellar clouds may be hindered because of the existence of the bimolecular reaction channel. Other alkyl cations besides CH3+ should react in a similar manner to produce more complex alcohols, amines, nitriles, and aldehydes. Hiraoka and Kebarle38 have demonstrated the equivalent of reaction (3b) for C2H5+, i-C3H7+, and t-C4Hg+ ions. The C2H5+ ions can be produced in the Jovian atmosphere by reaction (68). These workers have also demonstrated gas-phase carbonylation via the equivalent of the solution-phase Koch-Haaf synthesis for organic acids using C2H5+ and i-C3H7+ ions.The equivalent for CH3+ ions produces acetic acid [reactions (89) and (95)-(97)l. Hiraoka and Kebarle38 also demonstrated the H,CH3+ + CO + CHsCO+ (89) H,CH3CO+ + H2O + CH3COOH2+ (95)I e-+CH3COOH + H (96)lNH1+CH3COOH + NH4+ (97) formation of protonated formic acid by a similar condensation of HCO+ ions with H2O. The HCO+ ions could be formed in the Jovian atmosphere by reactions (44) and (94). The or-aminonitriles have been identified as important precursors along with HCN and the aldehydes in prebiotic synthesis experiments using electrical dis~harges.~These compounds may possibly be produced by ionic association reactions such as (98) and (99), where CH2NH2+ ions are produced by reaction Huntress (76a).Subsequent hydrolysis in the liquid phase then yields amino-acids, glycine CH2NH2+ + HCN + NH2CH2CNH+ (98) lLNHzCHzCN (99) in the above case. It seems likely that ion-molecule association reactions could be responsible for the formation of a large number of the simple organic com- pounds produced in experiments on prebiotic organic synthesis, and that these processes are now taking place in the atmospheres of Jupiter and the other gas giant members of the solar family of planets. 6 Conclusion Ion chemistry is ubiquitous in the universe, and permeates the environment in regions as tenuous as the thin wisps of matter that make up interstellar clouds and in regions as dense as the deep, turbulent atmosphere of the Jovian planets. Gas-phase ion-molecule reactions are probably responsible for the synthesis of most of the molecules observed by radio astronomers in interstellar clouds, and probably contribute greatly to the synthesis of simple organic compounds in the atmospheres of the Jovian planets.Ion-molecule reactions may also have participated in the initial stages of chemical evolution, leading from the simplest hydrides in the gas phase to large organic molecules in solution, and thence possibly to chemical life itself. 5 323

ISSN:0306-0012    

DOI:10.1039/CS9770600295

出版商:RSC    

年代:1977

数据来源: RSC

摘要:

I(ELVIN LECTURE Across the Living Barrier By David E. Fenton DEPARTMENT OF CHEMISTRY, THE UNIVERSITY, SHEFFIELD, S3 7HF The purpose of the Kelvin Lectureship of the British Association for the Advancement of Science is ‘to convey in non-specialist language to intelligent people who are not experts’ the subject area under discussion. In order to retain the essence of the lecture only key references are given as it is hoped that this article, closely based on the above lecture, might act as a preface to further study. 1 Introduction The essential metals of the body may be considered in two categories; bulk metals and trace metals. The bulk metals (sodium, potassium, magnesium and calcium) constitute about 1 % of human body weight, whereas the trace metals (manganese, iron, copper, cobalt, zinc, molybdenum, vanadium and chromium) represent less than 0.01% of the same.A 70 Kg man has approximately 170 g of potassium present in his body of which about 9 g are in the blood and 3 g in the tissue fluid; the same man requires only about 5 g of iron. Paradoxically it is the trace metals, of transition group origin, that have been studied most during the recent upsurge of interest in the involvement of metal ions in biological processes. This is because, due to the absence of d orbital electrons and suitable spin states, the bulk metals have few spectroscopic properties that may be studied and so have remained ‘out of sight, out of mind’,l despite their essentiality. They have no unpaired electrons and so cannot be studied by magnetic measurements or electron spin resonance spectroscopy, but it is possible to study sodium complexes directly by 23Na nuclear magnetic resonance spectroscopy.In general it is the simplicity of the alkali metal cations that makes them difficult to study. The alkali metals provide highly mobile, unipositive cations and usually form weak complexes with ligands having ‘hard’ donor atoms such as oxygem2 They are concerned in vivo with the maintenance of normal water balance and distribution, the conduction of nerve impulses, the maintenance of neuromuscular activity and potassium helps the heart relax between beats.3 These roles are related to the ionic character and mobility of the cations. Sodium and potassium are also required to activate l R.J. P. Williams, Nature, 1974, 248, 302. * R. J. P. Williams, Quarr. Rev., 1970, 24, 331. P. B. Chock and E. 0. Titus, Prugr. in Znorg. Chenr., 1974, 16, 287. Across the Living Barrier certain enzymes, but the mechanisms by which they activate enzyme catalysed reactions are not clearly defined.4 A closer look at one pair of bulk metals, the alkali metals sodium (Na) and potassium (K) reveals a pronounced difference in the location of these metals. Naf is the principal extra-cellular cation, and K+ is the principal intra-cellular cation. The relative concentrations of these ions are well illustrated for mam- malian blood cells where for the blood plasma the levels are 5 mM Kg-l (K+) and 143 mM Kg-l (Na+) compared with levels of 105 mM Kg-1 (K+) and 10mM Kg-l (Na+) for the red blood cells.5 There is, therefore, a discriminatory mechanism which controIs the selective uptake of K+ in the cell from its bathing fluid.The cell is surrounded by a membrane which separates its aqueous interior from the bathing fluids and selectively allows into the cell ions and nutrients, and allows out unwanted material, or material produced for use elsewhere. The membrane functions as a ‘living barrier’,5 and it is in this barrier, it is believed, that the process of cationic discrimination occurs. The membrane is about 70 8, thick and is composed, in general, of proteins and lipids. These may vary qualitatively from membrane to membrane, and although the building blocks for the membrane are known, the precise archi- tecture is not known.Early work with electron micrographs suggested a tri- lamellar structure formed from two electron dense layers (protein) separated by an electron lucent layer (lipid).* This view persisted for many years and the lipid has been shown to exist as a bilayer with the polar head groups of the lipids bedding into the protein, and the long hydrophobic, hydrocarbon tails giving it a central zone of low dielectric constant. The alignment of these chains has been confirmed by X-ray crystallographic studies on the bilayer formed by 1,2-dilauroyl-(k)-phosphatidyl ethan~lamine.~ A more recent approach views the membrane as consisting of a lipid bilayer in which globular proteins float ‘as icebergs’. This model is called the Fluid Mosaic Model,8 (see Figures 1 and 2). To reach the inside of the cell the alkali metals must traverse this barrier and in so doing must convey their charge across a medium of low dielectric constant.This is an unfavoured step because of the high electrostatic energy required to transfer an ion from the high dielectric aqueous bathing fluid into the low dielectric hydrocarbon. In order to overcome this it has been proposed that the cations move across in association with organic molecules, either individually (a carrier), by relays of carriers, or by associated pores present in the membrane (see Figure 3). The carrier is postulated as encapsulating the cation and thus * C.H. Suelter, Science, 1970, 168, 789. R. Levin, ‘The Living Barrier’, Heinemann, London, 1969. R. Harrison and G. C. Lunt, ‘Biological Membranes’, Blackie, Glasgow and London, 1975. P. B. Hitchcock, R. Mason, K. M. Thomas, and G. Shipley, Proc. Nut. Acad. Sci. U.S.A., 1974, 71, 3036.* S. J. Singer and G. L. Nicolson, Science, 1972, 175, 720. Fenton PROTEl Fc LIPID PROTEIN Figure 1 The trilamellar structure of the membrane effectively disguising it by presenting an organic, lipid soluble surface to the membrane. In this form the complex containing the alkali cation may cross the barrier either individually, by passing the cation from carrier to carrier, or by passing the cation from donor site to donor site in a pore.It is therefore necessary to know more about the nature of such systems, and their mode of operation. This discussion is concerned with the carrier-assisted passive transfer of material independent of energy source-the so called facilitated diffusion. It is also possible for the cell to accumulate cations by working against concentration Across the Living Barrier -B C D E protein __O anionic carrier - 2 - phosphatide neutral carrier 7 cholesterol g oxygen 0 cation Figure 3 Some modes of ion transport (A, Bare cation permeation; B, Negatively charged pore; C, Anionic carrier molecule; D, Neutral pore; E, Neutral carrier molecule) gradients. This process requires energy and is called active tran~port.~~~ The hydrolysis of adenosine triphosphate (ATP) to adenosine diphosphate (ADP) and inorganic phosphates is believed to provide the energy source for this process, generally known as the ‘sodium pump’.2 Macrocyclic Antibiotics and their Alkali Metal Complexes In 1964, the American physiologist, B. C. Pressman found that certain antibiotics could induce the selective movement of K+ into rat liver mitochondria.9 These antibiotics, now termed collectively ionophoric (or ion-bearing) antibiotics could also increase the permeability of black-lipid films (synthetic lipid bilayers) to K+. Such compounds are neutral at physiological pH and so may act as discriminatory cation carriers by forming complexes with the alkali metal cations. This discussion concerns these species although a further class of antibiotics containing carboxylic acid functions have also been found to exercise discriminatory powers.10-13 This class can react with alkali metal cations to give 1:1 salts and so act as anionic carriers, or provide negatively charged pores. Valinomycin is a cyclic dodecadepsipeptide, an atypical peptide having alternate amino- and hydroxy-carboxylic acids linked by amide and ester bonds, [L-valine-D-hydroxyisovalericacid-D-valine-L-lactic acid13.It was first isolated from streptomyces sources and has been shown to have a high ion selective C. Moore and B. C. Pressman, Biochem. Biophys. Res. Comm., 1964, 15, 562. lo L. K. Steinrauf, M. Pinkerton, and J. W. Chamberlin, Biochem. Biophys. Res. Comm., 1968, 33, 29.l1 L. K. Steinrauf and M. Pinkerton, J. Mol. Biol., 1970, 49, 533. laL. K. Steinrauf, E. W. Czerwinski, and M. Pinkerton, Biochem. Biophys. Res. Comm., 1971, 45, 279. l3Yu.A. Ovchinnikov, V. T. Ivanov, and A. M. Shkrob, ‘Membrane-Active Complexones’ Elsevier, Amsterdam, 1974. complexation of K+ versus Na+.13 This selectivity has been retained, to a lesser extent, in numerous synthetic analogues of valinomycin. It was first proposed that the ion selectivity arose as a consequence of the size of the cavity in the antibiotic molecules as observed in models. The cavity was large enough to accommodate the hydrated alkali metal ion, however, X-ray structural deter- minations of the K+-complex showed that it is the water free K+ ion that is incorporated.The conformation of uncomplexed valinomycin in solution was established by the composite use of ORD, CD, infra-red, ultraviolet and n.m.r. spectro- scopic studies.13J4 Valinomycin being in the form of a 36 membered ring presents numerous conformational possibilities and it is interesting to note that it is the first large peptide to have its structure established without recourse to X-ray techniques. The latter have, however, provided the final arbiter as subsequent X-ray studies, necessarily carried out in the solid state, on crystals recovered from different solvents and having different modifications, have shown quite different conformations from that proposed for valinomycin in solution.15J6 The gross conformational features for the different crystalline modifications are nearly identical. The conformation of valinomycin in solution has been found to be solvent dependent, and is an equilibrium of three major conformers (Figure 4).In cA 7 B C Figure 4 The solution equilibrium of valinomycin solvents such as heptane or CC14 a bracelet-like structure is determined in which all NH groups are involved in hydrogen bonding (A). In polar solvents the hydrogen bonding breaks down such that only the D-valyl NH are involved, (B), and that at elevated temperatures and in highly polar solvents the NH 14 D. J. Pate1 and A. E. Tonelli, Biochemistry, 1973, 12, 486. G. D. Smith, W. L. Duax, D. A. Langs, G. T. de Titta, J. W. Edmonds, D. C. Rohrer, and C.M. Weeks, J. Amer. Chem. SOC.,1975, 97, 7242. l6 1. L. Karle, J. Amer. Chem. SOC.,1975, 97, 4379. Across the Living Barrier groups are H-bonded to the solvent (C). The crystal structure15 shows a different conformation (Figure 5) and although six H-bonds are maintained, only four are of the type present in the solution state, the remaining two being between ester carbonyls and NH groups. A possible mode of metal incorporation is @ OXYGEN 0 NITROGEN Figure 5 The crystal structure of valinomycin(Reproduced by permission from J. Amer. Chem. Soc., 1975, 97, 7242) suggested as two pairs of oxygen atoms lie in exposed positions on the molecular periphery (* in Figure 5). Either of these pairs could interact with an incoming K+ ion, partially removing its hydration shell, and enticing it in to the hydro- philic interior.Stepwise removal of the hydration sheath could then occur as the sinuous ligand enfolded the metal. The studies concerning the potassium complex have shown that it exists in essentially the same conformation in both solid and solution states (Figure 6) n Figure 6 The valinomycin-potassium complex (Reproduced by permission of Prof. W.L.Duax) Fenton with inwardly orientated ester carbonyl binding to the metal.13 The skeletal atoms wrap round the central cation as would three sine waves and are held in position by hydrogen bonds (Figure 7) which impart a rigidity to the molecular Figure 7 Schematic diagram of a cyclododecapeptide folded to produce a cavity lined with six carbonyl donors (arrows) and stabilized by hydrogen-bonding (broken lines) frame.The metal cation interacts with the ester carbonyl oxygen atoms to give a distorted octahedral geometry, but the carbonyls do not point precisely at the metal cation to give maximum ion-dipolar interaction. The X-ray structural studies have been carried out on the KAuCI~~~and K13-KIs1* complexes. In solution the structure has been shown to be independent of the nature of the solvent. The structure of the Na+-complex, however, has been found to be solvent dependentlg, and the i.r. spectrum in solution shows an asymmetry of the ester carbonyl stretching frequency indicating non-equivalence of these carbonyls in the structure. The smaller size of Na+ would lead to this observation as, if the cavity size is kept constant, it could not interact with all six carbonyl groups at once.This indicates that the availability of a precisely tailored cavity is essential to selectivity. When the H-bonded framework is destroyed by N-methylation of valinomycin selectivity is lost, again showing the necessity of retaining a precisely- tailored cavity for specific complexation. Experiments with small bilayer phospholipid residues have led to the proposal that valinomycin is mainly embedded towards the inner, non-polar portion of l7 M. Pinkerton, L. K. Steinraur, and P. Dawkins, Biochem. Biophys. Res. Comm., 1969, 35, 512. K. Neupert-Laves and M. Dobler, Helv. Chim. Acta, 1975, 58, 432. Yu.A. Ovchinnikov and V. T. Ivanov, Tetrahedron, 1975, 31, 2177. Across the Living Barrier the bilayer.20 Complex formation is believed to occur at the interface of the phospholipid bilayer and the aqueous phase. Precise details of the behaviour of valinomycin are, however, still unclear, some authors give preference to a single carrier modelz1 whilst others refer to a relay mechanism.22 The ability of valinomycin and related compounds to act as potential carriers of alkali metal ions leads to the conjecture that similar species might exist in membranes. A cyclopeptide has been isolated from beef-heart mitochondria1 membranes, and has shown transport propertiesz3, but the poor reproducibility so far obtained shows that more conclusive studies are required in this area.A second group of antibiotics neutral at physiological pH and capable of alkali metal complexation is the enniatins.13 They are isolated from fusarium sources and it is interesting to note that fusarium species have been shown to attack silicate minerals such as orthoclase releasing potassium.24 These soil fungi not only liberate metallic ions from minerals, but can also provide anti- biotics which will complex with alkali and alkaline earth metals. Enniatin B is a cyclohexadepsipeptide, [D-hydroxyisovaleric acid-N-methyl-L- valineI3, and the structure of the enniatin B-KI (Figure 8) complex resembles a 0 Figure 8 l%e enniatin B-potassium complex charged disc with lipophilic b~undaries,~~ the metal being found at its centre.The molecules stack one above the other and if such a process should occur in a membrane then it is possible to envisage the formation of an ion-carrying pore. Solution studies on enniatins have shown the existence of 2: 1 and 3: 2 com-2o E. Grell, Th. Funck, and F. Eggers, in ‘Molecular Mechanisms of Antibiotic Action on Protein Biosynthesis and Membranes’, ed. E. Munoz, F. Garcia-Ferrandiz, and D. Vasquez,Elsevier. Amsterdam, 1972, p. 646. *l P. Lauger, Science, 1972, 178, 24. 9aYu.A. Ovchinnikov, F.E.B.S. Letters, 1974, 44, 1. 23 G. A. Blondin, A. F. Decastro, and A. E. Senior, Biochem. Biophys. Res. Comm.,1971, 43, 28. F. J. Stevenson, in ‘Soil Biochemistry’, ed A. D. McLaren and G. H. Peterson, Edwin Arnold Ltd., London, 1967, p.139. ab M. Dobler, J. D. Dunitz, and J. Krajewski, J. MoI. Biol., 1969, 42, 603. Fenton plexes, and taken to the limit it becomes quite possible to conceive of a pore mechanism for cation transfer using these antibiotics. There is also evidence from solution studies that in the 1 :1 complexes the metal lies at the centre of the cavity but some authors doubt that this is so. especially as the conclusions of the X-ray analysis had not been carried through to the determination of atomic co-ordinates.26 There is a reduced K+/Na+ selectivity for the 1:l complexes relative to valinomycin, which is ascribed to the greater flexibility of the latter; whereas the 2:l complexes are reported to have a modest K+ selectivity. In the latter the cation is effectively shielded from any interaction with its anion and so begins to resemble valinomycin more.Other naturally occurring cyclodepsi- peptides such as the angolides, serratamolides and sporidesmolides which have features resembling enniatins have no reported alkali metal complexes.13 Antamanide, a cyclodecapeptide isolated from Arnanita phalloides mushrooms has a pronounced Na+ selectivity but manifests little ionophoric activity.27 The enniatins show reduced ionophoric activity relative to valinomycin, and this may be related to the incomplete encapsulation of the metal and so reduced lipophilicity of the complex.13 The third group of neutral ionophorous agents consists of the macrotetrolide actins which are isolable from actinomyces species.13 Nonactin, so-named, it is said, because of its inactivity in early tests for biological activity, is depicted in Figure 9.It has a selectivity for K+/Na+ intermediate to valinomycin and enniatin B. The X-ray structure of free nonactin28 shows a large cavity to be present and in the K+-complexZ9 this is occupied by the metal. The ligand wraps 7 CH3 R1 = Rz = R3 = R, = CH3 R, = Rz = R3 = CH3, R, = C2H, Nonactin Monactin R1 = R3 = CH3,Rz = R4 = CZH, R1 = CH3 ,Rz= R3 = R4 = CzH5 Dinactin Trinactin ~ R1 = Rz = R3 = R4 = CzH5 Tetranactin Figure 9 The actins J. A. Hamilton, L. K. Steinrauf. and B. Braden, Biochem. Biophys. Res. Comm., 1975, 64, 151. p7 Th. Wieland, H. Faulstich, and W. Burgermeister. Biochem. Biophys.Res. Comm., 1972, 47, 984. 18 M. Dobler, Helv. Chim. Ada, 1972, 55, 1371. 19 M. Dobler, J. D. Dunitz, and B. T. Kilbourn, Helv. Chim. Ada, 1969, 52, 2573. Across the Living Barrier itself round the metal as the seam of a tennis ball (Figure lo), with the four furanyl and four carbonyl oxygen atoms interacting with the metal. In the sodium complex a similar conformation is found but the metal-oxygen distances Cubic arrangement@ KF Figure 10 The nonactin-potassium complex and the schematic representation of this molecule (Reproduced from B. T. Kilbourn, J. D. Dunitz, L.A. R. Pioda and W. Simon. J. Mol. Biol., 1967, 30, 559, and H. Diebler, M. Eigen, G. Ilgenfritz, G. Maass and R. Winkler. Pure. Appl. Chem., 1969, 20, 93, with permission) differ showing a non-symmetric binding of Na+;30 K+--O (furan), 2.82 A, K+-O (carbonyl), 2.77 A, Na+--O (furan), 2.77 8, and Na+-0 (carbonyl), 2.42 A.This is a manifestation of the cavity-metal misfit due to the smaller ionic radius of the sodium ion. Similar observations have been made for alkali metal complexes of tetranactin.31 As with valinomycin it is possible to suggest a mode of incorporation of the metal ion.2s Upon complexation the cation must be stripped of its hydration 30 M. Dobler and R. P. Phizackerley, Helv. Chim. Ada., 1974, 57, 664. al I. Sakamaki, Y. Jitaka, and Y. Nawata, Acra Cryst., 1976, B32, 768. 334 Fenton shell, a process requiring loss of about 322 KJ mole-1 for K+. This must be compensated for by interactions between the cation and the donor ligands in the complex. Complex formation has been shown to be very fast, and stepwise incorporation of the cation is es~entiaI.~zJ3 Models show that a hydrated K+-cation could be inserted into the nonactin cavity and held there by hydrogen bonding. Through a series of ligand conformational changes the water molecules could then be removed and the cation held in the cavity by ion-dipole interactions with the donor oxygen atoms of the ligand.Certain common features may be found for the alkali metal complexes of the three groups of ionophores; (1) the alkali metal sits in the ligand cavity at a centre of optimal electron density provided by the donor atoms of the ligand, (2) a lipophilic exterior is presented to facilitate cation transport, (3) a flexible ligand is required to effect the energetically favoured stepwise removal of the solvation sheath, (4) a best-fit situation is required with regard to the cavity diameter and the diameter of the incoming ligand, but ligand-ligand repulsions must be minimized, (5) the difference between the energy of ligation and energy of solvation must be maximized.On fulfilment of these requirements the metal ion is given the appearance of a large organic moiety and so the lipid bilayer becomes the victim of a confidence trick in which it sees and transports an 'organic' cation across the barrier. It has been possible to provide evidence for both carrier and pore mechanisms for transport and investigations into the temperature dependence of ion selectivity have provided further insight into the problem.At 25 "C the selectivity ratio for K+:Na+ has been found to be of the order lo4:1 whereas at 0 "Cthis ratio is reduced to only 2:1.34 This dramatic diminishment has been interpreted as indirect evidence for a carrier mechanism as a pore mechanism would be unimpaired by freezing. In a carrier mechanism which necessarily involves a mobile ligand to effect incorporation and transfer of the metal freezing would cause loss of ligand mobility and so severely impair the mechanism. This would not be as serious in a pore. Other evidence to support this comes from experi- ments concerning the role antibiotics play in mediating the ionic conductance of lipid bilayers35 There is an abrupt loss of effectiveness in mediation for the presumed carriers nonactin and valinomycin occurring at the same temperature as the loss of membrane fluidity on cooling.This suggests that as in the selectivity experiments the mobile carriers are rendered inoperable on cooling. In contrast R. Winkler, Structure and Bonding, 1972, 10, 1. ss M. Eigen and R. Winkler, in 'The Neurosciences: Second Study Program', ed. F. D. Schmitt, Rockefeller University Press, New York, 1970, p. 685. E. Eyal and G. A. Rechnitz, Anal. Chem., 1971, 43, 1090. s6 S. Krasne, G. Eisenman, and G. Szabo,Science, 1971,174,412. Across the Living Barrier the effects of an alleged pore-forming polypeptide, gramicidin, on mediation were unaffected by freezing.These observations do not preclude the presence of both mechanisms in a membrane as the alleged carriers were seen to act as carriers, and the alleged pore-former as a pore-former. If, however, gramicidin whose structure is proposed as a helix,22 does form a pore, and if this is the route across the barrier, then it is plausible to suggest that in the fluid mosaic model where the protein striates the membrane, probably in helical array, the ionic selectivity and transport ability are inherent properties of the membrane components. This then alleviates the problem of a secondary system being present over and above the membrane’s protein constituents. For both the ionophorous antibiotics and the synthetic molecules which have been devised to mimic them a simple experiment may be carried out to illustrate both ion transport and ion selectivity.36~37 A system is constructed, using a U-tube, consisting of two aqueous layers separated by a semipermeable medium, or membrane, (CHC13).Chloroform is chosen because its dielectric constant is similar to that for the membrane. A coloured alkali-metal salt such as potassium orthonitrophenolate is introduced on one side of the barrier and this is set aside as a control experiment [Figure ll(a)]. A second tube is then chloroform chloroform only plusin carrier barrier in barrier Figure 11 The U-tube experiment: the shaded portions represent the colour imparted to the solvent phases by the potassium o-nitrophenolate; the right hand columns contain water prepared, having a carrier species in the chloroform layer, and the movement of colour on transport of the metal salt by the carrier may be observed.If the colour transfers rapidly into the organic layer, but no further, the added molecule is acting as an ion receptor [Figure ll(b)]; if the colour is readily transferred through the CHC13 layer and on into the second aqueous layer then the added molecule is acting as an ion-carrier [Figure ll(c)]. The presence of K+ in 36 R.Ashton and L. K. Steinrauf, J. Mol. Biol., 1970, 49, 547. 37 M. Kirch and J. M. Lehn, Angew. Chem., Internat. Edn., 1975, 14, 555. 336 Feirton either layer may be determined by spectrophotometric analysis (of the anion), or by atomic absorption spectrometry.The difference in behaviour is related to the stability constant of the K+-carrier complex; too high a value leads to ion-reception, too low a value inhibits movement, and a middle range value leads to ion-carriage. This type of experiment may be carried out either qualitatively, or with a detailed analysis of transfer rates. A wide range of com- parative results is obtained by changing the nature of the carrier or by altering the cation. 3 Cyclic Polyethers and their Alkali Metal Complexes The rigours of obtaining macrocyclic antibiotics from fungal sources, or of synthesizing polypeptides, causes a requirement for more accessible probes for the processes of transfer.C. J. Pedersen, in 1967, reported the syntheses of a group of macrocyclic pol yet her^,^^ now colloquially termed crown polyethers, which have to an extent filled this role. Pedersen exploited an observation made during research into the development of ligands for use in the preparation of vanadium catalysts for the polymerization of olefins.39 In attempts to synthesize the phenolic derivative (A), in the reaction scheme depicted, he isolated a fibrous white material (B), dibenzo-18-crown-6. 0A0a: QJo wowo OW0W0 (B) Arose from the reaction of residual unprotected catechol with the chloro ether. It was found to be insoluble in hydroxylic solvents, but, in the presence of added alkali a solubilization occurred. It was the recognition that this process was due to complexation of the metal by the polyether that opened up a new area of chemistry.Pedersen termed his compounds crown ethers because of their resemblance, in models, to royal crowns, and because of their ability to ‘crown’ alkali metal cations. The nomenclature adopted is as follows :dibenzo-38 (a) C. J. Pedersen, J. Amer. Chem. SOC.,1967, 89, 7017; (b) C. J. Pedersen and H. K. Frensdorff, Angew. Chem. Znternat. Edn., 1972, 11, 16. 38 C. J. Pedersen, AIdrich Chim. Acra (Aldrich Chem. Co., Milwaukee), 1971, 4, 1. Across the Living Barrier describes the non-ethyleneoxy content, 18, the total number of atoms in the crown ring and 6, the number of heteroatoms in the ring. A detailed study of the complexation properties of these polyether ligands led to the discovery of certain ~orrelations.3~ Three types of complex were isolated; 1:l (doughnut shaped), 2:l (sandwich) and 3:2 (club sandwich), (Figure 12).0 I:1 2: 1 3:2 ,.--.2s...--Figure 12 Schematic representations for the predicted shapes of metal complexes o crown ethers The 3:2 complex has not yet been confirmed as existing in the solid state. A relationship between the ionic radii of the metals and the number of crown oxygen atoms was noted: Li+ and 4 oxygens, Na+ and 5 oxygens, K+ and 6 oxygens, Cs+ and 8 oxygens. This resembles the close-fit parameter for ionophore complexation, but did not prove absolute due to the flexibility of some ligands, and the genera1 donor requirements of the metals.The stability constants40 for the 5 oxygen crown complexes, (benzo-l5-crown-5), with Na+ and K+ were similar and this was due to the stability of the 2: 1 complex with K+. The Na+- cation fits into the 5 oxygen cavity but the K+-cation is too large. However in order to sate its donor capacity a 2: 1 sandwich complex is formed. A very strong complex was found for K+ and the 10 oxygen crown (dibenzo-30-crown- 10) and this was due to encapsulation of the metal by the crown ether, as occurs 4oH.K. Frensdorff, J. Amer. Chem. SOL,1971, 93, 600. Fenfon for valinomycin or nonactin. Several X-ray structural determinations are avail- able for alkali metal complexes41 of cyclic polyethers and representative struc- tures are depicted below (Figure 13).Br0 Na0 0H10 (b) 0 0 0 C Figure 13 Structures of some crown ether complexes: (a), Dibenzo-18-crown-6,RbNCScomplex; (b), Dibenzo-1 8-crown-6,NaBr,2H20complex; (c),Benzo-1 5-crown-5,NaI,H20 complex; (d), (Benzo-l5-crown-5),,KI complex; (e),Dibenzo-30-crown-10; (f), Dibenzo-30-crown-10,KI complex(Reproduced by permission from Structure and Bonding, 1973, 16. 1 ; 1973, 16, 71) In dibenzo-l8-crown-6,RbNCS,(a), the rubidium cation is co-ordinated to six coplanar oxygen and to the nitrogen from the anion to give an ion pair. The rubidium cation lies slightly below the plane of the oxygen atoms, in the direction of the NCS-, and the shape may be likened to that of an inverted umbrella. K+ is slightly smaller than Rb+ and it may be predicted that it would lie closer to the centre of the six oxygen atom torus.Superficially these 18- membered crowns may be likened to the enniatins which also have 18-membered 41 M. R. Truter, Structure and Bonding, 1973, 16, 71. D.Bright and M. R. Truter, J. Chem. Soc. (B), 1970, 1544. 339 Across the Living Barrier rings, but whose lipophilicity is enhanced by the periferal groups making them more efficient operators in biological systems.43 The smaller cavity diameter in benzo-15-crown-5 leads, in general, to 1:l complexation with Na+ and 2:l complexation with K+. The structure of benzo-15-crown-5,NaI,H~0,44(c), follows the pattern of (a); the sodium ion lies beneath the plane of the oxygen atoms but is here pulled towards the residual water molecule.The tenacity of water for sodium is further indicated in the complexes 18-crown-6,NaNCS,H~O45 and dibenzo-l8-~rownd,NaBr,2H20, (b).4s In (b) one molecule in the unit cell has sodium centrally disposed in the crown cavity with axial water ligands above and beneath it, whilst the second molecule has sodium pulled towards a bromide ion, the other axial site containing non-interacting water, The problem of removing water from the solvent shell before complexation can occur is well illustrated in the complex 12-crown- 4,Mg(H~O)sC12,~~in which the structure contains octahedral Mg(H20)c2+ units, no dehydration having occurred, together with 12-crown-4 molecules hydrogen bonded to the complex cations.This again suggests a mode of incorporation of the crown, the first stage being hydrogen-bonding to the solvation shell followed by successive removal of the shell on conformational change of the ligand. 2: 1 Complexes arise when a small cavity ligand is reacted with a large radius cation. (Benzo-l5-crown-S)zKI, (d), is a sandwich complex48 as is [(12-crown- 4)2Na+][CI-,5H20].49 In the latter complex it is the halide anion that is hydrogen bonded to the water molecules, the Na+-cation only interacting with the crown. The macrocyclic dibenzo-30-crown-10 has perhaps the closest comparison with the ionophores as it resembles nonactin; the latter has a 32-membered ring and 8 metal-oxygen contacts in its complex. The structure of the free ligand (e)50 has the form of a long closed loop, and on complexation of K+ encapsulation occurs to give the 1:1 complex (f).50 The visual resemblance with nonactin-K+ is striking (see Figure lo), however, from solution data it has been shown that whilst both have the same selectivity pattern of K+ > Na+,51 the dibenzo-30-crown-lO-K+ complex is stronger than the nonactin -K+ complex (log Kdb = 4.57; log Knon = 3.58 (MeOH solvent)).In biological experiments nonactin is more efficient than the polyether and this has been ascribed to the greater lipophilicity of the antibiotic, A better synthetic model is therefore required ; reduction of the benzo-groups to cyclohexyl-groups, or the introduction of bulky side chains could help achieve this objective.One stoicheometry observed for the crowns but not for the antibiotics, to 4a D. C. Tosteson, Fed. Proc., 1968, 27, 1269. 44 M. A. Bush and M. R. Truter, J.C.S. Perkin 11, 1972, 341. 46 J. D. Dunitz, M. Dobler, P. Seiler, and R. P. Phizackerley, Acta Cryst., 1974, B30,2733. 46 M. A. Bush and M. R. Truter, J. Chem. SOC.(B),1971, 1440. M. A. Neumann, E. C. Steiner, F. P. van Remoortre, and F. P. Boer, Inorg. Chem., 1975, 14, 735. 48 P. R. Mallinson and M. R. Truter, J.C.S. Perkin 11, 1972, 1818. 49 F. P. van Remoortre and F. P. Boer, Inorg. Chem., 1974, 13, 2071. 6o M. A. Bush and M. R. Truter, J.C.S. Perkin II, 1972, 345. 61 P. B. Chock, Proc. Nut. Acad. Sci. U.S.A., 1972, 69, 1939. 340 Fenton date, is 1:2. Dibenzo-24-crown-8 has been found to encapsulate two K+ ions52 in the complex (dibenzo-24-crown-8), (KNCS)2 and several other bimetallic complexes are now The stoicheometry of one natural active trans- port system is that three sodium ions and two potassium ions move in opposite directions for every molecule of adenosine triphosphate hydrolysed and so the above observations indicate the possibility of a single carrier for more than one cation. 4 General Applications of Crown Ethers Twenty years ago the alkali metals were ‘out of sight, out of mind’ to the co- ordination chemist.A prolific growth has followed the planting of seeds by physiologists ten years ago and now an extensive chemistry has developed which has spread to encompass the whole domain of the subject.Movement away from two-dimensional, monocyclic ligands to three-dimensional ligands such as the macroheterobicyclic ‘~ryptands’~~9~~ (Figure 14) leads to the avail- [1,1,1] rn = n = 0 [2,1,1] m = 0;n = 1 [2,2,1] m = 1; n = 0 [2,2,2] m = n = 1 [3,2,2] m = 1; n = 2 [3,3,2] m = 2; n = 1 [3,3,3] m = n = 2 Figure 14 Cryptands ability of rigid cages of finite diameter which give much more specific metal complexation than crown ethers.57 This is because the tailoring of the cavity diameter to cation diameter is more precise. Cations of diameter larger than the optimal fit may be excluded from the cage, and smaller cations just rattle around forming only weak complexes. An extensive chemistry of this area has now been developed.Crown ethers have also found application in organic synthesis5* On surround- ing a cation with a crown ether solubility is conferred on the cation via hetero-atom solvation and accompanying enhancement of lipophilicity. The anion 62 D. E. Fenton, M. Mercer, N. S. Poonia, and M. R. Truter, J.C.S. Chem. Comm., 1972, 66. 53 N. S. Poonia and M. R. Truter, J.C.S. Dalton, 1973, 2062. j4 D. L. Hughes, J.C.S. Dalton, 1975, 2374. 56 B. Dietrich, J. M. Lehn, and J. P. Sauvage, Tetrahedron Letters, 1969, 2885, 2889. sE J. M. Lehn, Structure and Bonding, 1973, 16, 1. 57 J. M. Lehn and J. P. Sauvage, Chem. Comm., 1971, 440. 58 G. W. Gokel and H. D. Durst, Synthesis, 1976, 168. Across the Living Barrier may be carried into solution at the same time, and so a previously organic insoluble base such as KOH may be rendered soluble (in aprotic media for example).This procedure is used effectively in phase transfer catalysis59 where the reagent may be transferred from aqueous, or solid, phases into organic media by the formation of transient complexes. Homogeneous solutions of crown ether complexes have been used to study mechanisms, rates, and product distribution in nucleophilic substitution reactions in non-polar media.58t60 Furthermore the movement from contact ion-pairs to solvent-separated ion-pairs allows for the provision of activated nucleophiles for synthetic purposes, e.g. KF releases F-61 and K(0Ac) releases (OAC-)~~, the so-called ‘naked anions’. The solubilization process is not restricted to salts, and alkali metals have been made soluble in THF,ethers, and amines, in the presence of crowns and crypt and^.^^ One fascinating feature of this work has been the isolation and characterization of the crystalline species [cryptand 2,2,2]Na+,Na- as a stable compound.64 Further applications have been made in inorganic chemistry, to the preparation of substituted phosphazenes and the stabilization of unusual anions.Hexachlorotriphosphazene reacts with KF, or KSCN, in the presence of 18-crown-6 to produce the corresponding hexafluoro-, or hexathiocyano- deri~ative~~;andspecies such as Sng4- and Pb& have been stabilized, and characterized by X-ray techniques using [2,2,2-cryptand]Na+ as the counterion.66 Cyclic polyethers have also been used in organometallic chemistry to investigate reaction pathways,67 to prepare Group VI pentacarbonyl halides,Gs and as rr-electron donors in complexes such as dibenzo-l8-crown-6,Cr(C0)3.69 The versatility of synthetic polyethers is further exemplified by their use in chiral recognitioii studies.Optically pure crowns have been found to complex optical isomers of amino acids, selectively effecting resoIution.?O This has led to their use as models for enzyme-substrate reactions, the area of ‘host-g~est’,~~ or ‘lock and key’ ~hemistry,~~ discussed recently by Professor D. J. Cram in his 1976 Centenary Lecture.72 68 J. Dock, Synthesis, 1973, 441. 6o D. J. Sam and H. E. Simmons, J. Amer. Chem. SOC.,1972, 94, 4024.C. L. Liotta and H. P. Harris, J. Amer. Chem. SOC.,1974, 96, 2250. 62 C. L. Liotta, H. P. Harris, M. McDermott, T. Gonzalez, and K. Smith, Tetrahedron Letters, 1974, 2417. 63 J. L. Dye, M. G. de Backer, and V. A. Nicely, J. Amer. Chem. Sac., 1970, 92, 5226, 64 F. J. Tehan, B. L. Barnett, and J. L. Dye, J. Amer. Chem. SOC.,1974, 96, 7203. 66 E. J. Walsh, E. Derby, and J. Smegal, Inorg. Chint. Acta, 1976, 16, L9. J. D. Corbett and P. A. Edwards, J.C.S. Chem. Comm., 1975, 984. J. P. Collman, J. N. Cawse, and J. I. Braumann, J. Amer. Chem. SOC.,1972, 94, 5905; J. P. Collman and S. R. Winter, J. Amer. Chem. SOC.,1973, 95, 4089. 68 J. L. Cihonski and R. A. Levenson, Inorg. Chem., 1975, 14, 1717. O8 K. H. Pannell, D. C. Hambrick, and G. S. Lewandos, J.Organometallic Chem., 1975, 99, c21. 70 D. J. Cram, R. C. Helgeson, L. R. Sousa, J. M. Timko, M. Newcomb, P. Moreau, F. de Jong, G. W. Gokel, D. H. Hoffman, L. A. Domeier, S. C. Peacock, K. Madan, and L. Kaplan, Pure Appl. Chem., 1975, 43, 327. 71 W. D. Curtis, D. A. Laidler, J. F. Stoddart, and G. H. Jones, J.C.S. Perkin Z, 1977, 1756. 72 D. J. Cram, 1976 Centenary Lecture, Chemical Society. Fenton Despite the seemingly large amount of activity, this field of chemistry is still in a growth phase and some areas remain relatively unexplored. It is apparent that serendipity has played an important role in the development of models for antibiotics and that the spin-off areas, away from the immediate biological context of this report are rich in unmined chemistry.The interaction of alkali metal cations with acyclic species, polyethylene oxides73 and linear polypeptides may yield useful information pertaining to the interaction of such cations with the protein in the fluid mosaic model of the membrane. Certainly complexes, albeit weak, are formed with those systems. There remains much more to observe and discover, and many results and applications to harvest from this field of study. 73 D. E. Fenton, J. M. Parker, and P.W. Wright, Polymer, 1973, 14, 589.

ISSN:0306-0012    

DOI:10.1039/CS9770600325

出版商:RSC    

年代:1977

数据来源: RSC

摘要:

Metal-ion-promoted Reactions of Organo-sulphur Compounds By D. P. N. Satchell DEPARTMENT OF CHEMISTRY, KING’S COLLEGE, STRAND, LONDON WCZR 2LS 1 Introduction Certain metal ions, especially Hg2+ and Ag+, have long been known to react readily with organo-sulphur compounds.l Often these reactions facilitate changes in the organo-sulphur compound which can also occur slowly, either spontaneously or under catalysis by hydrogen acids. In a loose sense, the metal ions catalyse such reactions, but since the metal is also frequently consumed as a covalent and often insoluble sulphide the processes are usually better des- cribed2 as ‘metal-ion-promoted.’* A typical example is the hydrolysis of thiol esters in aqueous solvents. This reaction occurs spontaneously, is (slightly4) catalysed by hydrogen ions [reaction (l)], and is greatly facilitated by Hg2+ or Ag+ ions5 [reaction (2)].In the metal-ion-promoted process the thiol is formed HIO+ R1COSR2 + H20+R1C02H + R2SH (1) R’COSR2 + 2Hz0 + Ag++RlCOzH + R2SAg + H30+ (2) as a relatively insoluble metal derivative, which must subsequently be decomposed if the free thiol is required. Many such reactions are known, but the information is scattered in the literature and has been only fragmentarily reviewed. The present review will draw examples mainly from reactions of thiols, of disulphides, and of thio-ethers, -acetals, -esters, -acids, -anhydrides, and -amides. In these com- pounds the S atoms are divalent and can interact directly with the metal.Some of the reactions have been studied kinetically. A difficulty frequently met in kinetic work throughout the field is the low solubility of the metal sulphide products, since the presence of a precipitate can lead to (unwanted) auto-catalytic effects.6 Homogeneous systems can, however, usually be obtained by careful *The notiona that if a species accelerates a reaction by its effect on AS*, rather than by its effect on AH*, then that process should be called ‘promotion’ rather than ‘catalysis’ runs counter to current usage: catalysis covers both effects. The term promotion is normally used in the sense employed in the present article. E. E. Reid, ‘Organic Chemistry of Bivalent Sulphur’ vols. 1-5, Chemical Publishing Co., New York, 1958. * M.L. Bender, Adv. Chem. Series, 1963, 37, 19. R. W. Hay and P. J. Morris, J.C.S. Dalton, 1973, 56. J. R. Schaefgen,J. Amer. Chem. SOC.,1948, 70, 1308. 1970, 1306. (B),SOC.Chem.J.Secemski,1.I.N. Satchell and P.D. .5 * A. J. Hall and D. P, N,Satchell,J.C.S. Perkin 11, 1975, 1273 ;A. J. Hall, Thesis, UniversityQf London, 1976. Metal-ion-promoted Reactions of Organo-sulphur Compounds control of reaction conditions. Almost all the reactions reported involve class B (soft7y8) metal ions and have heterolytic (i.e. Lewis acid-base) mechanisms. The reactions covered are of interest in a surprising number of contexts. They offer simple examples of catalysis by class B metal ions, and their quanti- tative study allows some assessment of class B character; they have valuable applications in organic synthesis and in quantitative and qualitative analysis; they are relevant to biochemical processes and they throw light on ligand- substitution reactions in metal ions and upon reactions of co-ordinated ligands.All these points will be illustrated. The reactions of all sulphur compounds are much less susceptible to catalysis by hydrogen acids than are the reactions of their oxygen analogues when protonation of S (or 0)underlies the catalysis (H+ being a hard acid). That is one important reason why the corresponding metal- ion-promoted reactions are of interest for the sulphur-containing compounds. 2 Reactions of Thiols Thiols readily form compounds (mercaptides) with metals, e.g.reactions (3) and (4). Mercaptides of the heavier, and especially class B, metals are poorly soluble. RSH + KOH -+RSK + HzO (3) 2RSH + Hg(NO3)z -+ (RS)zHg + 2HN03 (4) They have characteristic melting points and are useful for the isolation, identi- fication, and quantitative analysis of thi01s.~ The early isolation and purification of coenzyme-A was based on its precipitation as the mercury or copper derivative, followed by regeneration of the thiol with hydrogen sulphide.1° Mercaptides are normally stable, but can decompose to give mainly sulphides, e.g.reaction (5). Sometimes, on heating, some disulphide, or olefin, is also formed, heat (MeS)zPb __+ MezS + PbS (5) heat (EtS)zHg __+ Hg + EtS-SEt heat (Me3CS)zHg --+Me3CSH + MezC=CHz + HgS (7) e.g.reactions (6)and (7). Mercaptide formation, followed by processes (5) or (7), represents metal-promoted conversion of thiols into thioethers or olefins. Similar reactions occurll when the thiol is heated with certain metal sulphides [e.g.reactions (8) and (9)], when the changes may also proceed via the mercaptide. S. Ahrland, J. Chatt, and N. R. Davies, Quart. Rev., 1958, 12, 265.* ‘Hard and Soft Acids and Bases’, ed. R. G. Pearson, Dowden, Hutchinson, and Ross Inc., New York, 1973. F. Challenger, ‘Aspects of the Organic Chemistry of Sulphur’, Butterworths, London, 1959. 10 J. D. Gregory, G. D. Novelli, and F. Lipmann, J. Amer. Chem. Soc., 1952, 74, 854; H. Beinert, R. W. von Korff, D. E. Green, D. A. Buyske, R. E. Handschumacher, H.Higgins, and F. M. Strong, ibid., 1952, 74, 854. 11 D. S. Tarbell and D. P. Harnish, Chem. Rev., 1951, 49, 1. Satchel1 However, the detailed mechanisms of all these essentially heterogeneous reactions are unknown. The conversion of thiols into thioethers has been found to occur more readily, and under more controlled conditions, when a mercaptide is treated with an alkyl halide12 [reaction (lo)], and especially13 when the thiolate group is part of a chelate [reactions (11) and (12)]. Normally the resulting thioether is isolated as the metal complex, from which it can be displaced by more strongly co-ordinating ligands [e.g. reaction (12)]. So far, such reactions have always involved class (EtS),Pt + 2 Etl -----+ Pt(SEtJa1, (10) Me Mc 2-I PhC-S\ ,S-CPh PhC --S -CPh II Pt II +2MeI -PhC-S' 'S -CPh PhC -Sa 'S -CPh i PhC -SXlr PN ,S-CPh il PhC -SMe Ph/\Ph €3-like metals.This type of alkylation has been used13 to synthesize large cyclic ligands [reaction (13)]. The reactions are conducted in solution, or suspension, la S. Smiles, J. Cheni. Soc., 1900, 77, 160. Is D. H. Busch, J. A. Burke, D. C. Jicha, M. C. Thompson, and M. L. Morris, Adv. Chem. Series, 1963, 31, 125. Metal-ion-promoted Reactions of Organo-sulphur Compounds I1 Br in DMF, chloroform, or methanol. The final complex is normally more soluble than the starting material. In some examples, which can be examined under homogeneous conditions throughout, kinetic studies have been made.13 The main conclusions are: (i) each alkylation step is first-order in the thio-complex and first-order in the alkyl halide, (ii) complexes of palladium are much less reactive than those of nickel, (iii) the rate constant for alkylation of the second S atom in an overall process such as (12) is ca.30-fold larger than that for alkylation of the first, and in reaction (13) the closure of the ring is more than 300-fold faster than the initial alkylation, (iv) the reactivity of alkyl halides (RHal) is in the sequence RI > RBr > RCl, and (v) the difference in reactivity between benzyl and p-nitrobenzyl chloride arises mainly from a change in AS*.These findings, and others subsequently reported14 for the alkylation of substituted 8-mercapto- quinoline derivatives of nickel [reaction (14)], seem best explained by a mechan-ism such as reaction (15).A slow intermolecular SN2 displacement of halide from free RHal by co-ordinated sulphur, or a prior slow dissociation of the metal-sulphur bond in what are essentially square-planar complexes (and there- fore unlikely to display dissociation mechanismsl5) seem less probable. Reaction (1 5) involves five-co-ordinate intermediates, but the fact that equilibrium constant K2 is likely to be greater than K1 readily explains the otherwise surprising rapidity of the second alkylation step. For examples such as reaction (13) the ring-closure stage is entirely intramolecular, and this accounts for its relatively great speed. Certain Friedel-Crafts (HCI-AlCI 3-catalysed) alkylations by thiols have been reported.16 However, it is not yet known if they involve direct interaction between the S and A1 atoms.Some sulphur derivatives of ketones that are otherwise difficult to isolate can be obtained by making use of the stability of their chelates with soft metals. For example, gem-dithiols, formed in the reaction of hydrogen sulphide with ketones, are very unstable even at low temperatures, but can be isolated as their lead derivatives17 [reaction (16)]. With p-diketones, hydrogen sulphide leads normally to the monothio-derivative only. However, in the presence of suitable l4 J. C. Schoup and J. A. Burke, Inorg. Chem., 1973, 12,1851. 15 F. Basolo and R. G. Pearson, ‘Mechanisms of Inorganic Reactions’, J.Wiley and Sons, New York,2nd edn., 1967. 16 J. R. Meadow, Chem. Abs., 1946,40, 6502; U.S.P. 2 403 013. l7 R. Mayer in ‘Organosulphur Chemistry’. ed. M. J. Jenssen, Wiley, New York, 1967, Satchel1 Me i I I Me metal ions the dithiodiketone is formed as its anion. Thus thioacetylacetonate (SAA) can be converted into dithioacetylacetonate (SSAA) via its chelated palladium complex [reaction (17)]. Mercury, rhodium, platinum, and cobalt effect similar transformations.18 -H2S/ ;c=s Pd(SAA)2 + H2S +Pd(SSAA)2 + Hz0 (1 7) S. Kawanishi, A. Yokoyama, and H. Tanaka, Chem. and Pharw. Bull. (Japan), 1973, 21, 2653; 1972,20,262. Metal-ion-promoted Reactions of Organo-sulphur Compounds The great stability of metal chelates involving thiolate ligands has important consequences in many contexts.Two further examples are (i) the preferential co-ordination of cysteine esters as in (1) (rather than viatheir ester groups) and the consequent relatively feeble catalysis by metal ions of their hydrolysis in aqueous solution,l9 and (ii) the displacement of thiazolhe-Schiff base equilibria, which normally lie on the thiazoline side in alcohol or DMF, towards the Schif€ base in the presence of Ni2+, Cd2+, or similar ions20 [reaction (18)]. These processes sometimes involve a prior, rate-limiting tautomerism and sometimes a direct attack of the metal on the thiazoline. The latter case constitutes a metal- promoted rearrangement. ,CHI-S, ,S-Cy (11-21~ ROIC -CH I\ irMZ/CH-CO,R NH, NH* (18) 2H+ As a final example of the influence of metal ions on the reactions of thiols, their conversion into disulphides must be mentioned.This reaction has long been known. It can occur simply on heating the mercaptide [equation (6)],when the metal is reduced, or in aqueous solutions of metal-thiol complexes (especially chelate complexes) in the presence of oxygen, when the latter takes up hydrogen. Complexes of Fez+ and Cu2+ have been widely studied, but other metals, especi- ally transition metals, can be used.21s22 Although the picture is by no means clear, kinetic results for the Fe2+-mercaptoacetate system21 suggest a reaction between a rapidly formed metal chelate and oxygen, probably involving electron transfer and radical formation, as in the outline scheme shown by reactions (19)-(21).Other steps are doubtless involved, and the details seem to depend upon the pH and upon the relative reactant concentrations. Disulphide formation is of wide- spread interest in protein chemistry, since protein stability can be critically dependent on sulphur-bridging. It is found too that proteins are protected against radiation damage on conversion of their free thiol groups into disulphides with added thi~ls.~~ lS L. J. Porter, D. D. Perrin, and R. W. Hay, J. Chem. Soc. (A), 1969, 118. Po L. F. Lindoy, Coordination Chem. Rev., 1969, 4,41. I1 D. L. Leussing and T. N. Tischer, Adv. Chem. Series, 1963, 37, 216. I* S. H. H. Chaston and S.E. Livingstone, Austral. J. Chem., 1967, 20, 1065. H. Sakurai, A. Yokoyama, and H. Tanaka, Chem. andPharm. Bull. (Japan), 1971,19,1416. Satchel1 (3) + SCH,CO,-+ SCH,CO,-I SCH &O0- HSCHSCOs-(2) 3 Reactions of Disulphides Most reactions of these compounds involve S-S bond fission. As noted above, disulphides are of special interest in biochemistry, and an example which nicely illustrates both the dependence of protein structure on disulphide bonds, and their cleavage by soft metal ions, is the disruption of the protein sheath of the TZ bacteriophage by complexes of Zn2+, Cd2+, and Hg2+.In nature, this phage is able to inject its DNA into the bacterium E. coli after an analogous disruption of the sheath by zinc species in the bacterial cell wall.24 In view of the importance of cleavage reactions of disulphides, surprisingly little work has been done on their promotion by metal ions.Most studied has been the hydrolysis in aqueous solvents. This occurs via a heterolytic path, perhaps as shown in reactions (22) and (23). Despite some kinetic w0rk,~5 this fast-slow (22)RSSR+ Ag+ .--RS[R -RS+ + RSAg Ag+ fast RS+ + 2H20 --+ RSOH + H30+ (23) scheme remains tentative. The sulphenic acid (RSOH) is unstable, and it dis- proportionates rapidly to sulphinic acid (RSOZH)and RSH. The hydrolysis is known to be promoted by ions other than Ag+, notably Hg2+and Cu2+.Disul-phides with chelating possibilities tend to react rapidly, and sometimes render one metal ion very much more effective than others.26 Cleavage occurs via some variety of electron-transfer route when disulphides L. M.Kozloff, Records Chem. Progr., 1960, 21, 49. as R. Cecil and J. R. McPhee, Biochem. J., 1957, 66, 538. a6 I. M. Klotz and B. J. Campbell, Arch. Biochem. Biophys., 1962, 96, 92. Metal-ion-promoted Reactions of Organo-sulphur Compounds displace carbon monoxide from metal carbonyl c0mplexes,~7 and reaction (24), which occurs in alkaline aqueous alcohol solution,28 may represent the reverse of a scheme such as that shown in reactions (19)-(21) or may result from initial S-S bond cleavage by hydroxide ion, followed by chelation of the thiolate product. PhC=O O=TPh PhC =oNi,+ CH CH 4 CH (24)OH-II It II /Ni\ r PhC --S----S----CPh PhC-S S-CPh 4 Reactions of Thioethers Thioethers have been more extensively studied than disulphides. Most thioethers are very stable towards attack by electrophiles.They are even less readily cleaved than are their oxygen-analogues by hydrogen acids, and they form stable addition compounds with suitable electrophiles, including soft metal ions [reactions (25) and (26)]. Carbon-sulphur bonds are thus difficult to break by EtzS + EtBr 3Et&+ Br- (25) 2EtzS + HgClz +(Et2S)zHgClz electrophilic attack on sulphur. It is found, however, that ethers R1SR2,in which R1 is an aryl group and/or R2 is an aliphatic group having some stability as a carbonium ion, can be cleaved by various metal derivatives, especially under forcing conditions.The ability of R2to leave as a cation is probably of particular importance, and various pieces of evidence point to unimolecular C-S bond heterolysis in the metal-ether complex as the usual mechanism of decomposition. Thus, with aqueous mercuric chloride, optically active a-methylbenzylmercapto- acetic acid leadsz9 to racemic a-methyl benzyl alcohol [reaction (27)] ;with aqueous silver nitrate, alkyl alkenyl sulphides undergo allylic rearrangement [reaction (28)];3O and the fission of aryl benzyl thioethers by aluminium bromide in a chlorobenzene solvent [reaction (29)] is accelerated by electron-withdrawing substituents in the aryl half of the ether.11131 This reaction also leads to some benzylation of the solvent, although this may occur after the formation of the PhCH 2Br.We have seen earlier how chelating thiol compounds can be alkylated at the S atom. [e.g. reaction (ll)]. The reverse20932 also occurs, but normally less E. W. Abel and B. C. Crosse, J. Chew. SOC.(A), 1966, 1377. 28 R. K. Y. Ho, S. E. Livingstone, and T. N. Lockyer, Austral. J. Chem., 1966, 19, 1179. B. Holmberg, Arkiv. Kenii Mineral. Geol., 1940, 14A, No. 2 (Chew. Abs., 1941, 35, 4364). B. Saville, J. Chem. SOC.,1962, 4062. 31 H. F. Wilson and D. S. Tarbell, J. Amer. Chew. SOC.,1950, 72, 5200. sa S. E. Livingstone and T. N. Lockyer, Znorg. Nuclear Chem. Letters, 1967, 3, 35. Satchel1 PhCH(Me)SCHzCOzH + HgClz + HzO -+ PhCH(Me)OH + ClHgSCHKOzH + HCI (27) MeK=CHCH(Me)SR + AgN03 + H2O -MezC=CHCH(Me)OH + MezC(OH)CH=CHMe + AgSR + HN03 (28) PhCl ArSCHzPh + AIBr3 -+ ArSAlBr2 + PhCHzBr (29) readily.Forcing conditions often seem necessary, and also conditions under which it is dficult for the cleaved group to re-alkylate the S atom [reaction (30)]. Aryl alkyl ethers again seem to be those most readily decomposed. These reactions are not yet well understood, but recent work33 using ligands similar to those used in the alkylation studies13 suggests that the usual mechanism of dealkylation may well be the microscopic reverse of processes such as reaction (15). As expected, a trityl group is more easily cleaved than is a benzyl group. The dealkylating ability of metal ions is in the sequence Co2+ c Ni2+< Hgz+. Studies which uncover the factors controlling the equilibrium position in these alkylation-dealkylation systems would be welcome.Ph, Ph, (30) t 2 MeOEt + ZHCIO, Compounds somewhat analogous to thioethers are the phosphonothiolates. The replacement of SEt-in diethyl phosphonothiolate by F-or OH- in aqueous solution is promoted by silver ions34 [reactions (31) and (32)]. A kinetic study shows that the rate equations take the forms (33) and (34), respectively. Equation (33) suggests the scheme outlined in reactions (35)-(37) for the fluoride reaction. EtOP(Et)SEt + 2H20 + Ag+ -EtOP(Et)OH + AgSEt + H30+ (32)II II0 0 33 R.W. Hay, A. L. Galyer, and G. A. Lawrence, J.C.S. Dalton, 1976, 939. 34 B. Saville, J. Chem. SOC.,1961, 4624. Metal-ion-promoted Reactions of Organo-sulphur CompoundLy ,(4) + Ag+eEtOP(Et)S-Et .--Ag+ fast (35) 0 Agf 0 (5) + F--+(7) + AgSEt slow (36) (6) + F--+ (7) + [AgzSEt]+ SIOW (37) Evidence for species containing two silver ions attached to sulphur is found in other contexts, and it seems that attack by neutral water molecules requires powerful promotion of this sort, since no kinetic term that is first-order in Ag+ is detected in this case [equation (34)].Ag+ Me Ag Me AgI Me \ 3 3 C0,Me C0,Me O H Two interesting examples of silver-ion-promoted reactions of cyclic thioethers may also be mentioned. Processes such as reaction (38) presumably involve cleavage first of thioether and then of thio1.35 The formation of olefinic rather than of solvolysis products is perhaps aided by the conjugated carbonyl groups, but the detailed mechanisms of these reactions can, at present, only be guessed at. The aminolysis of episulphides in aqueous solution is also promoted by silver ions, and provides a convenient route to aminoethanethiols.36 Aminolysis is slow in the absence of silver, and it has been suggested that promotion probably proceeds via silver-amine complexes [reaction (39)].This is in line with other work on promoted aminolyses (see p. 360). s6 D. Gravel, R. Gauthier, and C. Berse, J.C.S. Chem. Comm., 1972, 1322. sB R. Luhowy and F. Meneghini, J. Org. Chem., 1973, 38, 2405. Satchel1 R3 fast RlNH, + Ag+ R'NH2 +Ag+ fast (39)Islow R3 I -Hi AgSCH ,CNHR' f---fast S +NH, IIR2 Ag R' 5 Reactions of Thioacetals Thiols, like alcohols, react with carbonyl compounds, but more readily.The resulting hemi-thioacetals (8), thioacetals (9), and their cyclic analogues (10) and (11) are of great synthetic interest. They are much more stable towards dilute solutionsof hydrogen acids than are their oxygen analogues, but can be hydrolysed under mild conditions, using metal-ion promotion [e.g. reaction (40)]. They provide, therefore, a convenient means of protecting either -SH or -C=O groups during synthesis. Employed for many years in protein and sugar chemistry, interest has sharpened recently owing to their use by Seebach3' in indirect acylation, the -C=O group being generated finally by a process such as reaction (40). Species (10) and (11) are also readily reduced to the methylene derivatives by Raney nickel, but this reaction, which has proved particularly useful in the steroid field, is outside the scope of the present review.R'\/OH R' R' \c/SR3 R".pl'''slR2' 'SR3 R2' 'SR3 RZ' 's R2' \s R' ySR2+ 2Ag+ + 3H20 + R'CORI + 2R2SAg + 2H30+ (40) R' ''SR2 The synthetic importance of the metal-ion-promoted hydrolyses has led to the publication of numerous preparative recipes38 (mostly involving salts or oxides 37 D. Seebach, Angew. Chem. Internat. Edn., 1969, 8, 639. s8 E.g., D. Gravel, C. Vaziri, and S. Rahal, J.C.S. Chem. Comm., 1972, 1323; T.-L. HQand C. M. Wong, Canad. J. Chem., 1972, 50, 3740. Metal-ion-promoted Reactions of Organo-sulphur Compounds of soft metals), but to little kinetic work, so that the mechanistic details are in most cases unknown.One unusual (but apparently suitable) catalyst is ceric ammonium sulphate. Some kinetic information is available for catalysis by mercuric chloride. This was one of the earliest salts to be used,ll often in con- junction with hydrogen chloride, the combination being claimed11 (although on what evidence is not clear) to be more effective than mercuric chloride alone. The initial kinetic ~0rk~~,4~ followed this pattern. Reactions (41) and (42) were studied, using aqueous solvents containing various concentrations of mercury salt and a fixed, relatively large concentration of hydrogen chloride. Under these conditions the hydrolyses are first-order in the thioacetal and in the total mercury concentrations.A recent study41 of reaction (42) has shown that a reduction in the concentration of chloride ion leads to a marked increase in rate, and the (approximate) relative reactivities of the different mercury species in solution are Hg2+(1) N HgCl+(1) 2: HgClz (1) > HgC13-(0.1) $= HgCh2-(< 0.001).The rate constant for reaction via Hg2+ is ca. 106-fold larger than that for hydrolysis in the presence of H30+ alone, and the addition of hydrogen chloride leads to a reduction, rather than an increase, in rate. Since the hydrolysis proceeds to completion, if relatively slowly, even in the absence of the metal, it is unlikely that the metal’s catalytic role is to remove thiol as it is formed, via an equilibrium such as reaction (43), as was once believed.ll Direct, rapid, pre-equilibrium H,% /OHR12 C /SR3 + H20 R’ZC + R3SHL-(43)‘OR2 ‘OR2 co-ordination of the acetal’s S atom to mercury is likely, and substituent effects and values of AS* sugge~t~~#40that reaction (42) involves a slow unimolecular breakdown of the acetal-mercury complex, whereas reaction (41) proceeds via the slow attack of water on this complex.99 L. R. Fedor, J. Amer. Chem. SOC.,1968, 90, 7266. 40 L. R.Fedor and B. S. R. Murty, J. Artter. Chem. SOC.,1973,95, 8407. 41 D. P. N. Satchel1 and L. Z. Zdunek, unpublished results. Satchel1 That the metal is interacting directly with the thioacetal and assisting the removal of RS- is supported by reactions (44) to (46) observed in non-hydroxylic solvents :42-44 all suggest carbonium ion intermediates.Reaction (46) represents Cu2+-promoted alkylation, and other nucleophiles can replace anisole. CUClZ + EtSCuCl + HCldioxan.)\SEt anisole OMe 6 Reactions of Thioesters Relatively little work exists on the reactions between thioesters and soft metal ions, but a substantial portion of it is of a quantitative and kinetic nature. Virtually all the work refers either to hydrolysis or aminolysis, both these reactions of thioesters being only feebly catalysed by hydrogen ions. The metal- promoted hydrolysis of both thiolesters [reaction (47)] and thionesters [reaction (48)] has been known for some time;45~4~ studies of the aminolysis of thiolesters [reaction (49)] are more re~ent.~7 Early qualitative reports are somewhat in conflict concerning the relative efficiencies of different metals, but it is clear that R1COSR2 + H2O + HgClz -+ RlCOzH + HCI + R2SHgCI (47) R1CSOR2 + H2O + 2AgN03 -+ R1COOR2 + AgzS + 2HN03 (48) R1COSR2 + R3NHz + AgN03 -+ RlCONHR3 + AgSR2 + HN03 (49) 4aT.Cohen, G.Herman, J. R. Falck, and A. J. Mura, J. Org. Chem., 1975, 40, 812. 43 T. D. Lee, M. V. Pickering, and G. D. Daves, J. Org. Chem., 1974, 39, 1106. 44 T. Mukaiyama, K. Narasaka, and H. Hokonok, J. Amer. Chem. SOC.,1969, 91, 4315. 46 G. Sachs, Chem. Ber., 1921, 54, 1849. 46 T. Matsui, Mem. CON.Sci. Eng. Kyoto IMP. Univ., 1912, 3, 247. 41 R. Schwyzer and C.Hurlimann, Helv. Chim. Acta, 1954, 37, 155. 357 Metal-ion-promoted Reactions of Organo-sulphur Compounds Hg2+ and Ag+ can promote hydrolysis, Hg2+ being particularly effective. For aminolysis, preparative experiments by S~hwyzer*~suggest the sequence Pb2+ c Cu2+ < Hg2+ < Ag+. A recent kinetic study5948 of the hydrolysis of esters p-RCaH4COSEt, under homogeneous conditions in aqueous solutions of various metal perchlorates, leads to the following conclusions. (9 with Hg2+ the rate equation is -d[S-ester]/dt = k[Hg2+] [S-ester], but with Ag+ it is -d[S-ester]/dt = (kl[Ag+] + k2[Ag+I2} [S-ester]. (ii) substituent effects, solvent isotope effects, and the activation parameters indicate that the mechanism of the Hg2+-promoted hydrolysis changes from one [reaction (50)] that is analogous to an A1 mechanism, when R = OMe, to a mechanism (51) analogous to an A2 scheme, when R = NO2.With Ag+ the mechanism remains A2-like for all substituents for both the routes involving one [reaction (52)] and two [reaction (53)] silver ions. Route (53) is important for electron-repelling substituents (e.g. R = OMe). This kinetic behaviour in the presence of silver resembles that found for the phosphonothiolate reactions (33) and (34). r””” slow p-RCeH4COSEt + Hgz+ ,fast p-RCsH4CQS. +p-RCaH46.O + EtSHg+ ,-, \ (50)Et 2H:O fast \ I Hga+ fast f p-RCsH4COSEt + Hgz+ 4.p-RCaHaCOS \ Et (12) slow fast (12) t HzO --3 ~J-RC~H~C!~ + EtSHg+ -p RCeHdCOzH + HaO+ + EtSHg+ (51b)w\OH2+ 4* D.P. N. Satchel1 and I. I. Secemski, Tetrahedron Letters, 1969, 24, 1991. Satchell slow fast (14) + HzO +p-RCeH4C02H2+ + EtSAg2+ -p-RCsH4COzH + HsO+ + EtSAgz+ (53b)H*O (iii) Hg2+ and Ag+ are ca. 106 and ca. 103 times more effective, respectively, in accelerating the hydrolyses of these esters than the same concentration of hydrogen ions. Other soft ions tested (Pb2+, Cu2+, Cd2+, and Ni2+) are less effective than the hydrogen ion. A similar pattern of metal ion efficiencies was also found in a recent kinetic of the related thionesters p-RCgHKSOEt. These esters are hydrolysed ca. 106 times more rapidly in the presence of metal ions than are their thiol analogues. This is probably because the pre-equilibrium (54a) lies further to the right than in reaction (Sla), and because the positive charge on the metal can more readily be transmitted to the carbonyl carbon atom.This great difference in reactivity has been used to estimate the compositions of thiol-thionester mixtures.50 OEt slow I (15) + H,O p-RC,H,-C-SHg+ ] -p-RC,H,COOEt (54b) +AH2 2Hzo + HgS + 2H,O+[ 49 D. P. N. Satchell, M. N. White, and T. J. Weil, Chem. and Znd., 1975, 791. so S. A. Karjala and S. M. McElvain, J. Amer. Chem. SOC.,1933, 55, 2966. Metal-ion-promoted Reactions of Organo-sulphur Compounds The kinetics of the homogeneous n-butylaminolysis of ethyl thiolbenzoate in aqueous solution over a range of pH values have recently been e~amined.5~ It is found that aminolysis proceeds via a slow reaction of the ester with the rapidly formed complexes [Ag(BuNH2)]+ and [Ag(BuNH2)2]+.Transition states such as (16) and (17) are likely, similar to that suggested by S~hwyzer.~~ The kinetic BuBu 0 benefits arising from the intramolecular nature of the aminolysis step apparently more than compensate for the inevitable loss of nucleophilicity of the N atom when co-ordinated to Ag+. How it is that Ag+ is more effective than is Hg2+ in thiolester amin~lyses,~~ whereas the reverse is true in hydrolysis,5 and how Cu2+ and Pb2+ are effective in aminolysis,47 but almost entirely without effect in hydrolysis,5 is not yet clear. It is to be remembered, however, that preparative experiments47 often involve heterogeneous systems in which powerful surface catalysis may also be present.More work is needed on the kinetics of aminolysis. The fact that the silver-ion-promoted reactions of thiolesters proceed readily at neutral pH has been exploited in a recent52 O-ester synthesis [reaction (55)], using such conditions. Another class of S-esters to have had their metal-ion-promoted hydrolyses studied kinetically are thiolbenzimidate esters (1 8). These esters are relatively basic and exist in dilute aqueous solutions of H-acids predominantly as (19). Hydrolysis proceeds slowly under these conditions, and leads53 initially to R1CsH4COSR2 [see reaction (56)], which undergoes further slow hydrolysis to R1CsH4C02H. The sulphur atoms of these thiolbenzimidate esters are apparently 61 B.Boopsingh and D. P. N. Satchel], J.C.S. Perkin 11, 1972, 1702. 6s H. Gerlach and A. Thalmann, Helv. Chim. Acta, 1974, 57, 2661. Is R. K. Chaturvedi, A. F. MacMahon, and G. L. Schmir, J. Anier. Chem. SOC.,1967, 89, 6984. SatcheII /SR2 /SR2 ] +R'C~H~-C \\ [ R1c6H4-iHR3NR3 SR2 I HlO R'CsH4C + HzO R1CsH4C-NHR3 --+ R1CeH4COSR2 + H30+ + R3NH2 (56)\\ I NHR3 'OH2 + very feebly basic, since the usually powerful silver ion has a negligible effect54 on their rate of decomposition at low pH. The addition of Ag+ does, however, affect the products that can be isolated, since in its presence5 the species R1CsH4COSR2 are very rapidly desulphurized [reaction (52)]. Addition of Hg2+ ions to a solution of a suitable thiolbenzimidate ester in the presence of hydrogen ions does lead to a considerable increase in the rate of disappearance of the ester.54 With (20) the product is benzonitrile; with (21) it is the corresponding 0-amide.The rate equations for these two reactions have the forms (57) and (58), respectively. A mechanism for (20), compatible with the reaction orders, is outlined in the reactions (59)-(64). For (21) the same mechan- SEt 7 + PhC / Ls A. J. Hall and D. P.N. Satchell, J.C.S. Perkin II, 1976, 1274. 361 Metal-ion-promoted Reactions of Organo-sulphur Compounds fast (59) NH ugz+ fast(22)+ Hg2+ PhC-S ---+ \NH S i' PhC-\;"+/-1 N Hsob+ (24) (23) + HnO fast slow + HI0 (23) +PhC NH + EtSHg+ -PhCN + HaO++ EtSHg+ slow (24) PhCN + EtSI-Igf RN\ slow (24) + Hga+ Tk,h'g+ +PhCN + EtSHg+ + Hg2+PhC fast \ fS ism is satisfactory, except that, since now only one proton is available on nitrogen, it necessarily reduces to three steps, the analogues of reactions (59),(60), and (62).Satchell These steps again account for the observed rate equation, now equation (58). For (21), the analogue of step (62) will necessarily produce the 0-amide, via attack of water on the carbon atom, rather than the nitrile via attack on the proton; again as found. The essential role of the protons bound to nitrogen in these reactions is confirmed by the total unreactivity of compound (26), which cannot lose its charge by the dissociation of a proton.The mechanisms of these imidate ester hydrolyses are analogous to those of thioamides to be discussed below. Reaction (64), which emphasizes the value in thiolester reactions of a leaving group with more than one positive charge, is reminiscent of reaction (53). The difference in reactivity between mercury and silver in the benzirnidate ester reactions is rather surprising. Gold(1rr) and thallium(r1r) ions also promote S-imidate ester hydrolyses.55 7 Reactions of Thiocarboxylic Acids and Anhydrides There has been little relevant work of any kind in this area. One kinetic study examined the hydrolyses of thiolbenzoic acid and of thiobenzoic anhydride, (PhCO)zS, in dilute solution in aqueous perchloric acid in the presence of Hgz+ and other soft ions.5 Hydrogen ions alone catalyse these hydrolyses relatively little.56 When Ag+ or Cu2+ ions are added to solutions of thiolbenzoic acid it is rapidly precipitated as the corresponding salt.With Ni2+, Cd2+, or PbZ+ there is little precipitation, but no detectable increase in hydrolysis rate. Only for Hg2+ ions is an increase in rate observed. With this metal the initial precipitation of the salt is negligible when [PhCOSH] 2 rnol 1-1 and [Hg2+] 3: 10-3 mol 1-l. The ensuing stoicheiometric reaction (65)is firstorder in [HgZ+] and in [PhCOSH], and is independent of the ratio [PhCOS-]/[PhCOSH] (which can be controlled by choosing different initial values of [H30+]). The value of AS* N 0. These results suggest the scheme outlined in reactions (66)-(69).The benefit PhCOSH + Hg2++ 3Hz0 -+ PhCOzH + HgS + 2H30+ (65) PhCOSH + H20 + PhCOS-+ H30+ fast (66) PhCOS-+ Hg2+ PhCOSHg+ fast (67) 66 A. J. Hall and D. P. N. Satchell, J.C.S. Pcrkin ZZ, 1976, 1278. 56 J. Hipkin and D. P. N. Satchell, Tetrahedron, 1965, 21, 835; J. Chem. SOC.(B), 1966, 345. Metal-ion-promoted Reactions of Organo-sulphur Compounds Hg2+ I’ PhCOSHg+ + Hg2+,?,PhCOS \ slow + 3H,O(27)+PhCO + HgS + Hg2++ fast PhCOzH + 2H30+ + HgS + Hg2+ (69) from the involvement of a second metal ion when the first-formed substrate-metal complex carries only one positive charge is again indicated. The hydrolysis of the anhydride5 occurs in two stages [reactions (70) and (71)], the second being much the slower and, as expected, having kinetics identical (PhC0)zS + 2H20 + Hg2+---t PhCOSHg+ + PhCO2H + H30+ (70) PhCOSHg+ + 2Hz0 -+ PhCOzH + HgS + H30+ with those found for the hydrolysis of thiolbenzoic acid.Reaction (70) is kinetically first-order in [Hg2+] and in [(PhC0)2S] , and may proceed via reaction (72) and then reaction (73). For Hg2+-promoted hydrolysis, under (PhC0)aS + Hg2+ _-Zr (PhC0)zS +Hg2+ fast (72) (PhCO)aS+Hg*+ + H2O PhCOSHg+ + PhCOaHaf slow (73) comparable conditions, the relative reactivities of PhCOSEt, PhCOSH, and (PhC0)sS (first stage) are ca. 1 :2.3 x lo2:1.4 x lo4. 8 Reactions of Thioamides Thioacetamide held a particular interest in the 1950’s when it was found that it could be used as an alternative to hydrogen sulphide for precipitating certain soft metal ions as their sulphides.The reactions leading to precipitation depend57 upon the pH and upon the metal. Normally, thioacetamide undergoes prior hydrolysis to hydrogen sulphide and/or thioacetic acid5’*58 [reaction (74)]. Either b7 E. H. Swift and E. A. Butler, Analyt. Chem., 1956, 28, 146; E. H. Swift in ‘Advances in Analytical Chemistry and Instrumentation’, ed. C. N. Reilly, Interscience, New York, 1960; S. Washizuka, Bunseki Kagaku, 1961, 10, 580. 68 0. M. Peeters and C. J. De Ranter, J.C.S. Perkin IZ, 1974, 1832; A. J. Hall and D. P. N. Satchell, ibid., P. 1077. Satchell MeCONHz + HzS 7 MeCSNHz + H2OI (74) MeCOSH + NH3 compound can then lead rapidly to the metal sulphide.However, under some conditions a direct reaction between metal ion and thioacetamide is detected, whose rate falls as the pH falls. The early works7 on these reactions involved heterogeneous conditions, and was primarily concerned with rates of precipit- ation. The organic product was never identified. Recently, the mechanisms of some similar reactions have been studied, using homogeneous conditions. It is fo~nd~*5~-~1 that, at 25 "C in dilute aqueous perchloric acid, under which conditions the rate of the hydrogen-ion-catalysed hydrolysis is negligible, thiobenzamide undergoes direct reactions with Hg2+, Ag+, Cu2+, and Tl3+ ions [e.g.reaction (75)], always producing the metal sulphide and benzonitrile (not the O-amide). The kinetic details differ with each metal.The mercury reaction PhCSNHz + 2H20 + Hg2++.PhCN + HgS + 2H30+ (75) will serve as an example.59 Here a 2: 1 S-amide-Hg2+ complex is formed rapidly and stoicheiometrically under all concentration conditions. With a ten-fold excess of Hg2+ over thioamide, the rate equation, at any fixed value of [H30+], takes the form (76). As [H30+] is raised, kobs falls, reaching a constant, minimum -d(2: 1 complex]/dt = kobs[Hg2+][2:1 complex] (76) value when [H30+] > 1.0 moll-l. These facts, the details of the dependencies of kobs and of the spectrum of the 2: 1 complex on [H30+], and the behaviour of N-substituted S-amide~~~ (see below) all point to the mechanism of reaction (77)-(85). The fall in kobs with increase in [H30+] cannot be attributed to S Hg2+ fast (77) 6s A.J. Hall and D. P. N. Satchell, J.C.S. Perkin ZZ, 1975, 778. eo A. J. Hall and D. P. N. Satchell,J.C.S. Perkin ZZ, 1975,953; J.C.S. Chem. Comm., 1975, 50. A. J. Hall and D. P. N. Satchell, J.C.S. Perkin IZ, 1975, 1351; ibid. 1977, 1366. Metal-ion-promoted Reactions of Organo-sulphur Compounds fast fast (29)+ HzO PhC-S-Hg+-S=C-Ph + H3Ot fast \\ /NH NHz fast (81) (29) + Hg2+-+ 2 (28) slow (82) (31) + Hg2+-(28) + (30) slow (83) (32) + Hg2+-2 (30) slow (84) + Ha0 (30) -+ PhCENH + HgS +PhCN + H30+ + HgS fast (85) reduction of attack by OH-on the thiocarbonyl group since in such a mechanism the product would have to be the O-amide. For Hg2+-promotion only the (low concentration) deprotonated 1:1 adduct (30) leads to decomposition.Con-firmation of the requirement of N-bound protons for rapid reaction comes from the study of (33), whose decomposition is, by comparison, negligibly promoted by Hg2+ ions. Satchel1 The reaction of the mono-N-substituted compound N-cyclohexylthiobenzamide displays a kinetic form c0mpatible5~ with a scheme like that of reactions (77)--(85), modified to include only one N-H ionization, and a product-forming step (86); N-substituted S-amides lead, of necessity, to the O-amide as product (cf. thiobenzimidate esters, p. 363). S PhC-N” 7 (33) NR/s-H8+PhC-+ He0 4PhC//+ HgS H,O PhCONHR + HsO+ + HgS (86)\\ \NR +OH2 The reactions of Ag+ and Cu2+ with these thiobenzamides are broadly analogous to those of Hg2+.With Ag+ only 1:l complexes are formed. For thiobenzamide the stoicheiometrically formed complex (34) loses just one N-bound proton in an equilibrium like reaction (79), and the slow steps are (87) and (88). The reaction with copper6* is similar to that of silver, except that only /s-Ag + PhC dPhCwNH + AgS-\\ NH +/i”-Ag*PhC + H2O PhCsNH + HsO+ f A&-(88)\ NH2 Metal-ion-promoted Reactions of Organo-sulphur Compounds a small concentration of (probably N,S-chelated) 1:1 complex is formed. Because of the differences in the details of the mechanisms, exact comparisons are impos- sible, but for equal initial concentrations at 25 "Cthe reactivities of the metals are Hg2+> Ag+ S Cu2+.The ions Pb2+, Ni2+, Cd2+, and TI+ have no detectable effect at 25 "C, and are less effective in decomposing thiobenzamides at moderate ambient hydrogen ion concentrations than is the hydrogen ion itself. This is probably because these ions form little 1 :1 complex and because the equilibria analogous to (79) lie far to the left. Some small degree of catalysis by Pb2+ and Cd2+ can be detected at very low %of concentrations.s The long-known11 desulphurization of thioureas [(RzN)zC=S, where R = H or alkyl] by mercury and lead species, under mildly alkaline conditions, displays a qualitative behaviour pattern with interesting parallels to the thiobeniamide reactions discussed above. Again the ability to lose N-bound protons is apparently a crucial feature. The direct reactions of Tl3+ and of AuCl4- ions with thiobenzamides in dilute aqueous perchloric acid are also of considerable interest.s1 These metal species have reactivities comparable with that of Hg2+.Under conditions involving an excess of metal ion over thiobenzamide, they both lead to effectively stoicheio- metric 1:1 complex formation [reactions (89) and (%)I. In contrast to the cases AuCla i-PhCSNHR + c1-PhC/NHR\\ S-AuC13 NHR 1a+ Tl(H&)n3+ + PhCSNHR [Phi i-Ha0 (90) S TI(Hz0)n-1 of Hg2+, Ag+, and Cu2+, the water molecules co-ordinated to TP+ undergo significant dissociation [e.g. reaction (91)] at the hydrogen ion concentrations Tl(H20)n3++ H20 + [Tl(HzO)n-1OH]2+ + H30+ (91) used, and the same is true of the water molecule in the species AuCls(H20) formed from AuC14- at low ambient chloride concentrations. In spite of such dissociations, and in spite of the behaviour found with Hg2+, Cu2+, and Ag+, the rate of decomposition of thiobenzamides by TP+ and AuCl4- is always indepen- dent of [H30+].Clearly, the prior ionization of an N-bound proton either does not occur or does not facilitate the decomposition of the complexes in these instances. Satchel1 'OH2 *OH2 slow PhC=S----bAuCls + 2H20 PhC-S-AuC13 NHR LHRI transition state 1 PhC=;)H + H30+ + AuC13S2-(92)I NHR PhCONHR AUzS 3 For the gold reaction the steps following (89)are thought to be given by reaction (92).Variation of the ambient chloride ion concentration shows61 that reaction also occurs via (35)at low and via (36) at high chloride ion concentrations, both these complexes being more reactive than (37).For the thallium reaction,61 not only is kobs independent of [H30+], but it is inversely related to the excess thallium ion concentration. This effect, not found in other systems studied so far, suggests that the reactive complex is the 2:l S-amide-T13+ species, whose low equilibrium concentration at high thallium ion concentrations falls proportionately as [T13+] is increased. For thiobenza- mide, both the 1:1 and the 2: 1 complexes contribute to the decomposition, but for N-substituted thiobenzamides thecontribution of the 2 :1complex is dominant. The reason is not obvious; the involvement of a second substrate molecule does not appear beneficial to the decomposition step in any other S-substrate system studied.The fact that N-H ionization is kinetically unimportant in the TP+-and A~Cl4~--promotedreactions permits the tertiary amide N-thiobenzoylpiperidine to react with these ions at a rate comparable to those found for thiobenzamide and N-cyclohexylthiobenzamide.The gold and thallium ions are therefore especially useful for the promoted hydrolyses of tertiary S-amides. A final point of interest concerns the AuCh2-reaction. For all the other metal ions studied the rate of formation of the initial complex between the S-amide and metal species is too fast to be studied, even by the stopped-flow method.Metal-ion-promoted Reactions of Organo-sulphur Compound& However, for AuCh2- (where CI-rather than H2O is being displaced when the complex is formed), the establishment of the pre-equilibrium (89) can be studied.62 It proceeds via a five-co-ordinate gold intermediate, and the kinetic and mechan- istic details are similar to those e~tablished~~ for the replacement of C1-by SCN- [reaction (93)]. Both the initial uptake of SCN- (i) and the rate at which (93)It Cl-SCN-+ AuCl52-[AuC15SCN]3-+AuCl3SCN-+ 2C1-it enters the square plane (ii) are catalysed by added chloride ions. These processes are, therefore, easier when octahedral complexes are involved. 9 General Summary, and Consideration of the Effectiveness of Different Metals In nearly all metal-ion-catalysed reactions, some special feature is present which renders the concentration of the metal-substrate intermediate kinetically significant.Thus most of the metaI-ion-promoted reactions discussed in this review have as an important part of their driving force the exceptional covalent affinity of the soft metal for sulphur; the size of the metal ion and the magnitude of its charge are of secondary importance. Neither soft nor hard metal species will normally promote reactions of the oxygen analogues of the sulphur-containing substrates considered, unless the oxygen-analogue can behave as a chelating ligand. All the reactions considered probably involve a rapid pre-equilibrium between the metal centre and the sulphur atom of the substrate.In those reactions (the majority) which involve subsequent cleavage of the substrate, the sulphur atom is, or is part of, the leaving group. The products therefore normally include either the metal sulphide or a mercaptide. When the sulphur atom is removed by the metal, the remainder of the substrate either loses protons (as in nitrile formation from primary thioamides and in olefin formation from thiols) or is attacked by an ambient nucleophile. This nucleophile has usually been water, it being convenient to use metal ions in aqueous solution. Most of the known reactions are therefore promoted hydrolyses. Few aminolyses have been reported. There is evidence nevertheless that attack by a variety of nucleophiles occurs readily in non-aqueous solvents, but this type of reaction has been little exploited.Virtually any thio-compound in which the sulphur atom is available to co-ordinate to a metal can be cleaved. The metal species that have been used most are those derived from silver and from mercury(r1). It is clear that, of these two metals, mercury is by far (usually 62 A. J. Hall and D.P.N. Satchell, Chent. and Ind., 1976, 373. 63 A. J. Hall and D.P.N.Satchell, J.C.S. Dalton, 1977, 1403. Satchel1 ca. lO3-fold) the more reactive under comparable conditions. Two other metals, so far rarely used, with reactivities similar to mercury(n), are thallium(Ii1) and gold(m). In quantitative studies using non-chelating thio-compounds and homogeneous solutions, all other metal ions tested have been found to lead to negligible reaction rates under conditions where silver, mercury, thallium, and gold ions react very rapidly.For example, in hydrolyses where, under similar conditions, the HgZ+-promoted reaction is usually ca. 106-fold faster than the hydrogen-ion-catalysed process, the ions Cu2+, Ni2+, Cd2+, Pb2+, and Zn2” all lead to rates either comparable with, or sometimes less than, that produced by the (hard) hydrogen ion. There is evidently another large step after silver in the reactivity scale of the soft metals. This conclusion is not compatible with various theoretical of softness, and Cd2+, in particular, seems surprisingly unreactive. Although preparative (and some kinetic) work suggests that Cu2+, Pb2+, Cd2+, and other ions can be very effective with sulphur-containing sub- strates, it seems at present probable that they are so only under heterogeneous conditions or at high temperatures, or when the substrate is a chelating species (so that a significant amount of metal-substrate complex can form), or when the metal acts as a template (thus providing an intramolecular route for the process). The successful applications of catalysts based on copper, nickel, and lead can be explained this way, but then even hard metals can be effective in such circumstances.It appears that although soft metal ions (by definition) prefer a substrate that contains sulphur to an analogous one containing oxygen, nevertheless the separ- ate hard and soft reactivity scales for acids towards a particular class of base can show considerable overlap, for, as we have seen, only four of the soft metals tested display significantly greater reactivities towards sulphur-containing substrates than does hydrogen in promoted hydrolyses.This is not the sort of result that the hard-soft principle leads us to expect, at least superficially. Although it could be urged that the comparison of the hydrogen ion reactions with the others, if close, is not exact mechanistically, it may be countered that hard-soft comparisons rarely will be exact, Many more systematic quantitative measurements with simple, non-chelating systems are needed before a reliable picture of metal ion acidity can be drawn towards either hard or soft bases.It is unfortunate that the term ‘Super-acid catalysis’, which arose from the very large catalytic effects occasionally displayed by metal ions, is sometimes used65 to describe acidic catalysis by metal ions as a whole. This usage gives the misleading impression that metaI ions always provide ‘super’ catalysis; in reality, their catalysis with both hard and soft substrates is usually feeble compared to that provided by the hydrogen ion, for the special effects which give rise to powerful catalysis are by no means always available. O4 M. Misono and Y. Saito, Bull. Chem. SOC.Japan, 1970,43,3680;G.Klopman, J. Amer. Chem. SOC.,1968,90,223;S. Ahrland, Chem. Phys. Letters, 1968,2, 303;A.Yingst and D. H. McDaniel, Znorg. Chem., 1967,6, 1067. e6 M.L.Bender, ‘Mechanisms of Homogeneous Catalysis from Protons to Proteins’, Wiley, .New York, 1971,

ISSN:0306-0012    

DOI:10.1039/CS9770600345

出版商:RSC    

年代:1977

数据来源: RSC

摘要:

Corrigendum Vol4 No 3 1975 “Co-ordination Chemistry of Aryldiazonium Cations” by D. Sutton. Section 2 Nomenclature (pp. 445, 446) :the proposed I.U.P.A.C. rule, where- by the ArNN ligand was named aryldiazenato,has not been adopted. Instead, the I.U.P.A.C. Commission now recommends reversion to the original aryl-diazenido terminology.

ISSN:0306-0012    

DOI:10.1039/CS9770600372

出版商:RSC    

年代:1977

数据来源: RSC