|
1. |
Proceedings of the Chemical Society. June 1961 |
|
Proceedings of the Chemical Society ,
Volume 1,
Issue June,
1961,
Page 185-228
Preview
|
PDF (4977KB)
|
|
摘要:
PROCEEDINGS OF THE CHEMICAL SOCIETY JUNE 1961 ONE HUNDRED AND TWENTIETH ANNUAL GENERAL MEETING THE One Hundred and Twentieth Annual General Meeting was held in the Department of Inorganic and Physical Chemistry the University of Liverpool at 9.15 a.m. on Thursday April 13th 1961. The Chair was taken by the President Sir Alexander Todd who called upon Professor A. W. Johnson Honorary Secretary to read the notice convening the meeting. Dr. J. Chatt Honorary Secretary presented the Report of Council. He referred to the Society’s steady progress during the year drawing attention to the large increase in Fellowship and particularly to the encouraging increase in the number of Fellows under the age of 27 years. He also referred briefly to the honours gained during the year by a number of Fellows.In his comments on the Society’s publications Dr. Chatt reported that despite the dislocation of work caused by the removal of the Editorial Office and by staff difficulties the arrears of publication had not increased and he hoped that the position would be greatly improved in the current year. He referred to the fact that the proportion of physical and inorganic papers published in the Journal had increased and he expressed the opinion that the various branches of chemistry were now represented in a better balance. In dealing with the Library he reported on the success of the scheme for Library Subscribers under which some 86 subscribers had already availed them- selves of the opportunity to borrow books for a modest annual fee direct from the Library.He spoke of the burden placed upon chemists by the increasing amount of literature as instanced by the continuous growth of periodicals in the Library. He also referred to the keen personal loss felt by so many Fellows of the Society at the sudden death during the year of Mr. J. Bird the Deputy Librarian who in his long service with the Society had acquired a unique knowledge of the Library. Mr. M. W. Perrin Honorary Treasurer in presenting the accounts reported that the position of the General and Publications Fund Accounts were sound. He drew attention to the increased income from Fellows’ subscriptions which were the life- blood of the Society and to the windfall from the Inland Revenue.He pointed out that the accounts showed clearly that only one-third of the total in- come on General Purposes Account had been spent on administration and that the remainder had been used to meet the expenses of meetings Local Representatives’ Expenses and Capital Improve- ments at Burlington House. He referred to the sur- plus on Publications Account which he described as not excessive bearing in mind the number of papers still to be printed and the unforeseen difficulties that the printing trade can produce. He reminded Fellows that the cost of producing the Journal was over El0 per page but added that the Council was satisfied that it was possible for the Society to continue to finance its publications without having to go capin- hand to Industry.He referred to the Trust Funds the investments of which were consolidated in a com-bined pool under a scheme prepared by the Charity Commissioners and expressed the hope that presen- 185 tation of these accounts in this logical way would make them easier to comprehend. Mr. Perrin went on to deal with the Society’s financial position during his period of office and stated that the last seven years had seen a change from a state of impending bankruptcy to solvency. He said that two major decisions made during the period of office of his predecessor the late Sir Wallace Akers had laid the foundations of this transformation ;the decision to separate the Fellow’s subscription from his entitlement of publications and the removal of the financial burden which the publication of British Chemical Abstracts had im- posed.Subsequent steps which included the presenta- tion of Accounts separating the running costs of the Society from the costs of publications enabled the Society to have better financial control over its activities. He regretted that it had been found neces- sary to terminate the Joint Subscription Scheme in 1954 but this had produced a significant saving in administrative costs whilst the introduction of the new form of Proceedings in 1957 had proved popular and had contributed to the increase in membership in recent years. The Treasurer stated that the intro- duction of substantial increases in the selling price of the Journal in 1958 and 1959 had been overdue and were the basis of the Society’s present financial stability.He referred to the change in 1957 of the financial year which now ended at September 30th a change which had improved the efficiency of the Society’s administration by spreading the volume of work and thus making it possible to keep income and expenditure under continuous review without in- crease of staff. He added that since 1959 the super- vision of the Society’s investments had been en- trusted to Messrs. Helbert Wagg and Company and that this had given great cause for confidence in investment policy. He paid tribute to the General Secretary and his staff for their work and in summing up stated that whilst in 1954 every new Fellow was a liability the Treasurer can now welcome new Fellows without reservation and that so far as could be foreseen there appeared to be no need to change the subscription rate for some years.In moving the adoption of the report the President remarked that the Society was in sound condition both scientifically and financially and acknowledged the great debt that the Society owed to Mr. Perrin for the way in which he had so skilfully handled the Society’s finances during his period of office. The motion seconded by Professor A. W. Johnson was approved unanimously. The President then announced the names of the following new members of Council elected to fill vacancies caused by retirement PROCEEDINGS Vice-president who has not filled the Ofice of President Dr.E. J. Bowen Honorary Treasurer Dr. J. W. Barrett Elected Ordinary Members of Council Constituency I (South East England) Dr. L. J. Bellamy Professor D. P. Craig Dr. W. Gerrard Mr. H. M. Powell Constituency 11 (Central and South West England and South Wales) Professor D. H. Everett Constituency 111 (North West England North Wales and the Isle of Man) Dr. A. K. Holliday Dr. J. Honeyman Constituency V (Scotland) Dr. G. 0.Aspinall Constituency VI (Ireland) Professor W. Cocker On the motion by Professor K. W. Sykes seconded by Professor A. W. Johnson Messrs. W. B. Keen and Company Finsbury Circus House London E.C.2 were appointed as auditors of the Society’s Accounts for the year ended September 30th 1961.A vote of thanks to the President Officers Council Local Representatives and the permanent officers and staff for their services during the year was proposed by Professor C. E. H. Bawn and carried with acclamation. The President in replying to Professor Bawn thanked him for his remarks and expressed his warmest appreciation of the support which he had received during the past year. He then declared the formal business of the Annual General Meeting terminated. After an interval the meeting was resumed and 53 Fellows were formally admitted to the Society. The President then presented the Corday-Morgan Medal and Prize for 1959 to Dr. A. R. Battersby in recog- nition of his outstanding work on the stereochemistry of emetine and its congeners and his contributions to the chemistry of curare alkaloids and the bio- genesis of papaverine.The President then gave his Presidential Address entitled “Some Current Problems in Polynucleotide Chemistry.”* At the conclusion of the meeting a vote of thanks to the President for his address proposed by Professor G. W. Kennet was carried with acclamation. * See page 187. JUNE 1961 187 PRESIDENTIAL ADDRESS* Some Current Problems in Polynucleotide Chemistry By Sir ALEXANDER TODD IT is now almost ten years since the general chemical structure of the ribonucleic acids (RNA) and the deoxyribonucleic acids (DNA) was recognised and a satisfactory explanation of their hydrolytic behaviour became avai1able.l Since that time a great deal of work-chemical biochemical biophysical and bio- logical-has been carried out in the polynucleotide field and this seems an appropriate time to review the position from a chemical standpoint and to identify some at least of the major problems which call for attention if we are to progress further in our understanding of the function of these remarkable compounds in Nature.For although at times popular reports about their importance have been rather sensational in character there is no doubt that in the structure of the nucleic acids lies the key to many of the problems of genetics and indeed of the growth development and reproduction of living cells. Clear evidence that nucleic acids can and do act as carriers of genetic continuity is available from work on the transforming factors of bacteria (DNA) and from studies on bacteriophages andviruses (DNA and RNA).Particularly impressive in this connection are the beautiful experiments of Fraenkel-Conrat and of Schramm on the tobacco mosaic viruses.2 These workers have not only shown that it is the RNA of the virus nucleoprotein that is the infective agent and the carrier of genetic continuity but by chemical modification of the RNA with nitrous acid they have brought about mutations of the virus. Genetic varia- tion in viruses and in living cells seem therefore to be due to differences in the structural pattern of nucleic acids. It is of interest that although in the viruses both DNA and RNA can carry genetic con- tinuity this property in living cells seems on present knowledge to be restricted to DNA; RNA is of course present in living cells but it appears there to play a subsidiary role to the chromosomal DNA its production being in some way controlled by the latter although RNA itself plays a major role in the synthesis of proteins and protein enzymes in the cell.Even from this brief statement of the position it is clear that some consistent replication procedure must exist for the nucleic acids and through them for other cell constituents and furthermore that there must be an enormous number of individual nucleic acids to account for the almost infinite variety of Nature. It is therefore pertinent to ask whether the structural picture as we now see it is capable of accommodating these requirements.I believe that incomplete as it is in a number of ways present knowledge suggests that these requirements can indeed be met. All subsequent chemical work has reinforced the original postulate of Brown and Todd’ that both types of nucleic acid are 3’,5’-linked polynucleotides. They are polydiesters of phosphoric acid and there is no evidence that they contain any triester group- ings or that they are other than linear polyesters. Chain branching is theoretically possible in poly- ribonucleotides but as yet no unequivocal example of a branched-chain RNA has been reported. Until about ten years ago it was believed that only four nucleosides were present in RNA’s (adenosine guanosine cytidine and uridine) and four in DNA’s (deoxyadenosine deoxyguanosine deoxycytidine and thymidine) but since then a number of other nucleosides have been detected in a variety of natural nucleic acids of both types.These “unusual” or “abnormal” nucleosides have been found in sub- stantial amount in only a few instances; as a rule they occur in relatively small amount in preparations of nucleic acids from a variety of sources but frac- tionation experiments indicate that in many cases they may be significant constituents of some com- ponents of the nucleic acid preparations used. For it should be remembered that in all cases known natural nucleic acids as isolated are not homogeneous in- dividuals but mixtures of polynucleotides.Broadly speaking however the unnatural nucleosides (of which about half a dozen are known) are quite closely related to the major nucleosides mentioned above and for the purpose of our present exposition of the physical structure of the DNA molecule and its implications we can regard DNA as effectively built up of four nucleosides. As to the number of different DNA’s which are possible on this basis it will only be noted that if current estimates of mole- cular weight (10-20 million) are valid then the possible number of individual DNA’s differing from one another in sequence of nucleoside residues and size is almost astronomical. Preparations of sodium deoxyribonucleate can be drawn into fibres permitting X-ray crystallographic examination and a study of them by this means has revealed that undenatured DNA (except for some phage DNA’s) normally exists as a kind of double * Delivered before The Chemical Society’s Anniversary Meeting at Liverpool on April 13th 1961.Brown and Todd J. 1952 52. For general review see Schramm Proc. 11th Solvay Conference Interscience Publ. Inc. New York 1960. PROCEEDINGS ~ ~~ ~~ molecule consisting of two opposed helical poly- nucleotide chains about a dyad axis. The two chains separate in the first phase of denaturation and may recombine to reform the original double helix. The general form of the DNA molecule is too well known to require detailed description here; we will merely note that the two helical strands are formed by the sugar residues of the nucleosides linked by phosphate groups the attached heterocyclic nuclei lying within the double helix at right angles to the main axis.The stability of the system is considered as a result of a brilliant suggestion by Watson and Crick,3 to be due to hydrogen bonding between the heterocyclic bases in the two chains; Watson and Crick suggest that this hydrogen bonding is of a very specific type adenine pairing only with thymine and guanine with cytosine. I cannot here go into the detailed evidence for this view but it will suffice to say that there is much in its favour and that it has been widely accepted and used by biologists and others as pro- viding at least a basis for understanding some of the functions of nucleic acids.For it is clear that the Watson-Crick hypothesis offers an explanation of how DNA could be a self-replicating molecule and so provide for genetic continuity. Since the two poly- nucleotide chains making up the double helix must because of the obligatory base-pairing be comple- mentary it would be expected that if the two chains were separated and placed in a medium where active polynucleotide synthesis was going on then each would build on itself a new complementary chain thereby regenerating the original double molecule. In this way each chain would be a kind of template carrying a code of genetic information from one generation of cells to the next and there is at least some microbiological evidence which supports this view? By extrapolation the same template idea is usually carried over to cover the functions of nucleic acids in the synthesis of proteins and much mathe- matical ingenuity has been expended on trying to elucidate the nature of the polynucleotide “code” which would permit precise selection from twenty or so different amino-acids.But I doubt myself whether we have yet enough knowledge of the processes in- volved to derive such a code by simple application of information theory. A final point to note is that all the biophysical work I have mentioned refers strictly to DNA. The physical picture is by no means so clear in the case of RNA where detailed X-ray analysis has not yet been possible and there is no definite evidence that any natural RNA exists in anything but a single-stranded form.On the basis of present knowledge it is abundantly clear that the nucleic acids occupy a central position in the economy of living cells and the extension of our knowledge both chemically and biologically represents one of the most challenging fields of scientific investigation at the present time. There is of course a multitude of problems awaiting study but I would select three as outstanding-outstanding in the extent of the repercussions that their solution would have on the field. These are sequence deter- mination in polynucleotides synthesis of poly-nucleotides and elucidation of the mechanism of protein synthesis through intervention of poly-nucleotides. These are essentially chemical problems and I shall endeavour in the course of this address to discuss at any rate the first two.Sequence Determination in Polynuc1eotides.-Since the difference between specific polynucleotides must rest primarily in their differing arrangements of nucleotide residues the discovery of methods whereby nucleotide sequence in polynucleotide chains can be determined is obviously a matter of great importance. No such method can however be applied until homogeneous nucleic acids are available for study. As has been mentioned previously nucleic acids obtained from natural sources are manifestly hetero- geneous and no method is yet known by which pro- ducts which are truly homogeneous can be obtained. It cannot be too strongly emphasised that methods which can give homogeneous materials are desperate- ly needed and offer a great challenge to physical chemists interested in macromolecular separations.Success is most likely to be achieved in the first instance with the so-called “soluble RNA” or with the virus nucleic acids which on current views should be chemical individuals heterogeneity in the isolated materials being presumably due to a variable amount of degradation during the process of isolation. In protein studies the sequence of amino-acids in quite large polypeptides has been determined-a notable triumph in this direction being the sequence determination of the protein sub-unit of tobacco mosaic virus containing 158 amino-acid residue^.^ Generally speaking the method employed in poly- peptide studies rests on disruption of the chain into smaller pieces by use of specific enzymes and the determination of the structure of these smaller pieces by partial hydrolysis and stepwise degradation.The piecing together of the information so obtained depends upon the overlaps between the various frag- ments obtained by disrupting the original polypep- tide chain. The problem is very much more com- plicated in the case of polynucleotides. For one thing the small number of individual nucleotides in a given nucleic acid as compared with the much larger number of different amino-acids usually found in Watson and Crick Nature 1953 171 737. E.g. Meselsohn and Stahl Proc. Nat. Acad. Sci. U.S.A. 1958 44 671.Tsugita Gish Young Fraenkel-Conrat Knight and Stanley Proc. Nut. Acad. Sci. U.S.A. 1960 46 1463. JUNE 1961 proteins clearly gives much less scope for the pro- duction of reasonably large fragments with struc- turally useful overlaps by enzyme action. In addition it must be admitted that we have as yet few specific enzymes that can be used for this purpose. In the case of polyribonucleotides we have available such en- zymes as pancreatic ribonuclease which is specific for internucleotidic linkages attached to C,. in pyrimidine nucleoside residues.6 In recent years too other types of ribonucleases with rather different specificities have been reported such as the enzyme of Sat0 and Egami’ which splits guanosine-3’ phos- phate linkages; this suggests that enzymes specific for each of the four nucleosides and for 3’- or 5’-linkages may become available in due course.For polydeoxyribonucleotides we have deoxyribonuclease whose specificity is unfortunately ill-defined despite some interesting work on the subject by Khoranas using small synthetic polydeoxyribonucleotides. In addition to these we have such enzymes as spleen and venom phosphodiesterases which can be applied to either type of polynucleotide; these show no particu- lar nucleoside specificity but split internucleotidic linkages at Cgp and C3’ respectively and seem to attack at least small polynucleotides stepwise starting at one end of the chain. Use of these enzymes so far has yielded-in addition to end-group determination-only rather gross information indicating that different nucleic acids differ in their sequence and that there appears to be little observable regularity in their make-up.Some- what similar information has been obtained in the DNA field by chemical degradation via the so-called apurinic and apyrimidinic acids and by the study of DNA hydrolysis with formic acid in presence of di- phenylamine. There are a number of indications in the literature suggesting that linkages between different nucleotide pairs in RNA vary in their ease of hydrolysis with alkali; whether such differences could be exploited for sequence determination is as yet unknown. Non-enzymic methods for the stepwise degrada- tion of polynucleotides have also been sought and so far two potentially useful procedures have been reported-ne applicable to polyribonucleotides only and the other although devised with polydeoxy- ribonucleotides in mind theoretically applicable to both types.Such stepwise methods of degradation are likely to be valuable when applied to oligo- nucleotides such as might be obtained by preliminary enzymic breakdown of nucleic acids rather than on the intact acids themselves; but it must be stressed that until such oligonucleotides can be prepared and separated in a homogeneous state stepwise degrada- tion methods will be of very restricted value. The most thoroughly studied of the stepwise degradation procedures is that which depends on the periodate oxidation of the 2’,3’-glycol system in the terminal nucleoside residue of a polyribonucleotide chain.g The product of oxidation (which can only affect the terminal group) is a dialdehyde bearing in the /?-position to one of the carbonyls (i.e.at posi- tion 5’) the terminal internucleo tidic phosphate group. Under basic conditions which do not affect the normal internucleotidic linkage (i,e. at pH 9-10) the oxidised terminal residue is eliminated leaving a polynucleotide containing one less nucleo- side residue than the starting material but carrying a phosphate group at C, on the terminal residue. Removal of this phosphate group with a phospho- monoesterase provides a polynucleotide on which the whole process can be repeated. This method has been applied successfully to several small oligonucleotides isolated from partial hydrolysates of ribonucleic acids and the recent work of Ogur and Smalllo who have modified and developed the method on a quantitative basis has greatly enhanced its potential value as a stepwise degradation procedure.An interesting adaptation for end-group and molecular-weight determinations has also been introduced by Dulbecco and Srnith’l who “mark” the oxidised terminal residue by condensation of the liberated aldehyde groups with carbonyl reagents labelled with radioactive isotopes. The absence of a 2’-hydroxyI group from deoxy- ribose precludes the use of periodate oxidation methods in the polydeoxyribonucleotide series but recent work on the oxidation of nucleoside deriva- tives suggests at any rate a possible basis for develop- ing a degradation procedure applicable to the DNA series.Catalytic oxidation of nucleosides and nucleo- tides by molecular oxygen in the presence of platinum converts the terminal CH,-OH group in the sugar residue into carboxyl.l2 Although in such an oxidised nucleoside-3’ phosphate the phosphate group is attached to a carbon atom in the /%position with respect to the carboxyl group elimination does not occur readily; further work on this topic is being pursued since it is probable that elimination could be effected much more readily after conversion of the Brown Dekker and Todd J. 1952,2715; Brown and Todd J. 1953 2040. Sato and Egami J. Biochem. (Tokyo) 1957 44 753.Khorana Symposium on Enzyme Reaction Mechanisms J. Cell. Comp. Physiol. 1959 54 Suppl. 1 5. Brown Fried and Todd Chem. and Id. 1953 352; J. 1955 2206; Whitfield and Markham Nature 1953 171, I 151 ;Whitfield Biochem.J. 1954 58 390. lo Ogur and Small J. Biol. Chem. 1960 235 PC60. l1 Dulbecco and Smith Biochim. Biophys. Acta 1960,39 358. l2 Reese Schofield Shapiro and Todd Proc. Chew. Soc. 1960 290. PROCEEDINGS free carboxyl into an ester or amide group. Given such an elimination reaction the catalytic oxidation procedure might be applied to oligonucleotides in a manner analogous to the periodate oxidation method already described. But much remains to be done before the usefulness of the method can be estab- lished; for the moment it must be regarded as an interesting possibility only.that goal synthetic work in the field is necessary for a variety of reasons. For one thing a “library” of oligonucleotides of known structure would be of immense value in connection with sequence studies and the definition of enzyme specificities for physico- chemical studies and possibly also in therapeutic work. Again the production of larger polynucleo- tides of known structure would help towards a -Base--C,--C,--C,~ - P/ Base-CHCHO - OHCCH-CH P / <s@ Base-C,~--C,t--C,~ /-/P -+ Base--C,--C3t-C,. /-/ n n Base--C,~--C,~-C,~ P/ P P Base-C~-C3-- / - P‘ - P /-B~s~-C,.-C~~-C, P/ f- B~s~-C,~-C~+--C,~ -/-n-1 Base4,!-C ,,--c5 From this brief review of the current position of work on sequence determination it is evident that slow progress is being made although we are still far from the ultimate goal.The problem is indeed a formidable one and may call for a combination of biological physical and chemical methods before the sequence of even one of the smaller nucleic acids can be elucidated. It cannot be too strongly emphas- ised however that one of the greatest stumbling blocks to progress is the lack of adequate methods for separating mixtures of oligo- and poly-nucleo- tides;the discovery of really effective separation procedures would probably do more than anything else to open up the field. Synthesis of Oligonucleotides and Polynucleotides. -The artificial synthesis of physiologically active polynucleotides is of course an attractive if still distant prospect for the chemist.But quite apart from -/p Jn-1 Base-C,-C -C,t P/ clearer understanding of some of the more remark- able features of nucleic acid behaviour-problems such as base-pairing in helical structures denatura- tion etc. It is not my purpose here to discuss enzymic synthesis of polynucleotides but it should be noted that as yet only enzymic methods have been able to provide reasonably large polynucleo tides. The bril- liant work of Ochoa and his on the poly- nucleotide phosphorylase of micro-organisms which synthesises polyribonucleotides from nucleoside-5’ pyrophosphates and of Kornberg14 on the enzymic synthesis of polydeoxyribonucleotides from deoxy- nucleoside-5’ triphosphates is well-known and has provided very effective synthetic methods for the larger molecules.Enzymes synthesising polyribo- nucleotides from nucleoside-5’ triphosphates are also available. The stepwise synthesis of small oligonucleotides la For review see Ochoa Nobel Lecture reproduced in Angew. Chem. 1960,72,225. 14 For review see Kornberg Nobel Lecture reproduced in Angew. Chem. 1960,72,231. JUNE 1961 (di- and tri-) has been achieved by applying in the nucleotide field the general phosphorylation pro- cedures for alcohols developed during the past fifteen years or so and these methods could in theory be extended to higher members of the oligonucleo- tide series. That they have not been more extensively used is undoubtedly due to the extreme technical difficulty of work in this field which makes stepwise synthesis a very laborious matter.Before discussing the actual chemical procedures employed it is worth- while to note the general nature of these difficulties since they are common to all presently known methods. The major problems are associated with the need to work in essentially anhydrous solvents in which nucleotides and even more so oligonucleo-tides are notoriously insoluble and with the Iabor- ious methods which are employed to separate and isolate pure products from the complex reaction mixtures usually obtained. Additional if less acute difficulties frequently arise from the acid-lability of the glycosidic linkage notably in the purine deoxy- ribonucleotides and the reactivity of amino-groups in the heterocyclic nuclei.The solubility problem has been to some extent overcome by using nucleotides in the form of their salts with complex organic bases and in some instances by using also non-hydroxylic ionising solvents such as dimethylformamide but there is still much scope for improvement. The separation of reaction products which rapidly becomes a limiting factor on synthesis as one ascends the oligonucleotide series is nowadays usually effected by ion-exchange chromatography on modi-fied celluloses (ECTEOLA types). The introduction of the modified celluloses has vastly improved isola- tion procedures in this field of work and indeed it is probable that the apparently better results achieved in some of the most recent work owes at least as much to this factor as to any improvement in the reaction procedures; but the development of more refined techniques of separation remains an outstanding problem.All oligonucleotide syntheses are similar in principle in so far as they amount to the condensa- tion of a suitably protected nucleotide derivative which is an active phosphorylating agent (and this may be prepared in situ by “activating” a nucleotide derivative) with a second nucleotide or nucleoside in which only the hydroxyl group intended to parti- cipate in the internucleotidic linkage is unprotected. After condensation protecting groups are removed to give a dinucleotide. Alternatively a dinucleoside phosphate may be prepared by an analogous pro- cedure and the second (or terminal) phosphate group introduced into it by simple phosphorylation.l5 Michelson and Todd J. 1955 2632. l6 Michelson Szabo and Todd J. 1956 1546. The phosphorylation methods which have been used to form internucleotidic linkages fall into two clearly defined groups (a) those in which the active reagent derives from a diester of phosphoric acid and which yield initially triesters and (b) those in which the reagent derives from a monoester of phosphoric acid and the product is a diester. These two groups differ fundamentally from one another and it is useful to discuss them first in general terms. Methods of type @.-These methods are of the “acid anhydride” type and are analogous to the common acylation procedures employed in the carboxylic acid series.The reagents are of the general type (RO),POX and include phosphorohalidates (e.g. dibenzyl phosphorochloridate) tetraesters of pyrophosphoric acid or mixed anhydrides of di- esters of phosphoric acid with other stronger acids (e.g.,arene- or alkane-sulphonic acids). In general in mixed anhydrides derived from phosphodiesters it is the weaker acid of the two components of the an- hydride which acylates the main driving force being the expulsion of the component yielding the more stable anion. Thus Pl-diphenyl P2-dibenzyl pyro- phosphate and the mixed anhydride of dibenzyl phosphate and benzenesulphonic acid will each yield with an alcohol in presence of base the di- benzyl alkyl phosphate as main product.The non- additive nature of the P=O group ensures that this result will be obtained even when the nucleophile employed reacts with the un-ionised reagent (as well normally be the case); under such circumstances carboxylic anhydrides normally react to acylate with the stronger of the two component acids. The first reported oligonucleotide syntheses were of this type dithymidine-3’-5’-dinucleotide15and adenosine-2’ uridine-5’ phosphate,16 being synthes- ised by condensation of appropriately protected nucleotide or nucleoside benzyl phosphorochlori- dates with the appropriate protected nucleosides. This type of method has a drawback when applied in the ribonucleotide series to form either a 2’,5’-or 3’,5’-internucleotidic linkage in so far as any loss of the protecting group at 3’ or 2’ respectively before conversion of the initially formed triester into diester will yield a product extremely labile to hydrolysis (phosphotriester with vicinal &-OH).But methods of type (a)have on the other hand the theoretical advantage that structural variation of the phos- phorylating entity could yield variation in selectivity. Methods of type @).-In most of the earlier work on the phosphorylation of alcohols reagents based on phosphomonoesters were avoided since it was known that such compounds as P1P2-dialkyl pyro- phosphates do not act as phosphorylating agents presumably because the negative charges on the anion prevent attack on phosphorus by nucleophiles.But it is now well-known that conversion of a phosphomonoester into a mixed anhydride with a strong acid such as a sulphonic acid or a diester of phosphoric acid or treatment with an excess of one of the reagents which are known to convert them into pyrophosphates via imidoyl phosphate inter- mediates (Le. carbodi&nides keten hides tri-chloroacetonitrile etc.) all yield products which will phosphorylate alcohols effectively in the absence of other nucleophiles. In all these cases the weight of evidence is in favour of the view that the effective reagent is a monomeric metapho~phatel~ and that these various procedures are only different methods for converting a monoalkyl phosphate into the cor- responding alkyl metaphosphate which in the (hypo- thetical) monomeric form is and must be a powerful phosphorylating agent.It will be recalled that forma- tion of monomeric metaphosphate has been postu- lated in the hydrolysis of phosphomonoesters.l* The annexed scheme indicates the general picture. 0 R’-S02X I1 + RO-P--O.SO,R’ I 0-R’N= C= NR’ 00 etc. It I1 RO-P-0-P-OR I I RO*P\ PROCEEDINGS procedures by methods of type (b) can hardly be expected. They have however the merit that they yield diester linkages directly. It is clear that in principle at least any of the methods used for oligonucleotide synthesis could be employed in a polycondensation procedure to yield polynucleotides. So far only methods of type (b)have been examined in any detail perhaps because the starting materials particularly in the deoxyribo- nucleotide series are much more readily accessible than those needed for type (a)reactions.Thus deoxy- ribonucleotides give polymeric materials with carbodi-imides trichloroacetonitrile sulphonyl hal- ides dialkyl enol-phosphates diary1 phosphoro- chloridates and even phosphoryl ch10ride.l~ In all cases complex mixtures of oligonucleotides and poly- nucleotides containing up to 10-12 nucleotide units are produced. That higher polynucleotides do not seem to be formed to any noticeable extent is doubtless due to a combination of solubility diffi- culties and the fact that reactions of the type used are not by their nature obligatory chain-forming Bas< -+ RO*P02+ -O*S02R‘ Base -42RO.POZ fR‘.NH*CQ.NHR’ .0-O-C-NHR’ 0-II K’-NH+ 0 / (R’O),P 0 0 OR’ \ X 11 /I/ + RO-P-0-P I \ 0-OR’ Various methods of this type have been applied successfully to the preparation of small oligo-nucleotides and in any given case the advantage which one method may show over another probably resides mainly in accessibility of the reagent and the nature of the intermediates used. If the general mechanism outlined above is accepted it is clear that the active phosphorylating agent is in all cases the same so that the development of highly selective 0 Base il RO.PO2 + -O-P(OR’) reactions and therefore tend to produce as major products small oligonucleotides.An important side reaction which adversely affects such polycondensa- tion processes is the formation of cyclic oligo- nucleotides (notably di- tri- and tetra-) by esterifica- tion of the terminal primary phosphate group with the free hydroxyl on the sugar residue at the other end of the chain. As a result overall yields of poly- nucleotides by such methods are extremely low l7 Todd Proc. Nut. Acad Sci. U.S.A. 1959 45 1389. l8 Butcher and Westheimer J. Amer. Chem. Soc. 1955 77,242U. lS For a recent review see Cramer Angew. Chem. 1%1,73,49. JUNE 1961 despite improvements made by Khorana and his colleagues20 in separation methods and in minimising cyclic oligonucleotide formation by polymerising mixtures of nucleoside-5’ phosphate with up to 25 % of 3’-acylnucleoside-5’ phosphate chains based on the latter as end-group being protected against cyclic oligonucleotide formation.The production of oligo- and poly-ribonucleotides is rendered more complicated than that of the cor- responding deoxyribo-compounds by the presence of 2’-and 3’-hydroxyl groups. This means that before unambiguous production of oligonucleotides con- taining the 3’,5’-internucleotidic linkage character- istic of the natural ribonucleic acids can be realised it is necessary to have a convenient method of pro- tecting the 2’-hydroxyl by some group which can later be removed by very mild acid treatment. A neat solution to this problem has been provided by Smith and Khorana21 who prepared the 2’4-tetrahydro- pyranyl derivatives of nucleoside-3’,5’ cyclic phos- phates and subsequently opened the cyclic phosphate group with alkali giving a separable mixture of 2’-O-tetrahydropyranyl3’-nucleotide and 2’-O-tetra- hydropyranyl 5’-nucleotide ; these can be used for synthesising 3’,5’-linked polyribonucleotides since the tetrahydropyranyl group is readily removed by very mild acid treatment.An ingenious procedure leading to mixtures of 2’,5’-and 3’,5’-linked polyribonucleotides has been devised by Michelson.22 It involves reaction of ribo-nucleoside-2’,3’ cyclic phosphates with diphenyl phosphorochloridate; this reaction which may be fundamentally a type (b) procedure is simple in operation and seems to give somewhat larger poly- nucleotides than the other methods reported but the lack of physical separation methods for 2’,5‘-and 3’,5’-linked polynucleotides severely limits its use at present.When one surveys the present position of poly-nucleotide synthesis it is evident that here as in the problem of sequence studies the lack of really efficient separation methods is perhaps the major stumbling block to rapid development. But the prob- lem of solubility is almost as important and if it were possible to devise a suitable phosphorylation pro- cedure which would operate in aqueous media the difficulty of polynucleotide synthesis would be con- siderably reduced. For the development of polymeri- sation methods much beyond their present stage it would seem necessary to devise a true chain-forming procedure such as is available in the Leuchs method of polypeptide synthesis.At present however with enzymic procedures available to produce poly- nucleotides of substantial chain-length it seems that chemical methods of synthesising smaller oligo- nucleotides of known sequence are those likely to be of most practical value. A start has been made and although the difficulties are formidable there is no reason to believe that with further development chemical synthesis will not play its full part and help open the way in concert with biological and bio- chemical work to a clearer understanding of the function of nucleic acids in Nature. 2o Khorana Fed. Proc. 1960 19,931. 21 Smith and Khorana J. Amer.Chem. SOC.,1959 81 291 1 a2 Michelson,J. 1959 1371 3655. THE ANNIVERSARY MEETING AT LIVERPOOL APRIL 11-14th MIXINGwith those attending the Society’s meeting at Liverpool in April one heard general approval. The accommodation in the halls of residence was com- fortable and the food excellent. The social events were enjoyable and the scientific events popular. It was indeed a feature of the year that the two symposia (which are fully reported elsewhere in this issue) were fully attended to the capacity of the large lecture theatres and at times beyond it up to the last day; those who had taken the trouble to prepare the addresses earned the gratitude of their fellows. The Presidential address (see p. 187) was on a highly complex organic topic but was so graded that every chemist could follow its broad sweep and general survey-there was no bogging down in detail.The Society’s Annual General Meeting was perhaps most notable for the success story that the Honorary Treasurer could tell before relinquishing his office the Society had come through the period of reliance on grants and is now fully self-supporting with even the prospect of being able to face the prob-able continuance of rising prices with some equani-mity and enough left over to consider ncw ventures. If this has entailed that many Fellows can no longer buy their own copies of the Journal that is regret- fully only the general trend of the present day. The continuing rapid expansion in membership shows that what the Society has to offer is nevertheless increasingly attractive.Much hospitality was enjoyed. The visits to works were widely attended; the Reception Buffet and Dance by invitation of Unilever Limited at Hulme Hall Port Sunlight was immensely appreciated by a largeparty :and theluncheon given bytheDirectors of Imperial Chemical Industries General Chemicals Division introduced us not merely to a feast but also to the most interesting Bluecoat Chambers. The Chemical Society’s Anniversary Dinner was well patronised Satisfactory gastronomically and livened by speeches that rose to the occasion. The environs of the Chemistry Department at PROCEEDINGS Liverpool do of course leave room for improve- ment but the perceptive who went to the new Cathedral were well rewarded and cannot fail to have been impressed by the new springing arch.All those at Liverpool who contributed to making the visit a success can justifiably feel happy that their hard work earned its reward in the gratitude of the large number who were there. CHEMICAL SOCIETY ANNIVERSARY MEETING 1961 SYMPOSIA * Developments in the Chemistry of Boron Compounds DURINGthe Anniversary Meeting of the Society a most successful symposium was held in three sessions to discuss recent advances in Boron Chemistry. In spite of a tendency on the part of certain speakers to discuss mainly published and therefore familiar work many new and fascinating aspects of the subject were revealed. The symposium was opened by the chairman Professor C.E. H. Bawn. After welcoming the Society to the Chemistry Department of Liverpool University and announcing a change in the pro- gramme due to the absence of Professor Mikhailov (Moscow) the chairman introduced the first speaker Professor Anton B. Burg from the University of Southern California. Professor Burg who had chosen as his title “Some New Problems concerning Boron Hydride Deriva- tives,” presented a detailed lecture describing the effect of bond structure on the reactivity of certain groups in the higher boron hydrides. On this basis Professor Burg masterfully explained the formation of polymers from B4H1 and B5Hll the production of intermediates such as B3H and B,H, and the synthesis of B,H, and B9Hl,.Advancing to the substitution products of the boron hydrides Professor Burg covered a host of reactions such as the formation of the six-membered rings (Me2N,BH,) and (Me,P,BH,),. Just one of the many interesting problems raised in this section of the address was the structure of (Me,N),B,H,, for which Professor Burg suggested a tetrahedron of boron atoms with opposite edges formed by nitrogen or hydrogen three-centre bonds. As a final section to the lecture Professor Burg pointed out how the important synthetic tech- nique of hydroboronation of carbon-carbon multiple bonds could be extended to such interest- ing compounds as (CF,),PCH =CH-P(CF,), CF3-P-[CH=CH-P(CF3),I2 and (CF3),P-C_C-P(CF.J2. The second lecture was delivered by Professor H.Steinberg from Anaheim California. Entitled “The Chemistry of Diboron Compounds,” the lecture dealt exclusively with compounds containing a boron-boron sp2 hybrid covalent bond with no bridging groups. The study of diboron compounds has hitherto been restricted by the fact that the usual preparative methods yield only small quantities so that the description by Professor Steinberg of re-actions giving good yields of diboron compounds on a macro-scale must rank as one of the notable features of the meeting. The preparation of tetra- kisdimethylaminodiboron by treatment of bisdi-methylaminohalogenoboranes with highly dispersed molten sodium in an inert solvent and of diboron tetrachloride by the action of boron trichloride on boron monoxide at 200” were reported.The re- actions of these and related diboron compounds were then described from the viewpoint of rupture or retention of the boron-boron bond. Also on the subject of diboron compounds Dr. A. K. Holliday of Liverpool University continued the symposium with a lecture entitled “Diboron Tetrahalides.” Dealing principally with diboron tetrachloride Dr. Holliday capably described the variety of different addition reactions which this compound undergoes with conventional ligands. The highlight of the lecture concerned the .rr-bonded addi- tion of olefins to diboron tetrachloride which addi- tion is followed by insertion of the olefin between the two boron atoms yielding a B-C-C-B system. The reactions of this system towards oxygen and amines were shown to be remarkably similar to those of the original diboron compound the presence of the C-C bonding making the corresponding adducts more stable.Formation of a similar 7r-bonded adduct followed by insertion was described as occurring during the autoxidation of diboron tetrachloride and during the reaction with nitric oxide. The surprising properties of a green compound B,Cl,NO were * Held at Liverpool University April 12-14th 1961 JUNE 1961 described and apparently a similar compound is not formed from diboron tetrafluoride. A general discussion on the preceding papers followed. Three topics of importance were discussed the synthesis of diboron compounds in reasonable quantities the exact nature of the bonding in adducts of diboron compounds and the two forms of boron monoxide.Professor Burg and Dr. Turner (National Chemical Laboratory) were interested in the possibility of pre- paring diboron tetrachloride from tetra-amino- or tetrahydroxy-diboron but according to Professor Steinberg no success has been achieved in such reactions. The possibility of the formation of boron subhalides in the reaction between boron monoxide and boron trichloride was raised by Dr. Finch of Holloway College and Professor Steinberg reported some evidence for their formation. The discussion then swung to the structure of certain derivatives of diboron compounds. Dr. Aylett of Aberdeen University raised the question of the structure of the [B2C1J2- anion and Professor Burg put forward the possibility of a mixture of [B,CIQI2- and [B,C1,]2-.On this question Dr. Waddington of Cambridge University stated that the same similarities existed between the infrared spectra of the tetramethylammonium salt of the anion and the BC1,- ion as exists between the spectra of hexa- chloroethane and carbon tetrachloride. On the sub- ject of adducts formed from olefins and diboron tetrachloride Dr. Sharp of Imperial College queried the existence of a four-centre v-bonded system sug- gesting simple donation to one of the boron atoms. Dr. Holliday and Dr. Davidson (Akers Labora- tories) both thought this unlikely because of the chemical reactions of the adduct and on grounds of molecular-orbital overlap.Furthermore Dr. Lappert of the Manchester College of Science and Tech- nology reported that he had been unable to find evidence for the co-ordination of cyclohexene to boron halides. The discussion finally dealt with the two forms of boron monoxide which exhibit quite different chem- ical activity. Despite contributions by Dr. Nicholls Liverpool University Professor Stein berg and Dr. Anderson of Imperial Chemical Industries Limited it was evident that knowledge of this most interest- ing compound is very far from complete. The second section of the meeting opened with Dr. D. B. Clapp in the chair and the first lecture “Synthesis and Reactions of Organic Boron Hetero- cycles,” was presented by Dr. R. Koster from Miilheim.The main synthetic routes described were (a) displacement reactions between trialkyl boranes and di- or tri-olefins (b) hydroboronation of di- or poly-olefins and (c) pyrolysis of trialkylboranes and alkyldiboranes. The property of these boron hetero- cycles stressed by Dr. Koster was their ability to undergo thermal isomerisation leading to changes in ring size a reaction which is catalysed by traces of alkylaluminiums or by the presence of boron- hydrogen bonds. This reaction which is also charac- teristic of simple alkyl boranes was shown to have potentialities as a synthetic route in organic chemistry. Dr. Koster finally surveyed the formation of cyclic alkyldiboranes and pointed out how these com-pounds did not undergo characteristic reactions such as causing the polymerisation of diazomethane or exchange reactions in other alkylborons.In the next lecture “Thermochemistry of Boron- containing Addition Compounds,” Dr. N. H. Greenwood of Nottingham University showed how the acceptor strength of boron halides increased F < Cl < Br owing to decreasing back-donation from filled orbitals on the halogen to the vacant boron orbital. Dr. Greenwood then surprised many people by quoting the unusual system liquid boron trihalide-liquid inert gas as an example of two com- ponents capable of addition. Since the ionisation potentials of krypton or zenon were suitable for complex-formation with boron trichloride it was with regret that the audience heard experimental evidence against the existence of such complexes.An exciting new compound was introduced to the audience in the third lecture of the session “Poly- meric Borazynes,” by Dr. H. S. Turner (National Chemical Laboratory). The lecture dealt solely with the tetramer formed when one attempted the syn- thesis of borazoles containing bulky substituents on the nitrogen atom. These compounds such as @utNBC1) were described as being much more stable than similar borazoles. Dr. Turner pointed out however that only those borazynes which did not form borazoles gave these derivatives. Evidence was presented suggesting an eight-membered ring structure for these compounds. Dr. M. F. Lappert of the Manchester College of Science and Technology in a lecture entitled “Some Displacement Reactions of 3-Co-ordinate Boron Compounds,” introduced a kinetic aspect when he considered steric effects upon the rates of the four steps constituting the nucleophilic displacement reaction >BX + HY + >BY + HX via an ionised transition state.The usefulness of such a reaction in the preparation of isocyanates and iso- thiocyanates was made fully apparent. Dr. Lappert concluded with a stimulating survey of the evidence for double-bond character in the boron-nitrogen bond in compounds with both boron and nitrogen 3-co-ordinated. A brief discussion on the papers of this session followed. Dr. Lappert expounded upon the prepara- tion of the boron isocyanates and Dr. Turner who had taken the bold and refreshing step of reporting uncompleted work discussed with Professor Burg Dr.Clapp and Dr. Greenwood further experiments to be carried out on the tetrameric borazyne. The final session with Professor G.E. Coates in the chair opened with a lecture “Interaction of Boron Halides with Alkoxy-compounds of Phos- phorus Sulphur Silicon Titanium and Tin” by Dr. W. Gerrard (Northern Polytechnic). The reactions of boron trichloride with the many oxy-compounds of these elements were covered in a systematic fashion and cases where mutual replacement does not occur were pointed out. However no informa- tion on reactions with compounds of tin was made available. Dr. Gerrard finished with a description of the reactions between chloroborazoles and phos- phates from which polymers can be obtained.During the short discussion on this paper Dr. Greenwood exhibited interest in the reaction of phosphate with trichloroborazole. Dr. Gerrard postulated hydrogen bonding of the phosphate to the borazole-nitrogen as the first step of the reaction and Professor Burg contributed information on the complexing of BH and BF with phosphorus com- pounds. Dr. Gerrard also commented that the poly- mers formed in this reaction were not very resistant to hydrolysis. Dr. R. C.Anderson (Imperial Chemical Industries Nobel Division) delivered a lecture at short notice choosing as his subject “Reactions of alkylalumin- ium Halides with Boric Oxides.” Dr. Anderson showed how in analogy with the reactions of silica methylaluminium halides or alkylaluminium chlor- PROCEEDINGS ides with amorphous boric oxide yielded alkylborons.Bromoalkyl or large alkyl groups apparently exert an unfavourable steric effect. “The Reaction of Diazomethane with Boron Compounds” by Dr. A. G. Davies (University Col- lege of London) concluded the scheduled lectures. Dr. Davies concisely surveyed the two possible mechanisms (cationic chain reaction or nucleophilic 1,2-rearrangement) which could lead to the forma- tion of polymethylene. Suitably chosen reactions of nucleophilic reagents with possible intermediates in the cationic chain process showed that both pro- cesses probably occurred simultaneously. Interest in Dr. Davies’s stimulating presentation was evident from the discussion which followed.Dr. Lockhart (Holloway College) and Dr. Brown (Nottingham University) wondered whether there was carbene formation in these reactions but Dr. Davies reported evidence against this. Professor Bawn was interested in the termination mechanism of the polymerisation and Dr. Davies postulated proton-or boron-transfer of which the former seemed more reasonable to Professor Bawn although not to Dr. Gerrard. Dr. Lappert and Dr. Davies also discussed certain objections to the cationic chain mechanism and how these might be overcome. The symposium was then formally closed by the chairman. After so many authoritative lectures in- volving such highlights as the problem of bonding between a boron atom and boron or nitrogen and the discovery of many fascinating new compounds there can have been no doubt in anybody’s mind that the study of boron compounds represents a fascinat- ing and fertile area of chemistry.ALASTAIR M. NORTH Some Aspects of the Chemistry of Natural Products A SYMPOSIUM on this topic formed part of the Anniversary Meetings of the Chemical Society held this year for the first time in Liverpool. It was divided into three sessions two on April 12th and the third on April 14th. The first session was con- cerned with newer applications of physical methods the second with photochemical transformations and the third with porphyrins and related compounds. All three sessions were held in the new 350-seat Muspratt Lecture Theatre of the Donnan Labora- tories; and evidence of the widespread interest in the topics being discussed was shown by the fact that extra seating had to be provided.Four of the con- tributors came from abroad Professors G. Buchi (M.I.T. Boston Mass.) W. G. Dauben (Berkeley Cal.) C. Djerassi (Stanford Cal.) and E. Havinga (Leiden Holland). At the first session the Chairman Professor G. W. Kenner welcomed Fellows from all parts of the British Isles and from many countries overseas. He then introduced Professor C. Djerassi who gave a stimulating review of the applications of optical rotatory dispersion to organic chemical problems. Rotary dispersion curves in particular those of the “anomalous” type had been shown to be of great assistance in the determination of absolute configura- tions and the detection of subtle conformational changes as well as in certain structural and analytical applications.It was now expected that the long- awaited paper on the octant rule would be published this summer. A newer application made possible by improved instrumentation had been the determina- tion of the rotatory dispersion curve of the coni- pound (I) (emax.-10,OOO) at shorter wavelengths than the first extremum. The study had also been extended to the develop- JUNE 1961 ment of “chromophoric” derivatives of “non-chromophoric” compounds such as dithiocar-bamates from amines and thioureides and nitroso- amides from carboxylic acids; for instance the di- thiocarbamates of L-amino-acids gave a positive Cotton effect and the D-series a negative Cotton effect.Curves of the osmate ligands of steroidal olefins could be related to the position of the double bond. Dr. L. M. Jackman (Imperial College London) the next speaker illustrated the value of nuclear magnetic resonance spectroscopy by describing some studies in the carotenoid field. The olefinic absorption of the polyene chain was usually rather diffuse but the sharper C-methyl absorption could be used to differentiate between saturated aliphatic and “in- chain-” and “end-of-chain”-polyene methyl groups thus enabling the end groups of various carotenoids to be characterised e.g. capsanthin capsarubin. Examination of the pattern of C-methyl absorption of the biogenetic intermediates between squalene and lycopene (Le.phytoene phytofluene neurosporene and &carotene) had confirmed that the extra double bonds in lycopene were introduced successively from the centre of the squalene molecule. One of the end- groups in both of the bacterial pigments “R” and “Y” was the same as in lycopene but the other end of Y was identical with those in spirilloxanthin (cf. partial formula IIa) whereas in R this grouping had been modified to the ketone (IIb). The last paper of the session by Dr. R. I. Reed (Glasgow) was concerned with the applications of mass spectrometry to the determination of the struc- tures of natural products. The fragmentation pat- terns given by fused ring-systems were described and the variations due to the cis-or trans-nature of the ring junctions were discussed.Small discon- tinuities in an otherwise smooth fragmentation pat- tern had shown that dolichol (supplied by Professor R. A. Morton Liverpool) was a straight-chain alcohol C1,,H2,,~0H with 20 C,-units. Double-focussing spectrometers were now available which allowed the precise determination of the molecular weights of the fragment ions and thus in conjunction with a study of isotopic abundances precise informa- tion on their molecular constitution could often be obtained. Dr. Reed’s last remarks served to high- light the overwhelming impression given by all three speakers of the tremendous amount of work which had been and could be carried out by the use of new physical methods once a reliable instrument with good resolution had been developed.I97 The second session (with Professor E. R. H. Jones in the Chair) began with a scintillating account by Professor D. H. R. Barton of new photochemical reactions of organic nitrites and azides. There was a distinct conformational requirement for these re- actions which involved a transition state containing a six-membered ring. Whereas the residue from cyclohexyl nitrite disproportionated on irradiation to give a mixture of the corresponding ketone and alcohol cyclopentyl nitrite underwent ring opening to give a very convenient synthesis of glutaraldehyde monoxime in high yield. In the steroid field photolysis of 17a-or 17p-nitrites (LII) gave rise to a hydroxy- amide (IV) probably through the intermediate shown The photolysis of azides had been applied to a new synthesis of proline from the 8-azide of ethyl hexanoate and extended to give a brilliant synthesis of the steroidal alkaloid connessine (VI) by photolysis of the diazide (V).The nitrite and azide photolyses Me __t Me N Reagents :Photolysis then LiAIHI then CH,O-HCO,H . were probably only two examples of a whole family of similar rearrangements ;e.g. hy pochlori tes under- went similar photochemical reactions. In discussion Dr. V. Petrow (British Drug Houses London) and Professor Barton discovered that they had both investigated the photolysis of a steroid 6-hypo- chlorite and found that the 19-methyl group was converted into chloromethyl.This had been utilised by Petrow in the conversion of a 19-methyl-steroid into the corresponding nor-compound. In the next paper Professor G. Buchi discussed the photochemical rearrangements of unsaturated cyclic ketones. In contrast to the behaviour of cyclohexa- dienones “bond-switching” did not occur with the cycloheptadienone eucarvone or with octatrienone. Eucarvone (VI9 gave the ketone (VLU) and on further irradiation the isomer (IX) ; octatrienone gave mainly bicyclo [4,2,0]octa-4,7-dien-2-one. The mechanisms of these reactions were discussed and compared with the pyrolytic reactions of these compounds and their transformation products. In conclusion some photochemical studies with benzo- nitrile and olefins were discussed; addition to the benzene ring occurred rather than to the cyanide group as had been hoped.Professor W. G.Dauben's contribution was also concerned with photochem- ical transformations of unsaturated ketones mainly with tropolones. Tropolone methyl ether (X) gave the bicyclic material (XI) which was then trans- formed into the isomer (XU) identified by hydrolysis to the oxocyclopentylacetic ester (XIII). A study of n @'") MCH2.C0,M. the similar transformations of the 4-and 6-methyl- and the 5-isopropyl-tropolone methyl ethers sug- gested that the second stage in the rearrangement was probably a concerted process and did not occur by the movement of any single group. In the final paper of the second session Professor E.Havinga summarised recent work in Leiden on photochemical transformations in the vitamin-D field. Calciferol is formed by thermal equilibration of precalciferol itself formed directly by irradiation of ergosterol. Tachysterol and lumisterol are formed in a side reaction from precalciferol (as summarised in the reaction scheme) and are not directly involved D I$A hv hv E+P +T Yhvd'hv L in the formation of calciferol as postulated in the original Windaus scheme. The mechanisms of the various transformations were discussed with par- ticular reference to the steric factors involved. The final session with Professor F. Bergel in the Chair was concerned with porphyrins and related compounds. Professor A. W.Johnson (Nottingham) gave a masterly survey of the present position in the field achieving the remarkable feat of compressing the chemistry of porphyrins from pyrroles to vitamin B12 into the space of fifty minutes. After PROCEEDINGS describing briefly the determination of structure and the classical methods of synthesis of porphyrins and bile pigments he went on to discuss the biogenesis of the porphyrins in relation to the laboratory poly- merisation of individual pyrroles. The main features of the recent chlorophyll syntheses were then sum- marised the Munich synthesis being based on a route developed by Hans Fischer whilst the Harvard synthesis involved the condensation of two pyrro- methane units (a method which had been used with conspicuous success by Macdonald in his recent synthesis of uroporphyrin 111; cf.J. Arner. Chern. SOC.,1960 82 4389). Finally Professor Johnson discussed the structure of vitamin B, and referred briefly to the problems involved in the synthesis of the basic nucleus. During the discussion Dr. A. H. Jackson stated that the nuclear magnetic resonance spectra of porphyrins dissolved in trifluoroacetic acid were useful in structural diagnosis; in some instances type isomers could be distinguished and this point was illustrated with the coproporphyrins I 11 and 111. Professor G. W. Kenner pointed out the limitations of the classical porphyrin syntheses. Even the most recently developed methods were limited by sym- metry considerations or could only be applied in special cases (as for example in the Harvard syn- thesis of chlorophyll which relied upon a most ingeniously directed ring closure of two pyrro-methanes).Recent work in Liverpool had been directed towards the development of truly stepwise syntheses. Methods had already been worked out for synthesising di- and tri-pyrromethanes but for various reasons these could not be extended to tetra- pyrrolic compounds and hence to porphyrins. More recently unsymmetrical dipyrroketones had been prepared by the Vilsmeier procedure from pyrrole amides and pyrroles in presence of phosphorus oxy- chloride. These ketones did not show carbonyl pro- perties but nevertheless they could in certain instances be reduced to pyrromethanes by boro- hydrides.Furthermore the terminal methyl group in bicyclic ketones could be oxidised to a reactive formyl group permitting further extension of the chain. Further work was in progress and it was hoped that it might soon lead to a new synthesis of por- phyrins. The final paper of the session (and of the Sym- posium) by Dr. V. M. Clark (Cambridge) was con- cerned with work at Cambridge aimed at the synthesis of the corrin nucleus the basic unit of vitamin B12. The condensation of 2-methyl-l-pyrroline 1 -oxides with 1-pyrroline 1 -oxides in presence of base gave dipyrrolinylmethanes whereas the self-condensation of the 2-substituted 1-pyrroline 1-oxides gave bipyrrolyl derivatives. Thus methods JUNE 1961 199 were available (in principle) for linking four pyrrole on this problem was described.In conclusion Dr. rings together by three methylene bridges and Clark gave a brief outline of the proposed route by one direct link to form the corrin nucleus. One means of which it was hoped to complete the of the problems remaining was the introduction of synthesis. further unsaturation in extended systems and work A. H. JACKSON COMMUNICATIONS The Mechanism of AUylic Bromination by N-Bromosuccinidde By B. P. MCGRATH and J. M.TEDDER (UNNERSITY OF SHEFFIELD) WE recently reported two pieces of evidence in centration. Professor Sixma has kindly drawn our support of the Goldfinger mechanism for allylic attention to a paper he published in 1958 in which bromination by N-bromosuccinimide.1 The first among other evidence he reported in greater detail pi-of evidence was concerned with allylic bromina- almost identicaI results.2 tion when inolecular bromine is used at low con- (Received April 29tl1 1961.) McOrath and Tedder Proc.Chem. SOC.,1961 80. Sixma and %em Prm. k. ned. Akacl. Wetemchap. 1958 By61 183. Addition of Bromine to trans-and cis-Cycldecene A Transannular Reaction By J. Z~VADA and J. SICHER (INSTITUTE CHEMISTRY OF ORGANIC AND BIOCHEMISTRY CZECHOSLOVAK OF SCIENCE CZECHOSLOVAKIA) ACADEMY PRAGUE REACTIONS On re- of medium-sized ring compounds in-heated with methanolic potassium i~dide.~ volving formation of carbonium ions (e.g. solvolytic action with sodium thiophenoxide in ethanol the epoxide-ring opening) are k.nown to proceed by a dibronzides gave not the cyclo-olefins and diphenyl di- transannular path.l The possibility has been con- sulphide but products of replacement the bisphenyl- sidered that addition of bromine to the medium- thio-derivatives,6 in practically quantitative yields.ring cycloalkenes (viewed as involving a cyclic Thus the bromine atoms in the dibromocyclodecanes bromonium ion intermediate) could similarly lead to are not vicinal; they may be in 1,6-positions as is transannular products ;2 addition of bromine to cis-probable if the reaction is transannular. To confirm cydo-octene has been examined from this point of this cyclodec- 1,6-ylene ditoluene-p-sulphonate,' m.p. view but no evidence for a transannular reaction 1 16-1 18O corresponding to the cyclodecane- 1,6- course could be obtaind2 We now report the isola- dial* of m.p.152-153' was allowed to react with tionof two crystalline "transannular" dibromocydo- sodium thiophenoxide. The crystalline product was decanes on bromination of trans- and cis-cyclo- identical with the bisphenylthio-derivative m.p. decene respectively. 103' obtained from the dibromocyclodecane of m.p. Addition of bromine to trans-cyclodecene in 121". The other 1,6-ditoluene-p-sulphonate,m.p. tetrachloromethane at -5* gave an oil from which 91-92' analogously afforded the other bis-phenyl- about 20 % of a crystalline dibromo-derivative m.p. thio-derivative m.p. 96-97 These reactions prove O. 121O were isolated.* cis-Cyclodecene analogously that the crystalline bromination products are trans- afforded about 15% of a dibromide m.p.92.5-and cis- 1,6-dibrornocyclodecane. 93.5". Neither dibromide liberated iodine when (Received April 17th 1961.) *The dibromide of m.p. 121" is almost certainly identical with a product obtained in unstated yield by Braude and Gofton3 by addition of bromine to a mixture of cis-and trans-cyclodecenes. Although this compound had been found to behave somewhat anomalously in not giving l-bromocyclodecene on treatment with dimethylamine* it was assigned3 the structure of a 1,2-dibrornocyclodecane. l Prelog Angew. Chern. 1958,70 145. Cope and Wood J. Amer. Chem. SOC.,1957,79 3885. a Braude and Gofton J. 1957,4720. Kohler Tishler Potter and Thompson J. Amer. Ch.SOC.,1939 61 1057. 'Weinstock Lewis,and Bordwell J.Amer. Cliern. SOC.,1956 78 6072. Eliel and Haber J. Org. Chem. 1959,24 143; Bergelson and Badenkova Izvesi. A.4ad. Nauk S.S.S.R. Otdel. khiin. NaHk 1958 793. Cope and Holman J. Amer. Chern. Suc. 1950 72 3062. a Plattner and Hulstkamp Hefv. Chiw. Actn 194.4. 27 21 I. PROCEEDINGS The Polymerisation of Styrene by Perchloric Acid By D. C. PEPPER and P. J. REILLY (TRINITY DUBLIN) COLLEGE ANHYDROUS perchloric acid (obtained from 72% and decreases with conversion according to the aqueous solution with an excess of concentrated relation sulphuric acid) is found to be an unusually repro- '/Fn = k4/k2 $-k3/k2(1nLM lo/ IM It)/(LM 10 -LM It) ducible initiator for the cationic polymerisation of styrene in chlorinated aliphatic solvents (nitro- Values of these rate constants have been measured solvents give anomalously low rates).A large excess at 25" 0" and -30" and give good Arrhenius rela- Solvent (CH,Cl) 0.43 9.8 8.3. 6.9 11.6 7.7 13-6 . 10.1 0.87 9.15 8.6 7.0 9 1 -74 8-05 9.6 7.6 15.5 10-7 8.3 4.5 (cH2cl)2-cc14 (4:I) 0-43 7.0 11.4 8.4 (3 :2) 0.43 5.2 13.0 8.7 C2H,CI9 0.43 c.ll.0 12.0 8.6 --of water inhibits the reaction but very small con- tions yielding the activation energies and frequency centrations (up to several times that of the catalyst) factors in the Table. have no detectable influence. Very reproducible At 25" and with [MIo = 0.43,over the range of kinetics can thus be obtained by dilatometry with dielectric constant (E) from 2-3 to 16.4 [CC14-simple semi-open reagent-mixing methods instead of (CH2C1),-Ph-NO2] the value of k increases some the rigorous vacuum-manipulations normally found 20,000 times according to the relation necessary in cationic polymerisation.In the systems listed below the rate equation log, k = 2.18 -12.41~ Above this range as the proportion of nitrobenzene is increased to above 30% the rates decline and are is accurately obeyed over the greater part of the zero for 3:7 (CH,Cl),-PhNO,. Jn nitroethane (E -time-course (in favourable cases up to > 95% con-25) low rates and departures from the first-order version) and a range of [HC104] 0-1 -5 x time-course are observed. molar. At lower [HCIO,] the rate constant dimin- The following interpretation may be given. Since ishes with time and limited yields of polymer are there is no appreciable termination the constants k obtained.At higher concentrations there are depar- and k measure the processes of spontaneous and tures from first-order kinetics believed to be due to monomer transfer respectively. Initiation is believed decomposition of the acid. Inthe range quoted above to be very fast (cf. H2S0 polymerisationl) and not the catalyst is not consumed during the reaction rate-determining. Hence the overall rate will be (further monomer added is polymerised at effectively limited only by that of chain propagation and equa- the original rate) i.e. there is no effective chain- tion (1) yields the rate constant for propagation termination reaction. directly. The value found at 25" with [MI = 0-43 The molecular weights of the products are low is 17.0 1.mole-l sec.to be compared with a value of (500-7000) and related to the intrinsic viscosity in 6.7 deduced for the H,SO,-initiated polymerisation benzene at 15" by under these conditions. The variation of the rate constants with the polarity of the solution is in general concordance with expectations for cationic reactions except for where 3 is determined cryoscopically in benzene. the anomalous behaviour in ethyl chloride and nitro- The degree of polymerisation F,,= zn/104 in-solvents. creases with initial monomer concentration [MI, (Received April 17th 1961 .) Hayes and Pepper Proc. Chem. SOC.,1958 228. JUNE 1961 201 Thermal Diffusion in Solutions of Carbon Tetrachloride in Benzene By H.J. V. TYRRELL and J. G. FIRTH (THEUNIVERSITY SHEFFIELD) BENZENE and carbon tetrachloride form approxi- mately regular solutions and it was as an example of this class of mixture that we recently undertook a study of the Soret effect in solutions of carbon tetrachloride in benzene. Two previous investigations have been reported. Tichachek Kmak and Drick- amer,’ using a stirred diaphragm cell found that over a wide range of concentrations the solution in the colder chamber became enriched in carbon tetra- chloride the denser component. Korsching2 had earlier reported that carbon tetrachloride migrated to the cold wall in a pure Soret-effect cell from a solu- tion with a molar fraction of carbon tetrachloride equal to 0-05.It is not uncommon for the heavier component of a binary liquid mixture to become con- centrated at the cold wall in such experiments but on both empirical3 and theoretical grounds4s5 the principal factor determining the direction of separa- tion at least for regular solutions seems to be the relative magnitude of the latent heats of vaporisation per unit volume (cohesive energies square of solu- bility parameters) of the two components. On this criterion carbon tetrachloride should migrate to the hot wall from a benzene solution a direction apparently contrary to that observed experimentally. We have now examined the change in refractive- index gradient for solutions with molar fractions of carbon tetrachloride of 0.1 and 0-2 in two modified Tanner cells6 with heights (a) of 10 and 2 mm.respectively and a mean cell temperature of 298”~. With the latter which has a characteristic time (8) (= a2/n2D,where D is the mutual diffusion co- efficient) of less than four minutes the earliest possible observations taken about five minutes after the temperature gradient had been first applied showed that the solution in the cooler part of the cell was becoming richer in carbon tetrachloride. The apparent Soretcoefficient [-N-l(l -N)-l(dlv/dT),t,t. where N = the molar fraction of carbon tetra- chloride] in the more concentrated solution was 3.9 x this figure being calculated from the final observed refractive-index gradient. It agrees quite closely with that recorded by Tichachek et aZ.l at a slightly higher temperature (4-4 x deg.-l at 313”~).However in the 10 mm. cell in which observations of the refractive-index gradient at the cell centre at a much earlier stage in the demixing process were possible it was found that when t < 0-28 the carbon tetrachloride migrated towards the hot wall. When t > 0.28 the refractive-index gradient began to change in the opposite sense and had the experiments been continued for a sufficiently long time it seems probable that the final state would have been similar to that observed in the 2 mm. cell. Analysis of the initial separation curves obtained from the 10 mm. cell showed that the observations were reproducible and that when t < 0.28 the observed deflection of the light beam was a linear function of erf c[a/4d(Dt)],as required by Agar’s solution7 of the diffusion equation for small values of t.From the slopes of these lines the Soret coefficient for carbon tetrachloride defined as above was calculated to be approximately -20 x lop3 deg.-l for both the solutions examined. This exceptionally large value coupled with the large difference in the molar volumes of the two components adequately explains the observed change in direction of separation. In a column of liquid heated from the top downwards the density is normally least at the top and the column has some degree of mechanical stability. In a binary mixture with a weight fraction w of solute and Soret co- efficient o the density gradient at a height x above the base of the cell is given in the steady state by dp/d( = -~[j6p+ OW(^ -w)~P/~w] where 6 is a reduced height defined as x/a p is the density 18 the coefficient of dilation and r the temperature interval applied across the column pro- vided the temperature gradient is assumed to be uniform.The first term within the brackets is posi- tive and the second is negative for this system. If 101 exceeds approximately 10 x deg.-l then dp/df will become positive with the result that the column is no longer mechanically stable. Remixing will take place and the results show that eventually the con- centration gradient has a sign opposite to that developed initially. This is in fact an example of the “forgotten effect”,8 often invoked in discussions of separation in thermogravit ational (Clusius-Dickel) cells.It is likely that even a solution with a molar fraction of carbon tetrachloride of 0.05 would show Tichachek Kmak and Drickamer J. Phys. Chem. 1956,60 660. Korsching 2. Naturforsch. 1955 10a 242. Prigogme de Brouckere and Amand Physica 1950 16 577 851. * Dougherty and Drickamer J. Phys. Chem. 1955,59,443. Bearman Kirkwood and Fixman Adv. Chem. Phys. 1958 1 1. Tyrrell Firth and Kennedy unpublished work. ’Agar Trans. Faraday SOC.,1960,56 776. de Groot Hoogenstraaten and Gorter Physica 1942 9 923; 1943 10 81. instability of this kind if the Soret coefficient is in- dependent of concentration and Korsching's result based upon a single observation of the apparent steady-state refractive-index gradient in a pure Soret cell is not surprising.It is harder to explain the diaphragm-cell results since the purpose of using a cell with a diaphragm is to mirzimisc remixing of all kinds. Possibly there was some gross streaming of the heavier component through the pores of the diaphragm. This analysis of the initial separation curves in a cell of suitable dimensions has shown that the Soret PROCEEDINGS coefficient for solutions of carbon tetrachloride in benzene has the sign predicted by theory. Preliminary calculations suggest that the experimental value is rather larger than that expected by summing the calculable terms in Bearman Kirkwood and Fixman's statistical mechanical eq~ation.~ The ob-served complications due to the "forgotten effect" may well be important for other mixtures containing carbon tetrachloride and a lighter component with a higher latent heat of vaporisation per unit volume.To detect these a careful study of the earlier stages of the separation process is essential. (Received March 29th 196 I .) Nitratopenbcarbonylmanganese(1 By C. C. ADDISON, M. KILNER,and A. WOJCICKI (DEPARTMENT THE UNIVERSITY OF CHEMISTRY NOTTINGHAM) ALTHOUGH many derivatives of the metal carbonyls have been prepared no derivatives involving the nitrato-group have yet been described. Particular interest attaches to such compounds since it has appeared possible that the nitrato- and the carbonyl group might be incompatible when present in one complex.We have now prepared nitratopenta- carbonyImanganese(r) the first representative of this new class. Attempts to prepare the compound by direct replacement of the halogen in pentacarbonyl-manganese halides by using silver and lead nitrate were unsuccessful. However the nitrato-carbonyl is the major product formed when decacarbonyldi- manganese reacts with an excess of dinitrogen tetroxide in light petroleum. In a typical preparation 4 g. of the carbonyl in 250 ml. of light petroleum (b.p. 40-60") were treated with 12 g. of the tetroxide in 5 ml. of light petroleum at 0".Separation of a flocculent yeilow precipitate began almost immedi- ately and continued for about 40hours. The product was purified by a series of fractional recrystallisations from chloroform-light petroleum at -40".The final crystabation appears to involve the separation of nitratopentacarbonylmanganese (I) from the corre-sponding nitro-compound. This gives the pure com- pound Mn(CO),NOa as a bright yellow powder (Found Mn 21.5; N 5.55; C 23.3 %; M,cryoscopic in nitrobenzene 254. C,MnNO requires Mn 21.4; N 5.5; C 23.4%;M 257). Covalent bonding of the nitrate group is confirmed by the infrared bandsf at 1486 1284 1010 and 805 cm.-l (Nujol and hexachlorobutadiene mulls). Car- bony1 absorption bands at 2060 and 2002 crn.-l (Nujol mull) are similar to those observed with the pentacarbonyl halides.g The compound is diamag- netic and stable in air at room temperature. In a vacuum it begins to decompose at about 55" giving sublimed Mnp(CO)l and a residue containing Mn(NO$ and oxides of manganese.No nitrato-complex was detected in the sublimate. Solution properties of the complex Mn(CO),NO indicate that it is highly polar. It is readily soluble in chloroform ethyl acetate alcohol acetone nitro- methane and nitrobenzene; sparingly soluble in water methyl cyanide carbon tetrachloride benzene and ether; and insoluble in aliphatic hydrocarbons and carbon disulphide. Its stability in solution is very low ;appreciable decomposition is observed after 30 minutes. Nitratopentacarbonylmanganese(1) is a non-electrolyte in fresh solutions of nitromethane nitrobenzene and ethyl alcohol but the conductivity increases rapidly on storage.This decomposition may involve partial replacement of the nitrato-group by the solvent. We thank the National Science Foundation for a Postdoctoral Fellowship (for A.W.) the D.S.J.R. for a maintenance grant (to M.K.) and the Ethyl Corporation for a gift of manganese carbonyl. (Received April 21st 1961.) Addison and Gatehouse J. 1960,613; Quagliano J. Amer. Clzein. Soc. 1959 81 3818. * AM and Wilkinson J. 1959 1501. JUNE 1961 203 The Crystal Structure of Primary Cupric Dithizonate By R. F. BRYAN and P. M. KNOPF (BIOLOGY MASSACHUSETTS OF TECHNOLOGY 39 MASS.,U.S.A.) DEPARTMENT INSTITUTE CAMBRIDGE DIPHENYLTHIOCARBAZIDE (dithizone)(H,Dz) is wide- ly used as an analytical reagent for the determination of metals but the structure of the metal-dithizone complexes has remained in dispute.We report the preliminary results of an X-ray diffraction study of the crystal structure of primary cupric dithizonate Cu(HDz),. Crystals of the complex were grown from chloro- form and identified by their absorption maximum at 545 mp. The red needles are triclinic with a = 13-36 b = 10.97 c = 4.29 A 01 = 85" 44' 18 = 84" 25' y = 90" 48'. There is one molecule of the complex per unit cell the space group is Pi,and the molecule has a centre of symmetry coincident with the copper atom. The structure was derived from a study of the Patterson function P(lrv0). A projection of the electron density viewed down the short c axis is shown in the Figure together with the proposed structural interpretation.The co-ordination around the copper atom appears to be square coplanar and the distances Cu-N and Cu-S are 2-00 and 1-85 A respectively at the present stage of the refinement. The current value of the reliability index R is 0-14 for 189 non-zero reflexions in (hkO). The extra-benzenoid hydrogen atom has not yet been located though this may be possible at a later stage in the refinement. On steric grounds the most likely position of attachment seems to be to the nitrogen atom labelled with an asterisk in the Figure. This work was supported in part by a grant from the National Cancer Institute of the National Institutes of Health (U.S.A.). U' Electron-density distribution in (hkO) for primary cupric dithizonate together with structural interpreta- tion.The full circles represent carbon atoms the small open circles nitrogen and the large open circle sulphur. The copper atom lies at the centre of symmeti V. Contours are drawn at arbitrary intervals. (Received March 27th 196 .I Complex Pseudohalides of Boron and Aluminium By F. KLANBERG (EDUARD-ZINTL-INSTITUT FUR ANORGANISCHE UND PHYSIKALISCHE CHEMIE DER TECHNISCHEN DARMSTADT, HOCHSCHULE GERMANY) RECENT methods1 for the preparation of compounds in which pseudohalogen ions (CN CNS CNO) function as ligands to boron or similar elements are usually based on replacement of halogen atoms in a metal halide by means of suitable alkali pseudo- halides in usually highly polar solvents. It has now been found that such compounds particularly iso- thiocyanates of boron are also conveniently obtained by the general route X X MBH4 j(scN)z +MBH4-dNCS)z 4-ZH2 In solvents such as ether Or tetrahydrofuran at or little below room temperature.When sodium borohydride is used the number of hydrogen atoms in the BH4- unit that are displaced by NCS groups does not exceed three; if the boro- hydride is in great excess the composition of the Lappert and Pyszora Proc. Chenz. SOC.,1960 350; Green Sowerby and Wihksne Chem. and Ind. 1960 1305 Brennan Dahl and Schaeffer J. Amer. Chem. SOC.,1960 82 6248. product shows a B :H ratio somewhat smaller than that required for NaBH(NCS),. This compound has been isolated as a white hygroscopic solid which decomposes at about 130" without melting [Found B 5.1; Na 10.6; CNS 81.4; H 0.5.Calc. for NaBH(NCS), B 5.2; Na 11.0; CNS 83.4; H 0.5 %I. Conductance measurements for NaBH(CNS) in dimethyl sulphoxide support an ionic structure (Ao = 30.5 at 25" which is in the range of a 1:l electrolyte in this solvent2). The compound forms a bis-diethy1 ether complex and a bisdioxan complex. At room temperature the former is an oil the latter a white solid. Evolution of hydrogen both from NaBH(NCS) and from its adducts is markedly slower than from sodium borohydride in acidic aqueous or alcoholic solutions. Lithium borohydride in ether gives with thio- cyanogen ultimately an ether complex of LiB(NCS), although there is evidence for the intermediate formation of hydrogen-containing species.It is not possible to isolate a solid tetrathiocyanatoborate from the ether complex obtained after removal of the excess of solvent in vacuo since,when the tightly bound ether is removed at 80"in vacuo the residual material starts to decompose. An ether-insoluble bis- dioxan complex LiB(NCS)4,2C4H,02 can be pre- PROCEEDINGS cipitated from the solution in ether and is stable up to 100". Interaction between lithium aluminium hydride and thiocyanogen in ether yields chiefly a very un- stable ether complex of LiAl(NCS) (decomp. 50"). Further the C-S bond of thiocyanogen is sensitive to reductive cleavage by lithium aluminium hydride so that hydrogen sulphide and hydrogen cyanide are also produced in this reaction.Selenocyanogen and cyanogen were also treated with borohydrides and lithium aluminium hydride under similar conditions the products being either insoluble polymers [with (CN),] or a mixture of the corresponding selenocyanate and its fission products. Infrared spectra of the more stable compounds have been recorded in the sodium chloride region. A preliminary inspection shows agreement with estab- lished spectroscopic evidence3 for bonding of the thiocyanate group through nitrogen. NaBH(NCS) has bands at 2335 2225 (vB-H) a broad band at 2082-1980 (vasN=C=S) and a band at 962 cm.-l (vsN=C=S) together with a few other peaks not yet definitely assigned. The data however strong- ly indicate that NaBH(NCS) is sodium tri-isothio- cyanatornonohydro borate.(Received Aprif 17th 1961.) Sears Lester and Dawson J. Phys. Chem. 1956 60 1433. * Mitchell and Williams J. 1960 1912; Miller and White 2. Elektrochem. 1960 64 701 ;and literature cited in ref. 1. The Addition of Methyl Radicals to Nitrosomethane in the Vapour Phase Trimethylhydroxylamie By L. PHILLIPS (EXPLOSIVES ESTABLISHMENT, RESEARCH AND DEVELOPMENT MINISTRY WALTHAM ABBEY, OF AVIATION ESSEX) THEimportance of reactions of nitrosomethane in the interpretation of the inhibition of methyl radical chain reactions by nitric oxide has recently been stressed by several authors? v2 Christie1 suggested that a substituted hydroxylamine was formed by addition of nitric oxide to initially formed nitroso- methane in the low-temperature photolysis of methyl iodide with an excess of nitric oxide.Hoare3 has shown that in the photolysis of acetone-nitric oxide mixtures methyl radicals and nitric oxide can react in ratios from 3 :1 to 1 :3 depending on the relative amounts of the reactants; no products were identi- fied. Batt and Gowenlock4 have shown that nitroso- methane and nitric oxide at about 400" form a pro- duct not completely identified by probably N-methyl-N-nitrosohydroxylaminenitrite. Methyl radicals and nitrosomethane are present in di-t-butyl peroxide-alkyl nitrite systems at 160-180" the abstraction of nitric oxide from methyl nitrite by methyl radicals in the vapour phase at 180" has been rep~rted:~ CH,. + CH,O.NO -+CH,*NO + CH,O-.The majority of the methoxyl radicals react with nitrosomethane to form methanol the nitrosomethane being converted largely into an unidentified involatile oily nitroso-compound. These reactions have been confirmed for a series of alkyl nitrites (results to be published elsewhere). Detailed investigation of the volatile products of these reactions by vapour-phase chromatography revealed in each case a major unidentified peak sug- gestive of a low-boiling (ca. 30") compound. The facts that yields of alcohol were always less than the Christie Proc. Roy. Soc. 1958 A 249 258. Gowenlock Trotman and Batt Chem. SOC. Special Publ. 1957 10 75. a Hoare,quoted by Batt and Gowenlock (ref. 4) and personal communication. Batt and Gowenlock Trans.Faraday SOC.,1959 56 682. Jest and Phillips Proc. Chem. SOC.,1960 73. JUNE 1961 nitrite consumed and therefore less than the nitroso- methane initially formed and that methyl balances (CH + 2C,H,) were low suggested that methyl radicals might add to some of the nitrosomethane to form methylated hydroxylamines e.g. 2CH3-+ CH,-NO + (CH&N.OCH,. Such reactions have been demonstrated by Gingras and Waters6 for 2 cyano-2 propyl radicals with aromatic nitroso- compounds and with nitric oxide in liquid systems; and Gowenlock and Liittke' pointed out that such reactions might be expected in vapour-phase reactions. Sufficient of the unknown compound for mass- spectrometric examination was separated from the volatile products by vapour-phase chromatography.The cracking pattern agreed well with that of an authentic sample of trimethylhydroxylamine (b.p. Gingras and Waters J. 1954 1920. Gowenlock and Luttke Quart. Rev. 1958 12 231. Jones and Major J. Amer. Chem. SOC.,1928,50,2742. 30") prepared by the method of Jones and Major? Chromatographic retention volumes of the unknown sample and trimethylhydroxylamine were identical. Yields in the methyl radical-alkyl nitrite reactions were about one-fifth of the nitrite consumed. This observation together with those referred to ab~ve,l~~~~ raises important questions in free-radical inhibition by nitric oxide in the vapour phase. Batt and Gowenlock4 consider that addition of nitric oxide might outweigh isomerisation of nitroso-methane; addition of methyl radicals can also occur and shouId be favoured at temperatures below about 1SO" where isomerisation is slow.Earlier concepts in which nitric oxide was assumed to remove only one radical require revision. (Received April 18th 1961.) Mono-and Di-tluorogermane By T. N. SRIVASTAVA and M. ONYSZCHUK OF CHEMISTRY MONTREAL (DEPARTMENT MCGILL UNIVERSITY 2 CANADA) OF the four fluoro-derivatives of germane only the mono- and di-fluoro-compounds have not yet been described. We have obtained these new compounds from the interaction of monobromogermane with anhydrous silver@ fluoride. In a typical preparation monobromogermane (7.59 mmole) was passed slowly at 25" in a vacuum through a column packed loosely with a mixture of silver@ fluoride and glass helices.The products of the exothermic reaction were (a) germane (0-56 mmole) (b)monofluorogermane (5-70 mmole 75 %) (Found F 19.7%; M 95.1. GeH,F requires F 20.1%; M 94-6) and (c) difluorogermane (0-43 mmole) (Found M 113.2. GeH,F requires M 112-6). In addition to the main reaction GeH,Br + AgF + AgBr + GeH,F germane and difluoro- germane were produced in almost equimolar amounts by the disproportionation reaction 2GeH,F +-GeH,F + GeH,. This disproportionation occurred to an extent of about 15 % when pure monofluoro- germane was kept at 25" for 16 hr. Tensiometrically pure monofluorogermane had vapour pressures in the range -45" to 0" given by the equation loglop (mm.) = 8.639 -1662/T which indicates an extrapolated b.p.of 15.6" a latent heat of vaporization of 7605 cal./mole and a Trouton constant of 26.4. The compound melted sharply at -22". Its infrared spectrum in the range 4000-650 cm.-l showed strong absorption bands at 2140,2120 875 and 702 cm.-l which have been assigned to the asymmetric Ge-H stretching the symmetric Ge-H stretching the GeH deformation and the Ge-F stretching vibration respectively. Monofluorogermane did not react with water to produce germanol GeH,.OH or digermoxane (GeH,),O. With ammonia at -78" it formed a 1:2 adduct GeH3F,2NH, which liberated ammonia at 25",leaving a homogeneous white non-volatile 1 :1 adduct which decomposed without melting at 180". We believe the 1 :1 adduct has the ionic constitution [GeH,NH,]+F- on the basis of infrared measure- ments the details of which will be described else- where.This result contrasts with that reported by Dennis and Work1 for the interaction of monochloro-germane with ammonia at -78" to -50" which gave ammonium chloride and amorphous german- ium(@ hydride. We are grateful to the National Research Council of Canada for generous financial support. (Received April 6tk 1961 .) Dennis and Work,J. Amer. Chem. SOC.,1933 55 4486. PROCEEDINGS ~~ The Synthesis of Conessine By D. H. R. BARTON jun. and L. R. MORGAN (IMPERIAL OF SCIENCE LONDON,S.W.7) COLLEGE AMD TECHNOLOGY WEhere report a new photochemical reaction of azides. Irradiation of n-butyll and n-octyl azide2 in ethanol ether or cyclohexane in a quartz flask by means of a bare mercury-arc lamp gives pyrrolidine (I; R = H) and 2-butylpyrrolidine (I; R = Bun) respectively in yields of up to 35%.Similar irradia- tion of 3-phenylpropyl azide3 in cyclohexane affords tetrahydroquinoline (21 %) but phenethyl azide4 in the same solvent furnishes only phenethylamine (42%). Irradiation of butyl azide in benzene gives N-butylaniline (22 %); similarly octyl aide affords N-octylaniline (31 %). Irradiation of cyclohexyl azide in ethanol gives on working up cyclohexylamine (45%) and cyclohexanone (21 %). In the same way 3/%azidocholest-5-ene m.p. 61-62' [a] -19' (c. 1.71; all [a], refer to CHCl solutions) Amax 270 mp (log 1-75; all ultraviolet spectra are for EtOH solutions) vmax.(Nujol) 21 10 cm.-l(prepared by treating cholesteryl toluene-p-sulphonate with rnethanolic lithium azide5 under reflux; configuration proved by reduction with lithium aluminium hydride) gives after reduction of the product with lithium aluminium h~dride,~l' cholest-4-en-3/kd (20 %) and 3~-aminocholest-5-ene8 (33 %). These results can be explained by the assumption that photolysis of azides affords "activated" nitrenes R-N :.9 When conformation and constitution permit these can rearrange to pyrrolidines [see (11) or equivalent]. if these conditions are not satisfied then the nitrene either abstracts hydrogen from the solvent to give an amine or undergoes conventional 1,2- hydrogen shift to furnish an imine.Reduction of the latter affords the amine whilst the usual ready hydrolysis furnishes carbonyl derivatives. With these preliminaries completed we were ready to synthesise ConessinelO (V) in the following simple manner. Pregn-5-ene3~,20~-dioll1 (111; R = OH) was converted into the ditoluene-p-sulphonate which with methanolic lithium aide5 afforded 3/3-20cy.-diazidopregn-5-ene (1V) (78 %) [cy.3 -27' (c. 1-25) Am,,. 270 mp (log E 2-75) vmax. (Nujol) 2110cm.-l (converted by lithium aluminium hydride and then formic acid-formaldehyde into 319-20 a-bis-dimethylaminopreg11-5-ene~~). Irradiation of the di- aide (IV) in cyclohexane reduction of the product with lithium aluminium hydride in ether N-methyla- tion with formic acid-formaldehyde and chromato- graphy over alumina (grade 111) [elution with chloro- form-benzene (1 :9)] gave conessine (V) (4.5 %) identified by m.p.mixed mp. [a] in two solvents and infrared comparison with an authentic specimen kindly provided by Professor R. D. Haworth F.R.S. We have also synthesised proline by photolysis of the azide N,. [CH,],CO&t followed by isolation as the ethyl ester toluene-p-sulphonate and saponifica-tion. Interestingly enough irradiation of t-butyl azide gave 2,2-dimethylethyleninine13(lS%). We thank the Research Institute for Medicine and Chemistry for a generous gift of 3&2Op-diacetoxy- pregn-Sene. One of us (L.R.M.)is indebted to the U.S. Public Health Service for a Post-doctorate Fellowship that made this work possible.(Received April 6th 1961.) Boyer Canter Hamer and Put,neg J. Amer. Chem. SOC.,1956 78 325. Lieber and Chao J. Org. Chem. 1957 22 238. Boyer and Morgan J. Amer. Chem. SOC.,1959,81 3369. Smith and Brown J. Amer. Chem. SOC.,1957 73,2438. Huisgen and Ugi Chem. Ber. 1957 90,2914. Boyer J. Amer. Chem. SOC.,1951 73,5865. Cf. Johnson Bauer and Franck Tetrahedron Letters 1961 No. 2 72. Julian Magnani Meyer and Cole f. Amer. Chem. Soc. 1948 70,1834. Cf. Smolinsky J. Amer. Chem. SOC.,1960,82,4717; ibid.,!n the press; J. Org. Chern.,in the press; we thank cordially Dr. G. Smolinsky (Bell Telephone Labs.) for sending us copies of these papers in advance of publication. lo Favre Haworth McKenna Powell and Whitfield J. 1953 11 15; and other papers by Professor Haworth Dr.McKenna and their colleagues. l1 Turner and Voitle J. Amer. Chem. SOC., 1951 73 2283. l2 Cerny Habler and gorm Chem. Listy 1957 51 2344. l3 Schatz and Clopp f. Amer. Chem SOC.,1955,77 5113. JUNE 1961 207 The Preparation of 5,9-Cyclosteroids By 0. GNOJ,E. P. OLIVETO, and C. H. ROSINSON (SCHERINGCORPORATION NEWJERSEY) BLOOMFIELD and D. H. R. BARTON (IMPERIAL OF SCIENCE LONDON,S.W.7) COLLEGE AND TECHNOLOGY REDUCTIONof 11(3,21-diacetoxy-9a-bromo-l7a-hydroxypregna- 1 ,4-diene-3,20-dione1 (I; R = Ac) with chromous chloride in acetone gives a bromine- free product which is not the expected 1,4,9(11)- triene2(V). The new compound has m.p. 186-191 " [a] + 254" (all rotations are for 1 % solutions in dioxan) Amax.272 mp (E 4900; all ultraviolet spectra for MeOH solutions) Vmax. 3.00 5.70 5.78 6.10 and 8-15 p (all infrared data refer to Nujol mulls). The infrared bands indicate an intact 1 lp-acetate and side chain and the presence of an cup-unsaturated ketone system. The ultraviolet absorption slpectrum is peculiar and closely comparable with thatS (Amax. 269 mp; E 8700)for 1-dehydrocycloartanone (as VLT). The 5,9-cyclo-structure (11; R = Ac) is established by these and the following facts. CHiOAc "w intensity ultraviolet absorption and infrared bands at 2.85 5.70 5-78 8.05 and 8.14 p but showing no vinyl hydrogen in its nuclear magnetic resonance spectrum. On treatment with hydrogen chloride it afforded the known 17 a,21 -dihydroxypregna-4,9( 11)-dime-3,20-dione 21-acetate4 (VI).Extension of the chromouschloride reaction to the 1 lp-formate (I; R = CHO) gave the corresponding 5,9-cyclo-steroid (11; R = CHO) m.p. 189-193" [a], + 254" Amax. 272 mp (E 4900) Vmax. 2.94 5.75 5.8 to 5-84 8.15 and 8.6 p. Similar reduction of the 1lb-hydroxy-compound (I; R = H) afforded the triene (V) (major product) and the 1,ll-oxide (IV) m.p. 177-180" [alD+ ll" having no high-intensity ultraviolet absorption and infrared bands at 2-95 5-72 5-76 5-84 and 8.12 p. This oxide was RoV\ 1. CrCI,. 2 H,-Pd. 3 Flavobacterium dehydrogenans. 4 HCI-CHCI, Treatment of the diacetate (II; R = Ac) with hydrogen chloride in chloroform or with zinc bromide in ethanol furnished the triene (V). Hydro-genation in ethyl acetate-ethanol over palladised calcium carbonate gave a dihydro-derivative (TII; R = Ac) m.p.163-165" [a] -59" having no high- also obtained by the action of Flavobacterium de-hydr0genans5 (ester hydrolysis) on the diacetate (11; R = Ac) to give the corresponding diol which at once isomerised to the oxide (as LV; 21-hydroxyl) m.p. 208-213" [a],-14" vmax. 2-96 5.82 and 5.86 mp reacetylation of which by pyridine-acetic Robinson Finckenor Kirtley Gould and Oliveto J. Amer. Chem. SOC.,1959 81 2195. Fried Florey Sabo Herz Restivo Borman and Singer J. Amer. Chem. SOC.,1955 77 4181. Irvine Henry and Spring J. 1955 1316. * Fried and Sabo J. Amer. Chem. Soc. 1957 79 1130. Charney Weber and Oliveto Arch. Biockm. Biophys. 1959 79,402. PROCEEDINGS anhydride afforded the 21-acetate (W).Treatment of catalysed transformation products of these com- this acetate (1V) with hydrogen chloride gave the pounds. usual triene (V). The mechanistic principle whereby these 5,9-The stereochemistry of the product (XI; R = Ac) cyclo-steroids are formed is possibly as already is defined6s7 by the lOP-methyl group as indicated in indicated (I; see arrows). In agreement the com- formulz-(U) and (111). Models of the alcohol (11; R pound (11; R = Ac) was also obtained from the = H) show that the 1 1p-hydroxyl group is placed in iodo- but not from the chloro-analogue of (I; R = space in close proximity to the 1-position so that the Ac). formation of a lp,ll/i?-oxide (IV) is to be expected. We have prepared in an analogous manner the We thank Professor J.Meinwald (Cornell) for 5,9-cyclo-steroid (VIII) m.p. 138-143 " [a] helpful discussion and Dr. J. W. Lown (Imperial + 184" Am,,. 272 mp (E 4900) Vmax. 5.78 6.02 College) for the nuclear magnetic resonance 8.04 and 8.45 p its dihydro-derivative m.p. spectrum 120-122" [a] -150" and the expected acid- (Received April 6th 1961 .) Cf. Barton Bernasconi and Klein J. 1960 51 1. Barton and Gilham Proc. Chem. SOC.,1959 391 ;J. 1960,4596. Proton Magnetic Resonance Spectra of Thallium Triakyis; Chemical Exchange and Formation of Mixed Alkyls By J. P. MAHERand D. F. EVANS COLLEGE OF SCIENCE (IMPERIAL AND TECHNOLOGY SOUTHKENSINGTON LONDON,S.W.7) THALLIUM has two isotopes both with nuclear spin 6 methane the collapse of the spin multiplets due to and with very similar nuclear magnetic moments splitting by the thallium nuclei with increase in 203Tl(29.5% abundant p = 1.5960) and 205Tl temperature enables1 the activation energy for the (70.5 % abundant p = 1.61 14).These isotopes pro- exchange to be estimated as 6 f 1 kcal. mole-l. This duce large spin-spin splittings in the proton reson- exchange which was found to be concentration- and ance spectra of dialkyl- and diaryl-thallium cations solvent-dependent presumably involves interchange in aqueous solution. The results obtained will be sub-of methyl (or ethyl) groups between two monomeric mitted for publication elsewhere. However tri- trialkylthallium molecules. methyl- and triethyl-thallium in solvents such as Tn agreement with this interpretation mixtures of Proton magnetic resonance results for trialkyl thalliums in dichloromethane at ca.-85 ". Me group Et group 2 0 5 ~ ~ ~ 205T1-H(CH& 205T1-H(CHJ Compound coupling 7 coupling r coupling 7 constant (p.p.m.) constant CH2 constant CH3 (cycIe/sec.*) (cycle/sec.*) (p.p.m.) (cycle/sec.) (p.p.m.) TIMe 250.8 9.48 -I -TIMe,E t 223~~ 9.58 242- 8-66 472. 8-20 TIMeEt 186~~ 9.73 218.8 8-7 1 441- 8-18 TIEt -198- 8.71 396*1 8.21 The H(CH,)-H(CH,) coupling constant was 7.7 f0.3cycle/sec. in all the ethyl derivatives. * These values will be slightly low (by < 1 cyclelsec.) since the zo5T1/20ST1 splitting was not resolved. dichloromethane and toluene give only a single trimethyl- and triethyl-thallium in dichloromethane broad proton-resonance line at room temperature.when cooled to -85" give proton-resonance lines At -60" to -100" the spectra resemble those for additional to those due to the pure trialkyls. By vary- the dimethyl- and diethyl-thallium cations indicating ing the initial concentrations of the two components that a rapid chemical exchange is occurring at room it was possible to detect the presence of the mixed temperature. For trimethylthallium in dichloro-alkyls TlMe2Et and TlMeEt,. The spin-spin Pople Schneider and Bernstein "High Resolution Nuclear Magnetic Resonance," McGraw-Hill New York 1959; Arnold Phys. Rev. 1956 102 136. JUNE1961 coupling constants and chemical shifts for all four trialkyls are given in the Table and a typical spectrum is shown in the Figure.The 205Tl-(H)CH coupling constants are uni-formly greater than the 20?%(H)CH coupling con- stants. This has previously been observed for ethyl groups attached to heavy metals such as lead2 and mercury which have nuclei with spin 9. It is also probable that the two coupling constants opposite signs. This technique should be useful in obtaining relative signs of coupling constants in spectra where the separations between the groups are large compared to the splittings within them. Chemical evidence for mixed alkyl formation was also obtained since hydrolysis of a mixture of tri- methyl- and triethyl-thallium gave a solution con- taining the stable unsymmetrical cation TlMeEt+ in addition to the TIMe2+ and TIEt,+ cations.This Top scale = r (p.p.m.). Lower scale = cycles/sec. Proton magnetic resonance spectrum of a mixture of trimethyl- and triethyl-thallium in dichloromethane at -85”. Et groups A,A’ CH in TlMe,Et; B,B’ CH in TIMeEt,; C,C‘ CH in TIMe,Et; D,D’ CH in TIMeEt,. Me groups E,E’ TIMe,; F,F’,TlMe,Et; G,G TIMeEt,. There is considerable overlap of the lines arising from the Et groups. Diferential splitting by the two thallium isotopes is visible for A,A’ and B,B’. X = CH,CI,. Y = SiMe,. 205T1-(H)CH3 and 20%-(H)CH, have opposite signs as in the diethylthallium cation. In this cation as in triethylthallium itself at low temperatures the thallium nuclei split the proton resonances into doublets. Together with the smaller splitting within each ethyl group this gives rise to two outer CH triplets and two inner CH quadruplets.By using the proton-spin decoupling apparatus constructed by T~rner,~ it was found that an inner CH quadruplet appeared as a single line when the distant CH triplet was simultaneously subjected to an intense radio- frequency field and vice versa. This shows unam- biguously that the two TI-H coupling constants have mixed species was also detected by its proton- resonance spectrum (at room temperature). Measurements were made at 56-4 Mc. on a Varian V4300B spectrometer with a Varian V-4340 variable temperature probe. The authors thank Dr. D. W. Turner for assistance with the double resonance experiment Dr. L. Pratt for helpful discussions and the Department of Industrial and Scientific Research for a maintenance grant (to J.P.M.).(Received March 30th 1961.) Baker J. Chem. Phys. 1957,26 960. Narasimhan and Rogers,J. Amzr. Chem. SOC.,1960 82 34; Dessy Flautt Jaffe and Reynolds J. Chem. Phys., 1959,30 1422. Turner unpublished work. Biosynthesis of Aromatic Substances from Acetyl- and Malonyl-coenzyme A By JOHN D. Bu’L~cK and H. M. SMALLEY (THE UNIVERSITY 13) MANCHESTER THE appearance of the Communication by Bentley and Keill on the function of malonate as a part- precursor of penicillic acid (I) prompts us to report our work on the biosynthesis of 6-methylsalicylic Bentley and Keil Pruc. Chem. SOC.,1961 111. a Mosbach Acta Chem. Scand. 1960 14,457. acid (111). Penicillic acid is one of a number of substances demonstrably2 or hypothetically derived from “acetate”-presumed to be in the form of acetyl-coenzyme A-by way of orsellinic acid (11) in the Penicillium cyclopium series.In the Penicillium urticae series 6-methylsalicylic acid is of correspond- ing importance as the demonstrated3 or presumed parent of a large number of derived metabolites. In this case the work of Bassett and Tanenbaum4 has shown conclusively that the precursor form of “acetate” must be acetyl-coenzyme A. We have now shown that in the conversion of “acetate” into 6-methylsalicylic acid three of the four C units are incorporated by way of “malonate” presumably in the form of malonyl-coenzyme A arising from the carboxylation of the obligatory intermediate acetyl-coenzyme A.Using shake cultures of a P. urticae strain during the relatively short period when 6-methylsalicylate production is not complicated by further reactions we have found up to 15% incorporation of added diethyl [2-14C]maionate in an incubation period of 6 hours. On degradation of the isolated and pursed 6-methylsalicylic acid by the Kuhn-Roth procedure all of the incorporated labelling appears as 14C0 and none remains in either carbon atom of the acetic acid. This is in sharp contrast to the incorporation of MeOC=CH II Me.$-C ,Co lo H2C HO (I Endogenous PROCEEDINGS with accompanying decarboxylation as indicated in the reaction scheme. The analogy between this pro- cess and that of fatty acid biosynthesis5 is therefore close the most essential difference being the omission of the reductive steps which characterise fatty acid synthesis; such an omission can be seen as a neces- sary prerequisite for cyclisation to the aromatic ring.The demonstrated origin of 6-methylsalicylic acid (HI) is identical with that inferred by Bentley and Keil for orsellinic acid (II) from degradations of the derivative (I) ;we suppose that the “acetate-derived” substances in general are formed from “acetate + n(malonate).” The difference between one C unit (acetate-derived) and the remainder (malonate-derived) in the aromatic metabolites whilst most clearIy brought out by the differential incorporation of labelled malonate is reflected to a smaller extent in the in- corporation of acetate itself.Birch has observed6 that in molecules such as griseofulvin7 and cur-vularin8 the incorporation of acetate into the “terminal” C2unit is higher than into the remaining C units; this is readily understood if in forming the Endogenous t-+ co I -+ CHiC02H CH;CO-CoA 2H02C*CHiCO-CoA --CH,(CC$Et) { I mol. + 3 mols. A--F-$.-CO :Groups lost j -CH,-C-FH-$-$H [CO2][01 [Cq] 0 [co,] bl I OH [l -14C]acetate under similar conditions which is virtually equal in the four constituent “CZunits” so that on Kuhn-Roth oxidation about three-quarters appears as lT0 and one-quarter as CH3.1T0,H. The aromatic substance 6-methylsalicylate is therefore built up from four molecules of acetyl-coenzyme A by the conversion of three into malonyl- coenzyme A and the linear combination of these three with the one remaining acetyl-coenzyme A Im> latter units the labelled acetate is diluted not only by endogenous acetyl-coenzyme A but also by endogenous malonyl-coeiuymz A.It is also of interest that in an (acetatefma1onate)- derived metabolite there will in general be a dis- tinguishable “end” to the assembly of c units which is in good accord with general views on the biogenesis of such compounds6 but conflicts with more elaborate variants e.g. formation of 6-methyl- Bu’Lock and Ryan Proc. Chem. Sac. 1958,222. Bassett and Tanenbaum Biochem. Biophys. Acta 1960,40 535. Eg. Homing Aartiu Karmen and Vagelos Biochem. Biophys.Research Comm. 1960,3 101. Birch Chem. SOC. Special Publ. 1958 No. 12 p. 1. Birch Massy-Westropp Rickards and Smith J. 1958 360. Birch Musgrave Rickards and Smith J. 1959 3146. JUNE 1961 211 salicylate from hydroxymethylglutarate and acetate replaced by other acyl units ; obvious examples or from two C units. The fact that the “end” C which should be capable of experimental verification unit has a different origin to the remainder leads to are replacement by a propionyl unit as in rutilanti- the concept that (as in fatty acid synthesis6) it may be none9 or by a cinnamoyl unit as in the flavonoids. (Received ApriZ 5th 1961.) !’ Ollis Sutherland Ccdner Gordcn and Miller Proc. Chem. SOC.,1960 347. The Stereochemistryof Columbid By K. H.OVERTON, N. G. WEIR,and A. WYLIE (CHEMISTRY ”HE UNIVERSITY, DEPARTMENT GLASGOW) THE recent disclosure2 of the constitution and stereochemistry of clerodin prompts us to publish our results concerning the stereochemistry of the structurally related bitter principle columbin? Hydrolysis of is~dihydrocolumbin~ (as I; R = H; but single bond at position 2 3 and 8/3-H) with N-sodium hydroxide at 100” afforded the dicarboxylic acid (II;R -C02H,R’ = OH) in analogy with the compound obtained (but not formulated) from methyl isocolumbin (as I; R = Me but 819-H) by Wessely and Jentzs~h.~ Oxidation of the diacid from isodihydrocolumbin with chromic acid in pyridine furnished the nor-ketone (LI; R + R‘ = 0). Similar hydrolysis of the octahydroisocolumbinic acid (111; R = H 8/3-H) m.p.186-188” pK 5.1. [Vmax. of methyl ester 1759 (boat s-lactone) and 1741 cm.-l (ester)] (obtained cleanly as the major acidic product from hydrogenation of isodihydro- columbin) afforded the “isolactone” (N; R = CO,H R’= OH) pK 3-98 [vrnax.of methyl ester 1742 cm.? (normal S-lactone and ester)] which was stable to acetic anhydride and pyridine at 100”. Oxidation of the “iso-lactone” with lead dioxide in acetic acid5 afforded the (oily) nor-ketone (N;R + R’ = 0),vmax. 1741 (&lactone) and 1712 (cyclo- hexanone) cm.-l characterised as the crystalline derived alcohol (IV;R = OH R = H) which was smoothly reoxidised to the nor-ketone. Hydrolysis of the nor-ketone and immediate methylation furnished the (oily) ketol ester (V; R + R’ = 0 R” = Me) (crystalline acetate).This was oxidised by chromic acid in pyridine to the dione ester (VI) m.p. 128-140” which with seieniurn dioxide in acetic acid gave the expected 2,3-unsaturated diketone m.p. 94-107” Amax. 228 mp (E 9,500). Both oxida- tion products appear to be stereoisomeric mixtures (presumably at C(lo)),which have not been resolved. The octahydrocolumbinic ester (ID; R = Me 8a-H) m.p. 129-1 3 1O obtained by hydrogenation of dihydrocolumbin and separated from its 13-epimer by extensive fractional crystallisation and chromato- graphy gave on hydrolysis as expected a dihydroxy- dicarboxyfic acid (V; R = C02H R’= OH R” = H) not a lactonic hydroxy-acid. F F =p-Fury1 ; F (4H) =p-Tetruhydrofuryl The formation of the “iso-lactone” in high yield under relatively mild conditions imposes stringent conditions on the stereochemistry of columbin which is best represented as in (I) or the enantiomer.In particular it requires a cis-fusion of rings A and B,* and attachment of the methyl group at position 5 to the side opposite the lactone A. The 19-methyl group is placed on the side of the molecule opposite the 18-methyl group since (a)this readily permits the nucleophilic displacement at C(l) leading to the * The stereochemistry of columhh proposed by Cava Weinstein and Malhotra6 is based on the biogenetic assumption of an AIB trans-fusion which is now shown to be inadmissible. Presented at a Chemical Society Symposium on Terpene Chemistry on February 23rd 1961 at Imperial College London.* Sim Hamor Paul and Robertson Proc. Chem. Soc. 1%1,75; Barton,Cheung Cross Jackman and Martin-Smith Proc. Chem. SOC.,1961 76. Barton and Elad J. 1956,2085. Wessely and Jentzsch Monatsh. 1937 70 30. li Conroy J. Amer. Chem. SQC.,1957,79 1726. @ Cava Weinstein and Malhstra Tetrdedron Letters 1959 No. 15 I. 212 PROCEEDINGS system (11) and (b) the columbin4socolumbin experiments implies a biogenesis which differs funda- change then becomes the change from a cis-to a mentally from that of clerodin. trans-fused lactone. The absolute stereochemistry at Infrared spectra refer to CCl, ultraviolet spectra position 12 (P-H) follows from application of the to EtOH and pK’s to water-tetrahydrofuran Hudson-Klyne lactone rule7 to the more easily solutions.hydrolysed lactone B. Taken in conjunction with sup- porting evidence the configuration at C!,, appears We thank Professor D. H. R. Barton for his to favour the structure (I) for columbin over its critical interest and the D.S.I.R. for a maintenance enanti0mer.t grant (to A.W.). The stereochemistry of columbin disclosed by our (Received March 30th 1961 .) t This assumption is implied in the structures and designations used throughout. Klyne Chem. and Id. 1954 1198; Barton and Gilham Proc. Chem. SOC.,1959,391. Free-radical Arylation and Aroylation of Aromatic Compounds Induced by Gamma-radiation By A. F. EVERARD and G. A. SWAN OF GEMISTRY,KING’S COLLEGE OF DURHAM), (DEPARTMENT (UNIVERSITY NEWCASTLE UPON TYNE 1) PREVIOUS work1 suggested that y-radiolysis of for the bromobiphenyl mixture = 0-638) compared bromobenzene gives bromine atoms and phenyl with that in irradiation 1 (0.812) are probably due to radicals and that the latter can attack aniline or the greater dose.The isomer ratio is consistent with pyridine (if present) giving rise to nuclear phenyla- the assumption of an initial free-radical attack of tion products in the isomer ratio expected of radical phenyl radicals on bromobenzene. attack. y-Irradiation of bromobenzene gave rise to a Although the arylation of aromatic compounds by complex mixture of products,2 including the three chemically generated aryl radicals is well known we isomeric bromobiphenyls. We have now determined are unaware of any similar case where aroylation by the ratio of these isomers by the isotope dilution free aroyl radicals has been discussed.However the method. Aliquot parts of irradiated bromo [14C]-formation of benzophenone by the action of zinc on benzene were treated with known weights of the a mixture of benzoyl chloride and benzene5 and of respective bromobiphenyl which was then isolated. 4-methoxybenzophenone by a similar reaction with 4-Bromobiphenyl was isolated as such but the 2-and anisoles may proceed by a radical mechanism. Also the 3-isomer were obtained as crystalline dinitro- benzoyl radicals have been postulated as inter-derivatives. The accompanying Table shows the G mediates in a variety of reactions. values and isomer ratios obtained in two such experi- We have now shown the formation of benzo-G Values of the isomeric Isomer ratio (%) of the Mode of generation of phenyl radicals bromo bipheny Is bromobiphenyls 2-3-4-2-3-4-Irradiation 1 :dose = 2-44 x 1021 ev/ml.0.387 0.295 0.130 47-7 36.3 16.0 , 2 dose = 4.22 x 1021 ev/ml. 0.303 0-221 0.114 47.5 34.6 17.9 Benzoyl peroxide (ref. 3) 49.3 33-3 17.4 9 Y? (ref. 4) 48-5 33.0 18.5 ments together with the values reported for the phenone by y-irradiation of a mixture of benzene and latter in the phenylation of bromobenzene by phenyl benzoyl chloride (37.5 1 by volume). This was radicals formed by the thermal decomposition of isolated after chromatography as the 2,4-dinitro- benzoy 1 per~xide.~~~ phenylhydrazone with G = 0-06(dose 1.2 x 1021 The lower G values obtained in irradiation 2 (totalG ev/ml.) and was accompanied by biphenyl (G = 0-2) Swan and Timmons J.1958,4669. Parrack Swan and Timmons unpublished work. Augood Cadogan Hey and Williams J. 1953 3412. Dannley Gregg Phelps and Coleman J. Amer. Chem. Suc. 1954,76,445. Zincke Ber. 1873 6 137. Kaufmann and Fuchs Arch. Pharm. 1924,262 119. JUNE 1961 and a gum. These G values and those given for the following experiment are almost certainly much too low on account of losses during isolation. We also irradiated a mixture of anisole and benzoyl chloride (37.5 :1 by volume). By chromatography followed by the use of the Girard T reagent a mixture of the three isomeric methoxybenzophenones was isolated (G = 0.22).The 2- 3- and 4-isomers as estimated in the mixture by the infrared bands at 762 727 and 744 cm.-l respectively in nitromethane solution were found to be in the ratio of 54 19 and 213 27%. These figures suggest a free-radical attack of benzoyl radicals on anisole. We thank the United Kingdom Atomic Energy Authority (Research Group Harwell) for financial support including the provision of a maintenance grant (to A.F.E.) during the earlier stages of this work which was completed under contract with the United States Air Force. (Received March 29th 1961.) Bonding of the Selenocyanate Ion in Complexes with Cobalt(1x) By A. TURCO,C. PECILE,and M. NICCOLDTI OF INORGANIC AND (INSTITUTES CHEMISTRY PHYSICALCHEMISTRY UNIVERSITY OF PADUA,ITALY) THE bifunctional nature of the thiocyanate ion as a Iigand in inorganic compounds is now well estab- lished.On the other hand information on seleno- cyanate co-ordination compounds is limited almost wholly to compounds with metals such as mercury platinum and silver for which it is easy to predict bonding to the selenium atom. Though this difference in bonding between thiocyanate and selenocyanate is surprising we know of no compound with seleno- cyanate in which the co-ordination is through nitrogen. Such compounds would be expected with the metals of class (a)character particularly where the binding is through nitrogen in the corresponding thiocyanate compounds but it has been reported that selenocyanate complexes are less stable than the corresponding thiocyanates with the central atom bonded to the nitrogen and also1 that "cobalt ion and KSeCN give blue solutions containing a very weak complex unstable in aqueous solution." It is known2 that in the Co(NCS),2- ion the cobalt atom is tetrahedrally co-ordinated to the nitrogen ends of four thiocyanate groups.We here report the isolation of the CO(NCS~),~- ion in the form of its tetraphenylarsonium salt. Blue-green crystals are obtained by precipitation with tetraphenylarsonium chloride from ethanol solutions of cobaltous chloride in the presence of a stoicheiometric amount of potassium selenocyanate (Found Co 4.7; Se 25.4. (AsPh,),Co(NCSe) re-quires Co 4-7; Se 25.4%). The molar conductivity in nitromethane at 20" is 168 mho in 5 x lO-*~-solut-ion and corresponds well with that of similar uni-bivalent electrolytes in the same solvent? The spec- trum of (AsPh,),Co(NCSe) in nitromethane is given in the Figure together with that of (AsPh4),CoC1,.The former can be almost perfectly superimposed in the 500-700 mpregion on that of (AsPh,),Co(NCS) (not shown) and both compounds give a sharp peak at 625-627 mp. This shows that the cobalt atom must be bound to nitrogen as it certainly is in the latter compound. Indeed the position of NCSe- and I I6001 9 '0 1200--4 '800-n 51 U 400- I Absorption 800 n { -600 1I II;-,{ '\ \ \ -400 I 't -200 / ,e' I / \ 1 I Wavelength (mp) spectra of 5 x 10"M-solution of (AsPh,),Co(NCSe) (solid line Iej? axis ofordinates) and (AsPh,),CoCI (broken line right axis of ordinates) in nitromethane.NCS- in the spectrochemical series is likely to be the same only when these ligands are linked to a metal through the nitrogen atom. The large value of Emax. (ca. 1700 at 627 mp) is not unexpected for tetrahedral cobalt(@ complexes. The magnetic moment at 20"is 4.5 B.M. and corresponds well with the value ex- pected for tetrahedral cobalt(r1) c~mplexes.~ It can be Toropova Zhur. Neorg. Khim. 1956 1 243. Zvonkova Zhur. fiz. Khim. 1950,24 1339. Gill and Nyholm J. 1959 3997. Holm and Cotton J. Clrem. Phys. 1960 32,1168. compared with the values 4.44 B.M. found for Co(NCS),Hg 4-5 B.M.for the CO(NCS),~- ion and the vaIue 4-4 as measured by us for Co(NCSe),Hg. This rules out the existence of possible co-ordination octahedra which might be completed by selenium ends of the selenocyanate groups. All the evidence given here indicates that in the Co(NCSe),"-ion the cobalt must be tetrahedrally co-ordinated through the nitrogen atoms showing PROCEEMNGS thus that SeCN- has similar co-ordinating properties to SCra-. One can then anticipate that SeCN- must also be able to bind as a bridge two different metal atoms and this may well be the case for Co(NCSe),Hg. One of us (M.N.) is indebted to the National Research Council of Italy (C.N.R. Rome) for a Research Fellowship. (Received April loth 1961.) Thiourea as a Leaving Group from Carbon in S Displacements By B.SAVILLE (THE NATURAL PRODUCERS' RESEARCH RUBBER ASSOCIATION WELWYN CITY,HERTS.) GARDEN WHEREAS the alkaline hydrolysis of the S-formami- m.p. 152" (cf. Werner') quantitatively on further dinium bond in organic thiouronium cations (I) treatment with iodine. This hydrolysis is almost is recognised as a standard procedure in thiol synthesis examples have not been reported previously of heterolysis in the sense (I> x:?R~-c~NH~).NH,-'X-R + SC(NH,) i.e. either an SJ ionisation proms of the R-S bond or a himdecular attack by a nudeophile on R dis-placing thiourea (SN2).Evidence on the existence of the above reaction types has now been adduced as folIows. S-1,3-Dirnethylbut-2-enylthiouronium bromide on being heated in dilute aqueous solution for 10 nlinutes at 75" gives acid and thiourea quanti- tatively RS-C(:NH2+)-NH2 4ROH + SC(NH2)2+ H+ .-. -. -This is shown when the aqueous solution no longer gives the corresponding thiouronium picrate with picric acid but gives dithioformamidinium dipicrate Werner (J. 1912 101 2166) gives m.p. 154". certainly of the SJ category gaining its driving force from the potentialities of the 1,3,3-trimethyl- ally1 system as a highly stabilised carbonium ion. When the thiouronium bromide (0.1 mole) is heated under reflux for 15 minutes with dimethyl- ammonium dimethyldithiocarbamate (0.2 mole) in 50% aqueous methanol (100 ml.) and the mixture diluted with water there results in 90% yield a material m.p.56-57" that appears to be S-1,3-di- methyltmt-2-enyl dimethyldithiocarbamate (I1 :R = Me& CHCHMe.) Me,N+CS- + R.SC( N H,i).N H,-+ RS.CS.NMe (It) + SC(NH2) It is not known by which mechanism this product is formed although S,l ionisation of the alkenyl-S bond followed bygttack of dimethyldithiocarbamate ion on the trimethylallyl carbonium ion intermediate seems probable on general grounds. (Received March 29th 1961.) Proton Magnelk Resonance Shifts in Steroid D-Homoannulation B. H. ARISON, By N. R. TRENNER D. TAUB,and N. L. WENDLER SHARP& DOHEJIE LABORATORIES, (MERCK RESEARCH RAHwAY NEWJERSEY,U.S.A.) THEformation of new substances by application of novel transformational conditions to 17-hydroxy 20-keto-steroids(I) often poses the problem whether the nod steroid D-ring has kretained or whether acyloin rearrangement to a D-homo-isomer 01 or 111) has occurred.A comprehensive study of thirty such D-homo-transformation products has shown that proton magnetic resonance spectroscopy provides a convenient means for this differentiation. The bases for differentiation are two-fold. First as would be expected in terms of the general theory of proton chemical shifts the protons of the 2l-mthyI ~QUPadjacent to the 20-keto-group (cf. I) whose proton resonances are to be found in the 7-7 T JUNE 1961 region move up-field about 0.9 unit on D-homoan- nulation. Similarly those of the 21-acetoxymethyl group resonances found in the 5.1 r region move 0.6 unit up-field.In all cases the variation of these D-homoannulation shifts was found to be less than 0.1 7Unit. (n> (m) R = H or OAc ; R'= H,OH,or Me Secondly analysis of our results has enabled us to arrive at a self-consistent set of 18-methyl incre mental shifts related to the known structural features of these D-homo-steroids (see Table) analogous to those found for the normal steroids by Shoolery and Rogem4 Two examples will be cited. For l7P-acetoxy-methyl- 17 a-hydroxy- 168 -methyl-D-homoandrosta- 4,9-diene-3,17a-dione6 the 18-methyl position would be calculated as follows 9-21 + (-0-18 + 0.06 -0.08 + 0-03) = 9.04. The observed 18-methyl position is 9-00. For 3a-acetoxy-17aa-hydroxy-17aP-methyl-D -homo -58 -androstane- 1 1 Incremental 1 %Methyl Chemical Shifts Associated with D-Homo Ring Substituents (Tau Units).' Normal unperturbed position of 18-methyl for D-homo class = 9-21 in deuterochloroform solution.Function 17 C=O 17 CX-CH Increment a1 Function shift (-downfield; + up field) Incremental Incremental shift Function shift (-down field; + up field) (-down field; + up field) 17a C=O 17a a-or p-CH3 17a a-or /3-CH2.0Ac 17a a-or /%OH 17a P-COCH,*OAc -0*18b +Om10 +Om05 +0*05 -015' 0.00 16 a-CH3 -0.03 -0.07 +0-0@ 16 p-CH3 16 a-OH +0.03 -0.04e -0.08 16 /%OH -0.05' -0.06 17 /3-CH3 or P-CH,.OAc 17 a-OH 17 P-OH a The incremental shifts for the indicated functions are preponderantly those to be associated with substituents not in close proximity to the 18-methyl group.b Note that a 1Zketo group in the normal steroids is associated with a -0-33 incremental shift (cf. Christensen Strachan Trenner Arison Hirschmann and Chemerda J. Arner. Chem. SOC.,1960 82 3995. C This is for the cortisone acetate-D-homocortisone acetate type of D-homoannulation where the C=O function does not enter the D-homo-ring (see ref. 3). d This value is similar to the incremental 18-methyl shift caused by a 16p-methyl group in a normal 17-keto-steroid which is = +0-08unit. 6 As applied to the 17a-keto-part-structure (11; R' = OH). The magnitude of these shifts was found to be essentially the same in both CDCl and deutero- acetone solutions. Indeed a value of 058 r unit was derived from our calculations based on the data recently published2 on D-homo-isomerisation of tri- amcinolone.Our results show that the gross use of the down-field shift of the 18-methyl group is at best only of very limited use as exemplified by cortisone 21-acetate vs. D-homocortisone acetate where only the D-ring size change^.^ While it is generally true that the 18-methyl resonance tends to move down-field in most steroid D-homoannulations the value is variable (-056 to +0.10) and depends on the detailed location and configuration of the substituents on the expanded D-ring. 17-diones we calculate 9.21 + (f0.10 + 0.05) = 9.36; the observed value is 9.40. In the same way the 18-methyl positions for thirty D-homo-annulation products have been calculated with as good or better agreement.The deduction by Shoolery and Rogers4 of -0.30 unit for the shift of the 18-methyl group on changing from a 5-membered to a 6-membered D-ring based on the difference between the positions of the un- perturbed 19-methyl and 18-methyl groups in steroids did not take into consideration the possible influence of the c-ring flanking the 19-methyl group in the D-homo-series. Our studies show that expan- sion of the D-ring of 3a-acetoxy-5~-androstane-11,17-dione 3a 1 lfl-dihydroxy- 16ar-methyl-5~-androstan-Tiers J. Phys. Chem. 1958,62 1151. Smith Man Garbarini Foell Origoni and Goodman J. Arner. Chem. SOC.,1960 82 4616. "NMR at Work," No. 51 Varian Associates Instrument Division Chem. and Eng. News Sept.22nd 1958 p. 59. Shoolery and Rogers J. Arner. Chern. SOC.,1958 80 5121. Taub Hoffsommer Slates Kuo and Wendler J. Arner. Chern. SOC.,1960 82,4012. Wendler Taub Dobriner and Fukushima J. Arner. Chem. Soc. 1956 78 5027. PROCEEDINGS ~~ 17-one and of 3a-acetoxy-l6~-methyl-5/3-andro-Associates Model 4300B high resolution spectro- stan- 1 7-one7 give D-homo-products whose 1 8-methyl meter equipped with superstabiliser and phase resonances are shifted down field by -0-12 -0.13 detector operating at 60 megacycles. Spectra were and -0.16 unit respectively. In these transforma- run on 5-10 % solutions in deuterochloroform tion; the only change is that due to D-ring expansion placed in a spinning Wilmad precision bore tube. by one CH group.Subtracting the average of these The resonance positions were determined relative to shifts from the normal steroid 18-methyl value a benzene capillary as an external reference and of 9~35~ leads to 9-21 as the average absolute position scaled by the use of side-bands8 generated by a of the unperturbed 18-methyl group in D-homo- frequency-counter-calibrated Hewlett-Packard steroids. audio-oscillator model 200 CD. The shielding Finally as would be expected the 19-methyl numbers were calculated by using the equation' incremental shifts characteristic of the normal T =dv/vo+ 3-60 wheredv is the observed resonance steroids4 hold equally well for the D-homosteroids as displacement from benzene in cycles per second and do those of the 18-methyl group derived from vo is the spectrometer frequency in megacycles.The substituents located on ring c. constant 3.60 was determined through the use of The nuclear magnetic resonance data reported tetramethylsilane as an internal standard. Our precision in dv is + I cycle/sec. herein were obtained through the use of a Varian (Received March 13tk 1961.) Wendler Taub and Graber Tetrahedron 1959,7 173. Arnold and Packard J. Chem. Phys. 1951 19 1608. A Synthesis of DL-Willardiine By G. SHAWand (in part) J. H. DEWAR (CHEMISTRY INSTITUTE BRADFORD, DEPARTMENT OF TECHNOLOGY 7) A NEW L-a-amino-acid willardiine isolated from A solution of the acylisocyanate2 the seeds of Acacia willardiana by Gmelin,l was EtOCH :CHCO.NC0 in benzene with aminoacetal formulated as a-amino-/% 1 -uracilylpropionic acid (I) gave the acylurea (11) which with warm dilute sodium by virtue of its stability to hot hydrobromic acid and hydroxide solution afforded 1-(2,2-diethoxyethyl)- by comparison of its absorption spectra with those uracil (Ill) and this with acid gave l-uracilylacet- of 1-methyluracil and uridine.We now report a aldehyde (IV) n1.p. 207" (decomp. with darkening synthesis of the racemic form of willardiine by an from 175"). Reaction of the aldehyde with potassium unambiguous route which confirms this structure. cyanide ammonia and ammonium chloride (Strecker reaction) followed by hydrolysis with hydrochloric acid evaporation and adjustment of the solution to HO,C*CH (NH,) .CH,-N m0(1) pH 4 gave an -50% yield of DL-willardiine m.p.YNH 205-209" (decomp.) Amax. 262-263 mp (E 8750), 0 Amin. 229-230 mp (E 1500) at pH 1 and Amax. CD;' EtO*CH:CH.CO.NH.CO.NH-CH,.CH(CEt~ 265-266 mp (E 6250) Amln. 242 mp (E 3150) at pH 12. The compound also gave a blue colour with the ninhydrin reagent and its physical properties no ?Yo were in excellent agreement with those recorded for iEtO),CHCH,.N YNH 0HCCH;N the natural material. YNH (@ 0 (JV) 0 (Received April 6th 1961.) Gmelin 2.physiol. Chem. 1959 316 164. Shaw and Warrener J. 1958 157. o-Electron Delocalisation and The Electronic Structure of Dinitrogen Tetroxide By R. D. BROWNand R. D. HARCOURT (CHEMISTRY MONASHUNIVERSITY VICTORIA, DEPARTMENT CLAYTON AUSTRALIA) THEelectronic structure of the N,04 molecule is of bond between the nitrogen at0rns.l Three models particular interest because of the difficulty in have previously been proposed for the electronic accounting for the long bond length small force structure of the molecule.constant and low dissociation energy of the central (1) McEwen2 suggested that the structure could be Coulson and Duchesne Bull. Classe Sci. Acad. roy. Belg. 1957 43 522. McEwen J. Chem. Phys. 1960,32 1801. JUNE 1961 described as a superposition of separate NO struc-tures and bonding charge-transfer structures. How- ever this proposal does not make it clear why the N-N distance should be so great that a strong mixing of the two NO functions does not occur correspond- ing to the formation of a normal N-N bond. (2) The long weak bond is incompatible with classical valency formulae which imply slight 77-electron delocalisation across the central bond imparting to it a total bond order slightly greater than unity? We have carried through a n-electron V.E.S.C.F.M.O.calculation4 on the “a + 7~’’ structure obtaining a total u + n bond order of 1-085. (3) The “n-only” model in which bonding unites the two nitro-groups together no u-bonding being present between the nitrogen atoms was first sug- gested by Smith and Hedberg? and elaborated more fully by Coulson and Duchesne,l and by Mason.g This structure involves six n-electrons. These result in two almost degenerate n-electron ground-state configurations @a and @ (see Table). Application n-only model configuration of n-electsons Configurations Orbital -____ n h c all4 61s ? .1 f.1 f ? @a @ Qrc 9 0 00 00 00 FIG.I. a-Orbitals included in the V.E.S.C.F. treat-ment (negative lobes are shaded). of a suitable extension of the V.E.S.C.F. procedure to a linear combination of these yielded a N-N bond order of 0.087 (cf. Mason 0.1; Coulson and Duchesne 0.3) which implies a weak bond. In addi- tion there is also a 3Blu configuration,@, which is almost degenerate with the above singlet ground state. We have considered quantitatively the factors which might increase the singlet-triplet energy separation including a complete configuration- interaction treatment of all the lA configurations and of all the 3Blu configurations. The resultant energy separation was so small that it would require specimens of N,O at room temperature to be para- magnetic in contrast to the observed diamagnetism.’ A further disadvantage of the n-only model is that it implies that rotation of one NO group relative to the other by n/2 about the N-N bond would result in dissociation.However the relative magnitude of the observed internal-rotation barrie? and the bond- dissociation energy show that the N-N bond energy is appreciable even in the perpendicular conforma- tion. We now propose a new electronic structure for dinitrogen tetroxide.* It is suggested that the n-electron structure corresponds to the classical ‘‘0+ n” model 8 electrons being accommodated in n-orbitals but that the lone pair 2pZ-electrons on the oxygen atoms delocalise appreciably into the anti- bonding a-orbital of the central bond.A V.E.S.C.F. treatment of the 10a-electrons [the two 2p~-electrons from each oxygen and the two electrons which are classically assigned to the N-N bond (see Fig. l) other electrons being omitted on energetic grounds] and 8 n-electrons leads to the orbital energies as shown in Fig. 2 and a N-N bond order of 0.76 (a = 0.67; n = 0-09),which implies a considerably stronger bond than does the 7-r-only model yet a N-N bond length longer than for a pure single bond. On the basis of these calculations each of the singlet 77-only configurations @a and Qb represents doubly excited configurations 17-37 and 17-31 ev above the ground-state configuration.On this model there are no triplet configurations with energies close to the singlet ground state so that there is no difficulty in accounting for the observed diamagnetism of N,O,. * Since this note was prepared we have received two papers (Green and Linnett J 1960,4959; Trans. Faraday Sac., 1961 57 10) in which the structure of N204is described from a different viewpoint on the basis of simple Huckel molecular-orbital calculations. These authors have also concluded .that a u-and .rr-bond exists between the nitrogen atoms but overlooked the significant effect of u-electron delocalisation upon N-N and N-0 bonds. Chalvet and Daudel J. Chim.phys. 1952 49 77. Brown and Heffernan Trans. Faraday SOC.,1958,54 757; Austral. J. Chein. 1959 12 319 330 543 554; 1960 13, 33 49.The technique used in the present calculations was based on method BJ in these references; it included n-a exchange and the resonance integrals were evaluated from the formula BCLv= -Spv (I b)/2 rather than from molecular spectroscopic data. Smith and Hedberg J. Chem. Phys. 1956 25 1282. Mason J. 1959 1288. Havens Phys. Rev.,1932,41 337. Snyder and Hisatsuma J. Mol. Spectroscopy 1957 1 139. PROCEEDINGS In Fig. 2 it can be seen that the 0-and n-molecular orbitals intermingle and hence are not energetically separable into two distinct groups. On the basis of the order of these N204orbitals it seems possible to interpret the observed molecular symmetry and (-I-.---.-_-+-10.04 central-bond properties of many other X,Y systems.(=I---..-*.---I! *9/ --,.,-:.-.---',.. -__ -The extensive lone-pair delocalisation in N20 w:-,.; \.'. If -94 r '\ introduces a further contribution to the N-0 bond-(s).',,' ,,-.,'-12.07 I ,, ing in addition to that from the 7r-and the (assumed) :*)' .,/' ,,- '-,,'\-12.22 \, I, localised NO a-electrons. We obtain a total N-O ' .. bond order of 2-02. However a considerable negative (&' ,,',/-\ .J\ ' '-16-42 '.. overlap exists between atomic orbitals such as an b 2g (Tr),,f,,,' \ '.-19.07 '\ oxygen 293;-orbital and the nitrogen hybrid orbital of 63 (+' '-20.76 the adjacent N-0 a-bond; hence delocalisation of Qg (&T)--* __---_---28-81 NO a-electrons should alter this value. FIG. 2. Eigenvalues of the Hartree-Fock operator for theground state of N204.The Iowest nine values repre- (Received March 27th 1 961 .) sent energies of occupied orbitals.Quinokidine Stereochemistry T._M. MOYNEHAN and K. SCHOFIELD (WASHINGTON LABORATORIES OF EXETER), SINGER UNIVERSITY R.A. Y. JONESand A. R. KATRITZKY (UNIVERSITY LABORATORY, CHEMICAL CAMBRIDGE) CONFORMATIONAL arguments attributing greater are illustrated by the examples of cis-and trans-1 -stability to the conformation (I;R = H) than to (II; methylquinolizidine set out. perchiorate gave both of the 3-methylquinolizidines in comparable amounts their configurations re mained ambiguous (but see below). Bohlmannl demonstrated that trans-fused quino- (rD lizidines in which the nitrogen lone electron pair was trans to at least two axial hydrogen atoms on In a study of the influence of structural factors carbon atoms adjacent to it showed a prominent band at 2800-2700 cm.-l; this criterion has been upon conformational equilibria of the type (I -+II) and of the course of quaternisation of such bases we amply confirmed? Such a band is shown by all the have synthesised cis-and trans-1- 2- 3- and 4-methylquinolizidines except the trans-4-methyl com- methylquinolizidine (the prefixes referring to the pound.This indicates that the cis-fused ring con- relative configurations of the 10-hydrogen atom with formation is preferred in the latter case and the respect to that at position 1 2 3 or 4). These com- trans-fused in all the others. pounds have been described previously though This conclusion is supported by nuclear magnetic sometimes probably as mixtures but only in two resonance data.The C-methyl group appears as a cases were the configurations known.2Our methods more or less well-resolved doublet for all the com- Cookson Chem. and Ind. 1953 337; Leonard and Ryder J. Org. Chem. 1953 18 598; Bohlmann Chem. Ber. 1958,91 2157. * Boekelheide and ROSS, J. Amer. Chem. Soc. 1955,77,5691; Nesmeyanov and Rybinskaya Chem. Abs. 1958,6349. Ohki and Noike Chem. and Pharm. Bull. (Japan) 1959,7 708. JUNE 1961 pounds; the degree of splitting (by the adjacent hydrogen) should be less for equatorial than for axial methyl groups.* For the 1- 2- and 3-methyl com- pounds the splitting is 2-49 cycles/sec. for equa- torial and 5.7-7.0 for axial groups.For the 4-methyl compounds in which the methyl groups are expected to be equatorial on the basis of the infrared evidence the splitting (5.4 and 6.2 cycleslsec. for cis- and trans-4-methylquinolizidine,respectively) is rather high for equatorial groups in both cases. Possibly this is a result of the proximity of the nitrogen atom. A second nuclear magnetic resonance criterion is provided by the position of the centre of the methyl doublet. Previous work indicates that axial and equatorial groups cause absorption at different posi- tions; however the conformational effect is not always in the same direction? In the present series for the 1- 2- and 3-methyl compounds the peak for an axial methyl group is in each case at a lower field (by 0.09-026 p.p.m.) than that for the correspond- ing equatorial methyl group.For cis-4-methyl-quinolizidine the methyl group (T = 8.98) is almost certainly equatorial yet the resonance for the trans- 4-methyl compound (T = 9.07) is at a higher field. This indicates that the methyl group in the latter compound is not in the axial position and hence that the rings are cis-fused. As regards the quaternisation of these bases the following observations have been made. Quino-lizidine gives a methiodide m.p. 320” (decomp.).6 Treatment of 2-4’-ethoxybutyl-l-methylpiperidine with hydriodic acid gives a second quinolizidine methiodide m.p. 214” (decomp.) (salts with other anions also differ in the two cases). We believe the following facts show that the first series of salts possess the trans-fused ring structure [as (I) with Me on N+J,and the second series the cis-fused ring structure [As (II) with Me on N+] (i) cis-1-Methyl- quinolizidine provides the same quaternary salt as is obtained from cis-2-4’-ethoxybutyl-1,3-dimethyl-piperidine.trans-1-Methy lquinolizidine gives a different quaternary salt from that obtained from trans-2-4’-ethoxybutyl- 1,3-dimethyIpiperidine. (ii) The 3-methylquinolizidines of ambiguous configura- tion and the corresponding piperidines likewise give in the case of one pair the same quaternary deriva- tives and in the case of the other pair different quaternary derivatives. The spectroscopic evidence (infrared and nuclear magnetic resonance) indicates the 3-methylquinolizidine which gives quaternary derivatives the same as those from the piperidine to be cis-3-methylquinolizidine and the isomer giving quaternary derivatives different from those from the piperidine to be trans- 1-methylquinolizidine.These observations suggest that the bicyclic bases having in the trans-conformation (see I) axial methyl groups hindering approach to the nitrogen lone-pair [cis-1- (I; R = Me) and cis-3-methylquinolizidine] give quaternary derivatives in which the rings are cis-fused whilst the bases in which the trans-conformation places the methyl group equatorially (trans-1-and trans-3-methylquinolizidine) and quinolizidine itself give quaternary salts in which the rings are trans-fused. Similar arguments have been applied to the quaternisation of (-)-lupinhe and (+)-epil~pinine.~ These results illustrate the lack of any necessary connection between the outcome of the kinetic pro- cess of quaternisation and the position of conforma-tional equilibrium in tertiary bases.sd (Received March 29th 1961 .) Musher Spectruchim.Acta 1960 16 835. Lemieux Kullnig Bernstein and Schneider J. Amer. Chem. Soc. (a) 1957,79 1005; (b) 1958,80,6098; (c) Baggett, Dobinson Foster Homer and Thomas Chem. and Ind. 1961 106; (d) Closs J. Arner. Chem. Suc. 1959 81 5456; (e) Jensen Noyce Werholm and Berlin J. Amer. Chem. Suc. 1960 82 1256; cf) Brownstein and Miller J. Org. Chem. 1959,24 1886. Leonard and Wildmann J. Amer. Chem. Suc. 1949,71 3100; Winterfield and Dunwald Naturwiss.1956,43 517. ’Crow Australian J. Chem. 1958 11 366. PROCEED~NGS The Hexachlorodiborate Anion B2Cls2-By A. K. HOLLIDAY (THE Ummw~~ LIVERPOOL) M. E. PEACH,and T. C. WADDINGTON (UNIVERSITY LABORATORIES, CHEMICAL CAMBRIDGE) THE salts of the halogenoboric acids can be prepared from the boron trihalides in the corresponding liquid anhydrous hydrogen halide.ls2 This Communication describes attempts to prepare salts containing the hexachlororodiborate ion B2C1G2- and the penta- chlorodiborate ion B2C15- by using diboron tetra- chloride in liquid anhydrous hydrogen chloride. The solvo-bases tetramethylammonium chloride and phosphorus pentachIoride were titrated conducto- metrically with diboron tetrachloride.The titration curve for the former is shown in the diagram. There is a sharp break when the molar ratio B2C14 Me,NCl is 0.55 but there is little or no indication of a break at a molar ratio of one. With phosphorus penta- HI chloride the conductometric titration curve showed a break at a molar ratio of Bp, PCI = 0.55; after this point precipitation occurred. The values of 0-55 1 I I I obtained for the molar ratios in the two cases can be 0 05 I *o 1.5 2.0 explained in terms of slight contamination of the Molar ratio B2CI, Me4NCt B,Cl by BCI, formed during transference opera- tions. Both curves thus give evidence for the forma- FIG.-Condurtivity titration of B2CI4 with Me,NCI. tion of the hexachlorodiborate ion B,Clc- in solu- tion but none for the formation of B2C1,-.F'umping 25.1 ;H 6-3;C1 55.6; N 7-3%). The presence of a off the solvent and the excess of diboron tetrachloride boron-boron bond made analysis for the element left a white solid in each case. Both these solids were difficult and the figures obtained were unreliable. somewhat unstable on storage particularly the pro- Analyses of the phosphorus product were unsatis- duct from phosphorus pentachloride which appeared factory and probably reflect the instability of the to be volatile at -80"; the same phenomenon was compound. The infrared spectrum of [Me,N ],B,CI observed with tetrachlorophosphonium tetrachloro- showed strong peaks at 694 665 and 600 cm.-l due Analysis of the product from tetramethyl- to the B2C1,2- ion as compared to peaks at 694 and b~rate.~ ammonium chloride showed it to be tetramethyl- 664 cm.-l for the BCl,-ion at 778 and 667 cm.-l in ammonium hexachlorodiborate (Found C 25.2; C2CIG and at 778 and 760 cm.-l in CCl,.H 6.6; C1 57.1; N 7-2. C8H,4B,ClGN2 requires C (Received March 23rd 1961.> Waddington and Klanberg J. 1960 2329. Waddington and White Proc. Chem. SOC.,1960 85 315. Waddington and Klanberg unpublished observation. Complexes involving an Olefhk Tertiary Arsine Chelate Group By H. W. KOUWENHOVEN J. LEWIS,and R. S. NYHOLM RAMSAY LABORATORIES COLLEGE W.C. 1) (WILLIAM AND RALPHFORSTER UNIVERSITY LONDON WE report the preparation and properties of a new are known the best example being o-phenylenebis- type of chelate group one donor centre of which is a dimethy1arsine.l h common are those which tertiary arsenic atom and the other a C=C double contain two double bonds suitably oriented for co-bond.Chelate groups with two donor arsenic atoms ordination; Chatt and Venanzi2 have shown how-l Chatt and Mann J. 1939 610. * Chatt and VenanZi Nature 1956 177 852. JUNE1961 ever that cyclo-octa-1 ,5-diene behaves in this way towards univalent rhodium both double bonds being used for co-ordination. In the expectation that co-ordination of a double bond to a metal atom might be enhanced by attach- ing the ligand by a second donor centre in its mole- cule we have prepared and studied some compounds of this type especially dimethylpent-4-enylarsine AsMe,. [CH,],-CH :CH (denoted pas below).This was prepared starting from tetrahydro-2-hydroxy- methyl furan ; the derived chloromethyI compound served here. As further proof if one treats the com- plex PtBrgas with diphenylmethylarsine the latter displaces the double bond forming the compound ~Cl,-pa~.Ph~AsMe]~* in which the C =C stretch-ing frequency is restored to 1640 cm.-l as expected. This is shown in the annexed scheme. The pure arsine reacts readily with bromine or methyl iodide with formation of C,Hl,AsBr2 or [C,H ,As Me]I respectively these reactions leaving the C =C stretching frequency virtually unaffected. In marked contrast with platinmi salts mercuric Properties of dimethylpent-4-enylassine (pas) complexes. Compound“ Colour M.p. p(C=C)(cm.-l) pas (free arsine) Colourless Liquid 1637 PtClgas White 196OC 1506 PtBrgas Pale yellow 196c 1504 Pt1,pas Yellow 1 37c 1506 PtBr,.pas.Ph ,As Me Yellow 132-1 34 1640 (HgBr,pas) White 95 1645 (Hgbpas)2 Pale yellow Green 93 1639 C,H9*A~Me2Br White 69 1639 (C,H,.AsMe3)I White 1 53c 1640 a Analyses for C H halogen and metal have been obtained for each of these compounds.Nuiol or hexachlorobutadienemull. Decomp. Decomp. at 75”. with sodium and then water furnished pent-4-en- 1-01. The latter was converted into the 1-bromo-derivative the Grignard derivative from which with iododimethylarsine gave the required arsine a colourless liquid b.p. 154”/760 mm. which is oxidised rapidly in air. This ligand reacts readily with the tetrachloro- platinate(@ ion in acidified aqueous alcohol in an atmosphere of carbon dioxide.A yellow precipitate is formed immediately but soon dissolves and the required complex PtCl,pas is precipitated after about 10 hours. The corresponding bromide and iodide were prepared by adding the appropriate lithium salt to the original solution. The properties of the complexes are shown in the Table. The corn- pounds are monomeric and this indicates that the ligand behaves as a chelate group completing the four-fold co-ordination of the platinum(~~) atom. The infrared data support this view. A band shown by the free ligand at 1637 cm.-l which is characteristic of a C=C double bond is shifted to about 1505 cm.-l in the co-ordinated complex; it has been stated by others3 that co-ordination of a double bond to platinum@) results in a decrease of the C=C stretching frequency by about 140 cm.-l as ob- salts give dimers [HgBr,pas],; the infrared spectra of these (Table) show that the C=C bond is not co-ordinated to the metal.These products are formulated as halogen-bridged dimers of tetrahedral mercury(@. The authors are indebted to the Koninklyke/Shell Laboratory in Amsterdam for leave for one of them (H.W.K.) to work on this project. (Received March 29th 1961.) * In the diagram this compound is shown as having a cis-configuration,but we have no information on this point at present. a See Cotton in “Modern Co-ordination Chemistry,” ed.J. Lewis and R. G. Willrins Intersc ience Publ. Inc. London, 1960 p.301. PROCEEDINGS Entropies and Heat Capacities of Activation in Solvolysis by S Mechanisms By G. R. Cow J. R. Fox M. J. H. FITCHES, K. A. HOOTON,D. M. HUNT G. KOHNSTAM, and B. SHILLAKER (UNIVERSITY LABORATORIES, SCIENCE SOUTHROAD DURHAM) THEsolvolysis of organic chlorides and toluene-p- ture or from simultaneous SNl and SN2solvolysisin sulphonates shows negative heat capacities of roughly equal proportions. activation (aC$ = dE/dT -R where E is the The present interpretation of the values of activation energy)? Bensley and Kohnstam2 sug- aCS/ASS is based on the assumption that the magni- gested that the ratio AC$/AS$ (where AS$ is the tude ofaC$ andnSS in S,l solvolysis is mainly if entropy of activation) should be independent of the not entirely controlled by the increase in solvation nature of the substrate in S,l solvolysis and that the on passage into the more polar activated complex ratio should have a lower value for solvolysis by while the additional negative contribution to these mechanism SN2 under the same experimental con- parameters which arises from the partial covalent ditions.Little relevant information was available at attachment of a solvent molecule in the transition TABLE1. ACS/AS for reaction with aqueous acetone at 50". (Mean values of dC$/ASS are reported for each series the errors are the standard errors of the final mean. All substituents are in the 4-position). Acetone (%) Mechanism Substrates A c::/As: 50 BunBr PrnBr Ph.CH@r 1.20 5 0.16 SN2 { X-C6H4.CH2Cl(x = H,* NO2) 0.90 0.11 50 SN1 PhCCI, BuC1 XCGH,CHCl (X = H Me) NO,C,H,.CHPhClt 2.89 0.17 70 SN2 Ph.CH,Cl PhCH2Br 0.95 & 0-09 70 ButBr NO,C,H,.CHPhBr 3.84 & 0.24 SN1 rButC1 X-C6H4*CHPhC1 (X = H,$ Cl,$ Br, I, NO,) 3.62 & 0.13 ButBr 2.80 5 0-37 sN1 ButC1 Ph,CHCl 2.81 & 0-27 85 SNl XC6H4CHPhCl(X = H Ph But Me) NO2~C,H4CHC1C,H4.OPh 2-77 5 0-14 * Bensley and Kohnstam.a t Brittain Kohnstam Queen and Shillaker.' # Kohn~tam.~ the time but the measurements on reactions of known state of SN2solvolysis is much larger forAS1 than mechanism which are now reported in Table 1 con-for Act (@C andaS$ are both negative in the firmthe validity of these predictions for the solvolysis present reactions).2 The results are consistent with of a variety of chlorides and bromides in aqueous this view which is also supported by the fact that acetone.The value ofaC$/@S$ therefore appears to aC$/aSlfor S,l solvolysis in the present solvents afford an additional method for recognising sol- has the greatest value in 70% acetone (see Table I) volytic mechanism which may be useful in the neigh- the solvent in which water has the greatest partial bourhood of the mechanistic "border-line region" molar heat capacity and the lowest partial molar where the classical tests do not always yield unam- entropy-this is readily deduced from thermal4 and biguous answers (cf. ref. 2). vapour-pressure data5 for acetone-water mixtures. The application of this test to the reactions of A study of activation parameters for solvolysis in 4-substituted benzyl chlorides with aqueous acetone water has led Robertson and his co-workers to con- (see Table 2) shows that only the most powerful clude6 that the structure of the substrate affects AC$ electron-donating substituents (MeO.and PhO.) andnS+ in different ways contrary to the present yield completely S,l solvolysis. The abnormally low solvation hypothesis but it may well be that the value ofnC$/nS$ for the hydrolysis of the 4-methyl factors controlling the magnitude of these para- derivative in 50% acetone could arise either from a meters are not the same in water as in aqueous highly temperature-dependent transition-state struc- acetone where one of the solvent components can Brittain Kohnstam Queen and Shillaker J. 1961 2045 and references there cited.Bensley and Kohnstam J. 1957,4747. Kohnstam .I.. 1960 2066. Kister and Waldman J. Phys. Client. 1958 62 245. Ewert Thesis Brussels 1936. Robertson Heppolette and Scott Cunad,J. Chem. 1959 37 803; Robertson Suurnen Ken?.,1960 33 A 44. JUNE 1961 223 TABLE 2. A C$/dS$for the reaction of p-X.C6H,-CH2Clwith aqueous acetone at 50’. Acetone (%) S,l value* X = NO2 H Me Ant PhO Me0 50 2-89 0.84 70 3-68 - 80 2-81 - * From Table 1. t An = p-MeOCBH4-. “solvate” an organic solute. Further discussion of this point is deferred. Added in Proof Robertson and Scott have recently criticised mechanistic interpretations based on values of nCi/ASS on the grounds that AS$for solvo-lysis in water passes through zero and becomes pos- itive as the temperature is red~ced.~ Such a change Robertson and Scott J.1961 1596. 0.90 0.34 --0.91 1.19 2.31 3.60 3-83 --2.55 is claimed for two compounds within the experi- mentally accessible temperature range but in both cases these authors’ method of calculating LSf.leads to standard errors which are larger than the positive values of ASSat 0’. Further discussion of any differ- ences beetween solvolysis in water and in mixed sol- vents is deferred. (Received April 20th 1961 .) Alkaloids of Calabash Curare The Structure of Macusine-A By A. T. MCPHAIL,J. MONTEATH G. A. SIM ROBERTSON (CHEMISTRY DEPARTMENT w.2), THEUNIVERSITY GLASGOW A. R. BATTERSBY and D. A. YEOWELL H. F. HODSON (CHEMISTRY THE UNIVERSITY DEPARTMENT BRISTOL) THEstructure of macusine-A Cz2H2,N203+ a new quaternary alkaloid isolated in small quantity from Strychnos toxifera,l has been shown to be (I)by the X-ray crystallographic work of the Glasgow group.Chemical studies at Bristol support this constitution. Thus the ultraviolet and infrared spectra of macusine- A showed the presence of a 2,3-disubstituted indole system and ester hydroxyl and imino-groups; the molecular formula of the alkaloid when considered with the acylation study reported below allows there to be only one of each of these groups. Acetylation of macusine-A gave a product having the ultraviolet spectrum of an N-acylindole; the imino-group in the alkaloid is therefore the indolic >NH and since the second nitrogen atom is quaternary both are accounted for.Modified Kuhn-Roth oxidation2 of macusine-A yielded only acetic acid showing that the alkaloid does not contain a C-R group where R is ethyl or a higher homologue. Selenium dehydro- genation afforded a base recognised spectrally as an a-pyridylindole (cf. 11); the relationship of the in- dolic system to N(b) is thus shown. The picrate of the dehydrogenation base corresponds with that of alstyrine (11) but the quantity available precluded complete identification. Thermal fission of macusine- A chloride produced a tertiary base m.p. 240-241 O (picrate m.p. 239-241 ”) which was unaffected by HO.H,C ,CO,Me H ,CH,.OH palladium in boiling aqueous maleic acid;3 this evidence points to a structure for macusine-A in which it is sterically impossible for dehydrogenation to occur [cf.normacusine-B4 (tertiary base corres- ponding to III)]. The foregoing results are fully consistent with the structure (I) for macusine-A but alone do not establish it. The X-ray study by the Glasgow group was carried out with macusine-A iodide the crystals of which belong to the orthorhombic system with cell dimen- sions a = 13-82,b = 9.06 c = 17.43 A. The space group is P2,2,21 and there are four molecules of C22H,7N,0,1 in the unit cell. In all 1145independent Battersby Binks Hodson and Yeowell J. 1960 1848. Garbers Schmid and Karrer Helv. Chim. Acta 1954 37 1336. Wenkert and Roychaudhuri J. Amer. Chem. Soc. 1958 80 1613. Battersby and Yeowell Proc.Chem. SOC.,1961 17. structure amplitudes were obtained inclination Weissenberg photographs intensity estimates. from equi-and visual The third three-dimensional electron-density distribu- tion over one molecule of macusine-A iodide shown by means of superimposed contour sections parallel to (010). The co-ordinates of the iodide ion were obtained from Patterson syntheses. Successive three-dimen- sional Fourier syntheses with increasing numbers of PROCEEDINGS atoms included in the phasing calculations as they became clearly defined on the maps then served to locate the remaining atoms other than hydrogen in the asymmetric crystal unit. Independently of the chemical studies the results define the constitution and stereochemistry (apart from absolute configura- tion) of macusine-A iodide to be as in (I); the absolute stereochemistry shown is the more probable one.The average discrepancy between observed and calculated structure amplitudes at the present stage is 17% and refinement is proceeding. Superimposed contour sections illustrating the third three-dimensional electron-density distribution over one molecule are shown in the Figure. The structure of macusine-A has features similar to e~hitamine,~ and the stereochemistry of the ethyl- idene system is the same in both alkaloids. Experi- ments are in progress to determine the absolute configuration of macusine-A by correlation with macusine-B4(III). The calculations were carried out on the GIasgow University DEUCE computer with programmes devised by Dr.J. S. Rollett and Dr. J. G. Sime and we are indebted to the Director of the Computing Laboratory Dr. D. C. Gilles and his staff for facilities. We also thank the Medical Research Council and the Tropical Products Institute for financial support. (Received ApriZ loth 1961.) Hamilton Hamor Robertson and Sim Proc. Chern. SOC.,1961 63; Birch Hodson Moore and Smith Proc. Chem. SOC.,1961,62; Conroy Bernasconi Brook Ikan Kurtz and Robinson Tetrahedron Letters 1960 No. 6 1; and refs. cited in these papers. NEWS AND ANNOUNCEMENTS Chemical Society Liaison Officers.-The following additional Fellows have agreed to act as Liaison Officers U.K.A.E.A. Dounreay Experimental Reactor Establishment Mr.P. Lees Reactor Group Risley Dr. H. C. Dunn Unilever Limited Sharnbrook Mr. D. Welti Library.-From July 17th until September 30th 1961 the Library will close at 5 p.m. instead of at 7.30 p.m. It will not be open on August 7th and 8th. Corday-Morgan Medal and Prize for 1960.-This award consisting of a silver medal and a monetary prize of 400 guineas is made annually to the chemist of either sex and of British nationality who in the judgment of the Council of The Chemical Society has published during the year in question the most meritorious contribution to experimental chemistry and who has not at the date of publication attained the age of thirty-six years. Copies of the rules governing the award may be obtained from the General Secretary of the Society.Applications or recommendations in respect of the award for the year 1960 must be received not later than December 31st 1961 and applications for the award for 1961 are due before the end of 1962. MeIdola Medal Award 1960.The Meldola Medal for 1960 has been awarded to John Newton BradZey for his work in the field of physical chem- istry with special reference to the kinetics of re-actions involving free radicals and reactions in shock waves studied by mass spectrometry. The Meldola Medal which is the gift of the Society of Maccabaeans is awarded each year to the chemist JUNE 1961 who being a British subject and under 30 years of age at December 31st in that year shows the most promise as indicated by his or her published work.Awards are made by the Council of the Royal Institute of Chemistry with the concurrence of the Society of Maccabaeans on the recommendation of a specially appointed advisory committee. Dexter Award in the History of Chemistry.-Professor J. R. Partington has been chosen as the recipient of the Dexter Award of the American Chemical Society for his work in the history of chem- istry. The Award consists of a money prize and a plaque and it is anticipated that it will be made at a meeting of the Chemical Society in London in the Autumn by Dr. Edelstein President of the Dexter Chemical Corporation New York. “Chemistry in the Service of Medicine.”-The Faculty of the History of Medicine and Pharmacy of the Society of Apothecaries of London has chosen “Chemistry in the Service of Medicine” as the theme of its second annual congress which is to be held in London on September 28th and 29th 1961.The Chemical Society is co-operating in this meeting and the last of the four sessions will be held in the Chemical Society’s rooms in Burlington House. Application forms for membership of the Congress and copies of the programme may be obtained from the Honorary Secretary Dr. F. N. L. Poynter The Wellcome Historical Medical Library Euston Road London N.W.l. Symposium on Carbohydrate Chemistry.-An International Symposium on Carbohydrate Chem- istry sponsored by the Chemical Society in associa- tion with the University of Birmingham will be held in Birmingham during the period July 10-20th 1962.Further particulars will be published by the Society in due course. Copies of the announcement will be sent when available to those who apply to the General Secretary The Chemical Society Burlington House London W. 1. Papers read at this meeting will not be published in full in collected form. International Congresses etc.-The Third Inter- national Congress on Catalysis is to be held in Amsterdam from July 20-25th 1961. Enquiries should be addressed to Dr. D. M. Brouwer P.O. Box 3003 Badhuisweg 3 Amsterdam-N Nether- lands. The Fourth European Peptide Symposium will be held in Moscow on August 18th-21st 1961. En- quiries should be addressed to Dr. W. K. Antonow Institute of Chemistry for Natural Products 1 Akademichesky Proesd.18 Moscow B-134 U.S.S.R. An International Conference on Chemical Physics of Non-Metallic Crystals will be held in Evanston Illinois on August 28th-31st 1961. Enquiries should be addressed to Mr. 0. C. Simpson Con- ference Secretary Argonne National Laboratory 9700 S. Cass Avenue Argonne Illinois U.S.A. A Symposium on Oxidation Processes in Chemical Manufacture will be held in London on September 28-29th 1961. Enquiries should be addressed to the Honorary Secretary The Society of Chemical Industry 14 Belgrave Square London S.W. 1. The 33rd International Congress of Industrial Chemistry will be held in Bordeaux on October 1st-Sth 1961. Enquiries should be addressed to the Soci6tb de Chimie Industrielle 28 rue Saint-Dominique Paris 7e France.The First European Symposium on Food Tech- nology 34th event of European Federation of Chemical Engineering is to be held in Frankfurt on October 26th and 27th 1961. Enquiries should be addressed to DECHEMA Deutsche Gesellschaft fur chemisches Apparatewesen Rheingau-Allee 25 Frankfurt (Main). An International Symposium on “Reversible Photochemical Processes” will be held at Duke University Durham North Carolina on April 16-18th 1962 under the sponsorship of the U.S. Army Research Office (Durham). Participation by invitation only will be limited to scientists who have been active in this area of research. Enquiries should be addressed to Dr. George M. Wyman Director Chemistry Division U.S.Army Research Office (Durham) Box CM Duke Station Durham North Carolina. The Fourth Rubber Technology Conference will be held in London on May 22nd-25th 1962. Enquiries should be addressed to the Institution of the Rubber Industry 4 Kensington Palace Gardens London W.8. The Third Congress of the European Federation of Chemical Engineering will be held in London on June 20-26th and in Paris on June 27th-July 3rd 1962. The London sessions of the congress will be held in conjunction with the Chemical and Petroleum Engineering Exhibition. Enquiries should be ad- dressed to Mr. J. B. Brennan London Office of the Federation c/o Institution of Chemical Engineers 16 Belgrave Square London S.W. 1. Election of New Fellows.-1 18 Candidates whose names were published in Proceedings for April have been elected to the Fellowship.Deaths.-We regret to announce the deaths of the following Mr. Charles L@shaw (16.12.60) of St. Annes-on-Sea and Professor W.G. T. Qvist (8.2.61) of Ah Akademi Finland. Personal.-Dr. J. A. Ballantine has been appointed Lecturer in Chemistry at University College of Swansea. Mr. R. R. Butler formerly Principal of Liverpool College of Technology has been appointed Director of Studies of the British Institute of Management Residential Course of Studies on “Management Practice” at Wadham College Oxford in July next. Mr. W. G. Carey has retired recently as Chief Chemist and Bacteriologist to the Sunderland and South Shields Water Company.He is continuing as Public Analyst to various authorities at his New- castle upon Tyne Laboratories (J. & H. s.Pattinson). Professor J. C. Craig of the San Francisco Medical Centre has been awarded the degree of Doctor of Science by the University of Sydney for his work in Organic and Medical Chemistry. Dr. R. J. W. Cremlyn has been appointed Senior Lecturer in Organic Chemistry at Hatfield College of Technology from September 1st next. Professor G. Fodor of the Stereochemical Labora- tory of the Hungarian Academy of Sciences Budapest is spending three months in Cambridge as an Overseas Fellow of Churchill College. The British Industrial Biological Research Associa- tion has appointed Dr. Leon Golberg as its Director.Dr. Golberg is at present Medical Research Director of Benger Laboratories Ltd. It is expected that he will take up his appointment with the Association towards the end of the year. Dr. N. N. Greenwood who is at present Senior Lecturer at Nottingham has been appointed to the newly established Chair of Inorganic Chemistry in the Newcastle Division of the University of Durham from August lst 1961. Sir Cyril Hinshelwood is among ten new members appointed by the Pope to the Pontifical Academy of Sciences. Mr. R. A. Y. Jones has been appointed Assistant Lecturer in Chemistry at the University of Sheffield. PROCEEDINGS The Honorary Degree of Doctor of Science of London University has been conferred upon Mr. J. D. Kendall Director of the Minnesota Mining and Manufacturing Company.Mr. D. H. Mansfield has been appointed to the Board of General Tin Investments. Dr. P. J. Padley of Jesus College has been ap- pointed to a University Demonstratorship in the Department of Physical Chemistry Cambridge. Dr. A. 0.Pittet formerly of the Department of Chemistry Ohio State University Columbus is now with the Natural Gum Laboratory Stein Hall and Company New York. Mr. E. N. Pullom formerly Editor of the Pharma-ceutical JournaZ has taken up a new appointment as Pharmaceutical Editor with Intercontinental Market- ing Services Ltd. Dr. G. S. Rao formerly Assistant Professor of Chemistry University of Saugar India is now a Research Officer Chemistry Division Atomic Energy Establishment Trombay Bombay.The Directors of Laporte Industries Limited have announced that Dr. F. S. Spring has been appointed a Director of Howard and Sons Limited. Mr. B. Shelton has transferred from Queen’s University Ontario to the Department of Organic Chemistry University of New South Wales. Mr. G. V. Taylor Works Manager at the Newport factory of Monsanto Chemicals Limited retired on June 1st. Sir Alexander Todd will be admitted to the Honorary Degree of D.Sc. of University of Wales at a Congregation of the University to be held in Cardiff on July 21st. Dr. D. T. A. Townend C.B.E. Director-General of the British Coal Utilisation Research Association is to receive the Honorary Degree of D.Sc.(Tech.) of the University of Sheffield on July 1st next.Dr. D. M. S. Wheeler has been appointed Assistant Professor of Chemistry at the University of Nebraska. FORTHCOMING SCIENTIFIC MEETING Nottingham Tuesday July 18th to Friday July 21st 1961. Symposium on Inorganic Polymers to be held at the University. Details have been circulated to Fellows. JUNE 1961 227 APPLICATIONS FOR FELLOWSHIP (Fellows wishing to lodge objections to the election of these candidates should communicate with the Honorary Secretaries within ten days of the publication of this issue of Proceedings. Such objections will be treated as confidential. The forms of application are available in the Rooms of the Society for inspection by Fellows.) Atkinson George Francis M.A. Ph.D. Department of Chemistry University of Waterloo Waterloo Ontario Canada.Bird Michael Charles John A.R.I.C. 16 Trysull Gardens Merry Hill Wolverhampton Staffs. Boivin Jean-L. B.A. Ph.D. 131 Avenue Rochette Quebec 10 Quebec Canada. Byrne Geoffrey Arthur. 6 Holland Park Drive Rhyl N. Wales. Campbell Michel Jules Malcolm BSc. 42 Wavertree Road London S.W.2. Christmas Susan B.Sc. 10 Harewood Road Bedford Beds. Clark Neville John B.Sc. 33 Park Street Parkville N.2 Victoria Australia. Clayton Colin John B.Sc. 3 Wingrove Road Newcastle upon Tyne 4. Cochrane Gordon Campbell. 7 East Main Street, Menstrie Clackmannanshire. Coley Noel George B.Sc. 70 The Broadway Dudley Worcs. Cotson Sydney Ph.D. A.R.I.C. 3 College Mansions 45 Sydenham Park Road London S.E.26.Dahl Thor B.A. 26 Waltersland Road Stillorgan Co. Dublin. Davies Robert Edward. 48 Crosslands Avenue Norwood Green Southall Middlesex. Drewer Robert James B.Sc. 10 Cranbrook Avenue Millswood Estate South Australia. Driver Ronald Michael. 500 Dunstable Road Luton Beds. Duffield Alan Malcolm B.Sc. Department of Chemistry, University of Western Australia Nedlands Western Australia. Ellestad George Alfred M.S. Department of Chemistry University of California at Los Angeles California U.S.A. Emsley John B.Sc. Department of Chemistry University of Manchester Manchester 13. Feairheller Stephen Henry B.S. 17 Raymond Avenue Somerville 44,Massachusetts U.S.A. Field Lamar Ph.D. Box 1507 Substation B Vanderbilt University Nashville 5 Tennessee U.S.A.Fielding Brian Crosbie Ph.D. Dawn Lea Field Lane Liverpool 10. Gordon Edmund B.Sc. 5 Blenheim Road High Barnet Herts. Gray John Stewart B.Sc. 14 Cemetery Road Pudsey Yorks. Griffey Paul Fernall B.Sc. A.R.I.C. 26 Federation Road London S.E.2. Guise Geoffrey Bruce. 3 Loftus Street Ashfield New South Wales Australia. Haas Alois Dr.rer.nat. 107 Chesterton Road Cam- bridge. Hall Peter Charles. 3 Brandon Road Esh Winning Co. Durham. Hall William Pettifer. 42 Etheldene Avenue London N.lO. Hanna William John Wesley B.Sc. Riverside Millvale Bessbrook Co. Armagh. Hester Ronald Ernest B.Sc. 1094 Harvard Place Ithaca New York,U.S.A. Hofmann Noel L. B.S. 3800 Woodford Road Cincin- nati 13 Ohio U.S.A. Honeywood Roy Ivan William.42 Hillcroft Crescent Oxhey Watford Herts. Howarth Michael B.Sc. 21 Gloucester Crescent London N.W.1. Howell Gilbert Leslie B.Sc. 172 Waghorne Street Ahuriri Napier New Zealand. Jeacocke Gilbert James Ph.D. 86 Monkswood Avenue Waltham-Abbey Essex. Johnson Nigel Peter B.Sc.6 Habgood Road Loughton Essex. Kaufman Harold B.Sc. 4909 Centre Avenue Pittsburgh 13 Pennsylvania USA. Keighley John Hewitt M.Sc. P.O.B. 287 7 Roy Camp Valcartier Quebec P.Q. Canada. Keldsen Glenn Lloyd B.Sc. Department of Chemistry, University of Kansas Lawrence Kansas U.S.A. Kirby John Alexander B.Sc. 45 Jenny Lane Woodford Cheshire. Koop Donald A. B.A. M.S. 14B South Apartments Orono Maine U.S.A. Lynch Michael Felix Ph.D. M.I.C.I.A.R.I.C. 78 Blinco Grove Cambridge. MacDonald Roderick M.Sc. D.Sc. Box 1427 Quebec City P.Q. Canada. McGinnety John Anderson. 5 Baroness Drive Newcastle upon Tyne 5. McKay Arthur Ferguson M.Sc. Ph.D. F.R.S.C. 215 Stonehenge Drive Beaconsfield W. Montreal 33 Canada. McLauchlan Eric John M.Sc. A.R.I.C. 98 Copthall Road West Ickenham Middlesex. McRae. John Alastair. B.Sc. 122 Summerhill Road Glen Iris Victoria Australia. Milwidskv. Benjamin Max. B.Sc. Ramat Shaul 4317 Haifa Israel. -. Obaid Rida M. S.,B.Sc. Department of Chemistry The University Edgbaston Birmingham 15. Olszewski Edward John M.S. 509 East John Street Champaign Illinois USA. Overbury Richard Douglas. 173 Boroughbridge Road York. Padget Geoffrey. The United Steel Company Ltd.Swinden Laboratories Moorgate Rothei ham Yorks. Peters Stanley T. B.S. 161 Calderon Avenue Apt. 62, Mountain View California U.S.A. Reggel Leslie Ph.D. U.S. Buxeau of Mines 4800Forbes Avenue Pittsburgh 13 Pennsylvania U.S.A. Robinson George Kenneth B.Sc. 5 Burnside Avenue Mossend Bellshill Lanarkshire. Robinson Ralph Nigel B.Sc. 33 Heaton Road Withing- ton Manchester. Ross Euan James Ferguson. 11 Garden Terrace, Stanley Co. Durham. Said Ahmad a1 Haj B.Sc. Department of Chemistry, The University Edgbaston Birmingham 15. Sartori Guido. Istituto di Chimica Generale Universita di Roma Roma Italy. Schaleger Larry Leroy B.A. Ph.D. Department of Chemistry Cornell University Ithaca New York U.S.A. Schut Robert N.A.B. Ph.D. Chemical Therapeutics Research Laboratory Miles Laboratories Inc. Elk- hart Indiana U.S.A. Shepherd George Robbins M.Sc. 1 154 College Avenue Palo Alto California U.S.A. Simlot Mahavir Mohan M.Sc. Department of Chem- istry Kansas State University Manhattan Kansas State U.S.A. Smith Leslie Harold. 16 Woodlands Avenue Cheadle Hulme Cheshire. Spencer-Young Russell. World’s End Great Kimble Rondebosch Cape South Africa. Stear Adrian Nigel B.Sc. University Chemical Labora- toIy Lensfield Road Cambridge. Stiles David Austin B.Sc. 12 Sherbourne Road Acocks Green Birmingham 27. Stripp Irene Marian B.Sc.2 Church Lane Hedge End nr. Southampton Hants. Taggart Herbert Andrew. 344 Oldpark Road Belfast 14. Teh Soon Hoe. 53 Nithdale Road London S.E.18.Thompson William Stanley. 4 Lias Cottages Porthcawl Glam. Sykes Alfred Geoffrey Ph.D. 9 Skipton Avenue, Fartown Huddersfield Yorks. Tremblay Meude D.Sc. 7325 Avenue Germain Charles- bourg 7 P.Q. Canada. Ward Brian Godfrey. 25 Glentworth Road Radford Nottingham. Wardle Ronald B.Sc. 25243 Hopkins Avenue Dearborn 9 Michigan U.S.A. Wilson Richard D. B.Sc. 7258 Pone Avenue Canoga Park California U.S.A. Wright Derek Ernest A.R.I.C. 12 Holland Road Withycombe Raleigh Exmouth Devon. Wuskell Joseph P. B.A. 634 East First Avenue Roselle New Jersey U.S.A. Wye Richard. Blackboys Uckfield E. Sussex. ADDITIONS TO THE LIBRARY A bibliography of the Honourable Robert Boyle. J. F. Fulton. 2nd edn. Pp. 217. Clarendon Press.Oxford. 1961. Chemical crystallography an introduction to optical and X-ray methods. C. W. Bunn. 2nd edn. Pp. 509. Clarendon Press. Oxford. 1961. (Presented by the publisher.) Selected studies in chemical kinetics. F. Daniels. Sponsored by the Pennsylvania State University. 35th Annual Priestley Lecture. Pp. 123. Pennsylvania Uni- versity. Pennsylvania. 1961. Cahiers de synthkse organique mkthodes et tableaux d’application. J. Mathieu and A. Allais. Edited by L. Velluz. Vol. 6. Pp. 417; Vol. 7. Pp. 309; and Vol. 8. Pp. 233. Masson. Paris. 1960-61. Die Praxis des organischen Chemikers. L. Gattermann. Pp.323.4th edn. Veit & Comp. Leipzig. 1898. (Presented by R.Lessing.) Formaldehyd. J. E. Orloff. Translated into German by C.Kietaibl. Pp. 327. Johann Ambrosius. Leipzig. 1909. (Presented by R. Lessing.) Die Technologie der Cyanverbindungen. Dr. W. Bertelsman. Pp. 332. R. Oldenbourg. Miinchen und Berlin. 1906. (Presented by R. Lessing.) Pyridine and its derivatives. Edited by E. Klingsberg. (The chemistry of heterocyclic compounds. Edited by Arnold Weissberger.) Part 2. Pp. 547. Interscience Publ. Inc. New York. 1961. Heterocyclic systems with bridgehead nitrogen atoms. W. L. Mosby. (The chemistry of heterocyclic compounds. Edited by A. Weissberger. Vol. 15. Part 1.) Pp. 747. Interscience Publ. Inc. New York. 1961. The enzymes. Edited by P. D. Boyer H. Lardy and K. Myrback. Vol. 5.2nd edn. Pp. 645. Academic Press. New York. 1961. Treatise on analytical chemistry.Edited by I. M. Kolthoff P. J. Elving and E. B. Sandell. Vol. 2. Part 1. Pp. 1308. Vol. 3. Part 2. Pp. 380. Interscience Publ. Inc. New York. 1961. Gas sampling and chemical analysis in combustion processes. G. Tine. Pp. 94. Pergamon Press. Oxford. 1961. (Presented by the publisher.) IP Standards for petroleum and its products. Part 1. Methods for analysis and testing. 20th edn. Pp. 793. Institute of Petroleum. London. 1961. Preservation of documents by lamination. W. K. Wilson and B. W. Forshee. (National Bureau of Standards Monograph No. 5.) Pp. 20. U.S. Government Printing Office. Washington. 1959. Technical data on fuel. Edited by H. M. Spiers. 6th edn. Pp. 360. British National Committee World Power Conference. London. 1961.(Presented by €5. M. Spiers.) Computing methods and the phase problem in X-ray crystal analysis report of a conference held at Glasgow 1960. Edited by R. Pepinsky J. M. Robertson and J. C. Speakman. Pp. 526. Pergamon Press. Oxford. 1961. (Presented by Prof. J. M. Robertson and Dr. J. C. Speakman.) Comparative effects of radiation. Edited by M. Burton J. S. Kirby-Smith and J. L. Magee. Report of a con-ference held in San Juan at the University of Puerto Rico 1960 and sponsored by the National Academy of Sciences National Research Council. Pp. 426. John Wiley and Sons. New York. 1960. (Presented by the publisher.) Gibberellins a symposium presented before the Division of Agricultural and Food Chemistry at the 138th National Meeting of the A.C.S.New York 1960. (Advances in Chemistry Series No. 28). Pp. 167. A.C.S. Washington. 1961. Continuous culture of micro-organisms:a symposium organised by the Microbiology Group of the Society of Chemical Industry held in London 1960. (S.C.I. Mono- graph No. 12.) Pp. 272. S.C.I. London. 1961. (Presented by the publisher.) NEW JOURNALS Postepy Biochemii from 1961. 7. Zhurnal Strukturnoi Khimii from 1961,2.
ISSN:0369-8718
DOI:10.1039/PS9610000185
出版商:RSC
年代:1961
数据来源: RSC
|
|