|
1. |
Proceedings of the Chemical Society. November 1961 |
|
Proceedings of the Chemical Society ,
Volume 1,
Issue November,
1961,
Page 397-440
Preview
|
PDF (4619KB)
|
|
摘要:
PROCEEDINGS OF THE CHEMICAL SOCIETY NOVEMBER 1961 WILLIAM WARDLAW 1892-1958 WARDLAW WILLIAM was born at Newcastle-on-Tyne on March 29th 1892 the elder son of William and Margaret Wardlaw and was educated in Newcastle first at Rutherford College (1904-10) and then at Armstrong College (now King’s College) University of Durham (1 9 10-1 5). Although Wardlaw left Newcastle in 191 5 and lived for the rest of his life south of the Trent he always remained a loyal Northumbrian and had a strong attachment both to his old school and to the University of Durham; he was a lifelong member of the Old Rutherfordians’ Association and its President at the time of his death. He showed his aptitude for inorganic chemistry early winning the Freire-Marecco Prize and Medal in that subject in 1913.But for the war he would probably have gone to Germany for postgraduate research; as it was he volunteered for military service but was transferred to the Army Reserve and em- ployed as a chemist by the Ministry of Munitions. In almost every gas-works in the country at that time coal gas was washed with coal tar to remove toluene which was needed for the manufacture of explosives and the necessary analytical control was exercised by laboratories in various University centres including Newcastle where Wardlaw was put in charge of the coal-tar testing (which was actually done in a laboratory in the Medical School). In 1915 Wardlaw applied for a post as Assistant Lecturer and Demonstrator in the Chemistry Department of the University of Birmingham [which was vacant through the appointment of C.K. Tinkler to a Readership at King’s College of House hold Science (now Queen Elizabeth College)) and was appointed at a stipend of El50 p.a. He started work on November 3rd 1915 and remained in Birmingham for 22 years being promoted to Lecturer in 1921 and Senior Lecturer (then called Lecturer Grade I) in 1929. The work which the head of the department Prof. P. F. Frankland allocated to him on appointment was to help Dr. A. Parker (later Director of Water Pollution Research) in the running of the first-year inorganic practical class which was conducted in the Chemistry Department of the Municipal Technical College in Suffolk Street the new University buildings at Edgbaston having been taken over as a military hospital on the out- break of war.In addition to his teaching Wardlaw carried on with work for the Ministry of Munitions but the tar testing being in the hands of S.R. Carter Wardlaw and Parker became responsible under Frankland for various tests on TNT which was manufactured by Chance and Hunt at Oldbury. Towards the end of the war largely through the influence of Sir John Cadman then Professor of Mining and Petroleum Technology in the University he was also involved in some work on petroleum. All this meant that Wardlaw was considerably over- worked and perhaps as a result of overwork inade- quate food and exposure to bad weather during his frequent journeys to explosives factories he con- tracted tuberculosis.Fortunately he had a good con- stitution and after a six-month stay in a sanatorium he had recovered completely but for some years afterwards he lived on a farm at Packington in Warwickshire and travelled in to Birmingham daily partly by bicycle and partly by train. It was during his stay in the sanatorium that Wardlaw developed the love of gardening which provided him with 39 7 relaxation through all his later life. It was charac- teristic of him that though he was genuinely devoted to and knowledgeable about gardening (and his roses in particular) he kept the subject in perspective and was quick to see the ridiculous in the attitudes of the more solemn and pretentious type of amateur gardener; I remember for example how one day on our way home from Leamington where we had been walking round the Jephson Gardens admiring the flowers he invented a tale of a wonderful new variety of rose with which he was going to excite the envy of one of our colleagues whose roses tended to be bigger and better than any others.But his leg-pulling was always entirely free from malice and usually although he could on occasion maintain a poker-face “cheerfulness was always breaking in” and his twinkling eyes would give the game away. Wardlaw’s first paper published in 1914 with J. A. Smythe was on the oxidising properties of sulphur dioxide as manifested in its action on the lower chlorides of Sn Ti Hg and Fe and as soon as he was free from war work he returned to and extended this topic on which he published about half-a-dozen papers in the Society’s Journal in the early twenties and obtained the D.Sc.degree of the University of Durham (1 920). Wardlaw was always a true inorganic chemist and although no one had a better apprecia- tion of the contributions which physical methods could make to inorganic chemistry he preferred to leave their exploitation to others while he sought for new ideas by the application of the techniques of inorganic preparation and analysis of which he was a master; his note-books were full of details of in- numerable preparations and analyses which he had done personally all written in his clear and attrac- tive hand and it gave considerable aesthetic pleasure to a young scientist to see him at the bench trying out his latest idea for the preparation of an interest- ing new inorganic compound.In the case of the SO work when Wardlaw had explored the inorganic chemistry and his investigations began to throw up interesting physicochemical problems he was con- tent to leave the further development to his colleague S. R. Carter and turned himself to a new topic the chemistry of molybdenum on which he published seventeen papers in the Journal in the period 1924-31 making a substantial addition to our knowledge of the subject. Wardlaw always took great pleasure in the colours of inorganic compounds and was curious to understand the stereochemical and other differences between series of salts of different colours.This interest too provided a link between his research and his teaching of qualitative analysis (on which with F. W. Pinkard he pub- lished a book in 1928). The choice of some of his investigations was probably decided by this interest in colour-for example his study with R. G. James PROCEEDINGS (J. 1928 2726) of the complex thiocyanates which are formed in a well-known colour test for molybdenum. Wardlaw served under P. F. Frankland for only four years but the two had a great regard for each other and remained friends until Frankland’s death However scientifically Wardlaw was much more influenced by G. T.Morgan Frankland’s successor even though he only stayed in Birmingham for five years.Morgan was a man of very wide interests who for example made substantial contributions to the chemistry of dyestuffs (a subject on which he regularly lectured to undergraduates for two hours at a time) but he is perhaps chiefly remembered for his impact on inorganic chemistry and particularly for his recognition of the high importance of organic chelate groups for the study of inorganic stereo- chemistry. He also recognised the importance of the new physical methods for studying the structure of complexes W. T. Astbury’s early X-ray work on acetylacetonates of tervalent metals and on basic beryllium acetate having been done at his instigation. Morgan’s influence is discernible in the turning of Wardlaw’s interest while he was still working on molybdenum to complex compounds and problems of isomerism but apart from its immediate effect on his researches there is no doubt that association with Morgan helped to prepare Wardlaw’s mind for the stereochemical investigations which were to follow.(He was certainly also much influenced during this period by Sidgwick’s writings and no doubt by the work of his colleague J. D. Main Smith.) Wardlaw was very quick to see that the advent of wave-mechanics heralded a renaissance in inorganic chem- istry and that its applications as they were develop- ing in the hands of such men as Pauling on the one hand gave extremely useful indications as to what experimental research was likely to be significant and on the other hand required the guidance of new experimental evidence for their further development.He quickly made in his investigations of the co- ordination compounds of a number of metals a major contribution to inorganic stereochemistry a subject which remained his principal scientific interest for the rest of his life. Though the resources available to him were meagre indeed by present-day standards through his own researches and those of his col- leagues and pupils whom he encouraged so un-selfishly Wardlaw probably did more than anyone except N. V. Sidgwick in the pre-war decade to raise British inorganic chemistry towards the high position which it now enjoys. Wardlaw’s research was always done in a relatively small proportion of his time. No sooner had he cast off the burden of war work than he had to take a major part in dealing with the big influx of ex-service students and soon after W.N. Haworth was ap- NOVEMBER 1961 pointed Mason Professor in 1925 Wardlaw became responsible for all the inorganic work in the depart- ment. He was a magnificent lecturer still remembered by many who attended his classes as undergraduates. He gave the first-year lectures himself in these days of problems posed by large classes it is perhaps worthy of record that (if memory serves) the first- year class in the middle thirties was about 180 strong containing budding medicals engineers etc. as well as future scientists. Undergraduates between the wars were not as docile as they are today and I remember that strong official intervention was re- quired at one time to quell riots which had been occurring in the first-year lectures in physics botany and mathematics.The same students largely attended Wardlaw’s lectures where they behaved like angels. His material was of course very well prepared and equally well presented but the respect and admiration which students gave him was evoked by something more than outstanding competence; they recognised a man of great character and per- sonality. Undergraduates might writhe under his sarcasm or shrink from the rare blaze of anger which dishonesty or slackness could spark off but they would never doubt his justness nor fail to perceive his essential kindness. In 1937 the Chair of Physical Chemistry at Birkbeck College fell vacant through Sugden’s trans- lation to University College and Wardlaw was appointed.One of his old friends knowing his devo- tion to inorganic chemistry said to him “Some are born to physical chemistry some achieve it and others have it thrust upon them”; I remember that having by chance to go to London on the same day that Wardlaw was called for interview I went with him to the Senate House and met one of the other candidates a physical chemist who clearly thought that Wardlaw’s candidature constituted unfair com- petition. But no one could accuse Wardlaw of narrowness of outlook and his appointment far from being to the disadvantage of physical chemistry was greatly to the benefit of chemistry as a whole More- over knowing with hindsight what qualities in addition to eminence in chemistry were to be re- quired in a Birkbeck professor during the war and after one finds it difficult to believe that the College could have made a better choice.Wardlaw found himself the joint head (with F. Barrow reader in organic chemistry) of a depart- ment in which quality and enthusiasm had perforce to compensate for spaciousness of accommodation or magnificence of equipment. Besides himself and Barrow the staff consisted of F. J. Thomeycroft D. J. Ives and later A. J. E. Welch. They had to teach a large number of day and evening students in very inadequate laboratories and with their post- graduate students do research as best they could in such small comers of the department as could be spared from teaching.Nevertheless Wardlaw soon settled down very happily in his new quarters which approached through the legal atmosphere of Chancery or Fetter Lane and with a blazing open fire to greet one on arrival gave the tea-timz visitor on a November day an irresistibly Dickensian im- pression He continued with some of his Birmingham researches but had scarcely had time to open up new lines of investigation when the war came. During the early part of the war he was responsible for some work in the College for the Ministry of Supply but because of his other war-time commitments and his subsequent preoccupation with the problems of re-building the ColIege on its present site it was not untiI about ten years after his appointment that he was able to engage in active research work on new problems.He then took up the study of metal alkoxides and related compounds and in the years before his retirement published a considerable number of papers jointly with D. C. Bradley in this field. Wardlaw and Bradley studied a wide range of alkoxides starting with those of zirconium extend- ing them to the other Group IV metals and later to some Group V elements. They established methods for the preparation of many new compounds studied their physical properties especially their volatility and molecular complexity and speculated about their structures. As in his earlier work Wardlaw here displayed his knack of picking on a fertile topic one which is still being actively pursued and which in its structural aspects is giving rise to vigorous controversy.In 1940 Wardlaw accepted an appointmmt on the staff of the Central Register of the Ministry of Labour and National Service. The main function of the Register during the war was the allocation and distribution of scientists and technologists in accord-ance with the national interest; Wardlaw had charge of the sections of the Register which dealt with all scientists other than mathematicians and physicists. He also acted on behalf of the Ministry of Produc- tion as Joint Secretary of the Scientific Advisory Council of the War Cabinet but his energies were mostly devoted to the work of the Central Register where he achieved great success because of his wide knowledge of scientists and of industry his entire freedom from any touch of bureaucracy and above all because he enjoyed the complete trust both of his own profession and of the Ministry.When he re- turned to academic life in 1944 he was invited to accept a part-time appointment as Scientific Adviser to the Technical and Scientific Register a post which he held until his death. In 1949 he was appointed C.B.E. in recognition of his services to the Ministry. Wardlaw believed strongly that individual scientists should contribute to the well-being of their subject and their profession through membership of scientific societies and professional bodies. He him- self was a member from his very early years of the Chemical Society The Institute of Chemistry and the Society of Chemical Industry and later became a member of the Faraday Society.He served the Chemical Society first as local representative in Birmingham (1932-33 then as a member of Council (1934-37) and for the very difficult period 193948 as Secretary when in spite of heavy duties at the Ministry of Labour and responsibilities to Birkbeck College he devoted much time and trouble to the Society’s affairs. He served as Vice-president 1949-52 and finally as President 1954-56 during which time he was largely instrumental in bringing about the refurnishing of the Library and the renova- tion of the Meeting Room and other parts of the Society’s premises. Wardlaw’s record of service to the Royal lnstitute of Chemistry is very similar; he was a member of Council from 1929 to 1936 a Censor from 1948 to 1957 and President from 1957 until his death.He won many friends for both bodies by his integrity his charm of manner and his unfailing courtesy. Wardlaw was in great demand as an examiner and had served as external examiner to more than half the Universities in the country as well as those of Alexandria and Cairo. He was examiner for the Associateship of the Institute of Chemistry for 15 years and an Assessor for National Certificates in Chemistry for 17 years; in his earlier days he did a good deal of examining for the Northern Universities Joint Matriculation Board. He was interested in many aspects of science and education and gave voluntary service to committees governing bodies and so forth too numerous to specify individually but his long connection with the British Association deserves special mention; he was for many years secretary to Section B,becoming its President for the Belfast meeting in 1952 and from 1955 until his PROCEEDINGS death he was one of the General Secretaries of the Association.Wardlaw’s first wife died at an early age and he married secondly in 1932 Doris Whitfield who had been one of his pupils and had herself taken a first class honours degree in chemistry. He was devoted to his wife and to his only child Margaret and in Iater years took great delight in his grandson Jonathan. His last few days were cheered by the news of the birth of a second grandson Nicholas.Wardlaw did not attain the highest honours either as a scientist or as an administrator but he did more for British science and British scientists than most men of his generation. He was an effective man in everything he undertook because he understood him- self and knew of what he was and was not capable because he understood other human beings and could judge their qualities accurately and because he was not actuated by self-interest and was uni- versally trusted. Nearly everyone who met him liked him,and all who had the opportunity to know him better respected and trusted him; some came to love him as the staunchest of friends never failing in cheerfulness and wise advice. He was a happy man and a better companion for a day’s journey would be hard to find.He was kindly and understanding always ready to help those who were in trouble but he was not sentimental and knew too much about human nature to be easily deceived. He was not only a regular church-goer but a true Christian endued with real humility but rarely hesitant or uncertain because he saw clearly what he should do. His life was regulated very much in accord with a line of George Meredith’s found in one of his old diaries- “Our life is but a little holding lent to do a mighty labour.” Life to Wardlaw was certainly not a vale of tears but a field of battle for the Happy Warrior “Who comprehends his trust and to the same Keeps faithful with a singleness of aim.” E.G. Cox. CHRISTMAS COMPETITON THISyear’s competition was inspired by the appearance of a second edition by J. E. Gowan and T. S. Wheeler of the invaluable “Name Index of Organic Reactions” (Longmans London). We offer a prize (book token €2 25.) for brief definitions of the nature and scope of any three of the following (which escaped attention by Gowan and Wheeler) Eve’s method Picasso replication Adam’s reaction Miss Arden’s colour test Cain’s reagent Busoni transposition Borgia solution Methuselisation Entries must be received by the Editor not later than first post on December 29th. The author’s name and if desired a pseudonym for publication should be given. It is hoped to publish a report in Proceedings for January 1962. The Editor’s decision will be final.NOVEMBER 1961 401 PEDLER LECTURE* Some Problems in the Chemistry of the Gallotannins By R. D. HAWORTH (THEUNIVERSITY, SHEFFIELD) THEterm “tannin” wasintroduced by Sequin in 1796 to denote substances with a capacity to convert animal skins into leather. The vegetable tannins are plant extracts composed of complex mixtures but it is not possible to single out any one component of the extract and to regard it as the tannin of the plant concerned. The extracts have other applications such as ink ingredients astringents mordants in the dye industry colloidal sta bilkers for boiler-wa ter treat- ment and oil-well dressings but these are all secondary to the main use in the leather industry. The constituents of the vegetable tannins are fre- quently phenolic in type varying from simple phenols such as gallic acid and catechin with little if any tanning properties to macromolecular in- tractable types and it has been suggested (Whitel) that the major leathering action of the extracts is derived from components with molecular weights between 500 and 3,000.It is only since the introduc- tion of chromatographic methods that the com-plexity of tannin extracts has been fully realised and much of the earlier work requires repetition and re-investigation with the aid of recently developed techniques. In 1920 Perkin and Everest,2 and Fre~denberg,~ divided the vegetable tannins into (a) the non-hydrolysable or condensed tannins and (b) the hydrolysable tannins.The non-hydrolysable tannins including extracts of quebracho (wood of Schinopsis lorentzii or S. halansae),wattle (bark of Acacia mol-lissima) hemlock (bark of Tsuga canadensis) and many others of practical value contain little if any carbohydrate but are converted by acids into in- soluble amorphous products known as phlobaphens. The hydrolysable tannins on the other hand are esters which are converted by acids alkalis or enzymes into glucose or other polyhydric alcohols and into gallic acid or other phenolic acids such as m-digallic (I) ellagic (IT) and chebulic (111) acid which are structurally related to gallic acid. The hydrolysable tannins include extracts of Chinese tannin (galls produced by Aphis chinensis on leaves of Rhus semialata) Turkish tannin (galls produced by Cynips tinctoria on twigs of Quercus infectoria) sumach (leaves of Rhus coriaria R.fyphina and other species) tara (pods of Caesalpinia spinosa) myrobalans (fruit of Terminaha chebufa) valonea (acorn cups of Quercus aegilops) and algarobilla (pods of Caesalpinia brevifolia). A further subdivision of the hydrolysable tannins was suggested on the basis of the products of hydrolysis; tannins yielding phenolic acids in which gallic acid predominates are known as gallotannins and those giving ellagic or structurally related acids are classed as ellagitannins. On this basis the gall tannins sumach and tara are gallotannins and valonea and algarobilla are ellagitannins whilst myrobalans contain chebulinic acid (which ranks them as gallotannins) together with corilagin and chebulagic acid (which rank them as ellagitannins).The chemistry of the gallotannins which is the main concern of this Lecture formed the subject of the classical researches of Emil Fischer (I 908-1 8). As a result of analytical. methylation and hydrolysis experiments FischeI-‘ and Freuden berg5 concluded that the amorphous Chinese gallotannin was a mixture of isomeric and closely related compounds whose average composition corresponded to that of a P-penta-0-m-digalloylglucose (IV; R = m-digalloyl). A synthetic specimen prepared from penta-0-acetyl-m-digalloy1chloride and 6-glucose with subsequent hydrolysis at 0” by sodium hydroxide showed many similarities with the natural product a1 though differences in water solubility and optical rotatory values were reported.Karrer and his co-workers6 discovered that precipitation of Chinese gallotannin with aluminium hydroxide gave fractions * Delivered before The Chemical Society on January 12th. 1961 at the Royal Institution London W.1 on February 20th at Aberdeen on February 21st at Edinburgh and on February 27th at Swansea. ‘.Vhite “The Chemistry of the Vegetable Tannins,” Society of pther Trades’ Chemists Croydon 1956 p. 13. Perkin and Everest “The Natural Organic Colouring Matters. Longmans Green & Co.,London 1918. a Freudenberg “Die Chemie der naturlichen Gerbstoffe,” Verlag Chemie Berlin 1920 p. 107. Fischer Ber. 1913,46 3253; 1919,52 809. Freudenberg “Tannin Cellulose and Lignin,” Verlag Chemie Berlin 1933 1938.Karrer Salomon and Payer Helv. Chim.Acta 1923 6,17. differing significantly in optical rotatory values and suggested the presence of polygalloyl chains in the tannin and Freudenberg pointed out that the experi- mental evidence could be consistent with many arrangements varying from the /?-penta-O-m-digalloylglucose structure to the extreme 1,2,3,4-tetra-0-galloyl-6-O-m-hexagalloy Iglucose form 0. Similar analyses of Turkish gallotannin led Fischer and Freudenberg7 to suggest an average composition corresponding to /3-penta-O-galloyl-glucose (IV; R = galloyl) and a synthetic specimen of the latter prepared from tri-0-acetylgalloyl chloride and /3-glucose with subsequent hydrolysis by sodium hydroxide showed many similarities with the natural tannin.But Fischer and Freudenberg7 and later workers8 have drawn attention to the presence of significant quantities of ellagic acid (11) in Turkish gallotannin. With regard to sumach Lowe9 claimed identity with Chinese gallotannin but Karrer and his co- worked suggested a close resemblance with Turkish tannin. More recent work involving chromatography has thrown doubt upon these earlier conclusions. Thus AsquithlO showed that synthetic p-penta-0-galloyl- glucose prepared by Fischer’s method was impure and Whitel claimed that Turkish gallotannin con- tained little if any /3-penta-O-galloylglucose.White and his colleaguesl~ll also showed that Fischer’s Chinese tannin was impure and on the basis of new analytical methods suggested that the main gallo- tannin constituent was a polygalloylated trisac- charide containing 4-5 galloyl groups per hexose molecule.This idea of a polysaccharide core for the tannins was also suggested by Grassmann Stiefen- hofer and Endres12 for Sicilian and stagshorn sumach; in these cases a tetrasaccharide nucleus composed of arabinose rhamnose and two mole- cules of glucose was advocated. FKocEEDINas Obviously further work was required to resolve the conflicting claims and the following are some of the points which have been investigated by us during the last five years (1) the homogeneity of the gallo-tannins (2) the nature of the carbohydrate core (3) the presence of polygalloyl chains and (4) the extent of esterification of the carbohydrate core.FIG. 1. Chromatogram of Chinese gallotannin. All the chromatograms reproduced were obtained with the solvent systems (a) 6% acetic acid and (b) butan-2-ol-acetic acid-water (14;1 ;5). Key for Figs. 1-5 1 gallotannin; 2 gallic acid; 3 m-digallic acid; 4 trigallic acid (?); 5 shikimic acid; 6 quinic acid; 7 glucose. I i 1 +L b FIG.2. Chromatogram of extract of Turkish gallo- tannin. Chromatographyf3 showed that extracts of Chinese Turkish Sicilian and stagshorn sumach tannins (Figs. 1-4) were mixtures. After purification by extraction with ethyl acetate from solutions buffered to pH 6.8 by sodium phosphate or by chromatography on cellulose or perlon powder the Fischer and Freudenberg Ber.(a) 1912 45 915; (b) 1912 45 2709; (c) 1914 47 2485. Karrer Widmer and Staub Annalm 1923,433 288. 9 Lowe Z. anulyt. Cham. 1873 12 128. lo Asquith Nature 1951 168 738. l1 Kirby Knowles and White J. SOC.Leather Trades’ Chemists (a) 1951 35 338; (b) 1952 36 148. 12 Grassmann Stiefenhofer and Endres Chem. Ber. 1956 89 454. l3 Armitage Baylias Gramshaw Haslam Haworth Jones Rogers and Searle J. 1961 1842. NOVEMBER 1961 extracts of Chinese and sumach tannins yielded purified gallotannins as white amorphous hygro- scopic substances which were chromatographically homogeneous (Fig. 5) and gave the analytical results shown in Table 1. The combustions were particularly diacult to carry out because of the hygroscopic b FIG.3.Chromatogram of extract of Sicilian sumach. o7 06 +---+-b FIG.4. Chromatogram of stagshorn sumach. nature of the substances and high hydrogen values could not be avoided but the results suggest that Chinese gallotannin and Sicilian and stagshorn sumach tannins are indistinguishable and approxi- mate to octa- or nona-galloylated glucoses. Turkish gallotannin shows discrepancies in glucose content and specific rotation which may be due to impurities or to basic structural differences and further experi- ments on this tannin are in progress. On hydrolysis with acids alkalis or enzymes all tannins gave glucose and gallic acid; no polysaccharides were detected and .he idea of a polysaccharide core must +-b FIG.5.Chromatogram of purified gallotannin. I PQ Tube no FIG.6. Graded elution of crude tannase 1 Maltase; 2 invertase; 3A-D esterase; 4A-C p-glucosiduse. Tannin Chinese Sicilian sumach Stagshorn sumach Turkish NonagaIIoylgIucose TABLE 1. Analyses of some tannins. Liberated on hydrolysis C H Gallic acid (%) Glucose [a] in (%I (%) (a) (b) (%) COMe 53.2 53.1 4-0 4.0 102 102 99.4 99-8 12.1 12.1 12.1" 53.4 53-5 3.5 4.0 -98.2 12-4 12.2 12.4 12.2 53.3 4.0 -97.8 12.1 11.7 12.1 53.8 53-2 4.1 4-0 -97.5 16.4 16.5 23.2 27.7 53.5 3-0 -98-9 11.6 be abandoned as a result of experiments with tan- nase.13s14 This enzyme obtained by growth of the mould Aspergillus niger 106 on a gallotannin medium was absorbed on the basic resin Dowex 2 and separated by graded elution with a buffer of decreasing pH into 160 fractions (Fig.6) some of which had specific activities; e.g. a maltase an in-vertase two /?-glucosidases and an esterase fraction were isolated. The last fraction apart from the very PROCEEDINGS model depside-ester structures such as (VI-X) were synthesised along the lines shown in the chart for the ester (VI); the phenolic protecting groups were removed by hydrogenolysis. It was then incidentally discovered that many of these compounds were de- composed during attempted recrystallisation from rnethan01.l~ Thus ester (VI) gave methyl gallate (VII) gave methyl gallate and methyl benzoate (VIII) gave methyl gallate and methyl p-hydroxy- TABLE 2.Action of esterase fraction from tannase. (a) Esters. Attacked Me gallate 3,4dihydroxybenzoate and 3,5-dihydroxybenzoate; 3,6-di-O-gal!oylglucose;Chinese Turkish and sumach tannins. Not attacked Me u- m- and p-hydroxybenzoate 2,4-dihydroxybenzoateY and 2,5-dihydroxybenzoate. (b) Carbohydrates. Not at tacked Sucrose maltose cellobiose salicin trehalose melibiose turanose lactose quercitrin iso- maltose nigerose gentiobiose maltotetrose starch and inulin. small and insignificant residual invertase activity did not attack any of the numerous glucosides or poly- saccharides tested (Table 2) but it rapidly hydrolysed methyl gallate and galloylated glucoses. It rapidly attacked Chinese and Turkish gallotannin and the tannin from both Sicilian and stagshorn sumach yielding gallic acid and glucose only.Any polysac- charides would have been detected and it is concluded that these tannins are polygalloylated glucoses as suggested originally by Fischer and not galloylated polysaccharides as proposed by later workers. Dr. White15 and Dr. Endre@ have since withdrawn their claims concerning the polysaccharide cores of Chinese gallotannin and sumach tannin respectively. It was hoped that further fractionation of tannase might produce enzyme fractions capable of hydrolys- ing depside linkages without attacking ester linkages as such enzymes would reveal the presence of the polygalloyl chains proposed by Karrer and Freuden- berg. In order to test the fractionation a number of benzoate and (IX) gave methyl protocatechuate and methyl benzoate but (X)was unaffected.Tri-0- galloylglycerol (XI) and glucogallin (XII) were also unattacked and it appeared that an o-hydroxy-depsidic structure is required for the methanolysis. A mechanism involving a cyclic orthoester of type (XIII) as intermediate is tentatively suggested. Alter- native mechanisms are possible but the methano- lysis whatever its mechanism may be clearly achieves the selective hydrolysis for which the enzyme fractionation had been contemplated. The methanolysis has been applied to Chinese and sumach gal10tannins.l~ At first reaction with 90% methanol at pH 5-6 at 37" in absence of air for 48 hours was favoured but recently boiling with methanol at pH 5-6 for 24 hours has proved equally satisfactory and more convenient.Fig. 7 shows typical results. After 48 hours the main products were methyl gallate and a substance shown to be /%penta-O-galloylglucose(IV ; R = galloyl) x4 Haworth Jones and Rogers Proc. Chem. SOC.,1958 8; Haslam Haworth Jones and Rogers J. 1961 1829. l6 White and King Chem. and Ind. 1958 683. l6 Statement by Dr. Endres at Symposium of Plant Phenolics Group at Egham April 21st-22nd 1960. x7 Haslam Haworth Mills Rogers (in part) Armitage and Searle J. 1961 1836. NOVEMBER 1961 together with a trace of gallic acid; small amounts of other substances including P-2,3,4,6-tetra-O- galloylglucose (XIV; R = galloyl) were sometimes present and the origin of this artefact is discussed OH OH CH2.O.COQ0H OH 0 CH-OH OH OH -Me0,C I + + below.The /%pen ta-0-galloylglucose was readily isolated as a white amorphous powder but separa- tion of the /3-tetra-O-galloylglucose was more difficult as it was produced in small amounts only. However by using chromatography on cellulose columns the latter was also isolated as a white fi -GIucw + A-2,3,4,6-Tetra-o-gailoyl giucose OCH2Ph amorphous powder; it gave positive reactions in the silver nitrate and aniline hydrogen phthalate tests indicative of the presence of free aldehydic groups. These galloylated glucoses were synthesised from tri-0-benzylgalloyl chloride and S-D-glucose with subsequent hydrogenolysis of the benzyl groups as shown in the chart.In the synthesis of p-tetra-0-galloylglucose p-pentakis-0-(tri-0-benzylgalloy1)-glucose (IV ; R = tri-0-benzylgalloyl) was partially hydrolysed by passage down a column of acetic acid- washed alumina ; the crystalline p-tetra-O-(tri-O-benzyIgalloyl)glucose (XIV; R = tri-0-benzyl-... 0 i..:3 40 'r 0 +' ___r b FIG.7. Chromatogram of products of methanolysis of gallotannin after 48 hours. 1 p-Penta-0-galloyl-glucose; 2 methyl gallate; 3 gallic acid; 4 S-2,3,4,6-tetra-0-galloylglucose. 3.-t @I ,I I i LJ FIG.8. Chromatogram of products of methanolysis of gallotannin after 5 hours. 1 /3-Penta-O-galloylglucose; 2 methyl gallate; 3 gallic acid; 4 S-2,3,4,6-tetra-O-galloylglucose ; 5 methyl m-digallate ; 6 m-digallic acid; 7 gallotannin.galloyl) was finally subjected to hydrogenolysis. The synthetic p-penta-0-galloylglucose was identical with the major methanolysis product analytically chromatographically and optically (Table 3). The /htetra-O-galloylgIucoses were also identical. Similar results are obtained from Chinese gallo- tannin and from Sicilian and stagshorn sumach tannin and the isolation of p-penta-0-galloylglucose proves that these tannins must be based on a /3-penta- O-galloylglucose core to which several (probably 3-5) additional galloyl groups are attached as depside linkages. When however the methanolysis FH.O.Ga1loyl AU was interrupted after 5 hours (Fig. S) the chromato- grams showed that the reaction was incomplete and that methyl m-digallate was now present; this has been confirmed by the isolation of this depsidic ester.PROCEEDINGS but in the presence of air it is gradually converted into ,8-tetra-O-galloylglucose (and other artefacts). Surprisingly it was foundls that the reaction was not accompanied by formation of either methyl gallate or gallic acid. Instead ellagic acid was formed; this was detected chromatographically 3.nd was isolated and characterised. Consequently it is suggested that 19-tetra-O-galloylglucoseis produced from the gallo-tannin by oxidation of the l-galloyl to the ellagoyl group which is then preferentially removed; the oxidation may occur either in the plant itself or under aerobic met hanoly sis. Since the time of Fischer there has been a tendency to isolate the gallotannin in modified ways and some workerslg have employed extraction with aqueous ethanol and subsequent removal of the solvents by TABLE3.Galloylglucoses. C H Glucose RF* in p-Penta-O-galloyl-p-Tetra-O-galIoyl-glucose glucose Synthetic Synthetic Methanolysis1Methanolysis (%) 51-2 52.7 52-4 52.1 52.4 51.4 (%) 4-0 3.8 3.7 3.6 3.9 4.0 224 22.5 19.1 18.9 19.0 19.2 22.1 22.3 (a) 0.20 0.02-4.14 0.024-14 0.20 (b) 0.60 0.58 0.58 0.60 COMe 55.4 17.5” 17.7 54-6 * For solvents see Figs. 14. The isolation of methyl m-digallate at an inter- mediate stage suggests that at least two of these additional galloyl groups are attached chain-wise to one of the galloyl groups of the core as indicated in formula (XV) although there is no evidence that attachment is at the terminal grouping of the glucose.Nothing further is known about the remaining 1-3 galloyl groups and in fact they may be distributed at random so that the tannin would be heterogeneous and composed of a mixture of isomeric and closely related structures which present techniques have failed to resolve. Electrophoretic processes are at present under examination. Consideration of the origin of p-2,3,4,6-tetra-O- galloylglucose raises a point of interest. There are chromatographic indications of the presence of other artefacts during the methanolysis but p-tetra-0- galloylglucose only has been isolated so far. The amounts of the artefacts are greatest when old tannin extracts are used or when the methanolysis is carried out in the presence of air; a gallotannin sample pre- pared from freshly collected sumach leaves when subjected to anaerobic methanolysis gave methyl gallate and p-penta-O-galloylglucose free from /3-tetra-O-galloylglucoseor other artefacts./3-Penta- O-galloylglucose is stable to anaerobic methanolysis Unpublished observation. l9 E.g. Ilgin Ber. 1914 47 485. evaporation. Now although ethanolysis is slower than methanolysis there is no doubt that ethyl gallate is produced in considerable amounts under these conditions and consequently any criticisms of Fischer’s work that are based on tannins isolated by these methods are not valid. Extraction with methanol and ethanol must be avoided and evapora- tion of aqueous solutions must be carried out as rapidly as possible and at room temperature if reproducible results are desired.Crude tara tannin isolated from the pods of CaesaZpinia spinosa has usually been classified as a hydrolysable tannin although it is more acidic than gallotannin (Burton and Nursten20). Fig. 9 shows the complexity of the extract ;13 gallic quinic (XVII) shikimic (XIX) and m-digallic acid (I) are present together with carbohydrates theogallin glucogallin (XII) etc. The major component can be purified either by counter-current distribution in ethyl methyl ketone-water systems or better by precipitation from ethereal solution by benzene. A colourless chromatographically pure amorphous powder was obtained which on hydrolysis with acid alkali or tannase gave gallic acid and quinic acid.Acid hydrolysis was complete after 8 hours and quinic and gallic acids were isolated and characterised 2o Burton and Nursten “The Chemistry of the Vegetable Tannins,” Society of Leather Trades’ Chemists Croydon 1956 p. 61. NOVEMBER 1961 chemically as well as chromatographically. The analytical figures are shown in Table 4;the molecular weight was found by isothermal distillation to be 800-900 and the equivalent and quinic acid con- tent were determined by titration to pH 5.3 which was selected after experiments with standardised mixtures of quinic acid and glucogallin. After titra- tion to pH 5.3 tannase was added and when hydrolysis was complete the gallic acid was titrated -“as graphy (Fig.11); these have not been completely separated and further work is in progress. When the methanolysis was interrupted after 1 hour methyl m-digallate was identified chromatographically and it is concluded that at least three galloyl groups must be present as in the structure (XVI). 0 + -b FIG.10. Chromatogram of purified tara tannin. The replacement of glucose by quinic acid in tara tannin is interesting biogenetically and the bio- synthesis of gallic acid in the mould Phycomyces blakesleeanus has been examinedz1 as a preliminary to a study of its formation in the plant. Several workers have reported the formation of gallic acid TABLE 4.Analysis of purified tara tannin. Liberated on hydrolysis -.- C H Gallic Quinic (%I Purified tara tannin 52.6 Calc.for penta-O-galloy lquinic acid 52-8 to pH 6.5. The figures are consistent with a penta-0- galloylquinic acid structure ; however the gallic- quinic acid determination is not altogether satis- factory and new methods may lead to a revision of the structure. At pH 5-6 methanolysis of tara tannin was complete in 48 houm and methyl gallate and two “basic-tannin” residues were detected by chromato- HO ,CO,H HO OH Haslam Haworth and Knowles J. 1961 1854. 22 Albrecht and Bernard Helv. Chim.Acta 1947 30 627. (%) Equiv. acid (%) acid (%) 3.6 969 87.0 21.0 3.4 952 89-3 21.2 by the mould when grown on a glucose medium and Albrecht and Bernard22 isolated both gallic and protocatechuic acid from old cultures of the mould.We have confirmed these early observations and shown in addition that pyrogallol and catechol are also present; and the isolation of shikimic acid (XTX) and 5-dehydroshikimic acid (XX) and the chromatographic detection of 5-dehydroquinic acid nu (W (XVIII) after 60 days’ incubation of the mould sug- gested these acids as intermediates in the biosyn- thetic sequence. It was found that Phycomyces blakesleeanus grown on glucose supplemented by (XVII) (XVIII) (XIX) or (XX) gave the greatest yield of gallic acid when the last (XX) was used. The medium replacement technique23 was also em- ployed in this growth on the glucose medium was interrupted when gallic acid formation was estab- lished and the medium was then replaced by glucose media supplemented by one of the four possible pre- cursors.Immediate resumption of gallic acid forma- tion was obtained from 5-dehydroshikimic acid (XX) but with the others induction periods were observed [36 hours for 5-dehydroquinic acid (I); 5 days for glucose or quinic acid (XVII)] before gallic acid was again detected. In addition paper- chromatographic analysis of the replaced media showed (a) that 5-dehydroshikimic acid (XX) gave no acid except gallic and protocatechuic (b) 5-de-hydroquinic acid (XVIII) gave also 5-dehydro-shikimic (XX) and shikimic acid (XIX) and T b FIG.11. Chromatogram of products of methanolysis of tara tannin afer 48 hours.1 Methyl gallate; 2 gallic acid; 3 methyl m-digallate; 4 and 5 mixture of poly-0-galloytated quinic acids. (c) shikimic acid 0gave also 5-dehydroshikimic (XX). All these observations suggest 5-dehydro- shikimic acid as the immediate precursor of gallic acid. The alternative involving conversion of 5-dehydroshikimic (XX) into protocatechuic acid and thence into gallic acid is excluded because (i) gallic acid appears before protocatechuic acid and PROCEEDINGS (ii) addition of [14C]protocatechuic acid did not lead to isotopically labelled gallic acid. Generally labelled [14C]protoca techuic acid was obtained from [14C]quinic acid (11),produced in leaves and petals of roses grown in [14C]carbon dioxide as described by Weinstein Porter and Laurencot.24 [14C]Quinic acid (11) was then oxidised with platinum and oxygen to [14C]dehydroquinic acid (I) which with hydro- chloric acid gave [14C]protocatechuic acid (V).HA& (XVI 1) Attention should be drawn to the relation between the gallotannins and the ellagibnnins. Strictly speak- ing chebulinic acid a crystalline constituent of myrobalans is a gallotannin although it is closely associated with the ellagibnnins. Chebulinic acid (XXI) first isolated by Fridolin26 in 1884 was shown by Freudenberg et a1,2s to give on acid hydrolysis glucose three molecules of gallic acid and “split acid,” the last being renamed chebulic acid by Schmidt and his colleagues27 who advanced for it structure (XXII). Schmidt et a1.= also isolated cori- lagin (XXrII) and chebulagic acid (XXIV) as crystalline components of myrobalans ; these are ellagitannins and the former may be regarded as a dehydro-l,3,6-tri-O-galloylglucose,and the latter acid as a dehydrochebulinic acid.The relationship with the gallotannins becomes more marked on con-sideration of the structure of chebulic acid (XXII). The structure was supported by oxidation with per- manganate to the hydroxybutanetetracarboxylicacid lactone (XXV) which Schmidt and his colleagues29 synthesised from methyl fumarate and 1-acetoxy-butadiene-1,3 as indicated in the chart. 23 Evans Ann. Reports 1956 53 282. 24 Weinstein Porter and Laurencot Contrib. Boyce Thompson Inst. 1959 20 121. Fridolin Dissertation Dorpat 1884. 26 Freudenberg et a!.Ber. 1919,52 1238; 1920,53 1728; Annalen 1927 452 303. 27 Schmidt and Mayer Annalen 1951,571 1; Schmidt Lademann and Himmele Chem. Ber. 1952 85,408. Schmidt and Lademam Amah 1950 569 149; 1951 571 232; Schmidt and Schmidt ibid. 1952 578 31; Schmidt and Nieswandt ibid. 1950,568 165. Zg Schmidt “The Chemistry of the Vegetable Tannins,” Society of Leather Trades’ Chemists Croydon 1956 p. 68 “Fortschrete der Chemie der organischen Naturstoffe,” 1956 VoI. XIII p. 78. NOVEMBER 1961 Oxidation of the trimethyl ether of chebulic acid by ferricyanide to 3,4,5-trimethoxybenzoic acid:* and its pyrolysis31 to succinic acid and 5,6,7-tri- methoxyisocoumarin-3-carboxylic acid (XXVIII) provide conclusive confirmatory evidence particu- larly as the nature of acid (XXVIIr) has been con- firmed by independent synthesis32 involving Claisen t HO ICH CH,*O:I C OH O W OH HOaC$ HO\ ,CH*CO,H HO C.H $H .CH,-CO~H CO H (XXII) OH *OCO 1 \ OH -CH I OH ao Haworth and de Silva J.1951 3511. 81 Haworth and de Silva J. 1954 3611. a2 Haworth Pindred and (in part) Jefferies J. 1954 3617. 33 Schmidt and Bernauer Annalen 1954,588 21 1. 409 condensation of ethyl oxalate with the ester 0. The product was the diethyl ester (Xxvrr),which in hot dilute sulphuric acid yielded 5,6,7-trimethoxy- isocoumarin-3-carboxylic acid (XXWI) differing from the isomeric 4-carboxylic acid prepared by condensationof the ester (XXVI)with ethyl formate. It is interesting that chebulic acid (Xxrr) may be ,CH-CH k)C=CH +CH.OAC \ PAC C H2Cti ,C02Me H2C\ _Ic ,c=c HF-$ 'H Me0,C 'H MQO,C C02Me O-CH*C02 H H025 502H I I CH*CO,H 2 4% I c-OC-CH.CH,C02H HF-FH H02C C02H (XXW 1 Ozonisation of each of the two (A)-forms.2 Resolution by strychnine fotlowed by lactonisation. derived hypothetically from ellagic acid (II) by hydrolysis oxidation and reduction these reactions involving formally three molecules of water. Similar arguments may be applied in the case of brevifolincarboxylic acid which was isolated by Schmidt and Bernaue? on hydrolysis of brevilagin a constituent of algarobilla extract. This acid behaved as a p-keto-acid readily losing carbon dioxide to give brevifolin the structure (xxxr) of which was Me01 C0,Et Me0 \ CH C02Et Me 8 (XXVI) C0,Et MQ0&'% (xxvr I) (xxvI I I ) M~O&&~C-CO,H I MQO HO,C WOH $CO,H CH*C02H PROCEEDINGS established by two independent of the attractive,= particularly because unlike the /3-keto- trimethyl ether.The first approach involved oxida- acid structure it could be derived rationally from tion of ellagic acid (II)to flavellagic acid 0, ellagic acid (11) by hydrolysis oxidation and re- which was further oxidised by alkaline hydrogen peroxide to the acid (XXX; n = 1; R = H). The Ho\ / I OH II H2C-CH (XXX III) trimethyl ether (XXX; n = 1 R = Me) was con- verted by the Arndt-Eistert technique into the homo- logue (XXX; n = 2 R = Me) which with phos- phorus pentoxide yielded the trimethyl ether of brevifolin.In the second synthesis the Gomberg reaction between the diazonium compound (Xxxrr) and the methoxycyclopentenone (XXXTII) was used. Of the alternative formulae 0and (XXXV) for brevifolin carboxylic acid Schmidt and Bern-sue? first suggested the /3-keto-acid structure (XXxrV) but IateP the vinylogous structure (XXXV) was established by synthesis from the diazoniurn salt (XXxTl) and the methoxyoxocyclo- pentenecarboxylic acid (XXXVI). This structure (XXXV)has always been more HO II H,C -CH.CO,H (XXXIV) ?Me (xxxv; H02C*CH- CH (XXXVI) arrangement of the benzilic acid type involving addition of two molecules of water and removal of two atoms of hydrogen and one molecule of carbon dioxide.Equally clear relations can be traced between the gallotannins and the ellagitannins of the Spanish chestnut (Castania vesca) and of valonea. The main components of these tannins are dehydrodigallic acid3' (XXXVJI) and valoneic acid dilactoneSS (XXXVIII) which represent dehydrogenation pro- ducts of gallic acid and of gallic and ellagic acid respectively. HO&OD2 CO H HO /OH HO,C OH (XXXVII) (mxvrII HT Unfortunately knowledge concerning the struc- tures of the condensed tannins is in a much more primitive state and all that can be said is that they seem to be derived from flavan types by polymerisa- tion involving the coupling of reactive aromatic nuclei possibly by condensation (Fre~denberg~~~~~~) and oxidative processes (Hathway et ~21.4~).34 Grimshaw and Haworth Chem. and Ind. 1955 199; J. 1956 418; Schmidt and Mayer Z. ungew. Chem. 1956, 68 103. 35 Schmidt and Bernauer Annalen 1955 591 133. 36 Schmidt and Eckert 2.Naturforsch. 1956 llb 757. 97 Mayer Annalen 1952 578 34. s8 Schmidt and Komarck Annalen 1955,591 156. 39 Freudenberg Annalen 1925,444 135; 1934,510 193; 1954,590 140. 40 Hathway and Seakins Biochem. J. 1957 67 239; Hathway ibid. 1958 70 34; J. 1958 520. NOVEMBER 1961 411 CHEMICAL SOCIETY MEETING THE following papers were read and discussed at a Scientific Meeting held at Burlington House on Thursday October 19th 1961. Stereochemical Control in Oxidative Cyclisation. By F. M. DEAN and H. LOCKSLEY. OXIDATIVE cyclisations giving spirans related to griseofulvin picrolichenic acid etc.have only recently attracted wide attention although the funda- mental reaction was recognised in 1914. Thus the oxidation of methylenebis-/haphthol affords such a spiran although some of its chemical features have never been fully explained. During a renewed in- vestigation the oxidation of benzylbis-/%naphthol was also studied. Here the derived spiran has two asymmetric atoms and can therefore be obtained in two (racemic) forms in the past only one racemate had been encountered but conditions have now been established in which the other is formed. The chemical and spectroscopic properties of the isomers prove that they are not structural isomers and con- figurations have been allocated on the basis of chem- ical and nuclear magnetic resonance studies.The chemical method depends on reduction to the related unsaturated alcohols and the rates at which these are transformed into the (same) unsymmetrical dibenzo- xanthen an alternative and definitive synthesis of which had to be evolved as accepted methods proved inadequate. Preliminary studies of the way in which the oxidising agent and the conditions of the oxida- tion determine the isomer produced suggest that in some cases the hydrogen may be abstracted from one side of the bisnaphthol molecule only thus supplying a converse to the cis-addition of hydrogen in catalytic reduction. A New Group of Antibiotics. By W. D. OLLISand I. 0. SUTHERLAND. AKLAVIN and rutilantin belong to a new group of antibiotics which includes the cinerubins isolated by Prelog et al.,l and pyrromycin the rhodomycins and isorhodbmycins studied by Brockmana2 These substances are basic nitrogen-containing glycosides which are hydrolysed by acid to hydroxyanthra- quinone derivatives with the general structure (I; R1 = OH; Ra and R3 = H or OH).The other products of hydrolysis include a basic sugar and in some cases neutral sugars. The 7-deoxy- derivatives of the aglycones (I; R1 = H) are fre- quently congeners of the antibiotics. It is probable that the aglycones have a similar biosynthetic origin from acetate and propionate the differences in the oxygenation pattern being the result of secondary oxidation and reduction proce~ses.~ The structures of rutilantinone (I; R1 = R2 = OH; R3= H) and aklavinone (I; R1 = OH; R2 = R3= H) obtained from rutilantin and aklavin were originally determined by a combination of chemical degradation and infrared ultraviolet and visible spectroscopy.The nuclear magnetic resonance spectra of these and related compounds and their acetates provide support for the suggested structure. A compound isolated in low yield from an aklavin- producing fermentation has been identified as 7-deoxyaklavinone by chemical degradation and physical methods. 7-Deoxyrutilantinone has not been isolated as a congener of rutilantin but its acetate is readily produced by the catalytic reduction of rutilantinone acetate. Conclusions regarding the relative stereochemistry of the 7-deoxy-compounds have been made and these may be extended to rutilantinone and aklavinone.Rotenoids and Their Relatives. By L. CROMBIE R. PEACE, and D. A. WHITING. THE problem of linear versus angular fusion of the D/E rings of natural rotenoids is discussed. Natural sumatrol and malaccol are shown to be angularly fused (I and II respectively) though (&)-malaccol of the literature is a linear form. Optical rotatory Ettlinger Gaumann Hiitter ,Keller-Schierlein Kradolfer Neipp Prelog Reusser and Zahner Chem. Ber. 1959, 92 1869. Brockmann and Tknk Chem. Ber. 1959,92,1164 and other papers in the series. * Ollis Sutherland and Veal Proc. Chem. Soc. 1960 349. dispersion correlates the configuration of the two natural compounds with that recently deduced for natural rotenone.Nuclear magnetic resonance can also be employed to decide the geometry of the B/C fusion on an absolute basis. Examination of the bark of Neorautanenia (Doliclzos) pseudopachyrrhizus reported to give colour reactions of rotenoids has led to the isolation of two new compounds neotenone and in small PROCEEDINGS yield dolichone. Degradation and physical evidence shows the first to be the isoflavanone (III) and in- formation on the structure of dolichone will be dis- cussed. Dolichone can also be isolated from yam beans (Pachyrrhizus erosus) and neotenone detected by thin-layer chromatography. The new compounds form a biogenetically connected sequence with known natural products from this source (pachyrrhizin pachyrrhizone and erosnin).COMMUNICATIONS Alkaloids of Calabash Curare :Caracurine-I1 and C-Alkaloid D By A. R. BATTERSBY G. V. RAO,and D. A. YEOWELL H. F. HODSON (THEUNIVERSITY, BRISTOL) CARACURINE-II was isolatedl from Venezuelan methochloride requires the action of both oxygen and Sfrychnos foxifera whereas S. foxifera from British acid on toxiferine-I in a non-photochemical reaction; Guiana yielded caracurine-II meth~chloride.~s~ hemitoxiferine-I is unaffected under the same con- Treatment of caracurine-V (I) with aqueous acid ditions. When toxiferine-I in dilute acid is shaken yields (by way of caracurine-Va) caracurine-VII and with oxygen and active platinum it yields 86% of 42R '2 RI 2x--__c_ 2 2 2L (VI 0 2 cara~urine-II;~ in the quaternary series t0xiferine-1~9~ caracurine-11 methochloride (theoretical oxygen up (U; R = OH) affords hemitoxiferine-I and cara- take).Unlike toxiferine-IY6 caracurine-II metho- curine-I1 methochloride* under the same conditions.6 chloride is unaffected at 120" by N-hydrochloric or In the strict absence of oxygen toxiferine-I is cleaved glacial acetic acid. almost quantitatively to hemitoxiferine-I7 and it has Pyrolysis of the methochloride gives caracurine-11 now been shown that the formation of caracurine-I1 base (70 %) m.p. 248-249" which was converted by * All the quaternary salts are bis-derivatives. Asmis Schmid and Karrer Helv. Chim. Acta. 1954 37 1983. King J 1949 3263. Battersby Binks Hodson and Yeowell J.1960 1848. Asmis Bachli Schmid and Karrer Helv. Chim. Acta 1954 37 1993. Bernauer Berlage von Philipsborn Schmid and Karrer Helv. Chim. Acta 1958,41 2233; Battersby and Hodson Proc. Chem. SOC.,1958 287; Boekelheide Ceder Crabb Kawazoe and Knowles Tetrahedron Letters 1960 No. 26 1 ; A,mold Hesse Hiltebrand Melera von Philipsborn Schmid and Karrer Helv. Chim.Acta 1961 44 620. Battersby and Hodson J. 1960 786. 'Battersby and Rao quoted in ref. 12; Karrer Schmid and Waser I1 Farmaco 1960 15 126. NOVEMBER 1961 methyl iodide into caracurine-11 methiodide. Ana- lyses of the base and its derivatives agree with a formula C3,H,,N,02 for caracurine-II and this was rigorously confirmed by mass-spectrometric deter- mination of its molecular weight (Found 582) kindly carried out by Dr.R. I. Reed (Glasgow). The infrared spectrum of caracurine-II shows strong absorption corresponding to ortho-disubstituted benzene rings and to ether linkages. Hydroxyl I imino > C =N- >C =C-N ,and carbonyl groups are absent (cf. recovery of caracurine-11 after mild acetylation). The ultraviolet spectrum of the alkaloid (in 1:1 H,O-EtOH) shows the presence of two in- doline residues (Amax. 246 291 mp log E 4.22 3.73) and the spectrum is virtually unaffected when measured in 1 :1 2~-hydrochloric acid-EtOH. The basic strength of the indoline nitrogen atoms is thus much lower than that of N(a) in the Wieland- Gumlich aldehyde.e Catalytic hydrogenation of caracurineII yields tetrahydrocaracurine-II which contains two indoline residues (Amax.250 295 mp log E 4-29 3-76); it follows from this and the molecular formula that caracurine-ll[ contains fourteen rings three more than toxiferine-I (11; R = OH) from which it is formed. Caracurine-11 is reduced by zinc and concentrated sulphuric acid in methanol to a diol (strong infrared OH bands) which had ultraviolet absorption showing the presence of two indoline systems (Amax. 246 294 mp); quaternisation of this product yielded material which differs from deoxy-C-Alkaloid A? The spectrum of the diol changed when measured in the above acidic ethanol proving considerable pro- tonation of one N(a)-nitrogen atom; thus the base- :Battersby and Hodson unpublished work.Bertho and Loebmann. Annalen. 1954. 588. 182. weakening structural feature in caracurine-II has been removed in the diol. Hot acetic acid readily converted the diol into its diacetate in keeping with the expected (see toxiferine-I; 11 R = OH) allylic nature of the hydroxyl groups. In concentrated sulphuric acid caracurine-11 shows ultraviolet absorption corresponding to two -t Ar-N =C< residues [spectrum compared with that of tde salts (VI; R = Bun) in same solvent] and caracurine-II was recovered from the diluted acid. The spectrum of caracurine-XI diol in concentrated sulphuric acid showed considerable protonation of both N(a)-nitrogen atoms. These results prove that caracurine-II contains two carbinolamine ether systems (HI) which are not con-verted into the protonated form (IV) by aqueous acid but which are so protonated in concentrated sulphuric acid; the weak basicity of nitrogen in such systems is known.1° The systems (IU) account for two of the three new rings in caracurine-11 and the third demands a new carbon-carbon bond.Cara- curine-11 diol then has the partial structure (V) and tetrahydrocaracurine-II the partial structure (111; double bonds reduced) both with a new carbon- carbon bond in the complete molecule. Further this knowledge when considered with known relations in this field,l1Sl2 allows the partial structure (VII) to be written for C-Alkaloid D again with one new carbon-carbon bond. The authors thank the Medical Research Council and the Tropical Products Institute for financial sup- port and Drs.A. Furst and H. Els (F. Hofmann-La Roche Basle) for generous help. (Received September 25th 1961 .) lo Kupchan Johnson and Rajagopalan; Tetiahedron 1959,7 47. l1 See Bernauer Furschr. Chern. org. Naturstofle 1959 17 184. IZ See Batteisby and Hodson Quart. Rev.,1960 14 77. AIkaIoids of Calabash Curare The Structure of Caracurine-H and C-Alkaloid D By A. R. BATTERSBY and L. M. JACKMAN D. A. YEOWELL (CHEMISTRY DEPARTMENTS UNIVERSITY OF BRISTOLAND IMPERIAL COLLEGE OF SCIENCE AND TECHNOLOGY, LONDON) M. HESSE,H. HILTEBRAND H. SCHMID and H.-D. SCHROEDER W. VON PHILIPSBORN and P. KARRER* (ORGANISCH-CHEMISCHES DER UNIVERSITAT, INSTITUT ZURICH) INDEPENDENT researches at Zurich1 and at Bristo12 have established the partial formulae (ID)and (VII),? respectively for caracurine-U and C-Alkaloid D and the presence of a new carbon-carbon bond in both alkaloids.The complete struc- tures (VTU) and (IX),respectively have now been elucidated. Nuclear magnetic resonance data (56.4 Mc./sec. in * 52nd Paper on Calabash Alkaloids. 1. One set of formulae is used for this and the preceding Communication. Schroeder Hiltebrand Schmid and Karrer Helv. Chirn. Acta 1961 44 34. ’ Battersby Hodson Rao and Yeowell preceding Communication. PROCEEDINGS CDCl,) for caracurine-V caracurine-II and their derivatives are given in Table 1. Assignment of the absorption near 5.0 to the protons at positions 17 and 17' is based on the absorption of analogous protons in related compounds of known structure viz.carbogeissoschizoline3(X; R = H) (4-80, 5.43 AB system) the carbinolamine ether4 (XI) (4.69 5.30AB system) and geis~ospermine~ (X; R = com-plex indole) (4.81 doublet). In the spectrum of HC Toxiferi-Toxiferine -1' or C-dlhy C-dihydrotoxiferine -I caracurine-V5 (I) the absorption of the carbinol-amine ether protons 17 and 17' appears as an incompletely resolved doublet having line width equivalent to a coupling of ca. 3 c./sec. with the protons at the adjacent positions 16 and 16'; when the spectrum is determined in CCl (60 Mc./sec.) the doublet (5.44) is well resolved (J ca. 2 c./sec.). The corresponding absorption in the spectrum of cara-curine-11 (VIII) is a very sharp singlet showing the absence of protons at positions 16 and 16'; the sharp singlet at 5.80 superimposed on other absorption is assigned to the protons at positions 2 and 2'.In the spectrum of tetrahydrocaracurine-II2 (VIU ; olefinic Janot Tetrahedron 1961 14 113. double bonds reduced) the latter absorption is a sharp singlet at 6-08 and is completely separated from interfering absorption which has been moved to higher fields by reduction of the 19 20-and 19' 2W-double bonds. Caracurine-11diol does not show absorption in the 5-0 region. These observations confirm the presence of two carbinolamine ether systems (III)in caracurine-LI and what is important prove the absence of protons at positions 16 and 16'.H+ Ho&; ,,o I ':& or H,O &N\ HO-y' H (XII) I t Table 2 collects the (60Mc./sec.) nuclear magnetic resonance results for C-Alkaloid D (IX)and cara-curine-Vdimethofluoride [I;N(b),N(b') methylated]. The former shows a sharp singlet for the protons at positions 17 and 17' which appears in the spectrum of the latter alkaloid as a broad peak. Further proof for the presence of protons at positions 17 and 17' of caracurine-II was obtained by measurements on the dimethofluoride in concentrated D,SO in which the absorption of these protons appears at much + lower fields as expected for the type :N=CH-+ (Table 2). That the Ar-N=C system is present Battersby and Le Count unpublished work. Bernauer Berlage von Philipsborn Schmid and Karrer Helv.Chim.Ada 1958 41 2293. NOVEMBER 1961 under these strongly acidic conditions is shown by the ultraviolet spectra (Table 2). Thus the combined evidence establishes the presence of one proton at each of the positions 2 2' 17 and 17' and proves that carbon atoms 16 and 16' are fully substituted. 415 C-Alkaloid D respectively. The relative and absolute stereochemistry follows from that5s6 of their parent alkaloids toxiferine-I (II; R = OH) and C-dihydro- toxiferine-I (II; R = H) and the fact that the new rings can only be closed in the illustrated way unless TABLE 1. r-Values and corresponding proton equivalents (inparentheses). 18 18' Aromatic 19 19' 17 17' 2 2' and 21 21' Caracurine-V 2.7-3.7 (8) 4.15 (2) 5.31 (2) a a Caracurine-I1 2.7-3.7 (8) 4.76 (2) 5.14 (2) 5.8' a Tetrahydrocara- curine-I1 2.7-3.7 (8) -(0) 5.15 (2) 6.08 (2) 6.10 Caracurine-XI diol 2.7-3.7 (8) 4.29 (2) -(0) a a 0 Absorption between 5.5 and 6.6integrated to ten protons.TABLE 2. Nuclear magnetic resonance* (D,O)' 19 19' 17 17' 2 2' 18 18' C-Alkaloid D Q (J 6 c./sec.) 377 (2) S 316 (2) S 291 (2) D (J 6 c./sec.) 126 (6) Caracurine-V dimet hofluoride -B 309 (2) B 279 (2) - Nuclear magnetic resonance* (D,SO,)b Ultraviolet (H,SOJ 17 17' 2 2' Caracurine-I1 - 236 mp (log E 3-87) 322 mp (log E 4.35) C-Alkaloid D - 236 mp (log E 3-86) 322 mp (log E 4.32) Caracurine-II dimethofluoride B 532 B 378 - Compound (VI; R = H with 19 20 saturated) B 534 231 mp (log E 3-75) 291 mp (log E 3.97) * Since the chemical shifts cannot be related precisely to the 7-scale they are quoted a5 c./sec.from tetramethylsilane (marked a) or from tetramethylsilane in CCll (marked b) both as external standards. Proton equivalents are in parenthesis. S = singlet B = broad band or incompletely resolved doublet D = doublet Q = quartet. c Acetone was added to D20to avoid gel formation. The new bond known112 to have been formed as these alkaloids are produced therefore joins posi- tions 16 and 16'. This conclusion when taken with the chemical evidence,lS2 leads unequivocally to the structures (VIU) and (uc) for caracurine-II and * Battersby and Hodson J. 1960 786. McPhail and Sim following Communication. impossible strain is introduced.The complete struc- ture and stereochemistry of caracurine-I1 were com-municated to Dr. G. A. Sim (Glasgow) who kindly informed us that the identical conclusions had been derived independently by X-ray analysis.' PROCEEDINGS The mechanism whereby these alkaloids are formed must explain the necessity for both acid and oxygen and the formation of hydrogen peroxide during the reaction in low yie1d.f Oxidation could proceed as indicated to the intermediate (XII) which by transannular reaction could generate the required systems; primary attack at nitrogen to give (XLU) could also lead as shown to the caracurine-11-C-Alkaloid D system. There are other possibilities related in principle to these. The authors thank Drs.A. Fiirst and H. Els (F. Hofmann-La Roche Basle) and Professor M.-M. Janot (Paris) for valuable materials and the Tropical Products Institute (London) and the Schweizerischen Nationalfonds fur wissenschaftliche Forschung for financial support. (Received September 25t11 1961.) $ The reasons for the low yield have been studied and the results together with details of the control experiments will be published later. The Structure of Caracurine-I1 By A. T. MCPHAIL and G. A. SIM DEPARTMENT GLASGOW, (CHEMISTRY THEUNIVERSITY W.2) IN order to elucidate the molecular structure of caracurine-11 an alkaloid isolatedl from Strychnos toxifera we have carried out a detailed X-ray analysis of caracurine-II dimethiodide. Independently of the chemical studies,2 which lead to identical con- clusions our results demonstrate that caracurine-I1 dimethiodide has constitution and stereochemistry as defined in (I)(or its mirror image).The dimethiodide crystallises in the orthorhombic system space group P2,212, with four molecules of C40H4412N402 in a unit cell of dimensions a = 18.59 b = 27.44 c = 7.52 A. Three-dimensional X-ray diffraction data (1 300 independent structure ampli- tudes) were obtained from equi-inclination Weissen- berg photographs taken with Cu-Ka radiation. The co-ordinates of the iodide ions were deduced from Patterson syntheses. Thereafter the remaining atoms other than hydrogen were located by com- puting successive three-dimensional Fourier syn-theses with increasing numbers of atoms included in the phasing calculations as they became clearly defined on the maps; after the third round of calcula-tions structure (I) was established.Before we notified Dr. A. R. Battersby of our findings we were informed by him that structure (I) had been deduced from the chemical investigatiom2 -F Superimposed contour sections parallel to (001) showing the fourth three-dimensional electron-density distribution over one molecule of caracurine-I1 dime thiodide. Superimposed contour sections illustrating the fourth three-dimensional electron-density distribu- tion over one molecule are shown in the Figure. The value of R is 24% and refinement is continuing. The distance between the two quaternary centres in the molecule is 8-6 A whereas in toxiferine I and d-tubocurarine the corresponding separations have been estimated3 at about 14 A; this distinction is of King J.1949 3263; Asmis Schmid and Karrer Helv. Chim. Acta 1954 37 1983; Battersby Binks Hodson and Yeowell J. 1960 1848. Schroeder Hiltebrand Schmid and Karier Helv. Chim.Acta 1961,44 34; Battersby Hodson Rao and Yeowell also Battersby Yeowell Jackman Schroeder Hess Hiltebrand von Philipsborn Schmid and Karrer preceding two Communications. Battersby and Hodson Quart. Rev. 1960,14 77. NOVEMBER 1961 417 interest in view of the relatively low physiological of material. The calculations were performed on the activity of caracurine-II dimethochloride. Glasgow University DEUCE computer with pro- grammes devised by Dr.J. S. Rollett and Dr. J. G. We are grateful to Professor J. Monteath Robert- Sime. We also thank the Department of Scientific son F.R.S. for his encouragement and to Dr. A. R. and Industrial Research for a maintenance grant (to Battersby for suggesting the problem and for supplies A.T. McP.). (Received,September 25th 1961.) D-Xylothiapyranose A Sugar with Sulphur in the Ring By J. C. P. SCHWARZ and K. C. YULE (DEPARTMENT UNIVERSITY OF CHEMISTRY OF EDINBURGH) MANY sugar derivatives in which oxygen has been replaced by sulphur have been preparedl and some occur in Nature. However compounds in which the ring oxygen is replaced by sulphur have not yet been described although comparison of the physical pro- perties and reactivity of such compounds with those of their oxygen analogues would be of considerable interest in understanding the behaviour of carbo- hydrates.D-Xylothiapyranose (I) and some of its derivatives have now been synthesised. 0.001M-potassium hydrogen phthalate2 (pH ca. 4-4) but it was very much faster (half life ca. 10 min.) in an 0*005~-phthalate buffer of pH ca. 6.6; this remarkable behaviour is being examined in detail. Although the thiol (11; X = SH) reacts rapidly with iodine and with acidified dichlorophenolindophenol the thio-sugar (I) reacts relatively slowly with the former and very slowly with the latter. This together with the mutarotation results suggests that the sugar exists mainly in the thiapyranose form shown.With acetic anhydride in pyridine the thio-sugar gave a tetra-acetate m.p. 99-100' [a]:0 +219" (c 2.2 in CHCI,); with acetic anhydride in the presence of sodium acetate it gave the same tetra- acetate with a smaller quantity of an isomeric tetra- acetate map. 157-158.5" [a]kl -49" (c 2-3 in CHCI,). Ultraviolet and infrared spectra showed Heating 1,2-O-isopropy~idene-5-O-tosy~-~-xy~ofura-nose (11; X = OTs) with sodium thiocyanate in acetone or preferably in a eutectic melt containing potassium thiocyanate gave the thiocyanate (11; X = .SCN). With aqueous sodium sulphide this yielded the thiol (11; X = SH) m.p. 85-87' [a]","-52" (c 1-99 in CHCI,) together with a little of the corresponding disulphide. The thiol was obtained more conveniently by heating the toluene- p-sulphonate (11; X = OTs) with sodium thiosul- phate in aqueous dimethylformamide followed by reduction of the resulting Bunte salt (11; X = S,O,Na) with potassium borohydride.Hydrolysis of the thiol under mild acidic conditions (Amberlite IR-120 at room temperature) gave chromato-graphically pure D-xylothiapyranose (I) m.p. 122-123" [a]? f202' -+ + 178" (c 2 in H,O). The mutarotation was very slow (half life > 10 hr.) in that neither of the acetates contained an S-acetate or disulphide group; they must therefore be the two anomeric tetra-acetates of sugar (I). Although the usual relation between optical rotation and anomeric configuration need not necessarily hold for such unusual compo~nds,~ it seems likely that the dextro- and lzvo-rotatory tetra-acetates are respectively the a-anomer (i.e. ~-gulo-2,3,4,5-tetra-acetoxythian) and the /3-anomer (ix.,~-ido-2,3,4,5-tetra-acetoxy-thian). This is supported by the relative mobilities of the compounds on paper chromatography with di- methyl sulphoxide as the stationary and di-isopropyl ether as the mobile phase.4 We are grateful to Dr. L. N. Owen for informing us of the work described in the following com- m~nication.~ (Received,September 21sr 1961.) Reid "Organic Chemistry of Bivalent Sulfur," Chemical Publ. Co. Inc. New York 1958-1960 Vols. 1-111; Raymond Ah. Carbohydrate Chem. 1945 1 129; Iqbal and Owen J. 1960 1030 and earlier papers Hall Hough and Pritchard J. 1961 1537. lsbell and Pigman J.Res. Nat. Bur. Stand. 1938 20 773. Lemieux and Hoffer Canad. J. Chem. 1961 39 110. Wickberg Acta Chem. Scand. 1958 12 615. j Adley and Owen following Communication. PROCEEDINGS Thio-sugars with Sulphur in the Ring By T. J. ADLEYand L. N. OWEN (DEPARTMENT IMPERIAL LONDON, OF CHEMISTRY COLLEGE S.W.7) D-XYLOTHIAPYRANOSE, the first example of a thio- sugar in which sulphur takes the place of oxygen in the ring system is described in the preceding Com- municati0n.l We have independently obtained the same compound by a somewhat different route. (I R=OTs R=OH) 1" R= SAC R'=OH) a R=SH R'=OAc) IV R= SH R'=OH) ' OAc CH,*OAc I (Vl I I :X =C.H20Ac) HC-SAc Reaction of 1,2-0-isopropylidene-5-O-tosyl-~-xylose (I) with potassium thiolacetate2 in boiling dimethylformamide gave a mixture b.p.1 30'/103 ~~lfn., nE3 1.4877 A,,,. 2300 A (E 1060),Vmax. 1745 1698 cm.-l of the S-acetyl compound (U) and (mainly) its rearrangement product the 0-acetate (ILI). Deacetylation of this mixture with sodium methoxide in methanol gave 5-deoxy- 1,2-0-iso- propylidene-5-mercapto-~-xylose (IV) m.p. 80-82" in which the thiol group titrated normally with iodine. Hydrolysis of this product (IV) with aqueous acid gave a-D-xylothiapyranose (V),m.p. 122-123' Schwarz and Yule Proc. Chem. SOC.,1961 417. Cf. Chapman and Owen J. 1950 579. Creighton and Owen J. 1960 1024. Cali5 $202 -+ +173' (c 1-6 in H,O); mutarotation was extremely slow presumably because of the rela- tively high stability of thioacetals and the final equilibrium value was obtained after a trace of ammonia had been added and the solution heated for 30 min.at 60". The sugar showed no thiol band in its infrared spectrum and gave no immediate re- actions for a thiol group but it slowly reacted with aqueous iodine and by titration during several hours with intermittent heating about 85% of the theoretical uptake for one thiol group was observed. With acetic anhydride-sulphuric acid it gave the tetra-0-acetyl derivative m.p. 98-100" [&]El +210° (c 2.8 in CHCl,) which showed no absorp- tion characteristic of an S-acetyl group at 2300 A; this compound therefore must also have the thiapyranose structure. A hexose derivative of similar structure has also been obtained.5,6-Dideoxy-5,6-epithio-1,2-0-iso-propylidene-~-idofuranose~ (VI) on reaction with acetic anhydride-acetic acid-potassium acetate at 130" gave 3,6-di-O-acetyl-5-acetylthio-5-deoxy-l,2-0-isopropylidene-L-idofuranose(VII) m.p. 69-70' Am,,. 2300 8 (E 4200) +10' (c 0.6 in CHCI,); hydrolysis with aqueous acid furnished a non-crystalline sugar which on acetylation gave penta-0- acetyl-L-idothiapyranose (VIJI) m.p. 9@-92" [a]z3 3-41" (c 1 in CHCI,). The ultraviolet and infrared absorption spectra confirmed the absence of an S-acetyl group. We much appreciate Dr. J. C. P. Schwarz's courtesy in informing us of his work on xyIothia- pyranose. (Received September 21st 1961.) Ring Inversion in Cyclohexane and Related Molecules as Studied by Nuclear Magnetic Resonance By R.K. HARRIS and N. SHEPPARD (THE UNIVERSITY LABORATORY, CHEMICAL CAMBRIDGE) THEtemperature dependence of the proton nuclear At room temperature the spectrum consists of a magnetic resonance spectrum of cyclohexane has single sharp line since the chair-chair interconversion been studied qualitatively by Jensen et al? at is very rapid. Below -9O"c. however this inversion 60Mc/sec. and by Moniz and Dixon2 at 40Mc/sec. has been slowed so that the spectrum appropriate to Jensen Noyce Sederholm and Berlin J. Arner. Chem. Suc. 1960 82 1256. Moniz and Dkon ibid. 1961 83 1671. NOVEMBER 1961 a rigid chair isomer is obtained. This takes the form of an unsymmetrical broadened quartet (at 40 Mc/sec.).From the coalescence temperature Tc Of -66.5"~at 60 Mc/sec. Jensen and his co-workers1 calculated a value of 10.1 kcal./mole for the free energy of activation AF* of the inversion (revised However this assumes that the correlation time T,at the coalescence temperature is given by rc = 1/2/2578 where 8 is the chemical shift in cfsec. This formula is only correct for averaging between two single uncoupled lines and when line widths in the absence of exchange are effectively zero. The cyclohexane case would seem to approximate better to the averaging of the chemical shift of an AB system* with J/8 M Q,where the correct equation can be shown to be T~ = l/n& Moreover the consider- able width of the lines when treated in this way leads to an apparently lower value of Tc.t An estimated correction to Jensen's value for AF* is 4-03 kcal./mole.This value has been checked by a more rigorous investigation of the rate of broadening of the single peak over the temperature range -20" to -70" at 40 Mc/sec. similar to the work of Tiers3 on per- fluorocyclohexane. We have used Eyring's equation in the form log rT = AH*/4.59 (1/T) -10-620-AS*/4.59.T was obtained by the method of Piette and Anderson5 at the fast-exchange approximation. The line width was measured directly whenever practic- able; at higher temperatures when the width was very small the exponential decay of the line at fast passage was measured instead. Piette and Anderson's equation would appear to be valid for an AB system especially since in this case the second moment of the spectrum is the same as for two single lines with the same chemical shift.Moreover the validity is confirmed by the results. The plot of log TTagainst l/Tgives a very good straight line down to within 10" of Tc. The slope gives a value of the enthalpy of activation AH* = 9.0 f 0.2 kcal./mole. 8 being taken as 18-2 c/sec. (the value obtained by Jensen corrected to 40 Mc/sec.) we obtained the entropy of activation from the intercept AS* = -7.9 f1-0 e.u. This compares with the value of -10 f1 e.u. found by Tiers3 for perfluorocyclohexane and contrasts 419 with the value of dS* M 0 for cyclohexane assumed by Jensen ef aZ.l Our results give dF* = AH* -T'S* = 10.6 kcal./mole at -66.5 c in excellent agreement with the corrected value obtained by Jensen et a1.l We can also report preliminary results of low- temperature work on cyclo-octane.This compound also shows a single-line spectrum at room tempera- ture which starts to broaden only below -85"c. At -113"c the line is 5.8 c/sec. broad with no sign of structure. We could not obtain the true low-temperature spectrum (a rough calculation shows that Tc is about 120"~). The results indicate a low enthalpy of activation of 2.6 f0.9 kcal./mole and a chemical shift of the same order as cyclohexane being assumed an unusually large negative entropy dS* M -30 e.u. Moniz and Dixon2 reported that cis-decalin shows no indication of spectral changes even at -120"~.We have confirmed this result and have further looked at the spectrum of 1,l- cis-1,2- and trans-1,2-dimethylcyclohexane to -120"c; only very slight changes are observable and we conclude that these compounds are probably inverting rapidly even at such low temperatures in contrast to the halogenated cyclohexanes studied by Reeves and Str~mme.~ Moniz and Dixon2 have suggested that strain in the ground state may have lowered AH* for cis-decalin. Tiers3 postulated a similar effect for perfluorocyclohexane and this could also apply to the dimethylcyclohexanes but there are two alterna- tive or possibly complementary explanations for the absence of an experimentally observable temperature effect above -120"~. Firstly the chemical shift in- volved may be considerably smaller than expected; little is known about its precise magnitude.Secondly AS* may be positive; a simple calculation shows that if AH* !z 10 kcal./mole a value of dS* of + 15 e.u. would explain the observed results. Such variations in AS* could be associated with differenctes in the path of the inversion. A negative value (as found for cyclohexane and perfluorocyclohexane) would be consistent with a rigidly defined intermediate such as the planar form; a positive value could arise with a more flexible and unsymmetrical transition state for example the boat form.' (Received August 22nd 196 I .) t By Tcis here meant the temperature at which the maximum at the band centre changes to a minimum on cooling. Tiers Proc.Chem. SOC.,1960 389. Pople Schneider and Bernstein "High-resolution Nuclear Magnetic Resonance," McGraw-Hill Book Co. 1959. ti Piette and Anderson J. Chem. Phys. 1959 30 899. Reeves and Strsmme Camd. J. Chem. 1960,38 1241; J. Chem. Phys. 1961,34 1711; Trans. Faraday Soc. 1961, 57 390. Hazebroek and Oosterhoff Discuss. Faraday SOC.,1951 10 87. PROCEEDINGS Internal Rotation of Methyl Groups as Detected by Band Contours in Infrared Spectra of Solutions and Gases By W. J. JONES and N. SHEPPARD CHEMICAL LENSFIELD (UNIVERSITY LABORATORY ROAD CAMBRIDGE) PREVIOUSLY~ we made a comparison of the widths and shapes of infrared vibrational band contours of certain molecules in solution in non-polar solvents and in the gas phase. It was concluded that a consider-able degree of rotational freedom is retained in the condensed state.In order that such conclusions can be drawn realistically from infrared spectral evidence it is necessary that the molecule should have at least one very small nioment of inertia. This condition is required so that rotational motions if present will 1600 1500 1400 I I 1 I I 1600 1500 1400 cm3 FIG.1 (a) Nitromethane in solution in carbon tetra- chloride (0.9 molar in 0.1 mm. and 0.0125 mm. cells). (b) Nitromethane vapour (8 mm. pressure in a 10 cm. cell). give rise to band contours in solution of notably greater breadths than those expected from other causes. The perpendicular vibrations of the methyl halides fulfil this condition ;their considerable band widths in both phases (with resolvable structure in the gas phase) are principally caused by rotational motions about the three-fold symmetry axis.We next considered whether it might be possible to detect free intemal rotation of a small part of a molecule with respect to the remainder from infrared band widths in solution. If the small part has a much lower moment of inertia about the internal axis than the larger part then the reduced moment of inertia for internal rotation approximates to that of the small part. Thus the effective moment of inertia for internal rotation of a methyl group attached to a heavy framework is approximately the same as that for overall rotation of a methyl halide molecule about the three-fold axis; for free intern21 rotation the 05-(a) 'skeletal 5 0.4-vi bration v-C 1500 140y 1300 ern.-FIG.2 (a) Toluene in solution in carbon tetrachloride (1.1 molar in a 0.1 mm.cell). (b) Neopentane in solution in carbon tetrachloride(0-7molar in a 0.1 mm. cell). perpendicular vibrations of the methyl group should therefore show band-width phenomena analogous to those caused by overall rotation of the methyl halide molecules in solution. In order to test this possibility we studied the infra- red spectra in solution in carbon tetrachloride of nitromethane CH,.NO (with the very low barrier of 6-0 cal./mole for internal rotation in the gas phase2) and toluene (barrier 500 & 500 cal./mole Jones and Sheppard Trans. Faraday SOC.,1960 56 625.Tannenbaum Johnson Myers and Gwinn J. Chem. Phys. 1954,22,949. NOVEMBER 1961 i.e. possibly less than RT at 300"K3**). Portions of these spectra together with a part of the spectrum of neopentane which has the high barrier to internal rotation of ca. 4000cal./mole,5 are shown in Figures 1 a and 2. It is seen that for nitromethane and toluene the perpendicular bands of the methyl group near 1450 cm.-l (half-widths ca. 30 cm.-l) are notably broader than those of the neighbouring parallel vibration of the methyl group (as for the methyl halides) but the analogous perpendicular band of neopentane (Figure 2b) is a much narrower one. The breadth of the perpendicular band of the methyl group in toluene is particularly noteworthy in com- parison with those other bands which represent vibrations of the benzene part of the molecule.It is considered that this evidence lends strong support to the hypothesis that relatively free internal rotation of the methyl group occurs for both nitromethane and toluene in this solvent and that the presence or absence of broad bands in such cases probably pro- vides a measure of whether the barrier is consider- ably less or considerably greater than RT. The broad bands in solution imply that additional fine structure caused by internal rotation should be found in the corresponding bands for the gas phase. A1 though for these more complex molecules over- lapping bands in the spectra of the vapour state cause some difficulties we have carried out measure- Pitzer and Scott J.Amer. Chem. SOC.,1943 65 803. Pitzer Discuss. Faraday Soc. 1951 10 66. Pitzer J. Chem. Phys. 1937 5,473. ' Naylor and Wilson J. Chem. Phys. 1957 26 1057. 421 ments on the relevant bands of nitromethane and methyldifluoroborine (which also has a low barrier6) using a diffraction-grating spectrometer. It has been found that the overall band-widths for the gas phase are indeed considerably greater than would be pre- dicted on the basis of overall molecular rotation and that additional fine structure is present (Figure lb). Although the spectra are of considerable complexity it has in several cases been possible to pick out the strcng weak weak strong pattern of intensity alter- nation to be expected for the presence of a pseudo-threefold symmetry axis in the molecule resulting from free rotation.It should be noted that the symmetry properties of nitromethane and toluene' are such that the de- generacies of the perpendicular bands of the methyl groups may be slightly split; even if the barrier to internal rotation is virtually zero such splitting might occur as a result of specific vibrational inter- action with in-plane and out-of-plane vibrations of the nitro or phenyl groups. However the high-resolution studies in the gas phase mentioned above and the very close correspondence between the band contours of these methyl bands in solution with those of methyl bromide in the same medium leads us tr the conclusion that these band-width phenomena are caused by internal rotation with a barrier much less than RT.(Received July 3rd 1961.) Fuson Carrigou-Lagrange and Josien Spectrochim. Acta 1960 16 106. n-Bonding and Hindered Rotation in Inorganic Systems By P. A. BARFIELD and J. LEE M. F. LAPPERT (CHEMISTRY DEPARTMENT FACULTY UNIVERSITY OF MANCHESTER) OF TECHNOLOGY IT has recently been shown by Ryschkewitsch Brey and Sajil (for Me,B*NMePh) and almost simul- taneously2 [for Ph(C1)B-NMe (I)] by us (in col- laboration with Dr. J. K. Becconsall) that in a three- co-ordinate boron-nitrogen compound there is a substantial potential barrier to rotation about the B-N bond. This could be due either to restriction to rotation for steric reasons or to a large measure of p,-p,-bonding (h.,that there is a significant con- -+ tribution of canonical structures of type :B = N:) (see for example ref.3). Our further experiments show unequivocally that the latter explanation is correct and have also refined our results on com- pound (I). The compounds on which we report are (I) and Ph.B(NMea (II). Their lH nuclear magnetic resonance spectra were examined at 40 or 60 Mc./sec. in the range 20-150". For compound (I) methyl absorption occurs at T = 7-2 at 23"and con- sists of a doublet in accord with the two sets of lH nuclei being in two chemically different and equally abundant environments (cis or trans to the phenyl group). At progressively higher temperatures band broadening and decrease of maxima separation were evident the separation becoming zero at 118O ("coalescence") and above this temperature the single band became increasingly narrow.The association of this kind of spectral behaviour with Ryschkewitsch Brey and Saji J. Amer. Chem. Soc. 1961 83 1010. Results reported by M. F. Lappert at the Anniversary Meeting of the Chenlical Society Liverpool April 1961. Aubrey Lappert and Pyszora J. 1960 5239. variable isomeric interconversion rate is well estab- li~hed.~ Energy-barrier calculations have been made by four different approaches and afford a value of 18 & 2 kcal. mole-l. A numerical treatment based upon the Bloch phenomenological equations has been used. This relates first-order rate constants to the three spectral features (i) band maxima separa- tion before band coalescence (ii) maximum minimum ratio before band coalescence and (iii) band-width at half-height after band coalescence.With input datum of the product of (a) band maximum separation at infinitely slow exchange and (h) transverse relaxation time T (assumed equal for the two environments) the calculations have been performed by a Ferranti Mercury Computer and indicate a greater sensitivity to the choice of T of (ii) than of (i) and (iii). The value of T2 obtained experimentally from the band widths at ambient temperature is 0.19 sec. This is confirmed by the fact that (i) (ii) and (iii) gave closest agreement with * Pople Schneider and Bernstein “High Resolution 1959 pp. 218 365.PROCEEDINGS T value of the order of 0.2 sec. The fourth method was based on a single rate constant and the transition- state theory. The spectrum of compound (11) shows that the coalescence temperature for that compound is <20° which implies a barrier of <10kcal. mole-l. Steric effects with respect to rotation about the B-N bond would undoubtedly be greater in com- pound (11) than in (I) and the presence of a higher barrier in the latter therefore proves that restriction to rotation must be due to electronic (namely wbonding) rather than steric factors. Furthermore our results show that cis-trans isomerism due to restricted rotation about inorganic atomic pairs can be extended beyond the only hitherto established case that of N = N.We thank Dr. J. K. Becconsall for early nuclear magnetic resonance spectra and the Air Research and Development Command of the U.S. Air Force for support through its European Office. (Received August 21st 1961.) Nuclear Magnetic Resonance,” McGraw-Hill New York Solvolysis of the Toluene-p-sulphonate of 3#l-HydroxymethyI- A-norcholest-5-ene By G. H. WHITHAM (DEPARTMENT OF CHEMISTRY UNIVERSITY OF BIRMINGHAM) CONSIDERABLE speculation has been devoted to the precise structure of non-classical ion intermediates in solvolysis reacti0ns.l In the case of cholesteryl toluene-p-sulphonate(I; R = p-Me-C,H,.SO,-) it is not clear whether formulation (11 “unsymmetrical ion”) implying only delocalisation of the n-electrons of the 5,6-double bond or (111 “symmetrical ion”) implying also delocalisation of the electrons of the 4,5-bond is the better representation of the product- determining intermediate.l& We now report evidence which is considered to favour formulation (111).Solvolysis of the toluene-p-sulphonate of 3p-hydroxymethyl-~-norcholest-5-ene(IV; R = p-MeC,H,*SO,-) in aqueous acetone buffered with potassium acetate gave a 85% yield of hydroxylic product comprising 3 a,5-cyclo-5a-cholestan-6~-ol (V) (82%) and cholesterol (I; R = H) (18%). This result is identical within experimental error with that obtained with cholesteryl toluene-p-sulphonate under the same conditions. The inference drawn is that in both cases the solvolysis proceeds via the same intermediate viz.(111). The alcohol (IV; R = H) m.p. 98-100” [aID-26” (c 1-6 in CHCl,) is RO& assigned the P-configuration of the hydroxymethyl group on the basis of the solvolysis results outlined above. What is presumably the 3a-isomer has been described by Dauben and Ross.2 (Received September 8th 1961.) 1 (a)Winstein and Kosower J. Amer. Chem. SOC.,1959 81 4399; (b) Mazur White Semenow Lee Silver and Roberts ibid. p. 4390. Dauben and ROSS,ibid. p. 6521. NOVEMBER 1961 423 Intensification of Absorption and Rotation of the Carbonyl Chromophore by Mixing of n + 7~*with n -+ n* Transitions By R. C. COOKSON and SCOTTMACKENZIE (THEUNIVERSITY AMPTON) SOUTH THEintensification of the n -+ n* absorption of certain unsaturated ketones1B2 comes from the mixing of the forbidden n -+ T* transition (which lacks an electric dipole moment) with the allowed n -+ T* transition in which an electron is transferred from the C=C double bond (n)to the antibonding orbital of the carbonyl gro~p~-~ (T*).This is possible when the n orbital overlaps both the oxygen p (or n) orbital and the n* orbital of the carbonyl group an arrangement that places the donor T orbital asym- metrically to the symmetry planes of the carbonyl group and confers rotatory character on the mixed transition. Thus asymmetric ketones with such intensified n + T* transitions have high optical activity,ly5 and the sign of the Cotton effect is deter- mined only by the asymmetry of the PIT ~verlap.~ For a given geometrical arrangement the smaller the difference in energy between the n + n* and the n -+ T* transition the more intense the mixed n -+n* transition and the larger its rotatory power (when asymmetric).The borrowing of intensity by the n -+ n* from the T -+n* transition of course reduces the intensity of the latter. I' 0boH3 *Lb In considering possible molecular frameworks in which the double bond and carbonyl group might be held in a suitable disposition we thought that the electronic and geometrical factors might be particu- larly favourable in 5-phenylborn-5-en-2-one (11; R = H XY = 0).Also the ionisation potential of the styrene donor-group could be varied by para-substitution without disturbing the geometry. In the event these compounds showed the most extreme manifestation of this sort of interaction yet reported eclipsing even the santonides.lV6 The ketones (11; R = H Me OMe; XY = 0) Cookson and Wariyar J.1956 2302. a Birnbaum Cookson and Lewin J. 1961 1224. were obtained by oxidation with chromic oxide in pyridine of the phenylbornenoIs (U; R = H Me OMe; X = H; Y = OH) formed by reaction of (-)-5-oxoborneo17 (I) with the appropriate phenyl- lithium. Their elementary analyses and infrared spectra are consistent with their structures and that of the phenyl compound (11; R = H; XY = 0)was proved by Wolff-Kishner reduction to 5-phenylborn- Sene (11; R = H X = Y = H) identical with a sample made from epicamphor (111) and phenyl- lithium. / In cyclohexane the spectrum of phenylbornenone (U; R = H; XY = 0)showed the following maxima (E in parentheses) 217 (12,000) 222 (13,000) and 229 mp (1 1,230) ;265 mp (1 5,400 )(styrene chrorno- phore) ; 274 mp (14,900) (charge-transfer band) ; 307 (4,400) 318 (6,170) and 330 mp (5,150) (n -+n*).The spectra of the ketones (11; R = Me and R = QMe; XY = 0)were very similar but the intensity of the mixed n -+ T* band was increased because of the reduction in the energy difference between the n -+ n* and the T -+ n* transition by the electron-releasing para-substituents (for the central vibrational component of the n -+ n* band E = 6,250 and 7,750 respectively). For the same reason in a more polar solvent the mixed n -+ T* band was further increased in intensity at the expense of the band at 274 mp.Camphor itself in which any Labhart and Wagniere Helv. Chim. Acta 1959 42 2219. * Cookson and Lewin Chem. and Ind. 1956 894; Cookson Hill and Hudec ibid. 1961 589; Winstein de Vries and Orloski J. Amer. Chem. Soc. 1961 83 2013. Cookson and Hudec J. in the press. Mitchell and Schwarzwald J. 1939 889; Woodward and Kovach J. Amer. Chem. Soc. 1950,72 1009. ' Asahina Ishidate and Tukamoto Ber. 1936 69 349. PROCEEDINGS small charge-transfer contribution to the It -+n* transition can come only from saturated groups with ionisation potentials almost 3 ev higher than that of the ap-disubstituted styrene double bond hass A,,,. 292 mp (E 23). The optical rotatory dispersion of the phenyl- bornenone (11; R = H; XY = 0)([a],,+ 1060" in ethanol) was kindly measured by Professor W.Klyne and Miss J. Jackson. It showed a positive Cotton effect having a peak at 345 mp with a mole- cular rotation (aM/lOO) of 108,000". Readings could not be taken below 325 mp at which the rotation had fallen to zero; the amplitude is therefore probably about +200,W0. The optical rotatory power of an electronic transition is proportional to the scalar product of the electric and magnetic dipole transition moments (pe and pm respectively) the rotational strength being the imaginary part of the product pe-pm.9 The rotatory transition of an electron from the 2p oxygen orbital to the n*,orbital provides a magnetic moment along the C-0 axis (z) the linear charge- transfer transition of an electron from the styrene orbital to the r*,orbital of the carbonyl group pro- vides an electric moment of pe at an angle 8 to the C-0 axis and therefore of p.,cosO along the axis (IV).? Thus the electron is transferred with a spiral motion from the mixed p-T orbital to the n*orbital and the sign of the Cotton effect is determined by the direction of the twist or thread of the spiral.A left-handed thread (V) as in the transition (IV) produces a positive Cotton effect and vice versa5 The size and sign of the Cotton effect then fall nicely into the generalised octant rule recently ~uggested.~ S.M. is indebted to the Research Corporation for a grant and to the University of Rhode Island for sabbatical leave.We also thank Dr. S. F. Mason for interesting correspondence and unpubhhed infor- mation. (Received Jury 26th 1961.) t So as not to obscure the view in diagram (IV) the exo-lobe of the styrene n orbital the bonding ns orbital of the carbonyl group and the distant lobe of the 2pv oxygen orbital (which overlaps the styrene n orbital very much less), are all left out. The mixed p-n orbitals are shaded. a Cookson J. 1954 282. @ Kauzmann Walter and Eyring Chem. Rev.! 1940,26,339; Moscowitz in Djerassi's "Optical Rotatory Dispersion," McGraw-Hill 1960 Chap. 12 and references given there. Crystal Field Stabilisation as a Factor in Surface Processes By J. HABERand F. S. STONE (DEPARTMENT OF PHYSICAL AND INORGANIC CHEMISTRY UNIVERSITY OF BRISTOL) THEsuccess of ligand field theory in understanding the mechanisms of inorganic reactions in solution has aroused an interest1 in the possibility of inter- preting in similar terms the reactions of transition- metal ions at the solid-gas interface.We now report an experimental study which we believe to be among the first on heterogeneous processes amenable to direct consideration in these terms. The configurations which afford the greatest stabilisation in a cubic environment are d3 and ds,so it follows that among the common oxides of the 3d transition metals Cr203 and NiO merit special attention. We selected nickel oxide partly in view of its rather higher symmetry and partly because its absorption spectrum is somewhat better understood. The dissociative chemisorption of oxygen on nickel oxide may be regarded as completing the six-fold co-ordination of oxygen about the Ni2+(ds) ions in the surface.The increase in crystal field stabiIisation (CFS) contributes to a strong adsorption which is normally irreversible at 20". If the surface is now irradiated in the appropriate d-d bands the Ni2+ ions may be sufficiently destabilised to release adsorbed oxygen. The analogy is with an S,1 reaction. The absorption spectrum2 of NiO shows a funda- mental absorption edge at ca. 350 mp with several weak absorption bands in the visible and near-infrared region corresponding to the d-d transitions. We have accordingly examined the influence of light on nickel oxide carrying adsorbed oxygen from 200 mp to the near infrared.Photo-desorption was readily observed and experiments with filters revealed that activity was almost completely confined to a region of the spectrum from 650 to 900 mp. Since the only absorption bands in the active region are those corresponding to the d-d transitions 3A2g-+ 3T1gand 3A2q-+ lE, the spectral sensitivity result immediately implies a connection between crystal field stabilisation and chemisorption. A simple consideration of the co-ordination number of Ni2+ ions at a cleaved surface of nickel oxide shows that of the three principal planes (100) Cf. Dowden Actes du 2me Congrks International de Catalyse (Paris 1960) Vol. 11,p. 1499 Technip Paris 1961. Newman and Chrenko Phys. Rev.,1959 114 1407. NOVEMBER 1961 ~~~ ~ (110) and (lll) the (110) plane affords by far the greatest gain in crystal field stabilisation during the chemisorption of oxygen.If we make the approxima- tionl that for a clean (110) surface the four-fold co-ordination of Ni2+ is tetrahedral and if we assume3 that Dqtet = $Dqoct the gain in stabilisa- tion energy is 5.2 Dqoct per atom of oxygen ad- sorbed. If Dqoct for NiO is taken2 as 910 cm.-l it follows that the energy gain is 27 kcal. per mole of oxygen adsorbed on this plane. At low overall coverages of adsorbed oxygen such as are commonly encountered on nickel oxide at room temperature the differences in CFS energy may be sufficient to favour chemisorption predominantly on the (1 10) plane. We suggest therefore that the observed photo- desorption proceeds by an excitation of Ni2+ from Dunitz and Orgel J.Phys. and Chem. Solids 1957 3 20. the octahedral ground state 3A2ffto the excited 3T1ff state which may be regarded as the activated com- plex. Desorption of two adjacent oxygen atoms from the surface of a (110) plane then allows the d8ion to assume tetrahedral configuration in the 3T1ground state which is extremely stable with respect to the 3Tlsexcited state of octahedral ds. Sustained irradia- tion in the 3A2g-3T1g band thereby serves to maintain a new steady state with a lower coverage of adsorbed oxygen. The self-consistency of the interpretation favours the assumption implicit in the model that the posi- tive holes (Ni3+ ions) produced during oxygen adsorption are delocalised as indeed conductivity studies imply.The d7 case may in any event be treated similarly. (Received September 27th 1961.) Unsymmetrically Substituted Borazoles and Some Bicyclic Derivatives thereof By M. F. LAPPERT and M. K. MAJUMDAR (CHEMISTRY FACULTY UNXVERSITY DEPARTMENT OF TECHNOLOGY OF MANCHESTER) UNSYMMETRICALLY substituted borazoles are rare.l By making use of the steric factors that predominate in displacement reactions of three-co-ordinate boron compounds2 we have prepared some unsymmetrical B-amino- and B-alkoxy-borazoles. For example the tris(diethylamino)-triethylbor-azole (Ia) and ethylamine in equimolar proportions gave the ethylaminobis(diethyIamino)-compound (Ib) and diethylamine in nearly quantitative yield.That the product (Ib) is an unsymmetrical individual and not a mixture [(Ia) and (Ic)~] of its possible dis- proportionation rests on its physical constants molecular weight specific infrared spectrum and conversion into the bicyclic compound (11) by pyrolysis at 300" (94 hr.) with elimination of diethy lamine. I XI X2 x3 z X' a NEt NEt NEt Et b NHEt NEt NEt Et ZNOB\YZ c NHEt NHEt NHEt Et x28,N,Bx3 d NHEt NEt NEt H e NEt But0 But0 H z f But0 But0 But0 H g Bun0 NHEt NHEt Et In a similar way the unsymmetrical borazoles (Id e and g) were obtained by partial aminolysis or alcoholysis of the appropriate symmetrical borazolesp but their condensation products have not yet been fully investigated.An attempt to distil compound (Id) led to elimination of diethylamine and formation of a polymeric white solid. The di-t-butoxyborazole (Ie) was obtained in high Yield but mono-n-butoxY- borazole (Id only in low Yield since alcoholYsis of (Ic) gave also the fission product n-but yl bisethyl-aminoborate? BunO*B(NHEt)2. Further evidence for structure (Ie) was provided by the reaction with t-butyl alcohol which afforded tri-t-butoxyborazole (If) diethylamine and a trace of t-butYl borate. Et# $t $t YHEt ,B-N ,N-8 Et? 7-y' B-y-8 NEt Et 'y-f Et,N Et Et NU 4) Compound (11) is one of the very few bis borazoles to be characterised and is of a novel class.4 (Received August 14th 1961.) Sheldon and Smith Quart. Rev.,1960 14 200; Mikhailov Uspekhi Khim.1960,972; Smalley and Stafiej. J. Amer. Chem. SOC.,1959 81 582; Ryschkewitsch Harris and Sisler J. Amer. Chem. SOC.,1958 80,4515. Aubrey and Lappert Proc. Chem. Soc. 1960 148. Lappert Proc. Chem. SOC.,1959 59; Aubrey and Lappert J. 1959 2927. Laubengayer Moews and Porter,J. Amer. Chem. SOC.,1961,83,1337; Harris J. Org. Chem. 1961,26,2155. PROCEEDINGS Spin Decoupling and Relative Chemical Shifts of Dissimilar Nuclei Measured by Using Frequency Related Oscillators By J. A. GLASEL and D. W. TURNER L. M. JACKMAN (IMPERIAL ORGANICCHEMISTRY LABORATORIES, COLLEGE RESEARCH LONDON,S.W.7) To measure precisely the chemical shift between the The present communication reports an experi-resonant frequencies of two different spin-coupled ment such as has been outlined above for double nuclei by the observation of decoupling requires irradiation of spin-coupled nitrogen and hydrogen.either an accurate measurement of the individual A Varian 56.4 mc./sec. nuclear magnetic resonance frequencies of separate oscillators adjusted to bring spectrometer was the starting point and use was both nuclei to resonance simultaneously in a given made of the fact that the basic crystal oscillator of magnetic field or more simply the measurement of this unit operates at 56.4/4 = 14.1 mc./sec. A the ratio of their frequencies. 14.1 mc./sec. signal was extracted and divided by 45 in three steps locked transitron sinusoidal oscil- lators being used. The last division stage synchro- -nised a harmonic generator.The output of this AAAAA-8llcps 812 813 814 815 passed to a harmonic selector and a buffer stage and was modulated by an audio-oscillator (frequency -15cpS fMOD c.P.s.) in a suppressed carrier modulator. After final power amplification the sideband signals were directed to the Varian 56.4 mc./sec. probe. A visual display of each division and multiplica- 816 817 -818 819-820 tion factor by means of their Lissajous figures was provided on four small oscilloscope tubes. The lock- in stability of the unit was of the order of a day. The circuit itself will be described fully in another publicat ion. 821 822 823 824 825 TABLE Spectrum of the amine protons in an acid Simplest Final modulation fraction (kc./sec.) solution of methylamine as a function of audio- Nucleus frequency modulation when the exact nitrogen lH to 15/98 -10 (-1,OOO p.p.m.) resonance frequency is reached (-821 c.P.s.).The 2H 7/18 10 (-500 p.p.m.) pattern due to spin-spin splitting of the amine protons 'Li 9/28 1 (-40 p.p.m.) by the methyl group is clearly visible with approxi- IlB mately the expected 1:3 ;3:1 intensities. 1°B 3/28 10 (-2,000 p.p.m.) 13C 1 /4 100 (-10,OOO p.p.m.) 13/180 1 (-250 p.p.m.) If instead of constructing two separate oscillators 14N 16/17 10 (-200 p.p.m.) of high stability the decoupling RF signal could be l9F 13/32 30 (-1,500 p.p.m.) generated from the first by locked frequency division 31P or multiplication the frequency ratio would be deter- mined precisely and the requirement for high-An example of a decoupling experiment is shown frequency stability would be confined to one oscil- in the Figure.Here the effect of sweeping through the lator only. Since the actual ratios of the resonance nitrogen resonance upon the proton spectrum is frequencies of dissimilar nuclei do not however illustrated. In this case the modulation necessary correspond to exactly simple fractions interpolation was about 821 C.P.S. (the precision of the experiment by modulation at some low frequency will be is such that in a more careful measurement the required (Table). The oscillator providing the value was found to be 821.6 f0-3 c.P.s.). In con- modulation need then have only a relatively low trast for NH,+ in aqueous ammonium chloride stability in most cases about 0.1 % since this repre- the modulation was 875 C.P.S.Thus these experi- sents a variation of only 1 C.P.S. in 1 kc./sec. In ments present the possibility of relating the chemical practice audio-oscillators can have stabilities much shifts of all nuclei to a proton standard. higher than this. The ways in which these modulation values can be Note Added in Proof.-It has been brought to the authors' attention that a recent publication' emphasises the possibility of using spin decoupling to determine chemical shifts. Baldeschwieler and Randall Proc. Chem. Soc. 1961 p. 303. NOVEMBER 1961 related to nitrogen-14 chemical shifts will also be fully discussed in a later publication. One consequence of using this method is that if a field stabilising system controlled by nuclear resonance is used the stabilisation is satisfactory for any other nuclear resonance with an oscillator frequency related as above to the master oscillator (which need not be the proton resonance oscillator).One of us (J.A.G.) acknowledges the award of a Fellowship from the National Science Foundation 1961-1962. (Received September 12th 1961 .) A New Method for Deriving Molecular Wave Functions By D. M. HIRWand J. W. LINNETT (INORGANIC OXFORD CHEMISTRYLABORATORY UNIVERSFI-Y) FORtwenty-five years the two methods for construct- ing wave functions of polyatomic molecules have been the valence-bond (V.B.) and the molecular- orbital (M.O.) method. These have been improved in various ways (for example by employing resonance and configuration interaction respectively) but in their essential form they have changed very little.As a result of certain general considerations,’ we have been led to test another procedure which like the other two makes use of combinations of atomic orbitals. This method which we have called the “non-pairing” (N.P.) method has been applied to the .n-election systems of the ally1 positive ion radical ion and negative ion involving two three and four electrons respectively. In all cases the non-pairing method is found to be superior to the M.O. and the V.B. method in providing a basis for constructing the wave functions for the ground states. This note summarises the calculations for the positive ion. The V.B.method describes the ground state of the ion as a hybrid of one structure in which the two electrons are as a pair in one bond (I) and another in which the two electrons are as a pair in the other bond (11).The M.O. method assigns the electrons as + (I) CH2=CH-CH bH2-CH=CH (11) (111) CH,-CH-CH~ CH~LCH-ICH (IV) a pair to one bonding molecular orbital (III). The new method regards the structure as consisting of two one-electron bonds (IV);hence the term “non- pairing.” The antisymmetrised wave function for the singlet ground state consists of two determinants one in which the spins are disposed in one way and the other in which they are exchanged. For the one- electron bond between A and B the bond orbital 2prA+ k.2p.nBhas been used.In the simpler treat- ment k was put equal to one but in the later more accurate treatment it was adjusted to minimise the energy. The performance of the functions provided by the three methods has been compared with the best function obtainable by combining together all possible combinations of the 2p functions on the three atoms. Such “best” functions have been de duced for all three species.2 The derived energy has been used most for the comparison but in addition the value of the overlap integral [jt,h@dT where $ and @ are respectively the “best” function and the approximate one (V.B.,M.O. and N.P.)] has been used to judge the degree of similarity of the functions being tested to the “best” one. In the first place tests were made for functions of the three types in which no adjustable constants were allowed.It was found that the V.B. method with a Heitler-London function gave an energy 1-718 ev greater than that obtained by the “best” treatment; the M.O. method with Huckel orbitals gave for this 1.566ev; the N.P. method with k = 1 gave 0-700ev. When one adjustable constant was allowed the results were as follows :V.B.,with Coulson-Fischer pair-bond function^,^ 1.094 ev; M.O. with the coefficient for the orbital on the central atom vari- able 0.962 ev; the N.P. method with k adjustable but the same in both one electron bonds 0.195 ev. With the N.P. method it was found to be better to use the orbitals (2p.nA + k.2~71-Jtogether with (2~77~ in one pair of determinants along + k.2~~3 with (2p7rC + k.2p7rB) and (2p.nB + k.2~77,) in another pair with equal weight i.e.rather than a combination of just (2prA+ k.2p.nJ and (2p7rC+ k.2p.nd in one pair which gave an energy 0.358 ev greater than the “best”; in these formula the three carbon atoms are labelled A B and C (I3 being in the centre). It is seen that at both levels of approxi- Linnett Nature 1960 187 859; J. Amer. Chem. SOC.,1961 83 2643. Chalvet Compt. rend. 1952 234 2369; Ann. Chim. (France) 1954 9 97; Chalvet and Daudel J. Chim. Phys., 1952,49 629; Leflovits Fain and Matsen J. Chem. Phys. 1955,23 1690; Hunt Peterson and Simpson ibid. 1957 27 20; Higuchi ibid. 1957,26 151; 1957 27 825. Coulson and Fischer Phil. Mag. 1949 40 386. mation the N.P.function is better than those pro- duced by the M.O. and V.B. methods. The overlap integrals confirmed this conclusion. Using functions with one adjustable constant gives the energy figures for the negative ion as V.B. 1-300 ev M.O. 1.068 ev N.P. 0.408 ev; and those PROCEEDINGS for the radical M.O. 1.790 ev V.B. 1-065ev N.P. 0-010 ev. So in every case at this level of approxi- mation the non-pairing method produces an energy considerably closer to the "best" value than do the M.O. and the V.B. method. For each species the overlap integral is greatest for the N.P. function. (Received August 14th 1961 .) Polysulphuryl Fluorides By R. J. GILLESPIE and E. A. ROBINSON J. V. OUBRIDGE (DEPARTMENT OF CHEMISTRY MCMASTER ONTARIO,CANADA) UNIVERSITY HAMILTON DISULPHURYL FLUORIDES205F2,first prepared in 1951 by Hayek and Kollerl by reaction of sulphur trioxide with antimony pentafluoride has since been reported as a product of the reactions of several fluorides with sulphur trioxide2 and of the thermal decomposition of some fluorosulphates.3 Lehmann and Kolditz4 passed gaseous boron trifluoride into liquid sulphur trioxide and obtained a white solid mixed with an excess of sulphur trioxide; on adding 70% sulphuric acid to this mixture they obtained an upper layer of sulphuric acid and as lower layer a colourless oil (I) which on distillation gave a product boiling at 120"which they identified as trisulphuryl fluoride S,O,F,.Ill I I I 125.5318109 80 0 J23 c./sec.at 56.4 Mc./sec. We have found that the nuclear magnetic reson- ance spectrum of the oily lower layer (I) includes at least six fluorine resonances (cf. Figure). When a small quantity of disulphuryl fluoride (prepared by Hayek and Koller's method1) was added to the layer (I) the intensity of the line at highest field (S,) was increased and no new line appeared in the spectrum. Thus this line is due to disulphuryl fluoride. On distillation of the layer (I) at atmospheric pressure trisulphuryl fluoride S308F2was obtained boiling at 120" as reported by Lehmann and Kolditz and having a fluorine nuclear magnetic resonance spectrum consisting of a single line. When a small quantity of this material was added to the layer (I) the intensity of the line S3was increased and no new line appeared; this line is therefore due to trisulphuryl fluoride.It seemed reasonable to suppose that the remaining lines in the spectrum were due to higher polysulphuryl fluorides. The mixture (I) was dis- tilled under reduced pressure trisulphuryl fluoride was obtained at 30°/17 mm. and further fractions (II) and (III) were obtained boiling at 57"/6 mm. and 65"/1 m. respectively. The nuclear magnetic resonance spectrum of fraction (11) had a single peak with a chemical shift of 109 c./sec. from disulphuryi fluoride a small quantity of which was added to one sample of this fraction; this shift is identical with that of peak S in the spectrum of layer (I). The spectrum of fraction (111) also consisted of a single peak but with a chemical shift of 118 c./sec.from disulphuryl fluoride which is identical with the shift of peak S in the spectrum of the layer (I). We con- clude that fractions (11) and (111) are tetrasulphuryl fluoride S,01,F2 and pentasulphuryl fluoride S5014F2,respectively. It seems reasonable to suppose that peaks S6 and S are due to S6017F2and S7O2oF2 respectively although these compounds have not yet been obtained pure. The relative amounts of the individual polymers in a typical sample of layer (I) obtained from the areas under the various peaks were S20,F 25 % S308F2SO% 5x3 S*O,,F 16% S5014F2 S6°,7F2 3% and S,O20F2 1%. Hayek and Koller Monatsh. 1951 82 942. a Clark and Emeleus J. 1958 190; Schmidt Monatsh. 1954,85,452; Muetterties and Coffnian J.Amer. Chem. Soc. 1958 80 594. a Hayek Aigensberger and Engelbrecht Monatsh. 1955,86,735; Hayek Czaloun and Krismer ibid. 1956,87,741. Lehmann and Kolditz 2.anorg. Chem. 1953,272 73. NOVEMBER 1961 429 The compositions of tri- tetra- and penta-each of the polysulphuryl fluorides is consistent with sulphuryl fluoride were confirmed by the determina- their having the acyclic structures which have been tion of molecular weights by cryoscopic measure- proposed4 for di- and tri-sulphuryl fluoride. ments in fluorosulphuric acid by titration of the acid produced on hydrolysis and by determination of F\ ,o 00\....../O\s/' fluorine from the relative areas of the two peaks in ss the nuclear magnetic resonance spectra of mixtures db db d'"o of weighed amounts of one of the substances and disulphuryl fluoride (see Table).The fluorine nuclear magnetic resonance spectra The decreasing chemical shift between successive of the products of the reactions between sulphur tri- 6* B.p. Mol. wt. F (%I Equiv. wt.? Found Reqd. Found Reqd. Found Reqd. S2W2 0 51O 182 182 --91 S3°8F2 -80 120" 251 262 144 14.5 64.9 65.5 S*O,lF -109 57/6 mm. 321 342 10-4 10.5 56.7 57.0 S5°14F2 -118 65/1 IT~III. -422 52.5 9.0 52.5 52.8 S6°17F2 -123 7O20F2 -125.5 * Chemical shift from S,O,F (internal) in c./sec. at 56.4 Mc./sec. t By titration of the acid produced on hydrolysis SnOBn-lFz+ (n -1)H,O = 2HS0,F + (n -2)HaS04. peaks in the spectrum of layer (I) is consistent with oxide and antimony pentafluoride arsenic tri-their being due to polysulphuryl fluorides of in-fluoride potassium tetrafluoroborate and calcium creasing chain length.The large shift of 2.5 c./sec. fluoride have shown that polysulphuryl fluorides are between S6017F2 and S,02,F2 is a striking illustra- formed also in these cases. tion of the sensitivity of the chemical shift to slight changes in the electronic environment of the fluorine The research for this paper was supported by the nucleus. The single line observed in the spectrum of Defence Research Board of Canada. (Received,September 22nd 1961.) Sulphur Substitution Compounds of Amino-sugars.Part 1111 By WOLFGANG zu RECKENDORF A. BONNER MEYER and WILLIAM OF CHEMISTRY STANFORD STANFORD U.S.A.) (DEPARTMENT UNIVERSITY CALIFORNIA INcontinuing our syntheses of amino-sugars contain- (corr.) which with a slight excess of 0-IN-sodium ing a thiol group we have studied replacement by methoxide in absolute methanol under reflux for thiol of the 3-hydroxyl group of D-glucosamine.The 15 min. afforded a 70 % yield of the thiazoline (III) low reactivity of a secondary sulphonate rendered it m.p. 108-109" (corr.) (Found C 54.8;H 5.6; impossible to use Chapman and Owen's convenient N,4-2; S 17.85. C,,H,,N04S2 requires C 544; H method2 of displacing primary tosyl by thioacetyl 5.4;N,4-0; S 18.1%). This showed the character- groups. An alternative was utilisation of a neighbour- istic infrared C=N absorption at 6.2 p. The ing group in this instance a substituted 2-amino- presence of the thiazoline ring was proved by the group? Christensen and Goodman4 recently failed following reactions.Reduction with aluminium in such an attempt starting from a 3-amino-altroside. amalgam' afforded the crystalline thiazolidine (IV) By Crawhall and Elliott's method5 we obtained with liberation of methanethiol ;the thiazolidine was from the appropriate D-glucosamine derivative,6 the converted into methyl 2-amin0-4,6-benzylidene-Z crystalline but unstable dithiocarbamate (I) and deoxy-3-thio-~-alloside(V) mp. 175" (decomp.) by thence the methanesulphonate (11) m.p. 190-191 " treatment with aqueous mercuric ~hloride,~ followed Part TI Chem. Ber. in the press. a Chapman and Owen J, 1950 579. Winstein and Boschan J. Amer. Chem. SOC.,1950,72,4669. Christensen and Goodman J.Amer. Chem. SOC.,1960 82,4738. ti Crawhall and Elliott J. 1956 2071. Meyer zu Reckendorf and Bonner Chem. Ber. in the press. 'Cook,Hunter and Pollock J. 1950 1892. PROCEEDINGS 0-CH, cy PhHC/o-cH2 -~~ \o 0" YH YH (I) Mescs / (u) MeSG Fodor and &vos Chem. Ber. 1956,89 701. by decomposition of the resulting mercury salt with hydrogen sulphide and this product was readily <o~~ soluble in water gave the characteristic dour re- action with sodium nitroprusside and decolorised iodine solution. It has not yet been analysed because of scarcity of material. The main reason for the different behaviour of to be trans-diequatorial. In the same conformation maintained rigidly by the 4,6-benzylidene group the 2-hydroxyl and 3-amino-group of Christensen and Goodman's altrosamine are trans-diaxial.The authors are indebted to the U.S. Army Medical Research and Development Command (Contract DA-49-193-MD-2070) for its generous support of this investigation. (Received September 4th 1961.) The Synthesis and Stereochemistryof Mycaminose By A. C. RICHARDSON (THEUNIVERSITY, BRISTOL) MYCAMINOSE, from the acid hydrolysates of the anti- biotics magnamycin and magnamycin B,l is2 a 3,6-dideoxy-3-dimethylaminohexose. Woodward on the basis of differences in the pK values for various mycaminose derivatives assigned a relative stereo- chemistry to positions 2 3 and 4 which is com- patible only with the ~-altro-isomer.~ Comparison of the molecular rotations of the two isomeric mycami- nose tri-O-acetates1 (+6,020" and +30,100') with those of Is-and a-r>-altrose penta-acetates5 (-17,200" and +24,500' respectively) did not support the assignment of the D-altro-configuration.However the molecular rotations of the tri-O-acetates were in fair agreement with those of 3-acetamido-l,2,4-tri-O-acetyl-3,6-dideoxy-~-~-glucopyranose and its a-isomer6 (+7,450" and +36,800' respectively cal- culated from the L-forms). It was thus of interest to synthesise 3,6-dideoxy-3-d~ethylamino-~-glucose resulted in a 71 % yield of the dimethylamino- for comparison with mycaminose. glucoside (11). Acid hydrolysis afforded 3,6-dideoxy- Hochstein and Murai J. Amer. Chem. Soc. 1954 76 5080. Hochstein and Regna J Amer.Chem. SOC.,1955 77 3353. Woodward Angew. Chem. 1957,69 50. * Foster and Horton Adv. Carbohydrate Chem. 1959 14 231. Richtmyer and Hudson J. Amer. Chem. SOC.,1941 63 1727. Richardson Proc. Chem. Soc. 1961 255. Clark Gillespie and Weisshaus J. Amer. Chem. Soc.,1933 55 4576. Richardson unpublished xesults. Clark and his co-workers' have reported that selective dimethylation of primary amines can be achieved with formaldehydeformic acid and by use of this method dimethylamino-pyranosides have Me Me OH H bH (U) JHCI (I) HC[ (m) HO NMe2H H6H been prepared in good yield! In particular methyla- tion of methyl 3-amino-3,6-dideoxy- a-D-glucoside (I) prepared as described6 for the L-enantiomorph NOVEMBER 1961 3-dimethyla~no-~-~-glucose hydrochloride (111) which crystallised as a monohydrate m.p.116-1 18" [a]D + 11 -f + 31 O (H,O). Hochstein and Regna2 report m.p. 1 15-1 16O [a]D + 3 1O (H20 24 hr.) for mycaminose hydrochloride monohydrate. The iden- tity of the synthetic and the natural sugar was con-firmed by their infrared spectra and the m.p. 1 17" of a mixture. The L-enantiomorph of (I) similarly yielded L-mycaminose hydrochloride as a monohydrate 431 m.p. 116-117" [aID -13" -+ -30" (H,O). The infrared spectrum of this isomer was identical with that of the D-isomer. The author thanks Dr. L. Ho~ighfor helpful discussions Dr. I. 0. Sutherland for first bringing the formaldehyde-formic acid method of N-methyla- tion to his attention and Drs.F. A. Hochstein and P. P. Regna for a sample of mycaminose hydro- chloride. (Received September 7th 1961.) Solvent Effects on the Infrared Spectra of Inorganic Complex Hydrides By D. M. ~AMS (IMPERIAL CHEMICAL INDUSTRIES LIMITED ORGANIC CHEMICALS DIVISION, HEAVY AKERS LABORATORIES WELWYN, RESEARCH THEFRYTHE HERTS.) MANYmolecular vibration frequencies are sensitive iodides corresponding to (I) (IV) and (VII) have to the nature of the solvent and a considerable dv smaller by 20-30%. Compound (LII) is insoluble literature exists on solvent-sensitive vibration fre- CL quencies of organic molecules. Apart from a recent paper by Barraclough Lewis and Nyholml on metal carbonyls similar studies of inorganic com- plexes have not been reported.This note presents in summary solvent effects observed in the infrared H PEt3 spectra of several inorganic complex hydrides and indicates the value of the solvent-induced shift as a (11) CL method of studying the steric and electronic structures of inorganic complexes. The preparative properties of the compounds studied have been de~cribed~,~,~ and their structures and configurations determined by analysis dipole moments and other established methods. Each was examined in about ten solvents (depending upon solubility). In every case the lowest frequency of the metal-hydrogen stretching vibration (v~)was ob- served in hexane. Solvents of greater polarity shifted vm to higher frequency the shift being greatest for chloroform. The maximum frequency shift observed (dV)is Vchloroform-Vhexane.When the hydrogen atom is trans to a chlorine atom dv is 30 or greater but when it is trans to a trisubstituted phosphine or arsine VIrH is almost solvent-insensitive. In compound (IV) both types of behaviour are found and it is likely that the large shift (34 cm.-l) is associated with vI~H(~).The CL (V) PqTP Av (cm.-l) v, in hexane (cm.-l) 2183 2090 Insol. 2099 2032 * Estimated-see text. ? (CH2*PMe2),. Barraclough Lewis and Nyholm J. 1961,2582. Chatt Duncanson and Shaw Proc. Chem. Soc. 1957,343. Chatt and Shaw Chem. and Ind. 1960 931; 1961,290. Chatt and Hayter J. 1961 2605. in hexane butdv' =vchloroform-Qenzene =21 cm.-l. For each of the other complex hydrides dv' z 0.6 dv,which leads to an estimated dv of 35 cm.-l.From this limited number of examples general conclusions should not be drawn but the above results show a clear difference in behaviour between vMHfor a hydrogen atom trans to a halogen and vMH for one trans to a phosphine or arsine. The different magnitude of dv in the two cases may be explained in terms of the ability (or lack of it) of the atom or group trans to the hydrogen atom to attract d-electrons across the metal atom by dative r-bonding. A correspondence exists between dv and the PROCEEDINGS nuclear magnetic resonance chemical shift (0)of the hydridic proton but not between cr and the intensity or frequency of the vm band. The interaction between solvent and complex solute is unlikely to be due mainly to dielectric forces since plots of relative frequency shift against (E -1)/(2~+ 1) give no indication of a correlation.The rise in v, in polar solvents can be explained by postulating a large contribution to the electronic structure from wave functions involving ionic character as was done by Barraclough Lewis and Nyholm.' This also explains the relative positions of vm of the series of iridium complexes (LL)-(VI). (Received September 15th 1961 .) Dimeric Phosphinoborine Derivatives By W. GEE,R. A. SHAW and G. J. BULLEN B. C. SMJTH (BIRKBECK COLLEGE UNIVERSITY OF LONDON, W.C.1) FOLLOWING the discovery of the dimethylphos-phinoborinesl several trimeric tetrameric and poly- meric phosphinoborine derivatives were prepared and monomeric compounds have also been rep~rted.~ The first examples of dimeric phosphinoborines are now described.Diphenylphosphine and boron tribromide in pentane formed a colourless crystalline addition compound Ph,PH,BBr, map. 133-1 34" (Found C 32.8; H 2-6; B 2-1; Br 54.6%; M by ebullio- metry in benzene 439. Cl,Hl,BBr,P requires C 33.0; H 2-5; B 2.5; Br 54.9%; M 437). Its solution in benzene on treatment with one equivalent of tri- ethylamine gave a quantitative precipitate of tri- ethylamine hydrobromide. Evaporation of the filtrate and recrystallisation from benzene gave colourless laths of a BB-dibromo-PP-diphenylphosphinoborine m.p. 183-184" (Found C 40.9; H 3.0; B 2-75; Br 435; P 9.0. C1,HloBBr,P requires C 40.5; H 2.8; B 3.0; Br 44.9; P 8.7%).Similarly a colourless crystalline addition com- pound Ph,PH,BI, decomp. 137-139" (Found I 66.1. C12Hl,B13P requires I 65.9 %) and colourless crystals of a BB-di-iodo-PP-diphenylphosphino-borine m.p. 194-195" (decomp.) (Found C 31.7; H 2.5; 13 2.6; I 55.8; P 7.3. C12HloBI,P requires C 32.0; H 2.2; B 2.4; I 56.4; P 6-9%) were prepared from diphenylphosphine and boron tri- iodide. The two phosphinoborine derivatives are iso-morphous. They are monoclinic and the unit-cell dimensions and densities (measured by flotation) are bromide a = 12.16 rt 0.05,b = 13.45 f0.1 c = 17.4 f0.1 A /3 = 111*0"& 0.5" Dm = 1.76; iodide a = 12.26 f0.05,b = 13.80 f0-05 c = 17.8 f0.1 A /3 = 110.4" f0+5",Dm = 2.13. Systematic absences of X-ray reflexions are (hOl) when I is odd and (OkO) when k is odd.The space group is therefore P2Jc. In both cases the crystals examined were twinned on (100). This is not easily recognisable because the angle between [401] and [Ool] is very close to 90" (certainly within 1") so that the crystals at first sight appear single with a larger unit cell. The unit cells specified by the parameters given above contain eight Ph,P.BBr or Ph,P-BI units (calculated 7.91 4 0.14 for the bromide and 8.04 & 0.1 1 for the iodide). The minimum number of molecules placed in general positions in the unit cell required to satisfy the symmetry P2Jc is four dimeric units but the possibility of two centrosymmetric tetramers (placed in special positions) or eight monomers cannot be rejected.Trimeric molecules are definitely excluded. Ebullioscopic measurements show that the mole- cules are dimeric in a number of different solvents over a wide concentration range :M,for (Ph,P.BBr,) (in benzene toluene carbon tetrachloride respec- tively) 740 738 677 (calc. 712); for (Ph2P.BIJ2 862 and 929 (in benzene) 962 (in carbon tetrachloride) (calc. 900). (Received September 224 1 96 1 .) Burg and Wagner J. Amer. Chem. SOC.,1953 75 3872. a Burg and Brendel J. Amer. Chem. SOC.,1958,80 3198; Burg J. Znorg. Nuclear Chem. 1959,11,258; Wagner and Caserio ibid.,p. 259. * Gates and Livingstone J.,1961 1O00. NOVEMBER 1961 433 FORTHCOMING SCIENTIFIC MEETINGS London Thiircclav-Decemher 14th-at 7.30 n.m-Liversidge Lecture “Stereospecific Polymerisation,” by Professor C.E. H. Bawn C.B.E. Ph.D. F.R.S. To be given in the Lecture Theatre The Royal tnstitution Albemarle Street W.l. Aberdeen Friday December 8th at 8 p.m. Lecture “Inorganic Heterocycles,” by Dr. N. L. Paddock B.A. Joint Meeting with the Royal Insti- tute of Chemistry and the Society of Chemical Industry to be held at Marischal College. Aberystwyth Thursday December 7th at 5 p.m. Lecture “Microcalorimetry and the Thermogenesis of Living Species,” by Dr. H. A. Skinner B.A. Joint Meeting with the University College of Wales Chemical Society to be held in the Edward Davies Chemical Laboratories. Birmingham Friday December lst at 4.30 p.m. Lecture “New Reactions in Dinitrogen Tetroxide,” by Professor C.C. Addison D.Sc. F.R.I.C. Joint Meeting with the University Chemical Society to be held in the Chemistry Department The University. %&to1 Thursday December 7th at 6.30 p.m. Lecture “Mining Smelting and Refining of Nickel,” by Dr. G. L. J. Bailey. Joint Meeting with the Royal Institute of Chemistry the Society of Chemical Industry and the Institute of Metals to be held in the Department of Chemistry The University. Cambridge (Meetings will be held in the University Chemical Laboratory.) Friday December lst at 8.30 p.m. Lecture “Biphenylene and Related Compounds,” by Professor W. Baker D.Sc. F.R.S. Joint Meeting with the University Chemical Society. Monday December 4th at 5 p.m.Lecture “Some Hydrogen Transfer Reactions,” by Professor H. B. Henbest Ph.D. F.R.I.C. Cardiff Monday December 4th at 5 p.m. Lecture “Transfer Reactions of Oxygenated Radi- cals,” by Professor A. F. Trotman-Dickenson Ph.D. To be given in the Department of Chemistry University College Cathays Park. Durham Monday December 11 th at 5 p.m. Lecture “Seeing Molecules with Microwaves,” by Dr. J. Sheridan M.A. Joint Meeting with the Durham Colleges Chemical Society to be held in the Science Laboratories The University. Glasgow Friday December 8th at 7.15 p.m. Lecture “The Structure of Natural Products by Direct X-Ray Analysis,” by Professor J. Monteath Robertson D.Sc. F.R.I.C. F.R.S. Joint Meeting with the Royal Institute of Chemistry the Society of Chemical Industry and the Society for Analytical Chemistry to be held in the Royal College of Science and Technology.Leicester Tuesday December 5th at 4.30 p.m. Lecture “Some Recent Observations on the Activity of Metal Catalysts,” by Professor C. Kemball M.A. Ph.D. F.R.I.C. Joint Meeting with the University Chemical Society to be held in the University. Manchester Thursday December 7th at 6.30 p.m. Official Meeting and Lecture “The Structure of Proteins,” by Dr. M. F. Perutz F.R.S. To be held in Room F1 Manchester College of Science and Technology. Newcastle upon Tyne Friday December lst at 5.30 p.m. Bedson Club Lecture “The Gibberellins a New Group of Plant Hormones,” by Dr. P. W.Brain F.R.S. To be given in The Chemistry Department King’s College.NEWS AND ANNOUNCEMENTS Library.-The Library will close for the Christmas holiday from 7.30 p.m. on Friday December 22nd until 9.30 a.m. on Thursday December 28th 1961. Election of New FeUows.46 Candidates whose names were published in Proceedings for September have been elected to the Fellowship. New Members of D.S.I.R. Research Council.-The Minister for Science Lord Hailsham has appointed five new members of the Council for Scientific and Industrial Research as from October lst 1961. They are Mr. L. H. Bedford Director of Engineering Guided Weapons Division English Electric Aviation Ltd. Mr. G. B. R. Feilden Managing Director Hawker Siddeley Brush Turbines Ltd. Professor E.R. H. Jones Waynflete Professor of Chemistry Oxford University Prqfessor 0.A. Saunders Pro-fessor of Mechanical Engineering Imperial College London and Mr. H. C. Tett Chairman and Manag- ing Director Esso Petroleum Co. Ltd. The former Chairman Sir Harry Jephcott Professor C. E. H. Bawn and Sir Willis Jackson who completed their five year term of office retired on September 30th. A fourth member Sir WaEter Drummond who was also due to retire has been re-appointed for a period of 3 months. The present constitution of the Research Council is Sir Harold Roxbee Cox (Chairman) Mr. L. H. Bedford Professor B. Bleaney Professor C. F. Curter Dr. J. W. Cook Mr. Frank Cousins Sir Walter Drumrnond Mr. G. B. R. Feilden Professor E. R. I€. Jones Vice-Admiral Sir Frank Mason Professor 0.A.Saunders Dr. C. J. Smithells Mr. H. C. Tett Mr. L. T. Wright with Sir Harry Melville as Secretary. Meldola Medal for 1961.-The Meldola Medal is the gift of the Society of Maccabaeans and is normally awarded annually. The next award will be made early in 1962 to the chemist who being a British subject and under 30 years of age at December 31st 1961 shows the most promise as indicated by his or her published chemical work brought to the notice of the Council of the Royal Institute of Chemistry before December 31st 1961. No restrictions are placed upon the kind of chem- ical work or the place in which it is conducted. The merits of the work may be brought to the notice of the Council either by persons who desire to recom- mend the candidate or by the candidate himself by letter addressed to The President The Royal Insti- tute of Chemistry 30 Russell Square London W.C.1 the envelope being marked “Meldola Medal”. The letter should be accompanied by six copies of a short statement on the candidate’s career (date of birth education and experience degrees and other qualifications special awards etc. with dates) and of a list of titles with references of papers or other works published by the candidate independently or jointly. Candidates are also advised to forward one reprint of each published paper of which copies are available. Directory of Consultants.-The Royal Institute of Chemistry has issued the fourth edition (1961) of its Directory of Independent Consultants in Chemistry and Related Subjects.The general plan of the Directory is similar to that adopted in earlier edi- tions but the basis of the subject guide has been revised to take account of the increased number and variety of fields of specialisation. The cover in the PROCEEDINGS Directory extends from chemistry to its borderlands with physics and engineering and with the biological sciences. Biochemical Journal.-The Editorial Office of the Biochemical Journal has moved from Oxford Street to 20 Park Crescent Regent’s Park London W. 1. Symposium.-The Second International Sympos- ium on Fluorine Chemistry will be held at Estes Park Colorado U.S.A. on July 17-20th 1962. Enquiries should be addressed to Dr.0. R. Pierce of the Dow Corning Corporation Midland Michigan U.S.A. Glassware and Glassblowing.-Fellows may be interested to know that two 16 m. instructional films for students “Volumetric Glassware” and “Elementary Glassblowing,” have been made by the University of Leeds in association with the Yorkshire Film Company. The films may be bought or hired from the Yorkshire Film Company 12 Queen Street Huddersfield. Deaths.-We much regret to announce the death of Mr. Stanley E. Carr who retired as General Secretary of the Society in 1945 after having served as Assistant Secretary and then General Secretary for 43 years. We also regret to announce the deaths of Dr. G. Barr (13.8.61) formerly of the National Physical Laboratory Teddington; Mr.P. J. Burgess (11.10.61) a Fellow for more than 60 years; Mr. Lewis Eynon (18.10.61) Founder and Senior Partner of Eynon & Lane Consulting and Analytical Chem- ists London E.C.3; and Mr. J. H. Haynes (21.9.61) formerly Senior Chemistry Master at Fairfield Grammar School Bristol. Personal.-The following appointments have been made at Westfield College University of London Dr. B. J. Aylett Lecturer in Inorganic Chemistry; Dr. J. J. Throssell Lecturer in Physical Chemistry; Dr. P. M. Scopes Assistant Lecturer in Organic Chemistry; and Dr. A. Walton Assistant Lecturer in Physical and Inorganic Chemistry. Dr. W. Blakey Joint Managing Director of British Industrial Plastics Limited has been ap-pointed Deputy Chairman of the Company.Professor E.J. Bourne and Mr. P. T. Clothier have been appointed official visitors by D.S.I.R. to the British Leather Manufacturers’ Research Associa- tion in place of Professor R.M. Barrer and Mr. H. W. Cremer who have completed their terms of office. Dr. J. A. Cade formerly of A.E.R.E. Harwell has been appointed Principal Scientific Officer at the CENT0 Institute of Nuclear Science Teheran Iran. Mr. J. S. Clarke formerly Chemistry Master at Lancaster Royal Grammar School has taken up the appointment of Senior Chemistry Master at Alleyn’s School. NOVEMBER 1961 Dr. J. W.Clark-Lewis Reader in Organic Chem- istry in the University of Adelaide has been awarded the degree of D.Sc. by the University of London for his work in the field of Organic Chemistry.Dr. R. GZyn Davies has been appointed Leverhulme Professor of Geology at the University of the Punjab Lahore West Pakistan. Dr. J. Ferguson Research and Development Director is retiring from the Main Board of Im- perial Chemical Industries Limited on November 30th after 33 years’ service with the Company. Mr. T. C. Gallant has recently joined J. & J. Colman Ltd. Norwich as Head of Packaging Research. Dr. D. E. Hathay formerly of the British Leather Manufacturers’ Research Association has taken up an appointment as Head of the Division of Biochem-istry at Tunstall Laboratory (Toxicology) Shell Research Ltd. Sittingbourne. Professor E. R. H. Jones has been awarded the Fritzsche Award for 1962 by the American Chemical Society.It has been made for Professor Jones’ work in the field of terpenoid chemistry and is to be pre- sented at the Spring Meeting of the American Chemical Society in Washington in May 1962. Dr.D. N. Kirk has been appointed to a Lecture- ship in Chemistry at the University of Canterbury New Zealand. 43s Mr. P.J. March has been appointed to the Board of Styrene Co-Polymers Ltd. in succession to Mr. N. A. Ilif who has resigned. Dr. N. J. McCorkindale has been appointed Lecturer in Chemistry at the University of Glasgow. The title of Professor of Organic Chemistry in the University of London has been conferred on Dr. L. N. Owen in respect of his post at the Imper- ial College of Science and Technology.Dr. E. W. Randell has taken up a post as Lecturer at Queen Mary College London on completion of a Research Fellowship at Harvard University. Mr. C. E. J. Reynolds formerly Assistant Research Manager has been appointed Research Manager of Vinyl Products Ltd. Carshalton Surrey. Mr. W.V. Sherman has been awarded a Fulbright Travelling Scholarship and will be spending one year as post-doctoral research associate at Brandeis University Waltham Mass. Dr. G. Tesi formerly of the U.S. Naval Ordnance Laboratory has been appointed Assistant to the Associate Director for Research at the Naval Propel- lant Plant Indian Head Maryland. The title of Reader in Chemistry in the Univer- sity of London has been conferred on Dr .C.A. Vernon in respect of his post at University College.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.) Addy. John Keith B.Sc. Ph.D. A.R.I.C. Lynmouth 8 Shay House Lane Stocksbridge Sheffield. Ager Margaret Elizabeth B.A. Department of Biochem- istry South Parks Road Oxford. Albon Colin Paul. 44-D The Gardens London S.E.22. Baldwin Jack Edward B.Sc.,A.R.C.S. Department of Chemistry Imperial College South Kensington S.W.7. Barker John Morris B.Sc. 54 Lisbon Avenue Twicken- ham Middlesex.Bass Kenneth Cyril B.Sc.,Ph.D. A.R.I.C. 42 Hornsey Road London N.7. Batty Walter Edward M.A. B.Sc. 68 The Avenue West Wickham Kent. Baum Edward Joseph B.Sc. 3372 Utah Street Riverside California U.S.A. Bhattacharya Arun Kumar M.Sc. Ph.D. Department of Chemistry University of Saugar Sangor M.P.,India. Boi t Arthur G. B.Sc. A.R. A.C.I. Department of Pharmacy Chelsea College of Science Manresa Road London S.W.3. Boston David Joseph. 14 Mentmore Road St. Albans Herts. Brignell Peter John B.Sc. 282 Horns Road Ilford Es3ex. Brown Peter B.Sc. 54 Milbury Crescent Bitterne Southampton. Brown Robert James. 18 Pearl Road Walthamstow E.17. Buckle Ernest Roy Ph.D. F.R.I.C. Department of Chemical Engineering and Chemical Technology, Imperial College London S.W.7.Carter Richard Powell Jr. B.S. Department of Chem-istry Carnegie Institute of Technology Pittsburgh 13, Pa. U.S.A. Carper William Robert B.S. Box 2616 University of Mississippi Mississippi U.S.A. Chadbourne David John B.Pharm. 167 Lichfield Lane Berry Hill Mansfield Nottingham. Chaloner Robert Eyton B.A. 13 Windsor Avenue Grays Essex. Chang Raymond. Woolwich Polytechnic London, S.E.18. Coats Anthony William B.Sc. 81 Wakeley Road Rainham Gillingham Kent. Cook David Branston. 18 Windsor Road Crowle, Scunthorpe Lincs. Crawford John B.Sc. 15 Cowpen Lane Billingham Co. Durham. Cresswell Michael Alan. Rayfield Ray Park Avenue Maidenhead Berks. Crum John Kistler B.S. 1613-112 North Congress Avenue Austin 1 Texas U.S.A.Davidson Charles Martin B.Sc. Department of Chem-istry Queen’s College Dundee Scotland. Davies Alun Rosser B.Sc. Department of Chemistry University College of Swansea Glamorgan. De Vries Juan X. Dr. Chem. Whiffen Laboratories Department of Chemistry Imperial College London s.w.7. Dobson Geoffrey Robert B.Sc. A.R.C.S. Department of Chemistry lmperial College London S.W.7. El-Bayoumi Mohamed Ashraf M.Sc. Ph.D. Depart- ment of Biology Massachusetts Institute of Tech-nology Cambridge 39 Massachusetts U.S.A. Fallows Walter Betton. 68 Roslyn Road Davenport Stockport Cheshire. Faulkner David John. 36 Lincoln Avenue Bourne- mouth Hants. Fewell Anthony. “Dartmouth Villa,” Wycombe Lane Wooburn Green Nr.High Wycombe Bucks. Fletcher Rodney. 12 Churchfields Road Frodsham Via Warrington Lancs. Franks Felix B.Sc. Ph.D. F.R.I.C. Department of Chemical Technology Institute of Technology, Bradford 7. Friesen Reginald Jacob M.Sc. London House Guilford Street London W.C.l. Furlani Claudio Dr. Chem. Via Marangoni 1 Rome 1taly. Goldsmith John David. Worcester College Oxford. Gordon Ian Reginald B.A. Magdalen College Oxford. Greenhalgh Colin William B.Sc. Ph.D. 23 Sefton Drive Worsley Manchester. Hafner Klaus. Dr. Dozent. Marburg/Lahn Bahnof- strasse 7 (Chemical Institut) Germany. Hales Leonard Austin Winser B.Sc. 171-A Hill Farm Estate Salmon Street Kingsbury N.W.9. Hall Reginald David B.A. Merton College Oxford. Harper Robert James B.Sc.“Harryville,” Ballynahinch Co. Down Northern Ireland. Heard David Dunn B.Pharm. 136 Roman Road, Birstall Leicester. Hickey Bernard Daniel. 277 North Woolwich Road London E. 16. Hillier Ian Harvey B.Sc. A.R.C.S. Department of Chemistry Imperial College London S.W.7. Hillman Robin Alastair Hubert. 9 Westland View Luston Leominster Herefordshire. Hills Kenneth B.Sc. A.R.C.S. 60 Hawthorn Road Hornsey N.8. Jarman Michael B.A. 86 West Wycombe Road High Wycombe Bucks. Jensen Wesley Eugene B.A. 322 Hagley Road, Edgbaston Birmingham 17. Johnston Brian Tisdall B.Sc. Newtownbutler Co. Fermanagh Northern Ireland. Jones Elwyn Herbert B.Sc. 43 Bruce Grove Shotgate Wickford Essex. Jones Neville Murray B.Sc. Westfield College Hamp- stead N.W.3.Jones Richard B.Sc. 43 Water Reserve Road North Balgowlam New South Wales Australia. Jurinski. Neil B. B.S. Box 2616 University of Mississippi, Mississippi U.S.A. Kan Robert O. B.A. M.S. Ph.D. Department of Chemistry Frick Laboratory Princeton University Princeton N.J. U.S.A. Kime David Edward. 6 Castleton Crescent Skegness Lincs. Kirman. John B.Sc. 7 Birch Grove Rusholme Man- Chester 14. ’ Klett Donald S. M.S. Department of Chemistry Uni- versity of Texas Austin 12 Texas U.S.A. PROCEEDINGS Lee Brian Edward B.Sc. 16 Woodland View Thornton Liverpool 23. Lee Brian Nicholls B.Sc. 56 Whitehorse Lane South Norwood S.E.25. Low Beng See,B.Sc. Department of Organic Chemistry University of Adelaide Adelaide South Australia.Mallick Subhash Kumar B.Sc. B.Pharm. 49 Macaulay Court Macaulay Road London S.W.4. Marsh Kenneth Neil B.Sc. Department of Physical and Inorganic Chemistry University of New England, Armidale New South Wales Australia. Miki Takuichi Dr.Pharm. Research Laboratories, Takeda Chemical Industries Limited Osaka Japan. Millington James Peter. Chemistry Department Gordon Hall Queen’s University Kingston Ontario Canada. Mitchell Stephen. “Crud-Y-Gan,” The Handfords, Derrington Nr. Stafford. Mitchell William Norman B.S. Department of Chem- istry University of California Riverside California U.S.A. Mryao Kohei Ph.D. Eisai Research Laboratory, Takehayacho Bunkyoku Tokyo Japan. Mode Vincent Alan Jr. 622 E. Main Walla Walla Washington U.S.A.Moore David James B.Sc. 47 Corinne Road London N.19. Morgan Paul Felton. 13 Upper North Road Bargoed Glamorgan. Moseley William David Jr. A.B. Fulmer Hall Washing- ton State University Pullman Washington U.S.A. Moss Ernest Kent B.S. Department of Chemistry, University of Arizona Tucson Arizona U.S.A. Page Clifford Barrie. 65 Rhydhelig Avenue Heath Cardiff. Page Jane Muriel Julian B.Sc. A.R.C.S. East Lodge Betchetts Green South Holmwood Dorking Surrey. Patel Amrutlal Nathubhai B.Sc. 86 Percy Road, London E. 16. Paul Ian A.R.C.S.T. c/o Keeley Benachie Belmont Avenue Bangor North Wales. Pavia Donald Lee. 4214 S.E. 28th Place Portland 2 Oregon U.S.A. Payne John Weston. 7 Kenwardley Road Willerby Hull Yorks. Pews Richard Garth B.Sc.Apartment 9 295 St. George Street London Ontario Canada. Pickles Allan. 207 Cop Lane Penwortham Preston Lancs. Poole Teresa Maria B.Sc. 103 Walsworth Road, Hitchin Herts. Powell David Guy. B.Sc. 8 Park Road West Curzon Park Chester Cheshire. Powers George Arthur B.Sc. Department of Chemistry Birmingham University Edgbaston Birmingham 15. Putman Ernest John B.Sc. 47-B Castle Gate Notting- ham. Richards Jeffrey B.Sc. 4 Chaucer Street Castleton Rochdale Lancs. Rosswog Werner. Freiburg/Brg. Eschholzstrasse 82 West Germany. Rothwell Barry Stephenson. 167 Stubbins Lane, Ramsbottom Via Bury Lancs. Rule Laurence B.Sc. 66 Enys Road Camborne Cornwall. Sastri Kodavanti Mallayya M.Pharm. Andhra Univer- sity Waltair India.Schulman Jerome M. M.A. Department of Chemistry, Columbia University Havemeyer Building New York U S.A. Scregg Gerald B.A. Wadham College Oxford. Sheppard Brian. 29 Manor Rise Bearsted Maidstone Kent. NOVEMBER 1961 Speight James Glassford B.Sc. 32 Longford Road Chorlton-cum-Hardy Manchester 21. Spence David Hugh B.Sc. 31 Richmond Hill Road, Edgbaston Birmingham 15. Spillett Robert Ernest B.Sc. Department of Chemistry University of Leicester University Road Leicester. Srinivasan &manujam M-s~.Department of Physics, University College of North Staffordshire Ke& North Staffordshire. Stevenson ROY B.Sc. National Chemical Research Laboratory P.O. Box 395 Pretoria South Africa. Stone Thomas John B.A. 240-B Iffley Road Oxford.Tahk Frederick Christopher S.B. Brickett Hill Road RFD No. 4 Concord New Hampshire U.S.A. Treherne Bryan Leslie. 11 Manor Gardens Larkhall Rise Clapham S.W.4. ADDITIONS TO The molecular basis of evolution. C. B. Anifinsen. Pp. 228. John Wiley & Son. New York. 1961. New thinking in school chemistry report on the OEEC seminar on the status and development of school chemistry. 1960. Pp. 215. OEEC. Greystones Ireland. 1961. (Presented by the publisher.) Rontgenographische Chemie. E. Brandenberger and W. Epprecht. 2nd edn. Pp. 272. Birkhauser. Basle. 1961. States of matter. E. A. Moelwyn-Hughes. Pp. 91. Oliver and Boyd. Edinburgh. 1961. (Presented by the publisher.) Atomic energy waste its nature use and disposal. Edited by E.Glueckauf. Pp. 420. Interscience. New York. 1961. Physical chemistry. E. A. Moelwyn-Hughes. 2nd edn. Pp. 1333. Pergamon Press. Oxford. 1961. Progress in reaction kinetics. Edited by G. Porter. Vol. 1. Pp. 276. Pergamon Press. Oxford. 1961. Gas chromatography. E. Bayer. Pp. 240. Elsevier. Amsterdam. 1961. Reference electrodes theory and practice. Edited by D. J. G. Ives and G. J. Janz. Pp. 651. Academic Press. New York. 1961. lnfrared absorption of inorganic substances. K. E. Lawson. Pp.200. Reinhold. New York. 1961. Separation of heavy metals. A. K. De. Pp. 308. Pergamon Press. Oxford. 1961. (Presented by the publisher.) Rare metals handbook. Edited by C. A. Hampel. 2nd edn. Pp.715. Chapman and Hall. London. 1961. Rare earth alloys.K. A. Gschneidner. Pp. 449. Van Nostrand. London. 1961. 7 he identification of organic compounds. S. Veibel. 5th edn. Pp. 426. Gad. Copenhagen. 1961. Rules and methods for calculating the physico-chemical properties of paraffinic hydrocarbons. V. M. Tatevskki V. A. Benderskii and S. S. Yarovoi. (Trans- lated from the Russian by M. F. Mullins.) Pp. 128. Pergamon Press. Oxford. 1961. Organic peroxides. E. G. E. Hawkins. Pp.434. Spon. London. 1961. (Presented by the publisher.) Organic peroxides. A. G. Davies. Pp. 215. Butter-worths. London. 1961. (Presented by the author.) Organic sulfur compounds. Edited by N. Kharasch. Vol. 1. Pp.623. Pergamon Press. Oxford. 1961. Heterocyclic compounds. Edited by R. C. Elderfield. Vol. 7. Pp.878. John Wiley and Son. New York. 1961. Benzoles production and uses. Edited by G. Claxton. Pp. 979. National Benzole and Allied Products Associa- tion. London. 1961. Ware Michael John B.A. Jesus College Oxford. Wei Min-Min B.Sc. Department of Chemistry Univer- sity of British Columbia Vancouver B.C. Canada. Whan David Alexander B.Sc. 8 Stranmillis Road Belfast 9 Northern Ireland. WhateleY Tony Louis B-A. 22 Waltonwell Road Oxford- Williams Hugh Jeffrey B.A. Department of Chemistry, The Polytechnic Regent Street London W. 1. Williams John Roderick B.Sc. 22 Herdsman Parade Wembley Perth Western Australia. Wilson Keith Victor B.A. Department of Chemistry, University College of North Staffordshire Keele North Staffordshire. Young David Edward Michael.Hertford College, Oxford. THE LIBRARY Viscoelastic properties of polymers. J. D. Ferry. -Pp. 482. John Wiley and Son. New York. 1961. Die Schwfelsaurefabrikation. B. Waeser. Pp. 480. Vieweg und Sohn. Braunschweig. 1961 A textbook of quantitative inorganic analysis including elementary instrumental analysis. A. I. Vogel. 3rd edn. Pp. 1216. Longmans. London. 1961. (Presented by the author.) Trace elements in plants. W. Stiles. 3rd edn. Pp. 249. University Press. Cambridge. 1961. The chemistry and mode of action of herbicides. A. S. Crafts. Pp. 269. Interscience. New York. 1961. Methods for the analysis of non-soapy detergent (NSD) products. G. F. Longman and J. Hilton. (S.A.C. Monograph No. 1.) Society for Analytical Chemistry.London. 1961. (Presented by the publisher.) The optimal design of chemical reactors a study in dynamic programming. R. Ark. Pp. 191. Academic Press. New York. 1961. Industrial water treatment practice. Edited by P. Hamer J. Jackson and E. F. Thurston. Pp. 514. Butter-worths. London. 1961. Paint technology manuals. Part 1. Non-convertible coatings. Edited by I. C. R. Bews. Pp. 326. Chapman and Hall. London. 1961. Printing ink manual ; commissioned by the Technical Training Board of the Society of British Printing Ink Manufacturers. Edited by R. F. Bowles. Pp. 746. Heffer. Cambridge. 1961. Proceedings of the symposium on chemical process hazards with special reference to plant design held in Manchester 1960. Edited by J. M. Pirie.Pp. 117. Institution of Chemical Engineers. London. 1961. Symposium on quinones in electron transport held in London 1960. Edited by G. E. W. Wolstenholme and C. M. O’Connor. Pp. 453. Churchill. London. 1961. Protein biosynthesis a symposium held at Wassenaar 1960 under the auspices of UNESCO and the Council for International Organisations of Medical Sciences. Edited by R. C. Harris. Pp.409. Academic Press. London. 1961. The McCollum-Pratt Institute. Symposium. Baltimore. 1960. A symposium on light and life sponsored by the McCollum-Pratt Institute of the Johns Hopkins Univer- sity. Edited by William D. McElroy and Bentley Glass. Held at The Johns Hopkins University 1960. Pp. 924. Johns Hopkins Press. Baltimore. 1961. NMR and EPR spectroscopy papers presented at Varian’s third Annual Workshop on Nuclear Magnetic Resonance and Electron Paramagnetic Resonance held at Palo Alto California by the NMR-EPR staff of Varian Associates.Pp.288. Pergamon Press. Oxford. 1960. Proceedings of the International symposium on distil- lation organised by the Institution of Chemical Engineers held at Brighton 1960. Edited by P. A. Rottenburg. (European Federation of Chemical Engineering 24th Meeting.)InstitutionofChemicalEngineers.London.1961. PROCEEDINGS Separation processes in practice a collection of papers originally presented in Philadelphia 1960 under the auspices of the Philadelphia-Wilmington Section of the American Institute of Chemical Engineers and the School of Chemical Engineering University of Pennsylvania.Fdjted by Robert F. Chapman. Pp.209. Reinhold. New York. 1961. ELECTRON-PAIR REPULSIONS A MECHANICAL ANALOGY By H. R. JONES and R. B. BENTLEY (CENTRAL OF FURTHER CARLETT WIRRAL) COLLEGE EDUCATION PARK EASTHAM CONSIDERABLE progress has been made in recent years in the explanation of the shapes of simple covalently bonded molecules and ions in terms of an essentially electrostatic interaction between electron- pairs in the valency shell of the central atom. This concept originated with Sidgwick and Powell) has been discussed at length by Gillespie and NyholmS and recently elaborated by Gillespie? Much of the success of this concept lies in the assumption that repulsions between electron-pairs diminish in the order :-lone pair-lone pair > lone pair-bond pair > bond pair-bond pair.Where multiple bonds are concerned it is assumed that repulsions diminish in the order:- multiple bond-multiple bond > multiple bond-single bond > single bond-single bond. Pro-ceeding from these assumptions a very satisfactory though essentially qualitative explanation of the shapes of many molecules and ions is available. There are some cases however where it is not easy to decide between the relative stabilities of two differ- ent shapes. This is to some extent the case with the molecule of chlorine trifluoride and more particu- larly with that of iodine heptafluoride. We have found that this electrostatic effect can be simulated mechanically by the use of the so-called modelling balloons sold in toy shops.* These balloons elongated in shape are treated in such a way that they may be twisted after inflation (Fig.1) FIG. 1 and then intertwined with each other at the narrow necks thus formed. Two balloons thus twisted form four lobes directed to the corners of a regular tetrahedron (Plate 1). Three balloons form six lobes directed to the corners of a regular octahedron (Plate 2). The balloons take up these positions spontane- ously because of mechanical repulsion and the analogy with the behaviour of electron-pairs is clear. FIG.2 When it is desired to simulate the arrangement of an odd number of electron pairs it is only necessary to twist an extra balloon by its neck or tail into an assembly containing an even number of lobes.In this way we find three lobes arranging themselves trigonally (Fig. 2) five lobes directed to the corners of a trigonal bipyramid (Fig. 3) and seven lobes arrange themselves in the known configuration of the mole- cule of IF (Fig. 4). We have therefore an illustra- tion of the shapes of mdecules such as BF (and FIG.3 other 3-co-ordina ted boron molecules and ions) CH4 (and other 4-co-ordinated carbon molecules) PF, SF, and IF,. It has been suggested that other shapes for PF and IF are feasible (Figs. 5 and 6) * The authors are aware of two brands of modelling balloon s (a)Pucks Modelling Balloons (sold in 11-packetsof 12 and also loose); {b) Arie! Modelling Balloons (sold in 2/6 packets of 12).The latter are larger but all give models about 2-3 feet high. It ISadvisable to use fully and freshly i nflated balloons for demonstration purposes as the shapes deteriorate after shrinkage. Sidgwick and Powell Proc. Roy. SOC.,1940 A 176,153. Gillespie and Nyholm Quart. Rev. 11 339. Gillespie,J. Amer. Chern. SOC., 1960,82 5978. NOVEMBER 1961 and that there is little to choose in terms of stability between these and the known shapes. It is interesting to find,therefore that the balloons as in Figs. 3 and 6 can be manipulated into shapes shown in Figs. 5 and 8 respectively and will remain in these positions until the whole assembly is shaken. The balloons then return spontaneously to their original arrange- ments thus demonstrating the metastability of the arrangements in Figs.5 and 6. FIG.4 n FIG.5 FIG.6 The shapes of molecules involving multiple bonds can be reproduced by tying these lobes together (in pairs for double bonds). This treatment applied to the tetrahedron gives a planar arrangement (Fig. 7) representing the shape of for example COF and the linear arrangement (Fig. 8) representing the FIG.7 FIG.8 shape of COP We have also the tetrahedral arrange- ment (Fig. 9) derived from the octahedron and corresponding to S02CI,,and the planar arrangement (Fig. 10) corresponding to monomeric SO,. FIG.9 Fro. 10 Even more satisfying is the way in which lone pairs and bond pairs can be simulated by the use of balloons of different sizes.The packets of these balloons on sale usually contain two kinds one size being shorter and what is more important fatter than the other. The greater width of the shorter balloons is analogous to the greater repulsive effect of a lone pair of electrons (perhaps the analogy is even closer since an orbital occupied by a lone pair is probably more widely diffused in space compared with a bonding orbital which is more likely to be elongated and narrow). When a tetrahedron is formed from a fat balloon and a thin balloon (Fig. 11) the angle between the thin lobes is less than 109" FIG.11 n FIG.12 and that between the fat lobes (representing lone pairs) exceeds 109". We have thus simulated the situation in the water molecule though it is im- portant to realise that an exact agreement with the known H-0-H bond angle would be purely coinci- dental.With the situation in the H2S molecule in mind it can be pointed out that the thinner the thin balloons the smaller the angle between them. A particularly interesting model reproduces the situa- tion in the ClF molecule (Plate 3). If a single thin balloon is twisted by one end into the assembly representing H20 we find that the five lobes point to the corners of a distorted trigonal bipyramid and that the three thin lobes (bond pairs) are coplanar forming the known distorted T shape of the CIF molecule (Plate 3). If the balloons are manipulated into the more symmetrical shape (Fig. 12) and then shaken they return spontaneously to the original shape.It will be clear that structures involving multiple bonds and lone pairs can also be repro- duced as for example in Plate 4 where the shape of a thionyl halide molecule is shown. FIG.13 The technique described above not only provides some physical support for the theoretical arguments (though only by analogy) but is obviously of great value as a teaching aid and has been used as such by the authors. The balloons are cheap (some are only Id. each) and the models are very large. The models are particularly convincing to students because of the fact that they take up their shapes spontaneously. Several other stereochemical features may be illustrated with these balloons and may be used in lecture demonstrations with students at the appropriate level.Some examples of these will now be given. (i) The Walden Inversion.-An additional un-inflated balloon is tied into a tetrahedral assembly (Fig. 13a) and to represent an incoming group inflated in situ. The shape of the transition complex (Fig. 13b) is reproduced. The lobe representing the outgoing group is then removed with a pin and a tetrahedron is re-formed but with an inverted con- figuration (Fig. 13c). It should be emphasised that for the best effect the original tetrahedron should be formed from four balloons of different colours tied in by their necks. Also if separate balloons are not used the junction should be secured by thread to prevent complete disintegration of the model when one lobe is burst.RG.14 (ii) Octahedral Substitution.-The stereochemica changes are reproduced by tying an uninflated’ balloon into an octahedral arrangement inflating it in situ and puncturing the “outgoing group”. There are two possible points of attack when the SN2mechanism is involved and both can be simu- lated. Again for demonstration the colours of the balloons should be carefully chosen. (iii) cis-trans-Zsornerism in AZkenes.-Two long balloons of different colours each twisted at two points and twined together at these two points reproduces the stereochemical features of the alkenes. (Figs. 14a and 14b). The coplanarity of the atoms in the ethylene molecule is demonstrated and there is a perceptible analogy to activation energy in convert- ing from the cis- into the trans-form.The apparent nature of the carbon-carbon bond may meet with some disapproval but on reflection it is either old- fashioned or new-fangled according to taste. The two balloons may be tied together as for double bonds mentioned earlier. The shape of the tied pair does indeed correspond to the accepted shape of a wbond but it is not possible to give any repre- sentation of a o-bond in addition. One of the authors (H. R. J.) records his thanks to his son Nicholas for whom the balloons were originally purchased.
ISSN:0369-8718
DOI:10.1039/PS9610000397
出版商:RSC
年代:1961
数据来源: RSC
|
|