|
1. |
Proceedings of the Chemical Society. March 1963 |
|
Proceedings of the Chemical Society ,
Volume 1,
Issue March,
1963,
Page 73-100
Preview
|
PDF (4014KB)
|
|
摘要:
PROCEEDINGS OF THE CHEMICAL SOCIETY MARCH 1963 PUBLICATIONS SURVEY THE COUNCILof the Chemical Society has authorised a Publications Survey the objects of which shall be to study and report on the needs problems and possible developments in the publication of chemical information and to recommend actions on an informed basis whereby the pattern of publications can more adequately meet present and future needs of those who publish and those who read alike; the survey and its recommendations are to be em- bodied in a report that will be published; the survey to be complete within two years. Further the Council has appointed Dr. R. S. Cahn (see p. 98) to conduct this survey thereby taking advantage of his wide experience in chem- istry and unique knowledge of the various aspects of publication.Dr. Cahn has undertaken academic research and teaching in this country and abroad has experience of industrial research as well as academic has run an industrial in- formation service and was an abstractor for many years and ultimately Associate Editor of British Abstracts. As Editor to the Chemical Society for the last 14 years he has seen not only the steady climb in volume of original publica- tion but the demise of the main British activityjn abstract work. He inaugurated the first major Chemical Journal devoted to current awareness and as an original member of the Editorial Board of the I.U.P.A.C. Journal “Pure and Applied Chemistry” has become acquainted with the problems of international publications.The need for such a publications survey is indicated by the fact that for several decades past the volume of chemical publication has doubled in geometrical proportion every 8-10 years. The increase takes the form of larger and more journals of original research more patents and larger abstract journals and compendia. Whilst problems resulting for the learned Societies that act as the principal publishers are difficult those for the user are already almost beyond control. These users include academic and industrial research workers research directors and man- 73 agers information officers and teachers; it is only with great difficulty that they can keep abreast of research even in their own fields.The Chemical Society believes that it owes a duty to its Fellows and to chemists and chemistry in general to replace the present drifting develop- ments by a rational general policy. It is rather generally realised that new tech- niques are required and some work is being done in this direction. There are however in Britain few detailed and no correlated data on the com- plex and inter-related issues of production and use of chemical publications. The survey must provide such data and outline methods whereby the increasing amount of information may be disseminated efficiently to those who need to use it. The Three Stages in the Dissemination of Informa-tion.-To be of value experimental results and theoretical progress require dissemination which involves three stages (a) the work must be made easily and permanently accessible in detail; (b) the work must be brought to the attention of the relevant working specialist (“current aware- ness”); and (c) the total of work must be cor- related and individual items later retrieved.Publication of Original Research.-The increasing size and cost of journals issued by learned Societies and commercial publishers alike have so far been met by increasing the price by splitting one large into several smaller journals or by a page charge. Each method has its own disadvantages. The first two in particular lead to less availability of the work to the individual potential user thus partially stultifying the very object of publication.In 1962 Current Chemical Papers which deals with “pure” chemistry only listed 31,101 titles taken from 2415 issues of journals. In its much wider field Chemical Abstracts now reports more than 160,000 published items a year. Current Awareness.-Originally the chemist read all original chemical work later all in his own field only; for long however he has relied on abstract journals; but particularly in the last ten years even abstract journals have become too large for scanning and too tardy for present competitive conditions and there has resulted a rapid expansion in review journals and in periodic correlated lists of titles new compounds and spectra. In addition individual industries PROCEEDINGS firms and research teams require their own scanning services.Correlation and Retrieval.-The classical methods of correlation and retrieval involve the abstracting journals and great compilations such as Beilstein and Gmelin. The volume of current research has outstripped these methods. Thus for instance the last decennial index of Chemical Abstracts comprised 19 volumes and future indexes must be still larger; retrieval of a fact from such indexes by individual search is already difficult and will soon become often well-nigh impossible. Again the supplement to Beilstein now being issued covers the literature only to 1939; the still more prolific post-war years remain wholly to be included in later supple- ments. The potentialities of electronic methods are being studjed by Cheniical Abstracts on a large research budget; and similar methods are being studied elsewhere notably in industry.The potentialities of these and other modern methods for data are obvious; their usefulness for con- cepts is still to be determined. Interlocking of the Tlzree Stages.-The three aspects above cannot be considered separately. Current awareness might perhaps be achieved more readily with co-operation of the publishers of original research so might correlation. The methods used for current awareness correlation and retrieval are clearly in the melting pot and the changes that are inevitable there may well have notable effects on the publication of original research which however has problems whose solution may in turn affect the other two aspects.If for example an international information service included an adequate copying service then much of the detail now in journals of original research might be deposited for supply only on demand; one can make a first estimate that this would reduce the size of journals to one half or one quarter; but one cannot yet estimate the effects on the user and other objections can be envisaged. Journals coiitaining original papers shorn of detail might aid the reader to survey the whole to find items of particular interest to him but they might hamper his assessment of the new research. Again realignment from journals deal- ing with general chemistry to more specialist journals perhaps organised on an international basis might benefit the individual in avoiding scatter of related work and provide a more MARCH1963 rational basis for expansion and for development as a whole but equally this might be damaging in its restrictive effect.In none of these cases can the conflicting factors yet be weighed; the examples serve to illustrate some of the types of problem that await and require study. The difficulties of course affect all branches of science; but they are particularly acute for chemistry and it is hoped that any remedies found for chemical publications may incidentally benefit also other disciplines. Co-ordination and Collaboration.-It must be made clear that there is no intention to duplicate the research being done by Chemical Abstracts which is directed specifically to recommend changes in the methods of abstracting and re- trieval; it is manifest however that research in those fields may have great influence on develop-ments in methods of original publication and that the Chemical Society’s actions in the latter field must be compatible with and may need to be dovetailed into progress in the former.It is believed that a programme restricted to “pure” chemistry would be inadequate and that as indicated above it should embrace chemical publication as a whole pure and applied and indeed also aspects of the bordering sciences. The Chemical Society is very glad therefore of the interest shown by other learned Societies DSIR and ASLIB. Further inasmuch as science is international and as conditions in different countries are not all the same it is earnestly hoped that collabora- tion will be forthcoming also from organisations and individuals of other countries.To be of real value such a survey must clearly be based on the maximum amount of evidence. Individuals and organisations having informa- tion or comment relating to the objects of the survey are therefore cordially invited to contact the Director of Publications Research The Chemical Society (Dr. R. S.Cahn). PROCEEDINGS PEDLER LECTURE* Amino-acid Sequences in the Active Centres of Certain Enzymes By F.SANGER RESEARCH LABOMTORY BIOLOGY, (MEDICAL COUNCIL OF MOLECULAR CAMBRIDGE) FROMthe point of view of the chemist one of the most interestingfeatures of living matter is the large number of high-molecular catalytic protein mole-cules which are responsible for controlling and directing the metabolismand chemical reactions that go on in the body.These are the enzymes and the question of what they are like and how they work is one of the fundamental problems we have to solve in order to understand the nature of living matter. There can be little doubt that the catalytic activityof an enzyme depends on its exact chemical structure so that the problem is intimately connected with protein chemistry. Presumably an enzyme exercises its catdytic function by coming into contact with its substrate in a specific manner and exerting an appropriate force which will labilise the substrate bringing about the chemical change.Whereas enzymes are very large molecules the substrates on which they act are fre-quentlysmall. It thus seems probable that only a part of the enzyme may be involved in combining with the substrate and this part is usually known as the “active centre.” It does not necessarily follow that the rest of the molecule is unimportant for activity, but it seems that particular importance should be attached to the active centre and its exact chemical structure. Since enzymes are proteins the active centre must be composed of amino-acid side chains arranged in a specific three-dimensionalarray which will depend both on the sequence of amino-acids in the polypeptide chains and on the method of folding of these chains.Frequently side chains forming an active centre may be distantly spaced from one another along the line of the peptide chain but brought into close contact by the foldingof the chain. A number of enzymes contain non-proteinprosthetic groups which are involved in the catalytic activity. However these are bound to the protein and suitably activated by amino-acid side-chainsso that in these cases too the problem of the active centre is essenti-ally the same namely the specific arrangement of amino-acid side-chains. One way to determine the structure of an active centre which may eventually prove to be the only reliable way is to deternriine the whole structure of the protein molecule. This is clear-ly a formidable task and has not yet been accom-plished for a single enzyme.The amino-acid sequences of a number of the smaller proteins have now been determined and the configuration of the chains in myoglobin and haemoglobin has been de-duced by the X-ray method.’ Thus there seems to be a possibility that we may have a complete picture of the structure of an enzyme molecule in the not too distant future and much work is at present in progress in this direction. In certain cases it has been possible by using suitable labelling techniques which depend on en-zyme activity,to obtain a limitedamount of informa-tion about parts of the active centres of some enzymes without determining the structure of the whole of the molecule and in this lecture I shall describe some of these techniques with which we have been concerned.Although only a limited picture of amino-acidsequence in a part of the active centre can be drawn it is at least a beginning and the methods can in some cases be applied to relatively large proteins whose complete structure could prob-ably not be determined by present techniques. The enzymes to be discussed can all be labelled with 32Pin their active centres and advantage has been taken of this to determineamino-acidsequences around the bound 32P. 32P-Labelled enzymes.-There are two ways in which enzymes can be labelled with 32P.First there are enzymes which contain in their active centre a phosphate group that actually enters into the cata-lysed reaction. Thus phosphoglucomutase (PGM), which catalyses the interconversion of glucose 1-phosphate and glucose 6-phosphate contains a phosphate group that becomes labelled if the enzyme is incubated with radioactivesubstrate.2The reaction of labelled glucose I -phosphate with phospho-glucomutase may be summarised as follows Glc-l-P*04 ++ PO,PO $$ [,,,.[,,,.1 -P*04 JGM I kP‘o,I -I I + p*o4 + GIC- &-PO4 ~CM In this way phosphoglucomutase containing labelled phosphate in its active centre may be pre-* Delivered before The Chemical Societyat The Royal Institution Albemarle Street London W.1 on October I1th, 1962 ar The University Bristol on November Ist and at the University Hull on November 8th.Kendrew Dickerson Strandberg Hart Davies Phillips and Shore Nature 1960 185 422; Perutz Rossmann, Cullis Muirhead Will and North ibid.p. 416. Jagannathan and Luck J. Biol. Chem. 1949,179,569. MARCH1963 pared. In this case there can be little doubt that the active centre is actually labelled since the fact that the 32P comes from the substrate indicates that this must be the area of the molecule that comes into con- tact with the substrate and is primarily responsible for the catalytic effect. Another way in which enzymes can be labelled with 32P is by use of the specific inhibitor di-iso- propyl phosphofluoridate (DFP). This compound reacts specifically and stoicheiometrically with a number of hydrolytic enzymes in such a way that they are inactivated when one equivalent of the phosphofluoridate has reacted with them to form the di-isopropylphospho-enzyme(DIP-enzyme).If the DJP-protein is then subjected to acid hydrolysis serine phosphate (SerP) can be isolated. The re- actions are swnmarised as follows Pri-0 0 Pri-0 \pH +enzyme-+ \$0 .‘ /\ PrI-0’ ‘F Pri-0 enzyme DFP DIP-enzyme HO 0 HCI \p/ -4 HO’ b-CHp I NH,-~H-CO~H SerP Although these enzymes may contain many serine residues only one reacts with di-isopropyl phospho- fluoridate in this way and it seems likely that this unique serine is in fact a part of the active centre and that its hydroxyl group becomes acylated by the substrate during the formation of the enzyme- substrate complex. If di-isopropyl phosphofluoridate containing radioactive phosphorus C2P) is used a radioactive enzyme @132P-enzyme) is formed in which the label is attached to the active centre serine.After acid hydrolysis of both types of 32P-labelled enzyme the radioactivity is found in the form of phosphate and serine phosphate. If however hydro- lysis is only partial the radioactivity is found in pep- tides of serine phosphate and may thus be used as a marker for the isolation and identification of such peptides. Turba and Gundlach and Schaffer and his colleagues? applied partial acid hydrolysis to Dp2P-chymotrypsin and isolated phosphopeptides by means of a DowexSO (sulphonic acid) resin. Owing to the strongly acidic nature of the phosphate *Turba and Gundlach Biochem. Z. 1955,327 186. group these peptides are not adsorbed on the resin whereas other normal peptides are.The phospho- peptides were then fractionated on ion-exchange columns and in this way Schaffer et aL4were able to isolate and identify the peptides SerP-Gly Asp-SerP Asp-SerP-Gly and Gly-Asp-SerP-Gly and from this it was concluded that the sequence around the active serine residue in chymotrypsin was Gly-Asp-Ser-Gly. This was amply supported by the work of others though there was some doubt about some larger peptides from acid hydrolysates. Similar experiments with trypsin indicated that it had the same tetra- peptide sequence. Partial acid hydrolysis is not very suitable for the determination of longer sequences in this way since if mild conditions of hydrolysis are used to give longer radioactive peptides the mixtures produced are extremely complex owing to the non-specific nature of the splitting and purification is then diffi- cult.Better results can however be obtained by using proteolytic enzymes for the hydrolysis and this method has been used extensively by Cohen and his collaborator^.^ Thus for instance they hydrolysed D132P-chymotrypsin with a crude pancreatic extract causing extensive breakdown of the molecule but yielding only two major radioactive peptides; these were pursed by ion-exchange chromatography ionophoresis and paper chromatography. The struc- ture of one of these was shown to be Gly-Asp-SerP- Gly-Gly-Pro-Leu.g In this case the 32P was still in the form of the di-isopropyl derivative (SerDIP) so that the peptide did not have the strong negative charge characteristic of the phosphate group which had proved useful in isolating the SerP peptides from acid hydrolysates.A more rigorous purification was there- fore necessary and the isolation was greatly helped by the fact that there seemed to be very few other peptides as large in the hydrolysate. Similar studies were carried out on other enzymes and these are summarised in the Table. By using more specific enzymes larger peptide fragments can be isolated thus the sequence of 15 amino-acids re- ported by Dixon et al.’ was deduced from the struc- ture of a peptide obtained by the action of trypsin on D132P-trypsin. The Use of Radioautograph Techniques.-In con-nection with attempts to develop techniques for determining amino-acid sequences in larger proteins by isotopic methods we were anxious to develop a method for deducing the sequence around a given radioactive amino-acid.Enzymes of this type labelled 4 SchafFer Simet Harshman Engle and Drisko J. Biol. Chem. 1957,225 197. ICohen Oosterbaan Warringa and Jansz Discuss. Faraday SOC.,1955,20 114. *Oosterbaan Kunst van Rotterdam and Cohen Biochim. Biophys. Acta 1958 27 556. 7 Dixon KaufYman and Neurath J. Biol. Chem. 1958,233 1373. PROCEEDINGS in specific positions with 32P seemed to be ideal model only peptides present in significant amounts in such systems with which to work out such methods. The a hydrolysate were SerP-Gly Asp-SerP Asp-SerP- studies were thus initiated on chymotrypsin which Gly and Gly-Asp-SerP-Gly.Moreover when the was known to have the sequence Gly-Asp-Ser-Gly various peptides from the ionogram were investigated and it was found to be possible to work out sequences it was found that several were interconvertible and of this type by using radioactive techniques and with- appeared to have the same structure. This was out depending on the ninhydrin colour reaction as eventually found to be due to an interconversion of had hitherto been the case. This had several ad- the aspartyl residues that was occurring during the vantages for determining small sequences especially acid hydrolysis. Thus the normal a-aspartyl residue Sequences near the active centres of sume enzymes.DFP-inhibited enzymes Ref. Trypsin AspNH2-Ser-Cys-Glu-Gly-G1y-Asp-Ser-Gly-Pro-V~-Cys-~r-Gly-L~ 7 Chymotrypsin GIy-Val-Ser-Ser-Cys-Me t -Gly-Asp-Ser-GI y-GI y-Pro-Leu-Val-Cys-Lys 4 6, Elastase GI y- Asp-Ser- G1 y 9 Thrombin Asp-Ser-Gly b Liver aliesterase (horse) GI y- Glu-Ser- Ma-GI y-GI y 11 Pseudocholine esterase Phe-Gl y- Glu-Sev- Ma-GI y C Acetylcholinesterase (electric tissue) Glu-Ser-Ma 10 Subtilisin Thr-Ser- Met-Ma 15 Mold protease (Aspergillus oryzae) Thr-Ser-Met-Ma 10 Serum albumin (rabbit human) Axg- Tyr-Thr-Lys 16 Serum albumin (bovine) Arg- Tyr-Thr-Arg 16 Phospho-enzymes Phosphorylase a Lys-GIuNH2-Ileu-SerP-Val-Arg d Phosphoglucomutase Thr-Ma-SerP-His- Asp 13 Refs. :(a) Hartley Roc. 5th Internat. Biochem. Congress Moscow,Symp.IVYin the press. (6) Gladner and Laki, J. Amer. Chem. Soc. 1958,80 1263.(c) Jansz Brons and Warringa Biochem. Biophys. Acta 1959,34,573. (d)Fisher, Graves Crittenden and Krebs J. Biol. Chem. 1959,234 1698. in the case of the =P-enzymes which were then in the peptides was being converted in acid to the studied in more detail as described below. a/!-ring form which by hydrolysis could give rise to The main technical difficulty in this type of work the /!-form and indeed in any work on amino-acid sequences CHZ-CO H lies in the purification of the peptides. The method CH,-CO most commonly used for fractionating the small SerP ?+ ;hi-j peptides obtained from acid hydrolysates was ion- -NH-CH-CO exchange chromatography which has proved so -NH-CH-CO-NH-apAspartyl successful in work on the structure of ribonuclease* a-AspartyI 11 and other proteins.We had been using high-voltage CHS-CO-N H-paper ionophoresis for the separation of peptides and this seemed to offer considerable advantages for the -NH-CH-CO,H fractionation of serine phosphate peptides from the @-Asparty1 point of view of simplicity and of resolution. Fig. 1 shows a radioautograph of an ionogram of a partial This reaction appears to be a general one and will acid hydrolysate of D132P-chymotrypsin.9 This shows have to be taken into account in experiments in considerably more components (thirteen) than ap- which partial acid hydrolysis is used. pear on an ion-exchange column of a similar An ionogram of the type shown in Fig. 1 can be hydrolysate suggesting more efficient resolution.regarded as a very specific characterisation for a However the large number of bands raised a serious given radioactive amino-acid in a particular amino-difficulty since Schaffer et aL4 had shown that the acid sequence. The position of the bands is deter- s Him Moore and Stein J. Biol. Chem. 1960,235 633. Naughton Sanger Hartley and Shaw Biochem. J. 1960,77 149. FIG. 1. Radioautograph of ionogram (pH 3-5 40 vlcm. 2.5 hr.) of partial acid hydrolysate of D132P-chymotrypsin. (Reproduced by permission from Biochem. J. 1960,77, 149.) FIG. 2. Radioautograph of ionogram (pH 3-5 40 vlcm. 2 hr.) ofpartial acid hydrolysates of various 32P-labelled proteins ;DIP-Ch D132P-chymotrypsin ; DIP-S D132P-subtilisin B.DIP-LAE D132P-liver aliesterase;PGM 32P-phosphoglucomutase; O1and O, peptides from chymotryptic hydrolysate of 32P-labelled ovalbumin. (D. Shaw unpublished work.) FIG. 3. Radioautograph of ionogram (pH 3.5 40 v/cm. 2 hr.) of partial acid hydrolysates of D132P-derivatives of chymotrypsin elastase and trypsin. (The strong fast-moving bands from trypsin are derived fvom DF32P which contaminated the preparation.) (Reproduced by permission from Biochem. J. 1960 77 149.) FIG. 7. Radioautograph of ionogram (pH 3.5 40 v/cm. 2 hr.) of partial acid hydrolysate of D132P-derivative of rabbit-serum albumin (RSA) compared with similar hydrolysat e of D132P-~ hymotrypsin(Ch). MARCH1963 mined by the amino-acids that are bound to the SerP residue.Thus the pattern shown in Fig. 1 is charac- teristic for the sequence Gly-Asp-SerP-Gly .Patterns obtained for other enzymes are shown in Fig. 2. We were interested in the enzyme elastase which like trypsin and chymotrypsin is a pancreatic proteinase and was found to react with di-isopropyl phospho- fluoridate. Fig. 3 shows a radioautograph of an iono- gram of a partial acid hydrolysate of D132P-elastase.Q This was run on ionophoresis together with similar hydrolysates of the DIP derivatives of trypsin and chymotrypsin. It can be seen that the patterns ob- tained from the three enzymes are identical showing that elastase also contains the sequence Gly-Asp- Ser-Gly around its reactive serine residue. In a similar manner it was shown1* that electric tissue acetylcholinesterase contains the sequence Glu-Ser- Ma since it gives the same pattern as liver aliesterase (Fig.2) which had already been shown to have this sequence?' Determinationof Sequenceby Radioactive Methods. -In the case of phosphoglucomutase preliminary resultsf2 had suggested a sequence similar to that found in chymotrypsin and trypsin. However Fig. 2 shows that the pattern of radioactive peptides is entirely different from that obtained with chymo- trypsin and also different from that obtained with liver aliesterase or subtilisin. Thus the question arises how can we determine an amino-acid sequence repre- sented by a pattern which is not the same as that shown by any known sequence? To the classical organic chemist the correct way to find the structure of an unknown compound is to start by determining its melting point and elementary composition.On the small amounts of material we are working with this is out of the question and in any case it would not provide much useful informa- tion about the structure of a peptide. Thus the pro- tein chemist has come to rely more on spots on chromatograms and on quantitative analyses of amino-acids by the ninhydrin method as an initial step in studying the structure of peptides. Thus one way to determine the sequence around the active serine in a labelled enzyme such as phospho-glucomutase would be to cut out the radioactive bands shown in Fig. 2 elute the material from them and hydrolyse and identify the amino-acids by paper chromatography or ion-exchange chromatography.This is the method that has been used by Schaffer Cohen and others. However in this case certain difficulties are involved. One is the small amount of material available; but a more serious difficulty is lo D. C.Shaw. unmblished work. that the radioactive bands are usually not pure being contaminated by non-radioactive peptides. Thus a partial hydrolysate of a protein may contain several hundred different peptides which may contaminate the few radioactive ones. Normally the serine phos-phate peptides may be almost completely separated from the other peptides by virtue of the strong acidic phosphate residue by ion-exchange chromato- graphy or by paper ionophoresis at pH 3.5 (as in Fig.2). In the latter system they migrate towards the anode whereas most other peptides are neutral or basic at this pH and stay near the origin or move towards the cathode. However in the case of phos- phoglucomutase most of the radioactive peptides are near the origin because they contain a basic residue as well as the acidic phosphate group and they are thus heavily contaminated with non-radioactive pep- tides. It was thus necessary to develop a method for determining the amino-acid sequence of the radio- active peptides that did not depend on the use of the ninhydrin method but used only radioactive tech- niques. Such techniques would apply only to the few radioactive peptides that can readily be separated from one another and are not affected by the presence of non-radioactive impurities.The first step in this study was to work out the inter-relations of the various radioactive sp0td3 Samples of each band were eluted and subjected to two treatments (a) Partial hydrolysis with acid followed by iono- phoresis at pH 6.5 in parallel with a sample of the complete hydrolysate to see which were the break- down products of each peptide. (b) Treatment with phenyl isothiocyanate by the Edman method followed by ionophoresis. This method splits off the N-terminal residue of the pep tide as a phenylthiohydantoin derivative. If serine PhNCS + NHiCHR-CO-NH phosphate is N-terminal the radioactivity would be expected to appear as the thiohydantoin of serine phosphate; however this is unstable and phosphate is in fact produced.If the serine phosphate is not N-terminal the radioactivity would appear in the peptide from which the N-terminal residue was removed. The results with the peptides from phos- phoglucomutase are summarised diagrammatically l1 Jansz Posthumus and Cohen Biochim. Biophys. Acta 1959 33 396. Koshland and Erwin J. Amer. Chem. SOC.,1957,79 2657. Milstein and Sanger Biochem.J. 1961 79 456. in Fig. 4. In this case ionophoresis was carried out at pH 6.5 since better separation of the peptides is obtained than at pH 3.5 (Fig. 2). Portiol hydrolysis PCM eEdmon-7A 1 FIG.4. Diagram of ionograms (PH 6.5) of the pro- ducts formed from partial acid hydrolysis or the Edman degradation of radioactive peptides obtained from partial acid hydrolysis of 32P-PGM.A hydro-Iysate of 32P-PGMwas run in parallel to identgy bands (column labelled PGM). If a dipeptide is subjected to partial acid hydrolysis only two radioactive products serine phosphate and phosphate are produced besides the unchanged di- peptide. Larger peptides would give more breakdown products. There should be two dipeptides containing the serine phosphate residue and these are clearly peptides 4 and 7A (Fig. 4). Peptide 7A when sub- jected to the Edman degradation gave only phos- phate indicating that serine phosphate is N-terminal. There was not sufficient of peptide 4 to subject it to the Edman method but by difference it must be the dipeptide with serine phosphate C-terminal.Peptide 5 on degradation gave peptide 4,showing it is the tripeptide formed by addition of a residue to the N-terminal group of peptide 4. Peptide 6A had N-terminal serine phosphate and on partial acid hydrolysis gave peptide 7A as well as phosphate serine phosphate and unchanged 6A. Such be- haviour would be expected from a tripeptide formed by addition of a residue to the C-terminal end of the dipeptide 7A. If the sequence around the serine phos- phate is represented as A-B-SerP-X-Y the relations and size of the various radioactive peptides as de- duced from the results shown may be summarised as in Fig. 5. PROCEEDINGS The next problem was to identify the unknown residues A B X and Y.Since peptide 7A was neutral on ionophoresis at pH 3.5 (Fig. 2) it was concluded that it contained a basic amino-acid which must of course be X. Further details about the I-jA1 r7"1 r57 141 A. B. Sere X. Y. -78--7c--6s I FIG.5. Relationshes of the =P-peptides obtained from partial acid hydrolysate of 32P-PGM. (Repro- duced by permission from Biochent.J.,1961,79,458.) nature of $he various charged groups could be ob- tained by studying the imophoretic rates of the various radioactive peptides at different pH values. The results for peptides 6A and 7A are shown in Fig. 6 compared with those for synthetic SerP-Gly. These curves are essentially analogous to titration curves. Comparing the curve for 7A (SerP-X) with that for SerP-Gly it can be seen that whereas 7A contains an extra basic group at low pH's this basic group becomes discharged around pH 6-7 so that at pH 8 peptide 7A has almost the same mobility as SerP-Gly.The only basic amino-acid that ionises in the range 6-7 is histidine so it could be concluded that X was histidine. By comparing the pH-mobility 2 4 6 8 PH FIG.6. pH-mobility curves for peptides 6A and 7A from 3"-PGM and for SerP-Gly. Mobilities were expressed relative to a standard serine phosphate marker. (Reproduced by permission from Biochem J. 1961 79 458.) MARCH1963 curves of peptides 6A and 7A it can be seen that whereas at low pH's they are similarly charged pep- tide 6A takes on an extra negative charge at pH 4-5.This must be a carboxyl group indicating that residue Y must be one of the acidic amino-acids (glutamic and aspartic acid). Similarly from the ionophoretic rates of peptides 4 and 5 it could be concluded that residues A and B were neutral amino-acids. Further identification could be made by carrying out specific chemical reactions on the radioactive peptides and seeing if changes in mobility occurred. Thus histidine residues are uniquely sensitive to photo-oxidation. After photo-oxidation those pep- tides containing residue X became more acidic where- as other peptides were unaffected. In this way the identification of X as histidine was confirmed. Another useful test was the use of periodate. Only those peptides containing serine or threonine as the N-terminal residue react with this reagent to become more acidic.Peptide 5 was attacked whereas 4 was not showing that acid A is either serine or threonine but B is not. At this stage the sequence could be reduced to ik;}B-SerP-Hit {zi where B was a neutral amino-acid other than serine or threoni ne. Clearly these specific reactions are somewhat limited and it seemed more desirable to have a general method that could be used to identify any residue. The most hopeful approach to this problem was by studying the RFvalues of the different radioactive peptides on paper chromatography in different sol- vent systems. It has been suggested by Pardeef4 that the RFvalue of a peptide (RF(P))in a given solvent system can be related to the RF values of the com- ponent amino-acids (RF(a)) by the following formula RT In[(l/RF(p~)-1J = (n -1)A + B + CRTIn[(lI&(a)) -1I in which R is the molar gas constant Tthe absolute temperature and n the number of amino-acid residues in the peptide and A and B are constants.if for a given solvent system the values of the con- stants and the RFvalue of a peptide with one un-known amino-acid are known the RF vahe of the unknown amino-acid can be deduced. The RFvalue of an unknown amino-acid is not in most cases enough to characterise it but from a combination of RF data on more than one system this should be possible. In order to identify residues A and B two solvent systems were studied. The RF values of the known amino-acids were first determined then the RF l4 Pardee J.Biol. Chem. 1951 190 757. values of a number of synthetic peptides; from these the values of the constants A and B could be cal-culated. From the RFvalue of peptide 4 (B-SerP) on one system it could be deduced that residue B was either threonine alanine or proline. The periodate results had already indicated that it was not threo- nine. The RFvalues of peptide 4 on another system (phenol) on which alanine and proline are very widely separated was then determined and from this it could be concluded that residue B was alanine. Similarly from the RFvalues of peptide 5 residue A was shown to be threonine rather than serine. The sequence could then be written as lU Thr-Al a-Se rP-H is-(ASP The eRcacy of this method clearly depends on the accuracy of the Pardee formula.It is certainly not absolutely accurate and some peptides may show greater deviations than others but for most purposes it is probably sufficiently accurate. This approach is likely to improve as more experience is gained both with regard to the known rates of peptides and as to whether to expect deviations from the formula or from any other improved formula that may be developed. In the case of the above sequence it was not possible to decide whether residue Ywas aspartic or glutamic acid and in order to do this and to con- firm the above results it was necessary actually to isolate peptides 5 and 6A in a pure form and to identify the amino-acids by the standard method.This purification was greatly helped by the fact that their structure was already almost known so that the purification was continued until a rational result was obtained. In the case of peptide 6A it was necessary to fractionate it on five different systems and even then small amounts of impurities were still present. These results confirmed the above results and showed that the sequence was Thr-AIa-Ser P-His-Asp. Serum AIbumin.-In all cases where di-isopropyl phosphofluoridate has been found to react with the active centre of an enzyme serine phosphate has been obtained on hydro1y.k of the DIP-protein indicating thdt the reiction has been with the hydroxyl group of a serine residue. On the basis of early results it was suggested that all DIP-enzymes contained the same sequence (Asp-Ser-Gly) at the reactive site.When however horse liver aliesterase was found to have the sequence Glu-Ser-Alall the generalisation had to be modified by saying that the sequences were similar the serine being bound to an acidic residue through its amino-group and a small neutral one through its carhoxyl group. a2 Fig 2 shows the pattern obtained with the bacterial proteinase subtilisin. This is entirely different from the other patterns showing that this enzyme does not have the sequence Asp-SerP-Gly of chymotrypsin or Clu-SerP-Ala of afiesterase or Ala-SerP-His of phosphoglucomutase. By studying the radioactive peptides by the methods described above it was shown that the sequence was Thr-Ser-Met-Ala.lS This is entirely different from those found in the other DFP-inhibited enzymes and indicates that a serine residue in at least three different sequences is capable of reacting with di-isopropyl phospho- fluoridate.The most likely explanation of this seems to be that reactivity of the serine is not in fact con- ditioned by this sequence but rather by other amino- acid side chains which come near the serine residue. Such side chains may be far removed from the serine according to the primary structure along the poly- peptide chains but brought near to it in space owing to the secondary and tertiary folding of the chains. Alternatively it may be that the amino-acids linked to the serine in the chain are concerned in its re- activity and that somewhat different mechanisms are involved in different cases.It seems that the only thing that the DIP-enzymes have in common is a serine residue. Recent work by D. C. Shaw,16 however shows that even this is not entirely true. In an attempt to study new DIP-enzymes the reaction of the phosphofluoridate with various biological preparations was studied and an extensive reaction was found with blood. On hydro- lysis and ionophoresis of the D132P-derivative a pat- tern was obtained which was different from these of the known DIP-enzymes. Further investigation showed that the activity was due to serum albumin. The reaction was considerably slower than with the DFP-inhibited enzymes but was nevertheless virtually specific for a single active site.Fig. 7 shows the radioautograph obtained with DIP-albumin. It is entirely different from that ob- tained with other 32P-labelled proteins and in par- ticular no serine phosphate is present. Instead the main end-product of hydrolysis is a band (R8,Fig. 7) which was identified as tyrosine phosphate (TyrP) showing that the phosphofluoridate had reacted with an activated tyrosine residue. The peptides obtained by acid and enzymic hydrolysis of the D132P-albumin were studied by the methods outlined above and it was concluded that the sequences in rabbit- and human-serum albumin were A-B-C-D-Arg-Tyr-Thr-Lys where A,B,C and D are neutral residues. In bovine material the sequence was Arg-Tyr-Thr-Arg and l5 Sanger and Shaw AJature,1960 187 872.I6 Shaw in the press. PROCEEDINGS there were further differences in the neutral residues. These results show the presence of a uniquely re- active tyrosine side-chain in serum albumin and the question arises whether the reactivity has any physio- logical significance. It could of course be a mere coincidence that a particular centre binding di-iso- propyl phosphofluoridate is present. Serum albumin is not normally considered to be an enzyme though it does possess certain unique reactivities. The best known of these is its ability to bind anions and to transport them in the blood and some preliminary results suggest that the reactive tyrosine residue may in fact be involved in this ion-binding reaction.If indeed this were the case the presence of two basic side chains near the active tyrosine residue would fit the hypothesis. However further work is clearly necessary on this problem. General Conclusions.-In the Table are listed the various sequences that have been found to exist in the region of the reactive site on DFP-inhibited enzymes or serine phosphate enzymes. While there are certain similarities there are also considerable differences and no general conclusion seems possible as to what residues are responsible for the activation of the serine. That there are differences is not al- together surprising in view of the different activities of the enzymes and it seems that more significance should be attached to the similarities as in the case of the pancreatic enzymes and thrombin.One ex- planation for these similarities is that the sequences are necessary for the biological activity. Another is based on the biological origin of the proteins thus it may be considered that all three pancreatic pro- teinases may have evolved from a single protein which was present in some more primitive organism and controlled by a single gene; as evolution pro- ceded duplication of the gene has occurred and the different enzymes have developed separately giving rise to molecules with different specificities but re- taining the common sequence Gly-Asp-Ser-Gly. In fact both explanations may be correct. The enzymes may have evolved from a common precursor but have retained the common sequence since this was suitable for the catalytic activity.The techniques described above if generally applicable make it possible to determine an amino- acid sequence around a given labelled amino-acid and can therefore be used in any enzyme whose active centre can be labelled. It may also be possible to use them more generally for the determination of longer sequences in proteins. For instance by bio- logical incorporation methods proteins may be pre- pared in which a single amino-acid is labelled. If such a protein were digested with an enzyme and the MARCH1963 digest fractionated peptides would be isolated each containing one or a few residues of the labelled residue and the sequence around each of these resi- dues could be determined.By using a number of labels it might be possible to deduce complete sequences. At present such a method would be more tedious than other techniques available; however if a routine system could be developed the possibility of working entirely with paper-fractionation tech- niques could offer a considerable advantage. The reaction of di-isopropyl phosphofluoridate has certainly proved to be the most profitable method for determining sequences in active centres; the method is clearly limited to a small group of enzymes and to only a part of the active centres of these enzymes. However other techniques of labelling are available and have been much less explored. Certain enzymes contain prosthetic groups that are sufficiently strongly bound to the protein chain for the peptides containing the prosthetic group to be isolated.The classical example of this is the work of Tuppy and Paleusl’ on cytochrome c. Here the ham group is bound to two cysteine residues. From enzymic hydrolysates it was possible to isolate haem peptides and by studying their structure the sequence around the “active” cysteine residues was deter-mined. In general most other prosthetic groups are less firmly bound but it may be possible to apply this approach more generally by the careful use of en- zymes under mild conditions. Many enzymes contain thiol groups in their active Tuppy and Paleus Acta Chern. Scand. 1955,9 353. l8 J. I. Harris and J. Park unpublished work. Icentres.These groups are very reactive and offer the possibility of labelling with coloured or radioactive markers. Thus glycerophosphate dehydrogenase con- tains a unique thiol group which reacts stoicheio- metrically with iodoacetate causing inhibition of the enzyme activity. Harris and Parkla have used i~do[~~C]acetate to label this site and by isolating labelled peptides have determined the sequence around this active centre. Another approach is suggested by recent work of Grazi et a!.19on the enzyme aldolase. In this case the substrate (dihydroxyacetone phosphate) is bound to an amino-acid group on the enzyme as a Schiff base in the enzyme-substrate complex. By reducing this complex with borohydride it was possible to stabilise this linkage and after hydrolysis a compound was isolated in which the substrate was bound by a secondary amino-group to a lysine residue.The possibility of using bifunctional reagents has also been suggested. In such a reagent one function should resemble the substrate molecule and thus attract the reagent to the active centre while the other function should be a reactive group such as an acid chloride and thus become attached to groups in the vicinity of the active centre. Clearly there are many possibilities in this field. All enzymes have a unique reactivity which is the basis for their catalytic effect and as we learn more about these reactivities it may be possible to devise suitabIe labelling techniques. l@Grazi Rowley Cheng Tchola and Horecker Biochem.Biophys. Res. Conini. 1962 9 38. COMMUNICATIONS A New Route to Thione Esters By G. R. BANKSand D. COHEN (DEPARTMENT THEUNIVERSITY, OF CHEMISTRY KEELE,STAFFS.) ALKOXYACETYLENES have been used previously to prepare 1-alkoxyvinyl esters of carboxylic (I) and phosphoric acids (II).ls2 We have investigated the use of ethoxyacetylene with and without mercuric ion as catalyst in the formation of ethoxyvinyl esters of various acids. By reaction of the appropriate thiolic acid with an excess of ethoxyacetylene we have isolated 1-ethoxyvinyl thiolacetate 011; R = Me) (Found C 49-15; H 6.7; S 22.1. C,H,,O,S requires C 49.3; H 6.9; S 21.9%)and thiolbenzoate 011; R = Ph) (Found C 64.0;H 5-5; S 15-3. C11H1202Srequires C 63.45; H 5-8; S 15.8%) both having the characteristic double peaks between OEt (I) X = OCOR CHL=C/ (11) X = O*PO(OR) ‘x (111) X = SCOR 5.7 and 6.0 p in the infrared spectrum.(The double peak between 5-7 and 6-0p appears to be a charac- teristic of all the alkoxyvinyl esters and this absorp- tion may be used quantitatively in kinetic ~tudies.~) Wasserman and Wharton Tetrahedron 1958 3 321; J. Amer. Chem. SOC.,1960 82,661; Cohen and Springall, in the press. a Wasserman and Cohen J. Amer. Cheni. SOC.,1960 82,4435. These intermediates act as acylating agents in the same way as the enol esters of carboxylic acids (I).133 Thus the thiolknzoate with aniline gives benzanilide in >70% yield. The other product of these acyla- tions is ethyl thionacetate.Wasserman and Cohen unpublished work. Matsui Ber. 1909,42,423. PROCEEDINGS CH&(OEt)*S*COR + R'*NH2+ R'.NH.COR + CH,-CS*OEt The ethyl thionacetate was identified by reaction with ammonia to form thionacetamide and by com- parison with an authentic sample? (Received January 28th 1963.) Aziridino-derivatives of Carbohydrates By R. D. GUTHRIEand D. MURPHY (DEPARTMENT THE UNIVERSITY OF CHEMISTRY LEICESTER) D. H. Buss L. HOUGH,and A. C. RICHARDSON (DEPARTMENT CHEMISTRY BRISTOL) OF ORGANIC THE UNIVERSITY THE preparation of non-toxic aziridines is important because of their possible applications in cancer chemotherapy? Christensen and Goodman2 have recently outlined the preparation of an aziridino- sugar namely methyl 2,3-aziridino-4,6-O-benzyl-idene-2,3-dideoxy- a-~-allopyranoside (I1 ;R' = H) with m.p.153-154" and 143-145". We now report three new methods developed independently in the above two laboratories which have greatly facilitated the preparation of these derivatives. The D-do- isomer (11; R = H) has been prepared in Bristol from methyl 3-benzamido-4,6-O-benzylidene-3-de-oxy-2-0-methanesulphonyl-a-D-altrop yranoside (I ; R = NHBz) by treatment either with lithium alumin- ium hydride in tetrahydrofuran or hot methanolic sodium methoxide? and in Leicester by treatment of the corresponding 3-azido-derivative* (I; R = N,) with Raney nickel in methanolic hydrazine. The very weakly basic imine (XI; R = H) m.p. 152-153" [a] + 150" (in CHCL,) was characterised by con- version into the N-acetyl and N-benzoyl derivative.The D-manno-isomer m.p. 145-146" [a] + 105" (in CHCI,) was similarly prepared by these methods and gave a crystalline N-acetyl and N-benzoyl deriva- tive. The benzoate was directly prepared from the amide (111; R = NHBz) by treatment with sodium methoxide at room temperature demonstrating the ease of imine-ring formation. Treatment of the benzoate (IV; R = Bz) with hot methanolic sodium methoxide or lithium aluminium hydride removed the acyl group with the formation of the free base (IV; R' = H). The loss of the N-benzoyl group with lithium aluminium hydride is obscure but it is note- worthy that the aziridino-N-acetyl derivative (IV; R = Ac) is similarly converted into the aziridine (IV; R = H),albiet in poor yield.Both the D-ah-and the D-manno-aziridine were stable to lithium aluminium hydride in contrast to the corresponding epoxides and episulphides. How- ever ring opening of the D-manno-aziridine was effected by the Leicester group using sodium azide in aqueous 2-methoxyethanol to give the trans- diaxial azido-amine (V) which was converted into the syrupy diamine (VI) characterised as the crystalline di-N-acetyl derivative. Ring opening has also been achieved by using the thiocyanate ion. These experiments should give derivatives of amino-sugars difficultly accessible by conventional methods. The Leicester studies were supported by the U.S. Army through its European Office and one of us (D.H.B.) thanks the Department of Scientific and Industrial Research for a Research Studentship.(Received,January 15th 1963.) Internat. Cancer Congress Moscow Angew. Chem. Internat. Edn. 1962 1 600. Christensen and Goodman J. Amer. Chem. Soc. 1960,82,4738. Cf. Taguchi and Kojima J. Amer. Chem. Soc. 1959 81,4316. Guthrie and Murphy Chem. and Ind. 1962 1473. MARCH1963 The Crystal Structure of Trimethyltin Fluoride By H. C. CLARK R. J. O’BRIEN and J. TROTTER (DEPARTMENT UNIVERSITY VANCOUVER, OF CHEMISTRY OF BRITISH COLUMBIA 8 B.C. CANADA) INTEREST in the structures of organo-tin compounds has been stimulated by the recent report of five- co-ordination of the tin atom in the compound Me3SnC1,Py? and by the suggestion that trimethyl- tin acetate contains bridging acetato-groups again giving five co-ordination to the tin atom.It therefore seemed valuable to determine the structure of tri- methyltin fluoride which on the basis of its infrared spectrum3 has been considered to consist of planar Me3Sn+ and F-ions. A number of unusual structural features have been observed. Crystals of trimethyltin fluoride Me3SnF are orthorhombic with four molecules in a unit cell of dimensions a == 4.32 b = 10.85 c = 12434 A space group Pmcn. The structure was determined by Patterson methods and refined by three-dimensional Fourier and difference syntheses. R is 0.12 for 230 observed reflexions. The structure consists of chains of trimethyltin groups and fluorine atoms along Q with only weak van der Waals forces between the chains.In the three- dimensional electron-density distribution the tin atom and one of the carbon atoms (C,) are repre- sented by spherical peaks in the mirror plane at x = 1/4 but both of the other carbons (C and C,) are split into two half-atoms one on either side of the plane. The electron density at the fluorine atom is even more remarkable being spread out over two parts of a surface of a sphere; the density is highest in positions above the C,-Sn-C and C1-Sn-C angles. One possible interpretation of these rather unusual features is that within any one chain the fluorine atoms are disordered occupying any posi- tion on parts of a spherical surface about 2-1 A from one tin atom so that Sn-Fa-Sn is not linear while the atoms C and C3are ordered and displaced from the plane x = 1/4 in the opposite direction to the Sn-F bond (see Figure).The disorder of Cz and C3 and the further disordering represented by reflexion of a fluorine density in the mirror plane at x = 3/4 are a consequence of the chains’ being able to align themselves in either direction along a. The consider- able amount of disorder may allow interpretations which differ in detail from this description but one definite conclusion is that the structure is not that of a purely ionic solid; indeed the non-linear Sn-Fa-Sn arrangement is a definite indication of some covalent interaction. Moreover each fluorine atom is not equi- distant from the two Sn atoms so that there is some small tendency towards the formation of discrete molecules.The apparent lack of planarity of the tri- methyltin groups should produce an infrared absorp- tion at -515 cm.-l due to the Sn-C symmetric stretching vibration. Although this has not been previously reported it can be observed as a weak band in the spectrum obtained from a very con- centrated Nujol mull. Again in this structure the tin atom is five-co-ordinate although its stereo- chemistry is not simple. Infrared spectroscopic studies of compounds such as Me3SnC104 and Me3SnBF suggest4 that their structures consist of planar trimethyltin groups interacting strongly with the anionic group so as to make the tin atom essentially five-co-ordinate and to destroy the regular tetrahedral symmetry of the anion.The available evidence therefore indicates that five-co-ordination of tin may be common and that the free planar Me3Sn+ ion is rarely encountered. The authors thank the National Research Council of Canada for financial support and Canadian Industries Limited for a Fellowship (to R.J.O’B.). (Received. December 14th 1962.) Beattis McQuillan and Rulme Chem. and Znd. 1962 1429. Van der Kerk Luijten and Janssen Chimiu 1962 16 10. * Okawara Webster and Rochow J. Amer. Chem. SOC.,1960,82 3287. Clark and O’Brien unpublished results. 86 PROCEEDINGS Novel Spectra and Some Sterically Uncomplicmted Basicities for Metal Halide-Nitrogen Base Equilibria By D. P.N. SATCHELL and J. L. WARDELL OF CHEMISTRY COLLEGE OF LONDON,STRAND, (DEPARTMENT KING'S UNIVERSITY LONDON,W.C.2) PREVIOUS studies1 of the interaction in inert solvents similar bases with much stronger Lewis acids,lc or of halides of the non-transition metals (M& e.g.perhaps in more polar solvents. SbCl, BF, SnC1,) with various classes of indicator When large reagent concentrations are used base (B,e.g.,aromatichydrocarbons ketones amines) coloured complexes are precipitated. For bases nos. have been considered to reveal one or both of two 2 4 and 6 these have a 2B 1SnC14 stoicheiometry distinct spectral effects (1)effects similar to those pro- but dissolve in an excess of solvent to give the duced on protonation of the base and probably due to equilibria noted above. formation of the species +B-M&-and (2) effects Values of pK (K = [C]/[Bj[SnCl,]) are included arising from the presence of the free radical ions B+ in the Table.They appear to be the first determined and M&-. Here paramagnetic phenomena are also for stannic chloride and in fact comprise the bulk observed. For amines we now report a new effect of all the available data not concerned with Lewis not previously recognised in these contexts indica- acids derived from boron. K may be calculated from tive of a weaker type of interaction. the stoicheiometry and the changes in absorption of Spectral details and dissociation constants at 20" f2'. E [B] x 104~. [SnCI,] w lo4 to 1~. = extinction coefficient at band maximum (Amax*). No. Aniline derivative lWe ATx-(mp) AFax.(mp) pK PKa (at 25") 1 3-Nitro 1.81 Small at > 310* 366 -3.60 2.50 2 4-Nitro 15-7 347 -3.18 0.99 1 3 &Met hyl-2-nitro 7.93 412 -2.50 0.45 9) 4 2-Nitro 5.02 397 -2.30 -0.29 Y9 5 2-Chloro4nitro 12.6 7.81 425 352 -0.70 -0.94 6 4-Chloro-2-nitro 4.77 1-67 41 8 410 -0.7 1 -1 *03 7 5-Chloro-2-nitro 5.5 1 3.2 444 393 -0.34 -1*54 8 2,5-Dichloro-4-ni tro 9.26 10.1 410 347 0.15 -1.87 9 N-Phenyl-4-nitro 21.6 22.6 476 384 -0.96 -2.48 10 2,6-Dichloro-4-nitro 11.5 16.9 415 350 0.53 -3.02 * Stannic chloride absorption prevented measurements at shorter wavelengths.Anilinium ions absorb below 310 mp. Nitroanilines in o-dichlorobenzene react re-either the base or the complex. K may be determined versibly with stannic chloride to give 1:1 complexes for successive bases either directly or by stepwise (C).A wide basicity range can be covered. Solutions comparison. A comparison of thermodynamic of the complexes have negligible radical character basicities is therefore accomplished.ld yet their spectra are not always similar to those of Since all previous studies (not subsequently shown solutions containing the corresponding anilinium suspectle) have either deliberately courted steric ions. A new absorption band is often found between effects3or have covered only a small basicity range1d$3 410 and 480 my whose relative intensity to that of it has not been possible hitherto to state with con- the base tends to vary inversely with the latter's fidence whether relative basicities of unhindered strength (see Table).We interpret the new bands as bases towards metal halides may be expected to charge-transfer spectra and the complexes as charge- parallel their basicities towards protons (e.g. as transfer species2 possessing an appreciable non-defined by pKa determined in aqueous acids). There bonding component (BySnC1,) coupled with a usual- seem sufficient sterically uncomplicated comparisons ly less appreciable ionic component (+B-SnCI,-) ;for available in the Table to show that although the the stronger bases (nos. 1-4) the latter component general sequence remains much as in aqueous solu- may be more important making their spectra more tion relative basicities are often notably altered- akin to those of the anilinium ions the charge- and altered more than would be expected from the transfer band being weak.However it seems that the solvent extreme of radical-ion formation is only found for (Received January 23rd 1963.) l (a)Aalbersberg Hoijtink Mackor and Weijland J. 1959 3055; (b) Shuba and Zenchelsky J. Amer. Chem. Soc. 1960,82,4136; (c) Kainer and Hawser Chem.Ber. 1953,86,1563; (d)Moodie Chem.ctnd Znd. 1961,1269; (e) Manelis, Vinnik and Chirkov Zhur. fiz. Khim. 1959,33,1030; see however Gerrard and Mooney J. 1960,4028. * Mulliken J. Amer. Chem. Soc. 1952,74,811; Jorgensen. Acta Chem. Scand. 1957,11 166. Brown,J. Chem.Educ.,1959,36,424; McLaughlin Tamres Searles and Nukina J.Znorg. Nucl. Chem. 1961,17,112. * Satchell J. 1958 1916. MARCH1963 The Crystal Structure of tbe /%Formof Triglycerides By Kh LARSON (CRYSTALLOGRAPHY INSTITUTE BIOCHEMISTRY OF GOTEBORG, GROUP OF MEDICAL UNIVERSITY SWDEN) THEstable crystal form of simple fatty-acid tri- glycerides the ,&form has been studied by Vand and Bell.l They analysed the chain packing thoroughly but a complete structure determination is necessary to see the shape of the whole molecule and such knowledge is fundamental for discussions on the solid-state behaviour of triglycerides.The present author has therefore reinvestigated the /%form of trilaurin; experience has shown that this chain length is suitable for obtaining good single-crystals. A study at this Institute of long-chain compounds containing different heavy-atoms has shown that a bromine atom can often replace a terminating methyl group isomorphously.2 The triglyceride of 1 1 -bromo- a jections.Only the (010) projection showed acceptable resolution and three-dimensional refinement was started with the reflexion data of the unsubstituted acid. A composite drawing of the first three-dimensional electron density series is shown in the Figure. The R-value was 0.37 for the 680 reflexions used in the summation. The directions of the chains in the molecule correspond to the proposed "tuning-fork" form with the chains in 1- and 3-position pointing opposite to the chain in 2-position. The molecules are arranged "head-to-head" in double layers. The methyl groups in a chain layer do not lie in one plane as they do in earlier known long-chain structures but form ter- The /&form of trilaurin viewed along the b-axis shown as a composite drawing of the three-dimensional electron-density series.Contours are given at intervals of leA4 starting with 2eAW3. undecanoic acid was therefore prepared and crystals of its /3-form gave the following X-ray data a = 12.40 & 0.10 b = 5.52 Zt 0.05 c = 31.6 f 0.3 A; a = 90.5" -+ 0-6" = 96.2" If lo 7 = 101-9" Ifi 1". The corresponding X-ray data for tri- laurin were a = 12.35 f0.08,b = 5.44 rt 0.04 c = 31.75 & 0.10 A; a = 94.0" & 0.5" 18 = 96.7" f 05" y = 99.2" f0.5". The triclinic unit cell contains two centrosymmetrically related mole-cules. The structure for the bromo-glyceride was solved from bromine-phased electron-density projections with bromine positions derived from Patterson pro- Vand and Bell Acta Cryst.1951 4 465. Larsson Acta Gem. Scand. 1962 16 1751. races (cf. Figure). This is a consequence of the fact that the chains do not possess translational freedom in relation to each other. The unit-cell dimensions and the x-and z-co-ordinates of the carbon atoms in the chains are in general accordance with Vand and Bell's results1 on trilaurin. There is however an error in their matrix s which gives the transformation between the subcell and main-cell co-ordinates. This is obvious if the main-cell co-ordinates are calculated by using this matrix as the lateral packing of the chains then becomes irregular. The element denoted sfshould be -0,290 instead of -0.241. Their calculation of the PROCEEDINGS ~ ~___~ subcell dimensions is based on this matrix and there- This work is financially supported by a grant from fore contains smaller errors.the National Institute of Arthritis and Metabolic The structure is now being refined by anisotropic Diseases US.Public €kalth Service. least-squares methods. (Received December 1Oth 1962.) The Three-co-ordinate Boron-Nitrogen Four-membered Ring System (1,3-Diaza-2,4-boretane) By M. F. LAPPERT and M. K. MAJUMDAR (DEPARTMENT FACULTY UNIVERSITY OF CHEMISTRY OF TECHNOLOGY OF MANCHESTER) OF the borazynes (XB=NY) and their oligomers The diazaboretanes are of interest also because only the cyclic trimersl (borazoles) and recently the formally they may be regarded as isoconjugate with tetramers2 (borazocines) have been well character- the cyclobutadienes.The instability of the latter has ised;polymeric presumably linear derivatives have been explained in terms of molecular-orbital theory.6 We now report the preparation Compound (I) proved to be diamagnetic implying also been de~cribed.~ of a cyclic dimer (I) b.p. 72-74"/0405 mm. ng that delocalisation within the ring is not significant. 1.4582 di0 0-8666. Typical reactions leading to its Indeed we believe that n-bonding is largely exocyclic formation are indicated below. Compound (IV) b.p. and that this is responsible at least in part for the 72-74'110 mm. n2,0 1.4330 obtained so far in only stability of compound (I). N 90% purity was made from tris-t-butylamino- R borane* and boron trichloride ;an alternative method of preparing compound (111) b.p.82-86"/0.02 mm. n2,0 1.4548 dp 0.9014 was from the trichloride and R t-butylamine; compound 01)had b.p. 98-100"/0~01 mm. m.p. 52-54" nz 1.4630. Compounds (I)--(IV) represent novel classes of boron compounds (see also ref. 5). The evidence is spectroscopic and indicates that CL NHBU' compound (I) is a mixture of geometrical isomers (Ia) and (Ib). Thus the infrared spectrum shows a (poorly resolved) doublet at -3460 crn.-l (NH stretching frequency) and the lH nuclear magnetic resonance spectrum indicates that the methyl absorption consists of a quartet. (Ed-NH.8. NBu' -)* A H H (1) [-B (NH ELI (a a compound having a B-N-B skeleton; (ii) an intra- molecular 1,3-nucleophilic rearrangement ; and (iii) We are grateful to Messrs.R. A. Saunders A. E. polymerisation of the borazyne. This is illustrated Williams and Imperial Chemical Industries Limited for the case of a trisaminoborane. Dyestuffs Division for the mass-spectral data Mr. Evidence for structure (I) rests on full elemental P. A. Barfield and Dr. J. Lee for the nuclear- analysis molecular-weight determinations (cryo- magnetic data and Mr. N. Paddock for the magnetic scopic and ebullioscopic in benzene) and the mass measurement. spectnun. (Received November 28th 1962.) Cf. Sheldon and Smith Quart. Rev. 1960,14 200; Mikhailov Uspekhi Khiin. 1960,29 972. a Turner and Warne Proc. Chem. Soc. 1962,69. Burg and Banus J. Amer. Chem. SOC.,1954,76 3903; Bissot Campbell and Parry ibid.1958 80 1868; Burch Gerrard and Mooney J. 1962,2200. Aubrey and Lappert J. 1959,2927. 6Noth 2. Naturforsch. 1961 16b 618. 6 Longuet-Higgins and Orgel J. 1956 1959. MARCH1963 The Structure of Elsinochrome A By T. J. BATTERHAM and U. WEISS (NATIONALINSTITUTE AND METABOLIC INSTITUTES OF ARTHRITIS DISEASES,NATIONAL OF HEALTH,PUBLIC DEPARTMENT EDUCATION BETHESDA,MARYLAND) HEALTHSERVICE OF HEALTH AND WELFARE A NUMBER of species of the fungus Elsinot? (Asco-mycetes) and its conidial stage Sphaceloma,produce mixtures of red pigments with absorption spectra similar to that of erythr0aphinfb.l We propose the name “elsinochromes” for these compounds. Thin- layer chromatpgraphy of crude extracts from E. aiznonae showed the presence of at least five related pigments.After separation by chromatography on silica gel the least polar member of the series elsino- chrome A (I) crystallised from benzene-hexane as dark-red plates C26H,,0,(OMe), m.p. 255”. We propose the rapidly interconvert ing taut omeric structures (Ia) and (Ib) for this pigment. Compound (I) shows infrared bands at 1712 and 1623 cm.-’ which we assign respectively to the two identical methyl ketone groups and the hydrogen- bonded extended quinone system. The presence of the 4,9-dihydroxyperylene-3,lO-quinonenucleus is shown by comparison of the absorption spectra of compound (I) [Amax. (in CHCl,) 445 460 532 572mp] and its leuco-acetate [Xm,x. (in CHCl,) 277 294 437 469 500 mp] with those of erythroaphin- fb2 [Amax.(in CHCI,) 447 485 520 560 586 mp] and its leuco-acetate [Amax. 278,436,465,498mp]. The strikingly simple nuclear magnetic resonance spectrum of compound (I) with only six single peaks confirms the presence of two methyl ketone groups (T 7.95) and shows the presence of two pairs of methoxyl groups (T 5.68 6*00) two ring protons (T 3.48) two alicyclic protons (T 4.86) and two hydroxyl groups (T -5.9). The extremely sharp signal from the hydroxyl protons can be explained as a combination signal from the rapidly interconverting tautomers (Ia and Ib). Methylation (dimethyl sul- phate and potassium carbonate in acetone) confirms this interpretation by producing the isomeric di- methyl ethers (IIa) [vmax. 1715 1632 cm.-l; T (quinonoid) 3.821and (IIb) [vmax.1715,1632 cm.-l; T (aromatic) 3.101. In hot polar solvents (e.g. methanol or acetic acid) the ethers (Ha) and (IIb) disproportionate readily into the corresponding dehydro-ethers (IIIa) [Vmax. 1685 1632 cm.-l; T (CO.CN,} 7.36 (quinonoid) 3.751and (IIIb) [vmax. 1700,1630cm.-l; T (COCHa 7.30 (aromatic) 3.31 and quinols. The latter in methanol are converted into purple insoluble pro- ducts; in acetic anhydride-acetic acid however they can be trapped as acetates. Peaks at T 3-75 7.36 and T 3.25 7-30 in the nuclear magnetic resonance spectra of the dehydro- compounds (IIIa) and (IIIb) respectively can best be explained if in the former both ring protons and methyl ketone groups are attached to the quinonoid system while in the latter the same groups are attached to a condensed aromatic system Conver- sion of hydroxy-quinone (I) into the methoxy- #H..COMe COMe Me0 o.H,o quinone (IIa) does not alter the absorption spectrum greatly. However the isomer (IIb) shows a shift of the main band of the phenol (I) (445mp) to 435 mp (reduced intensity) and disappearance of bands at longer wavelengths. These effects are due to twisting of the extended quinone by the alicyclic group. In the Weiss Flon and Berger Arch. Biochem. Biophys. 1957 69 311. Calderbank Johnson and Todd J. 1954 1285. PROCEEDINGS ether (IIa) the ring proton is quinonoid (T 3.82) while easily; hence permanent optical activity can best be in the isomer (IIb) it is aromatic (T 3.10).These con- ascribed to a trans-arrangement of groups in the siderations place the methoxyl groups as shown in alicyclic portion of the molecule. (Ia) and (Ib). On biogenetic grounds compound (I) could have Compound (I) is optically active (negative Cotton been formed by dimerisation of an acetate-derived effect centred at -425 mp). Dreiding models show unit related to javanicin (cf. the biosynthesis of crowding of the perylene ring system by substituents hypericin3); the necessary removal of carboxyl-at positions 1 and 12 and 6 and 7. This could produce derived carbon and introduction of oxygen are optical isomers which however should racemise unexceptional. (Received,January 30th 1963.) Brockman Pohl Maier and Haschad Aitnalen 1942 553 1.Addition of Dichlorocarbene to Norbornylene L. GHOSEZ and P. LAROCHE (LABORATOIRE UNIVERSITY OF LOUVAIN, DE CHIMIE MACROMOL~CULAIRE BELGIUM) IT is known that dihalogenocarbenes and olefins ketone (VI) v 1692 1597 cm.-l (2,4-dinitrophenyl-readily yield gem-dihalogenocyclopropanes,l that hydrazone). cyclohexenel and cycloheptene2 give very stable ad- Methyl endo-norbornylene-5-carboxylategave a ducts but that dibr~mo-~ and dichloro-carbene2 product whose analytical and spectroscopic proper- adducts of cyclopentene undergo ring expansion ties indicated a mixture of the isomeric compounds under mild conditions. We find that the reaction of (VII) and (VIII); with silver nitrate in water or dichlorocarbene with a strained olefin such as nor- methanol it yielded a $-unsaturated chloro-lactone bornylene yields a rearranged addition product2 (IX),m.p.84-85" trmax. 1787 cm.-l. (111). r-Adding methyl trichl~roacetate~ to sodium meth- oxide and norbornylene in pentane at 0" gave a liquid C8H10C12 b.p. 86-87"/3 mm. n2,5 1.53227 in yields up to 50%. The rearranged structure (III) was proved as follows. The product was unsaturated (permanganate; vmax. 1624 cm.-l). Proton magnetic resonance peaks occurred at r 8.27 (six CH,) 7.29 (two bridgehead H) 5.88 (one tertiary allylic H at a rather low field because of the chlorine atom bonded to the same carbon atom) and 3.92 (one olefinic H); the expected spin splitting patterns were found for protons 2 and 4 both adjacent to a single hydrogen atom.Sodium in moist methanol1 reduced the com- pound to bicyclo [3,2,1 Ioct-Zene (IV). As expected C0,Me for (111) only one chlorine atom was reactive towards dilute aqueous silver nitrate giving an unsaturated (VlO (Vl1 r) chlorhydrin (V) m.p. 40-42"C Vmax. 1627 cm.-l By formation of compound (111) could Cp-nitrobenzoate) obtained also by refluxing the be explained by the rearrangement of an initial adduct in aqueous acetone. The infrared spectrum of adduct such as (II) but we have not yet been able a dilute solution (CCl,) of the chlorhydrin showed a to prove the existence of this intermediate. single peak at 3600 crn.-l; in concentrated solutions intermolecular association was comparatively low. We thank Professor G. Smets Dr. G. Slinckx the These observations suggest an intramolecular inter- Fonds National de la Recherche Scientifique and action between the hydroxyl group and the chlorine the Institut pour 1'Encouragement de la Recherche atom which could only occur if they are vicinal as Scientifique dans 1'Industrie et I'Agriculture (P.L.) in (V).The chlorhydrin (V) was smoothly converted for help and support. by chromic anhydride in pyridine5 into a conjugated (Received December 3 lst 1962.) Doering and Hofman J. Amer. Chem. SOC.,1954,76,6162. Bergman Abs. Papers 142nd Meeting Amer. Chem. SOC. Atlantic City N.J. Sept. 1962 p. 790. Sonnenberg and Winstein J. Org. Chem. 1962 27 748. Parham and Schweizer J. Org. Chem. 1959,24 1733. Poos Arth Beyler and Sarett J. Amer. Chem. Suc. 1953 75 422.MARCH1963 Kinetics of Cross-linking Reactionsin Solids By D. H. EVERETT and E. REDMAN (DEPARTMENT AND INORGANIC UNIVERSITY OF PHYSICAL CHEMISTRY OF BRISTOL) WHENpoly(viny1idene chloride) [-CH,~CCl,-], is have now carried out experiments to test our hypo- pyrolysed in an inert atmosphere at about 170°c thesis. We assume for simplicity that the cross- decomposition tends to a limit corresponding to the loss of about 50% of the hydrogen chloride although over long periods a slow subsequent reaction can be detected. If the temperature is raised say by 40° rapid decomposition sets in again only to reach another limit around 65% of reaction. Further in- creases in temperature lead to a series of steps in the graph of hydrogen chloride loss against time and the reaction is not completed in a reasonable time until a temperature of at least 700”is rea~hedl-~ (Fig.1). The first stage of reaction corresponding to the loss of 50% of the hydrogen chloride is an accurately first-order process3 and is thought1 to be the elimination of HCl between adjacent carbon atoms to form a vinyl chain [-CH=CCl-I,. Further loss probably occurs by cross-linking which may involve a Diels-Alder mechanism. I I I 1 I c _1 10 20 3040 50 Time (hours) FIG. 1. Percentage decomposition of poly(viny1- fdene chloride) as function of time 50 decom-position reached at (0)175” then three samples of this material further decomposed at (A) 211*5” (B) 22705”~ and (C) 295.7”. Dacey and Cadenhead2 suggested that the behaviour shown in Fig.1 might be explained by a shift in the equilibrium position of the cross-linking reaction with temperature. This seems unlikely and we put forward an alternative explanation3 which assumed a rapid change in the activation energy of the cross-linking reaction with extent of reaction. We linking reaction is of the first order in vinyl units and that the free energy of activation (dGS) increases linearly with the number of cross-links already formed. If the reaction can be divided cleanly into two stages then when the second stage has proceeded to a fractional extent 6 the number of remaining vinyl units is proportional to (1 -f)and the number of cross-links to t.The rate equation is then d&dt = A(l -5) exp[-(dG$ + af)/RT].(1) A graph of log{ ($)/(1 -4)} against 6 should be linear with a slope of a/2*303RT.A typical graph of our results is shown in Fig.2; d&dt was obtained by seven-point numerical differentiation.* The 3.5-B ‘r‘ t. 5 FIG.2. Graph of log { (d[/dt)/( 1 -5) against 5 for second stage of reaction at 227.5”-dominating effect of the change of activation free energy with extent of reaction makes it impossible to test the order of the reaction equally good lines are obtained by assuming second- or third-order kinetics. Nor will other details of the mechanism such as the possible need in the condensation for the simul- taneous interaction of two adjacent vinyl groups on Winslow Baker and Yager Roc.1st and 2nd Buffalo Conference on Carbon Buffalo N.Y. 1956 p. 93; Winslow, Matreyek and Yager “Industrial Carbon and Graphite,” Soc. Chem. Ind. London 1958 p. 190. Dacey and Cadenhead Proc. 4th Buffalo Conference on Carbon Buffalo N.Y. 1960 p. 315. Everett Redman Miles and Davies 4th Internat. Conference of Coal Science LeTouquet 1961 ;to be published in Fuel. * Comrie in Chamber’s “Shorter Six-Figure Mathematical Tables,” W. and R. Chambers Ltd. 1961 p. 349. one chain with a vinyl group on another chain be detectable in the overall kinetics. Precise determination of the slope d&dt is difficult and a better test of eqn. (1) is to compare its inte- grated form alRT where B == A exp{-(dGo$ + a)[RT),with the actual experimental results.The integrals are standard exponential integrals whose values are tab~lated.~ For a given value of a Bt can be cal-culated as a function of 6 and this curve should superimpose on the experimental plot of 4' against f when the abscissa is divided by B. Fig. 3 shows this comparison with a = 1.79 x 1oQ cal.mole-l and B = 8.77 x 10-lO min.-l for the experiment in which polymer decomposed to 50% at 175" was subse- quently heated to 227". Eqn. (2) represents the experi- mental data very closely. The above value of QI implies that the activation free energy changes by 17.9 kcal. between 4 = 0 and 6 = 1. Analysis of the PROCEEDINGS experiments at other temperatures is needed to separate this into changes in the energy and entropy of activation.t Time (hours) FIG. 3. Comparison of experiment with curve calculated from eqn. (2) for second stage of reaction at 227.5'. Crosses experimental; closed circles calculated. It is suggested that the above method of analysis may be applicable to other cross-linking reactions in solids and in particular to the graphitisation of carbon. (Received January 30th 1963.) Jahnke and Emde "Tables of Functions," Dover Publns. Ltd. N.Y. 4th edn. 1945 p. 1; these tables had to be extended by series summation to cover the range needed in this analysis. The Structure of Bromoisotenulin By D. ROGERS and MAZHAR-UL-HAQUE (CHEMISTRY DEPARTMENT IMPERIAL COLLEGE LONDON,s.w.7) TENULIN, which was isolated by Clark1 from various Helenium species and studied by Barton and de Mayo? Was assigned fOrmUh (I) that of isotenub being @I) but nuclear mametic ~~Onance studies3 have thrown doubt on these fOrmUl2- The COfleCt constitutions and relative stereochemistry have now ~o~ Oq-$ 0 Brq?J-&@ r+" OAc been determined by an X-ray study of bromoiso- tenulin4(111).The carbon skeleton is biogenetically abnormal a methyl group having migrated from position 4to 5; both five-membered rings are trans-fused to the seven-membered ring ; non-bonded repulsions be- tween the angular methyl (a to C-5) and the a-hydrogens on C-8 and C-10 cause the molecule to be appreciably folded. A Dreiding model shows this interaction clearly and readily folds into the form found. Bromoisotenulin (C,,H,,Br05) crystallises in the monoclinic system with 4 molecules in a cell of dimensions a = 8.75 b = 23.15 c = 10.28 A ,8 = 121"; the space group being P2, the asymmetric unit comprises two crystallographically unrelated molecules.Some 2300 independent reflextions were measured visually and the positions of the two sets l Clark J. Amer. Chem. SOC.,1939,61 1836; 1940,62 597. a Barton and de Mayo J. 1956 142. a Herz Watanabe Miyazaki and Kishida J. Amer. Chem SOC.,1962 84 2601. Clark J. Amer. Chem. Soc. 1939,61 1840. MARCH1963 __ ___ -__ __ __~_ of bromine atoms deduced from the three-dimen- between the two unrelated molecules in this study and sional Patterson function were such as to avoid partly because of a recent chemical study by Hen phase ambiguity.The first electron-density synthesis et a23 which though devoid of stereochemistry revealed all 44 carbon and oxygen atoms. Refine- accords with our findings. ment not yet complete has reduced R to 0.21 for all observed reflexions and shows satisfactory agree- We are indebted for assistance and support to ment between bond lengths in the two molecules and Professor D. H. R. Barton to Dr. J. T. Pinhey to identical stereochemistry. We publish this Communi- Woolwich Polytechnic to the British Council and to cation at this stage partly because of the concordance the Department of Scientific and Industrial Research. (Received December 21st 1962.) ti Hen Rohde Rabindran Jayaraman and Viswanathan J. Amer. Chem. Soc. 1962,84,3857.The Trifluoroacetyl Radical By B. G. TUCKER and E. WHITTLE DEPARTMENT COLLEGE, (THECHEMISTRY UNIVERSITY CARDIFF) THEacetyl radical is well known as an intermediate CF3.CO 4 CF + CO * .(3) in the photolysis of acetone but the analogous tri- fluoroacetyl radical has not been detected in the In the presence of bromine the trifluoroacetyl photolysis of hexafluoroacetone; the primary re- radicals may react to give trifluoroacetyl bromide action is thought to be particularly at lower temperatures. It can be shown that CF,CO*CF + hv -2CF3 + CO . . (1) R(CF,COBr) = R(CF,.Br) -2R(CO) .(4) If reaction (1) is correct we should expect the photolysis of a mixture of hexafluoroacetone and where R denotes rate of formation. In the Table we bromine to yield only trifluoromethyl bromide and compare R(CF,COBr) calculated by using equa- carbon monoxide in the ratio 2 1 provided that the tion (4) with the observed values; the agreement is trifluoromethyl radicals react rapidly with bromine.reasonable as the trifluoroacetyl bromide analyses The results of such photolyses are shown in the were not of high accuracy. Table. The main conclusions are (1) the ratio Apparently the trifluoroacetyl radical exists even CF,Br/CO is greater than 2 and there is very little at 200" and the failure to detect hexafluorobiacetyl Temp. (OK.) Products ( molecules ern.- sec.-l) co CF3Br (CF3.COBr)exp. (CF3- COBr)calc. 294 0.076 0.71 0.42 0.56 357 0.162 1-57 1.15 1-25 385 0.405 1.91 1-06 1.10 449 1-54 3.99 0.82 0.91 524 2.59 5.08 0.44 0 P(Br-& = 5 mm.; P(HFA),,.= 22 mm. Photolysis times 30-250 min. carbon monoxide at the lower temperatures and from the photolysis of hexafluoroacetone alone is (2) the products contain trifluoroacetyl bromide probably because of the instability of the compound. which was identified and estimated by means of its If the hexafluoroacetone is photolysed in the presence infrared spectrum. of nitric oxide the only products are trifluoro- These observations strongly suggest that the nitrosomethane and carbon monoxide in the ratio primary process is 2 1 ;l presumably the trifluoroacetyl radical does not react with nitric oxide. CF,*CO*CF,+ hv 4 CF,.CO + CF3 .(2) We thank D.S.I.R. for a grant (to B.G.T.). followed by (Received February 8th 1963.) Charles and Whittle unpublished results.PROCEEDINGS Photolytic Generation of Aromatic Radical-anions:EIectron-spin Resonance Studies By P. B. AYSCOUGH and F. P. SARGENT (DEPARTMENT CHEMISTRY LEEDS,2) OF PHYSICAL TEEUNIVERSITY NUMEROUS observations of the electron-spin reson- ance spectra of the radical-anions of aromatic hydro- carbons ketones and nitro-compounds have been reported recently. Preparative methods include (1) re-duction by an alkali metal in an ethereal ~olvent,l-~ (2) electrochemical reduction in both aprotic and protic solvents,3-s and (3) reduction by zinc or glucose in alkaline solutions. The use of ethanol methanol or water as solvent in the latter procedure has the advantage of reducing the line width and permitting more detailed interpretation of the spectra some new features of which are reported elsewhere.6 Here we report that these same species can be produced in high yield by ultraviolet irradia- tion of aromatic nitro-compounds or ketones in con- centrated sodium ethoxide methoxide or hydroxide solutions.This provides an even more convenient method of preparation especially where kinetic studies are to be undertaken since the rate of reduc- tion and the concentration are readily controlled by means of the light intensity. All the compounds studied (twenty nitro-com- pounds and six ketones) have strong absorption bands in the ultraviolet region corresponding to n-+r* or T-T* transitions. No electron-spin resonance spectra are observed at room temperature in the absence of ultraviolet irradiation.During photolysis the concentration of radical-anions rises to a maxi-mum in 10-30 minutes for most of the nitro- compounds but in a few seconds only in the case of benzophenone. When the light source is removed the fall in concentration follows a similar time-scale (compare ref. 7). Reducing the temperature increases the net rate of formation of radical-anions increases the limiting concentration and decreases the rate of decay. Removal of oxygen from the solution has a similar effect. The electron-spin resonance spectra are identical with those observed during chemical or electrochemical reduction in the same solvents and are attributed to the primary radical-anion in each case.We therefore presume that we are observing the transfer of an electron from the alkoxide or hydroxide ion to the vacant orbital of the electronically excited aromatic compound e.g. Ph-NO,* + OR-+ Ph*NO,-+ *OR Ph,CO* + OR-+ Ph,CO-+ .OR The thermally activated reaction of nitrobenzene with alkoxide ions which can be observed above a" gives the same species but increases rapidly in rate with temperature and is therefore easily distinguished. Similar effects are observed on using higher alcohols though thermal reactions become increasingly im- portant and the spectra are not so well resolved. Irradiations were carried out inside the cavity of a Varian V-4500EPR spectrometer with light from a 250-w medium-pressure mercury lamp from which infrared radiation had been removed by means of a water filter.We are grateful to D.S.I.R. for financial assistance. (Received,January 25th 1963.) Ward J. Chem. Phys. 1959,30 852; 1960,32,410; 1962,36 1405. * Atherton and Weissman J. Amer. Chem. Suc. 1961 83 1330. Ayscough and Wilson Pruc. Chem. Soc. 1962 16. I; Geske and Maki J. Amer. Chem. Suc. 1960 82 2671; 1961 83 1852. Piette Ludwig and Adams Analyt. Chem. 1962,34 916. Ayscough Sargent and Wilson in the press. Porter and Wilkinson Trans. Faraduy Soc. 1961,57 1686. Direct Preparation of Some Functional Fluoroaromtic Compounds By G. M. BROOKE J. HEYES and W. K R. MUSGRAVE R. D. CHAMBERS (THEDURHAM IN THEUNIVERSITY COLLEGES OF DURHAM) OFthe routes to hexafluorobenzene,l the only one passage over a heated iron gauze giving hexafluoro- suitable for large-scale work requires fluorination of benzene and polyAuorobenzenes.2 benzene with cobalt trifluoride and dehydrofluorina- We now report the first simple preparation of fully tion of the mixed products to polyfhorocyclo-substituted chlorofluorobenzenes as well as hexa- hexadienes.These dienes have been defluorinated by fluorobenzene by a method which can be used on a Johncock Mobbs and Musgrave Znd. Eng. Chem. Process Design and Development 1962,1,267; Banks Birchall Hazeldine Simm Sutcliffe and Umfreville Proc. Chem. Soc. 1962 281 ;and references therein. * Gething Patrick Tatlow Banks Barbour and Tipping Nature 1959 183 586; Gething Patrick Stacey and Tatlow ibid.p. 588; Coe Patrick and Tatlow Tetrahedron 1960 9 240. MARCH1963 large scale. Passage of a three-molar proportion of fluorine diluted with nitrogen through a stirred slurry of hexachlorobenzene in 1,1,2-trichlorotrifluoro-ethane at room temperature gave a mixture of per- chlorofluorocyclohexanes of average composition C&&F, id. C&l,F,,- where X = 4-7 in high yield whereas only small amounts of materials C6Cl6F6 and C,CI,F were isolated previously from a similar reaction (under different experimental con- dition~).~ Dehalogenationof the mixture of perchloro- fluorocyclohexanes by passage over hot iron gave hexafluorobenzene chloropentafluorobenzene and dichlorotetrafluorobenzenes in high overall yield. CCI,FCCI F Fe The method may be adjusted to give a larger propor- tion of hexafluorobenzene by further fluorination of the mixture of perchlorofluorocyclohexanes at higher temperatures before dehalogenation.Variable results were obtained by the iron-gauze technique but remarkably consistent dehalogenation was effected by using a cobalt fluoride reactofl in which the cobalt fluoride had been replaced by iron filings. The filings were discarded when spent (after the passage of over 1 kg. of material) so avoiding the uncertain process of regeneration. We have shown that chlorofluorobenzenes are useful functional compounds e.g. chloropenta-fluorobenzene formed a Grignard reagent in diethyl ether when activated by ethylene dibromide and gave 67 % of pentafluorobenzene on hydrolysis.Reaction of the Grignard reagent with methylmercuy(Ii) iodide gave 50 %of pentafluorophenylmethylmercury and carbonation gave 41.X of pentafluorobenzoic acid after preforming the reagent in diethylether and then changing the solvent to tetrahydrofuran before the passage of carbon dioxide therefore showing similar reactivity to penta.fluorophenylmagnesium br~mide.~ When formation of the Grignard reagent in tetrahydrofuran was attempted high molecular weight fluoro aromatic material was obtained in high yield which did not melt below 360". We thank the Imperial Smelting Corporation for a maintenance grant (to J.H.) and the Imperial Chemical Industries Limited for the loan of a fluorine cell. (Received,January 16th 1963.) Bigelow and Pearson J.Amer. Chem. Soc. 1934,56,2773. Barbour Barlow and Tatlow J. Appl. Chem. 1952,2 127. Chambers Coates Livingstone and Musgrave J. 1962,4367; Harper and Tamborski,Chem. and Ind. 1962,1824. Electron Spin Resonance Spectrum of the XeF Radical By W. E. FALCONER and J. R. MORTON OF APPLIED NATIONAL OTTAWA (DIVISION CHEMISTRY RESEARCHCOUNCIL 2 CANADA) BARTLETT'S recent report1 of the first stable xenon compound xenon hexafluoroplatinate has revived interest in the chemistry of the inert ga~es.~,~ We here report the existence of a xenon-containing radical XeF the electron spin resonance spectrum of which has been obtained at 77"~. The crystalline nature of xenon tetrafluoride2 sug- gested the possibility of stabilising in the lattice xenon-containing paramagnetic fragments which could be studied by electron spin resonance.Xenon tetrafluoride was prepared as recently described2 and a small single crystal was grown by sublimation. This had the form of an elongated hexagonal plate with prismatic ends; it was irradiated at 77"~ with a 5 Megarad dose of 1.3 MeV s°Co prays. The ir- radiated crystal was blue and exhibited a powerful paramagnetic resonance spectrum when examined at 77"~ in the cavity of a Varian V-4500 spectrometer. The spectra were attributed to the species XeF and the spectrum obtained with the in-plane transverse axis of the crystal parallel to the magnetic field of the spectrometer is reproduced in the Figure. The two strongest lines in the spectrum are due to XeF radicals containing xenon isotopes of zero spin;* they are separated by the hyperfine interaction of the 19Fnucleus (i = 1/2).Hyperfine interaction with xenon nuclei 129Xe and 131Xe (i = 1/2 and 3/2, respectively) is also apparent although there is some * The symbol lseXe is used to represent all the zero-spin isotopes which together constitute 52.4% of naturally occurring xenon and have the even mass numbers 124 to 136 inclusive. Bartlett Proc. Chem. Soc. 1962 218. Claassen Selig and Malm J. Amer. Gem. Suc. 1962 84 3593; Chernick Claassen Fields Hyman Malm Manning Matheson Quarterman Schreiner Selig Sheft Siegel Sloth Stein Studier Weeks and Zirin Science 1962, 138 136. Weeks,Chernick and Matheson J.Amer. Chenz. Soc. 1962 84 4612. overlapping of the lines from the different isotopic species. The relative intensities of the three groups of lines are consistent with the known isotopic distribu- tion in xenon. Moreover the lzsXe splitting exceeds that of 131Xe in accord with their relative magneto- gyric ratios. PROCEEDINGS crystal. The lack of centrosymmetry apparent in the above spectrum and the deviation from the free-spin g-value suggest that second-order terms in the Hamiltonian are imp~rtant.~ The application of second-order theory requires data as yet unobtained from spectra for other crystal orientations and there- 1 1 =XeFetc. Electron spin resonance spectrum of XeF h a y-irradiated single crystal of XeF at 77"~.The spectra were virtually isotropic provided the fore precise hyperfine interaction constants and g-magnetic field explored a plane perpendicular to the values cannot at present be given. However from the longitudinal axis of the crystal but were otherwise above spectrum approximate hyperfine splittings for highly anisotropic in the magnetic-field direction. the three nuclei 19F,129Xe and 131Xe are respectively These observations imply that the Xe-F bond is 180 425 and 125 Gauss. probably parallel to the Iongitudinal axis of the (Received January 14th 1963.) Horsfield Morton and Whiffen Mol. Phys. 1961 4 475; Chantry Horsfield Morton Rowlands and Whiffen ibid. 1962 5 233. Preferred Formation of the cis-Olefin in BimolecularElimination By J. ZAVADAand J.SICHER (INSTITUTE CHEMISTRY CZECHOSLOVAK OF ORGANIC AND BIOCHEMISTRY ACADEMY PRAGUE) OF SCIENCE INbimolecular eliminations compounds of the type It is recognised that the mechanism of bimolecular (I)generally give an olefin mixture in which the trans-elimination on simple open-chain 'onium salts may isomer (II) pred0minates.l We now report several range from the fully coupled E2processes (transifion- cases of elimination from 'onium compounds (I; R state conformations A or B) to E,cb-tike processes = Prn R = Bun; X = Me,N+ or Me&+) which (transition state conformations C and 0). Banthorpe preferentially give the cis-olefin 01). The reaction Hughes and Ingold2 concluded on the basis of conditions and results are tabulated. theoretical considerations that for 'onium salts an (I) RCH,-CHX*R' RCH =CH*R' (11) E,cb-like mechanism should be more favoured in Reaction of C3H,CH2CHX-C4H Me,N+ Me2S+ A X f \\ Expt.1 2 3 4 5 Base/Solvent Bu@-Bu@H EtO-EtOH MeO-MeOH Bu@-ButOH EtO-EtOH cis-(11)(%)* 26 74 81 9 64 * Percentage of cis-(@ in the cis-trans-mixture determined by vapour-phase chromatography on a silver nitrate- triethylene glycol column at 40"by Mr. L. BaStAf. The faster-moving isomer was shown to have the trans-configuration in agreement with expectation. Reviews Ingold Proc. Chem. Soc. 1962,265; Bunnett Angew. Chem. 1962,74,731. a Banthorpe Hughes and Ingold J. 1960,4054; we are indebted to Sir Christopher Ingold for a manuscript of this paper before its publication. MARCH1963 comparison with the fully coupled E2mechanism in ethanol than in t-butyl alcohol and in any one sol- vent more for ammonium than for sulphonium compounds.Since this is the order in which the per- centage of the cis-isomer increases in our experiments (i.e. expt. 1-2-3; 4-5; 5-2; 4-1) the preferred forma- tion of the cis-olefin can be regarded as a consequence of an E&-like mechanism. A fully coupled E2 reaction should give pre- ferentially the trans-olefin (eclipsing between R and R’ being smaller in A than in B). If the transition states C and D in which the leaving groups are anti- periplanar represented the only alternatives for an E&li ke process then this mechanism would-for the same reason as with the fully coupled E2reaction -again give preferentially the trans-isomer.It is hence unlikely that the reaction in experiments 2 3 and 5 proceeds to a major extent by way of the transi- tion states C and D. A reaction path proceeding by way of transition state E which could lead to the cis-isomer must therefore be considered. In E the non-bonded interactions are undoubtedly H smaller than in C or D the bulky ’onium group here being flanked by hydrogen atoms.* The choice of transition state E would hence indicate that in E,cb-like processes advantages accruing from antiperi- planarity are no longer necessarily decisive; in our reactions the molecule apparently sacrifices the advantages of antiperiplanarity (cf. C,D) for a sterically less encumbered transition state such as E.(Received January 1 lth 1963.) * Ukaji and Bonham (J. Amer. Chem. Soc. 1962 84 3631) recently found by electron refraction that for s-butyl chloride there is essentially no difference in energy between the conformers of the type C and D (R = R‘ = Me X = Cl); it seems very unlikely that this would be the case where X is the bulky (and solvated) ’onium group. An Intermediate in Homolytic Aromatic Substitution By W.T.DIXON and R. 0. C. NORMAN (DYSON PERRINS LABORATORY, OXFORD UNIVERSITY) BENZENEreacts with titanous ion and hydrogen peroxide to give phenol and biphenyl. We here report evidence that the resonance-stabilised radical (I)is an intermediate. H OH n (1) Warm acidified aqueous solutions of titanous ion and hydrogen peroxide each saturated with benzene were alluwed to react in a flow system less than 0.02 sec.before entering a cell in the cavity of a 100 kc./sec. Varian V4500 electron spin resonance spectrometer. The resulting spectrum (Figure) re- placed the signal attributed to the hydroxyl radical which we have observed in the absence of benzene.l The reconstruction shown is based on coupling with two single protons (coupling constants 36.0 and 13.4 gauss) (& 5%) and two pairs of protons (coupling constants 9.3 and 2.9 gauss) (f5 %). This is consistent with the adduct (I) in which coupling would be expected with the six protons on carbon atoms but not with that on oxygen? When phenol was used in place of benzene the formation of the phenoxy-radical was confirmed by its spectrum3 (coupling constants o-H 6.4; m-H 1.7;p-H 9.7 gauss).This is almost identical in form with half the spectrum ascribed to radical (I), as expected from examination of the appropriate canonical structures and leads to the assignment of three of the coupling constants for radical (I) (9.3, 2-9,and 13.4 gauss) to 0-,m-,and p-protons respec- tively. The smaller couplings for the phenoxy-radical arise from the additional possibility of delocalisation of the unpaired electron on to the oxygen atom. The largest splitting observed for radical (I) is evidently due to the proton on sp3-carbon which is favourably situated for interaction with the n-electron system in which the unpaired electron is found.Dixon and Norman. Nature. 1962. 196. 891. * Ingram “Free Radicals as Studied by’Electron Spin Resonance,” Butterworths Scientific Publications London 1958 p. 174. Stone and Waters Proc. Chem. SQC.,1962 253. Our assignments for radical (I) receive further support from the similarity of the coupling constants I I I I I I I Spectrum of the intermediate and reconstruction based on the coupling constants given in the text. PROCEEDINGS to those of the related radical cyclohexadienyl (CH, 50; o-H 10.6; m-H 2.6; pH 10.6 gauss)? The formation of radical (I) is consistent with the absence of a hydrogen isotope effect in the formation of both phenol and biphenyl when benzene is oxi- dised with Fenton's reagent which behaves similarly to the system of Ti% and H202.s Finally the radical (I) is analogous to the intermediate postulated in the phenylation of benzenoid compounds on the basis of kinetic isotope effect measurements6 and product analysis.' One of us (W.T.D.) thanks the D.S.I.R.for a main~~nance @ant* (Received December mth 1W2.) Fischer J. Chem.Phys. 1962,37 1094. Lindsay Smith and Norman unpublished observations. Convery and Price J. Amer. Chem. Soc. 1958,80,4101; Chang Shih Hey and Williams J. 1959 1871. DeTar and Long J. Amer. Chem. Soc. 1958,80,4742. NEWS AND ANNOUNCEMENTS Editorial Appointments.-Dr. R. S. Cahn the Society's Editor since 1949 has been appointed to the newly created post of Director of Publications Research with the task of conducting a Survey of Chemical Publications (see page 73).Dr. L. C. Cross Deputy Editor has been appointed Editor as from September Ist 1963 when Dr. Cahn's new post will become a full-time appointment. Mr. G. P. Pullard has been appointed Senior Assistant Editor and will take up his duties on April 17th. Dr. N. A. Keen was appointed as an additional Assistant Editor from January lst 1963. LibnUy.-The Library will close for the Easter Holiday from 9 p.m. Thursday April 1 lth until 9.30 a.m. Wednesday April 17th 1963. Local Representatives.-The Council has approved the following changes of Local Representatives Birmingham . . Dr. E. J. Forbes in place of Dr. A. B. Foster Bristol . . . . Dr. W.D. Ollis in place of Dr. R. Parsons Cardiff ,.. . Dr. J. H. Thomas in place of Dr. A. R. Pinder Durham . . . . Dr. H. M.M. Shearer in place of Dr. F. Glockling Northern Ireland . . Dr. H. G. HeaE in place of Dr. M. F. Grundon Tees-side . . . . Dr. L. A. Duncanson in place of Dr. I. J. Faulkner The Corday-Morgan Medal and Prize.-The Council of the Chemical Society has awarded the Corday-Morgan Medal and Prize to Professor Franz Sondheimer Professor and Head of the Organic Chemistry Department at the 'Weizmann Institute of Science Rehovoth in consideration of his contribu-tion to the chemistry of natural products including his notable synthesis of the steroidal sapogenins and his studies of the synthesis and properties of un-saturated macrocyclic compounds. The award is made in respect of the year 1961 This Award consisting of a Silver Medal and a monetary Prize is made annually to the chemist of either sex and of British Nationality who in the judgement of the Council of the Chemical Society has published during the year in question and in the immediately preceding five years the most meritor- ious contribution to experimental chemistry and who has not at the date of publication attained the age of thirty-six years.Copies of the rules governing the Award may be obtained from the General Secretary of the Society. Applications or recommendations in respect of the Award for the year 1962 must be received not later than December 31st 1963 and applications for the Award for 1963 are due before the end of 1964.Election of New Fellows.-289 Candidates were elected to the Fellowship in February 1963. Deaths.-We regret to announce the deaths of the following Dr. F. M. Brewer (11.2.63) Reader in Inorganic Chemistry at the University of Oxford; Professor A. Nasini (21.1.63) Director of the Chem- ical Institute of the University of Turin; and Dr. S. W.Smith (30.1.63) formerly Chief Assayer at the Royal Mint. Aspects of Molecular Dissymmebry.-A Chemical Society Symposium on this subject will be held in the MARCH1963 afternoon and early evening of Thursday March 19th 1964 at Battersea College of Technology London S.W.1 1. As the topic is close to the life-long interests of the late Dr. J. Kenyon for many years Head of the Chemistry Department in the College this meeting has been chosen as the occasion for the unveiling of a memorial plaque as part of the tribute of his many friends to his work and inspiration.Full details will be published later. International Symposia etc.-A Eurochemic Sym-posium on Nuclear Fuel Reprocessing will be held in Brussels on April 23rd-26th 1963. Further enquiries should be addressed to O.E.C.D. European Nuclear Energy Agency 38 boulevard Suchet Paris l6eme. A Congress on “Immediate Separation and Chromatography” will be held in Milan on June 14-16th 1963. Further details can be obtained from the Secretary of Societa Italiana per lo studio delle sostanze grass via Lauro 3 Milan. An International Conference on Magnetism will be held in Nottingham on September 7-1 lth 1964.Further enquiries should be addressed to Mr. N. Clarke The Institute of Physics and The Physical Society 47 Belgrave Square London S.W. I. Joint British Committee for Vacuum Science and Technobgy.-A news bulletin has recently been issued by the Committee giving details of forth- coming meetings and other matters of interest to those concerned with vacuum science and technology. For the time being the bulletin is available free of charge from the Secretary Joint British Committee for Vacuum Science and Technology 47 Belgrave Square London S.W.l. Personal.-Dr. P. G. Ashmore has been appointed to the newly created Chair in Chemistry in the Faculty of Technology at Manchester College of Science and Technology.Dr. J. Chatt is being released from Imperial Chemical Industries Ltd. at the request of the Agricultural Research Council to take charge of a team to work on the fundamental chemistry and other aspects of the biological fixation of nitrogen. Dr. S. K. Deb formerly with the National Research Council Ottawa is now a Research Chemist with the Central Research Division of the American Cyanamid Company Stamford Connecticut. Dr. K. Folkers has been elected President of the Stanford Research Institute. Dr. G. G. Freeman Head of Silicones Research Nobel Division Imperial Chemical Industries Ltd. since 1954 has retired. Mr. J. H. Greaves formerly of Younghusband Stephens and Co. Ltd. has joined Proprietary Perfumes Ltd.as Head of the Analytical Laboratory. Mr. I. Greenfield formerly of F. W. Berk and Company Limited has been appointed Managing Director of Cayford Technical Service and Cayford Chemicals Limited. Dr. R. Hurst Director of the Dounreay Experi- mental Reactor Establishment of the Atomic Energy Authority has been appointed Director of Research of British Ship Research Association. Mr. B. S. Jackson Chief Chemist of Building Chemicals Division of Evode Limited has been appointed to the Board of Evomastics Limited. Dr. R. A. Y. Jones of the University of Sheffield has been appointed Lecturer in the School of Chemistry in the University of East Anglia as from September 30th 1963. Dr. R. A. Mitchell has joined the staff of the Department of Physiological Chemistry University of Minnesota Medical School Minnesota.Dr. F. H. C. Stewart formerly of Weizmann Institute of Science Israel is now with the Division of Protein Chemistry C.S.I.R.O. Wool Research Laboratories Parkville Victoria Australia. Mr. W.F. A. Thorp a Director of Authur Holden and Sons Ltd. has been elected Vice-chairman of the Surface Coating Resin Manufacturers’ Associa- tion. Lord Todd has been pre-elected Master of Christ’s College in succession to Professor B. W. Downs whose term of office expires on July 11 th 1963. Mr. A. H. Waddington has been appointed Con- sultant to the Patterson Engineering Company Limited on all chemical and bacteriological matters from May lst when he relinquishes the post of Chief Chemist.Dr. W. Wilso~,formerly Research Manager of B.I.P. Chemicals Limited Oldbury is now Director of Research and Development of CIBA (A.R.L.) Limited. Mr. J. Wright formerly of Alkali Division Imperial Chemical Industries Ltd. is now Senior Research Chemist Carreras Ltd. Basildon Essex. FORTHCOMING SCIENTIFIC MEETINGS London Thursday May 9th at 6 p.m. Hugo Muller Lecture “The Biogenesis of Phenolic Alkaloids,” by Professor D. H. R. Barton D.Sc. F.R.S. to be given in the Lecture Theatre The Royal Institution Albemarle Street W.1. Birmingham Friday May loth at 4.30 p.m. Hugo Miiller Lecture “The Biogenesis of Phenolic Alkaloids,” by Professor D. H. R. Barton D.Sc. F.R.S. Joint Meeting with the University Chemical Society to be held in the Chemistry Department The University.Liverpool Thursday May 9th at 5 p.m. Lecture “Magnetism and Stereochemistry of First Row Transition Elements,” by Professor J. Lewis Ph.D. D,Sc. Joint Meeting with the University Chemical Society to be held in the Donnan Labora- tories Chemistry Department The University. ManChester Thursday April 25th,at 10 a.m. Symposium “Chemical Product Development.” Joint Meeting with the Royal Institute of Chemistry, the Society of Chemical Industry and the Institute of Petroleum to be held in Theatre RIC9 Renold Building Manchester College of Science and Technology. Thursday May 2nd at 6.30 p.m. Hugo Muller Lecture “The Biogenesis of Phenolic Alkaloids,” by Professor D.H. R. Barton D.Sc. F.R.S.,to be given in Room Fl Manchester College of Science and Technology. North Wales Thursday May 9th at 5.45 p.m. Lecture “Some Problems Experienced in the Manu- facture of Pure Beryllium,” by J. A. Dukes. Joint Meeting with the University College of North Wales Chemical Society to be held in the Chemistry Department University College Bangor. Reading Tuesday May 7th at 5.45 p.m. Lecture “Simple and Complex Metal Nitrates and Nitrites,” by Professor C. C. Addison D.Sc. F.R.T.C. Joint Meeting with the Royal Institute of Chemistry and University Chemical Society to be held in the Large Chemistry Lecture Theatre The University. St. Andrews (Joint Meetings with the University Chemical Society to be held in the Chemistry Department St.Salvator’s College.) Friday April 19th at 5.15 p.m. Lecture “Surface Radiochemistry,” by Dr. S. J. Thomson. Friday April 26th at 5.15 p.m. Lecture “Structure Stereochemistry and Biosyn- thesis,” by Professor A. R. Battersby Ph.D. Southampton Wednesday April 3rd at 7 p.m. Lecture “Recent Trends in the Study and Utilisation of Coal,” by A. R. Middleton. To be given at the Portsmouth College of Technology. ADDITIONS TO THE LIBRARY Scientific books and collectors. J. L. Thornton and R. I. J. Tully. 2nd edn. Pp. 406. Library Association. London. 1962. Index to reviews symposia volumes and monographs in organic chemistry for the period 1940-1960. Edited and compiled by N. Kharasch W.Wolf and E.C. P. Harrison. Pp.345. Pergamon Press. Oxford. 1962. Theory and applications of ultraviolet spectroscopy. H. H. Jaffe and M. Orchin. Pp. 624. J. Wiley and Sons. New York. 1962. The interpretation of NMR spectra. K. B. Wiberg and B. J. Nist. Pp. 593. Benjamin. New York. 1962. Mass spectrometry organic chemical applications. K. Biemann. Pp. 370. McGraw-Hill. New York. 1962. Organic chemical crystallography. A. I. Kitaigorodskii. Pp. 541. Consultants Bureau. New York. 1961. Introduction to ligand field theory. C. J. Ballhausen. Pp. 298. McGraw-Hill. New York. 1962. The irreducible tensor method for molecular symmetry groups. J. S. Griffith. Pp. 134. Prentics-Hall Inter- national. London. 1962. Chemical thermodynamics. J. A. V. Butler. 5th edn.Pp. 601. MacMillan. London. 1962. Reaction heats and bond strengths. C. T.Mortimer. Pp. 230. Pergamon Press. Oxford. 1962. Azeotropic data-11. Compiled by L. H. HorsIey. (A.C.S. Advances in Chemistry Series No. 35.) Pp. 100. American Chemical Society. Washington. 1962. (pre sented by the publisher.) Chemistry of combustion reactions. G. J. Minkoff and C. F. H. Tipper. Pp. 393. Butterworths Scientific Publica- tions Ltd. London. 1962. Autoxidation and antioxidants. Edited by W. 0. Lundberg. Vol. 2. Interscience Publishers Inc. New York. 1962. (Presented by the publisher.)
ISSN:0369-8718
DOI:10.1039/PS9630000073
出版商:RSC
年代:1963
数据来源: RSC
|
|