|
1. |
Proceedings of the Chemical Society. May 1961 |
|
Proceedings of the Chemical Society ,
Volume 1,
Issue May,
1961,
Page 153-184
Preview
|
PDF (3536KB)
|
|
摘要:
PROCEEDINGS OF THE CHEMICAL SOCIETY MAY 1961 CENTENARY LECTURE* Chemical Constitution and immunological^ Specificity By MICHAEL HEIDELBERGER UNIVERSITY NEWJERSEY, (COLUMBIA ;RUTGERS,THE STATE UNIVERSITY U.S.A.) WHENa microbial or viral agent of disease penetrates the barriers of defence of an animal organism or when certain foreign substances are injected re- actions are set up which often result in the appearance of new proteins in the blood. These new materials which are globulins,l are called “antibodies” because they are more or less specifically directed against the disease agent or foreign substance injected. This in turn is called “antigen” because it stimulates the formation of antibodies. Antibodies may often be distinguished from the other proteins of the blood serum because they agglutinate or clump together suspensions of the disease agent which stimulated their production; or if the antigen is a soluble pro- tein or polysaccharide antibodies may combine with it to yield an insoluble precipitate.Precipitating anti- bodies or precipitins and agglutinating antibodies or agglutinins may often be measured quantitatively in units of weight.2 Out of such measurements was evolved the quantitative theory of the precipitin reaction which F. E. Kendall and I published in I 9353and which furnishes the basis in immunological theory for much of what I wish to say. In agreement with Marrack and Smith,* antigen and antibody are considered to be multivalent with respect to each other a view based on much experimental evidence.When antigen is added to homologous antibody even though single molecules of each combine with single molecules of the other there are still available on the antigen other groupings capable of combining with other molecules of antibody and there are avail- able on the antibody even if already combined with a molecule of antigen other groupings capable of such combination. Such a mechanism represents a dynamic process which goes on until enormous aggregates are formed. These either fall out of solu- tion because of their weight or as Marrack5 has suggested because of loss of affinity for water owing to the juxtaposition and consequent discharge of vast numbers of positively and negatively charged groups.According to the theory then the immunological specificity evidenced in the precipitin reaction or in bacterial agglutination is due to the interaction of multiple reactive groupings on both antigen and antibody. Application of the theory to the all-important question of the specificity of proteins is unfortunate- ly difficult because of our present rudimentary know- ledge of the chemistry and geometry of the effective surfaces of proteins. We know that a drastic change * Delivered before the Society at Burlington House London on June 2nd 1960 and elsewhere. Heidelberger and Kendall J. Exp. Med. 1936 64 161 ;Heidelberger and Pedersen ibid. 1937 65 393; Goodner and Horsfall ibid. 1937 66 437. Heidelberger and Kendall J.Exp. Med. 1929 50 809; 1935 61 559. Heidelberger and Kendall J. Exp. Med. 1935 61 563. Marrack and Smith Brit. J. Exp. Path. 1931 12 182; 1932 13 394. ti Marrack “The Chemistry of Antigens and Antibodies,” Special Report Series No. 194,2nd edn. H.M.S.O.,London, 1938. 153 154 of such surfaces as by denaturation results in an almost completely altered specificity.6 We also know that crystalline hen's egg albumin is slightly degraded by an enzyme of Bacillus subtilis with removal of the peptides indicated in Fig. 1 to form a crystallisable derivative plakalbumin,' which precipitates most but not all of the antibodies in an antiserum to in- tact egg albumin.8 Perhaps the clearest indication is furnished by tobacco mosaic virus (TMV).When this is treated with carboxypeptidase some 2000 C-terminal residues of L-threonine are split off. The resulting degradation product CTMV not only fails to precipitate all of the antibodies in antiserum to TMV but shows a new specificity possibly charac- teristic of the newly exposed 2000 residuesof L-alanine since an antiserum to CTMV contained antibodies which could not be removed by intact TMV9 (Table 1). Additional useful information has been obtained by chemical alteration of available polar groups such as partial deaminationlO or acetylationll of amino groups or esterification of carboxyl groups.12 More often however one encounters situations such as the immunological cross-reactivity of hen's egg albumin and duck-egg albumin.Each of these precipitates Oval bumin /I 3. //Plakalbumin I + Ala-ala L l'' Plakalbumin I1 + Ala-gly-Val-asp -ala-ala FIG. 1. Removal of peptides from ovalbuniin by enzyme of B. subtilis. only a portion of the antibodies in antisera to the other and yields precipitin curves such as in Fig. 2.13 The chemical reasons for this are completely un- known for the art of protein chemistry has not yet progressed sufficiently to provide answers to such questions. It is for this reason that I shall now turn to the polysaccharides the chemistry of which is in a far more advanced state thanks to the develop- ment of far less laborious techniques. Even these however have their limitations especially from a quantitative standpoint and these may limit the uniqueness of some interpretations as we shall see.PROCEEDINGS TABLE 1. Immunological behaviour of tobacco mosaic virus and carboxvpeptidase-treated virus. Antisera to AntigenTMV TMV +++ CTMV +++ CTMV +++ +++ Absorbed with CTMV TMV TMV ++ - CTMV - ++ That carbohydrates may take part in the immune reactions of precipitation and agglutination was definitely established in 1923 when it was shown1* that the capsular slimes of virulent pneumococci responsible for both virulence and type-specificity were composed of polysaccharides which differed for each of the serological types of Pneumococcus studied. This led to the recognition of polysac- charides as determinants of the immunological specificities of many varieties of micro-organisms.The utility of these new polysaccharides in the study of the mechanisms of immune reactions became 002 0.06 0.1 0.15 Ea,N or Ead N added (mg.) FIG.2. Precipitation of antibody nitrogen at 0"from rabbit anti-Eac serum 791 by increasing amounts oj' Eac N (A) and Ead N (A). Antibody N:Ea N ratios in the precipitate are given for the homologous reaction (a) and for the cross-precipitation with Jhj(0). (Reproduced from J. Immunol. 1948 60 329.) * MacPherson and Heidelberger J. Amer. Chem. SOC. 1945 67 585. Linderstrom-Lang and Ottesen Compt. rend. Trav. Lab. Curlsberg 1949,26,404; Ottesen ibid. 1958,30 14 31 1. Grabar and Kaminski Bull. SOC. Chim. biol. 1950 32 620. Harris and Knight J. Biol. Chem. 1955 214 215.lo Maurer and Heidelberger J. Amer. Chem. Soc. 1951 73 2076. l1 Marrack and Orlans Brit. J. Exp. Path. 1954 35 389; Ehrenpreis Maurer and Ram Arch Biochem. Biophys. 1957 67 178 196; 1959,79 13. l2 Ram and Maurer Arch. Biochem. Biophys. 1959 83 223. l3 Osler and Heidelberger J. Immunol. 1948 60 327. l4 Heidelberger and Avery J. Exp. Med. 1923 38 73; 1924 40 301. MAY1961 155 evident when nitrogen-free pneumococcal polysac- knowledge was due in part to the necessity of learn- charides were used to simplify the problem in our ing something of the fine structures of these immuno- initial quantitative studies on the precipitin reaction logically specific polysaccharides. In 1937 and 1941 but it is only comparatively recently that the realisa- the Type 111 pneumococcal capsular substance was tion has come of the ideal nature of sugars oligo- shown to be a polycellobiuronic acid16a in which each and poly-saccharides for tests of the validity of the unit was probably p-linked to the next D-glucuronic ensuing theory3 and for predictions based upon it.acid at position 3.lS The chemical basis for the long- Polysaccharides are usually considered to be known cross-reactivities of Types I11 and VIlI polymers of definite repeating units composed of pneumococci was established in 1935 when it was one or more sugars. That such units or small shown that both capsular polysaccharides contain FIGURE 3. Structural formulae of pneumococcal polysaccharides (all sugars are apparently in the pyranose form). Type L1* -3)-~-Rha-(l-+3)-~-Rha-( 1+3)-~-Rha-( 14)-~-Glu-( 14)-~-GluA-( 1+.. 6 f 1 D-GIuA X I 1 Type 11I -[3;-/3-~-GluA-( 1+4)-/3(?)-~-GIu~( I+ I1 ' Ix Type VI -[2)-cx-~-Gal-( 1+3)-D-Glu-( 1+3)-~-Rha-( l-+3)-Ribitol-l[5]~-0~PO(OH)*O-I 1-+3)-~-Gal-( 1+6)-~-GluNAc-(1+3)-~-Gal-( 1+6)-~-GluNAc-( 1+3)-~-Gal-( 1- 4 4 1 f a-D-Gal 1 I. * Not uniquely determined. ? Or -2(4)-. 3 Lactose has now been isolated among the products of partial hydrolysis.laa S XIV is susceptible to the action of both a-and /I-galactosidases (Kabat) ;most other linkages are probably j?. Modified from Progr. Chem. Natural Products 1960,18 507. multiples of them might function immunologically the same aldobiuronic acid,17 but the he structure of as the multiple reactive groupings postulated by our the Type VIII substance was not elucidated until theory was soon shown.15 We found that partial more than twenty years later.l* This now enables us hydrolytic products of the specific capsular polysac- to discuss both the original instance of cross-charides of Types I and III pneumococcus having an reactivity and another major one in terms of the average molecular size of 700 to 2000 could still multiple reactive units involved.precipitate a portion of the antibodies in Type I and In Fig. 3 are given the structures of the two poly-Type I11 antipneumococcal horse sera. The multi- saccharides. It will be noted that while the Type III valence of the intact specific polysaccharides was substance has alternate 3-and 4-linkages all of thus clearly indicated.The delay in exploiting this which appear to be p- all of the linkages in the l6 Heidelberger and Kendall J. Exp. Med. 1933 57 373. (a) Hotchkiss and Goebel J. Biol. Chem. 1937 121 195; (6) Reeves and Goebel J. Biol. Chem. 1941 139 511. Goebel J. Biol. Chem. 1935 110 391. la Jones and Perry J. Amer. Chem. SOC.,1957 79 2787. 'Barker unpublished data. Type VIII substance are 1 -+ 4 and some are a-.As far as the methods of carbohydrate chemistry are capable of deciding this point both ply saccharides are unbranched. The probability of a linear structure was even deduced on immunological groundslg long before the chemical structure was worked out which should give some idea of the power of quantitative immunochemical methods.It is also clear from Fig. 3 that such instances of cross-reactivity do not depend on an “antigen in common,” as the bacteriologists so often and so loosely express the reason for cross-reactivity S111 and S VIII are different antigens which do however possess multiple reactive groups in common. The groupings in question in Fig. 3 are marked off by the dashed lines. Another group of cross-reactions in Type VXII antiserum is referable to multiple units of cellobiose and I have marked these off with x’s barley and oat glucans and Iles glucomannan are the substances which give precipitates. Indeed the positive sero- logical test20 preceded the chemical isolation of cel- lobiose from barley and oat glucans by Professor I.A. Preece and confirmed the deductions made from their degradation by enzymes.21 Another variety of cross-reaction with poly-glucoses became interpretable after Butler and Stacey22 cleared up the main features of the capsular polysaccharide of pneumococcal Type 11 a possible formula of which is also given in Fig. 3. All of the glucose was found to be in the form of 1,4,6-branch points. From the quantitative theory of the precipitin reaction it was predicted that all polysaccharides with multiples of such branch points would cross- precipitate in Type IIantipneumococcal horse serum particularly if the anomeric linkages of the branched glucose were the same. The most obvious of such carbohydrates is glycogen but since it is so ubi-quitous a component of animal tissues it should not be capable of reacting with antibodies according to the older immunological theories.The new theory demanded a trial however and glycogen was im- mediately shown as predicted to be an immuno- logically reactive carb~hydrate.~~ Amylopectin in which the branch points are more widely spaced precipitated less antibody. In both of these poly- glucoses all linkages are a- but in another polysac- charide that of tamarind seed two of the three PROCEEDINGS glucose units are also linked 1,4,6-,24 but j%25 Tamarind-seed polysaccharide also precipitates Type I1 antiserum but the fraction of antibody pre- cipitated is not the same as that brought down by glycogen. It therefore seems to matter little when precipitation is due to multiple interior linkages whether these are a-or p- in contradistinction to sharp differences in specificity due to anomeric forms of the same sugar when these are terminal.26 Another instance in which both anomeric forms of glucose cause precipitation is the reactivity of Type VI antipneumococcal serum with oat and barley glucans presumabfy because of their multiple residues of P-l,3-linked glucose and the antiserum’s equal reactivity with a dextran containing a high proportion of ~ll-1~3-linkages.2~ The structure of the specific polysaccharide of Type VI pneumococcus is given in Fig.3. Reactivity of the oat and barley glucans with Type VI antisera is thus due to different portions of the molecule from those reactive with Type VIII antiserum.To return to Type 11 one should recall that it has long been known that dextrans with their pre- dominant a-1,6-linkages also precipitate Type I1and other antipneumococcal sera.28 Since however the 1,6-linkages in the Type 11 specific polysaccharide are at the 1,4,6-branch points this need not seem surprising. That the reactivity with the glycogens occurs at these points would seem to be borne out by the data in Table 2 from which it is evident that as the exterior chains of a glycogen are shortened by TABLE2. Precipitation of antipneumococcal horse serum 11 513 at 0” by a glycogen fraction and limit dextrins derived from it. Oyster glycogen Alb Antibody nitrogen pre- (mg.) cipitated from 1.0 ml.of serum (pg.) 3 172 5 184 Phosphorylase limit dextrin 3 290 p-Amylase limit dextrin 2 363 4 410 l9 Heidelberger Kabat and Mayer J. Exp. Med. 1942 75 35. 2o Heidelberger and Rebers J. Amer. Chem. SOC. 1958 80 116. 21 Aitken Eddy Ingram and Weurman Biochem. J. 1956,64 63. 22 Butler and Stacey J. 1955 1537. 23 Heidelberger Aisenberg and Hassid J. Exp. Med. 1954 99 343. e4 White and Rao J. Amer. Chem. SOC. 1953 75 2617. 25 Khan and Mukherjee Chem. and Ind. 1959 1413. 26 Cf. for example Avery Goebel and Babers J. Exp. Med. 1932 55 769. 27 Heidelberger and Rebers J. Bacteriol. 1960 80 145. 28 Zozaya J. Exp. Med. 1932 55 353; Sugg and Hehre J. Immunol. 1942 43 119; Hehre Sugg,and Neill Ann. N. Y. Acad. Sci. 1952 55 467.MAY1961 phosphorylase or p-amylase the amount of antibody precipitated goes up as would be expected if access to the branch points by antibody became less hindered sterically. It is only fair to point out however that there is an alternative interpretation of these results with the polyglucoses. The larger part of the antibodies in Type I1 antipneumococcal sera is reactive with a wide variety of gums containing glucuronic acid29 in the form of multiple non-reducing end-gro~ps,~~ a feature also possessed by the Type I1 polysaccharide (Fig. 3). It is not only possible that non-reducing glucose end-groups of glycogen and amy lopectin could fit into antibody-spaces complementary to glucuronic acid end-groups but also conceivable that a few per cent of glucose end-groups undetectable by present methods available to carbohydrate chemists could even exist in the Type I1polysaccharide and so confuse the apparently clear-cut results involving the 1,4,6-branch points.If this were true the increased activity of the limit dextrins derived from glycogen might be due to a larger number of such short glucose side-chains. This alternative explanation however is unsupported by direct evidence and rests upon the present imperfect state of the methods of carbohydrate ~hemistry.~~ Glycogen was also found to precipitate Type IX and Type XI1 antipneumococcal sera and as we had on hand highly purified samples of the specific poly- saccharides of Types IXand XU pneumococci these carbohydrates were hydrolysed and the solutions chr~matographed.~~ As might have been predicted from these cross-reactions both substances con- tained glucose although this prediction was unsafe because of lack of information as to whether or not glucosamine differs from glucose in its specificity.Both polysaccharides also contained amino-sugars so that more work will have to be done before the entire basis of the cross-reactivity of the two antisera is known. The precipitation of the polyglucoses in Type XVIII antiserum is also presumably attributable to antibodies stimulated as a result of the known glucose content of the capsular polysaccharide of Type XVIII pneumococcus although the glucose units must be in very different linkage from those in the Type I1substance since there is no cross-reactivity in either direction between Type I1 and Type XVIII pneumococcus.33 A group of specificities about which one can be more definite is that due to multiple Occurrences of galactose.A possible structure for the specific poly- saccharide of Type XIV pneurnococcus is given in Fig. 3. Before the main features of this structure were worked out the prediction was made that non- reducing end groups of galactose would be found since all polysaccharides containing such end groups including those tested in which all of the galactose was present as end groups gave precipitates in Type XIV antipneumococcal serum.= Verification of this prediction soon f0llowed.3~ It was also possible by this potent immunochemical method to point to several possible linkages for the internal galactose residues of the Type XIV polysaccharide one of which turned out to be correct.A peculiarity of the specificity due to multiple galactose end-groups came to light when the fine structure of the specific polysaccharide of Type VI pneumococcus was elucidated. Even though the polysaccharides with galactose end-groups precipi- tated Type VI antipneumococcal sera there was no evidence for side-chains or end-groups of galactose as will be noted from Figure 3. However the phos- phate groups which bind the repeating units into a long chain are attached on one side to position 2 of the galactopyranose residues leaving positions 3 4 and 6 free.Such an arrangement evidently suffices to stimulate the antibody-forming mechanism in much the same way as do true end groups of gala~tose.~’ This is rather unfortunate because we shall have to qualify our erstwhile carefree prediction since pre- cipitation of an antiserum by galactomannans and tamarind-seed polysaccharide may indicate either multiple end-groups of galactose or multiple 1,2-linked galactose residues in the antigenic deter- minant. The capsular poly saccharide of pneumococcal Type VI has been mentioned several times and I should like to tell you why we are interested in it and something of the results of our study. In 1915,O. T. AveryX noted that occasional strains of pneumococci isolated from patients were difficult to assign to the already recognised types but agglutinated “weakly and incompletely” in Type I1antipneumococcal sera.He therefore called these strains IIA IIB etc. but as the first two appeared with alarming frequency in human pneumonia they were eventually given the Marrack and Carpenter Brit. J. Exp. Pathol. 1938 19,53. 30 Heidelberger and Adams J. Exp. Med. 1956 103 189; Heidelberger ibid. 1960 111 33. 31 Cf.Goodman and Kabat J. 1mmuno1. 1960 &I 333. 32 Heidelberger Barker and Stacey Science 1954 120,781. 33 Markowitz and Heidelberger J. Amer. Chem. Soc. 1954 76 1317. 34 (a) Heidelberger Dische Neely and Wolfrom J. Amer. Chem. Soc. 1955 77 3511; Heidelberger ibid. 4308; (b) Heidelberger Barker and Bjorklund ibid. 1958 80 113. 35 Barker Heidelberger Stacey and Tipper J.1958 3468. 36 Avery J. Exp. Med. 1915 22,804. ical structure and immunological specificity. Part of TABLE 3. Precipitation of types II and VI antipneumococcal horse sera by hamolytic streptococcal group and variant (V)carbohydrates* (pg. antibody nitrogen from 1-0ml. at 0'). Poly saccharide Wt. (mg.1 Group A substance 0.02 0.03 0.05 0.08 0.1 0-15 0.2 0-5 1.0 Group A V substance 0.03 0-05 0.08 0.1 0-16 0-2 0.3 0.5 0-6 * Modified from ref. 27. a Data from ref. 38. the information was already at hand for a beginning had been made on the he structure of the polysac- charide of Type IIz2 and the study of its partial specificitie~.~~~~~~~ S.A.Barker's interest was enlisted in the Type V substance which has shown some most unusual features,39 and P. A. Rebers and I have been studying the polysaccharide of Type VI the structure of which you will note again in Figure 3. Com-parison will show at a glance the great differences from the Type 11 substance and the possession in common of only two sugars D-glucose and L-37 Cooper Rosenstein Walter and Peizer J. Exp. Med. 1932 55 531. 38 Heidelberger and McCarty Proc. Nat. Acad. Sci. (U.S.A.) 1959 45 235. 39 Barker and Williams Proc. Symposium on Glucides G$-sur-Yvette 1960; Barker Brimacombe How and Stacey Nature 1961 189 303. McCarty and Lancefield J. Exp. Med. 1955 102 11; McCarty ibid. 1956 104 629. PROCEEDINGS type designations V and VI.37 While it was obvious rhamnose.Only the latter is similarly linked in both that the capsular polysaccharides of these three polysaccharides. D-glucose can scarcely be involved types the determinants of their type specificity in the 11-VI relationship since in general the poly- would turn out to be chemically different it was also glucoses which react well in anti-I1 precipitate anti- apparent that V and VI must also be chemically VI poorly or not at all while those which give good related to I1 and not necessarily in the same way. It precipitates in anti-VI fail to do so in anti-11. This therefore seemed likely that a correlation of these would leave only the 1,3-linked L-rhamnose residues differences and similarities with the immunological in both as the culprit.Let us see if there is any less behaviour of the three types would make a useful circumstantial evidence. addition to knowledge of the relation between chem- In the study of the partial specificities of TypeIIi it was found that the carbohydrate characteristic of A Type IJ Type VI antiserum antiserum 513a 614 (pg.) (pg.1 14 16 3 0 0 1 66 63 0 43 0 I Type V] Type VI antiserum antiserum 681C 771C (p.8.) (pg.) 8 12 11 13 21 22,26 32 15 68 12 63 38 the Group A haemolytic streptococcus gave weak precipitation with Type I1 antipneumococcal sera and this reactivity was ascribed to multiple residues of L-rhamnose the only common component in the two antigens3O An opportunity to test this conclusion further arose when a mutant of the Group A strepto-coccus (called V for variant) was discovered in which much of the N-acetylglucosamine of the carbo- hydrate was stripped off and the proportion of rhamnose was much higher.4O It was reasoned that the V substance should precipitate more antibody MAY1961 from Type 11 antipneumococcal sera than the A substance since its rhamnose would either be more nearly terminal or else be more easily accessible to reactive antibody because of removal of N-acetyl- glucosamine side-chains.Experiments showed this assumption to be justified.% This not only strengthened the belief that the cross-reaction occurred because of the multiple residues of L-rhamnose in the Type I1 pneumococcal and A V streptococcal polysaccharides but rendered it highly likely that at least a portion of the rhamnose in the streptococcal substances will be found to be linked in the same way as in the Type I1 substance namely 1 3.We are anxiously but confidently awaiting the results of the chemists who are even now working on this problem. Well what has all this to do with pneumococcal Type VI? Merely that if the 11-VI relationship were indeed due to the similarly linked residues of rham- nose one would expect the streptococcal A and V substances to behave in the same way in anti- pneumococcal VI sera as in anti-I1 sera. As you will note in Table 3 they do,27 so that I believe we now know the chemical reason why Avery called the Type VI “pneumococcus IIB.” There are still many other things that quantitative immunochemistry has done for the chemistry of polysaccharides but I will only hint at them.It has for example provided two new criteria of inhomo-geneity one by measurement of the ratios of com- ponent sugars before and after precipitation of a cross-reacting polysaccharide with an antiserum (identical ratios do not necessarily indicate homo- geneity),34as41 and the other by determination of the polysaccharide-antibody ratios of successive frac- tional precipitations with purified antibody in com- parison with the ratios of blanks of the original material in amounts similar to those of each remain- ing fraction.42 In listing these possibilities in addi- tion to the ones which I have described in greater detail I am hopeful that before long every well- equipped laboratory for the study of polysaccharides will rely for short cuts and guidance upon a small and well-selected library of antisera and that the usefulness of immunochemical methods and the scope of their application will steadily increase.41 Beiser Kabat and Schor 1952 J. Immunol. 69 297; Heidelberger Adams and Dische. J. Anrer. Chern. SOC. 1956,78,2853; Rebers Barker Heidelberger Dische and Evans ibid. 1958,80 1135. 42 Heidelberger Jahrmarker Bjorklund and Adams J. Immunol. 1957 78 419. CHEMICAL SOCIETY AUSTRALIAN LECTURE TOUR INMarch and April 1961 under the auspices of The Chemical Society Professor A. Albert of the Australian National University Canberra gave two lectures each in Hobart Melbourne Adelaide Sydney Brisbane Armidale and Newcastle.Sum- maries of these two lectures are given below. A New Approach to Heterocyclic Chemistry Lecture I. Overcoming Difficulties by Familiar Analogies.-The ever-growing importance of hetero- cyclic substances in biology medicine and industry has led to the search for a logical framework that will correlate the often bewildering facts of hetero-cyclic chemistry. These lectures describe such a framework,l which is derived from the most familiar concepts of general organic chemistry aliphatic and aromatic. Thus the whole of heterocyclic chemistry is conveniently divided into three classes :hetero-paraffinic hetero- ethylenic and hetero-aromatic.* Except for three-membered rings which are highly strained and reactive heteroparafiic substances have properties essentially similar to their paraffinic analogues.Thus tetrahydrofuran resembles diethyl ether and pyrrolidine resembles diethylamine. Similarly piperazine resembles a simple aliphatic di- amine the lactams resemble simple amides and cyclic anhydrides such as succinic anhydride are very similar to acyclic aliphatic anhydrides. Some second- order differences are known and many of them arise from the more compact character of the hetero- paraffinic substances. The biggest departure from type is shown by some lactones (5-and 6-membered rings) which are hydrolysed much faster than ali- phatic esters and have larger dipole moments these differences have been traced to the twist imposed on the ester-group in these small rings2 Hetero-ethylenic substances i.e.those which have one or more double bonds but lack aromatic pro- perties have been little investigated in comparison * By hetero-paraffinic Professor Albert refers to hetero-substituted cycloalkanes by hetero-ethylenic to hetero- substituted cycloalkenes and by hetero-aromatic to heterocyclic compounds having some of the properties of the aromatic ring.-ED. Albert “Heterocyclic Chemistry an Introduction,” The Athlone Press London 1959. Huisgen Angew. Chern. 1957 69 341. with the other two heterocyclic classes. In particular insufficient attention has been given to the positions finally occupied by the double bonds which have more tendency to migrate than is found in carbo- cyclic chemistry.Hetero-aromatic substances form a large class that is conveniently divided further. One division re- sembles nitrobenzene in that the hetero-atom attracts electrons from the dayer; this division of which pyridine is an example has been termed .rr-deficient,l the deficiency referring to the carbon atoms of the ring. The other division resembles aniline in that the hetero-atom releases electrons from its lone pair(s) to the .rr-layer of the ring; this division of which pyr- role is an example has been termed n-excessive. The aromaticity of pyrrole arises from a sextet formed from two electrons contributed by the nitrogen atom and two contributed by each of the two double bonds.Cyclopentadiene the carbon analogue of pyrrole has no such possibility of forming a sextet hence it lacks the stability and the possibilities for substitu- tion which pyrrole possesses. The rr-deficient hetero-aromatics are in general resistant to hydrolysis. But when the ratio of doubly bonded nitrogen atoms to carbon atoms becomes high (as in triazine and pteridine) the ring opens with remarkable ease. This is attributed to the large proportion of n-electrons sequestered by the ring- nitrogen atoms resulting in a loss of aromatic stability for the ring. 7-r-Deficient N-hetero-aromatics undergo nucleophilic substitution very readily and electrophilic substitution with great difficulty just as would be expected from the electron-density dia- grams.The insertion of one strongly electron- releasing group (NH, SH OH) per ring-nitrogen atom is necessary for easy electrophilic substitution. The phenomenon of covalent hydration is often found in those n-deficient N-hetero-aromatics which have a high N:C ratio and is attributed to the emergence of an isolated double bond through the partial loss of aromatic character? Thus the quin- azoline cation (but not the neutral molecule) is tenaciously hydrated in the 3,4-p0sition.~ 6-Hydroxy- pteridine is likewise hydrated in the 7,8-p0sition.~ (A methyl group can hinder these hydrations ~terically.~~~) The double bonds that add water can add a variety of Michael reagents instead also acetone and hydroxylamine.6 Many other examples of covalent hydration have come to light since its original discovery in my Department where the kinetics of the reaction are being explored.' Albert Armarego and Spinner J.in the press. Albert J. 1955,2690. Brown and Mason J. 1956 3443. Albert and Reich J. 1961 127. Perrin and Inoue Proc. Chem. SOC.,1960 342. Badger and Christie J. 1956 3438. PROCEEDINGS The n-excessive hetero-aromatics are stable to hydrolysis but are polymerised by acid. Thiophen requires more vigorous conditions than pyrrole and furan. For pyrrole this polymerisation is a con-sequence of the loss of aromatic stabilisation when the lone pair becomes involved in cation formation. ?r-Excessive hetero-aromatics undergo electrophilic substitution very readily but nucleophilic substitu- tion is extremely rare all of which would be expected from the electron-density diagrams.That the electron-releasing effect of a singly bonded nitrogen atom is quantitatively greater than the electron- attracting effect of two doubly bonded nitrogens can be deduced from the behaviour of tetrazole and the triazoles. Lecture II. The Help which Physical Properties can Give.-In recent years a very high degree of correla-tion has been obtained between the physical pro- perties and constitution of heterocyclic substances. Even solubilities can be diagnostic in .rr-deficient hetero-aromatics hydrogen-bonding groups (e.g. NH,) lower the solubility in water. Again doubly bonded nitrogen atoms are water-attracting whereas singly bonded nitrogen in aromatic rings is water- repe1ling.l Ionisation constants are closely correlated with structure.The hetero-paraffinic bases are strongly basic. They have pKa values about 11 as have secondary aliphatic amines (only the highly strained ethyleneimine forms an exception it is lo00 times weaker and has pK 8). The various wdeficient heterocycles (pyridine quinoline acridine etc.) have pKa about 5 when only one nitrogen atom is present. Additional ring-nitrogen atoms drastically lower this figure because each has an inductive effect equivalent to that of a nitro-group. The singly bonded nitrogen atom in n-excessive heterocycles (e.g. pyrrole in- dole carbazole) is only very feebly basic.When both kinds of nitrogen atom are present in the same nucleus (as in imidazole and diazaindenes) the possibilitiesof extra ionic resonance in the cation can greatly increase basic strength. Ultraviolet spectra also afford many interesting correlations. Among these may be mentioned the high degree of equivalence of =N-to =CH- so that the spectra of most r-deficient hetero-aromatics closely resemble those of the corresponding aromatic hydrocarbons. Recognition of the isosterism between carbazole and phenanthrene led to the formulation of similar ruless for -NH- -0- arid -S-. Rules MAY1961 161 based on similarity between the spectra of aniline hydrochloride and benzenes make it possible to say whether an amino-derivative of an N-heterocycle has accepted a proton on the exocyclic nitrogen or on the ring-nitrogen atom? The similar spectra pro- duced by pyrones and pyridones has been useful for assigning constitutions to oxygen-heterocycles.The predominant tautomer present at equilibrium in aqueous solutions of amino-derivatives has been solved by comparing ionisation constants and ultra- violet spectra of these substances with those of analogues in which the mobile hydrogen atoms are replaced by methyl groups (it is essential in such work to examine methyl derivatives of bath tautomers in case they should not differ greatlyll). From studies of this kindlo it can be said beyond doubt that the amino-derivatives of all n-deficient hetero-aromatics are mainly in the form of the true primary amines e.g.much (I) is in equilibrium with very little (11). The same methods applied to hydroxy-derivatives of n-deficient hetero-aromatics has revealed that the a-and y-hydroxy-derivatives are mainly in the form of (cyclic) amides and vinylogous amides e.g. (111). Some /3-hy droxy-derivatives have a high concentra- tion of zwitterion e.g. (IV) whereas others are true phenols.12 Similarly a-and y-mercapto-derivatives were shown to be (cyclic) thioamides.13 The use of nuclear magnetic resonance,14 infrared spectroscopy,15 dipole moments,16 and oxidation reduction potentials1 has also led to valuable cor- relations between physical properties and molecular structure. Waterman and Harberts Bull. SOC.chim.France 1936 3 643. lo Angyal and Angyal J. 1952 1461 ;Anderson and Seegar J. Amer. Chewi. SOC.,1949,71 340. l1 Arndt and Eistert Ber. 1938 71 2042. l2 Albert and Phillips J. 1956 1294. l3 Albert and Barlin J. 1959 2384. l4 Gronowitz and Hoffman Arkiv Kemi 1960 15 499. l5 Mason J. 1957 4874 5010. Le Fkvre and Le Fkvre J. 1937 1088; Pauling Forfschr. Chem. Org. Nafursfoffe,1939 Vol. 111 p. 219. COMMUNICATIONS Reduction of Aromatic Nitro-compounds by Sodium Borohydride Catalysed by Transition-metal Complexes By ANTON~N and ALBERT A. VL~EK RUSINA INSTITUTE CZECHOSLOVAK OF SCIENCES CZECHOSLOVAKIA) (POLAROGRAPHIC ACADEMY PRAGUE A REDUCTION process may proceed either as an electron-transfer or as a group-transfer reaction. The experimental material at present available indicates that electron-transfer processes are not greatly in- fluenced by structural effects so that the ability to react is mostly governed by the redox potentials of the reacting systems.Many systems which could react only by atom- or group-transfer do not react at all in spite of the fact that reaction would be expected from the value of their redox potentials. The presence of a third redox system which reacts as electron-transfer agent and has a redox potential between the potentials of the other systems might catalyse such a process assuming that the redox system used as catalyst reacts rapidly and accurateIy reversibly. The group-transfer reaction specifically hindered is thus replaced by a set of reactions all or some of which are of the charge-transfer type.This principle is analogous to that proposed by Shafferl for charge- transfzr reactions in which the number of electrons that may be supplied by the reductant is not the same as that required by the oxidant. To study this possibility we chose the reaction between aromatic nitro-compounds and sodium borohydride in water or ethanol. Hitherto only a few reductions of aromatic nitro- and nitroso-compounds with this reagent in non-aqueous solvents have been Shaffer,J. Amer. Gem. SOC.,1933.55 2169; see also Clark “Oxidation-Reduction Potentials of Organic Systems,” Williams and Wilkins Baltimore 1960 pp. 340-342. described,2 whereas in aqueous or ethanolic solution the reduction does not proceed.On the other hand aromatic nitro- and nitroso-compounds are very easily reduced at a mercury electrode3 by an electron- transfer mechanism. As the catalytic system cobalt-bipyridyl com-plexes have been used. The complexes Co(dipy),3+ and Co(dipy),2+ react very rapidly with sodium boro- hydride with formation of a complex of univalent cobalt4 Co(dipy),+ which is paramagnetic (corres- ponding to two unpaired electrons5) and is easily and reversibly oxidised to the complex C~(dipy),~+. The redox potential of the system C~(dipy),~+-Co(dipy),+ is -0.91 v with respect to the saturated calomel electrode in 50% ethanolic solution or -1.2 v in aqueous solution? Sodium borohydride does not react in a solution containing 0.1 Swphosphate or borate buffer (PH between 6 and 8.5 was followed) with nitrobenzene p- 0- or rn-nitrophenol p-nitroaniline p-nitro- benzoic acid 2,3-dinitrophenol 2,4-dinitrotoluene and 3;Sdinitrobenzoic acid even after several hours.Addition of a small quantity of Co(dipy),(ClO,) to this solution causes reduction of the nitro-compound at a rate dependent on the amount of cobalt complex present. When equimolar quantities of nitro-compound and cobalt complex are used the reduc- tion is complete in 1-2 minutes in solutions with a molar ratio 10 1 the reaction is complete in 10-30 minutes (depending on the pH and the constitution PROCEEDINGS of the nitro-compound). The reaction proceeds also at molar ratio 100 1 though very slowly. The course of the reaction has been followed polarographically in dilute solutions (-2 x molar).After the completion of the reaction no reduction wave was observed up to a potential of -1.4 v this shows that nitroso azo- and azoxy-compounds are not formed as reduction products. The chemical evidence shows that at least in the case of mononitro-compounds hydroxylamine and mine derivatives are formed as reduction products. The preparative reduction has to be carried out in well-buffered solutions (the best pH being 6-5-7) and with the exclusion of oxygen the presence of which causes formation of other reduction products and decreases the amount of amine formed. In dinitro-compounds both nitro- groups are reduced (probably successively) a polaro- graphically inactive product being formed.The ion Co(dipy),+ represents the reducing agent proper the reducing action of which on organic compounds is very rapid as has been deduced by direct observation of its reaction with aromatic nitro- compounds as well as from the fact that the deep blue colour of univalent cobalt is observed only after complete reduction of the organic material. The use of other transition-metal complexes as electron-transfer catalysts is now being studied. The full results will be published in CoZZ. Czech. Chern. Comm. (Received February 23rd 1961 .) Weill and Panson J. Org. Chem. 1956,21 803; Shine and Tsai J. Org. Chem. 1958,23 1952; Boyer and Ellzey jun. J. Amer. Chem. SOC.,1960 82 2525. See Kolthoff and Lingane “Polarography,” Interscience Publ.New York 1952. Vol. 11. VliSek Nature 1957 180 753. Rusina Thesis Prague 1960. Week 2.phys. Chem. (Leipzig) 1958 Sonderheft,p. 164. Steric Effects in Substituted Benzoic Acids By G. FERGUSON and G. A. SIM DEPARTMENT GLASGOW, (CHEMISTRY THEUNIVERSITY W.2) FORovercrowded aromatic hydrocarbons theoretical calculations’ of the deviations from planarity appear to be in reasonable agreement with the deviations determined experimentally.2 For polysubstituted benzenes however the situation at present is not so satisfactory. Electron-diffraction investigations3 of polyhalogenobenzenes suggest that such molecules are markedly non-planar whereas X-ray studies4s5 indicate that in the solid state such molecules are very closely or even exactly planar.Theoretical Coulson and Senent J. 1955 1819. Mclntosh Robertson and Vand J. 1954 1661. Bastiansen and Hassel Acia Chem. Scand. 1947 1 489. Tulinsky and White AcfaCryst. 1958 11 7. Gafner and Herbstein Acfa Cryst. 1960 13 706. Coulson and Stocker Mol. Phys. 1959 2 397. models moreover do not yet appear to be sufficiently precise to yield results in good agreement with physical meas~rements.~ One of the major troubles associated with diffraction studies of polyhalogenobenzenes is that in the presence of a number of much heavier halogen atoms the accurate location of the lighter carbon atoms is difficult. Even when the positions of the halogen atoms have been very precisely determined the inaccuracies associated with the positions of the MAY 1961 carbon atoms may prevent a really detailed descrip- tion of the molecular geometry.This difficulty can be minimised by studying polysubstituted benzenes containing not more than one halogen atom and we placement of these bonds away from one another. These out-of-plane displacements however are con- siderably smaller than those claimed for poly-halogenobenzenes on the basis of electron-diffraction are therefore engaged at present in a number of studies? In 2-chloro-5-nitrobenzoic acid the carboxyl group is inclined at a greater angle to the plane of the m;;en benzene ring than in o-chlorobenzoic acid. This may be due to a buttressing effect of the bulky nitro-group three-dimensional X-ray Wstal-structUre mdYSeS on the hydrogen atom at position 6.In support Of of such COmPO~ds- unitell data have this may be cited the angle of 7.0"between the plane been published elsewhere.' of the nitro-group and the plane of the benzene ring. Extensive least-squares refinement of positional In rn-dinitrobenzene where a buttressing effect should and thermal parameters has been completed for also occur the molecule is non-planar each nitro- O-C,H~CI*CO~H o-C,H,BICO,H 5,2-N02*C,H3ClCO2H Angle between the plane of the carboxyl group 13.7" 18.3" 23.0" and the plane of the benzene ring O Valency angle I 122.5 123.4" 125.5" Valency angle I1 124.7O 124.9O 121-9" Displacement of exocyclic carbon atom from the plane of the benzene ring -0.058 A -0.057 A -0-051 A Displacement of halogen atom from the plane of the benzene ring +0.036 A $0.064 A +0.001 A o-chlorobenzoic acid (R =0.105 for 1034 reflexions) group being inclined at about 11O to the plane of the o-bromobenzoic acid (R =0.1 32 for 1 145 reflexions) benzene ring whereas in nitrobenzene where the and 2-chloro-5-nitrobenzoic acid (R = 0.091 for effect is absent the molecule is planar.g lo00 reflexions) and the relevant results are listed in the annexed Table.Standard statistical tests with the The calculations were carried out on the Glasgow estimated standard deviations in positional para- University DEUCE computer with programmes meters show that the deviations from ideal planar devised by Dr. J. S. Rollett and Dr. J. G. Sime and structures with valency angles of 120" are highly we are indebted to the Director of the Computing significant.Laboratory Dr. D. C. Gilles and his staff for In all three molecules the steric strain is relieved by facilities. We also thank Professor J. Monteath small out-of-plane displacements of *he exocyclic Robertson for his advice and interest and the valency bonds in addition to the larger in-plane dis-Carnegie Trust for a Scholarship (to G.F.). (Received March 15th 1961.) Ferguson and Sim Acta Cryst. 1959 12,941. Trotter Acta Cryst. 1961 14,244. Trotter Acta Cryst. 1959 12 884. The Structure of Biphenylene By T. C. W. MAKand J. TROTTER OF CHEMISTRY OF BRITISH VANCOUVER (DEPARTMENT UNIVERSITY COLUMBIA 8 CANADA) PREDICTIONS by resonance theory and by molecular- tions but contrary to expectations on the simplest orbital calculations of the chemical behaviour of bi- resonance theory.2 Since the two methods also pre- phenylene are in marked disagreement and the dict different bond-length variations it should be pos-chemical reactivity of biphenylene and its derivatives sible to decide between them by an accurate measure is in accordance with the molecular-orbital calcula- of the bond distances and we have now carried out Longuer-Higgins,Proc.Chem. SOC.,1957 157. Baker McOmie Preston and Rogers J. 1960 414; Baker and McOmie Chem. SOC.SpecialPubl. No. 12 1958 49; Chem. andhd. 1958 53. a detailed X-ray investigation of crystalline bi- phenylene since previous analy~es~.~ were not sufficiently precise to detect the small differences.The measured distances at the present stage of the Al/? (I) rn Calculated and measured bond distunces (A) in biphenylene. Calculated I-h---7 Measured Bond Simple resonance theory A 1.41 B 1-38 C 1.41 D 1.41 E 1*45 Molecular-(X-ray orbital diffraction) theory 1-38 1.36 1 -40 1-42 1*38 1.38 1.41 1.38 1-47 1-52 PROCEEDINGS analysis (further refinement is unlikely to alter these significantly) are compared in the Table with the calculated lengths,15 and these measured distances support the molecular-orbital predictions. In valence- bond terms the preferred Kekul6 structure is (I) as suggested by Baker et although this of course represents an extreme fixation of double and single bonds.Reasonable agreement between calculated and measured distances can be obtained by giving maximum weight to form (I) (that is by considering biphenylene as a cyclobutane derivative) and less weight to the two cyclobutene structures and includ- ing only a negligible amount of cyclobutadiene character. These more detailed comparisons will be postponed until refinement of the structure is complete. We are indebted to the National Research Council of Canada for grants-in-aid of this work and for the award of a research studentship (to T.C.W.M.). (Received February 27th 1961.) Waser and Schomaker J. Amer. Cheni. SOC.,1943 65 1451. Waser and Lu J. Amer. Chem. SOC.,1944 66 2035. Cf. Ali and Coulson Tetrahedron 1960 10 41.A New Type of Addition to 1-AlkylpyrroIes By R. M. ACHESON, A. R. HANDS,and J. M. VERNON (DEPARTMENT UNIVERSITY OF BIOCHEMISTRY OF OXFORD) THE addition of acetylenedicarboxylic acid to 1-methyl-l and 1-benzyl-pyrrole2 gives the 2-pyrryl- fumaric acid and the corresponding maleic an-hydride. A minor product in the case of l-benzylpyr- role is the Diels-Alder adduct (I). Dimethyl acetylenedicarboxylate gives a yellow 2 1 adduct with 1-methylpyrrole,’ for which structure (II) has been proposed. Oxidation with bromine and methanol gave an indole triester which on hydrolysis and decarboxylation yielded 1 -methylindole. The remarkable loss of one ester group was not explained. CO H We have prepared the 2 1 adduct but have found that the reaction can also give a number of other products including trimethyl 1-methylpyrrole-2,3,4-tricarboxylate (VI).This was first detected by Mr. N. J. Earl and proved identical with an authentic specimen for which we are greatly indebted to Professor R. A. Nicolaus. Heating the 2 1 adduct with dimethyl acetylenedicarboxylate gave the same pyrrole (VI) and trimethyl hemimellitate (V). The indole triester was obtained and hydrolysed to a tricarboxylic acid according to the directions of Diels et a1.l This acid on selective decarboxylation yielded 1-methylindole-4-carboxylicacid which was identified by comparison of both the acid and the methyl ester with synthetic specimens. An immediate Ehrlich colour was given by these last indoles but not by the triester or the triacid so the carboxyl groups were probably lost from the 2-and the 3-posi tion.The 2:l adduct can now be formulated as (IV) and Mr. P. Higham informs us that the nuclear magnetic resonance spectrum of the compound is consistent with this structure. The loss of one ester group on aromatisation and the formation of the aromatic esters (V and VI) by addition of dimethyl acetylenedicarboxylate across the 4,7-positions of the adduct (IV) followed by scission of the resulting molecule are now understandable. The adduct is Diels Alder and Winckler Annalen 1931 490 361. a Mandell and Blanchard J. Amer. Chem. Soc. 1957. 79 6198. MAY1961 165 probably formed by Diels-Alder addition to 1-correspond to the addition of the acetylenic ester to methylpyrrole yielding compound (111) which then compound (111) in an alternative sense.A similar reacts with a second molecule of the ester as indi- reaction occurs between benzene and bistrifluoro- cated. Other products have been isolated which -\ C02Me Me02C C02Me 4 Q:: ‘C-C0,Me CA -c ‘\C02Me Me (IV) Me02C C0,Me C0,Me (Vl) Me methyla~etylene~ to give 2,3,6,7-tetrakistriuoro-me th ylnap hthalene . 1-Benzylpyrrole and dimethyl acetylenedicarboxyl- ate gave a 1 :2 adduct and an indole triester whose ultraviolet adsorption spectra are extremely similar to those of the corresponding compounds derived from 1 -methylpyrrole. We thank the D.S.I.R. for a studentship (to J*M.V.)* (Received March 13th 1961.) Krespan McKusick and Cairns J.Amer. Chem. SOC.,1960 82 1515. The Stereochemistry and Conformation of Flavan Derivatives By J. W. CLARK-LEWIS and L. M. JACKMAN (IMPERIAL OF SCIENCE S. KENSINGTON, COLLEGE AND TECHNOLOGY LONDON S.W.7) AN examination of the nuclear magnetic resonance spectra of a number of flavan-3-ols flavan-3,4-diols 3-hydroxyflavan-4-ones and their derivatives has shown that spin-spin coupling constants between the 2-and 3- and the 3- and 4-protons define the stereochemistry and conformation of the hetero- cyclic ring. The pertinent results which however are not applicable to cyclic derivatives of the diols (car- bonates and isopropylidene derivatives) are sum- marised in the Table.J2.3 a 2(ax)3(ax) 2(ax)3 (eq) Flavan-3-01s 76-8.3 0-1.5 Flavan-3,4-diols 7.1-8.5 0-1.0 The 2,3-cis-coupling constant has a value close to zero the minimum value for spin-spin interaction across a carbon-carbon single bond,lV2 and this sug- gests that in 2,3-cis-derivatives one conformation is predominantly populated. It is also expected that in 2,3-trans-flavan derivatives the conformation in which both the 2-and the 3-substituent are quasi- equatorial will be highly favoured at room tempera- ture. If it be assumed that the 2-aryl group is always predominantly equatorial a self-consistent inter- pretation of the observed values is possible in terms of the established relation between coupling constant correspond to dihedral angles between the two C-H bonds of ca.90” and 150” respectively consistent with a half-chair conformation. Further the ob-served values for the two 3,4-coupling constants in trans-flavan-3-01s require the analogous interactions in the cis-isomers to be 5.6 [3(eq)4(ax)] and ca. 1.0 [3(eq)4(eq)] cycle/sec. and although these two coupling constants cannot be measured separately because the two methylene protons have the same chemical shift their observed mean value (3.0-3.3 J3,Q r \ 3(ax)4(ax) 3(ax)4(eq) 3 (eq)4(ax) 3(eq)4(eq) 9.0 5-6 6.5-6.7 3.94.3 0-1.0 cycles/sec.) is in good agreement with these predicted values. The observed values of 3,4-coupling constants in flavan-3,4-diols and their derivatives are in reason- able agreement with those determined or predicted for flavan-3-01s.Values for 3(eq)4(ax)- and 3(ax)4(eq)- interactions are somewhat lower and higher respec- tively than predicted and correspond to distortions of the dihedral angles by about 10” in the cis-cis- and 2,3-trans-3,4-cis-flavan-3,4-diols. It has been suggested3 that the O-ethyl derivative obtained by ethanolysis of isomelacacidin and the and dihedral angle in the system XH-CH:? 4-sulphone derived from either melacacidin or iso- Thus the observed cis-and trans-2,3-interactions melacacidin have the 2,3-cis-3,4-trans-stereochem-Karplus J. Chem. Phys. 1959 30 11. Conroy “Advances in Organic Chemistry,” Vol. 11 Interscience Publishers Inc. New York 1960 p. 31 1. Clark-Lewis and Mortimer J. 1960,4106.PROCEEDINGS istry. This is now confirmed in both compounds the 2,3-coupling constant is ca. 0-1 cycle/sec. showing that there has been no alteration in the 2,3-configura- tion and the 3,4-coupIing constant is also approxi- mately 1.0 cycle/sec. (compared with 4-0 found for the 3,4-cis-diols) in good agreement with the value predicted above for the 3 (eq)4 (eq) coupling constant. The 2,3-coupling constant in trans-3-hydroxy-flavan-4-ones is 12 cycles/sec. which is slightly lower than the value observed for the 1,2-diaxial coupling constant in 2a-bromochoIe~tan-3-one.~ The chemical shifts of the 2- 3- and 4-protons are Karplus J. Phys. Chem.,1960,64 1793. useful for making spectral assignments and the absorptions of aromatic protons can often be used to determine the hydroxylation patterns of the aromatic rings.These aspects together with a more critical discussion of the coupling-constant data will be reported later. The award of a Royal Society and Nuffield Foundation Commonwealth Bursary (to J.W.C.-L.) is gratefully acknowledged. (Received March 9th 1961.) The Structure of Bipyridyl-(1-acetylacetonyl)trimethylplatinum(1v) A. G. SWALLOW and MARYR. TRUTER (DEPARTMENT THEUNIVERSITY, OF INORGANIC AND STRUCTURAL CHEMISTRY LEEDS,2) THE empirical formulz of several co-ordination compounds of platinum(Iv) suggest that the central atom may have a co-ordination number other than six. For example complexes of trimethylplatinum(1v) with /?-diketones which from the empirical formula Me,Pt(R-CO-CH-CO-R’) appear to be derivatives of 5-co-ordinated platinum have been shown1 to consist of centrosymmetric dimeric molecules in which the platinum atom is actually 6-co-ordinated being linked to three methyl groups two oxygen atoms of one chelated p-diketone and to the “active methylene” carbon atom of the other p-diketone in the dimer.We now report our findings on an apparently 7-co-ordinatd platinum(1v) complex a compound first prepared by Lile and Menzies2 from a mixed solution of bipyridyl and the acetylacetone- trimethylplatinumdimerinbenzene; it isayellow crys- talline solid of formula Me,l?t(CHAc&C,,H,N& and the seven atoms which could be co-ordinated to the central platinum atom are the three methyl carbon atoms the two nitrogen atoms and the two oxygen atoms.Three-dimensional X-ray crystal-structure analysis reveals that the molecule is monomeric with the stereochemistry depicted in Fig. 1 and the structural formula (I). Each platinum atom is octahedrdly co-ordinated to the three methyl groups the two nitrogen atoms and to the “active methylene” carbon atom of the p-diketone. The p-diketone is not planar as it is in its chelated complexes and the carbonyl groups have quite different configurations (1) with respect to the Pt-CH bond. In this the first example of a p-diketone acting as a monodentate ligand it is the carbon atom which is co-ordinated. The preference of the platinum atom for co-ordina- tion to activated carbon atoms instead of to carbonyl oxygen atoms may underlie its behaviour as a hydrogenation catalyst i.e.for carbon-carbon double bonds rather than carbon-oxygen double bonds. Only the stereochemical results are presented here; refinement on three-dimensional observations col-Swallowand Truter,Proc. Roy. SOC.,1960 A 254,205; Hazel1 and Truter Proc. Roy. SOC.,1960 A 254 218. Lile and Menzies J. 1949,1168. my 1961 lected at 120"~is still proceeding. There are four molecules in the monoclinic unit cell a = 14-863 b = 8-480,c = 13.749 A fi = 99" 24' space group P2,/,. The platinum atoms were located from Patterson projections and the carbon oxygen and nitrogen atoms from a three-dimensional difference Fourier synthesis in which the oxygen atoms had the highest electron densities and the nitrogen atoms had higher electron densities than the carbon atoms.The agreement index R has fallen from 0.34 for the platinum atom alone to 0-084 for all the atoms ex- cluding hydrogen. Four cycles of least-squares re- finement have shown that no atoms have con-tinuously-increasing temperature factors and the refined co-ordinates lead to reasonable bond lengths and angles. We thank Mr. W. J. Lile for the crystals and the Royal Society and Imperial Chemical Industries Limited For the loan of some of the apparatus used. (Received March 6th 1961.) Rearrangement of Certain Quinoxalinecarboxyadlides. Isolation of an Intermediate in a Related N-Oxide Rearrangement By M. S.HBIB and C. W. REES (KING'SCOLLEGE LONDON, STRAND W.C.2) THEvery rapid transformation of the 1-oxide of the amide (I) into the base (In)by cold sulphuric acid proceeds1 by intramolecular electrophilic substitu- tion of the anilide ring by C(2)of the quinoxaline ring. Although the N-oxide function is partly responsible for this electrophilic centre at and is necessary for rapid decarboxylation by a cyclic mechanism,l it was expected that the base (I) would also be sensitive to acid. This is so; ice-cold sulphuric acid rapidly converted it into the isomeric spiro- lactam (LT) though the yield was low (25%). How-ever boiling ethanolic hydrogen chloride effected quantitative rearrangement. The structure of (11) Me H+ Me Me follows from appearance of an N-H absorption band at 3300 cm.-l shift of the amide carbonyl absorption to 1712 cm.-l (characteristic of 1 -methyloxindolesz) and conversion into the base (111) by prolonged boiling with hydrochloric acid.The 6-chloro-derivative of (11) has been prepared by treatment of the 1-oxide of the amide (I)with ethanolic hydrogen chloride and its structure similarly dem~nstrated.~ The methylanilide (IV; R1= H R2= Me R3= Ph) is similarly rearranged by cold concentrated sul- phuric acid or hot ethanolic hydrogen chloride to the and the pyrazine methylanilide analogous to the quinoxaline (I) were all unchanged by concentrated sulphuric acid at room temperature. Thus the scope of this rearrangement is very similar to that of the corresponding 1-oxides1 and a similar mechanism is indicated; an anilide with a free ortho-position is necessary and electron-withdrawal from this ring inhibits reaction.The following mechanism is there-fore suggested MQ IVa; R1= Me R2 = R3 = H b; R1 = $22 = R3 = Me c; R1 = R2 = Me R3= 2 6-C6H,Me d; R1 = R2= Me R3= pNO2C,H Isolation of the spirolactams and their much slower formation when sulphuric acid is replaced by ethanolic hydrogen chloride led us to attempt the isolation of a corresponding 1 -hydroxy-com- pound (e.g. VI) postulatedl as intermediates in the corresponding spirolactam which forms @I)rearrangement of the 1-oxides (e.g. V). This was on methylation. However the amides (IVa b c and d) successful with the oxide (V) which when heated with Habib and Rees J.1960 3371 a Wenkert Bose and Reid J. Amer. Chern. Soc. 1953 75 5514. Clark-Lewis and Katekar J. 1959 2825. PROCEEDINGS ethanolic hydrogen chloride for 3 hr. gave the N-hydroxy-spirolactam (VI). Me Me Me ('1"HCI [:Io/o / H,SO [:,% \+' Y yo 0-NMePh Ho'O ke / (v> (VI) (VI I> The spirolactam (VI) like the N-oxide (V) itself was converted by sulphuric acid into the amine (VII) thus providing strong support for our mechanism of acid-catalysed reaction of the N-oxides through the corresponding spiro-compounds We thank Professor D. H. Hey F.R.S. for his kind encouragement. (Received March 3rd 1961.) The LFynthesis of Derivatives of Corrole. An Amendment By A. W. JOHNSON and I.T. KAY (THEUNIVERSITY, NOTTINGHAM) JOHNSON and PRICE^ recently described the cyclisation of the palladium derivative of a bi(dipyrromethen-5-yl) (I; R = Br) with formaldehyde and hydro- chloric acid to the palladium corrole derivative (11). As a result of a suggestion from Professor R. B. Woodward whom we thank for his interest this reaction has been re-examined with the consequence that it is necessary to re-formulate the product as the palladium derivative of the oxide (111). Me Me R Me I O Thus when the palladium derivative of (I; R = Br) is heated under reflux in ethanolic hydrochloric acid in the absence of formaldehyde the palladium derivative of compound (111) is obtained in approxi- mately 45 % yield.When an alcoholic solution of the cobalt derivative of the dibromide (I R = Br) was heated with formaldehyde and hydrochloric acid a red compound previously formulated as (I; R = CH2.0H)was obtained. This has now been prepared from the metal-free derivative (I R = Br) by treat- ment with ethanolic hydrochloric acid without formaldehyde and the product which does not form Johnson and Price. 1.. 1960. 1649. metallic derivatives has been re-formulated as the diamide (IV). In addition to (IV) a second product has been isolated from the reaction and this is represented as the monochloro-monoamide (V). Et Et Et Et Treatment of the monoamide (V) with metal salts gave ca. 30% yields of the metal derivatives of the cyclic oxide (111) and it is assumed that in the presence of salts (V) tautomerises to the lactim form in order to permit the formation of the mononuclear metallic complex.This difference in reaction between (JV) and (V) finds a parallel in the bile pigment series where the verdins form metal complexes whereas the rubins do not. The cyclisation of the metallic deriva- tives of (V) to (111) is simulated inter al. by the formation of the vitamin B12 lactone2 and the selective fission of tryptophan units from pep tide^.^ When the copper but not the nickel or the palladium derivative of (111) was treated with concentrated hydrochloric acid it slowly reverted to (V); this sug- gested that the action of cold concentrated sulphuric acid 011 the copper derivative of (111) might give the parent metal-free macrocycle which was found to be the case.The analysis of the parent compound (111) proved that the oxygen atom was part of the macro- cycle and not attached to the metal; treatment with metal salts in chloroform-methanol rapidly gave the metallic derivatives which were detected by their characteristic spectra. Bonnett Cannon Clark Johnson Parker Smith and Todd J. 1957 1158. Patchornik Lawson Gross and Witkop J. Arnev. Chern. Soc. 1960 82 5923. MAY1961 169 Evidence has been obtained for the existence of We thank the Department of Scientific and macrocycles corresponding to (111) containing a Industrial Research for the award of a maintenance nitrogen bridge and experiments are also in hand to grant to one of us (I.T.K.).prepare the sulphur and carbon bridge analogues. (Received March 6th 1961.) Synthesis and Structure of Complexes of the Type [Co,(CO),(CY)] By W. T. DENT,L. A. DUNCANSON, R. G. GUY,H. W. B. REED,and €3. L. SHAW CHEMICAL LIMITED CHEMICALS (IMPERIAL INDUSTRIES HEAVYORGANIC DIVISION RESEARCH BILLINGHAM) DEPARTMENT ON the basis of degradative work Kriierke and Hubell recently suggested that the trinuclear cobalt complexes [Co,(CO),(C,H,R)] (R = H alkyl aryl) formed by the treatment of the binuclear acetylene complexes [Co,(CO),(RC i CH)] with acids2 have the structure (I); (Y = CH,R). We have reached the same conclusion as a result of a novel synthesis and spectroscopic results. Dicobalt octacarbonyl or tetracobalt dodecacar- bony1 reacts in ethanol with trihalogenomethyl derivatives of the type CX,Y (X = halogen) to form derivatives [Co,(CY)(CO),] to which we assign the characteristic of the CY group; for example when Y = F there is a band at 1167 cm.-l and when Y = C1 at 903 cm.-l which can be assigned to C-F and C-Cl stretching modes raised in frequency from their values in the alkyl halides (1100-lo00 cm.-l for fluorides and 750-700 cm.-l for chlorides3) by the unusual intramolecular environment of the carbon-halogen bonds.The compound prepared by the reaction of di- cobalt octacarbonyl with 1,l ,1-trichloroethane in ethyl alcohol (I; Y = Me) is identical with that pre- pared by the method of Markby et al. from Frequency range No. of bands 1Suggested assignment 2110-2116 crn.-l 1 (m) 2047-2070 2 6) c-0 stretch 2016-2040 1 (m) 650-600 1 (variable intensity) i 527-5 3 2 1 6) 503-509 1 6) 470-480 1 (w structure (I).Compounds having Y = Me Cl F Ph CO,Et CO,H CH(OAc), and CF were pre pared and were purified by steam-distillation or recrystallisation. They form dark purple crystals giving deep purple solutions in organic solvents. The structure (I) was assigned on the basis of this method of synthesis and on their infrared spectra charac- teristic bands in which are given in the Table. Y All the carbonyl groups must be terminal because there are no bands which can be attributed to bridging carbonyl groups and the Co-C frequencies are of two types those in the 650-600 cm.-l region which are susceptible to the groups Y and so are presumably mainly Co-CY stretching modes and those in the 527-532 cm,-l region which are almost independent of Y and are probably mainly Co-CO stretching frequencies.In addition there are bands Kriierke and Hubel. Chem. and Znd.. 1960. 1264. Co-C stretch [Co,(CO),(CH i CH)] by treatment with acid. All the higher-frequency bands in its spectrum can be assigned on the basis that the molecule contains a methyl group 2924m 2882m 2817m (Me stretching mode and probably an overtone of the band at 1420 cm.-l enhanced in intensity by Fermi reson- ance) 1420m 1359m (Me deformation) 1161m and 1006s cm.-l (Me rocking and C-Me stretching mode). Similarly the spectrum of the compound prepared from cobalt octacarbonyl and CBr,ClCF is consistent with structure (I) where Y = CF,.Thus there are strong bands at 1284 (vCF as.) 1117 (vCF3 sym.) and 957 cm.-l (vC-CF,) the first two bands being assigned by analogy with CClF which has CF stretching frequencies of 1210 and 1107 ~m.-l.~ The compounds described by Markby et af. are therefore of the same type as those we have pre- pared and probably have the structure suggested by Kruerke and Hube1.l The authors are grateful to Drs. J. Chatt and D. M. Adams for valuable discussion. (Received February 14th 1961 .) Markby Wender Friedel Cotton and Sternberg J. Amer. Chem. SOC.,1958 86 6529. Bellamy “Infra-red Spectra of Complex Molecules,” Methuen and Co. London 1958 pp. 329 330. Thompson and Temple J.1948 1522. 170 PROCEEDINGS Stereochemistryof a Five-coadma ted Piatinurn Compound By G. A. MAIR H. M. POWELL, and L. M. VENANZI (CHEMICAL LABORATORY CHEMISTRY , CRYSTALLOGRAPHY AND INORGANIC LABORATORY OXFORD UNIVERSITY) A NEW ligand tris-(u4phenylarsinophenyl)arsine (I = QAS) reacts with platinum(1r) salts to form complexes' of the type WX(QAS)]Y where X = C1 Br I or NCS and Y = X C104 or BPh,. Their properties indicate that the platinum is five-co- ordinated. To confirm this and to determine the stereochemistry of platinum in this type of com- pound the crystal structure of [PtI(QAS)]wPh,] has been examined. Results. [PtI(QAS)][BPh,] A4 = 1631.9 triclinic pinacoidal a = 10-1 b = 18.1 c = 19.0 A 01 = y 73+",18 = No = 91" U = 3335 As Dm = 1-63 (by flotation) 2 = 2 D = 1.625.Space group 5 (Cil No. 2). Cu-K radiation single-crystal photo- graphs. The structure was solved from a three-dimensional Patterson synthesis evaluated with 5000 independent F& as coefficients. 0. S. Mill's programme for the Mercury computer was used. Owing to the large sizes of both anion and cation the high peaks within 5 A of the origin of necessity represent vectors be- tween heavy atoms of the same cation. Their number heights and positions when considered in relation to the known chemical composition establish that the structure is centrosymmetric and give positions for all the heavy atoms. An approximately linear I-Pt-As group in the cation causes overlap of the corresponding Pt-I and Pt-As vectors.This arsenic atom and the iodine atom were originally assigned co-ordinates as though their distances from platinum were equal. The R value for Fhkl,calculated on the co-ordinates deduced was initially 39 % and by least- squares refinement was reduced rapidly to 22% these Pt-As and Pt-I distances becoming distinct. J. S.Rollett's least-squares programme was used. So far only heavy atoms have been considered in the calculations and only these atoms are shown in the figure representing co-ordination of the platinum atom. The platinum bonds point towards the corners of a trigonal bipyramid which has iodine at one apex and what is presumed to be the central arsenic at the Brewster Savage and Venanzi personal communication.a Harris Nyholm and Phillips J. 1960. 4379. Mair Powell and Henn Proc. Chem. Soc. 1960 415. other. From the present heavy-atom co-ordinates which are expected to undergo little change in the course of structure refinement the interatomic distances are approximately those that would be expected for single bonds to platinum and the angles do not differ greatly from those of a regular trigonal bipyramid. The platinum atom is displaced slightly from the plane of the three equatorial arsenic atoms towards the iodine atom. Co-ordination of the platinum atom. No other five-co-ordinated platinum complex has a knownstereochemical form. Usually platinum@) is four-co-ordinated but evidence has been obtained for the existence of five-co-ordinated compounds in solution.2 In the closest analogue of known struc- t~re,~ the molecule N~B~,,M~As(CH,CH,CH~-ASM~~)~ the arrangement of nickel bonds is that of a distorted tetragonal pyramid but the palladium and platinum compounds of similar formula contain four-co- ordinated metal atoms.The adoption of the trigonal bipyramidal form in the present instance is deter- mined by steric factors including the potential trigonal symmetry of the ligand which makes possible the attachment of the three remaining arsenic atoms in an identical manner when the central arsenic atom is placed at an apex. We thank D.S.I.R. for a grant (to G.A.M.) and the Oxford University Computing Laboratory for facilities. (Received,March loth 1961.) MAY196 1 171 The Nature of the First Absorption Band of Nitrobenzene By M.GODFREY and J. N. MURRELL (DEPARTMENT SHEFFIELD, OF CHEMISTRY THE UNIVERSITY 10) THEabsorption spectrum of nitrobenzene starts with a shoulder at 3300 A. Plattl suggested that this is due to an n+n* transition (classified as lAtl U)similar to that of the 2700 A band of nitromethane (E 20). A theoretical analysis of the spectrum by the method of Nagakura and Tanaka? and MurrellY3 suggests that the band is in fact due to a n+n* transition. We now submit evidence which supports our interpretation. Wolf and Herold4 showed that the 3300 A band of nitrobenzene was shifted towards red on changing from heptane to methanol as solvent.This is contrary to the characteristic blue shift by polar solvents which is used to identify n+n* bands.5 Wavelength (s) 2250 2500 2750 30003?503;500,4?00 4.51 Absorption spectra of (I) nitrobenzene in heptane? (2) 2-nitropyridine in cyclohexane and (3) 2-nitro-pyridine in ethanolic HC1. Nitrobenzene n+n* states would have an excess of electrons in the benzene ring since the n orbital is localised on the oxygen atoms whereas the n* orbital is delocalised over the whole molecule. On the other hand the charge-transfer T+T* states would be electron-deficient in the benzene ring since the niko-group behaves as a n-electron-acceptor. It follows that replacing one of the benzene -CH groups by the more electron-attracting groups -N Platt J.Chem. Phys. 1951 19 101. Nagakura and Tanaka J. Chem. Phys. 1954 22 236. Murrell Proc. Php. SOC.,1955 68 A 969. Wolf and Herold 2.phys. Chem. 1931 B 13 201 McConnell J. Chem. Phys. 1952 20 700. Murrell Mol. Phys. 1958 1 384. 'Braude Jones and Rose,J. 1947 1104. and -NH+ will produce a fmt-order red shift in an n+n* band. The spectra of 2-nitropyridine and its conjugate acid (Figure) show a definite blue shift of the 3300 A band for the pyridinium cation although there is very little difference between the spectra of nitrobenzene and 2-nitropyridine in this region. The latter is not too unexpected since the second-order inductive effect of a nitrogen atom will give rise to a red shift in this band if it is n+n* which will tend to counteract the first-order blue shift.6 The n+n* band of the carbonyl group undergoes a large blue shift when an amino- or a hydroxy-substituent is introduced into the a-position.Thus acetic acid and acetamide have bands around 2000 A compared with the 2810 8 band of acetaldehyde. This effect is due to the donor property of the substituent group which makes it energetically more difficult to move electrons from the oxygen n orbital into the C=O n* orbital. On substituting a phenyl or vinyl group into the a-position of nitromethane one is again introducing a group which behaves as a donor relative to the nitro-group since the latter is a strong electron-acceptor. By analogy with the behaviour of the carbonyl band we again expect some tendency towards a blue shift in the nitro- group n+n* band in these cases.Nitro-olefins do not show low-intensity absorption which could be interpreted as being an n+r* band.' Likewise crotonic acid shows no n+n* band since the car- boxyl group like the nitro-group is an electron- acceptor. Our theoretical calculations predict a band at 3100 A which is mainly charge transfer (82%) in which an electron is excited from the highest occupied symmetric orbital of the benzene nucleus into the lowest vacant n-orbital of the nitro-group. The intensity of the band (E -165) is rather large for a forbidden n+n* but small for an allowed n+T* transition. However our calculations show that this will be weak because of an out-of-phase combination of the transition moments of the benzene locally excited and the charge-transfer con- tributions to the wave functions.By comparison the corresponding band of aniline (Amax. 2350 A; E 8000) in which the amino-group is acting as a donor is strong because here the transition moments are in-phase. (Received March 20th 1961.) PROCEEDINGS Monomethylenation and Polymethylenation by Diazometbane in the Presence of Boron Compounds By ALWYN D. G. HARE,0. R. KHAN and J. SIKORA G. DAVIES RAMSAY LABORATORIES COLLEGE W.C. 1) (WILLIAM AND RALPHFORSTER UNIVERSITY LONDON A WIDE variety of boron compounds (e.g. the halides alkoxides and alkyls) in minute amount catalyse the conversion of diazomethane into poly- methylene and nitr0gen.l Two mechanisms have been proposed for the reaction.In both the first step is the nucleophilic attack of diazomethane on boron to give the intermediate (I); the first then assumes a cationic chain propagated by a diazonium or carbonium ion (e.g. eqn. l) and the other a repeating nucleophilic rearrangement of the group X*[CH2],(n = 0 1 2 . . , in order) from boron to carbon (eqn. 2). The latter mechanism is now widely accepted.2 Mechanism (2) requires that the average degree of polymerisation for the reaction of diazomethane with a trialkylboron e.g. (C5H11),B should be given by the equivalence ratio of the reagents CH,N :3R,B because the reactivity of the first and subsequent X,B-CH,X x,B;cH,-,, I -..I X-CH, -0 X,B CH,N R-[CH212.B R.[CH,],.B etc.and poly-methylene can be built up by his second route. We have investigated the partitioning of the diazomethane between these three types of product [CH,A X*[CH,];B (n = 1 or more) and poly- methylene] as the nature and concentration of the reactants are varied. We suggest that the inter- mediate (I) can react by both paths (1)* and (2) at comparable rates but that subsequent steps to poly- methylene by route (1) are relatively very fast and account for most of the diazomethane so that the yield of methylenation product by route (2) is usually very small. The reagent AH competes with diazo- methane for nucleophilic attack on the electrophile (I) blocking the cationic chain (1) and producing CH,A by reaction (3); a small amount of the capture of the chain carrier may in some cases be delayed to the second or third stage of the chain giving -N7_ X,B.CH~CH,X etc (2) X,$*CHih CH2N2+ X,BCH~CH$J X,B.CH,.CH~CH,JJ,, -N etc ( 11 -+ products (C5Hl1),B-CH2C5Hll etc.should be similar to that of the reactant. Yet we find that even with an excess of trialkylboron only polymethylene of high molecular weight is formed and that little or no monomethylenation product can be detected. The species propagating the polymerisation must there- fore be much more reactive towards diazomethane than is the original trialkylboron and the commonly accepted mechanism (2) must be of negligible importance in producing polymer. If a nucleophilic reagent AH (e.g.,an amine water or an alcohol) is added to the system the overall rate of disappearance of the diazomethane is decreased and no polymer is formed.Some of the diazo- methane then reacts with AH to give CH,A and the rest methylenates the RB group to RCH2.B:. With a larger amount of diazomethane further methylene groups can be introduced giving structures accompanying low yields of CH,CH,.A and CH,-CH,-CH,A.3 As most of the diazomethane is now not being lost to reaction (l) the methylenation (2)can be observed. These reactions are useful preparatively for the methylenation of AH (e.g. amines and alcohols) under mild conditions and for the conversion RB -+R-CH,.B:; for example we have prepared the first neopentylboron compounds by this reaction. If our interpretation of these reactions is correct it should be possible to choose conditions under which linear polymethylene of preselected molecular weight could be prepared by reaction (1).We are grateful to Dr. A. Ledwith for a number of stimulating discussions on this subject. (Received March 13th 1961.) * We would not exclude a number of possible variants of equation (l) for example where the chain is propagated Phraugh the ion pair X,B.[CH,] ",+BX4- again by an SN2or SNlprocess. Meenvein Angew. Chem. 1948,60 78. The present position is reviewed by Bawn Ledwith and Matthies J. Polymer Sci. 1959,34,93. Bawn and Ledwith Chem. andhd. 1958,1329. MAY1961 173 The Mechanisms of Diazotisation at htermediate Acidities By B. C. CHALLIS and J.H. RIDD COLLEGE STREET W.C.1) (UNIVERSITY GOWER LONDON THEmechanisms of diazotisation have already been discussed for media of very low acidity1 and of very high acidity; this Communication is concerned with part of the intermediate region involving concentra- tions of peichloric acid between 0.5~ and 3.0~.In these conditions the dependence of the reaction rate on acidity appears to be very complex,S but our results suggest that this complexity arises mainly from a medium effect combined with one new kinetic term. Ionised perchlorates (0+-3-0~)have a catalytic effect on diazotisation in dilute perchloric acid the ratio of the rates in the presence and in the absence of the salt depends exponentially on the salt concen- tration.The form of this catalysis and the dependence on the nature of the metal ion do not suggest that nitrosonium perchlorate is an intermediate in the reaction and this accords with other evidence4 that The kinetics of diazotisation of o-chloroaniline (intermediate in basicity between aniline and p-nitroaniline) can be analysed into the sum of con- tributions from the kinetic terms in equations (1) and (2); these terms are therefore considered to repre- sent two independent mechanisms of diazotisation not two stages of a continuously variable dependence on acidity. The first kinetic term has already been assigned to a reaction of the free amine with the nitrous acidium ion,lV5 the second kinetic term has not been discussed before. It is unlikely that the new kinetic term corresponds to a reaction between the free amine and the nitro- soniwn ion for as the acidity is increased the incur- sion of this mechanism should be first observed with the least basic amines.In contrast the new kinetic term is first observed with the most basic aniines [while the diazotisation of p-nitroaniline is still Diazotisation at 0" in aqueous perchloric acid containing sodium perchlorate. Ionic strength = 3.0. Initial [HNO,] = 10-3-10-5~. Initial [ArNH3+] = 10-2-10-5~. Units k in rnole-'sec,-ll. [HClO,I(M) 0-5 1.0 ho 2-52 5-11 k (p-nitroaniline) 80.0 81.1 0-158 0.146 '3 {E:tEiine) 0.790 0.836 nitrosonium perchlorate is fully ionised. The reason- able assumption that perchloric acid has a medium effect similar to that of sodium perchlorate greatly simplifies the interpretation of the acid catalysis.This is illustrated by the kinetics of diazotisation in aqueous perchloric acid containing sufficient sodium perchlorate to maintain the ionic strength at 3.0; the diazotisation of p-nitroaniline then agrees well with equation (1) and that of aniline and p-toluidine with equation (2). Rate = k,[ArNH,+] [HNO,] . . . (1) Rate = k,[ArNH,+][HNO,]h . . . (2) In considering the constancy of k and k (see Table) it should be appreciated that in perchloric acid alone the rate of diazotisation of aniline increases by more than a hundred-fold over this range of acidity. :k3 in mole-2sec.-11.2. 1.5 2-0 2.5 3.0 7.91 11.0 14.9 17.0 - 84-1 - 93.0 0.142 0.146 0.140 0-153 0.848 0.872 0.785 0.865 following equation (1)1.The new kinetic term is also associated with an unusual discrimination between different amines forp-toluidine is more reactive than aniline by a factor of six; at somewhat lower acidities the more basic amines are less reactive because of the lower concentration of free amine in the solution. From these and related considerations we have con- cluded that the new kinetic term is unlikely to arise from a rate-determining nitrosation of the free amine and we suggest that the reaction effectively involves nitrosation of the protonated amine in such a way that the proton being displaced is still present in the transition state. This is equivalent to an SE~ reaction at a protonated nitrogen atom.It is probably significant that the rate of diazotisation can con- siderably exceed the calculated encounter rate between amine molecules and nitrosonium ions? This approach is not a return to the old theory that the protonation of the amine is necessary before l Hughes Ingold and Ridd J. 1958 58 and following papers. a Challis and Ridd Proc. Chem. Soc. 1960 245. Schmid and Essler Monatsh. 1959 90 222. Singer and Vamplew J. 1956 3971. Larkworthy,J. 1959 3304. Challis Ph.D. Thesis London 1960. PROCEEDINGS diazotisation. Reaction occurs much more readily We thank Professor E. D. Hughes F.R.S. Sir through the free amine but if the amine is almost Christopher Ingold F.R.S.,and Dr. C. A. Bunton completely protonated interaction of the nitrosating for many helpful discussions.One of us (B.C.C.) agent with the protonated amine can apparently thanks the D.S.I.R. for a maintenance award. become significant. (Received March 22nd 1961.) The Electron-spin Resonance Spectrum of thep-Xylene Positive Ion By J. R. BOLTONand A. CARRINGTON OF THEORETICAL UNIVERSITY (DEPARTMENT CHEMISTRY OF CAMBRIDGE) 1~ is now well known that when large conjugated electrons. This effect is reflected in the reduced ring-aromatic hydrocarbons are dissolved in con-proton splitting and the large methyl-proton centrated sulphuric acid they lose an electron to splitting. form fairly stable positive ions which give well- resolved electron-spin resonance spectra.]e2 The smallest hydrocarbon which behaves in this manner seems to be anthracene.Naphthalene and benzene are not oxidised by concentrated sulphuric acid but it is the purpose of this Communication to report that fairly stable positive ions of the xylenes are formed by dissolution of the hydrocarbon in con- centrated sulphuric acid containing potassium per- sulphate. We have analysed the proton hyperhe structure of the p-xylene positive ion whose spectrum is shown in the Figure. The six methyl protons give rise to seven lines which are split by the four ring protons. The splitting constants are acHs= 3-89 gauss a .-3-00 gauss and the Figure shows a reconstruction of the spectrum based on these splitting constants. Further confirmation of the assignments was obtained by carrying out the pre- paration in concentrated dideuterosulphuric acid.The ring protons exchange rapidly with deuterium and the resulting spectrum consists of seven broad lines with traces of unresolved structure. The separation between these lines is still 3-89 gauss. There does not appear to be any information in the literature on the hydrogen exchange of p-xylene in D,SO, but it has been shown that in deuterium Top E.S.R.spectrum of p-C,H,Me$. Centre Re- bromide exchange of ring protons of mesitylene and construction based on the splitting constants given in durene is complete in 30 minutes whilst the methyl the text. Bottom E.S.R. spectrum of p-CsD,Me,+. protons do not exchange ~ignifkantly.~ The values of the proton hyperfine splitting con- We have obtained strong reasonably well-stants are extremely interesting and quite different resolved spectra from other substituted benzenes from those obtained from the p-xylene negative dissolved in concentrated sulphuric acid with per- Voevodsky et al.found no splitting from the sulphate notably toluene o-and m-xylene mesityl- methyl protons and a hyperfine constant of 4-00 ene and durene. Benzene itself gives a green solution gauss for the ring protons. Our results demonstrate which displays a poorly resolved seven-line spectrum the ability of the methyl group to help fill the hole that could conceivably arise from six equivalent in the benzene n-electron system by donation of protons. However the unexpected feature of this Weissman de Boer and Conradi J.Chem. Phys. 1957,26 963. a Carrington Dravnieks and Symons J. 1959 947. Kaliuachenko Varshavskii and Shetenshtein Doklady Akad. Nauk S.S.S.R.,1953,91 577. Tuttle and Weissman J. Amer. Chem. SOL 1958 80 5342. Voevodsky Solodovnikov and Chibrikin Doklady Akad. Nauk S.S.S.R. 1959 129 1082. MAY1961 spectrum if it is due to the benzene positive ion is that the splitting between the lines is only 2.89 gauss compared with 3.75 gauss for the benzene negative ion? Hitherto it has been found that the positive and the negative ions of even alternant hydrocarbons have very similar electron spin resonance spectra the splitting for the positive ions being slightly larger than for the negative ions. This is in accord with theoretical predictions.In the absence of any obvious reason why this splitting constant should be so small we cannot conclude without further evidence that the spectrum arises from the benzene positive ion. The spectra were obtained by using a Varian 100 kc. spectrometer. One of us (J.R.B.) is indebted to the Shell Inter- national Petroleum Company for the award of a Post-graduate Research Studentship. We thank the D.S.I.R. and General Electric U.S.A. for generous financial SUPPOrt- (Received March 2 1st 1961.) NEWS AND ANNOUNCEMENTS New Members of CounciI.-The following new appointments to the Council were announced at the Annual General Meeting Vice-president who has not filled the Osee of President Dr. E. J. Bowen Honorary Treasurer Dr.J. W. Barrett Elected Ordinary Members of Council Constituency I Dr. L. J. Bellamy Professor D. P. Craig Dr. W. Gerrard Mr. H. M. Powell Constituency 11 Professor D. H. Everett Constituency III Dr. A. K. Holliday Dr. J. Honeyman Constituency V Dr. G. 0. Aspinall Constituency VI Professor W. Cocker Chemical Society Lectureships.-The Council has made the following appointments for 1961-62 Faraday Lectureship Sir Christopher Ingold Liversidge Lectureship Professor C. E. H. Barn Tilden Lectureships Dr. J. Chatt Professor H. B. Henbest Centenary Lectureships Professor G. B. Kistiakowsky Professor H. Schmid LocaI Representative.-Dr. L. M. Venanzi has been appointed as the Local Representative for Oxford in succession to Dr.M. L. Tomlinson who has resigned. Chemical Society Liaison OfEcers.-The following additional Fellows have agreed to act as Liaison officers Aero Research Ltd. (Ciba) Cambridge Ashburton Chemical Works Manchester Bolton Technical College Dr. R. F. Webb Dr. F. R. Basford Dr. G. W. Wood Clayton Aniline Company Manchester Dr. E. N. Abrahart Chesterford Park Research Station Fisons Pest Control Ltd. Saffron Walden Dr. G. T. Newbold Robert Gordon’s Technical College Aberdeen Dr. M. B. Watson Hatfield Technical College Dr. R. F. Robbins Imperial Chemical In- dustries Limited Dye- stuffs Division Blackley Dr. G. de W. Anderson Imperial Chemical In- dustries Limited Phar- maceutical Division Alderley Edge Manchester University Petrochemicals Ltd.Urmston The Technical College Plymouth Royal Technical College Salford Stockport College for Further Education Dr. R. Hull Dr. G. F. Smith Dr. A. V. Mercer Dr. B. L. Tonge Dr. H. Suschitzky Mr. H. H. Armstrong Chemical Society Symposia.-Three symposia will be arranged in conjunction with the Anniversary Meetings of the Chemical Society to be held in Sheffield during the period April 3rd-5th 1962 “Some Aspects of the Chemistry of Natural Phenols.” “Reactivity and Structure in Inorganic Chem-istry.” “The Transition State.” The papers given at the organic and the inorganic discussions will not be published in full but abstracts will be available to those who register for the meet- ing.The symposium on “The Transition State” will be published by the Society and full preprints of the contributions will be available before the meeting. Full details of these Symposia and of the Anniver- sary Meetings of the Society will be available in December 1961. These will be sent to all Fellows of the Society and will be available to others on application from the General Secretary The Chemical Society Burlington House London W. 1. Index Chemicus.-The Council has agreed to collaborate in a proposal submitted by the Institute of Scientific Information of Philadelphia whereby Fellows of the Society may obtain at preferential rates Index Chemicus-a register and index of new chemiczl compounds.Further details will be sent to Fellows in due course. Royal Society.-Pro fessor Alexander Nicolai Nesmeyanov President of the Academy of Sciences of the U.S.S.R. Moscow has been elected a foreign member of the Royal Society. New Chairman of the D.S.I.R. Research Council.- The Lord President of the Council and Minister for Science Lord Hailsham has appointed Sir Harold Roxbee Cox as Chairman of the Council for Scientific and Industrial Research for five years from October lst 1961. He will succeed Sir Harry Jephcott who has been Chairman of the Council since its formation under the D.S.I.R. Act of 1956 and who will complete his term of office in September. Sir Harold has been a member of the Council since 1957.British Association for the Advancement of Science. The British Association is to hold its Annual Meeting in Nonvich from August 30th to September 6th 1961. The Annual Meeting is the largest scientific gathering of its kind in the year and the only one which members of the general public can join on equal terms with scientists. It provides not only a platform on which scientists can discuss their work with their colleagues in their own language and one on which scientists in separate but related fields can consider the “growing points” of science but affords an unrivalled chance for the layman to learn some- thing of the progress of science from the scientists concerned. Further details of this meeting can be obtained from the offices of The British Association at 18 Adam Street Adelphi London W.C.2.PROCEEDINGS International Congresses etc.-An International Conference on the Metallurgy of Beryllium will be held in London on October 16-18th 1961. En- quiries should be addressed to the Secretary The Institute of Metals 17 Belgrave Square London S.W.1. A British Insecticide and Fungicide Conference will be held in Brighton on November 7-10th 1961. Enquiries should be addressed to the Secretary The Association of British Manufacturers of Agricultural Chemicals Cecil Chambers 86 Strand London w.c.2. An International Symposium on Purest Sub-stances in Science and Technology will be held in Dresden on November 30th-December 12th 1961. Enquiries should be addressed to the Secretary Chemisches Gesellschaft in der Deutschen Demo- kratischen Republik Unter den Linden 68/70 Berlin W.8 Germany.An International Colloquium on Ionic Bombard- ment will be held at Bellevue on December 4-8th 1961. Enquiries should be addressed to National Scientific Research Centre 15 quai Anatole France Paris 7e France. An International Conference on Spectroscopy organised by the Society for Applied Spectroscopy will be held at the University of Maryland on June 18-22nd 1962. Enquiries should be addressed to Bourdon F. Scribner Conference Chairman Inter- national Spectroscopy Conference National Bureau of Standards Washington 25 D.C. U.S.A. An International Symposium on the Chemistry of Natural Products sponsored by the International Union of Pure and Applied Chemistry will be held in Prague in August 1962.Enquiries should be addressed to Chemicky ustav CSAV Na cvicisti 2 Prague-Dejvice Czechoslovakia. An International Colloquium on Organometallic Derivatives will be held in Paris in September 1962. Enquiries should be addressed to Professor Henri Normant FacultC des Sciences UniversitC de Paris A la Sorbonne 47 Rue des Gcoles Paris 5e France. The Second International Congress and Exhibition of Laboratory Measurement and Automation Tech- niques in Chemistry will be held in Bade on October 15-20th 1962. Enquiries should be addressed to the Secretary ILMAC Clarastr. 61 Basle Switzer- land. The Third World Metallurgical Congress will be held in Chicago on November 5-9th 1962.En- quiries should be addressed to Chester Wells Exposition Manager American Society for Metals 7301 Euclid Avenue Cleveland 3 Ohio U.S.A. Van’t Hoff Fund.-The Committee of the Van? Hoff Fund for the endowment of investigations in the field of pure and applied chemistry invites applications for grants from the fund. The amount MAY1961 available for next year is about 2,000 Dutch guilders. Applications should be sent by registered post to Het Bestuur der Kon. Ned. Akademie van Weten- schappen bestemd voor de Commissie van het “Van’t Hoff Fonds,” Trippenhuis Kloveniers-burgwal29 Amsterdam before December lst 1961. The purpose for which the grant is required the reasons for the application and the amount desired must be stated.Grants from the Fund for 1961 were awarded to Dr. 2. Brada (Brno) and Professor T. Urbanski (Warsaw). Election of New Fellows.-1 57 Candidates whose names were published in Proceedings for March have been elected to the Fellowship. Deaths.-We regret to announce the deaths of the following Dr. T. Malkin (25.4.61) Reader in Organic Chemistry University of Bristol; Mr. F. Owen (31.1.61) of Manchester; Mr. A. F. Hollis (2.4.61) of the University of Sydney; Sir David Rivett (1.4.61) former Chairman of the Common- wealth Council for Scientific and Industrial Research ; and Mr. E. A. Tyler (5.4.61) of Swansea and a Fellow since 1899. Personal.-Dr. A. Couper has been appointed Senior Lecturer in Inorganic and Physical Chemistry at the University of Bristol.Dr. J. A. W. Dalziel of Imperial College has been appointed Reader in Inorganic and Analytical Chem- istry at Chelsea College of Science and Technology as from September next. Dr. J. D. Dunaldson of the University of Aberdeen has been appointed to the post of Lecturer at Chelsea College. On May 13th at Antioch College Yellow Springs U.S.A. the Patterson award was presented to Dr. G. Malcolm Dyson of Loughborough Leicester- shire. The award is made every two years in honour of the late Dr. Austin M. Patterson who was for a time editor of Chemical Abstracts and for many years scientific editor of “Webster’s Dictionary.” This is the first time the award has been given to a non-American.Dr. Dyson acts as consultant director of research to Chemical Abstracts. He is carrying out an extensive programme of research in the storage and retrieval of scientific data by computer tech- niques and has devised many special technical advances in this subject. Dr. U. R. Evans has received the Honorary Degree of Doctor of Metallurgy at the University of Sheffield. Dr. N. A. C. Friend has been appointed European Technical Manager of the Canadian Chemical Com- pany Ltd. of Montreal. Mr. W. Hunter has been appointed a Director of British Celanese Ltd. Dr. D. W.Kent-Jones has been made an Honorary Member of the American Association of Cereal Chemists. Dr. M. F. Lappert has been appointed Senior Lecturer in Chemistry Faculty of Technology University of Manchester as from October 1st.Dr. S. I;. Mason has been appointed Senior Lecturer in the Department of Chemistry of the University of Exeter. Professor D. M. Newitt who retires on September 30th 1961 as Pro Rector and Head of the Depart- ment of Chemical Engineering and Chemical Tech- nology of the Imperial College of Science and Tech- nology has been appointed Senior Research Fellow of the College. Professor P. M. S. Blackett is to become Pro Rector and Professor A. R. Ubbelohde is to be Head of the Department of Chemical Engineering and Chemical Technology. Mr. G. Nonhebel has been elected a Vice-president of the Institute of Fuel. Professor P. L. Pauson has been elected to the Fellowship of the Royal Society of Edinburgh.Dr. Carl F. Pruttun has been awarded the Perkin Medal of the American Section of the Society of Chemical Industry. The title of Senior Lecturer in Chemistry at the Southampton University has been conferred on Dr. A. C. Riddiford. The title of Reader in Spectroscopy has been con- ferred upon Dr. E. M. F. Roe; and Dr. F. J. C. Roe has been appointed Reader in Experimental Path- ology at the Institute of Cancer Research Royal Cancer Hospital. It is regretted that Dr. E. M. F. Roe’s appointment was incorrectly stated in Proceed-ings for April 1961 page 147. Dr. Pentti Salomaa Associate Professor of Phys-ical Chemistry of the University of Turku Finland has been appointed Professor of Physical Chemistry at the same University as from May lst 1961.Sir Alexander Tudd has received the honorary degree of D.Sc. at the University of Sheffield. Dr. D. T. A. Tuwnend has been elected a Fellow of the Imperial College of Science and Technology. Dr. S. Skidmore has been appointed Vice-President of the Harris College Preston. Dr. W. B. Whalley Reader in Organic Chemistry in Liverpool University has been appointed to the London University Chair of Chemistry tenable at the School of Pharmacy. Dr. T. D. Whittet Chief Pharmacist University College Hospital London has been awarded a World Health Organisation Fellowship to study hospital organisation in Europe. Mr. R. J. Wuodwardhas been appointed a Director of Shell Trinidad. I78 PROCEEDINGS FORTHCOMING SCIENTIFIC MEETINGS London Thursday June 8th 1961 at 6.30 p.m.Reception and Conversazione. To be held by kind permission of the Director in the Science Museum South Kensington S.W.7. (Tickets 10/6d. each- full details have been circulated.) Cambridge Monday June 5th 1961 at 5 p.m. Lecture “Some Rearrangements in the Pyrimidine and Related Series,” by Dr. D. J. Brown. To be held in the University Chemical Laboratory Lensfield Road. APPLICATIONS FOR FELLOWSHIP (Fellows wishing to lodge objections to the election of these candidates should communicate with the Honorary Secretaries within ten days of the publication of this issue of Proceedings. Such objections will be treated as confidential. The forms of application are available in the Rooms of the Society for inspection by Fellows.) Abrash Henry Ivan B.A.Ph.D. Chemistry Department The University of Wisconsin Madison Wisconsin U.S.A. Ackers John Philip. 17 Martin Rise Bexleyheath Kent. Anastassiou Apostolos George B.Sc. M.S. 679 Orange Street New Haven Connecticut U.S.A. Azim Shaukat M.Sc. Department of Chemistry The University Southamp ton. Baker Joseph Thomas M.Sc. University College of Townsville Townsville Queensland Australia. Bell John Alexander B.Sc. 19 Lidyard Road London N.19. Bhatnagar Vijay Mohani M.Sc. Department of Chem- istry The University of Western Australia Nedlands Australia. Bobtelsky Miriam M.Sc. Antebi Street 4 Jerusalem Israel. Buehler Charles Allen B.Sc. 1569 Spartan Village East Lansing Michigan U.S.A.Carey John Gerard P. B.A. 18 Claude Avenue Lin- thorpe Middlesbrough Yorks. Cartlidge John Ph.D. Department of Chemistry King’s College London W.C.2. Catterall Ronald. 41 Lancaster Gate Nelson Lancs. Ciula Richard Paul M.S. Ph.D. Department of Chem- istry Fresno State College Fesno 26 California U.S.A. Courtot Pierre. 37 rue d’Amsterdam Paris 8e France. Crandall Jack Kenneth B.S. Department of Chemistry Cornell University Ithaca New York U.S.A. Crook David John B.Sc. 16 Coolhurst Road London N.8. Davies Robin Havard B.A. New College Oxford. Dougherty Ralph Clifford B.S. Department of Chem- istry The University of Chicago Chicago 37 Illinois U.S.A. Emmett John Colin B.Sc. 31A Lansdowne Road London E.18.Galton Maurice James B.Sc. 7 Camer Street Meopham Kent. Grady Lee Timothy B.S. 2842-W. Fletcher Street, Chicago 18 Illinois U.S.A. Griffiths Trevor Redston Ph.D. A.R.T.C. N.E. Essex Technical College and School of Art Sheepen Road Colchester Essex. Hamlin Clive Richard. 435A Alexander Road P.O. Greendale Salisbury S. Rhodesia. Harris Thomas A.B. Department of Chemistry Brown University Providence 12 R.I. U.S.A. Harris William Merl B.S. 12333 Charnock Road Los Angeles 66 California U.S.A. Hawkes John Crosfield B.Sc. F.1nst.P. Hollytops, Vicarage Road Llangollen Denbighshire. Hewitson Richard E. B.S. College of Pharmacy, University of Illinois Chicago 12 Illinois U.S.A. Hughes Clifford Arthur M.Sc. 281 Burger Street Pietermaritzburg Natal S.Africa. Imhof Violet Ilene A.B. Department of Chemistry University of Illinois Urbana Illinois U.S.A. Johnson Brian John B.Sc. 37 Loose Road Maidstone Kent. Jones Maitland M.S. Sterling Chemistry Laboratory Yale University New Haven Connecticut U.S.A. Jones Michael Edward Benet B.Sc. 3 Aylesford Way Stapleford Cambridge. Kasturi Thirumalai R. Ph.D. A.R.I.C. Department of Organic Chemistry Indian Institute of Science Bangalore 12 India. Kornet Milton Joseph B.S. 47 N. Lockwood Chicago 44,Illinois U.S.A. Kucera Miloslav Ph.D. Tkalcovska 2 Brno Czecho- slovakia. Laufer Daniel Allon B.S. 1806 Beacon Street Brookline 46 Massachusetts U.S.A. McCann Vincent Bernard. 15 Simons Croft Ford, Liverpool 4. Martin Russell Niall Dickson.15 Lauder Road, Edinburgh 9. Molyneux Russell John B.Sc. Chemistry Department The University Nottingham. Moore Theron Langford B.S. 90478 Carson Street Culver City California U.S.A. Moser Armin P. Institute of Inorganic Chemistry Swiss Federal Institute of Technology Zurich Switzerland. Neiss Edward Samuel B.S. 809 South Henderson Street Bloomington Tndiana U.S.A. Ohline Robert Wayne M.S. Ph.D. 5466 Neosho Street St. Louis 9 Missouri U.S.A. Orloski Raymond F. B.S. Department of Chemistry University of California Los Angeles 24 U.S.A. Ottery Christine Mary B.Sc. 68 Pakefield Road King’s Norton. Birmingham 30. Pariaud Jean-Charles D.es.Sc. Faculte des Sciences I7 ter Rue Paul Collomp Clermont-Ferrand France. Pass Geoffrey Ph.D.12A Manor Road Cuddington Cheshire. Paul Norman Charles. 136 Huxley Road London N.18. Persky Avigdor M.Sc. 5 Baharan Street Jerusalem Israel. Randell Donald Richard Ph.D. A.R.T.C. 19 Newhaven Road Chaddesden Derby. MAY1961 Sandhu Gurmit Kaur M.Sc. 56 Windsor Road London N.7. Schnetzinger Richard W. B.S. M.A. 1523 Hollywood Avenue Bronx 61 New York U.S.A. Seelye Ralph Nicholson M.Sc. 9 South Parade Oxford. Shea Joseph Lawrence B.S. Kedzie Chemical Labora- tory Michigan State University East Lansing, Michigan U.S.A. Staley Stuart Warner B.A. 126 Meriden Drive Hockes- sin Delaware U.S.A. Staveley Alan Keith Randall. Lynton Stevington Lane Stevington Wigan Lancs. Stocks Kenneth B.Sc. St. Anselm Hall Victoria Park Manchester 14.Suherwardy Peerzada Barkatali M.Sc. Chemistry Department Government College Larkana West Pakistan. Taylor Stephen Colin. 17 Colchester Avenue Penylan Cardiff. Thomas Gordon H. S. B.Sc. Ph.D. Ontario Research Foundation 43 Queen’s Park Crescent East Toronto 5 Ontario Canada. Todd James S. B.A. Chemistry Department University of Rochester Rochester 20 New York U.S.A. Vandi Antonio Francesco Ph.D. 350D. Noyes Labora- tory Urbana Illinois U.S.A. Veigel Jon Michael B.S. Department of Chemistry, University College of Los Angeles Los Angeles 24, California U.S.A. Walters Philip Arthur B.Sc. 17 Norfolk Road Romford Essex. Williams Frederick Paul B.Sc. Ph.D. 181 Cross Flatts Grove Leeds 11 Yorks. Willis Martin Richard Ph.D.Chemistry Department The University Nottingham. Young James Francis M.Sc. South Makirikiri Marton. New Zealand. OBITUARY NOTICES JAMES YOUNGER MACDONALD 1902-1960 JAMES MACDONALD YOUNGER was born in Glasgow on October 29th 1902 and died at St. Andrews on April 9th 1960 after a lengthy illness borne with great fortitude. His father George (later Sir George) Macdonald was assistant to Professor Gilbert Murray in the Department of Greek in Glasgow University; later he entered the Scottish Education Department of which he became Secretary and James was therefore brought up in Edinburgh; the family consisted of himself and an elder sister. His father was a classical scholar and an authority on numismatics archaeology and Roman Britain.He wished his son to become a classicist and sent him as a day-boy to the Edinburgh Academy a school with a strong classical tradition. Here he remained from the age of five until eighteen; and it was only in the last year or two of this period that he was allowed to follow his decided bent towards chemistry His was a very literary home full of books and James imbibed a love of English literature; he developed a neat turn of phrase with a gift for writing comic verse and parodies which later delighted his family and students. Nobody who heard it will ever forget the coruscating and delightful whimsicality of his speech on May 25 th 193 1 at the dinner celebrating the twenty-fifth anniversary of the St. Andrews Uni- versity Chemical Society this was entitled “On Buffers coupled with the toast of the External Examiner” (a distinguished authority on buffers and buffering).In 1921 Macdonald won a Scholarship in chem-istry at Trinity College Oxford where he studied from 1921 to 1925 his tutor being C. N. (later Sir Cyril) Hinshelwood. He took second-class honours in chemistry presenting in 1925 a thesis dealing with the thermal decomposition of silver oxalate. He spent the next two years in postgraduate research work under F. G. Donnan at University College London and was awarded the Ph.D. degree of London Uni- versity for a thesis entitled “The Photochemical Decomposition of Nitrous Oxide and Nitric Oxide.” In 1927 he was appointed to a Lectureship in chemistry in the United College of St.Salvator and St. Leonard University of St. Andrews. Throughout his long association with the St. Andrews school of chemistry lasting for the rest of his career he rendered loyal and invaluable service particularly in the development of the teaching of physical chem- istry which at the time of his appointment occupied a very subordinate position in a Department with strong organic chemical traditions. He remained the only Iecturer in this branch of the subject until 1947. and the many successful chemists who had their initial training in the Department during these twenty years owed much to his devoted teaching. Although his heavy burden of teaching took up so. much of his time and energy Macdonald maintained a keen interest in research; he published occasional contributions in the Journal of the Chemical Society the Transactions of the Faraday Society the Journar of the American Chemical Society etc.dealing with the thermal decomposition of silver oxalate the hydrations of ions the constitution of ferric thio- cyanate the formation and growth of nuclei in the decomposition of solids and the molecular mechan- ism of rate processes in solids. Most of this work was individual but some was conducted in collaboration with postgraduate students. His last paper published in 1960 dealt with the influence of design of apparatus and nature of suspension on measurements of turbidity with special reference to serological work. Macdonald’s activities as teacher and research worker were blended with a sympathetic and discern- ing interest in students whether in the Chemistry Department or outside.In his hospitable home in South Street students were always sure of a warm welcome from himself and his gracious wife. He was Adviser of Studies in recent years in addition to being a Regent and he rendered services of outstand- ing value as a member of the Faculty of Science especially in connection with the new Science Ordinance. He took part also in University Exten- sion work and often talked to the naval cadets at Arbroath. During the war Macdonald was Group Warden for St. Andrews and district with charge of all A.R.P. He organised group report posts trained wardens and took command at the main report post during air-raid alarms.He lectured to the troops and took American and foreign leave parties around Fife once every week. His games were tennis and badminton. Tennis is handicapped in the Scottish Universities by the early end of the summer term; but Macdonald fostered the game at St. Andrews and he and his wife gave cups for the men’s and women’s singles. Macdonald had a passionate love for the Scottish hills and he was an ardent and skilled mountaineer. He went to the hills at every opportunity and never missed the New Year Meet of the Scottish Mountain- eering Club. He also made many ascents in the Alps. He took an active interest in the students’ Mountain- eering Club from its foundation in 1935 onwards; when the Club was re-formed in 1946 he was elected honorary president and became the guide comforter and friend of all who aspired to climb.He turned out PROCEEDINGS at night to drive to Glencoe when students got lost or failed to return; and on one occasion in March 1949 he spent a long night in heavy rain leading a rescue party up Ben Nevis. He and his eldest son climbed all of the 277 “Munros” (mountains in Scotland over 3000 ft.) and finished them together on Ben More Mull in August 1958 with all the family present to celebrate. Here it may be said that in 1929 he married Nan daughter of Dr. D. Roxburgh of London and they had three sons and one daughter. With mountaineering he combined a great interest in photography and some of his most beautiful pictures were snowscapes.In 1953 he went as photo- grapher on Mr. T. B. Mitford’s expedition to Cyprus and here his early training in archaeology stood him in good stead. Before this in 1937 he organised and led a University scientific expedition to the north- west peninsula of Iceland in order to undertake a geological botanical and zoological survey of this little-known region. His interest in photography and drama led naturally to an interest in cinematography and he was a member of the Film Society from its foundation. He belonged to a play-reading Society in St. Andrews of which he was twice president. To the townsfolk of St. Andrews he was well-known for his work on behalf of the St. Andrews Preservation Trust particularly in photography.He was a member of the Chemical Society the Faraday Society and the American Chemical Society. At meetings of the Faraday Society he was a regular attendant; but in 1959 he and his wife were prevented from taking part in the Canadian meeting by the serious illness which led to his death. “J.Y.” as he was familiarly known among his wide circle of friends has been truly described by a colleague as the salt of the earth than whom the University had no more devoted or loyal servant. Kindly generous and understanding he was indeed a loyal colleague and friend whose memory will be cherished by all who had the privilege of knowing him and working with him. JOHN READ. ROBERT CROSBIE FARMER 1877-1 960 THEdeath of Robert Crosbie Farmer O.B.E.DSc. Ph.D. on July 30th was a grievous loss to his wide circle of friends and relations and no less to the scientific world. He attained the good age of 82 and to the end happily enjoyed health vigour and lively interests. Tall slim and alert he was a striking figure. Widely read witty modest and intensely scrupulous and loyal he was dear to all and a man without an enemy. Deeply religious he met adversity of which he had his share with faith and fortitude. Of his hobbies walking was almost a passion. Until he was well over fifty years of age he liked to walk on at least one occasion each year x miles in one day where x is equal to his age in years; in later years he changed x to lOO-x although he was somewhat disappointed when his wife was no longer able to keep up his high standard.He made many marathon treks recording events and reactions in great detail MAY1961 which provided marvellous reading for the favoured few. Not long ago he complained “Iam not the man I was-I do only four miles a day now.” After a distinguished academic career at Liverpool Wiirzburg (with Hantzsch) and Birmingham he joined the War Office Experimental Establishment at Woolwich (later the Research Department) where he remained for almost the whole of his working life. He became an authority of world-wide repute on the chemistry and manufacture of explosives and he made many notable contributions to the subject. During the first World War he was Chief Chemical Adviser to Lord Moulton and was awarded the O.B.E.and the Order of St. Anne. He served as a Member of Council of the Royal Institute of Chem- istry and was elected to the Fellowship of the Chemical Society in 1903. After his retirement in 1945 Dr. Farmer spent more than ten years at King’s College London carrying out fundamental research on problems con- tinued from his days at Wurzburg under Hantzsch. The results of these investigations were published shortly before his death (J. 1959 3425 3430 3433). During this time he endeared himself to a new range of friends young and old. He really became a student again and bought a College tie which he was proud to wear on every occasion. In spite of the disparity in their ages the students soon came to regard him with admiration and affection and he set them a wonderful example in many of those great virtues which are now sometimes referred to as old-fashioned.It was characteristic of him that he started each morning by noting his time of arrival (accurate to the nearest minute) the room temperature and the barometric pressure. He then polished his already spotless bench and cleaned his shoes. He often made use of simple home-made pieces of apparatus because he did not wish to involve the College in any expense on his behalf. When he re-started experi- mental work he realised that his knowledge of more recent developments in organic chemistry was inade- quate and he at once set out with determination to acquire this new knowledge making voluminous neat notes with a steel pen.Robert Crosbie Farmer was truly one of the last of the old school. He was admired and respected by his acquaintances loved by his friends and his long life was one of dis- tinguished and devoted service. A. FORSTER D. H. HEY. LEONARD JAMES GOLDSWORTHY 1890-1960 GOLDSWORTHY, LEONARD son of William Alfred Goldsworthy was born in York on December 12th 1890 and died in Oxford on August 26th 1960. He attended Ripon Grammar School from 1902 becoming head-boy in 1910 and gaining a de Grey Exhibition. He played in the school rugby team cap- tained hockey and shone also in athletics tennis and golf. Winning in 1910 an open Exhibition in Natural Science at Magdalen College Oxford he became a pupil of T.S. Moore who was much impressed by his keenness persistence and great self-control. In 191 3 he took his B.A. Degree (with Second Class Honours in a year in which no First was awarded) and became a research pupil of the Waynflete Professor (W.H. Perkin jun.). It was the hey-day of the college laboratories and Goldsworthy became a part-time Demonstrator at the Christ Church and the Queen’s College laboratories. This made for a strenuous life but he obviously enjoyed it and his stories of Perkin Sidgwick Moore. Harold Hartley A. F. Walden B. Lambert Chattaway Applebey and D. L. Chapman confirmed the stimulating influence of these out- standing men on the young scientist. At Magdalen Sir Herbert Warren and the Prince of Wales must have enlivened college life.In 1914 at the age of 24 he was appointed to the Indian Educational Service as Professor of Chem- istry Victoria College of Science Nagpur (affiliated to the University of Allahabad). He would have pre- ferred to delay taking up the post in order to serve in the war. Accordingly during 1914-1917 he was active in the Volunteers Indian Defence Force be-coming Commandant in Nagpur. He was commis- sioned in 1917 and saw active service with the Deccan Horse 16th Cavalry from 1917 until 1920 in Egypt and Palestine. Returning to India he became Inspector of Schools and Science Central Provinces then Assistant Director of Public Instruction and he held other administrative posts until 1922. Then longing to return to laboratory work he applied for a transfer to another province.He accepted a Senior Lecture- ship at University College Rangoon and the Wardenship of a residential hall in that College. He was twice Acting-Professor of Chemistry. Professor D. H. Peacock (to whom the writer of this notice is much indebted) pays tribute to his devoted help in all aspects of the work of the University. Despite the heavy burden of teaching he managed to carry on his own research work play tennis and polo and serve as a Company Commander in the 6th (Burma) Battalion University Training Corps becoming in 1935 Commanding Officer of this Battalion. On the introduction of the Montague-Chelmsford Reforms Members of Indian Government services were permitted to retire prematurely on pensions proportionate to their service.In 1935 Goldsworthy took advantage of this and with his wife and two young sons returned to Oxford and began the fruit- ful association with the Dyson Perrins Laboratory. He was obviously happy to be back in Oxford and at bench-work. Sir Robert Robinson (the Waynflete Professor in succession to Perkin) was able to sug- gest an array of exciting topics and perhaps even more welcome provided a succession of overseas students requiring varying degrees of help. (Sir Robert has already written of Goldsworthy’s success with these young students.l) The writer remembers that in 1935-36 Goldsworthy and 19 post-graduates were at work in one large laboratory.The swing- doors were rather noisy and on each entry of a visitor 19 heads turned towards the doors but not the head of L.J.G. His upright well-groomed figure remained facing the bench but any necessary inter- ruption was met with courtesy patience and a twinkle in the eye. Off duty he was good company full of fun and rich in anecdote. It was fascinating to hear his account of life in those far-off pre-1914 days. In his first research with W. H. Perkin he succeeded in resolving trans-cyclopentane-l,2-di-carboxylic acid as the brucine salt.2 This was the third of the cyclic dicarboxylic acids to be re-solved. Next he made use of ethyl a!-dibromo- propionate to prepare a series of cycloalkane-polycarboxylic acids from disodium salts of the type (EtO,C),CNa.[CH,],CNa(C0,Et)2.3 While in Ran- goon he succeeded in resolving trans-cyclobutane- dicarboxylic acid using the quinine salt and he measured a series of optical rotations.* Then he noticed a flaw in Haywood’s5 much quoted work on the reactivities of alkyl iodides with sodium benzyl- oxide and showed conclusively6 that under the conditions used by Haywood the rates measured were those of the reaction with sodium ethoxide. This was followed by a study of the reactivity of Nature 1960 188,536. Goldsworthy and Perkin J. 1914 105 2639. Goldsworthy and Perkin J. 1914 105 2665. Goldsworthy J. 1924 125 2012. Haywood J. 1922 121,1904. Goldsworthy J. 1926 1102. Goldsworthy J. 1926 1254. Goldsworthy J. 1931 482. Goldsworthy.J.. 1934. 377. lo Goldsworthy i,1936 1148. PROCEED~NGS alkyl iodides with the sodium salts of u- rn- and p-chloro- and -methyl-phen~ls.~ He concluded that the order of reactivities was usually the inverse of the order of the ionisation constants of the correspond- ing phenols the exception being the u-methyl analogue. In 1931 he returned to cyclic compounds with a study of the use of tetramethylene dibromide along with diethyl sodiomalonate or tetraethyl di- sodioethanetetracarboxylate. From tetramethylene dibromide and tetraethyl disodiopropanetetracar- boxylate the (unexpected) products were diethyl cyclopentane- 1,l-dicarboxylate (showing the prefer- ence for the formation of 5-membered rings) and a polymer of diethyl methylenemalonate.* Ethyl sodio- acetoacetate and tetramethylene dibromide again reacted to form a cyclopentane derivative:9 2 NaCHAcC0,Et +-Br[CH,],Br -+ CH,-CH CO-Me + CH,AcCO,Et I /-\ C!.H,-CH C0,Et + 2 NaBr On his return to Oxford in 1935 Goldsworthy began his collaboration with Professor Robert Robinson in a study on the series “The Relative Directive Powers of the Groups RO and RR’N in Aromatic Substitution.” Part X.For nitration he found the relative powers Bur0 PrBO Me0 = 328 229 100 in an elegant piece of work which rounded off the series.1° Then followed some papers on the synthesis of flavones tangeritinll and herbacetine12 (5,6,7,8 :4’-pentamethoxyflavone and 3,5,7,8 :4’-pentahydroxyflavone respectively). During a period in the nearby Sir William Dunn School of Pathology Goldsworthy collaborated with Dr.(now Professor) E. Chain on a study of the anti- fermenting principle of the venom of the black tiger- snake.13 Returning to the Dyson Perrins Laboratory he joined Professor Robinson’s team working on the synthesis of polypeptides (1938-39). He studied the application of the Leuchs reaction to D- L- and DL-alanine.14 When war came he threw himself with great industry into work on the war-gases and later he joined the penicillin team. He served in the Home l1 Goldsworthy and Robinson J, 1937 46; Chem. and Id. 1957,47. l2 Goldsworthy and Robinson J. 1938 56. l3 Chain and Goldsworthy Quart. J. Exp. Physiol. 1938 27,375. l4 Goldsworthy Robinson Abraham Baker and Chain C.P.S.Report of the Medical Research Council (C.P.S. No. 650) 1946. MAY 1961 Guard Summertown and Wolvercote Platoon of the Oxford Company. After the war he took part in a varied series of researches including studies on quinamine15 and brasilin; some of these have not yet been published. In 1955 he turned his attention to gas-liquid chromatography. Collaborating with A. E. Thomp-son he greatly speeded up the development in the Dyson Perrins Laboratory of this analytical method. During the post-war period Goldsworthy had become more than ever the philosopher and guide to overseas students. It was a surprise to most people that he had reached retiring age (sixty-seven) in 1958 his activity and his quick wit belied his age.Charac- teristically he stayed on to help in a period of shortage of staff and he was lecturing until June of last year. Realising that his heart was failing he had arranged a late holiday away from the mountains this time but the end came suddenly on August 26th when he retired for an afternoon’s rest. Hospitality at the Goldsworthy home in Apsley Road was generously “Yorkshire” and has left pleasant memories alike to friends colleagues and students. His loss will be deeply felt in many parts of the world and all who knew Leonard Golds- worthy will wish to express sympathy with his family. He married in Ripon Cathedral in 1927 Rose daughter of Joseph Elston Cawthorn. There are two sons. J. C. SMITH. l6 Bendz,Culvenor Goldsworthy Kirby and Robinson J.1950 1130; Nature 1950 166 105 CHARLES SAMUEL GARLAND 1887-1960 THEdeath occurred on December 6th 1960 of Charles Samuel Garland at his residence in Bromley Kent. He had been a Fellow of the Chemical Society for 50 years. Mr. Garland was a man of wide interests and although he pursued with great purpose and efficiency work in the fields of politics scientific education and commerce he will be remembered chiefly for his important achievements as a chemical engineer. Born in London on June 23rd 1887 he was educated at Camberwell Grammar School and the Royal College of Science Kensington where as a student and then as a demonstEator he early showed his planning and organising ability. He was active in the negotiations for the erection of the Imperial College Union building and was an early President of the R.C.S.Old Students’ Association. Mr. Garland had a flair for the practical applica- tion of scientific discovery; his first excursion into that field was in managing the Volker Mantle Factory in 1912. His later interests to name only a few out of many were in the establishment of a com- pany in Worcester to produce high-voltage insulators in the manufacture of equipment for the radio industry and notably in the development of Hele- Shaw’s principles of stream-line filtration. He served for many years as Hon. Treasurer and from 1956-58 as President of the National Union of Manufacturers. In the General Election of November 1922 Mr. Garland was returned as Unionist M.P.for Islington South. When that short-lived parliament resigned only 12 months later he did not seek re-election but nevertheless continued his interest in parliamentary affairs throughout his life and sat on various com- mittees dealing with scientific development and finance. Mr. Garland was an active founder of the Chemical Engineering Group of the Society of Chemical Industry and later of the Institution of Chemical Engineers. He was a Crown Governor and Hon. Fellow of the Imperial College of Science and Technology and of the City and Guilds Engineering College. He is survived by his wife and daughter. J. A. CRANSTON. ADDITIONS TO THE LIBRARY Biographisch-LiterarischesHandworterbuch der exak- ten Naturwissenschaften.J. C. Poggendorf. Vol. 7a. Part 4; S-Z. Section 5. .Pp.560. Akademie Verlag. Berlin. 1960. Commercial Alphabetical Index. (Loose leaf pages.) Sadtler. Philadelphia. 1960. The radiochemistry of manganese. G. W. Leddicotte. Sponsored by the United States Atomic Energy Com- mission. (Nuclear Science Series NAS-NS 3018.) Pp. 23. Subcommittee on Radiochemistry. Washington. 1960. (Presented by the National Academy of Sciences.) The interpretation of X-ray diffraction photographs. N. F. M. Henry H. Lipson and W. A. Wooster. 2nd edn. Pp. 282. Macmillan. London. 1960. Standard X-ray diffraction powder patterns. H. E. Swanson et al. National Bureau of Standards Circular 539. Vols. 4-10. U.S. Government Printing Office.Washington. 1955-60. Organic electronic spectral data. Vol. 1. 1946-1952. Edited by H. E. Ungnade. Vol. 2. 1953-1955. Edited by M.J. Kamlet. Interscience Publishers Inc. New York. 1960. Fatty acids their chemistry properties production and uses. Edited by K. S. Markley. Part 1. Pp. 714. 2nd edn. Interscience Publishers Inc. New York. 1961. Chemistry of the amino acids. J. P. Greenstein and M. Winitz. 3 vols. Pp. 2872. John Wiley & Sons. New York. 1961. Die atherischen ole. E. Gildemeister and F. Hoffmann. Vol. 6. 4th edn. Pp. 644.. Akademie Verlag. Berlin. 1961. Die terpene Sammlung der Spektren und physikal- ischen Konstanten. J. Pliva M. Horak V. Herout and F. Sorm.Sponsored by Deutsche Akademie der Wissen- schaften zu Berlin.Teil 1. Sesquiterpene. (Loose leaves.) Akademie Verlag. Berlin. 1960. (Presented by the sponsors.) A review of syntheses of organic compounds containing the ferroin group. F. H. Case. Pp. 36. G. Frederick Smith Chemical Co. Columbus Ohio. 1960. (Presented by the publisher.) Organogermanium compounds a survey of the literature from January 1950 to July 1960. F. Rijkens. Pp. 97. Germanium Research Committee. Utrecht. 1960. (Presented.) Advances in analytical chemistry and instrumentation. Edited by C. N. Reilley. Vol. 1. Pp. 445. Interscience Publishers Inc. New York. 1960. Die coulometrische Analyse. K. Abresch and I. Claassen. (Monographien zu Angewandte Chemie und Chemie-Ingenieur-Technik No. 71.) Pp. 228. Verlag Chemie GmbH.Weinheim. 1961. (Presented by the publisher.) Analytical chemistry of the rare earths. R. C. Vickery. Pp. 189. Pergamon Press. Oxford. 1961. Approved methods for the physical and chemical examination of water. Recommendations of a joint committee of representatives of The Institution of Water Engineers the Royal Institute of Chemistry the Society for Analytical Chemistry and the Society for Water Treatment and Examination. 3rd edn. Pp. 74. Instn. Water Engineers. London. 1960. A rapid method of analysing cements and rock pro- ducts. C. E. s. Davis. (Chemical Research Laboratories Technical Paper No. 3.) Pp. 30. Commonwealth Scientific and Industrial Research Organisation. Melbourne. 1961. (Presented by the publishers.) Advanced paint chemistry.P. M. Fisk. Pp. 164. Leonard Hill. London. 1961. The biochemistry of insects. D. Gilmour. Pp. 343. Academic Press. New York. 1961. Specifications and criteria for biochemical compounds prepared by the Committee on Biological Chemistry Division of Chemistry and Chemical Technology with support by The National Institutes of Health. (Loose leaf pages.) National Research Council National Acad. Sci. Washington. 1960. (Presented by the publisher.) The chemistry and technology of edible oils and fats proceedings of a conference arranged by Unilever Limited Port Sunlight 1959. Edited by J. Devine and P. N. Williams. Pp. 154. Pergamon Press. London. 1961. Symposium on spectroscopy presented at the third Pacific Area National Meeting of the American Society for Testing Materials San Francisco October 1959.(A.S.T.M. Special Technical Publication No. 269.) Pp. 245. A.S.T.M. Philadelphia. 1960. Pure food and pure food legislation papers of the 1960 Centenary Conference. Edited by A. J. Amos. Sponsored by the Society for Analytical Chemistry Food Group of the Society of Chemical Industry Association of Public Analysts Royal Institute of Chemistry Ministry of Agriculture Fisheries and Food Ministry of Health Department of Health for Scotland and Ministry of Health and Local Government in Northern Ireland. Held at the Royal Institution 1960. Butterworths. London. 1960. Boron synthesis structure and properties proceed- ings of the conference on boron. Sponsored by the Institute for Exploratory Research The U.S.Army Signal Research and Development Laboratory For Monmouth New Jersey. Edited by J. A. Kohn W. F. Nye and G. K. Gaule. Pp. 189. Plenum Press. New York. 1960.
ISSN:0369-8718
DOI:10.1039/PS9610000153
出版商:RSC
年代:1961
数据来源: RSC
|
|