ORGANIC CHEMISTRY.1. INTRODUCTION.THE following Report on Organic Chemistry commences with a brief reviewof work published during the past two years concerning the theoreticalmethods and chemical applications of quantum organic chemistry. There-after, the theoretical section is concerned mainly with (a) the interplay ofsteric and electronic factors, including d-orbital resonance in organic com-pounds containing sulphur and phosphorus, and its effect on the reactionsof organic compounds, (b) transfer of hydride ion, (c) intermediates contain-ing bivalent carbon, (a) intramolecular rearrangements, and (e) electrophilicand nucleophilic substitution processes in the aliphatic field.Physical methods are finding ever wider application to structural prob-lems. The structure of de(oxymethy1ene)lycoctonine has been determinedby X-ray analysis and, in the steroid field, nuclear magnetic resonance hasbeen used for the first time in the structural assignment of helvolic acid.Among the more striking features of recently published work are thedetermination of the structures of methymycin, erythromycin, and magna-mycin, the synthesis of reserpine, and Isler’s industrial synthesis of 8-carotene.Complexes of aromatic hydrocarbon with metal atom or ion,especially dibenzenechromium(0) which can be sublimed and decomposesonly slowly at 300°, will intrigue all chemists. They are included in thesection on Aromatic Compounds though they are discussed in the Report onInorganic Chemistry.In the search for therapeutic agents having increased or selective activity,a great deal of interest has centred on the synthesis of modified hormones.In the terpene field, sesquiterpene lactones and complex diterpenes haveattracted much attention.The Report concludes with a brief review of the chemistry of lignins,which was last reported in 1942.Topics which it has not been possible tocover this year, such as the chemistry of proteins and stereochemistry, willbe reported in Annual Reports for 1957.G. B. rr. G. H.2. QUANTUM ORGANIC CHEMISTRY.THE function of quantum chemistry may be divided into three parts : thequantitative interpretation of physical properties of molecules, the inter-pretation of relative reactivities of related molecules, and the prediction ofabsolute rates of reaction.The third of these targets is not yet in sight, butconsiderable progress has been made towards the other two.Nearly all the work in progress in this field has been based 011 themolecular-orbital (M.O.) approximation and the assumption that 0- andx-electrons can be treated separately. Since the properties of o-bonds areadditive, most of the work has been directed to the study of x-electroDEWAK : QUANTUM ORGANIC CHEMISTRY. 127systems ; and usually some version of the L.C.A.O. (linear combiriatioii ofatomic orbitals) M.O. method has been used.The original Hiickel method involves a number of additional approxim-ations of which the use of product wave functions, the neglect of electronrepulsion, and the neglect of electron correlation have the most seriousconsequences.Moreover, the method is by its nature semi-empirical sincethe integrals involved in it cannot readily be calculated from first principles.The approximations in the Hiickel method can be overcome by usingdeterminantal wave functions (antisymmetrised M.O., A.S.M.O., method) ,by allowing explicitly for electron repulsion using the full x-electron Hamil-tonian and carrying out a self-consistent field (S.C.F.) treatment,2 and byincluding configuration interaction (C.I.). In these refined M.O. treatmentsall the integrals are clearly defined and are evaluated from first principlesby means of explicit A.O. functions (usually Slater functions).The Roothaan S.C.F. treatment is very laborious, and the further stepof including configuration interaction has proved possible only 'for verysimple molecules. Pariser and Parr pointed out that a complete C.I.treatment must give the same result no matter what set of orbitals is takenas a basis ; they therefore used simple Hiickel molecular orbitals to constructdeterrninantal wave functions, and then included configuration interaction.It is not usually possible to carry out a complete C.I.treatment; however,Pariser and Pam found very good results in many cases by including only afew of the least excited configurations. In this form the treatment iscertainly simpler than the S.C.F. method and seems on the whole to givebetter results.A further difficulty arises from the use of the orbital approximation itself,which is known to be inaccurate even for atoms.Two methods have beendevised to minimise these errors.divides the expression for the total energy of a molecule intotwo parts; the first is the energy of the component atoms widely separatedand in their appropriate valency states; the second is the energy liberatedwhen the atoms are allowed to come together to form the molecule. Thesecond is the part of major interest to chemists and is usually only a smallfraction of the total energy of the molecule; Moffitt calculates it directlyby a perturbation method. In this way the energy difference is founddirectly and so the results should be relatively insensitive to errors in thewave functions used. Moffitt describes this method as the " atoms inmolecules '' method; here the abbreviation A.I.M.will be used.The second method, due to Pariser and Parr,l is frankly semi-empirical;they adjust the values of various electron-repulsion integrals to give agree-ment with experiment. The success of their treatment is largely due to thisrefinement.Complete C.I. calculations have been reported for hydrogen cyanide 4For a summary and references see R. Pariser and K. G. Parr, J . Chew. Phi*s.,Moffitt1953, 21, 466.2 C . C . J. Roothaan, Rev. Mod. Phys., 1951, 23, 69.U'. Moffitt, PYOC. Ii03). SOC., 1951, A , 210, 216; 1953, A , 218, 4d6, i V . 3Ioffitt anctI<. Tguchi, J. Chem. Phys., 1955, 23, 1983; 1956, 25, 217.J . Scanlan, ibid., p. 464; 1953, A , 220, 530128 ORGANIC CHEMISTRY.and b~tadiene.~ The dipole moment calculated for hydrogen cyanide(2-66 D) agreed well with experiment (2-77 D).Comparisons of the S.C.F.and the C.I. method have been reported for simple diatomic systems andfor naphthalene. The inclusion of configuration interaction improvescalculations of dipole moments.8 The A.I.M. method has been found betterthan a complete S.C.F.C.I. treatment for a~etylene,~ and the inclusion ofPariser and Pads corrections to integrals improves the S.C.F. treatment ofbutadiene.lo Straightforward S.C.F. calculations have been reported forferrocene,ll various alternant hydrocarbon radicals and ions l2 (allyl,benzyl, etc.), hydrogen fluoride,13 and ammonia and the NH, and the NHradi~al.1~ Good agreement with experiment was obtained for the dipolemoment of hydrogen fluoride, the ionisation potential of hydrocarbon radi-cals, and the bond angles in the N-H compounds.The S.C.F. method hasalso been used successfully to interpret the indirect electron-coupling ofnuclear spin l5 in nuclear magnetic resonance spectroscopy, and to show thatthe negative ion-radicals obtained by single-electron addition to aromatichydrocarbons are stable to disproportionation.16 Fulvene has been syn-thesised and its dipole moment shown to agree with the value calculated bythe S.C.F. method.17 The C.I. method has been applied to acetylene,polyacetylenes, vinylacetylenes, and cumulenes,l* aromatic hydrocarbons, '9 l9quinone,20 fulvene, heptafulvene, and tropylium.21 The general basis oforbital theories, configuration interaction, etc., has been discussed byvarious authors.22 A very interesting development is the calculation ofmolecular properties directly via the densityC.M. Moser, J., 1954, 3455.R. D. Brown and A. Penfold, J . Chem. Phys., 1956, 24, 1259; cf. R. G. Parr andR. Pariser, ibid.,1955, 23, 711; Y. Haya, J . Chern. SOC. Japan, 1956, 77, 314.C . M. Moser, J . Chem. Phys., 1955, 23, 598; J . China. phys., 1955, 52, 24; C. &I.Moser and R. Lefebvre, J., 1956, 1557, 2734.M. Wolfsberg, J . Chem. Phys., 1955, 23, 793.9 J. Serre, Compt. rend., 1956, 242, 1469.lo A. Pullman and H. Berthod, ibid., 1954, 239, 812; J . Chim. phys., 1955, 52, 771.l1 Y . Yamazaki, J . Chern. Phys., 1956, 24, 1260; cf. J. D. Dunitz and L.E. Orgel,ibid., 1955, 23, 954; F. 0. Ellison and H. Shull, J . Chem. Phys., 1955, 23, 2348.l2 A. Brickstock and J. A. Pople, Tmns. Favaday SOC., 1054, 50, 901 ; N. S. Hushand J. A. Pople, ibid., 1955, 51, 600; G. Berthier, J . Chim. phys., 1955, 52, 141; H. C.Lefkovits, J. Fain, and F. A. Blatsen, J . Chem. Phys., 1955, 23, 1690; Y . Mori, ibid.,1956, 27, 1253; C. Komatsu, Y . Mori, and I. Tanaka, J . Chem. SOC. Japan, 1956,7'7, 643. f13 H. Hamano, ibid., p. 985.l4 J. Higuchi, J . Chem. Phys., 1056, 24, 535.l6 H. M. McConnell, ibid., 1955, 23, 760, 2454; 1956, 24, 460.l6 N. S. Hush and J. Blackledge, ibid., 1955, 23, 514; N. S . Hush and J . R. Row-la J. Serre, J . China. phys., 1955, 52, 331; 1956, 53, 284.lQ R. Pariser, J . Chem. Phys., 1956, 24, 250.20 M.Okuda, ibid., 1956, 25, 1083.21 A. Julg and B. Pullman, J . Chim. phys., 1955, 52, 481.22 W. E. Moffitt, J . Chem. Phys., 1954, 22, 1820; R. K. Nesbet, Proc. Roy. SOL,1955, A , 230, 312; R. Lefebvre, Compt. rend., 1965, 240, 1094; A. Laforgue, Cahiersphys., 1955, 57/58. 23; R. G. Parr, F. 0. Ellison, and P. G. Lykos, J . Chew. Phys.,1956, 24, 1106; P. G. Lykos and R. G. Pam, ibid., p. 1166.23 R. McWeeny, Proc. 2 2 0 ~ . SOL, 1955, A , 232, 114; " Electronic Structures ofBlolecules," Techn. Rep. No. 7, Massachusetts Tnst. Technology, 1955.lands, ibid., 1956, 25, 1076.J. Thiec and J. Wiemann, Bztll. SOC. chim. France, 1956, 177DEWAR : QUANTUM ORGANIC CHEMISTRY. 129These methods seem to give good results for the properties of moleculesin their ground states and a tolerable correspondence with experiment in thecase of light absorption.The A.I.M. method seems the most promising butit has not yet been used for any but the simplest molecules. The C.I. methodis the easiest for use with desk machines but rather hard to programme forcomputers. It must be admitted that calculations, by any of these methods,using desk machines are exceedingly laborious, and that there is little imme-diate prospect of using them generally for the solution of chemical problems.They may be useful in specific cases for solving structural problems in simplemolecules; for example, a comparison of the observed light absorption ofmelamine with that predicted by a C.I. treatment suggests very strongly 24that melamine is s-triaminotriazine, rather than the isomeric s-tri-imino-hexahydrotriazine, both structures being consistent with the availableexperimental evidence.In view of the difficulty of A.S.M.O.calcuIations much work has beendone by the simple Hiickel method. Calculations of this kind can be carriedout very simply z5* 26p 27 with digital computers ; so much so that it is prob-ably a waste of time to attempt such calculations for large molecules unlessone has access to a digital computer. There has been much discussion in thepast concerning the inclusion of overlap in Huckel calculations; to includeoverlap is probably pointless, since it hardly ever makes any significantdifferences in practice and its neglect is much less serious than the otherapproximations inherent in the Hiickel method.Overlap 28 can be includedif desired by a simple perturbation method,2g and simple expressions havealso been given for calculating resonance energies 3O and polarisabilities 31by contour integration.The Hiickel method has been used in two distinct ways ; either to obtaindetailed information (bond orders and lengths, charge distributions anddipole moments, self-polarisabilities and mutual polarisabilities) for speciticmolecules, or to obtain general information concerning the way some propertychanges with structure in a group of related molecules. Examples of thefirst kind are provided by calculations for various a~o-cornpounds,~~ all thepossible unsubstituted monocyclic azoles and a ~ i n e s , ~ ~ various polycyclicaromatic hydrocarbon^,^^^ 26s 34~ 35 polynuclear heterocyclic cornpounds,27~ 2980 M.J . S. Dewar and L. Paoloni, Trans. Faraday SOC., in the press.26 H. 0. Pritchard and F. H. Sumner, Proc. Roy. SOL, 1954, A , 226, 128; F. H.20 H. 0. Pritchard and F. H. Sumner, ibid., p. 457.27 H. 0. Pritchard, Proc. Roy. Sot., 1956, A , S, 136; cf. E. Bak, D. Christensen,2* Cf. I. M. Bassett and R. D. Brown, Austral. J . Chew., 1956, 9, 305, 315.D. W. Davies, Trans. Faraday Sot., 1955, 51, 449.80 C. A. Coulson, J., 1954, 3111.81 R. D. Brown, J., 1956, 767.32 B. Pullman and J. Baudel, Ccmpt. rend., 1954, 238, 2529; 0. Chalvet, itrid., 1954,38 S. Basu, Proc. Nut. Inst. Sci. India, 1955, 21, A , 173; R. D. Brown and M.L.34 B. Pullman, Cahiers Phys., 1954, 48, 42; M. A . H. Zarza, Anales Fis Qziim.,36 B. Pullman and G. Eerthier, Compt. rend., 1956, 242, 2563.Sumner, Trans. Faraday SOC., 1955, 51, 315.J. Rastrup-Andersen, and E. Tannenbaum, J . Chem. Phys., 1956, 25, 892.259, 1135.Heffernan, Austral. J . Chem., 1956, 9, 83.1955, 51, B, 305; 0. Chalvet and J. Peltier, J . Chinz. phys., 1056, 53, 402.REP.-VOL. LIII r130 ORGANIC CHELIISTKY.c o u ~ ~ ~ a r i i i , ~ ~ various porpliiiis,3i vinyl ~hloride,3~ benzoic acid,39 the nitro-a n i l i n e ~ , ~ ~ the nitro naphthalene^,^^ the benzotropones,42 and paraffins.43The properties so calculated do not agree well with experiment. Calculateddipole moments are commonly several times too large, and similar errorsappear in calculated spectra. The agreement between calculated andobserved bond lengths is tolerable but not good; 26 it must be rememberedthat an error of 0.02 A in a C-C bond length is relatively large, for the wholerange of variation from pure single to pure double bonds is only 0.2 A.Onemust conclude that calculations by the Hiickel method are of very littlequantitative value.On the other hand, the relative values for a series of related compoundsdo follow experiment, at any rate in a qualitative way; and the Huckelmethod can be very useful in such cases. One of the most striking applic-ations of this kind was described by Heilbronner andSchmid 44 who showed by a comparison of a variety ofphysical and chemical properties with those calculatedby the Hiickel method for possible isomers that thenatural terpene derivative, lactaroviolin, almost cer-tainly has the structure (I).This is the first time thatthe structure of a natural product has been indicated by the use of quantumtheory; previously a different structure had been ascribed on the basis ofthe limited chemical evidence available.A similar comparison indicates 45 that the thermochromism of dianthronyland dixanthylidene is due to thermal excitation to perpendicular tripletstates ; and similar studies have been reported of hyperconjugation 46 andof substituent effects in para-substituted aniline~.~' It has been pre-dicted 48 that cyclobutadiene, the cyclopentadiene radical, and the cationC,H, do not have a symmetrical ring.In most of the simple M.O.calculations listed above various theoreticalquantities 49 (charge distribution, free valency, localisation energy, self-polarisability) have been calculated for comparison with observed chemicalreactivity. This procedure is not strictly justifiable, since the rate of areaction is determined, not by any property of the reactants alone, but bythe difference in energy between the reactants and the transition state;nevertheless, good correlations are found between the various theoreticalquantities and rates of reaction for reasons which will be considered presently.36 I . Samuel, Compt. rend., 1955, 240, 2534.37 S. Basu, Proc. Nut. Inst. Sci. India, 1955, 21, A , 269; S. L. Matlow, J . Chew.38 J. H. Goldstein, J .Chem. Phys., 1956, 24, 507.39 T. H . Goodwin, J., 1955, 4451.40 J. I. F. Alonso and R. Domingo, Andes Fis. Quim.. 1955, 51, B, 321.4 1 G. Favini and S . Carr2, Gazzrtta, 1955, 85, 1029.4 2 T. Gaumann, R. W. Schmid, and E . Heilbronner, HeZv. Chim. A d a , 1956,39, 1985.43 G. Sandorfy, Canad. J . Chem., 1955,33, 1337.44 E. Heilbronner and R. W. Schmid, HeZv. Chim. Acta, 1954, 37, 2018.4 5 S. L. Matlow, J . Ckem. Phys., 1955, 23, 152.46 N. Muller, L. C. Pickett, and R. S . Mulliken, J . Amer. Chem. SOC., 1954, 76, 4770;4 7 J. D. Roberts and D. A. Semenov, J . Amer. Chem. SOC., 1955, '97, 31524 8 A. D. Liehr, 2. phys. Chem., 1956, 9, 338.49 Cf. R. D. Brown, Quart. Rev., 1952, 6, 63.( I )OHC qPhys., 1955, 23, 673; J. R. Barnard and L.M. Jackman, J . , 1956, 1172.Y . I'Haya, J . Chem. Phys., 1955, 23, 1165, 1171DEWAK : QUANTUM ORGANIC CHEMISTRY. 131The correlation of methyl affinities of hydrocarbons with localisation energiesis a striking example.50Since the Huckel method is so inaccurate, and the S.C.F., the C.I., andthe A.I.M. method are so difficult, attempts have been made to find some-thing in between. The S.C.P. method can be simplied 51 in the case ofalternant hydrocarbons, and this simplified treatment has been appliedsuccessfully to a large number of polynuclear aromatic hydrocarbons. 52 Apromising method has been described 53 for including electron repulsionexplicitly in the Hiickel treatment. The free-electron approximatioil(F.E.M.O.) continues to attract interest ; it has the virtues of great simplicityand a relative freedom from empirical parameters.It is, however, equiva-lent to the Huckel treatment and much less flexible; it cannot for examplebe applied rigorously to compounds other than hydrocarbons. Nearly allapplications of the F.E.M.O. method have been to the interpretation ofspectra; attempts 55a to apply it to problems of chemical reactivity are opento criticism.55b The method can however be greatly improved by inclusionof electron repulsion, which can be done by a C.I. method; 56 the resultsfor the spectra of simple hydrocarbons seem superior to those given by theS.C.F. L.C.A.O. treatment.A com-plete V.B. treatment of the nitrogen molecule gave good agreement withexperiment for the bond energy and internuclear distance ;57 and good resultswere obtained for bond energies in hydrocarbons.58 ,4 simplified versionof it has been devised.59 Various methods, including the V.B., have beenused in studies of the hydrogen bond; 6o it is now agreed that the hydrogenbond is normally unsymmetrical, although exceptional cases with sym-metrical bonds (e.g., HF,-) are known.Other work by less popular methodsincludes studies of the carbonyl group by the semilocalised orbital method,61of benzene by the alternant orbital method,62 of simple bonds by thesmoothed potential method,63 and of bond orders, charge distributions, etc.,by the standard excited state method.64 A new potential function 65 hasThe valence bond (V.B.) method has been little used recently.60 C.A. Coulson, J., 1955, 1435; M. Szwarc and F. Leavitt, J . Amer. Chew. Soc.,b2 F. A. Matsen, J . Chem. Phys., 1956, 24, 602.63 L. Goodman and H. Shull, ibid., 1955, 23, 33.64 K. Ruedenberg, ibid., 1954, 22, 1878; 1965, 23, 401; A. A. Frost, ibid., p. 310.5 5 ( a ) S. Basu, ibid., 1954, 22, 1270, 1776, 1952; 1955, 23, 1964; (b) L. S. Bartell66 S. Olszewski, Acta Phys. PoEon., 1955, 14, 419 ; N. S. Ham and K. Ruedenberg,H.-J. Bruchner, J . Chenz. Phys., 1956, 25, 367.6* W. Heitler, Helv. Chim. Acta, 1955, 38, 5.59 G. W. Wheland, J . Chem. Phys., 1955, 23, 79.80 C. A. Coulson and U. Danielsson, Arkiv Fys., 1964, 8, 239; -4. N . Baker, J . ChewPhys., 1954, 22, 1625; T . Oshida, Y . Ooshika, and R. Miyasaka, J . Phys.SOC. Japa?,,1955, 10, 849.61 J. M. Cahill and C. R. Mueller, J . Chew. Phys., 1956, 24, 513.Li2 H. Yohizumi and T. Itoh, ibid., 1955, 23, 412; J . Phys. SOC. Japan, 1955,10, 201.c3 J . R. Arnold, J . Chem. Phys., 1956, 24, 181.64 G. G. Hall, Proc. Roy. SOC., 1955, A , 229, 251 : P. P. Manning, ibid., 1955, - 4 , 230,6 6 E. R. Lippincott, J . Chew. Phys., 1955, 23, 603.1956, 78, 3590.J. A. Pople, Trans. Faraday SOC., 1953, 49, 1375.and R. A. Bonham, ibid., 1956, 24, 909.J . Chem. Phys., 1956, 24, 1, 13; H. Labhart, Helv. Chim. Ada, 1956, 39, 1320.415, 424; H. D. Deas, Phil. Mag., 1955, 46, 670132 ORGANIC CHEMISTRY,been deduced theoretically and is claimed to be superior to the Morse func-tion. The effects of hybridisation and non-orthogonality have been dis-cussed.66 The empirid rules found by Brooker for the light absorptionof cyanine dyes have been interpreted by a generalisation of the resonancetheory.67Many chemical problems are concerned, not with the absolute values ofmolecular properties, but with the manner in which they change withmolecular structure ; the effect of substituents on the properties of a moleculeforms an obvious example.In such cases the required information can becalculated directly by representing the change in structure as a perturbationand using perturbation theory.68v6Q A general treatment of the effects ofsubstituents on light absorption has been developed in this way,70 and similarmethods have been used in investigations of the Hammett relation,?I n--x*tran~itions,'~ and the spectra of cyclooctatetraene derivatives.73Similar methods can be used very effectively in studying chemicalreactivity.69 The activation energy of a reaction can be written in theform :where AEaL, AEa, AE, are the energy differences for electrons in inner shells,in localised bonds, and in multicentre molecular orbitals respectively,between the initial and the transition state. For a series of related reactionsthe first two terms are often constant if additivity of energies of localisedelectrons is assumed; the relative rates are then determined by the delocal-isation energy difference AE,, which can be found very conveniently by aperturbation method.69 The potentialities are shown by recent work onsubstitution in polycyclic aromatic hydrocarbons. 74If Wheland's model 75 for the transition state is assumed, equation (1)becomes :where Cx is a constant characteristic of the reagent X, Nt is the reactivitynumber of atom t at which substitution takes place, and p is the carbonresonance integral.N t can be found by a very simple calc~lation.~~ Thepartial rate factors for nitration at various positions in a number of hydro-carbons have been measured 74 and agree well with equation (2). PartialAE =AE,,+ AE,+ AEw . . . . . (1)AE = Cx + N$ . . . . . . . (2)66 C . A. Coulson and G. R. Lester, Tvans. Favaday SOC., 1955, 51, 1605; R. S.Mulliken, J . Chem. Phys., 1955, 23, 1833, 1841, 2338, 2343.~37 J. R. Platt, ibid., 1956, 25, 80.68 C. A. Coulson and H.C. Longuet-Higgins, Proc. Roy. SOC., 1947, A , 191, 39;1947, A , 192, 16; 1948, A , 193, 447, 456; 1948, A , 195, 188; H. C . Longuet-Higgins,J . Chew. Phys., 1950, 18, 265, 275, 283.69 M. J . S. Dewar, J . Amer. Chem. SOC., 1952, 74, 3341,3345, 3350, 3353, 3355, 3357.70 J. N. Murrell and H. C . Longuet-Higgins, Proc. Plays. Soc., 1955, 68, A , 329, 601 :J . , 1955, 2562; H. C. Longuet-Higgins, Proc. Roy. SOL, 1956, A , 235, 537.7 1 H. H. JaffC, J . Amer. Chem. SOC., 1955, 77, 274.73 L. E. Orgel, J., 1955, 121.73 K. L. McEwen and H. C . Longuet-Higgins, J . Chem. Phys., 1956, 24, 771.74 P. M. G. Bavin and M. J. S. Dewar, J., 1956, 164; M. J. S. Dewar and T. Mole,ibid., p . 1441; M . J. S. Dewar and E. W. T. Warford, ibzd., p. 3570; M.J. S. Dewar,T. Mole, D. S. Urch, and E. W. T. Warford, ibid., p . 3572; M . J . S. Dewar, T. Mole, andE. W. T. Warford, ibid., p . 3576, 3581.7 5 G. IT. Whrland, J . Amev. Chew. SOC., 2942, 64, 900DEWAR : QUANTUM ORGANIC CHEMISTRY. 133rate factors are not known for other reactions; but plots 74 of overallreactivities of hydrocarbons to other reagents against their reactivities fornitration show that chlorination 74 and free-radical methylation andtrichloromethylation 76 also follow equation (2).There is, however, an anomaly; not only is the appropriate numericalvalue of p (3-12 kcal./mole) much less than that (-20 kcal./mole) normallyascribed to the carbon resonance integral, but the value of fi also varies fromone reaction to another. This can be reasonably explained74*75p77 byassuming that the Wheland structure (IV) is not a transition state but astable intermediate, the transition state (111) lying between the initial state(11) and (IV). In that case equation (2) becomes :when Pt* is the value of the resonance integrals for the bonds between atom tand the adjacent carbons in the transition state.If the transition state hasthe same general configuration for various reactions of X, equation (2) willhold, but the a in it will be not the true resonance integral but the numericallysmaller quantity (p - p,*) ; moreover, this will vary with the reagent.Now Pt* should be numerically greater, and so (a -- pt*) numerically smaller,the more reactive the reagent; this relation holds 73* 74 in practice, and canbe used to account for the empirical rule 78 that the selectivity of sub-stituting agents is smaller the more reactive they are.The Wheland structure corresponds to Pt* = 0 : the correspondingdifference in x-energy, AEv, is then by definition the localisation energy 49 ofatom t.Consequently the localisation energy will be an effective measureof reactivity. Since similar situations probably occur generally and it hasbeen shown 69 that the localisation energy is correlated with free valency,self-polarisability, and charge density, it is clear why these quantities alsoshow correlations with r e a ~ t i v i t y . ~ ~ The perturbational transition statetreatment is of course much more powerful, since it not only providescorrelations of this kind but can also throw light on the structure of thetransition state.A similar study has been made 79 of the solvolysis of arylmethyl chlorides.The variation of rate with structure can be interpreted quantitatively bythe P.M.O.(perturbational M.O.) method. The treatment again involves aparameter whose value vanes in a predictable manner with the structure ofthe transition state; p should be smaller the greater the participation of thenucleophile. In practice, p has its maximum value (30 kcal./mole) for the7 6 E. C. Kooyman and E. Farenhorst, Tmns. Faraduy SOL, 1953, 49, 58. '' H. C. Brown and K. L. Nelson, J . Amer. Ckem. SOL, 1953, 75, 6292.7g M. J. S. Dewar and R. J. Sampson, J., 1956, 2789 and unpublished work.M. J. S.Dewar and T. Mole, in the press134 ORGANIC CHEMISTRY.limiting (pure SN1) solvolysis in moist formic acid, dropping to a minimumvalue (5 kcal./mole) for the pure S N 2 reaction with iodide ion in acetone.80No theory of reactivity can be exact at present in view of the impossi-bility of including solvent and entropy effects ; the Huckel P.M.O. methodmay therefore prove adequate in spite of its simplicity. It is possible toapply perturbation theory to the S.C.F. method ; the resulting expressions *lshould be more accurate than those in the Huckel treatment but they are ofcourse much more complicated.Radiospectroscopy, thesubject of a recent Faraday Society Discussion,s2 seems likely to prove ofmajor importance to quantum chemistry. High-resolution nuclear magneticresonance spectra and nuclear quadrupole spectra can provide informationconcerning charge distribution in conjugated 83 and para-magnetic resonance spectra of radicals can give information concerning thedistribution of the odd electrons.84 Nuclear quadrupole spectra can alsogive information concerning double-bond character.82.85 Secondly, somerecent papers seem to suggest that the delocalisation of electrons in moleculesfor which only one classical structure can be written may be much lesssignificant than has been commonly supposed. The heats of formation ofparaffins,86 the spectra of p o l y e n e ~ , ~ ~ and the effects of alkyl substituents 88have been explained on this basis. It would be unwise to ignore thispossibility simply because it clashes with current theory ; the existingquantum-theoretical treatments are by no means so well based that theirconclusions can be accepted without reserve.9.HETEROCYCLIC COMPOUNDS.Small Rings.-The synthesis of simple optically active epoxides has beenstudied; for example, (+)-styrene oxide has been obtained from (-)-mandelic acid.l An ingenious application of ethylene and propyleneoxide as proton-acceptor solvents in certain brominations prevents thedevelopment of acidity.2 Pentaerythritol has been converted into 3 : 3-disubstituted oxacy~lobutanes,~ e.g., (1). Cationic polymerisation ofoxacyclobutanes, e.g., with boron trifluoride, affords linear pol yet her^.^^ 4Ethyleneimines are fairly weak bases (pKb 5.99-5.36) ; the basic strengthsof cyclicimines increases with increasing ring size.Five-membered ringcompounds (2) tend to lose carbon dioxide when heated and react as potential12G Y . Raoul, N. Le Boulch, C. Baron, R. Bazier, and A. Guerillot-Vinet, Bull. Soc.Chim. biol., 1956, 38, 495.1 2 7 Idem, ibid., p. 885.128 D. J. Cram and N. L. Allinger, J . Amer. Chem. Soc., 1956, 78, 5275.129 H. S. Burton, E. P. Abraham, and H. M. E. Cardwell, Bioclzem. J.. 1956, 62,171.1 E. L. Eliel and D. W. Delmonte, J . Org. Chem., 1956, 21, 596.2 D. N. Kirk, D. K. Patel, and V. Petrow, J., 1956, 627, 1184.3 A. C. Farthing, J., 1956, 3648.4 J. B. Rose, J., 1956, 542, 546.5 C. E. O'Rourke, L. B. Clapp, and J. 0. Edwards, J . Amer. Chem. Soc., 1956,78,6 S. Searles, M.Tamres, F. Block, and L. A. Quarterman, ibid., p. 4917.3159WILSON : HETEROCYCLIC COMPOUNDS. 229three-membered ring compounds ; thus, ethylene carbonate (2 ; X = 0)and oxazolid-%one (2; X = NH) exhibit some of the reactions of ethyleneoxide and ethyleneimine respectively. 7 v * Oxazirines (3) have been madefor the first time; they are formed from certain azomethines and peraceticacid, contain active oxygen, and can be assayed iodornetri~ally.~Five-membered Ring Compounds.-Applications of furans and pyrans asintermediates in organic synthesis have been reviewed. lo The rapidlygrowing chemistry and biochemistry of cr-lipoic (6 : 8-thioctic) acid havebeen summarised; l1 the properties of the acid have been investigatedsystematically, and a number of derivatives described.12 Polymers obtain-able from a-lipoic acid by oxidation under certain conditions appear to belinear di~u1phides.l~ The naturally occurring acid has been related to(+)-methylglutaric acid.14 Full details of the Imperial College synthesis l5of a-lipoic acid and an unambiguous synthesis of the isomeric (&-)-5 : 8-thioctic acid l6 (4) have been published.The rates of hydrolysis of N-acyl heterocyclic compounds have beenstudied ; such compounds are involved in certain biological transacylations.l7A number of 2 : 3-dioxopyrrolidines 15) have been made.by a useful one-stepsynthesis from ethyl acrylate, ethyl oxalate, and primary amines.l* Hydro-genation of @-oxo-ester cyanohydrins, e.g., (6)y provides a new route to pyrro-lones, e.g., (7).19 N-Arylpyrroles (8) have been obtained by a new methodfrom arylhydroxylamines and dimethyl acetylenedicarboxylate.20 Im-proved yields of dihydroxyiminoalkanes are obtained from pyrroles by usinga 2 : 1 ratio of hydroxylaniine to hydrogen chloride.21 Quaternary salts(9), obtained from oxazoles and methyl toluene-9-sulphonate, are stableto acids, but with cold alkali rapidly yield a-(acyl-N-methylamino)-ketones(10) .22 The chlorination of pyrazole and several methylpyrazoles has been7 W.H. Carlson and L. H. Cretcher, J . Amer. Chem. Soc., 1947, 69, 1952.8 J. I . Jones, Chem. and Ind., 1956, 1454; S. Sonnerskog, Acta Chem. Scand., 1956,9 W. D. Emmons, J . Amer. Chem. SOC., 1956, 78, 6208.10 R. Paul, Bull. Soc. chim.France, 1956, 838; C. H. Schmidt, AnEew. Chem.. 1956.10, 467.- . ,68, 175.11 H. Grisebach, Anpew. Chem., 1956. 68, 554.1 2 A. F. Wagner; E. Walton, G:E. Boxer; M. P. Pruss, F. W. Holly, and K. Folkers,13 R. C . Thomas and L. J. Reed, ibid., p. 6148.14 K. Mislow and W. C. Meluch, ibid., pp. 2341, 5920.16 E. A. Braude, R. P. Linstead, and K. R. H. Wooldridge, J., 1956, 3074.16 A. Campbell, J . , 1955, 4218.17 H. A. Staab, Chem. HEY., 1956, 89, 1927.18 P. L. Southwick, E. P. Previc, J. Casanova, and E. H. Carlson, J . Org. Chem.,19 H. Plieninger and M. Decker, Annalen, 1956, 598, 198.20 E. H. Huntress, T. E. Lesslie, and W. M. Hearon, J . Amel.. Chem. Soc., 1956, 78,2 l S. P. Findlay, J . Org. Chem., 1956, 21, 644.22 D. G. Ott, F. N.Hayes, and V. N. Kerr, J . Amer. Chem. SOL, 1956, 78, 1941.J . Amer. Chem. SOC., 1956, 78, 5079.1956, 21, 1087.419230 ORGANIC CHEMISTRY.studied ; 23 pyrazole with chlorine in carbon tetrachloride at 0” gives the 4-chloro-compound in 55% yield; at higher temperatures, the trimer (11) isformed, whilst with chlorine water the product (12) is obtained.(10) ( 1 1 )(Ts = p-C,H,Me.SO,.)2-Carboxymethylthiodihydroglyoxaline (13) on treatment with hydro-chloric acid undergoes an interesting rearrangementJ2* probably involvingthe bicyclic intermediate (14) , and forms the thiazolidine derivative (15).0NHThe properties of Hector’s bases, obtained by oxidation of thioureas, areconsistent with the 1 : 2 : 4-thiadiazolidine formulation (16). Ammoniarearranges them to non-basic compounds, first obtained by Dost in 1906,which have been shown 2s to have the structure (17).In a careful study of0the methylation of 5-hydroxytetrazole by means of diazomethane, fourdimethyl derivatives were isolated, and the structures of three of theseestablished unequivocally.26Meso-ionic Compounds.-Sydnones are readily halogenated at position 4,for example, by aqueous hydrobromic acid-potassium br~mate.~’ Theadducts from sydnones and quinones are believed to be indazolequinones,e.g., (18).28 It was concluded that dipole moments are not reliable forQNH.;.... ON o\ MeN-Nl $ + s lN N9 1 1 I II0-co o-c-o-( 1 9 ) (20) ( 2 1 ) ( 2 2 )the diagnosis of meso-ionic structures in the tetrazole series; thus themeso-ionic compound 5-imino-1 : 3-dimethyltetrazole (19) has a smaller dipolemoment than several tetrazoles which can be given conventional covalentre23 R.Hiittel, 0. Schafer, and G. Welzel, Annalen, 1956, 598, 186.24 J. A. Van Allan, J . Org. Chem., 1956, 21, 193.26 F. Kurzer, Chem. and Ind., 1956, 526; J., 1956, 2345.46 K. Hattori, E. Lieber, and J. P. Horwitz, J . Amer. Chem. Soc., 1956, 78, 411.27 J. C. Earl, Rec. Trav. chtim., 1956,75, 1080; H. Kata, K. Nakahara, and M. Ohta,28 D. L1. Hammick and D. J . Voaden, Chem. and I n d . , 1968, 739.J . Chem. SOC. Japan, 1956, 77, 1304WILSON : HETEROCYCLIC COMPOIJNDS. 231~tructures.~~ On being heated, the ethoxycarbonyl compound (20) cyclisesto a product (21), which can also be satisfactorily formulated as the betaine(22) .m cc-Arylazothioalkanoic acids (23) have been treated with aceticanhydride and a tertiary base, to give a series of very stable 3-aryl-4-oxo-l-thia(SIV)-2 : 3-diazolines.31 The unusual structure (24) proposed for theseproducts involves a dsp2-hybridised sulphur atom ; 32 however, the altern-ative meso-ionic formulation (26) has not been convincingly disproved.N=S N- 5( 2 3 )Ph (j 0O ( 2 4 ) -O ( 2 5 )Ph PhSix-membered Ring Compounds.-Condensation of certain pyrones,e.g., (26), with tertiary aromatic amines or with diarylethylenes in phosphorusoxychloride affords pyrylium salts of types (27) and (28) respectively.=2 : 4 : 6-Triarylpyrylium salts (29) are obtainable by a new synthesis fromchalcones and ketones.= Treatment of the products with sodium sulphidein acetone gives the corresponding thiopyrylium salts.35 Hydrochloric acidconverts the oxazine (30), obtainable from a-methylstyrene, formaldehyde,and ammonium chloride, into the base (31); dehydrogenation of the latterprovides a convenient new source of 4-phen~lpyridine.~~ 2-Trifluoromethyl-pyridine3’ is obtained in 88% yield from trifluoromethyl cyanide andbutadiene at 474”.Cleavage of the ring occurs on exposure of pyridine toultrasonic vibration; in 5% aqueous silver nitrate solution, a mixture ofsilver acetylide, diacetylide, and cyanide is produced.38 A new species,believed to be a “ charge transfer complex,’’ e g . , (32), has been identifiedin solutions of N-methypyridinium iodides ; such complexes may be29 M.H. Kaufman, F. M. Emsberger, and W. S. McEwan, J . Amer. Chem. SOC.,30 A. R. Katritzky, J . , 1956, 2063.31 G. F. Duffin and J. D. Kendall, J., 1956, 3189.32 Cf. E. B. Knott, J., 1955, 918; A. Mangani and R. Passerini, Expeuientia, 1956,33 R. Wizinger, A. Grune, and E. Jacbbi, Helv. Chim. A d a , 1956, 39, 1.34 R. Wizinger, S. Losinger, and P. Ulrich, ibid., p. 5.35 R. Wizinger and P. Ulrich, ibid., p. 207.3 6 C . J. Schmidle and R. C. Mansfield, J . Amer. Chem. SOL, 1956, 78, 1702.37 J. M. S. Jarvie, W. E. Fitzgerald, and G. J. Janz, ibid., p. 978.38 L. Zechmeister and E. F. Magoon, ibid., p. 2149.39 E. M. Kosower and P. E. Klinedinst, ihid., p. 3493.1956, 78, 4197.12, 49232 ORGANIC CHEMISTRY.important in the biochemistry of , for example, diphosphopyridinen~cleotide.~~ Treatment of certain pyridine homologues with benzylalcohol and alcoholic potassium hydroxide often causes benzylation of side-chain methyl groups.This unusual reaction has been shown to involve theformation of benzaldehyde, which condenses with certain side-chain methylgroups; reduction to C-benzyl products then occurs. The conversion ofisoquinoline into its 4-benzyl derivative proceeds by a similar mechanism.4lThe ionisation constants of a large number of hydroxypyridines and relatedcompounds have been ~urveyed.~2 In general, tautomeric equilibria in a-and y-hydroxypyridines and analogous compounds favour t he lact am form.The ratio lactam : enol is 340 for 2-hydroxypyridine, and 1 x lo7 for5-hydro~yacridine.~~ Ultraviolet absorption measurements 43 indicate that2 : 4-dihydroxypyridine exists in the a-pyridone form (33).2-Amino-pyridine is obtained in small yield from pyridine and chloramine.uOHA monograph on the chemistry of dioxans has been published.45 1 : 4-Dithiins (34) fairly readily give mono- and di-substitution products withelectrophilic reagents, and readily lose one sulphur atom to yield thiophens.46The yeast pigment pulcherrimin is the chelated ferric complex of pulcherrimicacid ; new analytical and degradative evidence supports the revised structure(35) for this acid.475-Alkylamino-4 : 6-dihydroxypyrimidines have been obtained by theRemfry-Hull synthesis from a-alkylaminomalondiamides and carboxylicesters.48 Several workers have studied the synthesis of 5-hydroxypyrim-idines , including divicine (36).Persulphates introduce a 5-hydroxyl groupdirectly into the pyrimidine ring, if at least one electron-releasing substituentis present.49 In other syntheses of 5-hydroxypyrimidinesJ ethyl glycollate isconverted into the benzyl or tetrahydropyranyl ether ; C-formylation fol-lowed by condensation with guanidine and removal of the benzyl 50 or tetra-hydropyranyl51 group completes the synthesis. The methylation and thetautomeric state of several aminopyrimidines have been investigated ; pK,40 E. M. Kosower, J . Amer. Chem. SOC., 1956, 78, 3497.4 1 M. Avramoff and Y . Sprinzak, ibid., p- 4090.4 p A. Albert and J. N. Phillips, J., 1956, 1294.4 3 H.J. Den Hertog and D. J. Buurman, Rec. Trav. chim., 1956, 75, 257.44 R I . E. Brooks and B. Rudner, J . Amer. Chern. SOL, 1956, 78, 2339.4 5 W. StumDf. “ Chemie und Anwendungen des 1 : 4-Dioxans,” Verlag ChemieWeinheim, 1956. ‘46 W. E. Paxham, I. Nicholson, and V. J. Traynelis, J . Amer. Chem. SOL, 1956, 78,850 ; cf. L. H. Szmant and L. M. Alfonso, ibid., p. 1064.4 7 A. H. Cook and C. A. Slater, J . , 1956, 4130, 4133.- -4 8 D. J. Brown, J., 1956, 2312.49 R. Hull, J . , 1956, 2033.5O J. F. W. McOmie and J. €3. Chesterfield, Chem. and Ind., 1956, 1453.5 1 J. Davoll and D. H. Laney, J., 1056, 2124WILSON : HETEROCYCLIC COMPOUNDS. 233and ultraviolet and infrared measurements indicate that 2- and 4-amino-pyrimidines exist largely in the amino-form.52 The unusual reactionof sodiomalonic ester with 4 : 6-dichloro-5-nitropyrimidine to form the5-amino-compound (37) is regarded as a nucleophilic attack at C,,, withsimultaneous reduction of the nitro-group. 53 Uracil with chlorosulphonicacid gives the 5-sulphonyl chloride 54 (38). On ultraviolet irradiation 1 : 3-dimethyluracil affords a 60-75% yield of the water adduct (39), whichreverts to dimethyluracil with acids or alkalis. 55 Work on s-triazines hasbeen continued, and several simple triazines isolated ; 56* 57 2-methyltriazineis not stable to acids and alkalis, but unlike s-triazine it does not undergoring cleavage with amines. 57 Condensation of arylamine hydrochlorides anddicyandiamide or its N-methyl derivative with a carbonyl compound pro-vides a new route to 1 : 2-dihydro-s-triazines; 58 some of the products arerearranged by alkali or by heat, e.g., (40) + (41).Condensed Ring Systems.-Several penicillin analogues, e.g., (42), havebeen synthesised.59 Benzothiepin 3-dioxide has been made from S-phen-ethylthioacetyl chloride ; the sulphur-containing ring does not have aromaticproperties, and the structure (43), analogous to that of benzotropone, appears,s,I I l lMe2fN?NHR Me2fN\INHR CO-NH-CH-CH CMe2Ar.NwN HN PhO.CH2 HlC, ,N-CH*COzHQN co NHAr(43)NH2(43) (41) ( 4 2 )to be excluded.60 The ultraviolet absorption spectra of a number of aromaticaza-hydrocarbons 61 and related polynuclear systems containing a condensedthiophen, furan, or pyrrole ring 62 have been discussed; they are somewhatsimilar to those of the corresponding aromatic hydrocarbons.Spectro-scopic studies indicate that 2-aminoindole exists in the tautomeric 2-amino-indolenine form.63 Some indoles are readily converted into the 3-aldehydesby hexamine in hot acetic acid.6452 D. J. Brown, E. Hoerger, and S. F. Mason, J . , 1955, 4035.53 F. L. Rose and D. J. Brown, J., 1956, 1953.54 R. R. Herr, T. Enkoji, and T. J. Bardos, J . Amer. Chem. Soc., 1956, 78, 401.55 S. Y. Wang, M. Apicella, and B. R. Stone, ibid., p. 4180.56 C. Grundmann and E. Kober, J . Org. Chem., 1956, 21, 641.5 7 C. Grundmann and E. Kreutzberger, J . Amer. Chem. SOC., 1955, 77, 6559.5 8 E. J . Modest, J. Org. Chem., 1956, 21, 1 ; E. J. Modest and P.Levine, ibid.,p. 14.69 J. C. Sheehan and P. A. Cruickshank, J . Amer. Chem. SOC., 1956, 78, 3677, 3680,3683; H. H. Wasserman, B. Suryanarayana, R. C. Koch, and R. L. Tse, Chem. andInd., 1956, 1022.6O W. E. Truce and F. J. Lotspeich, J . Amev. Chem. SOC., 1956, 78, 848.6 1 G. M. Badger and I. S. Walker, J., 1956, 122.62 G. M. Badger and B. J. Christie, J., 1956, 3438.83 J. Kebrle and K. Hoffmann, Helv. Chim. Acta, 1956, 39, 116.64 S. Swaminathan and S. Ranganathan, Chem. and Ind., 1955, 1774234 ORGANIC CHEMISTRY.The reactions of benziminazole have been discussed from the molecular-orbital viewpoint ; electrophilic substitution occurring in acid solutioninvolves the free base, not the cation, whilst substitutions which occur atC(2) in alkaline media probably involve the benziminazole ani0n.6~ Theantifungal substance 6-methoxybenzoxazolone (44) recently isolated frommaize and wheat plants has been synthesised from 4-phenylazoresorcin01.~6Benzoxazolone itself, which also has antifungal properties, has been isolatedfrom rye seedlings6’The basic strengths and ultraviolet and infrared absorption spectra of aconsiderable number of amino-derivatives of isoquinoline, cinnoline, andquinazoline have been discussed.68 The action of alkali on 3 : 4-dihydro-isoquinolinium salts readily affords 2-acylstyrenes.69 Studies on the form-ation of hexahydropyridocolines and related iminium salts, and on the re-actions of the latter with nucleophilic reagents, have been continued.70 Anovel route to the dehydropyridocolinium ion (46) is provided 71 bydehydration of the ketone (46), obtained from 2-cyanopyridine.Severalnaphthoquinolizinium and more complex compounds (type 47) have beendescribed. 72A valuable new monograph on acridines has been published during theyear. 73 The red pigment polystictin, isolated from the wood-rotting fungusCoriolus sa’lzguineus, is identical with cinnabarin and contains a phenox-azone nucleus (48). The actinomycins have been extensively investig-ated, particularly in Brockmann’s laboratory ; 75-79 the phenoxazone65 R. D. Brown and M. L. Heffernan, J., 1956, 4288.66 P. K. Hietala and 0. Wahlroos, Acta Chem. Scand., 1956,10, 1196; A. I. Virtanen,6 7 A. I. Virtanen and P. K. Hietala, Acta Chem. Scand., 1955, 9, 1543.68 A.R. Osborn, K. Schofield, and L. N. Short, J . , 1956, 4191.69 W. Gensler, E. M. Healy, I. Onshuus, and A. L. Bluhm, J . Amer. Chem. SOC.,1956, 78, 1713.70 N. J. Leonard, R. W. Fulmer, and A. S. Hay, ibid., 1956, 78, 3457; N. J. Leonard,L. A. Miller, and P. D. Thomas, ibid., p. 3463; N. J. Leonard and A. S. Hay, ibid.,p. 1984.7 1 E. E. Glover and G. Jones, Chem. and Ind., 1956, 1456.72 C. K. Brandsher and L. E. Beavers, J . Amer. Chem. SOC., 1956, 78, 2459.73 R. M. Acheson, “The Chemistry of Heterocyclic Compounds. Vol. IX.Acridines,” Interscience Publ. Inc., New York, 1956.74 G. W. K. Cavil1 and J. R. Tetaz, Chew. and Ind., 19.56, 986; J. Grippenberg,E. Houkanen, and 0. Patoharju, ibzd., p. 1505.7 5 H. Brockmann, G.Bohnsack, and C. H. Siiling, Angew. Chem., 1956, 12, 66;H. Brockmann and H. Grone, ibid., p. 66; H. Brockmann and H. Muxfeldt, ibid.,p. 67; H. Brockmann and B. Franck, ibid., pp. 68, 70.7 6 H. Brockmann, G. Bohnsack, B. Franck, H. Grone, H. Muxfeldt, and C. Siiling,Angew. Chem., 1956, 12, 70.7 7 H. Brockmann and K. Vohwinkel, Chew&. Bey., 1956, 89, 1373; H. Brockmannand M. Muxfeldt, ibid., pp. 1379, 1397.78 Idem, Angew. Chem., 1956,12, 69.7 9 G. G. Roussos and L. C. Vining, J., 1956, 2469.P. K. Hietala, and 0. Wahlroos, Suomen Kem., 1956, 29, B, 171WILSON HETEROCYCLIC COMPOUNDS. 235structure, e.g., (49) for actinomycin C,, has been established. 76 Depeptido-actinomycip is the acridonequinone (50) and has been synthesised; 77 itsformation from actinomycins by the action of barium hydroxide is depicted.as involving an unstable intermediate (51).78I Sar:osineL-ProlineIIcoD-alloisoLeucine I IMe- CH - CH - N H -Purines and Pteridines.-Details have been published of the isolation andsynthesis of the plant-growth factor kinetin (6-furf~rylaminopurine),~~ and anumber of kinetin analogues have been synthesised.81 The structure of" 6-succinaminopurine (a-6-purinylaminosuccinic acid) (52 ; R = H),which arises in the hydrolysis of an unusual nucleotide, has been confirmedby synthesis from 2 : 6 : 8-trichloropurine; 82 the latter was condensed withaspartic acid to yield the dichloro-compound ( 5 2 ; R = Cl), which wasdechlorinated by means of phosphine and hydrogen iodide.An alternative synthesis of pteridines is provided by the acylation andcyclisation of 2-aminopyrazine-3-carbox yamides (53), which are obtainablefrom 4 : 5-diamino-3-hydroxypyrazoles (54) by condensation with a-dicarb-onyl compounds and hydrogenation of the resulting pyrazolopyrazines(55).*3' 84 The physical and chemical properties of pteridine and its simplehydroxy- and amino-derivatives have been investigated sy~tematically.8~Pteridine is a weak acid (pK, 12*2), and in dilute hydrochloric acid is hydro-lysed to 2-amino-3-formylpyrazine.s5 Spectroscopic studies indicate thatthe four monohydroxypteridines exist in the lactam form.86 2- and 6-Hydr-oxypteridine bind water strongly, it is believed by addition to the 3 : 4- and7 : %double bond respectively.86 The isolation and characterisation ofbiopterin from human urine has been de~cribed,~' and this pterin has been80 C.0. Miller, F. Skoog, F. S. Okumura, M. H. Von Saltza, and F. M. Strong,81 M. W. Bullock, J. J . Hand, and E. L. R. Stokstad, ibid., p. 3693.8 2 J. Baddiley, J. G. Buchanan, F. J. Hawker, and J. E. Stephenson, J . , 1956, 4659.83 E. C. Taylor, R. B. Garland, and C . F. Howell, J . Amer. Chem. SOC., 1956, 78, 211.84 T. S. Osdene and E. C. Taylor, ibid., p. 5451.86 A. Albert, D. J. Brown, and H. C. S. Wood, J . , 1956, 2066; A. Albert, J . 13.86 D. J. Brown and S. F. Mason, J . , 1956, 3443.87 E. L. Patterson, M. H. Von Saltza, and E. L. R. Stokstad, J . Amer. Chem. SOC.,J . Amer. Chem. Soc., 1956, 78, 1375.Lister, and C.Pedersen, J . , 1956, 4621.1956, 78, 587 1 236 ORGANIC CAEMISTRY.synthesised from 2 : 4 : 5-triamino-6-hydroxyyrimidine and 5-deoxy-~-arabinose.88 The structure assigned to urothione is partially confirmed bythe synthesis of the simple analogue (56).89and theuse of isoindolenines as intermediates for synthesis of phthalocyanines 91have been discussed. Chlorin (57) has been synthesisedg2 in 3.974, yieldPorphyrins.-Recent developments in porphyrin chemistryfrom 2-dimethylaminometh~lpyrrole and ethylmagnesium bromide ino-dichlorobenzene at 180”. Dehydrogenation of chlorin gives porphin ; 93these results are important in connection with the problem of the location ofthe two “ extra ” hydrogens in chl~rophyll.~~ The structure of vitamin B,,has been further refined by X-ray crystallographic studies,95 and more workhas been done on the related “factor 111,” which contains a 5-hydroxy-benziminazole nucleoside fragment .96Complex Cyclic Oxygen Compounds.-The condensation of phenols withcinnamic acid provides a new route t o 3 : 4-dihydro-4-phenylcoumarins.97The reaction products from phenol ethers and malonyl chloride are 4-hydr-oxycoumarins, and not indanediones as hitherto supposed.98 It has beenestablishedg9 that Dianin’s compound, first made in 1914 from phenol,mesityl oxide, and hydrogen chloride, is the phenylchroman derivative (58),and this has been confirmed by independent synthesis.100 Dianin’s com-pound forms inclusion products with a remarkable variety of substances,including iodine, organic solvents, and some inorganic gasesggThe structure of the antibiotic novobiocin (streptonivicin) (59) has been8 8 E.L. Patterson, R. Milstrey, and E. L. R. Stokstad, J . Amer. Chem. SOC,, 1956,89 R. Tschesche and G. Heuschkel, Chem. Bey., 1956, 89, 1054.90 K . Zeile, Angew. Chem., 1956, 68, 193; R. J. P. Williams, Chem. Rev., 1956, 56,9 1 F. Baumann, B. Bienert, G. Rosch, H. Vollmann, and W. Wolf, Angew. Chem.,92 U . Eisner and R. P. Linstead, J . , 1955, 3742.93 Idem, ibid., p. 3749.94 G. E. Ficken, K. B. Johns, and R. P. Linstead, J., 1956, 2272.95 D. C. Hodgkin, J. Kamper, M. Mackay, J. Pickworth, K. N. Trueblood, and J. G.White Nature 1956, 178, 64.96’C. H. ;hunk, F. M. Robinson, J. F. McPherson, M. M.Gasser, and K, Folkers,1. Amer. Chern. SOG., 1956, 78, 3228; W. Friedrich and K. Bernhauer, Angew. Chem.,78, 5868.299.1956, 88, 133.i956, 68, 439.97 J. D. Simpson and H. Stephen, J., 1956, 1382.98 J. F. Garden, N. F..Hayes, and R. H. Thomson, J., 1956, 3315.99 W. Baker, A. J. Floyd, J. F. W. McOmie, G. Pope, A. S. Weaving, and J. H. Wild,J . , 1956, 2010.100 W. Baker, J. F. W. McOmie, and A. S. Weaving, J . , 1956, 2018WILSON : HETEROCYCLIC COMPOUNDS. 237deduced in two laboratories,lOl and the degradation products cyclonovobiocicacid (60) and dihydronovobiocic acid have been synthesised. lo2There has been continued activity in the study of flavonoids and relatedcompounds ; the proceedings of the symposium held last year in Dublin havebeen pub1i~hed.l~~ The natural tannins have been reviewed.lo4 Infraredspectroscopy is a simple method for the identification of anthocyanin pig-ments, especially when only small samples (ca. 1 mg.) are available.lo5 Theultraviolet absorption characteristics of a large number of aurones have beenreported.lO6 3-Aroylcoumarones, a number of which have been synthesised,are labile to acids; they are believed to be formed as intermediate duringthe rapid resinification of 2’-methoxyisoflavones on treatment with acids.lo76 : 8-Dihydroxyflavone has been synthesised for the first time, by two groupsof workers.108 A new synthesis lo9 of flavonols of the quercetagetin seriesinvolves the introduction of the 5-hydroxyl group by treating 5-formylcompounds with alkaline hydrogen peroxide : the formyl group is introducedin the first place by means of hexamine in acetic acid.This procedure wasemployed 110 in the synthesis of oxyanin-B (61). A crystalline Zeuco-anthocyanidin isolated from wattle wood appears to be a trihydroxyflavan-3 : 4-diol,ll1 and the synthesis of flavan-3 : 4-diols, which are important asthe likely precursors of both flavonoids and anthocyanins, has been in-l01 C. H. Shunk, C. H. Stammer, E. A. Kaczka, E. Walton, C. F. Spencer, A. N.Wilson, J. W. Lichter, F. W. Holly, and K. Folkers, J . Amer. Chem. SOC., 1956, 78.1770; H. Hoeksema, E. L. Caron, and J . W. Hinman, ibid., pp. 1072, 2019.l02 C. F. Spencer, C. H. Stammer, J. 0. Rodin, E. Walton, F. W. Holly, and K.Folkers, ibid., p.2655.103 Sci. Proc. Roy. Dublin Soc., 1956, 27, 75-192.1°4 0. T. Schmidt and W. Mayer, Angew. Chem., 1956, 68, 103.105 K. C. Li and A. C. Wagenknecht, J . Amer. Chem. SOC., 1956, 78, 979.106 T. A. Geissman and J. B. Harborne, ibid., p. 832.107 W. B. Whalley and G. Lloyd, J . , 1956, 3213.108 T. H. Simpson, Chem. and Ind., 1955, 1672; J . E. Gowan, S. P. &I. Riogh,G. H. McMahon, B. R. O’Farrell, S. O’Cleirigh, E. M. Philbin, and T. S. Wheeler, ibid.,p. 1672.lo9 A. C. Jain, T. R. Seshadri, and K. R. Sreenivasan, J., 1955, 3908.110 R. N. Goel, A. C,. Jain, and T. R. Seshadri, J . , 1956, 1369.ll1 H. H. Keppler, Chem. and Ind., 1956, 380238 ORGANIC CHEMISTRY.vestigated. 112, 113 Periodic acid oxidation of melacacidin tetramet hyl ether(62) yields the benzofuran derivative (63) by ring contraction.ll* Thesulphonation of chromone derivatives has been studied ; 2 : 3-dimethyl-chromone gives the 6-sulphonic acid, and 7-hydroxyflavone and the iso-flavones give 8-sulphonic and 6 : 8-disulphonic acids.l15 The nuclearmethylation of flavonoids and the natural occurrence of the C-methylderivatives have been reviewed.n6The Wesseley-Moser rearrangement of flavonols, chromonols, andxanthones has been studied : 117 contrary to earlier reports, 5 : 8-dihydroxy-compounds can be rearranged, e.g., (64) + (65), although drastic concli-tions are often required. The structure of the hardwood extractive lapo-chenole (66) has been elucidated 118 and confirmed by synthesis.l19 Thetrimethyl ether of wedelolactone, a new product isolated from Wedeliacalenddacea, is believed 120 to have the structure (67).Details of workreported briefly last year on grevifolin,121 fuscin,122 and usnic acid 123 havenow been published. The oxidation of monohydric phenols, e.g., 9-cresol toYummerer’s ketone (68), by alkaline ferricyanide has been studied by severalworkers : 124 the first stage in the synthesis 122 of usnic acid involves a reactionof this kind. The reactions of porphyrilic acid,125 recently isolated from the112 R. Bogn&r and M. Rtikosi, Chem. and Ind., 1956, 188.113 A. B. Kulkarni and C. G. Joshi, ibid., p. 124.114 W. Bottomley, ibid., p. 170.115 D. V. Joshi, J. R. Merchant, and R. C. Shah, J . Org. Chem., 1956, 21, 1104.116 A.C. Jain and T. R. Seshadri, Quart. Rev., 1956, 10, 169.11’ D. M. Donnelly, E. M. Philbin, and T. S. Wheeler, J., 1956, 4409; E. 1L1. I’hilbin,118 R. Livingstone and M. C. Whiting, J . , 1955, 3631.119 R. Livingstone and R. B. Watson, J . , 1956, 3701.120 T. R. Govindachari, K. Nagarajan, and B. R. Pai, J . , 1956, 629.121 J. Grimshaw and R. D. Haworth, J . , 1956, 418.122 D. H. R. Barton and J . B. Hendrickson, J . , 1956, 1028.123 D. H. K. Barton, A. M. Deflorin, and 0. E. Edwards, J . , 1956, 530.124 V. Arkley, F. M. Dean, A. Robertson, and P. Sidisunthorn, J . , 1956, 2322; C. G.125 11. Erdttnan and C. A . Mrachtmeister, Chem. and Iitd., 1956, 960.J. Swirski, and T. S. Wheeler, J . , 1956, 4455.Haynes, A. H. Turner, and W. A. Waters, J., 1956, 2823SMITH ALKALOIDS. 239lichen Haematomma coccin.eum, Dicks, suggest the structure (69).Flavo-gallol, C21HS02, obtained from gallic acid by oxidation with arsenic acid-sulphuric acid is probably represented by (70) ,126 although this structurehad been considered and rejected many years ago. The constituents ofAmmi visnaga Linn. have been actively studied, and revised structuresproposed for visnagane and k h e l l a ~ t o n e . ~ ~ ~ Visamminol has formula (71) .128The synthesis of (&)-sesamin and of (j-)-asarinin has been accomplished intwo laboratories.129 w. IV.10. ALKALOIDS.A REVIEW of the recent rapid developments in the chemistry and pharmaco-logy of the Rauwolfia alkaloids and another on alkaloids related to anthra-nilic acid have been published.An article in Quarterly Reviews deals withrecent work on indole alkaloids excluding harmine and ~trychnine.~ Fulldetails have now been published of the total synthesis of lysergic acid andof m~rphine.~ Admirable work on 400 mg. of the difficultly accessiblemuscarine has led to the proposal of a probable structure (1) for the alkaloid;one point for which there is as yet no adequate explanation is the failureof the alkaloid to undergo Hofrnann degradation.6 The absolute configur-ation of s'trychnine has been established by a direct X-ray crystallographicmethod. The formation of N-2-carboxyethyl-~-apartic acid by oxidationof (+)-laudanosine leads to the absolute configuration of this and relatedtetrahydroisoquinoline, aporphine, and tetrahydroberberine alkaloids.8 Alecture largely devoted to a review and discussion of the role of biogeneticschemes in alkaloid chemistry has been p~blished.~ Tracer work on thebiogenesis of alkaloids continues : although lysine is a precursor of thepiperidine ring of anabasine, it has been shown not to be a precursor of thepyridine ring of nitotine and anabasine in the species studied.1°Tropane Group.-The total synthesis of scopolamine has been com-pleted : l1 the difficulty surmounted was the epoxidation of (a), the synthesisof which had been announced earlier.12 Interesting results have been126 J.Grimshaw and R. D. Haworth, J . , 1956, 4225.12' L. Fabrini, Ann. Chim. (Italy), 1956, 46, 130, 137; L. W. Bencze, 0. Halpern,and H.Schmidt, Experientia, 1956, 12, 137.128 W. Bencze, J. Eisenbeiss, and H. Schmidt, Helv. Chinz. Acta, 1956, 39, 923.I29 M. Beroza and M. S. Schechter, J . Amer. Chem. SOC., 1956,78, 1242; K. Freuden-berg and E. Fischer, Naturwiss., 1956, 43, 16; Chew&. Ber., 1956, 89, 1230.A. Chatterjee, S. C. Pakrashi, and G. Werner, Forfschr. Chem. org. Natz~rsfofle,1956, 13, 346.J. R. Price, ibid., p. 302.J. E. Saxton, Quart. Rev., 1956, 10, 108.E. C. Kornfeld, E. J. Fornefeld, G. B. Kline, M. J . Mann, D. E. Morrison, R. G.5 &I. Gates and G. Tschudi, ibid., p. 1380.C. H. Eugster, Helv. Chim. Acta, 1956, 39, 1002, 1023.A. F. Peerdeman, Acta Cryst., 1956, 9, 824.8 H. Corrodi and E. Hardegger, Helv. Chim. Acta, 1956, 39, 889.R. B. Woodward, Angew.Chem., 1956, 68, 13.lo E. Leete, J . Amer. Chem. SOC., 1956, 78, 3520.l 1 G. Fodor, J. T6th, I. Koczor, P. Dobo, and I. Vincze, Chem. and Ind., 1956, 764.l2 G. Fodor, J. T6th, I. Koczor, and I. Vincze, ibid., 1955, 1369.Jones, and R. B. Woodward, J . Awer. Chem. SOC., 1956, 78, 3087240 ORGANIC CHEMISTRY.obtained in connection with the stereochemistry of quaternisation of tropanederivatives : thus tropan-a-ol ethiodide has been found to be different fromN-ethylnortropan-a-ol methiodide.13 The degradation of dioscorine is pre-senting considerable difficulties ; a tentative structure (3) has been proposedfor the a1kal0id.l~+H2C-CH*NMe3H? I IMe*CH - HC, ,CH20( 1 ) ( 2 )Lupinane Group.-A second and independent synthesis of (-)-cytisinehas been achieved : l5 the vinylpyridine (4) reacted with sodiomalonic esterto give the Michael adduct, hydrogenated (Raney Ni) to the pyridocoline (5),which with concentrated hydrobromic acid, then ammonia, and finally ringclosure yielded a mixture of stereoisomeric tetrahydrocytisines (6).De-hydrogenation of the acetylated mixture, followed by hydrolysis and distill-ation, gave in low yield (&)-cystkine (7), isolated as the picrate : the basewas resolved as the (+)-camphorsulphonate. The total synthesis of(&)-lupanine (12) has been achieved l6 through a relay (lo), itself preparedfrom naturally occurring lupanine (12) : the pyridocoline derivative (8) washydrogenated to yield, by ring closure, a mixture of stereoisomers of struc-ture (9) which successively with lithium aluminium hydride, thionyl chloride,trimethylamine, and alkali gave, after chromatography, the relay (10) ; thiswas characterised as the crystalline picrate and hydriodide and was thenconverted into the opened derivative (11) by benzoyl chloride in alkali,which with hypobromite, then ethanolic hydrochloric acid, and finallylactamisation yielded (-J-)-lupanine (12).A much neater total synthesisl3 G. Fodor, K. Koczka, and J. Lestyan, J., 1956, 1411.l4 A. R. Pinder, ibid., p. 1577.l5 F. Bohlmann, A. Englisch, N. Ottawa, H. Sander, and W. Weise, Angew. Chern.,16 G. R. Clemo, K. Raper, and J. C. Seaton, J., 1956, 3390.1955, 67, 708; Chem. Ber., 1956, 89, 792SMITH Z ALKALOIDS. 241yields (&)-anagyrine (13) and (&)-thermopsine (a diastereoisomer of 13)as well as (5)-lupanine : compound (14) was condensed with 2 : 3 : 4 : 5-tetrahydropyridine to give the acid (15) which, after epimerization, wastreated with lithium aluminium hydride followed by concentrated hydro-bromic acid; quaternisation in boiling benzene then gave the salt (16),alkaline ferricyanide oxidation of which led to ( &)-anagyrine (13).Thishas already been reduced to (&)-lupanine.18 The synthetic anagyrine wasepimerized to ( f)-thermopsine by oxidation with mercuric acetate, followedby hydr0genation.1~a- and P-Diplospartyrines have been shown to have structure (17) differ-ing by the configuration at C*; the formation of these compounds finds ananalogy in the dimerisation of 2 : 3 : 4 : 5-tetrahydropyridine; l9 supportfor structure (17) comes from a rational synthesis from two sparteine oxid-ation products of known structure.**Nordehydro-cc-matrinidine (1 8) has been synthesised.21 Leontine, analkaloid of Leontice ewersnzannii Bge., has been shown to be closely relatedto matrine by reduction to (-)-matridine (19) 22 by lithium aluminiumhydride.Pyrrole and Pyridine Group.-Bellaradine has been identified withcus~ohygrine.~~ Pinidine is ( -) -cis-2-methy1-6-prop-l -enylylpiperidine ; inPinus sabiniana Dougl.it occurs with (+)-a-pipec~line.~~ Further attemptsto synthesise p-Z-piperidylpropionaldehyde, the supposed pelletierine, havefailed; 25926 a specimen of natural " pelletierine " has been shown to beidentical with isopelletierine : 25 it would seem that the aldehyde alkaloiddoes not exist.The new alkaloid, (+)-sedridine, has been shown to bel i E. E. van Tamelen and J. S . Baran, J . Amer. Chew Soc., 1956, 78, 2913.IR H. R. Ing, I . , 1933, 504.19 C. Schopf and H. L. de Waal, with K. Keller, Chem. Bey., 1956, 89, 909.2o C. Schopf and K. Keller, Naturwiss., 1956, 43, 325.21 K. Tsuda, S. Okuda, S. Saeki, S. I. Mura, Y . Sato, and H. Mishima, J . Org. Chem.,z2 T. F. Platonova and A. D. Kuzovkov, Zhur. obshchei Khim., 1956, 26, 283.23 E. Steinegger and G. Phokas, Phavm. Acta Helv., 1955, 30, 441.z4 W. H. Tallent and E. C. Homing, J . Awer. Chem. SOL, 1956, 78, 4467.25 J. P. Wibaut and M. I. Hirschel, Rec. Tvav. chinz., 1956, '95, 225.2 6 R.E. Bowman and D. D. Evans, J . , 1956, 2553.1956, 21, 598242 ORGANIC CHEMISTRY.%2'-hydro~ypropylpiperidine.~~ Two independent syntheses of ethylcarpyrinate, a dehydrogenation product of ethyl carpamate, have been29 The structure of nicotelline (20), arrived at by oxidativedegradati~n,~~ has been confirmed by an elegant synthesis.31 The structureof gentianine (21), proposed on a new analysis of available data,32 has beenconfirmed by the synthesis of dihydr~gentianine.~~Quinoline Group.-Flindersine 34 (22), dictamnine 351 35a (23 ; R = H),and y-fagarine 3 5 9 35b (23 ; R = OMe) have been synthesised.#( 2 0 )'N'0isoQuinoline Group.-A synthesis of ( &)-isothebaine ethyl ether 36finally proves the structure proposed for the alkaloid itself by Klee.37Structure (24) has been assigned to tasyine, which is thus a most interestinginstance of a " degraded " aporphine system.3s The cactus alkaloid pilo-cereine has structure (25), and is thus the first alkaloid with a C-isobutylIndole Group.-A stereospecific synthesis of (&) -yohimbane 40 has con-firmed the ~ ~ L Z ~ S - D / E ring-junction assigned by Witkop *l to yohimbine on27 H.C. Beyermann and Y . M. F. Muller, Rec. Trav. chim., 1955, 74, 1568.28 T. R. Govindachari, N. S. Narasimhan, and S. Rajadurai, Chem. and Id., 1956,53.29 H. Rapoport and E. J . Volcheck, J . Amer. Chem. Soc., 1956, 78, 2451.30 F. Kuffner and N. Faded, Monatsh., 1956, 86, 71.31 J. Thesing and A. Miiller, Angew. Chem., 1956, 68, 577.32 N.F. Proskurnina and V. V. Shpanov, Zhur. obshchei Khim., 1966, 26, 936.33 T. R. Govindachari, K. Nagarajan, and S. Rajappa, Chem. and Ind., 1956, 1017.34 R. F. C. Brown, (the late) G. K. Hughes, and E. Ritchie, Austral. J . Chern., 1956,35 M. F. Grundon and N. J . McCorkingdale, Chem. and Id., 1956, 1091.35a H. Tuppy and F. Bohm, Monatsh., 1956, 87, 720.358 Idewz, ibid.. p. 774.36 K. W. Bentley and S. F. Dyke, ibid., p. 1054.37 W. Klee, Arch. Pharm., 1914, 252, 211.38 T. F. Platonova, A. D. Kurzovkov, and Yu. N. Sheinker, Zhur. obshchei Khim.,39 C. Djerassi, S. K. Figdor, J. M. Bobbitt, and F. X. Markley, J . Amer. Chem. Soc.,40 E. E. van Tamelen and H. Shamma, ibid., 1964, 76, 950; E. E. van Tamelen,41 B. Witkop, ibid., 1949, 71, 2659.9, 277.1956, 26, 2651.1956, 78, 3861.H. Shamma, and P.E. Aldrich, ibid., p. 4628SMITH ALKALOIDS. 243the basis of the formation of trans-decahydro-2-methylisoquinoline fromchanodeoxydihydroyohimbol : traas-hexahydroindan-2-one (26) was oxidisedby perbenzoic acid to the lactone (27), which was then converted by hydro-bromic acid into the acid (28); this with tryptamine gave the lactam (29)( 2 6 ) 4converted by phosphorus oxychloride into the quaternary pentacyclic base(30), whence hydrogenation gave a high yield of pure ( &)-yohimbane. Noneof these reactions can affect the critical trans-arrangement originally presentin the hexahydroindanone. The same school announced a stereospecificsynthesis of ( &)-dihydrocorynantheane (31), whence it follows that coryn-antheine and dihydrocorynantheine have a 3 : 15 : 20-cis-trans- and coryn-antheidine a 3 : 15 : 20-cis-cis-c0nfiguration.4~ The non-stereospecific syn-thesis of ( &-)-16-methylyohimbane has been unexpectedly achieved as theresult of a most remarkable Wolff-Kishner reduction involving the con-version of the keto-acid (32) into the lactam (33).43Perhaps the most outstanding achievement in the alkaloid field this yearhas been the total synthesis of reserpine by Woodward and his co-workers 44(see chart); the starting material (34) was prepared by the addition ofmethyl penta-2 : 4-dienoate to benzoquinone; the structure of the epoxy-lactone (36) does not follow from the reaction sequence shown, but hadalready been established by a longer series of reactions ; the very remarkablefeature of this compound is that it contains all five of the asymmetric carbonatoms of ring E of reserpine, properly oriented; the conversion into (38) ofthe bromo-keto-lactone, formed by oxidation of the bromo-hydrosy-lactoneJY E.E. van Tamelen, P. E. Aldrich, and T. J. Katz, Chem. aiad Ind., 1956, 793.43 F. L. Weisenborn and H. E. Applegate, ibid., p. 2021.44 R. B. Woodward, F. E. Bader, H. Bickel, A. J. Frey, and R. W. Kierstead, ibid.,p. 2023, 2657244 ORGANIC CHEMISTRY.(37), involves hydrogenolysis both of the lactone 0-C and of the Br-C bond;the position of the carbonyl group in this product (38) is of course of criticalimportance for the success of the synthesis; the diol (39) was converteddirectly in high yield into (41) without the isolation of labile intermediatessuch as (40) ; reduction of the pentacyclic quaternary salt (42) unfortunately( 3 4 )4, O%MeOQ( 4 1 ) OMe - O M e 42) O M e( 4 3 )Reagents : I , AI(OPri),-PriOH.2, Br,-MeOH, then NaOMe. 3, N-Bromosuccinimide.4, Cr0,-AcOH, then Zn-AcOH. 5, CH2N2, then Ac,O-pyridine, then Os04-KCI0,. 6,HIO4. 7, CH,N2, then 6-methoxytryptamine, then NaBH,. 8, POCI,. 9, NaBH,.gave a derivative of isoreserpine (43); this compound was resolved to the(-)-form; the final stages involve an epimerisation at via isoreserpicacid, to isoreserpic lactone, isomerised to reserpic lactone by pivalic acid inboiling xylene, and thence converted by obvious steps into reserpine.Thesynthesis of reserpine is somewhat simplified by the observation that3-dehydroreserpine is reduced to reserpine by zinc and aqueous acid.45 Theinteresting generalisation has been made that in the yohimbine-reserpinegroup, only compounds with an axial hydrogen atom at Col are dehydro-genated by mercuric acetate to the 3-dehydro-deri~atives.~~ isoReserpineis converted into reserpine by equilibration in boiling acetic a ~ i d . ~ 6 Addi-tional evidence for the configuration assigned to the 17-methoxyl group ofreserpine has been presented ; the 17-imethoxyl group of deserpidine hasbeen shown to have the same ~rientation.~’ The Rauwolfia species continueto be the subject of intensive chemical investigation. PseudoReserpine, a45 F. L. Weisenborn and P.A. Diassi, Chem. and Ind., 1956, 2022.46 C. F. Huebner, M. E. Kuehne, B. Korzum, and E. Schlittler, Ex+evientia, 1956,4 7 c. F. Huebner and D. F. Dickez, ibid., p. 250.12, 249SMITH : ALKALOIDS. 245new alkaloid from RazczuoZJia canescens, is an ester of 3 : 4 : 5-trimethoxy-benzoic acid and methyl Pseudoreserpate ; the latter is converted by boilingacetic anhydride into methyl 17-norisoreserpate, and can thus be givenstructure (44) .48Although its structure has been known for seven years, alstyrine (45),the important product of dehydrogenation of many indole alkaloids, has onlyrecently been synthesised : 49 the final step in the synthesis was a Fischerreaction on the phenylhydrazone of 2-butyryl-4 : 5-diethylpyridine. Robin-son's structure for ajmaline has been modified to (46) both on biogeneticgrounds and on the results of the following reactions9 of deoxydihydro-ajmaline (47) : (i) oxidation to a cyclopentanone (identified by its infraredspectrum) ; (ii) oxidation to the aldehyde (48), a most interesting reaction ;OH6H& Me(51)(iii) dehydrogenation under mild conditions to ind-N-methylharman, theharman derivative (49), and the indolylmethylpyridine derivative (50).Both the last two compounds have been synthesised.mTetraphyllicine, from RauwoZJia tetraphylla, has been shown to havestructure (51) ; 51 rauvomitine, from R.vomitoria, is its 3 : 4 : 5-trimethoxy-benzoate; 51 and ajmalidine, from R. seZZozwii is probably correctly repre-sented as (52).51The work of the Zurich group on the calabash-curare alkaloids continues,but there is no major progress to report : curarine has been shown to be aC,, and not a C,, base by the formation of a rnon~methiodide.~~ Alstoniline4 8 M.W. Klohs, F.Keller, R. E. Williams, and G. W. Kusserow, Chem. and Ind.,50 R. B. Woodward, quoted in ref. 1.61 C. Djerassi, M. Gorman, S. C. Pakrashi, and R. B. Woodward, J. Amer. Chem.63 W. von Philipsborn, H. Schmid, and P. Karrer, Helv. Chim. Ada, 1956, 39, 913.1956, 187.T. B. Lee and G. A. Swan, J., 1956, 771.Soc., 1956, 78, 1259246 ORGANIC CHEMISTRY.oxide has formula (53), for alkaline degradation yields 7-methoxynorharmanand toluene-2 : 6-dicarboxylic acid.= The two products of the degradationof folicanthine have been shown to be 9-methylnorharman and l-methyl-3-2'-methylaminoethylindole : this leads to two possible structures for thealkaloid.54 Iboluteine (54) has been shown to be an indoxyl alkaloid related(54) ( 5 5 )to ibogaine (55) : the two are interconvertible.55 Gelsemine has provided asurprise, for what had been considered for some years to be an exocyclicmethylene group has now been proved to be a vinyl group by oxidation toan aldehyde containing one carbon atom less.56 This, and the demonstrationby oxidation with diethyl azodicarboxylate that the grouping *NMe*CH,* ispre~ent,~T is still the only definite information available on the aliphaticportion of the molecule.Erythrina Group.-Direct chemical proof for the presence of theerythrinan skeleton in the aromatic Erythrina alkaloids has been obtainedboth degradatively and synthetically : dihydroerysotrine (56) has beendegraded by the cyanogen bromide method to a secondary amine formulated( 5 9 )as (57) because of its oxidation to the diphenic acid (58) ; 5* two independentsyntheses of ( -J-)-hexahydroapoerysotrine (59), obtained from erysotrine intwo steps not likely to cause skeletal rearra~~gement,~~ have beenannounced.6o> 61 One of them involves a remarkable formation of the53 R.C. Elderfield and 0. L. McCurdy, J . Org. Chem., 1956, 21, 295.54 H. F. Hodson and G. F. Smith, Chem. and Ind.', 1956, 740.5 5 M. Goutarel, M.-M. Janot, F. Mathys, and V. Prelog, Helv. Chim. Acfa, 1966, 39,6 6 L. Marion and K. Sargeant, J .Amer. Chem. SOC., 1956, 78, 5127.5 7 T. Habgood and L. Marion, Canad. J . Chem., 1955, 33, 604,5* V. Prelog, B. C . McKusick, J. R. Merchant, S . Julia, and M. Wilhelm, Helv.59 M. Carmack, B. C. McKusick, and V. Prelog, ibid., 1951, 54, 1601.60 B. Belleau, Chem. and Ind., 1956, 410.6 1 A. Mondon, Angtw. Chem., 1956, 68, 578.742; cf. Ann. Reports, 1955, 52, 245.Chim. Acta, 1956, 39, 498SMITH : ALKALOIDS. 247erythrinan skeleton by the simple heating together of dimethoxyphenethyl-amine and the ethylene ketal of 2-oxocycZohexylacetic ester.61 The completestructure of erythraline (60) has been worked out by X-ray crystallo-graphy.62 This very interesting achievement confirms the results of chemicalstudy and in addition fixes the position and conformation of the aliphaticmethoxyl group.Pyrrolizidine Group.-Work in this field has been limited to the elucid-ation of structure of trichodesmic, junceic, grantianic, sceleranecic, andjaconecic acid, which has led to proposals for the structures of tricho-d e ~ m i n e , ~ ~ j ~ n c e i n e , ~ ~ grantianine,G5 sceleratine, 65* 6G j acobine, andj a ~ o n i n e .~ ~Phenanthridine Group.-This fascinating and rapidly expanding field isbeing intensively studied. As many as 16 new bases have been isolatedthis year by one group alone.68 Structures have been proposed for thefollowing fifteen bases : pseudolycorine, methyl$seudolycorine, 69 galanthine, 7oand ~ a r a n i n e , ~ ~ which are closely related to lycorine ; h ~ r n a n t h i d i n e , ~ ~which is N-demethyltazettine ; crinine 73 (61), powelline, buphanidrine, andbuphanamine, 74 all closely related ; hippeastrin, 75 neronine, clivonine,krigeine, and alboma~uline,7~ which are related to homolycorine.The finaltouches have been put to the structure of lycorine (62; K = H) : the doublebond has been finally proved to be in ring c by allylic oxidation of the mono-acetyl-lycorine (62; R = Ac) to an +unsaturated ketone.77 A convincingOH OH OHconformational analysis 78 of dihydrolycorine results in the stereochemistryof this base being represented by (63). The structure of tazettine (64) nowseems to have been satisfactorily worked out : the most fascinating chemical62 W. Nowacki and G. F. Bonsma, quoted in ref. 58.63 R.Adams and M. Gianturco, J. Amer. Chem. SOC., 1956, 78, 1922.64 Idem, ibid., p. 1926.6 5 Idem, ibid., p. 4458.6 8 H. L. de Waal and B. L. van Duuren, ibid., p. 4464.6 7 R. B. Bradbury and J. B. Willis, Austral. J. Chem., 1956, 9, 258.68 H.-G. Boit, Chem. Ber., 1956, 89, 1129; H.-G. Boit and H. Ehmke, ibid., p. 163;H.-G. Boit and W. Dopke, ibid., p. 3462.69 H. M. Fales, L. D. Giuffrida, and W. C. Wildman, J. Amer. Chem. Soc., 1956, 78,4145.70 H. M. Fales and W. C. Wildman, ibid., p. 4151.7l E. W. Warnhoff and W. C. Wildman, Chem. and Ind., 1956, 348.72 H.-G. Boit and W. Stender, Chem. Ber., 1956,89,161; W. C. Wildman, Chenz. and73 W. C. Wildman, J. Amer. Chem. SOC., 1956, 78, 4180.74 Idem, Chem. and Ind., 1956, 1090.7 5 H.-G. Boit and H.Ehmke, Chem. Ber., 1956, 89, 2093.7 6 C. K. Briggs, P. F. Highet, R. J . Highet, and W. C. Wildman, J. Amer. Chern.7 7 Y. Nakagawa, S. Uyeo, and H. Yajima, Chem. and Ind., 1956, 1238.Ind., 1956, 123.SOC., 1956, 78, 2899.K. Takeda and K. Kotera, ibid., 347248 ORGANIC CHEMISTRY.transformations have come to light in the rationalisation of the degradationsequences.79% 8o For example, there is the methylation of tazettine (64)under alkaline conditions, which leads to (69) : 79 tazettine is visualised asbeing in eqnilibrium with (65) by a series of prototropic changes; methyl-ation of (65) and a prototropic shift lead to (66), which ring-opens to (67), ismethylated to (659, which loses trimethylamine with migration of the acetylOMeOMe 8 L O( 6 9 )H- \-0 (70)OMeNMe2FH2F05)-CH2group to give (69).On the other hand, if tazettine methohydroxide isprepared by the action of silver oxide on the methiodide, the reactionproceeds through (70) to give tazettine methine (71)1 : 2-Benzophenanthridine Group.-The first synthesis in this group,that of chelerythrine chloride, has been achieved. 81Diterpene Group.-Structures assigned earlier to atisine and isoatisinehave received further support from oxidative studies and partial synthesis.82Pyrolysis of diacetylatisine under mild conditions has been found to lead toa volatile fragment from which acetaldehyde 9-nitrophenylhydrazone isformed : the reaction is represented as a concerted cyclic elimination,(72) + (73).s3 The most striking result in this field has been the X-raycrystallographic determination 84 of the structure of de(oxymethy1ene)-lycoctonine (74) : lycoctonine itself is thus represented by (75), which hasbeen found to account for the known reactions of the alkaloid.85 A struc-tural relation between lycoctonine and the diterpenoid atisine group hasbeen pointed out; 86* 87 this relation, not obvious in (74), is brought out in79 K.Wiesner and 2. Valenta, Chem. and Ind., 1956, R 36.T. Ikeda, W. I. Taylor, Y. Tsuda, and S. Uyeo, zbid., p. 411; T. Ikeda, W. I.Taylor, Y . Tsuda, S. Uyeo, and H. Yajima, J., 1956, 4749.81 A. S. Bailey and C. R. Worthing, J . , 1956, 4535.82 S. W. Pelletier and W. A. Jacobs, J . Amer. Chem. SOC., 1956, 7'8, 4139, 4144.83 D. Dvornik and 0.E. Edwards, Chem. and Ind., 1956, 248.84 M. Przybylska and L. Marion, Canad. J . Chem., 1956, 34, 185.85 0. E. Edwards, L. Marion, and D. K. R. Stewart, ibid., p. 1315.R. C. Cookson and M. E. Trevett, J . , 1956, 3121.Z . Valenta and I<. WieLner, Chenz. and Ind., 1956, 354SMITH ALKALOIDS. 249(75). With (75) for lycoctonine as a basis, structure (76) has been proposedfor delpheline ; 8 6 s 88 this structure rationalises the reactions of the alkaloidincluding a very interesting oxidative degradation sequence. A study ofthe reactions of delphinine has shown that they cannot be rationalised interms of either a lycoctonine or an atisine skeleton, and, as the result of avery bold analysis of available data, structure (77) has been proposed for thealkaloid.89 Independent work on delphinine has led to the conclusion thatit must contain system (78) : Rapid this is in conflict with structure (77).( 7 2 ) (73)OAc OAcEtYO-H ..OMeOMe MeO'< 75)(74) OH OH HOsH2; OMe(76)progress is being made in the elucidation of the structure of annotinine byWiesner, Valenta, and their co-workers. That the position is still fluid isobvious from the fact that as many as three different tentative structureshave been proposed this year,g1 the latest being (79) with the carbon marked *attached to position 1, 2, or 3 (2 being preferred). A very important oxid-ation product (80), containing all but two of the carbon atoms of the alkaloid,has been synthesised as the r a ~ e m a t e .~ ~ If partial structure (79) is correct,then the substance (80) must be the product of an obscure rearrangement, forthe ring carrying the epoxide group must be broken and somehow rebuiltwith the -CHMe*CH,- bridge.Morphine Group.-The first conversion of a morphine derivative, dihydro-codeinone, into thebaine has been acc~mplished.~~ With the total synthesisof codeine and the conversion of thebaine into neopine,94 this constitutes a88 R. C. Cookson and M. E. Trevett, J., 1956, 2689, 3864.89 W. A. Jacobs and S. W. Pelletier, J . Amer. Chem. Soc., 1956, 78, 3542.90 W. Schneider, Chern. Ber., 1956, 89, 768.91 2. Valenta, F. W. Stonner, C. Bankiewicz, and K. Wiesner, J . Amer. Chem. SOC.,1956,7$, 2867; K. Wiesner, 2. Valenta, and C. Bankiewicz, Chem.and Ind., 1956, R 41;K. Wiesner, 2. Valenta, W. A. Ayer, and C. Bankiewicz, ibid., p. 1019.9a 2. Valenta, K. Wiesner, C. Bankiewicz, D. R. Henderson, and J . S. Little, ibid.,p. ~ 4 0 .93 H. Rapport, H. N. Reist, and C . H. Lovell, J . Amer. Chem. Soc., 1956,78, 8128.s4 H. Conroy, ibid., 1955, 77, 5960250 ORGANIC CHEMISTKY.formal total synthesis of the last two bases. Thermal decomposition ofnon-phenolic methine N-oxides in this group has been found to be a valuablealternative to Hofmann degradation : the olefin and NN-dimethylhydroxyl-amine are produced.95 An ingenious biogenetic scheme has been proposedto account for the formation of morphine alkaloids directly from a benzyl-isoquinoline alkaloids precur~or.~~ With the observation that all morphinealkaloids with an ether bridge to position 5 carry no oxygen function atposition 7, the suggestion has been made that ether-bridge formation involvesan allylic expulsion of an oxygen group a t position 7 ; in vitro analogies arepre~ented.~' This has been discussed further by different workers, and acyclopropanone intermediate is considered to be a possible alternative toallylic expulsion .98G.F. S.11. SUGARS.General Methods.-Some advantages are obtained by using glass-fibresheets instead of filter paper for the ionophoresis of carbohydrates in boratebuffer. Non-reducing compounds are more readily detected and the greaterelectro-endosmotic flow facilitates the separation of sugars of similar absolutemobilities. 1All of a number of aldohexoses gave, after being heated with aqueoussulphuric acid, the same final, stable ultraviolet spectrum.The ultravioletabsorption is mainly due to ether-soluble compounds which from D-glucose,D-mannose, and D-galactose were found to be formaldehyde, acetaldehyde,propaldehyde, 5-hydroxymethylfurfuraldehyde, and an unidentified com-pound.2 Similar studies with the four aldopentoses showed an identicalultraviolet spectrum from all of them : the ether-soluble products includedformaldehyde, acetaldehyde, and crotonaldehyde as well as f~rfuraldehyde.~Epimerization as a method for preparing methyl ethers of rarer sugarshaving C(2) substituted is illustrated by the preparation of 2 : 4 : 6-tri-O-methyl-D-mannose by treatment of the glucose epimer with dilute aqueousbarium hydr~xide.~Reduction of aldonolactones to the polyhydric alcohols may be carriedout by using sodium borohydride in aqueous or ethanolic solution^.^Whereas reduction of acetobromoglucose with zinc dust gives n-glucal, useof a palladium catalyst in the presence of a tertiary amine gives 1 : 5-anhydro-n-sorbitol, identical with the naturally occurring polygalitol.Hydrogeno-lysis of methyl p-L-arabopyranoside at 240"/250 atm. for about six hours inpresence of copper chromite yields mainly optically inactive forms of tetra-9 5 K. W. Bentley, J. C. Ball, and J. P. Ringe, J., 1956, 1963.96 T. Cohen, Chem. and Ind., 1956, 1391.9 7 K. W. Bentley, Experientia, 1956, 12, 251.M. Gates and G. M. K. Hughes, Chem.and Ind., 1956, 1506.1 E. J. Bourne, A. B. Foster, and P. M. Grant, J . , 1956, 4311.F. A. H. Rice and L. Fishbein, J . Amer. Chem. Soc., 1956, 78, 3731Idem, ibid., p. 1005.J N. Prentice, L. S. Cuendet, and F. Smith, ibid., p. 4439.5 H. L. Frush and H. S. Isbell, ibid., p. 2844.1,. Zervas and C. Zioudrou, J . , 1956, 214HONEYMAN : SUGARS. 25 1hydrodihydroxypyrans (1 and 2), together with 5-methoxypentane-1 : 2-and -2 : 3-dio1, and 1-methoxypentane-2 : 3-di01.~The degradation of sugars through the mercaptals and their oxidationproducts 899 has been further investigated. Oxidation of pentose mercaptalswith peroxypropionic acid in acetone or ether yields 5 : 5-dialkylsulphonyl-pent-4-ene-1 : 2 : 3-triols, which are converted by alkali into dialkyl-sulphonylmethane and aldotetrose.lo This provides a method for preparingD-erythrose from D-arabinose, alternative to the periodate oxidation of4 : 6-O-ethylidene-~-glucose.~~ Oxidation of the diethyl dithioacetals ofn-galactose and D-glucose with aqueous peroxypropionic acid gives, respec-tively, D-hb-2 : 6-epoxy-1 : 1-diethylsulphonyl-3 : 4 : 5-trihydroxyhexane(3) and the D-manno-analogue. When treated similarlyD-mannose diethyl dithioacetal gives the same compoundas D-glucose but, in addition, 1 : l-diethylsulphonyl-and Falk l3 obtained the analogous acyclic compoundby treating D-galactose diethyl dithioacetal withammonium molybdate and hydrogen peroxide but did not find the abovecyclic modification which Hough and Taylor also obtained by recrystallizingthe acyclic form.12 Perlin and his co-workers used the oxidation of hexoseswith lead tetra-acetate in acetic acid solution to prepare sugars with fewercarbon atoms.The mechanism of the stepwise reaction is uncertain but theoxidation appears to involve only the cyclic forms of the sugars. In aldoses,C(l) is eliminated, yielding the formate of the sugar with one carbon atomfewer. Finally the cyclic form of this is attacked, yielding the diformate ofthe aldose with two fewer carbon atoms. This enables D-erythrose to beobtained from ~-glucose.l~ With ketohexoses the bond between C,, and C(3)is broken, yielding glycollic acid; next, is removed to yield as the finalproduct the formate glycollate of the triose. In this way D-glyceraldehydemay be prepared from D-fructose and the enantiomer from ~-sorbose.l5 Thereaction has been applied to hexuronic acids.l6 Degradation of D-glucoseto D-arabinose is achieved by decarboxylation with bromine of the silver saltof the acetylated D-gluconic acid l7 or, better, by treating the correspondingHO H ~ ! ~ ~ ~ 0 2 E t ~ 2 ~-unanno-2 : 3 : 4 : 5 : 6-pentahydroxyhexane.12 Zinner( 3 37 H. F. Bauer and D. E. Stuetz, J . Amer. Chem. Soc., 1956, 78, 4097.8 D. L. McDonald and H. 0. L. Fischer, Biochim. Rio$hys. Acfa, 1953, 12, 203.B L. Hough and T. J. Taylor, J . , 1955, 1212.10 H. Zinner and K.-H. Falk, Chem. Ber., 1956, 89, 2451.11 C. E. Ballou, H. 0. L. Fischer, and D. L. McDonald, J . Amer. Chem. Soc., 1956,12 L. Hough and T.J. Taylor, J . , 1956, 970.13 H. Zinner and K.-H. Falk, Chem. Ber., 1955, 88, 566.14 A. S. Perlin and C . Brice, Canad. J . Chem., 1956, 34, 541.16 Idem, ibid., p . 85.16 Idem, ibid., p. 693.1: F. A . H. Rice and A. R. Johnson, J . Amer. Chem. SOC., 1956, 78, 428.77, 5967252 ORGANIC CHEMISTRY.acid chloride with silver oxide and bromine.l* The initial product of thesereactions, aldehyde-l-bromo-D-arabinose penta-acetate (as 4) may be con-verted into aldehyde-D-arabinose hexa-acetate (as 5) by treatment with silveracetate, or into the pentitol (as 6) by reduction with lithium aluminiumhydride. D-Arabinose also results from decarboxylation of D-galacturonicacid with heavy-metal ions in aqueous or pyridine s o l ~ t i o n s . ~ ~Derivatives of D-Glucosamine.-For the preparation of glycosides, 2-acet-amido-2-deoxy-~-g~ucosy~ chloride triacetate which is stable in dry solventsis particularly useful.20 A study of the acidic hydrolysis of some D-glucos-amine derivatives throws light on the different modes of hydrolysis ofheparin and hyaluronic acid.21 Preparation of 1 : 6-anhydro-~-glucosaminefrom the l-fluoro- or 1-azido-compounds is described.22 N-Acyl derivativesof D-glucosamine are obtained by treating its supersaturated methanolicsolution with the acid anhydride : fully acylated compounds result fromtreatment with acid chloride or anhydride in ~ y r i d i n e .~ ~ Deamination ofD-glucosamine hydrochloride has received further attention 24 and thestructure of the product, chitose, has been firmly established as 2 : 5-anhydro-D-mannose.The reaction follows a course similar to that of the deaminationof tra.ns-2-aminocycZohexano1, suggesting that the sugar reacts in the pyranoseform with the amino-group in the equatorial position.25Compounds obtained from Sugars and Amines, Hydrazines, etc.-Further crystalline derivatives have been isolated illustrating thatreactions of N-arylaldosylamines involve the pyranose form.26 Re-actions of aldehyde-D-galactose and -2-deoxy-~-glucose acetates withprimary aromatic amines give none. of the Schiff 's base but amorphouscompounds of the same type as ~-arabo-3 : 4 : 5 : 6-tetra-acetoxy-1 : l-di-9-toluidinohexane, RvCH(NHR'),.~' N-Glycylglucosylamine, which showsno mutarotation, is oxidized by periodate as a pyranose compound,28 butaldose isonicotinylhydrazones which mutarotate in water are acetylated inthe acyclic forms.29 The structures of the two hexa-acetyl derivatives ofD-fructose oxime (7 and 8) have been elucidated with the help of infraredl8 F.A. H. Rice and A. R. Johnson, J . Amer. Chem. SOC., 1956, 34, 3173.l9 G. Zweifel and H. Deuel, Helv. Chim. Acta, 1956, 39, 662.20 D. H. Leaback and P. G. Walker, Chem. and Ind., 1956, 1017.2 1 A. B. Foster, D. Horton, and M. Stacey, ibid., p. 175.Z 2 F. Micheel and H. Wulff, Chem. Ber., 1956, 89, 1521.23 Y. Inouye, K. Onodera, S. Kitaoka, and S. Hirano, J . Amer. Chem. Soc., 1956,24 A. B. Grant, New Zealand J . Sci. Technol., 1956, 37, 509.2 5 B. C. Bera, A. B. Foster, and M.Stacey, J . , 1956, 4531.26 J. G. Douglas and J. Honeyman, J., 1955, 3674,2 7 J. L. Barclay, A. B. Foster, and W. G. Overend, J . , 1956, 476.2 8 J. Baddiley, J. G. Buchanan, R. E. Handschumacher, and J. F. Prescott, ibid.,29 H. Zinner and W. Bock, Chem. Ber., 1956, 89, 1124.78, 4722.p. 2818HONEYMAN : SUGARS. 253spectra.30, 31 Amadori rearrangement products (1 -arylamino-1-deoxy-D-fructoses) of N-aryl-n-glucosylamines have a characteristic band at 3570 cm.-lin the infrared spectra.32 D-Fructose with anhydrous ethylamine gives N -ethyl-D-fructosylamine but this rearranges very readily, even in methanol atCH~*OAC CH2.OAcI IC = N.OAC C -NAc(OAc)II I1IAcO - C! -\ ACO-C- HH -C- OAc H-C-OAC 0H- C - OAc H- C-OAc ' ' ( 8 ) ( 7 CH2'0Ac CH 225", to the isomer which is considered to be 2-deoxy-2-ethylaniino-a-D-gluco-pyranose. Rearranged products are obtained directly in poor yield fromD-fructose and n-propylamine, n-b~tylamine,~~ or certain amino-acids.=* 35Both types of compound are isolated when benzylaniine is used.36 Therearranged products readily brown when heated and produce the odours ofcooking.Their importance in the non-enzymic browning of foods is furtherillustrated in work with simpler compounds.37Formazans 38 prepared from aldoses through phenylhydrazones have beensuggested as suitable derivatives for characteri~ation.~~ Unfortunately themelting points fall in a narrow temperature range. The structure ofD-glucosazonef ormazan ' ' (1 -phenylazo-D-glucosazone) has been proved,confirming the view that D-glucosazone has reacted as an acyclic com-pound.40 The ultraviolet absorption spectra of the necessarily acyclic3-O-methylglycerosazone is very similar to those of the sugar osazones butappreciably different from those of glyoxal and methylglyoxal osazones.Thus the sugar osazones are acyclic, the difference in their spectra from thoseof methylglyoxal being due, not to the presence of a ring, but to the oxygenatom at C(3).41 Interesting formazans have been obtained fromD-glucosone,401 42 1 : 2-O-~sopropylidene-~-xylofurano$e~~ D-xylotrihydroxy-gl~tardialdehyde?~ and periodate-oxidized cellulose (9), starch, inulin,xylan, and dextriaM The bright red product obtained from oxidizedcellulose, for example, has the structure shown (10) and may be convertedinto the colourless tetrazolium compound (11) by mild treatment with30 H.Bredereck and W. Protzer, Chem. Bey., 1954, 87, 1873.3 1 H. Rredereck, A. Wagner, D. Hummel, and H. Kreiselmeier, ibid., 1956, 89, 1532.32 F. Micheel and B. Schleppinghoff, ibid., p. 1702.33 J. F. Carson, J . Anzev. Chem. SOC., 1955, 77, 5957.34 P. H. Lowey and 13. Borsook, &id., 1956, 78, 3175.3 5 F. Micheel and A. Klemer, Chew. Bey., 1956, 89, 1238.36 J . F. Carson, J . Amev. Chem. SOC., 1956, 78, 3728.3 7 C. D. Hurd and C . M. Buess, ibid., p. 5667.3 8 G. 0. Aspinall and J. C. P. Schwarz, Ann. Repovts, 1955, 52, 257.39 L. Mester and A. Major, J . Amer. Chem. Soc., 1956, 78, 1403.4O Idem, J., 1956, 3227.4 1 J.C. P. Schwarz and (in part) M. Finnegan, ibid., p. 3979.42 G. Henseke and M. Winter, Chem. Ber., 1956, 89, 956.43 L. Mester and E. Mbczbr, J., 1956, 3228.44 L. Mester, J . Amev. Ckem. SOC., 1955, 77, 6452254 ORGANIC CHEMISTRY.N-bromosuccinimide. The reverse process is achieved with L-ascorbicacid.45 Cellulose oxidized with nitrogen dioxide also gives a bright redformazan d e r i ~ a t i v e . ~ ~- .. - .I1 I - P h N+- N P hEsters.--Whereas 2 : 3 : 4 : 6-tetra-O-acetyl-l-O-(2 : 4 : 6-trimethylbenzoy1)-p-D-glucose is converted by alkali into 1 : 6-anhydro-(3-~-glucose,~~ thea-anomer undergoes deacetylation accompanied by migration of the tri-methylbenzoyl group and yields 2-0-(2 : 4 : 6-trimethylbenzoyl)-~-glucose.~~When treated with methanolic hydrogen chloride this ester affords thep-D-glucopyranoside, Le., the bulky group makes the entering methylgroup go into the less hindered equatorial p-position.An example of thegreater reactivity of the hydroxyl group on C(21 of methyl 4 : 6-0-benzyl-idene-X-D-glucoside is also encountered : treatment of this with 2 : 4 : 6-tri-methylbenzoyl chloride in pyridine yields the 2-0-(2 : 4 : 6-trimethylbenzoyl)derivative only.48Further investigation of the compound previously considered to be theortho-acid,has shown it to be D-ribose 1 : 3 : 5-triben~oate.~~ This has also been identi-fied as the 2 : 3 : 5-triben~oate,~~ but this more recent proof shows that duringpreparation a benzoyl group migrates from C(2) to C(l>.Further orthoesterderivatives of D-fructofuranose 509 53 and of a-D-glucopyranose s4 have beendescribed.Use of complexes of the methyl D-glucopyranosides with boron oxide or acertain crystalline form of metaboric acid has enabled the acetates of the2 : 3- and 2 : 6-dibenzoates and of the 2 : 3 : 6-tribenzoate to be obtained,together with smaller quantities of other esters.55Whereas 1 : 6-anhydro-2-0-toluene-~-sulphonyl- and 1 : 6-anhydro-3 : 4-di-O-toluene-$-sulphonyl-~-D-altrose resist conversion into epoxides the3-O-toluene-~-sulphonate slowly reacts with alkali to give 1 : 6-3 : 4-di-anhydro-P-D-altrose. The conformational aspects are discussed 56 and arealso applied to the ready production of methyl 4 : 6-0-benzylidene-2 : 3-3 : 5-di-0-benzoyl-1 : 2-0-( hydroxybenzylidene)-cc-~-ribose,~~~45 L.Mester and E. M6czAr, Chem. and Ind., 1956, 848.46 Idem, ibid., p. 823.47 H. B. Wood, jun., and H. G. Fletcher, jun., J . Ameu. Chem. SOC., 1956, 78, 207.48 Idem, ibid., p. 2849.49 R. K. Ness and H. G. Fletcher, jun., ibid., 1954, 76, 1663.60 G. 0. Aspinall and J. C. P. Schwarz, Ann. Reports, 1955, 52, 259.51 R. K. Ness and H. G. Fletcher, jun., J . Amer. Chem. SOC., 1956, 78, 4710.62 F. Weygand and F. Wirth, Chem. Be?.., 1952, 85, 1000.53 R. K. Ness and H. G. Fletcher, jun., J . Amer. Chem. SOC., 1956, 78, 1001.54 R. U. Lemieux and J. D. T. Cipera, Canad. J . Chem., 1956, 34, 906.6 5 J. M. Sugihara and J. C . Petersen, J . Amel.. Chenz. SOC., 1956, 78, 1760.5 6 F. €I. Newth, J . , 1056, 441HONEYMAN : SUGARS.255didehydro-2 : 3-dideoxy-a-u-glucoside by treatment of methyl 4 : 6-O-benzyl-idene-3-deoxy-3-iodo-2-O-toluene-~-sulphon~l-~-~-glucoside with sodiumiodide in acetone.57The preparation and reactions of a number of nitrate esters have beendescribed, including selective conversion of secondary nitrates into thecorresponding alcohol by their reaction with aqueous sodium nitrite. 58The reductive removal of nitrate groups with hydrazine gives high yields ofthe alcohols,59 but catalytic hydrogenation with Raney nickel, or treatmentwith methylmagnesium iodide or lithium aluminium hydride only partiallydenitrates cellulose nitrate.60Benzenesulphinates, prepared by treating a sugar derivative or its acetatewith sodium benzenesulphinate and acetic anhydride, are hydrolysed by" acyl "-oxygen fission, i.e., without inversion or epoxide formation.5sSome S-methyl dithiocarbonates (xanthates) of D-glucose have beenprepared by treating the sodium alkoxide with carbon disulphide followedby methyl iodide : 61RONa + CS, ROCS,Na 4 RO*CS,MeMethyl 3 : 4-O-isopropylidene-p-~-arabinoside %O-(S-methyl dithiocarb-onate) gives on distillation, not the expected olefin,62 but the isomeric%S-(S-methyl dithiocarbonate) which is converted by reductive desulphur-ization into the 2-deoxy-D-ribose derivative in low yield.63 Methyl 3 : 4-isopropylidene-p-D-arabinoside 2-O-(S-sodium dithiocarbonate) gives theS-methyl and S-triphenylmethyl derivatives normally but with ethyliodide, n- or iso-propyl bromide, or tert.-butyl chloride the product is di-(methyl 3 : 4-O-isopropylidene-p-~-arabinoside) %thionocarbonate, also ob-tained by reaction of thiocarbonyl chloride and the D-arabinose com-pound.640ligosaccharides.-Full details are given of the chemical synthesis ofsucrose together with a discussion of the mode ofCHz'oAc reaction with alcohols of 1 : 2-anhydro-x-u-glucopyr-AcO I\ anose triacetate in its half-chair conforrnatibn~.~~ Withmethanol the p-D-glucoside is obtained, but with bulkyalcohols, as in the sucrose synthesis, the chief productis the a-D-glucoside because the axial CH,*OAc group OAc Y T g,prevents normal trans-addition (12).Stevioside 66 is shown to be a s~phoroside.~~Oxidation of disaccharides by lead tetra-acetate enables the position of57 F.H. Newth, J . , 1956, 471.58 J. Honeyman and J. W. W. Morgan, J., 1956, 3660.59 K. S. Ennor, J . Honeyman, and T. C. Stening, Chem. and Ind., 1956, 1308.E. P. Swan and L. D. Hayward, Canad. J . Chern., 1956, 34, 856.A. K. Sanyal and C. B. Purves, Canad. J . Chem., 1956, 34, 426.62 L. Chugaev, Ber., 1899, 32, 3332.63 M. L. Wolfrom and A. B. Foster, J . 4?)8er. Chem. Soc , 1956, 78, 1399.-4. B. Foster and M. L. Wolfrom, zbid., p. 2493.65 R. U. Lemieux and G. Huber, zbid., 4117.G. 0. Aspinall and J. C. P. Schwarz, Ann. Repovfs, 1955, 52, 258.G i E. Vis and 11, G . Flttclier, jun., J . ,411zcr. Chrnz. Soc., 1956, 78, 470:)256 ORGANIC CHEMISTRY.the linkage between the monosaccharide units to be determined.68* 69 Theoxidation of trisaccharides and higher polymers has also been studied.70The degree of polymerization of reducing oligosaccharides with up to aboutseven monosaccharide units may be found by use of the anthrone-concen-trated sulphuric acid reagent before and after reduction of the sugar withsodium borohydride. 71" Hydrol," the residual mother-liquor from the manufacture of D-ghCOSeby the acid hydrolysis of maize starch, has been examined and 5-O-P-D-glucopyranosyl-D-glucose has been isolated.This is believed to be formedas a reversion product.72 Leucrose, a by-product of the preparation ofbacterial dextran from sucrose, is shown to be 5-O-a-~-glucopyranosyl-D-fructose. 73Miscellaneous.-Phosphorus pentachloride reacts with 1 : 2-5 : 6-di-0-kopropylidene-D-glucose to give, in low yield, not the expected 3-chloro-3-deoxy-derivative, 74 but the 6-chloro-6-deoxy-isomer, the reaction involvingmigration of an isopropylideneCarboxymethyl ethers of n-glucose have been prepared and correlatedwith the products obtained by hydrolysis of O-carboxymethylcellulose. 76Reduction of these D-glucose ethers by lithium aluminium hydride yields thecorresponding hydroxyet hyl compounds.Separations of D-glucose, D-fructose, and sucrose, and of a number ofpartially methylated derivatives of D-glucose and D-xylose, have shown that" Celite " columns have advantages for preparativePreparations of 5-O-methyl-~-glucose 79 (and %O-methyl-~-glyceron-amide from it) and of 2 : 5 : 6-tri-O-methyl-~-glucosePure, but nevertheless non-crystalline, sedoheptulose (D-altroheptulose)has been obtained from its crystalline hexa-acetate.At 20" in dilute acidthe equilibrium solution contains the sugar and 91 % of sedoheptulosan(2 : 7-anhydro-p-~-nltroheptulopyranose). 81 The isomeric anhydro-com-pound 82 has now been crystallized and shown to be 2 : 7-anhydro-P-D-altroheptulofuranose. 81Acetone, like aldehydes, has been found to react with methyl X-D-glucoside to give the 4 : 6-O-isopropylidene compound.83 D-Ribose withacetone yields 1 : 5-anhydro-2 : 3-O-isopropylidene-~-ribofuranose and 2 : 3-O-isopropylidene-D-ribose. The latter yields two isomeric 1 : 5-O-benzyl-are described.6 8 A. S. Perlin, Analyt. Chew., 1955, 27, 306.69 A.J. Charlson and A. S. Perlin, Canad. J . Chew., 1956, 34, 1200.70 A. S. Perlin and A. R. Lansdown, ibid., p. 451.?l S. Peat, W. J. Whelan, and J - G. Roberts, J., 1956, 2258.72 J. C. Sowden and A. S. Spriggs, J . Amer. Chem. Soc., 1956, 78, 2503.7 3 F. H. Stodola, E. S. Sharpe, and H. J. Koepsell, ibid., p. 2514.74 J. B. Allison and R. M. Hixon, ibid., 1926, 48, 406.'5 D. C. C. Smith, J . , 1956, 1244.76 W. P. Shyluk and T. E. Timell, Caaad. J . Chem., 1956, 34, 575.7 7 Idem, ibid., p. 671.78 R. U. Lemieux, C. T. Bishop, and G. E. Pelletier, ibid., p. 1365.79 J. K. N. Jones, ibid., p. 310.C. T. Bishop and J. Schmorak, ibid., p. 845.81 N. K. Richtmyer and J. W. Pratt, J . APner. Chew. SOG., 1966, 78, 4717.82 L. P. Zill and N. E. Tolbert, ibid., 1954, 76, 2929.83 J.K. N. Jones, Canad. J , Chew., 1956, 34, 840SMITH : NATURAL MACROMOLECULES. 257idene derivatives. 84 Methyl D-ribopyranoside condenses with acetone togive a mixture of methyl 2 : 3-O-~sopropylidene-~-ribofuranoside and methyl3 : 4-O-isopropylidene-D-ribopyranoside. The conversion of the pyranoseinto the furanose is believed to occur after condensation with acetone onJ. H.12. NATURAL MACROMOLECULES.Po1ysaccharides.-Since last year's Report, fractionation of naturallyoccurring polysaccharides has continued to attract attention. Fractionalprecipitation with cupric acetate and ethanol has resolved linseed mucilageinto three distinct components.l Acidic polysaccharides form precipitateswith cetyltrimethylammonium bromide which redissolve in salt solutionsand can be fractionated by this rneans.2,3 In the presence of oxalic acid,which suppresses dissociation of carboxyl groups, only polysaccharides con-taining sulphate groups are pre~ipitated,~ and cliff erential estimation ofcarboxyl and sulphate groups on a microgram scale by spectrophotometricestimation of the excess of cetylpyridinium chloride has been des~ribed.~A globulin from Jack-bean meal forms precipitates with glycogens, yeastmannan, and heparin but not with amylopectin, chondroitin sulphate,hyaluronate, or lung gala~tan.~ Gum arabic recovered from its specificprecipitate with type I1 antipneumococcus serum contains only one- thirdto one-fifth as much rhamnose as does the native gum.6 Separation ofamylose and amylopectin by paper ionophoresis in borate solution is hinderedby adsorption of polysaccharides on paper : adsorption is very much lesswhen glass-fibre sheets are used in place of paper,* and detection of thepolysaccharides is ~implified.~A technique for determining the position of 1 : 6-linkages in aldohexosepolymers otherwise containing only 1 : 2-, 1 : 3-, or 1 : 4-linkages utilisesdegradation from the reducing end-groups by buffered periodate at pH 8.- 0- CH,I 1 CHOThe process oxidises away anhydrohexose residues until a 6-linked residue isencountered, and the amount of formaldehyde released indicates the positionof the 1 : 6-linkage relative to the reducing end-group.Clinical dextran84 G. R. Barker and J. W. Spoors, J., 1956, 1192.8 5 Idem, ibid., p.2656.1 A. J. Ersltine and J . K. N. Jones, Canad. J . Chem., 1956,34, 821.3 J. E. Scott, ibid., 1955, 18, 428.A. S. Jones, Biochim. Bioplzys. Acta, 1953, 10, 607.Idem, Chem. and I n d l 1955, 168.J. A. Cifonelli, R. Montgomery, andF. Smith, J . Amer. Chem. SOL, 1956,7$, 2488.M. Heidelberger, J . Adams, and 2. Dische, ibid., 1956, 78, 2853.7 A. B. Foster, P. A. Newton-Hearn, and M. Stacey, J., 1956, 30.E. J. Bourne, A. B. Foster, and P. M. Grant, J., 1956, 4311.D. R. Briggs, E. F. Garner, and F. Smith, Nature, 1966, 173, 164.REP .-VOL. LIII 258 ORGANIC CHEMISTRY ayields no formaldehyde, suggesting lo that the reducing end-group is linkedthrough position 6. Formic acid released in periodate oxidation measuresthe number of 1 : 6-linked residues in the dextran of Leucofiostocmesenteroides.ll Methylation studies l2 on a bacterial dextran indicatepredominantly 1 : 6-linked residues with 1 : 3-linked branches.Partial hydrolysis of polysaccharides to oligosaccharides grows in im-portance as a means of determining structure.A simple technique forisolating disaccharides from partial hydrolysates by adsorption on charcoalis described.13 Care must be taken to ensure that disaccharides are notformed by reversion. Reversion of arabinose gives 3-0- and 4-O-p-arabino-pyranosyl-L-arabinose and another disaccharide ; reversion of xylose gives4-O-~-~-xylopyranosyl-~-xylose and two other disaccharides ; and reversionof mannose gives 3-0- and 6-0-cc-D-mannopyranosy~-~-mannopyranose.~~Treatment of D-arabinose with sulphuric acid gives rise to D-arabinopyrano-syl-~-arabinopyranoside.l~ Condensation of xylose catalysed by hydrogenchloride yields a polysaccharide l6 containing pyranose residues with 1 : 4-,1 : 2-, and 1 : 3-linkages in the ratio of 13 : 6 : 2.Glucose monohydrate,when heated with Amberlite IR-l20(H+), yields a polysaccharide containing60% of a-1 : 6-1inkages.l' Use of the Koenigs-Knorr reaction for controlledsynthesis of gentiodextrins is reported ; 2 : 3 : 4-tri-~-acetyl-a-D-gluco-pyranosyl bromide with silver oxide, iodine, and a dehydrating agentyielded a mixture containing levoglucosan, and the polymeric series ofoligosaccharides containing the p-1 : 6-linkage.Polysaccharides associated with wood cellulose have been reviewed.lgAs an alternative to sodium hydroxide, liquid ammonia 20 and dimethylsulphoxide have been used for extracting hemicellulose from holocellulose.The use of alkali in isolating polysaccharides, while not invalidating thestructures assigned, may profoundly affect biological activity and variousmethods of chain-length determination if a terminal saccharinic acid residueis formed.22 Maize-cob xylan is stable to oxygen-free alkali, presumablybecause the terminal reducing groups were oxidised by sodium chloriteduring delignification.22Xylans containing variously arabinose, galactose, glucuronic, and 4-0-methylglucuronic acid residues from wheat straw,23* 24 wheat bran,25 oat10 L.Hough and M.B. Perry, Chem. and Ind., 1956, 768.11 R. J. Dimler, I. A. Wolff, J. W. Sloan, and C . E. Rist, J . Anzer. Chem. SOL., 1955,12 J. W. Van Cleve, W. C . Schaefer, and C . E. Rist, ibid., 1956, 78, 4435.13 P. Andrews, L. Hough, and D. B. Powell, Chem. and Ind., 1956, 658.14 D. H. Ball, J. K. N. Jones, W. H. Nicholson, and T. J. Painter, T A P P I , 1956,1 5 F. A. H. Rice, J . Amer. Chem. SOC., 1956, 78, 6167.1 6 C. T. Bishop, Canad. J . Chem., 1956, 34, 1255.1 7 P. S. O'Colla and E. Lee, Chem. und Ind., 1956, 522.1 8 S. Haq and W. J. 'Whelan, .J., 1956, 4543.19 W. J. Polglase, Adv. Carbohydrate Chem., 1955, 10, 283.20 J. E. Milks and C. B. Purves, J - Amer. Chem. Soc., 1956, 78, 3738.21 E. Hagglund, B. Lindberg, and J. McPherson, Acta Chew. Scand., 1956,10, 1160.22 R.L. Whistler and W. M. Corbett, J . Amer. Chem. SOC., 1956, 78, 1003.23 G. 0. Aspinall and E. G. Meek, J . , 1956, 3830.24 C. T. Bishop, J. Amer. Chem. Soc., 1956, 78, 2840.z 5 G. A. Adams and C . T. Bishop, ihid., p. 2842.77, 6568.39, 438SMITH : NATURAL MACKOMO1,ECULES. 259straw,26 oat hull,27 maize fibre,"> 29, 30 maize hull,319 3y tamarind andwoods of aspen,2* western hemlock,343 35, 3G and Norway spruce 37 have beenexamined. Methylation studies 23* 26 and the nature of the oligosaccharidesand aldobiuronic and aldotriuronic acids isolated accord with the patternsof structure already sumrnarised,38 though the isolation of a crystallineL-galactopyranosyl-( 1 __t 4) -D-xylopyranosyl-( 1 + 2) -L-arabinose frommaize-fibre hemicellulose is the first indication of such a combination ofsugars in Nature.28 m-Galactose occurs as non-reducing end groups 30 inmaize-fibre gum.Glucomannans have been isolated from western heml0ck,~6lily and AmorPhoPhaZZzGs plants.40 Hemicellulose of plum leaf giveson hydrolysis D-galactose, L-arabinose, D-xylose, L-rhamnose, L-fucose,D-mannose, 2-O-methylxylose, and one other O-methylmonosaccharide ;this is the first discovery of an O-methylpentose occurring in Nature.13* 41Polysaccharides from algz continue to present novel structural features ;L-guluronic acid has been reported for the first time as a natural product inbrown a l p ; 42 and the polysaccharide from a red seaweed contains 6-0-methyl-D-galactose, also previously unknown among natural products, andboth enantiomers of galactose, which incidentally were separable by crystal-lisat ion .43Gums of Khaya grandifoZia,@ Hakea acic~laris,~5 and lemon *6 have beeninvestigated. A polyglucose of Zen mays has been found indistinguishablefrom glycogen of animal and microbial origin.47 Additional evidence fromenzymic degradation indicates that the molecule of p-limit dextrin and henceof the parent amylopectin is multiple branched on the random patternsuggested by K. H.I~Ieyer.~~ Further work on the periodate oxidation ofamylopectin suggests that 1 : 3-linkages are present,49 and this is confirmedby the isolation of nigerose (3-0-a-~-glucopyranosyl-~-glucose) on partialhydrolysis of amylopectin under conditions unfavourable to reversion ; 5 0 ~ 51t 6 G.0. Aspinall and K. C. B. Wilkie, J . , 1956, 1072.27 E. L. Falconer and G. A. Adams, Canad. J . Chem., 1956, 34, 338.28 R. L. Whistler and W. M. Corbett, J . Amer. Chew. SOC., 1955, 77, 6328.2B Idem, J . Org. Chem., 1956, 21, 694.30 R. L. Whistler and J. N. B. Miller, J . Arner. Chem SOC., 1956, 78, 1163.31 R. Montgomery, F. Smith, and H. C. Srivastava, ibid., p. 2837.32 Idem, ibid., p. 6169.33 G. R. Savur, J., 1956, 2600.34 G. G. S. Dutton and F. Smith, J . .41ner. Chem. SOC., 1956, 78, 2506.3G Idem, ibid., p. 3744.36 J. K. Hamilton, H. W. Kircher, and N. S. Thompson, ibid., p. 2508.37 G. 0. Aspinall and M. E. Carter, J., 1956, 3744G. 0. Aspinall, Ann. Reports, 1955, 52, 261.P. Andrews, L. Hough, and J.K. N. Jones, J., 1956, 181.40 F. Smith and M. C. Srivastava, J . Amer. Chern. SOC., 1956, 78, 1404.41 P. Andrews and L. Hough, Chem. and Ind., 1956, 1278.42 F. G. Fischer and H. Dorfel, 2. physiol. Chem., 1955, 302, 186.43 J. R. Nunn and M. M. von Holdt, Chem. and Ind., 1956, 467.44 G. 0. Aspinall, E. L. Hirst, and N. K. Matheson, J . , 1956, 989.4 6 A. M. Stephen, J., 1956, 4487.46 G. G. S. Dutton, Canad. J . Chem., 1956, 34, 406.4 7 S. Peat, W. J. Whelan, and J. R. Turvey, J., 1956, 2317.48 S. Peat, ' A T . J . Whelan, and G. J . Thomas, J . , 1956, 3025.49 J. K. Hamilton and F. Smith, J . Amer. Chem. SOC., 1956, 78, 5910.so M. L. Wolfrom and A. Thompson, ibid., 1955, 77, 6403.Idem, ibid., 1956, 78, 4116260 ORGANIC CHEMISTRY.on the other hand, it is suggested that 1 : 3-linkages may be artefacts ofreversion since, otherwise, new enzymes must be po~tulated.~~ Partialhydrolysis of glycogen has yielded a small amount of isomaltotriose, suggest-ing that some of the 1 : 6-linkages exist on adjacent units.53The termination of fructosan chains by sucrose linkages is further con-firmed by the isolation of kestose and of inulobiosylsucrose from partial acidhydrolysates of inulin under conditions where the formation of reversionproducts such as 6-O-~-~-fructosylfructose is negligible.54 A structure hasbeen proposed for the glucofructan of Cordyline terminaZis. 55 Partialhydrolysis of sugar-beet araban yields 5-O-~-arabinofuranosyl-~-arabino-furanose and 3-0-~-arabinofuranosyl-~-arabinose. l3 It is confirmed bymethyiation that the polysaccharide luteose, produced as its malonic esterby Penicillium Zzdeztm Zukal., is a p-1 : 6-linked glucosan with many 1 : 3-and 1 : 4-linked branches.56 The extracellular polysaccharide of A erobacteraerogenes contains 1 : 4-linked glucose, 1 : 2-linked L-fucofuranose, andglucuronic acid units.67 The chemistry of heparin has been reviewed.5sAn aldoheptose, D-glycero-D-galactoheptose, and a new amino-sugar,C,H,,O,N,H,O, possibly a 3-O-carboxyethylhexosamine, occur in thespecific polysaccharides of Chromobacteriwm violaceum.59 An unidentifiedheptose is the main sugar component of a lipopolysaccharide from Pasteurellapestis. 6ONucleic Acids.-l955 saw the publication of a comprehensive , two-volumetreatise on nucleic acids,61 and the subject has been reviewed severalWith the establishment of the 3’ : 5’-phosphodiester structure of nucleicacids, further work has been mainly directed towards synthesis of oligo-nucleotides, structure in relation to enzyme specificity, degradative workwhich might indicate the sequence of base residues, study of hydrogen bondinteractions between base residues, and indications of the mode of combin-ation between nucleic acids and proteins.Details have been given of the synthesis of thyminenucleosides using a mercury derivative of thymine and poly-O-benzoyl-glycosyl halides.63 Adenylic and guanylic acids-a and -b have been identi-fied by acid hydrolysis to ribose 2’- and S’-phosphate respectively. Similarhydrolysis of pyrimidine nucleotides is precluded by the stability of theirglycosidic linkages,64 but proof of the structures of cytidylic acids-a and -bRibonzccleic acids.52 G.T. Cori, Makrornol. Chem., 1956, 20, 169.63 M. L. Wolfrom and A. Thompson, J . Amer. Chem. Soc., 1956, 78, 4182.54 D. S. Feingold and G. Avigad, Biochim. Biophys. Actu, 1956, 22, 196.6 5 L. A. Boggs and F. Smith, J . Amer. Chem. Soc., 1956, 78, 1880.66 P. F. Lloyd, G. Pon, and M. Stacey, Chem. and Ind., 1956, 172.5 7 G. 0. Aspinall, R. S. P. Jamieson, and J. F. Wilkinson, J., 1956, 3483.59 M. J. Crumpton and D. A. L. Davies, Baochenz. J., 1956, 64, 2 2 ~ ; R. E. Strange,6o D. A. L. Davies, ibid., 1956, 63, 105.61 E. Chargaff and J. N. Davidson, ‘’ The Nucleic hicds,” Academic Press, New62 Sir Alexander Todd in “ Perspectives in Organic Chemistry,” Interscience Publ.63 J.J. Fox, N. Yung, J. Davoll, and G. B. Brown, J . Amer. Chem. Soc., 1956,A. B. Foster and A. J. Huggard, Adv. Carbohydrate Chew.., 1955, 10, 335.ibid., p. 2 3 ~ ; A. P. Maclennan and D. A. L. Davies, 1956, 63, 3 1 ~ .York, 1955.Inc., New York, 1956; Makromol. Chem., 1956, 20, 87 ; Cht~% and Ind., 1956* 802-78, 2117.J. X. Khym and W. E. Cohn, ibid., 1954, 76, 1818, 5523SMITH : NATURAL MACROMOLECULES. 261as the 2‘- and 3’-phosphate respectively has now been obtained by degradingthese pyrimidine nucleotides with hydrazine hydrate, pyrazolone beingobtained with ribose 2- and 3-phosphate respectively. By this same treat-ment uridylic acid-b was converted into 3-aminopyrazole and ribose 3-phos-hate.^^ The ribose phosphates have also been obtained from both uridylicand cytidylic acid via the dihydrouridylic acids which were hydrolysed bydilute alkali to the N-ribosyl phosphates of p-ureidopropionic acid which, inturn, were hydrolysed by dilute acid to the ribose phosphates withoutappreciable migration of the phosphate group66The structures of uridylic acid-a and -b have also been confirmed bysynthesis of the former acid from 3’ : 5’-di-O-acetyluridine, the structure ofwhich was proved by conversion into 3’ : 5’-di-O-acetyl-Z‘-O-toluene-$-sulphonyluridine (1) and O2 : Z’-cycZouridine (2) ; this on hydroly~is,~~ gave3-P-u-arabofuranosyluracil (3), identical with spongouridine isolated fromsponges? This sequence of reactions parallels the conversion of 3’ : 5’-di-0-methanesulphonylthymidine into 5’-0-methanesulphonyl-O2 : 3’-cyclo-thymidine. 69AcO OTs AcO H HO H(Ts = p-C,H,Me-S02)An improved preparation of the crystalline 3’ : 5’-di-@acetyladenosineconsists in fusing equimolar quantities of 5’-0-acetyladenosine and 2’ : 3‘ : 5‘-tri-0-acetyladenosine (resulting in transacetylation), followed by separationof the di-0-acetyl compound from starting material.By starting from thissubstance, adenosine-2’ uridine-5’ phosphate has been synthesised.70 Inview of the labilising effect of the 3‘-hydroxyl of adenosine on the inter-nucleotide linkage the attainment of this dinucleosideI t phosphate by synthesis represents a notable achieve-ment. 0-Benzylphosphorous 00-diphenylphosphoricanhydride converted 3‘ : 5’-O-acetyladenosine into3’ : 5’-O-acetyladenosine-2’ benzyl phosphite, which1 1 with N-chlorosuccinimide yielded 3’ : fi‘-O-acetyl-I i adenosine-2’ benzyl phosphorochloridate. This,when condensed with 2’ : 3’-O-acetyluridine, gave amixture from which the product (4) was separated after removal of theprotecting groups.Conditions for survival of the internucleotide linkageduring removal of the protecting groups are that (a) the benzyl group mustbe removed by hydrogenolysis before the adjacent free 3’-hydroxyl is pro-OAc OAcAdenine - C ,- C3,- Cs,1 2O\ P”OPh. CH,. 6 \U r a c i I - c2 - C’ - c5‘OAc OAc 4 )66 F. Baron and D. M. Brown, J., 1955, 2855.W.E. Cohn and D. G. Doherty, J . Amcr. Chem. SOC., 1966, 78. 2863.6 7 D. M. Brown, Sir Alexander Todd, and S. Varadarajan, J., 1956, 2388.68 W. Bergman and D. C. Burke, J. Org. Chem., 1955, 20, 1501.69 A. M. Michelson and Sir Alexander Todd, J., 1955, 816.70 A. M. Michelson, L. Szab6, and Sir Alexander Todd, J., 1966, 1546262 ORGANIC CHEMISTRY.duced and (b) conditions for removal of the acetyl groups must be sufficientlymild (pH 9.6). Partial acid hydrolysis of cytidine-3’ benzyl phosphate ispreceded by some migration of the phosphate group, since 25% of therecovered diester consists of cytidine-2’ benzyl phosphate. Also, adenos-ine-2’ uridine-5’ phosphate gives some adenosine-3’ uridine-5’ phosphate.Hence a symmetrical transition state, probably a protonated form of (5Aor B) is involved.By contrast, partial alkaline hydrolysis of cytidine-3’benzyl phosphate is not accompanied by migration of the phosphate group,suggesting that there is an unsymmetrical transition state (6). Hence, ifpartial acid hydrolysis of ribonucleic acid gives rise to oligonucleotides con-taining a ribonuclease-resistant 2‘ : 5’-linkage, this will not be evidence forsuch a linkage in ribonucleic acids.’l The properties of uridine-3’ dimethyland dibenzyl phosphate make it unlikely that there are any phosphotriestergroups in ribonucleic acid.72The reversibility of ribonuclease action enables this enzyme to synthesisepyrimidine nucleoside-3’ methyl and ethyl phosphate from either pyrimidinenucleoside-2’ : 3’ cyclic phosphates 73* 74 or pyrimidine nucleoside-3’ benzylphosphates 73 and methyl or ethyl alcohol.An analogous reverse reactioninduced by spleen phosphodiesterase is the replacement of the benzyl groupsin cytidine-3’ benzyl phosphate or adenosine-3’ benzyl phosphate by amethyl or an ethyl group. This enzyme differs from ribonuclease in itsability to utilise the purine nucleoside benzyl phosphate and in its inabilityto utilise nucleoside-2’ : 3‘ cyclic phosphate.73 The ability to act as phosphateacceptors in ribonuclease-catalysed exchange is limited to primary alcohols. 74Cytidine-2’ : 3’ cyclic phosphate can act as an acceptor, leading to thesynthesis of di- and tri-nucleotides of cytidylic acid having a terminal2’ : 3’-cyclic phosphate group.Cytidine can also act as the acceptor, lead-ing to the dinucleoside phosphate and trinucleoside diphosphate. Adenosineand adenosine-2’ : 3’ cyclic phosphate can also act as acceptors, thoughnucleoside-3’ phosphates are very poor acceptors. With ribonuclease underfavourable conditions the synthesis of 3’ : 5’-internucleotide linkages fromnucleoside-2’ : 3’ cyclic phosphates exceeds the hydrolysis to nucleoside-3’phosphates. The reactions are essentially traiisesterifications and neitherribonuclease nor spleen phosphodiesterase is capable unaided of synthesisingphosphodiester linkages from nucleoside-3’phosphates. 75 However , pyrim-idine nucleoside-2’ : 3’ cyclic phosphates are produced in the early stages71 D. M.Brown, D. I. Magrath, A. H. Neilson, and Sir Alexander Todd, Nature,72 D. M. Brown, D. I. Magrath, and Sir Alexander Todd, J., 1955, 4396.73 L. A. Heppel and P. R. Whitfield, Biochem. J., 1955, 60, 1.74 G. R. Barker and M. A. Parsons, Chem. and Ind., 1955, 1009.7 5 L. A. Heppel, P. R. Whitfield, and R. Markham, Biochem. J., 1955, 60, 8 ; M.Holden and N. W. Pirie, abid., p. 39; W. S. Pierpoint, BiocAim. Biophys. Acta, 1966,21, 136.1956,177, 1124SMITH NATURAL MACROMOLECULES. 263of ribonuclease digestion of ribonucleic acid, and Heppel et ul. point out thatpart at least of the dinucleotides with terminal-2’ : 3’ cyclic phosphategroups released under these conditions could be synthetic. Alkaline degrad-ation of ribonucleic acid with sodium tert.-butoxide yields nucleoside-2’ : 3’cyclic phogphates in 65% yield.76The suggestion has been made that one of the active centres in enzymescapable of transferring nucleotide or phosphate groups is the glyoxalinenucleus of histidine ; synthetic 1-phosphoglyoxaline and its esters have beenprepared which can transfer phosphate groups to alcohols, amines, carboxylicacids phosphoric esters, and inorganic phosphate.77Perhaps the most significant advance in the last two years was the dis-covery of an enzyme (polynucleotide phosphorylase) capable of synthesisinghighly polymerised ribonucleic acid-like polynucleotides from nucleoside-5‘pyrophosphates with the release of orthophosphate. The enzyme alsocatalyses the phosphorolysis of polynucleotides to yield nucleoside 5’-pyro-phosphates.Uniform and mixed synthetic polynucleotides have beenprepared. 78 Polyadenylic and polyuridylic acid prepared in this way havealready been used, with striking results, in physicochemical studies : thetwo combine immediately when mixed, with an increase in viscosity and adecrease in optical density at 260 mp. The product forms tough glassyfibres which have been shown by X-ray diffraction to possess a two-strandhelical structure with a helical pitch of 32-36 A, and about ten nucleotideresidues per turn. These results show for the first time that it is possiblefor ribopolynucleotides to assume a configuration similar to that found inundenatured deoxyribonucleic acid. 79 For spontaneous formation of thedouble helix it is probably necessary that each polynucleotide should containonly one type of base residue.There is not much evidence available to showwhether any two-strand helical structures are present in naturally occurringribonucleic acid, but a ribonucleic acid isolated by mild procedures fromA erobacter aerogenes gives a titration curve which is irreversible toward acidand alkali to an extent of 25-30% of that shown by undenatured deoxy-ribonucleic acid. The authors suggest that this is because ribonucleic acidowes part of its structure to hydrogen bonds between the 6-amino- and6-keto-groups of the base residues.s0 Polyadenylic acid has also been usedin experiments with an enzyme preparation from liver nuclei which degradesit to adenosine and a mixture of oligonucleotides in which mono-, di-, trGJand tetra-nucleotides have been identified, all having a terminal 5’-phosphategroup.81Chromatographic separation of the dinucleotides released after brief acidhydrolysis of ribonucleic acid indicates that all the sixteen possible sequences76 D.Lipkin and P. T. Talbert, Chem. and Ind., 1955, 143.77 J. Baddiley, J. G. Buchanan, and R. Letters, J., 1956, 2812.78 M. Grunberg-Manago, P. J. Ortiz, and S. Ochoa, Science, 1955,122,907; Biochim.7B A. Rich and D. R. Davies, J . Amer. Chem. SOC., 1956, 78, 3548; R. C. Warner,Biophys. Acta, 1956, 20, 269.Fed. Proc., 1956. 15, 379.R. A. Cox; A:S. Jones, G. E. Marsh, and A. R. Peacocke, Biochim. Biophys. Acta,L. A. Heppel, P. J. Ortiz, and S. OChQa, Science, 1956, 123, 416.1956, 21, 676264 ORGANIC CHEMISTRY.of dinucleotides are to be found.82 Full details have been published of amethod suitable for stepwise degradation of homogeneous polyribonucleo-tides.83 The most serious obstacle to determination of base sequences isthat ribonucleic acids occur as complex mixtures of moleculesu Difficultyof establishing complete purity hinders analytical work aimed at establishingregularities of base composition.It has been pointed out that ribonucleicacid isolated under mild conditions usudy has a low phosphorus contentand does not give a theoretical yield of mononucleotides on chromatographyof alkaline hydrolysates on anion-exchangers. Additional products of thehydrolysis have now been shown to be recoverable from the anion-exchangeron elution with hydrochloric acid, and to yield on drastic acid hydrolysis theamino-acids, glutamic acid , threonine, valine, serine, alanine, glycine, andleucine or isoleucine, together with adenine and guanine.The amino acid-nucleotide compounds give no ninhydrin reaction and a phosphoamidestructure is tentatively proposed. Their amino-acid : nucleotide ratio isunity.85 The attachment of amino-acids to ribonucleic acid by mixedanhydride linkages has been suggested as a possible stage in peptidesynthesis. G2Deoxyyibonucleic acid. That the sugar of deoxyadenosine , deoxy-guanosine, thymidine, deoxycytidine, and deoxy-5-methylcytidine is%deoxy-~-ribose has been confirmed bypreparation of the benzylphenylhydrazone.863’-Deoxy-3’- and 5’-deoxy-5’-iodothymidine,heated with silver acetate in acetonitrile0 N+ containing a trace of base, yielded halogen- ‘73 free compounds formulated as O2 : 3’-cycZo-thymidine (7) and O2 : 5’-cycZothymidine (8).On hydrolysis, the former gives 3-deoxy-~-xylose whereas the latter gives 2-deoxy-~-ribose.This establishes thestructure of thymidine as 3-p-2’-deoxy-~-ribofuranosylthymine, a conclusionwhich was confirmed by X-ray analysis of 5’-bromo-5’-deoxythyidine. 87The 3‘- and 5’-phosphates of deoxyadenosine and deoxyguanosine 88 andthe 5‘-phosphate of thymidine have been synthesised by unambiguousmethods. The intermediate 3’- and 5’-O-acetylnucleosides were identifiedby conversion into 3’- and 5‘-0-acetyl-~-ribose, respectively, by mild acidhydrolysis. The synthetic 5’-phosphates were identical with the productsobtained in enzymic degradation of deoxyribonucleic acids.Rattlesnakevenom only dephosphorylates the 5’-phosphates. 88The dinucleoside phosphate and the dinucleotide of thymidine having thenaturally occurring terminal 5’-phosphate group have been synthesised. 895’-O-Acetylthyniidine was treated with 0-benzylphosphorous 00-diphenyl-phosphoric anhydride, then with N-chlorosuccinimide, yielding the corre-HO HHO*H,C Hf--& H;aMe(8) 0-TdMe( 7 ) 0-82 W. E. Cohn and R. Markham, Biochcwz. J., 1956, 62, 1 7 ~ .89 D. M. Brown, M. Fried, and Sir Alexander Todd, J., 1955, 2206.84 R. Markham, Biochem. J., 1956, 62, 3 9 ~ .8 5 J. L.Potter and A. L. Dounce, J . Amer. Chem. SOC., 1956, 78, 3078.8 6 I. G. Walker and G. C. Butler, Canad. J . Chem., 1956, 34, 1168.87 A. M. Michelson and Sir Alexander Todd, J., 1955, 816.89 A. M. Michelson and Sir Alexander Todd, J . , 1966, 2632.D. H. Hayes, A. M. Michelson, and Sir Alexander Todd, J., 1955, 808SMITH NATURAL MACROMOLECULES. 265sponding 3’-(benzyl phosphorochloridate) ; this was condensed with 3’-0-acetylthymidine and freed from protecting groups by treatment with acidand alkali and by hydrogenolysis. The resulting dinucleoside phosphate(T3‘-P-5’T according to the symbolism proposed by these authors) wasseparated from some dinucleoside pyrophosphate (T3’-P-P-3’T). Simi-larly, starting with 5’-(dibenzyl phosphoryl) thymine, the dinucleotide(T5’-P-3’T5’-P) was synthesised, accompanied by some dinucleotide pyro-phosphate (P-5’T3‘-P-P-3’T5’-P).Apurinic acids, the non-dialysable products of carefully controlled acid-hydrolysis of deoxyribonucleic acids with the removal of most of the purineresidues and negligible loss of phosphorus, are said to retain the originalinterpyrimidine ratios.g0 Reduction of the free reducing groups of the sugarwith sodium borohydride to the 2-deovyribitol stage renders apurinic acidssufficiently stable for the determination of titration curves.These reveal ahigh proportion of secondary phosphoryl dissociations, suggesting that theapurinic acids are considerably degraded.g1 Simultaneous hydrolysis of de-oxyribonucleic acids and condensation of the free reducing groups of the sugarresidues with mercaptoacetic acid yields aldehydoapurinic acid di(carboxy-methyl) dithioacetals.Use of mercaptoacetic acid in the presence of zincchloride and anhydrous sodium sulphate gave a product with the theoreticalsulphur content and with only 6.2% of the phosphorus rendered dialy~able.~~Wherever a purine residue has been displaced, both the 3’- and the 5’-phosphoester group are situated adjacent to a free hydroxyl group on whatwas the 4’-position of the purine nucleotide. As a result, such phospho-ester residues are labile to alkali, and the dialysable substances produced byalkaline hydrolysis have been separated by chromatography and iono-phoresis into at least twenty components : besides thymidine and deoxy-cytidine, thesc include the dinucleoside phosphates T-P-C, T-P-T, andC-P-C, and the trinucleoside diphosphate whose probable structure isT-P-T-P-C, and sulphur-containing oligonucleotides whose probable struc-tures are given as T- P-S-P, C-P-S-P, T-P-T-P-S-P, T-P-T-P-C-P-S-P,T-P-C-P-S-P, and C-P-S-P, where T = thymidine, C = deoxycytidine,P = phosphate, and S = 2-deoxy-aldehyde-D-ribose di(carboxymethy1) di-thioacetal.Reasons are given for believing that the sulphur-containingcomponents such as (9) and (10) have a terminal 4’-phosphate group blockingthe 4’-hydroxyl group and so removing its labilising action on the adjacentphosphodiester linkage.93 Evidence is given that the sulphur-containingcomponents occur as hitherto inseparable isomeric pairs (9) and (10).Theformer, (9), but not the latter should give formaldehyde when treated suc-cessively with phosphomonoesterase and sodium periodate ; the observedbehaviour of the sulphur-containing components is intermediate in thisrespect, a fraction of a mol. of formaldehyde being formed.94Proposal of the double-helical structure for deoxyribonucleic acids in90 M. E. Hodes and E. Chargaff, Biochim. Biophys. Acta, 1956, 22, 349.91 E. Hurlen, S. G. Laland, R. A. Cox, and A. R. Peacocke, Acta Chew. Scand.,O2 A. S. Jones and D. S. Letham, J., 1956, 2573.O3 Idem, J., 1956, 2579.O4 A. S. Jones, D. S. Letham, and M. Stacey, J., 1966, 2584.1956, 10, 793266 ORGANIC CHEMISTRY.which adenine residues adhere by two hydrogen bonds to thymine residuesof the other helix, and likewise guanine residues adhere by two hydrogenbonds to cytosine, 5-methylcytosine, or 5-hydroxymethylcytosine residues\ ?H2C Ha;( 9 ) HO h(Py = pyrimidine residue)of the other helix,95 has stimulated many physical investigations that havetended to confirm this structure.Titration of deoxyribonucleic acids withacid results, at pH 2.4 in 0.1M-sodium chloride or at pH 3.3 in O-OlM-sodiumchloride, in an abrupt and irreversible increase in optical density at 260 mp.This change has been associated with acceptance of a proton by the amino-group of deoxyguanylic acid 96-99 with concomitant enolisation and simul-taneous breaking of both hydrogen bonds [(ll) --t (lZ)].99 Deoxy-ribonucleic acids so treated are said to be denatured.lW Denaturation hasGuanine Cytosine Guanine + protonalso been induced lol by dialysis against water which lowers the pH to 2.6.Removal of the salt by dialysis at pH 6.5 also causes denaturation becauseit raises the pK, of guanylic acid to this value, whereas removal of the saltat pH 8 does not cause denat~ration.~~ Titration of deoxyribonucleic acidswith alkali produces the same irreversible changes since, contrary to previousobservations,g7* lo2 the back-titration curves after mild acid- and alkali-treatment are found to be coincident with one an0ther.10~ A method has95 G.R. Barker, Ann. Reports, 1954, 51, 279.96 L. F. Cavalieri and A. L. Stone, J . Amer. Chem. Sot., 1955, 77, 6499.97 D. 0. Jordan, A. R.Mathieson, and S. Matty, J., 1956, 154, 158.9s L. F. Cavalieri, M. Rosoff, and B. H. Rosenberg, J . Amer. Chem. Sot., 1956,78,L. F. Cavalieri and B. H. Rosenberg, Biochim. Biehys. Ada, 1966, 21, 202.6239.100 R. Thomas, ibid., 1954, 14, 231.101 C. A. Thomas and P. Doty, J . Amer. Chem. Sac., 1956, 78, 1854.109 W. A. Lee,and A. R. Peacocke, J., 1951, 2361.10s R. A. Cox and A. R. Peacocke, J., 1956, 2499SMITH : NATURAL MACROMOLECULES. 267been described whereby the fraction of hydrogen bonds which have beenruptured is determined by progressive displacement in the forward titrationcurve over pH 7-3; denaturation by heat only occurs above 75” and thereis a linear relation between the fraction of hydrogen bonds permanentlybroken and the increase in extinction at 260 mp.104 X-Rays cause anincreased susceptibility to denaturation by heat,lo5 and there is evidence forformation of hydroperoxides.lo6 The kinetics of enzymic digestion of deoxy-ribonucleic acids have been examined and the results indicate a two-strandstructure in which the number of pre-formed gaps before enzymic attack isnot more than one in three thousand nucleotide residues.106aLignins.-Since our last Report on this subject,lo7 interest in lignins hasincreased with the publication of about six hundred further papers, andseveral recent reviews.lO*-lll At no stage has a point been reached at whichit is possible to deduce a reliable structure for lignins.Lignins occur in their most concentrated form in woodand nearly all investigations have used this source. Methods of extractionfall into two classes, those dissolving the accompanying polysaccharide, andthose dissolving lignins.In the former, alkaline copper solution,lf2 periodicacid,l13? 11* and strong mineral acids 115 have all been shown to react withlignins, so that the preparations are not chemically identical with lignins asthey occurred in wood. In the latter class, reagents which dissolve thegreater part of lignins-sodium hydrogen sulphite, sodium hydrogen sul-phide, mercaptoacetic acid, sodium hydroxide, and mineral acid in organicsolvents-all combine with them and induce partial depolymerisation. Itwas shown recently that extraction of lignins from wood meal with ethanolichydrogen chloride can be carried out under milder conditions than werepreviously used.l16Of great practical significance was the discovery by Brauns l 1 7 that coldalcohol, without any acidic or basic reagent, extracts from wood mealssubstances having all the chemical properties hitherto associated with lignins.Extractions.lo4 R.A. Cox and A. R. Peacocke, J . , 1956, 2646.lo5 K. V. Shooter, R. H. Pain, and‘J. A. V. Butler, Biochim. Biophys. Acta, 1956,lo6 G. Scholes, J . Weiss, and C. M. Wheeler, Nature, 1956, 173, 157.lo60 V. N. Schumaker, E. G. Richards, and H. K. Schachman, J. Amer. Chem. SOC.,lo7 E. G. V. Percival, Ann. Reports, 1942, 39, 142.Io8 R. D. Haworth, “ Thorpe’s Dictionary of Pure and Applied Chemistry,” 4thedn., 1946, Vol. VII, p. 307; E. Hagglund, “ Chemistry of Wood,” Academic Press Inc.,New York, 1951; L.E. Wise and E. C. Jahn, “ Wood Chemistry,” Reinhold Publ.Corp., New York, 1952; K. Freudenberg in “ Moderne Methoden der Pflanzenanalyse,”K. Paech and M. V. Tracey, Springer, Berlin, 1955, Vol. 111, 509; K. Freudenberg,Angew. Chem., 1966, 68, 84.20, 497.1956, 78, 4230.loQ H. Erdtman, Research, 1950, 3, 63.l10 F. E. Brauns, “ Chemistry of Lignin,” Academic Press Inc., New York, 1952.ll1 K. Freudenberg, Fmtschr. Chem. org. Naturstofle,’ 1954, 11, 45.112 I. Pearl, J. Amer. Chem. SOL, 1950, 72, 2309.113 D. Pennington and D. M. Ritter, ibid., 1947, 69, 187; P. F. Ritchie and C. B.114 E. Adler and S. Hernestam, Acta Chem. Scand., 1955, 9, 319.116 H. G. Arlt, K. Sarlranen, and C. Schuerch, ibid., 1956, 78, 1904.117 F.E. Brauns, ibid., 1939, 61, 2120.Purves, Pulp and Paper Mag., Canada, 1947, 48, No. 12, p. 74.D. E. Read and C. B. Purves, J. Amer. Chem. Soc., 1952, 74, 120268 ORGANIC CHEMISTRY.The term " native lignins " has been transferred to these preparations and,though the yield is very small, they have been selected for investigation inmuch subsequent work. Native lignins have been prepared from blacks p r ~ c e , l l ~ - ~ ~ ~ western hemlock,122 a ~ p e n , l ~ ~ * 124 Scots pine,l25~ 126 oak, birch,maple,127* la8 bagasse,lzs9 129 wheat,130* 13l horse chestnut, and Douglas fir.131The question of whether native lignins are truly representative of lignins intheir insoluble condition in wood has been discussed.128 Brown rots, fungiwhich degrade polysaccharide more rapidly than lignins, have been shownto liberate to solvent extraction increasing amounts of material from wood,similar to native lignin~,l~~* 133 but differences between native and fungus-released lignins are sometimes observed.lls The method of extraction doesnot separate native lignins from other macromolecular phenolic substancesand so is unsatisfactory with plant tissues that contain such substances.laThorough crushing of wood meal has been reported to release no morelipins to solvent extraction,llg but Bjorkman has reported that if conditionsfor grinding give the minimum particle size and do not favour swelling of theparticles then two-thirds of the lignins of wood meal become extractable bycarefully chosen solvent mixtures." Milled wood lignins " obtained in thisway are less soluble than, but show only small quantitative differences from,the native lignins of the same wood.121 A native lignin has been separatedinto fractions of varying molecular weight, showing slight systematic vari-ations in composition but essentially the same infrared spectra.120 Nativelignin, as usually prepared, is only one of several fractions in the cold-alcoholic extract of wood ; the ether-soluble fraction rejected during isol-ation of native lignin has been found to resemble native lignin in somerespects at least, and it may contain lignins of low molecular weight.12*Despite the objection that native lignin constitutes only a small percentageof the total lignin of wood, it remains the purest and most convenientmaterial for degradative work and, once a point has been established fornative lignin, it can often then be repeated, providing polysaccharides donot interfere, with whole WOO^.^^^^ 135These are particularly important since they offerthe best hope of establishing the structuie of lignins.Degradation methods.118 A.Apenitis, H. Erdtman, and B. Leopold, Suertsk Kent. TidsFzr., 1951, 65, 195.119 F. E. Brauns and H. Seiler, TAPPI, 1952, 35, 67.121 A. Bjorkman, Nature, 1954, 174, 1057.l2% F. E. Brauns, J . Org. Chew., 1945, 10, 211.l a 3 M. A. Buchanan, F. E. Brauns, and R. L. Leaf, J . Amer. Chem. SOL, 1949, 71,u 4 D. C . C. Smith, J., 1935, 2347.125 W. J. Schubert and F. F.Nord, J . Amer. Chem. SOL, 1950, 72, 977, 3835.126 S. F. Kudzin, R. M. DeBaun, and F. F. Nord, Z%d, 1951, 73, 4615.127 S. F. Kudzin and F. F. Nord, ibid., p. 4619.128 G. DeStevens and F. F. Nord, ibid., p. 4622; 1953, 35, 305.129 D. C. C. Smith, Nature, 1955, 176, 267.130 J. E. Stone and K. G. Tanner, Canad. J . Chem., 1952, 30, 166.131 D. C. C. Smith, Nature, 1955, 176, 927.132 WT. J. Schubert and F. F. Nord, J . Amer. C h m . Soc., 1950, 72, 977, 3835.133 S. F. Kudzin and F. F. Nord, abid., 1951, 73, 4615, 4619.134 R. M. DeBaun and F. F. Nord, ibid., p. 1358.135 J. C. Pew, ibid., 1952, 74, 5784.C. L. Hess, ibid., p. 312.1297SMITH : NATURAL MACROMOLECULES. 269(a) Mild alkaline oxidation. Under these conditions, p-hydroxybenz-aldehyde , vanillin (4-l-iydroxy-3-methoxybenzaldehyde), and syringaldehyde(4-hydroxy-3 : 5-dimethoxybenzaldehyde) are produced, owing their survivalunder oxidising conditions to the stability of their mesomeric anions.Nitro-benzene is most frequently used as the oxidising agent 136 but amine oxides,sulphones , sulph~xides,~~~ cupric salts,138 and cobalt salts 139 have beenused. When silver oxide is used as the oxidising agent vanillic acid ratherthan vanillin is produced 140 and, with mercuric oxide, 5-hydroxymercuri-vanillin is 0btained.1~1Optimum conditions for oxidation of wood with alkali and nitro-benzene have been 143 and this reaction has been adapted forthe determination of lignins.la? 145 Lignins of twenty-seven conifers havebeen found to give mainly vanillin (-30% yield).146 This accords with theirmethoxyl content (-15%) which indicates mainly guaiacyl (4-hydroxy-3-methoxyphenyl) nuclei in these lignins.Minor constituents detected inthe oxidation product of spruce lignin inciude substances linked throughposition 5 to carbon such as 5-f0rmylvanillin,~~~ dehydrodi~anillifi,~~~5-formylvanillic acid ,147 5-carboxyvanillin,136* 148 and dehydrodivanillicacid.138 It is not known whether this linkage through position 5 is anintegral feature of lignin or is established during alkaline oxidation. Oxid-ation of model compounds by alkali and nitrobenzene shows that nearly allsubstances having a guaiacyl or veratryl nucleus give considerable amountsof ani ill in.^^^^ 149-152 It has been established that the yield of aldehydes isdetermined by the mode of linkage of the aldehyde-yielding residues ratherthan by any instability of the aldehydes under the conditions of 0 ~ i d a t i o n .l ~ ~The view has been expressed that there are ‘ I open units ” which are notlinked to carbon through C , and give rise to vanillin, and I ‘ condensed units ”which give rise to small amounts of 5-formylvanillin on oxidation by nitro-benzene in alkali.142 Evidence for some “ open units ” in lignins is affordedby the coupling of lignins with aryldiazonium salts,l10 and the formation of5-iodovanillin by alkaline oxidation of spruce lignin first mercurated withmercuric acetate and then treated with iodine.139 It has also been suggestedthat the majority of aromatic units in spruce lignin are of the uncondensedtype.lB138 K.Freudenberg, W. Lautsch, and K. Engler, Bey., 1940, 73, 167.137 P. Lagally, U.S.P. 2,547,913; Chem. Abs., 1951, 45, 8043.138 I. Pearl, J . Amer. Chem. SOC., 1942, 64, 1429; 1950, 72, 2309.139 W. Lautsch and G. Piazolo, Ber., 1940, 73, 317.140 I. Pearl, U.S.P. 2,483,559; Chew. Abs., 1950, 44, 1706.141 H. F. Lewis and I. Pearl, U.S.P. 2,489,380; Chem. Abs., 1950, 44, 1706.142 B. Leopold, Acta Chem. Scand., 1950, 4, 1523.143 K. €3. Kavanagh and J. M. Pepper, Canad. J . Chem., 1955, 33, 24.Ip4 J. E. Stone and M. J. Blundell, AnaZyt. Chem., 1951, 23, 771.145 F. E. Roadhouse and D. MacDougall, Biochem. J., 1956, 68, 33.146 B. Leopold and I.-L. Malmstriim, Acta Chem. Scand., 1962, 6, 49.lS7 B.Leopold, ibid., p. 38.lp8 K. Freudenberg and F. Klink, Ber., 1940, “3, 1369.140 A. V. Wacek and K. Kratzl, ibid., 1944, 77, 516.150 K. Kratzl and I. Khautz, Monatsh., 1948, ‘98, 376.151 B. Leopold and 1.-L. Malmstrdm, Acta Chem. Scand., 1951, 5, 936.15* B. Leopold, ibid., p. 1393.168 J. C. Pew, J . Amer. Chem. Soc., 1055, ‘97, 2831270 ORGANIC CHEMISTRY.Oxidation of hard-wood lignins by nitrobenzene in alkali yields, underoptimum conditions, mainly syringaldehyde (about 36%) and vanillin(15%).1437 146 The total yield of pure aldehydes reaches its highest in theselignins and accounts for about 60% of the original lignin substance. Theratio of syringaldehyde to vanillin in the product of oxidation is twice aslarge as the ratio of syringyl to guaiacyl residues indicated in the originallignin by its methoxyl content; since syringyl residues cannot form con-densed units by linkage to carbon through position 5, this lends furtherweight to the evidence for condensed units involving guaiacyl residues.The lignins of grasses and cereals yield $-hydroxybenzaldehyde as well asvanillin and ~yringaldehyde.1~~ The yield of P-hydroxybenzaldehyde varieslittle, that of vanillin increases slightly, and that of syringaldehyde increasesmarkedly with the age of the ~lant.14~ Sphagnum moss contains a ligninof very small methoxyl content, yielding mostly 9-hydroxybenzaldehyde onoxidation with nitrobenzene in alkali.155(b) Drastic oxidation.Oxidation of methylated lignins by alkalinepermanganate yields a small amount of isohemipinic acid (1),156 additionalevidence for the presence of condensed guaiacyl residues in lignins.Ligninswhich have been previously exposed to the action of strong acid yield a littlemetahemipinic acid (2) on methylation and 0xidation.1~7 It has been sug-gested that strong acid induces coniferaldehyde groups to condense withguaiacyl residues 158 and the isolation of metahemipinic acid seems to con-firm this. Small amounts of benzene-penta- and -1 : 2 : 3 : 4-tetra-carboxylicacid have been obtained in alkaline permanganate oxidation of lignins, butthe amounts obtained from whole wood were very small, indicating thatthese products result from oxidation of condensed structures producedmainly during the isolation of lignins.l15Alcoholysis of lignins produces products containing C-methyl groups (seebelow) but, since chromic acid yields negligible amounts of acetic acid, thesecan be considered absent from lignins themselves.159Hydrogenation of lignins at elevated temperatureand high pressure renders them entirely soluble in chloroform. Besidesmethanol and a complex residue,lw9 161 compounds (3)-(5) ,162 (6),160(7),161 (8)-(1O),l6O and (11)-(13) 163 have been isolated and characterised.(c) Hydrogenolysis.154 R. H. J. Creighton and H. Hibbert, J . Amer. Chem SOC., 1944, 66, 37.156 B. Leopold and 0. Theander, Acla Chem. Scand., 1952, 6, 311; V. C. Farmer,156 K. Freudenberg, A. Janson, E. Knopf, and A. Haag, Ber., 1936, 69, 1415.lB7 H.Richtzenhain, Acta Chem. Scand., 1950, 4, 589.15* J. C. Pew, J . Amer. Chem. SOC., 1952, 74, 2850.159 W. S. McGregor, T. H. Evans, and H. Hibbert, ibid., 1944, 66, 41.160 C. P. Brewer, L. M. Cooke, and H. Hibbert, ibid., 1948, 70, 57.161 C. Schuerch, ibid., 1950, 72, 3838.162 E. E. Harris, J. D’Ianni, and H. Adkins, ibid., 1938, 60, 1467.163 J. M. Pepper, C . J. Braunstein, and D. A. Shearer, ibid., 1951, 73, 3316.Research, 1953, 6, 475SMITH : NATURAL MACROMOLECULES. 27 1Together with methanol, compounds (3)-(5) accounted for 62% of thesubstance of aspen lignin. 162 Hydrogenolysis under acid conditions yieldsphenylpropane derivatives whereas in alkaline conditions phenylethanederivatives predominate. 163Cleavage of lignin by sodium in liquid ammonia has yielded the phenols(14),164 probably (15),165 and (16)J166 all in small yield together with a com-plex residue.Using potassium in liquid ammonia, Freudenberg et al. wereable to reduce the methoxyl content of spruce lignin from 15% to 6%.16'( d ) Acid-catalysed alcoholysis. Treatment of wood with warm alcoholichydrogen chloride yields three fractions derived from lignin : one ether-soluble and of low molecular weight, one soluble in alcohol but not in ether(" alcohol lignin "), and one insoluble and adhering to the residual poly-saccharide. The ether-soluble fraction has been shown to consist largelyof substances RCHO R*CO*COMe, R*CH,*COMe, and R*CO*CHMe*OR',where R = P-hydroxyphenyl, guaiacyl, or syringyl, and R is the alkylresidue of the alcohol used.In the reaction of ethanolic hydrogen chloridewith maple wood these substances (R = guaiacyl and syringyl) amount to9.8% of the original lignin. 168 Paper-chromatographic methods of separat-164 N. N. Shorygina, T. Ya. Kefeli, and A. F. Semechkina, Zhur. obshchei Khim.,165 N. N. Shorygina and T. Ya. Kefeli, ibid., 1950, 20, 1199.lB6 A. F. Semechkina and N. N. Shorygina, ibid., 1963, 23, 593.ld7 K. Freudenberg, K. Engler, E. Flickinger, A. Sobek, and F. Klink, Bey., 1938,1949,19, 1558.71. 1810.lS8 M. Kulka, E. Fisher, S. B. Baker, and H.66, 39.Hibbert, J . Amer. Chem. Soc., 1944272 ORGANIC CHEMISTRY.ing these substances (R = guaia~yl,16~ 9-hydroxyphenyl, and syringyl I7O)have been described,The lignin of spruce wood has been recovered in 60% yield as " ethanol-lignin " under mild conditions.171 It has a ratio of methoxyl to carbon closeto the theoretical value for a propylguaiacol polymer,172 and has beenallocated an average empirical formula (24).The acquired alkoxyl groups of '' alcohol lignins " are stable to alkaliand hydrolysed in acid.173 There is strong evidence that the alkylatablegroups in lignins are $-hydroxy- and P-alkoxy-benzyl alcohol and -benzylether groups (17)-(20), the group OR' in the ethers, like the hydroxyl groupof the benzyl alcohols, being replaced by the alkoxyl group derived from thes01vent.l~~ Since such benzyl ether linkages may connect parts of thelignin molecule it is not surprising that substances of lower molecular weightare released in alcoholysis.Study of model substances shows that com-pounds of type (17) and (18) react more rapidly with methanolic hydrogenchloride than do their ethers (19) and (2O).17*(e) Sulphonation. Treatment of spruce wood with a solution of sodiumsulphite and sodium hydrogen sulphite (pH 5.25) at 135" dissolves part ofthe lignin as a sulphonic acid derivative and leaves the remainder, alsocontaining sulphonic acid groups, insoluble and adhering to the residualpolysaccharide. 110 Holmberg found that phenylmethanols react withsulphite as ROH + NaHSO, _t R*SO,Na + H,O and suggested thatbenzyl alcohol groups in the lignin react with sulphite in this way.175 Modelcompounds having structural units (17)-(20) have been sulphonated andcompared with lignin with respect to the pH requirements and velocity of~ulphonation.~~~? 177 Lindgren concludes that structures of all types(17)-(20) are present in lignin and that all the sulphonatable groups inlignin have this type of structure.177The water-soluble ligninsulphonic acids have been separated into frac-tions of widely differing molecular weight by dialysis, 17* fractional precipit-a t i ~ n , l ~ ~ $ lSo and fractional elution 181 of their salts.The ratio of sulphoniclo@ K. Kratzl and W. Schweers, Monutsh., 1954, 85, 1046, 1166.170 Idem, ibid., 1956, 89, 186.171 H. G. Arlt, K. Sarkanen, and C. Schuerch, J . Amer. Chem. SOL, 1962, 73, 4996.17a C. Schuerch, ibid., p. 4996.178 F. E. Brauns, ibid., 1946, 68, 1721.17' E.Adler and J. Gierer, Acta Chem. Scand., 1955, 9, 84.176 S. Heden and B. Holmberg, Svensk hem. Tidskr., 1936, 48, 207.176 B. 0. Lindgren, Acta Chem. Scund., 1948,1, 779; 1949, 3, 1011; 1950, 4, 1365.177 I d c m , ibid., 1951, 5, 603.178 K. Schwabe and L. Hasner, Cellulosechem., 1942, 20, 61.179 W. Lautsch and G. Piazolo, ibid., 1944,22, 48.lSo K. Schwabe and L. Hahn, Holzi-foorsch., 1947, 1, 42, 79; E. D. Olleman, D. E.Pennington, and D. M. Ritter, J . Colloid Scz., 1948, 3, 185; A. E. Markham, Q. P.Peniston, and J. L. McCarthy, J . Amer. Chem. Soc., 1949, 71, 3599; J. Moacanin,V. F. Felicetta, W. Haller, and J. L. McCarthy, ibid., 1955, 77, 3470.V. F. Felicetta, A. hhola, and J . L. McCarthy, i b i d . , 1966, 78, 1899SMITH : NATURAL MACROMOLECULES.273acid to guaiacyl groups varies 179 between 0.5 and 1.0, and the lignin sul-phonic acids of lowest molecular weight can be separated by ionophoresisinto four groups of anionic substances of distinct mobilities, presumablycorresponding to simple values of this ratio. In accordance with this, theionophoresis patterns increase in complexity with fractions of increasingmolecular weight The prospect of isolating some of the simpler sulphonicacids by chromatographic procedures seems to be quite good.SulphidesJ1s2 mercaptoacetic acid,lB* l g 4 and thiols la5 have also beenshown to react analogously to sulphite.Reactive Groups in Lipins.-(a) Phenolic groups. Lignins owe theirability to dissolve in aqueous alkali (pH >10-4) to the possession of phenolicgroups.Attempts to carboxylate these according to the Kolbe-Schmittprocedure were unsuccessful. l86 Chemical methods of determining phenolicgroups include reaction with l-fluoro-2 : 4-dinitrobenzeneJ1l1 reaction withtoluene-9-sulphonyl chloride followed by reaction with hydrazine,ls7 anddifferential methylation with trimethylphenylammonium hydroxide.ls8Differential methylation by diazomethane is not successful.lll The absorp-tion band at 280 mp characteristic of phenolic ethers and phenols moves tolonger wavelength in the case of phenols when they ionise. This has beenused to determine free phenolic groups,189v l90 but it is only accurate whenonly one type of phenolic chromophore is present in lignin.The splitting off of methanol from guaiacyl residues when they aredehydrogenated with periodate (cf.reactions of 21) has been made the basisof a method for estimating such groups in lignin.114 Model compounds give90% yields of methan01.l~~Gierer lgl has used the reaction of quinone monochloroimide with @-hydr-oxyphenylmethanol groups (22) to determine the proportion of such groupsin lignins, the amount of indophenol (23) released being estimated colori-metrically.Determination of phenolic groups by titration in acet~ne-ethanol,~~~1 8 2 D. L. Brink, R. L. Hossfeld, and W. M. Sandstrom, J . Amer. Ghem. SOC., 1949,183 B. Holmberg and N. Gralen, Ing. Vetenskaps. Akad. Handl., 1942, No. 162.184 B. Holmberg, Finska Kemistsamfundets Medd., 1945, 54, 124.lS5 F. E. Brauns and M.A. Buchanan, Paper Trade J . , 1946, 122, No. 21, 49.lS6 M. M. Yan and C. B. Purves, Camad. J . Chem., 1956,34, 1582.K. Freudenberg and H. Walch, Ber., 1943, 76, 305.lB8 K. Freudenberg, Das Papier, 1947, 1, 209.lSB G. Aulin-E~dtman, Svensk Papperstidn., 1954, 57, 745.lB0 0. Goldschmid, Analyt. Chem., 1954, 26, 1421.lgl J. Gierer, Acta Chem. Scand., 1954, 8, 1319.lg2 K. Sarliancn and C. Schuerch, Aizal3.t. Chem., 1955, 27, 1245.71, 2275 ; T. Enkvist and E. Hagglund, Svensk Papperstidn., 1950.53, 85274 ORGANIC CHEMISTRY.dimethylformamide, dimethyl sulphoxide,lYd and ethylenediamine 194 hasbeen described. It is also possible to titrate them in aqueous solutionprovided the end-point is approached from the alkaline ~ide.12~Empirical formulae for spruce lignins have been derived from analyses ofethanol lignin (24) ,116 mercaptoacetic acid lignin (25) ,lS3 ligninsulphonicacid (26),lo9 and native lignin (27).loS Native lignin from spruce generally(24) CuH,.,O,.,(OMe)o.a,(OEt)~.~(25) CpH,.o,Oz-es(OMe)o.oz + HS*CHz*CO2H - H,O(26) Cu&.,0z.6(OMe)o.a~ + H,SO, - H,O(27) C B H ~ .~ O ~ . ~ ( O M ~ ) ~ . ~contains 0.5 phenolic hydroxyl group per guaiacyl residue, and 0.9 aliphatichydroxyl group 111 of which approximately two-thirds are primary as judgedby their reactivity.lsS Together, these account for 1.4 oxygen atoms perguaiacyl residue. Most of the oxygen not accounted for as hydroxyl ormethoxyl (1.0 atom per guaiacyl residue) is thought to occur in etherlinkages. 111(b) Carbonyl groups.These have been estimated in lignins by reactionswith hydroxylamine hydrochloride and titration of the liberated acid. 174The presence of 9-hydroxybenzoyl residues (28) has been inferred from theincrease in absorption at 350 mp when lignin is made alkaline; 131* lg5 suchgroups show a shift of the main absorption band from 278-307 to 330-364 mp on ionisation. 9-Alkoxyphenyl ketones (29) ionise in strong sul-phuric acid with a similar shift of absorption, and it has been claimed thatR'= a l k y l ; R'=H or Methe greater increase in absorption at 350 mp when lignin is dissolved instrong sulphuric acid is a measure of the presence of these groups.lg5 Thiscannot be taken as reliable since, at the temperature used, acid of thisstrength induces irreversible changes in the absorption spectrum of lignins.The most characteristic and universal colour reaction of lignins, a redcolour with phloroglucinol and strong acid (" the Wiesner reaction "), hasbeen shown to be given also by coniferyl (ferulic) aldehyde (3O),lg6 which canbe split off from lignins in small yield by heating them with stannic chloridesolution. lg6* lg7 Colorimetric estimations of the number of bound coniferylaldehyde groups in spruce lignin give one per forty 198 and one per thirty-six lg9 guaiacyl residues.The coloured complex given by lignin, resorcinol,and strong acid, and the coloured product formed on acidifying the productla3 J. P. Butler and T. P. Czepiel, AnuZyt. Chem., 1966, 28, 1468.lS4 K.Freudenberg and K. Dall, Nuturwiss., 1955, 42, 606.195 0. Goldscmid, J . Arner. Chem. Soc., 1953, 75, 3780.lS6 E. Adler, K. J. Bjorkqvist, and S . Naggroth, Actu Chem. Scand., 1948, 2, 93.la' F. Czapek, 2. physiol. Chem., 1899, 287, 141.1Q8 E. Adler and L. Ellmer, Actu Chem. Scund., 1948, 2, 839.10a J. C. Pew, J. Amer. Chem. Sot., 1951, 75, 1678SMITH NATUKAL MACROMOLECULES. 275of catalytic hydrogenation of 7 : 4'-dihydroxy-3'-methoxyflavanone, havebeen shown to have identical absorption spectra due to a mesomeric cation(31).19!) The green colour given by lignins in strong acid has been shown toconsist of two absorptioii bands. One absorption max. at 453 mp is due tothe conjugate cation of bound coniferyl aldehyde groups (32) ; the other, at628 mp, fails to appear at -30" and has been attributed to condensation ofconiferyl aldehyde groups with other guaiacyl residues.lS8H o o ; ~ . c H : w O o - H;:CH.CH : m o o -(31) OH OMe OMe (32)That the phenolic oxygen atom is engaged in binding the coniferylaldehyde group to the rest of the molecule is indicated by the reaction of thisgroup with cold dilute alkali. Acetaldehyde is split off in a reversed-aldolreaction, leaving an etherified vanillin residue.200 The reaction can befollowed spectrophotometrically.131 The reaction would not have beenpossible if the phenolic hydroxyl groups of the bound coniferyl aldehydehad been free; instead, a stable yellow anionic group would then have beenformed by ionisation. It has been suggested that the bound coniferylaldehyde residues constitute end-groups of lignin.lg6 They can be deter-mined by reduction with borohydride, followed spectrophotometrically ; theonly other carbonyl-containing residues that absorb at the same wavelengthare anionic in dilute alkali and therefore reduced only very s10wly.l~~ Thebound coniferyl aldehyde groups show a distinct band at 1657-1662 cm.-lin the infrared absorption spectrum of 1igni11s.l~~(c) Ester groups.9-Hydroxybenzoate groups have been detected inaspen lignin. They occur singly, involving aliphatic hydroxyl groups andamount to one-tenth of the 1ig11in.I~~ One-tenth of sugar-cane lignin con-sists of +-coumaric ester groups and there is evidence for +-coumaric andferulic ester groups in wheat-straw lignin.lZ9Biosynthesis of Zignin.Wheat plants began to incorporate 14C intolignin five hours after labelled carbon dioxide was supplied, as judged by theactivity of the vanillin and syringaldehyde obtainable on oxidation withalkali and nitrobenzene.201 The activity of labelled phenylalanine, tyro-sine, and cinnamic acid is incorporated efficiently in p-hydroxybenzaldehyde,vanillin, and syringaldehyde obtainable from wheat lignin.202 There is aparticularly good recovery of activity in the vanillin fraction after wheatplants have been fed with labelled ferulic acid.202 Of ten other speciesexamined, labelled tyrosine was incorporated in only one instance, thoughincorporation of the other substances mentioned was quite general.203Feeding shikimic acid (generally labelled) and phenylalanine (generallylabelled) to cut maple stems, followed by oxidation with nitrobenzene inalkali of the cell-wall fraction, yielded labelled vanillin and labelled syring-aldehyde ; the activity of protocatechuic acid (carboxyl-labelled) was not2oo K.Kratzl and G. Hofbauer, Monatsh., 1956, 87, 617.201 S. A. Brown, K. G. Tanner, and J. E. Stone, Canad. J . Chem., 1953, 31, 755.202 S. A. Brown, F. M. Claire, and M. D. Chisholm, Canad. J . Biochem. Physiol.,203 S. A. Brown and A. C. Neish, ibid., 1956, 54, 749.1955, 33, 948276 ORGANIC CHEMISTRY.incorporated.2a There can be little doubt that shikimic acid can act as adirect precursor of lignin, in view of the discovery 205 of a specific pattern oflabelling in vanillin (34) after sugar-cane plants have been fed with shikimicacid labelled in the 2- and the 6-position (33).$02H CHOS u g a rHO '0 HOH(33)Asterisks indicate '*C, numbers indicate relative radioactivity.Dehydrogenation of p-hydroxycimamyZ alcohols in vitro.Following thesuggestion by P. Klason %06 that lignin is related to coniferyl alcohol (4-hydr-oxy-3-methoxycinnamyl alcohol) which is present as the glucoside coniferin(35) in young tissue and in the cambial sap of trees, Freudenberg discoveredthat a crude extract of mushroom was capable of dehydrogenating coniferylI? OH (35)B: CHY* CH 2- OHalcohol in vitro to an amorphous substance of high molecular weight, whichhe termed " Dehydrierungspolymerisat " (DHP).207 The same mushroomextract had been previously found to have a similar effect on isoeugenol.208DHP of coniferyl alcohol, and the native lignin of black spruce wood, havean impressive similarity which takes in such properties as elementaryanalysis, colour reaction with phloroglucinol, solubility, oxidation to vanillin,hydrolysis in 28% sulphuric acid to small amounts of formaldehyde, methyl-ation followed by oxidative degradation to small amounts of isohemipinicacid, content of phenolic and aliphatic hydroxyl groups, acid-catalysedethanolysis, sulphonation with bisulphite, and ultraviolet and infraredspectra.209 DHP and native spruce lignin are said to differ in their behaviourtowards acetic anhydride in acetic acid.210 Native lignin also containsslightly less mcthoxyl than DWP.l1l 4-Hydroxycinnamyl alcohol,2114-hydroxy-3-methoxycinnamyl methyl ether,212 3 : 4dihydroxycinnamylalcohol, and 3 : 4 : 5-trihydroxycinnamyl alcohol 213 also give DHP's.Neither labelled 3 : 4-dimethoxycinnamyl alcohol nor labelled vanillin is204 S. A. Brown and A. C. Naish, Nature, 1955, 175, 688.206 G. Eberhardt and W. J. Schubert, J . Amer. Chem. SOL, 1956, 78, 283.eo6 P. Klason, Svensk kem. Tidskr., 1897, 9, 135.207 K. Freudenberg, Angew. Chem., 1949, 61, 228.* 0 8 H. Cousin and H. Herissey, Compt. rend., 1908, 147, 247.2oa K. Freudenberg, Sitzungsber. Heidelberg Akad. Wiss., 1949, No. 5; K. Freuden-*1O Idem, Naturwiss., 1954, 41, 232.211 K. Freudenberg and G. Gehrke, Chem. Bey., 1961, 84, 433.212 K. Freudenberg and F. Bittner, ibid., 1952, 85, 86.213 K. Freudenberg and W. Heel, ibid., 1953, 86, 190.berg et al., Chem. Bey., 1950, 83, 519, 530, 533; 1951, 84, 961SMITH : NATURAL MACROMOLECULES. 277incorporated in the DHP formed by coniferyl alcohol,214 demonstrating theimportance of both the free phenolic group and the double bond of the sidechain in DHP formation. Sinapyl alcohol (4-hydroxy-3 : 5-dimethoxy-cinnamyl alcohol) does not give a DHP with mushroom extract, forminginstead nL-synringaresinol (36) 2159 216 whose structure has been proved bysynthesis.217 However, sinapyl alcohol is incorporated in the DHP formedby a mixture of sinapyl and coniferyl alcohol.216Erdtman has suggested that the initial step in dehydrogenation of coni-feryl alcohol is removal of the phenolic hydrogen atom, leaving a phenoxideradical (38), the effect of the double bond in the side chain being to transferfree-radical reactivity to the p-position of the side chain as in (38c).lo9 Theradical (3%) would be expected to combine with another molecule of coni-feryl alcohol in one of the three positions p, 5, and 04, splitting off thephenolic hydrogen of the second coniferyl alcohol residue as a hydrogen atom,and yielding intermediate dimeric products (39), (40b), and (41) respectively.(4Ca) t HCH~.OH+ R'OH OoMe CHeOU'IThe reactive quinone methine structure possessed by (39)-(41) has beensuggested as an integral step in lignificati~n.~O~$ The quinone methinewould be expected to add a group O R at the cc-position of the side chain(cf. 42). In the context of lignification or DHP formation, O R could bethe phenoxide or the y-alkoxide group of another molecule of coniferylalcohol, or OH.214 K. Freudenberg and W. Fuchs, Chem. Ber., 1954, 87, 1824.216 K. Freudenberg et al., ibid., 1951, 84, 472.218 K. Freudenberg and H. €3. Hubner, ibid., 1952, 86, 1181.217 K. Freudenberg and H. Schraube, ibid., 1955, 88, 16.218 K. Freudenberg, Chew.-Ztg., 1960, 74, 12278 ORGANIC CHEMISTRY.Dimeric products representing three of the nine possible combinationsdescribed above have been isolated from the filtrate after DHP formationfrom coniferyl alcohol by mushroom extract. Dehydrodiconiferyl alcohol(43) (successive reaction with the 5-posi-tion of another molecule, and reactionwith the phenoxide group of that same OoMe CH.OH molecule) and m-pinoresinol (37) (suc-cessive reaction with the P-position ofanother molecule, and reaction with the Meo(y::2.0H y-alkoxide group of that same molecule):H have been obtained crystalline.219 InCHz'OH CH :H addition, syrupy cc-guaiacylglycerolaction with the 04-position of anothermolecule and with OH) was isolated by countercurrent distribution 219 andcharacterised by conversion into the 2 : 4-dinitrophenyl ether of its dihydro-derivative and by synthesis.220Freudenberg has suggested that these are intermediates (" sekundareBausteine ") in lignin formation. However, none of the three dimericproducts isolated has retained the reactive P-hydroxystyrene system and sothey may be by-products rather than intermediates in DHP formation.Each of them is, however, further dehydrogenated in mushroom extracts.During the formation of intermediate dimers (39), (40), and (41) a newasymmetric centre is created. If both reacting molecules of coniferylalcohol are engaged in an enzyme complex one would expect the new asym-metric centre to have a single enantiomeric configuration. With the natur-ally occurring lignans this is found to be the case.221 In contrast, both thepinoresinol and the syringaresinol synthesised in mushroom extracts areracemic. Such measurements as have been made l1ll 222 indicate thatnative lignin has no optical activity, suggesting that, though enzymes areconcerned in producing reactive intermediates from coniferyl alcohol, theactual polymerisation step of lignin formation is a more random process.Whether, as the weight of present evidence indicates, lignin and the DHP ofconiferyl alcohol as prepared in vitro have the same structure will be decidedonly when the structural pattern of these substances has been adequatelyestablished.OeMcMeo($: CH2.0H(1 3 ) CH2,0H P ( 4 4 ) p-coniferyl ether (44) (successive re-D. C. C. S.G. BADDELEY. T. G. HALSALL.P. BLADON. J. HONEYMAN.L. CROMBIE. D. C. C. SMITH.M. J. S. DEWAR. G. F. SMITH.R. F. GARWOOD. W. WILSON.A. R. BATTERSBY. H. B. HENBEST.zls K. Freudenberg and H. Schluter, Chem. Bey., 1955, 88, 617.220 K. Freudenberg and W. Eisenhut, ibid., p. 626.221 W. M. Hearon and W. S . MacGregor, Chem. Rev., 1955, 55, 957.232 33. Holmberg, Finska Kemistamfundets Afedd., 1945, 54, 124; D. C. C . Smith,unpublished observation