摘要:

J O U R N A L OF THE CHEMICAL SOCIETY. TRANSACTIONS. H. BRERETON BAKER M.A. D.Sc., F. R. S. J. N. COLLIE Ph.D. F.R.S. A. W. CROSSLEY D.Sc. Ph.D. F.R.S. B. G. DONNAN M.A. Ph.D. F.R.S. BERNARDYER D.Sc. M. 0. FORSTER D.Sc. Ph.D. F.R.S. A. HARDEN D.Sc. Ph.D. F.R.S. T. M. LOWRY D.Sc. J. C. PHILIP D.Sc. Ph.D. F. L. PYMAN D.Sc. PE.D. A. SCOTT M.A. D.Sc. F.R.S. G. SENTER D.Sc. Ph.D. S. SMILES D.Sc. @bitax : J. C. CAIN D.Sc. Ph.D. Sttb - @bifux : A. J. GREENAWAY. 2Jsrraistaut Snb-Qbifar : CLARENCE SMITH D.Sc. 1915. Vol. CVII. Part I. pp. 1-846. LONDON : GURNEY & JACKSON 33 PATERNOSTER ROW E.C. 1915 PRINTED IN GREAT B~ITAIN BY RICHAED CLAY & SONS LIMITED, BRUNSWICK BT. STAMFORD ST. S.E., AND BUNOAY SUFFOLK CONTENTS. PAPERS COMAITJNICATED TO THE CHEMICAL SOCIETY; 1.-Solutions of Bromine in Water Nitrobenzene and Carbon Tetrachloride.By ALFRED FRANCIS JOSEPH . 11.-z'soTlibenzoylglucoxylose. By FRANK TUTTN . 111.-A New Method of Preparing Alkylated Sugars. By WALTER NORMAN HAWORTH . 1V.-Organo-derivatives of Bismuth Part 11. The Stability of Derivatives of Quinquevalent Bismuth By FREDERICK CHALLENGER and CHARLES FREDERICK ALLPRESS . V.- -Conversion of E-Phenylchlorortcetic Acid into d-Dipheny l- succinic Acid. By ALEX. MOKENZIE HARRY DUGALD KEITH DREW and GERALD HARGRAVE MARTIN \-1.-2 4-Dichlorophenylhydrazine. By FREDERICK DANIEL CHATTAWAY and CHARLES FREDERICK BYRDE PEARCE VI1.-Investigations on the Dependence of Rotatory Power on Chemical Constitution. Part XI. The Go-ordination of the Rotatory Powers (a) of Menthyl Compounds ( b ) of the Menthones and (c) of the Borneols.By JOSEPE KENYON and ROBERT HOWSON PICKARD . VII1.-The Atomic Weight of Tin. By HENRY VINCENT AIHD BRISCOE . 1X.-Colorations Produced by Some Organic Nitro-compounds with Special Reference to Tetranitromethane. By ERNEST MAGOWAN HARPER and ALEXANDER KILLEN MACBETH . Y.-The Optical Rotatory Power of Derivatives of Succinic Acid in Aqueous Solutions of Inorganic Salts. Part 11. By GEORGE WILLIAN CLOUGH . XI.-Ammonium Perhaloids. By FREDERICK DANIEL CHATTA- XI.-Synthesis of Pinacones. Part 11. By WILLIAN PARRY . XII1.-Investigations on the Dependence of Rotatory Power on Chemical Constitution. Part XII. The Rotatory Powers of Some Eaters of Benzoic and of 1- and 8-Naphthoic Acids with Optically Active Secondary Alcohols.By JOSEPH KENTON and ROBERT HOWSON PICKARD X I V.-a-Hydroxy-P-phenylcrot onolactone. By NORMAN HALL, JAMES EDWARD HYNES and ARTHUR LAPWORTH . X V.- Studies in Optical Superposition. Part IV. 1-Menthyl- amine Tartrates and 1-Amy1 Dimethoxysuccinates. By THOMAS STEWART PATTERSON and DOR~THY CHRISTINA PATTERSON (late Carnegie Research Scholar) . . WAY . . . PAGE 1 7 8 16 26 32 35 63 87 96 105 108 115 132 14 iv CONTENTS. XV1.-The Rate of Saponification of Derivatives of Ethyl Benzoate. By HAMILTON MCCOMBIE and HAROLD ARCHIBALD SCARBOROUGH . . 166 XVI1.-The Behaviour of Colloids towards Pure and Mixed Liquids. Part I. The Systems Caoutchouc-Benzene-Alcohol and Caoutchouc-Benzene-Acetone.By WILLIAM AUGUSTUS CASPARI 162 XVII1.-Hydroaromatic Ketones. Part 111. 1-io Pr pylcyclo- hexan-%one. By ARTHUR WILLIAM CROSSLEY and WALTER RYLEY PRATT . . 171 XIX.-isoQuinoline Derivatives. Part VIII. The Constitution of the Reduction Products of Papaverine (continued). The Constitution of Pavine. By FRANK LEE PYMAN . . 176 XX.-Osmotic Pressure of Alcoholic Solutions. Part I. Vapour Pressures and Densities. By TUDOR WILLIAMS PRICE . . 188 XX1.-Addition of Auxochromes in the Flavone Group. By ARTHUR GEORGE PERKIN and EDWIN ROY WATSON . . 198 XXI1.-The Wet Oxidation of Metals. Part 111. The Corrosion of Lead. By BERTRAM LAMBERT and HERBERT EDWIN CULLIS . . 210 XXII1.-The Wet Oxidation of Metals. Part IV. The Ques- tion of Passivity.By BERTRAM LAMBERT . . 218 XX1V.-The Synthesis of p-Thiol-/3-phenylethylamine. By HAROLD KING . . 222 XXV.-The Oxidation of Aconitine. By GEORGE BARGER and ELLEN FIELD . . 231 XXV1.-Studies in Catalysis. Part 11. The Inversion of Sucrose. By ALFRED LAMBLE and WILLIAM CUDMORE MCCCLLAGR LEWIS . . 233 By EDWARD CHARLES CYRIL BALY and ROBERT ERNEST VICTOR HAMPSON . . 248 XXVII1.-Not e on the Nitroguaiacols. By DAVID CARDWELL and ROBERT ROBINSON . . 255 XX1X.-Azotisation by Chloroamine. By MARTIN ONSLOW FORSTER . . 260 XXX.-A Reaction OP Homopiperonyl and of Homoveratryl Alcohols. By (MRs.) GERTRUDE MAUD ROBINSON . . 267 XXX1.-The Dielectric Constants of Some Organic Solvents a.t their Melting or Boiling Points. By JOHN DOUGLAS CAIJWOOD and WILLIAM ERNEST STEPHEN TURNER .. 276 XXXIL-The Solubility of Carbon Dioxide in Water in the Presence of Starch. Ey ALEXANDER FINDLAY and OWEN RHYS HOWELL . . 282 PAGE XXVI1.-The Constitution of the Aminoazo-compounds CONTENTS. v PAGE XXXII1.-The Velocity of Ionisation at Low Temperatures. XXX1V.-The Firing of Gases by Adiabatic Compression. The Ignition-points of Mixtures of Electrolytic Ratio of the Specific Heat for Nitrogen XXXV.-The Firing of Gases by Adiabatic Compression. Part IV. The Ignition-points of Mixtures of Electrolytic Gas with Carbon Dioxide. Ratio of the Specific Heats for Carbon Dioxide. By JAMES ~\IURRAY CROFTS . . 306 XXXV1.-Volatile Oil from Tubers of hraenipfe~*ia ethela?. By ERNEST GOULDING and OSWALD DIGBY ROBERTS .. 314 XXXVI1.-Researches on Silicon Compounds Part VII 1. The Preparation and Constitution of Silico-oxalic Acids, and the Action of Methyl and Ethyl Alcohols on Trisilicon XXXVII1.-The Velocities of Flame in Mixtures of Methane XXX1X.-Condensation of Acetone and Eenzaldehyde with Glycerol. Preparation of Glycerol a-Methyl Ether. By JAMES COLQUHOUN IRVINE JAMES LESLIE AULD MACDONALD (Carnegie Fellow) and CHARLES W JLLIAM SOUTAR (Carnegie XL.-The Molecular Complexity of Acetic Acid. By GEORGE XL1.-The Organic Phosphoric Acid of Wheat. (Preliminary XLI1.-The Estimation of Potassium by the Perchlorate By RUSSELL GIBSON THIN and ALEXAXDER CHARLES XLII1.-The Removal of Oxygen from Triethylphosphine Oxide. XL1V.-Some Benzopyranol Derivatives. By JOHN NORMAN XLV.-Influence of Sucrose and Alkali Chlorides on the Solvent By JAMES CHARLES PHILIP and ARTHUR XLV1.-The Formation of Coumarin Derivatives and the Pre- paration of Stable Coumarinic Acids.By LOUIS ARNOLD JORDAN and JOCELYN FIELD THORPE . . 387 By ALEXANDER ROBERT NORMAWD . . 285 Part 111. Gas with Argon. and Hydrogen. By JAMES MURRAY CROFTS . . 290 Octachloride. By GEOFFREY MARTIN . . . . . 319 and Air. Part 11. By ALBERT PARKER .. . 398 Scholar) . . . . 337 MACDONALD BENNETT . . 351 Note.) By GEORGE CLARKE . 360 Method. CUMMINU . . 361 By JOHN NORMAN COLLIE and FRANK REYNOLDS . . 367 COLLIE and GERALD NOEL WHITE . . 369 Power of Water.. BRAMLEY . .* . 377 XLVI1.-The Preparation of Ally1 Alcohol. By FREDERICK DANIEL CHATTAWAY . . 407 XLV1II.-Blue Adsorption Compounds of Iodine.Parts 11. and 111. Derivatives of a- and of y-Pyrone. By GEORGE BARGER and WALTER WILLIAM STARLING . . 41 vi CONTENTS. XL1X.-Syntheses of -y-Benzopyrones and Flavones. Part I. By SERGE JACOBSON and EROJENDRA NATH GHOSH . L.-The Constitution of Allantoin and Allied Substances. By HENRY DRYSDALE DAKIR . L1.-Conversion of Racemic Acid into a Mixture of Itacemic and d-Tartaric Acids by Means of LMalic Acid. By ALEX. MCKENZIE . LIL-Configuration of the Stereoisomeric Diphenylsuccinic Acids. By ‘HENRY WREN and CHARLES JAMES STILL . LII1.-Studies in Phototropy and Thermotropy. Part VI. Poly- morphic Vanillylidenearylarnines Produced by Trituration and by the Influence of Actinic Light. By ALFRED SENIER and ROBERT BENJAMIN FORSTER .L1V.-Organic Derivatives of Silicon. Part XXIII. Further Experiments on the So-called Siliconic Acids By JOHN ARTHUR MEADS and FREDERIC STANLEY KIPPING . LV.-The Formation of Diallryloxyanilines in Reduction Pro- cesses. By EUSTACE EBENEZER TURNER . LV1.-Velocity of Crystallisation from Aqueous Solutions. By NORMAN PHILLIPS CAMPBELL . LVI1.-The Reaction Between Calcium Hydroxide and Sulphur in Aqueous Solution. By SAMUEL JAMES MANSON AULD . LVII1.-The Absorption Spectra of Various Halogen and Nitrile Derivatives of Benzene and Toluene as Vapours and in Solution. By JOHN EDWARD PURVIS . L1X.-The Action of Aldehydes on the Grignard Reagent. Part 11. By JOSEPH MARSHALL. . LX.-Derivatives of a New Form of Glucose. By JAMEB COLQUHOUN IRVINE ALEXANDER WALKER FYFE (Carnegie Fellow) and THOMAS PERCIVAL HOGG (Carnegie Scholar) .ANNUAL GENERAL MEETING . OBITUARY NOTICES . LX1.-Mangostin a Crystalline Substance Allied to the Resins. By JOHN ROBERTSHAW H.ILL . LXI1.-Dihydric Alcohols obtained by the Reduction of Mub- stituted Dihydroresorcins. By ARTHUB WILLIAM CROSSLEY and NORA RENOUF . LXII1.-A1 kylated Para-phenylenediamines and Derivatives. Qninoneimide-Ammonium Compounds. By RAPHAEL MELDOLA and WILLIAM FRANCIS HOLLELY . . . LX1V.-The Action of Stannous Chloride on Sulphuric Acid and on Sulphurous Acid. By REGINALD GRAHAM DURRANT. PRESIDENTIAL ADDRESS . PAC4 E 424 434 440 444 452 459 46 9 475 480 49 6 509 524 542 557 578 595 602 610 62 CONTENTS.v11 PAGE LXV.-Studies on the Walden Inversion. Part I. The In- fluence of the Solvent 011 the Sign of the Product i n the Conversion of Phet~ylchloroacetic Acid to Phenylamino- acetic Acid. By GEOEGE SENTER and HARRY DUGALD KEITH DREW . LXVI.-The Absorption Spectra of the Isomerides of Ammonium d-a-Bromocamphor-/3-sulphonate. By JOHN EDWARD PURVIS LXVIL-The Constitution of Internal Diazo-oxides (Diazo- phenols). n y GILBERT T. MORGAN and JAMES WALKER PORTER . LXVII1.-The AbsoL-ption Spectra of the Vapours and Solutions of Anisole Phenetole and Various Derivat'ives. Ey JOHN EDWARD YURVIS . LXIX.-The Relation between Viscosity and Chemical Con- stitution. P a r t IX. The Viscosity and Fluidity of the Aliphatic Acids. By ALBERT ERNEST DUNSTAN . JJXX.-A Method for Distinguishing Tautomeric Isomeric and Polymeric from Polymorphic Substances. By NEVII, VIXCENT SIDGWICK . LXX1.-The Rate of Hydration of Camphoric Anhydride. By BERNARD HOWELL WILSDON and NEVIL VINCERT SIDGWICK . LXXI1.-The Distribution of Energy in the Radiation from the Uviol-glass Lamp. By ARTHUR JOHN ALLMAND . LXXIII.-Derivstives Gf 2-Pyridylhydrazine and 2-Quinolyl- h ydrazine. 13y ROBERT GEORGE FARGHER aud REGINALD FURNESS . LSX1V.-Enantiomorphism of Molecular and Crystal Structure. By WILLIAM BARLOW and FVILLIAM JACKSON POPE . LXXV.-The Racemisation of Phenyl-p-tolylscetic acid. Ey ALEX. MCKENZIE and SIRYL TAITIE WIDDOWS . LXXV1.-The Constitution of Carbamides. Part 11. The Relation of Cyanamide to Urea. The Coustitution of Cyanamide and the Mechanism of its Polymerisation.By EMIL ALPIIOXSE WERNER . LXXVI1.-Steric Influence Static and Dynamic. Part 11. By OLIVER CHARLES MINTY DAVIS and FREDERIC WILLIA&f RIXON . LXXVII1.-Heneicosoic Acid. By HENRY RONDEL LE SUEUR and JOHN CHARLES WITHERS . LXX1X.-Synthesis of a-Naphthapyranols. By EROJENDRA NATH GHOSH . LXXX.-The Valency Volume Theory. By THOMAS VIPOND BARKER . 638 643 6 45 660 667 672 679 682 688 700 702 715 728 736 739 74 ... V l l l CONTENTS. LXXX1.-The Freezing-point Diagrams of Formamide with Water and the Aliphatic Acids and their Bearing on the Interpretation of Viscosity Measurements. By SOLOMON ENGLISH (1851 Exhibition Scholar) and WILLIAJI ERNEST STEPHEN TURXER . . 774 LXXXI1.-Syntheses with the Aid of Monochloromethyl Ether.Part 111. The Action of Monochloromethyl Ether on the Sodium Derivatives of Ethyl Ethane-app-tricarboxylate, Ethyl Butaneaa66-tetracarboxylate and Ethyl Pentane- aaca-tetracarboxylate. By JOHN LIONEL SIMONSEN . . 783 LXXXIII. -Condensation of Ethyl Cganoacetate and Acetyl- acetone. By JOHN LIONEL SINONSEN and MUDLAGIRI NAYAK . . 792 LXXX1V.-Resolution of Externally Compensated p-Toluene- sulphonylalanine into its Optically Active Components. By CHARLES STANLEY GIBSON and JOHN LIONEL SIMONSEN . 798 LXXXV.-Condensation of Acid Chlorides with the Ethyl Ester of ( a ) Cyanoacetic Acid (6) Malonic Acid and (c) Acetoacetic Acid. Part 11. Some Experiments on Ethyl Phthaliminoacetoacetate and Ethyl y-Ethoxyacetoacetnte.By JOHN BRADSHAW HENRY STEPHEN (185 1 Exhibition Scholar) and CHARLES WEIZMANN . . 803 LXXXV1.-The Rotation of isoButyl Diacetpl-d-tartrate. By THOMAB STEWART PATTERSON and DONALD NEIL MCARTEIUR 814 LXXXVII. -The Structure of Methyleneglycerol and Cin- namylideneglycerol. (Preliminary Note.) By DAVID HENRY PEACOCK . . 815 LXXXVIK-Negative Colloidal Ferric Hydroxide. By FRANK POWIS . . . 818 LXXX1X.-A New Method of Finding the Second Dissociation Constants of Dibasic Acids. By ASWINI KUMAR DATTA and NILRATAN DHAR . . * . 824 XC.-The Nitration of 3-Acetylamino-2-methoxytoluene. By JOHN LIONEL SIMONSEN and MUDLAaIRI NAYAK. . . 828 XC1.-The Constituents of Gloriosa superba. By HUBERT WILLIAM BENTLEY CLEWER STANLEY JOSEPH GREEN and FRANK TUTIN .. 835 Nitration Products of 3-Chloro- 5-bromotoluene. By JULIUS BEREND COHEN and WALLACE JENNINGS BZURRAY . . 847 XCII1.-Electromotive Forces in Alcohol Part TI. Absolute Potentials by the Capillary Electrometer.. By EDGAR NEWBERY . . 852 XC1V.-Some Aspects of the Theory of Acids. By ARTHUR LAPWORTH . 851 PAGE XCI1.-Studies in Orient ation CONTENTS. ix XC V.-isoQuercetone By MAXIMILIAN NIERENSTEIR . . 869 YCV1.-The Occurrence of Flavone as the Farina of the Primula. By Huao M~~LLER . . . 872 XCVIL-Phthalides of the Benzene Naphthalene and Carb- azole Series. Part I. By MAURICE COPISAROW and CHARLES WEIZMANN . 878 XCVII I.-The Disposition of the Bonds of the Quinquevalent Nitrogen Atom. Part I. By JULIUS BEREND COHER, XC1X.-A New Method of Estimating Bromine and Chlorine in Organic Compounds.By PHILIP WILFRED ROBERTSON . 902 C.-Studies on the Walden Inversion. Part 11. The Kinetics and Dissociation Constant of Phenylchloroacetic Acid. By GEORGE SENTER . . 908 C1.-The Preparation of Anhydrous Solids. By WILLIAM RINGROSE GELSTON ATKINS and EDITH GERTRUDE WILSON. 916 CI1.-Contributions to the Study of Acenaplithylene and its Derivatives. By EERTRAM CAMPBELL . . 918 (2111.-Studies on Alcoholysis. Part I. Dilatometric Deter- mination of the Velocity of Alcoholysis in the Presence of a Large Excess of Alcohol. By GOPAL BALKRISHNA C1V.-The Formation of Chlorinated Amines by the Reduction of Nitro-compounds. By WINIFRED GRACE HURST and JOCELYN FIELD THORPE . . 934 CV.-Metallo-compounds of Cobalt and Nickel.By SPENCER UNFREVILLE PICKERING . . . 942 CVI. -3letallo-compounds in Solution. By SPENCER UMFREVILLE PICKERIKG . 955 CV 11.-Syntheses of Benzo-y-pyrones and Flavones. Part 11. By SERGE JACOBSON and BROJENDRA NATH GHOSH . . 959 CVII1.-The Absorption Spectra of Various Derivatives of Benzoic Acid. By JOHN EDWARD PURVIS. . . 966 Deputation from the Royal Society and the Chemical Society to the Government re Memorials on the Position of Chemical Industries presented to the Prime Minister March 1915 . 974 C1X.-A Method for the Volumetric Estimation of Lead. By FRANK DOUGLAS MILES . . . . 988 CX.-The Properties of Cold-worked Metals. Part I. The Density of Metallic Filings. By THOMAS MARTIN LOWRY and REGINALD GEORGE PARKER. I . 1005 By MARIAN JONES and JAMES RIDDICK PARTINGTON .. 1019 PAGE JOSEPH MARSHALL and HERBERT ERNEST WOOUMAN . . 887 KOLHATKAR . . 921 (3x1.-Experiments on Supersaturated Solutions X CONTENTS. CXTI. -Yohimbine (Quebrachine). By 0 EORGE BARGER and ELLEN FIELD . . 1025 CXII1.-Dynamics of Isomeric Change. Keto-enol Transforma- tion of Cyanoacetic Acid and its Derivatives. By HARRY MFIDFORTH DAWSON REQINALD SUGDEN and ARTHUR TAYLOII 1030 CX1V.-Nitrocamphor and its Derivatives. Part VIII. The Action of Formamide on Nitrocamphor. By THOMAS MARTIN LOWRY and VICTOR STEELE . . . 1038 CXV.-Researches on Silicon Compounds. Part TX. The Action of A1 kalis and Water on Hexaethoxysilicoethane, Silico-oxalic Acid Mesosilico-oxalic acid and Chloropentn- ethoxysilicoethane together with some Remarks on their Constitution.By GEOFFREY MARTIN. . . 1043 CXV1.-Ryntheses of -y-Benzopyrones and Fhvones. Part 111. By SERGE JACOBSON and BROJENDRA NATH GHoarr . 1051 CXVII. -The Absorption Spectra of Mono-substituted Benzene Componnds and the Benzene Substitution Law. By EDWARD CHARLES CYRIL BALY and FREDERICK GERALD TRYHORN 1058 CXVlI1.-Reactivity of the Halogens in Organic Compounds. Part VIII. Interaction of Alkalis and Alkali Bromo- acetates and Broniopropionntes in Methyl-alcoholic Solution. By GEORGE SENTER and HENRY WOOD . 1070 CXIX.-The Formation and Stability of spiro-Compounds. Part I. spiro-Compounds from cycZoHexane. By RICHARD MOORE BEESLEY CHRISTOPHER KEI.K INGOLD and JOCELYN FIELD THORPE . . 1080 CXX.-A Thermal Study of the Carbonisation Process.By HAROLD HOLLINGS and JOHN WILLIAM COBB . . 1106 CXX1.-Preparation of Diahlorodinitromethane by the Simul- taneous Nitration and Chlorination of Acetone. By JITENDRA NATH RAKSRIT . . 1115 Part 11. Application of Traube’s Atomic Volume Method to Binary Mixtures. By WILLIAM RINGROSE GELSTOX ATRINS and KATHLEEN SHIPSEY . . 1117 CXXII1.-The Relation Between the Infra-red and Ultra-violet Absorption and the Variation in Absorption with Con- centration. By EDWARD CHARLES CYRIL BALY and FREDERICK GERALD TRYHORN . . 1121 CXXIV. -Acetyl Derivatives of the Diphenylthiosernicarbazides. By JAMES LYTTLE n%cKEe . . 1133 CXXV.-Nitro-derivatives of Dinaphthnthioxin. By BROJEN- DRA NATH GHOSH and SAMUEL SMILES . . 1144 PAGE CXXI1.-The Molecular Condition of Mixed Liquids CONTENTS.xi PAGE CXXV1.-The Resolution of Externally Compensated Tetm- hyd ro-/3-naphthaquinaldine into its Optically Active Com- ponents. By CHARLESTANLEY GIBSON and JOHN LIONEL SIMONSEN . . 1148 Part 11. Methods of Measuring Small Changes of Density produced by Annealing. By REGINALD GEORGE PARKER and THOMAS MARTIN LOWRY. 1160 Pilrt VTI. Polymorphic Anisylidenearylamines Produced by Trituration and by the Influence of Actinic Light. By ALFRED SENIER and ROBERT BENJAMIN FORSTER . . 1168 CXXIX.-The Rotatory Dispersive Power of Organic Com- pounds. Part VI. Complex Rotatory Dispersion in Ethyl Tartrate. By THOMAS MARTIN LOWRY and THOMAS WILLIAM DICRSON . . 1173 CXXX.-The Rotatory Dispersive Power of Organic Compounds.Part VII. Complex Rotatory Dispersion in Methyl Tnr- trate. By THOMAS MARTIN LOWRY and HAROLD HELLISG ABEAM . . 1187 CXXX1.-The Rotatory Dispersive Power of Organic Com- pounds. Part VIII. An Exact Definition of Normal and Anomalous Rotatory Dispersion. By THOMAS MARTIN LOWRY . 1195 CXXXI1.-The Solubility of the Nitrophenols and Other Isomeric Disubstitution Products of Benzene. By NEVIL VINCENT SIDGWICK WILLIAN JANES SPURRELL and THOMAS ELLIS DAVIES . . 1802 CXXXIII. -3-Thiolphenanthrene and Some of its Derivatives, CXXX1V.-An Oxidation Product of Yyrogallol. By MAXIMILIAN NIERESSTEIN . 1217 CXXXV.-The Influence of Configuration on the Condensation Reactions of Polyhydroxy-compounds. Part 11. The Effect of Boric Acid on the Conductivity and Specific Rotation of Metliylated Derivatives of Mannitol.By CXXX V1.-The Mechanism of Mutarotation in Aqueous Solu- tion. BY JAMES COLQUHOUN IRVINE and ErTIE STEWART STEELE . . . 1230 CXXXVIL-Action of Grignwd’s Reagent on Hydroxy- ketonic Dyestuffs and their Ethers. By ARABINDA SIRKER . 1241 CXXXVII1.-A New Method of Preparing Chloro- and Bromo- triammino-platinous Haloids (Cleve’s Salts). By LEO ALEXANDROVITSCH TSCHUGAEV . . 1247 CXXVIL-The Properties of Cold-worked Metals. CXXVII1.-Studies in Phototropy and Thermotropy. By ELLEN FIELD . . 1214 JAMES COLQUHOUN IRVINE and ETTIE STEWART STEELE . 122 xii CONTENTS. CSXX1X.-Interaction of Dimercuriammonium Nitrite and the Alkyl Iodides. By PRAFULLA CHANDRA RAY . . 1251 CXL.-Synthesis of a-Tetronic Acid.By LEONORE KLETZ and ARTHUR LAPWORTH . 1254 CXL1.-Studies on Cerium Compounds. Part I. Basic Ceric Sulphates and the Colour of Cerium Dioxide. By JAMES FREDERICK SPENCER . . 1265 CXLI1.-Condensation of Chloromethyl Ether with a-Alkyl- acetoacetic Esters. By ARTHUR LAP WORTH and BENJAMIN STANLEY MELLOR . . 1273 CXL1II.-Note on the Basic Copper Formates. By GEORGE FOWLES . . 1281 ANNUAL REPORT OF THE INTERNATIONAL COMMITTEE ON ATOMIC WEIGHTS 1916 . . 1282 CXL1V.-The Action of Bromine on the Alkali Iodides. By WILLIAM NORMAN RAE . . 1286 CXLV.-Non-aromatic Diazonium Salts. Part IV. Thiazole- 2-diazonium Salts. By GILBERT T. MORGAN and GENEVIEVE VIOLET MORROW . . 1291 CXLV1.-The Absorption Spectra of Certain Aromatic Nitro- amines and Nitroamides.Part 11. By GILBERT T. MORUAN HENRY WEBSTER Moss and JAMES WALKER PORTER . . 1296 CXLVI1.-The Composition of Coal. Part 111. By DAVID TREVOR JONES and RICHARD VERNON WHEELER . . 1318 CXLVII1.-Extinction Measurements of Solutions of Sul- phurous Acid Sulphites Hydrogen Sulphites and Meta- bisulphites and of Gaseous and Liquid Sulphur Dioxide. By CHARLESCOTT GARRETT . . 1324 CXL1X.-The Action of Alkalis on Dextrose and Laevulose. By CKARLES WTLFRID ROBERTS POWELL . . 1335 CL.-The Formation of Heterocyclic Compounds from Cyano- acetamide and Hydroxymethylene Ketones. Part I. By HEMENDRA KUMAR SEN-GUPTA . . 1347 CL1.-Complex Metallic Ammines. Part I. cis-Sulphonyldi- acetatodiethylenediaminecobaltic Salts. By THOMAS SLATER PRICE and SIDNEY ALBERT BRAZIER .. 1367 CLI1.-Syntheses of Derivatives of 3-0xy( 1)thionaphthen. CLII1.-New Halogen Derivatives of Camphor. Part I. a'-Chlorocamphor ; with a Note on Isomerism Static and Dynamic. CL1V.-Metallic Derivatives and Constitution of Guanidine. PAGE By SAMUEL SMILES and BROJENDRA NATH GHOSH . . 1377 By THOMAS MARTIN LOWRY and VICTOR STEELE 1382 By HANS KRALL . e . 139 .I. CONTENTS. X l l l PAGE CLV.-The Study of the Density and Viscosity of Aqueous Solutions with Special Reference to Nitric Acid. Part I. Densities. By WILLIAM RORERT BOUSFIELD K.C. . . 1405 CLV1.-The Dual Theory of Acid Catalysis. The Catalytic Activity of Monochloroacetic Acid in Presence of its Salts. By HARRY MEDFORTH DAWSON and CLARENCE KENWORTI-IY REIMAN . . 1426 OLVI1.-The Root Bark of Calotropis gigantea.By ERNEST GEORGE HILL and ANNODA PRASAD SIRKAR . . 1437 CLVII1.-Condensation of Aromatic Hydroxyaldehydes with Diketohydrindene. By SOSALE GARALAPURY SASTRY and Part IT. Some Derivatives of the Optically Active Diphenylsuccinic Acids. By HENRY WREN and CEARLES JAMES STILL. . 1449 CLX. -The Behavioiir of Nitriles towards Organometallic Deriva- tives of Magnesium. By EUSTACE EBENEZER TURNER . 1459 B! 1465 CLX1.-The Volatile Oil of Cgmbopogon sennaarensis Chiov. OSWALD DIGBY ROBERTS , CLXI1.-Contributions to the Theory of Solutions. The In- fluence of Change of Volume on Specific Refraction in Mixtures of Liquids. By JOHN HOLMES . . 1471 CLXII1.-Conversion of the Natural Flavone Colouring Matters into Pyranol Dyes. By EDWIN ROY WATSON KUMUD CLX1V.-The Preparation of Ethyl Bromide.By FRANK EDWIN WESTON. . 1489 CLXV.-The Action of Diazomethane on Some Aromatic Acyl Chlorides. By DOUGLAS ARTHUR CLIBBENS and MAXIMILIAN NIERENSTEIN . 1491 CLXV1.-Studies in Ring Formation. Part I. The Kaufler Formula for Derivatives of Diphenyl. By EUSTACE EBENEZER TURNER . . 2495 CLXVI1.-A Comparative Method for Determining Vapour Densities. By PHILIP BLACKMAN . 1500 CLX VIII. -Polymorphic Ph thalyl-halogen-substi tuted-phenyl- and -tolyl-hydrazides. By FREDERICK DANIEL CHATTAWAY and ERNST VONDERWAHL . . 1503 CLXTX.-'Tht Optical Rotatory Power of Derivatives of Succinic Acid in Aqueous Solutions of Inorganic Salts. Part 111. By GEORGE WILLIAM CLOUGH . 1509 CLXX.-Electromotive Forces in Alcohol.Part VII. Concen- tration Cells with Calomel Electrodes. By EDGAR NE WBERY . . 1520 BROJENDRA NATH GHOSH . 1442 CLTX.-Studies in the Phenylsuccinic Acid Series. BEHARI SEN and VISHNU RAM MEDEI . . 147 xiv CONTENTS, CLXX1.-Tapour-pressure Investigations of the Fusion Products of Iodine with Sulphip~ Selenium and Tellurium. By ROBERT WRIGHT . . 1527 CLXXI1.-Experiments on the Corrosion of Molybdenum Steel. By LESLIE AI'~CHISON . . 1531 CLXXII1.-The Hydrolysis of Sodium Phenoxides in Aqueous Solution. By DAVID RUNCIMAN BOYD . . 1538 CLXX1V.-Rotatory Power and Refractivity. Part 11. The Rotatory Powers Refractivities and Molecular Solution Volumes of Camphor Bromocamphor and Ethyl Tartrate CLXXV.-The Nature of the Vibrations Causing the Colour of Dyes.CLXXV1.-The Effect of Additional Aaxochromes on the Colour of Dyes. Part I. Phthalein and Benzein Dyes. By VISHNU RAM MEDHI and ~ D W I N ROY WATSON . . 1579 CLXXVI1.-The Thermal Decomposition of Hydrogenated Aromatic Hydrocarbons. Dy DAVID TREVOR JONES. . 1582 CLXXVI1I.The Influence of Constitution on the Basic Property of Oxygen. Part I. By BROJENDRA NATH GHOSH 1588 CLXX1X.-A Study in the Coumarin Condensation By BIMAN BIHARI DEY . . 1606 CLXXX.-The Determination of the Concentration of Hydroxyl Ions. By FRANCIS FRANCIS FRANCIS HENRY GEAKE and JAMES WILLIAM ROCHE . . 1651 CLXXX1.-The Action of Alkalis on the Nitrosoarnines of 4-Piperidone Derivatives. By ERIC DODDRELL EVENS, EDGAR CRANTHORNE GIFFORD and WALTER EDWARD LAMBORNE GRIFFITHS . . 1673 CLXXXI1.-The Interaction of Perchloric Acid and Potassium Sulphate as an Example of Reversible Change.By WILLIAM ALFRED DAVIS . . 1678 CLXXXIIL-The Displacement of Halogen in Optically Active Phenylhalogenacetic Acids by the Anilino-group. By ALEX. MCKENZIE and STAXLEY CEARLES BATE . . 1681 CLXXX1V.-Experiments on the Walden Inversion. Part X. Displacement Reactions with 1-Phenylbromoacetic Acid. By ALEX. MCKENZIE and NELLIE WALKER . 1685 CLXXXV.-The Formation and Preparation of Glucosmono- acetone. By JAMES COLQUHOUN ~ R V I N E and JAMES LESLIE AULD MACDONALD . . 1701 CLXXXV1.-A Laboratory Circulating Pump. Ry JOHN STANLEY MORGAN . . 1710 PAGE in Certain Solvents. By DAVID HENRY PEACOCK . . 1547 By EDWIN ROY WATSON and DAVID B. MEEK . 156 CONTENTS.xv PAGE CLXXXVII.-3-gem-IDiinethylpiperidine. By the late JOHN GUNNING MOORE DUNLOP . . 1712 CLXXXVIII. -Complex Metallic Ammines. Part 11. Addi- tive Compounds Formed from truns-Dichlorodiethylenedi- aminecobaltic Chloride. By THOMAS SLATER PRICE and SIDNEY ALBERT BRAZIER . . 1713 CLXXX1X.-Contribution to Our Knowledge of the Semi- carbazones. Part V. Sernicarbazones of Benzaldehyde and Some of its Substitution Products. By JAMES ALEXANDER RUSSELL HENDERSON and ISIDOR MORRIS HEILBRON . . 1740 CXC1.-A Decomposition of Certain o-Nitromandelic Acids. By GERTRUDE MAUD ROBINSON and ROBERT ROBINSON . . 1753 C YC1.-The Indirect Determination of Velocity of Hydrolysis by the Polsrimetric Mebhod. By JAMES CODRINGTON CROCKER . . 1762 CJXCI1.-Monotropic Polymorphic Anilides. By FREDERICK DANIEL CHATTAWAY and WILLIAM JAMES LAMBERT . . 1766 CXCIIL-The Transition Points of the Polymorphic Phthalyl- hydrazides. By FREDERICK DANIEL CHATTAWAY and WILLIAM JAMES LAMBERT . . 1773 CXC1V.-The Study of the Density and Viscosity of Aqueous Solutions with Special Reference to Nitric Acid. Part 11. Viscosites. By WILLIAM ROBERT BOUSFIELD K.C. . . 1781 CXCV.-The Vapour Pressures of Some Saturated Aqueous Solutions. By MALCOLM PERCIVAL APPLEBEY and WILLIAM HUGHES . . 1798 CXCV1.-N-Halogen Derivatives of the paru-Halogen-substi- tuted Benzenesulphonamides. By ROBERT REGIKALD BAXTER and FREDERICK DANIEL CHATTAWAY . . 1814 CXCVTI.-Coloi.ations produced by Some Organic Nitro-com- pounds with Special Reference to Tetranitrorrietharie. Part 11. By QLEXANDER KILLEN IfACBETEI . . 1884 UXCVII1.-The Catalytic Bleaching of Palm Oil. By SOSALE GaRALAPURY SASTRY . . 1828 CXC1X.-Influence of Glycerol Dextrose Alkali Nitrates and Sulphates and Ammonium Salts on the Solvent Power of Water. Lecture Delivered before the Chemical Society on November 18th 1915. By EDWARD JOHN RUSSELL . . 1838 CC.-The Isomerism of the Oximes. Part VII. 5-bOmO- vanillinoxime 5-Nitrovnnillinoxime and 6-Nitropiperonal- oxime. By OSCAR LISLE BRADY and FREDERICK PERCY DUNN 1858 By JAMES CHARLES PHILIP and ARTHUR BRAMLEY 1831 THE PRINCIPLES OF CROP PRODUCTION

ISSN:0368-1645    

DOI:10.1039/CT91507FP001

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

TUTIN : ISODIBENZO~YLGLUCOXYLOSE. 711. -isoDibelzzoyZgZucoxyZo~e.By FRANK TUTIN.IN recent communications (T., 1914, 105, 767, 1062) Power andSalway described the isolation from the leaves of Daviesiu latifolkof a new, crystalline, bitter substance, dibenzoylglucoxylose. Thissubstance proved to be the dibenzoyl derivative of a newdisaccharide, glucoxylme, and represented a class of compoundnot previously known to occur in nature.The present author recently had occasion to prepare a consider-able quantity of dibenzoylglucoxylose from the material containedin the mother liquors of the product isolated by Power and Salway.During the course of the purification of this material it was notedthat the crystalline product was not homogeneous. It was there-fore subjected to a prolonged process of fractional crystallisationfrom water, from dilute alcohol, and from ethyl acetate.By thismeans a substance was eventually isolated which melted 20°higher than dibenzoylglucoxylose, and, since it is isomeric with thelatter, i t has been named isodibenzoylglucosylose. The amount ofthis substance isolated in a state of purity was not large, owingto the difficulty of separating it from dibenzoylglucoxylose, but itwas evident that an appreciable quantity of i t was present.asoDibenzoylglucoxylo-se crystallises from water or ethyl acetatein colourless needles, and is more sparingly soluble in these solventsthan dibenzoylglucoxylose. It melts a t 173-174O, and a mixtureof i t with the latter (m. p. 152-153O*) melts a t 138-143O:0'1106 gave 0.2340 CO, and 0.0575 H,O.0.2102, made up to 20 C.C.with methyl alcohol, gave a, -Oo8'in a 2-dcm. tube, whence [aID -6-3O.* Power and Salway (loc. cit.) gave the melting point of dibenzoylglucoxylose as14'7-148", but the present author finds that after this material has been crystalli~edtwo or three times from ethyl acetate the melting point is raised to 152-153 .0=57.7; H=5*7.c125H28012 requires C = 57.7 ; H = 5.4 per cent8 HAWORTH: A NEW METHOD OFThe specific rotation of dibenzoylglucoxylose was determinedunder identical conditions, and found to be [a], -105.9O.It is evident, therefore, that the new substance is an isomerideof dibenzoylglucoxylose. Its properties are similar t o those of thelatter compound, and i t poesesses a bitter taste.Penta-ucetylisodib enzo ylglucoxylose was prepared, but the amountavailable was too small for analysis.It is insoluble in water, verysparingly so in alcohol, but very readily soluble in ethyl acetate.From a mixture of the last-mentioned two' solvents it crystallisesin small, colourless needleg, which melt a t 173-174O.I n their first communication (loc. cit., p. 773) Power and Salwayrecorde'd the isolation from Baviesia Zatifolia of a small quantityof a quercetin glucoside which was considered to be probablyidentical with rutin. This small amount of substance, however,appeamd to differ from rutin in two respeck, namely, it was statedto ble retadily soluble in alcohol, and to contain four, instead ofthree, molecules of wa,ter of crystallisation. A considerablequantity of this glucoside (about 25 grams) has now been isolated.After recrystallisation from hot water it was found to1 be verysparingly soluble in alcohol, and was identical with rutin in allits properties. On analysis, the air-dried substance1 was found t opossess the formula C,7H,,0,,,3H,0. (Found, C = 48.8 ; H = 5-5.Cdc., C=48*8; H=5*4 per cent.)TEE WELLCOME CHEMICAL RESEARCH LABORATORIES,LONDON, E.C

ISSN:0368-1645    

DOI:10.1039/CT9150700007

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

8 HAWORTH: A NEW METHOD OF111.-A New Method of Pq-eparing Alkylated Sugars.By WALTER NORMAN HAWORTH.THE investigations conducted in these laboratories during the pasttwelve years on the alkylated sugars have been directed to theelucidation of the constitut,ion of the simple carbohydrates andtheir derivatives.The utility of these sugar ethers has hitherto been restricted bythe comparative inaccessibility of the compounds, due t o the labourand expense involved in their preparation. For this reason thedetailed study of tho1 alkylated disaccharides has long been deferred,whilst the application of the alkylated monosaccharides for purposesof synthetic work has been difficult of realisation. Similarly, thegeneral investigation of the action of enzymes on methylatePREPARING ALKYLATED SUGARS.9sugars, on a scale adequate t o the importance of the subject, hashad t o be postponed until a simplification in the mode of prepara-tion could be introduced.The present research was undertaken with a view to render thesesugar derivatives more reladily available for the purposes enumer-ated. This intention has been realised in so far as it is nowpossible, by applying the working methods hereafter described, toexpedite the production of the alkylated sugars, and also to preparethem in greater variety and a t an expense which corresponds witha reduction of the former cost by about four-fifths.The procedure hitherto adopted for the preparation of thealkylated sugars depends upon thel employment of the expensivealkyl iodides and dry silver oxide (Purdie and Irvine, T., 1903,83, 1021); and, moreover, the necessity for finding a suitablesolvent for the sugar, in the initial stages of the reaction, furthercircumscribes the applicability and efficiency of the method.Thechoice of these reagents has, however, been abundantly justified bythe results, since it has been established that, under these mildexperimental conditions, profound changes, such as racemisation,the Walden inversion, or the interconversion of glucosides, do notoccur. Hence these results have been used as a standard for com-parison in the present work, wherein somewhat more drasticreagents have found applica,tion.It has now been observed that the substitution of alkyl sulphatesand commercial sodium hydroxide: in an aqueous medium, for theexpensive reagents formerly used, is attended with many advan-tages; and whilst the success of the operation is governed to aremarkable degree by the conditions of the experiment, no detri-mental effect due to the introduction of alkali hydroxide isapparent, the sugar derivatives isolated having the optimum rota-tion values in every case where a comparison with known values ispossible.I n adapting the experimental Conditions to the requirements ofthe sugar group the guiding principles to be followed were, first,that during thO alkylation process the existence, even locally, ofan acid system should 6e avoided, as this, even a t comparativelylow temperatures, tends t o induce changes of a hydrolytic character.Secondly, at no stage of the operations should there be a stronglyalkaline system, as in this case the possibility of the formationof complex enolic derivatives, resins, and polysaccharides presentsitself (Nef, AnnuZen, 1914, 403, 204).The conditions have beensatisfactorily adjusted t o meet these requirements, and the completeexperimental details are described on p. 11.Examples representing almost every type of optically activ10 HAWORTH: A NEW METHOD OFhydroxyl compound in the aliphatic series have been chosen, andthe derivatives obtmained from these are tabulated below:Sucrose + heptamethyl and octamethyl sucrose.Methyl glucoside + tetramethyl, trimethyl, and (probably)Methyl mannoside + tetramethyl, trimethyl, and probablyMethyl galactoside + tetramethyl, trimethyl, and probablySalicin + dimethyl, and probably trimethyl and tetramethylTartaric acid + a-hydroxy-P-me thoxysuccinic acid.Dextrose -+ alkylated methylglucosides.Lactose --+ alkylated lactosides.Mannitul‘-+ various alkylated mannitols up to the pentamethylderivative.The derivatives of mannitol will be referred to fully in a subse-quent paper, but it may here be mentioned that in this case, aswell as in others now described, i t has been substantiated that theintroduction of methyl groups in a sugar-chain proceeds in definitestages, certain hydroxyl groups displaying a tendency to undergoetlierification preferentially,.and that homogeneous products,representing the intermediate stages in the alkylation process, maybe isolated and definitely characterised.I n this respect the presentmode of preparation possesses an advantage, in that the use ofsilver oxide appears to result in a mixture of unchanged and partlyalkylated material, which is complex and difficult, to separate.Although the account of the method applied in the course ofthis work refers exclusively to methyl sulphate, it is intended toapply it t o “ethylation,” and to the use of the higher alkylsulphates, in the hope of obtaining the higher homologues of thesugar ethers, which may conceivably be capable of better character-isation than the methyl derivatives. The preparation of thesehigher ethers by the silver oxide method is excluded by the factthat a t the boiling point of the higher alkyl iodides silver oxidefunctions as an oxidising agent.For t.he first time the completely methylated sucrose, namely,octamethyl sucrose, has been obtained in a condition sufficientlypure to admit of its characterisation by physical methods.Curi-ously enough, the reagents recommended in this paper only carrythe alkylation of sucrose to the stage of the heptamethyl compound,and this has been carefully described. Whilst this is so, yet thisstage is reached in one alkylation, when the product becomessparingly soluble in sodium hydroxide, and forms an oily layerdimethyl methylglucoside.dimethyl methylmannoside.dimethyl methylgalactoside.salicinPREPARING ALKYLATED SIIGARS. 11above the solution.It is necessary, in order to advance thematerial t o the octamethyl stage, to have recourse to the use ofmethyl iodide and silver oxide for the introduction of the onerernaining me€hoxyl group. Reptamethyl sucrose readily dissolvesin methyl iodide, and only two treatments with this reagent arerequired.EXPERIMENTAL.The apparatus designed for the process of alkylation with methylsulphate and sodium hydroxide consists in a wide-necked flaskfitted with a cork, and provided with two dropping funnels, a watercondenser, and an arrangement for stirring mechanically. Thesugar or sugar derivative is dissolved in the minimum amount ofwater, and the flask surrounded by a water-bath, which is main-tained a t 70°. An excess of the alkylating reagents, amounting tothree times the quantity theoretically required, is then addedslowly through the two dropping funnels, one of which is reservedfor methyl sulphate, and the other for a 30 per cent.solution ofcommercial sodium hydroxide, the mixture being vigorously stirredduring the operation, which is complete in the course of an hour;subsequently the bmperature of the bath is raised to looo for half-an-hour. The rate of admission of either reagent t o the vessel inwhich the reaction is proceeding is determined by the fact that aslight alkalinity must be maintained throughout. This conditionshould likewise obtain a t the end of the reaction, in order thatany unchanged methyl sulphate may be destroyed during theboiling operation, and for this reason the proportion of alkali usedshould be slightly greater than that of methyl sulphate.The product, on cooling, is extracted twice with chloroform, theextract dried, and the chloroform distilled.The residue may besubjected to a similar alkylation process if this is necessary for Dhepurpose in hand, half the previous quantities being employed. I nmost cases the experience has been that the material remainingin the aquezus residues after extraction with chloroform is practi-cally negligible. This partly methylated residue may also be dealtwith by concentrating the aqueous solution which has previouslybeen rendered neutral, filtering the separated salt through linen,and submitting the filtrate t o a further treatment with the alkylat-ing reagents.The quantities of the reagents requish for thispurpose may be approximately judged by estimating the amountof material already recovered in the chloroform extracts12 HAWORTH: A NEW METHOD OFMetAylation of Sucrose : Heptarnethyl Sucrose,C,,€I,,O,(OH)(OMe),, and Octarnethtyl Sucrose, C,,H,,03(OMe),.From 20 grams of pure sucrose, 15 grams of the heptamethylether were obtained after one alkylation. This distilled constantlya t 191-195°/0.18 mm. as a colourless, viscid syrup, and gave thefollowing analytical results :0.1921 gave 0.3633 CO, and 0.1418 H,O. C= 51.58 ; H= 8-20.C,,H3601, requires C = 51.81 ; H = 8.18 ; OMe = 49.3 per cent.0.1874 ,, 0.7180 AgI. OMe=50*5.The physical constants were determined as below :nD 1.4656; Dfj 1.1654; M, 104.49 (M, calc.=103*49).[aJD + 68'5O in methyl alcohol ( c =5*585).Using methyl sulphate as the reagent it was not found possibleto increase appreciably the methoxyl content of heptamethylsucrose.This seems t o be due to the sparing solubility of the etherin sodium hydroxide. It was therefore found necessary to revert tothe use of methyl iodide, in which heptamethyl sucrose readilydissolves, and two treatments with this reagent and silver oxidewere necessary to methylate the remaining hydroxyl group in themolecule. The effect of the introduction of this last methyl groupwas markedly shown in the diminution of the boiling point and ofthe refractive index, and an increase in the mobility of the product.This was isolated as a slightly viscid, colourless liquid, distillingconstantly at 176O/0*05 mm.:0-1622 gave 0.3133 CO, and 0.1242 H,O. C=52.68 ; H = 8-50.0.1306 7 7 0.5355 AgI. OMe=54.12.C20H38011 requires C = 52-86 ; H = 8.37 ; OMe= 54.6 per cent.The physical constants determined are as follows :n, 1-4588; DY 1.1406; M, 108.77 (M, calc.=108-23).[ a ] , in methyl alcohol ( c =7.345) + 69.3O.[a], in acetone (c= 6.782) + 66.8O.It will be Seen by comparison of the optical and analytical datawith the value8 given in a previous communication by Purdie andIrvine (T., 1905, 87, l028), that the material has now beenobtained in a condition of purity. A larger quantity of the sub-stance has since been prepared with a view to the determinationof the constitution of sucrose from an examination of the hydro-lytic products of the octamethyl etherPREPARING ALKYLATED SUGARS.13llethylation of a-Methyl Glucoside : Tetramethyl a-Methylgluco-side, C6H70(OMe),, and the Corresponding Sugar, and ofTrimethyl a-Methylglucoside, C,H80,(OMe),.The chloroform eoluble extract (20 grams), obtained by a singlealkylation of 25 grams of methyl glucoside, was fractionated intothree portions. The first of these was pure tekamethyl a-methyl-glucoside, which distilled a t 108O/0*1 mm. as a colourless, mobileoil :nD 1.4454 ; D:" 1.1082 * ; M, = 60.14 (M, calc. = 60.65).The identity of this fraction with the substance previouslydescribed by Purdie and Irvine (T., 1904, 85, 1059) was provedby its conversion into the very characteristic, crystalline sugar,tetran?,ethyl glucose.The hydrolysis was effected in the usual way,and the free sugar obtained in long, colourless needles, melting a t84O. This showed mutarotation, and finally attained the value[a], +84*5G in aqueous solution (c=1*6), which is in good agree-ment with the rotation previously recorded (Zoc. cit.) (Found,OMe = 50.84. C1,Et2,06 requires OMe= 52.54 per cent.).The refractive index of the superfused sugar was determined,and this was identical with the value shown by a specimen preparedby Purdie and Irvine, namely, %=1*4588.The second fraction, distilling a t 130°/0*13 mm., was a viscid,colourless oil ; the analytical, physical, and other data indicate thatthis was trimethyl a-methylglucoside (compare Purdie and Bridgett,T., 1903, 83, 1037) (Found, OMe=51*21.C1,H,06 requiresOMe = 52.54 per cent.) :n,, 1.4583; D:' 1-158; M, 55.63 (MD calc.=55'91).[a], + 160.3O in methyl alcohol (c=2.823).The third fraction was very viscid, but quite colourless, anddistilled a t 140-150°/0~07 mm., and, from its methoxyl contentand refractive index (nD 1-4648), it would appear to be a mixtureof dimethyl and trimethyl a-methylglucosides (Found, OMe = 45.5.C,H,O,(OMe), requires OMe = 41.9 ; C6H80,(OMe), requires OMe=52.5 per cert.).Methylation of a-Methyl Mmznoside : Tetramethyl a-Methyl-mannoside and Trim e t hy 1 a-Me t h y lmannosid e.One alkylation of 14 grams of a-methyl mannoside yielded10 grams of material soluble in chloroform, which was twice frac-tionated under diminished pressure.The first fraction, distillingat 108-110°/0*1 mm. and having nD = 1.4497, crystallised corn-* Purdie and Irvine's value14 HAWOHTH: A NEW METHOD OFpletely on nucleating with a specimen of tetramethyl a-msthyl-mannoside, and melted a t 37--38O, no depression of the meltingpoint being observed when the crystals were mixed with those of thespecimen previously obtained by Irvine and Moodie (T,, 1905, 87,1465). Its rotation in ethyl alcohol ( c =4.475) showed [a], + 76*4O,the value previously recorded being + 75.5O.The second fraction distilled at 150°/12 mm. as a viscid, colour-less oil, showing nD= 1.4583 :0.1147 gave 0.4538 AgI. OMe=52*2.CBH80,(OMe), requires OMe = 52.5 per cent.The third fraction, distilling a t 155--162O/12 mm.and showingnD = 1.4650, corresponds with the similar fraction obtained onalkylating a-methyl glucoside. It consists of a mixture of dimethyland trimethyl a-methylmannosides.Met hylation of Met hylgalactoside : Tetrame thy1 Met hylgaluct oaideand Trimethyl Methylgalactoside.The initial material consisted of a mixture of about equalamounts of a- and &methyl galactosides. The chloroform extractyielded on distillation three fractions.The first of these distilled a t 135-140°/12 mm. as a colourlm,mobile oil, having n,=1.4520, and this gave analytical figurescorresponding with tetramethyl methylgalactoside (compare Irvineand Cameron, T., 1904, 86, 1071j (Found, OMe=60-2. C11HB06requires OMe = 62.0 per cent.).Fraction 11, distilling a t 150--154O/12 mm., was slightly moreviscid than the above, and showed n,=1.4568 and [a],, +48.7O inethyl alcohol (c=3*1).This material consisted of trimethyl a- andP-methylgalact;osides (Found, OMe = 52.6. . C,H80,(OMe), requiresOMe =52*5 per cent.).The third fraction, which was small in amount, distilled about165O/ 12 mm., and showed n, = 1.4640. A methoxyl estimation indi-cated that this contained dimethyl and trimethyl a- and &methyl-galactosides.Methylation of Sdicin : Uimethyl Salicin, C,,H,,02(0H),(OMe)2.The chief product isolat'ed as the result of one treatment ofsalicin (10 grams) with the alkylating agents appears to be thedimethyl ether. This crystallised on evaporation of the chloroformextaact, and after draining on porous tile and recrystallising froma mixture of ethyl acetate and light petroleum, it melted a t 122O.It crystallises in slender, colourless prisms, and is lavorotatory,having about the same specific rotation as salicinPREPARING ALKYLATED SUGARS.I50.0868 gave 0.1821 CO, and 0.0560 H,O. C = 57-21 ; I3 = 7.17.0'1310 ,, 0.2055 AgI. OMe=20'7.C,,H,,O, requiree Cr = 57.32 ; €1: = 7.01 ; OMe = 19.7 per cent.On extraction of the porous tiles with chloroform, and evapora-tion of the extract, a syrupy residue remained, which failed tocrystailim. This distilled under 0.2 mm. pressure (bath a t 2 2 0 O )as a colourless syrup, which slowly deposited crystals. These were,however, too small in amount to purify, but a methoxyl estimationindicabd b e presence of three or four methoxyl groups in thesalicin molecule, thus showing that alkylation had proceededbeyond the dimethyl ether stage (compare Irvine and Rose, T.,1906, 89, 814).Methylation of Tartaric Acid: a-Eydroxy-8-methoxysuccinic Acid,CO,H-CH (OH) *CH(OMe) *CO,H.The application of the method of alkylation recommended inthis paper tc, the case of tartaric acid yielded a somewhat unex-pected result, in that the only product isolated in a condition ofpurity was a-hydroxy-fl-methxyszlccinic acid.In order to isolatethe material formed in this alkylation process, it was necessaryto neutralise the alkaline solution with sulphuric acid and evaporateit on the water-bath. The concentrated liquid was acidified andextracted with ether for a prolonged period by means of theHagemann apparatus.The ethereal extract yieIded a syrup whicheventually crystallised. The solid was drained on porous tile andrecrystallised from much ether, in which it is sparingly soluble. Itforms flat, colourleas prisms, melting a t 174*, and having[a], + 45-4O in aqueous solution ( G = 1.85).The specific rotation is thus the mean of the values recorded fortartaric acid, +14.52O (Landolt, Ber., 1873, 6, 1075), and di-methoxysuccinic acid, + 76*63O (Purdie and Irvine, T., 1901, 79,962). The addition of ammonium molybdate t o the aqueous solu-tion effects an exaltation of the rotation value:0.C433 gave 0.0590 A@. OMe = 17.6.C5H806 requires OMe= 18.9 per cent.Titration .- 0.1850 required 0*0885 NaOH for neutralisation,whilst this amount of an acid, Ct,H,O,, ahould require 0.0896NaOH.Methyhtior of Dextrose and Lactose.As examples of the alkylation of the reducing sugars, dextroseand lactose were selected.The temperature in the initial stagesof the reaction was, in them cases, maintained a t 50°. Productsderived from both these sugars were isolated in the usual way, an16 CHALLENGER AND ALLPRESS :these were distilled and found to be glucosidic in character. It istherefore established that the alkylated glucosidea, probably mix-tures of a- and &forms, can be prepared from the parent sugardirectly by the methyl sulphate process, although the yield is notquite so good as in the case of the preparation of ethers from theglucosides. The product derived from dextrose distils a t the sametemperature as tetramethyl methylglucoside, and has the samerefractive index. The product obtained from lactose distils a tabout 170°/0*05 mm., and has' nD 1.4588, and therefore is com-parable with octamethyl sucrose. @Methyl glucoside has, appa-rently, been prepared by Maquenne (Bull. SOC. cfiim., 1905, [%I,33, 469) by the agency of methyl sulphate, but, so far as one isaware, no attempt to prepare the alkylated glucosides by thismethod has before been made.The author wishes to reserve the substances described in thispaper f o r further work. He also desires t o thank the CarnegieTrust for a research grant for materials, and to express his indebted-ness to Prof. Irvine for valuable advice, and to Miss Grace C.Leitch f o r her assistance.UNITED COLLEGE OF ST. SALVATOR AND ST. LEONAPD,UNIVERSITY OF ST. ANDREWB

ISSN:0368-1645    

DOI:10.1039/CT9150700008

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

16 CHALLENGER AND ALLPRESS :IV.-Organo-derivatives of Bismuth. Part I . . TheStability of Derivatives o f Quinquevalent Bismuth.By FREDERICK CHALLENGER and CHARLES FREDERICK ALLPRESS.IN a previous communication (T., 1914, 105, 2211) it was statedthat certain derivatives of quinquevalent bismuth are characterisedby the ease with which they can be degraded to compounds inwhich bismuth functions as a triad. For example, it was not foundpossible to prepare tetraphenylbisrnuthonium bromide or triphenyl-ethylbismuthonium bromide; moreover, triphenylbismuthine di-bromide and tri-a-naphthylbismuthins dibromide readily eliminatearyl bromide under the influence of heat.This led to an invMtigation of the stability of tertiary bis-muthine dihaloids containing dissimilar halogen atoms, since thequestion as to which of these halogens would be eliminated in unionwith an aryl radicle seemed to be of interest.Attempts were therefore made to prepare additive products oftriphenyl- and tri-a-naphthyl-bismuthine, with iodine, iodinOKGANO-DERIVATIVES OF I3ISMUTIi. PAK'P 11.17cliloricle, and iodine bromide, and also o€ triplienylbismutliiiie withcyanogen, cyanogen iodide, cyanogen bromide, and cyanamide.when triphenylbismuthine and iodine monobromide or iodinemonochloride interact, no additive product can be isolated, butiodobenzene and either diphenylbromobismuthine or diphenyl-chlorobismuthine (see p. 19) are produced in yields whichapproach closely to those demanded by thO equation:BiPh, + IX =BiPbX + PhI.Uhlorobenzene or bromobenzene could not be detected.Similarly,cyanogen iodide gives iodobenzene and diphenylcyano bismuthine,which decomposes a t a5out 2100.The reaction of triphenylbismuthine with iodine is somewhatcomplicated, and certain points in this connexion yet remain t o beelucidated, but owing to the inability of one of us t o cotntinue theinvestigation the results obtained up t o the present may briefly besummarised.I n ether 6r light petroleum solution no di-iodide appears to beformed (compare Gillmeister, Ber., 1897, 30, 2843), the productsof the reaction being diphenyliodobismuthine, iodobenzene, and ared powder containing much organic matter and more than 50 percent. of iodine.The red cow,pound is also obtained when iodine is added tjo asolution of diphenyliodobismuthine, so is probably a product of asecondary reaction, It cannot be triphenylbismuthine di-iodide,since this contains only about 26 per cent.of iodine.Moreover, the fact of tho formation of diphenyliodobismuthineand iodobenzene points to the extreme instability of such a sub-stance. The results obtained with iodine and tri-a-naphthylbis-muthine also support this conclusion. The evidence obtained upt o the present points strongly in one direction, but owing to theexistence of such compounds as triphenylarsine tetraiodide,Ph3As14, and tri-ptolylarsine tetraiodide, (C,H,),AsI, (Michaelis,ATznacZen, 1902, 321 , 203), further discussion is undesirable untilthese compounds have been prepared for the purposes of com-p arison.Gillmeister mentions that iodobenzene, bismuth oxyiodide, anddiphenyliodobismuthine are obtained when triphenylbismuthinedichloride and potassium iodide inte'ract and the mixture is pouredinto water.We have confirmed these results, but find that the redcompound referred to above is also a produd of the reaction.The action of iodine monochloride and monobromide on tri-a-naphthylbismuthine is similar to that on the triphenyl compound.No additive product was isolated, but a-iodonaphthalene in yieldssomewhat below the theoretical was obtained.VOL. CVlI. 18 CHALLENGER AND ALLPRESSWith iodine and tri-a-naphthylbismuthine the products atebismuth iodide, a-idonaphthalene (isolated as dichloride) , un-changed bismuthine, and (probably) di-a-naphthyliodobismuthine.The.results so far described, excluding those with iodine, showthat when two dissimilar halogen atoms, and, in the case of tri-phenylbismuthine, cyanogen iodide, enter into reaction with atertiary aromatic bismuthine, the more negative atom or groupremains attlached to bismuth, whilst iodine is eliminated as aryliodide. I n the case of tri-a-naphthylbismuthine, since the yields ofa-iodonaphthalenel are not quite quantitative, the formation ofsome a-bromo- or a-chloro-naphthalene is not absolutely excluded,but seems improbable.The action of iodine haloids on unsymmetrical (or mixed) bis-muthines is being investigated. It might be expected that iodine,being a comparatively positive element, would be eliminated inunion with the most strongly negative or unsaturated hydrocarbonradicle present.In a preliminary experiment diphenyl-a-naphthylbismuthine wastreated with iodine monobromide, and a-iodonaphthalene wasdetected in the solution.The interaction of diphenylbromobismuthine and iodine chlorideis also instructive, a yield of iodobenzene 77 per cent.of thatrequired by the equation PhzBiBr + IC1= PhBiClBr + P h I beingobtained.As regards the experiments with cyanogen and its derivatives itmight be expected that; the unknown compound Ph,Bi(CN), wouldbe very stable owing to the strongly negative nature of both thecyano- and the phenyl group. Under the conditions employed upto the present, however, cyanogen does not react with triphenyl-bismuthine; moreover, under conditions in which the iodide readilyreacts with triphenylbismuthine, cyanogen bromide has no action.Voii Braun (Ber., 1900, 33, 1438, 2728; 1902, 35, 1279; 1903,36, 1196; 1907, 40, 3914, 3933) has shown that cyanogen bromideis without action on triphenylamine even when heated in sealedtubes, although it reacts with tribenzylamine.With aliphatic oraliphatic-aromatic amines an unstable additive product is obtained,which decomposes so as t o yield alkyl or aryl bromide and a substi-tuted cyanamide.Unsaturated radicles such as ally1 are particularly easily elimin-ated as bromide.Cyanamide, which might have been expected to yield aniline anddiphe8nylcyanobismuthine when heated with triphenylbismuthine,polymerised to dicyanodiamide without otherwise reactingORGANO-DERIVATIVES OF BISMUTH, PART 11.19EXPERIMENTAL.Diphenylchlorobismuthine.Marquardt (Annaleiz, 1889, 25 1, 326) states that diphenylchloro-bismuthine, BiPh,Cl, appears to be formed by the interaction oftriphenylbismuthine and bismuth chloride in glacial acetic acid orethyl acetate, bint that he was unable to crystallise it.Since one of us has observed that triphenylbismuthine is decom-posed by glacial acetic acid, and since diphenylchlorobismuthineis decomposed by water or alcohol (both frequently present inordinary ethyl acetate) the reason for this non-success may not befar t o seek. An ethereal solution of 2.5 grams of triphenylbis-muthine (2 mols.) was treated with 1.3 grams of bismuth chloride,likewise dissolved in anhydrous ether.Three grams of colourlesscrystlals, melting a t 184--185O, were deposited, and on recrystal-lisation from benzene the melting point was unchanged.The results of analysis indicate the presence of traces of phenyl-dichlorobismuthine :0.3727 gave 0*2202 Bi,O,. Bi=52.94.*C,,H,,ClBi requires Bi = 52.33 per cent.CGH,Cl2Bi ,, Bi=58*43 ,, ,,Diphenylchloro bismuthine forms colourless crystals fairly readilysoluble in dry benzene, toluene, or chloroform, but much less 60in ether o r light petroleum.Alcohol and moist solvents cause decomposition as with all com-pounds of the type BiR,X and BiRX,. On keeping in a sealedtube gradual decomposition occurs, and benzene is produced.Con-centrated hydrochloric acid readily liberates benzene.Action of Iodine Monochloride on Triphenylbismuthke.Five grams of triphenylbismuthine in dry ether were treated with1.85 grams of iodine chloride (1 mol.) in the same solvent. Abrownish-yellow turbidity was produced, which quickly disappeared,pale yellow crysbals (A, 3.35 grams) separating eimultaneously.These melted a t 182-183O after being recrystallised, and did notdepress the melting point of diphenylchlorobismuthine. On evapor-ating the decanted ether and treating the oily residue with lightpetroleum, a further quantity of the crystals was obtained(B, 0.8 gram). The theoretical yield of diphenylchlorobismuthineis 4.5 grams. Deposits A and B contained traces of iodine.On removal of the light petroleum the residual oil, when treatedwith chlorine in chloroform solution, .gave 2.65 grams of iodo-benzene dichloride, whereas the theoretical yield is 3.1 grams.* A second estimation gave = 52.91.c 20 CHALLENGER AND ALLPRESS :In aiiotlier experiment; ti grams of triphenylbisniuthiiie weretreat,ed with slightly more than 2.3 grams (1.3 mols.) of iodinechloride'.Deposits A, (3.25 grams) and B, (1.2 grams) were obtainedas before, and the yield of iodobenzene dichloride was 3.9 grams.Deposit A, contained practically no iodine, but B, contained rela-tively large amounts in comparison with B.Actioiz of lodi?ze Monobromide on Trip~~enylbismuthine.Iodine bromide (2.35 grams; 1 mol.) dissolved in anhydrous etherwas gradually added to triphenylbismuthine (5 grams) in the samesolvent.A brownish-yellow turbidity, disappearing on shaking,was produced, and yellow crystals mere deposited. These melteda t 1540, and after recrystallisation from chloroform, sharply a t157O. The total quantity obtained was 4.6 grams, the theoreticalyield of diphenylbromobismuthine being 5 grams.Fractional extraction of the crystals with hot benzene gavedeposits meyting a t 156-15r0, 156-157O, 148O, and 157O. Thethird deposit after one 'recrystallisation melted a t 155-157O.Further recrystallisation of these deposits produced no change inmelting point, so that the preselnce of compounds other thandiphenylbromobismuthine (m. p. 157--158O), except in traces, isexcluded.The ethereal solution was distilled, and a further deposit ofdiplienylbromobismuthine removed by means of cold chloroform orether.The iodobenzene remaining in the chloroform was con-verted into the iododichloride. The yield was 2-95 grams.Action of Cyanogem Zodide on Triphelzylbismuthine.When 5 grams of triphenylbisnuthine and 2.2 grams of cyanogeniodide (1% mols.) were heated with benzene under a reflux con-denser, 3.2 grams of a yellow solid separated. On distilling off thebenzene and treatir,g the oily rwidue with light petroleum,0.7 gram of the yellow solid remained."On removing the petroleum, adding chloroform, and passingchlorine, 2.95 grams of iodobenzene dichloride were obtained.The1 solid substance was recrystallised from hot aIcoho1, only avery small, insoluble1 residue remaining :0.2396 + 10.1 C.C.0.13N-AgN0, required 7.3 C.C. N/10-NE4@NS.CN=6*29.C,,H,,NBi requires CN = 6.71 per cent.On crystal-lising from benzene, however, a very small quantity of yellow solid remained un-dissolved, whilst the solution deposited white crystals identical with the mainproduct. The formation of the yellow insoluble substance containiiig iodine isprobably due t o the oxcess of cyanogen iodide.* This deposit contained a certain amount of an iodine compoundOBGANO-DERIVATIVES OF BISMUTH. PART 11. 21Dip~enyEcyaizobismut~~ne forms colourless needles, resemblingglass-wool, and smelling faintly of hydrocyanic acid. These meltand decompose a t about 210°, and are sparingly soluble in hotalcohol or benzene, still less so in ether o r light petroleum.Onconcentrating the alcoholic solution, hydrocyanic acid is evolved(recognised by conversion into ammonium thiocyanate), and oncooling, triphenylbismuthine is deposited. Warm concentratedhydrochloric acid liberates benzene and hydrocyanic acid ; warmsodium hydroxide forms sodium cyanide and a yellow, insolublesubstance, which is being f urt'hesr investigated. With silver nitratesolution diphenylcyanobismuthine readily gives silver cyanide.Action of Iodine on Triphelzylbismuthine.A. In, Dry Ethereat Solation.-When 10 grams of triphenyl-bismuthine were treated with 5.8 grams of iodine (1 mol.) a redprecipitate was produced. The ether containing a yellow solid insuspension was poured off, and the red precipitate (3.5 grams)washed with dry ether.Di-phenyliodobismutkine forms yellow needles, melting a t 133O.)After distilling about half of the ether a solid separated, whichmelted a t 131O; this was removed, and the ether completely evapor-ated.The residual oil still contained bismuth. Its chloroformsolution was treated with a very slight excess of bromine, in thehope of removing any triphenylbismuthine as triphenylbismuthinedibromide (m. p. 122O). The crystals which separated, however,melted after one recrystallisation a t 201-202O, and did not depressthe melting point of a specimen of phenyldibromobismuthine.Their formation was obviously due to the action of bromine ondiphenyliodobismuthine thus : BiPh,I + Br, = BiPhBr, + PhI.The chloroform mother liquor gave rise t o 5.5 grams of iodo-benzene dichloride, corresponding with 4.1 grams of iodobenzene.The theoretical yield to be expected from the equationBiPh, + I, = BiPh,I + P h Iis 4-65 grams.Since much iodobenzene was probably producedthrough t8hel agency of the bromine, the quantity directly obtainedby the action of iodine on the bismuthine is seen to be far below thetheoretical.The red precipitate, melting a t about 194O, contained muchiodine and organic matter, and evolved benzene on treatment withhydrochloric acid, When heated with benzene or chloroform apurplish-black residue, still containing organic matter, and a yellowsolution were obtained, which on addition of light petroleumdeposited a red solid.(The yellow solid in ethereal suspension melted a t 134O22 CHALLENGER AND ALLPRESS :B.In) Ether-Light Petroleum So1utio.n.-Five grams of tri-phenylbismuthine were treated with 2.9 grams of iodine. Theusual brown turbidity, followed by the deposition of crystals melt-ing a t 131--134O, was produced, and the red precipitate, whichthis time melted a t 131°, was also deposited (4.5 grams). A furtherquantity of yellow crystals melting a t 133O was obtained by dis-tilling off the petroleum, after which the solution contained onlya trace of solid matter, and gave rise, in the usual way, t o 2.4 gramsof iodobenzene dichloride, corresponding with 1.8 grams of iodo-benzene. The theoretical yield is 2-33 grams.The red powder was heated with dry benzene, the yellow solutionfiltered, and allowed to deposit five separate crops of crystals :Original substance red, m.p. 131O.1st deposit. Deep red crystals, m. p. 194-198O.2nd deposit. Paler red crystals, m. p. 130O.3rd deposit. Orange crystals, m. p. 125O.4th deposit. Pale yellow crystals, m. p. 127O.5th deposit. Brown solid, m. p. above 2 1 0 O .The substance therefore decomposes on contact with hot benzeneand other solvents. I f it is extracted with hot dry carbon tetra-chloride until the extract is no longer coloured, the residue isbrownish-black, and consists principally of bismuth iodide.Action of Iodine Nonochloride OIL Tri-a-naphtthhylbismuthine.Five grams of tri-a-naphthylbismuthine were gradually treatedwith 1-95 grams (1.4 mols.) of iodine monochloride, both beingdissolved in a mixture of ether and chloroform.The solvent was then partly distilled off, and the yellow depositcollected.It decomposed a t about 240°, and probably consistedlargely of di-a-naphthylchlorobismuthine, Bi(C,oH,),Cl. Thechloroform-ether filtrate was distilled to dryness, the residual oilremoved from a small quantity of dark-coloured solid by means oflight petroleum, and the solution diluted with glacial acetic acid,cooled, and treated with chlorine. The weight of iodonaphthalenedichloride (m. p. 5 5 O ) obtained was 2-75 grams, thel theoretical yieldbeing 3.90 grams.Action of Iodine Mombromide on Tri-a-naphthtylbismuthine.Tri-a-naphthylbismuthine (2.5 grams) was treated with 0.85 gram(1 mol.) of iodine monobromide, both in chloroform solution.Thereaction mixture was worked up in the usual way, and found toyield in addition to 1.1 grams of a-iodonaphthalene dichloride (theo-retical yield 1'4 grams) a certain amount of unchanged bismuthine.The principal solid product, although containing traces of iodineORGANO-DERIVATIVES OF BISMUTH. PART 11. 23probably consisted largely of di-a-naphthylbromobismuthine. Itwas only superficially examined, and is reserved for further investi-gation.Action of Iodine o n Tri-a-naphthylbismuthine.One gram of tri-a-naphthylbismuthine in dry benzene was dowlytreated with a similar solution of 0.43 gram of iodine, and thebrown precipitate was collected. It was free from organic matter,and indistinguishable as regards the action of heat, water, sulphuricacid, or nitric acid, from bismuth iodide.The benzene was removed from the filtered reaction mixture, andthe reddish-yellow residue extracted with cold light petroleum.Onevaporation of the ext.rack, and removal of a small quantity oftri-a-naphthylbismuthine, the oily residue was treated with lightpetroleum and glacial acetic acid, boiled in order to decompose anytraces of bismuthine, cooled, and treated with chlorine, whena-iodonaphthalene dichloride (0.45 gram) was obtained.The residue insoluble in light petroleum weighed 0.5 gram, andcontained iodine. It almost certainly consisted of tri-a-naphthyl-bismuthine and di-a-naphthyliodobismuthine, since on allowing itto remain in the air it developed a strong odour of naphthalene.*On then extracting with cold benzene, tri-a-naphthylbismuthinewas obtained, and the residue was bright red, free from organicmatter, and indistinguishable from bismuth oxyiodide.I f the substance, when freshly obt'ained, was extracted with hotbenzene and the extract treated with alcohol, pure tri-a-naphthyl-bismuthine crystallised out.The colourless mother liquors gradu-ally deposited bismuth oxyiodide, indicating the decomposition byalcohol of di-a-naphthyliodobismuthine.A c tion of Zodin e Mono b ro mide om Diph em)-a-naph t hyl b ismut hine.Dip hen yl-a-nap ht h ylbismuthine (0.3 kr am) was treated withiodine monobromide (0.1 3 gram) in chlorof orm-ether solution.Instant decolorisation took place, and the solvent was distilled off.The residue was treated with light petroleum, the extract filtered,diluted with glacial acetic acid, and treated with chlorine, whenabout 0.05 gram of a-iodonaphthalene dichloride (m.p. 65-70°)was obtained.* Compounds of the type B i h X decompose on keeping even in closed vessels, ori n the presence of alcohol and moist solvents, giving a hydrocarbon and bismuthoxyhaloid (see page 19 and Gillmeister, Bw., 1597, 30, '2844)24 CHALLENGER AND ALLPRESS :Action of Iodine ~~~onochloride on Diphe,7ylbromobismuthine.Four grams of diphenylbromobismuthine were covered with dryether and slowly treated with an ethereal solution of 1.5 grams(1 mol.) of iodine monochloride. A brown turbidity was produced,which on vigorous shaking slowly disappeared.The1 insoluble residue (2.60 grams), which contained iodine,melted very indefinitely a t about 190°, and was not furtherexamined, since from its method of preparation it seemed veryprobable that it might contain phenylchlorobromobismutliine,BiPhClBr, bismuth dichlorobromide, BiCl,Br, and similar sub-stances.The oily residue from the ethereal solution and washings wasextracted with cold chloroform, and the solution yielded 1.95 gramsof iodobenzene dichloride, the theoretical amount (removal of onephenyl group) being 2.5 grams.The small residue insoluble inchloroform contained no organic matter.Cyanogen and Triph e n yl b ismu t hine.Cyanogeln was passed into 0.9 gram of triphenylbismuthine dis-solved in about 10 c .~ . of benzene, but no reaction appeared t ooccur. The liquid was diluted with an equal volume of ether, andthe gas again passed, but with the same result. On spontaneousevaporation of the solvent the bisrnuthine was recovered un-changed. Similar results were obtained in boiling benzene.Cyanogen Bromide and Triphenylbisrnuthine.Five grams of triphenylbismuthine and 1.5 grams of cyanogenbromide (la mol.) were heated in benzene solution for four hourswithout any separation of the sparingly soluble diphenylcyano-bismuthine taking place(. In ether there was no action, either onboiling or on allowing to remain for three weeks.Cyanamide and Triph en y 1 b ismu thine.When 5 grams of triphenylbismuthine and 0.9 gram of cyan-amide were heated with benzene a white solid separated, whichwas free from bismuth. It melted a t about 2 0 3 O , and afterrecrystallisation from alcohol and light petroleum, a t 205-207O.It was obviously dicyanodiamide, and was also obtained when cyan-amide alone was heated in benzene.Since the' cyanamide employed was deliquescent and containedtraces of dicyanodiamide (quite insufficient, however, to accounORGANO-DERIVATIVES OF BISMUTH.PART 11. 25for the large amount obtained as above), a portion was treatedwith ether, in which dicyanodiamide is insoluble, and with phos-phoric oxidg, filtered, the ether removed, and the dry cyanamideheated with benzene and triphenylbismuthine, but with the sameresultr. No change occurred when cyanamide and triphenylbis-muthine were heated in dry ether.Relative Stabilities of t h e Triphenylbismuthine Dihaloids.Owing t o the comparatively positive nature of the iodine atom,triphenylbismuthine di-iodide seems to be an extremely unstablesubstance> which decomposes as soon as formed, eliminating iodo-benzene, whereas the corresponding dibromide is much morestable.The question naturally suggested itself, whether, on account ofthe stronger negative nature of chlorine, the dichloride would befound to be more stable than the dibromide.On heating in a sealed tube at looo the dichloride was unaltered,but on boiling its solution in dry benzene partial decompositionoccurred, diphenylchlorobismuthine (see p. 19) separating. Themother liquors yielded unchanged dichloride. The difference instability is therefore only slight,Nevertheless, it seemed possible that although the action ofmagnesium phenyl bromide on triphenylbismuthine dibromide doesnot give rise t o tetraphenylbismuthonium bromide, the correspond-ing bismuthoniuni chloride might be obtained from triphenylbis-muthine dichloride.The action of magnesium phenyl bromide on the dichloride led,however, 60- the formation of derivatives of tervalent bismuth.The authors are indebted t o the Research Fund of the ChemicalSociety for a grant in aid of this investigation, and to Dr. T. S.Price for allowing the use of his laboratory in the later stages ofthe work.THE UNIVERSITY,BIRbf TNGITAM

ISSN:0368-1645    

DOI:10.1039/CT9150700016

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

J O U R N A L OF THE CHEMICAL SOCIETY. TRANSACTIONS. dontmittre of @ublicafion : H. BHERETON BAKER M.A. D.Sc., J. N. COLLIE Ph.D. F.R.S. A. W. CROSSLICY D.Sc. Ph.D. P.R.S: l+’. G. DONNAN M.A. Ph.D. F.R.S. BERNARD DYER D.Sc. M. 0. FOKSI’ER D.Sc. Ph.D. F.R.S. F. R. S. A. HARDEN D.Sc. Ph.D. F.R.S. T. M. LOWRY D.Sc. J. C PHCLIP D.Sc. Ph.D. F. I,. PYMAN D.Sc. Ph.D. A . SCOTT M.A. D.Sc. F.R.S. G . SENTER D.Yc. Ph.D. S. SMILES D.Sc. &bitor : J. C. CAIN D.Sc. Ph.D. S,ttb-Q%iar : A. J. GREENAWAY. 3misfrrnt Snb-&’bifor : CLAREKCE SMITH D.Sc. 1915. Vol. CVII. Part II. pp. 847-end. L 0 N*D 0.N : GURNEY & JACKSON 33 PATERNOSTER ROW E.C. 1015 PRINTED IN GREAT BRITAIN BY RICHARD CLAY & SONS LIMITED, BRUNSWICK ST. STAMFORD ST. S.E., AND BUNGAY SUFFOLK

ISSN:0368-1645    

DOI:10.1039/CT91507FP017

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

26 BlcKENZIE, DREW, AND MARTIN : CONVERSION OFV.-Conuemion of I.-PIEer~ylch loroacetic Acid intod-Diphen ylsuccinic Acid.By ALEX. MCKENZIE, HARRY DUGALD KEITH DREW, andGERALD HARGRAVE MARTIN.IN examining the action of thionyl chloride on the isomerica-hydroxy-a-phenylpropionic acids, McKenzie and Clough (T., 1910,97, 1016) observed that the change,Me Ph >C<E& + bJe Ph>"<g:2rr,which takes place a t the ordinary temperature, could be utilisedfor the preparation of the a-chloro-a-phenylpropionic acids with-out the necessity of isolating the corresponding acid chlorides.The chlorine in those chloro-acids is very loosely bound; it can bedisplaced with remarkable ease by the action of water. Withthese facts before us, it appeared likely that a coavenient methodmight be available for the preparation of compounds containingan asymmetric quaternary carbon atom ; for example, if a-chloro-a-phenylpropionic acid were acted on by magnesium ethyl iodide,there was a reasonable prospect that the change,Me Ph >Q<%,H ---+ Ble Ph >c<;;,H9would occur.Accordingly, with the object of studying the actionof Grignard reagents on compounds of the type of a-chloro-a-phenylpropionic acid containing a labile halogen atom, some pre-liminary experiments were conducted with the more readilyaccessible diphenylchloroacetic acid, prepared from benzilic acid.When this chloro-acid was acted on by magnesium phenyl bromide,and the additive compound then decomposed in the customaryfashion, it became clear that some other organic acid was presentin addition to triphenylacetic acid, and the conclusion was drawnthat the change,had in all probability also occurred.This particular action wasnot, however, thoroughly investigated, as it suggested the moreinteresting stereochemical problem, which is dealt with in thepresent communication.The existence of four diphenylsuccinic acids, namely, the d-, 7-,r-, and meso-forms, is indicated by theory, and of these acids t%etwo optically inactive modifications, the so-called a- and p-f orms,are described in the literature; no information is, however, availL-PHENYLCHLOROACETIC ACID INTO D-DIPHENYLSUCCINIC ACID. 27able as to which of these is the resolvable form. The authorshave now been able to obtain the dextrorotatory isomeride byacting on I-phenylchloroacetic acid with magnesium phenylbromide, This action proceeds, as might be expected, in variousdirections, since the chloro-acid presents a number of points ofattack for the Grignard reagent.I n the transformation involvingthe production of d-diphenylsuccinic acid, namely,C,H,* QH*CO,HCO,H*CH*C,H, ' d- I-C,H,* CHCI. CO,H +an acid containing two asymmetric carbon atoms is synthesisedfrom an acid containing one such atom, the latter acid combining,as it were, with itself, with the elimination of chlorine. Thestriking point is that the optical activity does not disappear in thissynthesis. The change may be represented as follows:The results obtained by the action of magnesium alkyl haloidson T-phenylchloroacetic acid and on r-phenylbromoacetic acid arerecorded in t,he experimental part.The following compoands wereobtained by the action of magnesium phenyl bromide on I-phenyl-chloroacetic acid : diphenyl, d-aP-dihydroxy-apP-triphenylethz:ie./3-diphenylsuccinic acid, diphenylacetic acid, and cl-diplienylsuccinic:acid. The latter acid was obtained only in very small yield fromthe mixture of acids resulting from the Grignard action, this mix-ture consisting of I-phenylchloroacetic acid, r- and I-mandelic acids,diphenylacetic acid, d- and Z- and meso-diphenylsuccinic acids.The acid is strongly dextrorotatory, having [aID +34S0 in ethyl-alcoholic solution ; we were, however, unable, with the smallquantity at our disposal, to prove the optical purity of this pro-duct, so that this value must for the present be given with reserve.Of the two inactive diphenylsuccinic acids, the a-form has tlhe lowermelting point.I n the case of tlhe isomeric dibromosuccinic acids,the Go-form m l t s a t a lower temperaimre than its inactiveisomeride, and it is resolvable into optically active components(McKenzie, T., 1912, 101, 1196); similarly with the dimethyl-succinic acids, the isomeride of lower melting point is resolvable(Werner and Basyrin, Ber., 1913, 46, 3229; compare also thecase of the ad-dimethyladipic acids examined by No yes andKyriakides, J . Amer. Chem. Soc., 1910, 32, 1057). It is accord-ingly desirable that the question as to which of the diphenyl-succinic acids is resolvable should be settled28 McRENZIE, DREW, AND MARTIN : CONVERSION OFAs already mentioned, d-up-dihydroxy-aPP-triphenylethane ++(lzevorotatory) was one of the neutral products of the Grignardact'ion on E-phenylchloroacetic acid.Since Z-phenylchloroacetic acidis obtained from Z-mandelic acid by means of thionyl chloride(McKenzie and Barrow, T., 1911, 99, 1910), it is accordinglypossible to convert Z-mandelic acid into both the d- and Z-glycolsaccording to the, scheme :(by C6H6'MgBr) LC,H5*CH( 0 H ) CO2H Z-C,H5 CH (OH)*C( C6H5),*OE(Dex trorotatory )(by 50C12) 1EXPERIMENTAL.Action of Magnesium Methyl Iodide o n r-PhenyEchEoroacetic Acid.Several experiments on this action were conducted in which theproportion of the Grignard reagent was varied ; thus, for phenyl-chloroacetic acid (1 mol.), magnesium methyl iodide in the pro-portions of 2, 3, 4, 6, and 7 mols.respectively was employed. I nother experiments, where methyl iodide was siphoned into anethereal solution of phenylchloroacetic acid in contact with anexcess of magnesium, there wits no improvement, so far as theyield of the diphenylsuccinic acids was concerned, on the customarymet4hod of applying the Grignard reagent; on the contrary, theyield of these acids was smaller than usual. The yield of glycolsdid not increase appreciably when an excess of magnesium alkylhaloid was used. With the proportion of 2, 3, or 4 mols. ofmagnesium methyl iodide to 1 mol. of the chloro-acid, there wasonly a slight evolution of gas on the addition of ice to the pro-duct of the reaction; on the othe'r hand, with the reagent in theproportion of 6 or 7 mols., the evolution of gas on the addition ofice was vigorous, the Grignard reagent being obviously in excessunder these conditions.From the resulting mixture of acids, /3-diphenylsuccinic acid tothe' extent of from 7 to 11 per cent.was obtained (the percentageyields given here and subsequently are calculated, for convenience,on the basis that the initial halogen acid is transfo'rmed into onecompound only, namely, the compound under consideratdon). Thea-isomeride was also present, apparently t o about the same extentas the @-form, but the amount of i t actually isolated in the pure* McKenzie and Wren (T., 1910, 97, 473) prepared the dextrorotatory glycolboth from methyl Z-mandelate and from Z-benzoin.A configurational change was im-possible here, and the compound was therefore designated as I-TA-PHENYLCHLOROACETIC ACID INTO D-DIPHENYLSUCCINIC ACID. 29conditioii was always very sniall, owing to the difficulty ofseparating it from the other acids present. Hydratropic acid wasobtained in yields varying from 12 t o 15 per cent. when the mix-ture of acids was subjected t o distillation in steam. Even whenthe reaction took place in the proportlion of 3 mols. of the mag-nesium alkyl haloid to 1 mol. of the chloro-acid, it was quite clearthat some of the latter was not attacked, since a comparativelylarge quantity of mandelic acid was subsequently isolated.More-over, this chloro-acid which had survived the attack of the Grignardreagent was not entirely hydrolysed during the various operations,and some of i t was extracted from the mixture of acids by meansof boiling light petroleum. It was observed that the etherealsolution of the mixture of acids was not rendered free from iodineeven after it had been shaken several times with sulphurous acid;this behaviour indicated the presence of some unstable iodine com-pound, possibly phenyliodoacetic acid. When the proportion ofmagnesium methyl iodide was 4 o r more mols. to 1 mol. o€ thechloro-acid, the iodine which was liberated a t the initial stages ofthe action eventually disappeared as the addition oZ the reagentproceeded, and there was no liberation of iodine when the additivecompounds were decomposed by ammonium chloride, A typicalexperiment may be described in detail.The Grignard reagent prepared from 34 grams of methyl iodide(4 mols.) was added drop by drop by means of a siphon to anethereal solution of 10 grams of r-phenylchloroacetic acid (1 mol.).The vigorous action was accompanied by the evolution of an in-flammable gas.An oil containing iodine separated a t an earlystage, and the colour of the iodine disappeased later. After de-composition with ice and ammonium chloride, the aqueoas solutionwas extracted with ether for the purpose of removing the glycolspresent, whilst the acids were obtained from the aqueousammoniacal solution by acidification with mineral acid and extrac-tion with ether.The ether was then removed from the latterextract, and hydratropic acid was obtained as a colourless oil bymeans of distillation in steam; it was converted into its calciumsalt. (Found, H20=9.8. Calc. for 2H20, 9-7 per cent. Ca inanhydrous salt = 11.4. Calc., Ca= 11.8 per cent.) Those acidswhich were not volatile with steam were extracted with ether, theether expelled, and the residue extracted with benzene. The whitesolid which remained undissolved after this treatment was crystal-lised from ethyl alcohol, from which i t separated in fine, silkyneedles. It consisted of /3-diphenylsuccinic acid, and melted a t228-230°. (Found, C=70*5; H=5*2. Calc., C=71*1; H=5.3per cent. Equivalent, by titration with baryta=135*7. Calc.30 McKEh’ZIE, DREW, AND MARTIN : CONVERSION OF135.1.) The silver salt was also analysed. (Found, Ag=44.6.Calc., Ag=44*6 per cent.) By the crystallisation of the oily acidresdting from the benzene solution, a small quantity ofa-diphenylsuccinic acid was eventually isolated ; this melted a t181--183O, and, after resolidification, at 219-221O. Small quanti-ties of mandelic and phenylchloroacetic acids were also separated.The following data regarding the melting point of P-diphenyl-succinic acid are recorded in the literature: 229O, Reimer (Ber.,1881, 14, 1803), Chdaney and Knoevenagel (Ber., 1892, 25,296); 245O, Anschiitz and Bendix (dnna.len, 1890, 259, 61); 252*,Rmer (AnmaZen, 1888, 247, 152), Ruhemann and Naunton (T.,1912, 101, 50).According to Reimer (loc. cit.), a-diphenylsuccinicacid fuses a t 183O, then solidifies, and then melts a t 220-221O.Action of Magnesium Plienyl Bromide on r-PhenylhalogenaceticAn ethereal solution of the Grignard reagent prepared from37 grams of bromobenzene (4 mols.) was added gradually to anethereal solution of 12.5 grams 0f T-phenylbromoacetic acid(1 mol.). The following products were isolated : diphenyl, phenol,aP-dihydroxy-aP8-triphenylethane, diphenylacetic acid, 8-diphenyl-succinic acid (1.9 grams), and a-diphenylsuccinic acid (1-8 grams).The latter acid was analysed. (Found, C=70*8; H=5.6 Calc.,C=71.1; H-5.2 per cent. Equivalent, by titration withbaryta8= 136.2. Calc.= 135.1.)The amounts of the isomeric diphenylsuccinic acids formed inthis experiment were greater than when phenylchloroacetic acid(see below) was employed.This is to be attributed to the fact thattlhe displacement of halogen occurs much more readily with thebromo- than with the chloro-acid.I n three experiments carried out under varying conditions withr-phenylchloroacetic acid, the following substances were isolated :triphenylethane. 8- Acid. a- Acid. acetic acid.Experiment 1. ..... 10 per cent. 13 per cent. 1 per cent. 1 per cent.$9 8 2s9 , 7 ss2.. 20 ,, 5 9 ,3 17 6 9 sA cids.uB-Dihydroxy-uB& Dip henyl-- .... - ...... ..After the removal of the neutral products of the action, amixture of acids was obtained as usual. When the ethereal solu-tion of the latter was allowed to evaporate spontaneously, a mix-ture of P-diphenylsuccinic acid and diphenylacetic acid crystallises.This mixture is easily separated into its components; the residualoil contains the a-acid, together with phenylchloroacetic andmandelic acidsL-PHENYLCfiLOROACETIC AClD IXTO D-DIPHENYLSUCCINIC ACID.31Action of Magnesium Phenyl Bromide o n l-Phe~2ylchlol.oacetic:Acid.Of the various experiments carried out on this action, thefollowing gave the best results.I-Phenylchloroacetic acid was prepared by resolving the r-acidwith morphine according to McEenzie and Clough's method (T.,1908, 93, 811). The Grignard reagent, prepared from 111 gramsof bromobenzene (4 mols.) and 550 C.C. of ether, was added withinan interval of fifty minutes to a solution of 30 grams of the 1-acid(1 mol.) in 80 C.C.of ether, the flask being immersed in icecoldwater during the operation. After warming f o r fifteen minutes,the mixture was left a t the ordinary temperature for twenty-fourhours, and then decomposed by a concentrated solution ofammonium chloride. The ethereal solution containing the neutralproducts was separated from the aqueous layer, which was furtherextracted twice with ether. ThO oil resulting from the etherealsolution amounted to about 30 grams, and consisted mainly ofdiphenyl. After Beven days, the crystals which had separated wereremoved and crystallised from ethyl alcohol. The resulting solid(2.2 grams), which had [alD - 1 8 6 O in acetone, and which possiblycontained some of the r-glycol, was crystallised from methyl alcohol,when 1.6 grams of the pure active glycol were obtained.d-aS-Dihydrozy-aaP-triphenylethalze (triphenyle thyleme glycol),OH*C.HPh*CPh,*OH, separates from methyl alcohol in colourlessneedles and melts a t 1 2 8 O , whereas the dextrorotatory I-isomeridemelts at 128-129O (McEenzie and Wren, T., 1910, 97, 480):0.192 gave 0.5817 CO, and 0'1104 3 0 .C20H1802 requires C=82-7; H= 6.3 per cent.The compound gave an emerald-green coloration with conceii-trated sulphuric acid.Its solution in chloroform was lzvo-rotatory :I = 2, c = 1.32, a, - 6.02*, [a]* - 228O.The aqueous solution of ammonium salbs, from which the neutralproducts had been removed in the manner indicated, was acidifiedby h'ydrochloric acid, and the organic acids were then extractedwith ether-.On expulsion of the ether, an oil (23 grams) wasobtained, which gradually solidified partly. The solid (5-5 grams)consisted of a 'mixture of S-diphenylsuccinic acid and diphenyl-acetic acid, the separation of which was easy, since the former ispractically insoluble in benzene. One gram oE P-diphenylsuccinicacid was isola@d. (Found, C=71*1; H=5*2. Calc., C=71-1;H=5*2 per cent.) The amount of diphenylacetic acid isolated was3.1 grams. (Found, equivalent= 213. C'alc. = 212.1.) The melt-C=82-6; H=6.432 CHATTAWAY AND PEARCE : 2 : 4-DICHLOROPHENYLHYDRAZJNE.iiig poilit observed was 14G0, wlieress Jetla ("1 ~ I I N I C I I , 1870, 155,84) gives 145--146O.The oil from which these acids had been partly separated wasdextrorotatory, having [aID + 17.9O in ethyl-alcoholic solution.Aslight excess of sodium hydroxide was added, and the solutionboiled, in order t o convert any phenylchloroacetic acid present intomandelic acid. After evaporation, the crystals of sodium salt (1-6grams) which separated had [a], +66.B0 in aqueous somlution. Twofurther quantities of sodium salt were removed from the filtrate,but these were only slightly dextrorotatory, the values obtainedbeing [a], + 4' and + 3.4' respectively. The sodium salt (1.6grams) was accordingly recrystallised from water, in which it dis-solved readily. 0.7 Gram with [a], +127O wilg obtained. Theacid obtained from this by acidification and extraction with etherhad [a]= + 287O in ethyl-alcoholic solution; it was crystallised frombenzene, in which it is very sparingly soluble.d-Diphenylsuccink acid, C16H1404, separates from benzene incolourless, prismatic needles. After drying at looo, it shranksharply to a thin core a t 170-171°, which was unchanged until211*5O, when it melted sharply to a clear liquid, from which gaswas not evolved.Yield, 0.2 gram.The acid is sparingly soluble in hot water :0.1107 gave 0.2881 CO, and 0.0553 H,O.Its rotation was determined in ethyl-alcoholic solution :C=71*0; H=5.6.C16H,,0, requires C= 71.1 ; H = 5 2 per cent.Z=1, ~=1*07, U: + 3 * 7 3 O , [a]: +348*.Part of the expense of this investigation was defrayed by agrant from the Government Grant Committee of the Royal Society,to whom our best thanks are due.UNIVERSITY COLLEGE, DUNDEE, ~JIRKBECK COLLEGE,UNIVERSITY OF ST. ANDREWS. LONDON, E.C

ISSN:0368-1645    

DOI:10.1039/CT9150700026

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

32 CHATTAWAY AND PEARCE : 2 : 4-DICHLOROPHENYLHYDRAZJ".By FREDERICK DANIEL CHATTAWAY and CHARLES FREDERICKBYRDE PEARCE.FROM the circumstance1 that 2 : 4-dichlorophenylhydrazine, whichshould bO obtainable from perhaps the most easily made chloro-substituted aniline, has never been described, it might be inferredthat its preparation offers some difficulty. This, however, is notthe case; it is easily made by the ordinary method, and is a stableCHATTAWAY AND PEARCE : 2 : 4-DICHLOROPHEKPLHYDRAZINE. 33well-crystallised compound showing the usual behaviour of halogen-substituted aromatic hydrazines, and yielding a series of well-crystallised hydrazides and hydrazones.2 : 4-DichEorophen.ylhydrazine, C,H,Cl,*NH*NH,.Forty grams of 2 : 4-dichloroaniline suspended in 200 C.C.ofconcentrated hydrochloric acid were diazotised by 22 grams ofsodium nitrite dissolved in 60 C.C. of water, and the diazoniuinsalt thus formed was reduced by 120 grams of crystallised staiinouschloride dissolved in 100 C.C. of concentrated hydrochloric acid.The hydrochloride of the base, which separated a t once as acrystalline precipitate, was washed several times with a saturatedaolubion of sodium chloride, and recrystallid twice from boilingwater slightly acidified with hydrochloric acid.2 : 4-Dichlorophenylhydrnzine hydrocldoride,C,H,Cl,*NH-NH,,HC~,crystallises from aqueous solution in tufts of thin, colourless plates,which decompose in the neighbourhood of 210O:0.3701 gave 0.7417 AgC1. C\=49*57.C6H6N,C1,,HC1 requires C1= 49.83 per cent.The base was o’btained by adding excess of potassium hydroxideto an aqueous solution of the hydrochloride and extracting withether, air being carefully excluded.On distilling off the ether, the base was left as a mass of slenderneedles.It was recrystallised from ether or light petroleum, inboth of which i t is readily soluble. It separates from either solventin slender, colourless, needle-shaped crystals, melting a t 94O :0.5349 gave 0.8631 AgC1. C1= 39.91.C6H,N,C12 requires C1= 40.0’7 per cenC.When strongly heated alone i t slowly decomposes into2 : 4-dichloroaniline, m-dichlorobenzene, nitrogen, and ammonia,and when oxidised by Fehling’s solution or alkaline chromateyields mdichlorobenzene and nitrogen.A c e ty l-2 : 4-die h.1 oroph e rty 1 hy drazine, C6H,C12.N~*NH*CO*~,H,.2 : 4-Dichlorophenylhydrazine reads vigorously with aceticanhydride when the materials are mixed in equivalent amounts,much heat is evolved, and the acetyl compound is formed. Thiscrystallises from alcohol, in which it is moderately soluble, inslender, colourless prisms, melting at 1 5 7 O :0.3511 gave 0.4608 AgCl.C1= 32.46.C,H,ON,Cl, requires C1= 32-38 per cent.VOL. CVTI. 34 CHATTAWBY AND PEARCE : 2 : ~PDICHLOROPEENYLEYDRAZINE.Pro piony l-2 : 4-dichl orophen y 1 hydrazine,C6H3C1,*NH~NH~CO*C,H,.This compound, prepared similarly from propionic anhydride,is very readily soluble in chloroform and sparingly so in lightpetroleum, It crystallises from g. mixture of the two in clustersof very small, colourlew needles, melting a t 102O:0.4911 gave 0.5958 AgC1.C1=30.01.C,H,oON2C12 requires C1= 30.43 per cent.Benzoyl-2 : 4-dkchlorop~~enylhydruz~~e,C,H,Cl,*NH*NH* CO=C,H,.This compound is formed when benzoyl chloride is allowed toact on 2 : 4dichlorophenylhydrazine suspended in water, theliberated hydrogen chloride being neutralised by dilute alkali. Itcrystallises from alcohol, in which it is moderately soluble, insmall, colourless, needle-shaped crystals, melting at 168O. (Found,GI = 25.07. C13H,,0NzCl, requires C1= 25-23 per cenb.) Ponzio(Gazzetta, 1909, 39, i, 661) describes the compound as crystallkingfrom benzene in yellow needles, melting a t 166O.Benzaldehyde-2 : 4-dichlorophemg lhydruzorce,C,H,Cl,*NH*N:CH 'C6H5.This compound is formed, with evolution of heat, when alcoholicsolutions of equivalent amounts of benzaldehyde and 2 : 4-dichloro-phenylhydrazine axe mixed. It is readily soluble in hot alcohol,from which it crystallises in colourless, rhombic plat&, meltingat 107O:0'4228 gave 0'4535 AgCl. Cl= 26.53.Cl,HloN2C1, requires C1= 26.76 per cent.Cinnamaldehyde-2 : 4-dichlorop~~enylhy~ruzorre,C,H,CI,*NE*N:CH*CE:CE*C,H,.This compound, which is similarly prepared from cinnam-aldehyde, is very sparingly soluble in boiling alcohol, but morereadily so in bailing acetic acid. It orystallisee well from eithersolvent in pale ydlvw, four-ided, rhombic plates, melting a t 163O :C1,H,,N2C1, requires c1= 24-36 per cent.0.3188 gave 0.3129 AgC1. C1= 24.28.UNIVERSITY CHEMI~AL LABORATORY,OXFOBD

ISSN:0368-1645    

DOI:10.1039/CT9150700032

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

1NVESTIGATIOhS ON DEPENDENCE OF ROTATORY POWEIL, ETC. 35VII. -Investigatioiis on the Dependence qj’ RotatoryPower on Chemical Constitution. Part XI.TJw Co-ordination of tlze Rotatoqy Powers (a) of’Merzthyl Compounds, (13) of the iMenthoYLes, and (c)of’ the Borneols.By JOSEPH KENYON and X~OBERT HOWSON PICKAKD.A SURVEY of the rotatory powers of the optically active carboncompounds shows that those of many derivatives of I-menthol andZ-menthylamine exhibit striking regularities.Thus homologous series of the Z-menthyl esters of normal fattyacids (Tschugaev, J. Russ. Phys. Chem. SOC., 1902, 34, 606), ofthe I-menthylamides of the same acids (Wallach and Biiiz,A rwalen, 1893, 276, 317), and of various derivatives of I-menthyl-carbamic acid (Pickard and Littlebury, T., 1907, 91, 300), havebeen shown to exhibit approximately a constant molecular rota-tory power [MID.So regular have these values proved to be thatsome investigabrs (compare Hilditch, inter ulia, T., 1909, 95,1570, and Frankland and O’Sullivan, T., 1912, 101, 287) havebased somewhat elaborate deductions on such a ‘‘ series constant.”Although this question has already been briefly discussed(Pickard and Littlebury, Zoc. cit.), it appears desirable to attemptto discover furhher relations among the rotatory powers of thesecompounds. This is all the more important, as the comparison ofthe rotatory powers of unit mass, that is, of the molecular rota-tory power, tends to give overwhelming importance to the factordue to the mass of the molecule with a corresponding obliterationof the effect due to the chemical constitution of the compounds.From tkieoretical considerations, Armstrong and Walker (PTOC.Roy.Soc., 1913, [A], 88, 388) have devised a method of plottingrotation values (known as “ characteristic diagrams ”) whichsuccessfully explains a t least some, if not all, of the cases of so-called anomalous dispersion, and the present authors haveelaborated this method in a somewhat empirical manner, so thatit has been applied in many instances in the more recent papersof this series of communications to the co-ordination of the specificrotatory powers of groups of compounds of allied chemical con-stitution.It has now been found that one characteristic diagram will serveapproximately to co-ordinate the specific rotatory powers ofI-menthol, I-menthylamine, and many of their simpler derivativs.Thus in Fig.1, after plotting the values for the mercury-green0 36 KENYON AND PICKARD : INVESTIGATIONS ON nEPENDENCE- 46'- 60'- 80'- 100'- 120'- 140'- 160'- 180'- 200'FIG. 1.I1. Ment hy la mi ne h yd ro-chloride in H20.2. Menthylamine at 20".3. Menthol in CBH~.4. Menthylamine hydro-chloride in CHCI:+12. Menthyl p-fluorobenzoate in13. Menthyl p-flzorobenzoate in C2H5'0H.14. ,, benzoate in CGHG.15. ,, menthylcarbamate in CHCI:%.16. Dirnenthyl oxalate in CGHG 9The broken lines are copied from the lines for sodium-yellow and mercury-violetFor the sake of clearness only B few points have been lights drawn in Fig.3.placed in each figureOF ROTATORY POWER ON CHEMICAL CONSTITUTION. 37light on a reference line of unit slope, it is found that the corre-sponding values for sodium-yellow and mercury-violet lights fallon two straight lines, of which the line for sodium-yellow intersectsthe reference line where [a] is zero, and that f o r mercury-violetwhere [a] is -3O. The compounds of which the specific rotatorypowers have been found to conform to this generalisation are:Z-menthol, the Z-menthyl esters of acetic, oxalic, benzoic, m-toluic,phenylacetic, pfluoro-, m-chloro-, m- and piodo-, 2-chloro-6-bromo-,pmethoxy-, p-ethoxy-, o-propoxy-, pisopropoxy-, and pallyloxy-benzoic acids ; * dimenthyl sulphite ; Z-menthyl hydrogen phthalate(in some solvents) and also its magnesium salt ; Z-menthylamineand its hydrochloride ; dimenthylcarbamide ; Z-menthyl Z-menthyl-carbamate and several other amides and esters of Z-menthyl-carbamic acid containing the f dlowing radicles : ethyl, n-propyl,mbutyl, allyl, m- and ptolyl, and m-4-xylenyl.Perhaps morestriking than the list of compounds conforming t o this generalisa-tion is, however, the list of compounds of allied constitution withspecific rotatory powers which do not lie on this characteristicdiagram for Z-menthyl derivatives. This includes the menthylesters of o-iodo-, the mono- and di-nitro-benzoic and 2-methoxy-naphthoic acids, 1- and 2-naphthyl esters and amides of Z-menthyl-carbamic acid, the sodium salt of Z-menthyl hydrogen pht'halate inaqueous solution, and (probably) compounds of the type of m-tolyl-menthylcarbamide when dissolved in pyridine.Many of the perplexing phenomena connected with the rotatorypowers of carbon compounds have been ascribed to the occurrenceof dynamic isomerism, and certainly the assumption of the exist-ence of two isodynamic forms (with rotations of opposite sign andof different dispersive powers) of an esterified carboxylic group orof a substituted a-naphthyl group has allowed of a consistent ex-planation of the cases of the so-called " anomalous " dispersiondescribed in this series of communications.I n the cases of theseinenthyl compounds, f the departures from the generalisation givenabove (and in the majority of cases the compounds are those ofwhich the molecular rotatory powers differ from the '' seriesconstant") can be accounted for in the foIIowiiig manner.Themethod of plotting rotation values known as " characteristicdiagrams " allows for the presence in the otherwiss homogeneous* The authors are very niuch indebted t o Prof. J. B. Cohen, P.R.S., who hasmost generously placed at their disposal much costly :end highly purilied materialprepared by him for a11 extended investigation of the nienthyl esters of substitutedbenzoic acids of which a connected account has been given recently (T., 1924, 105,1892).t None of which, however, exhibits under the available experiinental conditioiis aso-called anomalous dispersion38 KENYON AND PICKAKD : INVESTIGATIONS ON DEPENDENCEcompound of two dynamic isomerides, but fails when more thantwo are present (see Part X, T., 1914, 105, 2677, for instances ofthis already described).Thus in the cases of the menthyl com-pounds, the characteristic diagram for the menthyl group co-ordinates t'he specific rotatory powers of I-menthyl derivatives solong as these do not contain two dissimilar centres of possibledynamic isomerism. It can here be assumed that the amide groupGO-NHR is as much a possible centre of dynamic isomerism asthe ester group *CO*OR has been showii in all probability to be.The presence of two similar centres of possible dynamic isomerismin a compound does not seem to prevent the co-ordination of itsrotatory powers, but the evidence on this point is a t presentslender, being derived only from an examination of dirnenthyloxalate.It may be observed from Fig.2 that the yellow and green andviolet lines all intersect very nearly where [a] is zero; as it resultof this, any one of the three dispersion ratios violet/yellow,green-yellow, violet /green is very nearly constant' for all compoundsof which the rotation values fit on the diagram; the first aad lastratios, however, will tend t o be a function of the magnitude of therotation, and decrease with its magnitude.* I n thO particularcase, then, of lihese menthyl compounds, the approximate constancyof the dispersion ratio serves as a test of the normal character ofthe optical activity of the compound.+The various exceptions to this co-ordination of the specific rota-tory powers of the I-menthyl compounds may now be discussed.Itwill be observed that in the case of every exception the compoundcontains some modification of the carboxylic group, which itself isa possible centre of dynamic isomerism. Perhaps the most strikingcase is the mentliyl ester of o-iodobenzoic acid, since it offers sucha striking contrast to the corresponding m- and p-compounds. Itis suggested that in the o-compound there is a second possiblecentre of dynamic isomerism, owing to the possibility of theformation of the complexSomewhat in a similar manner it is suggested that the esters of* Since the violet line does not intersect the others exactly where [u] is zero,slight discrepancies from this statement among the figures quoted in Tables I.and 11.are probably due to errors in the "violet" readinge, which are the lesscertain of the three and are often very much affected by any tinge of yellow colourin the substance.i- Compare, however, remarks on the use of dispersion ratios in Part X., Zoc. citOF ROTATORY POWER ON CHEMICAL CONSTITUTION. 39the nitro-substituted benzoic acids may contain either the group-ing *N<L or "go, An assumption of the existence of suchisomerides, however intangible they may be, appears suEi$ent toexplain the large effect of solution on these esters of nitrobenzoicacids. Here again the high rotatory powers of the o-nitrobenaoic00100"80"60"40"20"0"- 20"- 40'- 60"FLQ. 2.I- - I Oontinuati of0krrcCsri.I~ Omgramfor Derlvativer ofand I-Neomenthol1.Mixture of hydrogenphthalates of d-neohen-tho1 and I-menthol (seeT., 1912. 101,128) in %He.2. Ditto in CHCIp3. Ditto in C2ks4OH.4. Menthyl 2-chloro-6-bromo-benzoate ih C&.5. Menltrylamine hydro-chloride in H2O. *ester, and inter alia of the 2 : 6-dinitrobenzoic ester, may bedirectly due to the influence of the o-nitro-group on one of theisodynamic forms of the carboxylic group. Some confirmation ofthis view is afforded by the similar case of the 2-chloro-6-bromo-benzoic ester, which when dissolved in carbon disulphide has anabnormally low rotatory power. This idea may be extended togive an alternative explanation of the successive variation of th40 KENYON AND PICKARD : INVESTIGATIONS ON DEPENDENCEmagnitude of rotatory powers of the menthyl esters of substitutedbenzoic acids put forward by Cohen (Zoc.cit.) t o connect positionisomerism and optical activity. Thus it may be suggested thatthe substituent lying nearest to the carboxylic group (not, asCohen suggests, to the active group, although in the case of theseesters the distance is measured from the same group) * producesthe greatest effect, which, according to the nature of the sub-stituent, may be either an increase or a decrease in the rotationvalue following upon an alteration in the proportion of the twoisodynamic forms of the carboxylic group present in the compound.The consistent application of these ideas to so many compounds ofvaried type favours the underlying idea of much research inthis field that the rotatory power of a derivative of (at least) asecondary alcohol is actually some function of that of the parentcompound.The abnormality of the dispersion ratios of the menthyl esterof 2-methoxynaphthoic acid and of the 1- and 2-naphthyl esters andamides of I-menthylcarbamic acid is readily explained by theassumption of the form of dynamic isomerism of the naphthylgroup, which has been suggested as the explanation of the complexdispersion of the 1-naphthylalkylcarbinols described in Parts VIand IX (T., 1914, 105, 1115, 2644).The hypothesis can be readily applied t o the case of the rotatorypower of the sodium salt of I-menthyl hydrogen plithalate inaqueous solution, as it is easy to suggest more than two isodynamicforms of the active compound when dissolved in water.It has already been shown (Pickard and Littlebury, Zoc.cat.)that the aryl amides of I-menthylcarbamic acid exhibit in pyridinesolution molecujar rotatory powers [MI, which are considerablyhigher than those observed in chloroform solution. Probably allof these, like the ?4dylmenthylcarbamide, the only one nowreexamined, have in pyridine solution dispersions different fromthat of the majarity of inenthyl compounds, and this difference maybe caused by these aryl menthyl carbarnides acting partly in theisocarbamide form when dissolved in this solvent.Additional evidence in favour of the underlying hypothesis toexplain these phenomena is afforded by the characte'ristic diagram(Fig.3) which, drawn in the usual manner, serves to co-ordinatethe rotations of the esters of mono- and di-nitrobenzoic acidsobserved under many conditions of temperature and solution. I twill be seen that the diagram, like the dispersion ratios, differsgreatly from that in Fig. 1.* Analogous caws of the derivatives of' optically active acids of simple constitu-tion where the optically active group is t h e one to which t l i u estmified carboxylicgroup is attached are under examination in this laboratoryOF ROTATORY POWER ON CHEMICAL CONSTITUTION. 41It has been shown by one of us and Littlebury (2, 1912, 101,109) that I-menthol and d-neomenthol correspond with I-rnenthone.F I G . 3.SO"- 140'- 200- 260'n %2& -320'u--ah -uhu 5.-880$0 -- 440- 500- 560- ($20for Manthyl Ester,of NitrobenzoicAci~~ ~- Characteristic DiagramMcnthyl Es1. Menthyl 2:g-dinitro-benzoate in C6H6.2. Menthyl o-nitroben-zoate i n C6H6.6. Menthyl o-nitroben- -roate i n CHC13./ 8. Menthyl o-nitroben-/// 7. Menthyl o-nitroben- -// Menthyl m-nitroben-zoate in pyridineMenthyl 3:5-dinitro- //3. Menthyl 6 - chloro - 2 -nitrobenzoate i n CSg.:5. Menthyl 5 - chloro - 2 -n i t r o b e n z o a t e inCHC13.12. Menthyl 2 :4-dinitro-benzoate in pyridine.14. Menthyl 2 ; 4-dinitro-benzoate i n CHCl13. Menthyl 4-chloro-3-~~tro-- benzoate in pyridine.15. Menthyl o-nitrobenzoateHomogeneous a t 20".16.Menthyl o-nitrobenzoile.Homogeneous a t 100".17. Menthyl m-nitrobenzoate.Homogeneous at 20".It is significant that the points, for exairilde, Nos. 2, 6, 7, 8, which do notThe diagiamTheexactly fit on this diagram, refer to derivatives of o-nitrobenzoic acid.covers a much greater range (from [u] 60" to 660") than the other diagrams.broken line is the violet line copied from Fig. 142 KENYON AND PICKARD : INVESTIGATIONS ON DEPENDENCEThus, as the two menthols possess corresponding configurations, itis not surprising that the characteristic diagram for I-menthylderivatives serves also to co-ordinate the rotatory powers ofci-neomenthoi and its derivatives (see Fig. 2).When I-menthone is treated with alkalis o r acids, it is partlyconverted into isomenthone, and yields a mixture which is gener-ally dextrorotatory.Since the rotatory powers of I-menthone and of various mixturesof ketones produced from it by inversion can be co-ordinated 011one characteristic diagram (see Fig.4), i t may be inferred that theinversion takes place without any appreciable amount of racemisa-tion (compare Tutin and Kipping, T., 1904, 85, 66).Further, the configuration of lzevorotatory borneol correspondswith that of dextrorotatory isoborneol, since each gives whenoxidised the same I-camphor (Pickard and Littlebury, T., 1907,91, 1973). As may be therefore expected, one diagram (seeFig. 5) serves to co-ordinate the rotatory powers of I-borneol,d-isoborneol, and their derivatives, whilst the irregular behaviourof the hydrogen phthalates of the borneols (like that of the corre-sponding menthyl compound under some conditions) suggests thatthe two groups *COOOR and *CO*OH can function as two dis-similar centres of possible dynamic isomerism.It may therefore be suggested that the extension of this methodwill result in the recognition of many of the more elusive formsof isomerism which have been postulated in so many instances t oexplain some of the more obscure reactions in organic chemistry.The great majority of the observed rotations recorded in thispaper conform approximately to the Drude equation with oneterm, a = k / ~2 - ~ 2 , .That the dispersive power of these compoundsis found thus to be simple is not perhaps very surprising, for thereappear t o be other instances known where a mixture of isodynamicforms of a substance exhibits simple dispersive power; thus, forexample, the solutions of fructose at various concentrations de-scribed by Armstrong and Walker (Zoc.cit.) exhibit simple rota-tory dispersion.However, some of our observed rotations do not conform t o theDrude equation with one. term, and the following list of these sub-stances which in the homogeneous state o r in solution exhibitscomplex dispersive power shows many analogies with the resultspreviously recorded : solutions of the menthyl esters of o-nitro-,2 : 4- and 3 : 5-dinitrobenzoic acids in benzene, pyridine, chloro-form, and alcohol; solutions of the hydrogen plzthalates andof the menthyl ester of 2-methoxynaphthoic acid.Furtlier,the following compounds a t high temperatures exhibit compleOF ROTATORY POWER ON CHEMICAL CONSTITUTION. 43dispersive power : the inenthyl esters of acetic, phenylacetic,p-methoxy- and o-nitro-benzoic acids, I-bornyl acetate, Z-menthyl-40"20"0"- 20"- 40"- 60"/ Characteristic Diagram fbrI-llllenthone and d-isohllenthone*d -1. imMenthone (A) a t 2 0 .2. i8oMenthone (B) at 19".3. Menthone (C) at 20".4. Z-Menthone in C6H6.6. ,, ,, C2H5'OH.6. ,, at 20".7. ,, ,, 140'.Points 6 and 7 refer to the rotations of a sample of pure I-mcnthone investigatedin the homogencons state in a glass polarimeter tube. The sample after heating to180" was found when cold to have a lower rotation and was doubtless partlyinverted owing to enolisation iaduced by the alkali of the glass.Points 3 and 5refer to I-menthone dissolved in benzene and alcohol respectively. Poiut 4-refers tothe rotatory power of a partially inverted I-menthone (C) which had been preparedby heating 2-nienthone for a short time with a solution of sodium hydroxide.The samples of isomenthone (A and B) to which points 1 and 2 refer were obtainedas follows : I.mcothone was partly reduced by the Sabatier process and the resultingmenthols removed by treatment of one preparation with benzoyl chloride and of Rsecond preparation with phthalic anhydride. From the residues of these two pre-parations were obtaincd the two samples (marked A and B) of isomenthone44 KENYON AND PICKARD : INVESTIGATIONS ON DEPENDENCEamine, and I-menthone (having then undergone some inversion).It seems probable that the rotlations of a substance displayingdynamic isomerism will conform to the law of simple dispersionif the two isodynamic forms have the same dispersive power,Frc.5.175'145"115"85"2 55"5& -5"0 F4Pa. 9 25"- 35"- 65'- 95"Characteristic Diagram forI-Borneol, d-isoBorneol,and their Derivatives,-6.7.9.I-Borneo1 in CzHs'OH.I-Bornyl acetate. Homogene-I-Bornyl acetate. Homogene-1-Bornyl hydrogen phthalateI-Bornyl hydrogen phthalaterl-isoBorneol in CHCI,.d-isoBornyl hydrogenphthalate in CHC13.I d-isoBornyl hydrogenphthalate in C2H,'OH.ous at 20".ous a t 180".in Cr,H6.111 CzHs'OH.I ,, C2Hs'OH.i n which case the lines of its characteristic diagram will intersectwhere [a] is zeroEXPER I M E N T A L .With two exceptions, all the compounds, which have now beensubmitted t o an extended re-examination in the polarimeter, havOF ROTATORY POWER ON CHEMICAL CONSTITUTION.45been described in the investigations of Cohen and his pupils (Zoc.cit.) and in those of one of us and Littlebury (Ioc. cit.).Di-1-menthyl sulphite, (C,,H,,O),SO, is readily obtained whenequivalent amounts of thionyl chloride and I-menthol, each dilutedwith much light petroleum, are slowly mixed and allowed t o reactat a temperature not exceeding 6O. The ester is separated fromunchanged menthol by fractional distillation. It boils at 210°/9 mm., and solidifies to a white mass, which crystallises readilyfrom dilute alcohol in colourless, glistening needles, melting a t 52O.It was analysed by titration with iodine of the sulphurous acidformed on hydrolysis.Found, S= 8-66.C,,H,03S requires S = 8.94 per cent.Magnesium 1-Menthy1 Phthalate.-This salt is precipitated onmixing solutions of magnesium chloride and sodium I-menthylphthalate, and crystallises very readily from dilute alcohol inglistening flakes.The menthyl esters on loan from Prof.Cohen were in themajority of cases used in the condition as received from him. Therotations observed for sodium light were generally in close agree-rnent with his results. Some of the esters were redistilled a t a lowpressure, and the very slight diminution of rotatory power whichresulted was not taken into account.The general experimentalconditions for the determination of density and of rotatory power(both in solution and in the homogeneous state) were the sameas previously described in this series of investigationsTABLEI.Rotations* of-[aID* 48.96" 49.55 50.02 49.97 49.48 48.78 48.15 47.58 47-09-1-79.41' 79.20 78-98 78.89 78.57 78.34 78-11 77-89 77.62VariousCompoundsintheHomogeneousStateatDifferentTemperatures.[abe*-51.39" 52.10 52.34 52-06 51-53 50.94 50.34 49-76 49.09-83.01' 82.73 82.47 82-31 81.98 81.74 81.51 81.26 81.00[algr* 58.80 59.30 58-95 58-21 57.48 56.78 56.14 55.59-57.90"-94-12' 93.58 93-50 93.28 92.90 92.64 92.34 92.04 91-70DispersionI-Menthol.[a]+ 97.61 98.15 97.97 97-03 95.82 94-66 93.46 92.04-96.61'-[MID.76-19'-77.29 78.03 77.94 77.18 76.10 75.11 74.23 73.45CM14.e. -80.13" 81-27 81-66 81.21 80.39 79-45 78-52 77-62 76.58I-Men t hy l A ce tat e.156.1"-155.7 155.2 155.1 154-4 153.9 153.4 152.9 152.4.157.2"-156-8 156-4 156.3 155.6 155-2 154.6 154.3 153.7164.4" 163.9 163.3 163.4 162.4 161.8 161-4 160.9 160.4CMlgr.-90.33" 91-73 92-51 .91-96 90.80 89.66 88-57 87.58 86.72-186.4' 185.8 185.1 184.7 184-0 183.4 182.8 182.3 181.6CM1vi.-150.7' 152.3 153.1 152.9 151.4 149-5 147.7 145.7 143.5-309.1" 308.2 307.3 307.0 305.7 304.8 303.8 302.8 301.8Hgviolet,2.978 1.970 1-962 1-961 1.961 1-964 1.966 1-963 1-954N&yelloWHggreenNagellow 1.9661.9651.9651.9661.9651.9651.9641.9631-963recordedinthispaperarenegativeunlessotherwisestated.recordedonpage60 etseq.ThevaluesinTableI.arededuce[(*ID.-93.46" 93.28 93-11 92.87 92-64 92.10 91.68 91.11 90.38-68.06" 67-41 66-65 65.66 64.74 63.89 62.95 62.07 61-55[alw 97.40 97.21 96.99 96.61 96-10 95.43 94.77 94.04-97.61"--71.34"-70.44 69-56 68.58 67.54 66.62 65.67 64-92 64.40[algr.110.4"-110.1 109.9 109.6 109-2 108.8 108.1 107-4 106.681.14"-80.09 '79.07 77.87 76.72 75.80 74.87 73.91 73.02TABLEI.(continued).I-Dimen t hylOxulat e.[ah. 1S2.0"-181.9 181.5 181.0 180.3 179.6 178.9 178.0 177.0[Jw". -342.1" 341-4 340-8 339.9 339-0 337-1 335.5 333.5 330.9CMIw -357.3O 356.6 355.8 355.0 353-6 351.8 349.2 346.9 344.2I-MenthtjlPhenylaceta te.-137.4"-135.6 133.8 131.9 130.0 128.1 126.1 123.9 121.9-186.5"-184.7 182.4 179-9 177:4 175.1 172.5 170.1 168.7195.4' 193.1 190.6 187.9 185.1 182.6 180.0 178.3 176.5[MI,=. -404.0" 403.1 402-3 401-0 399.5 398.2 395.7 393.2 390.1-222.3" 219.5 216.7 213.4 210.2 207.7 205.1 202.5 200.0[Ml,i* -666.3" 666.6 664.4 662.6 660-0 657.4 655.0 651.5 647.7-376.5" 371.6 366.5 361.3 356.2 350.9 345.4 339.7 333.9Dispersion1.9481.9491-9491.9491-9461.9501-9531.9541.9582.018 2.012 2.009 2.008 2-003 1-996 2.002 1.997 1.9801.190 1.191 1.18TABLEI.(continued).Wenthylp-Nethoxybenzoate.[a],,.[alvi-[MI,.[Mlye.[Mlp103.6175.7252.9262.1300.6101.4170.1247.7258.7293.999-04167-2242.2253.0287.296.70163.2236-8247.4280.494.64169.1231.5241.9274.592.47155.4226.3236.9268.190.28151.9221.2231.4261.888-35148.7217.1226.2256.3-10296"-174.7'-250.0"-261.3"-297.5"1-Menthylo-Zodob enzost e.-73.57"- 130.8"- 236.8"-248.2"-284.0"73.16130.4236.6246.7282.4I-Mentliylm-Zodobenzoate.I-Menthylp-Zodobenzoate.I-Menthylo-Nitrobenzoate.-72.67"-121.8"-236.3"-246.6"-280.5"-75.30"-126.7"-245.6"-256.7'-290.7"-161.3"-381.9"-392.4"-425.8"-492.0'157.9372-7384.7408.0481.5154.4362-9377.0399.5470.8161.0352.4369.6390.8460.4147.6341.8362.9381.8450.2145-5-357.9376.5443.9143.9-353.5372.5438.8142.1-349.0368.1433-4140.2-344.4363.8427.6Dispersion[alD* -86-22' 87.42 85-39 83.52 81-64 79.82 78-03 76.28 74.87[alue.- 90.12" 90.38 89.43 87-44 85.31 83.45 81.68 79.79 78.00[Mlvi. -505.7" 509.5 493.3 484.8 473.3 461-4 450.7 440.5 431.12.026 2.013 1.992 2.001 2.004 1.993 1.992 1,991 1.986- 61.35" 61.02-64.30" 63.90- 504.8' 503.22.1312.1361.989- 61.22"-63.90"-470.2"- 63-63'-66.60"-488.9"1.991-136.4' 133.8 131.0 128.1 125.2 123.4 122.1 120.7 119.3-1164.7" 1136.7 1106.3 1074.8 1042.5 -2.9682.9552.9352.9082.873-----128.7" 126.1 123.6 121.2 119.0 117.3 115.9 114.4 112.TABLEI.(contimed).I-Menthqim-Nitrobenzoate.Dispersion-82.52" 82.14 81-70 81-06 80.39 79.68 78-93 78.42 78.70-70.99" 70.86 70-73 70.60 70.34 70-18 69.98 69.08 67.33- 79.76"- 86.30' 86.93 85.43 84.76 84.00 83.28 82.59 82.27 82.56-73.98" 73.91 73.72 73.57 73.40 73-16 73.05 71.86 70.44-83.55"-98.44'-173.6"-97.90-97.25-96.44-95-72-94-95-94-18-93.66-93.65-.251.7' 250-5 249.2 247-2 246-2 243.0 240.7 239-1 240.1-263.2"-262- 1 260.6 268.6 266.2 254.0 261.9 250.9 251.2I-men thyp-Nitrobenzoate.-83.33"--83-45-83.28-83.12-82-83-82.59-82.47-81.64-79.50--216.5"-216.1 215.7 215.3 211-7 214.1 213.4 210.7 205.3,225.6"-225.4 224.9 224.4 223.9 223-1 222.8 219.2 214.8-300.2" 298.6 296.6 294.1 292.0 289.6 287.2 285.7 285.7.264*2" 264.5 254.0 253.5 262.6 261.9 251.6 249.0 242.8I-Menthylp-AIhJoxybenzoate.-94.14"-164.60"-252.1"-264.0"-297.5"1.192 1.190 1.190 1.191 1.192 1.194 1.1901.174 1-178 1.177 1.177 1-177 1.177 1.176 1-181 1-1821.18TABLEI.(continued).I-men thylamine.Dispersion[Qlw-39.97" 40-81 41-52 42.14 42-76 43.40 44.01 44-61 45.02"XIw - 41.63' 42.51 43.24 43.96 44.62 45-21 45.75 46-20 46-69[Qlvr.-47.04' 48-07 48-88 49.58 50.27 5141 51.69 52.32 52-81[a]ri.78.03 79.47 80.85 82-22 83.75 85-09 86-52 87.78-76.40"-[MI,,. .64*52" 65.89 67-02 68.14 69-15 70-06 70.91 71.61 72-36[Mlyr. - 72-91' 74.50 75.75 76.85 77.91 79.07 80.11 81.10 81.85[Mlvi. -118.4" 121.0 123.1 125.3 127.5 129.8 131.9 134-1 136.01.911 1.913 1.914 1.919 1.923 1.930 1.933 1.939 1.9501.179 1.177 1.176 1.174 61.95"-63.24 64.34 65.31 66.27 67.26 68.22 69.13 69.77 Wenthone.*-26.74" 27-66 28.36 29-32 30.43 31-41 31.63 31.48 31.77- 27-84' 28.89 29.79 30.67 31.77 32-88 33-13 33-06 33-09-32-07' 33.29 34.38 35-48 36-82 38-18 38-63 38.43 38-37I-57.15" 59-92 62.61 65-55 68.84 71.83 72-13 71.48 7 1.52-41.17' 42.57 43.68 45.15 46.86 48.37 48.71 48.48 48.91-42-87" 44.49 45.76 47-23 48.93 50.63 51-02 50.90 50.95-49-39" 51-24 52.93 54-64 56-70 58.79 59.48 59-19 59-09-88-0O0 92.28 96-40 100.9 106-0 110.6 111-1 110.1 110.12.137 2.168 2.207 2.236 2-262 2.288 2.280 2.271 2.2521.221 1.221 'SeefootnotetoFig.4,page43LaJw+ 3.74'+ 4.25'+ 11.56'+ 12.32'-18.74"- 19.49'-42.43' 40.75 39-06 37.64 36.44 35.22 34.00 33.04 32.30-44.16" 42.49 40.71 39.22 38-08 36.84 35-58 34.47 33-55TABLEI.(coiitinued).Menthone,4 .*[~]~p[aIvi.[MI,.[MJye.[MI,,.[Mlvi.t5.61"+20*55'+5.76"+6.55'+8*64'$31.65"MenthoneB.*+ 15.22'+40*10'+17.80"+18.97'$23.43"+61.76'-550.18'-48.19 46.18 44.52 43.16 41.80 40.46 39-09 37.67I-BorrtylAcetate.'84.33' 80.98 77.49 74.56 72.40 70.04 67-79 65.55 63.43-83.18'-79.87 76-58 73.79 71.44 69.02 66.65 64.76 83.33'86.56" 83.30 79.82 76.90 74-64 72.21 69.74 67.57 65.75* SeefootnotetoFig.4,page43.-98.36'-94.48 90.53 87.28 84-61 81.94 79.30 76.63 73.84163.5' 158-8 151-9 146.2 141.9 137.3 132.9 128.5 124.3Dispersionf-Hgviolet. Xayrl~ow 5.4943.4681.9351.987 1.987 1.979 1.981 1.986 1,989 1.994 1.985 1.963Hggreen1.500 NaWeightofTABLE11.DeterminationsofRotatoryPowerin(~pprox.)5percent.Solutions.solute, grams.a,.1.0328-5.11'1.05714.971.10645.551,03607.981,15829871.055610.330.95897.87098499.670.94999.791.04034-701.03688.981.00629791,01523.27aye- - 5*35O 5.19 5.798.42 10-3110.75 8-23 10.10 10.184.89 9-4010.203.42l-MenthoE.- 6*11°- 10*03'- 4946'- 51.80'- 59-14'- 97.09'- 76-77'- 80.28'- 91.86'- 15O.5'5.889.8547-0349.1155.6493.1972.9076.1286.24144.410.8044-6046.5252.7886.7869.1172.1181-83134.56.579.45 11.6912.20 9.27 11.44 11.585.55 10.8011.623-87I-Menthy 2 A ce tad e.15.6077-0181.2791.20150.6819.4085.2389.02100.93167.50I-DimenthylOxaEate.20.2897.86101.84115.57192.1215.1682.0885-8296.681574918.9998.18102.55116.2519241192193.6997.43110.83183.84I-MenthylBenzoate.9.2990.3896.01106.70178.6017.6086-6190.66102.24169.75I-Menthylm-Toluate.19.5588.4792.17104.99176.63I-MenthylPhenylacetate.6.5264.4367.39i6.26128.45151.7160.1167.9175.4358-1372.7300.4314.2359.4375.3342.9358.6234.1248.6224.3254.8242.4252.7176.0184.7179.7 198.9423.0 353.8 425.1 405-6276.5 264.8287.7208.9296.6 330.1703.2 578.3 705.7 672.9462.7 439.6484.1352.11*9831.9821.9451-9541.9651.9631.9251.9641-96!1.9761.9601-9961.99Weightofsolute,gram. 1.0624 0.9695 0.9477 0.9108 0.96541.13501.0674 1.2382162l13 1.14220.965904432TABLE11.(continued).DeterminationsofRotatoryPowerin(approx.)5percent.Solutiom.a,. 8-86" 7.92 8.80 4.11 8.389-076.79 3.967-04 3-880.767.09aye.902' 8.28 9-14 4-26 8-7610.117-09 4.107.98 4-120.807.35I-Menthylp-Fluorobenzoate.agr.a>l.[a],.[alye.[a]g1.[a],,.[MI,.10.21'17.06"81-45"84.W96.14160.58226~4~9.3615.5081.7086.4196.56159.87227.110.3617.2684.3781-6299.33165.57234.54-808.2690.2493-33106.72181.38250.89-9316.71864090.74102.86173.09241.3I-Menthylm-Chlorobenzoate.11.5019.5085.1989-07100.13171.80250.9I-Menthylrn-Zodobenzoate.8-0213-4058.3862-3768.96115-20225.44.637.9163.9766.2474-80127.77246.9I-Menthy1p-lodobenzoate.9-0214-8757.3659.9167.72111.61221.44.697.9869-6972-1482.13139.73269.0I-.JIe,zt?iyl2-Chloro-6-bromobenzoate.0.911.6315.6916.7419.0439.5858.7l-Menthylp-Ethox y benzoa.te.8-3813.8576-1577.938844146.84225.5"Ye. 237.2" 2374 234.5 259.4 262.3262.3240.7 255.7231.3 278.562.7236.9298.4506.0266.2444.7288.7t493.3261.4430.9317.1539.471.3127.0270.1448.51.972 1.957 1-081 2.010 1.9952-0171.973 1.9971-946 2.0042.1731.95Weightof-solute, grams.0.97940.06780.98460.99420.76311.0104 1.01900.9920Determinationsaye* 8.1'7.078.050.734.346.08 7.687-04agr. 6.89"0.069.1811%4.901-88 8.75BellTABLE11.(continued).ofRotatoryPowerin(approz.)5percent.Solutions.I-Menthy 1o-Propox y b en zoa t e.av1*[ah[a]~.[a]g~[o]vi.[MID.11.30"59.82"62.37"70.34"115.45"190.2"l-Menthylp-isoPropox9 benzoat e.15.4679.5683.2294.58161.41263.0Di-l-menthylSulphite.15.3578.8181.7593.22155.90282.1I-MenthylHydrogenPhthalate.20.4093-74W*88113.15205.19284.0Magnesium1-MenthylPhthalate.8.2055-2357.626546108.88180.0I-MenthylHydrogenSuccinate.130059.4362.2470.47116.27152.2144565.6568.6078.05128.91168.1Wenthy1Phenylcarbamate.15.2076-0180.0491.83153.22211.4CMIre- 198.4"264.6292.7296.6187.8160.3 176.4220.4300-7613.3333.7658.1342.8[62l.7212.0364.9180.4287.7199.7330.0262.4421.41.9292.0291.9782.1991.9721.0561.9631.99WeightOfsolute, gram.ah.1w4-09.0.99744.660.93688-660.78815-400.55573.720.88546.250.87055-660.39302-22aye -4-22" 4.758-956.623.860.506.102.30agt* 4.80" 5-4010.126.434-407.426-95243TABLE11.(contiimed).I-MenthylamineHydrochloride.avi.[alD.[a],,.[alp[a],i.[MID*[Mlye.[MIgr*1-81"36.29"37-66"42-60'69.85"43-38"8.9645-7249.8654.1489.8287.54Di-l-men t hylcarbamide.16-9891.2695.53108.02181.25306.6EthylI-Ment hylcapbamate.10.7968.5271.3181.58137-0155.5n-Propyl-1-ment hy lcarbamide.7-4066.9469.4779*20133.17159.3n-Buty1I-Ment h yl curbamate.12.5464.1866.7476.19128.76163.6Allyl-I-menthylcarbamate.11.7567.3370.0879,86135.98160.9p-Tolyl1-Menthylcarbamate.4.4456.4958.5266.92112.981633.244.76" 95.48320.9161.8185.4170.2167.5169.2fro-91" 103.6362.9185.2188.5194.3190.8193.483.49" 172-0609.0391.3317.0328.3322.7326.51.92Q 1.9641-9861.9981.9892.0062.0052.OOWeightofSOIUk, grams.or,.0.58781-65"0.77074.570.94978.551.03486.671.014314.091.030217.380.98487-721.11459.070.839915.321.17439-601.421912.65Cye- 1.74"4.728.897.0314.90 18.36 8.15 9.60 162510.06 13.26ggr.1.96"5.3910.108.0117.75 21-79 9-83 11.45 19.3411.48 15.14TABLE11.(continum').m-ToEyll-i@enthylcarbnniate.avi*[a],.[a],,.[algr.[alsi.Lar!w3.39'56.1.1"59.20"66-70"115.35'162.8"m-4-Xylenyll-Menthylcarbamate.9.1653.91554763.57103.05lG3.3l-Menthyll-Menthylcarbama t e.16.9390.0293.60106.31178.26303.3l-Menthylo-lodobensode.14.1758.5961.7670.34124-49226.1l-Menthylo-Nitrobenzoate.43.17138.91146.90175.00435.52422.752.35154.01162.01170.67461.97466.822.99156.78165-52195.57466.90478.028.60162.76172.27208.08513.34496.547.05182.40193.47230.27560.18556.31-Menthylm-Nitrobenzoate.20.3681.7585.6797.76173.34243.326.4688.9792.83106-48186.092il.4CWye. 171.1"168.7315.4238.4448.0 494.2 504.8 525-5 590.1261.3 284.4192.6327.3358.4600.7271-6480.5533.81299.1520.51409.0596.41424.0628.81565.7702.31708.6298.2528.8324.8567.52.0542.0041.9802.1243.064 3012 2.979 3.154 3.0712.121 2.09Weightofsolute,O*W 0.9328 1.09360.82480.92030.946104463 0*9660 0.9887 0.95990.92790.4453ED.13.71" 15.80 16-8111.933-348-785.48 8-44 6.72 &637.464.04+2.14.59" 16.87 17.7612.603-409-395.78 6.82 7.12 7-017.734-26agr. 17.55" 20.27 21.3514.993-9311.336.85 8-11 8-46 8.338.914-80TABLE11.(continued).I-Menthyl5-Chloro-2-nitro6enzoate.avi.[a]".[a]ye.[albr.*[a]Ii.[MI,.46.70'151.53'16145'193.97'516.14'514.2'52.87169.38180.86217.30586.79575.055-23153.71162.40195-23505.03521'7I-Menthy 14-Chloro-2-nitro b enzoat e .36.80144.64152.76181.74447.38491.1I-Menthyl4-Chloro-3-nitro b enzoa t e.7-1572.57I-Menthyl30.75184.051-M en t hyl:16.43116.019.47133.319.96135.920.431w11-Menth yl16.1380.3975.1985.41155.39246.32 :6-Dinitrobensoate.196.83237.49644.00644.32 :+Dinitrobensoate.122.6145.034742405.9141.2167.9403.11466.7144.01704403.76475.7146-1173.6425.68483.53 :5-Dinitrobenzoate.83.3296.03173.83281.4m-Tolyl-I-men t h ylca r barnide.8.5883-0785-0097-03171.32233.1547.3" 613.9 551 *2518.8255.3689.2ma 494.2 504.1 511.1291.6245.8658.5"1752.0'737.71924.0862.51714.0617.01518.9289.9527.4831.62256.0501.41116-7587.61410.9598-214l3.2607.41489.8336.1808.4280.4495.13.406 3.346 3.9853.0932.1413.5012.997 3.024 2.970 3.0812-1622.12Weightofsolute,grams.0.22630.02070.9894 1.00451-31401.0590aD.1.61"1.512.32 5-659.327.06aye* 1-89"1-812.44 5.909407.37agr. 1.91"1-822.79 G-713-3884-68"67.89"76-72'136*16"24B5"220-0"248.5'441.2"1-Naphthy11-Ment hy lcarbamate.3.1148.8651.8659-64100.20157.1167'5189.4323.6l-Menthyl2-Methoxynaphthoat e.4.4246.89493256.3989-33159.4167-7191.2303.811-0251-1853.4360.7599.79174.0181.7206.5339-3PyridineSaltof1-MenthylHydrogenPhthalate.11.3420.9970.9274.5786.30159.74264.5278.2321.9595.8SodiumSaltofl-MenthylHydrogenPhthalate.8-4314.4260.74634272'52123'79192'02004229.2392.12.1062.0601.905 1.9512.2522'043MixedHydrogenPhthalutesof1-Mentholandd-Neomenthol(T.,1912,101,126). 2.224 2.178 2.1831.0154+3.18+3*34+3*82+747+%a48+29*91+34*22+63*33----1.0307+3.37+3.51+4*04f7.34+29*72+30.95+3543+W73--04429+3*76+4*0!2$4.62+8*19+39*87+42*63+48%+86%----0.9543+4.52+4*70+6*40+9.M-1-47.37C49.26+56.59+104*37---WeightofTABLE11.(contiiizied).1-Menth0n.e.solute, grams.a,.aye.agr.avi.[.ID.[alp.[a]gr.[a]vi*[MID.[M]pIM]gr*[Mlvi.1.00371.88"1.72"1.99'3.53"16.73"17.13"19.83"35.17"25.8"20.2"30.3"63.8"I-OQQO2-552.643-046BI23.1924.0227.6647.3135.536.842.372.41.09442.622.723-186.4223-9524.8629-0758.6836.638.144-589.8d-iwBorneoZ.1.0328+345+4.02+4.50+7*42+33*89+35*38$39.61+65*31+52.2+54*5+60*9+100*60.W7+2*02+2.09+2-S+3-70+20*33+20.92+23.33+37G+3>2+32.2+am9+57.1I-BorheoE.1.10204.464.745-409.2236.7939.1044.5476.0566.660.268.6117.11.0663 1.03918.25 8.488-60 8.869.88 10.181-isoBornylHydrogenPhthala-te.17-40 18-0577.47 81.6280.74 85.2792.76 97.97163.34 173.71234.0 248.5d-BorltylHydrogenPhthalate.243-9 257.5280.2 2969493.3 524.70.9563+4*08+4*30+4*96+8-76+42*72+45*02+51*93+91*70+129+0+135*9+156.8+277*00.8300+1*52+1-&3+1*7l+3.20+40.06+47*28+51.82+96*96+1391+142.8+166.5+292.810304f5.88$6.19$7.09+12*81+56-72+59.71$68.39+123.60+171*3+180*3+206*54-373.22.101 2.039 2.4501.927 1.8312.0692.109 2.1282.147 2.105 2.1760 KENYON AND PICKARD : INVESTIGATIOXS ON DEPENDENCEFurther observations of density.Di. and rotatory power (alM ,,,,,,.)of substances examined in the homogeneous state .I-Men t hol .Temp ....... 35"Temp ....... 37"aye ............ -46.12Temp ....... 36"agr ............ -52.08Temp ....... 35"a v i ............ - 86-54LID ............ -43.9055"43.6042"46.0443"52.0244'86.1074"42.9654'45.6853'51.8052"86.8479" 104"42.62 41-2271" . 100"44.94 43-0671" 100'50.90 48.6672" 100"84.38 81-14I-Menthyl Acetate .126"39.68126"41-50126"46-90126O78.10Temp ..........17"Density ...... 0.9287a, ............ -73-82 73.20Temp ....... 16.5. -a,. ............ -77.16 -Temp ....... 16.5" 25"Temp ....... 16.5. -Temp ....... 16.5. -ugr ............ -87.54 -avi ............ -1145.06 -61"0.891460"70.3865"73-1462"83.1667"137.3696" 132"0.8617 0.829480" 108" 147"83" 108" 147"83" 108" 147"81-22 78.92 75.4685" 110" 147'134-82 131-50 125.5068.86 66-68 63.7671.62 69.70 66-64Di-l-m enthyl Oxala t e .Temp ....... 19" 54"Density ...... 0.9849 0.9566Temp ....... 20" 53" 56" 90"a, ............ -91.86 89.20 89.00 86.12Temp ....... 20" 69" 80" 125"aye ............ - 96.92 92.60 90.76 86.34Temp ....... 20" 60" 80" 126Oag, ............-108.52 104.64 102.50 97.82Temp ....... 20" 61" 80" 126"uVi ............ - 179.00 172.62 169.02 161.6891"0.9265125.5" 140'82-74 81.06140" 164"84.60 82.46140" 164"96-94 93-46140" 164'159.00 164.70159" 197"37.72 35-60159" 196"39-44 37-90159" 196"44.52 42.20159"74-16167"62-14167"64-64167"73-54167"121-30138"0.8905164" 194"79-18 76-00194"79.16194'89-64194"149.16l-Menthyl Phenytacetate .Temp .......... 21" 63" 99" 149O 173Oan ............... - 68.22 64-48 60-82 56.42 64-82Temp .......... 21" 60" 99" 149" 176"aye ............... - 71.32 67.60 63.52 58-98 6642Temp .......... 21' 60" 99" 149" 176"agr ............... - 18.18 76.86 72-24 76.20 64.44avi ............... -137.60 129.90 122.26 113.10 107.9OF ROTATORY POWER ON CHEMICAL CONSTITUTION .61Temp ..........Temp ..........Temp ..........ag. ...............Temp ..........a" ...............a y e ...............Qvi ...............I-Men thy1 p-M ethoxyb enzoa t e .19". 89.4619". 93.5219". 106.4819". 181.1460.5"84.3656"88-6657"100.7055"170.70123"77-20122"80.70123"91-44122"153.60150"73.94151"77.20151"87.60151"147.40175"71.34176074.56176"84-38176"141.60I-Men t h y l d o d o b enzoa t e .A t 19 " ...... aD -84.36. aye 88.40. agr 101.16. ayi 179.82At 70.5" ... aD -81.18. aye 85.00. a,, 97.32. a, i 173.38I-Ment hyl m-lodo b enaoat e .At 19O.. .... OD -42.12. a y e 43.96. ag 50.00 a v i 83.80I-Men t h yl p-lodo b enzoat e .At 19" ...... a, -41.74.aye 43.62. agr 49.40. a v i 83.10I-Menthyl o.Nitrobenzoate, b . p . 185O/2 mm .Temp ....... 22.5" 25" 38" 56" 104" 126" 153" 184"Temp ....... 22.6" 28" 58" 84" 97" 121" 133" 153" 186"aye ............ - 151.40 149.90 142.00 135.30 131.92 127-10 12518 122-00 116-26Temp ....... 22.5" 28" 34" 60" 84" 100" 121' 153" 186"agr ............ -178.90 177.64 175.10 166.80 159.80 154-70 150.08 143.66 136.46Temp ....... 22.5" 28" 33" 56" 80"aD ............ - 142.78 142.44 138.80 134.52 123.96 120.22 115.56 110.16a, i ............ -423.0 420.2 417.0 395.0 375.6I-Menthyl m.Nitrobenzoata, b . p . 186O/2 mm .Temp ....... 19" 56" 65" 87" 102" 149" 182"Temp ....... 19" 62" 65" 80' 109' 149" 182"Temp .......19" 62" 66" 80" 109" 149" 182"asp ............ -109.94 105-00 104-41 102.66 99.08 94.36 91-50Q D ............ - 92.22 88.66 88.00 85.56 84.00 79.04 76-98aye ............ - 96.44 92.20 91.82 90.24 86.84 82.76 80.72a, j ............... - 193.94".I-Menthyl plli'itrobenzoate. b . p . 192O/3 mm .Temp ....... 22" 25" 39" 63" 75" 118.5" 169" 187"Temp ....... 22" 30" 39" 73" 85" 118" 162" 187"Temp ....... 22" 32" 39" 74" 86" 118" 163" 187"U D ............ -79.32 79.20 78.10 76.10 75.18 71.86 68.70 66.78aye ............ -82.70 82.22 81.50 78.74 77.76 74-92 71.68 69.70a,, ............ -93.20 92-44 91.90 88.72 87.44 84.56 80.86 78.66I-Ment hy l p-A llyloay b enzoat e .At 20" ...... au -81.92 aye 85.80. agr 96.68 . a. 165.262 INVESTIGATIONS ON DEPENDENCE OF ROTATORY POWER. ETC .1-Men t hylarnine .Temp ....... 17" 54" 94" 133"Density ... 0.8667 0.8307 0.7976 0.7857Temp ....... 19" 47" 98" 137' ............ -34.20 34.28 34.02 33.58a y e ............ - 35.62 35.70 35.50 34.90~ g r ............ -40.26 40.36 40.00 39-44a v i ............ -65.38 65.58 65.52 64.48l-Menthofae.*Temp ....................Density ................Temp ....... 21"Temp ....... 21"aye ............ -25.00Temp ....... 21"a, ............ -24.00agr ............ - 28.80Temp ....... 21"a Y i ............ -51.34.. 18" .. 0.896834" 60"24.10 24.4034" 56"25-14 25.5234" 56"28.98 ' 29.4234" 56"52.06 63-6454O0.869276"24.6277"25.8477"29-8877"54-6494.5" 129"0.8362 0.8073100" 129" 145"100" 128" 145"100" 128" 145"30.60 31-14 30.64100" 128" 145'25.18 25.62 25-2226.38 26-86 26.4657.02 58.68 57-46167"167"167"30.20167"55.9424.7625.86.iK enthone A.*A t 19 ....... a . +3.36. a y e +3*82 . ayr +5*04 . ayi +18.26Menthone B.*At 20 ....... aD +lo038 . a y e +11*06 . ~ g r +13.66 . Ori +36*00Menthone C.** See footnote to Fig . 4, page 43 .~-BoTT$ A ce tat e .A t 20 ....... -16.82. a y e -17.50. agr -19.86- a v i -32.54Temp ................ 17.5"Temp .......... 16.5" 40"Temp .......... 16.5" 41"aye ............... -43.98 41.02Temp .......... 16.5" 42."agr ............... -550.00 46.36Temp .......... 16.5" 42"uyi ............... -84.10 78.00Density ............ 0.9879a, ............... -42.26 39.4252"0.959773"35.7272'37.4072"42.4471"71-1498"0.916994"34.0092"35.6092"40.3692"67.70133.5"0.8840135"30.30133O31-90133"36-24133"60-78167"28-02167"29-20167"33.00167"55-42The authors desire to express their thanks to Mr . John Ramonfor able assistance. and desire to state that many of the compoundsused in this investigation were prepared from material the cost ofwhich was defrayed b y grants received from the Government GrantCommittee of the Royal Society .MUNICIPAL TECHNICAL SUHOOL.BLACKBURN

ISSN:0368-1645    

DOI:10.1039/CT9150700035

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

BRISCOE: THE ATOMIC WEIGHT OF TIN. 63VIII.-The Atomic Weight of Tin.By HENRY VINCENT AIRD BRISCOE.NUMEROUS determinations of the athomic weight of tin have beenmade, and, as a necessary preliminary to any account of the presentinvestigation, the methods and results of these researches will bebriefly indicated.The investigations of Gay Lussac (Ann. Chim. Phys., 1811, 80,160), Berzelius (Ann. Phys. Chem., 1826, [ii], 8, 177), Mulder andVlaanderen (J. pr. Chem., 1849, 49, 35), and Dumas (Ann. Chim.Phys., 1859, Liii], 55, 154) were made by the oxidation of tin withnitric acid. Dumas also made two determinations of the ratioSnC14 : 4Ag by titration (the method of the present investigation),but regarded them merely as of confirmatory value. Vlaanderen',later (Jahresb., 1858, 183), found certaiii errors in the method ofoxidising tin with nitric acid, and, therefore, determined the ratioSn : SnO, by reducing stannic oxide in hydrogen.Van der Plaats(Cornpt. rend., 1885, 100, 52) employed both these methods forthe measurement of the oxide ratio.Two comparatively recent determinations of the ratio Sn : SnO,by oxidation of tin have been made incidentally in the course ofother work by Schmidt (Ber., 1894, 27, 2743) and by Meyer andKerstein (Ber., 1913, 46, 2882).The research of Bongartz and Classen (Ber., 1888, 21, 2900) isby far the most considerable attempt that has hitherto been madeto fix the atomic weight of tin. Their work involved the measureinent of five ratios, namely, Sn : SnO,; SnBr, : Sn; K2SnCl, : Sn;(NII,),SnCl, : Sn; and Sn : 2BaS04.Their starting material was Banca tin, from which the tetra-chloride was prepared by the action of dry chlorine.The productwas fractionated repeatedly, and the fraction boiling constantly a tl Z O o * collected f o r use.In the determination of the, oxide and barium sulphate ratios themetallic tin required for conversion into oxide or sulphide wasobtained as follows: The purified stannic chloride was dissolved inwater, and treated, with constant stirring, with a concentratedsolution of crystallised sodium sulphide until the precipitate redis-solved, and a solution of sadium hydroxide equal in amount toabout half the sodium sulphide used was then added. After somedays the solution had become quite clear; and it was then electro-* The boiling point here given is 6" higher than that determined by Thorpe(T., 1880, 37, 331)64 BRISCOE: THE ATOMIC WEIGHT OF TIN.lysed in a weighed platinum dish o r crucible.The deposit ofmetallic tin waa washed with water and alcohol, dried at looo,cooled over phosphoric oxide, and weighed.In the determination of the ratio Sn : 2BaSOl a very compli-cated analytical procedure was used, of which only an outline isgiven. The results are stated merely as percentages of sulphur,but, assuming that these are based on the atomic weights of Meyerand Seubert, it is possible to calculate the weights of bariumsulphate actually obtained.*The potassium and ammonium stannichlorides were also preparedfrom the same stannic chloride, and the percentage of tin theycontained was determined by electrolysis of their solutions inaqueous ammonium oxalate a t 30-40° and a t the ordinary tem-perature respectively.The ratio of tin to stannic bromide was determined by electro-lysis of a solution of stannic bromide containing ammoniumoxalate.TABLE I.No.Result:Ratio of Sn= AverageNo. Date. Investigator. measured. Detns. (0 = 16) error.123456789101811 G~Y-LUSS~C 1826 ~ ~ ~ ~ ~ l i ~ ~ } Sn : SnO, (oxidation)1849 Mulder &Vlaanderen Sn:SnO, ,,1859 Dumas ............... Sn:SnO, .. 1885 van der Plaats ... Sn : SnO, .. 1894 Schmidt ............... Sn : SnO,1885 van der Plrtats ...... Sn : SnO,.. 1913 Meyer & Kerstein Sn : SnO, ,,1868 Vlaanderen ......Sn : SnO, (reduction)1869 Dumas ............... SnC1, : 4Ag ..117.7 -116.3 0.32118.06 0.15118.08 0-04117-54 0-12118.16 0.08118.07 0-05117.97 0.04118.49 -11 1888 Bongmtz & Classen Sn : SnO, (oxidation) 11 119.06 0-14412 1888 Bongartz & Classen Sn : 2BaS0, 8 119.08 0.07513 1888 Bongartz & Classen Sn : (NH4),SnCl, 16 119.09 0.07614 1888 Bongartz & Classen Sn : &SnCl, 10 119.07 0.03516 1888 Bongartz & Classen Sn : SnBr, 10 118.98 0-040The raults of the determinations quoted above are collected intable I. Inspection will show that they may be divided into twoclasses; on the one hand, the determinations of Bongartz andClassen, giving a mean result of Sn=119.06, and on the other therest of the determinations, which all give values for the atomicweight less than 118.5.Of the latter results the analyses of GayLussac, Berzelius, and Mulder and Vlaanderen appear to be low ascompared with those of Dumas, Vlaanderen, and van der Plaats,* The antecedent dataused throughout this paper are : Ag= 107*880, C1=35 '457and Cl=35'460, Br=79 920, N=14*010, S=32-067, Ba =137*363, H=1'00779.Where the higher value for chlorine, Cl=35*460, is used, B statement to that el€ectis made ; in all other cases the use of the lower value is to be understoodBRTSCOE: THE ATOMIC WEIGHT OF TIN. 65which approximate tQ 118. If anything a t all can be inferred fromthe figures i t is that the atomic weight of tin is near 118.0.The five series of fairly concordant measurements by Bongartzand Classen, however, indicate that the atomic weight of tin isgreater than 119.0,Itl is extremely difficult to decide between these two values, butthe International Committee on Atomic Weights and Clarke in his" Recalculation of the At,omic Weights " reject all determinationsother than those of Boi1gartz and Classen, on whose value theybase the accepted figure Sn=119*0.A further discussion of these data will be made in connexionwith the statement of the results of this investigation, but enoughhas been said t o show the urgent need for new determinations ofthe atomic weight of tin.Outline of the Present Investigation.Tin forms few definite compounds that can be purified by distilla-tion or by crystallisation and fusion, and the chief of these-thechloride and bromide-present peculiar difficulties in manipula-tion on account of their sensitiveness to moisture.It is probablyfor this reason that these substances have been so little used inresearches on the atomic weight of tin, and that preference hasbeen given to the determination of the ratio SnO, : Sn. The varia-tion in the results obtained is a measure of the uncertainty of thelatter method.Of all the compounds of tin, stannic chloride would appear t obe the most suitable for use in the determination of the atomicweight of the metal, provided that proper steps are taken to protectit from atmospheric moisture. It is very stable, it can be preparedeasily, and it can be completely purified without great difficultyprovided water be excluded rigidly.Moreover, the proportion oftin to chlorine in this compound is such that a given error in thedetermination of the percentage of the latter element produces amuch smaller error in the atomic weight of tin than would an equalerror in the determination of the ratio Sn: SnO, (see table IV).The present re-determination of the atomic weight of tin hastherefore been effected by measurement of the ratio SnCl, : 4Ag.Stannic chloride was prepared by direct union of its elements,and purified by fractional distillation. A portion of the finalfraction of this purified material was further fractionated twice,and collected in weighed bulbs in a sealed and exhausted appara-tus. The solutions for analysis were prepared by breaking the bulbsin dilute nitric acid containing either oxalic acid or tartaric acid,and were then treated with a slight deficit of silver nitrate preparedVOL, CTJII.66 BRISCOE: THE ATOMIC WEIGHT OF TIN.by dissolving a weighed amount of silver in nitric acid. Ths slighLdeficit of silver was made up by additJon of a measured volumeof a dilute standard silver solution, and the titration was corn-pleted by the further addition of dilute standard silver or sodiumchloride solution. The progress of the titration and the attainmentof the end-point were ascertained by nephelometric measurements,which were carried out by the same method and in the sameapparatus as was described in a former communication (Briscoeand Little, T., 1914, 105, 1321).I n this way fifteen distinctfractions of stannic chloride were analysed.The chief difficulties arose from (1) the sensitiveness to moistureof the stannic chloride, and (2) the rapid hydrolysis of that corn-pound in dilute aqueous solution. Of these the first' was completelyand sativf actorily overcome by the special methods of distillationand weighing employed in previous work on the atomic weight ofvanadium (Briscoe and Little, Zoc. cit.), and the second was metby the addition of oxalic acid, or, in one case, tartaric acid(Rochelle salt) to the solutions.The conditions of the titrations, the amounts of the variousreagents used, and the details of manipulation were varied fromone experiment to another, with the object of eliminating possibleconstant errors due to such sources, and all the usual precautionswere carefully observed.Preparation and Purification of Reagents.The Purification of all reagents received very special attention,and those used in the analyses were frequently tested throughoutthe course of the work.I n the following paragraphs the methodsby which each reagent was purified or obtained are described indetail.TVater.-In the course of the determinations it has been foundthat water of an extremely high degree of purity as regardsfreedom from chlorine can be obtained in one distillation fromtap-water if a still of the domed-condenser type is used. Duringthe vacation the laboratory supply of distilled water, which isobtained from such a still, was free from chlorine, and was, accord-ingly, used in the purification of several of the reagents.Duringthe time students were in the laboratory in which the still issituated the water was found to be slightly contaminated withchlorine, derived, presumably, from the air of the laboratory, andat such times all the water used was further purified, In all casesexcept one, specially purified water was used in the preparation ofthe solutions for analysis.For the further purification of the water it was distilled froBRISCOE : THE ATOMIC WEIGHT OF TIN. 67a large copper vessel with the addition of a trace of sodiumcarbonate, and the steam was led through a wide copper tube inorder to diminish spraying, and condensed in a long tube of puretin.The water was kept during the brief period before use in well-st’oppered bottles of Jena glass.Hydrochloric A cid.-This reagent was required free from arsenicfor the preparation of pure hydrogen. It was therefore treatedwith a copper-tin couple, and distilled from copper gauze accord-ing to Thorne and Jeffers’ method (Analyst, 1906, 31, 101).Zinc.-Granulated electrolytic zinc, supplied through the kind-ness of Messrs. Brunner, Mond and Co., Ltd., prmed to be freefrom arsenic, and was therefore used without further purifica-tion.Hydrogen.-This gas was prepared by the action of dilute hydro-chloric acid on zinc in a Kipp’s apparatus. A small quantity ofplatinic chloride was added to the acid to facilitate the action.The hydrogen generated was bubbled through a long column ofpotassium hydroxide solution, and then passed over solid potassiumhydroxide and finally over phosphoric oxide.The gas issuing fromthe purifier was tested by passing it through a Jena-glass tubedrawn out to a capillary, and heated to redness in the wider portionimmediately adjacent thereto. The longest run-of five hours a tthe rate of 3 litres per hour-gave no visible deposit in the capil-lary, and the gas must, therefore, have been free from arsenic.Magnesia.-Pure magnesia was prepared by repeated precipita-tion of magnesium carbonate from magnesium nitrate solution bymeans of pure sodium carbonate; the precipitate was washed thor-oughly, dried, and ignited in an electrically-heated muffle furnace(compare Briscoe and Little, Zoc.cit.).I n making new magnesia boats for these determinations somelittle difficulty was experienced, as some of the boats cracked andwere very friable after ignition. Numerous trials ultimatelyshowed that good resultx could be consistently obtained if the boatswere moulded with the walls not less than 3 mm. thick. Thestrongest boats were those made with the powdered fragments ofold boats mixed with about one-sixth of their bulk of freshmagnesia. The last traces of sulphur were removed from the boatsby igniting them for several hours in a stream of hydrogen contain-ing a little water vapour.During all fusions of silver the hydrogen issuing from thefurnace was examined, and was found in every case to be quiteodourless. This fact affords good evidence of the purity of both thehydrogen and the magnesia as the probable dangerous impurities,namely, sulphur and arsenic, would be indicated by the presence inF 68 BRISCOE: THE ATOMIC WEIGHT OF TIN.the gas of arsine and hydrogen sulphide, both of which can bedetechd by smell when preaent in extremely minute quantities.Nitric A &.-The purest nitric acid of commerce was redistilledtwice from a glass flask with ground connexions, and in eachdistillation the first half of the distillate and a small end-fractionwere rejected.Oxalic Acid.-Two samples of oxalic acid were prepared byrecrystallisation of the purest commercial substance according tothe following plan.A boiling saturated solution of the acid inpure water was filtered through a hot funnel, reheated in order todissolve the crystals formed, and then cooled as rapidly as possiblewith violent agitation. The mother liquor was carefully decantedfrom the crystals, and the latter were then divided into twoapproximately equal parts and wrapped in two pieces of cleanlinen.These bundles of crystals were then placed on oppositesides of the copper basket of a centrifuge, and whirled a t 2000revolutions per minute, which, with a basket 28 cm. in diameter,corresponds with a peripheral speed of 1770 metres per minute.The crystals were then again dissolved, recryshllised, and drained,and the process was repeated three times in the case of one sample(oxalic acid I) and six times in another case (oxalic acid 11). Allthe operations were carried out in porcelain vessels.The linen cloths awed were cleansed originally by boiling withdilute ammonia for an hour and thsn washing with water in aSoxhlet extraction apparatas for forty-eight hours.Betweenrecrystallisations the cloths were similarly washed for a period ofa t least two hours (twenty-four t o twenty-eight washings).It was found impossible to detect by means of the nephelometerany trace of chloride in either sample of the oxalic acid.*Rochelle Salt.-Commercial recrystallised Rochelle salt wasrecrystallised four times by a method precisely similar to thatdescribed above for oxalic acid. Nephelometric tests showed theproduct to be free from chloride.ChloTine.-The chlorine used for the preparation of stannicchloride was obtained from the middle fraction remaining aft'er15.75 kilos.had been evaporated from 43.2 kilos. originally con-tained in a steel cylinder of the liquefied gas.Ammonia Solution.-This was prepared by dissolving in purewater ammonia expelled from the purest commercial ammonia* Centrifugal extraction of the mother-liquor under the conditions described wasremarkably effective. The crystals appeared loose and dry when taken from thocentrifuge and could have contained only a very small amount of water as the finalcrop some 1500 gramg, dried completely in six hours' exposure to air in a covereddishBRISCOE. THE ATOMIC WEIGHT OF TIN. 69(D O*SSO) by heating it to about 50°.The operation was performedin vessels of Jena glass.Sulphur Diozide.-Gaeeous sulphur dioxide was obtained fromthe middle fraction of a syphon of the liquid. I t s further purifica-tion was unnecessary as the presence of a trace of chlorine in thismaterial would not adversely affect the purity of the silver preparedby its aid.Formic A cid.-Kahlbaum’s purest formic acid was redistilled inplatinum and collected in a platinum bottle.Ammoibium formate was prepared in solution by neutralisingpure formic acid with ammonia solution prepared by the solutionin pure water of ammonia gas expelled by heat from ammoniasolution prepared as above described, the whole operation beingconducted in platinum vessels.Silver.-Five samples of pure silver were prepared, four byreduction of an ammoniacal solution of silver with ammoniacalcuprous sulphite solution, according to Stas’ method, and one byreduction of pure silver nitrate with ammonium formate asdescribed by Richards and Wells (J.Amer. Chem. SOC., 1905, 27,459). The main details of these preparations are as follows:SumpZe Z.-A solution of 200 grams of recrystallised silvernitrate was mixed with a solution of 15 grams of electrolytic copperin nitric acid, and treated with ammonia solution until the precipi-tate first formed had redissolved. This solution was diluted t oabout 10 1iFres.Meanwhile a solution of ammonium sulphite had been preparedby neutralising with sulphur dioxide a solution of approximately100 grams of ammonia in 2 litres of water.This solution, whilestill warm (approx. 40°), was added t o the prepared silver solution,and the whole was well shaken and kept in a stoppered bottle.After twenty-four hours all the silver had bem deposited as acrystalline precipitate, and the solution was nearly colourless.The supernatant liquid was poured off, and the silver was washedsix times with water, allowed to remain for twelve hours withdilute ammonia solution, then washed again six times with water,and finally transferred to a large platinum dish, in which it waswashed three times with water and, finally, drained as completelyas possible and dried in the dish by heating a t 250° in an dec-trically-heated air-bath.Sample 12.-Residues of silver chloride obtained in the deter-mination of a previous atomic weight were dissolved in diluteammonia, and the solution filtered and treated with ammoniumsulphite prepared as for sample I, the temperature being raised t 70 BRISCOE: THE ATOMIC WEIGHT OF TIN.70°.The silver produced was thoroughly washed and united witha quantity of silver, which had been prepared previously, and thewhole dissolved in nitric acid. The filtered solution was thenreduced as before, except that the reduction only became completeafter long heating to 70°.It may here be remarked that in this process of reduction therate of deposition of silver, the density of the product, etc., varyconsiderably with the concentration of free ammonia and copperin the solution. There can be no doubt that the product is inevery case silver of the same high order of purity, but an investi-gation of the reaction would probably be of interest.SampZe IIL' of silver was prepared by the same method assamples I and I1 from silver nit'rat'e which had been thricerecrystallised from water with centrifugal drainage.The excessconcentratiop of ammonia was kept low, and a finely crystallineprecipitate of silver quickly formed. The silver was washed anddried as before.San2pZe ZV.-A portion of sample I was dissolved in nitric acidand reprecipitated by the method described above. In this casethe excess concentration of ammonia was considerable, and theprecipitation of silver took about three times as long as in thepreparation of sample 111.SampZe V.-Silver nitrate which had been twice recrystallisedfrom water was reduced with ammonium formate. The product waswashed as in the case of the other samples, and dried a t 300'.I n every case the silver was prepared f o r weighing by fusionin a magnesia boat in a stream of hydrogen.The fused metal wascooled in hydrogen, etched for five to ten minutes in diluted nitricacid, washed thoroughly with water, dried a t 300' for twelve hours,and cooled in a desiccator over potassium hydroxide. The weightof silver was adjusted by adding pieces of silver wire drawn frombuttons of pure silver with the precautions previously described(Briscoe and Little, Zoc. c i t e ) .Stannic Chloride.-According to' the analyses of Mulder (J. pr.Chem., 1849, 49, 35) and van der Plaats (Compt. rend., 1885,100, 52) the common impurities of Banca tin are iron, copper,lead, and silica.I n the preparation of stannic chloride by theaction of chlorine on tin it is found that the1 silica is left un-attacked; whilst it seems probable that in the presence of anexcess of tin any copper chloride formed would be again decom-posed as copper is notably electronegative to tin.Supposing iron, copper, and lead chlorides to be formed theyshould be easily separated from the main product by reason ofthe differences between their boiling points and that of stsnniBHISCOE: THE ATOMIC WEIGHT OF TIN. 7 1chloride (SnCl,, 1 1 4 O ; FeCl,, 448O; FeC12, about 1100O; CuCl,about 1000° ; CuCl,, decomposes; PbCl,, 860O). Further, ifstannous chloride were formed it also should be completely separatedby fractional distillation, as it boils a t 606O.The following methodof preparing and purifying stannic chloride was thereforeemployed.About 1 kilo. of good commercial granulated tin was treated, inquantities of about 250 grams a t a time, with chlorine gas. Thereaction was carried out in a round Jena-glass flask provided witha long, air-cooled condenser, and the current of chlorine was soregulated as to keep the stannic chloride gently boiling. Noexternal source of heat is required.The product (about 1800 grams) was kept in contact withmetallic tin for forty-eight hours, and then distilled three timesin an apparatus constructed entirely of glass, rejecting in eachdistillation the first and last quarters of the distillate.The finalmain fraction of about 400 grams boiled constantly a tAfter being kept for some time all the fractions of stannicchloride, but especially the earlier fractions, deposited small quan-tities of a colo-urless, crystalline substance, which was probably thehydrate SnC1,,3H20.During the earlier stages of the fractionation, which involvedsome contact of the stannic chloride with air, it was found to beimpossible to1 eliminate this impurity completely. Attempts toremove this water by distillation with metallic sodium failed, asit was found that even in an atmosphere of carbon dioxide thematerial exploded as soon as the boiling point was reached.113.9-1 14' lo.The Deter miization of FV eigh t .All the weighings involved in the, determinations now underconsideration were made on a short-beam Bunge balance, whichwas set aside especially for this work, and was placed in a specialroom entirely contained within another .room, their walls and roofsbeing separated by an air space several feet wide.This room issituated in the interior of a section of the building which hasspecial foundations designed to insulate it from all vibration. Thetemperature of this balance-room did not vary half a degree duringnine months, and to this fact and to the absence of vibrationmust be attributed the remarkably constant behaviour of thebalance.The sensitiveness of the balance was so adjusted that the zeroshifted about nine scale divisions for a difference in weight of1 milligram, and the excursions of the pointer were read with th72 BRlSCOE: THE ATOMIC WEIGHT OF T1N.aid of a lens.In weighing, the rider wag adjusted to within two-or threetenths of a milligram of the apparent weight, and theremaining tenths and hundredths of a milligram to be added orsubtracted in order to get the true apparent weight were deter-mined by th0 method of oscillations.A set of gilt brass weights with platinum fractions by Becker,and a gold fivsmilligram rider were used; they were calibratedcarefully both before and during the course of the work in ordert o determine their relative weights in air. The relative weights ina vacuum were not calculated, for the reasons given in a formerpaper (Briscoe and Little, Zoc.cit.), and the vacuum correctionapplied to the apparent weight of the silver (D 10.49) was+ 0.000117 gram per gram.A11 weighings were made by reversal, and two separate approxi-mate weighings were also made, one on an ordinary balance andone on the special balance. A double check was thus providedagainst any error in the recording of the weight.I n many cases the precise weighings were made in duplicate,but in no instance did the separate values for the vacuum weightdiffer by more than 0*00002 gram. All weighings of bulbs weremade against a tare consisting of a similar bulb and stem in whichthe capillary between the bulb and stem was sealed. The air inthe balance case was dried by solid potassium hydroxide, and thebulbs were kept in this dried atmosphere for several hours previousto the determination of their weight.To ensure the dissipation ofany electrical charge induced on the. surface of the glass when itwas cleaned, a piece of pitch-blende was kept in the balance case.I n all cases the weight of stannic chloride was corrected for theweight of air displaced froin the bulb, the magnitude of thiscorrection being determined by the method described in a formercommunication (Briscoe and Little, Zoc. cit.). As the pans of thebalance were of glass, the buttons of silver were weighed directlyon t,hem.The weights of stannic chloride are stated to the nearest twen-tieth of a milligram; those of silver to the nearest tenth of amilligram. It will readily be understood that the actual weighingsof silver are more precisol than is stated, but that the accuracyof the determination of the total weight of silver required dependson the precision with which the end-point of the titration can befound.With the concentrations used it was possible, if the solutionswere kept a t about loo, to ascertain the end-point within one-tenthof a milligram of silver, and the weights of silver ar0, therefore,stated to that degree of accuracy.It is evident that the errors in the weights of stannic chloridBRISCOE: THE ATOMIC WEIGHT OF TIN. 73and silver are less than 1 part in 150,000, and, therefore, insigni-ficant in comparison with the other errors of the determinations.Purification of Stannic Chloride and its Collection in Bulbs.I n the final purification of stannic chloride about 150 C.C. (330grams) of the end-fraction previously obtained (p.71) wasplaced in a flask of 250 C.C. capacity, together with about 2 gramsof metallic tin. This flask had a long neck, which was closed byfusion a t the upper end after the introduction of the chloride, andhad two side-tubes, one of which carried a special joint (Briscoeand Little, Zoc. cit.), whilst the other was open, but constrictedFIG. 1. --4--Ga t one point. By means of a Gaede pump, the flask was ex-hausted through this second tube, which was then sealed off atthe constriction. The flask thus sealed was entirely immersed in aste'am-bath and heated to looo for four hours.After the heating a whih sublimate of tihe crystalline hydratepreviously referred to was observed in the upper part of the flask,and, although it was present in a very small amount, it seemeddesirable to remove it.This sublimate appeared to be firmlyatbche,d to the walls of the flask, and the greater part of it wasseparated from the liquid by decanting the latter into a secondflask. The apparatus used is shown in Pig. 1.By heating the first flask A the volatile hydrate was driven intothe upper part of the neck, and the special joint F of A was the74 BRISCOE: THE ATOMIC WEIGHT OF TIN.sealeld on a t C to the side-tube of the flask B, which in turn wasprovided with a special joint D. After the apparatus had beencleaned, the1 long neck of B was connected through the bulb E withthe pump, and then a piece of glass rod was inserted in the side-tube of the joint F ; this side-tube and the narrow tube of D weresealed, and the whole flask B was completely exhausted and sealedoff from the pump at G.By tilting the apparatus the jointa t F was broken, and the liquid in A was then poured into B.Finally, B was sealed off from A a t the capillary H .Some fragments of the crystalline hydrate had been carried intothe second flask, an'd the greater part of this slight impurity wasnext removed by heating the whole flask to looo, except the bulbE, which was kept cool. The sublimate was thus driven into thebulb, which was then sealed of€ a t the capillary K.The contents of the flask B were then distilled in a vacuum ina sealed apparatus into five approximately equal fractions, Y, A,B, C, and Z, of which the middle three, A, B, and C, were collectedin tubes provided with special joints, whilst fraction Y was col-lected in a fourth vessel and fraction Z was left in the flask.Each of the fractions A, B, and C was further distilled, A andC into ten fractions each, of which the middle eight were in eachcase collected in weighed bulbs, and B into four small fractionsand one large end-fraction, of which the second and third smallfractions (B 1 and B 2) only were collected in weighed bulbs.Thatonly two small fractions of main fraction B were collected inweighed bulbs was due to the accidental cracking of the apparatusafter the third small fraction (B 2) had been sealed off.These final fractionations of starinic chloride were carried out ina manner quite similar to that of the fractionations of vanadyltrichloride described in a former communication (Briscoe andLittle, Zoc.cit.), and need not be further discussed.The general course of the fractionation in a vacuum is indicatedin thp, following diagram:/ay Rejected,+A--+y. 1. 2. 3. 4. 5. 6 . 7. 8. z.SnC14 --.-+B-- + 3. 1. 2. 3. 2. (330 grams) --1-+c -- +y. 1 . 2 . 3 . 4 . 5 . 6 . 7 . 8 . z.*Z RejectedIt will be wen that nineteen fractions were ultimately obtainedThe fractions y were collected in unweighedOf these fractions fifteen were analysed, namely, A 4, A 5, A 6,in weighed bulbs.bulbs, and the fractions z were left in the main vessels.A 7, A8, B 1, B 2, and all the eight fractions of main fraction CBRISCOE: THE ATOMIC WEIGHT OF TIN.75Fraction B 3 was collected after the cracking of the apparatus,and contained a visible amount of the crystalline hydrate ; fractionsA y and A 1 also contained minute but visible traces of thissubstance, and, consequently, neither A1 nor A 2 was analysed.The bulb containing fraction A 3 was accidentally broken.No attempt was made to test the stannic chloride for possibleimpurities, as these could only be present in amounts so minuteas to be entirely beyond the range of sensitiveness of the testsavailable. The best possible evidence of the purity or otherwiseof the stannic chloride is found in a comparison of the results ofanalyses of fractions so widely separated as A4 and C8.The Solubility of Metastannic Acid in Aqueous Oxalic Acidand Tartaric Acid.The most serious problem arising in connexion with the directdetermination of chlorine in stannic chloride was the rapid hydro-lysis of this salt in dilute solution.A solution of the concentrationemployed in the analyses, that is, containing 1-5-2-0 grams ofstannic chloride per litre, deposits considerable amounts of meta-stannic acid within three days. It is hardly probable that theprecipitate would occlude much chloride, but its presence wouldrender i t almost impossible to make nephelometric tests owing t othe time (over ten days) taken for it to settle.Investigations have been made on the solubility of metastannicacid in solutions of oxalates, from which it has been concluded thatsalts of a complex oxalostannic acid are formed.It has also beenobserved that mebastannic acid is soluble in solutions of oxalic acid(Rosenheim and Platsch, Zeitsch. anorg. Chem., 1899, 20, 281),although the nature of the compound formed is not definitelyknown. It has been stated that it is possible to obtain the oxalo-stannic acid in the crystalline form (Phchard, Compt. rend., 1893,116, 1513), but this lacks confirmation. There is evidence thatsimilar action takes place with other organic acids, especially withhydroxy-acids (see, for example, Rosenheim and Schnabel, Ber.,1905, 38, 2777).It seemed probable, therefore, that by the use of oxalic acid orcertain hydroxy-acids it would be possible to keep in solution themetastannic acid produced by the hydrolysis of stannic chloride.From the point of view of the determinations it was necessary thatthe substance selected should fulfil certain conditions :(1) It must prevent precipitation for a period of a t least afortnight under the conditions obtaining in the analyses ;(2) It must be obtainable perfectly free from chlorine; an76 RlilSCOE: THE ATOMIC WEIGHT OF TIN.(3) It must not hinder the complete precipitation of the chlorineas silver chloride.It seemed likely that oxalic acid and perhaps tartaric acid wouldfulfil condition (2) satisfactorily.The prevention of the precipita-tion of chlorine by these reagents could hardly take place unlessthrough the formation of complex ions containing chlorine, ofwhich the investigations referred to above afford no evidencewhatever.It remained, therefore, to ascertain the extent to which oxalicand tartaric acids fulfilled the first requirement. Six solutionswere prepared, each containing 1-5 grams per litre of stannicchloride, to which were added various amounts of nitric acid andof oxalic acid or Rochelle salt as stated below.I n experiments IV,V, and VI, the concentration of tartaric acid equivalent to theRochelle salt taken is given in the fifth column, and allowance ismade for the nitric acid used in liberating this acid from the salt.The last column shows the time elapsing before a visible opalescencedue to precipitated metastannic acid was observed in the liquid.No.ofTest.I.11.111.IV.V.VI .RochelleNitric acid. Oxalic acid. salt.Grams per Grams per Grams perlitre. litre. litre.1.0 2.0 -1.0 10.0 -10.0 10.0 -0.3 - 4.05.7 - 1420 11.1 -Tartaricacid.Grams perlitre. Time.- > 10 weeks.> 10 wesks.- > 10 weeks.7.4 30 days.-2-1 10 days.10.6 45 days.No precipitation was ever observed in those solutions t o whichoxalic acid had been added.The results for Rochelle salt may fairly be said to suggest thatthe lapse of time before precipitation varies directly as the concen-tration of tartaric acid, and does not depend on the concentrationof the nitric acid within the range covered by these experiments.It was found that by recrystallisation both oxalic acid andRochelle salt Could be freed compleztdy from chlorides, but, as theexperiments above described show clearly that the former substancewas to be preferred on account both of the smaller quantityrequired and the longer time during which precipitation is pre-vented, Rochelle salt was used for one of the analyses and oxalicacid for the remaining fourteen.illethod of Analysis.The procedure followed in the analyses of the fractions of stannicchloride was essentially thel same as that employed by Richards anRRISCOE: THE ATOM10 WEIGHT OF TIN.77his collaborators in the determination of the ratio of silver t o achloride.A large bottqle of Jena glass provided with a well-fitting stopperwas very carefully cleaned. About 2-3 litres of pure water werethen introduced, together with 10-30 C.C.of pure nitric acid and10-30 grams of oxalic acid. I n the analysis of fraction C4100 grams of Rochelle salt and 110 C.C. of nitric acid were used.When the solution had become quite clear, the bulb containingthe required fraction of stannic chloride was lifted from its boxby means of a long, hooked glass rod, and washed extarnally withpure water and lowered into the bottle. The bottle was thenstoppered and shaken in order to break the bulb.It; was found that if the breaking of the bulb were accomplishedby shaking the bottle sideways the products of decomposition wereapt t o escape entire absorption in the solution, as was evidencedby the formation of a fog in the upper part of the bottle.Whilsti t is unlikely that any loss of chlorine would thus be caused if thefog were allowed to clear before the bottle was opened, it seemeddesirable to eliminate the risk entirely. Accordingly, the bulb wasbroken by holding the bottle vertical, and giving it a sharp jerkin a vertical direction; this action caused the bulb t o strike withsome force against the bottom of the bottle while remaining com-pletely submerged in the solution.*Aftm the bottle had stood for a time, the fragments, and especi-ally the capillary ends, of the bulb were broken with a flat-endedglass rod, which was carefully rinsed in the bottle before it wasremoved. Finally, the chloride solution was diluted to a bulk of3-5 litres.A quantity of pure silver one or two milligrams less than thatactually required for the stannic chloride taken wi~s next weighedout and dissolved in an excess of dilute nitric acid, its solution beingassisted by warming the flask t o about 35O.The solution was thenheated to about SOo in order to expel the greater part of the nitrousacid formed, cooled, and diluted t o 800-1100 c.c., and finallyadded to the chloride solution prepared as described above.I n effecting the transference of the silver solution the bottlecontaining the stannic chloride was shaken steadily with a rotarymotion until the solution was itself rotating rapidly. The bottlewas then set down, and the silver solution poured steadily into it.The flask which had contained the silver solution was rinsed five toeight times with quantities of 100 C.C.of water, and the rinsings* This apparently trivial point has been given prominence because it appearedpossible that the “accidental” errors in precise work might be attributable to a lackof refincment in the mechanical operations : this question is again referred to in thediscussion of the results78 BRIECOE: THE ATOMIC WEIGHT OF TIN.were added to the solution in the bottle. Finally, the small amountof dilute silver solution supposed to be required t o bring the totalweight of silver to the calculated amount was added from a burette,and the bottle was stloppered, shaken gently, and set aside. Aftertwo days the bottle was shaken vigorously on several occasions, andabout the sixth or seventh day a nephelometric test was made.Subsequently small quantities of silver or chloride solution wereadded as required until the end-point was reached.The type of nephelometer employed and the methods of using ithaving been dealt with in aformer communication (Briscoeand Little, loc.c i t . ) need not bedescribed here.All the analyses were carriedout in a small laboratory whichwm devoted entirely to this work,and was kept permanently dark.Careful precautions were taken tokeep the atmosphere of the roomfree from dust and fumes ofhydrochloric acid.The mixing of the silver andstannic chloride solutions waseffected under yellow light, andalthough some of the bottles wereexposed to the light of anordinary carbon filament lampduring the later stages of the titrations no discoloration of thesilver chloride was ever observed.FIG.5.,4 New Form of Solution Flask for Silver.The transference of the silver solution from the flask in whichit is prepared to the vessel containing the chloride1 solution isevidently the chief point of the whole series of operations whereaccidental loss may take place. I n the earliex experiments the flaskemployed was of the ordinary round type, but was fitted with astopper carrying a set of bulbs similar t'o those shown in Fig. 2 ( A ) ,and the pouring of the liquid was effected in the ordinary waywith the aid of a glass rod. This procedure is slow, and imposesconsiderable strain on the operator.A new form of solution flask was therefore devised and con-structed of Jena glass; its shape will readily be understood onreference to Fig.2. During the dissolution of the silver the cap Cclosed the side-tube B, and the bulbs prevented any loss of silveRRTSCOE: THE ATOMIC WEIGHT OF TIN. 79by spraying of thei solution. When all the silver had dissolved thetube A carrying the bulb was rinsed into the flask and removed,the cap C was taken off, and the diluted silver nitrate solution waspoured out through the side-tube. Then, while the side-tube waskept within the receiving vessel, it was possible to tilt the flask alittle back so that a stream of water from a wash-bottle directedinto the neck would wash the silver nitrate solution remaining inthe upper part of the flask and in the side-tube B either back intothe bottom of the flask or out through the side-tube.Then t>heoutside of the tube B could be washed, and the flask removedwithout. risk of loss of silver. The washings of the flask wereperformed by the same process.Oxalic acid.r WeigCtFraction used.of SnC1,. Sample. Grams.A.4 I1 15A.5 I1 15A.6 I1 10A.7 I1 30A.8 I1 20TABLE 11.Nitric acid.&Volumeused.Sample. C.C.I1 20I 15I1 20I1 30I1 13VolumeofSolution solution.flask.* C.C.B 4600B 5900B 6700a 7200P 3500Durationofanalysis.Days.1415201618B. 1 I 15 I 10 a 5900 47B. 2 I 10 I 10 a 3900 33c. 1c.2 c.3c.4 c. 5C.6 c. 7C. 8I 12 I 20 aI 10 I 10 a11 10 I 10 a - - - f I1 110 BI 10 I1 15 aI1 35 I1 15 BI1 20 I1 25 PI 12 I1 10 B560044005600550059005900680067002930223621311920* The flask denoted by a is the simpler type, whilst the ffask B is tlie specialj.One hundred grains of Xochelle salt were used in this analysis.vessel shown in Fig. 2.TABLE 111.WeightNo. of Sncl, Sampleof in Offraction. vacuum. silver.A.4 9.02435 I1A.5 9.39855 I1A.6 10.32760 IVA.7 9.17005 IVA.8 6.68995 I1Weightof silvervacuum. SnC1,/4Ag.14.9475 0.60373615.5666 0.60376417.1058 0.60374815.1880 0,60376911.0808 0.603742in Ratio :Atomicweightof tin35.460).118.685118.696118.689118.699118.686(C1=Atomicwaightof tin (el=35.457).118.697118-708118-701118.71 1118.698Devia-tionsfrom meanandaverage- 0.002 + 0.009 + 0.0023-0-012 - 0.0018rrOl.S.B.l 9.33130 I 15.4570 0.603694 118.666 118.678 -0.021B.2 7.77450 I 12.8764 0.603779 118.702 118.714 +0*0180 BRTSCOE: THE ATOMIC WEIGHT OF TIN.T.4sr.1~ 111 (cou fin‘l~ed).Devia-Atomic Atomic tionsWeight MTeight weight weight from meanNo.of SnC1, Sample of silver of tin of tinof in of in Ratio: (C1= ,C1=frmtion. vacuum. silver. vacuum. SnC1,/4Ag. 35.460). 35.457).C.1 9.22730C.2 7.55785C.4 9.22875C.5 9.21890C.6 11.29055(3.7 9.89635C.8 9.43385C.3 9.11535I1 15.2840I 12.5183I1 15.0983I1 15.2866V 15.2696V 18.7017I11 16.3912I1 15,62500.6037 230.6037440.6037320.6037 150.6037420-6037 150-6037600.603767118.678118.688118.683118.675118.686118-675118-694118-69611 8.69011 8.70011 8.695118.687118.698118.68711 8.70611 8.708General means.................. 0.603742 118.686 118.698General means, omitting B. 1and B.2 ..................... 0.603744 118.687 118.699Means of Series A ............ 0-603752 118.691 118.703Means of Series B ............ 0.603734 118.684 118-696Means of Series C ............ 0.603734 118.684 118.696andaverageerrors.- 0.009 + 0.001- 0.004-0.012- 0.001- 0.012 + 0.007 + 0.0090.0080.0060.0050.0380.006The Ratio of Sta?bnic Chloride t o Silver.Of twenty-four fractions of stannic chloride which should havebeen obtained in weighed bulbs eighteen were actually collectedand weighed; of these two were rejected on account of possiblecontamination with water (A1 and A2), and one met) with anaccident (A 3).The remaining fifteen fractions were analysed inthe manner which has been described. The details of the analysesare given in table 11, and the results obtained in table 111.The analyses of fractions B l and B 2 were of the nature ofpreliminary determinations, as they were the first to be performed,and in all such investigations as the present the liability to erroris unduly great until the technique has been thoroughly mastered.On examination of table I11 it’ will be found that the extremevariation in the values of the ratio SnC1, : 4Ag lies between thevalues for B 1 and B 2, and amounts to 0.000085, or 1 part in 7200,with a corresponding variation in the atomic weight of tin of 0.036or 1 part in 3300.If the analyses of fractions B 1 and B 2 areomitted from consideration and those of fractions A 4 to A8 andC 1 to C 8 taken together as the final series of thirteen determina-tions, the values of the ratio SnCl, : 4Ag vary by 0.000054 or 1 partin 11,000, which corresponds with a variation in the value for theatomic weight of 0.024 or 1 part in 4950.The general mean of all fifteen determinations of the ratioSnCl, : 4Ag is 0-603742 with a ‘‘ probable error” of 4.1 x 10-6,and the mean of series A and C only is 0.603744 with the probableerror ” 3-5 x 10-6BRISC'OE: THE ATOMIC WEIGHT OF TIE. 81I n view of the fact that the variations of the results of theanalyses of B 1 and B 2, from the mean, although larger than thevariations in tho other analyses, are in opposite directions, it is,perhaps, doubtful whether their rejection is justifiable.It will beseen, however, that in either case the mean value of the ratioSnC1, : 4Ag is practically the same, although the probable error isnaturally greater if the extreme valua are considered.I n discussing the value to be attached t o these results regardmust first be had to the purity of the stannic chloride and silverused. For the reasons given in the section on the preparation ofstannic chloride there seems t o be little doubt that the material wasobtained free from chlorides of otlher elements. The action of asmall quantity of water on a large excess of stannic chloride doesnot produce hydrogen chloride, but thO solid hydrate of stannicchloride previously referred to, and there is good reason to believethat all tzaces of this hydrate were eliminated by the processes ofdecantation and fractionation described.Supposing that a small amount of hydrogen chloride had beenformed it is conceivable that a mixture of this substance withstannic chloride, of constant, boiling point, might have resulted, butit seems improbable that the hydrogen chloride could have con-tinued to exist in the presence of the metallic tin used in the firstdistillation in a vacuum.Moreover, in the distillation of mainfraction A the earlier fractions, Ax-A5, were condensed a t thetemperature of liquid air, whilst the later fractions, A 6-A 8, werecooled only to -loo; and the attendant variation in pressurewithin the apparatus should have had an effect on the compositionof the distillate if a mixture were being distilled.Another possibility is that the vapour of stannic chloride mightundergo partial dissociation into stannous chloride and chlorine.The evidence that no such dissociat'ion occurred is (1) that thereduction of stannic chloride t o stannous chloride even by hydrogendoes not take place to an appreciable extent below 200° (Meyer andKersteiii, Ber., 1913, 46, 2882); (2) that the distillations werealways carried o u t a t temperatures below 30°; and (3) that whenthe final main fraction Y was evaporated to dryness in a vacuumthere was no visible residue of stannous chloride.No systematic variation can he observed in the values of theratio of stannic chloride to silver, but, on the contrary, the com-position of the earlier fractions of series A is substantially the sameas that of the later fractions of scrim C.This fact appears toindicate clearly that any impurity which may have been presentin the stannic chloride formed so small a proportion of the wholeVOL. CVII. 82 BRlSCOE: THE ATOMIC WETGHT OF TIN.as to produce an effect on the results which is negligible in com-parison with the experimental error.With regard to the silver a similar argument may be advancedto show that the variation in the amount of impurity present inthe five different samples employed must have produced effects onthe atomic weight small in comparison with the errors of experi-ment. These samples of silver were prepared by two differentmethods, in the one case with variations in the details of procedure,and the agreement of the results, therefore, affords strong presump-tive evidence of the substantial purity of all the samples.Afurther indication of this is afforded by comparative assays madea t the Royal Mint, wliich show that both the1 methods employedfor the purification of the silver are capable of yielding a productwhich is 999-97-999.98 fine as compared with the purest sampleof silver a t the Mint taken as 1000 fine. The effect of any impurityin the silver would be to cause the atomic weight deduced to belower than the true value.I n addition t o impurity in the stannic chloride or silver thefollowing possible sources of error in the determination of the ratioSnC1, : 4Ag have to be considered:(I) Presence of chloride in the reagents (water, nitric acid, oxalicacid, Rochelle salt) used in the titrations.(2) Mechanical loss of silver in transference.(3) Co-precipitation of silver oxalate or tartrate with silver(4) Mechanical loss of chloside during the titration.(5) Occlusion of soluble chloridel by precipitated silver chloride.(6) Prevention of complete precipitation of chlorine as silverchloride owing t o formation of complex ions containing chlorine.Errors due to causes (I), (2), or (3) would make the apparentvalue of the atomic weight less, whilst those due to causes (4), (5),or (6) would make i t greater than the true value.The precautions taken in the purification of the reagents andthe negative results of tests for chlorine in them render it improb-able that the error due to (I) is appreciable. With regard to (2)an attempt was made to ascertain whether loss of silver occurredin the process of transference by the use of the new form of solutionflask described, but the change in procedure’ had no appreciableeffect on the results.* Supposing co-precipitation of silver oxalate* In this operation and in the breaking of the bulbs (p.77) the methods ofmanipulation were varied in the course of the analyses with a view to ascertainingwhether the “accidental errors” of the determinations could be traced to loss orgain of silver or chlorine through lack of refinement in the manual details.Theresults clearly lead to the inference that such errors cannot be attributed to thissource. That such variations are due to errors of weighing is highly improbable ;chlorideBRISCOE: THE ATOMIC WEIGHT O F TIN. 83or tartrate with silver chloride (3) to take place, the amount ofsilver thus lost to the titration would, presumably, decrease as theconcentration of nitric acid was increased, but no such effect isapparent in the data. Further, the1 agreement of the result of theanalysis of fraction C4, in which tartaric acid was used, with theother results suggests that no perceptible error is due to this cause.All possible care was taken t o prevent mechanical loss of chlorideduring the analyses (see p.77), and it is evident that such asource of error as (4) cannot otherwise be eliminated. I n referenceto (5), it seems reasonable t o assume that any occlusion of silveror chlorine by the precipitate was inconsiderable, as variations i nthe duration of the anaIyses (between fourteen and forty-sevendays) and in the concentration of the solutions do not appear t ohave any influence on the result. Baxter and Moore, in the courseof analyses of phosphorus trichloride (J. A4mer. Chern. SOC., 1912,34, 1644), have shown that occlusion produces an error which isat most 1 part in 35,000, and is much reduced when the durationof the analyses is, as in the present case, fourteen days or more.The1 consideration of the probable magnitude of the uncertaintyin the results due t o (6) presents difficulties owing t o the lack ofknowledge of the precise manner in which oxalic and tartaric acidsa d in retaining metastannic acid in solution.There does notappear to be in the literature any evidence for the formation ofcomplex ions containing chlorine (see p. 76), and, moreover, weresuch complexes formed, the amount of chlorine thus retained wouldcertainly depend in some way on the concentration of the organicacid. That a large variation in the amount of oxalic acid usedproduced no corresponding variation in the ratio of stannic chlorideto silver is, therefore, good evidence that the chlorine was alwayscompletely precipitated. Again, the proportion of chlorine presentin a non-precipitable form would very probably vary with theacid used, but no such effect is to be observed on comparison of theanalysis of fraction C 4 with the others.Criticism may possibly be directed t o the fact that no analyseswere made by the gravimetric measurement of the ratioSnCl, : 4AgCI.It has, however, been shown repeatedly that thetitration with silver and the weighing of silver chloride give identi-cal results when applied t o the same material."they must, therefore, as Richards has sugqested, be mainly due to impurities in thereagents or to irregularities in the course of the chemical reactions involved in t h ianalyses.* For determinations in which the two methods were applied together to a seriesof experiment3 see Richards and Archibald, Proc.Amen Acad., 1903, 38, 413 ;Archibald, T., 1904, 85, 786 ; Baxter and Coffin, Zeitsch. anoTg. Chem., 1906, 51,171 ; Baxter and Hines, J. Amel.. CJLem Soc., 1906, 28, 1580. For the application6 84 BRISCOE: THE ATOMIC WEIGHT OF TIN.As the gravi-volumetric method involves much simpler manipnla-tion, and is therefore less liable to error, i t seemed preferable t,oadopt i t entirely. Moreover, the performance of the gravimetricanalyses almost necessitates the co-operation of two persons, as itis difficult for one alone to carry out with certainty such operationsas the filtration of large volumM of liquid through Gooch crucibles.The antecedent data involved in the calculation of the atomicweight of tin from the values of the ratio SnCl, : 4Ag are theatomic weights of silver and chlorine, for which the values 107.88and 35-46 respectively are adopted by the International Commis-sion.Probably the numbers Ag = 107.880 and C1= 35.457 deter-mined by Richards and his collaborators represent more nearly tlieratio between these atomic weigh&, although, in the light of recentinvestigations, it would seem that the true value of silver in refer-ence to 0 = 16.00 is probably slightly below, rather than above,11 7-880.I f the values Ag=107*880 and c1=35*457 are used, the atomicweight of tin deduced is about 0.012 higher than that obtained bytaking C1= 35.46.I n table 111 are given the mean values of the atomic weight oftin deduced from determinations of the ratio SnCl : 4Ag; it willbe seen that the mean of the fifteen determinations is 118.69 or118.70, according to the value adopted for the atomic weight ofchlorine.The “average error”+ of the values of the atomicweights from which each mean is deduced is given in the lastcolumn opposite that mean. The “probable error” of the valueSn= 118.698 (C1= 35.457) deduced from all determinations is0-0013, whilst that of the value Sn = 118.699 obtained by omittingthe analyses of B 1 and B 2 is O * O O l l . *Taking into consideration the possible sources of error in thedeterminations i t is clear that the most probable value of theatomic weight of tin is 118.70. On the basis of the experimentsnow under discussion it does not seem likely that the true valueof the atomic weight will differ by more than 0.01 from this value.of the two methods separately in different series of experiments, see, for example,Thorpe, Bey., 1880, 16, 3014 ; Richards and Stachlcr, Pub.Cnrnegie Inst.Washington, 1907, No. 69, 7 ; Richards and Wells, ibid., 1905, No. 28, 52 ;Baxter, Hines, and Frevert, J. Anter. Cha~n. SOC., 1906, 28, 770 ; Richards andHonigschmid, ibid., 1911, 33, 28.i t In the general statement of the results of this and earlier determinations theauthor has given in all cases the avccage error of the means, and not the “ probableerror,” as the former is IL more tiscfiil standard of comparison when the relativeaccuracy of the individual observations in different series is under consideration. Atthe same time, in accordance with custom, tlie “probable error” is also given incertaiu casesBRISCOE: THE ATOMIC WEIGHT OF TIN.8ETABLE IV.Error in the atomic weight of tinproduced by an error of 1 in 1000 of the ratio. Ratio measured.Sn:SnO, ............ 0.56Sn:2BaS04 ......... 0.053Sn : (NH,),SnCl, ... 0.174Sn : K,SnC16 ......... 0.129Sn :SnBr, ............ 0-162SnC1, : 4Ag ......... 0.25I n estimating the weight attaching to this result i t is iiecessaryto consider in what way, if any, the earlier determinations areunsatisfactory. The discordance in the values 1-7 and 11 intable I clearly suggesb that the oxidation of tin with nitric acidis capable, even in the hands of the most experienced analysts, ofgiving very vaziable results. The best determinations of the ratioSn : SnOz by oxidation (Nos. 4 and 5)) in which attempts weremade to ascertain and t'o avoid the chief errors, give valuesagreeing with those obtained by the reduction of stannic oxide inhydrogen, but if the results are considered in relation to thoseobtained by Bongartz and Classen, they suggest that stannic oxideis not, a suitable compound to employ in atomic-weight determina-tions, a conclusion the correctness of which seems inherently prob-able from the general character of that substance.From table IV, in which is given the error produced in theatomic weight of tin by an error of 1 part in 1000 in the deter-mination of each of the ratios here dealt with, i t will be seen thatthe measurement of the ratio Sn : SnO, is, on this ground also,open to serious objection.The results obtained by Bongartz and Classen cannot be so easilyaccounted for. Whilst the nature of some of the compoundsweighed (ammonium and potassium stannichlorides) and themethods of analysis employed (for example, in the determinationof the ratio Sn : ZBaSOJ are not in accordance with modernstandards, the results obtained in the different series of determina-tidns are fairly concordant.One very serious criticism of theirinvestigation that can be made is that the tin was in every cawweighed as an electrolytic deposit on platinum, and that this tinhad been deposited from a solution containing a number of differentcompounds in solution, and was dried only by washing with alcoholand heating to looo. It has been shown that electrolyticallydeposited silver contains included material amounting t o about0.02 per cent., which is not expelled by heating to any temperatureshort of the softening point of the metal (Richards and Heimrod,PTOC. Amer. Acad., 1902, 37, 415; compare Smith, Mather, andLowry, Phil. Trans., 1908, [ A ] , 207, 545; Richards, Proc. Amer86 BHISCOE: THE ATOMlC WEIGHT OF TIN.Acad., 1908, 44, 91). Moreover, i t has been found that, even insuch a simple case as the electrolysis of silver nitrate solutions,side-reactions occur ; in the deposition of tin such side-reactionsmight seriously affect the composition of the deposit. It is,therefore, not improbable that metallic tin, deposited from anelectrolyte of complex composition, may have contained includedsolution or gas sufficient to affect its weight to a serious extent,and! it does not appear that any steps were taken to ascertainwhether the tin was impure from this cause. Assuming that suchinclusion of impurity occurred, the weight o l metal found wouldbe greater than its true1 weight, and the atomic weight deducedfrom determinations affected by this error would be higher thant h e true value. It is worthy of note that in the case where thecomposition of the electrolyte was simplest, namely, the analysis ofstannic bromide, the result obtained was lower than the otherresults by 0.1, a quantity greater than the average error of anysingle series of their determinations except the first.The true value of this atomic weight cannot, of course, be fixedby any one series of determinations, and further work on thesubject is much t o be desired. The result here given is not improb-able, in that it lies within the range of earlier values. It. has, it istrue, been arrived a t by the determination of a single ratio, but theanalytical methods involved have repeatedly been shown to becapable of yielding results of the highest degree of accuracy. Allthe variable factors in the analyses have been varied over a com-paratively wide range, notwithstanding which fact the results alllie between 118.678 and 118.714, and eleven of the fifteen between118.690 and 118.711. The atomic weight finally deduced,Sn = 118.70, would seem t o be worthy of considerable confidence.I n conclusion, the author desires to express his thanks toProfessor H. Brereton Baker f o r the numerous facilities he has sokindly afforded for the prosecution of this investigation.ROYAL COLLEGE OF SCIEKCE,IXIPERIAL COLLEGE OF SCIENCE AND 'fECHNOLOGY

ISSN:0368-1645    

DOI:10.1039/CT9150700063

出版商:RSC    

年代:1915

数据来源: RSC

摘要:

HARPER AND MACBETH : COLORATIONS PRODUCED, ETC. 87IX.- Colorations Produced Some Organic Nitro -compounds with Special Reference to Tetra-nitromethane.By ERNEST MAGOWAN HARPER and ALEXANDER KILLEN MACBETH.IN a recent paper (P., 1913, 20, 161) Clarke, Macbeth, and Stewartextended the observations of Ostromisslenski ( J . p r . Chern., 1911,[ii], 84, 489), and showed that the production of colour in the caseof tetranitromethane was not limited to the influence of ethyleniccompounds; the phenomenon is of more general occcrrence, and, asa rule, a substance containing an atom which can exercise a higherdegree of valency gives a colour on the addition of tetranitro-methane.As the phenomenon is primarily a colour reaction it seemed moshreasonable to approach it along the line of absorption spectroscopy,and such an investigation was undertaken by us.The absorptionof the influencing substance was obtained, and from it the strengthof solution which would be fairly diactinic was found. Tetranitro-methane was photographed in an alcoholic solution of the influenc-ing substance of a suitable constant strength, and the effect on theabsorption of tetranitromethane was thus recorded.Tetranitromethane itself gives no colour in dry alcohol. With aninfluencing substance a yellow colour is produced on mixing thesolution, but the absorption spectrum is found to differ only veryslightly from that of the nitro-compound alonel. The colour, how-ever, deepens on keeping, and after some time the spectrum showsa great change ; the absorption becomes strongly selective, and aband appears a t the region 1 h 2800-1 / h 2900.The penetrationof the band increases as time goes on, and finally attains a maxi-mum. This can be seen from a study of the figures given.Carvene, allyl alcohol, and P-methyl-AP-butylene (trimetliyl-ethylene) were taken as represent'atives of substances containing theethylene linking. The results are expressed in Fig. 1. The increaseof the penetration of the band is clearly seen in the cases of carveneand P-methyl-L\@-butylene.The substances containing bivalent sulphur atoms, which werestudied, were ethyl, propyl, allyl, and pentamethylene sulphides ;also l:4-thioxan and ethyl mercaptan. The results obtained intypical cases are expressed in Fig.2.Amino-compounds were also used as influencing substances, andthe curves are given in Fig. 2.From a study of the curves it is seen that the general effec88 HARPER AND MACBETH: COLORATIONS PRODUCED BYobtained on mixing tetranitromethane and a subst,ance containingan ethylene linking, or an atom capable of exercising higher valency,FIG. 1.Oscillation frequencies.0 0 0 0 0 0 0 0 0 0 o o o o o c o o o oc . l w m c a E Z S 2 S E Z S I10 nzni. N/10010 112112. N/100010 min. Nf10010 112111. N/1000Upper czcrvcs.a , . . . . . .--Tetranitrometham + N/IO-allyl alcohol (after 14 days).Telranitromethanc + N 110-carvene (after 4 hours).9 2 , Y ), (after 3 days).¶ > Y Y ,, (afler 14 days).- _ _ _ _ . _ _ . _Lower curves.-- Tetraqtitronwlhane -!- N/IO-amylcne (after hydrogen chloride).,, 9 , ,, (after 5 days).Y V Y ¶ ,) (after 14 days).¶ , 9 , ,, (after mixing).- _ - - _ ._ . _- . . - . . - . . . ..... Nitroform.is the development of strong selective absorption with a maximumat3 or about, l / h 2850SOME ORGANIC NITRO-COMPOUNDS, ETC. 89FIG. 2.Odlntion frcgttc.rl.cies,0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0m ~ ~ c 9 o G a e w ~ o ~ m e 4 m m m m m m m + *10 mm. N/100010 mnt. N/10010 mm. N/1000Upper c l m e s .--- - _ __ . - . - . I Tetranitromothanc + N/10-dieth!jla?nine ( 15 minutes).1 Tetranitromethane + N/ZO-triethylainine (1 5 minoitca).Tetranitromethane + NJ10-ethyl nrercaptan (nfter 15 minutes).(after 4 hours).Tetranitromethane + N/10-methylniiainc (after 15 minutes).(after 30 minutes).Lower curves.-- Tetranilromethanc + N/10-~"itanaethylenc Sulphide (afler 2 days).Tetranitromethane + NJlO-propyl sulphide (ufter 5 days)., 7 9 , 1, ,, (qfler 5 days).9 9 9, 9 , ,, (after mixing).- _ _ _ _.-.- .. . . . . . .I n searching for the explanation of these colorations i t eeemedadvisable to examine other simple nitro-compounds. Substancescontaining one nitro-group (nitromethane, etc.) were tried, but wit90 HARPER AND MACBETH : COLORSTIONS PRODUCED BYnegative results, no change in the absorption being observed. Itappeared likely that the nitro-group in tetranitromethane wasinfluenced by the electronegative character of the other three. Toapproximate to this molecular condition in the mononitro-com-pounds, subsLances were tried in which some of the1 hydrogen atomswere replaced by electronegative atoms.Monochlosonitromethane,monobromonitromethane, dibromonihromethane, and chloropicrinwere photographed in the presence of an influencing substance, butno development of a band was observed. It appeared of interestnow to examine the behaviour of another type of moiionitro-com-pounds, namely, the alkyl nitrites. Colours were obtained (Harperand Macbeth, P., 1914, 30, 15) when these were treated withsubstances containing unsaturated centres or atoms capable ofexercising additional valency. The tints developed closelyresembled those obtained when tetranitromethans is similarlytreated.Amyleae and carveilel were taken as representative of the classcontaining ethylene linkings.These give faint, greenish-yellowcolours with propyl, isopropyl, isoamyl, and benzyl nitrites, Thecolours are intensified on warming.With bivalent sulphur compounds the colorations obtained arevery pronounced. Ethyl sulphide gives a deep red colour withthe nitrites mentioned. Ethyl mercaptan yields similar results, thecolour being slightly more intense. The production of colour is notconfined to the organic sulphides, but is also shown by someinorganic sulphides-sodium and potassium sulphides giving decidedorange colours on shaking with the alkyl nitrites or with tetra-nitromethane.Amino-compounds were also inveekigated, and yield the character-istic reddish-yellow and orange colours on mixing with the nitrites.The above cases of colorations were also investigated spectro-scopically. The curves are shown in Fig.3. It will be noticed thatall the nitsites examined have practically the same absorption. Allthe curves showing a strong selective effect with a maximum a t anapproximate frequency of 2850.The absorption spectrum of ethyl nitrite and ethyl mercaptanimmediately after mixing does not differ appreciably from that ofthe nitrite itself; after some time (twenty hours) the colour of thesolution deepens, and the penetration of the band increases; thegeneral absorption extends further into the visible, and thusproduces the change in colour. Similar effects were noticed withthe other nitrites.The case of the amines has also been studied,but the increase in the penetration of the band is here not nearlyso greatSOME ORGANIC NITRO-COMPOUNDS, ETC. 91From the foregoing i t is evident that the question of the tetra-nitromethane colorations is closely allied to the behaviour of theFIG. 3.Oscil latwn freqtmlcies.10LOt oNil0Nil0N/100Upp~r curves.-- EtJbyl nitrite.__ . __ . - Elhgl nitrite + N/IO-ethyl mercaplan (afm 60 hourb).AmyE nitrite + 9 9 ,, (after 14 hours). ---- . . . . . . . . Piperidinium nikrite.Lower CUTVPS.-- Phnyldinitromethaw in C H C&.2, 9 ) C2H6-0.$ 9 ¶ ? H,O-- - - A _ ........ K salt qf same in H,O. -.-._ Bromodinitromethane in H,O + 1 naol. HC1. - . . - . . -alkyl n h i t e s ; so that if it is proposed to account for the formerby the behaviour of the latter a clear understanding of th92 HARPER AND MACBETH: COLORATIONS PRODUCED BYprocesses involved in the latter is essential.A plausible explana-tion of the latter is here advanced. The spectroscopic results pointt o the formation of additive compounds of the nitrites and the othersubstances, tlie alteration of the absorption on keeping being dueto such cause. From the spectroscopic results it is impossible todecide whether the colour is produced by the final additive productof the nitrite and the influencing substance, or is due to an inter-mediate stage in the addition reaction; the question must thereforebe approached from another direction.VorlSnder's hypothesis to account for the formation of thecoloured intermediate product in the reaction of an ap-unsaturatedketone and an acid may be recalled and applied to tlie present case.With the alkyl nitrites and the bivalent sulphur compounds wehave :(1.1 (11.1Formula I represents the coloured intermediate product which is .quickly produced, and passes slowly into the true additive product11, which is colourless.A similar hypothesis covers the cases of the amino-compoundsaiid the unsaturateid hydrocarbons.(1.1Colourecl, quickly produced.CHR,(ONO) *CHRRCHRR, * CHR,*ONOCHR,-CHR, ~ CHRl:CHR2 + R*ONO -+ [R.ONO] or(1- 1 (11.) -Colonred, quickly produced.If Vorlander's theory is applied the1 existence of the finaladditive product must be accepted; proof of the isolation of suchcompound must therefore be forthcoming.RBy and Rakshit (T.,1911, 99, 1016, 1470) and RBy and Datta (ibid., 1475) haveprepared the additive products of nitrous acid and the alkylnitrites with primary, secondary, and tertiary amines. Such com-pounds are all colourless, or, a t mwt, weakly-colourFd compounds.Neogi (T., 1911, 99, 1598) has obtained additive products of cycliccompounds containing tertiary nitrogen with the nitrites. Withfew exceptions these are colourless compounds. The isolation of thefinal additive products of the nitrites with bivalent sulphur com-pounds has been undertaken by us. The existence of the finaladditive product of the nitrites is'thus established, and the applicaSOME ORGANIC XITRO-COMPOUNDS, ETC.93tion of Vorlander’s hypothesis to cover the cases of nitrite colora-tions is theref ore justified. The coloured substances obtained byNeogi (pyridine metlionitrib, etc.) are evidentdy cases where thesubstance has been isolated in the intermediate stage only.Ostromisslenski ( J . pr. Chenz., 1911, [ii], 84, 495) advancedseveral fornuke to account for the phenomenon observed. The‘‘ tetranitromethanates ” had a constitution which was one of fiv0indicated, only one nitro-group being involved. Considerable doubtis thrown on the validity of his assumption by our study of themononitro-compounds where no effect whatever is obtained whensuch substances are used instead of tetranitromethane. The onlycase observed m here colorations are obtained with a molecule con-taining one nitro-group is that of the alkyl nitrites.None of theformuh advanced by Ostromisslenski appears to cover this case.The question of the colorations produced by tetranitromethaneseems to us to be answered from the data we have obtained. Onemust first postulate an isomerisation of one, or more, of the nitro-groups of the tetranitromethane to a nitrite form. This is inducedby the influencing substance used. We have then an intermediateadditive product formed as in the case of the nitrites, such com-pound being the result of the potential of the unsaturated com-pound, or the atom capable of higher valency. This isomerisation isnot an extravagant demand, and affords a very simple explanationof the observed facts.The first fact tlhat warrants this assumptionis the similarity of the colorations obtained in the cases of tetra-nitromethane and the alkyl nitrites. The curves show clearly theparallel between the nitrites and tetranitsomethane. The charac-teristic nitrite band has a maximum a t an approximate frequencyof 2850, and it is noteworthy that all the curves obtained by theinteraction of tetranitromethane and widely diff eriiig compoundscontaining centres of residual affinity develop this same band. Ittherefore seems justifiable to postulate that such isomerisation doesoccur. The isomerised nitrite group need not be chemically identi-cal with the ordinary nitrite group.Hantzsch and Voigt (Ber., 1912, 45, SS), from it spectroscopicstudy of many nitro-compounds and their salts, classified thesecompounds under three types, namely, (a) the true nitro-group,( b ) the aci-nitro-group, and ( c ) the conjugated aci-nitro-group. Theabsorption spectra obtained with tetranitromethane closely resemblethe particular case of Hantzsch’s conjugated mi-type when the com-pounds contain two or more nitro-groups attached to the1 samecarbon atom.The application of this hypothesis to the case oftetranitromethane might theref ore appear likely. Such applicationmnst first postulate the decomposition of the tetranitromethane t94 HARPER AND MACBETTI : COLORATIONS PRODUCED BYnitroform in order t o supply the necessary waiiilering hydrogenatom. The application of the conjugated mi-nitro-hypothesistherefore involves an assumption as great as, if not greater than,that demanded by the nitrite isomerisation hypothesis.Moreover,the necassary experimental confirmatory evidence is wanting. Ostro-misslenski (Zoc. cit.) showed that' colours can be obtained withnitroform, but these are not nearly so intense as those obtainedwhen tetranitromethane is used. I f the coloration depends on theconjugated mi-form of nitroform i t is reasonable t o expect that) itwill be more intense when nitroform itself is used than when thedecomposition of tetranitromethane is depended on for the supplyof nitroform.It has been shown (Fig. 1) that tetranitromethane and amyleiiegive a band on keeping for ten days. After treatment with dryhydrogen chloride1 this band disappears, and the absorption spec-trum is found to be practically identical with that of tetranitro-methane. It is therefore not reasonable to suppose that the coloursare a result of the decomposition of tetranitromethane'.Direct esti-mations of the amount of nitro-compound present in the solutionsare now being made, and will form the subject of a further com-munication.Examination of the curves attributed by Hantzsch and Voigt(Zoc. c i t . ) to the conjugated mi-nitro-compound shows that they areof widely differing nature, having maxima a t different positionsextending over the photographic range. The only uniformity in thecurves is noticed in the case of those compounds which have twoor more nitro-groups attached t o the same carbon atom, and awandering hydrogen atom.The salts of all these have maxima a tllx2850. The same band is also found in their aqueous solutions.It would appear that the hypothesis of a conjugated nci-nitro-groupcould be rejected in these cases, and the observation accounted forby the nitritel isomerisation. To meet the cases of dinitro-com-pounds which contain no wandering hydrogen atom, and developno selective absorption, we are1 driven to the assumption that alabile hydrogen atom has an influencing effect on ths isomerisationin question.Baly and Rice (T., 1913, 103, 2085) examined the absorption oftrinitrophenol, trinitroanisole, and trinitrobenzene under differentconditions. Two mo,difications of trinitroanisole in alcoholic solutionare noted.The colourless form changes to a yellow variety onexposure t o sunlight, and the latter shows a selective effect with amaximum in the region 350 ,up (l/h 2860). These modificationscannot be accounted for by Hantzsch's hypothesis of a conjugat;edaci-nitro-group. Baly and Rice advance an hypothesis t o explaiSOME ORGANTC NITRO-COMPOTJXDS, E m . 95this phenomenon, and regard it as the first stage in the openingup of the force fields of the nitro-compound by the solveiit. Itwould appear t h a t the instance quoted may be regarded as a con-firmation of the nitrite isomerisation hypothesis.Baly and Rice (Zoc. cit.) suggested t h a t the colours observed byus are) due to similar causes, and are instances of “ t h e openingup of tlie closed system of the complex nitro-compound by the forcelines due to the1 various residixal affinities with which it is treated.”Their theory would cover the cases of the polynitro-compoundscoiitaiiiirig rz labile hydrogen which develop a selective effect inaqueous or in sodium ethoxide solution, observed by Hantzsch(loc.cit.) ar,d Hedley (Ber., 1905, 36, 1195, Fig. 3). Here alkalifacilitates the opening up, and acid inhibits it. A similar effect isobserved in the case of tetranitromethane, an alcoholic solution ofwhich in tho presence of iV/lO-sodium ethoxide shows a band withmaximum absorption a t 1 / A 2850, tlie head of the band being a t 21(log thickness in mm. of AT/ 100,000-solution). The hypothesis hasthe same weakness as the nitrite isomerisation hypothesis, whichnecessitates tlzs assumption of the influence of the labile hydrogenatom t o cover cases of the nitro-compounds wliicli liave no suchatom and do not develop a selective effect.If it is granted t h a t tlie first stage in the opening up of thenitro-compound is tantamount t o the formation of a grouping akinto the nitritic structure it is evident t h a t Baly and Rice’s theorycan be brought into conformity with our hypothesis.Sicmrnnry.(1 ) The colorations produced by tetranitromethane when influ-enced by unsaturated substances or compounds containing an atomcapable of exercising a higher valency were spectroscopically investi-gated. A resultant selective absorption was noted in all the casesexamined, with a maximum a t 1 / A 2550 and penetration dependingon the time elapsed since mixing.(2) The colours of the alkyl nitrites when treated with the sameinfluencing substances are found t o bear a close relation t o thecoloration produced with tetranitromethane.The nitrites them-selves have selective’ absorption a t l / ~ 2850. An increase of pene-tration occurs on keeping tho mixtures of the nitrites and influenc-ipg solvents.(3) An hypothesis to account for the tetranitromethane colora-tions is considered. This calls f o r the isomerisation of one, or more,of the nitro-groups of the nitro-compound to a structure similart o the nitritic form. The colours are then regarded as parallel tothose obtained with the alkyl nitrites, whether these are due t 96 CLOUGH : OPTICAL ROTATORY POWER OF DERIVATIVES OFVarlander’s intermediate compound or not’. The isomerised“ nitrite ” group need not be chemically identical with the ordinarynitrite group. Such isomerisation occurs in other cases when alarge number of nitro-groups are attached to the same carbon, andalso when a labile hydrogen is attached to the carbon along withtwo or more nitro-groups.(4) Baly nnd Rice’s “ opening up ” theory may be brought intoconformity with the views of the authors if the first stage in suchopening up of nitro-compounds may be regarded as equivalent tothe formation of a st’ructure akin to the nitritic form.I n conclusion, we wish t o express our best thanks to ProfessorLetts and Dr. A. W. Stewart for the interest they have taken inthis work and the advice they have given throughout its course.TIIE SIR DoN-kLD CURRIE LABORATORIES,THE QUEEN’S UNIVERSITY OF BELFAST

ISSN:0368-1645    

DOI:10.1039/CT9150700087

出版商:RSC    

年代:1915

数据来源: RSC