PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE 6/01 c:/CIRC/DaleDue.p65.p. 15 ABSTRACT DIPOSIT IVE CAR BONIUM IONS by Robert Roland Rafos The primary purpose of this thesis was to further investigate the sc0pe of dipositive carbonium ion chemistry. The aluminum chloride-catalyzed reaction of carbon tetrachloride with pentamethyl- benzene, which gave high yields of trichloromethylpentamethylbenzene (l), was extended to the preparation of 4-bromo-, 4-chloro-,. and 4-fluoro-2, 3, 5, 6-tetramethylbenzotrichlorides. Each of these 4-halo derivatives dissolved in 100% sulfuric acid to form intensely red solutions. Two moles of hydrogen chloride were quantitatively swept from these solutions, using dry nitrogen. Cryosc0pic measurements showed that five particles were produced in the reaction. Hydrolysis of the colored solutions gave nearly quantitative yields of the corresponding 4-halodurenecarboxy1ic acids, which were synthesized by independent routes. By analogy with earlier work, the best explanation for these data is the formation of stable dipositively charged carbonium ions . + -X O cc13+2sto,—->x C C-Cl+ZHC1+ZHSO4' The structures of these dipositive ions were investigated by ultraviolet, visible and proton magnetic resonance spectroscopy. Dipositive carbonium ions formed from single ionizations at two different sites in a molecule were also investigated. The three compounds Robe rt Roland Rafo 3 studied, which were in a sense analogs of triphenylcarbinol, were tetraphenyl-p-xylyleneglycol, tetra-p-anisyl-p-xylyleneglycol and 9, 10-dihydro-9, lO-dihydroxy-9, lO-diphenylanthracene. The first two dissolved in 100% sulfuric acid to form intensely red solutions. Hydrolysis gave nearly quantitative yields of the starting glycols. A study of the visible spectra in solutions of varying acidity showed that both'glycols undergo reversible stepwise ionizations to mono- positive and dipositive carbonium ions. 9, lO-Dihydro-9, lO-dihydroxy-9, lO-diphenylanthracene dissolved in 100% sulfuric acid to form a deep-blue solution. Hydrolysis did not yield the starting glycol, but rather a mixture containing 9, 10~diphenyl- anthracene (36%), 4-phenyl-Z, 3-benzofluoranthene (19%) and some I products which remain unidentified. ‘ In much less acidic solutions 9, lO-dihydro-9, lO-dihydroxy- 9, lO-diphenylanthracene undergoes the normal stepwise ionization to a yellow monopositive and a red dipositive carbonium ion, as shown by hydrolysis experiments. The pK'S for the process: + (ROH) + H+ ——> R+++ H20 (—— were determined spectrosc0pically for the three glycols and were found to be -8. 7 for tetraphenyl-p-xylyleneglycol and 9, 10-dihydro-9, 10- dihydroxy—9, 10-dipheny1anthracene and -3.4 for tetra-p-anisyl-p- xylyleneglycol. REFERENCES 1) H. Hart and R. W. Fish, J. Am. Chem. Soc., _§_2_, 5419 (1960). DIPOSITIVE CARBONIUM IONS BY Robert Roland Rafos A THESIS 1 Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1961 ACKNOWLEDGMENT The author wishes to express his sincere appreciation to Doctor Harold Hart for his guidance and encouragement during the course of this investigation. Grateful acknowledgment is also extended to Mr. J. S. Fleming for determining the proton magnetic resonance spectra and to the Petroleum Research Fund of the American Chemical Society whose fellowship program provided personal financial assistance from September, 1959 through September, 1961. ************ ii TABLE OF CONTENTS Page INTRODUCTION .......................... 1 PART A DIPOSITIVE CARBONIUM IONS ARISING FROM.MULTIPLE IONIZATIONS AT A SINGLE SITE IN A MOLECULE . . . 4 RESULTS AND DISCUSSION ................... 5 The Trichloromethylation Reaction ............. 5 Reaction of 4-Halo-2, 3, 5, 6-tetramethy1benzotrichlorides with 100% Sulfuric Acid. . . ............. 7 Cryoscopic Measurements . . . . ............. 8 Stoichiometry . . . . .................... 8 Ultraviolet and Visible Spectra ............... 11 Proton Magnetic Resonance Spectra ............ 17 EXPERIMENTAL ......................... 23 A. Syntheses and Reactions in 100% Sulfuric Acid ..... 24 Preparation of'Bromodurene . . . .......... 24 Preparation of 4-Bromo-2, 3, 5, 6-tetramethy1benzo— trichloride ....... . . . . ........ 24 Preparation of 4-Bromo-Z, 3, 5, 6-tetramethy1benzoic aCid O 000000000000 O O O O O O O O O O 26 Reaction of a Solution of 4-Bromo-2, 3, 5, 6-tetra- methylbenzotrichloride in 100% Sulfuric Acid with Water . . . ............... Z8 Hydrolysis of 4- Bromo- 2, 3, 5, 6- -tetramethy1benzo- trichloride ................... 28 Preparation of Chlorodurene . . . . ........ 30 Preparation of 4- Chloro- Z, 3, 5, 6-tetramethy1benzo- trichloride . . . . ...... . . . . . 30 Reaction of 4- Chloro- 2, 3, 5, 6-tetramethy1benzo- trichloride in 100% Sulfuric Acid with Water . 32 Hydrolysis of 4-Chloro-Z, 3, 5, 6-tetramethy1benzo- trichloride . . . . . . ..... . ...... . 32 Preparation of Dinitrodurene ............ l . 34 iii TABLE OF CONTENTS - Continued Page Preparation of Aminonitrodurene ........... 34 Preparation of 4-Fluoronitrodurene ......... 36 Preparation of 4-F1uoroaminodurene ......... 37 Preparation of Fluorodurene . . . . ......... 37 Preparation of 4-F1uoro- 2, 3, 5, 6 tetramethylbenzo- trichloride ................... 38 Hydrolysis of 4-F1uoro- Z, 3, 5, 6-tetramethylbenzo- trichloride . ............. . . . . . 40 Reaction of 4~F1uoro~ Z, 3, 5, 6-tetramethy1benzotri«- chloride in 100% Sulfuric Acid with Water . . . 40 Preparation of Acetoxymercuridurene ..... . . . 43 Preparation of Nitrodurene . .......... . . . 43 Attempted Preparation of 4-Nitro-Z, 3, 5, 6-tetra- methylbenzotrichloride ....... . . . . . . 43 Preparation of Iododurene ............... 44 Attempted Preparation of 4—Iodo-Z, 3, 5, 6-tetra- methylbenzotrichloride ......... . . . . 45 B. Cryoscopic Measurements ................ 45_ Apparatus ...................... ' . 45 Procedure ....................... 46 Stock Sulfuric Acid .................. 47 C. Quantitative Determination of Hydrogen Chloride from the Reaction of Trichloromethyl Compounds with Sulfuric Acid ....... . . . . . . ......... 47 Apparatus . . . .................... 47 Procedure ......... . ........... . . 48 D. Spectra ........................ . . 48 Ultraviolet and Visible ................ 48 Proton Magnetic Resonance . . . . .......... 48 Infrared ........................ 48 PART B ' DIPOSITIVE. CARBONIUM IONS ARISING FROM SINGLE IONIZATIONS AT TWO SEPARATE SITES IN A MOLE- CULE ..... . .............. . ........ 49 iv TABLE OF CONTENTS «- Continued Page Spectra ........................... 51 Interpretation of the Visible Spectra. . . . . . . . . . . . 67 Hydrolysis Products .................... 72 Cryoscopic Measurements ................. 79 EXPERIMENTAL ......................... 82 A. Syntheses . ....................... 83 Preparation of Tetraphenyl-p-xy1y1eneg1ycol . . . . 83 Preparation of Tetraphenyl-p-xylylenedichloride. . 83 Reaction of Tetraphenyl-p—xy1y1eneg1ycol in 100% Sulfuric Acid with Water .......... . 85 Reaction of Tetraphenyl- p- -xyly1enedichloride in 100% Sulfuric Acid with Water . . ..... 85 Preparation of the Dimethylether of Tetraphenyl- p- xylyleneglycol ...... . . . . . . . 86 Preparation of the Monomethylether of Tetraphenyl- p- xylyleneglycol ............... 86 Reaction of Tetraphenyl- p- xylyleneglycol in Sulfuric Acid (2. 52%)- Acetic Acid with Absolute Methanol .................... 88 Reaction of Tetraphenyl-p-xy1y1eneg1ycol-Methanol with Sulfuric Acid (2. 52%)-Acetic Acid. . . . 91 Reaction of the Monomethylether of Tetraphenyl-p- xylyleneglycol in 100% Sulfuric Acid with Water........... ....... .. 91 Preparation of Tetra- -p-anisy1-p- xylyleneglycol . . 91 Reaction of a Solution of Tetra- -p- anisyl- p- xylylene- glycol in 100% Sulfuric Acid with Water . . . 92 Preparation of 9, 10-Dihydro-9, lO-dihydroxy-9, 10- diphenylanthracene ............ . . 94 Preparation of 9,10-Diphenylanthracene . . . . 95 Reaction of 9, 10- Dihydro- 9, 10- dihydroxy-9, 10- diphenylanthracene in Conc entrated Sulfuric Acid with 15% Sodium Hydroxide Solution . - 95 Reaction of a Solution of 9, lO-Dihydro-9, 10- dihydroxy-9, 10-dipheny1anthracene in 6% Sulfuric Acid in Acetic Acid with Water .. .. 102 TABLE OF CONTENTS - Continued Page B. Solutions for Spectral Measurements .......... 102 Preparation of Weight Per Cent Sulfuric Acid» Water Solutions for the Visible Spectra . . . . 102 Preparation of Weight Per Cent Sulfuric Acid- Acetic Acid Solutions for the Visible Spectra . 103 C. Spectra .......................... 103 Ultraviolet-Visible Spectra .............. 103 Infrared Spectra .................... 103 Test of Beer's Law .................. 104 SUMMARY ............................. 106 MISCELLANEOUS ......................... 109 Results and Discussion ................... 110 Experimental ........................ 111 Preparation of 3, 7, 7-Trichloro-1, 2, 4-trimethy1- bicyclo[4, 2, 0]octa- l, 3, 5- triene . . . . . . 111 Preparation of 3- Chloro- 1, 2, 4-trimethy1bicyclo [4, 2,0]octa- 1, 3, 5- triene- 7- one ........ 111 LITERATURE CITED ....................... 115 TABLE II. III . IV. VI. VII. VIII. IX. XI. LIST OF TABLES . Freezing Point Data ................... Stoichiometry with Respect to Hydrogen Chloride Ultraviolet Absorption Maxima of Some Polychloro Alkylbenzenes in Cyclohexane . ............. Ultraviolet Absorption Maxima of Some Trichloro- methyl Compounds in Cyclohexane ......... . Ultraviolet and Visible Spectra of Trichloromethyl Compounds in 100% Sulfuric Acid . . . ......... Visible Absorption Maxima of Tetraphenyl-p-xylylene- glycol in Varying Concentrations of Sulfuric Acid- Acetic Acid ............ . ........... Visible Absorption Maxima of Tetraphenyl-p-xylylene- glycol in Varying Concentrations of Sulfuric Acid- Water ........................... Visible AbsorptionMaxima of Tetra—p-anisyl-p- xylyleneglycol in Varying Concentrations of Sulfuric ACid-Acetic ACid oooooo o o o ooooooo o o o Visible Absorption Maxima of Tetra-p-anisyl-p— xylyleneglycol in Varying Concentrations of Sulfuric Acid-Water ...................... . Visible Absorption Maxima of 9, 10-Dihydro-9, 10- dihydroxy-9, 10-dipheny1anthracene in Varying Concen- trations of Sulfuric Acid-Acetic Acid . ......... Visible AbsorptionMaxima of 9, lO-Dihydro-9, 10- dihydroxy-9, lO-diphenylanthracene in Varying Concen- trations of Sulfuric Acid-Water ............. vii Page 9 10 12 18 19 53 56 59 62 64 66 LIST OF TABLES - Continued TABLE Page XII. Ultraviolet-Visible Absorption Maxima of 4—Pheny1- 2, 3-benzof1uoranthene in Absolute Ethanol ....... 77 XIII. Freezing Point Data of Tetraphenyl-p-xylyleneglycol. . 81 XIV. Relative Amounts of Products Formed From the Reaction of a Solution of Tetraphenyl-p-xylyleneglycol in 2. 52 Weight Per Cent Sulfuric Acid in Acetic Acid with Absolute Methanol ................ . 90 XV. Test of Beer's Law for Tetraphenyl-p-xy1yleneg1ycol in 2. 52% Sulfuric Acid-Acetic Acid ........... 105 viii LIST OF FIGURES FIGURE Page I. Comparison of the ultraviolet spectra of 4-chloro- 2, 3, 5, 6-tetramethy1benzotrichloride, benzotrichloride and perchlorotoluene in cyclohexane ........... 14 II. Comparison of the ultraviolet spectra of 4-«halou-Z, 3, 5, 6- tetramethylbenzotrichlorides and trichloromethylpenta- methylbenzene in cyclohexane . . ............. 15 III. Comparison of the ultraviolet-visible spectra of 4—halo- 2, 3, 5, 6—tetramethy1benzotrichlorides and trichloro- methylpentamethylbenzene in 100% sulfuric acid. . . . . 16 IV. Proton magnetic resonance spectra of 4-halo-2, 3, 5, 6- tetramethylbenzotrichlorides in carbon tetrachloride . . 20 V. Proton magnetic resonance spectra of 4-halo-2, 3, 5, 6- tetramethylbenzotrichlorides in 100% sulfuric acid. . . 21 VI. Infrared spectrum of bromodurene ......... . . 25 VII. Infrared spectrum of 4~bromo-2, 3, 5, 6-tetramethy1- benzotrichloride . . . . . . . ............. . 27 VIII. Infrared spectrum of 4—bromodurenecarboxylic acid . . 29 IX. Infrared spectrum of Chlorodurene ............ 31 X. Infrared spectrum of 4-chloro-2, 3, 5, 6-tetramethy1- benzotrichloride .......... . ......... . 33 XI. Infrared spectrum of 4-chlorodurenecarboxylic acid . . 35 XII. Infrared spectrum of fluorodurene ............ 39 XIII. Infrared spectrum of 4-fluoro-2, 3, 5, 6-tetramethyl- benzotrichloride . . ............ . . . . . . . 41 ‘ XIV. Infrared spectrum of 4-f1uorodurenecarboxy1ic acid . . 42 ix LIST OF FIGURES - Continued FIGURE XV. XVI. XVII. XVIII. XIX. XX. XXI. XXII. XXIII. ' XXIV. XXV. XXVI. XXVII.’ XXVIII. Visible spectra of tetraphenyl-p-xyly1eneg1ycol in vary- ing concentrations of sulfuric acid—acetic acid ..... Visible spectra of tetraphenyl-p-xyly1eneglycol in vary- ing concentrations of sulfuric acid-water ........ Visible spectra of tetraphenyl-p-xy1yleneg1yccl in vary- ing concentrations of sulfuric acid-water ........ Visible spectra of tetra-p-anisyl-p-xylyleneglycol in varying concentrations of sulfuric acid—acetic acid. . . Visible spectra of tetra-p-anisyl-p-xylyleneglycol in varying concentrations of sulfuric acid—water ...... Visible Spectra of tetra-p-anisyl-p-xylyleneglycol in varying concentrations of sulfuric acid-water . . . . Visible spectra of 9, 10-dihydro-9, 10-dihydroxy-9, 10- diphenylanthracene in varying concentrations of sulfuric acid-acetic acid .................... . . Visible spectra of 9, 10ndihydro—9, 10-dihydroxy-9, 10- diphenylanthracene in varying concentrations of sulfuric aCid-Water 00000000000000000000 o o o o 0 Infrared spectrum of tetraphenyl-p-xylyleneglycol . . . Infrared spectrum of the dimethylether of tetraphenyl- p—xylyleneglycol. ..................... Infrared spectrum of the monomethylether of tetra- phenyl-p-xylyleneglycol . . . . . ............. Infrared spectrum of tetra-p-anisyl-p-xylyleneglycol . Infrared spectrum of 9, lO-dihydrou9, lO-dihydroxy- 9, 10-dipheny1anthrac ene ................ . Infrared spectrum of 9, lO—diphenylanthracene . . . . . Page 52 54 55 58 60 61 63 65 84 87 89 93 96 97 ' LIST OF FIGURES - Continued FIGURE XXIX. XXX .- XXXI. XXXII.’ XXXIII. Ultraviolet-visible spectrum of 4-pheny1-2, 3~benzo= fluoranthene ............ . .......... Infrared spectrum of 4-phenyl-2, 3-benzofluoranthene . Infrared spectrum of the polar residue from the. hydrolysis of the blue sulfuric acid solution ..... Infrared spectrum of 3, 7, 7-trichloro-l, 2, 4~trimethyl- bicyclo[4, 2, 0]octa-1, 3, 5-triene ........... Infrared spectrum of 3-chloro-1, 2, 4-trimethy1bicyclo— [4, 2, 0]octa-1, 3, 5-triene-7-one ............. xi Page 99 100 101 112 114 INTRODUCTION It was recently observed (1) that trichloromethylpentamethyl- benzene (1) dissolved in 100% sulfuric acid to give a deep red stable solution. The color was attributed to the dipositively charged penta- methylphenylchlorodicarbonium ion (Ia), formed by ionization of I according to the equation: ’ + O CC13+2HZSO4-——-> C C-C1+2HC1+2HSO4" (1) I Ia The cogent evidence included cryoscopic measurements, which established the production of five particles in solution per mole of solute. Two of these were hydrogen chloride, which could be rapidly and quanti- tatively swept from the solution. After sweeping, the remaining solu- tion showed as expected a three-fold molal freezing point depression. Conductance measurements supported the formation of two bisulfate ions. Hydrolysis of the red solutions produced pentamethylbenzoic acid quantitatively. The visible, ultraviolet, and nuclear magnetic resonance spectra of these solutions also support the structure la. The ionization of trichloromethylpentamethylbenzene to the dipositive carbonium ion was rapid and the ion itself was quite stable. It was one intent of the present work to examine the sc0pe of dipositive carbonium ion formation from trichloromethylbenzenes. Trichloromethylpentamethylbenzene had been prepared by the reaction of pentamethylbenzene with carbon tetrachloride and aluminum chloride (1). Under similar conditions rearrangement occurred with durene and the product formed was trichloromethylprehnitene II (2). A O +cc14 ——1C—13—> O CC13 (2) II It was thought that one might avoid the rearrangement and at the same time examine structural influences on dicarbonium ion formation by first preparing monosubstituted durenes and then carrying out the trichloromethylation reaction. A1c13 ——-—> x + cc14 x CC13 (3) The preparation of several such compounds and their reaction with 100% sulfuric acid is reported in the first part of this thesis. Previous work on dicarbonium ion chemistry (1, 2) centered about the formation of a dication by multiple ionization at a single carbon atom. One can also conceive of dipositive carbonium ions derived from single ionizations at two separate sites in a molecule. One example might be the ionization of tetraphenyl-p-xylylene glycol, (I36H5 C|36Hs C6H5 - (I: (I: - C6115 O O H H which by analogy with the well-known ionization of triphenylcarbinol in sulfuric acid to triphenylcarbonium ion might be expected to produce the analogous dication. If as would seem likely the ionization of such a glycol were a stepwise process, going first to a monocarbonium ion and then to the dicarbonium ion, one might perhaps measure the separate pK's and determine how much energy is required to introduce the second charge. One can visualize other interesting problems posed by systems of this type, and some work on their chemistry will be presented here. This thesis will be presented in two separate sections, Part A dealing with experiments on the monosubstituted trichloromethyldurenes and Part B on the glycol systems. PART A DIPOSITIVE CARBONIUM IONS ARISING FROM MULTIPLE IONIZATIONS AT A SINGLE SITE IN A MOLECULE RESU LTS AND DISCUSSION The T richloromethylation Reaction The reaction of benzene with carbon tetrachloride in the presence of aluminum chloride has been known since 1877 (3). The main products from this reaction are diphenyldichloromethane and triphenylchloro- methane or the corresponding hydrocarbons, depending on the reaction conditions (4, 5). Toluene and the xylenes behave similarly (6, 7, 8, 9). Application of this reaction to mesitylene, durene or pentamethyl- benzene (see references 1 and 2) gave trichloromethyl derivatives with- out the formation of condensation products. But in the case of durene rearrangement occurred and the only product isolated was trichloro- methylprehnitene. ' Isodurene when treated in a similar manner also gave trichloromethylprehnitene (10). There is precedent for such rearrangements in the work of Baddely and Pendleton (11) who have shown that ac etyldurene when heated to 1000 with an excess of aluminum chloride yielded a mixture of products, 80% of which was acetylprehnitene. There was no reaction between acetylprehnitene and excess aluminum chloride when refluxed at 1500 for three hours. Therefore acetylprehnitene was assumed to be the most thermodynamically stable product. The Jacobsen rearrangement (12) is an analogous reaction. When durene was treated with sulfuric acid a 70% yield of prehnitenesulfonic acid was obtained. ‘ Isodurene gave a 50% yield of the same acid and in the case of prehnitene itself only sulfonation occurred without re- arrangement. This again implies that the most stable product is formed. In each case the methyls migrate in order to have only one methyl group ortho to the bulky substituent, whether it be trichloro- methyl, acet'yl or sulfonic acid. This is strong evidence for steric strain which can be relieved by migration of one of the methyl groups. In the reaction of monosubstituted durene derivatives with carbon tetrachloride no rearrangement occurred. 4-Bromow, 4~chloro-—, and 4-fluoro-2, 3, 5, 6-tetramethylbenzotrichlorides were prepared. A O + CC14 —1—C—1—3—> x 0 col, (4) IV X = Br IVa = C1 IVb = F IVc Since these compounds were new, it was necessary to unequivocally establish their structure. The infrared spectra of the three compounds differed only in their carbon-halogen frequencies and the ultraviolet spectra were quite similar (Figure II). They were readily hydrolyzed to the corresponding monosubstituted durenecarboxylic acids by reflux- ing with aqueous acetone. The infrared spectra of the acids differed only in their carbon-halogen frequencies and lead one to conclude that the trichloromethyl and acid derivatives had the same structural features. p-Bromodurenecarboxylic acid was prepared by an independent synthesis. Dibromodurene was converted to the mono Grignard reagent and then carbonated. The acid obtained by this method was identical in all respects (neutralization equivalent,. m.p. , m.m.p. , and infrared spectrum) to the acid obtained from the hydrolysis of IVa. This con- clusively proves there was no rearrangement during the trichloro- methylation reaction of bromodurene. By analogy IVb and c were assumed to have the structures indicated. Attempts were also made to synthesize 4-iodo- and 4-nitro-2, 3, 5, 6— tetramethylbenzotrichloride. In the former case iodine was obviously I liberated and only a brown solid that remains unidentified was isolated. Apparently the carbon iodine bond is too labile to resist attack by the aluminum chloride catalyst. In the case of nitrodurene only starting material was recovered and it appears the compound is too deactivated for the Friedel-Crafts reaction to take place. Reaction of 4-Halo-2, 3, 5, 6-tetramethylbenzotrichlorides with 100% Sulfuric Acid Each of the 4-halo-2, 3, 5, 6-tetramethylbenzotrichlorides, when added to 100% sulfuric acid, dissolved with the formation of a deep red solution. Hydrogen chloride was evolved and the red color persisted for long periods of time if the solutions were kept anhydrous. When the red solutions were hydrolyzed on ice a better than 90% yield of the corres- ponding 4-halodurenecarboxylic acids was obtained. The acids were identical (m.p. and infrared spectra) with those obtained when the corres- ponding trichloromethyldurenes were hydrolyzed by refluxing with aqueous acetone. In the light of previous work, it is reasonable that the species responsible for the red color observed when IVa, b, or c is dissolved in 100% sulfuric acid are the corresponding dicarbonium ions (V). + x O CC13 + 2 sto,—> x Czc --01+ 2 HC1+ 2 Hso,‘ (5) IVa,b,c Va,b,c Because of the available analogy, the evidence for V need not be as rigorous as that presented in earlier work (1, 2). Nevertheless, the cryosc0pic, stoi'chiometric and spectroscopic measurements to be described seem sufficient to firmly establish the ionization depicted in equation 5. Cryoscopic Measurements The results from freezing point measurements are given in Table I. - Following Gillespie's conventions (12), 2 is the number of moles of particles (molecules or ions) produced in solution by one mole of solute, and _i_ is the observed multiple of the molal freezing point depression. . To become familiar with the technique, preliminary measurements were made on triphenylcarbinol, trichloromethylprehnitene, and trichloromethylpentamethylbenzene where the 7) -values were already known. Slightly aqueous sulfuric acid with freezing point range 9. 820 to 10. 1250 was used in order to suppress the selfndissociation of the sulfuric acid. The data in Table I show that 4-bromo-2, 3, 5, 6-tetra- methylbenzotrichloride (IVa) produces 5 particles when it dissolves in 100% sulfuric acid. ~ It did not seem necessary, in view of this and subse- : quent results, to also run the 4-chloro and 4-fluoro compounds. Stoichiometry One obvious product of the reaction of trichlorOmethyl compounds with 100% sulfuric acid is hydrogen chloride. Using a suitable apparatus, this gas can readily be swept from the red solutions and trapped in aqueous sodium hydroxide. The latter was then acidified and titrated for chloride ion by the Fajans' method. The results from these experiments are presented in Table II. ' It is clear that in each case two moles of hydrogen chloride were produced per mole of trichloromethyl compound. In each case prolonged (12-24 hours) sweeping produced only small additional amounts of hydrogen chloride which may be attributed to the slow hydrolysis of the dicarbonium ion due to trace amounts of moisture. In one experiment 4-fluoro-2, 3, 5, 6-tetramethylbenzotrichloride was dissolved in sulfuric acid that contained an excess (approximately 0. 5 molar) of sulfur trioxide. After sweeping for 26 hours only 26. 59% of .o>fl.mH:§So won???“ can find “cc/flawed 6.3m mogg HQ oau ”magnum pwom 0:316 ofimgm .m 8. 330m mo maoflfippm o>fimmooudm ohm omoAH U ..§o>floommon co .m paw mm .«c .mw .w o» commune“: 639/ may» £390: 0 new “v .N you?» .oudHOm 65. mo 5.330508? ma mg 6.833 55 NH .o\.H.< a 50.3 pofimfioofimOM 02.m mfio.2 mmcm.o 2V .4 me... .0 m2 .2 8 . 8 52 .0 328223853 om .m :m .o N2. .o mm is 32 .o 2.33823. 2 .m mom .0 ammo... .o -c .m .m £58854 m MN .m mom .0 $29 $.02 22... .o :oocozmmorm oo.m wd>.o coozm.o do .m New .0 dos .d N .3 32 .o 320223023 2 .m 220 uwcmmd 2338239....“ch m 2 .m com .0 mac .2 m .32 23. .c roomcozmmov nom.¢ 92w.o ccohm.o md.¢ Hom.o coocm.o 8.4 some $5 2.3 23.0 2 .m 23 .o 0E8 .o 2833 1:38 mm .m m 2 c .o cvwmm .o -Sccfisficfiocoficfls m S .m mom .0 mo .0 m .26 $3 .0 ”80.02628 2 :c 22 .N ceooo .2 28332332; d S .d 322 So .2 o .2. News .0 mooxmmsov flN cfl Do .eq 00 .5. 16%: .m .m :35 398.50 mama ucwom mnfiuoouh .H 2038 mudoa NH Hofluocm no.“ waaooBm Houmda .LO 6330306 on poodponm mammoospm mo mudofi w Hmcofifippm G4. .opwxofinu usmfidm A: .338 m .o Sfioumgwxonmmm pwom owhzfismm .nHO mo mofiog aw .N mo H133 m o» maflocommouhoo «53.330 LO mo .wE m .NN m .mudoa 2 am HO mo moHoE N: .N mo 133 m mnfifim LO mo .mg m .h poodpoum wcaoogm mo musoa Hmcoflflopm mo .LO 6336266 o: 92% waaoospm mo mudog Hmcowfippm mp .28: 3 cc -5 Ho 338 cm .N mo Havoc. m o» mpcommonuoo mEH. .LU mo .m5 042 Hmaofifippm cm poosvonm mndog @905 : MOM mcaoospmo 10 .623. no .EFGAE 3.2m .... SE Lon 63.53,. 30G mo moumu was “GOHHSHOm 6:» Amdofifi amok/m ammonia 22$ 6&3 mo Auwcodm mmm .o mm .2 m3 moom .o 6532.335 «do; o¢.mv mw mmom .o nonoEoEuonoghrm omo.N mm .Nv om mmomd oGoudflTfifioE god mode we «$310 rouoEofiuuonoEUua mmod melom N. mmom.o ocoufip oHo .N ow .om o¢ NM H m . o racfiofiouogownuoflsnmrm pcsomfiou TxbuoEOHoEoEu mo oHoE\LO moHoSH nnOZm< mo .32 ofmpcs 68TH. Amv .35 BGSOQEOO opcodfiu Gowofiocnm o» poommom 8:3 Knhuogofiflofiohm .HH QEMP 11 the theoretical amount of hydrogen chloride had been obtained. This was significant since in the previous experiments the hydrogen chloride was rather rapidly liberated. Evidently the free sulfur trioxide reacts rapidly with the hydrogen chloride to form chlorosulfonic acid. Reversal of this equilibrium and recovery of the hydrogen chloride is obviously very slow. Whereas the reaction of hydrogen chloride with sulfur trioxide is apparently rapid, reaction with sulfuric acid molecules which could also lead to chlorosulfonic acid, must be slow and incomplete. Ultraviolet and Visible Spectra Benzene absorbs in two main regions of the ultraviolet, one of high intensity near 198 mu (6 ca. 8, 000) and called the K-band’and the other of much lower intensity between 230 mu and 270 mu (6 ca. 230) called the B-band. These bands are characteristic of all benzene derivatives. The K- band arises from conjugation in the aromatic ring whereas the B-bands are ascribed to changes in the symmetry properties of the pi-electron system (13). - Substitution on the aromatic nucleus causes these bands to shift and vary in intensity due to conjugation and steric effects. . Recently Ballester and Casta'fier (14) reported the ultraviolet spectra of some polychlorinated alkylbenzenes. ~ In Table III the molar absorbancy indices and the maxima for some of these compounds are reported. The spectrum of A is referred to as normal since it has the essential features found in other benzene derivatives. The spectra of B, C, D, and E are abnormal compared with A because they exhibit large bathochromic shifts. ' In each "abnormal" compound at least one trichloromethyl group is flanked by two ortho chlorine atoms. Presumably substances having such structures are sterically strained. (Note that a trichloromethyl is nearly as large as a t—butyl group.) This strain raises the energy of both the ground and excited states of the Table III. Ultraviolet Absorption Maxima of Some Polychloro Alkylbenzenes in Cyclohexane Wavelength Molar Absorbancy Compound (mu) Index (6) C1 C1 298 390 C1 289 370 C1 C1 219 21, 500 Pentachlorobenzene (A) C1 C1 CC13 316 2, 100 C1 C1 221 30, 000 2, 3, 5, 6-tetrachloro-l- trichloromethylbenzene (B) C1 C1 C1 CC13 319 1, 420 C1 C1 223 35, 000 Perchlorotoluene (C) C C1 C1 CZC15 319 2,050 C1 C1 224 41, 000 Perchloroethylbenzene (D) C1 C1 365 3, 000 Cl3&{ CC13 280 6, 600 C1 C1 236 25, 500 Perchloro-p-xylene (E) 219 27, 000 l3 molecule. However, due to the greater anti-bonding character of the excited states their energies are less effected by distortion. Consequently, it can be expected that the ultraviolet spectra would show bathochromic shifts. Since the sizes of a chlorine atom and a methyl group are about the same one would expect the ultraviolet spectra of monosubstituted trichloromethyldurene derivatives to be quite similar to those of the "abnormal" polychlorinated alkylbenzenes. . Figure I shows a comparison between the spectra of benzotrichloride, perchlorotoluene and 4-chloro- 2, 3, 5, 6-tetramethylbenzotrichloride. One can easily see the displace- ment from a normal benzene derivative in the spectra of the latter two compounds. Similar steric strain is also suggested for these compounds. Figure 11 shows a comparison of the ultraviolet Spectra of trichloro- methylpentamethylbenzene and the 4-halo-2, 3, 5, 6-tetramethylbenzotri- chlorides. The curves are almost identical and suggest very common structural features. The maxima for the 4-bromo, 4-chloro, and 4-methyl derivatives occur at almost identical wavelength with almost equal intensity. This is expected due to very similar resonance and steric contributions of these substituents. - However, in the case of the 4-f1uoro group both maxima undergo a hypsochromic shift with a decrease in intensity. This can be attributed to release of steric strain in the ground state since the fluorine atom is considerably smaller than the other 4-position substituents. Figure III shows a comparison of the ultraviolet and visible spectra of some trichloromethyl derivatives in 100% sulfuric acid. The great difference between the spectra of the parent compounds and the corres- ponding dicarbonium ions is due to the effect of introducing a permanent charge into the chromOphoric system giving rise to charge-resonance spectra (15). Sufficient information is not available to discuss the shifts in wavelength and increased intensity of the maxima. Since the spectra 3. Log 6 2.0 1.0 ,. 14 200 Figure I. . Comparison of the ultraviolet Spectra of 4-Chloro-2, 3, 5, 6-tetramethylbenzotrichloride (———), benzotrichloride (- - . . ), and perchlorotoluene (----) in cyclohexane. coco-coco... y(o.. \ / ml J 250 300 356 Wavelength (mu) 15 .' \ ’. \ .. .\ .\\ .‘ \ ..\.\ 4.0 .. ‘\ -‘ \ ’0‘. \ t.‘\.\\_._, \ 000.. .. \ “'4 ,\ \\ O... \ \ 0.. \\ , / MT Kits“... Log 6 Figure II. . Comparison of the ultraviolet spectra of trichlo romethylpentamethylbenzene ( ). 4-bromo-2, 3, 5, 6-tetramethylbenzotrichloride- (----), 4-chloro-2, 3, 5, 6-tetramethy1benzo- trichloride (-- -- -), and 4-fluoro-2, 3, 5, 6-tetra- methylbenzotrichloride ( ----- ) in cyclohexane. h 1 200 250 Wavelength (mu) 350 050A: 5 3592380 Tfifiofiouozofiuu Show .05 930on manwmwcwruoHOfiEHfiS 33 mo acmwnmmEoO . .Hfioudmwm oom .Eom 03933.. .N .m 3 80q 16 38v ,fimcofioxfim? omo ooe 0mm oom . 0mmV cow 0mm oom 0mm / _ _ L _ A _ _ _ _ / / II o .N 0 / 0. / r. / 400.0 / .UI. II b Ii m /.... x + “wk. / 0?. / /... . / 1 c ./o 0 0 / Q‘o 0 0/000 / I 0 . 0 \// to I'll 1' I I 000 00 / Juqfllll 0 / l/ .0\. .00, l 0 O 0 00/. // 000 \o 000 0 04/000 .U'no‘ Q 000/ l/ 000 \O \ 00V’0O0 I00 )I . 000 / 000 0 \ {\0 00000 /0 I — Q U 000 / // 00 \ \ 00 0 0000 0'” . 000 0/ /’000 \o \ ’o\/0 0K , .0 0’0 000 / 0\ \ ’0 O..\Oe 17 of the monosubstituted trichloromethyldurene derivatives in 100% sulfuric acid are quite similar to the spectrum of trichloromethylpenta- methylbenzene it can be concluded that the dipositive ions arising from these species have nearly identical charge distribution and structural features. The molar absorbancy indices of the maxima and minima absorption peaks are given in Tables IV and V. Proton Magnetic Resonance Spectra Figure IV shows a comparison of the proton magnetic resonance spectra of the 4-bromo-, 4-chloro-, and 4-fluoro-2, 3, 5, 6-tetramethyl- benzotrichlorides in carbon tetrachloride. Figure V shows a comparison of the corresponding dipositive carbonium ions in 100% sulfuric acid. All peaks have been assigned ’7" - values (16) relative to tetramethylsilane (”Tl value = 10.00) as an internal standard. The spectra were determined at 60 Mc. The Spectra in carbon tetrachloride are very similar. All Show two peaks with relative areas 1:1. The low field peaks are assigned to the methyl groups ortho to the trichloromethyl group. . The high field peaks are assigned to the methyls meta to the trichloromethyl group because of the shift in’l" -values observed for the different halogen substituents. Whereas the low field resonance remains constant at about 7.43 ’r/ , successive replacement of the bromine atom with chlorine and fluor- ine causes the higher field resonance to shift from 7.59 ’T/ to 7.641/ to a Split peak at 7.81 ’7’ and 7. 86 ’7’ respectively. . The largest shift was observed for the methyls ortho to the fluorine atom and is probably due to the higher electronegativity of this atom. The high field peak of the 4—f1uoro substituent has been split due to coupling with the fluorine atom . 18 Table IV. Ultraviolet Absorption of Some Trichloromethyl Compounds in Cyclohexane Compound Wavelength Molar Absorbancy (mu) Index (6) Trichloromethylpentamethyl 302 (max.) 1, 560 benzene 283 (min.) 920 250 (max.) 7, 680 239 (min.) 6, 440 4-Bromo-2,3,5,6- 303 (max.) 1,611 tetramethylbenzotrichloride 284 (min.) 940 254 (max.) 10,320 239 (min.) 7, 390 4-Chloro-2, 3,5,6- 301 (max.) 1,503 tetramethylbenzotriChloride 282 (min.) 831 251 (max.) 8, 980 238 (min.) 6, 730 4-Fluoro-2,3,5,6- 291 (max.) 853 tetramethylbenzotrichloride 276 (min.) 650 246 (max.) 6,150 4-Chloro-2, 3, 5, 6-tetramethyl- 700 (min.) 41 benzotrichloride 520 (shoulder) 2, 095 406 (max.) 24, 550 399 (shoulder) 22, 100 330 (min.) 1, 270 299 (max.) 5, 730 283 (min.) 4,170 260 (max.) 7, 500 253 (min.) 7, 220 226 (max.) 11, 000 4-Fluoro-2, 3, 5, 6-tetramethy1- 670 (min.) 44 benzotrichloride 500 (shoulder) 1, 775 380 (max.) 18, 250 313 (min.) 1,370 282 (max.) 4, 950 270 (min.) 4, 000 244 (max.) 9, 200 236 (min. ) 8, 800 19 Table V. Ultraviolet and Visible Spectra of Trichloromethyl Compounds in 100% Sulfuric Acid = 1 = Compound Wavelength Molar Absorbancy (mp) Index (6 Trichloromethylpenta- 710 (min.) 260 methylbenzene 545 (max.) 2, 370 485 (min.) 1, 550 393 (max.) 34, 600 387 (min.) 32,200 385 (max.) 32, 600 327 (min.) 1,000 265 (max.) 7, 300 250 (min.) 5, 820 235 (max.) 9,160 4-Bromo-2, 3, 5-6-tetramethy1- 699 (min.) 123 benzotrichloride 530 (shoulder) 2, 130 419 (max.) 27,150 405 (shoulder) 21, 300 340 (min.) 1, 305 310 (max.) 4, 800 289 (min.) 2, 630 262 (max.) 7, 820 255 (min.) 7,615 243 (max.) 8, 150 232 (min.) 7, 720 4-Chloro-2, 3, 5, 6-tetramethy1- 700 (min.) 41 benzotrichloride 520 (shoulder) 2, 095 406 (max.) 24, 550 399 (shoulder) 22,100 330 (min.) 1, 270 299 (max.) 5, 730 283 (min.) 4,170 260 (max.) 7, 500 253 (min.) 7, 220 226 (max.) 11, 000 4-Fluoro-2, 3, 5, 6-tetramethy1- 670 (min.) 44 benzotrichloride 500 (shoulder) 1, 775 380 (max.) 18, 250 313 (min.) 1, 370 282 (max.) 4, 950 270 (min.) 4, 000 244 (max.) 9, 200 236 (min.) 8, 800 20 (A) ,. J LW 0. L ! If I Ho—b 7.43 7.59 4“ 10.00 (B) l J I Ho—"'" 7.43 7.64 ’1’ 10.00 (C) 1 II V I H,—> 7.45 7.817.86 I)“ 10.00 Figure IV. Proton magnetic resonance spectra in CC14. (A) 4-Bromo-2, 3, 5, 6-tetramethylbenzotrichloride. (B) 4-Chloro-2, 3, 5, 6-tetramethylbenzotrichloride. (C) 4-Fluoro-2, 3, 5, 6-tetramethylbenzotrichloride. 21 (A) _ 1 ’71 1 Ho—’ f' 7) 72 2(7))96 IO. 00 _ x I“, [HI | Thai) ¢ (B) _ I L ’ l v ‘— Ho_§ {7.76 (7.94 10.00 i ’T' (C) y 1 u 1“" 1“ ~ * Ho—D {(72.83 ‘8,,21.\8‘,25 10.00 r1 FigureV. Proton magnetic resonance spectra in 100% sulfuric acid. (A) 4-Chloro-2, 3, 5, 6-tetramethylbenzotrichloride. (B) 4-Bromo-2, 3, 5, 6-tetramethylbenzotrichloride. (C) 4-Fluoro-2, 3, 5, 6-tetramethylbenzotrichloride. 22 The spectra of the three ions V a, b, and c are consistent with their structures. Again only two peaks appear in each spectrum with relative areas of 1:1 and the high field peak of V c has been split by the fluorine atom. As was observed in the parent compounds the down field methyl groups of the three ions have very similar/Tl -values, whereas the ’T’ -values for the high field methyl groups Shift depending on the halogen substituent. EXPERIMENTAL 23 24 A. Syntheses and Reactions in 100% Sulfuric Acid Pr eparation of Bromodurene Bromodurene was prepared by the method of Smith and Moyle (17). In a 1-1. 3-necked round-bottomed flask provided with a dropping funnel, stirrer, and exit tube was placed 100 g.‘ (0.746 mole) of durene and 200 g. of carbon tetrachloride. A crystal of iodine was added and the flask was cooled in an ice bath. A solution of 124 g. (0. 775 mole) of bromine in 120 ml. of carbon tetrachloride was added dropwise over a 3 hour period, after which the mixture was stirred at room temperature for an additional hour. The organic layer was separated and washed with 100 m1. of 5% sodium hydroxide solution, then with 100 ml. of water. The solvent was removed on a Rinco rotary evaporator and the solid was refluxed for one hour with a solution of 8 g. of sodium in 200 ml. of ethanol and then allowed to stand overnight. Water (1800 ml.) was added and the solid filtered and subjected to a 7 hour steam distillation. The solid in the distillate was filtered and recrystallized once from 95% ethanol to yield 105 g. (66%) of white needles of bromodurene, m.p. 57-600. Preparation of 4—bromo-2, 3, 5, 6-tetramethylbenzotrichloride (l) In a 1-1. 3-necked flask equipped with a Trubore stirrer, ther- mometer, reflux condenser and dropping funnel was placed 27 g. (0. 2 mole) of powdered anhydrous aluminum chloride and 100 ml. of carbon tetra- chloride previously dried over calcium chloride and distilled. The slurry was stirred and heated between 37-420 while a solution of 20 g. (0.094 mole) of bromodurene in 100 ml. of carbon tetrachloride was added dropwise (two hours). After the addition, heating and stirring were continued for another two hours. The deep purple complex was slowly added with vigorous stirring to a mixture of 100 g. of ice, 10 ml. of concentrated hydrochloric acid and 25 mg .oaoudpofionn mo 525on poumuwcH .H> ouflmfim Annonoflhv Auwnofiocrm? . 3 o m N. o m A _ _ i _ . 26 150 ml. of carbon tetrachloride. (The hydrolysis of the aluminum chloride complex proceeds slowly.) The carbon tetrachloride layer was separated and the aqueous layer extracted with two 100~m1. portions of carbon tetrachloride. The combined organic layers were washed with 150 m1. of water, two 200-ml. portions of 5% sodium carbonate solution, and finally with another 100 ml. of water. The solution was dried overnight with calcium chloride and the solvent removed on a Rinco rotary evaporator. . The brown residue was dissolved in anhydrous n-pentane and treated with Norite. This procedure was repeated until 23 g. (75%) of pure 4-bromo-2, 3, 5, 6-tetramethylbenzotrichloride, m.p. 83. 5-84. 50, was obtained. fl}: Calcd. for CnleBrCl3: C, 39.98; H, 3.66; Br, 24.18; C1, 32.18, Found: C, 40.09; H, 3.78; Br, 24.37; Cl, 32.02. Preparation of 4-bromo-2, 3, 5, 6-tetramethylbenzoic acid The acid was prepared by a modification of the method described by Newman and Lloyd (18) for the preparation of durenecarboxylic acid. In a 1-1. 3-necked flask fitted with a dropping funnel, Trubore stirrer and a reflux condenser with a drying tube was placed 2. 5 g. (0. 10 g. - atom) of magnesium. A solution of 4. 6 g. of dibromodurene (17) and 1.45 g. of ethyl bromide in 180 m1. of ether was added to the flask and the reaction started immediately. It was maintained for three hours by the slow addition of 10.0 g. (0. 05 mole total) of dibromodurene and 4. 0 g. (0. 05 mole total) of ethyl bromide in 250 ml. of ether. After the addition was completed the reaction flask was warmed gently and most of the ether was removed. Three hundred m1. of anhydrous benzene was * All analyses by Spang Microanalytical Laboratory, Ann Arbor, - Michigan. ' Melting points are uncorrected unless otherwise indicated. 27 .303033.30Naofigflwoamnuouno .m .m .Nuofiouniv mo 5.9.30on ponmnwcm . .HH> oyswwh Amcon 38V praofiocfim? NH 3 S o m N. o m w 1' D D D -) _ _b _ _ 4 _ _ 7 _ 28 added and the solution refluxed for another hour. . The reaction mixture was cooled in an ice bath and carbon dioxide gas (from the evaporation of dry ice in a suction flask) was bubbled into the mixture for 20 minutes. Water (200 ml.) was added and the aqueous layer separated, extracted with 100 ml. of ether to remove any organic material, and acidified with dilute hydrochloric acid. The white precipitate was filtered, washed with water and recrystallized twice from aqueous acetone to yield 4 g. (47. 3% based on 5.0 g. of recovered durene) of 4-bromo-2, 3, 5, 6- tetramethylbenzoic acid, m.p. 224—225°. Anal. Calcd. for C11H13Br02: C, 51.38; H, 5.10; Br, 31.08. Neutralization Equivalent: 257'. 14. Found: C, 51.63; H, 5.20; Br, 30.89. Neutralization Equivalent: 257. 1. Reaction of a Solution of 4-Bromo-2, 3, 5, 6-tetramethylbenzo- trichloride in 100% Sulfuric Acid with Water A sample of 0. 9318 g. of 4-bromo-2, 3, 5, 6-tetramethylbenzo- trichloride was dissolved in 5 m1. of 100% sulfuric acid. The deep red solution was slowly poured onto 50 g. of ice and a fluffy white precipitate was collected. . The solid was washed with water and recrystallized twice from aqueous acetone. The yield was 0.6690 g." (91. 7%) of 4-bromo- 2, 3, 5, 6-tetramethylbenzoic acid, m.p. 224-2250. Hydrolysis of 4-Bromo-2, 3, 5, 6-tetramethylbenzotrichloride A sample of 2 g. of 4-bromo-2, 3, 5, 6-tetramethylbenzotrichloride dissolved in 50 m1. of 75% aqueous acetone was refluxed overnight. The solvent was removed on a Rinco rotary evaporation and the solid recrystallized once from aqueous acetone to yield 1. 33 g. (85%) of 4-bromo—2, 3, 5, 6-tetramethylbenzoic acid, m.p. 223. 5-2250. The melting points and infrared spectra of the acids from all three experiments were identical. 29 .Trruv Hugo? Jaw? V ocoudpofiounfip 22> 03mg Bum 033:5u03nogufluonconfffiofimnaou-Aw .m 0m .NuOEOknuw mo coflomon 0:”— 89¢ pad A mo comuodon Unmcmiw 0:» god p80 emonconfnauofidnuou-o .m .m .Nuofiounrv m0 .950on voyeur: 38¢ aumfimfiokrm? w; 2 NA 3 OH «0 w w o m w m _ _ a L a _ _ , _ q- # _ ‘2'" 30 Preparation of Chlorodurene Chlorodurene was prepared by the method of Smith and Moyle (19). A solution of 72 g. (0. 5 mole) of durene in 200 g. of chloroform was chlorinated at 00C. by addition of 39 g. (0. 55 mole) of chlorine in 210 g, of chloroform during a 15 minute period. The product was steam dis- tilled and the oil in the distillate was separated. The aqueous layer was extracted with 200 ml. of chloroform and the combined organic layers were dried over calcium chloride for 12 hours. The solvent was removed and the remaining oil distilled at reduced pressure (20 mm). A small amount of durene (7 g.) came over first followed by the main fraction of Chlorodurene, b.p. 120-1280. The residue was mainly dichlorodurene. The crude product was recrystallized once from 95% ethanol and yielded 41 g. (50. 3% corrected for recovered durene) of white plates of chloro- durene, m. p. 46-480. Preparation of 4-Chloro-2, 3, 5, 6-tetramethylbenzotrichloride In a 1-1. 3-necked flask provided with a Trubore stirrer, ther- mometer, reflux condenser, and dr0pping funnel was placed 37. 2 g. (0. 28 mole) of powdered anhydrous aluminum chloride and 100 ml. of carbon tetrachloride (previously dried with calcium chloride and distilled). The slurry was stirred and warmed from 37--42o while a solution of 23. 6 g. (0. 14 mole) of Chlorodurene in 100 ml. of carbon tetrachloride was added dropwise over a two hour period. After the addition heating and stirring were continued for two more hours. The deep purple complex was hydrolyzed by Slowly adding it with vigorous stirring to a mixture of 100 g. of ice, 10 ml. of concentrated hydrochloric acid and 150 m1. of carbon tetrachloride. The carbon tetra- chloride layer was separated and the aqueous layer extracted with two 50-m1. portions of carbon tetrachloride. The combined organic layers were washed with 100 ml. of water, two 100-ml. portions of 5% sodium 31 «Q Ma ad .oconSpOHofifio mo 5.950on popduwcH .XH oufiwfim 0H Amcon 35v AumcoHQ/m? o 1. w F } q L -t ‘ _ 32 carbonate solution, and finally with another 100 ml. of water. The solu- tion was dried overnight with calcium chloride and the solvent removed on a Rinco rotary evaporator. The dark brown oily residue was dissolved in anhydrous n-pentane and treated with Norite. The pentane solution was filtered, cooled in dry ice and the brown precipitate filtered under a nitrogen atmosphere. After four recrystallizations from pentane 22 g. (56. 5%) of white plates of 4-chloro-2, 3, 5, 6-tetramethylbenzotrichloride was obtained, m.p. 72-730. A231. Calcd. for C11H12C14: C, 46.18; H, 4.23; C1, 49.58. Found: C, 46.49; H, 4.35; C1, 49.58. Reaction of 4-Chloro-2, 3, 5, 6~tetramethylbenzotrichloride in 100% Sulfuric Acid with Water A sample of 1.008 g. (0.0035 mole) of 4-chloro-2, 3, 5, 6-tetra- methylbenzotrichloride was dissolved in 20 g. of 100% sulfuric acid cooled to 100. The sample was stirred (magnetically) and maintained at 100 until it had completely dissolved (ca. 5 min). . The red solution was slowly added to 100 g. of ice and the white precipitate filtered, washed with water and recrystallized twice from aqueous acetone. The yield was 0. 6904 g.“ (92%) of white needles of 4-chloro-2, 3, 5, 6-tetramethylbenzoic acid, m.p. 206-2070. _A_n_ail_. Calcd. for C11H13OZC1: C, 62.12; H, 6.16; C1, 16.67. Neutralization Equivalent: 212. 68. Found: C, 62.16; H, 6.34; C1, 16.56. Neutralization Equivalent: 209. 8. Hydrolysis of 4-Chloro-2, 3, 5, 6~tetramethylbenzotrichloride A sample of 2.00 g. (0.007 mole) of 4-chlor02, 3, 5, 6-tetramethyl- benzotrichloride was dissolved in 50 ml. of 80% aqueous acetone. .ughofiflowpuoucofiasuogmnuourc .m .m .Nuouozonv mo 55.3.0on poumch .X ouswwh $25quch Aumaofiocfim? 2 NH 3 0H m w h o m 33 _ _ _ i E _ _ _ 4 L 34 The solution was refluxed for 12 hours and the solvent removed on a Rinco rotary evaporator. The white solid was recrystallized from aqueous acetone to yield 1. 34 g. (90%) of 4-chloro-2, 3, 5, 6-tetramethy1- benzoic acid, m.p. 206—2070. A mixed melting point with the acid from method I gave no depression. The infrared spectra of both acids were identical. Preparation of Dinitrodurene Dinitrodurene was prepared by the method of L. 1. Smith (20). A solution of 13.4 g. (0.1 mole) of durene in 100 ml. of chloroform was added to 75 ml. of concentrated sulfuric acid in an 800-m1. beaker pro- vided with a stirrer and thermometer. The mixture was cooled to 100 and 16 g. of fuming nitric acid was added drOpwise at such a rate that the temperature did not go over 500 (approximately 15 minutes). As soon as all of the acid had been added the mixture was poured into a separatory funnel, the sulfuric acid layer was removed, and the organic layer immediately run into 500 ml. of 10% sodium carbonate solution. - Six portions were nitrated and the combined chloroform solutions were washed twice with 500 m1. of 2. 5% sodium carbonate solution, dried overnight with calcium chloride, and the chloroform distilled until crystals appeared. At this point four times the volume of 95% ethanol (2 l.) was added and the solution cooled to 100. The solid was filtered and washed twice with 100 m1. of cold (100) 95% ethanol. The pale yellow crystals of dinitrodurene were not recrystallized, m.p. 205-2080. The yield was 118 g. (88%). Preparation of Aminonitrodurene (21) A typical experiment is described. Forty grams (0. 178 mole) of dinitrodurene and 1.1. of 95% ethanol were heated to reflux, and a solution 35 .711; .833 4.33 300 otfifismuoguozumuu noncongfiofimnuouuo .m .m .Nnonozolv mo :oflomou 02.3 893 0:0 A v 03.8EuwuuonsonaaaoEmnuouno .m .m .m 19820.4. mo 39:95.»: on» God 300 gondoagsuogmbouao .m .m .Nnouozouw mo 0300mm 00.1935 .an 0.2,.me Amcou 38V aumnoaohm? a; 2 NH 3 S o w i P _ _ w _ L A E 36 of sodium disulfide (prepared by warming 140 g. [0. 583 mole] of NazS- 9 H20 in 400 m1. of water with 18 g. [0. 563 g. -atom] of flowers of sulfur) was added slowly with stirring. ' After addition, the mixture was refluxed and stirred for 9-10 hours- The excess alcohol (about 700 ml.) was removed by distillation and the remaining solution poured into 2 1. of cold water. The yellow precipitate was filtered, washed with water, and heated with 1500 ml. of 2 N hydrochloric acid. The solution was filtered hot to remove any unreacted dinitrodurene or sulfur and the filtrate made basic with ammonium hydroxide. The orange solid was filtered, washed with water and dried. The experiment was repeated until 170 g. (0. 76 mole) of dinitrodurene had been reduced. The over-all yield was 133.4 g. (90.5%) of aminonitrodurene, m.p. 158.5-1600. The reported m.p. after recrystallization from 95% ethanol was 161-1620. The material was used without further purification. ‘ Preparation of 4-Fluoronitrodurene (22) A total of 130 g. (0. 671 mole) of aminonitrodurene was diazotized in small batches.. Aminonitrodurene (10 g.) was added to 100 ml. of 6 M sulfuric acid. The mixture was cooled to 00 in an ice-salt bath and a saturated solution of sodium nitrite was added dropwise while maintain- ing the temperature between 0-50. The diazotization was complete when a positive test for nitrous acid was observed with starch-potassium iodide paper. Approximately 55 ml. of 40% Ltetrafluoroboric acid was added to the clear solution and a pale yellow solid immediately pre- cipitated. Three batches were run simultaneously. The diazonium fluoroborate salt was filtered, washed with 30 m1. of cold water, 50 ml. of 95% ethanol, and 200 ml. of ether. The solid (188 g. , 96.4% based on 130 g. of aminonitrodurene) was dried at room temperature in a dessicator for 48 hours. 37 The diazonium fluoroborate was decomposed by heating (Bunsen burner) 30 g. portions in a 1-1. flask under reduced pressure. After 60 g. of material had been decomposed in this manner the remaining solid which was in a large crystallizing dish decomposed spontaneously with a voluminous evolution of boron trifluoride. The brown oily residue was dissolved in 1500 m1. of ether, washed with three 200~m1. portions of 5% sodium hydroxide solution and 200 ml. of water. The ether layer was separated and dried with Drierite for 24 hours. The solvent was evaporated and the brownish-yellow solid dissolved in 95% ethanol, treated with Norite, filtered, and recrystallized. The yield was 65 g. (49. 3%) of 4-fluoronitrodurene, m.p.935-95. 50 (Literature value 96-970). Preparation of 4-Fluoroaminodurene (22) Fluoronitrodurene (60 g. , 0. 305 mole) dissolved in 600 ml. of boiling acetic acid was reduced with 120 g. (1.01 g. -atoms) of tin and 600 ml. of concentrated hydrochloric acid. The solution was refluxed and stirred for 4 hours, made alkaline with concentrated sodium hydroxide solution, and steam distilled for 7 hours. The precipitate in the distillate was filtered, washed with water and dried. The yield was 34 g. (66. 5%) of light fluffy needles of 4-fluoroaminodurene, m.p. 98-990. - Reported m.p. 101-~1020 after recrystallization from 85% aqueous ethanol. Preparation of Fluorodurene (22, 23) To a 1-1. Erlenmeyer flask containing 450 m1. of water and 18 m1. of concentrated hydrochloric acid heated to boiling was added 22. 5 g. (0. 135 mole) of 4-fluoroaminodurene with magnetic stirring. After two minutes the hot mixture was rapidly cooled to 100 and 22. 5 m1. of concentrated hydrochloric acid was added. The mixture was cooled to 38 0-50 and diazotized with a saturated sodium nitrite solution until an excess of nitrous acid was present (indicated by potassium iodide-starch indicator paper). The reaction mixture was maintained at O«=-5o for another thirty minutes and then 214 m1. (2.025 moles) of ice-cold 50% hypophOSphorous acid was added. The flask was loosely stoppered and stored in a refrigerator at or near 00 for three days. , The pale yellow solid was filtered dissolved in 100 ml. of ether, washed with two 50-m1. portions of 10% sodium hydroxide solution, once with 50 m1. of water and dried over magnesium sulfate for 24 hours. The ether was evaporated and the solid residue (about 17 g.) was dissolved in 20 m1. of petroleum ether and adsorbed on 80 g. of Fisher Adsorption alumina (80-200 mesh). The column was eluted with 1200 ml. of petroleum ether and on evaporation of the solvent there remained 15 g. (73. 3%) of fluorodurene, m.p. 53.5-55.00. Preparation of 4-F1uoro-2, 3, 5, 6=tetramethylbenzotrichloride In a 500—ml. 3-necked flask fitted with a Trubore stirrer, thermometer, reflux condenser, and dropping funnel was placed 26. 6 g. (0. 2 mole) of powdered anhydrous aluminum chloride and 100 ml. of carbon tetrachloride previously dried over calcium chloride and distilled. The slurry was stirred and heated at 37—420 while a solution of 15. 2 g. (0. 10 mole) of fluorodurene in 100 ml. of carbon tetrachloride was added dropwise (two hours). After the addition heating and stirring were con- tinued for an additional two hours. The purple complex was slowly added with mechanical stirring to a mixture of 100 g. of ice, 10 ml. of concentrated hydrochloric acid and 150 ml. of carbon tetrachloride. The aqueous layer was separated and extracted with two 100-m1. portions of carbon tetrachloride. The com- bined organic layers were washed with two 150-ml. portions of 5% sodium carbonate solution and dried over magnesium sulfate. The solvent .mconsmoonosfl mo EGAN—00mm voyeuwcH . .HHN ouswwh Amaou 35V Auwnoaohm? S o m N. o ? 39 _ _ d _ _ 40 was removed on the Rinco rotary evaporator and the brown oil was dis— solved in anhydrous n-pentane and treated with Norite. This procedure was repeated until 20 g. (74.4%) of white plates of 4-fluoro-2, 3, 5, 6- tetramethylbenzotrichloride were obtained, m.p. 56-57. 50. A mixed melting of the product with fluorodurene (m.p. 54. 5-55. 50) gave a definite depres sion. Anal. CaICd0 for C11H12C13F: C, 49001; H, 40 49; C1, 3.9046; F, 7.05.. Found: c, 49.08; H, 4.64; Cl, 39.55; F, 7.04. Hydrolysis of 4-Fluoro-2, 3, 5, 6-tetramethylbenzotrichloride A sample of 2.00 g. (0.0075 mole) of 4-fluoro-2, 3, 5, 6-tetra- methylbenzotrichloride was dissolved in 50 ml. of 75% aqueous acetone. .The solution was refluxed for 6 hours and the solvent evaporated. The white solid was recrystallized once from aqueous acetone to yield 1.40 g. (96%) of white needles of 4-f1uoro-2, 3, 5, 6-tetramethylbenzoic acid, m.p. 168-169. 5°. Anal. Calcd. for C11H13FOZ: C, 67.33; H, 6.68; F, 9.68. Neutralization equivalent: 196. 2 Found: C, 67.49; H, 6.77; F, 9.60. ~ Neutralization equivalent: 198.1 Reaction of 4-Fluoro-2, 3, 5, 6-tetramethylbenzotrichloride in 100% Sulfuric Acid with Water A sample of 1.0159 g. (0.00376 mole) of 4-fluoro-2, 3, 5, 6-tetra- methylbenzotrichloride was dissolved in 10 ml. of 100% sulfuric acid. The red solution was slowly added to 100 g. of ice and a white precipitate was collected. The solid was recrystallized twice from aqueous acetone to yield 0. 7300 g. (99%) of white needles of 4-fluoro-2, 3, 5, 6-tetramethy1- benzoic acid, m.p. 168-169. 50. A m.m.p. with the acid from the 41 a; .03Hofiflowhuoucoflfwaioamuuouuo .m .m .NsOhodfirv mo 593009..“ 60.32%; Ammonowev flamnoaoewmg 2 NH 3 OH 0 m N. P p 1!) b P (P .2? 632m w _ _ _ 2 2 _ _ _ h —p 42 .Awuunv .3nt £53 300 ownsfismuoguogo uwuuouaonrfifionfimbouno .m .m .Nuonogmuv mo 2.53000.“ 9.3 50.3 95 A V 0.3.8EownuoncoflanU—05mfiouuo .m .m .N ionosmue mo mwmzoupcwg 05 503 “500 0wonconafio5030uuo .m .m .Nuonosfluv mo 0.3009... 00.332: 62an oufimfih chouuw5v 595305.03 E 2 NH : S o m N. c m 0 m D _ 2 a 2 _ a h . a _._ , _ _ _ _ _ 43 previous hydrolysis of 4-fluorocZ,3,5, 6-tetramethylbe‘nzotrichloride gave no depression. The infrared spectra of both acids were identical. Neutralization equivalent calcd. for C11H13FOZ: 196. 2. Found: 195. 6. Preparation of Acetoxymercuridurene (Z4) To a l-liter round—bottomed flask was added 72 g. (0. 537 mole) of durene, 360 m1. of methanol, 171 g. (0.537 mole) of mercuric acetate and 50 ml. of glacial acetic acid. The solution was refluxed for 5 days and then filtered hot to remove any diacetoxydurene formed. The filtrate was cooled and the white precipitate collected. The crude product was recrystallized from methanol to yield 87. 5 g. (41. 5%) of acetoxymercuri- durene, m.p. 158-1590. Preparation of Nitrodurene (25) Seventy-five grams (0. 191 mole) of acetoxyrnercuridurene was heated and shaken with 800 ml. of nitric acid (sp. gr. 1. 26). , The sub- stance was converted to a material which floated (presumably nitrosodurene) and nitrogen oxides were evolved. At 700, all of the nitrosodurene had dissolved within 10 minutes. The solution was kept warm 2 minutes longer, and then it was poured into 200 g. of ice and 200 m1. of water, and filtered. The yellow precipitate was triturated and washed with water and dried. The crude product was recrystallized from absolute ethanol and yielded 32. 5 g. (95%) of pale yellow prisms of nitrodurene, m.p. 111—112.50. I Attempted Preparation of 4-Nitro-Z, 3, 5, 6~tetramethyl- benzotrichloride In a l-l. 3-necked flask equipped with a Trubore stirrer, ther- mometer, reflux condenser, and dropping funnel was placed 39. 90 g. 44 (0. 30 mole) of powdered anhydrous aluminum chloride and 150 ml. of carbon tetrachloride. The slurry was stirred and heated between 37-420 while a solution of 26.85 g. (0.15 mole) of nitrodurene in 150 m1. of carbon tetrachloride was added dropwise (two hours). After the addition, heating and stirring were continued for another two hours. The light purple complex was slowly added with vigorous stirring to a mixture of 150 g. of ice, 20 m1. of concentrated hydrochloric acid and 200 m1. of carbon tetrachloride. The carbon tetrachloride layer was separated and washed with 150 m1. of water, two ZOO—ml.) portions of 5% sodium carbonate solution, and finally with another 100 m1. of water. The solution was dried overnight with calcium chloride and the solvent removed on a Rinco rotary evaporator. The yellow solid was recrystallized from absolute ethanol and yielded 23 g. (86%) of nitrodurene. The m.p. , m.m. p. , and infrared Spectrum of the product was identical with that of the starting material. Preparation of iododurene (26) Into a round-bottomed flask fitted with a mechanical stirrer and condenser was placed 26.8 g. (0. 2 mole) of durene, 100 ml. of acetic acid, 20 m1. of water, 3.0 m1. of concentrated sulfuric acid, 20.4 g. (0. 16 g. -atom) of iodine, 8.2 g. (0.04 mole) of iodic acid, and 10 ml. of carbon tetrachloride. The mixture was heated and stirred for two hours at 850. The reaction product was separated by addition of 600 ml. of water and then extracted with 100 ml. of chloroform. , The chloroform layer was separated and washed with 100 m1. of 5% sodium hydroxide solution, 100 m1. of 5% sodium thiosulfate solution, and 100 m1. of water. It was dried for twelve hours over anhydrous sodium carbonate. The solvent was evaporated and the white residue was recrystallized once from 95% ethanol to yield 44 g. (85%) of white needles of iododurene, m.p. 78. 5-800. A m.m.p. with durene gave a definite depression. 45 ‘ Attempted Preparation of 4-Iodo-u2, 3, 5, 6~tetramethyl~ benzotrichloride To a la]. 3-necked round-bottomed flask equipped with a stirrer, thermometer, reflux condenser, and drOpping funnel was added 26. 7 g. (0. 20 mole) of powdered anhydrous aluminum chloride and 125 ml. of carbon tetrachloride previously dried over calcium chloride and distilled. The slurry was stirred and heated to 37-420. A solution of 26.0 g. (0.1 mole) of iododurene in 100 ml. of carbon tetrachloride was added dropwise over a two hour period, and the solution was then heated and stirred for an additional three hours. Iodine appeared to be liberated during this 5 hour period. The colored complex was hydrolyzed by slowly adding it with vigorous stirring to a mixture of 100 g. of ice, 15 m1. .of hydrochloric acid and 150 m1. of carbon tetrachloride. The complex hydrolyzed with some difficulty. The organic layer was separated, washed with 100 m1. of water, two 50-ml. portions of 5% sodium carbonate solution, again with 100 m1. of water and finally dried over calcium chloride. The solvent was removed on a Rinco rotary evaporator. A brown viscous oil (12 g.) that could not be purified and was not identified remained after evaporation of the s olvent . B. Cryoscopic Measurements Apparatus The apparatus was the same as that described by Fish (27). The thermistor consisted of a bead of metallic oxides, with lead wires sealed in a glass envelope, designed by Fenwal Electronics Inc. The glass envelope was sealed to a piece of lead glass tubing extending from the cap of the freezing point cell to within a half inch of the bottom of the cell. The thermistor was calibrated from 00 to 200 against a 46 U. S. Bureau of Standards platinum resistance thermometer (No. 1016073). A Leeds and Northrup 5-decade Wheatstone bridge and a Rubicon lamp- scale galvanometer with a sensitivity of 0. 0015 ua/mm.were used to measure resistance. The freezing point cell was equipped with ground-glass joints which were lubricated with silicone grease. The cell cap was equipped with three standard ground joints, one accommodating the stirrer, another the thermistor and the third was fitted with a stOpper and used to add solutes. The cell was surrounded by a Styrofoam insulated air-jacket pro- vided with an upward sloping side arm through which a piece of Dry-ice could be inserted to touch the cell wall and induce crystallization. Procedure The apparatus was always cleaned in hot sulfuric acid, rinsed with distilled water, and dried in an oven. The cell was assembled and approximately 90 g. of stock sulfuric acid was added. After placing the cell in the insulated air—jacket, the entire apparatus was surrounded by a cooling bath of crushed ice and the solution stirred. . The temperature of the sulfuric acid was reduced to about one to two degrees below its freezing point, and crystallization was induced by touching the side of the cell with a piece of Dry-ice. The steady maximum of temperature, obtained by one-minute temperature reading intervals, reached after crystallization was taken as the freezing point. Since the stock solution was maintained on the sulfur trioxide side, a small amount of concen- trated sulfuric acid was added and the freezing point determined. The addition of concentrated sulfuric acid was repeated until the desired freezing point was obtained. The final freezing point was rechecked after two hours to see that it had remained constant. Sufficient solute was then added to depress the freezing point approximately 0. 3O. 47 After the sample had completely dissolved the new freezing point was determined. Two to four weighed portions of solute were added and the freezing point determined after each addition. Two techniques were used to add the sample. One method was to weigh the sample on glazed paper and then pour it into the cell by means of a long funnel made of glazed paper. This method was not satisfactory since the cell had to be opened for each addition and undoubtedly the sulfuric acid picked up some atmospheric moisture. A better method for adding the sample was by means of a solid dropping funnel with a long stem extending almost to the surface of the sulfuric acid. . The sample and dropping funnel were weighed and then placed in the cell. ‘ A small sample was introduced into the cell the freez- ing point determined, the dropping funnel removed, and reweighed. This procedure was repeated until all of the sample had been added. - Stock Sulfuric Acid Stock sulfuric acid was always slightly on the sulfur trioxide side. It was prepared by suitably diluting J. T.. Baker's reagent grade fuming (30-33% sulfur trioxide) sulfuric acid with reagent grade concentrated (96-98%) sulfuric acid. - C. QuantitativeDetermination of Hydrogen Chloride From the Reaction of the Trichloromethyl Compounds with Sulfuric Acid Appa ratu s The apparatus consisted of three drying towers and two traps arranged in series. The traps and towers had ground glass joints and were connected by Tygon tubing. Each trap had a long inlet tube with a fritted glass tip protruding below the surface of the solution, and an exit tube connecting it to the next trap. The first tower contained reagent 48 grade concentrated sulfuric acid, the second and third towers contained slightly greater than 100% (maximum 0.001 molar sulfur trioxide) sulfuric acid. The first trap contained a solution of trichloromethyl compound in 100% sulfuric acid and the last trap contained a 2 to 5% sodium hydroxide solution for collecting the hydrogen chloride. Procedure The apparatus was assembled as described. A known weight of‘tri- chloromethyl compound was added to the first trap, followed by approxi- mately 50 ml. of 100% sulfuric acid. . The solution was stirred magnetically and a stream of oil-pumped nitrogen was bubbled through the system. At various time intervals the aqueous sodium hydroxide trap was changed, the solution acidified with dilute nitric acid to pH 7-8 (pH was determined by a BecMan pH meter), and the chloride ion determined by the Fajans' method (28). D. Spectra The ultraviolet-visible spectra were obtained with the Beckman DK-Z Recording Spectrophotometer and Beckman DU-Spectrophotometer using 1 cm. glass-stoppered quartz cells. The proton magnetic resonance spectra were obtained with a high- resolution nuclear-induction type N.M. R.. spectrometer, varian Associates (VA) Model V-4300-2. The infrared spectra were scanned using a Perkin-Elmer (Model 21) Recording Infrared Spectrophotometer with 0. 5 mm. thickness solution cells. All of the infrared spectra were run in carbon disulfide except for the region of 6. 2 mp. to 7. '1 mp. which was run in carbon tetrachloride. PART B DIPOSITIVE.CARBONIUM IONS ARISING FROM SINGLE IONIZATIONS AT TWO SEPARATE SITES IN A MOLECULE 49 50 RESULTS AND DISCUSSION This part of the thesis deals with the mode of ionization of three glycols related structurally to the triarylcarbinols. Each is in principle capable of ionization to a mono or dicarbonium ion. The ionization was studied by examining the spectra in solutions of varying acidity and by characterizing the hydrolysis products of acidic solutions of the glycols. The three compounds studied were tetraphenyl—p-xylyleneglycol (I), tetra-p-anisyl-p-xylyleneglycol (II), and 9, lOudihydro—9, lO-dihydroxye- 9, 10-diphenylanthrac ene (III). III 51 Spectra Each of the compounds dissolved readily in 100% sulfuric acid producing intensely colored stable solutions. - Compound I gave a deep red solution in 100% sulfuric acid. The visible spectrum of this solution had a single maximum at 455 mu (6 = 58, 700). ‘ Solutions of I in acetic acid showed no absorbance in the visible region. Addition of sulfuric acid to an acetic acid solution of I produced a yellow color which persisted to approximately 28% sulfuric acid. Above this concentration the solutions were red. The maximum absorbance occurs between 40 and 46% sulfuric acid in acetic acid, and the spectrum was identical to that of I in 100% sulfuric acid. Figure XV shows the visible spectra of tetraphenyl-p-xylyleneglycol in varying concentrations of sulfuric acid in acetic acid. The molar absorbancy indices of the maxima are given in. Table VI. At approximately 3. 5% sulfuric acid in acetic acid 6 at 455 mu (dipositive ion) is one-half of its maximum value. In the solvent system sulfuric acid-water I gave similar color changes. At 5. 9% sulfuric acid 6 for the 455 mu band was half of its maximum value and the extinction coefficient did not change above 70% sulfuric acid. (Actually, the position of xma shifts slightly, from 445 mp. in x. 56-64% acid to 455 mu in 70-100% acid.) Figures XVI and XVII show the visible spectra of tetraphenyl-p-xylyleneglycol in'varying concen- trations of sulfuric acid in water and Table VIII records the molar absorbancy indices of the maxima. Compound 11 dissolved in 100% sulfuric acid forming a cherry-red solution. The visible spectrum of this solution had two peaks at 525 mu (e = 88, 500) and 445 mu (6 = 52, 500). When 11 was dissolved in acetic acid the solution was yellow and the visible spectrum had two peaks at 500 mu (6 = 11, 900) and 430 mu (6 = 4, 160). Addition of very small amounts of sulfuric acid to II in 60,000 52,000 44,000 36,000 28, 000 20, 000 52 p-xylyleneglycol in sulfuric acid-acetic . acid. Per cent indicated is weight per / cent sulfuric acid. 21 74% I I 5. 89%/ 5_' 3. 91% 3. 50%» Ada/’ l 380 420 Wavelength (mu) _ Figure XV. Visible spectra of tetraphenyl- -/ 53 Table VI. Visible Absorption Maxima of Tetraphenyl-p-xylyleneglycol in Varying Concentrations of Sulfuric Acid-Acetic Acid Wt. Per Cent Molar Absorbancy Index ~Molar Absorbancy Index H2804 (e) at 425 mu (6) at 455 mu 2.52 15,740 19,700 3.50 24,100 31,000 3.91 27,100 33,700 5.89 29,000 35,700 13.96 30,700 37,000 21.74 31,400 38,200 27.60 --- 41,800 35.24 --- 48,200 40.68 --- 55,800 46.04 ——- 59,900 54 40,000 36, 000 32,000 28,000. 24,000 20,000 16,000 12,000. 8,000 4,000 _ 1 ' I ‘1 360 ~ 400 440 . 480 Wavelength (mu) Figure’XVI. Visible spectra of tetraphenyl—p-xylyleneglycol in varying concentrations of sulfuric acid-water. ~ Per cent indicated is weight per cent sulfuric acid. 56, 000 .- 52,000 — 70% 48, 000 +— I 44, 000 - 40, 000 — . 64% I 36,000 ‘— 32,000 .— 28, 000 1— 24, 000 _ I 20, 000 .— I I 16,000 J l l l 400 440 Wavelength (mp) 480 520 Figure XVII. Visible spectra of tetraphenyl—p-xylyleneglycol in varying concentrations of sulfuric acid-water. - Per cent indicated is weight per c ent sulfuric acid . 56 Table VII. Visible Absorption Maxima of Tetraphenylupuxylyleneglycol in Varying Concentrations of Sulfuric Acid-Water ‘:=======r - A1=================== Wt. Per Cent Molar Absorbancy Index Molar Absorbancy Index H2504 (6) at 420 mp (6) at 445 mp. 56 9,000 10,150 57 16,600 18,750 58 21,400 24,700 59 25,200 30,000 60 30,100 35,200 61 33,500 39,100 62 34,100 40,600 66 ---- 47,600a 68 ---- 57,200b 70 ---- 58,500C 100 ---- 58,700C :Maximum has shifted to 447 mp. CMaximum has shifted to 450 mp. Maximum has shifted to 455 mu. 57 acetic acid caused the solution to become red and produced a batho- chromic shift and a large increase in intensity of the maxima. Figure XVIII shows the visible spectra and Table VIII records the molar absorbancy . indices for tetra-p-anisylap-xylyleneglycol in varying concentrations of sulfuric acid in acetic acid. . The maximum absorbance occurs between 0. 2 and 0.4% sulfuric acid, and half of 6 maximum occurred between 0. O and 0. 02% sulfuric acid in acetic acid. . Figures XIX and XX show the visible spectra of II in varying concen- trations of sulfuric acid and water. 7 The spectrum and shifts in the maxima by varying the sulfuric acid concentration were analogous to those observed in sulfuric acid-acetic acid. At 60% sulfuric acid in water II showed two peaks at 520 mp and 455 mp with e = 96, 000 and 51, 800 respectively. ' In 31% sulfuric acid the bands were displaced to 505 mp and 450 mp with 6 equal to 50, 500 and 26, 200. The colors produced by I and II in acid media can be reversed by appropriately varying the sulfuric acid concentration. Beer's law was obeyed by both compounds. - Compound III dissolves in 100% sulfuric acid forming a deep blue solution. ‘ If III is dissolved in acetic acid the solution is colorless. Addition of a small amount of sulfuric acid causes the solution to become yellow.‘ Increasing the sulfuric acid concentration changes the color progressively from yellow to red to blue. Once the blue color has been formed it cannot be reversed by the addition of acetic acid, whereas the red solution can be converted back to a yellow solution by dilution with acetic acid. . Figure XXI shows the visible spectrum and Table X reports the molar absorbancy index for 9, 10-dihydro-9, lO-dihydroxy-9, 10- diphenylanthracene (III) in varying amounts of sulfuric acid in acetic acid. Figure XXII and Table XI record the comparable data for III in varying concentrations of sulfuric acid in water. 58 100,000.. 80,000. 60,0004— 0.7562 1 \0 \ 0 0.193% \ 40,000 . , \ 0.019%) \ O \O \ 0 \ 20,000+- o l l - 420 460 500 540 580 Wavelength (mp) Figure XVIII. Visible spectra of tetra-p-anisyl-p-xylyleneglycol in varying concentrations of sulfuric acid-acetic acid. Per cent indicated is weight per cent sulfuric acid. 59 Table VIII. Visible Absorption Maxima of Tetra-p-anisyl-p-xylylene- glycol in Varying Concentrations of Sulfuric Acid-Acetic Acid _.__. m Wt.‘ Per Cent Molar Absorbancy Index (6) Molar Absorbancy Index (6) H2804 of the 400 mp Region Band of the 500 mp Region Band Mmp) 6 Mmp) 6 0.00 430 4,160 500 11,950 0.019 470 (shoulder) 40, 600 510 71,800 0.193 456 45, 000 530 88,300 0.756 458 53, 000 530 101, 500 1.50 459 53,000 530 101,300 6.01 457 53,000 530 101,500 9. 33 457 53, 000 530 101, 300 100.0 445 52,500 525 88,500 60 80, 000 — 33% 60, 000 ..... 32% 31% 40,000 — 30% 29% 20,000 28% 26% 0 j I 1| . 440 480 520 - 560 Wavelength (mp) Figure XIX. Visible spectra of tetra-p-anisyl-p-xylyleneglycol in varying concentrations of sulfuric acid-water. Per cent indicated is weight per cent sulfuric acid. 100,000 61 80,000 60,000 20,000r— l 460 500 l 540 ' sulfuric acid: 35% ( Figure XX. Visible spectra of tetra-p-anisyl-p-xylyleneglycol in varying concentrations of sulfuric acid-water. 580 Weight per cent ). 40% (---—). 60% ("'°). 100% (-----). 62 Table‘IX. Visible Absorption Maxima of Tetra-p-anisyl-pwxylylene- glycol in Varying Concentrations of Sulfuric Acid-Water Wt. Per Cent Molar Absorbancy Index (6) . Molar Absorbancy Index (6) H2504 of the 400 rfli Region Band of the 500 mp Region Band X(mp) 6 Mmp) 6 26 425 5., 000 500 10, 000 28 427 12,500 502 26,000 29 430 14,200 504 28,900 30 450 (shoulder) 18, 300 505 38, 000 31 450 (shoulder) 26, 200 505 50, 500 32 460 (shoulder) 38, 000 510 69,100 33 455 42, 300 515 75, 000 45 455 47, 600 526 85, 600 40 455 49, 800 530 92, 150 60 455 51, 800 530 96, 000 100 445 52, 500 525 88, 500 63 28,000..— 24,000 - 25-°% 20,0004—- 8.00% 16,000. 7.00% 12,000 6.00%> 8,000 5.003 4,000 4.005 , l 1 l - 360 400 440 480 520 Wavelength (mp) Figure XXI. Visible spectra of 9, 10-dihydro-9, 10-dihydroxy-9, 10- diphenylanthracene in varying concentrations of sulfuric acid-acetic acid. Per cent indicated is weight per cent sulfuric acid. 64 Table X. Visible Absorption Maxima of 9, 10-Dihydro-9, 10-dihydroxy- 9, lO-diphenylanthracene in Varying Concentrations of Sulfuric Acid- Acetic Acid Wt. Per Cent Wavelength Molar Absorbancy H2804 (mp) Index (6) 4. O 420 to 455 5, 350 5. 0 420 to 455 8, 560 6. 0 420 to 450 12, 800 7. O 425 to 450 15, 500 8.0 426 to 449 18,150 9.0 427 to 448 18,750 20. 0 425 to 435 25, 200 25. 0 425 to 433 25, 800 65 30,000 _. 26, 0001—— 22,000 __. 68% 18, 0001—- 14,000__ 60% 10,000 __ ‘ 59% .‘ 58% 6, 0001 57% 2,0001 56% l ' 1 I 380 420 460 500 Wavelength (mp) Figure XXII. Visible spectra of 9, 10-dihydro-9, 10-dihydroxy-9, 10- diphenylanthracene in varying concentrations of sulfuric acid-water. Per cent indicated is weight per cent sulfuric acid. 66 Table XI. Visible Absorption Maxima of 9, 10-Dihydro—9, lO-dihydroxy- 9, 10-diphenylanthracene in Varying Concentrations of Sulfuric Acid-Water Wt. Per Cent Wavelength Molar Absorbancy H2804 (mp) Index (5) 56 445 4,800 57 447 7,330 58 447 10,000 59 447 12,800 60 445 15,500 61 445 18,450 65 445 23,000 68 445 26,000 67 The yellow and red solutions of III are stable and obey Beer's Law. However, the blue solutions when diluted for spectral studies (concentration of the order of 2 x 10"5 M) fade rapidly and do not give absorbance in the visible region. The maximum stable absorbance of III in acetic acid is reached at 25% sulfuric acid and the‘blue color appears at approximately 32% sulfuric acid in acetic acid. One-half of 6 maximum is attained at 6% sulfuric acid. In sulfuric acid-water the maximum absorbance occurs at 67% sulfuric acid, one-half of x maximum occurs at 59% sulfuric acid and the blue color arises at 70% sulfuric acid in water. Interpretation of the Visible Spectra Because of the reversible color formation observed for compounds I and II in acid media it is proposed that they ionize and establish. an equilibrium in accordance with equation 1. X X Base I + O H H H I X = -H Base la 11 X = -OCH3 Acid IIa / X C . C + + (1) lb IIb 68 In 1955 Deno, it a_._1. (29) defined an acidity function (Co) which provided an acidity scale for secondary bases that ionize according to the equation: ROH + 11+ = R+ + H20 (2) + . . . . where R 18 a carbonium ion. Co was defined as: Co = pKR+ - log (CR+/CROH) (3) When half the alcohol is ionized the concentration of R+ would be equal to the concentration of ROH and the pKR+ would then be equal to the acidity function Co. This acidity function was applied only to the sulfuric acid-water solvent system, but it was valid throughout the entire range of sulfuric acid concentrations. . Using this acidity function Deno, at 2:1. were able to determine the pKf 's of some substituted triarylcarbinols, + which ionized according to equition 2. The red color of solutions of I in sulfuric acid, and more particularly the band at 455 mp, is attributed to the dipositive carbonium ion (Ib), whereas the yellow color and the shoulder at 420 mp probably are due~ to the mon0positive carbonium ion (Ia). . This reasonable assignment is consistent with the principal absorption of the triphenylmethyl cation (30) at 431 mp (6 = 39., 800; secondary band at 404 mp) and is substantiated by methanolysis studies discussed in a later part of the thesis. Unfortunately, the absorption peaks for [the mono and dications from 'I and II are not sufficiently separated, nor are the first and second pK's sufficiently far apart to allow their determination from the spectra with- out certain assumptions. The first is that the only species present where 6 was one-half of 6 are the mono- and dipositive carbonium ions Ia and lb. . There is max considerable basis for this assumption in solvolysis studies described in detail in a later part of this thesis. When compound I was disso1ved 69 in acetic acid containing 2. 5% sulfuric acid (Ho = —2. 17) (31), and the resulting solution poured into cold absolute methanol, a maximum of 17% of unionized glycol was recovered. One-half Ema was attained x. in 59% aqueous sulfuric acid, a solution appreciably more acidic (Ho = -4. 46) (32) than used in the methanolysis experiment. ' It is therefore eminently reasonable to conclude that no unionized glycol remains in 59% aqueous sulfuric acid. . Co is .-8. 7, and in solutions where E is 50% of its maximum this value can then be given to pKhH- for I. pKRq‘q; is defined as the pK of the reaction described in equation 4. ++ (ROH)+ + H+ R + H20 (4) The second assumption is that pKR-i- for I is approximately the same as that of triphenylcarbinol. The monopositive ion Ia has two (— ------------ ‘1 I 1 I I I I 0 I I I I l C 'O ‘13“ + : O : l H | L .............. J " Ia unsubstituted phenyls and the third phenyl has a para substituent. This substituent, however, cannot conjugate with the deve10ping positive charge, and therefore should not seriously affect the first ionization. A comparison of the pK‘R-l-‘s of 4-t—butyltriphenylchloromethane (-6. l) and triphenylcarbinol (-6. 63) shows little effect on the pKR+ by a 4-t-buty1 substituent (29). ' It is therefore reasonable to assume that pKR+ for compound I is approximately -'6. 63, the value for triphenyl- carbinol. The same analogies were applied to compound II. This case was not as clear-cut as compound I, because the bands in the spectrum 70 shifted slightly in wavelength with varying sulfuric acid concentration. Also no particular peak could be attributed to the monopositive carbonium ion (IIa), whereas the dipositive carbonium ion (‘IIb) was assumed ' (reasonably) to give rise to the longer wavelength band. 1 The pK"R.H. of compound II was found to be -3.4. The pK'R+ for 11 was assumed to be approximately -1. 24, the value for 4, 4-dimethoxytriphenylcarbinol (29). The ApKR , the difference between the first and second ionization constants, was 2. 16 for compound II and 2.07 for compound I. Also the ApKR+ between the estimated first ionization constants of I and II was 5. 39 and the ApKR++ between the experimentally determined second ionization constants of compounds I and II was 5. 3. That these ApKR's are of the same order lead one to conclude that the original assumptions were quite valid. The decrease in intensity and hypsochromic shift of the maxima observed in the visible spectrum of II in 100% sulfuric acid was unusual. A possible explanation for this is the protonation of one or more of the methoxyl groups according to equation 5. 110+ CH, This would eliminate the methoxyl group(s) from participating in resonance stabilization of the i6ns and should decrease the wavelength of maximum absorption. 71 It is proposed that 9, 10-dihydro-9, 10-dihydroxy-9, lO-diphenyl- anthracene (III) also ionizes to give mono and dipositive carbonium ions, as depicted in equation 6. Ac1d Ac1d Base Base 111 IIIa C6H5 (6) C6H5 IIIb But, when a certain acidity is reached the dipositive carbonium ion IIIb undergoes further irreversible reactions to be discussed more fully later. - The yellow color of III in acidic solutions is attributed to the mono- positive ion‘IIIa and the red color is attributed to the dipositive carbonium ion IIIb. The valuegbf pK‘R++ for III was found to be -8. 7, identical to that for compound I. This is perhaps not too. surprising since the substituents are the same for both compounds, although one would have perhaps thought that the time-average closer proximity of the positive charges in IIIb would decrease its stability when compared with'Ib. " It is assumed that the pK'Rii for III is approximately the same as that for triphenyl- carbinol. 72 Hydrolysis Products When solutions of compounds I or II in 100% sulfuric acid were hydrolyzed by pouring them on ice the corresponding glycols were recovered in better than 90% yields. Solutions of the dichloride of I in 100% sulfuric acid gave a visible spectrum identical to that of the glycol in 100% sulfuric acid. Hydrolysis of this red solution also gave a nearly quantitative yield of glycol. In 1942 Weitz and Schmidt (33) reported the isolation of a brownish- red perchlorate salt from tetraphenyl-p-xylylenedichloride. They proposed that this was the salt of the dipositive carbonium ion, but no further work was done on it. Recently in this laboratory a red perchlorate salt was isolated (34) from the dichloride of 1. Solutions of this salt in chloroform were red and conducted an electric current. A quantitative yield of the glycol I was obtained when the salt was hydrolyzed. Obviously the red species isolated as a crystalline salt as well as that formed by the ionization of I in sulfuric acid-water or sulfuric acid- acetic acid was the dipositive carbonium ion Ib. In an attempt to more accurately determine the amounts of glycol, monopositive carbonium ion and dipositive carbonium ion formed in varying concentrations of sulfuric acid-acitic acid,:amie‘thanqu'sis studies were undertaken. . The general procedure was to dissolve a sample of the glycol I in dilute solutions of sulfuric acid (approximately 2. 5%) in acetic acid. These solutions were then added under anhydrous conditions to cold absolute methanol. The methanol would be expected to react with the different species to give the corresponding products in accordance with equation 7 . ('36H5 (I36H5 CH. O O H H I CH,0H No Reaction TsHs ('36H5 CfiHs-C O C-C6H5 + + CH,0H C6H5 C6H5 ‘ I | c6H5—c C-C6H5 o o . CH, CH, V 73 CH,OH Clszs C|36H5 C6H5- c O C-C6H5 6 0 CH, H (7) Some preliminary experiments were run to prepare compound IV (previously unknown) and to work out a suitable separation for I, IV, and V using column chromatography. Addition of a solution of the glycol in concentrated sulfuric acid to a cold solution of methanol and water in the molar ratio of 2 to 3 yielded a mixture of the three products. A suitable chromatographic scheme of separation was devised and the amounts of products obtained were 12% of the glycol, 36% of the dimethyl- ether (V) and 46% of the monomethylether-monoalcohol (IV), m.p. 130. 5-1310. The structure of the latter was proved by analysis and by 74 the fact that dissolution in 100% sulfuric acid followed by hydrolysis gave the glycol (I) in quantitative yield. A change in the molar ratio of methanol to water varied the relative amounts of products obtained. _ Three samples, each of the same concentration of glycol in 2. 52% sulfuric acid-acetic acid were added to cold absolute methanol and worked up in the same manner. . Chromatography of the products did not give consistent results, different relative amounts of I, IV and V being obtained for each sample. This indicated that the products were either exchanging hydroxyl and methoxyl groups on the column or that the alcohol groups were reacting with the methanol under the acid con- ditions of methanolysis. The latter undoubtedly did occur to some extent, because when the glycol was dissolved in cold methanol and the sulfuric acid-acetic acid was added to it, no color was formed, but after purification and chromatography a 10% yield of V and a 6% yield of IV was obtained. Although this observation precluded further quanti- tative methanolysis studies, some valuable information was obtained from the work already completed. First, the amount of compound IV obtained from the methanolysis of I in 2. 52% sulfuric acid-acetic acid, even though it varied from sample to sample (21 to 40%), was much greater than can be attributed to reaction of the glycol with methanol in acid medium. . This leads one to conclude that the'majority of compound IV was formed from the reaction of the ion Ia with methanol. The second conclusion was that at very low acidity (Ho = -2. 17) there was very little unreacted diol (12-17%) and that at higher acidities there should be negligible amounts of unionized glycol. The hydrolysis products of compound II in sulfuric acid were not studied as extensively as I, because of the convenient analogy with compound I. When the red solution of II in 100% sulfuric acid was hydrolyzed on ice a better than 90% yield of the starting glycol was obtained. Hydrolysis of the solution of II in acetic acid also gave very high yields of the glycol. 75 In 1904 Haller and Guyot (35) first reported the preparation of 9, 10-dihydro-9, 10~dihydroxy~9, 10-diphenylanthracene (III). . They observed an indigo color when III was added to concentrated sulfuric acid. Since that time a number of workers (36, 37, 38) have prepared this compound and observed the same color. However, no one had attempted either to explain the blue color or to isolate any hydrolysis products from sulfuric acid solutions of III. . In the present work the blue color was originally attributed to the dipositive carbonium ion IIIb. . By analogy to compound I and II, hydrolysis should produce the starting glycol. . However, when the blue solution of III in 100% sulfuric acid was added to ice only a brownishagreen solid could be isolated. The material could not be purified by recrystallization and was chromatographed on alumina. ~ Elution with 1:1 pentane-benzene yielded a light yellow solid, m.p. 247-2480, which fluoresced blue in solution; these properties correspond to those of 9, 10-dipheny1anthracene. 9, lO-Diphenylanthracene was prepared independently by the reduction of III in acetic acid with sodium iodide (39). The pale yellow solid isolated from the hydrolysis of the blue solution of III in sulfuric acid was identical (m.p. , m.m.p. and infrared spectrum) to 9, lO-diphenyl- anthracene. The yield of 9, lO-diphenylanthracene (VI) from hydrolysis of the blue solution was 36%. . Further elution of the column with benzene yielded an orange solid, m. p. 191. 2-192. 40, which fluoresced greenish-yellow in solution. The infrared spectrum indicated the compound was an aromatic hydro- carbon; it analyzed correctly for C26H16. . Recently Clar and Willicks (40) reported the preparation of 4-phenyl-2, 3-benzofluoranthene (VII), m. p. 185-1860, by the method depicted in equation 8. A comparison of its ultraviolet and visible spectra (reported by Clar) with that of the material isolated from hydrolysis of the blue solution showed the same 76 KOH 000 Refluxingv ‘00 (8) C5H5 Quinoline 6H5 VII maxima with almost identical extinction coefficients. Table XII shows this comparison. The compounds are apparently identical and the small difference in melting points may be due to the different solvents used for recrystallization. , Clar and Willicks reported using alcohol (presumably ethanol) for the solvent. When this author tried ethanol the orange solid appeared to be insoluble, whereas this material readily dissolved in benzene-pentane. . The yield of 4-phenyl-2, 3-benzofluoranthene (VII) from the hydrolysis of the blue solution of III in sulfuric acid was 19%. Continued elution of the column with 1:1 benzene-ether gave a very small amount of red solid which melted at approximately 3850. It was at first believed to be rubicene, but the carbon-hydrogen analysis was not correct. The material remains unidentified. 9 1,9 0 Rubic ene Finally the remainder of the material on the column was removed with methanol. A dark brown solid was obtained which did not melt sharply (110-1700) and appeared to be a mixture. ' Attempts to purify it by recrystallization failed.. The infrared spectrum of the crude material 77 Table XII. Ultraviolet-Visible Absorption Maxima of 4-Pheny1-2,3-benzo- fluoranthene in Absolute Ethanol W Clar and Willicks (40) This Work Wavelength -Molar Absorbancy Wavelength Molar Absorbancy (mp) Index (6) (mp) - Index (6) 430 13,500 430 12,600 366 6, 300 366 5, 830 310 11,500 310 12,000 270 100, 000 270 ‘94, 500 78 indicated that it contained both hydroxyl and carbonyl functions. None of the constituents of the mixture have been identified. . Hydrolysis of the yellow solutions of III in sulfuric acid-acetic acid gave between 90 and 95% yield of the starting material along with traces of VI and VII. It seems that in low sulfuric acid concentrations the glycol under- goes normal stepwise ionization to the dipositive carbonium ion IIIb. In more concentrated acid (32% sulfuric acid and above) further reactions I occur, presumably including internal alkylation and hydride transfer. 4-Phenyl-2, 3-benzofluoranthene may arise from the mechanism suggested in equation. 9. (9) VII The formation of 9, lO-diphenylanthracene is more difficult to explain, because it requires the transfer of a hydride ion as shown in equation 10 . 79 (10) VI Until all of the hydrolysis products have been isolated and identified the nature of the hydride donor remains unknown, except that it presumably ends up as the polar residue which is last to be eluted from alumina, and which contains hydroxyl and carbonyl functions. The species responsible for the blue color is as yet not known. Cryoscopic Measurements Cryscopic measurements were run on solutions of tetraphenyl-p- xylyleneglycol (I) in 100% sulfuric acid. When I ionizes in 100% sulfuric acid 7 particles should be produced in accordance with equation 11. $61—15 ('36H5 |C6H5 ('36H5 C6H5‘C': O 9-C6H5 ‘1" 4 H2504 —> C6H5-C O C‘CéHs O O + + H H (1 1) + mm," + 2 H,o+ 80 The results are reported in Table XIH. The data in general support a molal freezing point depression of 7, although the _i_-value is a bit high. It should be pointed out that the first cryoscopic measurements determined by the author were done on compound I and the experimental technique had not yet been mastered. Also compounds that exhibit large molal freezing point depressions can not be determined as accurately as compounds that have lower i_ factors. . Further cryoscopic measurements should be run on 1 before making a conclusive decision on the results. 81 Table XIII. Freezing Point Data on Tetraphenyl-p-xylyleneglycol Sample Wt. , g. g. sto, T1, °c. AT, °c. Time (hrs.)a 1b 1.5257 97.10 9.941 1.728 'XC 7.67 1.717 4 7.62 1.7174 80.00 9.929 2.333 x 7.55 1.7204 85.12 9.889 2.143 x 7.39 2.093 13 7.22 a Indicates time elapsed between addition of the sample to sulfuric acid and determination of the i-factor. b Calculated from i = AT/6. 12 ms, where ms is the molality of the solute. c i-Factor determined immediately after addition of the sample to sulfuric acid. EXPERIMENTAL 82 83 A . Synthesis Preparation of Tetraphenyl-p-xylyleneglycol (41) To a 1-1. three-necked flask fitted with a Trubore stirrer, dropping funnel, and reflux condenser with a calcium chloride drying tube was added 8 g. (0. 33 g. -atom) of magnesium and 50 ml. of ether. A solution of 50 g. (0. 318 mole) of bromobenzene in 150 m1. of anhydrous ether was added slowly. When the initial vigorous reaction had subsided the remainder of the solution was added at such a rate as to maintain refluxing. After the Grignard reagent had been prepared a solution of 15 g. (0. 077 mole) of dimethyl terephthalate in 350 ml. of benzene (previously dried over sodium and redistilled) was added dropwise to the refluxing liquid during a three hour period. , The reaction mixture became orange and a solid precipitated. After three hours of further reflux, the mixture was poured onto 200 g. of ice, acidified with dilute sulfuric acid, and the organic layer separated. The aqueous phase was extracted with two lOO-ml. portions of benzene and the combined organic layers were dried for 12 hours with magnesium sulfate.. The benzene was distilled and'the pale-yellow, oily crystal mass was filtered and washed with benzene and petroleum ether. There was approximately 25 g. of crude product, m.p. 165-1690. The solid was recrystallized from benzene-petroleum ether to yield 22 g. (64. 5%) of tetraphenyl-p- xylyleneglycol, m.p. 169-1700. 1 (Literature value 168-1690.) Its infrared spectrum is shown in Figure XXIII. Preparation of Tetraphenyl—p-xylylenedichloride (41) Tetraphenyl—p-xylyleneglycol (5 g. , 0. 0113 mole) was dissolved in 100 m1. of hot anhydrous benzene. Dry hydrogen chloride was bubbled 84 .2155 Hoormosofinfimxumugsoammuuou mo £3.30on poumuwg .HHCCA ounmwh Amsouowgv fimcoaocrm? : 3 o w h . o m. _ «4 4 L d _ _ q _ _ 85 into the refluxing solution and after approximately five minutes white crystals began to precipitate. The gas was bubbled into the solution for another five minutes and then the contents were cooled to 100. The white crystals were filtered and recrystallized from benzene to yield 4 g. (72%) of tetraphenyl-p-xylylenedichloride, m.p. 247-2489. (Literature value 240-2410,) Reaction of a Solution of Tetraphenyl-p-xylyleneglycol in 100% Sulfuric Acid with Water Tetraphenyl-p-xylyleneglycol (1. 5 g.) was dissolved in 15 g. of 100% sulfuric acid. . The deep red solution was slowly added to 100 g. of ice. The white precipitate was collected, washed with water, and dissolved in 50 m1. of ether. 7 The ether layer was washed with two 50 ml. -portions of 5% sodium bicarbonate solution and then dried with magnesium sulfate. The ether was evaporated, to yield 1. 38 g. (92%) of tetraphenyl-p-xylyleneglycol m. p. 166-1680. . One recrystallization from benzeneapetroleum ether raised the melting point to 169-1700. A m.m.p. with an authentic sample gave no depression. Reaction of a Solution of Tetraphenyl-p-xylylenedichloride in 100% Sulfuric Acid with Water Tetraphenyl-p-xylylenedichloride (1.0 g.) was dissolved in 15 g. of 100% sulfuric acid. The red solution was slowly added to 100 g. of ice and the white precipitate was filtered, washed with water and dis- solved in 50 m1. of ether. The ether layer was washed with two 40 m1. - portions of 5% sodium bicarbonate solution and-then dried over magnesium sulfate. The ether was evaporated to yield 0. 83 g. (90. 3%) of a white solid, m.p. 166-1690. » Recrystallization from benzene-petroleum ether raised the melting point to 169-1700. The product had a m.p. , m.m.p. , and infrared spectrum identical to tetraphenyl-p-xylyleneglycol. 86 .Preparation of the Dimethylether of Tetraphenyl-p— xylyleneglycol This compound was prepared by two different methods. - Method A Tetraphenyl-p-xylylenedichloride (3 g. , 0.0063 mole) was dis- solved in a solution of 50 m1. of absolute methanol containing 0.625 g. (0. 0126 mole) of sodium methylate. The mixture was refluxed for 12 hours and the precipitate filtered, washed with water and dried. The. solid was recrystallized from benzene-petroleum ether to yield 2. 6 g. (87%) of white crystals of the dimethylether of tetraphenyl-p-xylylene- glycol, m.p. (sinters 182-1830) 185?. (Literature value 181-1830.) (41). Method B Tetraphenyl-p-xylyleneglycol (2.0 g.) was refluxed with 50 m1. of absolute methanol and two drops of concentrated sulfuric acid for two hours. (A white solid formed almost immediately.) The crystals were filtered and recrystallized from benzene—petroleum ether to yield» 1. 8 g. (87%) of the dimethylether of tetraphenyl-p-xylyleneglycol, m.p.. (sinters 182-1830) 1850. The m.p. , m.m.p. , and infrared spectra of the product isolated from both reactions were identical. . The infrared spectrum of the dimethylether is shown in Figure XXIV. Preparation of the Monomethylether of Tetraphenyl-p- xylyleneglycol ‘ Tetraphenyl-p-xylyleneglycol (2.0 g.) was dissolved in 15 g. of 100% sulfuric acid. The red solution was added dropwise to a cold and well-stirred solution of 43. 2 g. (2.4 moles) of water and 51. 2 g. (1.6 moles) of methanol. . The white solid was collected, dissolved in 75 m1. of ether, and washed with two 50-ml. portions of 5% sodium carbonate solution, and dried over magnesium sulfate. 1 The ether was evaporated 87 .HoHormone???"naugcogmmhou mo uvfimfnzuogwp v.5. mo 8.930on 60.3.35 (3an onfimfim NH : ’ OH Announcing #9953225 my . 1’ w I, —-b\o fl m _ _ 88 and the solid was dissolved in 10 m1. of benzene and adsorbed on 70 g. of Fisher's Adsorption alumina (80-200 mesh). The column was eluted with 300 m1. of benzene and yielded 0.77 g. (36.45%) of the dimethylether of tetraphenyl-p-xylyleneglycol, m. p. 1850. . Further elution of the column with 500 m1. of tetrahydrofuran yielded 0. 946 g.- (45. 83%) of white crystals of the monomethylether of tetraphenyl-p-xylyleneglycol, m.p. 130. 0-130. 50. ~ Its infrared spectrum is shown in Figure XXV. -_A_n_a_1_. Calcd. for C33HZBOZ: C, 86.81; H, 6.18. Found: C, 86.84; H, 6.21. The glycol was removed by eluting with 200 m1. of absolute methanol and yielded 0. 24 g. (12%) of tetraphenyl-p-xylyleneglycol, m.p. 168-17o°. The total yield of recovered material was 94. 2%. Reaction of a Solution of Tetraphenyl-p-xylyleneglycol in Sulfuric Acid (2. 52%)-Acetic Acid with Absolute Methanol A sample of 0. 2500 g. of tetraphenyl-p-xylyleneglycol was dis- solved in 25 ml. of 2. 52% sulfuric acid in acetic acid. . The red solution was added dropwise to 100 m1. of cold (-100) absolute methanol. Cold water (100 ml.) was then added to the colorless alcohol solution totpre- cipitate all of the products. The solid was filtered, dissolved in 100 ml. of ether and washed with two 50-ml. portions of 5% sodium bicarbonate solution. . The basic washings were separated and extracted with 100 m1. of ether. The combined organic layers were dried withmagnesium sulfate for twelve hours and the ether was evaporated on a Rinco rotary evaporator. . The solid residue was dissolved in 10 m1. of benzene and adsorbed on 80 g. of Fisher's Adsorption alumina (80-200 mesh). The column was eluted successively with 200 m1. of benzene, 600 ml. of tetrahydrofuran, and 200ml. of absolute methanol. The experiment was run in triplicate using identical conditions and the same amount of sample for each run. 1 The results are recorded in Table XIV. 89 «L .HootfimocoHungxtmtacofimmhou mo nofiuoaguugocoe may mo 85.30on poumuwcu 698.3“ch sewage/.35 2 NH 0H «0 m N. o m .>xx euamE _,H . 0—4 m _ _ . _ fl _ _ 90 Table XIV. . Relative Amounts of Products Formed from the Reaction of a Solution of Tetraphenyl—p-xyly1eneglycol in 2. 52 Weight Per Cent Sulfuric Acid in Acetic Acid with Absolute Methanol Sample Compound I IV V Wt. (g.) Per Cent Wt. (g.) Per Cent Wt. (g.) Per Cent Yield Yield Yield I 0.149 55.8 0.119 44.57 0.198 74.50 11 0.084 32.4 0.103 40.76 0.055 21.20 111 0.038 14.95 0.032 12.76 0.04 16.45 91 Reaction of a solution of tetraphenyl-p-xylyleneglycol in Absolute Methanol with Sulfuric Acid (2.52%)-Acetic Acid A sample of 0. 100 g. of tetraphenyl-p-xylyleneglycol was dissolved in 40 m1. of absolute methanol. The solution was cooled to -100 and a solution of‘10 m1. of 2. 5% sulfuric acid in acetic acid was added dropwise while maintaining the temperature between -10 and 00. ' After the addition the acid was neutralized with 10% sodium carbonate solution. . The solid was filtered, dissolved in 50 m1. of ether, and dried with magnesium sulfate. The solvent was evaporated and the solid residue was dissolved in 5 m1. of benzene and adsorbed on 35 g. of Fisher's Adsorption alumina (80-200 mesh). The column was eluted successively with 100 m1. of benzene, 300 m1. of tetrahydrofuran and 100 m1. of absolute methanol. The amount of products obtained was 0.010 g. (9%) of the dimethylether of tetraphenyl-p—xylyleneglycol, 0.0072 g. (6%) of the mono- methylether of tetraphenyl-p-xylyleneglycol, and 0.0833 g. (83.3%) of tetraphenyl-p-xylyleneglycol. Reaction of Ia Solution of the Monomethylether of Tetra- phenyl-p-xylyleneglycol in 100% Sulfuric Acid with Water The monomethylether of tetraphenyl-p-xylyleneglycol (0. 50 g.) . was dissolved in 5 ml. of 100% sulfuric acid. The red solution was slowly added to 50 g. of ice and the white precipitate was filtered, washed with water and dried. The crude product (0.47 g. , 95%) was recrystal- lized from benzene-petroleum ether to yield white crystals of tetra- phenyl-p-xylyleneglycol, m. p. 167- 1690. Preparation of Tetra-p-anisyl-p-xylyleneglycol (42) To a .1-1. 3-necked round-bottomed flask equipped with a Trubore stirrer, dropping funnel and a reflux condenser with a calcium-chloride drying tube was added 14. 58 g. (0. 6 g. -atom) of magnesium and 150 m1. 92 of anhydrous ether. Approximately 25 m1. of a solution of 112. 2 g. (0. 6 mole) of 4-bromoanisole in 200 m1. of ether was added to the flask and the mixture was refluxed until the Grignard reagent began to form. After the initial vigorous reaction had subsided the remainder of the 4-bromoanisole solution was added at such a rate as to maintain reflux (2 hours). After the formation of the Grignard reagent a solution of 19.4 g. (0. 10 mole) of dimethyl terephthalate in 200 ml. of anhydrous benzene was added slowly (2 hours). When the addition was complete the reaction mixture was refluxed for an additional 8 hours after which the solution was hydrolyzed by adding it to 300 m1. of saturated ammonium chloride solution. The organic layer was separated and steam distilled until the distillate was clear. The residue was filtered and recrystallized from benzene and then acetone to yield 20 g. of a pale yellow solid, m. p. 163-1660, which was taken up in 250 ml. of benzene and chroma- tographically absorbed on 100 g. of Fisher's Adsorption alumina (80-200 mesh). Elution with 3 l. of benzene containing 15 m1. of absolute ethanol yielded 18 g. (32%) of cream~colored powder, m.p. 165-1670. Recrystallization from acetone raised the melting point of the tetra-p- anisyl-p-xylyleneglycol to 167-1680. (Literature value 170-1710.) Its infrared spectrum is shown in Figure XVI. Reaction of a Solution of Tetra-p-anisyl-p-xylyleneglycol in 100% Sulfuric Acid with Water Tetra-p-anisyl-p-xylyleneglycol (2.0 g.) was dissolved in 15 g. of 100% sulfuric acid. The cherry-red solution was slowly added to 100 g. of ice. The white precipitate was collected, washed with water, and dissolved in 50 ml. of ether. The ether layer was washed with two 50-ml. portions of 10% sodium carbonate solution and then dried with magnesium sulfate. . The ether was evaporated and yielded 1.79 g. (90%) of tetra-p- anisyl-p-xylyleneglycol, m.p. 166.5-1680. The m.p., m.m.p., and infrared spectrum were identical to those of the starting glycol. 93 NH .SHDEV H00>amocoa>a>xtmtfi>mwcmumnmuuoa mo gabomm popmnmsH . .H>NX 0.3..th Anson 38v nuwaofioxrm? H A 2 o m N. o m w _ _ m _ h _ _ _ 94 Preparation of 9, 10~Dihydro~9, lOudihydroxy—9, 10- diphenylanthracene (35, 38) To a lul. 3~necked round-bottomed flask equipped with a Trubore stirrer, reflux condenser and a dropping funnel was added 32. 9 g. (1. 35 g. -atoms) of magnesium, 100 m1. of anhydrous ether, and 30 g. of bromobenzene. After the initial vigorous reaction had subsided a solution of 185 g. (a total of 1. 37 moles) of bromobenzene in 200 ml. of anhydrous ether was added dropwise during a 3 hour period. In a second 141. 3-necked roundabottomed flask equipped with a stirrer and reflux, condenser with a calcium chloride drying tube was placed 57. 3 g. (0. 274 mole) of anthraquinone in 250 m1. of ether. . The previously prepared Grignard solution was transferred under a nitrogen atmosphere to a 500 ml. dropping funnel and was then slowly added (1 hr.) to the refluxing and stirred slurry of anthraquinone in ether. After the addition, the reaction mixture was refluxed for an additional three hours and then hydrolyzed on 400 g. of ice, 100 ml. of water and 20 m1. of concentrated sulfuric acid. The solution was filtered to remove unreacted anthra- quinone and the layers were separated. The aqueous phase was extracted with two 100-ml. portions of ether. The organic layer and the ether extracts were combined and the solvent was removed on a Rinco Rotary evaporater. The solid residue was combined with the previously filtered solid and added to 500 ml. of boiling ethyl acetate. The mixture was cooled and filtered and the solid residue was again added to 500 ml. of boiling ethyl acetate. . The mixture was cooled and filtered and the filtrate was combined with the previously filtered ethyl acetate. . The solution was concentrated until a solid began to precipitate. The solution was cooled to 0°, filtered and the solid was washed withligroin. ~ It was recrystallized from acetone and then benzene to yield 35 g. (35. 1%) of white needles of 9, 10-dihydro-9, lO-dihydroxy-9, lO-diphenylanthracene, m.p. 258-2590. Because of the wide deviation between the melting points reported in the literature for this compound a sample was sent for analysis. 95 Anal- Calcd. for CstzoOz: C, 85.71; H, 5.57. Found: C, 85.54; H. 5.66. Its infrared spectrum is shown in Figure XXVII. ~ Preparation of 9, lOeDiphenylanthracene (39) 9, 10—Dihydro-9, 10-dihydroxy-9, lO-diphenylanthracene (2. 0 g.) and 2.48 g. (0. 0165 mole) of sodium iodide were dissolved in 50 ml. of acetic acid. The solution was warmed on a steam bath for three hours. At the end of this time the solution was added to 50 g. of ice and 50 g. of 10% sodium thiosulfate solution. . The pink solid was filtered, dissolved in 50 ml. of benzene, washed with two 50-rnl. portions of 10% sodium thiosulfate solution, and dried with magnesium sulfate. The benzene was removed on a Rinco Rotary evaporator and the pale yellow solid was recrystallized from benzene-pentane to yield 1.42 g. (85%) of 9, 10- diphenylanthracene, m.p. 247-2480. (Literature value 2480.) Its infra- red spectrum is shown in Figure XXVIII. Reaction of a Solution of 9, lO-Dihydro-9, lO-dihydroxy- 9, lO-diphenylanthracene in Concentrated Sulfuric Acid with 15% Sodium Hydroxide Solution A sample of 3. 573 g. of 9,10-dihydro-9,10-dihydroxy-9,10- diphenylanthracene was dissolved in 35 ml. of 98% sulfuric acid. The solution was stirred and maintained at 00 for 15 minutes. The deep blue . solution was slowly added to 150 ml. of 15% sodium hydroxide solution 1 at -5 to 00. , The greenish-brown solid was filtered, washed with water and dried. The solid (3. 334 g.) was dissolved in 15 m1. of benzene and adsorbed on 100 g. of Fisher's Adsorption alumina (80-200 mesh). The column was eluted with 200 m1. of 1:1 benzene-pentane and yielded 1. 171 g. ' (36.6%) of a pale-yellow solid, m.p. 247-2480. The m.p. ,. m.m.p. , and infrared spectrum of this material were identical to those of an authentic sample of 9, lO-diphenylanthracene. .Asosflon ”Homov osooenflsoisohahoéh 55.83572 5-9.3572 .o co Sansone 6233 .55? Hanna Anson 02H: £umG0H0>0>> 2 2 S o m e c m _ _ _ _ _ A _ _ p4 —-J<1* 97 V odoudnnunmfwaogmmptoH .mtunxohotfiflp .713; GOEEOm 300 033.25 030. 0%. mo 3930.38»: Eonm was .A nod .Otonpifiptoa .o mo £030.30"; may Eon“. 0G00myaucma>coamwptofi .0 mo gnu—00mm pondnfifi . Ame/XX oufiwwh Amsouflav Aumcofioczo? E. . 2 Nth! : i 8 _..o m h c m a. m .. ud.. _ Z. . c... _ _ _ J _ { _ Z _ _ _ 98 Continued elution of the column with 300 m1. of benzene yielded 0.6016 g. (19%) of an orange solid, m.p. 188-1900. Recrystallization from benzene-pentane raised the m.p. to 191. 2~192.4O (corrected). The compound has an ultraviolet—visible spectrum (Figure XXIX) that is almost identical to that of 4~phenyl-2, 3~benzofluoranthene reported by Clar and Willicks (40). Because of the difference between the observed melting point (191. 2-192.4°) and that reported (185-1860) for this com— pound a sample was sent for analysis. Anal. Calcd. for C26H16: C, 95.09; H, 4.91. Found: C, 95.14; H, 5.05. Its infrared spectrum is shown in Figure XXX. . Continued elution of the column with 1:1 benzene-ether yielded 0. 040 g. of a red solid. The material was recrystallized from xylene- petroleum ether and had a m.p. of approximately 385°. 7 There was only enough material for one carbon-hydrogen analysis and the compound was found to contain some ash and possibly an element(s) other than carbon or hydrogen. £1131. Found: C, 91.81; H, 4.89; Ash, 1.24. The material was not identified. 7 Finally elution of the column with ether, ether-methanol mixtures or methanol yielded only a dark brown residue. Attempts to purify it by recrystallization from methanol were unsuccessful. . The material melted over a very wide range (110-1700) and the infrared spectrum, shown in Figure XXXI. of the crude material showed hydroxyl and carbonyl bonds. The identity and composition of this material was not determined. 99 100, 000 A Figure XXIX. Ultraviolet-visible spectrum of 4~phenyl- 2, 3—benzof1uoranthrac ene in absolute ethanol. 75,000 p. 50,000- 25, 000 __.. 14,000<——- 11,000 __ 8,000 5,0001—- , H l 1 l 2.40 300 360 420 480 F 540 Wavelength (mp) .mcoumhflucmHOSfioucofium .NnTfiSAQi. mo Efiuuoomm consume: . .VCCA 0n9mfim . Amcouoflhv AumsofioNrm? «L 2 NH 3 g m m N. o 100 _ 3. _ _ ._ _ _ _ _ dim 101 503300 300 0?.9390 093 05. mo 3930.28»: 0%. 50.3 05300." H.309 03p mo 5:50090 00.20.35 . .HXXX 0u5mwh Amcouflav 47903203 2 NH : 0H m m N. o m w m _ m _ w _ _ i _ _ ——( 102 Reaction of a Solution of 9, lO-Dihydro-9, lO-dihydroxyu 9, lO-diphenylanthracene in 6% Sulfuric Acid in Acetic Acid with Water 9, lO-Dihydro-9, lO-dihydroxy-9, lO-diphenylanthracene (0. 8705 g.) was dissolved in 15 m1. of 6% sulfuric acid in acetic acid. The yellow solution was slowly added to 100 g. of ice and the light yellow precipitate collected. The solid was dissolved in 50 ml. of ether, washed with two 30-ml. portions of 5% sodium carbonate solution and dried with magnesium sulfate. The ether was evaporated and the solid residue dissolved in 15 m1. of benzene and adsorbed on 60 g. of Fisher's Adsorption alumina (80-200 meSh). Elution with 200 ml. of 1:1 benzene-ether yielded a small amount of yellow solid (0. 023 g.) presumably a mixture of 9, lO-diphenyl- anthracene and 4-phenyl-2, 3-benzofluoranthene. ~ Elution with 250 m1. of absolute methanol gave 0. 8071 g. (92. 5%) of 9, lO-dihydro-9, lO-dihydroxy- 9,10-dipheny1anthracene, m.p. 257-258. 5°. A mixed m.p. with an authentic sample of glycol gave no depression. B. Solutions for Spectral Measurements Preparation of Weight Per Cent Sulfuric Acid-Water Solutions for the Visib1e_Spectra Stock solutions consisted of "Baker Analyzed" Reagent grade concentrated sulfuric acid. The per cent composition of the stock solu- tion was determined by titrating a weighed volume of sulfuric acid with standardized sodium hydroxide solution to the phenolphthalein end point. Calculated weights of water were then added to known weights of stock solution to give the desired weight per cent of sulfuric acid-water for the spectral measurements. 103 Preparation of Weight Per Cent Sulfuric Acid-Acetic Acid Solutions for the Visible Spectra Reagent grade glacial acetic acid was fractionally distilled in an all glass still protected from atmospheric moisture. Only acid boiling at ll8o/atm. press. was used. . Clear 100% sulfuric acid was prepared as previously described in the experimental section in Part A of this thesis. The sulfuric acid-acetic acid solutions were prepared by the method of Hall and Spengeman (6) at 250 1: 2°. Standard sulfuric acid solutions were prepared by directly weighing 100% sulfuric acid and diluting to a known volume with acetic acid. A The solutions were then made up by mixing known volumes of acetic acid with known volumes of stock solution. For solutions 3 M or less the stock solutions could be measured from a buret. For more concentrated solutions, however, it was necessary to weigh out the stock solution because of its viscosity. C. Spectra The ultraviolet-visible spectra were obtained with the Beckman DK-Z Recording Spectrophotometer using 1 cm. glass-stoppered quartz cells. . The infrared spectra were scanned using a Perkin-Elmer (Model 21) Recording Infrared Spectrophotometer with a 0. 5 mm. thickness solution cell. All of the infrared spectra were run in carbon disulfide except for the region 6. 2-7. 1 mp. which was run in carbon tetrachloride. This region was incorporated into the carbon disulfide spectrum to give one continuous spectrum. . For solubility purposes some compounds were run in chloroform or as mulls, as indicated on the spectra. 104 Test of Beer's Law Each of the three glycols studied was observed to obey Beer's Law. The results were obtained with the Beckman DU-Spectrophotometer using 1 cm. glass-astoppered quartz cells. A set of sample data is given for tetraphenyl-p-xy1yleneglycol in 2. 52% sulfuric aciduacetic acid in Table XV. 105 Table XV. Test of Beer's Law for Tetraphenyl-p-xylyleneglycol in 2. 52% Sulfuric Acid-Acetic Acid Conc. (moles) Optical Density Molar Absorbancy at 455 mu Index (6) 7.03 x10"6 0.138 19,650 1.406 x10"5 0.274 19,450 2.109 x10“5 0.405 19,200 SUMMARY 106 107 1. The aluminum chloride-catalyzed reaction of carbon tetra- chloride with polymethylbenzene derivatives was extended to‘mono- substituted durene derivatives to yield 4-bromo-, 4-chloro-, and 4-fluoro-2, 3, 5, 6-tetramethy1benzotrichlorides. The structures of the latter were established by hydrolysis to the corresponding acids and comparison with independently synthesized authentic specimens. 2. Each of the 4-halo-2, 3, 5, 6-tetramethy1benzotrichlorides dissolved in 100% sulfuric acid to form a deep red solution. ‘ In each case two moles of hydrogen chloride were quantitatively swept from these solutions. . Hydrolysis of the colored solutions gave better than 90% yield of the corresponding 4-halodurenecarboxylic acids. Cryosc0pic measurements showed that five particles were produced when 4-bromo-Z, 3, 5, 6-tetramethylbenzotrichloride dissolved in 100% sulfuric acid. By analogy to the reaction of trichloromethylpentamethyl- benzene with 100% sulfuric acid it was found that the 4-halo-2, 3, 5, 6- tetramethylbenzotrichlorides ionize in 100% sulfuric acid to form the corresponding dipositively charged carbonium ions. + x O cc13+2H,so,——>x C=c-c1+2HC1+2Hso4- IVa=Br Va,b,c b=Cl c=F The ultraviolet, visible and proton magnetic resonance spectra supported the structure asSignments for the dipositive carbonium ions. 3. The formation of dipositive carbonium ions arising from single ionizations at two separate sites in a molecule was observed for three glycols. . Tetraphenyl-p-xy1y1eneg1ycol and tetra-p-anisyl-p-xylylene- glycol dissolved in 100% sulfuric acid to form intensely red solutions. Hydrolysis gave better than a 90% yield of the corresponding glycols. 108 A study of the visible spectra in solutions of varying acidity showed that both glycols undergo reversible stepwise ionizations to monopositive and dipositive carbonium ions. 9, lO-Dihydro-9, lO-dihydroxy-9, lO-diphenylanthracene dissolved in 100% sulfuric acid to form an intense blue solution. . Hydrolysis did not yield the starting material, but rather a mixture containing 9, lO-diphenylanthracene (36%), 4-phenyl-2, 3-benzofluoranthene (19%) and some products which remain unidentified. In much less acidic solutions 9, lO-dihydro-9, lO~dihydroxy-9, lO- diphenylanthracene undergoes the normal stepwise ionization to mono- positive and dipositive carbonium ions, as shown by hydrolysis experi- ments. 4. The pKR++'s, the pK's for the process: + + ++ <__ were determined spectrosc0pically for the three glycols, and were found to be -8. 7 for tetraphenyl-p-xylyleneglycol and 9, lO-dihydro-9, 10- dihydroxy-9, lO-diphenylanthracene and -3.4 for tetra~p-anisyl-p- xylyleneglycol. ~ MISCELLANEOUS 109 110 Results and Discussion Recently Hart and Fish (1) reported an unusual thermal reaction for trichloromethylpentamethylbenzene (A). When A was heated slightly above its melting point (94. 5-95. 00), hydrogen chloride was evolved and a new compound 7, 7-dichloro-1, 2, 3, 4-tetramethyl-bicyclo[4, 2, 0] octa-l, 3, 5~triene was formed. Hydrolysis of B with ethanolic silver nitrate solution gave the ketone (C) cc13 c1 110-1251 / ' C1Ethanolic\ / l \ -HC1 ’ A NO ( ) \ 8 3 s}. (1) A B C In order to ascertain the applicability of this reaction to analogous systems 4-chloro-2, 3, 5, 6-tetramethylbenzotrichloride (D) was heated to 1750 for 4 hours. A new compound was isolated which analyzed correctly for 3, 7, 7-trichloro-l, 2, 4-trimethy1-bicyclo[4, 2, O]octa-l, 3, 5- triene (E). Compound (E) was then hydrolyzed with ethanolic silver nitrate to the corresponding ketone (F). cm, 1750 / If} Cl Ethanolic> \‘ éom) N2 c1 \ A3N03 c1 / 1 D E F The structural assignments of E and F rest only on their micro- analysis, infrared spectra, ready hydrolysis of E to F, and analogy with previous work (43). The unusual elimination of hydrogen chloride seems to be capable of further extension. lll Expe rimental Preparation of 3, 7, 7-Trichloro- l, 2, 4-trimethylubicyclo- [4, 2, Ojocta-l, 3, 5-triene To a 100-ml. round-bottomed flask equipped with a side arm for nitrogen inlet and a take-off for gaseous outlet, there was placed 5.0 g. of 4-chloro—2, 3, 5, 6-tetramethylbenzotrichloride. The outlet tube was connected to a trap containing a 3% sodium hydroxide solution. The reaction vessel was immersed in an oil bath heated to 1750 and nitrogen was passed through the system. After heating for 3 hours 0.75 equi- valents of hydrogen chloride had been liberated. (Chloride ion determined by the Fajans' method at pH 8.) The flask was heated for another hour at 1750 and then cooled to room temperature. The brown oily residue, 4. 0 g. , was dissolved in anhydrous n-pentane and distilled under reduced pressure. Three fractions were collected: 1 13.13. 94-100°/o.s mm, n25 1.5730, 1.68 g.; D 11 B.P. 102-104°/0.5 mm, n3 1.5709, 1.58 g.; III B.P. 104~114°/0.5 mm, n3 1.5746, 0.6 g. On standing fractions I and II solidified and had m.p. 's of 37-390 and 40. 0~40. 50 respectively. The infrared spectra of bothfractions were identical. . Fractions I and II were combined and recrystallized from anhydrous n-pentane to yield 3.10 g.» (69%) of 3, 7, 7-trichloro-l, 2, 4- trimethylbicyclo[4, 2, 0]octa-1, 3, 5-triene, m. p. 40.0-40. 5°. Anal. . Calcd. for C11H11C13: C, 52.94; H, 4.44; CI, 42.62. Found: C, 53.07; H, 4.46; Cl, 42.40. Its infrared spectrum is shown in Figure XXXII. Preparation of 3-Chloro- 1, 2, 4-trimethy1-bicyclo- [4, 2, Ojocta- l, 3, 5-triene-7none A sample of 0. 50 g. of 3, 7, 7-trichloro-l, 2, 4-trimethyl—bicyclo- [4, 2, 0]octa- l, 3, 5otriene and 0. 71 g. of silver nitrate were dissolved in illlll’lll'i 112 6:33am .m .Humuoomo .N .iofioaofiaflumgwpuav .N . H nonofigomuwnn .N. .m Ho 5930on coumuwcm , .HHXNX oudmfim Amcouuflhv Suwcmfimxrm? 4: NA 2 OH «0 m N. e m 2w m —D —> _ _ _ d m A _ _ 113 20 m1. of 80% aqueous ethanol. A precipitate of silver chloride formed immediately. After stirring the solution for one hour at room tempera- ture the silver chloride was filtered. The filtrate was evaporated to dryness, the residue dissolved in anhydrous ether and filtered again. The ether was evaporated and yielded 0. 35 g. (89%) of crude product. The material was recrystallized from aqueous-ethanol and then sub- limed to give white needles of 3-chloro-l, 2, 4-trimethyl-ubicyclo[4, 2, 0]- octa-l, 3, 5-triene-7-one, m.p. 182. 5-1840. £331. . Calcd. for CllHnClO: C, 67.87; H, 5.70; Cl, 18.21. Found: C, 67.81; H, 5.79; C1, 18.23. Its infrared spectrum is shown in Figure XXXIII. 114 «L 1m .m J smuoomo .N .iofioiwofignuogwuuuv .N £198.20an wo 8.9.30on ponmuwGH NH : OH Ammo.“ 35v numrodmcrm? w w h . emote 1653.3 .230? 25E q _ _ A _ _ —1>Ln 115 ' LIT ERATURE CITED 1) H. Hart and R. W. Fish, J. Am. Chem. Soc., _8_2_, 5419 (1960). 2) H. Hart and R. W- Fish, J. Am. Chem. Soc., in press. 3) C. Friedel and J.- M. Crafts, Compt. rend., 84, 1450 (1877). 4) C. Friedel and C.. Vincent, Bull. soc. chim., (2), 36, 1 (1881). 5) E. Fisher and 0. Fisher, Ann., 191, 242 (1878). 6) M. Gomberg and O. W. Voedisch, J. Am. Chem. Soc., _2_3_, 177 (1901). 7) N. E. Tousely and M. Gomberg, J.' Am- Chem. Soc., 26, 1516 (1904). 8) M. Comberg and J. D. Todd, J. Am. Chem.‘ Soc., 32, 2392 (1917). 9) J. Boeseken, Rec. trav. chim.,_2_4_, 1 (1905). 10)C. Buehler, unpublished results. 11) G. Baddeley and A. G. Pendleton, J.. Chem. Soc., 807 (1952). 12) R. J. Gillespie, Rev. Pure and Appl. Chem., _9_, 1 (1959). 13) E. A. Brande and F- C. Nachod, "Determination of Organic Structures by Physical Methods», " Academic Press Inc. , New York, N. Y. , 1955, pp. 150“].510 14) M. Ballester and J. Casta‘fier, J. Am. Chem. Soc., _8_2_, 4259 (1960). 15) E. A. Braude and F.. C.- Nachod, loc. 93.2" p. 158. 16) L- M. Jackman, "Applications of Nuclear‘Magnetic Resonance Spectrosc0py in Organic Chemistry, " Pergamon Press, New York, N- Y., 1959, p. 47. 17) L.‘ 1. Smith and C.-L.~Moy1e, J. Am- Chem.. Soc. , 25, 1676 (1933). 18) M. S.- Newman and H.~ Lloyd, J. Am. Chem. Soc., _7_4_, 2672 (1952). 19) L.-I. Smith and C. L-Moy1e, J. Am. Chem., Soc., _5_§_, 1 (1936). 20) L. I. Smith, Organic Syntheses, Coll. Vol.11,. p. 254. 21) C.‘ E.- Ingham and G..- C. Hampson, J.- Chem. Soc. , 981 (1939). 22) G- Grassini, G. Illuminati, and G.. Marino, Gazz. chim. ita1., 86, 1138 (1956). 23) R.‘ Adams and N.‘Kornb1um, J. Am. Chem. Soc., 6_3_, 188 (1941). 116 24) L.‘ 1. Smith and F.- L. Taylor, J. Am. Chem. Soc., 51, 2370 (1935). 25) L. 1. Smith and F. L. Taylor, J. Am. Chem. Soc. , _5_7_, 2460 (1935). 26) H. O. Wirth, Ann., 634, 97' (1960). 27) R. W. Fish, Ph. D.. Thesis, Michigan State University, 1960, p. 107. 28) W. C. Pierce and E. L.. Haenisch,~ "Quantitative Analysis, " John Wiley and Sons, Inc., New York, 1950, p. 299. 29) N- C. Deno, J. J. Jaruzelski, and A. Schriesheim, J. Am. Chem. Soc., 17, 3044 (1955). 30) V. Gold and B. W- V. Hawes, J. Chem. Soc., 2102 (1951). 31) N. F. Hall and W. F. Spengeman, J. Am. Chem. Soc., 62, 2487 (1940). 32) M. A. Paul and F. A. Long, Chem- Revs., _5_'_7_, 15 (1957). 33) E. Weitz and F. Schmidt, Ber., _7_5_B_, 1921 (1942). 34) J. Fleming, unpublished results. 35) A. Haller and A. Guyot, Bull. soc. chim., _3_1_, 798 (1904). 36) F. Kehrmann, R- Monnier and M. Ramm, Ber., 26B, 173 (1923). 37) C. K.‘ Ingold and P. C. Marshall, J. Chem. Soc., 3083 (1926). 38) E. Barnett, J. W. Cook and J.- L. Wiltshire, J. Chem. Soc., 1724 (1927). 39) C. F. Koelsch, J. Org. Chem., 3, 456 (1938). 40) E. Clar and W. Willicks, J.. Chem. Soc., 942 (1958). 41) J- Thiele and H. Balhorn, Ber., _31, 1468 (1904). 42) G. J. Sloan and W. R. Vaughan, J. Org. Chem., 22, 750 (1957). 43) H. Hart and R. W- Fish, J. Am. Chem. Soc., _8_2, 749 (1960). i21