DiCAREGNWM iONS FROM StSoARYL CARBENQLS Thesis for “M Dawn 0‘ DE. 0. MECH‘EGAN STATE U 'EVERSET? Theodore Sulzberg 1962 This is to certify that the thesis entitled IDICARBONIUM IONS FROM BIS-ARYL CARBINOLS presented by Theodore Sulzberg has been accepted towards fulfillment of the requirements for _Ph'_D°. degree in Mr Y 41 ‘ Date 4K1M¢ ‘7‘ 1661.— 0-169 Major professor LIBRARY Michigan State University w , ._..-.——_- _._ 4 L3334%ARY Méfihtm s. Lon-fl UfiWE’W‘Ey’ 3P3 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 cz/CIRCJDateDuepeS-p. 15 ABSTRACT DICARBONIUM IONS FROM BIS-ARYL CARBINOLS by Theodore Sulzberg The main purpose of this thesis was to investigate compounds which could form stable dicarbonium ions by ionizations at separate sites in molecules (1). By use of the Friedel-Crafts benzoylation and Grignard reactions on the appropriate aromatic molecules the following types of compounds were prepared where R = phenyl or hydrogen, Y = hydroxyl or chlorine and X = benzene, biphenyl or fluorene. All of the compounds studied dissolved in 98% sulfuric acid to form highly colored solutions ranging from yellow to blue. Investigation of the visible absorption spectra of solutions of these compounds in varying concentrations of sulfuric acid has shown that reversible ionization to dications can occur either stepwise or simultaneously and is, in general, a function of the moiety between the sites to be ionized. The pKR's for these processes have also been determined and are given in Table 1 along with the xmax and emax of the dications. In order to interpret the visible spectra of these dications, a series of compounds were prepared which ionized to give monocations: Y Theodore Sulzbe rg 2 Table 1. Spectral Data for the Bis-Aryl Carbinols in Sulfuric Acid w a a Compound xmax €max pKR (mix) 1 Z a,a,a', o.'-Tetraphenyl-p- 455 59,000 -8.1 -10.5 xylene-a, a'-diol a, a, a', o.‘-Tetrapheny1-m- 440 70, 000 -8. 0 - 10.1 xylene-0., a'-diol 419 73, 000 a, a, a', a'-Tetrapheny1-o- 455 44,000 N-8.0b -16.6 xylene-o, a'-diol 373 30, 500 0., e'- Diphenyl-p-xylene- 461 48, 000 c -a, o.‘-diol 0., a'-Dipheny1-m-xylene- 447 35, 000 c a, o.’-diol a, a, a', o.‘-Tetrapheny1 530 90, 000 -8. 2 d p, p‘-bitolyl-o., a' -diol 442 53, 000 4108 34,500 0., a, a', a'-Tetraphenyl- 430 67, 500 -8. 4 d m, m'-bitoly1-a, a-diol a,a,o.',o.'-Tetraphenyl- 428 35,500 N -8.3eN -15.0 o, o'-bitoly1-a, a.‘ -diol ‘ 390‘s 24, 000 0., o,'-Dipheny1-p, p' - 563 134, 000 c bitolyl-a, a'-diol 420 13, 000 a, o.‘-Dichloro-a,a'-dipheny1- 603 215,000 N -12.0 -16.6 2, 7-dimethylf1uorene 554 62, 000 0., a, a', a' -Tetraphenyl-2, 7- 568 152, 000 c dimethylfluorene-a, o.‘-diol 525 46, 000 440 35,000 aThe x and 6 data were obtained in 98% sulfuric acid. max max The first pK was assumed to be the same as for the corresponding meta and para dio s. The close proximity of the second -OH after mono-ionization permits cyclic protonated ether formation. This species, which was not detectable by visible spectrophotometry, required solutions of highacidity for further ionization. c Not determined. In these diols, ionization occurred simultaneously at both sites. This compound ionizes Similarly to the diol discussed in b. The first ioni- zation was assumed to be the same as the corresponding meta and para de- rivatives. Theodore Sulzberg Table 2. Spectral Data for the Aryl Carbinols in Sulfuric Acid w . a a Compound xmax Emax . pKR (mp) Triphenylmethanol 432 37, 500 - 7 . 4 408 36, 500 Diphenylmethanol 442 47, 000 - 14. 7 b 4-Biphenylyldiphenylmethanol 510 43 , 000 - 7 . 7 420 21,500 4-Biphenylylphenylmethanol 535 90, 000 c u-Phenyl-a-hydroxy-Z-methyl- 547 66, 000 N - 12. 0 fluorene 0., a-Diphenyl—a-hydroxy- 540 53, 000 -6. 5 Z-methylfluoreneb 404 17, 000 aThe x and 6 data were obtained in 98% sulfuric acid, except ma max where ndcted. These spectra were taken in 80% sulfuric acid because sulfonation was shown to occur in concentrated acid. c . Not determined. Theodore Sulzberg The spectral data for these compounds are given in Table 2. Comparison of the spectra of the mono- and dications has shown that the resonance interactions of the latter can be pictured as being composites of mono- cations. In addition to spectral studies, the hydrolysis products of the dication solutions in concentrated sulfuric acid were also investigated. In all cases studied the starting diols (or cyclic ethers in the case of ortho substituted diols) were isolated in nearly quantitative yields. The hydrolysis of two carbinols (4-biphenylyldiphenylmethanol and (1, a-diphenyl-a-hydroxy-Z- methylfluorene) in 98% sulfuric acid showed that sulfonation occurred after carbonium ion formation. The alcoholysis of solutions of a, a, a', a'-tetra- phenyl-o-xylene-a, o.‘-diol in concentrated sulfuric acid gave the corres- ponding ether of 9, 10, lO-tripheny1-9, 10-dihydro-9-anthrol. ‘ The study of trans-9, 10-diphenyl-9, lO-dihydroxy-9, 10-dihydro- anthracene in sulfuric acid was continued (1). It was demonstrated that in dilute sulfuric acid ionization of this diol gave the monocation which re- arranged to the cis-diol. By increasing the acid strength the cis-diol rearranged to 9, 10~dipheny1anthracene, which in more concentrated sulfuric acid gave 4-pheny1-2, 3-benzof1uoranthene. At no time, however, was the presence of a dication noted. REFERENCES 1. R. R. Rafos,.Ph. D. Thesis, Michigan State University, 1961. DICARBONIUM IONS FROM BIS-ARYL CARBINOLS BY Theodore Sulzberg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOC TOR OF PHILOSOPHY Department of Chemistry 1962 ACKNOWLEDGMENT The author would like to express his appreciation to Doctor Harold Hart for his encouragement and expert guidance in conducting this research. Grateful acknowledgment is also extended to the author's wife Edith for her patience and encour- agement at all times. Acknowledgments are extended to the National Science Foundation who provided financial support for this work and to Michigan State University for a Graduate Teaching Assistantship. >‘:<>i<>:<>i<>i<**>l<*>i<**>l<* ii TABLE OF CONTENTS INTRODUCTION ..... . . .................. RESULTS AND DISCUSSION . . . . . .............. I.. Syntheses and Structure Proofs ............ II. Visible Spectra in Sulfuric Acid and pKR Studies . . . A. The Xylene Diols ................ B. The Biphenyl Diols and Carbinols ........ C. The Fluorene Diols and Carbinols ........ D. The Anthracene Diols and Carbinols . ..... III. Interpretation of the Spectra .............. A. The Xylene Diols . . . ........... B. The Biphenyl Diols and Carbinols ........ C. The Fluorene Diols and Carbinols ....... D. The Anthracene Diols and Carbinols ...... E. Molecular Orbital Approach. .......... IV. Hydrolysis and Methanolysis Products ......... V- Cryoscopic Measurements ............... EXPERIMENTAL ........................ I.. Syntheses and Reactions ................ A. The Xylene Diols ................ Preparation of 0., a, a', o' -Tetrapheny1-o- xylene-u, a'-diol ........ . ..... Preparation of a, a, a', o.‘ -Tetraphenyl- phthalan.... ...... Reaction of 0;,u, a', o.‘-Tetraphenyl-o-xy1ene- a, a'-diol in 98% Sulfuric Acid with Water iii 26 27 32 33 43, 56 64 75 77 82 88 93 96 98 104 106 107 107 107 107 110 TABLE OF CONTENTS -» Continued Page Reaction of 0, 0., o', a'-Tetrapheny1-o-xylene-0., 0.'-diol in 98% Sulfuric Acid with Absolute Methanol ..... 110 Reaction of Absolute Methanol with 0., 0., 0.', 0.'-Tetra- phenyl-o-xylene-a, 0.'-diol in 98% Sulfuric Acid . . . 112 Reaction of 0, 0., 0' , 0.‘-Tetraphenyl-o-xy1ene-a, 0.'-diol in 98% Sulfuric Acid with Absolute Ethanol ..... . 112 Reaction of 0., a, a', a' -Tetrapheny1-o-xylene-0., 0.'-diol in 98% Sulfuric Acid with Anhydrous Tetrahydrofuran 112 Preparation of m- Dibenzoylbenzene ........... 114 Reaction of m- -Dibenzoy1benzene with Phenylmagnesium bromide ........................ 114 Preparation of 0., 0.'-Dichloro-0., 0., 0.', 0.‘-tetraphenyl-m-fi xylene ......................... 114 Preparation of 0., a, 0.', 0.‘-Tetrapheny1-m-xy1ene- £2 ................. Theoretical plot of A versus HR for stepwise dication formation where 62 >61, ................. The relative energies of the lowest antibonding, non- bonding and highest bonding 1r -orbitals of an aromatic hydrocarbon (ArH), and of the symmetrical diaryl- methyl (ArzCH) and triarylmethyl (Ar3C) systems . . . Infrared spectrum of 0, 0, 0', 0'-tetrapheny1-o-xylene- 0,0'-diol............... .......... Infrared spectrum of 0, 0, 0', 0'—tetraphenylphtha1an . . Infrared spectrum of the methyl ether of 9, 10, 10- tripheny1-9,10-dihydro-9 -anthrol ........... Infrared spectrum of the ethyl ether of 9, 10, 10-tri- pheny1-9, lO-dihydro-9-anthrol ............. Infrared spectrum of 0, 0'-dichloro-0, 0, 0', 0'-tetra- phenyl-m-xylene .................... Infrared spectrum of 0, 0, 0', 0'-tetrapheny1-m-xy1ene- 0, 0'-diol . . . . ............ . ........ Infrared spectrum of the dimethyl ether of 0, 0, 0', 0'- tetraphenyl-m-xylene-a, 0'-diol ..... . . . . . . . . Infrared spectrum of 0, 0'-dichloro-0, 0, 0', 0'-tetra- phenyl-p, p'—bitolyl ............. . . . . . Infrared spectrum of 0, 0, 0', 0'-tetrapheny1-p, p'- bitoylyl-a, 0'-diol ................... xi Page 76 78 79 97 108 109 111 113 116 117 119 122 124 LIST OF FIGURES - Continued FIGURE 35. Infrared spectrum of 4-biphenylyldiphenylmethanol . . 36. Infrared spectrum of 2-biphenylyldiphenylmethanol . . 37. Infrared spectrum of 4-biphenylylphenylmethanol . . . 38. Infrared spectrum of 4-bromo-4'-benzoy1bipheny1. . . 39. Infrared spectrum of 4-(p-bromophenyl)dipheny1- methanol ......................... 40.. Nuclear magnetic resonance spectrum of 4-(p-bromo- 41. 42 43. 44.. 45. 46 47. 48. phe nyl) diphe nylm ethanol ................ Nuclear magnetic resonance spectrum of the methyl ether of 4-(p—bromophenyl)diphenylmethanol ...... .0 Infrared spectrum of the solid isolated from the re- action of 4-(p-bromophenyl)diphenylmethanol and mag- nesium with benzophenone ...... . . . . . . . . . Infrared spectrum of 0-chloro-0-pheny1-2-methy1- fluorene ..................... Infrared spectrum of 0, 0-dipheny1-0-hydroxy-2- methylfluorene .............. . ..... Infrared spectrum of the hydrolysis product of 0, 0- diphenyl-a—hydroxy-2-methy1fluorene in 97% sulfuric acid .................... . . . . . . . .. Infrared spectrum of 0, 0, 0', 0'-tetraphenyl-2, 7—di- methylfluorene - 0, 0-diol ................ Infrared spectrum of 9, 10, 10-tripheny1-9, 10-dihydro- 9-anthrol ......... . .............. Infrared spectrum of 9—chloro-10, 10-diphenyl-9, 10- dihydroanthracene ................... xii Page 126 128 129 132 134 135 137 138 144 145 147 149 151 153 LIST OF FIGURES - Continued FIGURE Page 49. Infrared spectrum of trans-9, 10-dipheny1-9, lO-di- hydroxy-9, 10-dihydroanthracene ............ 155 50. Infrared spectrum of cis-9, 10-diphenyl-9, lO-dihydroxy- 9, 10-dihydroanthracene ................. 156 xiii INTRODUCTION INTRODUCTION Carboniurn ions have long been recognized as one of the most important species in organic chemistry. Their existence both as transient intermediates and as stable species has been well documented (1, 2). The most thoroughly studied of the latter are the diaryl-and triarylmethyl cations. 1.. Rever sible Ionization They can be formed in several ways: Ar Ar a. Ar — c'; - OH + 2H250,——-> Ar -c':+ + H3O+ + Hso,' (1) I v— I Ar Ar Air+ Ar Ar _ b.Ar-C-C17-—-> Ar-C ---------- C1—--‘~Ar-C++C1 (2) | v:— I x—— Ar Ar Ar 2. Protonation of Unsaturated Center a. 3. Protonation of Carbonyl Compounds 0 Ar-& Olefins Ar Ar {—— Ar ' Ar . Aromatic Hydrocarbons H+ __._\ f— + + :c — CH, + HSO,’ HH (3) (4) H -Ar+H.zSO4—-> Ar-C—Are Ar-g-Ar +HSO¢- (5) + It is with carbonium ions formed by the reversible ionization of alcohols and halides that this thesis deals. In fact, the major part of this thesis concerns molecules which produce doubly charged ions by single ionizations at two separate sites, i. e. , "ordinary" dipositive ions (3). Before discussing dipositive ions, it seems appropriate by use of selected examples to give a brief account of mon0positive aryl- methyl cations . I. Monopo sitive Cations The study of arylmethyl cations began at the turn of the twentieth century when interest in triphenylmethane dyes was high. The physical evidence for the existence of these stable species under equilibrium con- ditions came from several sources: 1. Conductivity Measurements In 1902, Walden (4) and Gomberg (5) established that various tri- arylmethyl halides ionize in liquid sulfur dioxide in a manner similar to methylammonium chloride or potassium iodide (see equation 2). By using electrical conductivity, later workers showed that the relative effectiveness of various groups in promoting the ionization of triaryl- chloromethane is p-methoxy > p-methyl > p-phenyl > hydrogen > p-nitro (6). Recently Lichtin and Glazer (7) studied by conductance in liquid sulfur dioxide, the ionization of triphenylchloromethanes with phenyl, t-butyl and methyl groups in the meta and para positions. Their findings showed that ionization was enhanced by para substitution. A number of possible sources of diarylmethyl cations have been studied in sulfur dioxide and benzhydryl chloride, m-chlorobenzhydryl chloride andp,p1-dimethylbenzhydryl chloride have been found not to conduct (8) but dimesitylmethyl chloride does (9). 2. Cryoscopy The extent of ionization of a solute in a particular solvent (one capable of producing ions with the solute in a reversible way) can be found because the freezing point depression of the solvent is propor- tional to the number of particles produced. Perhaps the most useful solvent for such determinations is sulfuric acid, because most com- pounds behave as bases in it and it has a convenient melting point (10. 371°) and a sizable molal freezing-point depression constant (6. 120 mole-1kg.-l). A review on the behavior of organic compounds in sulfuric acid has recently been published (10). Alcohols ionize in sulfuric acid in one of the following ways: (i = van't Hoff factor = number of particles formed per molecule of solute) + - + .. ROH 'l" ZHzSC)‘ —'_} RHSO4 ‘1‘ H3O 'l' H504 ( 1 = 3 ) (7) V'— ROH 'l" ZHZSO4 —_“ R+ ‘1' H30+ '1' H504- ( 1 = 4 ) (8) Since we are interested in the formation of arylmethyl cations, it is equation 8 that is important (see equation 1). This ionization may occur in a stepwise fashion: ,ROH+ sto, —-» ROH: + H50," (9) + += H20 + sto, ——->- H3O+ + H50; (11) _ <— The first reported cryosc0pic work on triarylmethanols was by Hantzsch in 1908 (11). . He obtained values slightly greater than three. Recently,. Newman and Deno(12) studied the following methanols in 100% sulfuric acid and their results strongly indicate the formation of stable carbonium ions: tri—p-tolyl, tri-o—tolyl-, tri-p-aminopheny1-, tri-p-dimethylaminophenyl, di(p-dimethylaminophenyl) pheny1-, tri-p- chlorophenyl- , di-p- chlorophenylm ethyl- . 3. Absorption Spectra In 1902, Walden (4) noted that the yellow solution formed by dis- solving triphenylmethanol in concentrated sulfuric acid had an absorp- tion spectrum identical with that of a solution of triphenylmethyl chloride in liquid sulfur dioxide. . It is reasonable that if a certain spectrum always appears in conducting but not in non-conducting solutions of a compound, and if the absorption intensity parallels the electrical conductivity, then the spectrum can be assigned to one of the ions present. Walden who had observed both of these phenomena, assigned the absorp- tion in the visible region to the triphenylmethyl cation, since it was common to both systems. Recently, Newman, Deno and co-workers (12, 13, 14) studied the visible spectra of various diaryl- and triarylmethanols in concentrated sulfuric acid. I In all cases the diarylmethyl cations absorbed at 16-25 mp. higher wavelength than the corresponding triarylmethyl cations. Newman and Deno explained this in either of two ways. First, there is steric inhibition to resonance so that the third ring contributes-little to the electronic structure of the carbonium ion and resonance interactions are restricted to only one or two rings. Secondly, each of the three rings in the arylmethyl cations makes an equal contribution to the resonance but this contribution is lessened because of an increase in the angle of ring twist). This twist is greater with triarylmethyl cations than with diarylmethyl cations. More recently, Deno and co-workers (15) have done LCAO-MO calculations (neglecting electron repulsion and overlap integrals) which predict that mono-, di-, and triphenylmethyl cations should possess nearly identical wavelength maxima in the 400-440 mu region. Grinter and Mason (16, 17) studied the steric and symmetry effects in the spectra of arylmethyl cations. I Within the limitations imposed by neglecting ion-pairing, solvent effects, steric hindrance to conjugation in the arylmethyl ions, and the contribution of electron repulsion changes to the transition energy, it is to be expected that corresponding di- and triarylmethyl ions, independent of the position of substitution of the methyl group in the aromatic nucleus, and of whether the ion is positively or negatively charged, should absorb at a wavelength approxi- mately twice that of the aromatic hydrocarbon from which the ion is derived. This can be stated since by use of the Hiickel approximation, Grinter and Mason showed that the highest bonding and lowest antibonding levels have the same energy as the respective levels in the aromatic hydrocarbon from which the ion is derived. An example can be seen in the absorption bands of some biphenyl substituted carbonium ions (wave- lengths in mp. Q+©@‘©@.\/©Q/‘ 0,, 0,, 13 m I oceégaotgo :0 “Es @ iz SZI m 5|O - 5l2 555 Since the position of the longest wavelength band of biphenyl lies at 250 mu, it is expected that the maxima for the above cations should occur at 500 i 10 mp. Grinter and Mason state that compound III}, as they expected, showed an absorption maximum at a wavelength lower than 500 :l'. 10 mp. . For a possible clarification of this point see page 45. The higher wavelength absorption of (VII) than (V) was explained by the better ground state stabilization of (VII). . More will be said about the biphenylmethyl cation in the discussion of the results of this work. 4. Isolation of Stable Salts In 1909, Hofman-n and Reimsreuther (18) isolated triphenylmethyl perchlorate as a yellow solid by treating triphenylmethyl chloride with 71% perchloric acid in either nitrobenzene or acetic anhydride followed by evaporation of the solvent in a desiccator. Gomberg also succeeded in isolating the perchlorate (19, 20). Treatment of triphenylmethyl chloride with acetyl tetrafluoro- borate in acetic anhydride afforded triphenylmethyl tetrafluoroborate (21). The same salt was prepared by the addition of boron trifluoride to triphenylmethyl fluoride (22). In 1957, as part of a study of complex fluorides, Sharp and Sheppard (23) allowed the appropriate silver salt to react in ether with triphenylmethyl chloride to form the following triphenylmethyl cation salts: hexafluorophosphate, hexafluoroarsenate, hexafluoroantimonate, hexafluoroniobate, hexafluorotantalate, fluorosulphonate and tetrafluoro- borate. By condensing an excess of stannic chloride or antimony pentachloride into a benzene solution of triphenylmethyl chloride, the pentachlorostannate and hexachloroantimonate salts of triphenylmethyl cation were prepared. , Since much discussion has appeared in the literature concerning the conformation of triphenylmethyl cation, these workers used infrared spectroscopy as a means of determining it. The structures considered were the D311, (all-planar), D3 (propeller-like) and unsymmetrical conformations. By examining the infrared region between 1700 and 650 cm'1 of several crystalline salts of triphenylmethyl cation, toluene, triphenylmethane, and triphenylboron, Sharp and Sheppard concluded that D3 symmetry best described the configuration of the triphenylmethyl cation. In the spectrum of toluene there is a strong band at 728 cm"1 which corresponds to the vibration in which all five aromatic C-H bonds bend out of the plane of the ring in unison. This band is quite sensitive to structural changes. . In triphenylmethyl cation this band is split into components at 810 and 770 cm'1 (774 and 1 is presum- 747 cm'1 for triphenylboron). This large splitting of 40 cm' ably caused by interaction of the aromatic rings with the cation. Karagounis (24) has considered the analogous problem with triphenylmethyl radical and has shown that D3h structure would have only one such infra- red band, the unsymmetrical configuration three such bands and the D3 two bands. This and several other features of his observed infrared spectra can only be accounted for by propeller-like symmetry. In 1958 Meerwein and co-workers (25) studied the preparation, complexing ability and use of silver and copper (I) tetrafluoroborates. They described the preparation of tri-anisylmethyl tetrafluoroborate by the treatment of trianisylmethane in ethylene dichloride with ethyl bromide and silver tetrafluoroborate. The red salt was obtained in 30% yield. Dauben and co-workers have recently reported simple and con- venient methods for obtaining the perchlorate and tetrafluoroborate of triphenylmethyl cation (26). Treatment of triphenylmethanol with acetic anhydride and 71% perchloric acid at room temperature gave the per- chlorate in 85% yield. The corresponding tetrafluoroborate was produced almost quantitatively by allowing triphenylmethanol to react with fluoboric acid in propionic anhydride at 200. The former salt was shown to be less stable since it turned dark and decomposed to 9-phenylfluorene (6%) during two weeks in the dark while the tetrafluoroborate was com- pletely stable. In 1961 Harmon prepared and characterized the first carbonium tetrabromoborate (27). The reaction of equimolar amounts of triphenyl- methyl bromide and boron tribromide in cyclohexane resulted in a nearly quantitative yield of triphenylmethyl tetrabromoborate. Unlike other triphenylmethyl cation salts, the tetriodoborate (28) was described as being so viciously hygroscopic and extremely light sensitive that a satis- factory analysis could not be obtained. The evidence for its existence was based on B11 n.m. r. spectra and on a rapid hydride exchange re- action with cycloheptatriene. The latter experiment gave a 57% yield of tropenium tetraiodoborate and a 95% yield of triphenylmethane. 5., Quenching Experiments Another important piece of evidence for the reversible formation of triarylmethyl cations in sulfuric acid is the quenching of such solutions with water and methanol. A Newman and Deno (12) have shown that various triarylmethanols are recovered unchanged when their sulfuric acid solu- tions are poured onto ice. When a solution of a diarylcarbinol in concen- trated sulfuric acid is poured onto ice, the corresponding bis-diarylmethyl ether is formed (29). . When a solution of a diarylalkylmethanol in sulfuric acid is hydrolyzed, the corresponding olefin is obtained (30): Ar + Ar \C - CH3 + H20 —._.> \ + Ar/ V'— Alf - CH2 '1' H30 (13) The methyl ether corresponding to the diaryl- or triarylcarbonium ion can be formed by pouring the sulfuric acid solution into cold methanol: + + Ar3C+ ‘l' CH3OH —-_} [ AI'3C " ? " CH3 l—A AI3COCH3 + H (14) 6. Nuclear Magnetic Resonance (31) Nuclear magnetic resonance has become a useful tool in determining the fine structure of molecules. One can ascertain the relative amount of positive charge on an atom in a given compound by studying the hydro- gen bonded directly to that atom. . By comparing the chemical shift of such hydrogens with various references, the degree of positive charge is discernable. The first detailed study of the n.m. r. spectra of triarylmethyl cations was carried out by Stewart and his group (32). By dissolving arylmethanols in a mixture of trifluoroacetic acid-trifluoroacetic anhydride, the appropriate carbonium ions were obtained (as determined by visible spectrophotometry). The carbinols studied are shown in Table 1: Table 1. . N.M. R. Data of Triphenylmethyl Cationsa’ b I R C R5 R1 R2 R3 Aromatic Shift Substituent Shift H H H -101. 0 p-Me H H -98.0 +102. 5 p-Me p-Me H -94. 75 +104. 5 p-Me p-Me p-Me -93. 5 +105. 5 p-OMe H H -96. 0 +42.0 p-OMe p-OMe p-OMe -87. 5 +47. 0 m-Me H _ H -95. 5 +112.5 m-Me m-Me m-Me —89.0 +112. 5 Anisole -68.0 +59. 5 Toluene -69. 0 +124.0 (a) All spectra run in trifluoroacetic acid-trifluoroacetic anhydride as solvent. (b) Measured against water as external standard at 40 mc. 10 As can be seen in Table 1, successive substitution of methyl or methoxyl groups at the para positions increases the shielding on the methyl groups. 7 Since the methyls appear as singlets in each compound, it is likely that the three rings are in the symmetrical propeller conformation. The meta methyls, as expected, carried a much smaller amount of positive charge. A study of the aromatic multiplet led Stewart to conclude that the para hydrogen is at lowest field with the meta next and the ortho at highest field. Vaughn and his co-workers (33) have also studied the question of charge distribution in triarylmethyl cations. They synthesized the follow- ing deute rated carbinol s: cw w w The n.m. r. spectra, which were run in liquid sulfur dioxide containing stannic chloride, showed conclusively that the greatest amount of positive charge is located at the para positions, since the para protons appeared at lowest field. The smallest share of the positive charge is at the ortho positions with the meta positions having slightly more. These findings are in excellent agreement with the prediction of self-consistent molecular orbital theory, as done by Pople (34), which gives the following negative charge densities: ortho, 0.95; meta 0.94; and para, 0.81. These are the methods that have been used to study the stable arylmethyl cations containing but a single positive charge. They demon- strate with a high degree of certainty the structure of these species. These types of systematic studies, however, have not been extended to molecules containing more than one positive charge. 11 II. Dipositive Cations A search of the literature has revealed that there is a limited amount of material on molecules that are polycarbonium ions. Representative examples of each group are: R 0 “'5 a. @219 0.5 0 0i W W och X11 + Eli H (Vie X=O orS :XII XIII Me Me ., Q. .Q Q ’—* Me @— -Cl —©—C * Me Me 6 Q 2321 /\ /\ XI XSZII l. Dixanthylium Salts and Related Compounds In 1901, Werner (35) reported that the reaction of xanthone with zinc andphydrobromic acid gave XI (X = 0) with the anion being the tri- bromide ion. A year later, Werner (36) showed that the following reactions each gave the same tetrabenzodixanthylium salt: 12 (15) Similar types of dications were reported by Arndt and co-workers (37, 38): (16) Schgnberg (39) has done more extensive work on the preparation of the salts and has isolated the explosive dixanthylium diperchlorate (XI, X = 0) by treatment of xanthone with zinc, acetic acid andsodium perchlorate. . He also reports the following reaction (40): écc'i 95..., 8 er / or bronze Cl Ci Cl 13 Sulfur and chlorine analyses that were each more than 1% low, were his only evidence for this product. . Wizinger and Al-Attar (41) reported the following dipositive carbon- ium ion. in the xanthylium series: 0.0 321:2 E .+ CH ’ 213- """" ' (18) These workers found that by varying the conditions, either a mono- or dication could be formed from dixanthylene: 2 Foo; Qfi Fcle,’ ECl2 0 ~ Fecl3 Q+0 FEET: ('19) 2. Bisacridinium Salts By analogy, it would appear that by replacing the oxygen in the dixanthylium salts with nitrogen, a new type of dipositive carbonium ion should be formed (XII). These salts are known and two methods for their preparations are (42): l4 (20) When a 0. 05% solution of XVIII (X = N03) in aqueous alkali is treated with hydrogen peroxide, a light stronger than daylight results. The resulting species is the most powerful chemiluminescent substance known. For this reason, most studies of bisacridinium salts have been along these lines and crystalline materials have usually not been prepared. 3. Triphenylmethane Dyes Of all the dications discussed, only in this category is there 'a cation formed by having a positive charge originating at a site other than a carbon atom. In 1939,. Lewis and Calvin (43) showed that Crystal Violet could be . . _ . x . : ionized stepWise ( max in mu) LQMQN HQ)Z(:::\I@C 2 (MeZNH{_>)c +1 end absorption 630 ”W (21) 15 The absorption of the dication is very similar to that of XIII (Malachite Green) which has a x at 623 mu. max Similar effects were noted by Morton and Wood with XIX (44) and MRI/Me by Theilacker and Schmid with XX (45). M M HC—OH “\Z/m inc-0H e G M XIX: 3K Me’N\Me M In an attempt to explain the visible spectra of dyes, Lewis and his co-workers (46, 47, 48) formulated rules for predicting the position of the main absorption. They stated that an excited state produced by the absorption of light is characterized by an oscillation of the electronic cloud along the axes of the molecules. The main current is attributed to the x band in the horizontal direction with the second band in the y 1 (vertical) direction Me Me Mc-N©+ N-Me XXII x \ 7 Since the x band has better resonance stabilization, it is of lower energy (higher wavelength) than the y band. Recently, it has been found by Anthony- Barbier and Rumpf (49) that the y band of Malachite Green (XXI) 16 is actually three separate and distinct bands. The ionization of XXI occurs in a stepwise manner ( x max in mu): Me Me M. M. . hi. _ hi, N MeNMe MeN e C-OH -> Q? —> O“ —% O (22) © o d: g? MeN Me e Me Me/N\M 425.3 447.26 454.6 A review by Barker on steric effects in triphenylmethane dyes appeared recently (50). 4. Tetraarylethylene Compounds It has been known for some time that the interaction of halogens with arylethylenes gives highly colored solutions (51). Buckles and Meinhart (52) studied the electrical conductivities of methylene chloride ' solutions containing bromine and various substituted tetraphenylethylenes. The tetraphenylethylenes with methoxyl and dimethylamino groups in at least "two oflthe para positions gave highly colored solutions and had molar conductances of the same order of magnitude as tetrabutylammonium iodide. . No attempt was made to interpret the data in terms of the compo- sition of the complex or complexes. Buckles and Womerr (53) reinvesti- gated the fact that when bromine in ethylene chloride was mixed with tetra bis-(p-methoxyphenyl)-ethylene a blue solution resulted immediately. This solution had an absorption peak at 575 mu which was not characteristic 17 of the starting materials. On standing a new absorbing species resulted (490 mu). This latter species was shown to be the trianisylmethyl cation (53). More recently, the former species has been shown to be the tetraanisylethylene dication (54). It is postulated that this cation was produced in the following ways: An An An + + An ‘0 - c/ + 331- ——-> ‘c c/ (Br') (23) " ' a " 3 2 An/ \An An/ \An An\ /An An\+ + lAn /C = C\ + 4ICI ——> /C - C\ (ICIi’); + I; (24) An An An An An An in /An—1 \ / \ + .. An OH OH An LAn + + An A|n —An\ /Ar? + An - C - C - An + 3HZSO4 -—> /C - C\ + H3O + 3HSO4' (26) An 0 ‘ An + + An (where An = anisyl) The dication was isolated as dark, blue-green trihalide salts and analyzed in the examples shown in equations (23) and (24). Attempts to study the formation of the tetraanisylethylene dication cryosc0pically were not successful due to sulfonation of the anisyl rings (55). In 1961, Anderson and co-workers (56) reported on the reaction of tetraki's-fip-dimethylaminophenyl)-ethylene with silver nitrate to form green crystals of the dinitrate. This in turn reacted with potassium iodide to give the green.tetraki'sr-(p-dimethylaminophenyl)-ethylene diiodide- monohydrate. This lost water at 75° to give the black anhydrous diiodide. 18 Me Me Me Me - -(27) XXDI “<1 > anode. XXII (28> (XXIII) and (XXIV) in water are purple and have identical visible and ultraviolet spectra. The same is true of the dinitrate in ethylene chloride. In ethylene chloride, however, the diiodide is green and is the only solu- tion to show an e.p. r. spectrum (20% radical character). The following reaction shows the apparent fate of the dication: Me Me MQN N'Me m _ . C==C ‘____ .XXEDIz Tel jg I<—I MQ’N We (29) 19 The reason that the diiodide undergoes the above reaction was attributed to the non-polar solvent (shift toward less charged species), the very good donor properties of the iodide ion as well as the good acceptor qualities of the quinoidal dication. 5. Trichloromethylbenzenes Recently, Hart and Fish (57) observed that the following trichloro- methylbenzenes ionize in 100% sulfuric acid to give red solutions with van't Hoff i-factors of five: Me Me ’ Me Me MeQCC l3 ‘ _ MQ‘QQC' (30) Me Me Me Me e Me Me C03 2 11.3%,.) ZHCH’ ZHSOZTMQ‘QEQC 1(31) Me Me e 2 Me ‘Me Me Q cm, . M Q ccu + (32) Me e The evidence for these "extraordinary" dipositive carbonium ions (58a), which in a formal sense lost. two anions from a single carbon atom, was based on cryoscopy, stoichiometry, electrical conductance, visible, ultraviolet and nuclear magnetic resonance spectroscopy. Confirmatory evidence for the existence of these "extraordinary" cations is the preparation of their crystalline bis-tetrachloroborate and tetrafluoro- borate salts(59). 6. 0, 0, 0', 0'-Tetraphenyl-p-xylene Dication The fact that 0, 0, 0', 0'-tetraphenyl-p— xylene:-0, 0'-diol gave a red color when dissolved in concentrated sulfuric acid has been known for a 20 long time (60). It was only recently that a study of this reaction was undertaken by Rafos (61). With the use of several of the methods dis- cussed earlier, he showed that the diol ionizes in the following way: Ho-CQ -OH + 4sto4——> +C©C+ +4 HSO’© © <33) 4 + + 2 H3O Hart and Wu (62) have studied the n.m. r. spectra of the following dications in order to determine the relative amounts of positive charge on the different carbon atoms. @\+/¢ ¢\+/Q§ ¢ 0!. + at at o( [3 ot+ ’8 ‘I C—‘¢ .6? o 1 P CA? F °‘ Cin .4 m on o p P P XXXZ XXVI m The dications were generated by dissolving the corresponding diols in 100% deuterosulfuric acid and the spectra compared to methane sulfonic acid as an internal reference. The diols studied had deuterium placed in the appropriate positions so that the chemical shift of the different 21 hydrogens could be determined. Their results indicate that: l. The positive charge located in the outer rings of all three compounds are in the same ratio as in triphenylmethyl cation, i. e. , p > m > o. 2. The percentages of total positive charge in the outer rings of XXVI and XXVH are about the same but in XXV the outer rings carry a larger share of the positive chargedensities. 3. The following are the relative positive densities for the dications: XXV - p > m > 0 Z 0 XXVI-[3: 7Q: p>0>m>o XXVII—[3>p>0>m>o It should be noted that the [3 positions in XXVI and XXVH, which carry a large amount of charge, are located para to the position where the charge originates and would be expected to have a high positive charge density. Sloan (63) measured the spectra of some homologues of XVI in concentrated sulfuric acid, but since the main purpose of his work was the study of aryl stabilized diradicals, no discussion of the various dications was included. 7. Tetraphenylcyclobutenyl Dication In 1960, it was postulated that the formation of an aromatic system might be a sufficient driving force to allow dication formation (58b). The two examples proposed were the four- and the eight-membered "Hiickel- aromatic" system s: “—9 '3’") (34) (35) 22 In fulfillment of this postulate, Freedman and Frantz (64) succeeded in preparing the tetraphenyl cyclobutenyl dication. Treatment of 3, 4-di- bromotetraphenylcyclobutene with either concentrated sulfuric or perchloric acid led to a deep red solution with simultaneous evolution of hydrogen bromide. ,\ (36) __ _ H280, or, (w \a \ / Br Br\ / HCIO4 /\\ \// Hydrolysis of the red solution gave tetraphenylfuran. Other evidence for the dication included the formation of solid salts and the visible and n.m. r. spectra. . Work on the formation of an eight-membered "Hiickel—aromatic" system is currently in progress (65). III. Acidity Function Theory Two extensive reviews on this subject have appeared recently, one by Paul and Long (66) and the other by Bushick (67). Since part of the present work involves the measurement of ionization constants of various carbinols and glycols by use of the HR acidity function it is important to describe briefly the development of this function. In the classical treatment of dilute aqueous solutions, the hydrogen or oxonium ion concentration expresses the acidity of the solution. In concentrated acids, however, the acid strength is not proportional to the oxonium ion concentration. Hammett and Deyrup (68) developed a method for expressing these acidities. This acidity function, H0, is derived 23 from the ionization equilibria of indicator bases such as the substituted anilines which behave asBr'dnsted- Lowry uncharged bases: + + B + H ——=~ BH (37) ‘___ The function H0, is defined in the following equation. C + BH where CBH+/CB is the observed ratio of the indicator in its protonated and unprotonated forms and K + is the ionization constant of the indicator BH referred to dilute aqueous solution. Since the Ho function measures the tendency for a given solution to transfer a proton to an uncharged base it has been quite useful for measuring the strengths of very weak bases and also for interpreting kinetics of various acid-catalyzed reactions. Bases of different charge types should have related acidity functions. Work with the H+ and H_ has not been extensive (69, 70), but the study of the carbinol-carbonium ion equilibrium has given rise to the HR function which is particularly pertinent to the present work. This equilibrium, H+ + Ar3COH ——> HOH + Ar3C+ (39) ‘__ involves a charge type different than the one involved in Ho studies and has water as a reactant; therefore, one would expect it to follow a dif- ferent acidity function. Gold (71, 72) and Hawes (73) were the first to define such a function, which they called Jo: Jo '3 H0 ‘l' 10g aHzo (40) (where i denotes activity) They assumed that the ionization of a carbinol would proceed via the following equilibria: 24 + sto, -———> H + HSO,’ (41) + - + ROH + H ——-\ ROHZ (42) V— + + ROH; ——-> HOH + R (43) J0 would then be equal to: Jo = Ho + log a = -pK — log — - log (44) HZO ROH CROH fROH2+ (where f refers to the molar-concentration activity coefficient). Gold and Hawes (73) suggested that the last term in equation 44 would be negligible since it is a ratio of activity coefficients of two similarly charged species. . Williams and Bevan (74) and Deno and his group (13, 14), using arylcarbinols as indicators, showed that this approximation is unsatisfactory for sulfuric acid solutions between 60-80% even though it is valid for the 80 to 90% region. Their approach was on the basis of the negative of the equilibrium 39 represented quantitatively by pK pK R+’ and they defined the H acidity function (initially called Co by ROH Deno and his group) as: R CR+ H = pK +10g (45) R R+ CROH This function can be shown to differ from the Jo function by the activity coefficient term fR+/fROHZ+ which Gold and Hawes assumed to be negligible. The fact that this ratio is not unity is striking evidence that the activity coefficient behavior of carbonium ions and of other types of organic ions is quite different. It also indicated that the nature and amount of solvation might be quite different for various types of organic ions. Recently, Bushick (67) studied the thermodynamics of the carbinol- carbonium ion equilibrium in aqueous sulfuric acid. He found that the difference between the Hammett and HR indicators is due mainly to the 25 difference in activity coefficients for the species involved and that the HR scale, like H0, is not very temperature dependent. IV. Present Work .This thesis is concerned with the preparation and study of com- pounds that ionize in sulfuric acid to give dications of a type similar to (XVI). The previous work on 0, 0, 0', 0'-tetrapheny1-p—xylene dication was extended to the diphenyl derivatives as well as to the meta and ortho compounds in order to study the effect of amount and position of substi- tution on the visible absorption spectra. Work was also done on the effect of having additional benzene rings located between the two positive charges. This was done by using substituted biphenyl and fluorenes. Since observation of the absorption spectra was sufficient in most cases to prove dication formation, the use of cryoscopy was limited. Another tool used to study dication formation was quenching experiments. With these three methods, it has been shown that it is possible to prepare various dications similar to (XVI). Their absorption spectra have been correlated with those of various monocations. RESU LTS AND DISCUSSION 26 27 I. Syntheses and Structure Proofs The majority of the diols and carbinols needed were known com- pounds and their syntheses were performed by the usual methods. There are several compounds,ihowever, whose synthesis is of special interest. These compounds and the new compounds prepared in this thesis will be discussed in this section. The reaction of phenylmagne sium bromide with methyl benzoate or with benzophenone has long been known to give triphenylmethanol (75, 76). The obvious way of preparing 0, 0, 0', 0'-tetraphenyl-o-xylene- 0, 0'-diol, therefore, was by the reaction of phenylmagnesium bromide with dimethyl phthalate. It has been shown, however, that when either o-dibenzoylbenzene (77) or diethyl phthalate (78) reacts with the phenyl Grignard reagent and is then heated to 3000, 10, 10-diphenyl-9-anthrone is obtained: ’/ I COZEEMQXs MgCr-O (4o) \ 0in \ =O Since the oil that results from treatment of the carbonyl compounds with phenylmagnesium bromide is hydrolyzed first and then heated to 300°, it is probable that this oil is a mixture of starting material and o-benzoyl- triphenylmethanol (XXVIII) . 28 Q5 06% C=O XXVHL Indeed, Barnett, Cook and Nixon showed that when XXVIII was heated to 3000 it gave 10, 10-diphenyl-9-anthrone even though concentrated sulfuric acid or hydrogen chloride in acetic, acid did not catalyze this transformation (77). The bulkiness of the Grignard reagent and the steric hindrance at the carbonyl of XXVIII prevents further reaction with phenylmagnesium bromide. Alkylation of the aromatic ring is favored, however, because of the proximity of the reaction sites. . Support for this steric effect is obtained from the work of Bennett and Wain (79) who prepared 0, 0, 0', 0' -tetramethyl-o-xylene-0, 0' -diol without difficulty from dimethyl phthalate and methylmagnesium iodide. \ COZMe \M\C\OH (47) l / + BMeMQI —‘9 )\ / C’OH COZMe MQMQ The formation of the anthrone, however, was of use later in this thesis in the synthesis of 9-chloro-10,lO-diphenyl—9, lO-dihydroanthracene. 0, 0, 0', 0'-Tetraphenyl-o-xylene-0, 0'-diol was prepared inhigh yield from dimethyl phthalate and phenyl lithium. . Being less bulky and more reactive, phenyl lithium allows the reaction to go to completion. 29 Sloan (63) has recently reported that the synthesis of some of the pure diols was quite difficult. The diols were obtained as oils, con- verted to the dichlorides in low yield, recrystallized, hydrolyzed back to the diols and recrystallized. This procedure was employed in this thesis for the preparation of 0., a, a', o.’utetraphenyl-m-xylene-a, a'-diol, 0., a, a‘, a' v-tetraphenylup, p'-bitolyl-a, a'udiol and a, a, a' , a'-tetrapheny1-m, m'-bitolyl-o., a'wdiol. One dichloro compound, (1, o.‘-dichloro-o., a'-di- phenyl-2, 7-dimethylf1uorene, was used without hydrolysis to the diol. The probable reason for the impurity of the diols is the inefficiency of the Grignard reaction. This results in the formation of a wide variety of possible products besides the diol: Q? 950 £80 COZMe C=O C02 Me (48) ->” G QOHQ COZMe (:02 Me C30 /CO $25 $@ Q3 Evidence for the occurrence of these compounds were obtained from infrared spectra of the crude products. Several paths to the synthesis of a, 0., o‘etriphenyl-p, p'-bitolyl-a, o' -diol were envisioned. XXIX The first method failed because the initial step in the following sequence 1 did not produce any ketone 3O @‘CQ’Q Q3COCI QCQ— C\‘IC‘E¥—4>@ LIAIH AICI3 (49) The second method also failed in the first step since the reaction of 4~benzoylbiphenyl with zinc cyanide and hydrochloric acid only produced an insoluble oil. 9 n Q CDC 0 0 2149122} WHO WQBEOM The third procedure involved four steps and was intended to form a, (1, (1'-triphenylr-amhydroxyua'-methoxy~p, p'ubitolyl as the end product. XXX The procedure was as follows: 31 BrH @COC' \ Bro—04345 AlCl3 / m XXX: LIA|H4 V 1. Mg (51) cog/5 . _ We 1.97% H 80 - 9” BFQQCWZ 5 5H 4 BrHQ-Qfi H ° 8 m H Mil This sequence of reactions proved more successful since compounds XXXI, XXXII, and XXXIII were isolated. The final step, however, gave a compound whose analysis was not the same as XXX. In order to demonstrate that the benzoylation of 4-bromobiphenyl gave the 4, 4'- derivative, XXXI was reduced to the bromohydrocarbon (by the Wolff— Kishner procedure) which was treated first with magnesium and then with water to give the known 4-benzy1biphenyl. B WWB WOW B ”Biz-B l. M, 2.HZO (52> @@C +42% 32 This demonstrates that the benzoyl group was in the 4uposition in com- pound XXXI. The structure of XXXH was shown by analysis, method of synthe~ sis and its infrared spectrum. The structure ct" XXXIII was demon- strated by its n.m. r. spectrum. The reaction of XXXIII with magnesium followed by benZOphenone gave a crystalline material which did not analyze correctly for XXX. Further work must be done to identify this product. II. Visible Spectra in Sulfuric Acid and pKR Studies The visible spectra of the carbinols and diols prepared in this thesis were studied in sulfuric acid. By varying the concentration of sulfuric acid, the pKR's of the triarylcarbinols were determined. It was not possible, however, to obtain accurate pKR‘s of the diarylcarbinols since their solutions tended to fade rapidly near the region of the pK This R° was presumed to be due to formation of bisudiarylmethyl ethers which tend to precipitate out of solution: (r I + I CHZOH + '—H a H-C o C? Burton and Cheeseman (80) have postulated a similar mechanism for the Q Q Q e (1} formation of dibenzhydryl ether from benzhydrol in acetic acid containing perchloric acid. . This section will present the results of the spectral studies and the following section will interpret them. A. The Xylene Diols The compounds whose spectra are discussed in this section are: oH @-C—@ Q It was shown previously (61) that XXXVII dissolved in concentrated sulfuric acid to form a red solution which gave a characteristic visible spectrum. It was of interest, therefore, to study the ortho and meta analogs of XXXVII as well as the corresponding diphenyl derivatives (XXXVIII and XXXIX). The red solution formed by dissolving compound XXXIV in concen- trated sulfuric acid gave a visible spectrum that had maxima at 455 mu (6 == 44, 000) and 373 mp. ( 5 =30, 500). Solutions of XXXIV showed no vi51ble absorbance in sulfuric acid whose concentration was less than 78%. . Figure 1 shows the visible spectrum of a, 0., (1', o'-tetraphenyl-o-xylene- ), OOO- 3,ooo~ o. 000 0,000 0,000— 34 para O a \ I. \ / . \ / o / k \ e. \ . I \ 1 : I J ‘. \ l 400 450 Wavelength (m 11) concentrated sulfuric acid. Figure l. . Visible spectra of the $1, a, a', a'-tetraphenylxylene-a, a'-diols in 35 a, u.’-.diol in 98% sulfuric acid. A solution of tetraphenylphthalan (XXXV) in 98% sulfuric acid gave the identical visible spectrum as the diol. Table 2 records the observed absorbances (A) at 455 mu for solu- tions of XXXIV in varying concentrations of sulfuric acid (between 78 and 97%). A plot of A versus the H of the solution is shown in Figure 2. R+ A differential plot of AA/AHR+ versus HR+ gives a value of -16. 6 for the pK of a, 0., u', o.’-tetraphenyl-o-xylene-d, a'-diol. In agreement with R this result, tetraphenylphthalan was found to have almost an identical pK -16.4 R! The meta glycol XXXVI also gave a red solution when dissolved in 98% sulfuric acid. The visible spectrum of this dication also had two peaks: 440 mp. (e = 70, 000) and 419 mp..(e = 73, 000). It was necessary to use acid of gmter than 50% in order to observe a visible spectrum. . Figure l shows,.the visible spectrum of a, a, a', a'-tetraphenyl-m-xylene- a, o.’-diol in 98%: sulfuric acid, and, for comparison purposes, that of the para isomer. The latter compound absorbed at 455 me with an e of 59, 000. The pK 's of compounds XXXVI and XXXVII were determined as R described above. Tables 3 and 4 give the appropriate data, Figures 3 and 4 the observed spectra, and Figure 2 shows the plot of A versus the H of the solution. Figure 5 shows the plot of AA/AH R+ R R+ used in determining the pK R of compound XXXVII. This is typical of the versus H + differential plots employed in determining pKR's in this thesis. The meta diol has pK 's at -8. 0 and -10. l and the paraisomer at R -8. 1 and -10. 5. This is in contrast to the ortho diol which has only one -. apparent pK A more detailed discussion of this and other points con- R' cerning the visible spectra of the a, a, a', o.‘ -tetraphenylxylene-a, a'-diols in sulfuric acid will be found in Section IIIA. 36 Table 2. Observed Visible Absorption Maxima at 455 mu of a, a, a', Ia'- Tetraphenyl- --o-xylene a, o.‘-diol (6. 2 x 10' '5 moles/liter) in Varying Concentrations of Sulfuric Acid M Wt. Per Cent H2504 ' HR+ Absorbance (A) 78.0 -l4.60 0.10 78.4 -14.70 0.11 80.0 -15.28 0.22 81.0 -15.60 0.30 82.0 -15.94 0.47 83.0 -l6.24 0.68 84.0 -16.58 1.00 85.0 -l6.87 1.31 86.0 -l7.21 1.58 87.0 -l7.50 1.61 88.0 -l7.82 1:63 89.0 -l8.10 1.66 92.0 -18.99 1.67 97.0 -20.40 1.67 37 0%.”: Odd: .oo©l +m .3037} .duoCoH>xH>co£mm~uouu .d ..d to .d 93 HOH m +m E mums“; .4 mo worm .N oudmfim m. .H 4 38 Table 3. Observed Visible Absorption Maxima at 419 ms of a, a, a', o'- Tetraphenyl-m-xylene-a, o.‘-diol (2.1 x 10'5 moles/liter) in Varying Concentrations of Sulfuric Acid Wt. Per Cent H2504 HR+ Absorbance (A) 50.0 -6.47 0.07 51.0 -6.62 0.08 52.0 -6.83 0.09 53.0 -7.02 0.12 54.0 -7.23 0.16 55.0 -7.46 0.23 56.0 -7.67 0.34 56.6 -7.80 0.44 57.0 -7.90 0.56 57.5 -8.00 0.67 58.0 -8.09 0.72 59.0 -8.37 0.80 60.0 -8.61 0.89 61.0 -8.90 0.95 62.0 -9.17 1.03 63.0 -9.45 1.11 64.0 -9.76 1.22 65.0 -10.06 1.37 66.0 -10.37 1.49 67.0 -lO.70 1.56 68.0 -11.00 1.62 69.0 -11.33 1.63 39 Table 4. Observed Visible Absorption Maxima at 455 mu of a, 0., a', (1'- Tetraphenyl-p-xylene-a, a'-diol (2. 8 x 10"5 moles/liter) in Varying Concentrations of Sulfuric Acid 1 M Wt. Per Cent I H2504 HR+ Absorbance (A) 53.0 -7.02 0.07 54.0 -7.23 0.08 55.0 -7.46 0.12 56.0 -7.67 0.18 57.0 -7.90 0.31 57.5 -8.00 0.41 58.0 -8.09 0.52 59.0 -8.37 0.83 60.0 -8.61 0.97 61.0 -8.90 1.01 62.0 -9.17 1.13 63.0 -9.45 1.18 64.0 -9.76 1.20 65.0 -10.06 1.20 66.0 -10.37 1.27 67.0 -10.70 1.41 68.0 -11.00 1.50 69.0 -11.33 1.50 40 68.0 °/o ‘ ...o. ‘ 65.0 °/o <2 ‘ \\ 64.0% ‘\\ 6|.O % \, 1.5— 1.0— \\ 57.0 %‘ “\\\ 590% l 58.0 °/o 400 ‘ 450 500 Wavelength (m p.) Figure 3. Visible spectra of 0., a, a', o.‘-tetraphenyl-m-xylene-a, a'-diol in varying concentrations of sulfuric acid. 41 1'55— 68.0 °/o 6 7.0 We 66.0 °/o 64.0 °/o 62.0 0/o \\\ 6:; \\‘ .. 460 450 500 ‘ 550 Wavelength (m (.1) Figure 4. Visible spectra of 0., 0., a', o.‘-tetraphenyl-p-xylene-a, a'-diol in varying concentrations of sulfuric acid. 25 "'“’ ““R+ 10 42 Figure 5. -8.0 Plot of AA/AH a, a'-diol. R+ -9.0 H versus H R+ R+ -10.0 ~-1l.0 for 0., 0., a', o.‘ ~tetraphenyl-p-xylene- 43 Compounds XXXVIII and XXXIX in concentrated sulfuric acid gave characteristicabsorption spectra» in the visible region. The former ~ absorbed at 447 mp. (6 = 35, 000) and the latter at 461 mu (6 = 48, 000). Figure 6 shows the visible spectra of these dications in 98% sulfuric acid. As explained above, no pKR studies were done on these compounds. B. The Biphenyl Diols and Carbinols The compounds‘whose spectra are discussed in this section are: e. a e o :XI; :th :XDI XLDI' 44 40, 000 F- H—para "\ / \\ / - \ 30,000 r— / \t" meta / \ / \ ‘“ / \ / \ / \ 20, 000 / - \ / \ / / \ / \ / \ \ \ 10, 000 ._ \ g l \ 1 400 450 500 Wavelength (mu) Figure 6. Visible spectra of meta and para a, a'-dipheny1xylene- a, a'-diols in concentrated sulfuric acid. 45 Triphenylmethanol (XL) dissolved in concentrated sulfuric acid to give ayellow solution. The visible spectrum showed two maxima: 432. mp. (6: 37, 500) and 408 mp. ( 6 = 36, 500). ‘These values are the same as those reported by Deno (it a}. (14). Figure 7 shows the visible spectrum of XL in 98% sulfuric acid. Table 5 records the observed absorbances (A) at 432 mp. for solutions of XL in varying concentrations of sulfuric acid between 51 and 68%. A plot of A versus the HR+ of the solu- tion is shown in Figure 8. The pKR of triphenylmethanol as obtained from a differential plot of AA/AHR+ versus HR+ is -7.4. The literature values are --6.4 (67) and -6.7 (14). When XLI was dissolved in concentrated sulfuric acid a red solution resulted. In less than a minute, however, the solution became orange. . The visible spectrum of the orange solution (98% sulfuric acid) showed maxima at 476 mp (6 = 49, 000) and 432 mp ( 6 = 34, 000). Mason reported a 1 max of 474 mp. and an 6 max of 51, 000 (see page 6 ). . In 78% sul- furic acid, where the solution was red, maxima were observed at 510 mp (e = 43, 000) and 420 mp. (e = 21, 500). Anderson and Fisher reported a visible spectrum for XLI that showed a maximum at 508 mp. (e = 39, 000) (81). The solvent used was 5 ml. of concentrated sulfuric acid in 95 m1. of glacial acetic acid. This spectrum, which is due to the carbonium ion, is depicted in Figure 9. Table 6 gives the A at 510 mp for solutions of XLI in sulfuric acid between 52 and 78%. . Figure 8 shows, the plot of A versus HR+° The pKR was -7. 7. Since in 93. 9% sulfuric acid the change from the red to the orange solution was slow, it was followed as a function of time. Figure 10 shows this change at 93. 9% sulfuric acid. The time for completion of one-half of the reaction was 135 minutes (k = 5. 1 x 10'3 min. '1). I Benzhydrol (XLII) dissolved in concentrated sulfuric acid to form a yellow solution. Figure 9. shows the visible spectrum of compound XLII in 98% sulfuric apid. Table 7 records the observed absorbances at the 46 ,000— / ’ \ 0., 0., a' , a'-tetraphenyl-m, m' - / \ bitolyl-a, o.'—diol / \ . ,ooo-—- /. \ ,OOOrjr V \ 0., 0., a', o.‘ -tetraphenyl- ‘ - f o, o'-bitolyl-o., a'-diol riphenylmethanol 000 l | l 400 450 506 Wavelength (m p) Figure 7. Visible spectra of triphenylmethanol, <1, 0., a', a'-tetraphenyl-m, m'- bitolyl-a, o.‘-diol and a, 0., a', o.‘—tetrapheny1-o-o'-bitolyl-a, a' -diol in concentrated sulfuric acid. 47 Table 5. Observed Visible Absorption Maxima at 432 mp of Triphenyl- methanol (5. 0 x 10"5 moles/liter) in Varying Concentrations of Sulfuric Acid ===r Wt. Per Cent HR+ Absorbance (A) H2504 51.0 —6.62 0.22 53.0 -7.02 0.49 54.0 -7.23 0.62 55.0 -7.46 1.19 56.0 -7.67 1.56 56.3 —7.73 1.64 58.5 -8.23 1.83 70.0 -11.64 1.91 97.0 -20.40 1.91 48 .36- .o 6.4333 .. .E ..E1H>conamhouu.d ...o .d .d was Homo: .6 6.4133313 duaqoflamnuounb . .d to :6 4099305 sisosmfigztfiuoafimfiuv .Hocdsuogtwcozmfiu .Hoamfioegaonmfiuu n8 3mm manner < mo worm .m oudmfim +m m 0.2.. 0.0.. o.m.. 0.5: — l— — .\ i— \ 11 A u \ \. Hog-.6 6-3339 u s. \. \.. u .E .EnH>co£QMHuouu.d ..d .d .divlua . \ K \ \ o x \o \ . XX 0 \ X . .\ 38-3 6-383 x x . .Q .mugcoxmmbounb ..d to .dIMIH'\ \\ Hocmnuoggcofifip 0.9a: 0.3- o.NH.. 49 .390 ownsfldm poem: acoocoo cw 351.0 76-13003:.mumngconmmnuouu.d ..d .d .d was HOGM£008H>G63QH>~>G0£93nv .Hocmfimegsozmflp pad 308 ofipdfidm 05w» 5 3523sSidesmfipgacozmz..v mo @30on 638 Cr 3 Ev guwcgocrm? .4333- .Q duaconmmfiounb . .d .d .0 000 000 000 000 I _ _ .\ /. \ .\1| ,/ , \ x. .. , \ . i \ .\ . l \ \ / / \ . . / \ .\ . / \ \ I1 / / Kr . / /. \ . Hocmfiogacnwgmfip . \ . /.\. . 3530.68 . 1321333002054 I1 .. . AaocmauoEgcmwgmagnofiafio...“v . , .. .\ l ./ ...\rvr/ Hon$;o.o .o ondwfm 000.0N 000.00 000.00 000.00 50 Table 6'. Observed Visible Absorption Mak’ima at 510 mp of 4-Biphenylyl- diphenylmethanol (5.4 x 10'5 moles/liter) in Varying Concen- trations of Sulfuric Acid — —’ g — L Wt. Per Cent HR+ Absorbance (A) PQSCM 52.0 -6.83 0.12 53.0 -7.02 0.32 54.0 -7.23 0.56 55. 0 -7.46 0. 94 56.0 -7.67 1.45 56.3 -7.73 1.62 56.6 -7.80 1.76 57.0 -7.90 1.92 58.0 -8.09 2.09 59.0 -8.37 2.22 60.0 -8.61 2.27 78.0 -14.60 2.27 51 Time en) 2.5 , , 31.5 475 , 75.5 L \\ __ ‘ 400 450 560 s50 - Wavelength (mp) Figure 10. . Change in the visible spectrum of 4-biphenylyldiphenylmethanol in 93.9% sulfuric acid with time. 52 Table 7. Observed Visible Absorption Spectra at 442 mp of Diphenyl- methanol (4. 7 x 10'5 moles/liter) in Varying Concentrations of Sulfuric Acid g I Wt. Per Cent HR+ Absorbance (A) PQSCM 71.0 -12.01 0.04 72.0 -12.40 0.06 73.0 -12.74 0.10 74.0 -13.11 0.14 75.0 -13.50 0.28 76.0 -13.82 0.48 77.0 -l4.21 0.68 78.0 -14.60 0.88 78.4 -14.70 0.99 80.0 -15.28 1:62 81.0 -15.60 1.76 87.0 -17.50 2.06 '88.0 -17.82 2.08 93.9 -l9.50 2.20 97.0 -20.30 2.20 53 maximum absorbance (442 mp; e = 47, 000) in varying concentrations of sulfuric acid. The previously reported values are 441 mp ( e = 43, 000) (82). A plot A versus H + is given in Figure 8. The pKR of R this carbinol was determined to be -l4. 7. The literature value is -13. 3 (14) . 4-Pheny1benzhydrol (XLIII) dissolved in concentrated sulfuric acid with formation of a red solution. Figures 9 and 11 depict its visible spectrum which showed a single maximum at 535 mp ( 6 = 90, 000). The pK of this compound was not determined because its solution in dilute R sulfuric acid were unstable. The spectrum of the cation from compound XLIV was not obtained because on treatment of a solution of XLIV in either acetic acid or methanol with sulfuric acid, a reaction occurred and the solutions were colorless. The sulfuric acid used was between 50 and 97%. The 4, 4'-biphenyl derivative XLV dissolved in concentrated sulfuric acid to form a deep red solution. . Solutions of XLV were colorless in sulfuric acid whose concentration was less than 54%. In Figure 9, which shows the visible spectrum of compound XLV in 98% sulfuric acid, it can be seen that three spectral maxima occur: 530 mp ( e = 90, 000), 442 mp ( e = 53, 500) and a shoulder at 410 mp ( e = 34, 500). Theilacker and Ozegowski reported maxima at 525 and 440 mp and a shoulder at 412 mp (6's were 89, 000, 50, 000 and 32, 000 respectively) (83). Table 8 records the A at 530 mp and 442 mp for solutions of XLV in sulfuric. acid whose concentration varied between 55 and 67%. , A plot of A versus HR+ is seen in Figure 8. The pKR of the diol was determined to be -8. 2. The corresponding diphenylcarbinol XLVI, which dissolved in con- centrated sulfuric to form a violet solution, showed two absorption maxima: 563 mp (5 = 134, 000) and 420 mp (e = 13, 000). This spectrum is shown in Figure 11. . Even though the pKR of this bis-secondary alcohol was not determined, it was observed that the position of the absorption 54 .Uwom 025346 macaw pad pouduuaoocoo GM Homo-.6 .d-TfioU—BJQ .m nacmfimflv-rd .d was “Jo-m 0:316 pouduucoofloo cm Hocmfioggaonmggconmfinv mo duuoomm odnwmfir . : madman 1.53 03903953 coo omm cam 03» 03» X .- _ _ . - / . ._ . / o \\. . IP’I/ \ 1 / . I - \ / \. \. \. \ . . \ . O \. . a \ 1..... 3.-.. .. p .2 .2. . \. IH>Hofin-.m duaaosmfipub .d .. .\ . . \ Hocmzuoggcogmf» 1:93 3 u w 3 \ . \ . .1 COO OOH \/ \\ 36- ..o 6-383- .0 0-35.180- .0 .0 \ \ / \ /\ 000.0ms 55 Table 8. Observed Visible Absorption Maxims. at 530 and 442 mp. of a, a, o', o'-Tetrapheny1-p, p'-bitoly1-o,, o'-diol (2. 4 x 10'5 moles/ liter) in Varying Concentrations of Sulfuric Acid Wt. Per Cent HR+ Absorbance (A) H2504 442 mp 530 mp 55.0 -7.46 0.03 0.04 56.0 -7.67 0.08 0.10 57.0 -7.90 0.18 0.29 57.5 +8.00 0.28 0.50 58.0 -8.09 0.44 0.80 58.5 -8.23 0.64 1.20 59.0 -8.37 0.88 1.65 60.0 -8.61 1.07 1.96 61.0 -8.96 1.11 2.05 63.0 -9.45 1.15 2.12 67.0 -10.70 1.15 2.12 80.0 -15.28 1.16 2.12 56 maximum shifts with decreasing acidity and in 83% sulfuric acid appears at 535 mu (6 = 66, 000). This, is also shown in Figure 11. The meta diol XLVII gave a yellow solution when diSsolved in concentrated sulfuric acid. Its solutions were colorless below 55% acid. In 98% sulfuric acid, compound XLVII showed one maximum at 430 mu (6 = 67, 500). This is depicted in Figure 7. Table 9 records the A at 430 mp. for solutions of XLVII in varying concentrations of sulfuric acid (between 57 and 72%). A plot of A versus HR+ is shown in Figure 8. The pKR was 78.4. . Solutions of the analogous ortho diol XLVIII in concentrated sulfuric acid were yellow. Its visible spectrum in 98% sulfuric acid. which is shown in Figure 7, had' a maximum at 428 mu ( 6: 35, 500) and shoulder at 390 mp. (6 = 24, 000). Solutions of XLVIII in sulfuric acid were color- less below 75%. The pK of this diol was approximately ~15. R C. The FluoreneDiols and Carbinols The compounds whose spectra are discussed in this section are: C 1 O. Q@ ¢_H MEH @ 57 Table 9. 7 Observed Visible Absorption Maxima at 430 mg of a, 0., u', o'- Tetraphenyl-m, m'-bitoly1-o., a'-diol (2. 5 x 10"5 moles/liter) in Varying Concentrations of Sulfuric Acid Wt. Per Cent HR+ Absorbance (A) H2504 57.0 -7.90 0.13 58.0 -8.09 0.26 59.0 -8.37 0.70 60.0 -8.61 1.34 61.0 -8.96 1.58 70.0 -1l.64 1.70 71.0 -11.95 1.70 58 The violet solution formed when compound XLIX was dissolved in concentrated sulfuric acid gave a visible spectrum that had a maximum at 547 mu ( 6 = 66, 000). As the acid strength was decreased, the posi- tion of the maximum shifted to 542 mu. . Figure 12 shows the visible spectrum of XLIX in 98% sulfuric acid. An exact determination of the pKR of this carbinol was not possible because of rapid fading of the solu- tions in dilute acid (see page 32) but an approximate value of -12.0 was found. Compoundeis solved in concentrated sulfuric acid to form a blue solution. Its visible spectrum showed maxima at 603 mu ( e = 215, 000) and 554 mp. ( 6 = 62, 000) and is depicted in Figure 12. As the acid strength was decreased, the intensity of the peak at 603 mu also de- creased and eventually it disappeared at 76% sulfuric acid. The peak at 554 mu, however, did not change in intensity until 80% sulfuric acid where the intensity increased slightly, reaching a maximum at 78% acid, and the xmax shifted to 547 mu. Below 78%, the absorption intensity decreased. The solutions were colorless below 65% sulfuric acid. Table 10 records the observed absorbances at 603 and 554 mp. for solutions of Lin varying concentrations of sulfuric acid and Figure 13 shows the plot of A versus the H of the solutions. The pK 's of the R like the pKR of R+ two ions are -l6.6 and -12.0. The latter pK compound XLIX, was approximate. R The visible spectrum of compound LI in sulfuric acid behaved similar to that of XLI in sulfuric acid (see page 45 ), i. e. , the spectrum in 80% sulfuric acid was different from that in 98% acid. In 98% sulfuric acid there were two absorption maxima at 505 mu (5 = 53, 000) and at 425 mu ( 6: 18, 000). In 80% sulfuric acid the peaks were at 540 mu ( 6 = 53, 000) and 404 mu ( 6 = 17, 000). Since the former spectrum is shown later to be due to sulfonation, the latter represents the spectrum of the genuine carbonium ion (see Figure 14). . Figure 15 shows the change 59 I"\ 200.000 #- a,c'-dichloro-a,a'-diphenyl-2,7i-- I I dimethylfluorene I \ I I I I I I ’ I I I 15.0,000 —— I I I I “ ’ I 0 l I ’ I I 100,000 __ I I I I I I h 1 hd -2- th 1- I I IDfluoi'Iene Y Y Y / I / I / \\ // \ 50,000 — / V I\ I I / I / I I \ 500 550 600 Wavelength. (m p.) Figure 12. . Visible spectra of a-phenyl-a-hydroxy-2-methylfluorene and a, o'-dichloro-o., o' -diphenyl-2, 7-dimethylfluorene in concentrated sulfuric acid. 60 Table 10. Observed Visible Absorption Maxima at 554 and 603 mu of o, o.‘-Dichloro-a, o.‘ -diphenyl-2, 7-dimethylfluorene (l . 2 x 10'5 moles/liter) in Varying Concentrations of Sulfuric Acid Wt. Per Cent H Absorbance (A) sto, M 554 m... 603 mg ‘70.0 -11.68 0.393 .0.00 71.0 -12.02 0.48a 0.00 72.0 -12.40 0.55a 0.02 73.0 -12.75 0.68a 0.03 74.0 -13.10 0.74a 0.03 75.0 -13.50 0.80a 0.03 78.0 -14.60 0.88a 0.10 79.0 -14.91 0.83 0.15 80.0 -15.28 0.77a 0.23 81.0 -15.60 0.77b 0.40 82.0 -15.94 0.70 0.60 83.0 -16.24 0.74d 0.82 84.0 -16.58 0.73 1.18 85.0 -16.87 0.69 1.62 92.0 -18.99 0.69 2.38 96.0 -20.08 0.75 2.60 a. 547 mu; b. 549 mu; c. 550 mu; d. 552 mu. 61 m . + .ocouodqgfifiofiflpup Nugcogmfioub 610.81.331.49 .0 HOH m 35.3; 4 mo no?" .mH madman + mm O42: ofiT. 067. 067. _ A _ IIIIIIII \ \ \ \ \ I \ m \ / . I o \ E «.mm \ \ I, m \ \AIJ. \ \ \ \ \ \ I. m I \ J .H .N .300 02.315 pouduucoocoo a: 331.0 .0noaouodflgfiogwpnn duaaoammflounb . .0 .0 .0 . pad “:00 03316 $03. awocouoaflgauogumI>xOup>£101H>ao£mfluud .0 mo 0300mm 033?? .3 oudmrm A105 ,aumnofigdg omm . com omw 00¢ 62 A odouoafl 11:381....” 1.33.5270 nanosmmo 10 .0 I 000 .om I \ I 80.2: ’ . \rlafloflpnb .0noaouo3maauoaflouh .Nnaqmfimmfiounb ..0 .0 .0 ooo.om~ (\ LI 2. 1. 1. 0. 63 O._. 00—— Tim (min 3I0.0 5 ._ I - I I 400 450 500 550 Wavelength (mp) Figure 15. . Change in the visible spectrum of a, a-diphenyl-c-hydroxy-2-methyl- fluorene in 93. 9% sulfuric acid with time. ’ 64 of the spectrum in 93. 9% sulfuric acid (note that the initial maxima, those attributed to the cation, occur at 545 and 408 mp. in this more con-V centrated acid). The time for completion of one-half of the reaction was 22 minutes (34.6 x 10'3 min. '1). Table 11 gives the-observed absorbance at 535 mp. and 404 mp. for solutions of LI in sulfuric acid between 45 and 65%. A plot of H The pK is) -6. 5. I R . The spectrum of the corresponding bis-diol LII in-concentrated Rd- versus A is shown in Figure 16. sulfuric acid is shown in Figure 14. . The purple solution has maxima at 568 m0 ( 6: 152,000), 525 m0 (e = 46,000) and 440 mp..(€ = 35,000). D. The Anthracene Diols and- Carbinols The compounds whose spectra are discussed in this section are: The assignment of the trans configuration for LV is based on infrared studies (115). ' Compound LIII dissolved in cOncentrated sulfuric acid to form a yellow solution. Its visible spectrum showed a maximum at 438 mp. ( e = 22, 000). .‘ Solutions of LIII showed no visible absorbance in (sulfuric acid whose concentration was less than 59%. . Figure 17 shows the visible spectrum of LIII in 98% sulfuric acid. Table 12 records the observed absorbances (A) at 438 mp. for solution of LIII in varying concentrations 65 Table 11. Observed Visible Absorption Maxima at 404 and 535 mu of 0., 0. -Diphenyl-o.-hydroxy-Z-methylfluorene (4. 0 x 10"5 moles/ liter) in Varying Concentrations of Sulfuric Acid — w Wt. Per Cent HR+ . Absorbance (A) sto, 404 m0 535 mg 45.0 -5.40 0.04 0.06 46.0 -5.61 0.06 0.10 47.0 -5.80 0.16a 0.17 48.0 -6.00 0.18a 0.26 49.0 -6.20 0.25a 0.48 50.0 -6.47 0.30a 0.70 51.0 -6.62 0.46 1.34 52.0 -6.83 0.56 1.75 53.0 -7.10 0.64 2.00 54.0 -7.23 ' 0.68 2.10 61.0. -8.90 0.70 2.14b 64.0. ~9.76 0.70 2.15b a. 400 mu; b. 538 mu. l. 1. 0. 5 66 I I I -6.0 -7.0 -8.0 H R+ Figure 16.. Plot of A versus HR+ for a, c-diphenyl-a-hydroxy-Z— methylfluor ene . 67 I 9-c'hloro-10,10-di- \ phenyl-9, 10-011»:de \ I, 000 _anthracene / \ ethyl ether of 9, 10, 10-triphenyl- \ 9, 10-dihydro-9-anthrol / \ I, 000 — / . ' ' \\ . ‘ I ‘ \ 0 7 \ , 000 -- j . / \ , / \ - '~;\\t‘rln8-9,10-diphenyl-9, 10- \ . ' dihydroxy-9, 10-dihydro- . ., 000 — . / anthracene \ \ - / \ .. o \ ‘ . . / \ / \\ . K \ \ 4 . - I g I . L 400 450 500 I Wavelength (mu) Figure 17. . Visible spectra of the ethyl ether of 9, 10, lO-triphenyl-9, 10-dihydro- 9-anthrol and 9-chloro-10, 10-diphenyl-9, lO-dihydroanthracene in concentrated sulfuric acid and trans-9, 10-dipheny1-9, 10-dihydroxy- 9, lO-dihydroanthracene in 62% sulfuric acid. 68 Table 12. . Observed Visible Absorption Maxima at 438 mu of the Ethyl Ether of 9,10, lO-Triphenyl-Q, lO-dihydro-9-anthrol (2. 6 x 10"5 moles/liter) in Varying Concentrations of Sulfuric Acid. Wt.. Per Cent HR+ Absorbance (A) H2504 60.0 -8.67 0.05 61.0 -8.90 0.08 62.0 -9.17 0.14 63.0 -9.45 0.22 64.0 -9.76 0.33 66.0 -10.37 0.53 70.0 -ll.67 0.57 75.0 -l3.49 0.57 69 of sulfuric acid (between 60 and 75%). A plot of A versus the HR+ of the solution is shown in Figure 18. The pKR was found to be -9. 6. . Compound LIV also gave a yellow solution when dissolved incon- centrated sulfuric acid. The visible spectrum of this cation, which is shown in Figure 17, had a peak of 425 mu (6 = 27, 000). . No pKR studies were done on this compound. . Compound LV dissolved in concentrated sulfuric acid with formation of a deep blue color. This color was formed with sulfuric acid 76% and stronger. In sulfuric acid below 70%, compound LV dissolved with formation of a yellow to orange color. . Figure 17 shows the visible spectrum of LV in 62% sulfuric acid. Table 13 gives the observed absorbances at 450 mp for solutions of LV in sulfuric acid between 51 and 65%. A plot of HR+ versus A is shown in Figure 18. The pKR was -8.4. In sulfuric acid between 60 and 74%, solutions of LV tended to fade. , The rate of change was very rapid at 74%. I In 62% acid this change was followed with time. The maximum at 450 mp. ( e = 20, 500) decreased and was replaced by maxima at 340, 360, 380 and 402 mp. ,( 5's of8, 300, 7, 000, 8, 300 and 8, 300 respectively) (see Figure 19). The visible spectrum of the blue solution formed when LV was dissolved in 78% sulfuric acid is shown in Figure 20. Also shown in Figure 20 is the spectrum of 9, 10-diphenylanthracene in 97% sulfuric acid. The latter compound was isolated upon hydrolysis of a solution of LV in 97% sulfuric acid (61). Figure 21 depicts the visible spectrum of LV in 97% sulfuric acid. This spectrum was taken within five minutes of the time that the solution was prepared. Figure 21 also shows the visible spectra of 9, lO-diphenyl- anthracene and 4-phenyl-2, 3-benzofluoranthene in 97% sulfuric acid. The latter compound was formed with 9, lO-diphenylanthracene in the hydrolysis experiment mentioned above. 70 Table 13. Observed Visible Maxima a 450 mu of 9, 10-Diphenyl-9, 10- dihydroxy-9, lO-dihydroanthracene (7.3 x 10"5 moles/liter) in Varying Concentrations of Sulfuric Acid Wt.. Per Cent HR+ Absorbance (A) EhSCh 51.0 -6.62 0.05 53.0 -7.02 0.10 55.0 -7.46 0.20 56.6 -7.80 0.44 57.5 -8.00 0.55 58.0 ~8.09 0.73 59.0 -8 37 0.95 60.0 -8.61 1.10 , 61.0 -8.96 1.30 62.0 -9.20 1.48 63.0 -9.45 1.46 64.0 -9.74 1.50 l. l. 0 0 x . 5 'T1 trans-9, 10-diphenyl- 9, 10- /‘\" dihydroxy- 9 , 10 - dihydroanthr ac ene / ethyl ether of 9, 10, lO-triphenyl-9, 10- dihydro- 9 - anthrol _ l / / / / / / / I l I , . -7.0 -9.0 -1l.0 -l3.0 HR+ Figure 18. Plot of A versus H + for the ethyl ether of 9, 10, 10-tripheny1- 9, lO-dihydro-9-anfiirol and trans-9, 10-diphenyl-9, lO-di- hydroxy-9, lO-dihydroanthracene. 72 0.8 0.6 I—" 0.4 — 0.2 -— L I I 350 400 ‘ 450 500 Wavelength (mu) Figure 19. . Change in the visible spectrum of trans-9, lO-diphenyl-9, 10- dihydroxy-9, lO-dihydroanthracene in 62% sulfuric acid with time. 73 .206 33:5 $3. 5 60006300023572 5-50.66.03.92 5-102892 .0 Imam: pad 300 0.35350 “0303:0230 cw oaoodnnucmaaonmflpuoa .0 mo 0300mm 03%ma .om onsmah and 83030440? cow 0.2. com com cow 09 I- v .\ L _ L Soon. 033 w 00.390033 ocoodngundonp>£wpuoa .m->Xoup>nfloao~ .ouacosmgw A: 5193.3. .ooo.o ooo.~. ooo.ms 000.0N 74 , Bum 02316 vaumbcmucou aw mcmomnnunmouv>£MVuoH .olaxouvifivuoa .ouacmsmfivuoa 6.933 98 maoficmuowdfionamnnfiewu1:23uv .meuMufi:d~>Gm£m€uoH .o «o «.30QO 033W? . . Hm mndmfim 38v Aumco~0>m3 . coo oom cgV . oom .1 Sad ’ r f . . mawficmuogmoucenum ..N..3cm£muv\.u\1‘ . . . .. \ I1“ 000 :2 mcmomuaucmouwkwfivuofi .olfinouvifivuoa .mvuacmgmflvnofi .oumcmuu . mcv unusucmgcofimflu .. OH . o \\|~|\'. . i go .2 ulM 000.3. 75 When a solution of LV in 97% sulfuric acid was allowed to stand for 16 hours, the visible spectrum underwent a change. This is seen in Figure 22. Also shown in Figure 22 is the spectrum of 4-phenyl- Z, 3-benzofluoranthene in 97% sulfuric acid. III. Interpretation of the Spectra It has been shown that in sulfuric acid, aryl carbinols and diOls undergo reversible color formation. Ionization of carbinols was dis- cussed on page 2 (see equation 1). With diols, however, ionization OH © _ QCIZ‘Q a > + C —X—C—OH (54) © © ©$® c . ‘ +?_X_?+ <55) OH (2 ii In all examples studied in this thesis, X was. aromatic. . Evidence, can either occur stepwise or all at once: mainly in the form of ionization constants, will be presented to show that the reactions represented by equations 54 and 55 can occur and that it is a function of the type of aromatic placed between the two sites to be ionized. .mudog A: nomad Bum 0:316 poomuucoocoo GM mcmomuaucmoupfnfipnoa .aufwxonpkwsfipuod .ongconaflpuod 5:98.: pad gum 0:315 bonanucoocoo a“ ocoficmuosaonconrm.Nnfixbozmuau mo «.50on 3333/ .NN ondmfim 38¢ Aumaoaokrm? oow ooh. cob oom oov oom _ _ _ 76 /, . .I 8....N / \ / / \ / \ \’\< . l ooo.m w ocoaucmuodfionconum JAE»??? 3 ocoomufianHpifivuoH .@I>XOHU>A€_..0H .ougcoamguofi 6125.3 _ 3/ loomc. a, ~~- ~~ ‘ / ].000 .0a 77 A. The Xylene Diols At first glance, it would appear that the xylene diols are ionized all at once (see equation 55) since the position of the absorption maximum does not shift with changing sulfuric acid concentration. This could indi- cate the formation of only one species on ionization and not the two required by the stepwise process. The meta and para derivatives, however, each showed two breaks in the plot of A versus HR+ (Figure 2.). If one makes the assumption that either the spectra of the monocation and the dication are indistinguishable or that the maximum of one masks the maximum of the other then it appears that ionization does occur in a stepwise manner. If the xylene diols ionize in a stepwise fashion (see equation 54) and if it is assumed that the dication is not formed until monocation formation is complete (this is reasonable since the cation is an electron withdrawing group and would make the second -OH more difficult to protonate) then several possibilities arise. The emax of the dication may be less than, greater than or equal to that of the monocation. The first two are depicted in Figures 23 and 24, respectively. . When the two are equal, there is only one break in the curve and only one apparent pKR. The experimentally observed pKR plots for the para and meta diols (see Figure 2 ) are very similar to Figure 24. It can be seen that the emax of both monocations are identical whereas the intensity of the absorption of the meta dication is greater than that of the para dication. This is in agreement with the n.m. r. data discussed on page 20 which showed that the central ring in the meta dication carried an appreciable portion of the positive charge, whereas in the para dication this was not the case. In other words, the ground state is more stabilized in the meta than in the para dication, i. e. , it is of lower energy. 78 HR+ Figure 23. . Theoretical plot of A versus H formation where 61 > £3. R+ for stepwise dication 79 / / / / H 3+ oi A versus EU wheoredca PX “hare 67' 7 61. WW“ {or StePWi se dication 80 It was assumed above that in order to explain the two pKR's, either the monocation and dication had similar absorption maxima or that one masked the other. , By careful examination of the spectra of the meta and para diols in the region between the pKR's it can be concluded that the latter effect occurs. In the para diol, the absorption spectrum is constant between HR+ of -8.00 and -10.00, i. e. , a peak at 445 mg and a shoulder at 405 mu (see Figure 4). Since this is the region assumed to contain only monocation then these peaks are attributed to this species. At acidities higher than -10. 00, the shoulder disappears and there is one peak at 450 mu. 7 It appears, therefore, that the dication absorbs at 450 mp. In the meta diol, this is reversed since two peaks appear at higher acidity where only one existed at lower acidity. The monocation, there- fore, has a peak at 420 my. and the dication has a peak at 420 mp. and a shoulder at 440 mp (see Figure 3). The fact that the para dication absorbs at a higher wavelength. (lower energy) than the meta dication is readily explained (Figure 1). When the dications absorb light their.1r electrons are elevated from the ground state to the first excited state. It is this energy absorption that produces the observed visible spectrum. 7 If it is assumed that the energy of the excited states of the meta and para dications are similar, then, since the meta derivative has a lower ground state energy, it must take less energy for the para dication to go to the excited state, i. e. , it absorbs at a longer wavelength.. It was observed that the ionization of the second -OH was more dif- ficult for the para than the meta diol, i. e. , it occurred at higher acidity (see Figure 2). This is consistent with the fact that the monocation can inhibit the ionization of the second -OH to a greater extent in the para diol than in the meta diol. 81 EC @Qj + COH<———> ,C Q-OH <56) 9Q: g g Q. +CQ/Q < _ A, Q//@ (57) QW OH g ‘OH The discussion thus far has not included the ortho diol. It is a special case, because of the proximity of the positive charges in the dication. . It has only one ionization point that is detectable by‘ means of visible spectrophotometry (Figure 2). One can envision the ortho diol ionizing to the monocation which is in rapid equilibrium with the colorless protonated phthalan. . This first ionization presumably occurs in the same region of acidity as the meta and para-diols, i. e. , at a pK of about -8. 0. The great stability of the protonated phthalan. (LVII) R makes the second ionization more difficult, therefore requiring more acidic conditions, i. e. , a pKR of -l6. 6. C\OH C/OH :OH::—u> 53 >0 Bi? Q‘C’Q 11H: ‘58) @9195 82 The dicationcan either be formed from the monocation. (LVI) by pro- tonation and subsequent loss of water or by further protonation of (LVII): @ Q5 \ + .‘+ C (59) $6 ' From the spectral data it is clear that the tetraphenyl xylene diols ionize in a stepwise fashion to give their respective dications. Since the spectra of the diphenyl xylene diols (compounds XXXVIII and XXXIX) were only studied in concentrated sulfuric acid the interpre- tations can be drawn only from the xmax and emax of the dications. - Compound XXXIX had a higher absorption intensity as well as a higher position of maximum absorption. . The conclusions, therefore, are the same as with the tetraphenyl dications, i. e. , the meta derivative must have a lower groundstate energy than the para compound. B. The Biphenyl Diols and- Carbinols- The pKR's and spectra of the carbinols XL-XLIV were studied in order to interpret those of diols XLV-XLVIII. In section III-A it was shown that if X was phenyl (see page 75) the diols ionized in a stepwise manner. In this section it will be demon- strated that when X is a biphenyl moiety only one pK is observed, i. e. , R ionization occurs simultaneously at both sites. It was observed that the p, p'-and m, m'-biphenyl diols (XLV) and (XLVII) had one pK whereas the corresponding xylene diols had two. R This is reasonable since after monoisioniza'tion ofthe latter compounds, a different type of carbinol remained, i. e. , one withan electron with- drawing group on the ring containing this carbinol (see equation 54). 83 With the biphenyl diols, however, mono-ionization would leave a. carbinol identical with the initial alcohol. It will therefore behave like two inde- pendent triphenylmethanols and ionize with the same pKR (see equation 55). The 0, o'-biphenyl diol (XLVIII) appeared to ionize in a stepwise manner (see below). Compound XLV’in 98% sulfuric acid had absorption maxima at 530, 442 and 410 mpkmax of 90, 000, 53, 500 and 36, 000 respectively). In order to identify the types of resonance interactions occurring in this dication, the visible spectra of the cations from XLI, XLII, and XLIII were taken (see Figure 9). Compound XLI absorbed at 510 Int—#06 = 43, 000) and 420 mu (6 = 22, 000), compound XLII at 442 my. (6 = 53, 000) and compound XLIII at 536 mu (6 = 90, 000). . The spectra of XLII and XLIII are due to the follOwing resonance interactions, respectively: 0,9 9&0 e op H H IIZ The electronic interactions of compound XLI can be explained as Lewis explained (the spectrum of malachite green, i. e. , with an x and a y band (47): 84 + _ Q... + /. . Q Hg, ©\Me—>Mc©m Q The resonance shown in equation 62. is similar to that depicted in equation ”Q 61 for compound XLIII whose x max is 536 mp. The low energy absorption in compound XLI is located at 510 mp. This difference of 28 mp. can be attributed to the decrease in conjugation caused by the increase in steric hindrance in going from a diphenyl cation (LIX) to a triphenyl cation (LX). The absorption at 420 mp. for compound XLI can be attributed to a diphenyl- methyl cation-like species (see equation 63). V With this information, we can better understand the spectrum of XLV in concentrated sulfuric acid (see Figure 9). x This dication which has absorption maxima at 530, 442 and 410 mu, canuhave resonance inter- actions very similar to those attributable to specieslike LVIII and LIX . 85 (64) Evidence supporting the type of resonance as depicted in 64 is obtained from the visible spectrum of LXII in concentrated sulfuric acid (83). The increase in. steric hindrance has decreased the possibility for the center rings becoming coplanar. This is clearly shown since the-dication from LXII had-maxima at 470 mp. ( 6 = 71, 000) and 420 mp. ( e = 56, 000). It was shown on page 45 that compound XLI gave a different visible spectrum in concentrated sulfuric acid than in 80% acid. . In acid of inter- mediate strength it was possible to follow this change and to observe an isosbestic point.(Figure 10). The time for completion of one-half of the change was found to be 135 minutes. A further discussion of this 86 phenomenon appears on page 91- where evidence is presented which indi- cates that the observed change is due to sulfonation. Compound XLVII like XLV, was found to have a single pKR (Figure 8). Unlike XLV, however, the dication from XLVII had only one visible absorption maximum, which occurred at 430 mu (see the com- parison to triphenylmethyl cation's maxima in Figure 7). It is reasonable to picture this dication as two separate monocations. .This view is sup- ported by looking at the possible resonance interactions of this dication in which the middle rings are substituted meta. ‘I \—/ 2 $-93 + | 2 +,(E“¢ ¢ ¢ ' Carbon atoms 1 represent the possible sites in the middle rings on which the positive charge on the left can reside and 2 is the same for the charge on the right. . The dication therefore, contains two triphenylmethyl cations per molecule and should have an emax about twice that of triphenyl- methyl cation. The €max for compound XLVII in sulfuric acid is 67, 500 and for XL is 37, 500. The reason that the pK of XL is greater than that of XLVII (-7.4 and -8.4, respectively) can 1: due to a - I inductive effect of the middle phenyl rings which are meta to the site of ionization. At first glance, the visible spectrum of compound LXVIII in concen- trated sulfuric acid might be expected to be similar to that of XLV. The type of resonance interactions that occur in LXI can not occur in the ortho dication because of steric hindrance. It was hoped that by comparing the 87 spectra in sulfuric acid of XLIV and XLVIII the spectrum of the latter could be interpreted. It turns out, however, that when XLIV was dis- solved in sulfuric acid it turned yellow but faded within, seconds. This was true of acid between 50 and 95%. This is reasonable since it is knowntzhat when XLIV is refluxed in acetic acid, 9, 9-dipheny1f1uorene is formed (84) . AcOH O O Reflux (65) gig—OH fl . Inspection of molecular models shows that the benzene rings in the dication from XLVIII should be less coplanar than the rings in triphenyl- methyl cation. This is manifested in the Emax of the single maximum at 428 mu which is 35, 500,or 17, 750 per triphenylmethyl Cation. This indicates a decrease in resonance stabilization. The pKR of XLVIII was about -15. This value is! similar to the pKR of XXXIV (-l6. 6) which apparently ionized in a stepwise manner (see page 81 ). The analogous iOnization of XLVIII would also give a protonated ether: KER—$918”? OO (6., W9\+/ \ C++C_@ a <2 <5 we e H H 88 The first pKR of this diol should be similar to that of the meta and para derivatives, i.e. , about U8. 3. Hydrolysis of this dication was shown to give the cyclic ether. . Compound XLVI, unlike XLV, appeared to have two pKR's. Since it was not possible to determine the pKR's of XLVI, it is from the visible spectra of this diol at several acidities that evidence is obtained for the occurrence of two ionization points. In concentrated sulfuric acid there are maxima at 563 and 420 me (e = 134, 000 and 13, 000). . Decreasing the acid strength shifts the position of the maximum as well as decreasing the absorption intensity. In sulfuric acid between 80 and 83% the peak is at 535 mu (6 = 66, 000) (see Figure 11). Compound XLIII in concentrated sulfuric acid was observed to have a maximum at 535 mu(€ = 90, 000). The tentative conclusion is that XLVI first ionizes to a monocation which resembles the ion from XLIII in its resonance interactions. The second ionization produces the dication in which the charges can resonate throughout the whole molecule: H\ __ /H pc \ / O + bf: Hgflce ee’rc. * (67) /H \ Q5 C. The Fluorene Diols and Carbinols Examination of the visible spectra of XLIX and L(see page 58) in varying concentrations of sulfuric acid has clearly shown that L ionizes in a stepwise manner to the dication (Figure 13). . Compound XLIX was‘ shown to have a pKR of about ~12. O, a xmax of 542 mu (shifts to 547 mp. in concentrated acid) and an €max of 66, 000. These data are almost identical with those for the first ionization of L which has a pKR of -12. O, 89 \ a x of 547 mp and an e of 66, 000 (at acid concentration ofu75%, max x i. e. , just prior to the onset of the second ionization). . From this it appears that the monocation of compound L has the same electronic interactions as the cation from XLIX: H H Rafi/gee R 0.5- <® é—> a b (68) Since the positions of the absorption maxima of these two ions are similar to those for the cation derived from compound XLIII (536 mp), they are best described by structures a-c. The dication formed from L showed absorption maxima at 554 mp (6 =62, 000) and 603 mp (e = 215, 000). The latter absorption can be pictured as being due to the following type of electronic interactions: (69) 90 The maximum at 554 mp can be assigned to a species similar to the one attributed to the monocation with the second charge localized on the side chain phenyl ring. @“i H m. The visible spectrum of the cation derived fromcompound LI (Figure 14) can be interpreted in the same way as Lewis explained the spectrum of malachite green (47). There are two modes of electronic interactions that this carbonium ion can undergo: QR} :8 HM / New, + Q5 \. . / (71) Q OC\¢“’ "ii The resonance depicted in equation 70 is similar to that described above for the cation of compound XLIX. Compound LI has an absorption maxi- mum (540 mp) and an absorption intensity (53., 000) that is in close agree- ment with compound XLIX. The absorption at 405 mp is attributed to a diphenylcation-like species, i. e. , resonance as depicted in equation 71. 91 The pK of compound L1 is -6. 5, as compared to -7. 7 for com- R pound XLI. This is further evidence that the fluorene molecule with its planar structure is better than biphenyl as a stabilizer of positive charge (85). It was also shown that compound L1 in sulfuric acid can undergo further reaction after carbonium ion formation. This reaction is instan- taneous in 97% sulfuric acid and non-existent (or very slow) in 80% acid. The rate of the reaction was followed in 93. 9% sulfuric acid (see Figure 15). The time for completion of one-half of the reaction was 22 minutes. On page 45 it was shown that the time necessary for compound XLI to undergo one-half reaction was 135 minutes. Since this reaction occurred more readily in concentrated sulfuric acid and was much faster with the fluorene than with the biphenyl derivative (the former being more susceptible to electrophilic attack) it appears that the observed reaction was one of sulfonation: CH,)n @ (CH2) Q3 HZSO4+O 0 {561—1033. Q {g (72) + H20 where n = 0 or 1 Further evidence for sulfonation is presented on page 102. By comparing the spectrum of LII in concentrated sulfuric acid with the spectra of the other cations studied, spectral assignments can be made (Figure 14). The major absorption at 568 mp is due to resonance throughout the entire molecule: 92 ‘8 gene} H fie <——> + S/ (73) ® + gcrcg <__> etc. The reason that this absorption occurs at a lower wave length. (higher energy) than that of compound L (568 mp 133. 603 mp) is attributed to the increase in steric hindrance relative to resonance stabilization when four instead of two phenyl groups are placed on the carbons in the 2 and 7 positions. The absorptions at 525 and 440 mp can be assigned to resonance structures resembling LI and diphenyl methanol in sulfuric acid, respectively: ¢ ® Q5 @ (74) \CW <<—> 9.0 C/ . <—> e *C- 93 <25 £5 “Q 93 At this point it is of interest to compare the spectra of the mono- and dications in the biphenyl and fluorene series (see Table 14). Correlations can be made between the major absorptions in. the spectra of the dications. ‘ Compound F absorbs at 563 mp. The addition of two phenyls (compound H) lowers the maximum absorption by 33 mp or -16. 5 mp per phenyl. The same effect is noted in going from E (603 mp) to G (568 mp). Here there is a 35 mp shift or -17. 5 mp per phenyl. . The effect of adding a methylene bridge to go from a biphenyl to a fluorene dication also has a. constant effect on the spectra. This change to a planar molecule causes a bathochromic shift. From E to F there is a change of + 40 mp and from G to H a change of + 38 mp is noted. The absorption intensities of the dications also fit a general pattern. The removal of 2 phenyl rings increases the intensities by a constant factor since E/G and F/H both are equal to 1.4. The methylene bridge enhances the intensity by l. 6 (E/F) and 1.7 (G/H). When one compares compounds A-D, no correlations are forth- coming. The cause of this might be attributed to the symmetry in the dication molecules and the lack of it in the monocations. D. The Anthracene Diols and Carbinols The position of the spectral maxima for LIII and LIV in concentrated sulfuric acid are typical of triphenylmethyl and diphenylrnethyl-like cations (see Figure 17). The low emax for these compounds is due to the extra energy needed to stretch the tetrahedral angle (109°28') in the 9 positions of LIII and LIV to 1200. This occurs because the two phenyls in the 10 position also move when this transformation occurs. 7 The expected similarity between the visible spectrum of LIV in sulfuric acid and the monocation from LV is seen in Figure 17. 94 Table 14. Visible Spectra of Some Mono- and Dications in The Biphenyl and Fluorene Series. ‘ Compound Wavelength Molar Absorbance (m H) (6) A 547 66, 000 B 536 90, 000 4. / . C 540 53, 000 ”’90 O 404 17,000 D 510 43,000 4.. /C 420 22,000 /® E 603 215 000 +9”. C + ’ \H 554 62,000 420 13, 000 @ g G 568 152,000 — y 525 46,000 + O \ x + ' \ 420 35,000 H 530 90,000 +M + 442 53,500 a 410 sh 36,000 /@ F 563 134,000 + \ 95 The spectrum of compound LV in sulfuric acid varied with the concentration of acid. In dilute acid (between 55 and 62%) it was possible to determine the pKR for ionization to the monocation. + 9‘ \ f— l . //l H (75) 0 m This pKR was higher than the one for LIV (-8.4 and - 9.6) (Figure 18). The difference in ease of ionization can be attributed to the second hydroxyl group in the molecule. . Since the -OH groups are trans, one can assist in ionization of the other and even stabilize the carbonium ion. Looking along the side of the molecule will help in seeing this. @HOMWQ ; "9H,: ”6’ Between 62 and 75% acid, the cis analog of LV is formed (see section IV and equation 85). . Figure 19 shows the change from the monocation to the cis diol in 62% acid. In 78% sulfuric acid LV forms a blue solution. Inspection of Figure 20 shows the striking similarity between the spectrum of this solution and that of 9, lO-diphenylanthracene (LXIV) in 97% sulfuric acid. LXIV is one of the hydrolySis products of LV in 97% sulfuric acid (see page 69). It is proposed that the cis-diol (LXV) does not undergo ordinary 96 ionization but, due to the proximity of the hydroxyl groups, is involved in an intramolecular hydride transfer reaction forming, LXIV. . It can be seen, therefore that dication formation does not occur. 0 ¢ ,——> see-m ”7’ $3.17 The visible spectrum ‘of LV in 97% sulfuric acid after standing for 16 hours is very similar to that of a sulfuric acid. solution of 4-phenyl- ' .2, 3-benzof1uoranthene (LXVI), another hydrolysis product (61) (see Figure 12). q i 11W The visible spectrum ofLV in 97% sulfuric acid when taken immediately after mixing looks like a combination ofwa and LXVI (Figure 21). t It appears, therefore, that LXIV is formed first and this in turn reacts further to give LXVI. Since there are unidentified products obtained from the hydrolysis experiment (61), no attempt will be made to propose a mechanism for'the formation of LXVI from. LXIV. - E. Molecular-Orbital Approach In recent years, the use of quantum mechanics has entered the field oforganic chemistry (86, 87). It is now used to explain the absorption 97 spectra of complex molecules, including those of carbonium ions (17). Simply, if one compares the relative energies of the highest bonding, non- bonding and lowest antibonding 1T orbitals of an aromatic compound and the arylmethyl ions derived from it, they appear as pictured in Figure 25 (17). Lowest antibonding _ — Non-bonding ENERGY \ ArH Arse AZCH" Highest bonding Figure 25. The relative energies of the lowest antibonding, non- bonding, and highest bonding n-orbitals of an aromatic hydrocarbon (ArH), and of the symmetrical diaryl- methyl (ArZCH) and triarylmethyl (Ar3C) systems. It can be demonstrated that the antibonding and bonding Tl' orbitals in aromatic compounds and in the derived arylmethyl ions are paired. In the ions there is an extra orbital which is non-bonding that bisects the other orbitals at the center of gravity of their energies (88). In the Huckel approximation, the highest bonding and lowest antibonding levels of the arylmethyl ions have the same energies as the respective levels in the aromatic hydrocarbon from which they are derived. On page 5 it was stated that by neglecting various factors, the ion should absorb at a wavelength twice that of the hydrocarbon from which it is derived. This is true since the A E needed for the excitation of an electron in a hydrocarbon is twice that of the ion and A E is inversely proportional to the wavelength. 98 Molecular orbital theory would predict, therefore, that the cations and dications studied in this the sis should absorb at a wavelength. twice that of the hydrocarbon's principal absorption band. The hydrocarbons involved are benzene, biphenyl and fluorene. They absorb at 207, 250 and 300 mp respectively. If a cation is derived from two different hydro- carbons, the spectrum should have absorptions due to electronic transitions from both of them. Table 15 gives the predicted and observed values for several of the cations and dications studied in this work. . The reason that it was not predicted that the second compound in Table 15 had an electronic transition due to biphenyl is caused by the position of the charges on the biphenyl molecule. At the point of juncture of the two rings making up biphenyl, the molecular orbitals cancel. . This is the same as saying in valence band terms that there are no resonance interactions between the rings. It is seen from Table 15 that at times molecular orbital theory pre- dicts the spectra of the cations quite well. At other times it is in error. It is possible that with more refined theories one will be able to predict these spectra more closely. IV. Hydrolysis and Methanolysis Products On 'pag e 8 , the importance of quenching experiments was dis- cussed. . In this connection, when a solution of a, 0., o), c'-tetraphenyl-o- . xylene-0., a'-diol in 98 % sulfuric acid was poured onto ice, tetraphenyl- phthalan (XXXV) re sulted: 01¢ (AC/0 0 e / ‘OH HO \ 0H“-> 4‘9 ' / 0’00 ¢’C\¢5 g5 05 X2901 [XVII XXXXZ + .4. (78') Table15. . Observed and Predicted (by the Huckel Approximation) Absorption Maxima for Several Cations ‘and Dications C ompounds X max Predicted Observed are? +C + 500 525 Q5: Q5” @@ 410 430 C§D CDC-(D w, He; :3: :i: Q3 EJCQ—OCE, “C $2 410 404 Cg 600 p 540 mg ' Q . Q) 410 554 QCQ QC H 600 .603 410 . 440 Q3 0 600 525 fag?) 5.. 410 .. 500 535 410 420 500 563 600 547 1 100 -Th.is is consistent with the fact that efforts to prepare the corresponding dichloro compound with reagents such as thionyl chloride,. acetyl chloride or hydrogen chloride also resulted in the phthalan (89). WhenzLXVII was poured onto‘ice-cold methanol or ethanol, an interesting product was formed: g OCH3 ©CHZCH3> p- CH3OH Tor\ LXV” CH3CHZOH’ (79) 0 0 LXVIII . Possible mechanisms, via the same intermediate, for; the formation of XXXV and LXVIII are: 7mm“ (8°) 30 O I \V ® LXVW _ where R 7- -H, -CH3 or r-CHZCH3 When R =IH, a protoncan readily be lost but when R a! H, an unstable methyl or ethyl cation would have to be eliminated. . When R )6 H, therefore, 101 it is reasonable, since the dication in sulfuric acid reacts instantaneously when added dropwise to excess alcohol (immediate disappearance of the red color), that the dication reacts with alcohol and not with another nucleOphile. The alkylation step is reasonable because of the close proximity of the benzene ring to the cation. An analogous type of intra- molecular alkylation was shown on page 2 7 in the formation of 10, 10- dipheny1-9-anthrone from o—benzoyltriphenylmethanol. If the mode of addition were reversed, sulfuric acid would be in excess and it should react with the alcohol as it was added: ROH + H2804 a ROSO3H + H20 (82) - . + H20 + H2504 —> H504 + H30 (83) Reaction between the dication and either water or hydronium ion should result in the formation of tetraphenylphthalan. This was demonstrated by addition of methanol to the dication, which resulted in formation of tetraphenylphthalan. An interesting but as yet unclarified reaction is the formation of tetraphenylphthalan in 90% yield when LXVII is poured onto anhydrous tetrahydrofuran. 9, 10, lO-Tripheny1-9, lO-dihydro-9-anthrol (83%) resulted from the hydrolysis of a sulfuric acid solution of LXVIII. When the red solution of a, a'-dichloro-o., a, a',n'-tetraphenyl-m- xylene in 98% sulfuric acid was poured onto water and methanol, the corresponding diol and di-ether were formed respectively. The respective diols were also obtained when sulfuric acid solutions of a, a, u', a'-tetra- phenyl-p, p'-bitolyl-a, a'+»diol and a, o.‘ -dichloro-a, a, o.‘ , (1' -tetrapheny1-m, m'-bit01yl were hydrolyzed. In a manner analogous to the formation of tetraphenylphthalan from LXVII, a sulfuric acid solution of a, a, u', a'-tetraphenyl-o, o'-bitolyl- a, a'-diol formed the anhydro derivative (98): 1102 @‘Q +C‘Q @‘,C /C:@ . 34 Q) Q3 @‘0 CD ( ) The other dications were not subjected to quenching experiments but two monocation we re: 9i; . 6+0 , ©fi . When a- solution of either LXIX or LXV in concentrated sulfuric acid was poured onto water, no precipitate resulted. If dilute sulfuric acid was used, the respective carbinols were obtained in high yield. This is‘fur- ther evidence that sulfonation is occurring in concentrated acid (see page ‘91 ). Additional proof that sulfonation, occurred was obtained by exami- nation of the infrared spectrum of a carbon tetrachloride extract of the hydrolysis solution of LXX... The presence of sulfonic acid bands were indicated (see Figure 45 ) (90). After refluxing this hydrolysis solution for several hours, the carbinol was isolated. The spectral (page 58) and hydrolysis studies, therefore, indicate that in concentrated sulfuric acid LXIX and LXX are sulfonated. 103 It was shown in 1923 that hydrolysis of the blue solution of trans- 9, lO-diphenyl-9, lO-dihydroxy-9, lO-dihydroanthracene (LV) in concen- trated sulfuric acid gave 9, lO-diphenylanthracene and an unidentified red- yellow compound (91). This work was recently repeated and 4-pheny1-2, 3-benzofluoranthene was isolated in addition to 9, lO-diphenyl- anthracene (61). . Rafos also found that hydrolysis of a solution of the diol in 6% sulfuric acid in acetic acid (at this acidity the diol was found to be half ionized to the monocation) afforded nearly quantitative yields of the starting diol. . Spectral studies of LV in dilute sulfuric acid indicated that the monocation (yellow solutions), once completely formed, reacts further to give an almost colorless species. . Since the monocation is completely formed in 62% sulfuric acid, and the species giving the blue solution is formed in 75%, it was decided to prepare a solution of LV in 65% sulfuric acid and study its hydrolysis products. The results of this experiment were a mixture of LV and its cis analog (LXV). The reaction sequence going from the trans to the cis diol can be illustrated as follows: (23 Q5 @\\ OH 104 In section III-D a further discussion of the reactions of LV in sulfuric acid was presented. V. Cryoscopic Measurements ‘ Cryoscopic measurements in this thesis were made only on tetra- phenylphthalan in 100% sulfuric acid since there was a possibility that the resulting dication would not form because of the proximity of the two positive charges. Since the same species was formed in concentrated sulfuric acid from either tetraphenylphthalan or 0., 0., a', a'-tetraphenyl-o- xylene-a, a'-diol (see page 35) only one was studied cryoscopically. The former was chosen because it should ionize to give 5 particles while the ‘latter' should form 7. A smaller number of particles results in more accurate i factor 3 . Q5 Q3 - QC? 0 + 3 HZSO4——> 6+H30 +3HSO; 0 Q5 Q/ @ i=5 (86) €23,395 o. 93 \OH » + — C(OH + 4HZSO4—.—> .: + 2H,).O+4HSO4 ¢ ¢ g/ g [=7 (87) 0+ +0 _ The data, which are given in Table 16, support the formation of a dication 7 from tetraphenylphthalan. 105 Table 16. Cryoscopic Data on Tetraphenylphthalan Sample Wt., g.a g. sto, T1,OC. T,°c ib’c 0.0 9 262 0.0 0.460 107.4 8.969 0.308 4.99 0.490 107.4 8.637 0.332 5.05 , 0.450 107.4 8.301 0.326 5.39 0.950 106.8 8 637 0.640 4.99 0.94 106.7 8.301 0.658 5.05 1.41 106.3 8.301 0.966 5.18 a The first three values are the actual weights of the samples. The other weights incorporate two or three samples. bCalculated from i = AT/6. 12 - ms, CThe average value for 1 is 5.11. where ms is the molality of the solute. EXPERIMENTAL 106 107 I. Syntheses and Reactions A. The Xylene Diols Preparation of a, a, a', (1'-Tetraphenyl-o-xylene-a, 0' —diol (89) To a mechanically stirred solution of 84 ml. (126 g. , 0. 8 mole) of bromobenzene in 300 ml. of anhydrous ether was added 9 g. (l. 3 g. . atoms) of lithium slices, all at once. Stirring was continued at room temperature for one hour under a dry nitrogen atmosphere. . The re- action mixture was then placed in an ice bath and 38. 8 g- (0. 2 mole, 37. 5 ml.) of dimethyl phthalate was added dropwise. After 3 hours the stirring was stopped and the reaction mixture was allowed to stand over- night under a nitrogen atmosphere. The free glycol was obtained by pouring the ether solution onto 300 m1. of an ice-water mixture and collecting the resulting solid by filtration. The crude glycol was taken up in 100 ml. of benzene, dried over magnesium sulfate, treated with Norite and filtered. . The glycol was precipitated by adding petroleum ether (35-600) and cooling in a dry ice-isopropyl alcohol bath. The yield was 70 g. (78%) of a, 0., (1', o.‘-tetraphenyl-o-xylene-o., o.'-diol, m.p. 2.03- 2040 (literature value, 203. 5°). Its infrared spectrum is shown in Figure 26. . Preparation of 0., cl, 0', a'-Tetraphenylphthalan (89) 0., a, a', a'-Tetraphenyl-o-xylene-a, cl'-diol (l. 8 g; , .4. l mmolés) was dissolved in 70 ml. of glacial acetic acid and 5 ml. of water. The solution was refluxed for one hour. As the solution cooled, white crystals began to precipitate. After standing overnight, 1.1 g. of solid was collected. Addition of water to the mother liquor afforded an additional 0. 2 g. of solid (total yield, 75%). The melting point of the tetraphenyl- phthalan was 174-1760 (literature value, 174-1750). Its infrared spectrum is shown in Figure 27. 4061.8 .duocoH>XtOIH>Gm£mmnuouu.o .78 .o to mo 85.30on UMHNHHGH om. ohsmfm a; nuwoofioim? 4. H o_._ m w _ _ _ _ _ _ 108 T I)! o :1: a) .9 db (\I U) U (I 109 «Q Ma Nfi AH OM .cmfimflfiaacoammuuout.d.o .d to Ho 6530on @932de A3 sumaoaourmg o m N. o m .3 6.33m 110 Reaction of 0., a, a', o.‘-Tetraphenyl-o-xylene~—c1, o.‘-diol in 98% Sulfuric Acid with Water 0., 0., a', d'-Tetrapheny1-o-xylene-a, d'-diol (0.119 g. , 0. 27 mmoles) was dissolved in 10 ml. of 98% sulfuric acid. The resulting orange—red solution was poured onto 50 ml. of ice-water. The white solid was collected, dissolved in 15 ml. of benzene, washed three times with lO-ml. portions of 5% sodium carbonate solution and then dried over calcium chloride. The solution was filtered and the benzene was distilled in vacuo to give 0.960 g. (84%) of tetraphenylphthalan, m.p. 164-1690. . Two re- crystallizations from benzene raised the melting point to 173-1740 (literature value, 174-1750). A m.m.p. with an authentic sample gave no depression. Reaction of 0., 0., a' , a'-Tetrapheny1-o-xy1ene-o., a'-diol in 98% Sulfuric Acid with Absolute Methanol 0., 0., a', o.‘-Tetraphenyl-o-xylene-a,o.'-diol (l . 00 g. , 2. 25 moles) was dissolved in 10 ml. of 98% sulfuric acid. The orange-red solution was poured onto 75 m1. of ice-cold absolute methanol with magnetic stirring. The resulting white precipitate was washed with 5% sodium carbonate solution, with water and dried to give 0.885 g. of a compound melting at 208-2110. Two recrystallizations raised the melting point to 220. 5-221°. The compound had a melting point almost identical with that of the methyl ether of 9, 10, 10-triphenyl-9, 10-dihydro-9-anthrol (92). £111.; Calcd. for C33H26O: C, 90.37; H, 5.97 Found: C, 89.86, 89.80; H, 6.21, 6.16. The yield of substituted anthrol was 89%. The infrared spectrum is shown in Figure 28. 1L ‘ 1., \ 111 .Houficmt Ha OH A3 #3953225 o w .ououpiiptoa .mougconmmnuuoa .oH .0 Ho .3530 T3308 2.3 m0 83302? 00.3.33 .mm 0.339th _ N8 i _ _ . _ NV C x 73031—7 I 112 Reaction of Absolute Methanol with a, 0., a', o.‘—Tetraphenyl- o-xylene-a, o.‘-diol in 98% Sulfuric Acid To the magnetically stirred orange-red solution of 0., 0., u', a'-tetra- phenyl-o-xylene-a, a.‘-diol (0.150 g. , 0. 34 mole)” in 50 m1. of 98% sulfuric acid, 50 m1. of absolute methanol was added dropwise. The color lightened slowly with the formation of a white solid. . When ad- dition was complete, the white solid was filtered, washed with 5% sodium carbonate solution, with water and dried at 1000. The m.p. was 175-”176O and a m.m.p. with tetraphenylphthalan was also 175-176°. The yield was 0.120 g. (80%). Reaction of 0., a, a', 0.‘-Tetraphenyl-o-xy1ene-a, o.‘-diol . in 98% Sulfuric Acid with Absolute Ethanol The procedure and amounts used were identical with those for the reaction with absolute methanol (page 10 ), except that absolute ethanol was used. The yield was 0.850 g. (84%) of the ethyl ether of 9,10,10- triphenyl-9, lO-dihydro-9-anthrol, m.p. after recrystallization from toluene—ligroin (90-1200) was 254-2550 (literature value, 2500) (92). Its infrared spectrum is shown in Figure 29. Reaction of a, a, u', o.‘-Tetrapheny1-o-xylene-a, o.'-diol . in 98% Sulfuric Acid with Anhydrous Tetrahydrofuran 0., a, a', o.‘-Tetraphenyl-o-xylene-a, ay'-diol (0:.20017g. ,, 0.45, 'zmmoles) was dissolved in 3 m1. of 98% sulfuric acid. J The red solution was poured into 50 ml. of anhydrous tetrahydrofuran with stirring. The resulting clear solution was neutralized with solid sodium carbonate, filtered and the tetrahydrofuran distilled to give 0. 180 g.. (90%) of a white solid residue. After one recrystallization fromligroin (90-1200) it melted at 174-1750 and gave no depression in melting point with an authentic sample of tetraphenylphthalan. Acufiucdtouonosnnmvuom .ougcgfimwnung .2 .a mo 99.30 3:30 0:» no 5530?? 00.3.35 .mN 0ndmfm 1.: Aumc0H0>m>> 2 NH 3 OH 0 m m. p m «4 m _ 3 4 A 3 _ _ _ _ a V? 113 mmol v m _Tllmmo .0101+ . 114 Preparation of m- Dibenzoylbenzene To a stirred mixture of 26.8 g. (0. 20 mole) of anhydrous aluminum chloride in 100 ml. . of dry benzene, 20. 3 g.. (0. 10 mole) of isophthaloyl chloride in 100 ml. of dry benzene was added dropwise. The temperature was maintained between 0-100 during the addition. . The reaction .mixture was refluxed for 5 hours, poured onto 300 ml. of iceéwater and the organic layer separated. After washing twice with 50-ml. portions of 2N sodium hydroxide and water, the benzene solution was dried over calcium chloride. Distillation of the benzene left 25. 5 g.. (89%) of m-di- benzoylbenzene, which melted at 1080 after three recrystallizations from 95% ethanol (literature value, 101-1020) (93). Reaction of m-Dibenzoylbenzene with PhenyMagnesium '. bromide 1 ‘ To a stirred solution of phenylmagnesium bromide (0. 25 moles) prepared in the usual way from 39. 3 g. (26. 2 ml.) of bromobenzene and 6 g. of magnesium in 100 ml. of anhydrous ether, 25 g. (0.087moles) of m-dibenzoylbenzene in 250 ml. of dry benzene was added dropwise. After being stirred for 4. 5 hours, the reaction mixture was decomposed by pouring onto 300 m1. of an ice-water mixture. . The organic layer was separated, dried over calcium chloride and the benzene distilled. An infrared spectrum of the resulting oil showed strong absorption in the carbonyl and hydroxyl regions. Attempts to recrystallize the product from various solvents were unsuccessful. Preparation of 0., a'-Dichloro-a, 0., 0', o.‘ - tetraphenyl-m-xylene Phehylmagnesium bromide (0. 90 mole) was prepared in the usual way from 141 g.. (94 ml.) of bromobenzeneand 21.6 g. of magnesium in 200 ml. of anhydrous ether. Dimethyl isophthalate (30 g. , 0.15 mole) 115 in 200 ml. of benzene was added dropwise to a stirred solution of the Grignard reagent. After refluxing for 8 days with stirring, the reaction mixture was decomposed by pouring onto ice-hydrochloric acid, the benzene separated and dried over magnesium sulfate. After distilling the benzene, the resulting oil was triturated with-ligroin. (90-1200) to remove any biphenyl and bromobenzene. The oil was then dissolved in 50 m1. of acetic acid-acetyl chloride (3: 1) and treatéd with anhydrous hydrogen chloride for 3 hours. After standing in the refrigerator for one week a brown solid was formed. One recrystallization from acetyl chloride gave 8 g. (11%) of a tan solid, m.p. 125-130°. Further recrystallization from acetyl chloride raised the melting point to 138-1400 (literature value, 140. 5°). The infrared spectrum is shown in Figure 30. Preparation of 0., a, o.', a'-Tetraphenyl-m-xylene- a, o.‘-diol (94) c1, a'-Dichloro-o., 0., a', a'-tetraphenyl-m-xylene (0. 2 g. , 0. 42 Mole) was dissolved in 6 m1. of 98% sulfuric acid. After the evolution of hy- drogen chloride had stopped, the red solution was poured onto 50ml. of ice-water. The white solid was filtered, washed with 5% sodium carbonate solution, with water, dried at 400 and recrystallized from ligroin (90-1200) to give a white crystalline solid which melted sharply at 870. This was apparently a complex between the glycol and the solvent because if the melt was resolidified or if the compound was dried over- night in a drying pistol, the melting point was 112-1130 (literature value, 112-113°). The yield of 0., a, a', o.‘-tetraphenyl-m-xylene-a, a'-diol was 0. 175 g. (94%). Its infrared spectrum is shown in Figure 31. Reaction of a, 0.‘ -Dichloro-a, 0., o.‘ , a'-tetraphenyl-m- xllene in 98% Sulfuric Acid with Absolute Methanol (95) a, u'-Dichloro-a, u', a', a'-tetraphenyl-m-xylene (0. 1 g. , 0. 21 mmole) in 5 ml. of 98% sulfuric acid was added to 50 ml. of ice-cold methanol. 116 4:03.300 ~93;0aoa>xlfium>n0£md30uub ...d..d 619333013 .0 mo 55.30090 00.3.35 . .om 0.193%: 43 £3wd0H0>0>> 0H ma NH 3 A: , o w -h o .m w. _ a _ _ _ _ a; _ _ 117 .Acoflgow nmUvAoflTB .duodmaxugugdwamduumau.d ..d to .6 mo afiuuovmm vmanwdH . .Hm 9:,th A3 suwcvfigd? MA A H A: a m N. o m w m _ A _ 1_ m _ _ u _ 118 The resulting clear solution was treated with 50 ml. of 5% sodium carbonate solution and the white solid that formed was filtered, washed with water and recrystallized from aqueous methanol. The dimethyl ether of a, 0., o', o.‘-tetraphenyl-m-xylene-o., o.‘-diol (0. 08 g. , 82%) melted at 1030 (literature value, 103-1040). Its infrared spectrum is shown in Figure 32. . Preparation of a, a'-Diphenyl-m-xy1ene-a, o.‘-diol (96) To a magnetically stirred slurry of 0. 05 g. (1. 2 mmoles) of lithium aluminum hydride in 30 m1. of anhydrous ether, 0. 11 g. (0. 38 mmole) of m-dibenzoylbenzene in 25 ml. of dry benzene was added dropwise. After refluxing for 12 hours the reaction mixture was decomposed by pour- ing onto an ice-sulfuric acid mixture. . The benzene layer was separated, dried over magnesium sulfate and the benzene distilled to give 0.08 g. (70%) of a, a'-diphenyl-m-xy1ene-a, a'-diol. After one recrystallization from toluene-ligroin (90-1200) its melting point was 153-1550 (literature value, 157°). Preparation of 0., d'-DLphenyl-p-xylene-a, a'-diol (96) Terephthalaldehyde (13.4 g. , 0.1 mole) in 300 ml. of an anhydrous benzene-ether mixture (5:1) was added to 0. 2 mole of phenylmagnesium bromide (from 47. 1 g. of bromobenzene and 7. 2 g. of magnesium) in 100 ml. of dry ether. After being stirred for 1. 5 hours, the reaction mixture was poured onto 300 ml. of an ice-hydrochloric acid mixture. The resulting solid was filtered to give 5. 1 g. (18%) of 0., a'-diphenyl- p-xylene-o, o.‘-diol. After two recrystallizations from ethanol it melted at 1730 (literature value, 171°). \~\a\~.. .Acofldfiow NmUv.Ho€n.d .duodoaxuguadonmdupouu.d ...o .d to no .350 15083 o5. Ho 5350on woumnmfl 1.: numdoaocrdg «1 ma . Na 3 3 my .w :3 8:sz —I[~ --u\o —)|,n _ _ _ _ _ _ _ 119 120 B. The Diols and Carbinols in the Biphenyl Series Preparation of 4, 4'-Dibromobiphenyl (97) An evaporating dish containing 15.4 g.. (0. 1 mole) of biphenyl was set on a porcelain rack in a desiccator. Below the rack was placed an evapor- ating dish with 39 g. (12 ml. , 0. 24 mole) of bromine. . The desiccator was closed but a small opening was left for hydrogen bromide evolution. After 12 hours, the top dish was removed and placed in the hood for 4 hours so that bromine and hydrogen bromide could escape from the crystals. Recrystallization from benzene gave 19. 3 g. (6279) of 4, 4'-dibromo- biphenyl, m.p. 163—164° (literature value, 162-1630). 7 Attempted Preparation of a, 0., a', o.‘-Tetraphenyl-p5p'-bitolyl- 0., o.’-diol To 1 g. (0. 14 g. -atoms) of lithium in 25 ml. of anhydrous ether, 15.6 g. (0. 05 mole) of 4, 4'-dibromobiphenyl in 50 ml. of anhydrous ether was added dropwise. After stirring for 3 hours, 18. 2 g.. (0.1 mole) of benzo- phenone in 50 ml. of dry ether was added dropwise and the reaction was refluxed for 24 hours. The reaction mixture was poured onto ice, extracted with toluene and driediover magnesium sulfate. . Most of the toluene was distilled and the residue was placed in the freezer for 2 days. The resulting viscous orange solid could not be recrystallized. Benzoylati on of BiphenLl To a stirred slurry of 79. 8 g. (0. 6 mole) aluminum chloride in 75 m1. of carbon disulfide, 84 g.. (0. 6 mole) of benzoyl chloride and 31 g. (0. 2 mole) of biphenyl in 75 ml. of carbon disulfide were added dropwise. After being refluxed for 25 hours, the reaction mixture was poured onto 500 ml. of ice-water and the resulting solid collected by filtration. The filtrate was set aside for further work-up. 121 The solid was recrystallized from isopropyl alcohol to-give 6. 8 g. (9%) of 4,4'-dibenzoylbiphenyl, m.p. 218o‘(literature value, 2169) (98). After the organic layer was separated in the filtrate, it was dried over magnesium sulfate, filteredand the benzene distilled. . Recrystal- lization 'of the residue from benzene-petroleum ether (35-600) gave 23.0 g. (45%) of 4-benzoylbiphenyl, m.p. 101° (literature value, 104°) (99). . Preparation of a, a'-Dichloro-a, 0., o.‘ , o.‘ - tetraphenyl- p, p' -bitol£ To a stirred solution of phenylmagnesium bromide (0.055 mole) prepared in the usual way from 8. 64 g.. (5. 8 m1.) of bromobenzene and 1. 35 g. of magnesium in 100 ml. of dry ether, 9.1 g. (0.025 mole) of 4, 4'-dibenzoy1biphenyl in. 100 ml. of dry benzene was added dropwise. After being refluxed for 3 hours, the reaction mixture was decomposed by pouring onto 200 ml. of an ice-sulfuric acid mixture. . The organic layer was separated, washed with 51% sodium carbonate and the benzene was distilled. The residue was steam distilled to remove side products and starting materials. a It was then taken upiin 501ml. of ether, dried over calcium sulfate and filtered. Acetyl chloride (10 ml.) was added and the solution was treated with anhydrous hydrogen chloride for 3 hours. The 0., aJ-dichloro-o, 0., 0', a'-tetraphenyl-p, p'-bitolyl (4. 1 g. , a 32%) was collected. After one recrystallization from acetyl chloride- benzene it melted at zz3-224° (literature value, 2239) (100). The infrared spectrum .is shown in' Figure 33. . Preparation of 0., 0., 0', 0' -Tetraphenyl-p£ - bitolyl-a, a'-diol ‘ " Method A: n-Butyllithium (0. 16 mole) (from 2. 9 g. of lithium slices and 22. 8 g. of n-butyl bromide) in 75 ml. of dry ether was added all at once to a mechanically stirred solution of 13 g.. (0.041 mole) of -' , . ‘ 122 mg a; -. a p a ,. w .AGOEEOm ~m0v T33R73 .mugcoammnuounb ...o o duouozofloub 6 mo 55.3.0on pandas: mm on: Tm 3v numcmaocfiwg Ma Ma .2 OH 0 m N. c D 4)Ln — _ _ _ a _ a _ _ 123 4, 4'-dibromobiphenyl in 125 ml. of anhydrous ether. After 12 hours, 15. 2 g- (0. 083 mole) of benzophenone in 150 ml. of dry ether was added dropwise and stirring was continued for another two hours. The reaction mixture was poured onto ice-water to give an oily solid which was collected by filtration. . Recrystallization from ethyl acetate-ligroin (90-120°) gave a very small amount of a solid, m.p. 177-1780. The literature value for 0., 0., 0', o.‘-tetraphenyl-p,p" -bitolyl-0., d'-diol is 177-1780 (63). Method B: a, 0.‘-Dichloro-a, 0., a', o.‘-tetrapherrylep, p'-bitolyl (0. 110 g. , 0.19 mmole) in 75 ml. of 25% aqueous acetone was refluxed for 15 minutes. After cooling to room temperature, the reaction mixture was poured Onto 150 ml. of ice-water and extracted with ether. After drying over calcium sulfate, the ether was distilled to leave a white residue which was recrystallized to give 0. 07 g. (70%) of a, a, (1', u'-tetraphenyl- p, p'-bitolyl-a, o.‘-diol, m.p. 177-1780 (literature value, 177-1780) (63). Its infrared spectrum is shown in Figure 34. . Reaction of a, a, 0', o.‘-Tetraphenyl-p, p'-bitolyl- a, o.‘-diol in 98% Sulfuric Acid with Water o, 0., a' ,0.! -.Te,traph.e.ny_1rp. plybitolyl-u,nil-.diollo '.,0.6..gi‘,‘ Q-.)12.mm'ole)was dis- solved in 10 m1. of 98% sulfuric acid. . The red solution was poured onto 75 ml. of ice-water and the resulting solid was collected. After being washed with 5% sodium carbonate, with water and dried, the crude a, (1,-pry, 0'- tetraphenyl-p; ‘p"-abitol,yli-n., a('J-di013‘(Q. 05- g. 284%) was recrystallized from toluene-ligroin (90-1200). The m. p. and m.m.p. were identical with that of an authentic sample of the glycol. Pr eparati on of 4 - Bipherylyldiphenylm ethanol (1 01) To 3.1 g.. (0.13 g. —atoms) of magnesium was added 30 g. (0.13 mole) of 4-bromobiphenyl in 175 ml. of anhydrous ether dropwise with stirring. 124 Mg . Acofidaom nHQmUV 331.6 51133313 .mufwconmdpuoulb ...o .d .8 mo Eduuoomm condemn: A3 nmmaofioxfim? 2 A a S . e m w o m a. .3 93th ._ _ _ _ A _ _ _ 125 After 5 hours, 18. 2 g- (0.10 mole) of benzophenone in 100 ml. of dry ether was added dropwise. The stirring was continued for 12 hours, at which time the reaction mixture was poured onto 500 ml. of an ice- sulfuric acid mixture. The ether layer was separated, washed with 5% sodium carbonate solution, with water and the ether distilled. . The residue was steam distilled for two hours, taken up in benzene, dried over mag- nesium sulfate and the benzene distilled. Recrystallization of the residue from cyclohexane gave 11. 2 g. (33%) of 4-Biphenylyldiphenylmethanol, m.p. 134-135O (literature value, 136°). Its infrared spectrum is shown in Figure 35. Reaction of 4-Biphenylyldiphenylmethanol in 98% Sulfuric Acid with Water 4-Biphenylyldiphenylmethanol (0. 08 g. , 0. 24 mmole), was dissolved in 10 ml. of 98% sulfuric acid. After being magnetically stirred for 10 hours, the red solution was poured onto 100 ml. of an ice-water mixture to give a clear solution. The aqueous solution was refluxed for 5 hours, cooled and extracted 3 times with 25-ml. portions of ether. The ether layer was washed 2 times with 15 ml. portions of 5% sodium carbonate, dried over magnesium sulfate and the ether layer distilled to give 0.06 g. (75%) of 4-biphenylyldiphenylmethanol, m.p. 135°. A.m.m.p. with an authentic sample gave no depression. . Reaction of 4-Biphenylyldiphenylmethanol in Dilute . Sulfuric Acid-Acetic Acid with Water I To 4-biphenylyldiphenylmethanol (0. 01 g. , 0. 3 mmole) in 5 ml. glacial acetic acid, 2 drops of 98% sulfuric acid was added. The red solution was cooled in an ice bath for 15 minutes and then poured onto 150 ml. of an ice-water mixture. . The resulting white precipitate was collected to give a quantitative recovery of 4-biphenylyldiphenylmethanol. . It was identi- cal in all respects with an authentic sample. 126 u: Ma Nd A3 aumcoaouwm? 3 OH 0 m N. o .__J(_n _ q a _ _ u 127 ‘ Preparation of 2-Biphenylyldiphenylmethanol 2-Bromobiphenyl (10 g. , 0. 043 mole) in 50 ml. of dry ether was treated with lithium (O. 6 g. , 0. 086 g. -atoms) to form 2-1ithiobiphenyl. Benzophenone (7. 8 g. , 0.043 mole) was added with stirring and the re- action was refluxed for two hours. The reaction mixture was poured onto 200 ml. of an ice-water mixture, extracted with benzene, steam distilled in the presence of sodium carbonate and the residue extracted with benzene. After being dried over magnesium sulfate, the benzene was distilled and the residue recrystallized from petroleum ether (35-600). to give 6. 2 g. (43%) of 2-biphenylyldiphenylmethanol, m.p. 91-92O (literature value, 87—880) (84). Its infrared spectrum is shown in Figure 36. . Preparation of 4-Biphenylylphenylmethanol (102) The procedure and amounts used were identical with that for the preparation of 4-biphenylyldiphenylmethanol (see page 123)except that 10. 6 g. (0. 1 mole) of benzaldehyde was used in place of the benzophenone. The yield was 23.0 g. (88%) of 4-biphenylylphenylmethanol which, after one recrystallization from cyclohexane, melted at 93-940 (literature value, 960). The infrared s ectrum is shown in Fi ure 37. P g Preparation of 0., 0' -Diphenyl-p, p'-bitolyl-a, 0' -diol Lithium aluminum hydride (0. 5 g. , 12 mmoles) was added all at once to a magnetically stirred solution of 1. 0 g.. (2. 8 mmoles) of 4,4'dibenzoyl- biphenyl in 40 ml. of pyridine. After being heated at 700 for 2 hours, water was added dropwise to decompose the excess lithium aluminum hydride. When the reaction mixture was poured onto ice—hydrochloric acid a white solid precipitated and was collected by filtration. Recrystallization twice from ethanol gave 0. 36 g- (35%) of 0., o'-diphenyl- p, p'-bitolyl-o., a'-diol, m.p. 183°. £131. Calcd. for C26H22023 C, 85. 21; H, 6.05. . Found: C, 85.01;. H, 6.17. 128 NH AH .Acoflgaum JUEUV HocmefioEfi>aonmwpa>fi>cogmfium Ho 5330on 60.3935 . .om 0.23er A3 Hanged/m? ca d w n o m a. m N _ _ _ a a . ._ . 129 MH HM .AcofldHOm NmUV HocmafioEKConmaacosQEuw mo 5.9.30on @2033 G: numnofiou/m? OH 9 m ~ 0 m w .2 $de _ _ _ a a _ 130 'Reaction of 4-Biphenylyldiphenylmethanol with Benzoyl chloride and Aluminum chloride To a mechanically stirred solution of 4—biphenylyldiphenylmethanol, (6. 3 g. , 0. 018 mole) in 75 m1. of methylene chloride was added 5. 3 g. (0. 04 mole) of aluminum chloride. To this red solution (presumably due to carbOnium ion formation) 2. 6 g. (0. 018 mole, 2. 2 ml.) of benzoyl chloride in 50 ml. of methylene chloride was added dropwise. . The reaction was refluxed for 3 hours, the organic layer separated and the methylene chloride evaporated in a stream of air. The residue was treated with 50 ml. of 5% sodium carbonate, extracted with benzene, dried over magnesium sulfate and the solvent distilled. Since an infrared spectrum of the residue showed only a small amount of carbonyl present, the reaction was repeated under more vigorous conditions. The procedure was the same as above with the following exceptions: 8.0 g. (0.06 moles) of aluminum chloride, 75 ml. of chloroform (in place of methylene chloride) and 8 hour reflux were used. After the same work-up, an infrared spectrum on the crude product still showed only a trace of carbonyl present. Reaction of 4-Benzoylbiphenyl with Zinc Cyanide and Hydrochloric Acid T Hydrogen chloride was bubbled into a stirred solution of 10 g. (0. 04 mole) of 4-benzoylbiphenyl and 7 g.. (0.06 mole) of zinc cyanide in 150 ml. of tetrachloroethane for 1. 5 hours. After aluminum chloride (23. 2 g. , 0. 18 mole) was added portionwise with ice-bath cooling, hydrogen chloride was bubbled in for 2 hours at 300 and 3 hours at 500. The reaction mix- ture was poured onto an ice-hydrochloric acid mixture, stirred for 3. 5 hours and heated to boiling for 2 hours. The organic layer was separated, washed with 100 ml. of a 5% sodium carbonate solution and steam dis— tilled. . The product was a dark oil which was insoluble in organic solvents. 131 Preparation of 4- Bromo-4' - benzoylbiphenyl To a stirred mixture of 14 g. (0.06 mole) of 4-bromobiphenyl and 9.8 g. (0.07 mole, 8.1 ml.) of benzoyl chloride in 150 ml. of carbon disulfide, 9. 3 g. (0. 07 mole) of aluminum chloride was added portionwise. After being refluxed for 28 hours, the reaction mixture‘was poured onto 500 ml. of ice-water, the organic layer separated and the carbon disulfide removed by distillation. » The residue was treated with 250 ml. of a boiling 5% sodium carbonate solution and the solid filtered. Recrystalli- zation twice from benzene-petroleum ether (35-600) gave 14. 1 g. (70%) of 4-bromo-4'-benzoylbiphenyl, m.p. 156. 5-157. 50. Anal. Calcd. for C19H13Br0: C, 67.67; H, 3.89; Br, 23.69. Found: C, 67.51; H, 3.89; Br, 23.95. The infrared spectrum is shown in Figure 38. . Wolff-Kishner Reduction of 4-Bromo-4'-benzoylbipheny1 A mixture of 9. 5 g. (0. 027 mole) of 4-bromo-4'-benzoylbiphenyl, Z. 3 g. (0. 057 mole) of sodium hydroxide and 4. 5 ml. of 99% hydrazine hydrate in 70ml. of triethylene glycol was refluxed for 1 hour. - The con- denser was removed and the water was driven off (flask temperature was 210-2200). After an additional hour of refluxing, the reaction mixture was cooled, poured onto 200 m1. of an ice-water mixture and extracted with ether. The ether solution was dried over magnesium sulfate, filtered and the ether distilled. . Since all attempts to recrystallize the residue were fruitless, the compound was used without further characterization. . Preparation of 4— Benzylbiphenyl This compound was prepared by two different methods. .Method A: A mixture of 6. 5 g. (0. 025 mole) of 4-benzoylbiphenyl, 1. 75 g. (0. 045 mole) of sodium hydroxide and 3 ml. of 99% hydrazine 132 .Emoxmfiaouaont.w..o§o.3...¢ Ho 8530on conduwcH .wm ousmfim A3 nuwcofiocfiw? f m E Z 2 a w s e m as . m _ _ _ d _ d a _ . _ . _ _ _ mmu (L . ( mmull TmSIo 133 hydrate in 50 ml. of triethylene glycol was treated in the same way as was 4-bromo-4'-benzoylbiphenyl (see above). After the ether was dis- tilled, 4.25 g.. (70%) of crude 4-benzy1biphenyl resulted, m.p. 79-810. One recrystallization. from isopropyl alcohol raised the melting point to 84--85° (literature value, 85°) (103). Method B: To a stirred solution of 0.65 g.(0.027 g. -atoms) of magnesium in 30 m1. of dry ether was added dropwise the residue from the Wolff-Kishner reduction of 4-bromo-4'-benzoylbiphenyl in 30 ml. of ether. After being refluxed for 0. 5 hour, water was added dropwise and the resulting solid was collected by filtration. . Recrystallization from isopropyl alcohol gave 3. 0 g. (45%) of lit-benzoylbiphenyl. The m.p. and m.m. p. of the products from method A and method B were identical. . Preparation of 4-(p-Bromophenyl)diphenylmethanol To a stirred solution of 9.4 g. (0.027 mole) of 4-bromo-4'-benzoyl- biphenyl in. 100 ml. of pyridine, 0. 5 g. (0. 013 mole) of lithium aluminum hydride was added all at once. After refluxing for 2 hours, water was added dropwise to decompose the excess lithium aluminum hydride. The reaction mixture was poured onto an ice-hydrochloric acid mixture and the resulting white solid was collected. :v Recrystallization from isopropyl alcohol gave 7.1 g. (77%) of 4-(p-bromophenyl)diphenylmethanol, m.p. 120-1210. _A_1_'1_a_1_. . Calcd. for C19H15Br0: C, 67.27; H, 4.46; Br, 23.56. Found: C, 67.40, 67.27; H, 4.66, 4.66; Br, 23.44. The infrared spectrum is shown in Figure 39 and the n.m. r. spectrum is shown in Figure 40. The n.m. r. spectrum is of interest since the apparent quartet at 4. 17? is due to spin-spin splitting between the hydro- gens on the tertiary carbon and. on the hydroxy group. The fact that they appear as a quartet and not two doublets is caused by the solvent, dimethyl sulfoxide (116). 134 ma NH .AGOSSHOm nHUEUv HocmnuoEfi>cosmfipAH>ao£QoEountmvtv mo 85.30on popmhfifi A3 AumGoHo>m3 : OH 0 m N. o m I .3 258a ‘51:} _ a _ _ _ _ _ 135 1.1 l I L l I y- . ’2-00 2432.50 2.67 3.00 . 3.50 f 4.00 4.17 4.50 T Figure 40 . . Nuclear ‘magnetic resonance. Spectrum of 4_ (P-bt‘omophenyl) _ diphenylm ethanol. 9 - , (CH3-S-CH3 solution) 136 Preparation of the Methyl ether of 4-(p-Bromophenyi)- diphenylmethanol To 4-(p-bromophenyl)diphenylmethanol (4. 0 g. , 12 mmoles) dissolved i n 200 m1. of carbon tetrachloride, 50 ml. of 99% sulfuric acid was added all at once. The violet mixture was poured onto 200ml. of ice-cold absolute methanol with stirring. The clear solution was poured onto 5.. ce, extracted with ether, dried over magnesium sulfate, filtered and the solvent distilled. The residue was recrystallized from isopropanol 't: 0 give 3. 1 g. (74%) of the methyl ether of 4-(p-bromophenyl)diphenyl- nethanol, m. p. 93-940. The n.m. r. spectrum is shown in Figure 41. fttempted-Preparation of a, 0., o'-Triphenyl-o..-I bydroxy-a' -methoxy-p, p' -bitolLl ' To a stirred solution of 0. 29 g. (12 mg. -atoms) of magnesium in 25 ml. of anhydrous ether, 3. 1 g. (8. 7 mmoles) of the methyl ether of 4-(bromophenyl)diphenylmethanol and 0. 2 g.. (1 mmole) of ethylene brom- ide in 30 ml. of dry benzene were added. After reaction. of themagnesium appeared to be complete, 1. 6 g.. (8. 7 mmoles) of benzophenone in 30 ml. of dry ether was added. The reaction mixture was refluxed for 3 hours, poured onto an ice-hydrochloric acid mixture, extracted with benzene, and dried over magnesium sulfate-sodium carbonate. . The benzene was distilled and the residue was recrystallized from benzene-ligroin (90-120°) to give 1.4 g. of solid material, m.p. 178-1790. Anal. . Calcd. for C33H3802: C, 86.81,. H, 6. 18. (Found: C. 84.86, 84.92; H, 6. 11, 6.16. The infrared spectrum is shown in Figure 42. - Preparation of Methyl 3élodobenzoate 3-Iodobenzoic acid (24.8 g. , 0.1 mole) in 125 ml. of anhydrous methanol saturated with dry hydrogen chloride was refluxed for 5 hours 137 I LIN"). W' 2.30 2-602.682.80 3.204.704.87 5.206.00 6.70 t Figure 41. . Nuclear magnetic resonance spectrum of the methyl ether of .4-(p-bromophenyl)diphenylmethanol.(CCL, solution). - 138 AGOSEOm «HOEUYococosmounon new? gmmocmdg was ~93:on ugconmngconmogounA:1v mo non—omen on» 50.3 @3303 @20m 0:» Ho 8330on ponmuwfi . .Nw madmfih A3 guwcofioawm? 2 2 S o m s o m . v m N _ _ _ _ _ _ a _ _ 139 with stirring. After the methanol was distilled, the resulting oil solidified and was dissolved in 100 ml. of ether. The ether solution was washed twice with SO-ml. portions of 5% sodium carbonate solution, once with 50- ml. of water and dried over calcium sulfate. Distillation Of the ether resulted in 25. 5 g.. (98%) of methyl 3-iodobenzoate. After one recrystallization from ether, the ester melted at 48—500- (literature value, 54°) (104) . Preparation of 3, 3'-Dicarbomethoxybipheny1 A. Activation of Copper Bronze (105) Copper bronze (70 g..) was treated with 1 1. of a 2% solutionof iodine i :n acetone for :10 minutes, filtered and treated for 15 minutes with 300 ml. of a hydrochloric acid - acetone mixture (151)). After being filtered, the copper bronze waswashed with acetone, dried and stored in a vacuum desiccator. B.. Preparation of 3, 3'-Dicarbomethoxybiphenyl This is a modification of Kornblum's procedure (106). . To.\a stirred solutionof 22. 1 g. (0.084 mole) of methyl 3-iodobenzoate in1001m1.. of dry dimethylformamide heated to reflux, 20 g. of copper bronze was added. After 5 hours, 25 g. ofcopper bronze was added and the refluxing was continued. for an additional 48 hours. . The reaction mixture (was filtered .hot, the filtrate poured onto 500- ml. of ice-water and the resulting precipitate filtered. . The solid was dissolved in benzene, dried over magnesium sulfate and the benzene distilled to give 8. 0 g.. (70%) of 3 , 3'-dicarbomethoxybiphenyl, m. p. 96-970 (literature value, 104°). Erepgration of 0., o.‘-Dichloro-a, a, a', a'-tetraphenyl-m, m' - , bitolyl ‘ A ‘ ° To a stirred solution of 0. 15 moles of phenylmagnesium bromide (from 23.. 5 g. of bromobenzene and 3. 6 g. of magnesium) in 50 m1. of dry 140 ether, 8. 0 g. (0. 029 mole) of 3, 3'-dicarbomethoxybipheny1 in 150 m1. of dry benzene was added dropwise. After being refluxed-for 24 hours, the reaction mixture was poured onto 700 m1.. of an ice-sulfuric acid mixture, the organic layer was separated and dried over magnesium sulfate. The benzene was distilled to a volume of 25 m1.. and. addition of petroleum ether (35-600) resulted in the formation of a- solid. On heating,the precipitate dissolved, but on cooling an oil resulted. 'I'he solvent was distilled, the residue was dissolved in 60 ml. of an acetic acid-acetyl chloride mixture (1:2) and anhydrous hydrogen chloride was Bubbled into the solution for 1 hour. The a, a'-dich1_oro—a, 0., 0', 0.‘-tetra- phenyl-m, m'-bitolyl (5.6 g. , 35%) was. collected. After one recrystal- lization from toluene-acetyl chloride it melted at 176-17701 (literature Value, 175-176°) (98). Sreparation of a, a, 0', o'-Tetraphenyl-m, m'-bitoly1-o.., 0.‘-diol 0., 0.‘ -Dichloro-o, a, a', a'-tetraphenyl-m, m'-bitolyl (1. 0 g. , 1. 8 moles) was dissolved in 97% sulfuric acid. The red solution was poured onto 100 ml. of ice-water with formation of a colorless precipi- tate. . The solid was filtered, washed with 5% sodium carbonate, with water and dried. . Recrystallizationi from toluene-ligroin(90-1200) gave 0. 65 g. (70 %) of a, a, a.‘ , a'-tetraphenyl-m, m'-bitolyl-a, a'-diol, m. p. 186-1870 (literature value, 183-1840) (98). Ereparation of 2,, 2'-Dicarbomethoxybiphenyl (107) Diphenic acid (15. 0 g. , 0.062 mole) in 100 m1.. of anhydrous methanol saturated with dry hydrogen chloride was refluxed. for 12 hours With stirring. After one-half of the methanol was removed by distilla- tion, as solid was deposited. . It was collected, washed twice with, 30-ml. Portions of 5% sodium carbonate, twice with 50-m1. portions of water 141 and recrystallized from methanol to give 11.2 g. (67%) of 2, 2'-dicarbo- methoxybiphenyl, m.p. 70-710.(literature value, 73-740). Preparation of 0., 0., 0', d'-Tetraphenyl-o, o' -bitolyl- '- a, a'-diol (89) 2, 2'-Dicarbomethoxybiphenyl (11. 2 g. , 0.041 mole) in 100 m1. of anhydrous ether was added dropwise with stirring to 0. 33 mole of phenyl-r lithium (from 51.8 g. of bromobenzene and 4.6 g. of lithium) in 150 ml. of anhydrous ether under a dry nitrogen atmosphere. After being stirred overnight, the reaction mixture was poured onto 700 m1. of ice-hydro- chloric acid, the ether was evaporated in a stream of dry air and the solid was filtered. . The yield of crude tetraphenyl-o, o'-bitolylyl glycol was 20.0 g. (94%). After one recrystallization from toluene, it melted at 247-250° (literature value, 252-253°). ) Preparation of the Anhydro Derivative of 0., 0., a', o'- Tetraphenyl-o, o'-bitoly1-o., o.‘-diol 0, a, a', o.'-Tetraphenyl-o, o'-bitolyl-0., o.‘-diol (1. 0 g. , 1. 8 mmoles) in 60 ml. of acetic acid-water (5:1) was refluxed for 2 hours. . The re- action mixture was poured onto 400 ml. of ice-water and the resulting solid was collected, washed twice with 20 ml. -portions of 5% sodium carbonate solution, once with 30 m1. of water and then dried. There re- sulted 0. 85 g. (88%) of the anhydro derivative of 0., 0, a', a'-tetraphenyl- o, o'-bitoly1-0., o.’-diol, m.p. 290-2920. (literature value, 290-2920) (98). Beaction of a, 0., 0', o.‘ -Tetraphenyl-o, o'-bitolyl-o., o.‘ - $01 in 98% Sulfuric Acid with Water. a, 0., o', o.‘-Tetraphenyl-o, o'-bitolyl-d, o.‘ -diol (0. 150 g. , 0. 29 mmole) was dissolved in 10ml. of 98% sulfuric acid. . The resulting yellow solu- tion was poured onto 150 ml. of ice-water. , The white solid was collected, 142 washed with 5% sodium carbonate solution, with water and dried to give 0. 125 g.. (86%) of the anhydro derivative of 0., a, a', 0.‘-tetraphenyl-o, o'- bitolyl-u, o.‘-diol. The m. p. and m.m. p. were identical to that of a known sample . C. The Diols and Carbinols in the Fluorene Series Preparation of 2-Benzoylfluorene (108) To a stirred solution of 33. 3 g. (0. 2 mole) of fluorene and 28. 2 g. (0. 2 mole) of benzoyl chloride in 250 ml. of carbon disulfide, 20.0, g. (0. 15 mole) of aluminum chloride was added. in small portions. After be- ing stirred for 4 hours at room temperature, the reaction was heated at reflux for 0. 5 hour to expel hydrogen chloride. . The reaction. mixture was poured onto ice-water, the carbon disulfide layer was separated and dried over calcium sulfate. . The carbon disulfide was distilled and a white solid resulted. Recrystallization from ethanol gave'34. 2 g.. (85%) of 2-benzoylfluorene, m.p. 123-124o (literature value, 124-125°). Preparation of o- Chloro- a- Phenyl- 2 -methylfluorene To a stirred solution of 2-benzoylfluorene (1.9 g. , 7 mmoles) in 25 m1. of pyridine, 0. 27 g.. (71mmoles) of lithium aluminum hydride was added all at once. After heating for 2 hours at 40°, the reaction mixture was poured onto 150 ml. of an ice-hydrochloric acid mixture, extracted with ether and dried over magnesium sulfate. . The ether was distilled and an attempt was made to recrystallize the resulting residue. Since all recrystallization efforts were fruitless, the oily residue was dissolved in 20 ml. of toluene-ligroin (90-1200) and anhydrous hydrogen chloride was bubbled throughfor 20 minutes. The brown solid was col- lected and recrystallized from toluene-petroleum ether.- (35-609) to give 0.67 g.. (35%) of a-chloro-a-phenyl-2-methylfluorene,. m.p. 115-~118o 143 0 (literature value, 122. 5-123.5 ) (109). Its infrared spectrum is shown in Figure 43. Preparation of a-Phenyl-a-hydroxy-Z -methylfluorene a-Chloro—o-phenyl-Z-methylfluorene (0.67 g. ,, 2. 5 moles) was dis- solved in 50 m1. of aqueous acetic acid (80%) and the mixture was heated to reflux for 1 hour. The acetic acid solution was poured onto an ice- water mixture, extracted with ether, dried over magnesium sulfate and .the ether distilled. The residue was recrystallized from aqueous ethanol to give 0. 50 g. (74%) of a-phenyl—a-hydroxy-Z-methylfluorene, m.p. 105-107° (literature value, 113-114°) (109). Preparation of a, u-Diphenyl-a-hydroxy-Z-methylfluprene To 0. 090 mole of phenylmagnesium bromide (from 14. 1 g. of bromo- benzene and 2. 2 g. of magnesium) in 50 m1. of anhydrous ether, was added dropwise 12. 3 g. (0. 0455 mole) of 1-benzoylfluorene in 245 m1. of benzene- ether (2. 5:1). After 4 hours of reflux, the reaction mixture was poured onto 500 ml. of an ice-hydrochloric acid mixture, extracted with benzene and dried over calcium sulfate. The benzene was distilled to a small volume and placed on a neutral alumina column. Elution was done with cyclohexane, benzene, ether and methanol. In all, 44 fractions were collected and fractions 19-35 (benzene and benzene-ether (1:1)) were used. The solvents were distilled and the combined residues were recrystallized from acetone-petroleum ether (35-600) to give 6. 5 g. (41%) of 0., a-diphenyl- o-hydroxy-2-methy1fluorene, m.p. 145-146O (literature value, 143-1440) (110). Its infrared spectrum is shown in Figure 44. . Reaction of a, a-Diphenyl-a-hydroxy-2-methy1fluorene in 97% Sulfuric Acid with Water a, o-Diphenyl-a-hydroxy-Z-methylfluorene (0. 067 g. , 0. 19 mmole) 144 MA .AcofldHOm NmUv oGoHOSCHsmau—ogtmtaconmtdtonozotd Ho 5530on @93qu NH 0 A3 Aumcofiourm? m N. P .2. Samara ‘ - _ .Acofidaoh nHUEUV oaonoaadgnuogtmt>xOup>£th>do£mfloto .8 mo 83:0on woumnmfi . .er oudwfih A3 numaoaocrm? NH 3 OH 0 w h o m as N 145 m a — — _ _ _ a _ _ _ 3 a 3 146 was dissolved in 25 m1. of 97% sulfuric acid. The purple solution rapidly . became orange-red. After stirring for several minutes, the solution was poured onto 300 m1. of an ice-water mixture. . No precipitate resulted but the solution was extracted twice with 25 ml. -portions of ether, dried over calcium chloride and the ether distilled. An infrared spectrumof the residue (see Figure 45) showed the presence of strong bands between 9 and 10 (.1, indicative of sulfonic acids. . Reaction of a, a-Diphenyl-a-hydroxy-2-methylfluorene '. in 85% Sulfuric Acid with water a, a-Diphenyl-o-hydroxy-2-methylfluorene (0. 07 g. , 0. 20 mmole) was dissolved in 25 ml. of 85% sulfuric acid. The purple solution was stirred for one minute, poured onto 300 ml. of an ice-water mixture and the resulting white precipitate collected by filtration. The solid was washed with 5% sodium carbonate, taken up in ether and dried over magnesium sulfate. Distillation of the ether gave 0.060 g.. (85%) recovery of a, a-diphenyl-a-hydroxy-2-methylfluorene. . It was identical in all respects with an authentic sample. . Preparation of 2, 7-Dibenzoylfluorene (111) To a stirred solution of 15.0 g. (0.056 mole) of 2-benzoylfluorene and 8. 0 g. (0. 056 mole) of benzoyl chloride in 100 ml. of carbon disulfide, 20. 0 g. (0. 15 mole) of aluminum chloride was added in small portions. After refluxing for 5 hours, 200 ml. of water was added and the resulting solid was filtered. . Recrystallization from isopropyl alcohol gave 14. 1 g. (67%) of 2, 7-dibenzoylf1uorene, m.p. 193-194o (literature value, 193- 194°). 147 .AGoSEOm «HUEUV .Bom 03535.6 amend Gm oaonozflfmnuoa 1Nt>xouv>£udtfi>co£mwvtd .8 mo oodponm mmmroupsfim 05 mo £3.30on pohmumfi A3 Aumcoaouwm? 2 2 o m a. w v .3. cheers #m __ _ n a 148 Preparation of a, a'-Dichloro—o, a'—diphenyl-2, 7- dimethylfluorene To a stirred solution of 1. 30 g. (3. 5 mmoles) of 2, 7-dibenzoyl- fluorene in 60 ml. of pyridine, 0. 14 g. (3. 5 mmoles) of lithium aluminum hydride was added all at once. After 1 hour, water was added to decom- pose the excess lithium aluminum hydride and the reaction mixture was poured onto an ice-hydrochloric acid mixture, extracted with ether and dried over magnesium sulfate. After the ether was distilled, the residue was dissolved in 50 ml. of acetic acid-acetyl chloride (2:1) and anhydrous hydrogen chloride was bubbled in for 1 hour. The brown solid was col- lected and recrystallized from ligroin (90-1200) to give 0.42 g.. (35%) of 0., o.‘-dichloro-o, a' «diphenyl-2, 7-dimethylf1uorene, m. p. 119-.121°. After a second recrystallization (from ligroin-acetyl chloride) it melted at121-122°. _I_\_n_a_l_. Calcd. for CNHzoClz: C, 78.07; H, 4.85; Cl, 17.07 Found: C, 77.88, 77.92; H, 5.03, 4.87; Cl, 16.84, 16.91. Preparation of a, o, a', a'-Tetrapheny1-2, 7-dimethylfluorene- 0., a'-diol 2, 7-Dibenzoylf1uorene (1.84 g. , 4. 9 mmoles) in 100 ml. of benzene- ether (3:1) was added dropwise to a stirred solution of 19. 6 mmoles phenylmagnesium bromide (prepared in the usual way from 0. 47 g. of magnesium and 3.08 g. of bromobenzene) in 30 m1. of anhydrous ether. The reaction mixture was refluxed for 2 hours, poured onto ice, extracted with benzene, dried over magnesium sulfate and the benzene distilled. . The residue was recrystallized three times from toluene-ligroin (60-900) to give 1. 6 g. of 0., a, a', o'-tetraphenyl-2, 7-dimethylfluorene-a, o.‘-diol, m.p. 218-2190.. .Ariil. Calcd. for C39H3003: C, 88.27; H, 5.70. Found: C, 87.04, 86.86; H, 6.03, 6.17. The infrared spectrum is shown in Figure 46. 40:73.8nonohosdgsuoefiptN. .thaosmduuoutb Jo .d to no 8530on UondprH ow oudmfim 1.: 5.951353 149 NH 2 2 a w e m ,m. v m _ _ _ a m _ _ a _ _ Pore v m a Peru (0, mu 150 D. The Diols and Carbinols in the Anthracene Series Preparation of the Methyl Ether of 9, 10, 10—Triphenyl- 9, 10-dihydro-9-anthrol - See page 110. Preparation of the Ethyl Ether of 9,10,10-Triphenyl- 9,10-dihydro—-9 anthrol See page 112. ‘ Preparation of 9, 10, 10-Triphenyl-9, 10- dihydro-9-anthrol The methyl ether of 9, 10, lO-triphenyl-9, 10-dihydro-9-anthrol (0.100 g. , 0. 23 mole) was dissolved in 10 ml. of 98% sulfuric acid. The red solution was poured onto 50 ml. of an ice-water mixture with stirring. The resulting white solid was filtered, washed with 5% sodium carbonate solution, with water and dried to give 0.80 g. (83%) of 9, 10, 10- triphenyl-9, 10-dihydro-9-anthrol. After one recrystallization from ligroin (90-120°) it melted at 204-204. 5° (literature value, 204°) (77). The infrared spectrum is shown in Figure 47. Preparation of 10, 10-Diphenyl-9-anthrone (77, 78) Dimethyl phthalate (14.4 g. , 0. 1 mole) in 40 ml. of anhydrous ether was added with mechanical stirring to 0. 4 moles of phenylmagnesium bromide (from 62.8 g. of bromobenzene and 9. 7 g. of magnesium) in 70 ml. of anhydrous ether. After the addition was complete, 100 ml. of dry benzene was added all at once and the reaction mixture was refluxed for :30 hours with stirring. It was then poured onto 300 ml. of ice-hydro- chloric acid, extracted with benzene and the organic layer separated. The benzene was distilled and the residue steam distilled to remove side 151 a; MA .Houauqmtotonpefifiptod .otfacoamfluuod dd .0 mo Eduuoomm pouwumaH . .34 ondmfim : 0H «0 43 ”imaged/m? w It mmole— m _ fl CID ——(>\o v. 152 products and unreacted starting materials. The residue was again taken up in benzene and the mixture was distilled. The portion boiling between 100- 2000 at 1 mm. was collected and recrystallized from acetone to give 11 g. (32%) of an impure yellow solid, m.p. 175-190O (literature value, 194- 1950). An infrared spectrum of the crude product showed strong carbonyl absorption at 6. 08 microns. The 10, 10-dipheny1-9-anthrone was used without fu rthe r purification . Preparation of 9-Chloro-10, lO-diphenyl-9, 10-dihydroanthracene (112) Crude 10, 10-diphenyl—9-anthrone (11 g. , 0.03 mole) was dissolved in 50 ml. of anhydrous benzene and added dropwise to 0. 95 g. (0. 025 mole) of lithium aluminum hydride in 25 ml. of anhydrous ether. The reaction mixture was refluxed for 3 hours with magnetic stirring. After decompo- sition with water and with dilute sulfuric acid, the benzene-ether layer was separated and dried over magnesium sulfate. The solvents were dis- tilled and the residue was dissolved in 15 m1. of benzene, 15 ml. of acetic acid and 25 ml. of acetyl chloride. Hydrogen chloride gas was passed through the solution for 3 hours and 5 g. (50%) of 9-chloro-10, 10-diphenyl- 9, 10-dihydroanthracene was collected, m.p. 2200. One recrystallization from benzene raised the melting point to 2240 (literature value, 2260). Its infrared spectrum is shown in Figure 48. . Preparation of trans-9, 10-Diphenyl-9, 10-dihydroxy- 9, lO-dihydroanthracene To 0. 25 mole of phenylmagnesium bromide (from 40.0 g. of bromo- benzene and 6.0 g. of magnesium) in 150 ml. of dry ether, a slurry of 15. 2 g.. (0.075 mole) of anthraquinone in 250 ml. of toluene was added with stirring. After addition was complete, the reaction mixture was stirred for one hour and refluxed for six. After the mixture was poured onto ice, the anthraquinone was collected, the organic layer separated, 153 a: .AcofiSHOm NmUV ocoomneficonPEfiptoH .mtaconmfiptod .oHtonoEoto mo 883on Bonsai: , .er madman» A3 gumcoHoNfim? ma NH 3 2 a m N. o m w. m . _ m _ _ _ _ 154 dried over magnesium sulfate and evaporated almost to dryness. The residue was extracted twice with boiling ethyl acetate and the filtrate evaporated to dryness. , Recrystallization three times from toluene- ligroin gave 7.6 g. (28%) of trans-9, 10-diphenyl-9, 10-dihydroxy-9, 10- dihydroanthracene, m.p. 260-2610 (literature value, 258-2600) (61). Its infrared spectrum is shown in Figure 49. Reaction of trans-9, 10-Diphenyl-9, 10-dihydroxy-9, 10- dihydroanthracene in 65% Sulfuric Acid withWater The red solution formed when trans-9, lO-dipheny1-9, lO-dihydroxy- 9, lO-dihydroanthracene (0. 280 g. , 0. 77 mmole) in 10 ml. of glacial acetic acid was reacted with 150 ml. of approximately 65% sulfuric acid, stirred for 10 minutes and then poured onto 200 ml. of an ice-water . mixture. The resulting solid was collected, dried and recrystallized twice from toluene-ligroin (60-900). The two types of crystals collected on the filter disk were separated by hand. The needle-like solid melted at 258-260° and was found to be identi- cal in all respects with the starting material. The cube-like solid melted at 188-1900 and had an infrared spectrum similar to that of trans-9, 10-dipheny1-9, 10-dihydroxy-9,‘10-dihydro- anthracene (see Figure 50). £231. Calcd. for CZ6H2002: C, 85.71; H, 5.57. Found: C, 89.24; H, 5.47. This analysis is similar to that calculated for 9, 10-diphenyl-9, 10-epoxy- 9, lO-dihydroanthracene (C, 90. 17; H, 5. 20). The melting point was identical to that reported for the epoxide (117) and for the cis-diol (118). . Since the solid shows the presence of intramolecular hydrogen bonding in its infrared spectrum, it is proposed that it is cis-9, 10- diphenyl-9, 10-dihydroxy-9, lO-dihydroanthracene. When one attempts to take its melting point, it dehydrates to the oxide. The same is true . o for its analysis since it was first heated at 100 for one hour. 155 .Godom umvfi ocoomyflucmoupifipnofi .olmxofiocfipuofi .otgcmfimflotod 665...: Mo 8350on poadufifi .ow oudmfim 1: sumcoaosfiw? «L ma NH 2 O“ o m N. Q m H. m _ 3 _ _ _1 _ _ _ _ _ a 156 .SoZoQ HMMV oaoomHQucmoupsfifiptoH .olmxonpsmfiptoa .mtfmcoamflotoa .otmfio mo 8:30on consume: 39 someones/mg 3. 2 NM 2 3 o m N. o m .8 Boss _ _ _ _ _ _ a _ _ 157 II. .Cryoscopic Measurements A . Apparatu s The apparatus was the same as that described by Fish (113). 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 U. S. Bureau of Standards platinum resistance thermometer (No. 1016073). A Leeds and Northrup 5-decade Wheatstone bridge andva Rubicon lamp- scale galvanometer with a sensitivity of 0. 0015 (ta/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. B. Procedure The apparatus was always cleaned in hot sulfuric acid, rinsed with distilled water and dried in an oven. The cell was assembled and approxi:- mately 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 158 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 concentrated 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 freez— ing point approximately 0. 30. The sample was added by means of a solid dropping funnel which had a long stem reaching close to the sulfuric acid surface. . The sample and funnel were weighed before and after the sample was added. The difference constituted the weight of sample. After the sample was dissolved, the freezing point was redetermined. . In order to eliminate such errors as freezing too rapidly or too slowly and improper stirring, the frozen mixture was allowed to melt and the freezing point determined again. This was repeated four to five times and the average of the closest three to four values were used. This was done until all of the samples were used. . C. 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. 111.. Preparation of Solutions for Spectral Measurements A. Aqueous Sulfuric Acid The stock solutions of sulfuric acid were obtained by the titration of "Baker Analized" Reagent grade with sodium hydroxide to a phenolphthalein end point. To obtain the aqueous solutions, calculated weights of distilled water were added to known weights of stock sulfuric acid. The water was added dropwise and the sulfuric acid was kept in. an ice-bath during addition. This was done to prevent loss of water by evaporation since the reaction between water and sulfuric acid is highly exothermic. B.. Spectral Solutions The solutions for the spectral studies were made up by dissolving a known amount of sample in 25 m1. of anhydrous glacial acetic acid (114) and taking 0. 1 m1.. (measured with a LaPine needle valve precision bore microburette which has a total capacity of 0. 50 ml. and which. can be estimated to 0. 001 ml.) of this solution and diluting it to 10 ml. with the appropriate sulfuric acid. The stock solutions of sample were made up so that the final solution had an absorbance of approximately 2. 0 in a 1 cm. light path. IV. Spectra The visible spectra were obtained mainly with the Cary 11 Recording (Spectrophotometer, with much of the early work being done on the Beckrnan DK-2 Recording Spectrophotometer. All of the spectra were run using 1 cm. glass-stoppered quartz cells. The proton magnetic resonance spectra were obtained with a High- Resolution Analytical N. M. R. Spectrometer System, Varian Associates A-60. The infrared spectra were obtained with a Perkin-Elmer (Model 21) Recording Infrared Spectrophotometer with 0. 5 mm. solution cells and 1.0 mm. cavity cells. SUMMARY 160 161 1. By use of the Friedel-Crafts benzoylation and Grignard reactions on the appropriate aromatic molecules the following types of compounds were prepared Y I c-x- I I C3 975 where R = phenyl or hydrogen, Y = hydroxyl or chlorine and X = benzene, R .. biphenyl or fluorene . 2. All of the compounds studied dissolved in 98% sulfuric acid to form highly colored solutions ranging from yellow to blue. Investigation of the visible absorption spectra of solutions of these compounds in vary- ing concentrations of sulfuric acid has shown that reversible ionization to dications can occur either stepwise or simultaneously and is, in general, a function of the moiety between the site s to be ionized. . The pKR' s for these processes have also been determined and are given in Table 17 along with the xmax and emax of the dications. In order to interpret the visible spectra of these dications, a series of compounds were prepared which ionized to give monocations: Y | c l . C5 The spectral data for these compounds are given in Table 18. . Comparison X- -R of the spectra of the mono- and dications has shown that the resonance interactions of the latter can be pictured as being composites of mono- cations. 3‘. In addition to spectral studies, the hydrolysis products of the dication solutions in concentrated sulfuric acid were also investigated. 162 In all cases studied the starting diols (or cyclic ethers in the case of ortho substituted diols) were isolated in nearly quantitative yields. . The hydrolysis of two carbinols (4-biphenylyldiphenylmethanol and a, a-diphenyl-a-hydroxy- 2-methylfluorene) in 98% sulfuric acid showed that sulfonation occurred after carbonium ion formation. The alcoholysis of solutions of 0., o, a', (1'- tetraphenyl—o-xylene-a, a'-diol in concentrated sulfuric acid gave the corresponding ether of 9, 10, 10-triphenyl-9, 10-dihydro—9-anthrol. 4. The study of trans-9, 10-diphenyl-9, 10-dihydroxy-9, 10-dihydro- anthracene in sulfuric acid was continued. It was demonstrated that in dilute sulfuric acid ionization of this diol gave the monocation which re- arranged to the cis-diol. By increasing the acid strength the cis-diol rearranged to 9, 10-diphenylanthracene, which in more concentrated sul- furic acid gave 4-phenyl-2, 3-benzofluoranthene. .At no time, however, was the presence of a dication noted. 163 Table 17. Spectral Data for the Bis-Aryl Carbinols in Sulfuric Acid ==== w a a ‘ Compound xmax emax pKR (mu) 1 2 0., 0., a', o'-Tetraphenyl-p- 455 59, 000 -8.1 - 10. 5 xylene-0., o'-diol 0., d, o', o'-Tetraphenyl-m- 440 70, 000 —8. 0 -10.1 xylene-0., o'-diol 419 73, 000 a, 0., o', a'-Tetraphenyl-o- 455 44,000 N-8.0b -16.6 xylene-0., o,‘-diol 373 30, 500 0., c'- Diphenyl-p-xylene- 461 48, 000 c -a, o.‘-diol 0., o'-Dipheny1—m-xylene- 447 35, 000 c 0., o'-diol 0., 0., d', o'-Tetrapheny1 530 90, 000 -8. 2 d p, p'-bitolyl—a, o.‘ -diol 442 53, 000 4108 34,500 0., a, o', o.‘-Tetraphenyl- 430 67, 500 -8. 4 d m, m'-bitolyl-o., o-diol o, o, o', o'-Tetraphenyl- 428 35,500 N -8. 36m -15.0 o, o'-bitolyl-a, o' -diol 390‘s 24, 000 a, o.‘-Dipheny1-p, p' - 563 134, 000 c bitolyl-o, o.‘-diol 420 13, 000 a,d'-Dichloro-o.,o'-diphenyl- 603 215,000 N -12.0 -16.6 2, 7-dimethylfluorene 554 62, 000 o, 0., o', c'-Tetraphenyl-2, 7- 568 152, 000 c dimethylfluorene-a, a'-diol 525 46, 000 440 35, 000 aThe x and 6 data were obtained in 98% sulfuric acid. max max The first pK was assumed to be the same as for the corresponding meta and para dio s. The close proximity of the second -OH after mono-ionization permits cyclic protonated ether formation. This species, which was not detectable by visible spectrophotometry, required solutions of high acidity for further ionization. c . Not determined. In these diols, ionization occurred simultaneously at both sites. This compound ionizes Similarly to the diol discussed in b. The first ioni- zation was assumed to be the same as the corresponding meta and para de- ri vati v». a - 164 Table 18. Spectral Data for the Aryl Carbinols in Sulfuric Acid m a a Compound xmax emax pKR (mp) Triphenylmethanol 432 37, 500 - 7 . 4 408 36, 500 Diphenylmethanol 442 47, 000 -14. 7 4-Biphenylyldiphenylmethanolb 510 43, 000 -7 . 7 420 21,500 4-Biphenylylphenylmethanol 535 90, 000 c o-Phenyl-a-hydroxy-2-methyl- 547 66, 000 N -12. 0 fluorene a, a-Diphenyl-a-hydroxy- 540 53, 000 -6. 5 Z-methylfluoreneb 404 17, 000 aThe ). and e .. data were obtained in 98% sulfuric acid, except ax where no ed. These spectra were taken in 80% sulfuric acid because sulfonation was shown to occur in concentrated acid. CNot determined. LITERATURE CITED 1.. J. E- Leffler,. "The Reactive Intermediates in Organic Chemistry, " 10. 11. 12. 13. 14. 15. 16. Interscience Publishers, Inc. ,1 New York, 1960,. Chaps... V-VII. .. E. S. Gould, "Mechanism and Structure in Organic Chemistry, " Henry Holt and Company, New York, 1959, p. 303ff., p. 514ff. . H. Hart and R. w. Fish, J. Am. Chem.Soc.,_8_2, 5419(1960), see footnote 21. .. P. Walden, Ber., _3_5_, 2018 (1902). . M. Gomberg, Ber.,_3_5, 2397 (1902). . K.. Ziegler and H- Wollschitt, .Ann. ,. 479, 90 (1930). .. N. N- Lichtin and H. Glazer, J. Am.,Chem. Soc., 13, .5537 (1951). . L..C..Bateman, E. D-Hughes and C. K. Ingold, J. Chem.- Soc. , 1012 (1940). . W. T. .Nauta and P. .J. .Wuis, Rec. trav. .Chem. , 56, .535 (1937). .R. J. Gillespie and J. A. Leisten, Quart-Revs., _8_, 40 (1954). A. Hontzsch, Z. physikal Chem. 6_1_, 257 (1908). M. S..Newman and N- C_.- Deno, J..Am.. Chem. Soc. , 13,, 3644 (1951). N. C. Deno, J. J. Jaruzelski andA. Schriesheim, J. Org. Chem. 19,155(1954) .N. C. Deno, J. J. Jaruzelski andA- Schriesheim, J. Am. Chem. Soc. , 77, 3044 (1955). N- C..Deno,. P. T. Groves and G- Saines, J. Am..Chem- Soc. , 8_1, 5790 (1959). . S. F. Mason and R. Grinter in ”Steric Effects inConjugated Systems, " G. W. Gray, Editor, Academic Press Inc. ,. New York, 1958, p. 52. 165 17 18. 19. 20. 21 22. 23. 24 25. 26. 27. 28. 29. 30. 31 32. 33. 34 35. 36. 166 S. F..Mason, Quart. Revs., l_5_, 287 (1961). _K.. A. . Hofmann and H. Reimsreuther, Ber. , fl, 4856 (1909). M. Gomberg, Ber., _4_0, 1847(1907). M. Gomberg and L. H.. Cone, Ann. , 370, 142 (1910). .. F. Seel, Z. anorg- Chem. ,, 250,. 331 (4943). C. R- Witschonke and C. A- Kraus, J. Am.. Chem.. Soc. , fl, 2472‘ (1947). D- W. A. Sharp and N..Sheppard, J. Chem. Soc., 674 (1957). .. G. Karagounis,. Helv. Chem Acta. , £11 994 (1951). H- Meerwein, V- Hederich and K. Wunderlich, Arch- Pharm. , 291, 541 (1958); C..A. 54, 5427b (1960). H- J. Dauben, Jr. , L.. R- Honnen andtK-M. Harmon, J. Org.. Chem. , 55, 1442 (1960). ’K. M. Harmon and A. B. Harmon, J..Am-Chem..Soc., 83, 865 (1961). . K. M. Harmon and F- E. , Cummings, J. Am.. Chem- Soc. , 84, 1751. (1962). C. M..Welch and H..A..Smith,. J. Am. Chem-Soc., 1_2_, 4748 (1950). V. Gold, B..W. V..Hawes, and-F. L. Tye, J..Chem...Soc., 2167-(1952). - L.-.M.. Jackman, "Nuclear‘Magnetic Resonance Spectroscopy, "‘ Pergamon Press,. New York, 1959. R.. B.. Moodie, (T. . M- Connor, and R..Stewart, Can- J. Chem. ,._3_Z, 1402 (1959). R. Dehl,.W..R. Vaughn, and R..S. Berry, J. Org.. Chem., 24, 1616 (1959). .J. A. Pople, J-Phys. Chem., _6_1, 6 (1957). A..Werner,. Ber. ,._3:4_, 3300 (1901). A. Werner, Ann.,, 322, 296 (1902). 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. .48. 49. 50. 51. 52. 53. 54. 55. 56. 167 F. Arndt, P. .Nachtwey, and E. Scholz, Ber., ‘51, 1903 (1924). F- Arndt and C.. Lorenz, Ber. , 63, 3121 (1930). A. Schenberg and S.. Nickel,. Ber. , fl, 1795 (1934). A..Scho‘nberg and w. Asker, J- Chem- Soc., 272 (1942). R. Wizinger and Y. Al-Attar, Helv.. Chem Acta. , _3_9_, 189(1947). R. .Acheson, "The Chemistry of Heterocyclic Compounds, " Interscience Publishers, Inc. New York, 1956, p. 280. ' G- N.. Lewis and M. Calvin, Chem. Revs. ,.2_5_, 273 (1939). A. A. Morton and A. H. Wood, J. Am. Chem.. Soc..,,_6_1_, 2902 (1939). .W. Theilacker and W. Schmid, Ber., '8_4_, 204 (1951). G. N- Lewis and I- Bigeleisen, J. Am- Chem..Soc. , 62, 2102(1943). G- N.. Lewis and 1- Bigeleisen, J. Am.. Chem.. Soc. , 62,. 2107' (1943). G. N. Lewis, J. Am. Chem..Soc., Q, 770 (1945). A. M. Anthony-Barbier, P. Rumpf and C. Viel, Bull..Soc-Chem. France, 1474 (1959). C. . C. Barker in. "Steric Effects in Conjugated Systems, " G.. W. Gray, Editor, Academic Press Inc. ,. New York, 1958, p. 34. R.. Wizinger and J- Fontaine, Ber. , 60B, 1377(1927). R.. E. Buckles and N. A. Meinhart, J. Am. Chem..Soc., 14, 1171 (1952). R. E. Buckles and W..D..Womer, J. Am. Chem-Soc., Q, 5055 (1958). R..E..Buckles, R..E..Erickson, J. D..Snyder, and w. B..Person, J. Am..Chem-Soc., _8_2_, 2444 (1960). R.. E. Buckles, R.. E. Erickson, T. W. McGee andW. B. Person, Abstracts of Papers 138t__h A. C. S. Meeting, New York, N. Y. ,1960, pSl-p D- H. Anderson, R..M..Elofson, H..S. Gutowsky, S- Levine, and R- B.. Sandin, J. Am.. Chem.. Soc. ,.8_3_, 3157 (1961). 168 57. H. Hart and R..W. Fish, J. Am. Chem. Soc., §_3_, 4460 (1961). 58.. (a) H. Hart and R..W. Fish, J. Am.. Chem. Soc. , _8_2_, 5419 (1960), footnote 21; (b) ibid. , footnote 24. 59. J. Fleming, unpublished results. 60. J. Thiele and H..Ba1horn, Ber., _31, 1468 (1904). 61. R. R. Rafos,.Ph. D. Thesis, Michigan State University, 1961. 62. H. Hart and C. Y- Wu, unpublished results. 63. G. J. Sloan,. Ph. D. Thesis, University of Michigan, 1953. 64. H. H. Freedman and A. M. Frantz, Jr., Abstracts of Papers 141—5.12. A.C.S. .Meeting, Washington, D..C., 1962, p. 26-0. 65. H. Hart and J.» Fleming, unpublished results. 66. M. A..Paul and F. A. Long, Chem. Revs., 51, 1 (1957). 67. R. D. Bushick, Ph. D. Thesis, University of Pittsburgh, 1961. 68. L. P. Hammett and A. J. Deyrup, J. Am..Chem-Soc., _5_4, 2721 (1932). 69. G..Schwarzenbach and R..Sulzberger, Helv.. Chem. Acta. , _21, 348 (1944). 70. J. C. D. Brand, W. C. Horning and M. B. Thornley, J..Chem..Soc., 1374 (1952). 71. V. Gold, Chem..andInd., 172 (1955). * 72. V. Gold, J. Chem..Soc., 1263 (1955). 73. V. Gold and B-M. V. Hawes, J. Chem..Soc., 2102 (1951). 74. G..Wi11iams and M. A. Bevan, Chem. and Ind., 171 (1955). 75. F. Ullmann and A. Miinzhuber, Ber., 36, 406(1903). 76..S..F. Acree, Ber., _31, 2755 (1904). 77. 78. 79. 80. 81. 82. . 83.. 84. 85. 86. 87. 88. .89 90. .91 92. 93.. 94.. 95.. 96. '97. £51“? F” .L. A. 169 Barnett, J- W- Cook- and I- G. . Nixon, J. Chem. . Soc. , 504- (1927). ..Shibata,. J. Chem.. Soc., 1449(1909). M. 1 Bennett and R. L..Wain,. J. Chem- Soc. , 1114‘ (1936). Burtonand R..W. H. Cheeseman, J. Chem.. Soc. , 986 (1953). C. Anderson and W. A. Fisher, J. Am- Chem..Soc., 66, 595 (1944). - Gold and-E- L. Tye,. J. Chem.. Soc. ,. 2172 (1952). . Theilacker and W- Ozegowski, Ber. , 13, 898(1940). G. . Clarkson and M- Gomberg, J. Am.. Chem.. Soc. ,._5_2, 2881 (1930). M..Stock and H- C. Brown, J. Am. Chem.. Soc. ,._8_4_, 1242 (1962). . J. D. Roberts, "Notes on MolecularOrbital Calculations, " W. A. . Benjamin,. Inc. ,. New York, 1961. Streitwieser, Jr. ,. "Molecular Orbital Theory for‘Organic Chemists, " John Wiley andeon, Inc. ,. NewYork, 1961. H. .. G. .L- .C- Longuet-Higgins, J. Chem.. Phys. , _1_8_, 265 (1950). .Wittig and M- Leo, Ber. , 64B,. 2395 (1931). J- Bellamy, "The Infrared» Spectra. of Complex Molecules, " John Wiley and Sons, Inc. ,. New York, 1958, p. 364. woes? R. .Kehrmann, R- Monnier and M. Ramm, Ber. , 56B, 169 (1923). Haller-and A- Guyat, Bull.. Soc. Chem. France, [3] 31, 981 (1904). Ador, Ber., 12,. 320 (1880). . Schlenk and M..Brauns,. Ber. , :48, 667 (1915). .. Stark. and O. . Garben, Ber. , :16, 2257' (1913). .. Deluchot, Compt- Rend., 190, 438(1930). E. Buckles and’N. G. Wheeler, Org. Syntheses, 3_l_, 29' (1951). 98. 99.. 100. 101.. . 102.. 103. 104.. 105 106. 107. 108. 109. 110 111 112. 113 114. 115. 116, 117. 118. 170 W. Schlenk and M.. Brauns, Ber. , 48, 723 (1915). € . Schlenk and E. . Bergmann, Ann. , .463, 120 (1928). P’ E- Tschitschibabin, Ber. , :12, 1810 (1907). . Schlenk, Ann. , (368,. 298(1909). TU € J. Montague, Rec. Trav- Chem., _21, 358(1908). .0 Goldschmeidt,. MonatshChem. , _2_, 433 (1881). '.> .VM..KellaS,. Z. phys..Chem‘. , 24, 244 (1897). .. N..Kornb1urn and D. L.. Kendall, J. Am- Chem.. Soc. , 3_4_, 5782 (1952). . F. Ullman and O.. Lowenthal, Ann. ,1 332, 72 (1904). H. W. Underwood, Jr. , and E. L. .Kochmann, J. Am. Chem.. Soc., 16, 2069 (1924). . W- E. . Bachmann and Sister M. Xaveria Barton, J. Org. Chem. ,. 3, 300 (1938). .W.. E. . Bachmann and G..Osburn,. J- Org.. Chem. , 5, 29 (1940). .. W- E. Bachmann, R. Hoffman and F- Whitehead, J- Org. Chem. , _8_, 320 (1943). .. K..Dziewonski and A. Obtulowicz, Bull. intern. acac. polonaise, 399(1930A); C..A. 22, 5157 (1931). -E. -D. . Bergmann and J- Hervery, Ber. ,. _6_2_,. 916 (1929). .. R- W. . Fish, Ph.. D- Thesis, Michigan State University, 1960. . L.. F. . Fieser,. "Experiments inOrganic‘Chemistry, "‘D. . C- Heath and Company, Boston, 1955, p. 281. Dr- L.. P. Kuhn, private communication. H. . Hart and T.. Sulzberg, unpublished results. G. Wittig, E. .Knauss, and K- Niteft‘h'amme‘r, Ann. , 63_0_, 10 (1960). Y. Lepage, Ann. chim- (Paris), [13], 4, 1137(1959). cliSTRY UBRAR“ (.MIFHIG?N ST‘AT‘E l) NNNNN TITY L'l‘ RRRRRR S WI)IIWIWI"(HM)MIIKIJHhUIWI 3 1293 02340 0033