Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The oxidation of secondary and tertiary aromatic alcohols by chromium (VI) and manganese (VII) Banoo, Fariza 1968

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1969_A1 B37.pdf [ 6.83MB ]
Metadata
JSON: 831-1.0059926.json
JSON-LD: 831-1.0059926-ld.json
RDF/XML (Pretty): 831-1.0059926-rdf.xml
RDF/JSON: 831-1.0059926-rdf.json
Turtle: 831-1.0059926-turtle.txt
N-Triples: 831-1.0059926-rdf-ntriples.txt
Original Record: 831-1.0059926-source.json
Full Text
831-1.0059926-fulltext.txt
Citation
831-1.0059926.ris

Full Text

THE OXIDATION OF SECONDARY AND TERTIARY AROMKTIC ALCOHOLS BY CHROJlIUM(vT) AND. MRNGAMESE (VII) by FARIZA BANCO M.Sc, Rajshahi University, 1962 A THESIS SUBMITTED IN PARTIAL FULFILtffiKT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of Chemistry We accept t h i s thesis as conforming.to the required standard THE UNIVERSITY OF BRITISH COLUMBIA Septenfoer, 1968 In p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the r e q u i r e m e n t s f o r an advanced degree a t the U n i v e r s i t y of B r i t i s h Columbia, I a g r e e t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and Study. I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department or by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Department Date ABSTRACT The mechanism of permanganate oxidation of benzhydrol • has been investigated between pH 7.00 and H Q -2.46. The deuterivtm isotope effect, k^/kp, obtained by studying the oxi-dation rate of benzhydrol a-d, was found to f a l l from 7.3 at pH 7.00 to 1.08 at H -1.22. Between pH 7.00 and H Q 0.20, the reaction i s of second order, f i r s t order i n each of the reactants. Beyond Ho-0.50 the reaction becomes f i r s t order i n carbinol and zero order i n permanganate, and i n this region the reaction i s strongly acid-catalyzed. .A study of eight substituted benzhydrols shows an excellent Hammett correlation with a +,p + being -1.02. The rate-determining step at higher acidities i s believed to be the scission of protonated carbinol to carbonium ion, which then reacts with permanganate i n a fast step. This idea i s supported by the results of a study of the rate of ionization of p-inethoxyben-zhydrol under conditions similar to those of the oxidation reaction. The mechanism of chromic acid oxidation of benzhydrol has been studied between H Q 0.50 and -4.20. The reaction i s acid-catalyzed i n this region and was found to be of second order, f i r s t order i n each of the reactants. The deuterium isotope effect with benzhydrol or-d at H Q -1.00 i s 6.81. i i A Hammett plot, obtained from the study of five sub-stituted benzhydrols, shows d to be the operative substituent constant, p being -0.54. The rate-determining step i s believed to be the unimolecular decomposition of a chromate ester. i The permanganate oxidation of triphenylcarbinol has been investigated i n the region of acidity of H Q -0.60 to -1.93. The reaction was found to be acid-catalyzed i n this region and to produce benzophenone and phenol. The order of reaction was found to be the sane as i n the case of benzhydrol, i.e., f i r s t order i n carbinol and zero order i n permanganate. A study of nine substituted triphenylcarbinols gave a good Hairmett correlation with a +,p + being -1.39. As i n the case of benzhydrol, the rate-determining step i s believed to be the ionization of the carbinols. The chromic acid oxidation of triphenylcarbinol has been investigated between H R -2.80 and -7.48. Acid-catalysis was observed for the reaction i n this.region. The reaction was found to be of second order, f i r s t order i n each of the reactants, as i n the case of benzhydrol. The reaction gave a quantitative yield of benzophenone, phenol also being formed. A Hararctt correlation with o+ was obtained from a study' of eleven substituted triphenylcarbinols, p + was found* to be -0.879. The rates of oxidation of the triarylcarbinols were also studied i n the presence of added manganous ions, which showed an i i i almost indform reduction i n rate for a l l the carbinols, the value of p + i n this case being -0.906. The migration aptitudes of the substituted aryl groups were determined and p + for migra-tion was found to be -1.44. A chromate ester mechanism, similar to that proposed for benzhydrol and other secondary alcohols, i s believed to be operative i n this case, except that the decom-position of the ester takes place by migration of the electron-rich ring. I t i s suggested that the chromic acid oxidation of primary and secondary alcohols may take place by an analogous rearrangement reaction of a chromate ester. In these cases the migrating group would be hydrogen. Permanganate oxidation of benzhydrol was also studied i n frozen system between pH 1.50 and 12.40. A large increase i n rate, compared to that i n the l i q u i d system, was observed. The deuterium isotope effect, k H/k D, of 7.2 to 7.5 was found i n this system. The reactions are of second order, as i n the l i q u i d system. The acceleration i n rate appears to be due to an increase i n con-centration of the reactants i n the l i q u i d phase between ice crystals. i v TABLE OF CONTENTS Page INTRDDUCTICN .. 1 SCOPE OF THE PRESENT RESEARCH 30 SECTION I : OXIDATION OF DIARSOJCARBINaLS EXPERIMENTAL 32 Materials .. . . . . .. .. .. .. 32 Kinetic Methods .. .. .. .. .. .. .. 33 Product Analysis .. .. .. .. .. .. .. 37 Determination of Acidity Functions .. .. .. .. 38 RESULTS .. 41 Product Analysis .. .. .. .. .. .. .. 41 DeterMnation of Acidity Functions .. .. .. .. 42 Order of Reaction .. .. .. .. .. .. .. 49 Dependence of Rate on Acidity .. .. 54 Substituent Effect .. .. .. 64 Isotope Effect .. .. .. .. 73 Activation Parameters .. .. .. .. .. .. 74 Ionization of 4-^thoxybenzhydrol .. .. .. .*. 77 DISCUSSION .. .. .. .. .. 85 V SECTION I I : OXIDATION OF T."nARYLCAJ<BrNOLS EXPEPJI42NTAL .. .. .. . . . . .. .. .. 96 Materials .. .. .. .. .. .. .. .. 96 Kinetic Methods 98 Product Analysis .. .. .. 100 RESULTS ' .. .. .. ..104 Product Analysis .. .. .. .. .. .. .. 104 Order of Reaction .. .. .. .. .. .. .. 112 Dependence of Rate on Acidity .. .. .. .. .. 117 Substituent Effect .. .. .. .. .. .. •• 128 Activation Parameters .. .. .. .. .. .. 135 DISCUSSION .. .. .140 A COMPARISON BETWEEN THE REACTIONS OF PERMANGANATE AND CHROMIC ACID .. .. .. .. 156 APPENDIX .. .. .. .. .. .. .. .. ..159 SUGGESTION FOR FURTHER WORK 171a BIBLIOGRAPHY .. .. .. .. 172 v i LIST OF TABLES Page I Values of K Q i n H2SO4 Diluted with 80 wt% Aqueous Acetic Acid .. .. .. .. 42 I I Values of HR i n H2SO4 Diluted with 80 wt% Aqueous Acetic Acid .. .. .. ...... .. 44 I I I Values of HQ i n H2SO4 Diluted with 70 v o l % Aqueous Methanol .. .. .. .. .. .. 47 IV Values of HR i n H2SO4 Diluted with 70 v o l % Aqueous Methanol .. .. .. .. .. .. 49 V Oxidation of Benzhydrol with Permanganate i n Aqueous S u l f u r i c Acid : Var i a t i o n of Rate-constant with A c i d i t y .. .. .. .. 58 VI Oxidation of Benzhydrol with Chromic Acid i n ,80 wt% Aqueous Acetic Acid : Variation of Rate-constant with A c i d i t y .. ... .. .. 61 VII Oxidation of Substituted Benzhydrols with Permanganate i n Aqueous S u l f u r i c Acid : Harnett P l o t . . . . .. 64 VIII Oxidation of Substituted Benzhydrols with . Chromic Acid i n 80 wt% Aqueous A.cetic Acid : Harmett P l o t . . . . .. .. .. 70 IX Oxidation of Benzhydrol with Permanganate i n Aqueous S u l f u r i c Acid : Deuterium Isotope E f f e c t .. 73 X Oxidation of Benzhydrol with Permanganate i n Aqueous S u l f u r i c Acid : Ac t i v a t i o n Parameters .. 74 X(a) Permanganate Oxidation of Benzhydrol : Vari a t i o n of Rate-constant with Tenperature .. .. 76 X(b) Chromic Acid Oxidation of Benzhydrol : Vari a t i o n of Rate-constant with Temperature .. .. 77 VX1 XT Ionization of 4-Methoxybenzhydrol : Variation of Rate-constant with Temperature .. .. .. 79 XII Analysis of Chromic Acid Oxidation Products of Triarylcarbinols .. .. .. .. .. .. 104 XIII Migration Aptitudes of Aryl Groups for Chromic Acid Oxidation of Triarylcarbinols .. .. 108 XIV Analysis of Permanganate Oxidation Product of Triarylcarbinols .. .. .. .. .. .. 112 XV Chromic Acid Oxidation of Triphenylcarbinol : Variation of Rate-constant with Acidity .. .. 117 XVT Permanganate Oxidation of Triphenylcarbinol : Variation of Rate-constant with Acidity .. .. 122 XVII Permanganate Oxidation of Tri(p-tolyl)carbinol : Variation of Rate-constant with Acidity .. .. 125 XVTII Chromic Acid Oxidation of Triphenylcarbinol : Effect of Substituent .. 129 * XIX Permanganate Oxidation of Triphenylcarbinol : Effect of Substituent .. .. .. .. .. 132 XX Activation Parameters for Oxidation of Triphenylcarbinol .. . . . . . .. .. .. 138 XX(a) Chromic Acid Oxidation of Triphenylcarbinol : Variation of Rate-constant with Temperature .. .. 138 XX (b) Permanganate Oxidation of Triphenylcarbinol : . Variation of Rate-constant with Temperature .. .. 139 v i i i LIST OF FIGURES Page 1. Acidity Function H i n H^ SO^  Diluted with 80 wt% Aqueous AceSic Acid .. .. .. .. .. 45 2. Acidity Function Hu i n H2SO4 Diluted with 80 wt% Aqueous Acetic Acid .. 48 3. Permanganate Oxidation of Benzhydrol : Typical Rate Plot, H = 0.25 " 51 4. Permanganate Oxidation of Benzhydrol : Typical Rate Plot, H = -0.69 52 5. Permanganate Oxidation of Benzhydrol : Typical Rate Plot, H 0 = -0.69 .. .. .. .. 53 6a. Permanganate Oxidation of Benzhydrol : F i r s t Order Rate Plot, KQ = -0.01 .. .. .. 55 6b. Permanganate Oxidation of Benzhydrol : Second Order Pate Plot, H Q = -0.01 .. ... .. 56 7. Chromic Acid Oxidation of Benzhydrol : Typical Eats Plot .. .. .. .. .. . . 5 7 8. Peripanganate Oxidation of Benzhydrol : Variation of Pate-constant with Acidity Function H Q .. .. .. .. .. .. 59 9. Penranganate Oxidation of Benzliydrol : Variation of Rate-constant with Acidity Function H R .. .. .. .. .. .. 60 10. Chromic Acid Oxidation of Benzhydrol : Variation of Rate-constant with A d d i t y Function K Q .. .. .. .. .. .. 62 11. Chromic Acid Oxidation of Benzhydrol : Variation of Rate-constant with Acidity Function H R .. .. .. .. .. .. 63 ix 12. Pe.inanqanate Oxidation of Diarylcarbinols : ' HaniTBtt Plot, pH = 7.00, 1.99 .. .. .. .. 67 13. Permanganate Oxidation of Diarylcarbinols : Hammetc Plot, H Q = -0.81 68 14. Permanganate Oxidation of Diarylcarbinols : Hamrrett Plot, HQ = -0.81 69 15. Chromic Acid Oxidation of Diarylcarbinols : Hammett Plot .. .. .. .. 71 16. Chromic Acid Oxidation of Diarylcarbinols : Hanmett Plot .. .. 72 17. Permanganate Oxidation of Benzhydrol : Variation of Rate-constant with Temperature .. .. 75 18. Chromic Acid Oxidation of Benzhydrol : Variation of Rate-constant with Tenperature .. .. .78 19. Ionization of p-?fetlioxybenzhydrol .. .. .. 80 20. Ionization of p-Methoxybenzhydrol : Variation of Rate-constant with Acidity Function H Q .. .. .. .. .. .. 81 21. Ionization of p-Methoxybenzhydrol : Variation of Rate-constant with Acidity Function H R .. .. .. .. .. .. 82 22. VPC Analysis of the Chromic Acid Oxidation Product of 4-Chlorotriphenylcarbinol .. .. .. 105 23. Chromic Acid Oxidation of Triarylcarbinols : Hairmett Equation Applied to the Migration of Aryl Groups .. .. .. .. .. 110 24. Chromic Acid Oxidation of Triphenylcarbinol : Typical Rate Plot 114 25. Permanganate Oxidation of Triphenylcarbinol : Typical Rate Plot .. .. .. .. .. . . 1 1 6 X 26. Permanganate O2ddati.cn of T r i (p-tolyl) carbinol : Typical Rate Plot .. 118 27. Chromic Acid Oxidation of Triphenylcarbinol : Variation of Rate-constant with Acidity Function H Q .. .; .. .. .. .. 119 28. Chromic Acid Oxidation of Triphenylcarbinol : Variation of Rate-constant with Acidity Function H R .. .. .. . . . . .. 120 29. Permanganate Oxidation of Triphenylcarbinol : Variation of Rate-constant with Acidity Function H Q .. .. .. .. .. .. 123 30. Permanganate Oxidation of Triphenylcarbinol : Variation of Rate-constant with Acidity Function H R .. .. 124 31. Permanganate Oxidation of Tri(p-tolyl)carbinol : Variation of Rate-constant with Acidity .. .. 126 32. • Chromic Acid Oxidation of Triarylcarbinols : Hartirett Plot .. 130 33. Gironiic Acid Oxidation of Triarylcarbinols : Hammett Plot 131 34. Permanganate Oxidation of Triarylcarbinols : Hammett Plot 133 35. Permanganate Oxidation of Triarylcarbinols : Haimett Plot .. 134 36. Chromic Acid Oxidation of Triphenylcarbinol : Variation of Rate-constant with Temperature .. .. 136 37. Permanganate Oxidation of Triphenylcarbinol : Variation of Rate-constant with Temperature .. .. 137 I v/ould l i k e to thank Professor Ross Stewart for suggesting this researcli problem, and his valuable guidance and encouragement throughout this work. I would also l i k e to thank the External Aid Office, Government of Canada for financial assistance. 1 INTEODUCTICT'J Potassium permanganate and chromic acid have been used extensively as oxidizing agents for both organic and inorganic substrates-. The mechanisms of the reactions depend largely on the substrate and on the experimental conditions. Permanganate Oxidation : Manganese can exist i n several oxidation states and the mechanisms of permanganate oxidation reactions vary to a great extent depending on the reaction conditions and the acidity of the medium, which determines the oxidation state to which per-manganate i s reduced. The oxidation state of manganese i s known to vary from +1 to +7, but only +2, +4 and +7 states are stable over a wide range of acidity. In highly alkaline solution the reduction product of permanganate i s manganese (VT), i n alkaline, neutral or weakly acidic solution manganese (IV) has been found to be the stable reduction product, and i n strongly acidic solution permanganate can be reduced to manganese(II). Stewart (1) has reviewed the properties and reactions of the different oxidation states of manganese i n d e t a i l ; a brief discussion of the same i s given here. Manganese(VI) i s stable i n basic solution and exists as manganate ion MnO~. Below pH 13, i t slowly disproportionates to manganese (VII) and manganese (TV), the rate of disproportionation 2 increasing with acidity of the medium (2,3). Usually oxidation by manganese(VI) i s slower than that by manganese(VII) (4,5). The only exception i s the oxidation of aromatic aldehydes i n which case the rates of oxidation by the too species are nearly same (6) . The reactions of manganese(VI) with 1,2-diols, phenols and olefins have also been studied (7). Stronger oxidizing agents oxidize manganate to permanganate. Manganese(V), or hypomanganate, i s f a i r l y stable i n cold concentrated a l k a l i solutions stronger than ION, but below 8N alkai i t disproportionates to manganese(VI) and manganese(IV) (7) . Primary and secondary alcohols are oxidized slowly by hypomanganate, but alkenes, tertiary alcohols and phenols are resistant to attack. Manganese dioxide, which i s the normal form of manganese(TV), i s the reduction product of permanganate i n weakly alkaline or acidic solution, except i n certain oxidations i n acid solution, such as with oxalic acid where the ultimate product i s manganese (II). Usually i t i s a brown insoluble powder which can exist i n soluble forms under certain circumstances. The precipitation of manganese dioxide i n solution i s delayed by phosphate buffers due to the formation of a complex. I t i s also soluble i n fuming sulfuric acid. Manganese dioxide, dispersed i n organic solvents has been found useful as an oxidizing agent for alcohols and diarylmethanes (8-12). Such heterogeneous reactions of manganese dioxide seem to follow radical mechanisms. 3 Manganese(III) i s stable as manganic ion i n concen-trated acid solution and also i n weakly acidic solution when complexed with such groups as fluoride or pyrophosphate, with-out which i t undergoes rapid disproportionation to manganese(II) and manganese(IV). However, In the presence of a large excess of manganous ion, manganese(III) i s f a i r l y stable even i n 1.5M perchloric acid (14). Pyrophosphate forms a stable complex with manganese(III) and the oxidation of malonic acid (15), pinacol (16), formaldehyde and formic acid (17) by manganic pyrophosphate have been studied. With strong reducing agents l i k e iodide, ferrous ion or oxalic acid, the end product of permanganate oxidation i n acid solution i s manganous ion, where the oxidation state of manganese i s +2. Permanganate reacts with manganese(II) i n neutral or weakly acidic solution, which i s the well known Guyard reaction used for the volumetric determination of manganese. The reaction taking place i s : 2MnO~ + 3Mn2+ + 2H"20 ^5Mn02 + 4H+ Ladbury and Cull i s (18) have reviewed the studies made on this reaction. In certain complexes manganese has been found to exist as manganese(I) (1). Permanganate Oxidation of Organic Substrates : Permanganate i s a powerful oxidizing agent for a wide variety of organic substrates. Only the reactions related to 4 the present investigation are discussed here. (i) Oxidation of Aromatic Rings; Aromatic compounds having electron donating ring substituents are readily degraded to carbon dioxide, whereas electron withdrawing groups appear to stabilise the ring. This i s obvious from the reactivities of phenols and anilines. Condensed polynuclear hydrocarbons are also degraded by permanganate (19). oxidation of aromatic aldehydes such as benzaldehyde (6,20,21), acetaldehyde and formaldehyde (20,21) has-been extensively studied. Stewart and Wiberg (6) studied the oxidation of benzalde-hyde and substituted benzaldehyde from pH 5 to 13. Two different mechanisms have been proposed by the authors for the reactions i n basic and acidic media. In basic solution from pH 11 to 13, specific hydroxy1 ion catalysis was observed according to the equation: 002H (ii ) Oxidation of Aromatic Aldehydes: The permanganate 1/2 5 Frcm pH 6.8 to 11, the rate was found to be independent of the medium. In more acidic solution, general acid catalysis was observed. In the basic region, the deuterium isotope effect with PhCDO was very small, the effect was observed to increase with decreasing pH and i n acidic medium i t s value was 7. In the basic region, a large positive "rho" was obtained from a Hammett plot. On the other hand, i n acidic medium "rho" has a small negative value. In neutral solution, the oxygen introduced into the alde-hyde, comes mainly from the oxidizing agent, whereas i n basic solution the solvent contributes a major part of this oxygen. For the reaction i n basic solution, a free radical chain mechanism was suggested and for the reaction i n neutral solution a fast ester formation between permanganate and aldehyde hydrate followed by rate-determining proton removal from the aldehydic hydrogen. In solution below pH 5, an autocatalytic reaction was found to take place between permanganate and benzaldehyde. ( i i i ) Oxidation of Aldehyde Hydrates; Aldehydes which are extensively hydrated i n aqueous solution are oxidized i n a way different from ordinary aldehydes. In basic solution, anions similar to those of alcohols are generated and are oxidized rapidly although formaldehyde has been found to be oxidized slowly (23). 6 Stewart and Mocek (24) studied the permanganate oxidation of fluoral hydrate in solutions ranging from 46% sulfuric acid to one molar alkali. They found four different reaction paths depending on the medium although in a l l cases the reaction is bimolecular between a fluoral hydrate and a manganese (VTI) species. The dominant reacting species, as the medium is changed from strongly basic to strongly acidic are : 0~ OH OH 1 I • I 'CF3- C -H .+ MnO~, CF 3— C—H + M n O ^ , C F 3 — C -H + MnO~ • 0" .. O" . ' OH : and . ' • '•' CF_-C-H + HMnO.. • 3 I 4 OH The large isotope effect with CF^CDCOH^ ^n ^ 1 cases suggests the breaking of C-H bond in the rate-determiriing step. It appears that either a hydride ion or a hydrogen atom abstrac-tion is occurring in this step. According to hydride ion tran-ter mechanism, the relative rate of oxidation of the different species should be: o" • 0"' OH •' ;;' 1 . 1 . 1 -CFo-C-II- >. ' CF, -C -H > CF0 -II . I 3 I • . 3 I • o ~ - . OH OH 7 and the oxidation by HMnO, should be faster than Mr£> • This fx is exactly what was observed. On the other hand, the effect of substituent Z in Z-CH-CF^  is much smaller than that expected for hydride ion o " abstraction. When Z is the strongly electron donating group 0~, the rate increases but at the same time, there i s an increase in charge density on the reductant which causes an electrostatic repulsion between the similarly charged reactants. This causes a less favorable entropy of activation which is more than com-pensated by the lowered enthalpy of activation. This suggests that hydrogen atom abstraction'.also can be a probable mechanism. The work of Halpern and Candlin (25) on the permanganate oxidation of the formate-Co(III) complex, which can only be interpreted in terms of hydrogen atom abstraction by permanganate, supports the latter mechanism as a possible route. (Iv) Oxidation of Formic Acid : Permanganate oxidation of formic acid to carbon dioxide has been extensively studied (23,26-29). In solutions of pH above 5, the rate i s almost independent of pH, since the reaction is with formate ion. Below this acidity, the rate is lower and follows the ionization of formic acid. In media containing more than 20% sulfuric acid, the rate increases due to the protonation of the permanganate ion (28,29). Above 8 50% sulfuric acid, the rate i s s t i l l higher due to the formation of MnOij. Large reactant isotope effects are observed i n these systems. A reactant isotope effect of 7 with DCC^ H (28,29) and solvent isotope effect of 0.38 (29) with was observed. By using permanganate-0"^ a considerable amount of 0"^ was found i n the product (28). Hydrogen atom abstraction mechanism, similar to that suggested by Candlin and Halpern (25) for the oxidation of formate-Co(III) complex, can be considered for the oxidation of formic acid. HCO~ + MnO~ — — ^ CO 2* + HMnO^  -. Mn(VTI) ^ C0 o .. ,Trr. "~ C0 9 2 or Mn(VT) ^ I t has been shown that both MnO^  and MnO~ oxidize formate anions (30). The oxygen transfer i s explained by the following mechanism : COl* + HMnO" *-0_C—CMnO_H2~ ^CO?" + Mn(V) In weakly acid solution, the unionized formic acid reacts with permanganate, although very slowly. The solvent isotope effect and reactant isotope effect are small i n such solutions. (slaw) (fast) 0 (v) Oxidation of Tertiary Hydrogen; Kenyon and Symons (31) found that carboxylic acids of the general structure r^CHfCT^^ 0-^ ^ s t r o n 9 a l k a l i solution can be converted into hydroxy--acids R2C(0H)(CH^^CCyi i n good yields. With dilute a l k a l i , the reaction i s slow and leads to extensive degradation. Permanganate oxidation of these acids i n concentrated a l k a l i solutions gives racemic products, whereas manganabe i n dilute a l k a l i , being equally effective as oxidant, gives products with retention of optical activity. The authors suggested that i n strong a l k a l i solution, permanganate may remove a hydrogen atom through hydroxy1 radical, although this does not explain the attack on the 7-H. (vi) Oxidation of Alcohols; Primary and secondary alcohols readily, undergo, oxidation by permanganate but tertiary alcohols are oxidized only under drastic conditions, and degrad-ation products are obtained. The rates of oxidation of primary and secondary alcohols are higher i n basic solutions than i n neutral or weak acidic solu-tions, although secondary alcohols are more readily oxidized i n acid solution than the primary alcohols (32). In more concen-trated acid the reactions are faster, due to protonation of the oxidant and probably also due to induced oxidation involving lower oxidation states of manganese (33). In basic solution, the rates increase with increasing pH, due to the ionization of the alcohol (4,34). 10 The permanganate oxidation of benzhydrol (CgHj^CHOH (4), h3xafluoro-2-propanol (CF3)2CHOH (35) and fluoral hydrate CF^GI (OK) 2 (24) have been studied extensively. The las t com-pound, although i s an aldehyde hydrate, can also be classed as a secondary alcohol; the oxidation reactions of this compound have been discussed before. The oxidation products of secondary alcohols are ketones. Enolizable ketones undergo further reaction to give cleavage products. Ketones which do not eriolize are stable i n alkaline permanganate. Primary alcohols produce aldehydes which are further oxidized to carboxylic acids when no a-hydrogen i s available for enolization. Enolizable aldehydes undergo degradation by permanganate. The reactions can be represented as : P^CHOH + OH" : — ^ r^CHO- + H 2 O „ VII Mn R^CHO" = *• P ^ O O Mn 7 1 1 • 0H-R C H 2 C H 2 O H «-RCH 2CHO RCH=CHOH OH" Mn Mn RCH 2CO2H RGO 2H + c o 2 The oxidation of alcohols i n neutral solutions i s very slow, and the rate i s proportional to the hydroxy1 ion concen-tration (4,34) : 11. Rate = k^CHOH J |^  MnO^  J [0H~ j The aldehydes are oxidized at a much higher rate than the alcohols i n neutral solution and the increase i n rate with the increase i n hydroxy 1 ion concentration i s not so much as i n the case of alcohols. Thus aldehydes can be isolated as the reaction product of oxidation of primary alcohols only i n the basic solutions, but not i n neutral or weakly acidic solutions. C u l l i s and Ladbury (36) have investigated the oxidation of benzyl alcohol and of 1- and 2-phenylethanol. With benzyl alcohol ring cleavage occurs to a great extent. I t has been shown by Stewart (4) that the alcohol-per-manganate reaction involves hydrogen abstraction, either as hydride ion or as a hydrogen atom, by permanganate ion from the anion of the alcohol. The rate-controlling step i s the reaction between the alkoxide and the permanganate ion. The deuterium isotope effect with R2CDOH for the oxidation of benzhydrol and fluorinated alcohols are large, shaving the rupture of C-H bond i n the rate determining step, which can be represented as: P^ CHO + Mn04 P^ C-O" + HTtiO~ The effect of substituents i n the phenyl or phenyltri-fluoromethyl carbinol on the rate controlling step i s very small, 12 shewing the mechanism involving hydride ion abstraction to be unlikely. If hydrogen atom abstraction i s taking place, i t would • mm produce the radical anion F^ C-O which i s rapidly oxidizexl to the ketone. This also explains the greater ease of abstraction from alkoxide ion than from neutral alcohol due to the relatively greater s t a b i l i t y of radical anion P^"-0 compared to the neutral ketyl PvjC-OH (37). Kurz (38) has examined these systems i n order to distinguish between different mechanisms. Chromic Acid Oxidation : Chromic acid oxidations have been reviewed i n detail by Westheimer (39) and recently by Wiberg (40). Only the aspects related to the present investigation are discussed here. Chromium i s known to exist i n oxidation state +2 to +6. Oxidation by chromic acid results the reduction of chromium (VI) to the +3 state Cr^ +, since the intermediate chromium (V) and chromium (TV) species are unstable and chromium (II) i s a very strong reducing agent under ordinary conditions. Chromium i n oxidation state of +6 exists as chromic oxide (CrO^^ or chromic acid, both containing O 0 — C r — 0 I O units (41). In dilute aqueous solutions (less than 0.05M) the predominating species i s HCrO., whereas at a higher concentraction dehydration occurs to a large extend according to the folia-zing equation : 2_ 2HCrO~ x * " r r 2 n 7 ~ + H 2 ° Polychromates are formed at s t i l l higher concentrations (42). Alkaline chromate i s devoid of oxidizing paver and i t 9— has been found that C^O^ has a much weaker oxxdizxng paver than HCrC>4 (43). The anhydride, CrC>3 and the acid chloride CrC^C^ are strong oxidizing agents. In strong acids, cations such as HCrO^ exist; this explains the sol u b i l i t y of CrC^ i n acetic acid. Moreover, chromium(VI) forms a new species with the anions added i n solution, the reactivity of this species largely depending on the identity of the added anion (44). Thus the chlorochromate anion, ClCrO^, formed by the following reaction 0 0 1 - - I H + HO— Cr 0 •+ Cl^=fcCl C r - — 0 + H00 I l 2 0 0 i s of much less oxidizing power than HCrO^. I t has been found that i n anhydrous acetic acid or i n acetic anhydride incomplete reduction of chrorrium(VT) occurs, leading to the formation of brown insoluble complexes of chromium(III), c±iromium(VI) and acetate ion (46). Chromium(V) can exist i n strongly basic solutions; Baily and Symons (47) prepared green K"3Cr04 by the action of KOH on K2Cr04. I t has been proved thtfc the compound contains cliromium(V) (47,48) . Solutions of chroinium(V) i n a l k a l i absorb oxygen to form chromium(VI) which on dilution disproportionates to chromium-14 (III) and dircsnium(VI). Barium hypochromate BajfCrO^) i s another known derivative of chromium(V) (49). The oxide CX2O5 i s also known (50). The known derivatives of chromium (IV) are CrF^, CrCl^ (51). Chromium dioxi.de CrC^ has also been prepared (52). Other known compounds of chromium (IV) are Ba CrO , Ba CrO . t ~ 2 4 3 5 Sr 2Cr0 4 (49b). Westheimer et a l . (39,53) have clearly shown that these intermediate chromium (V) and chromium(IV) species are stronger oxidizing agents than clirornium(VI). Chromic Acid Oxidation of Organic Substrates : Chromic acid oxidizes many classes of organic compounds. The reactions related to the present investigation are given i n the following discussion. (i) Oxidation of Hydrocarbons : (a) Aryl aikanes : Although benzene i t s e l f i s relatively resistant to oxidation by chromic acid (51) polycyclic aromatic hydrocarbons are easily oxidized to quinones (54,55). The reactivity of these conpounds depends on their electron a v a i l a b i l i t y and this i s consistent with attack by a cation such + . . . as HCrO-j. Chromic acid easily destroys ring systems containing electron-donating substituents such as -OH or -MI^. ^ i t h acidic ciiromic acid solution, the alkyl group of an alkylbenzene i s f i n a l l y oxidized to a carboxylic group (56). In diphenylmethanes the rrethylene group i s readily oxidized to a carbonyl group (57,58). Similarly the dibenzylbenzenes are oxidized to dibenzoylhenzoic acid (59). Further oxidation of the products leads to carboxylic acids (60). With different electron withdrawing groups rearranged products are also 15 obtained (61). Triarylmethaney on oxidation give the corresponding tertiary alcohols, which on further oxidation lose one aryl group and produce the diaryl ketones (62,63) as w i l l be ueen later. The chromic acid oxidation products of aliphatic alkanes usually undergo further oxidation and hence the reactions are not of much use i n synthesis. The primary carbon-hydrogen bonds are rather unreactive towards chromic acid. The oxidation of secondary carbon-hydrogen bonds leads to the formation of ke-tones which can undergo further oxidation (64). The tertiary carbon-hydrogen bonds on oxidation produce the corresponding tertiary alcohol, which also undergo further dehydration or oxidation (65). (b) Carbon-carbon double bonds: The oxidation of alkenes generally leads to several products obtained frcm the cleavage of the double bond, such as epoxides,ketols, acids or ketones and thus i t i s not of much use i n synthesis. The products of chromic acid oxidation of carbon-carbon double bonds are different i n sulfuric acid than i n acetic acid medium, the former giving mainly the rearrangement products, although the intermediate may be an epoxide i n both the cases (66,67). In some cases, a l l y l i e oxidation occurs (68). Thus cyclohexene yields cyclohexenone along with other products. This type of oxidation i s rather un-common with straight chain alkenes. ( i i ) Oxidation of Aldehydes: The kinetic studies of several aromatic aldehydes i n acetic acid as well as i n aqueous 16 solutions have been reported (69,70). The oxidation of benz-aldehyde i s of f i r s t order with respect to both |-PhCHoJ and j^HCrO" J and the rate constant follows the H Q acidity function (69,70). Comparison with PhCDO showed the cleavage of the C-H bond to be the rate-determining step. The effect of substituents on the oxidation rate indicates a positive value of p (70). The oxidation of aliphatic aldehydes i s found to be similar to that of aromatic aldehydes (71-74). I t has been shown with both aromatic and aliphatic aldehydes that the addition of manganous or oerous ions reduces the oxidation rates by 30-50% by eliininating secondary reactions due to chromium(V) and chrardura(IV) (69,72). A detailed study of such an induced oxidation and determination of the kinetics of the intermediate steps leading to the reduction of chromium (VI) to chromium (V) and chromium (IV) has been reported by Wiberg and Richardson (75), by using a connpetitive technique with substituted benzaldehydes. On the whole, the reactions of aldehydes are quite similar to those of alcohols, to be discussed later. ( i i i ) Oxidation of Ketones : Generally ketones on oxi-dation give two carboxylic acids derived from carbon-carbon bond cleavage (64,65). The kinetics of the chromic acid oxidation of cyclohexanone have been studied i n detail (76,77). The rate of reaction was found to be proportional to the concentrations of 17 ketone, c\cid chromate ion and acid. A kinetic isotope effect of 5.5. was obtained by comparison with cv'clohexanone-d^ and a solvent isotope effect of 4-5 was observed by using D20 (76). An enol intermediate has been suggested for the reaction (76-78). •- (iv) Oxidation of Alcohols : The chromic acid oxidation of alcohols has been investigated i n considerable d e t a i l . The reactions haw been studied i n different solvents such as water, aqueous acetic acid, aqueous acetone; mineral acids have also been used as catalysts. The chromic acid oxidation of isopropyl alcohol i n different solvents has been studied extensively (43-45, 79-82). In the absence of any other oxidizable group, a quantitative y i e l d of the corresponding ketone was obtained from secondary alcohols. The oxidation of primary alcohols i s complicated due to the formation of oxidizable products namely aldehydes (83). However, the reaction can give'satisfactory results i f the products can be removed from the reaction mixture as soon as they are formed. Vfestheimer and Novick (43) suggested the following rate law for the oxidation of isopropyl alcohol : v = k a j^ HCrO^ j j^aoiJJH +J + k b ^HCrO^P^CHOH^H"^2 This rate law has also been found to be applicable to the reactions of other alcohols (45, 76, 84). • Comparison with the, rate of oxidation of iscpropyl-a-d-alcohol showed that the reaction has a considerable kinetic isotope 1 8 effect (82), proving the cleavage of carbon-hydrogen bond i n the rate determining step. The existence of a pre-equilibrium step has been shewn by the fact that vhe reaction i s faster i n DjO than i n 1^0. In the presence of a large excess of manganous ion the rate of reaction was decreased by a factor of too (85), and one mole of manganese dioxide was produced for each mole of acetone formed. From these observations, the following steps were suggested for the oxidation i n the presence of manganese (II) R2CH0H + C r 7 1 — ^P^CO + Cr 3^ 2 0 ^ + Mn 1 1 MnO + 2 C r m Hence the probable reaction scheme i s : P^ CHOH + C r 7 1 ^P^CO + Cr 3 7 7 C r ^ + 0^ -2Cx V P^ CHOH + C r V e-P^ CO + C r 1 1 1 Since i t i s possible to prepare the chromate ester of isopropyl alcohol, Westheimer et a l . suggested the following mechanism for the oxidation : R2aiOH + HCrO + II R2CHOCr03H + H20 4 R2CHOCr03H + H + R2CH0CrO3H2+ R2CHOCr03H -^R200 + H 2Cr0 3 R2CH0CrO3Hj + H20 «^R2CO + H 2Cr0 3 + H 30 + The decomposition of the chromate ester was thought t o occur by proton loss t o any available base with elimination of a chromium(VI) ion: R O-Cr03H2 R C ^ CO + BH + H 2Cr0 3 H R :B folia-zed by Cr J" v + Cr V J- ^ 2Cr IV , ^ „ I f a s t n^,V fa s t C r V + P^ CHOH *-Cr E I I.+ P^ CO This mechanism v;as questioned by Ro&ek (79) who pointed out that the evidence i n d i c a t i n g the p a r t i c i p a t i o n of external base i n the reaction was inconclusive. He also suggested that since water, which i s the only base present, p a r t i c i p a t e s , i t i s u n l i k e l y that the reaction w i l l follow the h Q a c i d i t y function. 20 Graham and V7estheimer (87) showed that i f the too rate expressions are written, including the activity of the activated complexes, they become equal. This happens since water i s formed i n the f i r s t step of the ester mechanism. Thus the a c t i -v i t y of water that i s introduced i n rate expression w i l l be cancelled by the one as a result of the l a s t equation. Evidence for ester formation was also obtained from an investigation of the oxidation of the st e r i c a l l y hindered alcohol 38-28-diaoetoxy-68-hydroxy-18B-12-oleanen (40). Recently Wiberg and Schafer (88) have reported the direct spectrophotomstric observation of the formation of the intermediate chromate ester. Another objection to the ester mechanism i s the effect of substituents on the oxidation of phenylmethyl carbinols. Kwart and Francis (94b) have reported the reaction constant for such an oxidation to be -0.37 to -1.01, depending on the reaction conditions. Rocek obtained p* = -1.11 for the oxidation of aliphatic secondary alcohols (79) and -1.06 for primary alcohols (86). Similar results were obtained for a series of aryltrifluoroitethyl carbinols by Stewart and Lee (89) who found p = -1.01, and the recently reported oxidation of isopropyl alcohol i n aqueous acetone gives the value of p to be -1.16 (90). The results show a relatively electron poor reacting carbon i n the activated complex. According to Stewart and Lee (44,89,91) the proton abstraction should have a positive p, since the base catalyzed decomposition of analogous nitrate ester has a p value of +3.6. 21 This was also suggested by Rocek (79). However, the p value for elimination reactions i s known to vary widely depending on the degree of bond breaking i n the transition state. The effect of ring size on the rate of oxidation has been studied by Kuivila and Becker (84!). I t was observed that the 5-,7-, and 8-membered ring alcohols are more reactive than cyclohexanol. This i s due to the ease of change from tetra-hedral to a trigonal arrangement about the reacting carbon. In a recent publication (92) i t has been shown that the rates of oxidation of cyclic alcohols depend mainly on the r e l i e f of strain present i n the reactant molecule. The mechanism originally proposed by Rocek et a l . (79) involves a hydride transfer from alcohol to chromic acid or to i t s conjugate acid, with a simultaneous proton loss from the hydroxyl group as the carbon-hydrogen bond was cleaved. R O-H H-0 0-H ' : \ / j \ / •...• + C + yyQx RoCO + HCCTCM R IV 0 0-H The solvent isotope effect obtained from the rate con-stant i n D 20, shows this mechanism to be unlikely. The rate of oxidation i n D 26 i s much faster than that i n H 20 which arises from. + + the greater acidity of D3O ccmpared to H-^ O and the consequent larger concentration of the reactive conjugate acid. In the Rocek 22 mechanism the primary kinetic isotope effect due to the cleavage of oxygen-hydrogen bond should at least cancel the solvent isotope effect, since the rapid exchange of hydroxyl protons with the solvent produces deuterated hydroxyl groups. These observations show that the oxygen-hydrogen bond can neither be intact nor be broken in the transition state, since ethers are relatively inert to chromium (VT) and because of the positive solvent isotope effect. Hence i t can be reasonably concluded that the ester is an intermediate in the oxidation path. Rocek et al. have found supporting evidence for the ester mechanism from the study of the oxidation of highly hindered secondary alcohol 3$,28-diacetoxy-68-hydroxy-18S-12-oleanen (93). , It was observed that the isotope effect in 80% acetic acid is 1.0 and in less acidic solution, where the reaction is slaver, i t is 2.0. Presumably in the former case, ester formation becomes the rate-controlling step. Con-sidering a unimolecular deconposition mechanism for the chromate ester, the following rate controlling step is possible : Transition state 23 Kwart and Francis have alL-.o proposed a similar trans-i t i o n state (94a). Further support for an ester intermediate has been obtained by Wiherg and Shaefer (88) (see p 20). Stewart and Lee consider that this transi.tion state involves electron transfer to chromium through the C-K-0 bonds as well as through the C-O-Cr bonds, i.e., p a r t i a l bonds bind the hydro-gen to both carbon and oxygen i n the transition state (90). Thus the transition state becomes electron deficent and the dev-eloping carbonyl group w i l l be stabilized by electron donating groups. This explains the observed negative value of p. Kwart et a l . do not believe that the developing group has any influence on the reaction (94c). Tertiaty alcohols are relatively resistant to chromic acid oxidation but this can be effected i n the presence of sul-f u r i c acid. Sager (95) has reported the oxidation of t r i e t h y l -carbinol which proceeds by prior dehydration to the ol e f i n , the f i n a l product being a ketone. When the tertiary alcohol i s unable to dehydrate, i t under-goes oxidative cleavage (96-98). Cr .VI (CgH5) 3C0H HOAc (C6H5)2CO CrVT (C6H5)3CCHOH(C6H5) (C6H5)3COH + C6H5CHO HOAc Cleavage products also appear i n the case of phenyl-t-butyl carbinol (85,99); here the cleavage product consisting of 24 t-buvyl alcohol and benzaldehyde i s 60-70%, the phenyl-t-butyl ketone being the minor product. r CrVI C gH 5— C — C (CH3) 3 ^ C 6H 5—C — C (CH3) 3 > I I H • 0 + C6H5CHO + (CH3)3COH An interesting feature of this type of oxidation i s the virt u a l elimination of cleavage products i n the presence of manganous or cerous ions, which catalyze the disproportionation of chromium(IV) and chromium (V) (85). This shows that the ketone i s the oxidation product of the alcohol by chromium (VI) whereas the chromium species of intermediate valence produce the cleavage products. Oxidation of Alcohols by Other Reagents: Vanadium (V) : Pentavalent vanadium i n acid solutions exists as pentavanadyl ion VO^ . I t has been sham by Waters and his group (100) that the reactions of the oxidant with primary alcohols are the same as those with secondary alcohols, although the tertiary alcohols are much more resistant to attack. There i s evidence (101) that i n acid solution vanadium(V) forms a com-plex of the formula ROH.V(OH) ^  with an alcohol. The rate-determining step i n the oxidation of cyclohexanol to ketone i s the decomposition of the complex, as sham by the existence of 25 an isotope effect i n the oxidation of cyclohexanol-l-d. The kinetic equation corresponding to this slow decomposition of the complex to vanadium(XV) and an alcohol radical i s : d [V(V)] = kjpDnJ jjWO^ J |H30 J dt I t i s evident from the polymerization of added acrylonitrile i n such systems that the homolytic scissions, shown i n the following equation, take place: OH H VO+ + H 3 0 + + r Y ^—^r-x, ?. D H H .^V(OH) *-( Y+ ^^V(OH), + DOH D H 3 0 ^ + fast ( >=0 + 2 V ( I V ) •V(0H)2 V 0 J The presence of even small amounts of bromide ion changes the course of the oxidation of alcohols. The rate con-t r o l l i n g step i n this case i s the production of bromine atoms which act as the active oxidant (102): V(V) + Br"* V(IV) + Br* P^ CHOH + Br" P^ COH + Br" + H + 26 With certain alcohols, carbon-carbon bond fission giving rise to a stable radical is more favored than a carbon-hydrogen bond cleavage. Thus t-butylphenylcarbinol on oxidation produces benzaldehyde,, instead of t-butyl phenyl ketone. The ready forma-tion of the t-butyl radical assists in the scission of the carbon-carbon bond, which may accompany the hemolytic cleavage of metal-oxygen bond (103). 0* I CfiHr C — t-Bu I H OH I v v C 6 H 5 — C — t - B u H C6H5CHO + t-Bu C6H5CHO + t-BuOH Such carbon-carbon bond fission provides a route for the oxidation of tertiary alcohols and occurs when one group is split off as a stable radical. Cobalt(III) : Unlike vanadium(V) and cerium(IV), the oxidation of alcohols by cobalt(III) is faster than that of ketones. In a l l the cases of oxidation of alcohols, the rate of the main reaction varies inversely with the acid concentration and with cyclchexanol-l-d, a small isotope effect, k H/k D = 1.72 at 10°, was observed (104). The inverse acid dependence of the rate is explained to be due to the rapid exchange of a ligand with an alcohol molecule, occurring with the ion Co (OH) (n^P) | + but not with the ion Co(H 0 ) 3 + . 2 6 27 Co(H 20)g + a Co (OH) v'H20)|+ + H + According to Hoare and Waters (104) the i n i t i a l l y formed complex between cobalt(III) and the alcohol decomposes to a radical which i s oxidized rapidly : p H R R I .• . X I Co(III) 0 - — c R CO....0 C v *- Co(Il): + HO—c* R kR R The small kinetic isotope effect i s suggested to be due to the concerted breaking of a carbon-hydrogen group and the removal of an electron from the oxygen atom of alcohol by cobalt(III). Cerium (IV) : The oxidation of alcohols by eerie ions involves the formation of a cerium (IV)-alcohol complex which under-goes slow decomposition to products (105,106) : C e I V ( H 2 0 ) 8 4 + + C2H5OH « » Ce^O^O) ? (C^OH) 4 + + H20 Ce^O^O) (C^OH) 4 + — ^ products The deuterium isotope effect, for eerie sulfate oxidation of a-deutero-cyclohexanol was found to be 1.9, indicating 28 th.3 cleavage of carbon-hydrogen bond i n the rate-determining ster- (107). The la.-/ value of the isotope effect has been explained by a-cylic activated complex (107,108). For the kinetics of the oxidation of formaldehyde by eerie ion i n acid the follaving mechanism has been proposed (106): iv S L C W III X + Ce + CH2 (OH) ? ^ Ce + CH~ ' + H NOH 0> X IV F A S T TTT CH0 + Ce H^CO_H + Ce NOH Lead Tetraacetate; Because the hydrolysis of lead tetraacetate i s a fast reaction and because the exchange of acetate groups with other carboxylate groups takes place readily, i t can be anticipated that similar equilibria might be established with other substances having acidic hydrogens. Thus the observed oxidations with lead tetraacetate are probably effected by the species i n equilibrium with the reagent. Although alcohols are stable i n lead tetraacetate-acetic acid solution, (since the equilibria l i e on the side of the reactants), good yields of aldehydes or ketones have been obtained by the oxidation of alcohols i n boiling benzene (109,110). The following scheme has been sug-gested as a possible path: The heterolysis of the bond between the alcohol oxygen and the lead i s caused by the great electron a f f i n i t y of the lead. This leads to a simultaneous breaking of the carbon-hydrogen bond to replace the electron pair lost by oxygen. This mechanism also agrees with the fragmentation which can occur during the oxidation of alcohols (111-114). 30 SCOPE OF THE PRESEHT PESEAPCH The permanganate oxidation of alcohols has previously been studied i n basic and weakly acidic regions. In the case of the secondary alcohol, benzhydrol, the reaction was found to be of second order, f i r s t order i n each of the reactants, and the product was found to be benzophenone (4). No work has been reported on the permanganate oxidation of alcohols i n highly acidic regions. Hence i n the present investigation an attempt has been made to determine the course of the permanganate oxidation of benzhydrol . and substituted benzhydrols by studying the kinetics and the products of reaction. The chromic acid oxidation of a number of secondary alcohols i n various media has been studied (43-45, 79-82, 85-91), but not of diarylcarbinols. There has been controversy about the exact mode of transfer of the hydrogen atoms removed from the alcohol during chromic acid oxidation. In some cases, both oxi-dation and cleavage products have been obtained (99). In order to ascertain the mechanism of chromic acid oxidation of diary 1-carbinols, the kinetics and the products of the reaction of benzhydrol and i t s derivatives with chromic acid were studied. No work has been reported on the permanganate oxidation of triarylcarbinols. These compounds lack a hydrogen atom on the alcoholic carbon and any oxidation must produce carbon-carbon 21 bond cleavage, possibly via aryl migration. It seemed worth-while to examine this reaction closely. The chromic acid oxidation of triphenylcarbinol in acetic acid has been shewn to produce benzophenone and phenol (96,97). This reaction may occur via phenyl migration from carbon to electron-deficient oxygen and in order to investigate this and other possibilities, the reaction kinetics, substituent effects and the products of the reaction were examined. It has been reported that the addition of manganous ion reduces the rate of chromic acid oxidation of carbinols and in some cases changes the course of the reaction. These effects have been attributed to the scavenging action of manganous ion on the powerful oxi-dants chromium (V) and chromium (TV), formed as reaction inter-mediates (53,75). In order to find out i f any of these species are involved in the chromic acid oxidation of triarylcarbinols, the reaction was also studied in the presence of manganous ion. SECTION I OXIDATION OF DIAKYLCARBINOLS 32 EXPEPJ4ENTAL Materials t Benzhydrol and benzhydrol a-d, prepared previously in these laboratories, were crystallized from carbon tetrachloride. Both showed the constant melting point 64-65° (4). 4-MethyIbenzhydrol, 4,4'-diirethy Ibenzhydrol, 4-chloro-benzhydrol and 4,4'-dichlorobenzhydrol were prepared from the corresponding benzophenones by sodium borohydride reduction of the latter. The f i r s t two carbinols were crystallized from ether-ligroin and the last too from ethanol-water to constant melting points. The yields and the melting points of the carbinols prepared are given below: 4-Methy Ibenzhydrol: yield 90.5%; m.p. 50°, l i t . 49° (115a). 4,4'-DimethyIbenzhydrol: yield 90.1%; m.p. 68°, l i t . 69° (115b). 4-Chlorobenzhydrol: yield 95%; m.p. 59-60°, l i t . 59-61° (116). 4,4,-Dichlorobenzhydrol: yield 91.2%; m.p. 90.5°, l i t . 89-90° (115c). Using the same method, 9-fluorenone was reduced to 9-o fluorenol, crystallized from water; yield 91.2%, m.p. 157-8 , l i t . 158-9° (115b). 4,4'-Dimethoxybenzophenone was reduced to 4,4'-dimethoxy-benzhydrol by lithium aluminium hydride. The product was 33 crystallized from ethanol-water mixture to constant melting point 70-71°, l i t . 72° (117). The yield was 96%. Dimesityl carbinol and 4-methoxybenzhydrol, previously prepared i n these laboratories, were crystallized frcm ethanol-water to constant melting points 150° and 68° respectively. Benzhydrol-4-carboxylic acid was prepared by Mr. , J.A. MacPhee i n these laboratories by sodium borohydride reduction of the corresponding ketone. I t was crystallized from chloroform-ethanol to constant melting point 165-6°. Potassium permanganate solutions were standardised by t i t r a t i o n against aliquots of 0.1N sodium oxalate, both being prepared according to Vogel (118). Concentrated sulfuric acid (0.5 ml) was added to the flask along with sodium oxalate solution and permanganate was added to i t u n t i l the pink end point was obtained. Standard potassium dichrornate solution was prepared by direct weighing of reagent grade potassium dichromate into a stoppered volumetric flask and then made up to the mark with d i s t i l l e d water. Kinetic Methods : The kinetics of permanganate oxidation were followed by two procedures, namely t i t r a t i o n and measurement of li g h t absorption. In the f i r s t method an aqueous solution of approximately 0.006M benzhydrol was prepared by dissolving a weighed quantity 34 of benzhydrol i n hot d i s t i l l e d water which had been boiled to remove dissolved gases. In a typical experiment 10 ml of this solution, 20 ml of 1M KH^ PO^  buffer adjusted to the desired pH, and 68 ml d i s t i l l e d water were taken i n a 125 ml stoppered red Erlenmeyer flask which was then immersed i n a constant tem-perature bath, the temperature of which was adjusted to 25 ± 0.02°. Two ml of standard potassium permanganate solution was added to i t to start the reaction. The aliquots were withdrawn with a fast delivery pipette and added to a quenching solution containing 10 ml of 2M sulfuric acid, 3 ml of 5% sodium bicarbonate solution and an excess of potassium iodide. The liberated iodine was t i t r -ated with an appropriate thiosulfate solution using Thyodene as indicator. The t i t r a t i o n method was used for reactions i n solu-tions of acidity less than H Q 0.20, above which the reactions were too fast to be followed by this method, and hence the spectro-photometric methods were used to study the reactions. The ratio of substrate to permanganate used i n the reactions studied by ti t r a t i o n method was 3:2. This corresponds to the three equivalent change from Mn(VTI) to Mn(IV) and two equivalent change of the substrate. In the spectrophotorretric method the reaction was followed by observing the disappearance of permanganate at 526 my. Solu-tions of the alcohols were prepared by accurately weighing out the compounds directly into 10 ml volumetric flasks and made up to 35 the /nark with distilled methanol. In a typical run for per-manganate oxidation, 0.2 ml of 1.79 x 10" solution of benz-hydrol in methanol was taken in a 10 ml volumetric flask. The solvent was evaporated under vacuum; 2 ml of 18.01 M sulfuric acid was then added to the carbinol and water was added to i t slowly v/ith constant cooling t i l l the volume was up to the mark; 2.8 ml of this solution was taken in a c e l l of path length 1 cm and placed in the cell compartment of a B & L Spectronic 505 spectrophotometer, thermostated at 25 ± 0.02° by water circulating through the thermospacers. The contents of the ce l l were allowed to stand for 20-25 minutes' in order to attain the constant temperature. Next 30ul of 0.018 M potassium permanganate solution was injected from a 50ul syringe into the c e l l , and the reaction was follaved by observing the decrease in the absorbance of permanganate at 526 my. For the chromic acid oxidation of the diaryl alcohols only the spectrophotcmetric method was found to be suitable, since the reactions v/ere too fast for the titration method. In this case the solutions were made in the same way as in the permanganate oxidations; the only difference was that 80 wt% acetic acid in water was used as the reaction medium. The reason for using acetic was to obtain a higher concentration of the car-binols in solution since the reactions were rather slav for spectrophotcmetric measurements. In a typical run for chromic 36 acid oxidation, 2 ml of 2.13 x 10~^ M solution of benzhydrol in 80 wt% acetic acid in water and containing the required quantity of sulfuric acid was taken in a cel l and placed in the cell ccnparutK2nt of Cary 16 spectrophotometer. The ce l l com-partment was thermostated to 25 ± 0.02° by circulating water from a constant temperature bath. When the contents of the cell attained the constant temperature, 10yl of 1.42 x lO^M potassium dichrcmate solution was injected from a 50ul syringe. The reaction was followed by observing the disappearance of Cr(VI) at 350 my. In a l l of these reactions the ratio of the oxidant to the substrate used was 3:2, since the reactions correspond to the three equivalent change from Mn (VII) to Mn(IV) and Cr (VI) to Cr(III), and two equivalent change of the substrate. A spectrophotometry method was used to follow the rate of ionization of 4-methoxybenzhydrol by observing the increase in absorption of the corresponding carbonium ion at 466 my. Because of the low solubility of the carbinol in water 70% methanol (by volume) in water was taken as the reaction medium. In a typical experiment 2 ml of 18.005M sulfuric acid was taken in a 10 ml volumetric flask and made up to the mark xvith the aqueous methanol; 2.5 ml of this solution was taken in a cel l and placed in the thermos tated c e l l compartrrent of a Cary 16 spectro-photometer. Next 50yl of a 2.5M solution of 4-methoxybenzhydrol 37 was injected to the c e l l from a sy.-inge. The increase i n optical density; at 466 my showed the extent of ionization. Product Analysis : (i) Permanganate Oxidation : A weighed quantity of diarylcarbinol (0.001 mole) was dissolved i n 100 ml of 1.8M sulfuric acid. To the solution, 1.5 ml of 0.446H potassium permanganate solution was added. The mixture was allowed to stand t i l l the color of permanganate disappeared, after which the mixture was diluted with 100 ml d i s t i l l e d water and then extracted several times with petroleum ether. The extract was washed with water and dried over anhydrous sodium sulfate. The solvent was removed from the extract by evaporation and the infra-red spectrum of the residue observed. The residue was weighed and the melting point was ascertained. Following this method, the products of oxidation of benzhydrol and 4-methoxy-benzhydrol were isolated and analysed. In the case of benzhydrol, the 2,4-dinitrophenyl hydrazone of the product was prepared according to Vogel (119) and this served as an additional proof of the identity of benzophenone i n the product. (i i ) Chromic Acid Oxidation : The carbinol (0.001 mole) was dissolved i n 100 ml of 80 wt% aqueous acetic acid containing 15.9% sulfuric acid. To the solution 1.0 ml of 0.33M potassium dichromate solution was added. The mixture was allowed to stand for several hours t i l l the color changed to green. The products were isolated and analysed i n the same way as i n the case of 38 permanganate oxidation. Using this method, the oxidation products of benzhydrol, dimesityl carbinol and 4-methylbenzhydrol were isolated and analysed. Determination of Acidity Functions : (i) H c i n H7SO4 diluted with 80 wt% aqueous acetic acid (80g. acetic acid and 20g water) : The measurements were carried out with a Cary 16 spectrophotometer, provided with thermospacers, the temperature being kept constant at 25 ± 0.02°. The solution of acetic acid was made by mixing 100.0 ml of "Baker Analyzed Glacial Acetic Acid" (99.9%) with 26.25 ml of d i s t i l l e d water i n a glass-stoppered flask. For each measurement , calculated quantity of "Baker Analyzed Sulfuric Acid" (99.99%) was taken i n a 10* ml volumetric flask along with the required quantity of a suitable indicator and made up to the mark with 80 wt% aqueous acetic acid. The samples with varying quantities of sulfuric acid were thermostated at 25° and the absorbances of the indicators were measured at appropriate wavelengths. The indicators used were 4-nitroaniline, 3-methoxydiphenylamine, 3,4'-dichlorodipheny1-amine, 3-nitrodiphenylamine, 4-nitrodiphenylamine and 2-nitrodiphenyl-amine. Stock solutions of the indicators were prepared by dissolving weighed quantities of the cxxipounds i n 80 wt% aqueous acetic acid. Hammett's H q function relates the equilibrium between base and protonated base as (120) : H C = ? K B H + - ^ V 7 ^ 0 ( 1 ) 39 If AQ is the i n i t i a l absorbance of free base and A^ is the absorbance at any acidity, then - . <V*n>Ai = C B H + Z ^ ( 2 ) and H D = p K ^ - log [ < V V A i ] <3> The value of H Q at any acidity was calculated from equation (3) by substituting the values of A Q and A^ measured spectrophoto-metrically. (ii) Hp in HgSQ^  diluted with 80 wt% aqueous acetic  acid : The Steele solutions of acids and indicators were pre-pared and the measurements were carried out in the same way as in the case of the determination of the function H Q in the same system. The indicators used were tri(4-tolyl)carbinol, 4,4'-dimethyltriphenylcarbinol, triphenylcarbinol and tri(4-chloro-phenyl;) carbinol. The acidity function H is defined by the relation (121) H R = P V - l o g ^ / C ^ ) (4) where.Cj^ t- is the concentration of the carbonium ion and Cj ^ H is the concentration of the unionized alcohol. If A is the absorbance of the indicator at any acidity and A, the absorbance of the fully ionized alcohol, i t follows : Thus, IL^  = pKR+ - log^/fA-A^J (5) The value of H R at any acidity was calculated from eqn. (5) by substituting the observed values of A and A N at that acidity. 40 . ( i i i ) Hp i n H2SO4 Diluted with 70 vol% Aqueous  Methanol : Tlie values of the acidity function were calculated by using eqn. (3) 'following the similar experimental procedure. The indicators used were 2-nitroaniline and 4-chloro-2-nitroaniLine. (iv) HR i n H2SO4 Diluted with 70 vol% Aqueous  Methanol : Eqn. (5) was used to calculate the values of HR. The experimental procedures were similar to those followed for the detenrdnation of the acidity function i n the previously described madia. The indicator used was 4,4*-cu\irethoxyltriphenyl-chlororve thane. 41 RESULTS Product Analysis : (i) Permanganate Oxidation : The products of oxidation of a l l the diarylcarbinols contained carbonyl groups, as shown i n their infra-red spectra. In the case of benzhydrol, the meltincf point of the product was found to be 49°, l i t . 49° for benzophenone (119). The 2,4-dinitro phenylhydrazone melted at 238°, l i t . 239° (119). The yi e l d of the product was 99.2%. 4-Methoxybenzhydrol produced 4-irethoxybenzophenone i n 98.1% y i e l d , m.p. 62°, l i t . 61-62° (122). ( i i ) Chromic Acid Oxidation : The product obtained from the oxidation of benzhydrol was found to be the same as i n the case of permanganate oxidation, v i z . , benzophenone. The yi e l d was 98.5%. Dimesityl carbinol produced a compound of m.p. 139-40°, which was found to be diiresityl ketone from the analysis calc. C, 85.7%; H, 8.27%; found C, 85.4%; H, 8.3%; and also from the infra-red spectrum which shaved a strong carbonyl absorp-tion band. The yi e l d of dimesityl ketone was 99.7%. The crystallizable s o l i d product of oxidation of 4-methyl benzhydrol was found to be 4-nethylk>enzophenone, m.p. 52°, l i t , 53° (119). The yi e l d of the product was 96.4%. The non-crystallizable o i l y residue was probably a mixture of some 42 unreacted carbinol and the ketone produced, since i t s in f r a -red spectrum did not show the characteristic absorption bands -1 -1 of benzaldehyde at 2750cm and 2800cm . Since benzaldehyde i s the probable product i f phenyl-migration takes place, it-appears that no rearrangement due to phenyl-migration i s occurring. Determination of Acidity Functions; The results of spectrophotcmetric measurements are given i n Tables 1 to 5. The data from the f i r s t two tables are plotted i n Figs. 1 and 2. TABLE I Values of H Q i n H 2S0 4 diluted with 80 wt% aqueous acetic acid. (i) Indicator: 4-Nitroaniline, pK^-f = 1.00, = 370 %H2S04 H D 0.34 0.73 0.85 0.46 (ii ) Indicator: 3-Methoxydiphenylamine, pKgjj+ = 0.40, W = 2 8 2 ^ %H2S04 H c 1.80 -0.0196 4.2 -0.265 43 (iii) Indicator: 3,4'-Dichlorcdiphenylamine, pK^+ = -1.19, ^ = 288 my %EJSOA H 2 4 o 4.2 -0.27 5.8 -0.74 8.3 -1.12 12.9 -1.70 (iv) Indicator: 3-Nitxodiphenylamine, pKgjj+ - -1.61, = 280 my %H2S04 H c 5.8 -0.74 12.9 -1.66 18.0 -2.20 (v) Indicator: : 4-Nitrodiphenylamine, P 1^* = -3.13, \*ax = 4 0 0 m ^ %H2S04 H Q 18.0 -2.15 22.7 -2.78 35.6 -4.01 41.0 -4.50 44 (vi) Indicator : 2-Nitxodiphenyiandne, = -4.12, W = 4 3 3 ^ %H_SO. 2 4 H o 29.3 -3.38 35.6 -4.04 39.0 -4.37 41.0 -4.64 46.3 -5.25 51.8 -5.67 TABLE I I Values of KL^ i n H 2S0 4 Diluted with 80 wt% Aqueous Acetic Acid. (i) Indicator : Tri(4-tolyl)carbinol, pK^f = -3.56, \nax = 4 5 2 ^ %H2S04 H R 5.9 -2.50 8.3 -2.80 12.9 -3.55 15.9 . -3.95 %SULFURIC ACID 46 (ii) Indicator : 4/4'-Dirnethoxytilphenylcarbinol, pK + =-4.39, X = 456 mu * x. ' max %H2S04 H R 12.9 -3.60 15.9 -4.20 18.0 -4.40 22.7 -4.85 26.7 -5.50 (iii) Indicator : Triphenylcarbinol, pK + = -6.65, X = 433 my max %H2S04 H R 22.7 -4.80 26.7 -5.50 29.3 -6.01 31.6 -6.30 35.4 -7.00 36.6 -7.20 47 (iv) Indicator : Tri(4-chlorophenyl)carbinol, pKp+ = -7.74 F % H 2 S 0 4 ^ 37.7 -7.26 39.6 -7.48 TABLE I I I Values of K q i n H 2 S 0 4 Diluted with 70 vol% Aqueous Methanol (i) Indicator : 4-Giloro-2-nitroaniline, P&g£ = -1.07, W = 423 my M(H 2S0 4) • % 1.80 -0.15 2.16 -0.38 2.31 -0.50 2.88 -0.64 3.60 -0.82 ( i i ) Indicator : 2-Nitroaniline, pKgH+ = -0.26, = 412 M(H 2S0 4) H Q 1.80 -0.14 . 2. Acidity Function IIR in H2S04 Diluted with 70 2 0 3 0 % S U L F U R I C ACSD 49 TABLE IV . . . " r i Value of H R i n H2SO4 Diluted with 70 vol% Aqueous Methanol. Indicator : 4,4,-Diii^thoxytriphenylchloroinethane/. 3PKR+ r .-1.24, 500 my . .; . M(H2SO4) • HR . - .1.80 . -0.58 2.16 ' -0.98 . 2.30 •': -1.25 2.88 ' .' - -1.65 3.60 .' -1.94 ' : Order of Reaction : (i) Permanganate Oxidation : In a medium of lav acidity the reactions are of second order (4). For a 3:2 ratio of the substrate to permanganate, the following integrated second order rate expression was used (6) :.' .. 5 0 where, V C = volume of thiosulfate at t = 0 V T = volume of thiosulfate at time t 2 V — o = volume of thiosulfate at infinite time 5 [alcohol] = alcohol concentration at t = 0. Up to HQ 0 . 2 0 , good second order plots were obtained, a typical of which is shown in Fig. 3. At acidities greater than HQ - 0 . 5 0 the order of reaction changes to fi r s t order in carbinol concentration and zero order in permanganate concentration. Since the oxidant undergoes three-equivalent change whereas the substrate changes by two equivalents, which is apparent from the product analysis, i t is obvious that the rate of disappearance of permanganate i s two-thirds the. rate of disappearance of the carbinol, and the following f i r s t order rate expression was used : 2 *o i k t = 2 . 3 0 3 log 3 \ " A* where, A Q = absorbance of permanganate at t = O Aj. = absorbance of permanganate at time t An' = absorbance of permanganate at infinite time. For a 3:2 ratio of substrate to permanganate, A^ = O. Since AQ is a constant for a particular reaction, a plot of the logarithm of the absorbance of permanganate at any time against time gave a straight line. The f i r s t order rate constant was calculated from the slope of the line using' the above equation. A typical f i r s t order rate plot is shown in Fig. 4. Fig. 5 confirms that the rate is independent of per-manganate concentration. t in min 52 200 • 400 ~* 600 t in sec 53 54 In the intermediate region between H Q 0.20 and -0.50, the kinetics appear to change gradually from second order to fi r s t order and the rate plots f a i l to classify the kinetics clearly. Typical rate plots of this region are sham in Figs. 6a and 6b. (ii) Chromic Acid Oxidation : The rate of oxidation was studied in solutions containing 0.5 to 37.0% sulfuric acid in 80 wt% aqueous acetic acid. At these acidities the reaction was found to be of second order, f i r s t order with respect to each of the two reactants. Using the stoichiometric quantities of the substrate and Cr(VI), i.e. for the 3:2 ratio, the rate data were found to f i t the integrated rate expression (1) : 3H..- 1 ^ [Cr(V ! ) J t where and Cr(VI) Cr(VI) t = concentration of "cr(VI) at time t 0 = concentration of Cr(VI) at t = 0 Since the plot of 1/ Cr(VI) is a constant for a particular reaction, Cr(VI) t against time gives a straight line and the second order rate constant was calculated from the observed slope of the line. A typical second order rate plot is sham in Fig. 7. Dependence of Rate on Acidity: (i) Permanganate oxidation : In the range of 1.41M to 5.4M aqueous sulfuric acid, the oxidation of benzhydrol i s acid catalyzed. The concentration of permanganic acid is negligible 55 56 Fig. .6b. Permanganate Oxidation of Benzhydrol. H Q = -0.01 T =25 Second Order Rate Plot. O / O / / / k 2 = 9.76 x 10~1l.rrr)le~1sec~1 —i 1 1 r- r- , 400 1200 2000 2800 3600 4400 t in sec 58 in taxis region and i t seems reasonable. to ascribe the increase in ratv» to ionization of the carbinol. Table V gives the values of the rate constants at different acidities. The plot of the logarithm of the rate constants against acidity functions Ho and H R are shown in Figs. 8 and 9 respectively. TABLE V Oxidation of Benzhydrol with Permanganate in Aqueous Sulfuric Acid. Variation of Pate Constants with Acidity. -HQ (123) -HR (121) 103 k sec" 1 log k 0.30 0.70 1.07(*) -2.997 0.50 1.10 1.21 -2.916 0.69 1.40 2.00 -2.699 0.81 1.55 3.11 -2.507 0.98 1.80 4.23 -2.374 1.00 1.90 5.52 -2.258 1.22 2.25 8.63 -2.064 1.67 3.20 ' 18.97 ' -1.722 2.05 4.00 70.40 -1.152 2.46 4.90 248.40 -0.605 Not included in the plots in Figs. 9 and 10, because of the poor kinetics due to contribution from second order reaction. LGG K -3.200 -2.800 -2.1100 -2.000 -1.600 • -1.200 -.800 o ol From a comparison of Fign.l? and 9, i t i s apparent that the operative aciditv function i n the range studied i s H . o (i i ) Chromic Acid Oxidation : The rate of oxidation of benzhydrol increases with increasing acidity of the medium i n the range of 0.5% to 37.0% sulfuric acid i n 80 wt% aqueous acetic acid. The variation of the second order rate constants with acidity i s plotted i n Figs. 10 and 11 using the acidity functions H 0 and H R respectively. The data for the plots are given i n Table VT. TABLE VT Oxidation of Benzhydrol with Chromic Acid i n 80 wt% Aqueous Acetic Acid. Variation of Rate Constants with Acidity. H o HR k 2 l.nole ^ sec 1 log k. 0.50 0.088 -1.060 0.25 0.165 -0.783 -0.02 0.275 -0.560 -0.54 -2.15 0.740 -0.130 -1.12 -3.35 3.930 0.594 -1.70 -3.40 9.460 0.976 -2.00 -3.80 16.50 1.218 -2.15 -4.00 19.80 1.297 -2.70 -4.85 39.00 1.591 -3.00 -5.30 68.50 1.836 -3.30 -5.80 90.20 1.955 -3.40 -5.95 102.70 2.012 -3.60 -6.35 168.50 2.227 -4.20 -7.25 407.00 2.60S 64 Both Figs. 1 0 and 1 1 show a bend i n the i n i t i a l straight li n e . T.t appears that at acidities higher than 1 0 % sulfuric acid i n 80 wt% aqueous acetic acid the rate of increase of reaction rate with acidity i s lower than that at lower acidities. However, from a comparison of the i n i t i a l slopes of the two plots, i t seems that the acidity function H Q gives a better correlation with the logarithm of rate-constants than HR. Substituent Effect: (i) Permanganate Oxidation : The effect of different ring substituents on the rate of oxidation of benzhydrol was studied at pH 7.00 and 1 . 9 9 , and HQ - 0 . 8 1 . The measured rate-constants are given i n Table VII. TABLE VII Oxidation of Substituted Benzhydrols with Permanganate i n Aqueous Sulfuric Acid. T = 2 5 ° HAMMETT PLOT. ( 1 ) pH 7.00 Substituent a ( 1 2 4 ) a + ( 1 2 4 ) k 2l.mole" 1sec" 1 log K H 0 0 7 . 0 0 0.845 4-Me - 0 . 1 7 - 0 . 3 0 6 9 . 5 0 0.978 4-00 2H 0 . 1 3 2 0 . 1 3 2 5 . 8 5 0.766 (2) pH 1.99 Substituent a(124) H 0 4-Me -0.17 4-00^ 0.265 (3) Ho -0.81 Substituent a(124) H 0 4-Me -0.17 4,4*-diMe -0.34 4-C1 0.226 4,4'-diCl 0.452 4-OMe -0.268 4,4,-diCMe -0.536 2,4,6,2J4}6'-hexaMe 1.40 (* 4-CCLH 0.265 a+(±24) k2l.raole~1sec""1 log k, 0 21.0 1.322 -0.306 28.5 1.455 0.265 13.6 1.125 a + (124) 10^ sec" 1 log k 0 3.11 -2.507 -0.306 6.22 -2.207 -0.612 14.55 -1.837 0.112 2.42 -2.616 0.224 1.46 -2.834 -0.764 17.90 -1.747 -1.528 98.50 -1.007 ) -1.29 (**) 55.60 -1.253 0.265 1.43 -2.845 (*) Calculated from a comparison of pK values of benzoic acid and 2,4,6-trimethylbenzoic acid (125). (**) Calculated from a comparison of pK^. values of diary lcarbinols (121). 66 The rate constant of the oxidation of 9-fluorenol was found to be 3.09 x 10 Jsec , which i s very close to that of benzhydrol. Eecause of the similar structure of 9-fluorenol and benzhydrol and their similar pK^ + values { -13.3 and -14.0 (121) \ , i t can be anticipated that the rates of ionization of these compounds would also be similar. The close agreement between their permanganate oxidation rates suggests that the ionization of the carbinols i s the rate determining step i n the permanganate oxidation of these compounds. Further evidence on this point w i l l be discussed i n a later section. The Harcmett plots are sham i n Figs. 12, 13 and 14. At lower acidities the operative substituent constant i s a, which gives a better straight line when plotted against the logarithm of rate-constants than when a i s used. The plot i s sham i n Fig. 12. The substituted benzhydrols, other than 4-methyl and 4-carboxyl substituted ones, have very lav sol u b i l i t i e s i n water and thus they were found to be unsuitable for studying the rate of oxidation i n this medium. At higher acidity, the operative substituent constant i s o*+, as shown by a comparison of Figs. 13 and 14, obtained by plotting the logarithm of the rate-constants against a and a respectively. The values of a and a + for the hexamethyl substituted compounds are of opposite sign and this gives a clear indication of the operative substituent constant i n this case as well as i n the case of chromic acid oxidation of diarylcarbinols, discussed later. 67 Fig. 12. Permanganate Oxidation of SIGMA Fig. 13. Permanganate Oxidation of Diarylcarbinols, HMMETT PLOT. II = -0.81 o T = 25 CO CM CD ID LO i CM cn. i 1.1 -.6 -,35 -.1 .15 .65 .9 5IGMR •t-15 70 (ii) Chrome Acid Oxidation : The rate constants of oxi-dation of substituted benzhydrols with chromic acid at H Q -1.70 are given + in Table YTII. Figs. 15 and 16 show the Hammett plots, using a and a respectively. TABLE VIII Oxidation of Substituted Benzhydrols with Chromic Acid in 80 wt% Aqueous Acetic Acid. T = 25° HAMMETT PLOT H Q -1.70 + -1 -1 Substituent a (124) a (124) l^l.mole sec log H 0 0 9.46 0.976 4-Me -0.17 -0.306 11.20 1.049 4,4'-diMe -0.34 -0.612 13.70 1.132 4-C1 0.226 0.112 6.60 0.820 4,4'-diCl 0.452 0.224 4.17 0.620 2,4,6,2'4«6}-hexaMe 1.40 (*) -1.29 (**) 1.63 0.212 (*) and (**) : Calculated in the same way as in Table VII. It was not possible to study the chromic acid oxidations of 4-rtethoxybenzhydrol and 4,4'-dimethoxybenzhydrol under the same condi-tions as the other carbinols, since the methoxy group apparently forms 7 3 a colored complex with chromium (VI) , The complex was found to have a considerable absorbance at the wavelength used for kinetic studies, i . e . , 350 my. A similar complex was also found to be formed when c±irartium(vT) was added to an acetic acid solution of anisole. From the two Hammett plots shown i n Figs. 15 and 16, i t i s apparent that the better linear relation i s given when the substituent constant a i s used. Isotope Effect: (i) Permanganate Oxidation : Benzhydrol-a-d was oxidized at different acidities ranging from pH 7.00 to HQ -1.22. The observed deuterium isotope effects are given i n Table IX. TABLE IX Oxidation of Benzhydrol with Permanganate i n Aqueous Sulfuric Acid. T = 25° Deuterium Isotope Effect. • f l C i d i 1 y k K H ) A l ( D ) ^ ( H ^ D ) pH 7.00 - 7.3 pK 1.99 - 7.1 H 0.30 2.46 2.25 o H D - 0 . 0 1 1.20 1.08 H Q - 0 . 5 0 1.08 H Q - 0 . 8 1 1.09 -HQ - 1 . 2 2 1.08 74 Since in the region between H 0.20 and - 0 . 5 0 the kinetics are neither f i r s t nor second order, the isotope effects were, calculated in both ways, i.e. from the estimated f i r s t and second order rates. (ii) Chromic Acid Oxidation : Benzhydrol-os-d was oxi-dized with chromic acid at HQ - 1 . 0 0 ; the isotope effect at 2 5 ° was found to be 6 . 8 1 at this acidity. Activation Parameters; (i) Permanganate Oxidation : The activation parameters of permanganate oxidation of benzhydrol atpH 1 3 . 0 , 7 . 0 , 1.5 and H Q - 2 . 0 5 (in sulfuric acid) were determined by studying the reaction over a wide range of temperature. The enthalpies and entropies of activation were obtained from plots of log(k/T) against 1/T. Fig. 1 7 shows the plot for the reaction at H Q - 2 . 0 5 . Table X lists the values of the activation parameters at different acidities. TABLE X Oxidation of Benzhydrol with Permanganate. Activation Parameters. Acidity Reaction Order AH* (kcal/mole) AS* (e.u.) pH 1 3 . 0 0 Second 5.6 - 3 8 . 6 pH 7.00 Second 5.7 - 4 4 . 0 pH 1.50 Second 5.5 - 4 2 . 0 H Q - 2 . 0 5 First 1 7 . 5 - 1 3 . 5 75 CD I I p i . I CD' o i i i Fig, 17. Permanganate Oxidation of Benzhydrol. Variation of Rate Constant with Temperature. H = -2.05 o = 17.5 kcal/mole. = -13.5 e.u. 3*12 3.2 ~T~ 1 3.28 3.36 1/T ()(10 3 ) 3.44 3.52 76 The values of activation parameters obtained at pH 13.0 are i n good agreement, with the values of A H (5.7 kcal/ mole) and of AS (-33.4 e.u.) found by Stewart (4). The reaction at HQ -2.05 i s of f i r s t order and as w i l l be discussed later, the rate determining step i n the reaction i s the ionization of the carbinol. Hence the activation para-meters obtained for this reaction are actually those for the ionization of the carbinol rather than for the reaction with permanganate. The rate constants at H 0 -2.05 at different temperatures are given i n Table X(a). TABLE X(a) Permanganate Oxidation of Benzhydrol. Variation of Rate Constant with Temperature. o —1 Temp. °C 10Tc sec 20 3.00 25 7.04 35 18.10 45 34.00 (ii) Chromic Acid Oxidation : The activation parameters of cliromic acid oxidation of benzhydrol were determined 77 for the reaction at H -0.90. Fig. 18 shows the nlot of log(k/.n) against 1/r. The values of All and AS obtained are 5.9 kcal/mol and -37.1 e.u., respectively. The rate constants at different temperatures are given i n Table X(b). TABLE X(b) Chromic Acid Oxidation of Benzhydrol. Variation of Rate Constant with Temperature. Temp. °C k 2I .n iD le~ 1sec~ 1 15 1.50 25 2.20 27 2.44 43 4.11 Ionization of 4-Methoxybenzhydrol. Since carbonium ion formation seems to be an important part of the oxidation route of dLaxylcarbinols (as w i l l become apparent i n the Discussion), i t was decided to study the rate at which ionization of the carbinols to carbonium ions occurs. The rates of ionization of 4-methoxybenzhydrol were determined i n 70 vol% aqueous methanol containing sulfuric acid ranging from 1.8 to 3.6M. The reaction was found to be of f i r s t 79 order i n carbinol concentration. The following f i r s t order rats expression was used to obtain the rate constants : kt = 2.303 log A < x A« - At where . A« = absorbance at i n f i n i t e time , A = absorbance at time t . Since A« i s a constant for a particular reaction, log(A«* - A^was plotted against time to obtain a straight line. The rate constant was calculated from the slope of the li n e . A typical rate plot i s shown i n Fig. 19. In the range of acidity studied, the rate of ionization was found to be acid catalyzed. The variation of rate constant for ionization with acidity i s shown i n Figs.20 and 21, using the acidity function H Q and H R respectively. The data for the two plots are given i n Table XI. '•' : TABLE XI ' Ionization of 4-Methoxybenzhydrol. Variation of Rate Constant with Acidity. T = 25° - Ho -%(*•) ,'• .-%(**) 1 0 ^ sec" 1 •:' log k 0.15 0.58 0.63 3.22 -2.492 0.38 0.98 •:' 0.39 ; •; 5.99 -2.223' 0.50 1.25 1.02 . 7.94 -2.100: 80 F i g . 19. Ionization of p^fethoxylxinzhydrol i n 70 v o l % Aqueous Methanol Containing S u l f u r i c Acid. I I 2 S 0 4 = 2.16M H = -0.38 o T = 25° p-"fethoxybenzhydrol = 6. 0*;1. 1 / -3 -1 k = 5.99 x 10 sec ~l : 1 100 200 t in sec Fig. 21. Ionization of p-Methoxybenzhydrol. Variation of Rate Constant with Acidity Function II,,. T = 25° 0 0 r o Slope = -0.48 Coeff. of Correlation = 0.990 -2.000 -1.800 -1.600 - l . t o a -1.200 -l . o a a RCIDITY FUNCTION H R -,8oa -.600 83 TABLE XI (Cont.) -HQ "HR(*) -%(**) 1 0 ^ sec" 1 log k 0.64 1.65 1.78 10.60 -1.975 0.82 1.94 2.01 15.90 -1.799 (*) Determined by using 4,4'-dixnethoxv'txiphenylchloroirethane as indicator (Table TV, p.49). (**) Calculated from A* values, a typical calculation i s shown below : ~ In the solution containing 2.88M sulfuric acid i n 70 vol? aqueous methanol, the f i n a l absorbance of the carbonium ion (4-ai 30-C 6H 4)-C-C 6H 5, A<* i s 0.040. The acidity function H R i s H defined as (121) : = pKR+ - log CR+ CPDH CR+ = AVe where e, the extinction coefficient of the carlxmium ion i n the 4 -6 medium used, was found to be 3.5 x 10 . Thus, C + = 1.14 x 10 R moles, l i t r e \ CROH = I n^ t^- al concentration of the carbinol - CR+ (negligible) = 1.50M 84 — = 7.60 x 10"6 CROH leg _ = -5.12 pK^r = -7.90 (121) Hence, H R = -7.90 + 5.12 = -1.78 From a comparison of the slopes of the lines and the correlation coefficients shown i n Figs. 20 and 21, i t i s evi-dent that the operative acidity function for the reaction under investigation i s HQ. With this function a unit slope i s obtained with a correlation coefficient of 0.997. The H^ plot, on the other hand, yields a slope of -0.48 and a correlation coefficient of 0.990. 35 Discussiat It has been shown by Stewart (4) that the permanganate oxidation of benzhydrol in basic, neutral and weakly acidic regions is of second order, f i r s t order with respect to each of the reactants. The kinetics show the presence of hydroxyl ion catalysis in the basic region. In the present investigation the permanganate oxidation of benzhydrol and its derivatives in strongly acidic region is studied. The product analysis shaved that the conversion of the diarylcarbinols to the corresponding benzophenones is quantitative, and hence the stoichiortetry of the reaction is : 3Ar CHOH + 2MnC-T + 2H+ 9~3Ar.CO + 2MnC- + 4HL0 2 1 2 I I In addition, the permanganate oxidation of benzhydrol has been shavn to be acid-catalyzed, with the rate of this reaction being directly proportional to the carbinol concentration but independent of permanganate concentration. Thus i t appears that the acid-catalysis is not due to the formation of permanganic acid. It is likely that the increase in rate that accompanies a rise in acidity is due to the ionization of the carbinol and that this is the rate-determining step in these reactions. Furthermore, 86 the small isotope effect, k H A D = 1.08, shavs that the cleavage of the carbon-hydrogen bond does nor occur i n the rate-determining step. The rate of ionization of 4-iTethuxybenzhydrol has been found to be almost equal to the rate of oxidation of the carbinol at the same acidity ( k o x i d a t i o n = 1.79 x lO^sec" 1, k i o n i z a t i o n = 1.59 x 10- 2sec _ 1 at H Q = -0.81). Unfortunately, i t was not possible to use the same medium i n the study of both oxidation and ionization reactions. The aqueous medium used for the oxi-dation would not dissolve sufficient carbinol to enable the ionization study to be made and the aqueous methanolic system, used for the l a t t e r , was subject to oxidation by permanganate. However, the rates of the two reactions were compared where the two media, namely aqueous sulfuric acid and 70 vol% aqueous methanol contain-ing sulfuric acid, have the same values of the acidity function HQ. Since the two rates at the same acidities were found to be reasona-bly close to.each other, i t suggests that the ionization of the carbinols i s the rate-cetemining step i n the permanganate oxida-tions. Moreover, both oxidation and ionization reactions of the diarylcarbinols under investigation folia-; the acidity function H0. Since pKp+ values of the carbinols indicate that a much higher aci-dity i s required for a considerable extent of carbonium ion formation •than that used as the reaction media i n this investigation, i t 87 is probable that the carbinol undergoes a rate-controlling scission.of the probonated carbinol. The steps can be represented as -: ANCHOR + H30 :Ar2CHOH2 + H20 slew Ar2CHOH2 - [ A ^ C H .... 0H 2] fast -*^ Ar QI + H 0 2 2 Transition state fast fast Ar2CH + Mn04 A^aiQMnO-j Ar 2C = 0 + Mn03 + H The rapid decomposition of the ester may proceed as follows Ar Ar Ar CHOMnO. 6+ 5-T Q - — « M n O + C = OH + MnO Ar Ar H Ar Ar Ar + C = OH = O + H Ar .Ar 88 The disproporticrates to give MnC^  and MnO~. The manganate ester decomposition takes place after the rate-determining step and i t i s not possible to ascertain from the rate studies whether the hydride s h i f t i s 1,2-shift, as shown in I , or a 1,4-shift (see.later discussion on the analogous chromic acid reaction, p.92). . I t has been shown that the operative acidity function in this region i s H q. This i s i n agreement with the fact that the protonated alcohols, similar to oxonium ions, are involved in the rate-determining step. • I t i s apparent from the above mechanism that the presence of electron-donating ring substituents w i l l f a c i l i t a t e the reaction by s t a b i l i z i n g the positive centre of the cation. This-is supported by the observed value of p = -1.02. The operative substituent constant i n this reaction i s a which shows that the positively charged reaction centre i s stabilized by the electron-donating groups through resonance. I t i s usually found that the substituent constant a+ i s operative i n those cases where the reaction follows the acidity function H R rather than H Q. However, Nishida (126) has shown that the solvolysis 89 of monosubstituted benzhydryl chlorides also follows the substituent constant 0 + under conditions where f u l l ionization to carbonium ion is not possible, and hence H R cannot be the operative function. In the light of this observation of Nishida, i t appears that the proposed mechanism of the present investigation, which follov/s the acidity function HQ, is not in conflict with the observation that the opera-tive substituent constant in this case is a +. In neutral and weakly acidic regions, i.e., at pH 7.00 and 1.99 respectively, the reactions are of second order, f i r s t order in each of the reactants. The operative substituent constant in these cases is a. It has been shown that the reactions are almost indepen-dent of pH in these regions (4). In general, an alcohol molecule can ionize in two ways; in basic systems, the ionization leads to the formation of an anion : OH" P^ CHOH *-R2CH0 + H20 whereas in acidic medium, protonation, sometimes followed by carbonium ion formation, takes place : H + p^CHOH - RJCHGHJ »~ R2 C H + H2° In neutral and weakly acidic regions, neither of the two types of ionization can take place to an appreciable extent, and hence in these regions the permanganate-alcohol reactions are very slaw. A 90 hydride abstraction from the neutral alcohol molecule can be suggested for the reaction : • -. Ar 2C • >• Ar 2C = OH + HMn04 >• Ar 2C = 0 + H20 + Mn03 4 The kinetic isotope effects at pH 7.00 and 1.99 were found to be 7.3 and 7.1 respectively; this suggests the cleavage of carbon-hydrogen bond is the rate determining step. The Hammett p values for the reaction at these two acidities were found to be -0.72 and-0.78 respectively. These values are close to the observed p values for hydride abstraction reactions (86, 89, 94b, 127). . Between the acidities of H Q 0.40 and -0.50, the kinetics change from second order to fi r s t order and this is acccmpanied by a gradual change in the kinetic isotope effect (Table IX). The activation parameters, listed in Table X, reflect the changes in the reaction rates at different acidities. However, the values of AH (17.5 kcal/mole) and AS (-13.5 e.u.) for this reaction are somewhat different from those of the solvolysis reactions of benzhydryl chloride and bromide in acidic media, reported by Winstein and others (127a). These are also believed to proceed by Sjgl mechanism and have higher positive values of AH (~20 kcal/mole) and lower negative values of AS (-6 to -10.2 e.u.) than those found 91 in the present investigation. This difference may be due to the difference in solvent and leaving groups. In the work of Winstein et al . the leaving group is halogen, whereas in the present case i t is a water molecule. The carbon-halogen bond is more polar than carbon-oxygen bond. The work of Winstein et al. (127a) also reveals a change in AS when the leaving group is changed from chlorine to bromine; in the latter case a less + negative value of AS is obtained as the bond polarity is in-creased. In ethanol AS for benzhydryl chloride solvolysis is -7.9 e.u. and in the case of benzhydryl bromide i t is -2.6 e.u. Robertson has also reported activation parameters for solvolysis reactions of a large number of halides (127b). The results show a gradual decrease in the negative value of AS as the leaving halogen changes from fluorine to iodine. For example, the AS values for hydrolysis of methyl fluoride, methyl chloride, methyl bromide and methyl iodide in water are -26.2 e.u., -12.3 e.u., -10.1 e.u. and -8.1 e.u. respectively. The chromic acid oxidation of benzhydrol has been found to be of second order, f i r s t order in each of the carbinol and chromium (VI). The product analysis showed the formation of only one product, the corresponding benzophenone, as in the case of perman-ganate oxidation. The stoichiometry can be represented as : 3Ar2CHOH + 2Cr04_ + 10H+ -3Ar2CO + 2Cr 3 + + 8H20 92 Tha kinetic isotope effect, ky/kj-, = 6.81, obtained by using benzhydrol- a-d, shows tlie cleavage of carbon-hydrogen bond in the rate-determining step. The reaction has been found to be acid-catalyzed, and a better relationship is shown between the rate constant and the acidity function Hp, rather than HR, although a sharp change in slope is observed at HQ = -1.12. A similar change has been observed in the chromic acid oxidation of other secondary alcohols (44,91). According to previous workers, the mechanism of chromic acid oxi-dation of secondary alcohols involves the formation of a chromate ester intermediate, which decomposes unimolecularly to products (44,76,89,91,128) via the cyclic mechanism (see p.22, in "Intro-duction") : R,C" Cra>H ^F~c' C^rO-H^  R^~C = 0 *2 X f 2 -2 x / 2 2 <2 H 0 H~- 0 + II trcro + 3 3 At higher acidities diprotonation of the chromate part of the ester(II) occurs, and this would cause a decrease in rate (as observed with some secondary alcohols (44,91)), since there is no unprotonatsd oxygen atom available for participation in the hydroge transfer in (III) : 93 III It may be, however, that the intermediate chromate ester decomposes through the cyclic transition state (IV), by a 1,2-hydride shift : + 5 + 6 + + Ar 2C-—0—CrOjIL, «-Ar2C 0 C r 03 H2 ^Ar2C=OH \ H H + H 2C±0 3 IV + H2° + Ar 2C=OH — fc-Ar2C=0 + 1^ 0 The 1,2-hydride shifts occur with great ease in many rearrangement reactions, like 3aeyer-Villiger oxidation of aldehydes (128a).' The usual products are carboxylic acids, formed by hydride migration. Formate esters are formed only when the mobility of an aryl group is strongly increased by electron-releasing substituents (128b). The following mechanism has been proposed for the oxidation of carbonyl compounds to esters, or acids by means of peroxy acid (128c) : 91 OH R2C = OH ACC-O-OH 0—0—COA H H •sO—H"-0 £-A 0 The probability of a 1,2-hydride shift being the mode of oxidation of secondary alcohols was suggested by the results of the study; of oxidation of triarylcarbinols, to be discussed in Section II. In these cases, the evidence suggests that 1,2-aryl shifts occur. an aryl group for a 1,2-hydride shift agrees with the observations of the present investigation, where no detectable aryl migration, leading to the formation of aldehyde as the oxidation product of diarylcarbinols, was observed. aldehydes and the reaction under the present investigation seems reasonable, since in both the cases there is a migration of hydride ion or aryl group from carbon to the adjacent oxygen atom carrying positive charge. On protonation of the chromate part of ester (II), the electron attracting property of chromium increases, resulting in an The greater mobility of hydride ion compared to that of A comparison of Baeyer-Villiger oxidation of ketones and 95 increase in the rate of oxidation. At higher acidities, protonation of the central oxygen atom might occur, which causes a slower increase in rate, as shown in Fig. 10. Thus protonation affects the rate of oxidation in the same way as in the case of other alcohols investigated, and discussed before (44,91), i.e., by reducing the hydrogen abstracting property of the oxygen. According to the proposed mechanism electron-donating ring substituents are likely to enhance the rate of oxidation. The negative value of p = -0.54, obtained from the Hammett plot using substituted diary lcarbinols, is in good agreement with the mechanism. Also, the value of the entropy of activation for + the chromic acid oxidation of benzhydrol, AS = -37.1 e.u. is within the range for other secondary alcohols, i.e., -35.7 to -42.0 e.u., reported by D.G. Lee (129). The value for AH for the present reaction was found to be 5.9 kcal/mole, whereas those reported by D.G. Lee are between 8.15 to 10.38 kcal/mole (129). SECTION II OXIDATION OF TPJ^RYLCARBnraLS 96 EXPERIMENTAL Materials ; Triphenylcarbinol was crystallized from petroleum ether to constant melting point 164-5°. The following triarylcarbinols were prepared by Grignard reaction from available starting materials, as described below : Tri(4-tolyl)carbinol : from 4,4'-dimethylbenzophenone and 4-tolyL^gnesiurdbroru.de; y i e l d 40%, m.p. 95°, l i t . 96° (130). 4,4-Dimethyltriphenylcarbinol : from 4,4'-dimethylbenzophenone and phenylmagnesiumbroriu.de; yi e l d 59%, m.p. 76°, l i t . 76.5 - 77.5° (131) . 4-Methyltriphenylcarbinol : from 4-rtethylbenzophenone and phenylmagnesiumbromide; yi e l d 56.7%, m.p. 60 - 61°, l i t . 61 - 62° (132) . 4-Chlorotriphenylcarbinol : from 4-chlorobenzophenone and phenylmagnesiumbraTu.de; yi e l d 52%, m.p. 85°, l i t . 85° (133). 4,4'-Dichlorotriphenylcarbinol : from 4,4'-dichlorobenzophenone and phenylRiagnesiumbromide; y i e l d 49%, m.p. 86.5°, l i t . 87 - 88° (134). Tri(4-chlorophenyl)carbinol : from 4,4'-dichlorobenzophenone and 4-c±ilorophenylrnacjiesiumiodide; yie l d 70%, m.p. 93°, l i t . 93.5 - 94° (121). 97 T r i (3-chlorophenyl) carbinol : from 3-chlorophenylmagnesium-iodide and ethyl carbonate; y i e l d 35^, the product after p u r i f i -cation was found to be an o i l which tends to decompose on heating, and hence i t s boiling point could not be ascertained. The ultra-v i o l e t spectrum of concentrated sulfuric acid solution of the o i l showed a maximum at 412 my, e^^ = 2.92 x 10 4, l i t . 2.95 x 104 (121). Analysis : calc. C, 62.78%; H, 3.58%; C l , 29.35%; found C, 62.69%; H, 3.60%; C l , 29.40%. 4-Fluorotriphenylcarbinol : from 4-fluorobenzophenone and phenylmagnesiumbrc8ra.de; yield 53%, m.p. 122°,lit. 121 - 122° (135). T r i (4-biphenylyl) carbinol : from ethyl carbonate and 4-biphenylyl-magnesiumbromide; y i e l d 41%, m.p. 211°, l i t . 212° (136). 3- Trifluoromethyl-4'-methyltriphenylcarbinol : from 4-methy1-benzophenone and 3-trifluorcmethylphenylnagnesiumbromide; yield 41%, b.p. 184°/0.8mm. Analysis : calc. C, 73.7%; H, 4.97%; F, 16.7%; found C, 73.5%; H, 5.11%; F, 16.9%. The following compounds were prepared by Mr. I.Beheshti i n these laboratories : 4-Nitrotriphenylcarbinol : following the method of Stames (114), y i e l d 70%, m.p. 96°, l i t . 97 - 100° (114). 3-Trifluoromethyltriphenylcarbinol : from benzophenone and 3-trifluorcmethylphenylmagnesiumbrcmide; yi e l d , 61%, b.p. 120°/ 0.2mm. Analysis : calc. C, 73.17%; H, 4.57%, F, 17.37%; found C, 73.14%; H, 4.87%; F, 17.14%. 98 Standard solutions of potassium permanganate and potassium dichromate were prepared in the same way as mentioned in Sec. 1. A standard solution of manganous acetate was prepared by dissolving a weighed quantity of manganous carbonate in acetic acid. Manganous carbonate was prepared by adding a solution of sodium carbonate to that of manganous chloride in water t i l l the precipitation of manganous carbonate was complete. The precipitate was washed with water, dried under vacuum and tested for its purity by dissolving a known quantity of i t in dilute sulfuric acid and titrating the solution with standard potassium permanganate solution in presence of fluoride ions, foliating Zvenigorodskaya's method (137). Kinetic Methods : The methods employed were essentially the same as those used in the case of the diarylcarbinols, described in Sec. 1. In the case of permanganate oxidations, the reactions appeared to speed up with time due to the consumption of permanganate by phenol produced during reaction (1). For this reason only the i n i t i a l slopes of the rate plots were taken into consideration. The solutions for chromic acid oxidation of triaryl-carbinols were prepared in the same way as the diary lcarbinols, the only difference was that the reactions were studied as pseudo fi r s t order reactions, using an excess of carbinol. This was done 99 in order to avoid the following complications, as mentioned by Cohen and Westheimer (45) : the chromic acid oxidation reactions consume hydrogen ions, and thus the concentration of these ions changes as the reaction proceeds. This consumption of hydrogen ions occurs due to the reaction of acetic acid with chrcmium(III) which is obtained as the reduction product of chromic acid : 2HCrO~ + 2R2CHOH +• 8 H + *>- 2 C r 3 + + 3R2CO + 8H 20 Cr 3 4" + 3H0Ac , »-rr(nan) + 3H+ In the presence of a large quantity of acid or buffer, the hydrogen ion concentration remains almost constant during the reaction. With low concentration of chroraLc acid, the change in acidity is very small; moreover, unlike the case of diarylcarbinols, oxidation of triarylcarbinols produces phenols (as shown later) which also consume chromium(VI) and thus increase the concentration of chromium(III) in the system. For these reasons, an excess of carbinol was used. Most previous workers have used similar method (44, 45, 90, 91). The true second order rate constants were obtained by dividing the observed rate constants by the i n i t i a l concentration of the carbinol. 2— •» The extinction coefficient of CrO^ j at 350mu is 3.33 x 10 , + ' 2 ' whereas that of (C^ Rv) ^ C at the same wavelength is 1.03 x 10 , hence 100 i t i s apparent that the change of concentration of (CgH^J^C has very small effect on the decrease i n absorbance at 350mu. A l l the other triarylcarbinols investigated had a similar absorbance at 350mu. Product Analysisj (i) Chromic Acid Oxidation : A weighed quantity (0.001 mole) of the carbinol under investigation was dissolved in 85ml of 80 wt% aqueous acetic acid in a 250ml Erlenmayer flask. To the solution 15ml of 99.9% sulfuric acid was added slowly and with constant stirring. This gave a solution containing approxi-mately 23% sulfuric acid. Next 0.10g potassium dichrcmate dissolved in 3ml of 80 wt% acetic acid in water containing 0.5ml of 99.9% sulfuric acid was added. A slight excess of chromic acid was used in order to ensure complete oxidation of the carbinol, since some of the oxidant is consumed by the phenols produced. The mixture was allowed to stand for several days t i l l the yellow color of the solution changed to green. After this time the mix-ture was diluted with water and extracted several times with ligroin. The extract was treated with dilute sodium bicarbonate solution to remove acid. Next the extract was washed with 10% sodium hydroxide solution to remove any phenol formed. The ligroin extract was evaporated to dryness and the infra-red spectrum of the residue was taken. In the case of the unsubstituted carbinol, i.e. t r i -phenylcarbinol, the residue of the ligroin extract was dissolved 101 i n a small quantity of ether and treated with 2,4-dinitrophenyl-hydrazine and the derivative was separated and purified according to Vogel (119). The 2,4-dinitrophenylhydrazone was weighed and i t s melting point was determined. For the unsymretrically substituted triphenylcarbinols the residue of the l i g r o i n extract was dissolved i n ether and transferred quantitatively to a volumetric flask, the capacity of which depended on the quantity of the product. The ether solution was analyzed by vapor-phase chromatography using "Aerograph: Autoprep modal A-700" equipped with a stainless steel column 5 f t . x 25 i n . (o.d.), packed with SE-30 (15%) on clironosorb W (45-60 mesh) The carrier gas used was helium. Column temperature was varied * according to the constituents of the samples. The following l i s t shows the maximum column temperature employed to analyze the pro-ducts from the different triarylcarbinols : Triarylcarbinol Column Temperature 4HVfethyltriphenylcarbinol 230° 4 r4 ,-Dimethyltriphenylcarbinol 230° 4-Chlorotriphenylcarbinol 200° 3- Trifluoromethyl- 200° -4 1-nethyltriphenylcarbinol 4-Nitrotriphenylcarbinol ' 225° 3-Trlfluoromethyltriphenylcarbinol 170° 102 The exact quantities of ea-±i component of the products were determined by measuring the areas of the peaks and comparing them with those obtained from known quantities of the corresponding compounds. The alkali washings of the ligroin extract were acidified with dilute hydrochloric acid and extracted with ether. The ether extract was analyzed for phenols by preparing the corresponding brono-derivatives, according to Vogel (119). Since phenols are known to react with chromic acid to some extent, only a qualitative estimate of their presence was possible, and this shaved the exis-tence of the predominating species only. Chrcmic acid oxidations were also carried out in the presence of added manganous ions. The products were isolated and analyzed in the same way as in the case; of chromic acid oxidation without any added salt. (ii) Permanganate Oxidation : Since both the expected products of oxidation of triarylcarbinols react with permanganate to a certain extent, an exact quantitative determination of the products was not possible. However, a qualitative estimate was made. A weighed quantity (approx. 0.0002 mole) of triarylcarbinol was dissolved in 40ml of 18.01M sulfuric acid in a 500ml Erlenmeyer flask, kept immersed in an ice-bath. Water was added slowly to the solution with continuous stirring until the final volume was 200ml. 103 Next 0.011M potassium permanganate solution was added to i t slowly until the solution attained a faint pink color, indicating that the carbinol had reacted completely. The products were extracted with ligroin and the ketones produced were identified as their 2,4-dinitrophenyIhydrazones, although in this process only the predominating ketones were identified, the minor fractions being eliminated during the process of purification. Moreover, benzophenones with different substituents react with permanganate at different rates, which made the quantitative estimation of the products more difficult. The identification of the predominating benzophenone was also done by dissolving a small quantity of the residue of ligroin extract in con-centrated sulfuric acid and observing the ultra-violet spectrum of the protonated species. The phenol fraction of the products was not possible to isolate and identify because of the rapid reaction of phenols with permanganate (1). 104 RESULTS Product Analysis; (i) Chromic Acid Oxidation : The 2,4-dinitrophenyl-hydrazone of the ketone produced from triphenylcarbinol melted at 238°, and i t was identified to be the derivative of benzo-phenone, m.p. 239° (119). The bromo derivative of the phenol produced melted at 94-5°, identified as the tribromo derivative of phenol, m.p. 95° (119). The results of vapor phase chrcmatographic analysis of the ketones produced from unsymmetrically substituted t r i -phenylcarbinols are shown in the following table. A typical chromatogram is shown in Fig. 22. TABLE XII Analysis of Chromic Acid Oxidation. Product of Triarylcarbinols. Substituted Benzophenones Triphenylcarbinols Molar Ratio of Benzophenones (a) Without Mn" (b) With Mn + + A. 4,4'-diMe (i) 4,4'-diMe (ii) 4-Me (ii)/ ( i ) (a) 6.60 (b) 6.55 105 F i g . 22. VPC Analysis of the" Chrcsriic Acid Oxidation Product of 4-Chlorotriphenyl-carbinol. Column : SE 30 T = 200° 106 Substituted Triphenylcarbinols B. 4-Me Benzophenones (i) 4-Me (ii) Unsubstituted Molar Ratio of Benzophenones (a) Without Mn (b) With Mri** (ii)/(i) (a) 1.70 (b) 1.72 C. 4-C1 (i) 4-C1 ( i i ) unsubstituted (ii)/(i) (a) 0.38 (b) 0.37 D. 4-NCv, (i) 4-N02 ( i i ) Unsubstituted (ii)/(i) (a)0.039 (b)0.031 E. 3-CF3 (i) 3-CF3 (ii)/(i) (a) 0.18 (ii) Unsubstituted (b) 0.17 F. 3-CF3-4'-CH3 (i) 3-CF3 (i) : (ii) : (iii) (ii) S-CF^ ' - a i , (a) 3.1 : 1 : 0.34 (iii) 4-CH- (b) 3.0 : 1 : 0.35 The migration aptitudes of the substituted phenyl groups compared to the unsubstituted ones were calculated from the molar 107 ratios of the different products. The calculations are shown below and the results are l i s t e d i n Table XIII. A. 4,4 ,-Dimethyltxiphenylcarbinol : The migrating groups are two p-tolyl and one phenyl; hence the probability of formation of 4-methylbenzophenone due to the migration of a p-t o l y l group i s twice as much as the probability of formation of 4,41-dirnethyl-benzophenone which i s the result of migration of a phenyl group. The apparent migration aptitude of p-tolyl compared to phenyl group, obtained from the molar ratio of the corresponding product i s 6.6, and on correction by the s t a t i s t i c a l factor of . 2, following the above arguments, i t becomes 3.3. B. 4-Jfethyltriphenylcarbinol : The migrating groups are two phenyl and one p-t o l y l . Following the same arguments as i n the case of "A", the apparent migration aptitude of p-tolyl compared to phenol should be multiplied by the s t a t i s t i c a l factor of 2, and thus i t becomes 1.7 x 2 = 3.4. C. D and E. 4-Gilorotriphenylcarbinol, 4-nitrotriphenyl-carbinol and 3-trifluoromethyltriphenylcarbinol : In a l l these cases there are two unsubstituted and one substituted phenyl group that migrate. As i n the case of "B", the migration aptitudes are corrected by multiplying the apparent values, obtained from the molar ratios of the products by the s t a t i s t i c a l factor of 2. F. 3-Trifluoromethyl, 4'-methyltriphenylcarbinol : In this case there are three different migrating groups, hence there 108 i s no s t a t i s t i c a l correction required and the migration aptitudes were directly obtained from the molar ratio of the products, i.e. the quantity of the disubstituted benzophenone produced was directly proportional to the migrati.on aptitude of the unsubstituted phenyl group. The values for the other groups were similarly proportional to the quantities of the two other products. The addition of manganous ion was found to have very l i t t l e effect on the relative migration of the aryl groups. The results obtained were very close to those obtained without the added ion, and hence separate calculations were not done for these. TABLE XIII Migration Aptitudes of Aryl Groups for Chromic Acid Oxidation of Triarylcarbinols. Substituent X o+ (124) Mig.Apt. X-C6H4/CgH5 Log (Mig.Apt.) H 4-Me 4-C1 3-CF3 0.0 -0.306 0.112 0.415 1.0 3.3 3.4 Av.. 3.35 (3.1*) 0.76 0.36 0.0 . 0.525 -0.143 -0.444 (0.34*) 4-NO. 0.778 0.078 -1.102 109 * - Values obtained from "F", 3-trifluorornethyl, 4' -methyltri-phenyicarbinol, Table XIII. Fig. 23 shows the plot of the logarithm of the migra-tion aptitudes against the corresponding a+ values. The slope of the line is -1.44, which gives the value of p for the migra-tion. In a l l the cases discussed above, almost quantitative yields of the corresponding ketones were obtained. The bromo derivatives of the phenols produced were isolated and identified. The results are listed below : Substituted M.p. of the bromo deriva- Identified Triphenylcarbinols tives of products Phenols Obs. Unsubstituted 94-95° 4-Me 48.5° 4-C1 95° 4,4'-diCl 94° (ii) Permanganate 0>ddation : The products of oxida-tion of triarylcarbinols apparently consist only of ketones since the other possible product, phenols, are readily oxidized to carbon-dioxide and water by permangante (1). The evolution of carbon-dioxide from reaction mixtures has been observed and a Lit. (119) 95° Unsubstituted 49° 4-Me 95° Unsubstituted 95° Unsubstituted I l l quantitative estimate of the gas was made by passing the evolved gas into barium hydroxide solution and observing the increase in weight of the system. This shaved the production of six moles of carbon dioxide from each mole of triphenylcarbinol used. This indicates complete degradation of the phenol produced from the carbinol. The ketone from triphenylcarbinol was identified as benzophenone in the same way as in the case of chromic acid oxidation, although a quantitative yield was not obtained because of further oxidation of benzophenone by permanganate. With unsymmetrically substituted triphenylcarbinols the molar ratios of the ketones produced are rather inaccurate since the ketones with an electron-releasing group in the ring are destroyed by permanganate at a rate higher than those without. For this reason, a quantitative measurement of the migratory aptitudes of the different groups in this reaction could not be made. However, the spectrophotcmetric method and the iden-tification of the 2,4-dinitrophenylhydrazones of the ketones produced shaved that the ring containing the more electronegative group migrated preferentially. The results are given below : 112 TABLE XIV Analysis of Permanganate Oxidation Product of Triarylcarbinols . Substituted T r i - M.p. of 2,4-DNPH* phenylcarbinols of products Obs. Lit.(119) 4-Me 238° 4-C1 184° * DNPH Diriitrophenylhydrazone / The conjugate acid of benzophenone absorbs at 344 my (138). * # The conjugate acid of 4-dilorobenzophenone absorbs at 300 my and 367 my (138). Order of Reaction : (i) Chromic Acid 'Oxidation : As i n the case of the diary lcarbinols, discussed i n S e c I, a 3:2 ratio of the substrate to oxidant i s applicable and the reaction can be represented as : 3Ar3COH + 2H 2Cr0 4 »-3Ar2CO + 3ArOH + Cr 20 3+ 2H20 For the reasons mentioned earlier i n this section, the reactions were studied as pseudo-first order reactions, using an excess of the carbinols. Chromic acid oxidation of triarylcarbinols AJTBX O £  conc' Identified  H2SQ4 soln.of Benzophenones products. 239° 344 my7* Unsubstituted 185° 299 my, 367 my# 4-C1 113 unlike that of the cUarylcarbinolv,produces phenols, and this diromic acid. For this reason, only the i n i t i a l reaction (about 60%) was taken into account in t:e neasurement of rate. A typical pseudo f i r s t order rate plot is sha/m in Fig. 24. The true second order rate constant was obtained by dividing the observed f i r s t order rate constant by the i n i t i a l concen-tration of the carbinol. This method gave satisfactorily reproducible results over a wide range of i n i t i a l concentration. The following equation was used to calculate the rate constants : where JcarbinolJ o is the i n i t i a l concentration of the carbinol. (ii) Permanganate Oxidation : As in the case of the diarylcarbinols, a 3:2 ratio of substrate to permanganate is the stoichicmetry in this case, provided the i n i t i a l reaction pro-ducts undergo no degradation. The reaction can be represented as : When this stoichiometric ratio is used the plot of the logarithm of the absorbance at any time against time gives straight complicates the study because of their further oxidation by 2.303 x slope (obs.) 3Ar3COH + 2KMn04 +. H20 *-3Ar2CO + 3ArOH + 2K0H + 2Mn02 ro •r-< . I I i n i i Fig. 24. Chromic Acid Oxidation of Triphenylcarbinol. Typical Rate Plot. R^ = -3.55 Triphenylcarbinol = 7.44 x IO"4!-! Slope = 1.85 x 10~4 k 2 = 0.249 l.mole^sec" 1 "T 33.333 66-656 1 100.0 -1 133.333 166.S66 200.0 233.333 TtSEC) ( X 1 0 ) 7 115 line for approximately 50-60% reaction. A typical f i r s t order rate plot i s shewn i n Fig. 25. As i n the case of the diary 1-carbinols, discussed i n Sec. I, the rate of disappearance of permanganate i s two-thirds that of the carbinol, and hence the rate constant was calculated from the same equation as before : k = TJ: x 2.303 x slope For the permanganate oxidation of tri(p-tolyl)carbinol at higher acidities a second order rate equation was found to be suitable wheh the stoichiometric ratio (3:2) of substrate to permanganate was used. The rate data were found to f i t the second order rate expression (1) : 7 kt " [ VTIJ Mn 1 v r r Mn o Since Mnv L the permanganate concentration at zero time, i s a constant, a plot of t against l/j^Mn^ 1 1^ gives a straight l i n e , the slope being equal to 3k/2. The reciprocal o the absorbance of permanganate at any time was plotted against time, and i n this case the following equation i s valid : 2e k ~ — x slope 2 3 1 1 7 where e i s the molar extinction coefficient of Mn^11. A typical rate plo\ for the oxidation of tri(p-tolyl)carbinol i s shown i n Fig. 2 6 . Dependence of Rate on Acidity; (i) Chromic Acid Oxidation : The rate of chromic acid oxidation of triphenylcarbinol i n 80 wt% aqueous acetic acid containing sulfuric acid increases with increasing acidity of the medium i n the range of 8.3% to 3 9 . 6 % sulfuric acid. The variation of second order rate constant with acidity at 2 5 ° i s shown i n Table XV, and plotted i n Figs. 2 7 and 2 8 , using the acidity func-tions HQ and H R respectively. TABLE XV Chromic Acid Oxidation of Triphenylcarbinol. Variation of Rate Constant with Acidity. T = 2 5 ° - H Q - H R k 2 1. mole~ 1sec~ 1 log,k 2 1.12 2.80 0 . 0 3 1 - 1 . 5 0 6 1.15 3.20 . 0 . 0 7 8 - 1 . 1 0 6 1.70 3.55 0.249 - 0 . 6 0 4 2.00 3.95 0 . 5 4 8 - 0 . 2 6 1 2 . 2 5 4.10 0 . 6 4 5 - 0 . 1 9 1 118 121 TABLE XV (Cont.) k 2 l.mole~1sec~1 log k. 2.70 4.85 2.810 0.449 3.10 5.50 9.350 0.971 3.35 6.01 26.250 1.419 3.60 6.30 62.640 1.797 4.05 7.00 215.300 2.333 4.15 7.20 456.000 2.659 4.25 7.30 608.000 2.784 4.35 7.48 1023.000 3.010 In both Figs. 27 and 28, the plots of log k 2 against the two acidity functions give straight lines, but a comparison of the too shows a somewhat better correlation in Fig. 28, the slope -0.923 being closer to unity than -1.34. This suggests that an acidity function closely related to H R is operative. (ii) Permanganate Oxidation : The oxidation of triphenyl-carbinol by permanganate i s acid catalyzed in the range of 12.1% to 33.5% sulfuric acid in water. The concentration of permanganate acid is negligible in this region, and for this reason, as mentioned in the case of diarylcarbinols in Sec. 1, i t is likely that the increase in rate is due to the ionization of the carbinol. Table XVT lists the data for the variation of the rate constant with acidity in the region studied. .122 TABLE XTL Permanganate Oxidation of Triphenylcarbinol Variation of Rate Constant with Acidity . T = 25° - H Q (123) -HR (121) 103k sec" 1 log k 0.60 1.00 1.76 -2.755 0.67 1.15 2.04 -2.691 0.84 1.55 2.85 -2.545 1.00 2.00 4.14 -2.383 1.47 2.90 10.87 -1.964 1.93 3.75 31.05 -1.508 The plots of the logarithm of rate constants against acidity are shown i n Figs.29 and 30, using the acidity functions H q and EL^ respectively. Although both the plots gave reasonably good straight lines, only i n the f i r s t case was a line of slope close to unity obtained. This indicates that the operative acidity function i n this region i s HQ. In order to study the oxidation of a triarylcarbinol at acidities near and beyond i t s pKp+ t r i (p-tolyl) carbinol was investi-gated. Triphenylcarbinol has a much higher value of pKR+ than that of the trimethyl substituted ccmpound and i t s rate 'of oxidation i s too high to be measured at the acidities required to give a high degree of ionization. 125 At lower acidities the rate of oxidation of t r i (p-tolyl) -carbinol i s f i r s t order i n carbinol and zero order i n permanganate, as i n the case of triphenylcarbinol. As the acid approaches the concentration for inaximum ionization of the carbinol, the reaction changes to second order, f i r s t order i n each of the reactants, the stoichiometry remaining the same as before. Second order rate plots were obtained by using the 3:2 ratio of the substrate to permanganate. The rate of the oxidation was studied i n the range of 4.20M to 9.08M sulfuric acid. The data showing the variation of rate constant with acidity are given i n Table XVTI and plotted i n Fig. 31. TABLE XVTI Permanganate Oxidation of Tri(p-tolyl)carbinol. Variation of Rate Constant with Acidity. T = 25° -HR k 2 l.mole'^sec 1 log k 2 2.60 13.3 1.123 3.00 25.0 1.398 3.10 35.2 1.546 3.50 60.8 1.784 3.88 73.6 1.867 5.00 90.0 1.954 6.53 118.8 2.075 126 127 TABLE XV1T (Cent.) "*-R -1 6.80 120.1 2.080 7.20 121.0 2.083 8.00 158.6 2.200 8.20 246.0 2.391 Fig. 31 shows that the rate increases steadily with increasing acidity up to H R = -3.50, the slope of the plot of the logarithm of rate constant against the acidity function H R being very close to unity, and rises relatively slowly after t h i s . The pKp+ of the carbinol i s -3.56 (121). Assuming that ti o n curve, one can take the rate at f u l l ionization to be anti-log k 2 = 2.40. The rate at half-ionization w i l l be 2.08 - 0.30 = 1.78. This corresponds to an H R of-3.50, which i s very close to the reported pKR+ value of the carbinol. That i s , the rate constant at H„ = -6.90 i s i n good agreement with what i s expected from i t s value at H R = -3.50. However, the rates at the inter-mediate acidities are somewhat lower than what i s expected from an ideal ionization curve. the alterations i n rate up to H R = -7.0 are following an ioniza-128 Beyond H R = -6.90, the rate remains almost constant up to H R = -7.20, since the carbinol iy essentially f u l l y / ionized. At s t i l l higher acidities the rate rises again, pro-bably because of the formation of permanganic acid. Permanganate i s half-protonated i n 61% sulfuric acid (1), corresponding to H R = -9.20 (121) ,forming permanganic acid, which i s believed to be a stronger oxidizing agent than permangante ion. Substituent Effect: (i) Chromic Acid Oxidation : The effects of ring substituents on the rate of chromic acid oxidation of triphenyl-carbinol were studied. The effect was also studied i n presence of 3.045 x 10~\'l manganous ion, i n order to effect the induced oxidation of the added ion by chromium species of valence V and IV, formed as intermediates i n the reduction of chromic acid (53, 75, 99). A decrease i n the rates of oxidation of the car-binols was observed. Table XVTII l i s t s the rate constants of oxidation of triarylcarbinols with, and without, added manganous ion. The data are plotted i n Figs. 32 and 33, using the substi-tuent constants a and a+ respectively. 129 TABLE XVIII Chromic Acid Oxidation of Triphenylcarbinol. Effect of Substituents. Hp = -3.95 , T = 25? Substituent a(124) a+(124) 102k l.mole" 2 -1 -1 sec log k 2 Mn Mn 2 + 2+ Mn Mn2 + absent present* absent present' H' 0 0 54.81 37.77 -0.2611 -0.4229 4-Me -0.17 -0.306 108.21 71.39 0.0342 -0.1464 4,4'-diMe -0.34 -0.612 209.61 129.01 0.3214 0.1106 4,4)4''-triMe -0.51 -0.918 306.30 237.20 0.4857 0.3751 4-F 0.062 -0.073 59.88 41.21 -0.2227 -0.3850 4-C1 0.226 0.112 46.98 32.21 -0.3281 -0.4920 4,4'-diCl 0.452 0.224 39.84 -0.3997 4 , 4 J 4 " - t r i C l 0.678 0.336 28.56 20.73 -0.5443 -0.6834 3,3;3"-triCl 1.119 1.119 5.34 3.52 -1.2725 -1.4535 4-M02. 0.777 0.777 11.08 8.02 -0.9556 -1.1958 3-CF3,4'-Me 0.245 0.109 50.44 32.93 -0.2972 -0.4868 * 3.045 x 10 M manganous acetate added. J ' J O -1.8 -1.4 -1.0 CN CO o -0.6J 0.2H 0.2 Fig, 32. Chromic Acid Oxidation of Triary lcarbinols. •;; liAMT-ETT P L O T . 1 : Mn 2 + absent. .2+ 2 : Mn ..present.. II = -3.95 ' R T = 25° N \ V \ \ O \ \ \ \ s © \ \ O \ \ O \ \ \ o > -. . • A . A V SIGMA ;. -0.4 132 The two plots show that the better correlation between rate constant and substituent constant i s obtained when a i s used. From the two l ines i n F i g . 33 the values of p T for the oxidation reaction with and without added manganous ion are obtained; the former gives a value of -0.906 and the l a t t e r -0.879. ( i i) Permanganate Oxidation : The effects of different r i n g substituents on the reaction centre were studied. The rate constants of the substituted triphenylcarbinols are l i s t e d i n Table XIX. F i g . 34 shows the Hammett p l o t when the substituent constant o i s used and F i g . 35 when the substituent constant a + i s used. TABLE XIX Permanganate Oxidation of Triphenylcarbinol. Effects of Substituents. H Q = -0.65 , T = 25° Substituent 0(124) a+(124) 10 %L sec 1 log k H 0 0 20.70 -2.684 4-Me . -0.17 -0.306 58.10 -2.237 4,4'-diMe -0.34 -0.612 120.00 -1.921 4,4"4"-triMe -0.51 -0.918 448.00 -1.348 4-F 0.062 -0.073 26.90 -2.570 4-C1 0.226 0.112 12.90 -2.886 4,4'-diCl 0.452 0.224 8.11 -3.091 133 134 135 TABLE XIX (Cont.) Substituent a(124) a+(124) 104k sec" 1 log k 4,4,,4"-triCl 0.678 0.336 5.87 -3.231 3,3}3"-triCl 1.119 1.119 0.66 -4.181 3-CF3/4-r3e 0.245 0.109 10.90 -2.963 From a comparison of Figs. 34 and 35, i t appears that a better linear relationship is obtained when substituent constant a+ is used. The slope of the line gives the value of p +, which is -1.39. Activation Parameters ; The temperature effect on the rates of chromic acid and permanganate oxidation of triphenylcarbinol was studied. The plot of log(kA) against 1/T for the chromic acid oxidation is given in Fig. 36 and the same for the permanganate oxidation is given in Fig. 37. The results are given in the following Table. 13S 137 138 TABLE XX Oxidant Activation Parameters for Oxidation of Triphenylcarbinol. Acidity of the Medium Reaction Order AH* (kcal/mole) AS* (e.u.) Chromic Acid H R -3.95 Second Permanganate H -1.00 F i r s t 11.8 17.2 -25.1 -11.7 As i n the case of benzhydrol the permanganate oxidation reaction i s of f i r s t order i n carbinol and the rate-determining step i s the ionization of the carbinol, as discussed later. Thus i n thie case also the activation parameters calculated are those for the ionization reaction. The rate constants of the two reactions at different temperatures are given i n Tables XX(a) and XX(b) respectively. TABLE XX(a) Chromic Acid Oxidation of Triphenylcarbinol, Variation of Rate Constant with Temperature, Temp.°C 15 25 43 50 k 2 l.mole^sec"*1 0.24 0.55 . 1.44 2.10 139 TABLE XX (b) Permanganate Oxidation of Triphenylcarbinol Variation of Pats Constant with Temperature Temp.°C 103 k sec" 1 15 1.12 25 4.14 35 11.4 40 17.8 140 DISCUSSION The analysis of the products of permanganate oxidation of triarylcarbinols shov/ed that almost quantitative yields of diary1 ketones were obtained, although no phenol was detected. The formation of phenols during the reactions was inferred from the evolution of carbon dioxide from the reaction mixtures, quan-ti t a t i v e estimation of the gas showing the formation of one mole of phenol from every mole of carbinol. Since no other product was detected the following stoicliicmetry can be suggested for the reaction : 3Ar3COH + 2MnO~ + 2H+ 3 A r 2 C — 0 + 3ArOH + H20 + 2Mn02 As i n the case of the diary lcarbinols the permanganate oxidation of triphenylcarbinol was found to be acid-catalyzed and to be directly proportional to the substrate concentration but independent of the. concentration of the oxidant. As i n the case of the diarylcarbinols, discussed i n p.85, the increase i n rate i s believed to be due to the ionization of the carbinol, this being the rate-determining step i n the oxidation reaction. Recently, a direct observation of the ionization of a triarylcarbinol has been reported (139). The same rate law that was used i n the case of diarycarbinols was found to be suitable for the triarylcarbinols. 141 The negative value of P that i s observed, (P - -1.39) also supports the formation of a positive reaction centre i n the rate-determining step. The pK + of triphenylcarbinol i s -6.63 (121). There-fore at H R = -3.75, the maximum acidity used i n the present investigation, the ionization of the carbinol to the corres-ponding carbonium ion i s only 0.1%. This means that most of the substrate i s not present i n carbonium ion form i n the range of acidity employed. Furthermore, the operative acidity function within this rarae of aciclitv v;as found to be H and not Hp. o f- • Folia-ring the same arguments as i n the case of the diary lcar-binols, i t can be suggested that the permanganate oxidation of triarylcarbinols between 1.7M and 4.2M aqueous sulfuric acid proceeds through protonation and incipient ionization of the substrate; the steps can be represented as : Ar3COH + H30 slav Ar3COH2 Ar 3C Ar3COH2 + R^ C-OH. i •Ar3C + HO A r C + MnO, 3 4 fast Ar A r — ^ C £f 8-- O M n O . Ar H20 Ar 2C =OH + ArOH + MnO-Ar 2C==OH ^ A r 2 C =0 + H 142 As mentioned in the case of the diarylcarbinols, i t is not possible to distinguish between the two possible modes of aryl migration, namely 1,2- and 1,4-shifts, from the rate studies. How-. ever, as discussed in Section I, there are many evidences of 1,2-shifts, whereas no example of a 1,4-shift is available. From the above mechanism i t appears that the aromatic ring that migrates becomes the phenol. It has been observed in the cases of similar reactions, Like pinacol rearrangement (140), acid-catalyzed rearrangement of arylmethyl hydroperoxides (141, 142) etc., that the preferential migration of the ring containing the stronger electron releasing group occurs. A similar effect can also be expected in this case. This is supported by the product analysis which showed that the major fraction of the diaryl ketone formed contained the least electron-releasing group, indicating the migra-tion of the ring containing the strongest electron-releasing group. A quantitative estimate of the relative proportions of the ketones formed, as done in the case of chromic acid oxidation which is discussed later, was not possible, since the ketones were found to react with acidic permanganate to a significant extent. On increasing the acidity of the medium the order of reaction changes. This has been shewn by the experiments with t r i -(p-tolyl)carbinol. At higher acidity free carbonium ions are formed faster than their reaction with permanganate and thus, under such 143 conditions, the l a t t e r becomes the rate-determining step. The reaction then becomes one of second order, f i r s t order i n each of the carbinol and the oxidant. The PK r+ of t r i ( p - t o l y l ) c a r b i n o l i s -3.56 (121). The rate of oxidation was found to increase steadily up to H R = -3.50, which i s very close to the a c i d i t y required f o r the h a l f - i o n i z a t i o n of the carbinol. Beyond t h i s a c i d i t y the rate increases slowly and then remains f a i r l y constant up to H R = -7.2, above which the rate again r i s e s sharply as the ac i d i t y becomes s u f f i c i e n t l y high to form permanganic a c i d , which i s believed to be a stronger o x i d i s i n g agent than permanganate (24). This v a r i a t i o n i n the rate of oxidation with a c i d i t y c l e a r l y shows that the reaction proceeds tlirough the formation of carbonium ions at a c i d i t i e s near and beyond the pKp+ of the t r i a r y l c a r b i n o l , the rate-determining step being the decomposition of the manganate ester formed between the carbonium ion and manganate ion of the oxidant. The a c t i v a t i o n parameters f o r permanganate oxidation of triphenylcarbinol, l i s t e d i n Table XX, are s i m i l a r t o those f o r the permanganate oxidation of benzhydrol. This indicates that both the reactions proceed tlirough s i m i l a r mechanism. From the analysis of the products of chromic a c i d oxidation of t r i a r y l c a r b i n o l s the stoidiicmstry of the reaction was found to be : 3Ar3COH +'2H2Cr04 + 2H+ ^3Ar 2C=0 + 3ArOH + C r 2 0 3 + 2HgO+ 144 The chronic acid oxidation of triphenylcarbinol was also found to be acid-catalyzed and the operative acidity function in .' the range of 8.3% to 39.6% sulfuric acid in 80 wt% aqueous acetic acid to be K^ . The reactions were found to be of second order, fi r s t order in each of the reactants. There is evidence that the chromic acid oxidation of alcohols involves the formation of a chrcmate ester intermediate with the rate-determining step being the unimolecular decomposition of this intermediate (44,89-91, 128). Accordingly, the mechanism of the chromic acid oxidation of triphenylcarbinol can be represented as : 2H Ar^COH + HCrO $E>. Ar C 0 6+. Ar 0C Ar Ar„C + OAT ^s»--Ar2C = 0 + ArOH + H 145 1,2-shifts of aryl groups occur in the pinacol rearrangement of triphenylethylene glycol (140) and acid-catalyzed rearrangements of benzhydryl hydroperoxide (141) and triaryImethyl hydroperoxides (142). In the fi r s t case, the shift is from one carbon atom to another, the reaction may take either path (a) or (b), depending on the reaction conditons; this is discussed in a later section. In the latter two, the shift is from carbon to oxygen : (i) Ph2C CHPh OH OH OR Ph„C .OH OH Ph0C-Ph V + / H OH (a) (b) (ii) Ph2C H 0 OH H Ph. H" 6+ .0 -6+ •OH. (iii) Ph3C- 0 OH IT Ph2C; 6+ -.0-6+ •OH, 146 Baeyer-Villiger reactions are also examples of 1,2-shifts, ns mentioned in Section I. A 1,4-shift leading to a five-irembered ring in the transition state is not possible in any of th^ .se cases. Moreover, in a l l these cases, as in the present investigation, a preferential migration of the aryl group containing the stronger electron-donating substituent was observed. Further support for the 1,2-shift can be obtained from a comparison of the rates of chrcaiic acid oxidation of benzhydrol and triphenylcarbinol under identical conditions. It has been found that at the same acidity, the rate is higher for benzhydrol than for triphenylcarbinol; for example, at H Q = -1.12, the rate of oxidation of benzhydrol is 3.93 l.mole~""1'sec~1, whereas that for triphenylcarbinol at the same acidity is 0.31 l.mole "'"sec 1. This indicates that the 1,2-hydride shift is much faster than 1,2-phenyl shift, assuming for purposes of discussion that secondary alcohols are oxidized by chromic acid by such a mechanism. Collins has observed that in the pinacol rearrangement of triphenylethylene glycol a wide variation in the ratio of phenyl to hydrogen migration occurs, and this depends upon the acidity of the medium (140). In highly acidic medium preferential phenyl migration takes place, the ratio of phenyl to hydrogen migration being 7.33. In weak acid, such as in dilute sulfuric acid, hydrogen migration becomes pre-dominant and the ratio changes to 0.435. In strong acid, i t is 147 believed that carbonium ion (A) formation takes place. In this case the reaction proceeds through the path (ia), causing the preferred migration of phenyl over hydrogen, since the positive charge on carbon wil l attract the Tr-electton cloud of the ring more strongly than the hydrogen and consequently this effect overrides the steric effect, discussed below. Ph OH Ph (A) However, in less acidic medium, carbonium ion formation is negli-gible and i t is possible that a steric resistance to phenyl migra-tion might take place at the migration terminus and the smaller size of the hydrogen then permits its migration in preference to phenyl. The reaction then takes the path (ib). From the chromic acid oxidation of diarylcarbinols in the present investigation, discussed in Section I, i t has been found that the hydride-shift is faster than the phenyl-shift, assuming that the oxidation of secondary alcohols takes place through hydride-148 shifts, which is in agreement with the observation of the Baeyer-Villager oxidation of aldehydes (128a, 128b) (p.93). The Baeyer-Villiger oxidation reactions are believed to take place through 1,2-hydride shifts, and hydride-shifts are more favored than aryl-shifts in these reactions as well as in the reactions under the present investigation. •; The migratory aptitudes of aryl groups were calculated from the relative amounts of different products obtained. From the analysis of the products i t is apparent that a preferential migration of the electron-donating group occurs. However, the migratory aptitudes were found to be lover for electron-donating groups and higher for electron-attracting'groups than those reported by other workers for the acid-catalyzed rearrangements of d i - and triarylmethyl hydroperoxides (141, 142), which also proceed through aryl migration from carbon to an adjacent electron-deficient oxygen atom. The value of p for migration in the f i r s t case is -3.78 and in both cases the relative migratory aptitudes are very close to each other (141). The value of the reaction constant for migration was found to be -1.44 in the present inves-tigation. This is much lover than the value of the migration reaction constant in the hydroperoxide cases mentioned above. This indicates the absence of extensive aryl participation. In the acid-catalyzed rearrangements of arylmethyl hydroperoxides, a 149 considerable amount of aryl participation takes place, as shown in the mechanisms (ii) and ( i i i ) , and the reactions proceed mainly through phenonium ions. The lower value of reaction con-stant p in the present investigation indicates that the amount of positive charge on the central carbon atom in the cliramate ester (I) is less than that in the carbon atom of the intermediate (IV) of hydroperoxide rearrangement reaction. The possible path of the two mechanisms are shown below : Chromic acid oxidation of triarylcarbinol : Ar Ar Ar C ° — C*°3 H2 + Ar -C < 6+\ / 0 Ar I II Ar Ar + OAr + IL,CrO. 3 III 150 Rearrangement of triaryImethyl hydroperoxide (142) : Ar Ar Ar IV •0- + •OH, -*-Ar Ar 6+ 6+ - C r - 0 - - 0 H 6+V / ->Ar Ar -C =0-Ar + H 0 2 Ar A r — C O-Ar +° H2 VI The main difference between the hydroperoxide rearrangement reactions and the diromic acid reaction is the nature of the leaving groups. It is probable that the 0—Cr bond of II wil l be stronger than the 0—0 bond of V, and that bond rupture in the transition state wi l l be less advanced in II than in V. This will result in less positive charge accumulating on the central oxygen atom and the ring carbon atoms of II than of V. Thus the aryl participation in II is not as great as that in V, and the driving force due to aryl par-ticipation is less. This might be the cause of the more modest value of p found in the present study than in the rearrangement reactions of hydroperoxide where extensive aryl participation takes place. 151 Curtin and Crew (143) also found lower migration ratios in the rearrangement reactions of a variety of 2-amino-l, 1-diaryl-ethanols than in other cases of migration from carbon to nitrogen. These workers also explained the lower values by postulating a transition state with l i t t l e aryl participation. They believed that the rearrangement of the cortpound VI proceeds mainly through the structure VII, and not VIII, to the product IX. Ar Ar OH CH. * VI -NH, HNO„ Ar-Ar OH CH® VII Ar Ar C OH VIII •CH N 0 2 2 Ar C CH Ar 0 IX 152 However, this example is not parallel to the reaction under the present investigation. The reaction constant p+for chromic acid oxidation of triarylcarbinols was found to be -0.379. A negative p(usually near -1) is commonly found for chromic acid oxidation and this supports the idea that the central carbon becomes electron-defi-cient. The moderate size of the reaction constant shows that the amount of positive charge developed at this site is not great. This is also in agreement with the proposed mechanism that does not contain carbonium ion character to a great extent and consequently extensive aryl participation, as pointed out by Collins (140). The value, of p + for the overall reaction is less than that for aryl migration. In the f i r s t case, the observed value is the composite value, which includes the effect of electron withdrawing phenyl rings on the formation of the chromate ester, and thus i t reduces the overall value of p +. The higher acidity of the medium increases the ionization of the carbinol and consequently the concentration of the chromate ester, and thus accelerates the reaction. On addition of manganous ion a decrease in the rate of chromic acid oxidation of triarylcarbinols xvas observed. This is believed to be due to the oxidation of manganous ion to manganese dioxide by the intermediate chromium (IV) and chromium (V), induced 153 by the reacting carbinol. Similar reduction i n rates i n the presence,of manganous ions have also been reported by other workers (53, 75). According to Wiberg and Richardson (75) an organic substrate ZH^ can be oxidized by chromium (VI) folia-zing the possible paths (a) , (b) or (c) : (a) ZH2 + C r 7 1 + 2H+ + C r ^ C r ^ + Cr™ .2Cr V V « + i n ZH2 + Cr *Z + 2H + Cr VI + V (b) ZH2 + Cr ^ ZH* + H + Cr ZH* + C r 7 1 . — ^ Z + 2H+ + C r V V + i n ZH2 + Cr : «- Z + 2H + Cr (c) ZH2 + C r 7 1 -^ Z + 2H+ + C r 3 ^ ZH2 + Cr 3^ ^ZH* + H + + C r 1 1 1 ZH* + Cr71 --+-Z + H + + C r V ZH2 + C r V *-Z + 2H+ + C r 1 1 1 154 In a recent comrainication Pocek et a l . have ruled out the path (a) for the oxidative cleavage of cyclcbutanol by dxrornic acid i n presence of vanadium (IV) (144). This may be applicable to the chromic acid oxidation of other alcohols also even i n the absence of vanadium. These workers have proposed another • possible path (d), showing a rapid reaction between diromium(TV) and an organic substrate. This new mechanism has been based on Espenson's work (145), sham i n (e): IV f a s t V v (d) Cr + V — -*-Cr + V slav Cr" + V" *~Cr~' + V V . yEV b i t w IV . „v Cr" 0 7 + n-C4H?OH . ^ C r 1 1 1 + R* R* + V V — »-HOCH2CH2CH2aD. + where R* represents a free radical, (e) C^+V™ ^ ,Cr V + V V slav Cr' + V" — : >-Cr~" + V V . yTV » IV . „V IV IV I I I V Cr + \T : *-Cr + V 155 The added manganous ion, being a strong reducing agent, reduces diromium(V) and mrcmium(rv) thus eliminating the later steps i n a l l the above possible sequences of the oxidation reac-tion. This effectively decreases the rate of oxidation of the carbinol since the intermediate cliraTtLum(IV) acts as a strong reducing agent and reacts with a second chromium (VT) species to form two of diram.um(V), which then oxidize two more molecules of carbinol. The product analysis shaved that there i s no appreciable change i n the relative amounts of different products when manganous ions are added to the reaction mixture containing carbinol and chromic acid. This, together with the fact that the p + values i n * the absence and presence of manganous ion are very close to each other (Fig. 33), indicates that the mechanism of oxidation of triphenylcarbinol by chromium (V) or chrcmium(VT) i s essentially the same. The activation parameters calculated for the chromic acid oxidation of triphenylcarbinol, l i s t e d i n Table XX, are somewhat similar to those reported by D.G. Lee for chromic acid oxidation of secondary alcohols (129). 156 A COMPARISON BETv'IEEN THE REACTIONS OF  PERMANGANATE AND CHROMIC ACID Both permanganate and chromic acid oxidation of d i - and triarylcarbinols were found to be acid-catalyzed. This, together with the observed negative values of the respective reaction constants, indicates that i n a l l the cases the reaction proceeds through electron-deficient transition states which are stabilized by electron-donating ring substituents. In both the oxidation reactions the diary lcarbinols give only diaryl ketones whereas the t r i a r y lcarbinols produce the corresponding ketones along with phenols. Although the reactions of both chromic acid and acid permanganate are believed to proceed through an ester mechanism, the permanganate oxidation reactions are of f i r s t order i n carbinol and of zero order i n the oxidant, (except at very high acidities, as discussed later) whereas i n chromic acid oxidations the reactions are of f i r s t order i n each of the two reactants, i.e., of overall second order. The oxidation of triary lcarbinols i n acid i s believed to involve the formation and decorrposition of manganate and chromate esters. The decomposition of manganate ester i s much more rapid than that of drromate ester. This can be shorn from a comparison 157 of the rates of oxidation of t r i (p-tolyl) carbinol with the two oxidants at the saire acidity. Although the permanganate oxi-dation reactions of triarylcarbinols are of f i r s t order i n carbinol and zero order i n permanganate, they change to second order at higher acid i t i e s , and thus the rate can be compared at these acidities. Although the two sets of reactions were carried out i n different systems, the rates can be compared i n systems where the apparent acidity functions are the same. The comparison i s shown below : Acidity Rate of Chromic Acid Rate of Permanganate Oxidation Oxidation (Table XVIII) (Calculated from Fig.31) H R -3.95 3.06 l.mole~ 1sec" 1 75.86 l.mole^sec" 1 The rapid decomposition of manganate ester compared to chrcmate ester causes the step preceeding the reaction of carbinol and permanganate ion to be slower than the decomposition, and thus i n this case the scission of protonated carbinols i s the slaw step. On the other hand the decomposition of chrcmate ester, being slaver than the ionization of carbinols, becomes the rate-determining step. At higher acidities, where the ionization of 153 carbinols is faster, as shewn in the case of tri(p-tolyl)carbinol, the permangante reactions become second order, as in the case of diromic acid oxidation. APPENDIX 159 PERMANGANATE OXIDATION OF BENZHYDROL  IN FROZEN SYSTEMS During recent years i t has been found that many reac-tions proceed much more rapi d l y i n frozen systems than i n the corresponding l i q u i d systems (146-150). According to these workers an increase i n concentration of react ants i n i c e c r y s t a l s causes the acceleration. The permanganate oxidation of benzyl-amine has been studied and a s i m i l a r acceleration of rate was found i n the frozen systems (148). In order to study the oxida-t i o n of alcohols i n such systems, benzhydrol was chosen as the sub-st r a t e , since i t s reaction with permanganate i n l i q u i d systems has been studied by Stewart ( 4 ) , and a comparison of rates i n the too systems was convenient. EXPERIMENTAL For a t y p i c a l run a solution was made up with 10 ml of a standard solution of benzhydrol i n water, 20 ml of LM KH2PO4 buffer, and 68 ml d i s t i l l e d water which was boiled previously to expel dissolved gases, i n a 125 ml red Erlenmeyer f l a s k . For reactions at pH higher than 12, the required quantity of 0.1M carbonate-free potassium hydroxide was added instead of buffer t o maintain the desired pH. The t o t a l volume of mixture was kept 160 the same as before by adding water. To the mixture thus pre-pared, 2 ml of standard potassium permanganate solution was added and immediately after the addition 5 ml of the mixture was delivered into a screw-capped bottle, which was then immersed into a Dry Ice-ace tone bath. This sample was used as the blank. Eight more samples were prepared i n the same way and were frozen i n Dry Ice-acetone bath. After 5 minutes a l l the samples, except the blank, were taken out of the bath and immersed i n an ice-water bath for a minute and then transferred to an ethylene-glycol bath maintained at -10+0.02° with the aid of a cooling-unit. The baths were prepared i n covered Dewar flasks i n order to maintain a constant temperature. The i n i t i a l time was reckoned from the time the samples were placed i n the -10° bath. The blank was next warmed under tap water and was trans-ferred quantitatively to a quenching mixture containing an excess of potassium iodide and sulfuric acid. The liberated iodine was titrated with sodium thiosulphate using Thyodene as indicator. The other samples were also analyzed i n the same way. after definite intervals of time. For the reactions at pH less than 12, the ratio of sub-strate to permanganate taken was 3:2, according to the three 161 equivalent change of the oxidant to the two equivalent change of the substrate. A 1:4 molar ratio of benzhydrol to perman-ganate was taken for reactions at pH above 12. RESULTS • Order of Reaction : Good second order rate plots were obtained by using the rate expression V - V 1 o r k2 = ~JalcoholJo t X V q-2V 0/5 for the 3:2 ratio of substrate to permanganate and the rate expression V t " 9 V 1 0 V 4 V 5 for the 1:4 ratio of the reactants (4). A typical rate plot i s shown i n Fig. 1. Dependence of Rate on pH : The rates of reaction of benzhydrol and permanganate at -10° were studied between pH 1.50 and 12.40. The results are given i n Table I. Fig. 2 shows a comparison of the oxidation rates at 25° (4) and - 10°. 162 163 TABLE I Permanganate Oxidation of Benzhydrol i n Frozen System. Variation of Rate Constant with pH at -10° pH l.mole 1.50 34.50 3.48 23.01 7.00 15.00 8.30 13.90 9.40 20.00 11.30 26.20 11.80 65.80 12.20 95.50 12.40 164.00 Activation Parameters and Isotope Effect ; The oxidation of benzhydrol at pH 7.00 was studied at temperatures ranging from -20° to 30°, and that of benzhydrol-a-d was studied between the temperatures -14° to 25° under the same experimental conditions. The variation of rate constant of 1G4 165 the oxidation of benzhydrol with temperature is shown in Table II. The deuterium isotope effects at different temperatures are listed in Table III. Fig. 3 shows the change in rate con-stants with temperature for the two substrates. TABLE II Permanganate Oxidation of Benzhydrol in Frozen and Liquid Systems. Variation of P^te Constant with Temperature at pH 7.00. Temp.°C K2 l.mole~^min 30 8.05 25 7.00 20 5.80 10 3.95 0 2.80 -2 3.33 -3 11.50 -4 18.00 -5 22.89 -10 15.00 -15 8.20 -20 4.56 167 TABLE I I I Permanganate Oxidation of Benzhydrola -d i n Frozen and Liquid Systems . Variation of Rate Constant with Temperature at pH 7.00 .. Temp.°C k 2 l.nole'-Wn"1 25 0.95 7.3 10 0.57 7.7 0 0.37 1.1 -3 1.42 7.6 -5 3.01 7.4 -10 2.08 7.2 -14 1.20 7.5 From Fig. 3 i t can be found that the rnaximum rate i s obtained at -5° i n both the cases. The rate decreases on lowering as well as on raising the temperature. On one side, a steady decrease i n rate i s found up to -20°, this part shows the rates i n the frozen system. On the other side the rate reaches a mini-mum at 0° and then i t again rises steadily up to 30°, this part represents the rates i n the l i q u i d system. 168 Fig. 4 shows the plot of logQ^/T) against 1/T for the reactions of benzhydrol and benzhydrola-d. As i n Fig.3, i n this case also, linear parts, representing the reactions i n l i q u i d and frozen systems can be distinguished. From these portions the activation parameters for the reaction i n the two systems were calculated. Table IV l i s t s the results. TABLE IV Permanganate Oxidation.of Benzhydrol and Benzhydrola-d i n Frozen and Liquid Systems . pH 7.00 Activation Parameters. Substrate AH kcal;mole AH kcal.mole AS e.u. AS e.u. (frozen) (liquid) (frozen) (liquid) Benzhydrol 18.4 5.68 17.2 -44.0 Benzhvdrola-d 18.3 5.80 4.6 -47.7 169 F i g . 4 . '. Permanganate O x i d a t i o n o f 170 DISCUSSION The order of reaction in the frozen systems was found to be the same as in the liquid systems between pH 1.5 and .12.4, with the same stoichiometry : 3 ( C 6 H 5 ) 2 C H C H + 2 M n 04 = 3 ( C 6 H 5 ) 2 C O + 2 M n 02 + 2 H2° + 2 C H ~ In the frozen systems a large acceleration in rate compared to that in the liquid system at same pH was observed. At ,. pH 7.0 the rate constant in liquid systems at 25° was found to be 7.0 l-mole"^^*"! (4), whereas the same in the frozen system at -10° is 15.0 l.mole-^min--'-. Apart from this and the activation parameters, the dependency of rate on other variables is quite simi-lar to that for the liquid system. The rates are found to increase with pH in the same fashion as in the case of the liquid system, as shown in Fig. 2. The deuterium isotope effects are almost same for reactions in both the systems. These show that the "reaction mechanism in the frozen system is the same as that in the liquid system. The higher rate of reaction in the. frozen system is thus due to the higher 171 I t has been found i n some earlier work (146-150) that the rates of several types of bimolecular reactions i n frozen systems are higher than the corresponding rates of identical reaction at 25°. This acceleration i s believed to be due to the crystallization of water which results i n a higher concen-tration of solutes i n the remaining liq u i d phase. This increase i n concentration causes a higher rate of reaction i n frozen systems. The variation of rate constant with temperature i n frozen systems depends on two factors. In general, the second order rate constant for a bimolecular reaction i n solution decreases with decrease i n temperature. On the other hand i n frozen systems the concentrations of the reactants i n the liqu i d region increase, causing an increase i n rate. Thus the tempera-ture dependence of rates i n frozen systems i s the result of the competition between the too factors. I t has been observed (149) that the maximum rate i s obtained at a temperature be lav the freezing point of the solution. In the case of reaction between permanganate and benzhydrol, the same effect was found, the maxi-mum rate was found at -5°, below which the rate decreases steadily due to the normal temperature effect, as i n the l i q u i d system, following the Arrhenius Law. The minimum rate was obtained at 0°. Between 0° and -5° the concentration effect was stronger than the temperature effect and thus a gradual increase i n rate was observed i n this temperature range. 171a Suggestion for Further Work The rate-deterrninijig steps in the permanganate oxidation of benzhydrol and triphenylcarbinol have been found to be the ionization of the carbinol. Further support for this mechanism can be obtained by studying the racemization of ^symmetrically substituted carbinols in acidic media. The kinetics of the permanganate oxidation of benzhydrol in weakly acidic regions, between pH 7.00 and 1.50 requires further elucidation. The reaction has been found to be of second order in this region in both liquid (4) and frozen systems. The rate increases with acidity but the increase does not follow the pattern expected for simple acid catalysis. The slow increase in rate requires further investigation. The migratory aptitudes of different aryl groups during the chromic acid oxidation of triarylcarbinols have been deter-mined by a quantitative estimation of the products. The values have been found to be somewhat different than those found by other workers for migrations in analogous systems. This can be checked by labelling one of the rings with tritium and determin-ing the quantity of tritium in the two products, benzophenone and phenol. For the chronic acid oxidation of benzhydrol no migration of phenyl rings has been detected; that i s , no benz-aldehyde appears to have been formed. This can be checked by the method of isotope dilution. 172 BIBLIOGRAPHY 1. R. Stewart, Ch. 1, "Oxidation i n Organic Chemistry", Part A, K.B. Wiberg, ed., Academic Press, Nev; York, 1965. 2. F.R. Duke, J. Phys. Chem., 56, 882 (1952). 3. H.I. Schlesinger and H.B. Siems, J. Amer. Chem. Soc., 46_, 1965 (1924). 4. R. Stewart, J . Amer. Chem. Soc., 79, 3057 (1957). 5. R. Stewart and R. Van der Linden, Can. J. Chem., 38, 2237 (1960) . 6. K.B. Wiberg and R. Stewart, J . Amer. Chem. Soc., 77, 1786 (1955). 7. J.S.F. Pode and W.A. Waters, J. Chem. Soc., 717 (1956). 8. R.M. Evans, Quart. Rev., 61 (1959). 9. M. Harfenist, A. Baverly and W.A. Lazier, J . Org. Chem., 19_, 1608 (1954). 10. I.T. Harrison, Proc. Chem. Soc. 110 (1964). 11. E.F. Pratt and J.F. Van de Castle, J . Org. Chem., 26, 2973 (1961) . 12. E.F. Pratt and S.P. Suskind, J . Org. Chem., 28, 638 (1963). 13. H.F. Launer, J . Amer. Chem. Soc., 54, 2597 (1932). 14. D.R. Rosseinsky, J. Chem. Soc., 1181 (1963). 15. A.Y. Drummond and W.A. Waters, J . Chem. Soc., 2456 (1954). 173 16. A.Y. Druramond and W.A. Waters, J. Chem. Soc., 3119 (1953). 17. T.J. Kerrp and W.A. Waters, J. Ciiem. Soc., 339 (1964). 18. J.W. Ladbury and C.F. C u l l i s , Chem. Revs., 58, 403 (1953). 19. J.J. Ward, W.R. Kirner and H.C. Howard, J. Amer. Chem. Soc., 67, 246 (19.45). 20. Seijoo, A.C.S. Abstr., 18, 3517 (1924). 21. F.C. Tompkins, Trans. Far. Soc., 39, 280 (1943). 22. J. Holluta, Z. Physik. Chem., 106, 276, 324 (1923). 23. J. Holluta and A.-Mutschin, Z. Physik.Chan. (Leipzig), 150, 381 (1930). 24. R. Stewart and M.M. Mocek, Can. J. Chem., 41, 1160 (1963). 25. J. Halpem and J. Candlin, J. Amer. Chem. Soc., 85_, 2518 (1963). 26. J. Holluta, Z. Physik. Chem (Leipzig), 102, 32, 276 (1922). 27. D.R. Mann and F.C. Tompkins, Trans. Far. Soc., 37, 201 (1941). 28. R. Stewart and K.B. Wiberg, J. Amer. Chem. Soc., 78_, 1214 (1956). 29. S.M. Taylor and J. Halpem, J. Amer. Chem. Soc., 81, 2933 (1959). 30. J. Holluta, Z. Physik. Chem., 101, 34, 489 (1922). .31. J. Kenyon and M.C.R. Symons, J. Chem. Soc., 3580 (1953). 32. F.O. Ritter, J. Chem. Educ., 30, 395 (1953). 33. J.H. Merz, G. Stafford and W.A. Waters, J. Chem. Soc., 638 (1951). 174 34. K. Stewart and R. Van der Linden, Disc. Far. Soc., 211 35. R. Stewart and R. Van der Linden, Tet. Lett., 2, 28 (1960). 36. C.F. Cu l l i s and J.W. Ladbury, J. Chem. Soc., 555, 2850 (1955). 37. G. Porter and F. Wilkinson, Trans. Far. Soc., 57, 1686 (1961). 38. J.L. Kurz, J. Amer. Cham. Soc., 86_,2229 (1964). 39. F.H. Westheimer, Chem. Revs., 45, 419 (1949). 40. K.B. Wiberg, Ch.l, "Oxidation i n Organic Cliemistry", Part A, K.B. Wiberg, ed., Academic Press, New York, 1965. 41. A. Bystrom and K.A. Wilhelmi, Acta. Chem. Scand., 4, 1131 (1950). 42. M.L. Freedman, J. Amer. Chem. Soc., 80, 2072 (1958) 43. F.H. Westheimer and A. Novick, J. Chem. Phys., 11, 506 (1943). 44. R. Stewart and D.G. Lee, J. Amer. Chem. Soc., 86_, 3051 (1964). 45. M. Cohen and F.H. Westheimer, J. Amer. Chem. Soc., 74_, 4387 (1952). 46. R. Slack and W.A. Waters, J. Chem. Soc., 599 (1949). 47. N. Baily and M.C.R. Symon, J. Chem. Soc., 202 (1957). 48. A. Carrington, D.J.E. Ingram, D.S. Schonland and M.C.S. Symon, J. Chem. Soc, 4710 (1956). 175 49. (a) R. Scholder and W. Klemn, Angew. Chem., 66_, 461 (1954). (b) K. Wieczffinski and M. Andrelowicz, Roczniki Chem., 36, 1937 (1962), Chem. Abstr., 58_, 10953 (1963). 50. B. Takekazun and E. Hirota, Natl. Tech. Rept., 7, 372 (1961), Chem. Abstr., 56/ 15113 (1962). 51. W.A. Waters, Quart. Rev., 12, 277 (1958). 52. M.J. Udy, "Chrcmium", vol. 1, 128, Reinhold, New York 1956. 53. W. Watanabe and F.H. Westheimer, J. Chem.. Phys., 17, 61 (1949). 54. R. Wendland and J. La Londe, "Org. Synth", C o l l . vol. IV, 757 (1963) , R.P. Linstead and W.V.E. Doering, J. Amer. Chem. 'soc., 64, 1998 (1942). 55. L.I. Smith and I.M. Webster, J. Amer. Chem. Soc., 59_, 662 (1937). 56. R.D. Abell, J. Chem. See., 1397 (1951). 57. Y. Ogata and H. Akimoto, J. Org. Chem., 27_, 294 (1962). 58. R. Slack and W.A. Waters, J. Chem. Soc., 1669 (1948). 59. T. Zincke, Ber., 9, 31 (1876). 60. H. Stephen, W.F. Short and G. Gladding, J. Chem Soc., 520 (1920). 61. H. H. Shmant and J.F. Deffner, J. Amer. Chem. See., 81, 958 (1959). 176 62. W. "Hemlain, Ber., 1_, 1206 (1874). 63. A. Bistrzyeki and J. Gyr, Ber., 37, 1252, 3698 (1904). 64. F. Mares, J. Rocek and J. Sicher, °oll. Czech. Chem. Comm., 26, 2355 (1961). 65. W.F. Sager and W. Bradley, J. Amer. Chem. Soc., 78, 1187 (1956). 66. A. Behr, Ber., 5, 277 (1872). 67. W,J. Hickiribottcm and D.G.M. Wood, Nature, 168, 33 (1951). 68. F.C. Whitmore and G.W. Pedlow, J. Amer. Chem. See., 63_, 758 (1941). 69. G.T.E. Graham and F.H. Westheimer, J. Amer. Chem. See., 80_, 3030 (1958). 70. K.B. Wiberg and T. M i l l , J . Amer. Chem. Soc., 80_, 3022 (1958); K.B. Wiberg, i b i d . , 76, 5371 (1954). 71. J. Rocek, Tet. Lett., No. 5, 1 (1959). 72. A.K. Chatterji and S.K. Mukherjee, J. Amer. Chem. Soc., 80_, 3600 (1958). 73. T.J. Kemp and W.A. Waters, Proc. Roy. Soc., A274, 480 (1962). 74. J.W. Cornforth and G. Popjak, Nature, 164, 1053 (1940). 75. K.B. Wiberg and W.H. Richardson, J. Amer. Chem. Sec, 84, 2800 (1962). 76. P.A. Best, J. S. L i t t l e r and W.A. Waters, J. Chem. Soc., 822 (1962). 177 77. K. Urneda and K. Tarama, Nippon Kagaku Zasshi, 83_, 1216 (19t'2); Cham. Abstr., 58_, 12378 (1963). 78. G. Peti t , B u l l . Soc. Chim. France, 12_, 568 (1945). 79. J. Rocek et a l . , C o l l . Czech. Chan. Coram., 23, 2068 (1958). 80. A. Leo and F.H. Westheimer, J. Amer. Chem. Soc., 74, 4383 (1952). 81. F. Holloway, M. Cohen and F.H. Westheimer, J. Amer. Chem. Sec., 73, 65 (1951). 82. F.H. Westheimer and N. Nicolaides, J. Amer. Chem. See., 71, 25 (1949). 83. W.A. Mosher and D.M. Preiss, J. Amer. Chem. Soc., 75, 5605 (1953). 84. H.G. Kuivila and W.J. Becker, J. Amer. Cham. Soc., 74, 5329 (1952). 85. J.J. Cawly and F.H. Westheimer, J. Amer. Cham. Soc., 85, 1771 (1963). 86. J. Rocek, C o l l . Czech. Chem. Comm., 25, 1052 (1960). 87. G.T.E. Graham and F.H. Westheimer, J. Amer. Chem. Soc., 80, 3030 (1958). 88. K.B. Wiberg and H. Schafer, J . Amer. Chem. See., 89_, 455 (1967). 178 89. R. Stewart and D.G. Lee, Can. J. Chem., 42, 439 (1964). 90. D.G. Lee, W.L. Downey and R.M. Maass, Can. J. Chem., 46, 441 (1968). 91. D.G. Lee and R. Stewart, J . Org. Chem., 32_, 2868 (1967) . 92. A.K. Awasthy, J. Rocek and R.M. Moriarty, J. Amer. Chem. Soc., 89, 5400 (1967). 93. J. Rocek, F.H. Westheimer, A. Eschenrroser, L. Moldovanyi and J. Schreiber, Helv. Chim. Acta., 45_, 2554 (1962). 94. (a) H. Kwart and P.S. Francis, J . Amer. Chem. Soc., 81, 2116 (1959). (b) i b i d . , 77, 4907 (1955). (c) H. Kwart, Chem. Ind. (Lond.), 610-611 (1962). 95. W.F. Sager, J . Amer. Chem. Soc., 78_, 4907 (1956). 96. W.A. Mosher and E. Langerak, J . Amer. Chem. Soc., 73, 1302 (1951). 97. W.A. Mosher and E. Langerak, J . Amer. Chem. Soc., 7L, 286 (1949). 98. L.S. Levitt, J . Org. Chem., 20, 1297 (1955). 99. J. Hampton, A. Leo and F.H. Westheimer, J. Amer. Chem. Soc., 78,306 (1956). 100.. J.R. Jones and W.A. Waters, J . Chem. Soc, 2068 (1962). 179 101. W.A. Waters and J.S. L i t t l e r , J. Chem. Soc., 4046 (1959). 102. K. Julian and W.A. Waters, J. Chem. Soc., 818 (1962). 103. J.R. Jones and W.A. Waters, J. Chem. Soc., 2772 (1960). 104. D.G. Hoare and W.A. Waters, J. Chem. Soc., 965 (1962). 105. R.J. Meyers and R. Jacoby, Z. Anorg. Chem., 27, 359 (1901); M. Ardon, J. Chem. Soc., 1811 (1957). 106. G. Hargreaves and L.H. Sut c l i f f e , Trails. Far. Soc, 51, 1105 (1955). 107. J.S. L i t t l e r , J. Chem. Soc, 4135 (1959). 108. F.H. Westheimer, Chem. Revs., 61, 265 (1961). 109. V.M. Micovic and M.L. Mihailovic, Rec. Trav. Chim., 71_, 970 (1952). 110. R.E. Partch, Tet. Lett., 3071 (1964). 111. E.A. Braude and O.H. Wheeler, J. Chem. See, 320 (1955). 112. W.A. Mosher, C.L. Kehr and L.W. Wright, J. Org. Chem., 26_, 1044 (1961). 113. W.A. Mosher and H.A. Neidig, J. Amer. Chem. Soc, 72, 4452 (1950). 114. W.H. Starnes Jr., J. Amer. Chem. Soc, 89, 3368 (1967); i b i d . , 90, 1807 (1968). 115. (a) W.D. Cohen, Rec. Trav. Chim., 38, 118 (1919); (b) i b i d . , 38, 119 (1919); (c) i b i d . , 38, 116 (1919). .180 116. W.E. Bachrnan, E. Carlson and J.C. Moran, J. Org. Chem., 13, 916, 917 (1948). 117. H. Schnackenherg and R. Scholl, Ber., 36, 654 (1903). 118. A.I. Vogel, "Quantitative Inorganic Analysis", Longmans, Green and Co. Ltd., New York, N.Y. 1957. 119. A.I. Vogel, "Practical Organic Chemistry", Longmans, Green and Co. Ltd., London 1964. 120. L.P. Hairmett, "Physical Organic Chemistry", McGraw-Hill Book Co. Inc., New York, N.Y. 1940. 121. N.C. Deno, J.J. Jaruzelski and A. Schriesheim, J. Amer. Chem. Soc., 77, 3044 (1955). 122. W.D. Cohen, Pec. Trav. Chim., 38, 72-88, 113-131 (1919). 123. M.A. Paul and F.A. Long, Chem. Rev., 57, 1 (1957). 124. K.B. Wiberg, "Physical Organic Chemistry", John Wiley and Sons Inc., New York, 1964. 125. G. Kortum, W. Vogel and K. Andrussow, "Dissociation Constants of Organic Acids i n Aqueous Solutions", Butterworths, London 1961. 126. S. Nishida, J . Org. Chem., 32, 2692 (1967). 127. G.A. Russel, J. Org. Chem., 23, 1407 (1958). 127a. S. Winstein, A.H. Fainberg and E. Grunwald, J. Amer. Chem. Soc., 79, 4146 (1957). 181 127b. R.E. Robertson, "Solvolysis i n Water", National research Council Laboratories Publication, Ottawa, Canada. 128. R. Stewart, "Oxidation Mechanisms : Application to Organic Chemistry", W.A. Renjamin Inc., New York, N.Y. 1963. 128a. H. Kwart and M.J. Wegmer,.J. Amer. Chem. Soc., 83, 2746 (1961). 123b. E. Bamberger, Ber., 36, 2042 : (1903). 128c. P.A.S. Smith, "Molecular•Rearrangements", Vol. 1, edited by P. de Mayor, Interscience Publishers Inc., New York 1963, p.530. 129. D.G. Lee, Ph.D. Thesis, University of Bri t i s h Columbia,1963. 130. ,N.E. Tousley and M. Gcmberg, J. Amer. Chem. Soc., 26_, 1516 (1904). 131.. A. K l i e g l , Ber., 38, 86 (1905). 132. M.P. Balfe and others,. J . Chem. Soc., 797 (1946). 133. A.E. Tschitschibabin, Ber., 44, 450 (1911). 134. C.S. Marvel e t - a l . , J . Amer. Chem. Soc., 66_, 914, 917 (1944). 135. S.T. Bowden and T.F. Watkins, J. Chem. See., 1249, 1254 (1940). 136. . W.E. Bachman and F.Y. Wiselogle, J. Org. Chem., 1_, 354, 371 (1936). 137. V.M. Zvenigorodskaya, Zavoskaya Lab., 12_, 152 (1946); Chem. Abstr., 40, 5661 (1946). 182 138. R. Stewart, M.R. Granger, R.B. Moodie and L.J. Muenster, Can. J . Chem., 41, 1065 (1963). 139. H. Nicholson and P.A.H. Wyatt, J. Chem. Soc., 198 (1968). 140. C.J. Collins, Quart. Rev., 1£, 357 (1960). 141. G.H. Anderson and J.G. Smith, Can. J. Chem., 46_, 1553 (1968). 142. D.E. Hissing, C.A. Matuszak and W.E.McEwen, J. Amar. Chem. Soc., 86, 3824 (1964). 143. D.Y. Curtin and M.C. Crew, J. Amer. Chem. Soc., 76_, 3719 (1954). 144. J. Rocek and A.E. Radkowsky, J. Amar. Chem. Soc., 90_, 2986 (1968). 145. J.H. Espenson, J. Amar. Chem. Soc., 86_, 5101 (1964). 146. N.H. Grant, D.E. Clark and H.E. •Alburn, J . Amer. Chem Soc., 83, 4476 (1961). 147. T.C. Bruice and A.R. Butler, J. Amer. Chem. Soc.,86, 4104 (1954). 148. 1 T.C. 3ruice and A.R. Butler, J . Amer. Chem. Soc., 86_, 313 (1964). 149. R.E. Pincock and T.E. Kivosky, J. Amar. Chem. Soc., 87, 2072, 4100 (1965). 150. H.H. Wei and R. Stewart, J. Amer. Cham. Soc., 83, 1974 (1966). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059926/manifest

Comment

Related Items