@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Chemistry, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Albagli, Alain"@en ; dcterms:issued "2011-06-10T20:01:49Z"@en, "1969"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description """Supervisor : Professor Ross Stewart In the last decade there has been considerable interest in strongly basic solutions and Hammett H_ acidity functions have been defined for various solvent systems. However some doubts still remain about the ionization modes of certain indicators which were used to determine these H_ functions, and the pKa values assigned to weak carbon acids. A careful anchoring of the H_ scale for the system dimethylsulfoxide-water-0.011 molar tetramethylammonium hydroxide with 2,4,4'-trinitrodiphenylamine was undertaken. The lower region of this scale correlates well with the recently published H_ function for the more basic regions. To study the dependence of H_ functions on indicator structure, the ionization of nitrogen and oxygen acids were compared in strongly basic systems. The solvent systems used in the study were dimethylsulfoxide-methanol-0.01 molar sodium methoxide and dimethylsulfoxide-ethanol-0.01 molar sodium ethoxide. The shallow rise of the function for phenols as compared to that of the HN for diphenylamine is believed to be caused by the smaller degree of charge delocalization in the phenolate anions than in the amide ions. The dissociation constants of 15 substituted nitrodiphenyl-amines were determined in the two previously mentioned alcoholic dimethylsulfoxide systems, and in the aqueous dimethylsulfoxide system. The "similar colour indicator" postulate was elaborated on the observation that the relative acidity of Hammett-type indicators remains constant through changes in solvent environment only if their spectral characteristics are similar. Linear free energy relationships between the pK values of 30 substituted diphenylamines and the substituent constant, σ°, were drawn. The ρ value in excess of 4 for monosubstituted diphenylamines compared with ρ values of the order of 2 for nitrosubstituted diphenylamines, results from the low degree of conjugation between the central N atom and the ring substituent whenever a nitro group is present in the other ring. The base-catalyzed rates of detritiation of various hydrocarbons were determined in dimethylsulfoxide mixtures. The logarithm of these rates were correlated with the thermodynamic basicity of the medium as measured by the H N function and gave excellent straight lines with slopes around 0.8. The theoretical basis of such correlations is discussed, and it was concluded that the logarithm of the rates should be correlated with H_ + log aROH. In the case of methanolic dimethylsulfoxide the activity of methanol has been determined, giving corrected correlations with slopes around 0.9. The qualitative limitation due to the use of HN instead of HC functions is also discussed. The activation parameters for these detritiations have been determined. The enthalpies of activation were found to increase with decreasing acidity of the hydrocarbon. These variations in enthalpy of activation, coupled with variation in the kinetic isotope effect and the observed Bronsted relationship have been interpreted in terms of a mechanism involving a rate-determining pre-equilibrium step with a transition state of increasing asymmetry. A critical examination of the HC scale for carbon acids in dimethylsulfoxide-ethanol-0.01 molar sodium ethoxide was made. The failure of the scale probably results from the fact that, for a number of indicators, proton abstraction is followed by a reaction between the carbanion and the neutral molecule to give a radical and radical anion. The carbanions of three nitrophenylmethanes were prepared in hexamethylphosphoramide and unequivocally identified by high resolution nuclear magnetic resonance. The spectral characteristics of the carbanions were recorded and compared with previously published values. The radical anions of these nitrophenylmethanes were generated electrochemically and their e.s.r. spectra measured. The variation in ¹⁴N hyperfine coupling constant (aN) of these radical anions indicated a large solvent shift in the absorption maxima."""@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/35382?expand=metadata"@en ; skos:note "STRONGLY BASIC SYSTEMS by ALAIN ALBAGLI A THESIS SUBMITTED IN PARTIAL FULFILMENT 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 October, 1969 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 o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h C o lumbia, I agree 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 s t u d y . 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 p u rposes may be g r a n t e d by the Head o f my Department o r 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 o f 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 . Department The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8, Canada Date ABSTRACT Supervisor : Professor Ross Stewart In the l a s t decade there has been considerable i n t e r e s t i n strongly basic solutions and Hammett H a c i d i t y functions have been defined for various solvent systems. However some doubts s t i l l remain about the io n i z a t i o n modes of certain indicators which were used to determine these H functions, and the pK values assigned to weak carbon acids. A careful anchoring of the H scale for the system dimethylsulfoxide-water-0.011 molar tetramethylammonium hydroxide with 2,4,4'-trinitrodiphenylamine was undertaken. The lower region of t h i s scale correlates well with the recently published H function for the more basic regions. To study the dependence of H functions on indicator structure, the ion i z a t i o n of nitrogen and oxygen acids were compared i n strongly basic systems. The solvent systems used i n the study were dimethylsulfoxide-methanol-0.01 molar sodium methoxide and dimethylsulfoxide-ethanol-0.01 molar sodium ethoxide. The shallow r i s e of the function for phenols as compared to that of the for diphenylamine i s believed to be caused by the smaller degree of charge d e r e a l i z a t i o n i n the phenolate anions than i n the amide ions. The d i s s o c i a t i o n constants of 15 substituted nitrodiphenyl-amines were determined i n the two previously mentioned alcoholic dimethylsulfoxide systems, and i n the aqueous dimethylsulfoxide system. The \" s i m i l a r colour i n d i c a t o r \" postulate was elaborated on the observation that the r e l a t i v e a c i d i t y of Hammett-type indicators remains constant through changes i n solvent environment only i f t h e i r spectral characteristics are s i m i l a r . Linear free energy relationships between the pK values of 30 substituted diphenylamines and the substituent constant, a°, were drawn. The p value i n excess of 4 for monosubstituted diphenylamines compared with p values of the order of 2 for nitrosubstituted diphenylamines, results from the low degree of conjugation between,the central N atom and the r i n g substituent whenever a n i t r o group i s present i n the other r i n g . The base-catalyzed rates of d e t r i t i a t i o n of various hydro-carbons were determined i n dimethylsulfoxide mixtures. The logarithm of these rates were correlated with the thermodynamic b a s i c i t y of the • N medium as measured by the H function and gave excellent straight lines with slopes around 0.8. The theoretical basis of such correlations i s discussed, and i t was concluded that the logarithm of the rates should be correlated with H + log a D n u . In the case of — Run methanolic dimethylsulfoxide the a c t i v i t y of methanol has been determined, giving corrected correlations with slopes around 0.9. - i v -N C The q u a l i t a t i v e l i m i t a t i o n due to the use of H instead of H functions i s also discussed. The activation parameters for these d e t r i t i a t i o n s have been determined. The enthalpies of activation were found to increase with decreasing a c i d i t y of the hydrocarbon. These variations i n enthalpy of a c t i v a t i o n , coupled with v a r i a t i o n i n the k i n e t i c isotope effect and the observed Bronsted relationship have been interpreted i n terms of a mechanism involving a rate-determining pre-equilibrium step with a t r a n s i t i o n state of increasing asymmetry C A c r i t i c a l examination of the H scale for carbon acids i n dimethylsulfoxide-ethanol-0.01 molar sodium ethoxide was made. The f a i l u r e of the scale probably results from the fact that, for a number of indicators, proton abstraction i s followed by a reaction between the carbanion and the neutral molecule to give a r a d i c a l and r a d i c a l anion. The carbanions of three nitrophenylmethanes were prepared i n hexamethylphosphoramide and unequivocally i d e n t i f i e d by high resolu-t i o n nuclear magnetic resonance. The spectral characteristics of the carbanions were recorded and compared with previously published values. The r a d i c a l anions of these nitrophenylmethanes were generated electrochemically and t h e i r e.s.r. spectra measured. The v a r i a t i o n i n 1 4N hyperfine coupling constant (a^) of these r a d i c a l anions indicated a large solvent s h i f t i n the absorption maxima. - V -TABLE OF CONTENTS Page INTRODUCTION ...... . . .. 1 I. The H a c i d i t y function 4 I I . Carbon Acids 9 I I I . Nitrogen Acids 15 IV. Oxygen Acids ••• ^ V. Application of the H Function to Kinetics and Mechanisms of Base Catalysis 19 A. Kinetic-Thermodynamic relationships 23 B. Isotope effects 29 SCOPE OF THE PRESENT RESEARCH . 31 EXPERIMENTAL • . 33 I. P u r i f i c a t i o n of solvents 33 I I . Base \"•. • • 34 I I I . Preparation of solutions 34 IV. Spectral measurements 35 V. Treatment of data and calculations 35 VI. Probable error i n H and pK values ... 36 - a . VII. Kinetic measurements 37 VI I I . Treatment of the k i n e t i c data 39, IX. Reactions of nitrophenylmethanes- 43 X. Computer treatment of data ; 44* XI. Preparation of compounds used i n this work 44 - v i -Page RESULTS AND DISCUSSION '. . . . 51 Part I: Indicator measurements 51 N I. Diphenylamine indicators H . •.. 51 A. V a l i d i t y of the scale i n alcoholic DMSO 65 B. Correlation of Structure with A c i d i t y 8-2 I I . Phenol indicators 89 Part I I : Base catalyzed d e t r i t i a t i o n of hydrocarbons .... 100 I. Evaluation of the kinetics 109 I I . Activation parameters 117 I I I . K i n e t i c thermodynamic correlation f o r hydrocarbon ioni z a t i o n 122 IV. Change of isotope effect with the pK of hydrocarbons .... 124; Part I I I : Interactions of Nitrophenylmethanes with base . 127 I. N.m.r. investigation 129 I I . E.s.r. investigation . 138 CONCLUSION ....... 146 BIBLIOGRAPHY 147 APPENDIX A: Ionization Ratio Data . .. 154 APPENDIX B: Rate Correlation Data 162 APPENDIX C: Computer Program. 168 - v i i -LIST OF TABLES Table Page I Degree of ion i z a t i o n of weak acids, HA, i n 0.01 M OH solutions 2 II Degree of ion i z a t i o n of a n i l i n e , pK = 27.3, i n aqueous DMSO, [0H_] =0.01 2 III pK of carbon acids i n ethanolic DMSO and i n DMSO 11 a IV pK & values of hydrocarbons and \"best values\" 13 V pK of various diphenylamines as determined i n the aqueous DMSO system 54 N VI H values for the system DMSO-water-0.Oil M t e t r a -methylammonium hydroxide 55 VII pK of various diphenylamines as determined i n the methanolic DMSO system 58 N • VIII H values for the system DMSO-methanol-0.01 M sodium methoxide 59 IX pK of various amines as determined i n the ethanolic r a DMSO system. • • • • 62 N X H values for the system DMSO-ethanol-0.01 M sodium ethoxide 63 XI PKBH+ °^ w e a k bases i n s u l f u r i c acid media gathered from the l i t e r a t u r e 71 XII Summary of the pK values of diphenylamines determined i n this work , 77 XIII Absorption maxima and molar a b s o r p t i v i t i e s of the indicators i n absolute ethanol 78 - v i i i -Table Page XIV Absorption maxima and molar ab s o r p t i v i t i e s of indicator sanions i n DMSO-ethanol 79 XV Linear free energy correlation of diphenylamines sacidities computed by a least squares method ... .. 85 \"XVI values for the systems DMSO-methanol-0.01 M sodium AHiethoxide and DMSO-ethanol 0.01 sodium ethoxide 92 ;XVII pK a values of phenols determined i n aqueous solutions . .^and i n methanolic and ethanolic DMSO 94 'XVIII Absorption maxima and molar abso r p t i v i t i e s of phenolate anions ......... — 95 XIX Correlation of log k with for the base catalyzed d e t r i t i a t i o n of hydrocarbons .... 105 XX Correlation of log k with H_ and H + log a^Q^ for the base catalyzed isotopic exchange i n methanolic DMSO ... 115 XXI Activation parameters for the base catalyzed d e t r i t i a t i o n of hydrocarbons 118 XXII S p l i t t i n g constants a^ for BNPM and 3,4'-DNPM i n methanolic DMSO mixtures 140 XXIII Spectral data of nitrophenylmethanes i n strongly basic solutions 144 XXIV Experimental values of log I for diphenylamine indicators i n the system DMSO-water-tetramethylammonium hydroxide (0.0011 M) 154 - i x -Table Page XXV Experimental values of log I for diphenylamine indicators i n the system DMSO-methanol-sodium methoxide (0.01 M) 156 XXVI Experimental values of log I for amine indicators i n the system DMSO-ethanol-sodium ethoxide (0.01 M) 158 XXVII Experimental values of log I for the substituted phenol indicators i n the system DMSO-methanol-sodium methoxide (0.01 M) 160 XXVIII Experimental values of log I for the substituted phenol indicators i n the system DMSO-ethanol-sodium ethoxide (0.01 M) 161 XXIX Rates of d e t r i t i a t i o n (£. mole - 1sec 1 ) , k - at 25° for UK various hydrocarbons i n basic DMSO-ethanol solutions .. 162 XXX Rates of d e t r i t i a t i o n (5,. mole *sec , k^^- at 25° for various hydrocarbons i n basic DMSO solutions 164 XXXI Ionization rate constants (£. mole^sec 1) , k(DMSO), at 25° for DMSO i n basic DMSO solutions 165 XXXII Rate-temperature data for the base catalyzed d e t r i t i a t i o n of hydrocarbons i n DMSO-ethanol-0.01 M sodium ethoxide. 167 - X -LIST OF FIGURES Figure Page 1. Typical f i r s t - o r d e r rate plots for the d e t r i t i a t i o n of 9-phenylfluorene i n absolute methanol at 25° ......... 40 2. Typical f i r s t - o r d e r rate plot for the t r i t i a t i o n of DMSO 42 3. Plots of log I (A/AH) versus H for the system: DMSO-water-0.001 M TMAOH 53 4. Plots of log I (A/AH) versus H_ for the system: DMSO-methanol-0.01 M sodium methoxide 57 N 5. H a c i d i t y function for the system: DMSO-methanol-0.01 M sodium methoxide .......... 60 6. Plots of log I (A/AH) versus H for the system: DMS0-ethanoI-0.01 M sodium ethoxide . 61 7. a c i d i t y function for the system: DMSO-ethanol-0.01 M sodium ethoxide 64 8. Plots of log I (A/AH) versus mole percent DMSO for the system: DMSO-water-0.011 M TMAOH.. 66 9. Plots of log I (A/AH) versus mole percent DMSO for the system: DMSO-methanol-0.01 M sodium methoxide 67 10. Plots of log I (A/AH) versus mole percent DMSO for the system: DMSO-ethanol-0.01 M sodium ethoxide 68 11. Plots of log I (A/AH) versus mole percent DMSO for the amine indicators used to establish the function i n : DMSO-methanol-0.025 M sodium methoxide from reference 33. 81 Hammett correlation of the a c i d i t y of substituted diphenylamines Plots of log I (A/AH) versus H_ for the phenols used to establish the H^ scale i n DMSO-methanol-0.01 M sodium methoxide Plots of log I (A/AH) versus H for the phenols used to establish the H^ scale i n DMSO-ethanol-0.01 M sodium ethoxide H^ ac i d i t y functions for the systems: DMSO-methanol-0.01 M sodium methoxide and DMSO-ethanol-0.01 M sodium ethoxide 0 N H and H ac i d i t y functions f o r the system: DMSO-methanol-0.01 M sodium methoxide O N -H_ and H aci d i t y functions for the system: DMSO-ethanol 0.01 M sodium ethoxide Plots of log k versus H^ for the d e t r i t i a t i o n of hydro-carbons i n DMSO-ethanol-0.01 M sodium ethoxide Plots of log k ( d e t r i t i a t i o n of hydrocarbons) versus log k (DMSO exchange) i n DMSO-ethanol-0.01 M sodium ethoxide N Plot of log k versus H for the d e t r i t i a t i o n of fluorene-9t i n DMSO-water-0.011 M TMAOH. - x i i -Figure Page 21. Plot of log k versus for the DMSO isotopic exchange reaction i n DMSO-water-0.01 M TMAOH 106 N 22. Plot of log k versus H for the DMSO isotopic exchange reaction i n DMSO-ethanol-0.01 M sodium ethoxide ....... 107 N N 23. Plot of log k versus H (l i n e 1) and H_ + log a|^eQjj f o r the DMSO isotopic exchange i n DMSO-methanol-0.01 M sodium methoxide 108 N N 24. Plot of log k versus H_ (l i n e 1) and H_ + log ^ q H for the d e t r i t i a t i o n of 9-phenylfluorene-9t i n DMSO-methanol-0.01 M sodium methoxide 116 25. Plots of log k versus 1/T°K for the d e t r i t i a t i o n of hydrocarbons i n ethanolic DMSO (horizontal) 119 26. Plots of log k versus 1/T°K for the d e t r i t i a t i o n of hydrocarbons i n ethanolic DMSO (ve r t i c a l ) 120 27. Correlation of k i n e t i c and thermodynamic a c i d i t i e s of some hydrocarbons 123 28. N.m.r. spectra of tris-(p-nitrophenyl)methane and anion i n HMPT I 3 2 29. N.m.r. spectra of bis-(p-nitrophenyl)methane and anion i n HMPT 134 30. N.m.r. spectra of 3,4'-dinitrodiphenylmethane and anion i n HMPT ...'.... 136 31. Plot of the average s p l i t t i n g constant versus mole per-cent DMSO for the ra d i c a l anion of 3,4'-dinitrodiphenyl-methane i n methanolic DMSO 141 - X l l l -Figure Page 32. Plot of the average s p l i t t i n g constant versus mole percent DMSO for the r a d i c a l anion of bis-(p-nitrophenyl) methane i n methanolic DMSO 142 - x i v -ACKNOWLEDGEMENTS I would l i k e to thank Professor Ross Stewart for h guidance throughout t h i s research project. I would also l i k e to thank Dr. J.R.Jones and D.E. Kennedy and J.A.MacPhee for th e i r helpful suggestions and c r i t i c i s m s . INTRODUCTION 1 In a recently published review on H a c i d i t y functions, strongly basic systems have been a r b i t r a r i l y defined as those solutions which ionize acids with an a b i l i t y equal to or greater than 0.1 M aqueous a l k a l i metal hydroxide solutions. There are several methods of producing strongly basic systems. Of the two most common, the f i r s t consists of preparing a solution of a p r o t i c solvent and a s a l t of i t s conjugate base (e.g., sodium hydroxide water), the b a s i c i t y of the system increasing with increase of the base concentration. The second one, which produces s t a r t l i n g e f f e c t s , makes use of a binary solvent system: a pro t i c and a polar aprotic solvent with a small concentration of the conjugate base of the protic partner. In t h i s case the a b i l i t y of the base to remove a proton from a weak acid i s increased by increasing the concentration of the polar aprotic solvent 2 Indeed there have been cases reported i n the l i t e r a t u r e i n which these methods were coupled, i . e . increasing the base concentration i n a system consisting of a fixed percent polar solvent i n a p r o t i c one. The increase i n a b i l i t y of hydroxide ions to remove a proton from a weak acid by addition of a polar aprotic solvent l i k e dimethyl-sulfoxide (DMSO) can be i l l u s t r a t e d by the data i n Table I which shows the effect of DMSO on the degree of ion i z a t i o n of three aromatic amines i n solution each 0.01 M i n hydroxide ion. - 2 -TABLE I Degree of Ionization of Weak Acids, HA, i n 0.01 M OH Solutions HA Water 92 mole % DMSO 99.6 mole % DMSO °2 nA /r-m2 undetectable complete complete (C 6H 5) 2NH undetectable h a l f complete C,HCNH0 6 5 2 undetectable undetectable one-tenth 14 Ef f e c t i v e l y there i s a 10 fol d increase i n the b a s i c i t y of the system toward aromatic amines by replacing water by DMSO. Table II shows how the degree of ioni z a t i o n of a n i l i n e ( PK a = 27.3) varies with solvent enrichment i n DMSO. TABLE I I * Degree of Ionization of A n i l i n e , pK & = 27. ,3 i n aqueous DMSO, [OH-] = 0.01 wt - % DMSO mole - % DMSO H_ [C6H5NH-] [C 6H SNH 2] 0 0 12.0 i o - 1 5 20 5.45 12.6 i o - 1 5 40 13.4 13.6 i o \" 1 4 t — 60 25.7 15.1 1 0-12 80 48,0 17.3 i o \" 1 0 90 67.6 19.2 i o \" 8 95 81.5 20.8 i o \" 7 99.9 99.6 26.2 0.1 t i n 66 wt % DMSO [C6H5NH\"] = [C 6H 5NH 3 ] Table I and II are taken from reference (3) - 3 -3 4 This increase i n b a s i c i t y has been r a t i o n a l i z e d ' by considering the effect of addition of DMSO on the equilibrium shown i n equation (la) HA + OH (H 20) 3 *• A\" + 4H20 (la) By addition of DMSO the a c t i v i t y of water w i l l be lowered^. This lowering i n the a c t i v i t y of the pr o t i c partner by increasing the concentration of the aprotic solvent was also noted i n the sulfolane-7 8 water system and i n the DMSO-methanol system. But t h i s lowering i n the a c t i v i t y of water cannot account for a l l the increase i n b a s i c i t y when DMSO i s added to aqueous solutions of hydroxide ion. Another factor of importance i s the state of solvation of the species OH and A i n equation (la) as we go from aqueous solutions to DMSO. In water a small ion l i k e hydroxide ion i s surrounded by water molecules and i s s t a b i l i z e d by strong hydrogen bonds while i n DMSO the hydroxide ion i s weakly solvated and thus i t s a c t i v i t y w i l l increase with increasing DMSO content. This argument i s based on the premise that an unsolvated ion w i l l be more reactive than a solvated one which must be snatched from i t s sheath of solvent molecules before a reaction can occur. U n t i l recently, i t was widely accepted that large delocalized anions l i k e amide ions or carbanions were more solvated i n DMSO than 9 i n a pr o t i c solvent, but this premise was recently challenged. Fuchs et a l . \" \" ^ measured enthalpies of solution of lithium and potassium halides i n DMSO and found that the order of solvation of halide ions i n DMSO The species H^ O^ i n equation (la) corresponds to the secondary hydrate complex H„0.+ i n acid solutions.** r 9 4 - A -i s the same as i n water. They therefore postulated that differences of solvation between ions, whether large or small, i n DMSO i s minimal. The main effect of additing DMSO to the system shown i n equation (la) seems to be to desolvate OH . Besides these two large e f f e c t s , there are several minor ones. Lowering the concentration of water i n the system w i l l tend to s h i f t the equilibrium to the r i g h t . DMSO i t s e l f i s coordinated to two water molecules,*\"'' thus addition of DMSO further lowers the water concentration by complex formation. A f a i r estimate of the increase i n H up to 12 about 50 wt percent DMSO (20 mole percent) can be made by considering only the effect of DMSO on the concentration of the individual species i n equation ( l a ) . With further increase of the DMSO content, the vari a t i o n i n the a c t i v i t y of water w i l l be the predominant factor, and at very high DMSO content greater than 95 percent the desolvation of the hydroxide ion w i l l have a dramatic effect on H_ values. A l l these factors w i l l tend to s h i f t the equilibrium (la) to the rig h t as DMSO replaces water as a solvent. On the other hand, the a c t i v i t y c o e f f i c i e n t of the neutral acid HA w i l l tend to f a l l as water i s replaced by DMSO as indicated by . the generally higher s o l u b i l i t y of aromatic amines and hydrocarbons i n DMSO than i n water. This i s the only factor which w i l l tend to s h i f t the equilibrium to the l e f t . I. The H a c i d i t y function The H a c i d i t y function describes the proton-abstracting a b i l i t y of a solution. The following equilibrium exists between an acid - 5 -uncharged and i t s conjugate base A\" i n aqueous solution of a base; AH —^>- A + H + ( l b ) * The thermodynamic equilibrium constant for equation (lb) can be expressed as: \" . ~ aHA \" I>*1 ' fAH H ( ) or i n a logarithmic form as f -l 0 g { m } = l 0S KHA \" l 0S f ^ a H + C 3 ) The symbols [ ], a, f, refer to concentration, a c t i v i t y and a c t i v i t y c o e f f i c i e n t respectively; and are related by the equation a^ = f ^ [ i ] . In very d i l u t e solutions when the a c t i v i t y c o e f f i c i e n t s approach unity t h i s equation reduces to Since the simple term - log [H +] or pH measures the proton, abstracting power i n d i l u t e aqueous solution,then the equivalent terra i n equation (3) can be used i n strongly basic media, and thus defines the function * A l l these species are hydrated but are not so denoted, for the sake of s i m p l i c i t y . - 6 -V H = -log -f- a + (5) AH Hammett's a c t i v i t y c o e f f i c i e n t postulate states that for H_ to define £A~ the a c i d i t y of the medium,-5— should be independent of the structure AH of the indicator and only dependent on the solvent composition. Substituting (5) into (3) and rewriting we get: H- = P KHA + • C 6 ) The Hammett indicator-overlap or stepwise technique of determining t h i s a c i d i t y function H has been used by a number of workers i n the past years. I t involves the comparison of b a s i c i t y of indicators i n different media s t a r t i n g i n the water region. F i r s t l e t us consider an acid HA whose pK can be determined i n aqueous buffer solutions. We can rewrite equation (2) as P KHA = l o s [F] \" l 0 * V TT ( 7 ) AH Si m i l a r l y for another acid HB, another equation can be written: [BH1 _ V P KHB = l o s JS=f ~ l o g V f— ( 8 ) HB Subtracting (8) from (7) gives the following equation: * pK^, P^ H B are the pK & values of the acids HA and HB respectively - 7 -[HA] . [HB] . fHA fB\" . ... P HA \" P HB = *?* t H \" \" l 0 g WT + lGg (9) Now i f , and only i f , the la s t term i n equation (9) i s equal to zero (Hammett's a c t i v i t y c o e f f i c i e n t postulate) equation (9) becomes: PKHA - PKHB = l 0S W) ' l G g [Ff . ( 1 0 ) Hence, to establish an H scale referred to a standard state of i n f i n i t e d i l u t i o n i n water, an acid indicator HA i s required whose pK can be determined i n aqueous buffers and whose io n i z a t i o n range spans the upper end of the pH scale and the beginning of the H_ scale. This indicator i s then used to determine the H_ value for several different solutions e.g. , DMSO-water where the solvent character changes from aqueous to non-aqueous. I t i s then possible to determine i n the same solutions the f r a c t i o n of an indicator HB that i s ionized and by use of equation (10) determine i t s pK . I f the'Hammett postulate holds there should be a constant difference i n the pK a's obtained from equation (10), for two acids i n a series of solutions of diff e r e n t b a s i c i t y . This procedure can be applied repeatedly for successive indicators, and pK 's assigned to these indicators. These individual pK values are then used to calculate H values for each solution i n r a which the indicator's i o n i z a t i o n r a t i o can be measured with reasonable [A\"l accuracy, \"pj^j between 0.1 and 10. This procedure re s u l t s i n several measured H values for each solution and these values are then averaged. - 8 -It i s evident from equation (6) that a plot of log versus H should give a straight l i n e of unit slope and intercept equal to the pK of the indicator. The advantage of using a water-anchored H function, where H • = pH i n d i l u t e aqueous solutions, i s that the proton-abstracting power of any medium i s then always expressed on the same scale. I f i t were possible to determine experimentally a very small amount of i o n i z a t i o n , one should f i n d values s i m i l a r to those i n Table II for the degree of i o n i z a t i o n of a n i l i n e (of the order of 10 i n pure water with [OH ] = 0.01). The v a l i d i t y of t h i s treatment r e l i e s heavily on Hammett's assumption that the a c t i v i t y c o e f f i c i e n t r a t i o of a base and i t s conjugate acid depends only on the solvent composition and i s independent of the nature of the acid indicator. This assumption has been proven quite r e l i a b l e i n determining H functions i n the acid region, although i t has broken down i n certain cases. One of the l i m i t a t i o n s which has arisen i s that the indicators must be of s i m i l a r structure. It i s d i f f i c u l t to define the degree of s i m i l a r i t y required but i t i s known that even primary and t e r t i a r y amines produce di f f e r e n t H q scales. Bowden's review of a c i d i t y functions for strongly basic solutions* l i s t s the H data for 29 different solvent systems. But i n view of the dependence of H functions on indicator structure i t i s o advisable to distinguish between H a c i d i t y functions determined with diverse indicator types. The io n i z a t i o n behaviour i n basic solutions of each of the following type of indicators w i l l be compared: neutral carbon (H^), nitrogen (H1^) and oxygen (H^) acids. - 9 -I I . Carbon Acids Although carbon acids and t h e i r conjugate bases, carbanions, undoubtedly occupy a position of prime importance i n organic chemistry, i t i s only recently that a monograph*^ has been written on the subject of carbanion chemistry, with a detailed discussion on thermodynamic and k i n e t i c a c i d i t i e s of hydrocarbons. 17 C Bowden and Stewart published i n 1965 an H a c i d i t y function based solely on the ion i z a t i o n of carbon acids. The main d i f f i c u l t y i n C establishing a H scale i s that most hydrocarbons are insoluble i n water, even at spectroscopic concentrations 10 M), and thus the anchoring of the scale i n d i l u t e aqueous solutions i s impossible. These authors circumvented t h i s obstacle by f i r s t measuring the pK of malononitrile i n aqueous solution (pK = 11.14) and then i n d i l u t e solutions of 3 ethanolic sodium ethoxide. They then proceeded to measure the pK of two other carbon acids, 9-cyanofluorene and ..methyl fluorene-9-carboxy-l a t e , i n solutions of various concentration of sodium ethoxide i n ethanol. malononitrile 9-cyanofluorene methyl fluorene-9-carboxylate F i n a l l y they anchored the scale i n 0.01 M sodium ethoxide with tris-(p-nitrophenyl)methane / V 3 - 10 -.Subsequent work i n t h i s laboratory, much of which i s described i n t h i s t h e s i s , revealed that for some of these carbon acids the ;carbanion undergoes an unusually large spectral s h i f t (y 100 my) as the ^degree of ion i z a t i o n r i s e s . Indeed, t h i s behaviour was noticed by the -original authors for tris(p-nitrophenyl)methane. 18 C •Kroeger and Stewart i n attempting to set up an H_ scale for -^comparison to t h e i r H - scale found cases where d i s t i n c t l y non-parallel K \" i o n i z a t i o n slopes occurred. 19 •alndependently, Ritchie and Uschold suggested the existence C ^of-flaws i n the H scale. By use of the glass electrode they measured the a c i d i t i e s of a number of weak acids i n DMSO solutions r e l a t i v e to a -standard state i n the same solvent. They ess e n t i a l l y t i t r a t e d potentio-* -metrically these acids with dimsyl cesium solutions, after having -standardized the glass electrode with d i l u t e p-toluenesulfonic acid solutions. The pK 's of fluorene (20.5) and triphenylmethane (28.8) thus measured were i n agreement with values measured by Steiner and 20 ** Gi l b e r t , by a stepwise method with 4-nitroaniline as an anchor. They -measured the pK a's of four of Bowden and Stewart indicators (Table III) and concluded that the overlap of indicators used by Bowden and Stewart •was not v a l i d ; though i t i s not obvious how they arrived at t h i s conclusion. Dimsyl cesium i s the cesium s a l t of the conjugate base of DMSO. 4 4-Nitroaniline was found by Dolman and Stewart to be a non-Hammett indicator slope of log I.vs H 0.92, pK =18.9. - 11 -TABLE I I I 17 19 pKfl of carbon acids i n ethanolic DMSO and i n DMSO Acid ;malononitrile »tris-(p-nitrophenyl)methane '9-phenylfluorene f luorene •Bowden and Stewart r e f . 17. ^ Ritchie and Uschold r e f . 19. Bowden and C o c k e r i l l r e f . 23. 11.14 14.32 -18.59 .'22.1C solvent water DMSO-EtOH DMSO-H20 11.0 12.0 16.4 ,20.5 solvent DMSO Relative a c i d i t i e s of a number of compounds are reversed by the change of solvent from water to DMSO. The pK a's of acetic acid and hydrazoic acid for example, are 11.6 and 7.9 respectively, while i n water they -both have a pK of 4.8. On the other hand the pK of p-nitrophenol i s a a 10.4 i n DMSO and 7.1 i n water. -21 Strietwieser et al. determined equilibrium a c i d i t i e s of various hydrocarbons toward lithium and cesium cyclohexylamide i n cyclo-hexylamine. They measured the equilibrium constants between two hydrocarbons i n cyclohexylamine when treated with an i n s u f f i c i e n t amount of lit h i u m or cesium cyclohexylamide to produce the equilibrium shown i n Equation(11)where HA and HB are two hydrocarbons. HA + B ~ L i + HB + A L i (ID Since a l l the hydrocarbons studied were more acidic than cyclohexylamine equilibrium (12) lay far to the righ t and the spectrum - 12 -AH + Li+NHC6H1j A L i + + C 6H nNH 2 (12) of each hydrocarbon salt could be measured. The ultraviolet and visible spectra of the mixture (11) can then be compared with the spectra of the lithium salts of the two hydrocarbons in equation (12) and knowing the ini t i a l concentrations, the equilibrium constants determined. The main drawback of this procedure is that the conjugate bases involved are actually cesium carbanide ion pairs (contact) or lithium carbanide 21 solvent-separated ion pairs. Streitwieser et al. chose 9-phenyl-22 fluorene (pK 18.59) as their anchor, although they subsequently 3 reduced their values uniformly by 2.3 units to conform to those of 19 Ritchie and Uschold. 20a Steiner and Gilbert have used a similar method to determine the acidity of a few hydrocarbons in both methanolic and aqueous DMSO. Their anchor was 4-nitroaniline pK 18.4 (c*f. footnote page 10). They 3. 2 Ob later re-examined their pK& values and added 1.6 pK units to all the indicators less acidic than fluorene. Although the literature is in a state of some confusion as to the absolute pK& to be assigned to these 22 hydrocarbons, i t seems that a certain consensus has lately been reached between the four different group of workers. Table IV lists the different pK& values published and also the \"best values\". All these values are based on a pK value of 16.4 for 9-phenylfluorene except for Steiner's values which are based on 4-nitroaniline pK =18.4. 1 a 24 C Kuhn and Rewicki determined an H • scale for DMSO-glacial acetic acid - 0.01 M sodium acetate, with various olefinic hydrocarbons - 13 -TABLE IV pK& values of hydrocarbons and \"best values\" Compound Experimental pK& best value h 9-phenyIfluorene 16, .4 18 .5 18, .5 16, .2 18, .59 16, .4 Indene 18. .2 18. .2, 18, .5 20, .24 19, .93 17. .6 4,5-methylphenanthrene >19. ,5 >19, ,5 20. .0 22, .60 22. .63 20. .3 21. .19 2,3-benzofluorene >19, .5 >19, .5 23, .17 20. .9 20. 1 f luorene 20, .5 20, .5 20. .5 22, .83 22, .74 20, .4 22, .10 20, .5 9-phenylxanthene 24, .3 25 .9 28, .5 26, .2 26, .0 triphenylmethane _ 27. .2 28, .8 28, .0 31, .48 29, .2 29, .0 diphenylmethane , 28, .6 30, .2 33, .1 30, .8 30, .5 Steiner and Gilbert, ref. 20a, used as anchor 4-N0oAn., pK 18.4, from Z. 3 Stewart and O'Donnell ref. 13. Steiner and Gilbert, ref. 20b, increased a l l values above fluorene by 1.6, as they found an error in the relative acidities of 9-phenyl-xanthene and fluorene. Ritchie and Uschold, ref. 19, the validity of this scale is based on the agreement with: 20 13 1. Steiner and Stewart on the pK value of 2,4-dinitroaniline, 14.7. 20 a 2. Steiner on the pK& value of fluorene. ^ Streitwieser et al., ref. 21, solvent separated ion pairs. The pKa values were determined with lithiumcyclohexylamide in cyclohexylamine. Streitwieser et al., ref. 21, contact ion pairs. The pK values were determined with cesium cyclohexylamide in cyclohexylamine. f pK values presented by Streitwieser et al., at the symposium in San Francisco ref. 22. ^ Bowden and Cockerill, ref. 23. The pK& values were determined in aqueous. DMSO. h The author's judgement. - 14 -as indicators. The H values of the medium decrease with increase 25 in acetic acid content of the solution. They also measured the pl( of a number of other olefinic hydrocarbons in the above mentioned system, and also in the systems DMSO-tri-n-propylamine, and hexamethyl-25 phosphoramide-tri-n-propylamine. The major drawback of their technique is that the equilibrium between carbanions and hydrocarbon is not instantaneous. Kuhn and Rewicki chose 9-cyanofluorene, whose pK 3 17 (11.4) was determined in ethanol by direct overlap with malonomtrile. 24 Kuhn and Rewicki determined a pK of 19.6 for 9-phenylfluorene based on a pK value of 11.4 for 9-cyanofluorene in DMSO-tri-n-propylamine, 3 which is irreconcilable with the Ritchie and Uschold results. Indeed fluorene itself is half ionized in a 90 mole percent DMSO-water-0.01 M N TMAOH solution, which gives a pK value of 22.0 i f the H scale a — C 23 resembles the H scale, and since the relative acidities of tris-p-nitrophenylmethane, 9-phenylfluorene, and fluorene are the same whether measured potentiometrically or spectrophotometrically (stepwise), i t is not at a l l evident that the pK assignment by Bowden and Stewart is erroneous. Though this does not mean that their (Bowden and Stewart) pK& values are correct, since the various substituted CH-acids are not really similar to the various fluorenes. The nitrophenylmethane anions are much more delocalized than the fluorenes. Moreover, i t should be noted that Bowden and Stewart's extinction coefficients for 9-phenylfluorene, fluorene and tris-(p-nitrophenyl)methane anions do not agree with other reported values. It is thus of prime importance to determine the ionization behaviour of these - 15 -carbon-acids, and hopefully to find a series of hydrocarbons which ionize to give solely carbanions. . III. Nitrogen Acids An H function has been determined for both primary and secondary amines in aqueous dimethylsulfoxide containing 0.01 M TMAOH 4 (tetramethylammonium hydroxide) by Dolman and Stewart . This scale starts at 10 percent DMSO-water with 2,4-dinitrodiphenylamine as the reference indicator, the pK of which was measured using the stepwise 13 method by O'Donnell and taken as 13.84. It appears that a single acidity function will describe the ionization behaviour for both types of amines. The existence of only one H scale contrasts with the fact that separate H q scales are required for primary, and tertiary amines in sulfuric acid water mixtures. This is not surprising, i f the cause of the latter effect is, as suspected, hydrogen bonding between the protons on cationic nitrogen and water.\" H OIL + I 2 R.-N-H. . .OH- R- N-H OH. 3 2 +| 2 H....0H2 Hydrogen bonding of this type in amide ions should be of l i t t l e importance because of the negative charge on the nitrogen atom. Solvation will occur on the nitrogen atom because of its negative charge. - 16 -N 15 An scale has also been established by Bowden, Buckley and Stewart using anionic acids such as diphenylamine carboxylic or sulfonic acids, or aminobenzoic acids which ionize as follows: HA- —•- H+ + A2\" (13) (15) This scale appears to rise slightly less- rapidly than the previously evaluated H scale. In 97 mole percent dimethylsulfoxide-water-0.01 M TMAOH the H value is some two powers of ten higher than the correspond-ing H2 value. This may be caused by the higher solvation requirement of the dianion with respect to the single ion reflecting a relatively large increase in the activity coefficient of the dianion. IV. Oxygen Acids The only oxygen acids that are sufficiently weak to be incompletely ionized in strongly basic media are alcohols and stereo-chemically hindered phenols. The spectrophotometric method is unsuitable for pK determination of alcohols, as they do not undergo significant spectral shifts on ionization. 28 Coggeshall and Glessner studied the ionization of several phenols in ethanol-water solutions containing sodium hydroxide and found that those phenols with t-butyl groups in the ortho position required the highest base concentration to produce complete ionization. This is presumably due to steric inhibition to the solvation of the phenoxide anions. They pointed out that the difference of energy of transition between the ground and first excited state for the neutral molecule and the phenolate ion depends upon the type of phenols. * phenols A(l/A) unhindered 2100 cm 1 partially hindered (2-t-butyl-) 2300 cm 1 hindered (2,6-di-t-butyl) 3200 cm\"1 They tentatively explained this phenomenon by considering the energy of polarization of the large ortho groups. 29 Cohen and Jones measured the dissociation constants for a series of 4-substituted phenols and the corresponding 2,6-di-t-butyl-4-substituted phenols and found that the reaction constant p in water for the hindered phenols is 57% greater than for the unhindered phenols. Cohen and Jones suggested that the extremely weak acidity of hindered phenols may be attributed to the confinement of solvation on the phenolate oxygen, due to the concentration of negative charge on this atom. Cohen and Jones also correlated the ultraviolet spectral data of the 4-substituted and 2,6-di-t-butyl-4-substituted phenols and their corresponding anions with substituent parameters. Since p (ultraviolet * A(l/A) is the difference in frequency of maximum absorption between the molecules and the phenolate ion in each series expressed in wavenumbers. - 18 -spectral s h i f t ) for the 2,6-di-t-butyl series i s only s l i g h t l y greater than p for the hindered phenols they concluded that the phenolic 0-H bond i s coplanar with the aromatic r i n g . Further proof for the i n h i b i t i o n to solvation by ortho 30 substituents was provided by Fischer et a l . who measured the d i s s o c i a t i o n constants of 4-substituted-2,6-dichloro and 2,6-dimethyl phenols i n aqueous solutions. Since the chloro- and methyl-substituents are of o s i m i l a r size (van der Waals r a d i i are 2.0 and 1.8 A respectively) but have opposite inductive effects simple s t e r i c i n h i b i t i o n to solvation of the phenoxide anion should give s i m i l a r reaction constants P for both series i f the ortho substituents exhibits no, or very l i t t l e , polar effect. Fischer et al.found a p value of the same order of magnitude for both series and higher than for unhindered phenols. The order of decreasing values i s 2,6-di-t-butyl 3.50, 2,6-dimethyl 2.70, 2,6-dichloro 2.61, phenols 2.24, which i s consistent with the expected magnitude for purely s t e r i c e ffects, thus confirming the suggestion that i n h i b i t i o n to solvation and not the polar effect of the a l k y l group i s the cause for the acid-weakening effect of ortho substituents i n hindered phenols. 31 Rochester has recently studied the i o n i z a t i o n of s i x t-butyl substituted phenols i n methanolic sodium methoxide solutions, and found the results to be consistent with an a c i d i t y function which d i f f e r s from that obtained using anilines and diphenylamines. Indeed, he suggests that a small difference exists between the functions governing the i o n i z a t i o n of p a r t i a l l y hindered and hindered phenols. Rochester used *32 the method of O'Ferrall and Ridd and obtained pK values based on r a * p l o t t i n g log (I-[0Me ]) vs [OMe ] and extrapolation to zero concentra-t i o n of sodium methoxide. - 19 -methanol as the standard state. The H° function differs by up to 0.5 unit in 3 M sodium methoxide from the function, the latter being the greater. It would thus be interesting to compare the H acidity functions for amines and phenols. The most appropriate solvent system to use in this study would be methanolic and/or ethanolic DMSO since most phenols are completely ionized in 0.01 M aqueous base. V. Application of the H Function to Kinetics and Mechanisms of Base Catalysis As in the case of the HQ function, the main purpose of establishing an H function, besides equilibrium measurements of weak acids, is its possible correlation with rates of base-catalyzed reactions. These correlations are indeed a powerful tool in the elucidation of reaction mechanisms. Few such correlations have appeared in the l i t -erature and they in general involve breaking of a C-H bond in the rate determining step. The rate of racemization of (+)-2-methyl-3-33 phenylpropionitrile in dimethylsulfoxide containing sodium methoxide N has been correlated with the H function of the medium. A log k vs H plot is approximately linear with a slope of 0.87 over a range of rates 34 greater than six powers of ten. A correlation is also obtained between the logarithm of the rates for the methanolysis of chlorinated hydro-carbons and the H function of methanolic sodium methoxide solutions. A reasonable straight line is obtained with a mean slope of 0.8. It 35 was suggested however, that this correlation was fortuitous because of ion association in concentrated solutions of sodium methoxide in methanol. - 20 -36 Jones and Stewart have reported a correlation between the 3 logarithm of the ionization rate constant of (a- H) acetophenone and the H of the medium. The systems studied were: DMSO-water-0.01 M TMAOH, DMSO-methanol-0.01 M sodium methoxide, DMSO-ethanol-0.01 M sodium ethoxide. The slopes of these correlations though are only ^ 0.5 but this might be accounted for by the fact that the H function for the ionization of a ketone must be quite different from the ionization of an amine. Keto-enol tautomerism should render an H function more sensitive to DMSO addition than a hydrocarbon or an amine. Jones and 37 Stewart correlated also the logarithm of the rate of ionization of DMSO in a DMSO-water-0.01 M TMAOH solutions with the HN of the medium. They observed a linear relationship with a slops of 0.9. 38 Anbar et al. derived equations which correlate the rate constants of various base-catalyzed reactions with the H value of the medium. If the first step of the reaction is a proton abstraction, Anbar et al differentiate between the following cases: a) A rapid preequilibrium is established between the base and the substrate followed by a first order rate determining step to give the products R-H + 0H~ ^ R\" + H20 n- slow , R >- products In this case the logarithm of the rate constant correlates with the H . The above mentioned alkaline decomposition of chloroform is an example of such a case. (16a) (16b) b) A rapid preequilibrium i s also established as i n case a) but i t i s followed by a second order rate determining reaction with another reactant Y R- + Y s l 0 W> products (17) Here also there exists a linear relationship between the logarithm of the rate constant and the H . The reaction between chloramide and ammonia i n aqueous solutions to form hydrazine i s an example of such a correlation and has a slope of 0.9. b') I f Y i s a solvent molecule and thus the reaction i s pseudo f i r s t order,the logarithm of the rate constant should correlate with H +log Cj. Q. c) In the case where the proton abstraction i s the slow step, followed by a fast reaction to give the products, Anbar et a l suggested that the rate constants w i l l also correlate with H_ + log C„ n-2 S. H + OH\" S l 0 W> S\" + H20 (18a) S\" f a S t > products CI8b) This s i m i l a r i t y i n rate correlation between case b') and c) is not surprising since the t r a n s i t i o n states i n both of these cases should be si m i l a r . The decomposition of dl-serine phosphate to form pyruvate phosphate and ammonia i s an example of such a reaction type. OH\" + \"2P03OCH2CH(NH2)COO\" — - CH^OCOO\" + NH^ + PO3\" (19) - 22 -A plot of log k against H + log ^ for this decomposition has a 2 ' 2 slope of 0.98. More recently Cram and Kollmeyer have correlated the logarithm of the rates of potassium methoxide catalyzed isotopic exchange of carbon acids with the H_ of the medium. They proposed two mechanisms to account for the linear relationship observed. (a) The first one involves a slow ionization step RH + CH30~ ^y%- R + CH3OH f a S t> product (20) This mechanism occurs when APK (pK, , , -pK Jt, .) has a low r r hydrocarbon r methanol positive value. The transition state in this case can be represented by DOCH I i _ C H 0-CH, I 3 (b) The second mechanism involves a rapid preequilibrium followed by a rate-limiting reaction with the solvent RH + CILO Ke slow 3 «- R + CH30H ^ ' product (21) This mechanism occurs when ApK is large and the transition state in this case resembles H0CH3 — C'r '•DOCH. - 23 -For both of these reaction paths Cram et al.suggested that there i s a linear relationship between log ^ 0^ s and H . The H functions used 2 3 2 3 3 3 5 3 6 3 7 i n these correlations » » » » » have been derived using diphenyl-amine and a n i l i n e indicators, and thus r e f l e c t the effect of changing medium on nitrogen acids rather than on carbon acids. This i s reflected i n the fact that the slope e of the correlation was always found to be lower than unity. 39 Kresge suggests that k i n e t i c a c i d i t y dependence i s a property of the substrate and should not be interpreted as a mechanistic c r i -t e r i o n . He believes that such correlations can be used to provide an insight into t r a n s i t i o n state structure i n reactions of known mechanism, rather than to provide evidence useful i n choosing between two or more possible mechanisms. A. Kinetic-Thermodynamic relationships 'The rate of proton transfer i n the ioniz a t i o n of carbon acids has been looked upon as a measure of the k i n e t i c a c i d i t y of these acids. The ki n e t i c a c i d i t y i s defined as the forward rate of exchange., on a r e l a t i v e logarithmic scale R-H + B -r±-* R + BH , (22 ) k but u n t i l recently no systematic study of kinetic-thermodynamic relationships had been made for simple hydrocarbons. This i s because they are much too weak as acids to ionize i n common basic aqueous solutions, 4 3 Special experimental techniques had to be developed i n order to study isotope exchange rates of hydrocarbons i n l i q u i d ammonia and i t i s only - 24 -i n the l a s t decade that the use of non-aqueous solvents as reaction media has gained a certain popularity. Following Eigen's treatment, i t i s advantageous to consider that an intermediate (hydrogen-bonded) complex i s formed between the acid and the base as i n equation (23): - k l - k2 -R-H + B , . R-H---B B-H + R (23) k - l k-2 Normally the concentration of the complex R-H B i s very small compared with the concentration of either reactants or products. Using the steady-state assumption the forward (k£) and reverse (k r) rate constants can be expressed as follows: k., k _ k ^ k -k. = 1 .— k = ~* - 1 f i r k _ 1 + k 2 k - l + k 2 The equilibrium a c i d i t y i s related to the r a t i o of forward and reverse rate constants, usually as the negative logarithm, or log k £ - l o g k r = p K ^ - pK B H = A(pK) (24) The well-known Bronsted equation relates the k i n e t i c and e q u i l i -brium a c i d i t i e s . In i t s simplest form i t i s written: log k A = log G A - a log K & (25) where k. i s the rate constant for acid catalysed reaction and K the A . a - 25 -equilibrium acidity of the acid A, or of the conjugate acid of the base B. and a are constants. Considering equation (24), one may rewrite 40 the simple Bronsted relation (25) in the following form d log k f — — = a (26) d A (pK) where a is the Bronsted slope. It is evident from equation (26) that a is zero i f A (pK) has a large positive value and unity i f A (pK) is large negative. In the transition region between these extremes where A (pK) - 0, a has an intermediate value between 0 and 1.* The integrated form of the Bronsted equation (equation 25) was . proposed in 1924 to correlate the catalysed decomposition of nitramide as 41 shown in equation (27). Bronsted emphasized the fact that his equation is limited to a study of reactions catalysed by a small number of compounds H2 N2°2 > H2° + N2° ( 2 7) with comparable acid strength. Indeed Eigen^ pointed out that the decomposition of nitramide shows a continuous variation of the coefficient a i f studied over a wide range of reactivities. Even in the classical case of the base-catalysed halogenation of 42 ketones , Eigen showed that in a wider pK range a is not constant and that all the curves can be constructed on the basis of the generalized relationship (26). At high A (pK) value - whether negative or positive - the rate of proton transfer is diffusion controlled. In the vicinity of A (pK) = 0 * 55 Bordwell et al. recently measured Bronsted coefficients larger than unity and smaller than zero for proton abstraction of nitroalkanes. - 26 -there is a transition where log k is linearly related to A (pK) and takes a value between 0 and 1. Normal acid-base systems are those which show a sharp narrow transition region. Such acids are usually of the OH- (phenols), NH- (aniline), NH- (pyridinium ion) type, whose pK values lie in the range -1 to 15. CH-acids (or pseudo-acids) and some normal acids with special struc-tural characteristics (internal hydrogen bonding for example) have broad, transition regions. This transition from a = 0 to a = 1 will occur grad-ually so that a might appear to have a constant value. Carbon-acids show wide transition regions because usually deproton-ation is associated with resonance stabilization and the recombination rate may therefore be relatively low. Many examples of curvature of log k -A pK relationships have been found and i t is now a well accepted fact that Bronsted plots are usually non linear i f a wide enough range of reactivities is examined. The log k-A pK relation approaches ideal be-havior when: (i) both donor and acceptor belong to the classical hydrogen bond formers (0- and N-type). (ii) the electronic and spatial configurations of the acids and their con-jugate bases are similar. If the limiting values of k are identical (the extrapolated straight lines at a = 0 and a = 1 intersect at A pK = 0) then the slope of log k-A pK plot at A pK = 0 is often close to 0.5. If how-ever, the limiting values of k differ, then the center of symmetry--point of intersection of the straight line a =0 and a = 1 --is displaced along the A pK axis by the difference of the logarithms of these limiting value. The closer the value of k at A pK = 0 is to the limiting value, the nar-rower is the transition region, where a appears to be constant. This - 27 -symmetric behavior with respect to A pK = 0 i s only v a l i d for reactions of symmetric charge type. HX + Y~ * X\" + HY (28a) XH + + Y — X + YH + (28b) However i f the reaction shows an asymmetry of charge, the log k-A pK curve becomes asymmetric with respect to A pK = 0.' HX + Y — * X\" + YH + (28c) 22 Recently Streitwieser et a l . measured the k i n e t i c a c i d i t i e s of a series of hydrocarbons by hydrogen isotope exchange i n methanolic sodium methoxide. The thermodynamic a c i d i t y of these same hydrocarbons have already been measured i n cyclohexylamine by him, (cf. p. 12-13). A Bronsted plot of log k~ (second order rate constant) vs. pK i n cyclo-hexylamine showed d i s t i n c t curvature with a varying from about 0.4 near the fluorenes to 0.8 for d i - and tri-arylmethanes. Cram and Kollmeyer also published a kinetic-thermodynamic a c i d i t y correlation of carbon acids. Rate constants and activation para-meters for potassium methoxide catalyzed exchange of a few hydrocarbons i n methanol-O-d-dimethyl-d^-sulfoxide (~ 15 mole percent) were measured. These k i n e t i c a c i d i t i e s varied by a rate factor of seven powers of ten, and the thermodynamic a c i d i t i e s by a factor of ten powers of ten. The Bronsted plot was also curved. The slope of the log k vs. pK & curve was ~ 0. for the strongest acids (pK = 23 ± 3) and unity for the three weakest acids (pK a =28 + 2). - 28 -44 'Shatenshtein et al> have studied extensively isotopic exchange -reactions of hydrocarbons i n non-aqueous solvents, at f i r s t mainly i n 43 -liquid ammonia , but l a t e r also i n cyclohexylamine, dimethylsulfoxide 44 45 46 44 and others ' ' . Shatenshtein proposed a two-step mechanism s i m i l a r to the one shown i n equation (23). The f i r s t stage of t h i s reaction i s the l i m i t i n g stage, and thus k ^ = k^, i . e . the observed rate constants of the reaction i s equal to the rate constant for the .^ionization of the R-H bond. I t i s only i n t h i s case that the ki n e t i c s of hydrogen exchange can characterise the a c i d i t y of t h i s C-H bond. 44 ''\"^Shatenshtein suggested that there i s no direct r e l a t i o n between the .^observed rate constant of the exchange reaction and the ioni z a t i o n rate ^constant, i n the case where the second step i n equation (23) i s rate k ^determining; since k ^ g = 1 k., = Kk^ . k - l 44 45 Shatenshtein ' also pointed out that the r e l a t i v e k i n e t i c -acidities can vary with change i n the solvent medium. Very large d i f f e r -ences i n r e l a t i v e k i n e t i c a c i d i t i e s were found i n comparing the kin e t i c s of deuterium exchange i n DMSO and i n ammonia. For example the rate ^constants f o r the exchange of deuterium i n the methyl group and i n the r i n g of toluene i n DMSO d i f f e r by a factor of 10^ while they d i f f e r only 2 by a factor of 10 i n ammonia. He dif f e r e n t i a t e s between C-H bonds i n aromatic or heteroaromatic rings and those i n methyl groups attached to aromatic rings, when p l o t t i n g r e l a t i v e strengths of carbon acids i n one. solvent against the r e l a t i v e strength of the same acids i n another solvent. 44 Shatenshtein found that there i s no d i f f e r e n t i a t i n g effect when the solvents belong to the same category (aprotic or p r o t i c ) , for example DMSO ys. methyl-bismethoxymethyl-phosphine oxide, (CH^OQ^)2PO.CH.J, or 45 ammonia vs. cyclohexylamine. He found though, that i f the solvents - 29 -do not belong to the same group, DMSO vs. t-butanol or DMSO vs. ammonia, the points for the alkyl groups in alkylbenzenes lie on one line while those for benzene derivatives containing ring deuterium lie on another line. Shatenshtein concludes that the differentiating action of solvents on the kinetic acidity of carbon acids depends on the structure and the solvation of the transition states formed during the isotopic exchange reactions. He also points out that the problem of determining acidities (thermodynamic and kinetic) of hydrocarbons is s t i l l quite complicated due to our limited knowledge of solvent effects. B. Isotope effects Primary kinetic isotope effects for the base catalysed proton ab-straction from aromatic hydrocarbons have been measured by many workers. 47 Streitwieser et a l . have measured the kinetic isotope effect k^/k^ for the exchange of the methyl hydrogens of toluene in the cyclohexylamine r k cyclohexylamide system. They found values greater than 10 for H/k^ ,, 48 while in the DMSO - potassium-t-butoxide system Hofmann et al \\ measured k a D/k^ slightly less than unity. Hofmann et al believed that their results showed that the rate determining step was not ini t i a l ionization of the C-H bond, but the formation of a tight complex between the incip-ient carbanion and the alcohol. 49 50 This view was challenged by J.R. Jones . Both theory and exper-51 52 53 iment ' * indicate that the isotope effect for proton transfer be-tween acid-base systems has a maximum value when the A pK& value between the two acids is equal to zero. In Streitwieser ?s case the pK& of t-butanol is ^ 20, while the pK of toluene is ^ 40. Thus, we should expect to have a high isotope effect in cyclohexylamine, and a negligible - 30 -one i n dimethylsulfoxide. Kinetic isotope effects for proton abstraction reactions 2 22 have been measured for other hydrocarbons ' but these w i l l be discussed l a t e r i n t h i s thesis. - 31 -SCOPE OF THE PRESENT RESEARCH As emphasized i n the introduction, the indicators used to determine H functions can be di f f e r e n t i a t e d into oxygen, nitrogen and carbon acids. Part of th i s work was undertaken to determine the degree of divergence of a c i d i t y functions caused by v a r i a t i o n i n N indicator structure. The io n i z a t i o n behaviour of diphenylamines (H ) and phenols (H^) were therefore determined i n dimethylsulfoxide-methanol containing 0.01 molar sodium methoxide and dimethylsulfoxide-ethanol containing 0.01 molar sodium ethoxide. These weak acids exhibit a strong bathochromic s h i f t on i o n i z a t i o n , and t h i s provides a simple way of determining the position of t h e i r i o n i z a t i o n e q u i l i b r i a . N A redetermination of the H function i n low percentage dimethylsulfoxide i n water was also considered necessary as some doubts were raised on the ioni z a t i o n mode of the previously used anchor,. 2, 4 , 6 - t r i n i t r o a n i l i n e . . Preliminary results indicated that the p-nitro substituent had an overwhelming e f f e c t . I t was thus important to study the e f f e c t of substitution on the a c i d i t y of diphenylamines i n the presence of the n i t r o group i n the other r i n g . The main purpose i n establishing an H function, apart from i t s use i n equilibrium measurements of weak - 32 -acids, i s i t s possible correlation with rates of base-catalyzed reactions. Base-catalyzed rates of d e t r i t i a t i o n of various hydrocarbons were determined i n dimethylsulfoxide mixtures. These rates (on a logarithmic scale) were then correlated with the thermodynamic b a s i c i t y of the medium as measured by the H function. The shape of these correlations would then give us information on the applica-b i l i t y of these H functions, and on the structure of the t r a n s i t i o n state i n base-catalyzed proton abstraction reactions. Hopefully a kinetic-thermodynamic relationship might also be derived from these rate measurements. In view of the d i s p a r i t y between published pK and r e l a t i v e pK values of carbon acids, and the disagreements on spectral c h a r a c t e r i s t i c s of t h e i r carbanions formed on i o n i z a t i o n , a close examination of the ion i z a t i o n behaviour should be undertaken. I t was of prime importance to prepare and characterize the carbanions of these nitrophenylmethanes and determine t h e i r spectral c h a r a c t e r i s t i c s . One could then decide whether these carbon acids had ionized by proton abstraction i n previously reported work. Also, the cause of the dependence of spectral characteristics on the solvent should be investigated. EXPERIMENTAL Most of the handling techniques pertaining to the determina-4b tion of the H functions are described in full detail elsewhere. An outline of these procedures is given below. I. Purification of solvents DMSO was stirred in a closed vessel over powdered calcium * hydride for at least two days and then distilled in a stream of nitrogen at reduced pressure. A Perkin triangle was used to cut fractions, and the first and last 20% of the distillate were discarded. The DMSO was stored in glass stoppered flasks over molecular sieves 4A and under nitrogen. The water content of the DMSO was measured prior to its use,, by the Karl-Fischer method, and was shown to be less than 0.01%. Hexamethylphosphoramide (HMPT) was fractionally distilled twice from molecular sieves 13X. The first and last 20% of the d i s t i l -late were discarded each time. As HMPT dissolves most commercial vacuum greases, a distillation apparatus with \"clear-fit\" joints was used. The distilled HMPT was stored in stoppered bottles over molecular sieves 4A. Carbonate-free water was prepared by boiling distilled water and then bubbling nitrogen through as i t cooled. * The nitrogen gas used throughout was grade \"L\" (Liquid Air) as grade \"G\" contains small amounts of oxygen. - 34 -\"Super dry\" methanol and ethanol were prepared according to 54 standard procedures i n Vogel and were stored over molecular sieves 3A and under nitrogen. Dryness was checked p r i o r to use by Karl-Fischer t i t r a t i o n . toluene was Baker analyzed reagent grade and was d i s t i l l e d before use. I I . Base Tetramethylammonium hydroxide was obtained from Eastman Organic Chemicals either as a 10% aqueous solution or as the c r y s t a l l i n e penta-hydrate and was used without further p u r i f i c a t i o n . Sodium methoxide and ethoxide were prepared from the required amount of sodium and the corresponding alcohol to produce a 1 M solution. . This was done i n a rubber stoppered b o t t l e under a stream of nitrogen. After the reaction was complete aliquots of the stock solutions were t i t r a t e d against standard acid. The stock solutions were kept under nitrogen and stored i n the freezer compartment of a re f r i g e r a t o r . No solution older than 1 month was ever used. I I I . Preparation of solutions The DMSO-protic solvent (H 20, MeOH, EtOH) stock solutions were made up according to weight at approximately 5 mole percent intervals from 0 to 100 percent DMSO. Dry NO-SOL-VIT bottles with well f i t t i n g rubber stoppers - sleeve type - were weighed. Into each of these was syringed the required volume of DMSO, and the bottles were - 35 -reweighed to give the weight of DMSO. To this was syringed the appropri-ate amount of protic solvent and the bottles weighed for the final time to give the weight of the solvent. Nitrogen was then bubbled through these solutions for 3 minutes and they were then stored in a desiccator over Drierite. It was found that the amount of moisture which was picked up by the solvents during preparation was very small and did not warrant the extra precaution of using a drybox. IV. Spectral measurements Indicator measurements were made with a constant concentra-tion of base (0.011 M tetramethylammonium hydroxide or 0.01 M sodium alkoxide). All measurements of spectra were made on a Bausch and Lomb model 502 recording spectrophotometer, with the cells thermostated at 25° by means of a constant temperature cell holder. The procedure 13 developed previously by O'Donnell and used since then in these laboratories to protect the system from oxygen was followed. This involves the use of silicone rubber disks to seal absorption cells and syringes to introduce the indicators and the base. Dry nitrogen was bubbled through the solution for about 2 minutes. None of the indicator anions used seemed to react in any way for at least 15 minutes, as.their spectra showed no change during this lapse of time. V. Treatment of data and calculations From the spectral measurements on solutions of indicators in the aprotic-protic-base system, the ratios of the concentrations of ionized to unionized indicator (I =.(A )/(HA)) are obtained. From these - 36 -ra t i o s are then obtained the r e l a t i v e pK values of the amine indicators 1 a and the H values for the solutions i n which the measurements were made. The r e l a t i v e pK values were obtained by comparing the ion i z a t i o n ratios of overlapping indicators i n the same solution. The quantity log I was plotted against the solvent composition for each indicator and a smooth curve drawn through the points for each p l o t . Values of log I between -1 and +1 corresponding to 10% and 90% ioniza-t i o n , respectively, were used i n the plots. Where successive curves overlapped, differences were taken at regular intervals and the results averaged. These averages gave the differences between the pK a values of the indicators. The H values were then calculated from the pK values of - r a the indicators and the values of log I obtained from the smoothed curves mentioned above. This was accomplished by using equation (6). H- = P K a + ( 6 ) For a given s o l u t i o n , H i s generally an.average of the values obtained from two or more indicators. VI. Probable error i n H_ and pK a values With a l l a c i d i t y functions i t i s d i f f i c u l t to estimate the uncertainty i n the H values and i n the pK values of the indicators used to determine them. The further the a c i d i t y function i s from the usual pH range, the greater the li k e l i h o o d of errors because of the - 37 -stepwise procedure used to establish i t . The errors can be minimized by using a large number of overlapping indicators so that the H value for a given solution i s not dependent on the data from any one indi c a t o r , and by using indicators of the same structural type to ensure that the Hammett a c t i v i t y c o e t f i c i e n t postulate i s obeyed as closely as possible. 4 The previous estimate of the uncertainty as 0.05 pK unit appears to be on the low side. A more reasonable estimate would be around t 0.2-0.3. It i s to be noted that the precision of the UV measurement far exceeds the method's overall accuracy. A change i n absorbance of 0.02 alters log I by only ^ 0.04 i f -0.5 < log I < +0.5 and by 0.08 i f -0.5 > log I > +0.5. It was also noticed that solutions of indicators can be kept for an extended period of time (12 months) without any damage. . VII.Kinetic measurements ,. A. Hydrocarbon exchange A weighed amount (10-30 mg) of t r i t i a t e d hydrocarbon was weighed i n a 125 ml erlenmeyer with a B24 stopper, and 100 ml of the appropriate mole percent DMSO-alcohol solution was introduced into the fl a s k . After complete dissol u t i o n of the hydrocarbon was assured, the flas k was immersed i n a w e l l - s t i r r e d thermostated bath. After s u f f i c -ient time was allowed for temperature e q u i l i b r a t i o n , 1.0 ml of a 1 M solution of sodium ethoxide i n ethanol was added. Aliquots were withdrawn at fixed time intervals and introduced into a 125 ml separatory funnel containing 50 ml of cold water (to stop the reaction) and 15 ml of l i q u i d s c i n t i l l a t o r (3.5 gm of 2,5-diphenyloxazole per - 38 -litre of toluene). The funnel was shaken by hand and the aqueous layer run off. The toluene solution was dried over anhydrous magnesium sulfate. An aliquot (10 ml) of the dried toluene was transferred into a counting vial. The latter was stored for 20 min. then counted in a Nuclear Chicago Mark I model 6860 liquid scintillation counter. The counting efficiency was roughly determined for each sample by the external standard technique utilizing the built-in barium-133 gamma source. All vials which did not have the average efficiency were discarded. The usual count at time zero was around 500,000 counts per minute. In general the reaction was taken to ^ 70% completion. However for each hydrocarbon, at least one run was allowed to proceed to greater than 90% completion. The base concentration was determined by simple titration against standard acid using phenolphthalein indicator. The tritiated hydrocarbons were prepared (cf. preparation of compounds for details) by one of the following methods: 1. quenching of an ethereal solution of the sodium salt of the hydrocarbon with HTO, 2. catalytic exchange between the hydrocarbon and HTO*^ . B. DMSO exchange Solutions {y 30 mis) of appropriate mole percent DMSO-protic solvent were prepared as previously described and 0.1 ml of tritiated DMSO or protic solvent, as the basicity of the medium warrants, was added. After thermal equilibration 0.3 ml of 1 M base solution was added. If the reaction was fast the base concentration was determined - 39 -by simple titration. Aliquots (3 mis) were withdrawn and injected into tubes containing 7 ml of methanol. The CH^ OT thus formed was separated from the DMSO by fractional distillation, dried and then added to 7 ml of the liquid scintillator, cooled and counted. If the reaction was slow 3 ml aliquots were withdrawn into vials and sealed under nitrogen and kept in the constant temperature bath. At appropriate time intervals, the vial was broken and 7 mis of MeOH was added and the same routine followed. Tritiated DMSO was prepared by allowing 20 gm of DMSO to equilibrate with 0.1 ml of HTO (100 mcuries/ml) in the presence of a pellet of sodium hydroxide for 6 hours. The contents were neutralized with hydrochloric acid and 10 ml of methanol was added. The solution was dried, the methanol was removed by distillation and the DMSO was purified by vacuum distillation. Tritiated methanol and ethanol were prepared by allowing 10 ml of alcohol to equilibrate with 1 ml of HTO (4 mcurie/ml) for a half hour. The alcohol was then fractionally dis-tilled and dried. VIII.Treatment of the kinetic data The pseudo first order rates for the detritiation of hydro-carbon are described by equation 29. k = IWTt l o g [c] C29) where the [C] terms refer to the radioactivity of the hydrocarbons at various time intervals. No infinity correction was made as the hydrocarbon Figure 1. Typical f i r s t - o r d e r rate plots for the d e t r i t i a t i o n of 9-phenylfluorene i n absolute methanol at 25°. - 41 -loses almost a l l of i t s a c t i v i t y to the solvent sink. An i n f i n i t y point showed an a c t i v i t y of % 1000 c/min which i s negligible with respect to the average a c t i v i t y of 200,000 c/min. Eight points were obtained from each run and k was determined from the slope of the plot of log C against t . Duplicate runs showed the rate constants to be i n agreement within t 5%. A representative plot of a duplicate d e t r i t i a t i o n measurement - 9-phenylfluorene i n absolute methanol - i s shown i n figure 1. In the case of the DMSO exchange the equations are: V V 2-303 , [(MeOT)o - (MeOT)J t r i t [OR J t [ ( MeOT) o - (MeOT) ] V v 2.303 , [(MeOT)~ - (MeOT)p] • Writ \" [OlFTT l 0 * [ ( M e 0 T ) r o . ( M e 0 T ) t ] ( 3 1 ) A t y p i c a l plot for the t r i t i a t i o n of DMSO - i n 55% DMSO-water - i s shown i n figure 2. In the case where the base concentration dif f e r e d from the standard value 0.01 M a correction was made using the equation: H_ = H_ (0.01) > log J£jLl (32) The correction was never more than 0.1 unit. This correction i s warranted by the fact that the H_ function for various concentrations of pr o t i c solvent and i t s conjugate base follow t h i s equation for d i l u t e basic solutions. - 43 -IX. Reactions of nitrophenylmethanes a. Nuclear magnetic resonance (n.m.r.): A solution of the nitrophenylmethane in HMPT was prepared in a drybox under nitrogen atmosphere. An n.m.r. spectrum of this solution was then recorded on a 100 M.Hz. Varian H.A.-100 spectrometer. A small amount of solid sodium methoxide was added to the solution and another spectrum was recorded immediately. The solution was then filtered through glass wool into another n.m.r. tube and a subsequent spectrum was recorded. Aside from the fact that the filtered solution gave \"cleaner\" spectra than the unfiltered one, there was no difference between them. This same solution was then used for the determination of the spectral characteristics of the anionic species. Chemical shifts are reported in p.p.m. from an internal tetramethylsilane reference. A Cary spectrophotometer Model 16 was used for the measurements of the visible absorption spectra (up to a wavelength of 800 my), and a Cary recording spectrophotometer Model 14 was used in the near infrared region (beyond 800 mu). b. Electron spin resonance (e.s.r.) Tetra-n-propylammonium perchlorate (TPAP) was used as supporting electrolyte, and was prepared from tetra-propylammonium hydroxide and perchloric acid. The product was recrystallized from a (1:9) DMSO-water mixture. The electrolysis cell used in this work was designed by D.E. Kennedy^1and spectra were recorded on a Varian E. 3 spectrometer. The nitrophenylmethane and TPAP were weighed in. dry NO-SOL-VIT bottles, which were then fitted with rubber stoppers - 44 -(sleeve type). Aliquots of the appropriate mole percent DMSO-methanol solution were injected into these bottles. After complete dissolution of the solid nitrogen was bubbled through the solution for 10 minutes. The solution was then transferred to the electrolysis cell and degassed by connecting the cell to the vacuum system. The concentration of the supporting electrolyte was 0.1 M and that of the nitro-phenyl-methane ^ 0.01 M. The potential applied to the cell for electrolysis was measured with respect to the mercury pool electrode, and was 2^ volts with a current of ^ 50 u amp. i X. Computer treatment of data A least square program with a plotting routine was used for treatment of much of the data. I would like to thank Messrs. A.M.. Smolensky and D.E. Kennedy for help in writing the program. All the figures in this thesis are reproductions of plotting output. The program and a sample of input and output is listed in Appendix C for reference. XI. Preparation of compounds used in this work A. Amines a. 2,4-dinitrodiphenylamine series:-2,4,4'-Trinitrodiphenylamine: was prepared by nitration of 2,4-dinitrodiphenylamine with nitric acid-acetic acid mixture accord-57 ing to the method of M.P. Juillard. The product was recrystallized from toluene, acetic acid and finally from alcohol, melted at 187-188°. Lit. m.p. 188°. 5 7 - 45 -I 12 2,4,3\"-Trinitrodiphenylamine: has been prepared and purified previously in this laboratory. M.p. 192-194°. Recrystallized from pyridine. 2,4-Dinitro-4'-trifluoromethyldiphenylamine: has been synthesized and purified by Dr. A. Buckley in this laboratory by condensation of the chloro-dinitrobenzene with p-trifluoromethylaniline. M.p. 125-126.5°. Recrystallized from acetic acid and ethanol. C H . F N Analysis calcd. 47.71 2.46 17.42 12.84 found 47.57 2.62 17.16 12.01 2,4-Dinitro-31-trifluoromethyldiphenylamine: has been synthesized and purified by Dr. A. Buckley in this laboratory. M.p. 123-125°. Recrystallized from acetic acid and ethanol. C H F N Analysis calcd. 47.71 2.46 17.42 - 12.84 found 47.79 2.98 17.27 13.05 2,4-Dinitro-31-chlorodiphenylamine: has been prepared and purified by Dr. A. Buckley in this laboratory. M.p. 180-182°. 58 Recrystallized from ethanol. Lit. m.p. 182-183°. 2,4-Dinitrodiphenyiamine: obtained from Eastman Organic Chemicals and recrystallized from ethanol. M.p. 155-156°. Lit. m.p. 155°. 5 9 2,4-Dinitro-3'-methyldiphenylamine: obtained from Aldrich Chemicals and recrystallized from ethanol. M.p. 161-162°. Lit. m.p. 160°. 6 0 2,4-Dinitro-4'-aminodiphenylamine: obtained from Columbia Organics passed down an alumina column with a benzene/chloroform (1:1) mixture, then through another alumina column with chloroform as eluent. - 46 -M.p. 183-184°. Lit. m.p. 189°. 6 1 C H N Analysis calcd. 52.55 3.65 20.44 found 52.76 3.75 20.20 12 (cf. spectral data as they differ from published values). b. 4-Nitrodiphenylamine series:-4,4'-Dinitrodiphenylamine: has been prepared and purified 12 previously in this laboratory and was recrystallized from dioxane- . water pair. M.p. 213-216°. 4 3,4'-Dinitrodiphenylamine: has been prepared and purified previously in this laboratory and was recrystallized from benzene- . petroleum ether pair. M.p. 215-216°. 4-Nitro-3'-trifluoromethyldiphenylamine: was prepared by condensation of 2-chloro-5-nitrobenzene sulfonic acid with m-trifluoro-methylaniline followed by desulfonation according to the method of 6 2 F. Ullmann and R. Dahmen. (Another sample was previously prepared by Dr. A. Buckley in this laboratory). The product was from acetic acid and ethanol. M.p. 137.5-139°. C H N F Analysis calcd. 55.31 3.19 9.93 20.21 found 55.42 3.32 10.00 19.92 4-Nitro-3'-chlorodiphenylamine: prepared by the above mentioned method and purified by Dr. A. Buckley in this laboratory. Recrystallized from ethanol. M.p. 129-130°. Lit. m.p. 129°. 4-Nitrodiphenylamine: obtained from Aldrich Chemicals and was recrystallized from acetic acid. M.p. 131-133°. Lit. m.p. 133°. - 47 -4-Nitro-3'-methyldiphenylamine: synthesized and purified by Dr. A. Buckley in this laboratory. M.p. 127-128°. Recrystallized from ethanol. C H N Analysis calcd. 68.40 .5.30 12.28 found 68.24 5.29 12.12 4-Nitro-4'-aminodiphenylamine: prepared and purified by fV Dr. A. Buckley in this laboratory. M.p. 206-207°. Lit. m.p. 207-208°. c. Aniline series:-2,4,6-Trinitroaniline: was obtained from Eastman Chemicals and recrystallized from ethanol, acetic acid. M.p. 191-192°. Lit. m.p. 192-195°.63 2,4-Dinitroaniline: was obtained from Eastman Chemicals and 63 recrystallized from ethanol. M.p. 180-182°. Lit. m.p. 180°. 4-Chloro-2-nitroaniline: was obtained from Brothers Chemicals Co. and recrystallized from ethanol. M.p. 115.5-116.5°. Lit. m.p. 116°. 6 3 B. Phenols All the phenols used were commercially available and were recrystallized from absolute ethanol under a nitrogen atmosphere and sublimed before a final recrystallization under nitrogen. C. Hydrocarbons 2,4,2',4'-Tetranitrodiphenylmethane: was prepared in this 64 laboratory according to the method of Parkes and Morley and recrystal-- 48 -lized from acetic acid. M.p. 173-174°. Lit. m.p. 173°. 4,4' ,4\"-Trinitrodiphenylmethane (-tris-(p-nitrophenyl)methane) : was prepared by direct nitration of triphenylmethane with concentrated sulfuric and nitric acid mixture according to the method of Shoesmith, Sosson, Hetherington.^ After several recrystallizations from chloro-form-ether, the product was recrystallized from toluene and melted at 214-215°. Lit. m.p. 214.5°.65 4,4'-Dinitrodiphenylmethane (-bis-p-nitrophenyl-methane): was obtained from Eastman Chemicals and recrystallized from benzene. The compound melted at 183-183.5°. Lit. m.p. 183°. 6 6 3,4'-Dinitrodiphenylmethane: was prepared from p-nitrobenzyl 67 alcohol and nitrobenzene according to the method of Gattermann and Rudt The product was recrystallized from ethanol and melted at 104.5-105.5°. Lit. m.p. 103°. 6 7 9-Phenylfluorene: was prepared by dehydration, of triphenyl-68 carbinol in phosphoric acid according to the method of Kliegl. The product was recrystallized from ethanol and melted at 6 2 Lit. m.p. 145°. The tritiated compound was prepared by exchange 53 according to the method of J.R. Jones. 9-Phenylfluorene (1 gm) was dissolved in 5 ml of a (1:1) Dioxan-DMSO mixture. Tritiated water (0.1 ml of activity 10 mc/ml) was added and a few crystals of tetra-methylammonium hydroxide pentahydrate. After three days the solution was neutralized with acid, and the solvent evaporated. The solid was added to 100 mis of water, to dissolve the salt, and the solution filtered. This dissolution - filtration step was repeated thrice before recrystallization of the 9-phenylfluorene-t from ethanol. - 49 -2,3-Benzofluorene: was obtained from K § K Laboratories and was recrystallized from petroleum ether. M.p. 207-208°. Lit. m.p. 63 208°. The tritiated compound was prepared by Dr. J.R. Jones in this laboratory. Fluorene: was obtained from K § K Laboratories and recrystal-lized from benzene/petroleum ether (30-60°), then sublimed, then recrystallized. The product melted at 116.5-117.5°. Lit. m.p. 116-63 117°. The tritiated compound was prepared by direct exchange. 9-Ethylfluorene: was prepared by direct alkylation of fluorene in hexamethylphosphoramide in the presence of sodium hydride, according 70 to the method of T.Cuvigny and H. Normant. The product was vacuum 70 distilled twice. B.p.lf, 102-104°. Lit. b.p.1c 160° and then purified by gas chromatography on a 20% silicone SF96 column temp. 240°. The tritiated compound was prepared by direct exchange. 9-Phenylxanthene: 9-phenylxanthenol was prepared from phenyl-magnesium bromide and xanthenone in dry benzene. The xanthenol was then reduced by a formic acid/sodium formate mixture. The product was recrystallized from benzene-ethanol, then sublimed and recrystallized. 71 M.p. 144.5-145°. Lit. m.p. 145°. The tritiated compound was prepared by Dr. J.R. Jones by direct exchange. Triphenylmethane-t: triphenylchloromethane was prepared from triphenylcarbinol and acetyl chloride in benzene then treated with sodium amalgam in ether to give triphenylmethanesodium. The sodium salt was then neutralized with HTO to give the tritiated hydrocarbon. After filtering off the NaOH thus precipitated, the solvent was evaporated. The product was then redissolved in ether and refiltered to take off -. 50 -any remaining NaOH. This evaporation-filtration step was repeated thrice, before recrystallization of the triphenylmethane-t from ethanol. - 51 -RESULTS AND DISCUSSION PART I: Indicator Measurements N I. Diphenylamine indicators H From the derivation of the H function (pp. 4-6), i t i s evident that the anchoring of the scale i n the water region i s the most important step i n the establishment of th i s scale. 13 Stewart and O'Donnell chose 2,4 , 6 - t r i n i t r o a n i l i n e (pK = / a * N 12.20) as t h e i r anchor for the H scale i n DMSO-water-0.001 M TMAOH. 72 The pH of an 0.01 M aqueous sodium hydroxide solution i s 11.939. The pH of a 0.011 M aqueous TMAOH solution as measured on a Beckman pH meter model G (standardized at pH 12.45 and 10.00) i s 11.95. The N 13 above mentioned H function does not extrapolate to the expected value of 11.95 but to 11.65. I t i s now known that 2 , 4 , 6 - t r i n i t r o a n i l i n e forms a Meisenheimer 73 complex i n methanolic DMSO containing sodium methoxide. In thi s work no such complex was detected by n.m.r. i n aqueous DMSO, neverthe-less the u.v. spectrum of the ionized molecule shows some disturbing * A careful re-examination of the io n i z a t i o n of 2,4 , 6 - t r i n i t r o a n i l i n e i n aqueous buffers and i n sodium hydroxide solutions gave a pK value of 12.30. A subsequent determination of H values gave the following r e s u l t s . mole % D M S 0 o o.5 1. 5. 10. log I -.25 -.16 -.08 .30 .50 . \\ :/. H 12.05 12.14 12.22 12.60 12.80 - 52 -features. Besides the peak at 414 my- assigned to the anion a smaller 74 band appears at 470 my. Schaal also noticed a band at 470 my in concentrated ethylene diamine solutions, and treated i t as indicative of a second ionization for which he estimated a pK value of 17.55. r a This band (470 my) disappears after 12 hours in a 24 percent DMSO-water-0.3 M TMAOH solution, while the ion peak did not decrease in intensity and shifted to slightly higher wavelength 426 my. The 470 my band was not observed in sodium hydroxide solutions. These features led us to seek a more suitable anchor for the scale and our choice was 2,4,4'-trinitrodiphenylamine. Its pK was determined in aqueous glycine buffers and in sodium hydroxide solutions and was found to be 12.30 and 12.26 respectively. The value 12.30 was then used to determine H values for aqueous DMSO solutions up to 10 mole percent of DMSO. This same value was also used with the relative pK values of a series of substituted 2,4-dinitrodiphenyl-amines and 4-nitrodiphenylamines to calculate their thermodynamic pK& values. Table V contains a li s t of these indicators and their pK r a values determined in this work. These pK values were then used to a calculate the H values for DMSO-water solutions up to 50 mole percent DMSO (this was done by using equation 10). These H_ values are averaged and the results are summarized in Table VI. These H values are in 4 excellent agreement with the values of Dolman which are based on 2,4-dinitrodiphenylamine, pK 13.84. Figure 3 is a plot log I (log A /AH) versus H_ for the indicators in Table V and is a reproduction of a computer output for the least squares f i t for the lines. - 54 -TABLE V pK a values of various diphenylamines as determined i n the aqueous DMSO system. PKa reported 6 r N- T 2,4,4'-Trinitrodiphenylamine 12. 30 * 12.25 1, .002 .999 5. 396. ,6 2,4,3'-Trinitrodiphenylamine 12. 59 12.65* 1. .002 .999 4 341. ,3 2,4-Dinitro-4'-trifluoromethyl-diphenylamine 12. 87 0, .999 .999 3 205. ,8 2,4-Dinitro-3'-trifluoromethyl-diphenylamine 13. .06 1, .004 .999 3 409. .6 2,4-Dinitro-3'-chlorodiphenyl-amine 13. ,17 1, .004 .999 3 463. .8 2,4-Dinitrodiphenylamine 13. ,85 13.83* 1. .001 .999 4 394. .1 2,4-Dinitro-3'-methyldiphenyl-amine 13. ,90 0. .999 .999 3 168. ,9 2,4-Dinitro-4'-aminodiphenyl-amine 14, ,48 14.64* 1 .005 .999 3 452, .3 4,4'-Dinitrodiphenylamine 14.08* — - — 3,4'-Dinitrodiphenylamine 14, ,62 14.66+ 1 .005 .999 3 274, .9 4-Nitro-3'-trifluoromethyldi-phenylamine 14, ,90 14.96+ 1 .001 .999 4 383, .6 4-Nitro-3'-chlorodiphenylamine 15, .00 1 .004 .999 3 504, .2 4-Nitrodiphenylamine 15, .67 t 15.67 0 .997 .999 4 413 .8 4-Nitro-3'-methyldiphenylamine 15 .60 1 .005 .999 3 422 .5 4-Nitro-4'-aminodiphenylamine 16 .40 0 .926 .997 3 138, .1 0 slope of plot of log I vs. H r c o e f f i c i e n t of correlation N number of points T student's test * Stewart and J.P. O'Donnell, ref. 13. Dolman and Stewart, r e f . 4. 55 TABLE VI HN values for the system DMSO-water-0.011 M tetramethyl ammonium hydroxide mole % „N UN . * DMSO - - \"ported 0.4 11.95 1.0 12.08 2.0 12.20 5.0 12.54 10.0 13.04 11.3 13.19. 13.34 16.0 13.73 13.96 20.3 14.29 14.49 24.1 14.72 14.88 29.7 15.34 15.42 33.6 15.75 15.88 36.0 16.02 16.12 43.0 16.74 16.77 49.1 17.35 17.38-* Dolman and Stewart, ref. 4. - 56 -The same set of indicators were also used to determine H functions for the systems DMSO-MeOH-0.01 M sodium methoxide and DMSO-EtOH-0.01 M NaOEt. In alcoholic DMSO convergence of the scales at H = pH = 11.98 need not occur since all the pK& values refer to water as the standard state. The anchor for the methanolic DMSO scale : is the same as that for aqueous DMSO, i.e. 2,4,4'-trinitrodiphenylamine, and the anchor for ethanolic DMSO is 2,4-dinitrodiphenylamine. The pK values for the individual diphenylamines and the cl H values for different solutions were arrived at in exactly the same way as for the aqueous DMSO system. Tables VII and IX lis t the different pK values for these a indicators as arrived at in the two systems, methanolic and ethanolic DMSO. The acidity functions for these two systems are given in Table VIII and X. N It is to be noted that an H function for methanolic DMSO containing 0.025 molar sodium methoxide has already been determined by 33 Stewart, O'Donnell, Cram, and Rickborn. If we assume that the difference in H values of DMSO-MeOH solutions containing 0.01 M or 0.025 M sodium methoxide is constant with changing composition, N N then we can reduce the H_ (0.025) to an H (0..01) for comparison by using equation (32). H_ = H_ (0.01) .+ log (32) Figure 4 is a reproduction of the least square f i t for plots of log I against H_ for the indicators in Table VII. Figure 6 is the analogous plot for the indicators in Table IX. Figures 5 and 7 are plots of H- versus solvent composition for the methanolic and ethanolic DMSO systems. - 58 -TABLE VII pK values of various diphenylamines as determined in the methanolic DMSO system. P Ka reported e r N T 2,4,4'-Trinitrodiphenylamine 12. 25 12.35* 1, .001 .999 5 261 .9 2,4,3'-Trinitrodiphenylamine 12. 65 12.65* 0, .995 .999 4 372 .4 2,4-Dinitro-4'-trifluoromethyl-diphenylamine 12. 85 1, .000 .999 5 289 .6 2,4-Dinitro-3'-trifluoromethyl-diphenylamine 13. 03 1, .000 .999 5 289 .6 2,4-Dinitro-3'-chlorodiphenyl-amine 13. 20 1, .000 .999 4 310 .5 2,4-Dinitrodiphenylamine 14. 05 13.84* 1 .000 .999 4 333 .3 2,4-Dinitro-3'-methyldiphenyl-amine 14. ,10 0 .998 .999 5 557 .4 2,4-Dinitro-4'-aminophenylamine 14. ,95 14.64* 1 .001 .999 4 713 .0 4,4'-Dinitrodiphenylamine 14. ,42 14.08* 0 .994 .999 5 125 .4 3,4'-Dinitrodiphenylamine 15. .26 1 .001 .999 5 517 .4 4-Nitro-3'-trifluoromethyl-diphenylamine 15. ,69 1 .001 .999 5 544 .1 3-Nitro-3'-chlorodiphenylamine 15. .85 1 .022 .997 4 171 .6 4-Nitrodiphenylamine 16, .59 15.90* 1 .004 .999 4 493 .1 4-Nitro-3'-methyldiphenylamine 16, .50 1 .004 .999 4 557 .4 4-Nitro-4'-aminodiphenylamine 17 .35 1 .001 .999 4 299 .5 0 slope of plot of log I vs. H_ r coefficient of correlation N number of points T student's test * Stewart, O'Donnell, Cram and Rickborn, ref. 33. 59 TABLE VIII N H values for the system DMSO-methanol-0.01 M sodium methoxide Mole % „N UN ' * DMSO - H_ reported 0.3 11.95 11.83 1.0 12.0.2 11.92 2.0 12.10 12.01 5.0 12.28 12.23 10.6 12.68 12.57 15.0 13.08 19.2 13.38 13.27 25.0 13.80 29.7 14.17 13.90 ' 35.1 14.58 40.0 14.98 14.44 44.9 15.34 50.0 15.72 15.03 54.9 16.15 60.2 16.57 15.76 65.5 16.98 70.0 17.35 16.30 74.0 17.70 77.0 17.94 Stewart and O'Donnell, Cram and Rickborn, r e f . 33. 0.04 was substrated from each H_ value to account for difference i n base concentration - 62 -TABLE IX values of various amines as determined i n the ethanolic a DMSO system. Amine P Ka 6 r N T 2,4-Dinitrodiphenylamine 14. 05 1. .002 .999 6 508. .0 2,4-Dinitro-3 1-methyldiphenyl-amine 14. 12 1 .003 • .999 5 450. ,5 2,4-Dinitro-4 1-aminodiphenyl-amine 15. 05 0.992 .999 3 186, ,2 4,4'-Dinitrodiphenylamine • -3,4'-Dinitrodiphenylamine 15. .15 1 .002 .999 6 836, .1 4-Nitro-3'-trifluoromethyldi-phenylamine 15. 63 1 .003 .999 5 622, .2 4-Nitro-3 1-chlorodiphenylamine 15. ,73 1 .005 .999 3 222, .8 4-Nitrodiphenylamine 16. .63 1 .003 .999 5 599. .6 4-Nitro-3'-methyldiphenylamine 16. ,55 1 .009 .999 3 330, .1 4-Nitro-4 1-aminodiphenylamine 17. ,45 1 .003 .999 5 617, .5 2,4-Dinitroaniline 15. ,27 .960 .999 5 52, .7 4-Chloro-2-nitroaniline 18. ,05 1 .054 .999 5 39 .0 8 slope of plot of log I vs. H r. c o e f f i c i e n t of correla t i o n N number of points T student's test - 63 -TABLE. X values for the system DMSO-ethanol-0.01 M sodium ethoxide Mole % DMSO H_ 0.3 13.46 1.0 13.57 2.5 13.71 5.0 13.94 10.0 14.29 14.6 14.64 20.4 15.01 24.7 15.31 30.0 15.60 35.4 15.96 39.7 16.28 45.0 16.63 50.6 16.96 55.2 17.29 60.0 17.64 65.0 18.00 70.0 18.40 75.9 18.81 - 65 -N A. V a l i d i t y of the H scale i n alcoholic DMSO The v a l i d i t y of the stepwise procedure i s based e n t i r e l y on the r e l i a b i l i t y of Hammett's postulate. The postulate states that the a c t i v i t y c o e f f i c i e n t term i n equation 9 can be equated to zero. v •\" „ i [HA] i t H B ] , fHA fB\" : . P HA - P HB = l Q g [A=t \" l G g [B=] + l 0 g C 9 ) L J L J HB A The a c t i v i t y c o e f f i c i e n t r a t i o of the indicator and indicator anion for two overlapping acid indicators being equal to unity i n any given solution. • •'\" f f -HA B _ f - ' f A HB Applying this condition equation (9) reduces to (10) PKHA - PKHB \" l 0 g # r \" l 0 g T n • ( 1 0 ) One obvious experimental test of the v a l i d i t y of equation (10) i s that the term log -[^ry - log remains constant as the medium i s changed. This constancy can be best examined by the pa r a l l e l i s m i n the. plots of log j i ^ j (log I) versus solvent composition for the diffe r e n t indicators studied. . . In th i s work, figures 8, 9, and 10 are i l l u s t r a t i o n s of th i s t e s t , and i t i s evident that the plots of log I vs mole percent DMSO cv Figure 9. Plots of log I (A/AH) versus mole percent DMSO for the system: DMSO-methanol• 0.01 M sodium methoxide. CD o t 0.0 10.0 _ j I I I 20.0 30.0 40.0 50.0 MOLE PERCENT DMSO 60.0 70.0 80.0 CM , • , : , r 1 1 1 n [— 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 MOLE PERCENT DMSO - 69 -remain parallel for a l l the indicators. The reproducibility of any one pK determination using different solutions prepared over a period cL of several months was better than 0.05 pK unit, which exceeds the a accuracy of the H_ values (cf. experimental). There is a good agreement in the H values calculated using different indicators. The slopes of the lines of log I versus H given in the pK tables are calculated by a least square method and are exceedingly close to unity. Correlation coefficients are also given and are close to unity. Student's test T evaluates the significance of the correlation, as a high correlation coefficient may easily arise by chance i f the number of points plotted is small. These results would thus indicate that an H function can be defined in these systems. It has been a prevalent opinion that the apparent relative strength (log 1^ - log I B = pK^ - pK^g) of a series of Hammett 46 indicators must remain uniform through drastic changes in solvent. This requirement for the validity of an H function will now be examined and its applicability considered in strongly acidic media, since there is a larger amount of data published on pKg^+ than on pK&. The basis of Hammett's postulate is that i f the indicators f f A BH+ A and B are of similar structure one can expect -= . -~ = 1. AH+ B Thirty years of work on acidity functions resulted in more stringent limitations being placed on structural variations. First considered was the molecular size, then the aromatic or aliphatic character then also the functional group - amino, amide, etc. Amines themselves have been differentiated into primary, secondary and tertiary. The 75 existence of an H function in sulfuric acid systems has been asserted, - 70 -78 and primary an i l i n e s were found to be r e l i a b l e Hammett bases. Even i n such an \" i d e a l \" system, with r e l i a b l e indicators many discrepancies ar i s e . The e a r l i e r colorimetric work i s considerably less r e l i a b l e than more recent work i n which spectrophotometry has been used. 76 Bascombe and B e l l redetermined pKn,. values for a number of Hammett y BH+ indicators and found for example that, 4-methyl-2,6-dinitroaniline i s 75 0.48 pK unit away from the \"best value\" given by Paul and Long i n t h e i r review a r t i c l e . Table XI l i s t s pKg^ + values for primary a n i l i n e s , published between 1957 and 1964, a l l determined spectrophoto-* m e t r i c a l l y , and a l l i n s u l f u r i c acid media. Although a l l the authors start with the same value for t h e i r anchor, 4-chloro-2-nitroaniline, or 2-chloro-6-nitroaniline, pKDL, values bn+ 77 for the other indicators vary widely. Hogfeldt and Bigeleisen measured t h e i r plC,, values i n H»0-H„S0. and i n D»0-D„S0.. The r BH+ 2 2 4 2 2 4 average P^g^+ difference (ApK) between the deuterio acid i n heavy water solution, and the protio acid i n water solution i s 0.56 pK unit but some indicators show a much smaller ApK value e.g. 2,6-dinitro-4-methylaniline ApK =0.40. From table XI one can see that there i s a small difference between pKg^ + values i n ^ SO^-^O and i n D20-D2SO^, while there i s a marked difference i n pKg^ + between aqueous s u l f u r i c and ^SO^-sulfolane. This i s probably due to the fact that there i s * Paul and Long best values are only given as a reference, since 78 Jorgensen and Harter re-evaluated the pKg^ + values of 2-bromo-4,6-d i n i t r o a n i l i n e and 2 , 4 , 6 - t r i n i t r o a n i l i n e and found them dif f e r e n t from the best value by 0.23 and 0.38 pK unit respectively. - 71 -TABLE XI pK D U+ of weak bases, i n s u l f u r i c acid media, from the l i t e r a t u r e BH Anilines pKDIJ+ water D20 Sulfolane a b c d e f 2-nitro- -0.29 -0.29 +0.30 2-chloro-6-nitro- -2 .43 -2.42 4-chloro-2-nitro- -1.03 -1.02 -0.46 2,4-dichloro-6-nitro- -3.32 -2.73 -3.37 -3.14 2,6-dinitro-4-methyl- -4.44 -3.96 -4.04 2,4-dinitro- -4.53 -4.03 -4.36 -3.74 2,6-dinitro- -5.41 g -5 .54 -4.50 2-bromo-4,6-dinitro- -6.71 -6.64 -6 .68 -6.55 -6.06 2,4,6-trinitro- -9.41 -9.03 g -10 .10 -8.82 -8.21 a b c d e f g \"best value\" from Paul and Long, r e f . 75. Bascombe and B e l l , r ef. 76. Jorgensen and Harter, r e f . 78. Hogfeldt and Bigeleisen, ref. 77 these values are pK values of the deuterio acid i n heavy water. The average difference ApK between P KH 2S0 4 a n d P KD 2S0 4 1 5 °- 5 6* Arnett and Douty, r e f . 79a. Alder, Chalkley and Whiting, ref. 80. Value determined by Jorgensen and Harter, ref. 78 based on the H Q values of Paul and Long. - 72 -only a s l i g h t difference between water and heavy water while there i s an enormous difference between water and sulfolane as solvents. The vari a t i o n i n pKg^ + value does not go on increasing with decreasing b a s i c i t y of the indicators, and therefore i s a true difference and not due to a cumulative error i n p K R H + determination. For example: Indicator 2,6-dinitroaniline 2-bromo-4,6-dinitroaniline 2 , 4 , 6 - t r i n i t r o a n i l i n e Although most authors adopt a complaisant attitude towards the p r i n c i p l e of pK values of Hammett bases, the data i n Table XI do not warrant such a view. A not p a r t i c u l a r l y drastic change i n solvent system from aqueous to su l f o l a n i c can give r i s e to widely different pKg^ + values. Unfortunately few authors have published along with t h e i r results spectroscopic data of t h e i r indicators. 19,69 Only two groups of workers \" have measured a c i d i t i e s 19 of a n i l i n e s by a non-photometric method. Ritchie and Uschold measured potentiometrically the pK of four a n i l i n e type indicators i n DMSO and found that the r e l a t i v e a c i d i t i e s d i f f e r only s l i g h t l y from those measured spectrophotometrically i n aqueous DMSO by Dolman 4 69 and Stewart . B i r c h a l l and J o l l y measured the a c i d i t i e s of a series of substituted anilines r e l a t i v e to 2,5-dichloroaniline, by o n.m.r. i n l i q u i d ammonia at 31. For the f i v e anilines which have also 4 been determined by Dolman and Stewart none shows similar r e l a t i v e „ difference i n pK3^ + values \" BH+ between authors ^5 1.04 ^6 0.62 ^8-10 1.90 - 73 -acidities, in contrast with the above mentioned observation by Ritchie and Uschold. Therefore the relative acidities of anilines can vary 4,13 with change in solvent, contrary to accepted premises. . ' One must therefore study systematically Haramet type indicators to find a rationale for the fact that sometimes the acidity or the basicity of these indicators varies with solvent change and then sometimes does not. The pK of an acid whose ionization occurs in the non-a aqueous region is related to the standard state in water through a series of indicators with which i t overlaps and with which i t bears structural similarity. Ideally one would want to see a study of the relative strength of acids (HX, HX^ ...) with respect to a 81 fixed acid (HY) in different solvent systems. Grunwald and Price have noted that the relative strength of picric acid relative to acetic acid increases by almost two orders of magnitude in the solvent series HOH, CH^ OH, 02^011, while that of trichloroacetic acid remains nearly constant. Furthermore, the relative strength of picric acid with respect to phenol increases by the same amount while picric acid strength with respect to 2,4-dinitrophenol is constant. Therefore the increase in relative strength is not due to the chemical nature of the reference acid or phenol since, schematically we have: change in relative . . 5 „ increase constancy pK values 1 a __ compound picric acid picric acid trichloroacetic picric acid reference acetic acid phenol acetic acid 2,4-dinitro-phenol - 74 -81 Grunwald and Price postulated that this is due to disper-sion effects which occur between the delocalized virtual electronic oscillators of the solute and the surrounding medium. Their argument is that i f we compare a coloured indicator with a colourless reference compound (picric acid/acetic acid and picric acid/phenol) we can expect an increase in relative strength in the series HOH, CH^ OH, C^U^OH; while i f we compare a colourless acid with a colourless reference (trichloroacetic acid/acetic acid) or a coloured acid with a coloured reference (picric acid/2,4-dinitrophenol) we can expect constancy in relative strength. 81 These authors, by studying the system HOPic + BH+OAc\" ^ K 0 H > BH+OPic\" + HOAc (33) estimated the magnitude of these dispersion forces. The equilibrium constant K.,,. = f ^ ? ^ l>m.+^i C-i increases by almost an order of BH+ [HOPic][BH+OAc ] ' + + + magnitude in glacial acetic acid by varying BH from NH^ to (CH )gNH , while the corresponding equilibrium constant for trichloroacetic acid is insensitive to the change in BH+. Calculation using a model in which BH+ is hydrogen bonded to the anionic oxygen atom in the ion pair gives a dispersion energy of -1.6 kcal/mole. They also suggested that the difference in relative base strength of aliphatic amines obtained by measurements in water or with colour indicators in solvents of low dielectric constant could be accounted for by the existence of strong dispersion effects. - 75 -The importance of dispersion effects was also noted by 19 other workers. Ritchie and Uschold i n pointing out the f a i l u r e of C the H scale suggested that the ionization behaviour of hydrocarbon acids which have highly coloured conjugate bases i s not the same as 82 that of acids whose conjugate bases are not delocalized. J . J u i l l a r d studied the d i s s o c i a t i o n of a few carboxylic acids i n methanol and water and found a d e f i n i t e , though imprecise, correlation between the molecular p o l a r i z a b i l i t y of the acids and change i n pK with d i f f e r e n t solvent. A.J. Parker et. a l . ^ ' 8 3 i n t h e i r extensive studies on solvation of ions i n p r o t i c and aprotic solvents have also pointed out the necessity of taking into account solvent structure and mutual p o l a r i z a b i l i t y interactions (dispersion forces) i f we want to explain the s i g n i f i c a n t changes i n the chemistry of anions on transfer from water to methanol and dipolar aprotic solvents. I f we rearrange the table of pK of acids from reference 83 to take into account 1 a colour, we f i n d that although each acid has a different pK i n every a solvent, the r e l a t i v e strength of two acids ApK of s i m i l a r colour remains constant through change i n solvent. This i s not true,however, for any two acids i n two different colour d i v i s i o n s . Acid pK(H20) pK(MeOH) pK(DMF) pKDMSO ApK ApK ApK ApK dichloroacetic 1.29 1.57 6.4 1.30 7.2 1.8 chloroacetic 2.86 7.7 9.0 benzoic 4.20 1.34 9.1 1.4 10.2 1.2 11.0 acetic 4.76 0.56 9.6 0.5 11.1 0.9 11.4 0.4 2,4-dinitrophenol 4.10 7.9 6.0 5.2 4-nitrophenol 7.15 3.05 11.2 3.3 10.9 4.9 9.9 4.7 - 76 -They also noted that for all the acids listed (17 of them) acidities in water and methanol show some correlation, as well as in DMF and DMSO, but there is no correlation between the acidities in a protic and an aprotic solvent. And so there is enough evidence of the nonuniformity of the relative strength for a series of related acids or bases, through changes in solvent, except i f their similarity extends to their spectral characteristics. By extending Grunwald and Price's argument to its logical conclusion, we can put forward the postulate that the relative strengths of indicators will remain unaltered through changes in solvent environment only i f their spectral characteristics in both ionized and unionized forms are similar. Indicators which will comply with this limitation will be referred to as \"similar colour indicators.\" Table XII lists the pK measured in this work for the r a substituted 2,4-dinitro, and 4-dinitrodiphenylamines referred to here-after as 2,4-dinitro series and 4-nitro series. Their spectral characteristics are listed in Tables XIII and XIV. One can see that the visible spectra of the 2,4-dinitro compounds are a l l quite similar with the exception of their 4'-nitro- and 4'-amino-derivatives. The substituted 4-nitro compounds also show a similarity in their absorbances, With again the exception of their 4'-nitro- and 4'-amino-derivatives . The main difference is between the ionized form of these two classes of indicators, the 4-nitro compounds absorb at a longer wavelength and their molar absorptivities a r e twice as much as those of their 2,4-dinitro analogs. From Table XI one can also see that although - 77 -TABLE XII Summary of the pK values of diphenylamines determined in this work a H„0 MeOH EtOH 2,4,4'-Trinitrodiphenylamine 12. ,30 12 .25 2,4,3'-Trinitrodiphenylamine 12, ,59 12 .65 2,4-Dinitro-4'-trifluoromethyldiphenylamine 12, ,87 12 .85 2,4-Dinitro-3'-trifluoromethyldiphenylamine 13. ,06 13 .03 2,4-Dinitro-3'-chlorodiphenylamine 13. .17 13 .20 2,4-Dinitrodiphenylamine 13. .85 14 .05 14 .05 2,4-Dinitro-3'-methyldiphenylamine 13. .90 14 .10 14 .12 2,4-Dinitro-4'-aminodiphenylamine 14, .48 14 .95 15 .05 4,4'-Dinitrodiphenylamine 14, ,08(?) 14 .42 . -• 3,4'-Dinitrodiphenylamine 14, .62 15 .26 15 .15 4-Nitro-3'-trifluoromethyldiphenylamine 14, .90 15 .69 15 .63 4-Nitro-3'-chlorodiphenylamine 15, .00 15 .85 15 .73 4-Nitrodiphenylamine 15, .67 16 .59 16 .63 4-Nitro-3'-methyldiphenylamine 15, .60 \" 16 .50 16 .55 4-Nitro-4'-aminodiphenylamine 16, .40 17 .35 17 .45 - 78 -TABLE XIII Absorption maxima and molar ab s o r p t i v i t i e s of the indicators i n absolute ethanol Observed X (mp) max J Values e Reported Values X (my) e max^ V J Ref. 2,4,4'-Trinitrodiphenylamine 390 18300 385 21900b 13 2,4,3'-Trinitrodiphenylamine 354 20000 360 17850° 13 2,4-Dinitro-4'-trifluoromethyl . diphenylamine 354 17900 2,4-Dinitro-3'-trifluoromethyl diphenylamine 354 17500 2,4-Dinitro-3'-chlorodiphenyl-amine 356 17900 2,4-Dinitrodiphenylamine 361 16900 352 17700a 4b 2,4-Dinitro-3'-methyldiphenyl-amine 365 19200 2,4-Dinitro-4 *-aminodiphenyl-amine 378 14400 375 18300° 13 4,4'-Dinitrodiphenylamine 418 33000 403 34100 a 4b 3,4'-Dinitrodiphenylamine 390 22200 375 23200a 4b 4-Nitro-3'-trifluoromethyl-diphenylamine 394 21800 381 19200a 4b 4-Nitro-3'-chlorodiphenyl-amine 396 25500 385 22700a 4b 4-Nitrodiphenylamine 402 22000 400 19550 \"4b 4-Nitro-3'-methyldiphenyl-amine 410 24000 394 22300a 4b 4-Nitro-4'-aminodiphenyl-amine 420 19100 4-Chloro-2-nitroaniline 424 5300 417 3450® 13 2,4-Dinitroaniline 346 384 14400 6600 336 14900e 13 a i n 95.6% ethanol b i n pyridine or aqueous pyridine ° i n sulfolane or aqueous sulfolane e i n methanol - 79 -TABLE XIV Absorption maxima and molar abso r p t i v i t i e s of indicator anions i n DMSO-ethanol Observed X (my) maxv ' Values e Reported Values X (my) e max • Ref. 2,4,4'-Trinitrodiphenylamine 530 29900 520 27400b 13 2,4,3'-Trinitrodiphenylamine 432 23600 450 18700c 13 2,4-Dinitro-4'-trifluoromethyl-diphenylamine 435 474 20400 19500 2,4-Dinitro-3'-trifluoromethyl-diphenylamine 435 474 20000 18900 2,4-Dinitro-3'-chlorodiphenyl-amine 432 475 20400 18100 2,4-Dinitrodiphenylamine 432 485 19700 16500 434 19000a 4 2,4-Dinitro-3'-methyldiphenyl-amine 434 480 23000 19700 2,4-Dinitro-4'-aminodiphenylamine 496 17400 495 18300° 13 4,4'-Dinitrodiphenylamine 590 51000 580 37000b 13 3,4'-Dinitrodiphenylamine 324 504 15000 36100 4-Nitro-3'-trifluoromethyl-diphenylamine 506 36500 4-Nitro-3 1-chlorodiphenylamine 504 41800 4-Nitrodiphenylamine 505 34800 508 34600 13 4-Nitro-3 *-methyldiphenylamine 510 36200 4-Nitro-4'-aminodiphenylamine 518 29800 4-Chloro-2-nitroaniline 496 524 8300 8200 495 520 8440 a 8440 4 2,4-Dinitroaniline 388 392 502 17300 15400 12200 388 535 21100b 15300 13 a b e i n DMSO-water i n pyridine i n sulfolane - 80 -the pK of the 2,4-dinitro compounds remain the same i n the three a * systems, the 4-nitro compounds are some 0.8 units higher i n DMSO-MeOH and DMSO-EtOH than i n DMSO-H20. The importance of the anchor i n the establishment of an H function has been previously emphasized (pp.51-£2). 2,4,4'-trinitro-diphenylamine has been used as an anchor for the DMSO-water and DMSO-methanol systems, while 2,4-dinitrodiphenylamine was used for the DMSO-ethanol system. Therefore i f our assumption i s v a l i d we should f i n d the r e l a t i v e strength of a l l the substituted 2,4-dinitrodiphenyl-amines constant. Also we should f i n d that the r e l a t i v e strength of the substituted 4-nitrodiphenylamines constant. Only the r e l a t i v e strength of the 4-nitro series with respect to the 2,4-dinitro series should vary. Also i f a linea r free energy relationship i s constructed between the pK values versus Hammett substituent constants the p value for the 4-nitro series should remain of the same order of magnitude through change i n solvent system. Indeed t h i s i s the case as can be seen from Figure 12 and Table XV. As circumstancial evidence for the v a l i d i t y of the H scales defined, and the \" s i m i l a r colour indicator\" assumption, the i o n i z a t i o n of 2,4-dinitroaniline and 4-chloro-2-nitro-* N An H function i n DMSO-MeOH-0.025 M NaOMe has been determined by - 33 Stewart, O'Donnell, Cram and Rickborn, and these authors found that the pK & Values of a l l their indicators agreed with the determined values i n aqueous DMSO. But aside from the fact that t h e i r anchor 2 , 4 , 6 - t r i n i t r o a n i l i n e , has been shown subsequently to form Meisenheimer complex i n DMSO-MeOH-NaOMe, a plot of log I versus mole percent DMSO shows several cases of nonparallelism between the indicators, Figure N 11, These facts cast some doubt as to the v a l i d i t y of the H function 33 defined by Stewart et a l . , and t h e i r l i s t e d pK values. I I : ! I H I ! I : I 1 0.0 12.5 25.0 37.5 50.0 62.5 75.0 87.5 100.0 MOLE PERCENT DMSO - 82 -aniline was studied in the system DMSO-EtOH-0.01 M NaOEt. The pK& assigned are: pK pK reported in 3. 3. DMS0-H20, ref. 4. 3,41-dinitrodiphenylamine 15.10 14.66 2,4-dinitroaniline 15.27 15.00 4-chloro-2-nitroaniline 18.05 17.08 The pK of 3,4'-dinitrodiphenylamine is also included as i t has a pK 3. 3. similar to the one of 2,4-dinitroaniline in DMSO-EtOH, but its pK 3. in DMSO-water is 0.5 unit lower than in ethanolic DMSO. One can see that the spectral characteristics of 2,4-dinitro-aniline are somewhat similar to those of the 2,4-dinitrodiphenylamine series, while spectral characteristics of 4-chloro-2-nitroaniline are quite different. According to the similar colour assumption the pK 3. of the former should be comparable in both media (0.27 pK units apart is within the experimental accuracy) while the latter should be quite different (1.00 pK units away). a. B. Correlation of Structure with Acidity 13 Preliminary results on the acidity of substituted diphenyl-amines indicated the dominance of the nitro group, whose effect overwhelmed that of a l l other substituenfs. Enough data have now been gathered to make i t possible for us to compare the linear free 4 energy relationships of the A) substituted diphenylamines (D. Dolman results), B) 4-nitro substituted DPA, C) 2,4-dinitro substituted DPA, - 83 -D) 2,4,6-trinitro substituted DPA (A. Buckley, unpublished r e s u l t s ) . Figure 12 shows a plot of pK versus o°and i t i s obvious that there i s a a good correlation. Table XV l i s t s the p values,correlation c o e f f i c i e n t s and * Students test's T computed for the plot i n Figure 12. The substituent constant a°usually correlates r e a c t i v i t i e s for reaction i n which there i s no conjugation between the substituents and the reaction center. They were derived by interposing a methylene group between the benzene ring and the reaction center, so that the resonance effect would be ess e n t i a l l y constant for such a reaction series, o i s used for reaction i n which there i s direct conjugation, and a for systems that possess the limited amount of conjugation present i n benzoic acids. For the \"normal\" substituents m-methyl, m-chloro, m-trifluoro, p - t r i f l u o r o and m-nitro the a°values are very s i m i l a r to the a value so one cannot r e a l l y draw any conclusions from the correlation with o° except that there i s no direct resonance interaction between these groups and the amide ion. * In l i n e \"A\" the 4'-nitro substituent has been omitted from the calculation since i t requires an exalted a value. - 85 -TABLE XV Linear free energy correlation of diphenylamines a c i d i t i e s computed by a least squares method Z Ar System p+ Ap N r T 2,4,6-trinitropheny1 aqueous buffers -1 .88 0. .06 8 .997 32 .1 2,4-dinitrophenyl DMSO-.H20 -1 .79 . 0. .03 8 .999 60 .0 DMSO-MeOH . -2 .17 0. .08 8 .996 27 .4 DMSO-H20 -1 .75 0. .12 7 .987 13 .6 4-nitrophenyl DMSO-MeOH -2 • 17 0, .20 7 .979 10 .8 DMSO-EtOH -2 .12 0' .12 7 .993 17 .3 phenyl DMSO-H20 -4 .34 0. .14 7 .997 31 .0 N number of indicators used. r c o e f f i c i e n t of correlation. f T Student's test. - 86 -However the pK& values for the p-amino substituent do not correlate with a but with o°. This is because there is very l i t t l e resonance between the amino substituent and the amide ion, and correlation with a° (no resonance) is expected. For the p-nitro resonance interaction of the following type is of prime importance. The substituent constant value for the 4-nitro group is 0.78 in the ionization of benzoic acids (a); 0.82 in ionization of phenylacetic acids (o°); 1.27 in the ionization of phenols and anilinium ions (a ). 84 Recently a value of 1.70 was ascribed for a-naphthol dissociation and 4 1.81 for a-naphthoxide methylation. D.Dolman assigned a a value of 1.65 for the 4-nitro substituent in the ionization of diphenylamine, 1 8 -and D. Kroeger assigned a a of 1.73 to account for the effect of the p-nitro group in the a-ring on the Lewis acidity of a-cyanostilbenes. It is surprising that this same compound correlates well with the a° value in line b when i t is looked upon as unsubstituted 4-nitrodi-phenylamine instead of a diphenylamine substituted with a p-nitro group. Since the acidities of diphenylamines correlates with a°, we could conclude that there is very l i t t l e conjugation between the substituent and the central atom. This brings us to the dichotomy in the magnitude of P. between diphenylamines and nitro-diphenylamines. The magnitude of the p-value is a measure of the susceptibility of diphenylamine ionization on the polar effect of the substituent Z. - 87 -Nitro-substituted diphenylamines have a p-value around 2, with a minimal increase from the 2,4,6-trinitro series to the 2,4-dinitro series and the 4-nitro series; while diphenylamine itself shows a higher response to substituent changes and has a p-value of 4.34. These p-values should be compared with p's of 1.00 for benzoic acids, 2.23 for phenols and 2.89 for anilinium ions in aqueous systems at 25°, and 5.3 for anilines 69 in liquid ammonia. Thus the p-value for the ionization of diphenyl-amine is similar to the one for anilines and is twice as large as the P-value for the ionization of phenols; the latter being similar to the P for the ionization of nitrodiphenylamines. This dichotomy probably results from the greater enjoyment of charge derealization by the diphenylamines and aniline anions than by the anions of nitro-substituted diphenylamines and phenols. Thus the dominating feature of the p-nitro substituent can be put in its proper perspective. For nitro-diphenylamines a low degree of conjugation between the amino centre and the substituent is noted, since the negative charge on the amide ion is already delocalized into the nitrated ring, which is obvious from the need for an exalted a\"value for the p-nitro substituent in the diphenyl-amine correlation. For B, C, D, the orientation of the ring with the Z-substituent is such that much of its contribution is sacrificed so - 88 -that the p-nitro group can heighten i t s conjugation. While for A maximum overlap occurs between the nitrogen o r b i t a l s and the ring with the substituent Z, thus the greater p value. Remembering, that a c i d i t i e s of substituted diphenylamine correlate with cr°, there should be only a minimal and constant amount of resonance energy available for the ring and i t s substituents, 50 that i n the case of the n i t r o diphenylamines the \"pie\" has to be divided between both rings while for the un-nitrated diphenylamines a l l of i t goes to the ring with the substituent Z. 4 ' Dolman and Stewart have explained deviations from a d d i t i v i t y of the effect of n i t r o groups i n compounds with n i t r o groups i n both rings to the i n a b i l i t y of both rings to a t t a i n simultaneously maximum overlap with the nitrogen atom, which i s a s i m i l a r type of argument. - 89 -I I . Phenol indicators Many of the phenol indicators used i n establishing the scales i n DMSO-methanol-0.01 M sodium methoxide and DMSO-ethanol-0.01 M sodium 31 ethoxide are the same as those used by Rochester to construct his scale i n methanolic sodium methoxide. The methanolic dimethylsulfoxide scale was anchored with _2-isopropyl phenol (pK =10.5) and then checked with phenol i t s e l f (pK =9.98) a a which i s half ionized i n a 0.01 M sodium methoxide solution. The ethanolic dimethylsulfoxide scale was anchored with 2,6-diisopropyl phenol (pK =11.0). Figures 13 and 14 show computer output plots of the measured values of log I versus H for the phenol indicators i n methanolic and ethanolic DMSO solutions. The plots are p a r a l l e l a n d the slopes are very close to unity, as can be seen i n Table XVII. The pK values assigned to these phenols are also l i s t e d i n Table XVII. The thermodynamic a c i d i t i e s of the anchors were determined i n aqueous buffers and therefore the pK & values obtained are r e l a t i v e 31 to water as the standard state. The pK & values determined by Rochester 32 were deduced by the method of Ridd and O'Ferrall and are r e l a t i v e 29 to methanol as the standard state. Cohen and Jones values were determined spectrophotometrically i n aqueous solutions and agree with the values determined i n th i s work except for 2,4-di-t-butyl- and 2-t-butyl-phenol. The H° a c i d i t y functions for methanolic and ethanolic DMSO are given i n Table XVI,and are plotted i n Figure 15. A summary of Figure 14. Plots of log I (A/AH) versus H for the phenols used to establish the if* scale in DMSO-ethanol-0.01 M sodium ethoxide. 1 0 . 0 10.5 11.0 11.5 12.0 H MINUS 12.5 13.0 13.5 14.0 - 92 -TABLE XVI values for the systems DMSO-methanol 0.01 M sodium methoxide and DMSO-ethanol 0.01 sodium ethoxide DMSO H_(DMSO-MeOH) H_(DMSO-EtOH) 0.0 9.92 10.62 1.0 10.77 5.0 10.17 11.02 10.0 10.35 11.21 15.0 10.53 11.41 20.0 10.73 11.56 25.0 10.94 11.74 30.0 11.17 11.91 35.0 11.36 12.10 40.0 11.55 12.26 45.0 11.75 12.46 50.0 11.98 12.62 55.0 12.20 12.75 60.0 12.42 12.98 65.0 12.63 13.11 70.0 12.84 13.31 .94 TABLE XVII pK values of phenols determined i n aqueous solutions and i n methanolic and ethanolic DMSO Phenols unsubstituted 2-isopropyl 2,6-di-isopropyl 2-jt-butyl 2,4-di-t-butyl 2,6-di-t-butyl 2,6-di-t_-butyl-4-methyl 2 , 4 , 6 - t r i - t - b u t y l pK m water 9.98 10.51 11.00 11.70 pK„ i n (9) pK„ i n (9) DMSO-MeOH DMSO-EtOH PKa reported 9 . 9 7 t l 4 . 1 0 t 10.49(1.016) 10.98(1.000) 11.00( .999) 11.67(1.004) 11.67(1.008) 11.33*16.25t 12.10( .999) 12.12(1.002) 11.56*16.53t ? 11.70*17.08t 12.28( .982) 12.23*17.50t 12.28(1.081) 12.19*17.40t * Cohen and Jones, ref. 29 t Rochester, ref. 31 0 Slope of plot of log I vs. H. - ?5 -TABLE XVIII Absorption maxima and molar ab s o r p t i v i t i e s of phenolate anions i n i n ^ Phenols DMSO solutions 85% DMSO-EtOH i n MeOH-MaOMe X e x 10-4 X E X IO\" 4 X e x I O - 4 max max max m unsubstituted 312 .28 290 290 2-isopropyl 290 .42 290 2,6-di-isopropyl 300 .46 2-t-butyl 318 .39 300 291 .36 2,4-di-t>butyl 320 .37 306 296 .34 2,6-di-t-butyl 318 -.50 306 298 .52 2,6-di-t-butyl-4-methyl 325 .53 317 307 .49 2,4,6-tri-t-butyl 325 .50 312 302 .44 * Rochester, ref. 31. - 96 -the absorption maxima and molar abso r p t i v i t i e s for the phenol molecules and their conjugate base i s given i n Table XVIII. 2,6-di-t^butylphenol was not used i n determining the scale since the plot of log I versus * solvent composition was not p a r a l l e l to simi l a r plots for 2-t-butyl-and 2,4-di-t-butyl-phenols. In point of fact i t i s more ionized i n a 20 mole percent DMSO-ethanol solution than 2,4-di-t-butylphenol and less ionized i n a 55 mole percent solution. The other two hindered phenols 2,4-di-t-butyl-6-methylphenol and 2,4,6-tri-t-butylphenol had slopes of 0.98 and 1.08 respectively for plots of log I versus H . These two indicators determined only two values i n 65 and 70 mole percent DMSO-ethanol. Since the other values determined by these indicators did not d i f f e r from those determined by p a r t i a l l y hindered phenols, they (65 and 70 percent values) were also included. Figures 16 and 17 show c l e a r l y that the scales, based on the io n i z a t i o n of phenols, d i f f e r s markedly from those based on the io n i z a t i o n of nitrogen and carbon acids, a given solution being always more basic towards the l a t t e r than the former.This r e s u l t i s not unexpected since the negative charge of the anion i s mainly centered on the phenolate oxygen. We can therefore expect the phenolate anion to react to solvents changes much l i k e a small ion which i s s t a b i l i z e d by hydrogen bonds i . e . i t w i l l be desolyated as the concentration of DMSO i n the medium i s increased, causing an increasein i t s a c t i v i t y . T h i s w i l l s h i f t the equilibrium shown i n * 2,6-di-t-butylphenol i n DMSO-ethanol mole %DMSO 10 20 30 40 45 55 ' 11.21 11.57 11.91 12.26 12.47 12.75 ri log I -0.63 -0.22 +0.07 +0.28 +0.39 +0.53 - 99 -equation (34) to the r i g h t , t n u s p a r t i a l l y counterbalancing the increase i n a c t i v i t y of the alkoxide ion. AH + R0~ t A\" + ROH (34) Conversely the amide ion and carbanions enjoy a large degree of charge d e r e a l i z a t i o n and t h e i r solvation requirements are much less stringent, thus s h i f t i n g the above equilibrium to the r i g h t . Presumably alcohols with no formal charge d e r e a l i z a t i o n possible i n the i r anions would generate a shallower H scale i n which any increased b a s i c i t y would be due merely to a concentration effect on the equlibrium shown i n equation (34). C One can also note from figure 17 that the H scale, with N a l l i t s defects, p a r a l l e l s the H scale i n ethanolic DMSO. This fact N gives some v a l i d i t y to the use of the H scale i n the estimation of the a c i d i t i e s of weak carbon acids, and for the correlation with rates of reactions where the rate determining step i s an ioni z a t i o n process from a carbon atom. - 100 -PART I I : Base catalyzed d e t r i t i a t i o n of hydrocarbons Rate constants for the base catalyzed d e t r i t i a t i o n of 9-phenylfluorene, 2,3-benzofluorene, fluorene, 9-ethylfluorene, 9-phenylxanthene, and triphenylmethane have been measured i n various ethanolic DMSO mixtures. These rates are l i s t e d i n Tables XXIX, XXX. The logarithm of these rates, log k, were correlated with the N H value of the medium and the linea r relationships obtained are shown i n figure 18. 9-phenylfluorene 2,3-benzofluorene Fluorene 9-ethylfluorene 9-phenylxanthene Triphenylmethane - 103 -Table XIX l i s t s the slopes of each of these lines and the corresponding coe f f i c i e n t s of correlation and student's test T. The slopes are s i m i l a r i n magnitude for a l l the hydrocarbons and have an average value of 0.87. In the f i r s t approximation, the less-than-unit slope can be attributed to the fact that the H scale used was derived with amine indicators and not with hydrocarbons; therefore, the H_ scale r e f l e c t s the response of nitrogen acids rather than carbon acids to the change i n medium. Indeed, i t i s surprising that such a good c o r r e l a t i o n i s N C obtained, suggesting that the H_ and H scales are quite s i m i l a r . N C One way to compensate for the use of H_ instead of H scale would be to plot log k for the various hydrocarbons against log k for a reference carbon acid to be chosen a r b i t r a r i l y . I t was thought that DMSO would be the most appropriate reference acid, since the solvent mixtures contain DMSO. And indeed, i f we plot log k, , , vs ' r • hydrocarbon log k^gQ the slopes of the correlations are much closer to unity (average of 0.95). Figure 19 shows a plot of such a c o r r e l a t i o n . The rates of d e t r i t i a t i o n of fluorene i n DMSO-water-0.01 M TMAOH have been measured, and Figure 20 shows a plot for the N correl a t i o n between log k and the H value of the medium. ' - The rates of the base-catalyzed t r i t i a t i o n of DMSO and d e t r i t i a t i o n of DMSO-t i n the three media (aqueous, methanolic and N ethanolic DMSO) were correlated with t h e i r respective H values. Figures 21, 22, and 23 show plots of such a correlation. Figure (2) - 105 -TABLE XIX N Correlation of log k with H_ for the base catalyzed d e t r i t i a t i o n of hydrocarbons Compound 0(slope) r N T a) log k vs. H. 9-phenylfluorene DMSO-EtOH .873 .998 5 30.7 2,3-benzofluorene \" .809 .999 4 80.5 fluorene \" .782 .997 8 38.4 9-ethylfluorene \" .804 .999 9 62.3 9-phenylxanthene \" .848 .999 5 82.1 triphenylmethane \" .937 .999 3 53.4 DMSO \" .901 .995 9 26.6 fluorene DMSO-water .790 .998 4 21.1 log k vs. log k (DMSO) 9-phenylfluorene DMSO-EtOH .983 .999 5 35.3 2,3-benzofluorene \" .904 .999 4 39.1 fluorene \" .914 .999 8 47.0 9-ethylfluorene \" .908 .998 9 46.4 9-phenylxanthene \" .928 .999 5 56.6 triphenylmethane 11 1.029 .999 5 23.6 r i s the coefficient of correlation N i s the number of points T i s the student's test - 109 -shows a t y p i c a l run for the t r i t i a t i o n of DMSO One can notice that the points are much more scattered than for hydrocarbon d e t r i t i a t i o n . Although the re p r o d u c i b i l i t y of the run i s within ±10%, the r e p r o d u c i b i l i t y of any one point i s quite unreliable. This i s mainly due to the method of aliquot preparation. As explained i n the experimental section, 3.5 mis of methanol was d i s t i l l e d from a DMSO and methanol mixture of t o t a l volume 8 mis. These 3.5 mis were then dried. A 3 mis aliquot was withdrawn and 85 added to 7 mis of the s c i n t i l l a t o r solution. I t has been found that methanol acts as a d i l u t e r - a solvent which i n the presence of decreasing toluene concentration produces a gradual drop i n pulse height - while water i s a strong quencher. The counting e f f i c i e n c y i n the DMSO exchange runs were h a l f as much as the counting e f f i c i e n c y i n the hydrocarbon runs, and a small v a r i a t i o n i n the dryness of the methanol or i n i t s purity w i l l affect the r e p r o d u c i b i l i t y of any single determination. Nevertheless the rates of t r i t i a t i o n of DMSO i n aqueous DMSO determined i n th i s work compare favorably with the 37 values published by Stewart and Jones . The l i n e drawn i n figure 37 (21) i s the least square l i n e from Stewart and Jones results and the points are those determined i n t h i s work. I, Evaluation of the ki n e t i c s We are measuring i n these kinetics the rate of loss of a c t i v i t y of a t r i t i a t e d hydrocarbon. The radioactive isotope - 110 -concentration i s very small at the beginning of the run (in trace quantity) and almost a l l of the a c t i v i t y i s lost to the solvent sink. In a t y p i c a l run where the a c t i v i t y at time zero i s ~ 500,000 counts/ min the a c t i v i t y at time i n f i n i t y i s less than 1000 counts/min. Internal return was shown to be of small importance even 86 for the least a c i d i c of these hydrocarbons. . By i n t e r n a l return i s meant the reaction i n which the hydrocarbon i s ionized and the leaving group i s then recaptured by the newly formed carbanion. For a l l these reasons i t seems reasonable to associate the rate of d e t r i t i a t i o n of 44 87 these hydrocarbons with t h e i r rate of i o n i z a t i o n V . The slow step of the isotopic exchange i s undoubtedly the dissociation of the carbon acid i . e . the removal of the t r i t o n by the base. The simplest mechanism one can write i s a simple one step abstraction with a hydrogen bonded t r a n s i t i o n state as 6- 6^ ArT + RO\" —* [Ar...T... OR] » Ar + ROT (35) 6 - 6 -Ar\" + ROH y [kv...H...OR] —* ArH + RO\" (36) The reverse reaction of equation (35) i s written i n equation (36) to describe the isotopic exchange. This mechanism though, does not account for the p o s s i b i l i t y of isotopic exchange with retention of configuration, i f the carbon attached to the exchangeable hydrogen i s asymmetric t h i s base catalyzed isotopic exchange with high retention 86 of configuration has been noticed by Cram et a l . on o p t i c a l l y active - I l l -4-biphenylmethoxyphenylmethane OCH3 P - C 6 H 5 C 6 H 4 | - H ( D ) C 6 H 5 and 2-phenylbutane-2-d C 2H 5 C,H,.-*C-D(H) D O j CH 3 -• and 1-phenylmethoxyethane-l-d OCH I C6H5-*C-D(H) I CH 3 Another mechanism for proton transfer can be vi s u a l i z e d . I t involves an equilibrium step, giving r i s e to a hydrogen-bonded carbanion, followed by an isotopic exchange step, and f i n a l l y a collapse of the complex to give the non-radioactive hydrocarbon. Considering only the forward reaction, k l k2 ArT + OR\" > Ar~...TOR * Ar\"...HOR * ArH + OR\" (37) • k - i ' I I I I I I - 112 -we w i l l t r y to deduce the nuclear configuration of the t r a n s i t i o n state and i n p a r t i c u l a r where the proton (triton) l i e s . Applying a steady-state treatment d[Ar ...TOR] = k [ArT][OR\"] - k [Ar~...TOR] - k [Ar\"...TOR] = 0 (38) dt k l -[Ar-...TOR] = - [ArT] [OR ] (39) k - l + k2 d[Ar ...HOR] = k2[Ar\"...TOR] = — [ArT] [OR-] (40) dt k L l + k 2 k 2 k l • and therefore k , = (41) ° b S k _ 1 + k 2 2 Cram et a l have proposed a dual mechanism (cf introduction) d i f f e r e n t -iated according to the r e l a t i v e magnitude of k_^ and k 2 > I f k 2 » k ^ i. e . i f II goes to I I I faster than i t returns to I then equation (41) reduces to (42) kobs = k l <42> The proton transfer takes place immediately after the hydrogen bonded complex i s formed and before they can separate due to the thermal motion, i . e . i t s equivalent to having no preequilibrium as i n equations (35) and (36). I f k ^ » k 2 i . e . some kind of preequilibrium i s rapidly established followed by a rate determining step. In t h i s case equation (41) reduces to:, - .113 -k l kobs = — k2 = K k 2 W k - l Obviously the t r a n s i t i o n states corresponding to these two mechanisms are quite different from one another, and i t would seem quite unexpected to obtain a good Bronsted corre l a t i o n for a l l these hydrocarbons, i f the t r a n s i t i o n states are not s i m i l a r . Also the slope 0 of these correlations (log k vs H_) are of the same order of magnitude and therefore i t would seem somewhat unlikely that a change i n mechanism occurs. Cram also proposed a single mechanism for the i s o t o p i c exchange of hydrocarbons where the rate determining step i s I —> II and with a t r a n s i t i o n state of increasing asymmetry. In th i s case the rate equation can be expressed . = k [ArT] [HO\"] f A r l f R ° \" (44) dr a f + f f -which gives k Q b s = k a[RO \" ] . ' - M J « L (45) One can derive the H function for an equilibrium (46) by considering the AH + RO\" — ^ A\" + ROH (46) equilibrium constant _ aROHaA~ HA ~ a a 1-4/J H A AH RO - 11.4 -[ A \" ] LAH] f f -AH RO ROH [RO\"] (48) and therefore h f f -AH RO [RO\"] ROH (49) Combining equation (49) and (45) we get k , = k . ( h ) a n r, u obs a v - ROH taking the logarithms log k ArT obs log k a + H_ + log a R Q H + log ArT fA-AH AH (50) (51) The a c t i v i t y c o e f f i c i e n t term has t r a d i t i o n a l l y been assumed to be constant - i f not equated to zero - because of the s i m i l a r i t y between the indicator acids (AH) used to generate the H scale and the k i n e t i c substrate (ArT) and also the s i m i l a r i t y between the t r a n s i t i o n state (^ and the indicator anion (A ). Therefore equation (51) reduces to: log k , .= constant + H + log a-,-,, 6 obs - 6 ROH (52) ,38 Equation (52) resembles the one derived by Anbar et a l for base catalyzed reactions where the slow step i s a proton abstraction from the reactants (cf. introduction p.21 ), which i s s i m i l a r to the mechan-ism proposed for the d e t r i t i a t i o n of hydrocarbons and shown i n equation (37). - 115 -There are no published data for the a c t i v i t y of ethanol i n DMSO-ethanol mixtures, but a c t i v i t y c o e f f i c i e n t s have been determined 6 7 for water i n DMSO-water and sulfolane-water mixtures and also for 8 methanol i n DMSO-methanol mixtures. The rates of d e t r i t i a t i o n of 9-phenylfluorene (9-t) were therefore determined i n methanolic DMSO containing 0.01 M sodium methoxide. The l i n e 1 i n Figure 24 represents the corre l a t i o n between log k and H while l i n e 2 represents the correlation between log k and H + log a^eQ^. The lines 1 and 2 i n Figure 23 are the corresponding correlations for the isotopic exchange of DMSO i n methanolic DMSO. TABLE XX Correlation of log k with H and H + log a^Q^ f° r the base catalyzed i s o t o p i c exchange i n methanolic DMSO. . Compound System 6 a) log k vs. H 9-phenylfluorene DMSO-MeOH .770 DMSO \" .730 b) log k vs. H_ + log a^y.' 9-phenylfluorene DMSO-MeOH .859 DMSO \" .878 r N T .996 8 27.2 .993 12 27.1 .997 8 30.5 .994 12 28.1 - 117 -The net re s u l t of p l o t t i n g log k for 9-phenylfluorene and DMSO versus H_ + log a^gQj^ i s thus to steepen the curve as can be seen from Table XX. The value of log ^ qq^ i s p r a c t i c a l l y equal to zero up to 10 mole percent DMSO and decreases slowly to -0.21 for 30 mole percent DMSO then more rapidly as the DMSO content increases to 100 percent. Therefore since the slopes of log k, , , vs log k W^f,_ plots are r 6 hydrocarbon 6 DMSO r very close to unity (Table XIX) and that log k vs H_ + log a^gOH have C slopes which are 0.1 less than the former correlations, the H scale for DMSO systems should r i s e only s l i g h t l y more rapidly with the DMSO content than the corresponding scale. A j u s t i f i c a t i o n for the observed l i n e a r correlation between the logarithm of the rate constants and H has thus been obtained. I I . Activation parameters Activation parameters for these d e t r i t i a t i o n reactions have been measured ho r i z o n t a l l y and v e r t i c a l l y i n Figure 18. By horizon-t a l l y i s meant that an arbitrary rate constant at 25° was chosen for a l l the hydrocarbons (obviously the medium was quite different) and the enthalpy of activation was measured. V e r t i c a l l y means that the activa-t i o n parameters were measured i n one and the same medium (65% DMSO-EtOH) and the temperature range varied although the temperature l i m i t s are only 60° apart (from -5° to +55°). Table XXXII appendix B l i s t s the rate-temperature data for these d e t r i t i a t i o n reactions and Table XXI l i s t s the activation parameters deduced from these k i n e t i c s . - 118 -TABLE XXI Activation parameters for the base-cataiyzed d e t r i t i a t i o n of hydrocarbons. %DMSO kcal.mole AS T e, u. a) v e r t i c a l Triphenylmethane 62.5 23.3 ± 0.6 -2.3 ± 2.0 9-phenylxanthene 62.5 21.1 ± 0.9 0.5 ± 3.0 9-ethylfluorene 62.5 16.7 ± 0.7 5.9 ± 2.6 Fluorene 62.5 14.4 ±0.3 10.2 ± 1.2 b) horizontal 9-phenylxanthene 70 22.1 ± 1.0 5.6 ± 3.2 9-ethylfluorene 35 18.6 ± 0.8 -7.1+2.6 Fluorene 10 17.2 ± 0.3 -12.4 ± 1.1 4,4',4\"-Trinitrotriphenylmethane has an activation energy of 11.3 ± 0.2 k c a l . mole ' for i t s i o n i z a t i o n by sodium ethoxide i n an ethanol-toluene mixture One can see that as the hydrocarbon becomes less acidic the enthalpy of activation r i s e s . This general trend i n the enthalpy of activation has also been observed by two other groups of workers -2 22 Cram et a l • and Streitwieser et al'. - and would tend to indicate. that the amount of bond-breaking i n the t r a n s i t i o n state increases with increasing b a s i c i t y of the hydrocarbon. The enthalpy of - 121 -activation should increase as ApK increases, since the extent of the bond breaking i n the t r a n s i t i o n state w i l l increase as the R0~ moiety binds the proton more closely. Thus the greater the activation energy the more the t r a n s i t i o n state resembles the product I I . The significance of the rather small v a r i a t i o n i n activation entropy i s uncertain owing to the limited number of hydrocarbons and the high experimental error. One can notice a trend toward more pos i t i v e values as the hydrocarbon becomes less acidic and conclude that the charge must be more delocalized i n the t r a n s i t i o n state derived from •J* the weaker acids. Although Cram has found a change i n AS of 22 e.u. for the d e t r i t i a t i o n of hydrocarbons i n 75% DMSO-MeOH-KOMe at 75°, t Streitwieser found a very small change i n AS (6 e.u.) for the same reaction i n methanol at 45°. Also the rate - increase on going to higher DMSO concentration (higher H values) i s accounted for mainly t by the decrease i n enthalpy of act i v a t i o n ; AH for fluorene i n 10% DMSO i s 17.2 and i n 62.5% DMSO i s 14.4. This dependence has also been observed by Cram et a l . i n the catalyzed racemization of (+)-2-methyl-3-phenylpropionitrile i n DMSO-88 methanol solutions. Parker and coworkers also noted that p r o t i c -^dipolar aprotic)solvent effects on rate are usually re f l e c t e d i n the enthalpy rather than i n the entropy of activ a t i o n , and are quite strongly temperature dependent. This i s to be expected since these effects are mainly attributed to hydrogen bonding interactions -solvation. - 122 -I I I . Kinetic Thermodynamic correlation for hydrocarbon i o n i z a t i o n (log k vs ApK) In the case of hydrocarbons the electronic and s p a t i a l configuration of the acids and t h e i r conjugate bases are not i d e n t i c a l , and therefore the Bronsted plot w i l l show a broad t r a n s i t i o n from a = 0 to a = 1, which occurs over a wide ApK range. By extrapolation of the lines i n Figure 18 to an H of N 13.45 (H for a 0.01 M ethanolic sodium ethoxide solution) we can compute a k i n e t i c a c i d i t y for these hydrocarbons, and taking a value of 16.0 for the pK of ethanol, ApK values can be calculated. The 3. correlation of k i n e t i c and thermodynamic a c i d i t i e s i s shown i n Figure 27. The average slope a of the l i n e i s 0.5 but a curve w i l l f i t better the results with an i n i t i a l slope of 0.3 and a f i n a l slope of 0.6. The thermodynamic ac i d i t y of fluorene and 9-ethylfluorene 22 are s i m i l a r while t h e i r k i n e t i c a c i d i t y d i f f e r s by ~2 pK & units. Streitwieser correlated the rates of d e t r i t i a t i o n of 9-alkyl sub-s t i t u t e d fluorenes with the pK of the corresponding acetic acids :and 3. found that the r e l a t i v e rates are i n the prder F > 9-MeF > 9-EtF. But equilibrium measurements showed that the order of decreasing a c i d i t y i s 9-Me > F = 9-Et > 9-isopropyl > 9-t-butyl, which indicates a k i n e t i c . e f f e c t i n t h i s series. In Figure 29 at ApK = 0, the slope of.the log k -ApK curve i s not 0.5 but 0.3. This i s to be expected though since the acceptor ROH and the donor RgCH are of d i f f e r e n t acid-type, 0- and C- type respectively^. a o \" Figure 27. Correlation of k i n e t i c and thermodynamic a c i d i t i e s of some hydrocarbons, I 1 c O . I CD O a s . I o o •—». I ApK 9-phenylfluorene 0.4 2,3-benzofluorene 4.0 Fluorene 4.5 9-ethylfluorene 4.6 9-phenylxanthene 10.0 Triphenylmethane 13.0 o I -2.0 0.0 2.0 4.0 6.0 DELTA P K a.o 10.0 12.0 14.0 - 124 -IV. Change of Isotope effect with the pK^ .of hydrocarbons No measurements of isotope effects were made by us, i n the 2 ethanolic DMSO system, but both Cram et a l . and Streitwieser and 21 coworkers found a general trend i n the v a r i a t i o n of the magnitude of the isotope effect with the pK a values. Cram et a l . i n 75% DMSO-MeOH (by volume) - 0.0026 M potassium methoxide found an isotope effect ky/kp of 6.75 at 25° for 2-(N,N-dimethylcarboxamido)-9-methyl fluorene (pK ~20) and 2.5 at 126° for 4-biphenylmethoxy phenylmethane (pK ~30) a . 21 Streitwieser and coworkers found a primary isotope effect k^/k^, i n sodium methoxide of 2.5 (25°) for 9 phenylfluorene (pK 16.5) while 3-kp/k^, for triphenylmethane i s 1.3 (99°). The progressive decrease i n the isotope effect i s quite marked, and one should note that for triphenylmethane,pK 29.2, (ApK ~ 13) the isotope effect i s about h a l f that f o r 9-phenylfluorene for when pK 16.5 ( pK ^ 0). This lowering i n k i n e t i c isotope effect with increasing ApK w i l l tend to indicate an increase i n asymmetry i n the t r a n s i t i o n state 51 53 This i s i n l i n e with recent suggestions, ' that the isotope effect w i l l be greater when the t r a n s i t i o n state i s symmetrical and the symmetrical stretching mode v w i l l not involve any motion of the proton. This would probably happen when both anion A and B are (A....H....B (v, no motion of H) of si m i l a r strength i . e . ApK - 0. When on the other hand one anion i s stronger than the other, the proton w i l l be closer to the weaker base and the stretching w i l l become unsymmetrical and w i l l lead to a smaller A...H...B (imaginary - 125 -frequency iv^) isotope e f f e c t . This i s exactly what we notice when ApK > 0 and the proton i s closer to the RO- moiety than to the carbanion. This mechanism thus does not require the isotope effect to vanish for high ApK values and i s one more reason to prefer t h i s mechanism to Cram's dual mechanism. Of course i f at ApK ~ 0 there i s a maximum i n the isotope effect c o r r e l a t i o n with ApK one should be able to f i n d an isotope effect of less than 7 for reactions where the carbon acid i s a weaker base than ROH.. Thus a l l the k i n e t i c data are accommodated by mechanism (37). There i s no rea l way to discriminate - at the present time -between mechanisms (35) and (37) . ArT + RO\" — * Ar\" ....TOR ;^ =* Ar\" HOR - j — * . ArH + RO\" (37) k f ArT (H) + RO ; Ar + ROT (H) (35) k r In both cases the rate equation i s the same, and the configuration of the t r a n s i t i o n state of highest energy appears to be [Ar T.. . .OR]\" with T i n a symmetrical position i f ApK~ 0 and with k H A n ~7 and closer to RO i f ApK > 0 and with k H A n < 7 and closer to Ar i f ApK < 0 and with k^/k^ < 7. But of course the mechanism shown i n equation (35) cannot account for isotopic exchange reaction with retention of : . . 86 configuration - 126 -Ideally one would l i k e to have an H_ scale for carbon acids and use these same indicators for is o t o p i c exchange measurements, thus eliminating the uncertainty i n the value of the a c t i v i t y c o e f f i c i e n t terms i n equation (51) f f -ArT 2A t f f All Also i f one had values for the a c t i v i t y of ethanol i n DMSO mixtures that would eliminate the other unknown parameter i n equation (51), namely log a R Q H . - 127 -PART I I I : Interactions of Nitrophenylmethanes with base. An attempt to measure the degree of i o n i z a t i o n of 3-4'-dinitrodiphenylmethane i n ethanolic sodium ethoxide showed that the u.v. absorption maximum of the \"anion\" had shift e d to 430 mu from the 575 mu value observed i n DMSO containing sodium ethoxide. E.s.r. spectroscopy indicated the presence of a large concentration of radicals (~10 - 4 M i n -10 - 2 M indicator solution) i n these basic medium. A s i m i l a r behaviour was observed for two other carbon acids indicators, namely t r i s - and b i s - (p-nitrophenyl)methane. 17 19 Bowden and Stewart and Ritchie and Uschold also noted that the spectrum of the conjugate base of tris-(p-nitrophenyl)methane showed a pronounced solvent effect (A. of'582 mu i n ethanol, and - 128 -\"Since the formation of highly coloured solutions i n the presence of base has often been considered i n d i c a t i v e of proton abstaction (e.g. see reference 17) t h i s large s h i f t i n absorption maxima suggests that another process i s taking place. A variety of interactions can occur between aromatic nitro-compounds and the unshared electron p a i r of the base, with i n most cases, a conco-90 amitant formation of highly coloured solutions. Buncel, Norris and Russell i n t h e i r review l i s t e d the different species r e s u l t i n g from such reactions ^and the a n a l y t i c a l techniques used to define them. For example, a p a r t i a l ^transfer of electronic charge from the base to the aromatic nucleus can ^give r i s e to n-complexes (charge-transfer complexes). A radical-anion can -be produced i f an electron i s completely transferred from the base to the n i t r o compound. The electron p a i r of the base can also participate i n the formation of a covalent bond with an aromatic carbon atom to give a .o-complex or \" Meisenheimer \" type complex. Because the n i t r o group has a strong acid strengthening e f f e c t , the base could abstract a proton to form a carbanion. The possible existence of ion pairs as well as free ions i n the cases of proton abstraction and radical-anion.formation further complicates the picture. Ion pairs themselves can be d i f f e r e n t i a t e d into solvent separated and contact ion p a i r s . 91 Russell and Janzen have studied extensively the reactions of o- and p- nitrotoluenes i n strongly basic systems. They postulated that a spontaneous disproportionation reaction takes place to form the 91 radical-anion derived from the parent nitrotoluene. These authors suggested a mechanism for t h i s disproportionation reaction i n t-butanol containing potassium t-butoxide, i n which the rate l i m i t i n g step i s the - 129 -i o n i z a t i o n of the benzylic proton. p-N0 2C 6H 4CH 3 + B\" • p-N0 2C 6H 4CH 2\" + BH (53) p-N0 2C 6H 4CH 2\" + p-N0 2C 6H 4CH 3 • p-NO^H CH2\" + p-'CH CfiH N02* (54) Two molecules of the ra d i c a l p-NC^ C^ H^ H,^ '- then couple to give p-p'-dinitrobibenzyl (p-NO^^H^H^H^^H^C^-p) which can be isolate d from the reaction mixture . In 80 percent DMSO-t-butanol-potassium t-butoxide, the dispro-91 portionation reaction i s very rapid. Russell and Janzen postulated a different mechanism i n th i s case: The rate determining step i s now the one shown i n equation 54 and the io n i z a t i o n step shown i n equation 53 becomes a fast pre-equilibrium. This i s probably due to the higher b a s i c i t y of the DMSO solutions. Russell and Janzen suggested that t h i s disproportionation reaction i s quite general and would occur whenever hydrogen atoms are a to an eas i l y reducible group which also promotes the a c i d i t y of these hydrogens. I. N.m.r. investigation The d i s p a r i t y i n the pK values of the carbon acids reported i n Tables I I I and IV , the high degree of solvent dependence of the anions' spectral characteristics and the presence of appreciable amounts of radical-anions i n basic solutions of nitrophenylmethanes led to the u t i l i z a t i o n of high resolution n.m.r. as an independent method for the - 130 -characterization of the i o n i c species formed i n basic solutions. N.m.r. has been used previously to define the mode of ion i z a t i o n of amines i n 73 92 a l k a l i n e media . Crampton and Gold have shown that 2,4-dinitroaniline and 2,4-dinitrodiphenylamine i n DMSO as a solvent and i n the presence of methanolic sodium methoxide ionize to give an amide ion v i a proton abstrac-t i o n . 69 In 1966 B i r c h a l l and J o l l y established a scale of a c i d i t i e s f o r anilines i n l i q u i d ammonia* i n presence of sodium amide by an n.m.r. 69 93 method. The same authors ' have attempted to study some hydrocarbons using the same technique but have not yet published the results of t h e i r investigation. A basic system i n which, hopefully, only proton abstraction occurs was sought f o r the nitrophenylmethanes. Since hexamethylphosphora-mide (HMPT) containning sodium methoxide i s easier to handle than l i q u i d ammonia, i t was used for t h i s purpose.Preliminary results show that an HMPT-0.01 M TMAOH solution i s some 10 4 times more basic* than a DMSO-0.01 M TMAOH solution. Tris-(p-nitrophenyl)methane TNPM The n.m.r. spectrum of TNPM i n HMPT (0.042 M solution) consists of a sharp resonance at 6= -6.60 (r e l a t i v e area 1) which i s assigned to 94 Badoz-Lambling et a l . have developed a hydrogen electrode suitable for ac i d i t y measurements i n l i q u i d ammonia at -60°. Lagowski et a l have also used l i q u i d ammonia as a strongly basic system and determined spec-trophotometrically the a c i d i t y of o- and p- nitr o a c e t a n i l i d e and di-p-t o l y l - and di - p - a n i s y l - methane. In a 91.2 mole % HMPT-water solution containning 0.01 M TMAOH, Aniline (pK =27.3) i s one-tenth ionized and 3-trifluoromethylaniline (pK =25.4) i s completely ionized. These results would thus indicate an H y^alue of ~26.4. The H value for the corresponding DMSO solution is21.2 - 131 -the methine hydrogen and an AA'XX' multiplet centered around 6 = -8.02 (re l a t i v e area 12) due to the aromatic hydrogens (figure 28a ). The addi-t i o n of less than one equivalent of anhydrous sodium methoxide to th i s solution gives an intense green colouration. The int e n s i t y of the spectrum of the nitrophenylmethane decreases with the simultaneous appearance of another multiplet at higher f i e l d (figure 28b ). Both multiplets are quite sharp indicating a r e l a t i v e l y slow rate of exchange between the two species. When more than one equivalent of base was added the o r i g i n a l spectrum completely vanished leaving only the new multiplet centered around 6 = -7.56 (figure 28c ). This spectrum i s i n agreement with that expected for the exclusive formation of an anion by proton removal from the methine carbon. The v i s i b l e spectrum of the ion was determined d i r e c t l y i n a 0.0024 M solution and exhibited an absorbance at a X of 770 my max (e = 43,400). - 132 - • Figure 28. N.m.r. spectra of tris-(p-nitrophenyl)methane and anion in HMPT. •• '• 1 i • I • . . . . . I • I . 1 I . I 9 8 7 6 - 6 ( p p m ) — — » - 133 -Bis-(p-nitrophenyl)methane BNPM The n.m.r. spectrum of BNPM i n HMPT (0.058 M solution) consists of a single resonance at 6 = -4.45 (r e l a t i v e area 1 ) due to the methy-lene hydrogens and a AA'XX1 multiplet centered at 6 = -8.04 (r e l a t i v e area 4) assigned to the aromatic hydrogens (figure 29 a). Addition of anhydrous sodium methoxide to th i s solution gives a dark green colouration and causes a decrease i n the in t e n s i t y of the spectrum of BNPM along with the appearance of a new spectrum at higher f i e l d (figure 29 b). Both resonances are quite sharp and complete conversion to the spectrum at higher f i e l d i s possible with the addition of more base. The-.new spectrum consists of a sharp resonance at 6 = -5.44 (r e l a t i v e area 1) and a AA'XX1 multiplet centered around <5 = -7.27 (r e l a t i v e area 8) (figure 29 c). This new spectrum i s ind i c a t i v e of the formation of the BNPM carbanion. — The v i s i b l e spectrum of the carbanion was determined d i r e c t l y i n a 0.024 M solution and exhibited an absorbance at a A of 787mmy max (e = 58,300). - 134 -Figure 29. N.ni.r. spectra of bis-(p-nitrophenyl)methane and anion in HMPT. 9 8 7 6 5 -&(ppm) »-- 135 -3,4'-dinitrodiphenylmethane 3,4'-DNPM The n.m.r. spectrum of a 0.065 M solution of 3-4*-DNPM i n HMPT i s shown i n figure 30 a and consists of a sharp resonance at 6= -4.43 (rela t i v e area 1) due to the two methylene protons as well as a series of lines at low f i e l d C r e l a \" t i v e area 4) assigned to the aromatic protons. The resonances due to the protons from the 4'-nitro ring are d i s t i n c t from the resonances from the 3-nitro ring because of t h e i r symmetrical arrangement . They consists of a multiplet at 5 = -8.05 with J A A , = 2.2 - JXX' = 2 ' 2 ' J A X ' = J A ' X = 0 ' 5 ' J A X = JA'X< = 8\"2. A N D V A X = 46.1.cps. where v 6 A V represents the chemical s h i f t between nuclei A and X. o AX R Careful addition of an equivalent of sodium methoxide to th i s solution results i n the disappearance of the substrate spectrum with the ap-pearance of a new spectrum at higher f i e l d s . But unlike the two pre-vious nitrophenylmethanes studied, addition of less than one equivalent of base causes a broadening (specially at high substrate concentration ii 0,5 H) i n both spectrum at high and low f i e l d s . The second spectrum can be rat i o n a l i z e d i n terms of the forma-ti o n of the carbanion of 3,4'-DNPM. ,N0„ \\\\ / \\ H H NO, C-^H The resonances at 6 = -7.95 and -7.38 are assigned to the 2 and - 136 -Figure 30. N.m.r. spectra of 3,4 *-dinitrodiphenylmethane and anion i n HMPT. - 137 -4,5,6 hydrogens respectively. The hydrogens of the symmetrical ring give r i s e to four doublets of major s p l i t t i n g s ~8 c.p.s. and smaller s p l i t -tings - 2 c.p.s. centered at 6 = -6.02, -6.70, -7.26 and -7.44 ppm.. If t h i s i s indeed the case, then the doublet at 6 = -7.44 must be p a r t i a l l y hidden by the resonances of the second ri n g . I t i s possible to achieve p a r t i a l decoupling of the multiplet at 6 = -6.7 by i r r a d i a t i o n of the resonance at 6 = -7.44. S i m i l a r l y , the doublet at 5 = -6.02 i s decoupled by i r r a d i a t i o n of the doublet centered at 6 = -7.26 and vice versa. These decoupling experiments thus confirm the v a l i d i t y of the peak assignment. The v i s i b l e spectrum of the carbanion was determined d i r e c t l y i n a 0.0073 M solution and exhibited an absorbance at a A of 593 mp max (e = 28,000). No detectable amount of ion i z a t i o n by proton abstraction was found by n.m.r. for 3,4'-DNPM, when DMSO i s used as a solvent instead 96 of HMPT. Bis- and t r i s - (p-nitrophenyl)methane ionize i n DMSO only * to the extent of ^ 35 and - 75 percent respectively . Addition of 10 — 20 percent methanol to a solution containing a mixture of the anion and neutral BNPM i n HMPT causes substantial broadening of the substrate spectrum, but does not affect the anion spectrum. This probably results from rapid exchange between the substrate and the radical-anions produced i n the system. At high methanol concen-trations both spectra broaden and eventually disappear. * The apparent molar absorptivity of tris-(p-nitrophenyl)methane i n DMSO containning sodium methoxide i s =70 percent of the value obtain-ned i n HMPT containing sodium methoxide. - 138 -Both these observations would indicate that i f the solvent sys-tem i s not basic enough to ionize the nitrophenylmethane by proton abst-raction, side reactions become very important. This i s obvious from the fact that the r e l a t i v e amount of carbanion formation increases with inc-reasing a c i d i t y of nitrophenylmethanes i n the DMSO-sodium methoxide system. I I . E.s.r. investigation 91 Russell and Janzen noted that 2,4-dinitrotoluene radical-, anions are produced only when the solvent mixture i s deficient i n base, while o- and p- nitrotoluenes radical-anions are produced at a l l base concentrations. This observation can also be explained since 2,4-dinitro-toluene i s probably completely ionized i n the DMSO-t-butanol mixture while o- and p- nitrotoluenes are not. To examine the nature of the absorbing species at lower wave-length, the radical-anions of the nitrophenylmethanes were generated elec-14 trochemically i n DMSO-methanol mixtures. I t was found that the N hyper-97 fine s p l i t t i n g constant (a^) was strongly solvent dependent. Nitrobenzenes 99 and n i t r i c oxides radical-anions also exhibit strong solvent dependent 98 99 N s p l i t t i n g s . Adams et a l and Pannell reported several cases of a^ dependence on solvent composition. A quantitative treatment has been given by Gendall et al^P^ who assumed that the changes i n s p l i t t i n g s arise from r e d i s t r i b u t i o n of the n-electron spin density near the functional group. Gendall et a l 1 0 0 also assumed that the interaction between the solvent and the r a d i c a l i s sgnificant only at the s i t e of the functional group. Furthermore, i n a binary solvent mixture, i f the exchange between the two solvent-radical - 139 -complexes i s rapid enough, a model can be derived which can account for * the hyperbolic dependence of the hyperfine s p l i t t i n g s with the solvent composition r a t i o . Kennedy\"^ has shown that a simple extension of t h i s model w i l l give a linear dependence of the s p l i t t i n g on the mole f r a c t i o n of the solvent. Table XXII l i s t s the N s p l i t t i n g s a^ with the mole percent DMSO in methanolic DMSO mixtures and these values are plotted i n figures 31 and 32. I t i s obvious that a good straight l i n e can be drawn through these points, thus confirming the correctness of the model proposed by Gendall'''^ and altered by K e n n e d y i . e . the solvent complexes are long l i v e d with respect to t h e i r exchange time and the major effect on the spin density comes from the interaction of the solvent with the functional group. The e.s.r. spectra of tris-(p-nitrophenyl)methane radical-anions were rather broad and although a d e f i n i t i v e change i n a^ values was noted i n going from pure methanol to pure DMSO, the inaccuracy i n the measurements did not warrant t h e i r p l o t t i n g . 102 Chambers and Adams reported that the absorption maximum of the radical-anion of nitrobenzene s h i f t s to lower wavelength with increa-sing water concentration i n dimethylformamide (DMF). A blue s h i f t of 42 my. was observed..for an addition of 20 weight percent water to DMF. 102 Chambers and Adams correlated t h i s solvent s h i f t of the absorption maximum with the change i n a^. Although i n th i s work, no direct determination of the u.v. spectrum of electrochemically generated radical-anions was attempted, a change i n the colour of these radicals was observed v i s u a l l y . twice the difference of the average s p l i t t i n g from the arithmetic mean - 140 -TABLE XXII S p l i t t i n g constants a^ for BNPM and 3,4'-DNPM i n methanolic DMSO mixtures. mole % DMSO BNPM 3,4'-DNPM TNPM 0 1 24 53 73 97 100 13.3 colourless 13.3 colourless 12.7 f a i n t yellow 11.8 blue 11.24 green-blue 10.15 green 10.0 green 13.15 colourless 12.5 f a i n t yellow 11.6 l i g h t blue 11.0 blue 10.1 blue 13.6±1.6 colourless 12.5+0.6 9.611.6 green - 141 -Figure 31. Plot of the average splitting constants versus mole percent DMSO for the radical-anion of 3,4•-dinitrodiphenylmethane in methanolic DMSO. 2,0 4 0 6 0 8 0 M O L E P E R C E N T D M S O 100 - 142 -- 143 -Table XXII l i s t s the colour of the different DMSO-methanol solutions after the radical-anions were generated. One can estimate at -200 my the s h i f t i n absorption maxima between methanol and DMSO. I f we assume that there exists a quantitative c o r r e l a t i o n between the absorption maxima of the radical-anions and t h e i r s p l i t t i n g * constants , then we can account f o r the large solvent s h i f t s observed i n basic ethanolic DMSO solutions. The carbanions of the nitrophenylmethanes have been i d e n t i f i e d by ** n.m.r. and t h e i r u.v. spectra d i r e c t l y recorded i n the same solutions. Therefore the u.v. spectral characteristics of these carbanions are unequivocally defined and any absorption which occur at lower wavelength probably arises from other i o n i c species. Table XXIII l i s t s the spectral data determined i n t h i s work, 17 18 as wel as those determined by Bowden and Stewart and Kroeger The accumulated evidence thus suggests that i n the methanolic 18 sodium methoxide system, the u.v. absorption observed arise from the nitrophenylmethane radical-anions. In ethanolic DMSO the s i t u a t i o n i s more complex, since both carbanions and radicals are present i n the same solution. At high DMSO concentrations TNPM i s probably almost completely in the carbanionic form (X = 800 my) while the absorption at a X • max • J r . max of 580 my i n ethanol indicates a preponderance of rad i c a l s . As a rule of thumb, the absorptions around 750-800 my for TNPM and BNPM and around 600 my for 3-4'-DNPM can be attributed to the carbanions, while those _ i . e . a correlation s i m i l a r to the one observed by Chambers and Adams. ** u.v. c e l l s of path-length 0.1 and 0.01 mm have been used,but i t was necessary to d i l u t e the solutions by a factor of -15 to be able to determine the molar absorptivity on a Cary Model 16 spectrophotometer. - 144 -TABLE XXIII Spectral data of nitrophenylmethanes i n strongly basic solutions, Compound/solvent HMPT/NaOMec DMSO/EtOH MeOH/NaOMe TNPM BNPM 3,4'-DNPM X my e max 775 43,400 787 58,300 593 28,000 my max e X my max e 707 19,300 530 25,600 809 30,300d 810 63,000e 704 40,600 534 28,000 707 41,9008 550 31,700 430 f 570 28,800g values determined i n t h i s work i n HMPT containing sodium methoxide b 17 values determined by Bowden and Stewart i n ethanolic DMSO containing 0.01 M sodium ethoxide c 18 — values determined by Kroeger i n methanolic sodium methoxide 0.01 M[0R] d value determined i n t h i s work i n DMSO containing sodium methoxide EDMS0 ~ 7 0 % ° £ t h e £HMPT ° £* f o o t n o t e P.age 1 3 7 • e 19 value determined by Ritchie and Uschold i n DMSO f value determined i n t h i s work i n methanolic sodium methoxide, the molar absor p t i v i t i e s could not be measured as the medium i s not basic enough to completely ionize t h i s compound. g value redetermined i n t h i s work i n 90% DMSO-ethanol-0.01 M sodium ethoxide. - 145 -at lower wavelength arise either from the exclusive presence of radicals or a mixture of radicals and carbanions. Preliminary investigation of the interactions between n i t r o -phenylmethanes and basic DMSO solutions showed that a variety of synergic reactions occur. These should be investigated more c a r e f u l l y as t h i s 91 type of reaction appear to be quite general . The nitrophenylmethanes cover a wide range of a c i d i t i e s and are thus i d e a l l y suited for such an investigation. - 146 -CONCLUSION It i s obvious from the results i n th i s thesis that only carbon acids which ionize by proton abstraction should be used to determine a c i d i t y functions. In th i s area of study evidence should be obtained using more than one a n a l y t i c a l technique since the t r a d i t i o n a l method, 90 u.v. spectrophotometry, i s inadequate for the i d e n t i f i c a t i o n of ion i z a t i o n processes. The whole f i e l d of a c i d i t y determination, both thermodynamic and k i n e t i c , i s s t i l l i n i t s infancy as a result of our limited knowledge of solvent effects. Probably a step i n the right d i r e c t i o n would be to conceive a reference electrode whose response i s independent of the 103 medium. Recently Brenet considered the theoretical aspects of such a reference electrode i n non-aqueous systems. - 147 -Bibliography 1. K. Bowden, Chem. Rev., 66_, 119 (1966). 2. W.D. Kollmeyer and D.J. Cram, J . Am. Chem. S o c , 90_, 1784, 1791 (1968). 3. R. Stewart, Quart. Reports on Sulfur Chem., 3_, 99 (1968). 4. a) D. Dolman and R. Stewart, Can. J . Chem., 45_, 911 (1967). b) D. Dolman, Ph.D. Thesis, University of B r i t i s h Columbia, Vancouver, 1966. -5. M. Eigen, Angew.Chem. Intern. Ed., 3_, 1 (1964). 6. J . Kenttamma and J . J . Lindberg, Suomen Ke m i s t i l e h t i , B33, 98 (1960) 7. R.L. Benoit and G. Choux, Can..J. Chem., 46_, 3215 (1968). 8. K Quitzsch, H. Ulbrecht and G. Geiseler, Z. Phys. Chem. (Leipzig), 234, (1-2) 33 (1967). 9. A.J. Parker, a) Quart. Rev. London, 1_6, 163 (1962). b) Chem. Rev. 69, 1 (1969). 10. R.F. Rodewald, K. Mahendran, J.L. Bear and R. Fuchs, J . Am. Chem. Soc., 90, 6698 (1968) . 11. C. Agami, B u l l . Soc. Chim. France, 1021 (1965). 12. G. Y a g i l , J . Phys. Chem.,.73_, 1610 (1969) . 13. a) R. Stewart and J.P. O'Donnell, Can. J . Chem., 42_, 1681, 1694 (1964). b) J.P. O'Donnell, Ph.D. Thesis, University of B r i t i s h Columbia, Vancouver, 1962. 14. R..W. Taft J r . , J . Am.\" Chem. S o c , 82_, 2965 (1960). 15. K. Bowden, A. Buckley and R. Stewart, J . Am. Chem. S o c , 88, 947 (1966). 16. D.J. Cram, \"Fundamentals of Carbanion Chemistry\", Academic Press, New York, 1965. -. 148 -17. K. Bowden and R. Stewart, Tetrahedron, 21_, 261 (1965). 18. a) D.J. Kroeger and R. Stewart, Can. J . Chem., 4_5, 2163, 2173 (1967). b) D. Kroeger, Ph.D. the s i s , University of B r i t i s h Columbia, Vancouver, 1966. 19. CD. Ritchie and R.E. Uschold, a) J . Am. Chem. S o c , 89_, 1721,2752 (1967); b) J . Am. Chem. S o c , 90, 2821 (1968). 20. a) E.C. Steiner and J.M. G i l b e r t , J . Am. Chem. S o c , 87_, 382 (1965). b) E.C. Steiner and J.D. Starkey, J . Am. Chem. S o c , 89_, 2751 (1967) . ' 21. a) A. Streitwieser J r . , E. C i u f f a r i n , J.H. Hammons and J . I . Brauman, J. Am. Chem. S o c , 89, 59 (1967), b) A. Streitwieser, E. C i u f f a r i n and J.H. Hammons, J . Am. Chem. S o c , 89, 63 (1967). 22. Symposium on A c i d i t i e s of weak organic acids, 155th National meeting of the American Chemical Society, San Francisco, March 1968. 23. K. Bowden and A.F. C o c k e r i l l , Chem. Comm., 989 (1967). 24. R. Kuhn and D. Rewicki, a) Tetrahedron-Lett. 383 (1964); b) Tetra-hedron Lett., 3513 (1965). 25. R. Kuhn and D. Rewicki, Ann., 706, 250 (1967). 26. H. Rapoport and G. Smolinsky, J . Am. Chem. S o c , 80, 2910 (1958). . 27. H. Fischer and D. Rewicki, i n \"Progress i n Organic Chemistry\", v o l . 7, p. 5., Butterworths, London, 1968, 28. N.D. Coggeshall and A.S. Glessner J r . , J . Am. Chem. S o c , 7_1, 3.150 (1949) . 29. L.A. Cohen and W.M. Jones, J . Am. Chem. S o c , 85_, 3397, 3402 (1963). 30. A. Fischer, G.J. Leary, R.D. Topsom and J . Vaughan, a) J . Chem. Soc (B), 782 (1966); b) J . Chem. S o c , (B), 686, 846 (1967) . 31. C.H. Rochester, J . Chem. S o c , 676, 4603 (1965)-. - 149 -32. R.A. More O'Ferrall and J.H. Ridd, J . Chem. S o c , 5030 (1963). 33. R. Stewart, J.P. O'Donnell, D.J. Cram and B. Rickborn, Tetrahedron 18, 917 (1962) . 34. R.A. More O'Ferrall and J.H. Ridd, J . Chem. S o c , 5035 (1963). 35. J.R. Jones, Chem. Comm. 513 (1968). 36. J.R. Jones and R. Stewart, J . Chem. Soc (B), 1173 (1967). 37. J.R. Jones and R. Stewart, J . Am. Chem. S o c , 89, 5069 (1967). 38. M. Anbar, M. Bobtelsky, D. Samuel, B. Silverand and G. Y a g i l , J . Am. Chem. S o c , 85, 2380 (1963). 39. A.J. Kresge, R.A. More O'Ferrall, L.E. Hakka and V.P. V i t t u l l o , Chem. Comm., 46 (1965).-40. I. Andur and G.G. Hammes, \"Chemical Kinetics: p r i n c i p l e s and selected topics\", McGraw-Hill, New York, 1966, chapter s i x . 41. J.N. Bronsted and K. Pedersen, Z. Phys. Chem. (Leipzig), 108, 185 (1924) especially pp. 229-234. 42. R.P. B e l l , \"The proton i n Chemistry\", Cornell University Press, Ithaca, New York, 1959, chapter ten. 43. A.I. Shatenshtein, i n \"Advances i n Physical Organic Chemistry\", v o l . 1, p. 155, Academic Press, London 1963. 44. A.I. Shatenshtein and I.O. Shapiro, Russian Chemical Reviews, 37, 845 (1968). o r i g i n a l Usp. Khim, 37_, 1946 (1968). 45. A.I. Shatenshtein, I.O. Shapiro and.I.A. Romanskii, Dokl. Phys. Chem., 174, 447 (1967). o r i g i n a l Dokl. Akad. Nauk SSR., 174_ (5), 1138 (1967) . 46. A.I. Shatenshtein, Kinetics and Catalysis, £(5), 907 (1967). o r i g i n a l Kinet. Katal., 8(5), 1056 (1967). - 150 -47. A. Streitwieser J r . , W.C. Langworthy and D.E. Van S i c k l e , J . Am. Chem. S o c , 84, 251 (1962). 48. J.E. Hofmann, A. Schriesheim and R.E. Nickols, Tetrahedron Lett., 1745 (1965). 49. J.R. Jones, Chem. Comm., 710 (1967). 50. J . Bigelsein and M. Wolfsberg, i n \"Advances i n Chemical Physics\" v o l . 1, p. 15 Interscience, London 1958. 51. R.P. B e l l and D.M. Goodall, Proc. Roy. S o c , A, 294, 273 (1966). 52. J.L. Longridge and F.A. Long, J . Am. Chem. S o c , 89, 1292 (1967). 53. A.J. Kresge, D.S. Sagatys and H.L. Chen, J. Am. Chem. S o c , 90_, 4174 (1968) . 54. A.I. Vogel, \"A Textbook of P r a c t i c a l Organic Chemistry\", Longmans Green and Co. Ltd., London, 1964. 55. J.R. Jones, a) J . Label Compounds IV, 197 (1968); b) Internat. J. Appl. Radiation Isotopes, 1966. 56. F.G. Bordwell, W.J. Boyle, J r . , J . A. Hautala and K.C. Yee, J. Am. Chem. S o c , 91_, 4002 (1969). 57. M.P. J u i l l a r d , B u l l . Soc. Chim. France, [3], 33_, 1172 (1905). 58. F~.\" Reverdin and P. Crepieux, B u l l . Soc. Chim. France, [3] 29, 235 (1903). 59. E.J. Hoffmann and P.A. Dame, J. Am. Chem. S o c , 4J_, 1013 (1919) 60. B.B. Dey and Y.G. Doraiswami, J. Indian Chem. S o c , 1_0, 309 (1933) 61. CA. 35, 2868.3 (1941). 62. F. Ullmann and R. Dahmen, Ber. 41_, 3744 (1908). 63. Heilbron and H.M. Bunbury, Dictionary of Organic Compounds, Eyre and Spottiswoode, London 1946. - 151 -64, G.D. Parkes and R.H. Morley, J . Chem. S o c , 1478 (1936). 65. J.B. Shoesmith, C.E. Sooson and A.C. Hetherington, J . Chem. Soc. 2227 (1927). 66. W. Staedel, Ann. 153 (1894). 67. L. Gattermann and H. Rudt, Ber. , 27_, 2293 (1894). 68. A. K l i e g l , Ber., 38, 284 (1905). • 69. T. B i r c h a l l and W.L. J o l l y , J . Am. Chem. S o c , 88y 5439 (1966). 70. T. Cuvigny and H-. Normant, B u l l . Soc. Chim. France, 1872 (1965). 71. F. Ullmann and G. Engi, Ber., 37, 2367 (1904). 72. R.G. Bates, G.D. Pinching and E.R. Smith, J . Res. nat. Bur. Stand., 45, 418 (1950). 73. R. Foster and CA. Fyfe, Rev. Pure and Appl. Chem. 16, 61 (1966). 74. R. Schaal, J . Chim. Phys., 52, 796 (1955). 75. M.A. Paul and F.A. Long, Chem. Rev., 57, 1 (1957). 76. K.N. Bascombe and R.P. B e l l , J . Chem. S o c , 1096 (1959). 77. E. Hogfeldt and J.Bigeleisen, J . Am. Chem. S o c , 82_, 15 (1960). 78. M.J. Jorgenson and D.R. Hartter, J . Am. Chem. S o c , 85_, 878 (1963), 79. a) E.M. Arnett and C.F. Douty, J . Am. Chem. S o c , 86_, 409 (1964). b) E.M. Arnett and J.F. Burke, J . Am. Chem. S o c , 88, 4308 (1966). 80. R.W. Alder, G.R. Chalkey and M.C Whiting, Chem. Comm., 405 (1966). 81. E. Grunwald and E. Pr i c e , J . Am. Chem. S o c , 86, 4517 (1964). 82. J . J u i l l a r d , B u l l . Soc. Chim. France, 1727 (1966). 83. B.W. Clare, D. Cook, E.C.F. Ko, Y.C Mac and A.J. Parker, J . Am. . Chem. S o c , 88_, 1911 (1966). 84. A. Fischer, M.A. Riddolls and J . Vaughan, J . Chem.. Soc.(B), 106 (1966). - 152 -85. Handbook for the preparation of samples for l i q u i d s c i n t i l l a t i o n counting, Nuclear Chicago Corporation, I l l i n o i s . 86. W.T. Ford, E.W. Graham and D.J. Cram, J . Am. Chem. S o c , 89_, 689, 690 (1967) . 87. a) F. Hibbert, F.A. Long and E.A. Walters, J . Am. Chem. S o c , 91_, 2381 (1968) . b) E.A. Walters and F.A. Long, J . Am. Chem. S o c , 91_, 3733 (1969). 88. R. Alexander, E.C.F. Ko, A.J. Parker and T.J. Broxton, J . Am. Chem. S o c , 90_, 5049 (1968). 89. E.F. Caldin and J.C. T r i c k e t t , Trans. Farad. S o c , 49_, 772 (1953). 90. E. Buncel, A.R. Norris and K.E. Russell, Quart. Rev. London, 22, 123 (1968). 91. G.A. Russell and E.G. Janzen, a) J . Am. Chem. S o c , 84_, 4153 (1962). b) J . Am. Chem. S o c , 89_, 300 (1967). 92. M.R. Crampton and V. Gold, Proc. Chem. S o c , 298 (1964). 93. W.L. J o l l y , Inorg. Chem., 6, 1435 (1967). 94. a) J . Badoz-Lamb l i n g , M. Herlem and A. Thiebault, Anal, l e t t . , 2_ (1), 35 (1969) . b) M. Herlem, B u l l . Soc. Chim. France, 1687 (1967). 95. a) E.C. Fohn, R.E. Cuthrell and J . J . Lagowski, Inorg. Chem., 4, 1002 (1965). b) R.E. Cu t h r e l l , E.C. Fohn and J . J . Lagowski, Inorg. Chem., 5, 111 (1966) . 96. CA. Fyfe, personal communication. - 153 -97. L.H. Pi e t t e , P. Ludwig and R.N. Adams, a) J . Am. Chem. Soc., 83, 3909 (1962); b) J . Am. Chem. S o c , 84_, 4212 (1963).. 98. a) J.Q. Chambers, T. Lay1off and R.N. Adams, J . Phys. Chem., 68, 661 (1964) . b) P. Ludwig, T. Layloff and R.N. Adams, J . Am. Chem. S o c , 86, 4568 (1964). 99. J . Pannell, Mol. Phys., 7_, 317, 599 (1964). 100. J . Gendell, J.H. Freed and G.K. Fraenkel, J . Chem. Phys., 37, 2832 (1962). 101. D.EKennedy, Department of Chemistry, University of B r i t i s h Columbia seminars, (1967), and private communication. 102. J.Q. Chambers and R.N. Adams, Mol. Phys., 9, 413 (1965). 103. J . Brenet, J . Chim. Phys., 66, 1057 (1969). - 154 -APPENDIX A: Ionization Ratio Data TABLE XXIV Experimental values of log I for diphenylamine indicators in the system DMSO-water-tetramethylammonium hydroxide (0.011 M) <: < < a. a, < < a. a Q a, Q < < to to a Q CN a. a, p- ' r - l 0) oc Q Q u u S . to to -1 — • -1 < - i - i /—\\ /—\\ to to a, to 1^-CN CN 1 1 1 Q i i o o CN CN CN CN CN CN u /—% /—\\ r—s t—\\ /—\"\\ o CN CN CN CN CN CN •M 1 1 o O o O O o nj — z z 2 O to V / \\-—' v— J v—' v— j •H ** 1 1 i I I 1 -d ^ -C «\\ r - 1 CN CN CN CN CN CN CN CN Mole%. DMSO / p a 12.30 12.59 12.87 13.06 13.17 13.85 13.90 14.48 0.4 -.35 11.95 1.0 -.20 -.53 12.08 2.0 -.12 • -.38 .. 12.20 5.0 +.21 - . 1 0 . -.32 -.48 -.61 12.54 10.0 +.74 13.04 11.3 +.57 +.35 +.10. +.05 -.64 13.19 16.0 +.84 +.66 +.55 -.10 -.15 -.75 13.73 20.3 +.40 +.35 -.16 14.29 24.1 +.88 +.82 +.24 14.72 - 155 TABLE XXIV - Continued < < ^ < a, cv. <; t 2 IZ 2 . Z • Z . C •* i i i i i I—I to . • \"tf ™°in%/pK0 14.62 14.90 15.00 15.67 15.60 16.40 Assigned H_ 20.3 -.29 -.58 14.29 24.1 +.07 -.22 -.30 -.90 -.82 14.72 29.7 +.69 +.45 +.34 -.32 -.24 15.34 33.6 +.83 +.74 +.11 +.19 -.61 15.75 .36.2 1 +.34 . -.35 16.01 43.2 +.39 16.74 - 156 -TABLE XXV. Experimental values of log I for diphenylamine indicators i n the system DMSO-methanol-sodium methoxide (0.01 M) < < a. < < a, Q Q a, o, Q < < to to a Q CM Cl, a. a. tu, i-H I N to to a, •3-CM CM i 1 i Q 1 1 o O CM CM CM CM CM CM u Z, z , r—\\ —\\ ,—\\ /—\\ /—\\ o CM CM CM CM CM CM +-> I 1 O o . O O o o — — z z , • z z z , z , o to v / —J •H 1 1 1 1 1 1 TJ ^t- *t «\\ 1—1 CM CM CM CM CM CM CM CN DMSO % / p K a 1 2 , 2 5 1 2 , 6 5 1 2 , 8 5 1 3 , 0 3 1 3 , 2 0 1 4 , 0 5 1 4 , 1 0 1 4 , 9 5 A s s i g n e d H-0.3 -.30 11.95 1.0 -.23 12.02 2.0 -.16 -.54 12.10 5.0 +.10 -.30 -.64 -.84 12.31 10.6 +.50 +.12 -.22 -.40 -.60 12.68 15.0 +.51 +.22 -.02 -.18 13.08 19.2 +.56 +.39 +.24 -.73 -.78 13.42 25.0 +.100 +.80 +.62 -.28 -.36 13.81 29.7 +.12 +.04 14.20 33.0 -.52 14.43 35.1 +.58 +.46 14.59 40.0 +.87 -.08 15.00 46.0 +.51 15.46 49.0 +.75 15.70 - 157 -TABLE XXV - Continued < < O a, CN a x w I I Q I I O I I CM CN CN CN CN •H - - o • o o . o o TJ \"tf Z Z Z • Z 2 q * * i i i i i Mole% DMSO ' < < < a, a, ' to .4.42 15.26 15.69 15.85 16. 25.0 - -61 13.81 29.7 - . 2 2 14.20 35.1 +.14 - . 6 8 14.59 40.0 +.54 - . 2 9 - . 7 0 15.00 44 .9 +.91 +.07 - . 3 6 - . 4 9 15.34 50.0 +.41 +.06 - . 1 0 15.72 52.0 - . 5 7 54.9 +.93 +.45 +.29 - . 4 8 16.14 59.0 - . 0 5 60.2 +.93 +.71 - . 0 7 16.59 63.0 ' +.32 - . 5 7 65.5 +.39 16.98 69.5 +.76 +.81 - . 0 5 17.37 74.0 +.35 17.70 77.0 +.59 17.94 - 158 -TABLE XXVI Experimental values of log I for amine indicators i n the system DMSO-ethanol-sodium ethoxide (0.01 M) < < a. o o CM * Q 1 i CM CN CM - <: l—N /—\\ r—\\ /—s CN CM CM CM O O o o •z. Z z V * 1 1 i 1 CM CN CM CM 1^ o +J cd O •H X) R i—i !Ix°io%/pK 14.05 14.12 15.05 15.27 Assigned H DMSO r a -0.3 -.60 -.65 13.46 1.0 -.50 -.52 13.57 2.5 -.34 13.71 5.0 -.09 -.20 13.94 10.0 +.26 +.14 14.29 14.6 +.60 +.46 -.37 -.59 / 14.64 20.4 -.03 -.27 15.01 24.7 +.23 +.04 15.31 30.0 +.33 - 15.60 35.4 +.66 15.96 - 159 -TABLE XXVI - Continued < < a, a. a o to < < < a. CU a, Q G Q Q CN <; i—I CD X CN CJ 2 o -J < - i -1 2 to a, to *t i 1 a 1 1 CN CN CN CN CN 1 o o o o i—l 2 2 2 2 ] CJ | >* 5.73 16.63 16.55 17.45 18.0 u O CN +-> • o to rt 2 i CJ v-^ CN •H O t) ^ 2 G - ' n to \"sf ™ ™ % / p K Q 15.15 15.63 1 Assigned H UMbU • ** . \" 10.0 -.86 14.29 14.6 -.47 14.64 20.4 -.13 -.63 15.01 24.7 +.13 -.32 -.40 15.31 29.0 -.15 30.0 +.44 -.01 15.60 35.4 +.77 +.35 -.65 15.96 38.0 . +.45 16.18 39.7 +.66 -.35 16.28 45.0 +.01 +.08 16.63 50.6 +.33 -.49 16.96 52.2 +.52 17.07 55.2 +.66 +.80 -.16 -.81 17.29 60.0 +.19 -.41 17.64 65.0 +.55 +.01 18.00 70.0 +.95 +.38 18.40 75.9 +.80 18.81 - 160 -TABLE X X V I I Experimental values of log I for the substituted phenol indicators i n the system DMSO-methanol-sodium methoxide (0.01 M) rH >> cu rH o X u +J 4-> r-l • P< c o CD V) rH * l 1 o •H 4-> u I +-> •H •H •H •P o rQ -d VI U) 1 1 . i £1 •H \\D 3 1 n 1 co CN CN CN CN? DMSO ^ / p K a 1 0 , 4 9 1 0 , 9 8 1 1 , 6 7 1 2 , 1 0 Assigned H_ 0.0 -.60 9.92 5.0 . -.33 -.80 10.17 10.6 -.13 -.64 10.35 15.0 +.04 -.46 10.53 19.2 +.24 -.26 -.94 10.73 25.0 -.04 -.74 10.94 29.7 +.15 -.47 11.16 35.1 +.35 10.35 40.0 +.56 -.09 11.55 44.9 +.08 11.75 46.0 -.24 --50.0 +;30 12.97 53.0 +.03 • ' . r 12.13 59.0 +.26 12.36 63.0 +.44 12.54 69.7 +.69 12.79 - 161 -TABLE XXVIII Experimental values of log I for the substituted phenol indicators i n DMSO-ethanol-sodium ethoxide (0.01 M) -methyl Substituent 2,6-di-isopropyl rH X +J 3 £> i CN 2,4-di-t-butyl i i—i X 4-> 3 .£> M i •H 1 \\o CN rH X +-> 3 rd 1 •H u 4-> CN mole% . DMSO / p K a 11.00 11.67 12.12 12.28 12.28 Assigned 0.0 -.40 -1.03 10 .62 1.1 -.18 - .92 10 .78 5.0 + .02 - .66 11 .02 10.0 + .20 - .46 11 .21 14.8 + .40 - .26 11 .40 20.4 + .57 - .10 -.58 11 .56 24.7 + .72 + .07 11 .73 30.0 + .25 -.21 -.36 11 .91 35.4 - .02 -.20 12 .10 39.7 + .60 + .14 -.04 -.04 12 .25 45.0 + .35 + .17 + .21 12 .47 50.6 + .95 + .48 + .40 12 .63 55.2 + .63 + .46 + .56 12 .77 60.0 + .84 + .70 + .71 12 .98 65.0 + .80 13 .11 70.0 + .99 13 .31 - 1.62 -Appendix B: Rate correlation data TABLE XXIX Rate of d e t r i t i a t i o n (1. mole \"''sec '), ^ Q^ - at 25° for various hydrocarbons i n basic DMSO-Ethanol solutions Compound mole\" DMSO [OEt] M OR 9-Phenylfluorene 2,3-Benzofluorene Fluorene 9-Ethylfluorene 0.0 .01 13.46 7, .684 X 10 5.0 .0098 13.93 2, .327 X 10 10.0 .01 14.29 4, .600 X 10 14.6 .0105 14.66 8, .726 X 10 20.4 .01 15.01 1. .844 X 10' 0.0 .01 13.46 1. .535 X 10' 5.0 .0087 13.88 3 .516 X 10' 33.5 .0103 15.78 1, .250 X 10' 39.2 .0086 16.21 2, .523 X 10 10.0 .0102 14.30 2 .723 X 10 12.5 .0112 14.50 4 .186 X 10 20.2 .01 15.01 8 .746 X 10 24.4 .0095 15.26 1 .520 X 10 30.3 .01 15.60 2 .766 X 10 35.0 .0103 15.95 4 .808 X 10 39.2 .0086 16.21 7 .443 X 10 44.6 .01 16.63 2 .083 X 10 12.5 .0107 14.48 3 .241 X 10 29.9 .0091 15.56 2 .575 X 10 35.0 .0095 15.94 4 .218 X 10 39.2 .009 16.18 7 .563 X 10 44.5 .0091 16.55 1 .631 X 10 49.1 .01 16.88 2 .934 X 10 54.6 .0085 17.19 5 .540 X 10 59.6 .009 17.55 8 .884 X 10 64.9 .0092 17.92 1 .811 X 10 -3 -2 -2 -2 -1 -3 -3 -1 -1 -3 -3 -3 -2 -2 -2 -2 -1 -4 -3 -3 -3 -2 -2 -2 -2 -1 - 163 -TABLE XXIX - Continued N Compound mole% [OEt] H k -DMSO M 9-Phenylxanthene 49 .1 .0094 16. ,85 3, .514 X 10 64 .9 .01 17, ,96 2, .956 X 10 70 .3 .014 18, .42 6, .528 X 10 75 .0 .015 18, ,83 1, .645 X 10 .94 .0 .01 20, .63 5, .530 X 10 Triphenylmethane 83 .5 .01 19, .59 6, .448 X 10 87 .0 .01 19, .93 1 .287 X 10 20 .6 .01 20, .63 5, .990 X 10 164 -TABLE XXX Rates of d e t r i t i a t i o n (l.mole 1 sec ' ) , k - at 25° for various UK hydrocarbons i n basic DMSO solutions a) 9-phenylfluorene i n DMSO-methanol solutions mole % DMSO 0.0 5.3 9.7 15.7 19.4 28.8 35.7 40.1 [OMe] M .0093 .0095 .0104 .0108 .0105 .0108 .0105 .0099 H_ 11.91 12.30 12.68 13.18 13.42 14.15 14.63 14.98 1.525 x 10 4.849 x 10\" 1.035 x 10\" 2.25 x 10\"' 3.673 x 10\" 1.1 x 10\" 1 2.513 x 10 4.664 x 10' -1 b) fluorene i n DMSO-water solutions mole % DMSO 27.1 33.0 36.4 49.2 [TMAOH] M .011 .011 .011 .011 15.02 15.65 16.05 17.37 1.832 x 10 4.473 x 10\" 1.200 x 10 1.225 x 10 - 165 -TABLE XXXI Ionization rate constants (l.mole 1 sec \"*\"), k (DMSO), at 25° for DMSO in Basic DMSO solutions. mole % . [OR]\"\" d e t r i t i a t i o n t r i t r i a t i o n DMSO M H» k k. a) In water-DMSO solutions with TMAOH as base 0.0 .01 11.98 3.046 x 10\" 8 0.0 .05 12.71 7.235 x 10\" 8 0.0 .10 13.01 2.175 x 10\" 7 33.5 .01 t 15.9 9.63 x 10\"5 43.3 .01 16.84* 1.06 x 1 0 - 3 55.3 .01 t 17.88 5.23 x 10\"3 b) In Ethanol-DMSO solutions with NaOEt as base 0.0 • .01 13.46 5.077 x 10\" 7 34.6 .01 15.96+ 9.92 x 10\" 5 50.6 .011 17.00 1.00 x 10\" 3 50.6 .022 t 17.30 7.43 x 10\"4 65.5 .01 18.00 + 3.43 x 10\" 3 80.0 .098 19.21 4.57 x 10\" 2 80.4 .01 19.24 4.79 x 10\" 2 90.1 .01 20.05 3.61 x 1 0 _ 1 95.4 .098 20.62 2.88 .../continued - 166 -TABLE XXXI continued mole% DMSO [OR]\" M d e t r i t i a t i o n k t r i t i a t i o n k c) In Methanol-DMSO solutions with NaOMe as base 0.0 .01 11.95 9.14 x 10\" 8 30.0 .01 14.19 3.89 x 10\" 6 36.0 .0095 14.74 1.54 x 10\" 5 36.0 .01 14.76 9.55 x 10\" 6 50.1 .01 15.75 1.05 x 10\" 4 50.1 .013 15.85 1.07 x 10\" 4 60.1 .01 16.57 8.91 x 10\" 4 63.0 .0097 16.86 4.09 x 10\" 4 60.1 .024 16.95 4.26 x 10\" 4 74.0 .01 17.70 1.74 x 10\" 3 93.4 .01 19.50 3.15 x 10\" 2 94.6 .01 19.70 3.99 x 10\" 2 The H value of these solutions were determined spectrophotometrically, with amine indicators. - 1.67 -. TABLE XXXII Rate-temperature data for the base catalyzed d e t r i t i a t i o n of hydrocarbons i n DMSO-EtOH-0.01 M NaOEt Hydrocarbon Temp° 1/T°(K) kOR- ( mole sec ) loc ; k/T a 55 0. .003047 6 .630 X i o - 4 -5, .6946 triphenylmethane 45 0, .003143 2 .207 X ID\" 4 -6, .1589 35 0, .003245 6 .140 X i o \" 4 -6, .7009 45 0. .003143 2 .399 X i o \" 2 -4, .1227 9-phenylxanthene 35 0. .003245 9, .02 X IO\" 3 -4. .5337 25 0. ,003350 2 .955 X i o \" 3 -5, .0040 15 0. .003470 6 .692 X IO\" 4 -5, . 6332 25 0. .003354 1 .811 X i o \" 1 -3, .2165 9-ethylfluorene 15 0. .003470 6 .092 X i o \" 2 -3, .6747 5 0. .003595 2 .218 X i o \" 2 -4, .0984 fluorene . 15 0. .003470 4 .207. X i o \" 1 -2 .8358 5 0. .00359 1 .577 X i o \" 1 -3 . 2464 -5 0, .00373 5 .970 X i o \" 2 • -3 .6524 b 15 0, .003470. 1 .845 X IO\" 3 -5, .1937 9-phenylxanthene 25 0. .003350 6 .526 X i o \" 3 -4 .6603 70% DMSO 35 0, .003245 2 .019 X i o \" 2 -4, .1836 45 0. .003143 7 .995 X i o \" 2 -3 .5998 15 0. .003470 1 .444 X i o \" 3 -5 .2990 9-ethylfluorene 25 0. .003350 4 .218 X i o \" 3 -4 .8499 35% DMSO 35 0. .003245 1 .097 X i o \" 2 -4 .4486 45 0, .003143 3 .49 X i o \" 2 -3 .9597 15 0. .003470 1 .02 X i o \" 3 -5 .4510 fluorene 25 0, .003350 2 .776 X i o \" 3 -5, .0315 10% DMSO 35 0, .003245 7 .433 X i o \" 3 -4, .6176 45 0, .003143 1 .9 X i o \" 2 -4 .2239 - 168 -C C C C C\" c ALAIN /O.K./ A.M., S. I ll S PROORAM DOES A i s A s i C M i r i c y ^ i C L ; LEAST SqtJA^tS l : I T F OR TEST F iK A S1\" \\ I \",HT L TH E IN': T f IS A S lGN lP lCANC l : TI:ST FOR A STRAIGHT L IN!: CALL PLOTS DIMENSION LABELX(T 5 ) , L A B E L Y ( L5) DIMENSION XX (?) , YY ( 2 ) 0 I MEMS I_ON _X (_3_0_) »;Y (_30J 0 I ME N S I C N~ H TLJiG 12 0 ) DIMENSION TITLE (20) EXPRESSION Y = A + liX 31 READ(5,3,END-99) TITLEG 3 FORMAT ( 2 OA-'. ) WRITE(6, A) TITLEG 4 FOR,\", AT ( I H I » 30X » 20A 4) c o n Rt AD (5 , 1 ) M 00 12 READ!5,11) XU,XL,YU,YL 0013 11 FORMAT(4E 10,5) 0014 XMIN=XL .0015 YMIN=YL 00 I 6 DX='{ XU-XL ) / 8 . 0017 0Y={YU-YL )/6. 0018 READ (5, 6 )LX,LABELX • 0019 READ{5,6)LY,LATELY 0020 6 FORMAT( I 3, 15A4) 00 21 CALL AXIS (0.0,0 . 0 , L A \\'S E L X , -1. X , B ,0.0,x •11N,DX) C 0 2 2 . CALL A XI S ( 0. C , 0 , 0 , L A 2 ELY, LY,6. 9 0.0, Y •IN,I:Y) 00 2 3 CALL PLOT! 0. C O . 0, 3) 0024 CALL PLOT(0.0,6,3,2) C025 CALL PLOT ( >'• .5,6.3,1) C026 CALL PLOT(8.5 ,0.0,I) 0027 CALL PLOT (8.0,0,0, L) C028 . 0020 K = l , M 00 29 REAO(5,3)11TLH 00 30 WRITE(6,4 ) TITLE 003 L READ (5, I ) M . . . .' . CO 32 1 FORMAT( [ 2 ) 0033 RE AO(5,2,EN0^99) (X(I ) ,Y( I ) ,I=1 ,N) 0034 WRITE (5,2) ( X ( I) , Y ( 1 ) , I - 1 , N ) CC35 2 FORMAT(2E12.6) C036 CALL LEAST ( X , Y,N ,;\\A,BO) 00 37 X X.MAX--AM AX (X ,N) • 003 8 X X MIN- AM[N(X,M) CO 39 Y( N+l ) = AA+QB*XX.MIN 0040 Y(;'H-2 ) « A A + W1*.X.XM\"AX 00 41 X(N-M) =XXMIN C042 X (M<-2 ) = XX<••'AX 0043 DO 10 l = l , f l 004 4 X ( I ) = (X( I )-XMIN)/OX 0045 Y ( I ) = ( Y ( I )-YM l iD/OY CO 46 CALL S Y M i' 01. ( X( I ) , Y( I ) ,0 ,07 ,4, 0 .0 -1 ) 00 47 10 CON IINOE 004 n XX ( L ) - ( X ( N <- 1 )-XM IN )/0X C049 •XX( 2 ) =( X ( \\+2 )-XM[N)/D-X - 169 -C050 YY{1 ) = IY(N+\"l)-YMH)/l)Y C051 YY( J? ) = ( Y( + ? J-Y.M MJ/DY 0052 CALL LINE (XX,YY,?»+l) 00 53__ P = X( I )J .IS CC54 * ' 0 = Y(Tj-~. 15 0053 A.M-K 0056 CALL NUMBt:R(P»O,O.l't»AM,O.0,-l) C057 ( 20 Cl'W INUG C058 v CALL SYMBOL (0.5 ,9.4 r 0. l't ,'TI rLGG,0 .0, 80) CO 59 C A_LL_ JPL OT ( 1 1. 0. Q.C , - 3J 0060 GO rn 31 CC61 99 CALL PLOTND * 0062 STOP C063 END 0001 C002 0003 CQ. 04. CCO 5 0006 0007 0008 CC09 FUNCT ION DIMENSION B=X(l) DO 10 I-=2,N AMAX(X fN) X{30) IF C X( I.) . GT 1C CONTINUE AMAX-0 RETURN END B) B = X( I ) COOl CC02 C003 OOO't CC05 C006 0007 0008 0009 FUNCTION DIMENSION B = X( I ) DO 10 I=2,N AMlN(XtN) XC 30) IE ( X ( I ) .'L T . B ) 8=X(I) 10 CONTINUE AMIN=B RETURN END - 170 -SUBROUTINE I. C A S T { X F Y , N , A A , B R ) DI MENS 1CN X(30 ) ,Y(30) ,A(30) t B(30) ,C(30) VD(30) ,E(30) ,E(30) LEAST SQUARE FIT OF STRAIGHT LINE Y = A +• BX INJTALIZE s x = o . ' . \" . \" \" . : ~ \" * \" \" • ' • SY-0.0 Sxx=o„ SYY=0. CALCULATION OF SLOPE AND INTERCEPT DO 100 I - 1 ? N I T x ^ x T x T T T : ' : ; : ~ s x x = s x x + x ( I ) * x ( I ) ' • • SY=SY + Y ( I ) SXY=SXY \\- X { I ) *Y { I ) \" ' \" \" \" ' \" \" 7 ' ' SYY=SYY + Y{ T ) *Y'( I ) 1 0 0 - CONTINUE . ' • ' .\" \" •• — xT\\ = (T : ' ~ ~ BB=( XN*SXY-SX*SY)/(XN*SXX-SX*'SX) . : A A-{SY-BU*SX)/XN SDSQR=O.O ' \" \" • DO 1 0 1 1 = 1 p N f_U ) =_YJ I )-PO *X(I)-AA SDS\"0= S'OS'O'k + F ( 1 ) *FTTJ \"' : \" 1 0 1 CONTINUE : WRTTE ( 6 , 2 ) SDSQR , ' WRITE (6,3) A A , BB S I GMAS=(SD SQR/(XN-2 . 0 ) )/(XN*SXX-SX*Sx) SIG!'' AA = S0RT ( S I GMAS*S XX ) STT^Al'T^l iTrTTSl l^^rS-W) ~ ~ ~ ' : WRITE (6,6)SIGMAA,SIG.MAB C C CALCULATION OF COEFFICIFNT\"OF\"CORRELATION R C X3 = SX/XN Yli-SY/XN SC=o.o DO 1 0 2 I=1,N A ( I ) ~X { I )-XB 0 ( I ) = Y ( I-J-YB C( I ) = A( I )*B(I) . 102 •CONTINUE SE=SC/(XN-1.0 ) . SAXX=0.0 DO 1 0 3 I=1,N r_u17_A LL } ( J J S A X x = S A~>TxTDTn 103 CONTINUE SA xxp=SAXX/(XN-1 .0) C C r c c c c c - 171 -SBXX=0.0 DC lO'i 1 = 1 ,N E ( I ) = 3 ( I ) * B( I) SRxx=SBxx + F( i ) • 104 CONTINUE $Dxxp=SBxx/(XN-i.o} R=SF/SQRT ( SAXXP*SBXXP_) WRITE ( 6 , 4 ) R \" C C CALCULATION OF STUDENTS TEST T U = R*R T = AB S(R*SORT(XK-2.0)/SQRT(1. 0-U)) WRITE ( 6 , 5 ) T \"V\" * C . . C CALCULATION OF AVERAGE DEVIATION DELTA SD=0.0 -DO 105 I=1,N _ _ _. SD=SO + A B S ( F ( D ) \" 105 CUNT INUE ... DEL TA=SO/XN . ' WR ITF ( 6 , 1 9 ) DELTA • 2 FORMAT . (37H SUM OF THE SQUARES OF THE DEV I ATION=,E2 0.9/) 3 F 0 R M ATI I 2 H INTERCEPT = , 9X , E 1 5 .7 , 5X , 8H SLOPE = ,<3X,E15.7) 4 FORMAT(29H COEFFICIENT OF CORRELATION =,E20.9/) 5 FCRMAT (17H STUDENT TEST T =,F20.9/) 6 FORMAT(?1H ERROR IN INTERCEPT =,E15.7,5X t17H FRROR IN SLOPE =, 1 El'5.7) 19 FORMAT (7H DEL TA=,E20.9) -NR I T F ( * , 1 1 0 ) 110 FORMAT (62H0**** THIS SECTION ONLY RELEVANT FOR ACTIVATION STUD IE IS ****//) SACT= ( AA-10.3187) --1=4.574 HACT=-ftS*0.004574 ; WRITE(f-.,7)HACT, SACT 7 FORMAT( 31 HOT HE ENTHALPY OF ACTIVATION IS ,Gl0.3 , 32HKI LOGALORIES P IER MOLE PER 0GGPEF/3OH THE ENTROPY CF ACTIVATION IS , Gl0.3,13HENTR 20PY UNITS) DM ACT = SIGMAB»0. 004574 OSACT = SIG V>AA*4.574 WRI TE{6,76)DHACT,OSACT 76 FORMAT(1X,35H ERROR IN ENTHALPY OF ACTIVATION I S i E 1 0 . 5 , 2 1 UK ILCCALQ 1RIES PER M0LE/34H ERROR IN ENTROPY OF A C T I V U I O N IS,F 10 .5,13HFNTR0 2PY UNITS) RETURN * END Leaf 172 omitted i n page numbering. - 173 -SS1GN0N ALAN r-.i)?', P=20, PR10 = V t ; %'iiOH LSvj.*oPLof 9 = - P L O I' R A T E O F I O N I Z A T I O N O F D M S O - T VS H . ' M I N U S I N - METHANOL IC DMSO __2_ __. _• _. 20.0 1.2.0 \" -1 .00 -7.00\"\" \" 7H MINUS 22L0G__i000_0_ 15 . 8 4 9 9 9 9 16.5 6 9 9 9 2 1 6 . 8 5 9 9 8 5 1 6 . 9 4 9 9 9 7 1 7 . 6 9 9 9 9 7 19.5 0 0 0 0 0 19 . 6 9 9 9 9 7 SUM OF THF. - 7 . 0 3 9 1 0 0 - 5 . 4 1 0 0 0 0 - 4 . 8 1 3 0 0 0 - 5 . 0 2 0 0 0 0 jZ?>. 9 J 8 9 9 9 _ - 3 . 9 7 1 0 0 0 - 3 . 0 4 9 9 9 9 - 3 . 3 0 0 0 0 0 - 3 . 3 7 0 9 9 9 - 2 . 7 5 9 0 0 0 - 1 . 5 0 2 0 0 0 -1 . 4 0 0 0 0 0 S QUARES OF T H E METHANOL D E V I A T I O N ^ I N T E R C E P T = - 0 . 1 5 6 3 9 0 IE-ERROR IN I N T E R C E P T = 0 . 4 4 0 0 7 I 4 E C O E F F I C I E N T OF C O R R E L A T I O N = 0 . 3 8 5 2 7 8 5 2 3 E 0 0 02 S L O P E = 0 0 ERROR IN 0 . 9 9 3 2 6 4 1 9 8 E 0 0 S L O P E 0.7 29 32 8 0 E 0 0 0 . 2 6 9 2 2 1 7 G - 0 1 S T U D E N T T E S T T 0 . 2 7 L 0 7 3 6 0 8 E 0 2 D E L T A = 0 . 1 3 8 2 6 9 9 6 I E 0 0 * * * * T H I S S E C T I O N ONLY R E L E V A N T FOR A C T I V A T I O N S T U D I E S * * * * T H E E N T H A L P Y OF A C T I V A T I O N I S - 0 « 3 3 4 E - 0 2 K I L O C A L O R I E S PER MOLE P E R D E G R E E THE ENTROPY OF A C T I V A T I O N I S - 1 1 9 . ENTROPY U N I T S ERROR I_N E N T H A L P Y OF ACT I VAT ION IS 0 . 0 0 0 1 2K I L PC A L O R I E S PER MOLF E R R 0 R IN E N T ROPY OF ACT I VAT 1 O'N I S 2 . 0 1 2 8 9 E N T RO P Y U M T S - 175 -RATE OF IONIZATION OF DMSO-T VS H7MINUSQ LOG A 11.950000 -7.039100 13.980000 -5.410000 1.4.480000 -4.813000 14.500000 -5.020000 15.290000 -3. 9789 99 15.389999 -3.971000 15.959999 -3.049999 16.159908 -3.388000 16.339996 -3.370999 16.819992 -2.759000 _ 18. 32 9 9 87 U 50 2 000 > 18.4 799 9 h -1 .4 00 000 SUM OF THE SQUARES CF THE DFVI AT I ON- 0.358504772E 00 INTERCEPT = -0.1754129E 02 SLOPE = 0.8780563E 00 \"ERROR IN INTERCEPT = 0.49I4843E 00 ERROR IN SLOPE = 0.312299 LE-01 COEFFICIENT OF CORRELATION = 0.993734181E 00 STUDENT TEST T .•= 0. 281155701E 02 DELT A = 0.U0043244E 00 THIS SECTION ONLY RELEVANT FOR ACTIVATION STUDIES ***» THE ENTHALPY OF ACTIVATION IS -0.402E-C2KILOCALCRIES PER MOLE PER DEGREE THE ENTROPY OF ACTIVATION IS -127. ENTROPY UNITS ERROR IN EN THALPY OF _ AC J_l V AT I ON IS 0.00014K I L PC AL OR I E S PER HOLE ERROR IN f.NTROPY OF ACT IV AT I ON IS 2 . 24 805 ENTROPY UNITS\" PLOTTING KILL TAKE APPROX. 2 MIN 56 SEC SUCCESSFULL PLOT. STOP 0 EXECUTION TERMINATED SRUN *PLOTQ PAR=-PLOT EXECUTION BFGINS WHAT TYPE OF PAPER DO YCU WANT? REPLY \"BLANK\" OR \"LINED\" PLOT FILE IS OK \"AND~ IS IN SYSTEM EXECUTICN TERMINATED $SIGNOFF "@en ; edm:hasType "Thesis/Dissertation"@en ; edm:isShownAt "10.14288/1.0059885"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Chemistry"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en ; ns0:scholarLevel "Graduate"@en ; dcterms:title "Strongly basic systems"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/35382"@en .