TEMPERATURE DEPENDENCE OF SOME SECONDARY KINETIC DEUTERIUM ISDTDPE EFFECTS Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY UAN GEN KANG 1972 This is to certify that the thesis entitled TWP/“413*” Q‘s—(Wane of“ A’Dm Seaman/w} Kim’s/ac A/LWWN lawn-r4 £1193; presented by Dow (few K7): N‘éf has been accepted towards fulfillment of the requirements for Jh D dggreein Che/1M w} is ‘/ [[(UKKLRE I Majorptofessor - l A Date / ' ; / 0-7639 ’ nuox smnm mc. ‘ LIBRARY BINDERS gnlusropi. HM“ ABSTRACT TEMPERATURE DEPENDENCE OF SOME SECONDARY KINETIC DEUTERIUM ISOTOPE EFFECTS By Uan Gen Kang In an effort to understand the origin of isotope effects, Karabatsos and his coworkers1 measured several isotope effects as functions of temperature. They found that both AAH* and AAS* contribute to the isotope effects. They usually have the same sign, which results in smaller isotope effects due to cancellation. Their preliminary results indicated that in correlating isotOpe effects and reaction mechanisms the AAHt term is more significant than the AAG* term. Furthermore, within the limits of accuracy imposed by the assumptions of Bartell's procedure,2 a comparison of calculated isotOpe effects (AAEL due only to nonbonded interactions,with experimental values of AAHT indicates that in ordinary systems with hyperconjugation possible, hyperconjugation is the dominant contributor to AAH*. In the nucleophilic hydrolyses of acid chlorides 1b and esters, the isotope effects due to AAG* are nearly unity. However, in the activation process the force constants Uan Gen Kang should increase, because the demand for hyperconjugation decreases and nonbonded interactions increase. Therefore, an inverse isotope effect (RH/RD < l) is expected. Indeed, the AAHt term is reasonably large and positive leading to an inverse isotOpe effect. The experimental isotOpe effect is only near unity because of cancelling contributions to AAG* from AAST and AAHT. From the results of the present study, the negative AAHT value (-123 cal/mole in 65% acetone-water and -114 t 6 cal/mole in 70% acetone-water) is reasonable in view of the near limiting character of the solvolysis of the 8-methyl—l- naphthoyl chloride in 65% and 70% acetone-water. The much less negative values in 75%, 80% and 85% acetone—water (-23 i 14, -27 i 58 and -10 i 18 cal/mole respectively) are consistent with the borderline character of acid chloride solvolysis and indicate less limiting character in these solvents. However, the isotOpe effects due to AAG* (RH/RD = 1.014 t .003 in 85% A/W -20.60°; and 1.023 t .006 in 70% A/W, -20.60°) show their insensitivity to changes in reaction mechanism. They result from the cancelling contributions of H- the TAAS* and AAH"I terms (AAH* = -114 6 cal/mole and AAST = -0.4 i .0 e.u. in 70% A/W; AAHT = -10 i 18 cal/mole, AAST = 0.0 i .0 e.u. in 85% A/W) to AAG*. 2. Uan Gen Kang References (a) [i] G. J. Karabatsos, G. C. Sonnichsen, C. G. Papaioannou, S. B. Scheppele and R. L. Shone, J. Amer. Chem. Soc., fig, 463 (1967). [ii] G. J. Karabatsos and C. G. Papaioannou, Tetrahedron Lett., 2;” 2629 (1968). (b) G. J. Karabatsos and coworkers; unpublished results. [i] T. A. Evans (Ph.D. Thesis, 1968); [ii] U. G. Kang (M.S. Thesis, 1969); [iii] A. A. Effio Leon (Ph.D. Thesis, 1971); [iv] W. C. Sass (Ph.D. Thesis, 1970); [v] V. F. Smith, Jr. (Ph.D. Thesis, 1972); [vi] U. G. Kang (Ph.D. Thesis, 1972 present work). L. S. Bartell, J. Amer. Chem. Soc., 82) 3567 (1961). TEMPERATURE DEPENDENCE OF SOME SECONDARY KINETIC DEUTERIUM ISOTOPE EFFECTS By Uan Gen Kang A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1972 G’MEI To Jung ii ACKNOWLEDGMENTS The author wishes to express his sincere appreciation to Professor Gerasimos J. Karabatsos for his guidance and encouragement throughout this research. The financial support provided by the National Science Foundation is gratefully acknowledged. iii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . . LIST OF FIGURES. . . . . . . . . INTRODUCTION . . . . . . . . EXPERIMENTAL . . . . . . . . . . . . . . . KINETICS. . . . . . . . . . . . . . . . . Preparation of Solvents. . . . . . . . . Conductivity Water. Conductivity Acetone. Mixed Solvents. . . . . . . . Conductance Apparatus. . . . . . . . . . . Conductance Cell Measurement of Time. . . . . . . . . Constant Temperature Bath. Measurement of Temperature Rate Determination Treatment of Data. . SYNTHESIS . . . . . . . . . . . . . . . Preparation of Anhydro-8—hydroxymercuri- l-naphthoic Acid . . Preparation of 8-Bromo-1-naphthoic Acid. Preparation of 8—Bromo-l-naphthoyl Chloride. iv Page vi viii 12 12 12 12 12 12 13 13 13 14 14 15 16 24 25 25 26 RESULTS AND DISCUSSION MECHANISM OF ACID CHLORIDE SOLVOLYSIS THE ORIGIN OF THE ISOTOPE EFFECTS CONCLUSION REFERENCES APPENDICES TABLE OF CONTENTS Preparation of 8—Bromo-l-naphthylcarbinol a,a-d2 Preparation of 8-Bromo-1-bromomethyl- naphthalene-01,01-d2 Preparation of 1- -Methyl- d3-8- -bromo- naphthalene. Preparation of 8-Methyl-ds-l-naphthoic Acid. Preparation of 8- -Methyl- -d3 -1- -naphthoy1 Chloride Analytical Instruments The Effect of Solvent Changes on Mechanism . Correlation of Activation Parameters and Reaction Mechanism . IsotOpe Effects as Criteria of Mechanism . Relative Importance of Enthalpy (AAH*) and EntrOpy (AAS*) Term to IsotOpe Effects Relative Contribution of Hyperconjugation and Nonbonded Interactions to the Enthalpy Term (AAH* ). Page 27 27 28 29 31 32 33 33 33 36 4O 41 44 50 S3 60 TABLE 10 11 12 13 LIST OF TABLES Isotope Effects for l-Phenylethyl Chlorides Corrected to 50% Ethanol at 25°. . . . . . Deuterium IsotOpe Effects in Acidity . Solvolysis of ISOpropyl- B- '96 Derivatives in Water . . . . . . . . . . . . Binary Solvents. . . . . . . . . . . . . Rates and Isotope Effects of the Solvolysis of 8-Methyl- -d - and -d3 -1- -naphthoyl Chloride in 6g% Acetone- water (V/V). Rates and Isotope Effects of the Solvolysis of 8—Methyl-d - and -d -1-naphthoyl Chloride in 78% Acetong-water (V/V). Rates and Isotope Effects of the Solvolysis of 8-Methy1-d - and -d3-l-naphthoy1 Chloride in 79% Acetone- water (V/V). Rates and Isotope Effects of the Solvolysis of 8-Methy1-d - and -d3-1-naphthoyl Chloride in 88% Acetone-water (V/V). . Rates and Isotope Effects of the Solvolysis of 8- -Methyl- d - and -d -1-naphthoy1 Chloride in 89% Acetong- water (V/V). Melting Points of X-l—Naphthoic Acids. . . Entropies of Activation for Limiting and Nucleophilic Solvolyses in Aqueous-acetone at 50° . . . . . . . . . . . . . . . . EntrOpies of Activation (ASI) in e.u. for Some Acid Chlorides and Esters . . . . . . Activation Parameters for the Hydrolysis of 8-Methyl-gfl- and -d3-naphthoyl Chlorides. vi Page 13 17 18 19 20 21 30 37 38 39 TABLE 14 15 16 17 18 19 20 LIST OF TABLES a-Deuterium Effects (per D) on Rates of Solvolysis of Some Benzyl Derivatives at 25°. 0 I O O O O O O O O O O O O Hydrolysis of Compounds with Hyper- conjugation, Nonbonded Interactions and Inductive Effects Possible . . . . . . Hydrolysis of Compounds in Which Nonbonded Interactions are the Predominant Contributor to AAH* Term and Electronic Effects are Negligible . . . . . . . . . Temperature and Solvent Dependence of the IsotOpe Effect in the Hydrolysis of 8-Methy1-d0- and -d3-1-naphthoy1 Chlorides Activation Parameters Determined from the Temperature Dependence of the IsotOpe Effect of 8-Methy1-d0- and -d3-1- naphthoyl Chloride HydrolysiE. Calculated IsotOpe Effects Due to Nonbonded Interactions for 8-Methy1-1- naphthoyl Chloride and 8-Methy1-d3 Analog. Comparison of Calculated (AAE) and Experimental Isotope Effects (AAH‘) for Some Limiting Solvolyses . . . . . . . vii Page 42 45 47 48 49 51 52 LIST OF FIGURES FIGURE Page 1 Synthetic scheme . . . . . . . . . . . . . . 24 2 Plots of log (k/ko) against 0 for hydrolysis of substituted benzoyl chlorides in various solvent mixtures. . . . 34 viii INTRODUCTION Secondary kinetic isotOpe effects have been explained in terms of hyperconjugation, nonbonded interactions and inductive effects.1 Solvolysis studies of compounds with remote hyper- conjugating methyl-d3 groups support the importance of 2 hyperconjugation to the secondary isotope effects. For example, if the isotope effect caused by deuterium substitution at such remote positions is due to hyper- conjugation, then a decrease in the demand for hyper- C1 kH/kD = 1.10 (HOAc, 50°)3 2 CD3-CEC-E(CH3) c1 kH/kD = (80% EtOH, 25°) 1.092 5 C1 kH/kD = 1.058 (80% Acetone, 25°)4 CD3-CEC-C|IH-CH3 OBs CD3\\E==é//H H//’ \\\CH-CHI kH/kD = 1.109 (60% EtOH, 25°) 3 6 kH/kD 1.132 (95% EtOH, 25°)7 conjugation should decrease the B-secondary isotOpe effects. The results of Shiner and his coworkers8 confirm this prediction (Table 1). All a-isotope effects for the solvolysis of a series of substituted l-phenylethyl halides are about 1.154 1 .003, except those of the m-bromo (kH/kD = 1.133) and p-nitro (kH/kD = 1.098) compounds. The large a-isotope effects (kH/kD = 1.154) support the conclusion that the mechanism of the reaction is mainly limiting.9 The smaller a-isotope effects for the m-bromo and p-nitro compounds indicate some nucleophilic participation in the transition state. The observed 8- isotope effects show that the reduction of the demand for hyperconjugation, whether due to electron releasing para; substituents(p-methoxy and p-phenoxy), or to nucleOphilic participation in the transition state (m-bromo and p-nitro), decreases the B-isotOpe effect. Nuclear quadrupole coupling constant,10 nuclear magnetic 11 and dipole moment studies12 have resonance chemical shift, established that deuterium is more electropositive than hydrogen. The effect of deuterium substitution on acid strength (Table 2) has been interpreted in terms of the differences in the inductive effects of deuterium and hydrogen. In all cases deuteration decreases the acid strength. On the other hand, as expected, deuteration 13 increases the basicity of benzyl-u,a-d2-amine (KH/KD = 0.88. Table l. Isotope Effects for l-Phenylethy Corrected to 50% Ethanol at 25°. é Chlorides b- —.__. “.3312: ‘5‘ kH/kD (a) kH/kD (B) p-methoxy 1.157 1.113 p-phenoxy 1.157 1.164 p-methyl 1.157 1.200 p-fluoro 1.152 1.211 m-methyl 1.151 1.222 none 1.153 1.224 m-bromo 1.133 1.221 p-nitro 1.098 1.151 Table 2. Deuterium IsotOpe Effects in Acidity. ———-..— o —-.——4- -7-“ =-._..._ ._.__-__. .. : 6 S 2 Acid .'KH/KD Reference DCOOH 1.035 1 .002 14 CD3COOH 1.032 1 .002 15 (CD3)3COOH 1.042 1 .003 15 C6D5COOH 1.024 1 .006 15 c H CD coon 1.12 1 .02 13 Solvolysis of the following compounds shows that kinetic isotope effects due to the inductive effect are small and opposite in direction from those due to hyperconjugation and nonbonded interactions. D HCH3 (.311 C1 CD3 1 1) kH/kD = 0.976 kH/kD = 0.981 per D (80% Acetone, 62°)3 (80% Acetone, 25°)15 = + kH/kD 0 986 - .001 (60% EtOH, 45°)16 Bartell17 has explained secondary isotope effects in terms of nonbonded interactions. His argument is as follows: The vibrational amplitudes of the C-H bond are larger than those of the C-D bond. Therefore, in the activation process from a crowded ground state to a less crowded transition state, there is greater relief of nonbonded interactions for the unlabeled than for the deuterium labeled compound. Consequently, a normal isotope effect should be observed. An inverse isotope effect should be observed in cases where the transition state is more crowded than the ground state. The following conformational studies have shown that the effective steric requirement for the C-D bond is less than that for the C-H bond. Since the racemization (a, b, and c) and ring flipping (d) are purely conformational changes, it is reasonable to explain the observed isotOpe effects in terms of nonbonded interactions. CBS CBS kH/kD = 0.83 kH/kD = 0.85 (benzene, 42°)18 (EtOH, '20°)19 c d D ©© .. = + kH/kD - 0.83 to 0.88 kH/kD 0.83 - .004 (HCON(CH 20° and 65°)20 (CDCl 35°)21 3)3’ 3’ 1’22 studied the reactions of Brown and his coworkers methyl-dS-pyridines with Lewis acids and alkyl iodides. They found small or no isotope effects with the meta and para methyl-dS—substituted pyridines, but inverse isotope effects with the orthgfsubstituted ones. The inverse isotOpe effects increased as the steric requirement of the alkyl iodide and the Lewis acid became larger. For example, the isotope effects in the reaction of 2-methy1-d3-pyridine are kH/kD = 0.97 with methyl iodide and kH/kD = 0.935 with isopropyl iodide. For 2,6-dimethy1-d6-pyridine a kH/kD of 0.92 was observed with boron trifluoride. On the other hand, the much smaller boron hydride (BH3) gave no isotope effect. These results are consistent with the view that the secondary deuterium isotope effects in these cases are caused mainly by nonbonded interactions. There seems to be little question that hyperconjugation, nonbonded interactions and inductive effects all contribute to secondary deuterium isotOpe effects. In general, if effective hyperconjugation is possible it dominates the inductive effect. However, the relative contributions of hyperconjugation and nonbonded interactions are not well understood.1 Although secondary isotOpe effects are interpreted in terms of hyperconjugation, nonbonded interactions and inductive effects, molecular force constant changes56 on activation are generally accepted as the origin of isotope effects. Wolfsberg and Stern23 have calculated isotope effects from molecular geometries, atomic masses, and assigned force constants for initial and transition states. In general, lower force constants in the transition state than in the reactant cause normal isotope effects, kH/kD > 1, while higher force constants in the transition state cause inverse isotope effects, kH/kD < 1. If force constant changes determine secondary isotOpe effects, the temperature dependence of these effects should depend mainly on the differences in the corresponding enthalpies of activation (AAHI), and not on differences in the entrepies of activation (AASI). 1n(kH/k -AAH*/RT + AAST/R D) 1n(kH/k -Ea/RT + In A H/A D) D 24 Hakka and coworkers have shown that in the hydrolysis of tfbutyl chloride and its d9 analog AAH* = ~520 cal/mole a AAG* with no apparent contribution from AAST (= -0.2 e.u.). These results support the idea that force constant changes are the origin of the isotope effects. 4 However, Shiner and Verbanic's results on the solvolysis of p-methyl-d3-benzhydryl chlorides (A Ea = ~106 1 20 cal/mole per C-D bond, lnA /A -0.075 t 0.0015 per C-D bond), Shiner's H D = studies25 on the solvolysis of 2,3-dimethyl-2-chlorobutane-3-d (A Ea = -580 i 70 cal/mole, lnA A = -0.32 i 0.03) and Lewis H/ D and Coppinger's results3 on the solvolysis of p—methyl-d31a- phenylethyl chloride (A Ea = -lS4.S cal/mole per D, lnAH/AD = -0.084 per D) show that the isotope effects depend on both AAHI and AAST. Shiner and Hartshoron's studies26 on B-isotope effects of phenylethyl-d3 chloride (AAG* = -120 cal/mole, A E3 = ~77.2 cal/mole and AH/AD = 1.075) and p- methyl-phenylethyl-d3 chloride (AAGT = ~108.0 cal/mole, A E3 = -39.3 cal/mole and AH/AD = 1.123) show clearly the dependence of isotope effects on both AAH* and AAST. Mislow and coworkers' study of the racemization of 9,10-dihydro-4,5-bi§_(methyl-d3) phenanthrene18 gave similar results (AAH* = 240 cal/mole and AAS* = 0.53 e.u.) Carter and Dahlgren20 also found, for the racemization of 1,1'-binaphthy1-2,2'-d2, AAH = 270 t 140 cal/mole and AAS* = 0.54 t 0.43 e.u. (errors are three times the standard deviation). The results obtained by Leffek and coworkers on the 27 are the most solvolysis of iSOproyl-B-d6 derivatives unusual. The AAH* was approximately zero and the temperature independent isotope effect was due entirely to AAS* (Table 3). Five explanations have been given for the temperature independent isotope effect of this system. 1) Leffek and coworkers27 have suggested that the observed isotope effect is due to the rotational barrier difference in the ground state between methyl and methyl—d3, which results from the larger steric requirement of the protium compound. In the transition state the barrier to internal rotation is reduced considerably and leads to favorable acceleration of the protium analog. The negligible contribution to the isotOpe effect from the AAH* was explained by a cancellation of the effect from the rotational barrier difference and the one mo. + m6. mH + mm 1AA flashyemm.fi mooevnflm.fi on 6m-6w-6um-fi me. + mm. «H + HN mmm fioomvmem.a hoov~em.fi cmm mzo-6m-oag-fi H. + em. wN + A- com floomVAem.H momvmmm.fl on meo-6m-oam-d .=.o oHoE\Hmu oHoE\Hmu a z a : 4m<< 4m<< nom~veo<< Achy x\ x moev x\ x peo>Hom unconsou hope: :fl mm>flum>wwom om-m-meoumomH mo mflmxao>aom .m oHan 5N 10 1b has from the zero-point energy difference. 2) Halevi explained the data in terms of differences in the solvation of the protium compound and its deuterated analog. He assumed that charge dispersal is more effective in the deuterium compound, thus reducing the degree of solvation of the deuterio compound with reSpect to that of the protio compound. Consequently, the more effective solvation of the protio compound reduces its AH* with respect to that of the deuterio analog. However, this solvation energy gain for the protio compound is counterbalanced by the energy loss due to breaking the hydrogen bonds among the water molecules. Thus, the net effect is an entropy gain due to the breaking of the quasicrystalline water structure. 3) Wolfsberg and Stern23 have shown that it is possible to obtain reasonably large temperature-independent isotope effects over the usual temperature range by compensating changes in various force constants. Thus, if the C-D stretching force constant of the transition state was reduced relative to the reactant from 4.8 to 3.5 mdyn/A and the torsion and H-C-C bending force constants were increased from 0.15 to 1.0 mdyn/A and from 0.68 to 1.0 mdyn/A, respectively, then the calculated isotope effects were reasonably constant at 1.41 t 0.01 from 250° to 380° K. Shiner (2.p150) has suggested two possible explanations for this unusual temperature independent isotOpe effect: 11 4) The force constants of the transition state might be affected by solvation which is temperature dependent. Thus, if a reasonably large isotope effect was caused by lowering of a force constant in the transition state and if solvation were to increase this force constant, a smaller isotope effect would be obtained as a result of solvation. An increase in temperature would tend to lessen the solvation and, therefore, increase the isotOpe effect. This increase might be sufficient to cancel the usual decrease in isotope effect caused by the temperature increase. 5) With reservations due to inconsistancies with the a-isotope effect, the temperature independence of the isotope effect could be explained in terms of changes in mechanism, 145,, on the basis that the solvolysis of the ifpropyl compounds is borderline in character. Thus, as the temperature is raised, a more dissociative transition state could cause an increase in the B-effect which may balance the normal decrease eXpected from the zero-point energy term. All the data cited above show clearly the important influence that AAH* and AASt might have on secondary isotope effects. In order to understand further the origin of secondary isotOpe effects, we undertook to study the temperature and solvent dependence of isotope effects in the hydrolysis of 8-methy1-1-naphthoy1 chloride and its 8—methy1-d3 analog. EXPERIMENTAL KINETICS Preparation of Solvents Conductivity Water. -- Conductivity water was prepared by passing distilled water through a 5 x 80 cm column containing alternate layers of Dowex l-X8 (anion exchange resin) and Dowex SOW—X8 (cation exchange resin). Water treated in this manner had a specific conductance of less 6 than 2 x 10' mho/cm. Conductivity Acetone. -- Three liters of acetone (Baker Analyzed Reagent) was refluxed with 80 g of potassium permanganate and 10 pellets of sodium hydroxide for three hours and distilled. The acetone obtained in this manner had a specific conductance of less than 1 x 10'8 mho/cm. Mixed Solvents. -- Mixtures of acetone and water were prepared by weighing the water and acetone for the desired volume/volume ratios. The densities and ratios used are given in Table 4. A Torbal balance (model PL-12, Torsion Balance Co.) with a capacity of 2000 g and a stated accuracy of 0.1 g was used. 12 13 Table 4. Binary Solvents === an: :=-;__---.= T7 ““““““ V/V (%A/W) 85 80 75 70 65 A/w (%A/W) 81.67 75.89 70.24 64.74 59.37 Density of water at 25° = 0.997044. Density of acetone at 25° = 0.7844. Conductance Apparatus The conductance was measured by means of a Wayne-Kerr conductance bridge, accurate to i0.1% (model B 221, Wayne-Kerr Co., Ltd.), equipped with a Wayne-Kerr Autobalance Adaptor (model AA 221) with a P8109 power supply unit. Conductance Cell The two conductance cells used are essentially the same 34 The cells were stored with as described by Papaioannou. used reaction solvent at least two hours before use to avoid adsorbtion of ions during a kinetic run. They were rinsed three times with conductance water and twice with conductance acetone before use. Measurement of Time A precision Scientific Electronic Digital Timer (accurate to 1/100th of a minute) was used. 14 Constant Temperature Bath A geared-drive stirrer (model 382-66 Lapine Co.) with two small eleven-blade prepellers and a large eight-blade prepeller was placed in the upper center of the bath. The heating knife (125W, Central Scientific Co.) was placed at a distance of 9 cm from the stirrer. The precision thermo- regulator (micro-set Lapine Co.) and the Beckmann differential thermometer (Sargent 8 Co.) were placed side by side 4 cm from the heater. The conductivity cell on the submersible magnetic stirrer (Troemner Co.) was placed at a distance of 12 cm from the thermometer. The stirrer and the electronic relay (Precision Scientific Co.) were supported by plexiglas covering the bath. The bath was surrounded with styrofoam so as to be insulated from room temperature changes. After immersing the cell, the inside of the bath was insulated from the outside with a styrofoam cover which fitted the cell very well. The temperature control with this bath was better than :0.003° over a temperature range from -20 to +40°. Measurement of Temperature The temperature was determined by means of a quartz thermometer (2801A Hewlett-Packard) accurate to 10.01° and a platinum resistance thermometer (Portametric PVB 300 Electro Scientific Industries). The temperature was measured more than three times at each temperature. The quartz- and platinum-thermometer probes were attached 15 side-by-side in the ice point measurement and in the temperature bath to the Beckmann-thermometer. From each thermometer,ten readings were taken over a one hour period and the average temperature was corrected for linearity and for the deviation of the ice point for the quartz thermometer. The average resistance reading obtained was converted to temperature using the Werner and Frazer method.28 Results from the quartz and platinum thermometers agreed to better than 10.005°. During each kinetic run the temperature was monitored regularly by the quartz or the Beckmann thermometer and adjusted if necessary. Rate Determination After the conductance cell was filled with 260 m1 of solvent, it was immersed in the temperature bath and allowed to completely equilibrate for about 30 minutes. The solvent conductance, which was generally less than 8 mho, was recorded. About 20 mg (2.6 x 10-4 molar) 5 x 10' of acid chloride was dissolved in precisely one ml of conductivity acetone and injected into the cell with a one m1 syringe. At the same time, the timer was started and the solution was stirred for at least two minutes. Conductance readings, four per minute for fast reactions and usually two per minute and one per two minutes for slow reactions, were taken over the first three half-lives. The infinite value was obtained after 13 half-lives and exactly at the same time for all deuterated and undeuterated compounds. 16 Treatment of Data The first order rate constants were determined by a least squares solution of the integrated first order expression, using the program RATE (Tables 5-9). In (Cm-Ct) = -kt + lnC0° COD = conductance at thirteen half-lives Ct = conductance at time t. Conductance was taken as directly proportional to the concentration of hydrochloric acid formed. However, slight temperature independent deviations were observed in all solvents. Since we are mainly interested in the relative rate constants between deuterated and undeuterated acid chlorides under the same conditions, the observed slight deviations would not lead to appreciable error. The reported average rates are the mean average of the independent determinations. The uncertainty indicated is the standard error. The average rates and standard errors were calculated using program DEV written by the author (Appendix One). N - _ Z _- 2 a O - (l/(N 1) i=1 (x1 X) ) xi = observed rate R = mean rate. 17 Table 5. Rates and IsotOpe Effects of the Solvolysis of 8-Methyl-d0— and -dS-l-naphthoyl Chloride in 65% Acetone-water (V/V). 3 -l 3 -l o Isotope k x 10 sec k x 10 sec kH/kD Temp., C do 18.948 18.979 i .029 1.018 f 005 -15.45 18.983 19.006 d3 18.602 18.639 1 .086 18.737 18.578 do 9.613 9.618 1 .017 1.023 t .002 -20.60 9.636 9.604 is 9.397 9.398 t .001 9.399 18 Table 6. Rates and Isotope Effects of the Solvolysis of 8-Methyl-d0- and -d3-l-naphthoyl Chloride in 70% Acetone-water (V/V). 1:?“ :3-‘:-=T=‘-=:T= ’—"" """-‘ i 44-7-3 :-’—'3- ""—=: & rm 3 r2:=‘==8 twfi Isotope k x 103sec.1 E x 103’sec'1 kI/kD Temp.,°C do 14.914 14.821 1 .081 1.014 1 006 -10.39 14.764 14.784 d:5 14.597 14.610 1 .023 14.596 14.636 d0 7.840 7.852 1 .036 1.018 1 .010 -15.45 7.824 7.892 d3 7.663 7.717 1 .062 7.703 7.784 d0 4.135 4.136 1 .014 1.023 1 .006 -20.60 4.122 4.150 d3 4.038 4.041 1 .017 4.026 4.060 Table 7. IsotOpe 19 Rates and Isotope Effects of the Solvolysis of 8-Methyl- in 75% Acetone-water (V/V). _=_..-- .——-— o.-.» .———‘_..: r7: k x 10 sec- d - and -d —O 3 -l-naphthoyl Chloride —— 1 F x 103 sec.1 kH/kD Temp. .°C 90 20 20 20 20. 20. 20. MM“ 010404 000‘ 000‘ .667 .572 .727 390 487 175 .386 .408 .434 .292 .321 .340 .199 .202 .210 .146 O 141 .149 1.629 r-u—I FJPHH .687 .680 .648 .653 .669 20.655 l+ .078 20.351 H' .160 6.409 H- .024 6.318 H- .024 3.203 H- .006 3.145 I+ .005 1.686 H- .006 1.657 H- .011 1.015 H- .009 1.015 H- .005 1.018 H- .002 1.018 H- .008 -10. ~15. -20 .46 39 45 .60 20 Table 8. Rates and Isotope Effects of the Solvolysis of 8—Methyl-d0- and -d3-l—naphthoyl Chloride in 80% Acetone-water (V/V). r==': 1 T‘- .‘ :3 L: L: ;_ : :;_=|=-.-:.:=:..:::z:: ="—‘— " =’—.' -1:=" _'_4-—.‘s :- 71 {2:22.22 8 IsotOpe k x 103sec-1 k x 1035ec-1 kH/kD Temp.,°C d _0 .439 7.456 .450 .463 .509 .429 .477 .443 .442 I+ .026 1.014 1 .007 - 0.46 H- .304 7.380 .385 .376 .390 .451 .396 .396 .355 .045 \I\J\l\l\l\l\l\l \l\l\l\l\l\)\l\l .349 2.358 .355 .354 .375 H- .011 1.008 M- .009 -10.39 J33- .351 2.339 .347 .318 H- .018 NNN NNNN H- H- .146 1.145 .142 .147 .002 1.016 .005 -15.45 hfldhd .128 1.127 .121 .131 H- .005 hfldhd 0.6063 0.6078 .002 —20.60 0.6086 0.6078 0.6086 I+ .0011 1.017 H- 0.5986 0.5975 0.5972 0.5968 H- .0009 Table 9. Isotope 21 Rates and Isotope Effects of the Solvolysis of 8-Methy1- m...“- :. :12”: *- -..___- --- k x 103sec‘ d —O- and -d3-l-naphthoyl Chloride in 85% Acetone-water (V/V). 1 F x 10 3 -1 sec kH/kD -rm. Temp.,°C NN 2.811 2. 820 .787 .787 2.772 GOO GOO GOO COO .9254 .9257 .9275 .9124 .9134 .9108 .4348 .4339 .4282 .4289 .4292 .4282 0.2181 0 .2184 0.2184 COO .2158 .2146 .2152 2.815 2.782 0.9262 0.9122 0.4345 0.4288 0.2183 0.2152 t H- .006 .008 H- .0011 H- .0013 H- .0005 H'- .0006 H- .0002 H- .0006 1.012 1 1.015 1.013 1.014 .004 H- .002 H- .002 H- .003 - 0.46 -10.39 -1S.45 -20.60 22 Two rate determinations which deviated from the mean of the remaining rate constants by more than ten times the standard deviation were discarded. All rate constants used were determined with the same batch of solvent, except for the runs in 80% A/W at 0°. Rate ratios, kH/kD, (Tables 17) are the ratios of the corresponding means. The uncertainty indicated is the standard error obtained from the following relation: . 2 2 L 6 = (kH/kDMIbH/kb) + (OD/ED) 1’ Rate ratios and the standard errors were calculated by using program RRD (Appendix Two) written by the author. The activation parameters AHI and AS* (Table 13) were calculated by using program AKTIVE by a single least squares solution from the following equation: 16g(E/T) = (-AH*/Z.3026R)(l/T) + AS*/2.3026R + 10g(k/h) The difference in enthalpy (AAH* = AH; - AHB) and entropy (AAS* = A8; - ASE) of activation were calculated by using the following equation: 1n(kH/kD) = -AAH*/RT + AAS*/R In using the least squares curve-fitting program KINFIT (Table 18), written and provided by Professor Dye of this Department,29 the errors in the kH/kD determinations were weigHUuiin calculating AAH* and AAS*. Program HAND calculates AAH* and AASIII (Table 18) by a single least squares solution of lnkH/kD 12- 1/T. 23 Program RATE, AKTIVE and HANDS were written by Dr. Sonnichsen and program HAF (Appendix Three) was written by the author to calculate the half-live, and three- and thirteen half-lives. 24 SYNTHESIS The deuterated and undeuterated compounds were synthesized by the procedure summarized in Figure 1.30 Br 021 as m.p. 177 (lit. 177) SOClz 91% r DB Br CDZOH COCl crude SE “"3“. m.p. 67-9 1’? tP 775- 978N131; “Sjé-E— 3;“ (lit. 67-8) lit. 88p 9 LAD 91% ( )8 0 CD3 02“ SOCl2 D3 Mg C0 @‘:’ 634 99 first 60% 95% vac. dist'd m.p . 78.5-79.0 m.p . 155.5-156.0 m.p. 51.0-51.8 (lit. 77- -8) (lit. 155-156) (11t. 54) Figure 1. Synthetic scheme. 25 Preparation of Anhydro-8-hydroxymercuri-l-naphthoic Acid 31 The procedure of Whitmore E£.£l° was applied. In each of four 5-liter round-bottomed flasks, 49.5 g (0.25 mole) of 1,8-naphthalic anhydride was suspended in a solution of 35 g (0.875 mole) of sodium hydroxide in 1500 m1 of water. As each flask was heated, the color turned dark brown. After 30 min. of reflux all solid material had dissolved completely. When 87.5 g (0.275 mole) of mercuric acetate dissolved in 50 m1 of glacial acetic acid and 250 m1 of water was poured slowly into each flask, the Color turned orange. By adding 75 m1 of glacial acetic acid each solution was made acidic (pH 5). After 24 hrs. reflux, the solid was filtered and washed with two liters of water, one liter of ethanol and one liter of ether. Without further purification, the solid from the four reactions was used in the next step. Preparation of 8-Bromo-1-naphthoic Acid The procedure of Rule E£.El°32 was applied. Assuming a quantitative yield of anhydro-8-hydroxymercuri-l- naphthoic acid (1 mole, 371 g) in the previous step, we suspended the impure material in 1500 ml of glacial acetic acid and 250 ml of water in a 5-1iter three-necked round- bottomed flask. The mixture was vigorously stirred with a mechanical stirrer and cooled in an ice-bath. A bromine- sodium bromide aqueous solution (680 g sodium bromide in 1250 m1 of water and 56.6 ml of bromine) was added slowly from two dropping funnels so as to disperse as soon as 26 possible. The slurry formed was heated slowly to 100° and poured into a mixture of 2500 ml of water and 2500 g of ice with stirring. The white precipitate was suction filtered and washed with two liters of water. After drying in the hood, it yielded 84% of 8-bromo-l-naphthoic acid. Careful recrystallization gave acid which melted at 177° (lit.32 m.p. 177°). It was found that recrystallization was not necessary to get good yields in the next step. All the reaction apparatus described in the following sections was oven-dried overnight at 130° prior to use. Preparation of 8-Bromo-1-naphthoyl Chloride The thionyl chloride used was purified by Fieser's method.33 A mixture of 79 g (0.32 moles) of 8-bromo-l- naphthoic acid and 38.4 ml (0.48 moles) of thionyl chloride was refluxed in a 250 ml round-bottomed flask for two hours until evolution of hydrogen chloride and sulfur dioxide stopped. The reflux condenser was fitted with a barium oxide drying tube. The dark solution was kept overnight and then twice distilled under vacuum through a distilling head which was heated with electric heating tape to prevent crystallization. To protect the vacuum pump a drying tower filled with a mixture of barium oxide, potassium hydroxide and drierite and a liquid nitrogen vacuum trap were used. 34 B.p. 143-5°/0.06 mm, m.p. 67-9° (lit. m.p. 67-8°), yield 90.5%. 27 In all the syntheses that follow dry ether means ether distilled from lithium aluminum hydride immediately before USE. Preparation of 8-Bromo-l-naphthylcarbinol-a,a-d_2 In a 2-liter round-bottomed flask, 9.7 g (0.23 moles) of lithium aluminum deuteride (LAD) was suspended in 230 m1 of dry ether. A solution of 76.7 g (0.285 moles) of acid chloride in 900 ml of dry ether was added slowly over a one hour period with stirring while maintaining gentle reflux. After addition was finished, the mixture was further refluxed for 5 hours, cooled in an ice-bath for 30 min. and carefully quenched under stirring, with 16 m1 of water and 16 ml of 5% sodium hydroxide. The suspension was stirred overnight at room temperature and filtered. The ether layer, to which was added the ether used to wash the solid, was washed three times with 16 ml of water and dried over anhydrous magnesium sulfate. After rotary evaporation of the ether, a 96% yield of the alcohol was obtained. After recrystallization from cyclohexane the 30 35a 35b alcohol melted at 85-7° (lit. m.p. 86.8°, m.p. 88-9°, m.p. 87-88.5°). Preparation of 8-Bromo-1-bromomethy1naphthalene-a,a-d2 A solution of 58.2 g (0.244 moles) of 8-bromo-l- naphthylcarbinol-a,a-d2 in 1500 m1 of dry ether and 20 ml (0.25 moles) of pyridine dried over barium oxide36 was placed in a 5-liter three-necked round-bottomed flask fitted 28 with a mechanical stirrer and a condenser with an anhydrous barium oxide drying tube. To it was added with stirring 200 g (0.738 moles) of freshly opened phosphorus tribromide. The mixture was refluxed for 13 hours, cooled to room temperature and poured into 500 ml of water. The ether layer, to which was added the ether used to wash the water layer, was washed with 2 x 30 ml of water, 30 m1 of saturated sodium bicarbonate and 30 ml of water. It was treated over anhydrous magnesium sulfate with norit, filtered and rotary evaporated. Yield: 93.2%, crude m.p. 75-7° (lit.34 m.p. 78-9°). The nmr spectrum in carbon tetrachloride showed only an aromatic multiplet at T = 2.08-2.96. The product was used without further purification in the next step. Preparation of l-Methyl-d3-8-bromonaphthalene In a 5-liter three-necked round-bottomed flask fitted with a mechanical stirrer and condenser with anhydrous barium oxide drying tube was suspended 10 g (0.238 moles) of lithium aluminum deuteride in 600 ml of dry ether. To it was added over a 20-minute period, with stirring, 104.5 g (0.344 moles) of l-bromomethyl-8-bromonaphthalene-a,a-d2 dissolved in 1300 m1 of dry ether. After being refluxed for 8 hours, the mixture was cooled in an ice bath for one hour. It was very carefully quenched with 16 ml of water and 16 ml of 5% sodium hydroxide and stirred overnight. The solid was filtered and washed with ether. The filtrate was 29 washed with 2 x 16 ml of water, decolorized with norit and dried with anhydrous magnesium sulfate. After removal of the ether by rotary evaporation, a 90.5% yield of product was obtained. Careful recrystallization led to a product melting at 78.5-79.0° (11t.37 m.p. 77-8°,34 m.p. 76-7°). The nmr spectrum in carbon tetrachloride showed only an aromatic multiplet at r = 2.17-3.08. Preparation of 8-Methy1-d -1-naphthoic Acid 3 In a 2-liter three-necked round-bottomed flask equipped with a mechanical stirrer and a condenser with anhydrous barium oxide drying tube was placed a suspension of 21.3 g (0.875 moles) Domal high purity magnesium granules, previously heated in an oven for 30 minutes, in 32 ml of dry ether. To it was added slowly over a 20 minute period 40 g (0.178 moles) of l-methyl-d3-8-bromo- naphthalene dissolved in 65 m1 of dry ether and 5 m1 of ethyl bromide (dried over barium oxide). The mixture was refluxed and formed a greenish-milky suspension. To it was added dr0pwise 26 m1 of dry ethyl bromide mixed with 175 ml dry ether at such a rate as to keep the mixture refluxing. After the mixture was heated at reflux for an additional 15 hrs., 180 g of dry ice was added slowly, followed by 450 m1 of ether and an additional 180 g of dry ice. Stirring was continued until the excess dry ice was evaporated. Then 230 ml of 20% hydrochloric acid was added and the resulting solution was stirred overnight. The 30 ether layer was washed with 3 x 250 m1 of water. After that, 40 g of norit and 10g magnesium sulfate was added to it. The solid was filtered and the ether was removed by rotary evaporation. Yield: 94.5%. After careful recrystallization the solid melted at lS4-S° (lit.30 m.p. 155-6°,34 m.p. lSZ-3°,38 m.p. 153°). The nmr spectrum in carbon tetrachloride showed an aromatic multiplet at T = 1.90-2.39, a singlet at T = -2.70 and £2 methyl peak at T = 7.2. The mass spectrum had a peak at m/e 189 (96 1 3% labeled). The melting point of the relevant X-l-naphthoic acids (Table 10) confirm the conclusion that the compounds synthesized in this work have the substituent (bromo and methyl group) in the 8-position.39 Table 10. Melting Points of X-l-naphthoic Acids39 "' —" =3: -=F:.—c——=;:'—- m ~~~w— .3 23:!»- Position of x 3 4 5 6 7 857 Br 233-4 219-20 258-9 185-6 235-6 177 m.p. for X rep. for X CH3 --- 176—7 188-9 178-9 146-7 154-5 31 Preparation of 8-Methyl-d3-l-naphthoyl Chloride About 5 g of 8-methyl-d3-l-naphthoic acid was refluxed with 16 ml of purified thionyl chloride for one hour and kept at room temperature overnight. The excess thionyl chloride was distilled under vacuum, first at room temperature and then be heating to 60° (30 mm Hg) for three hours. The product acid chloride was distilled under vacuum (0.04 mm Hg) twice through a 15 cm glass spiral column with a distilling head that was heated to 60° with an electric heating tape, to prevent crystallization. The vacuum pump was protected with a drying tower containing a mixture of drierite, potassium hydroxide and barium oxide and a vacuum trap cooled with liquid nitrogen. The first 60% of the product which distilled was collected in three fractions. Of these, only the second and third fraction were used in the kinetic runs. The product boiled at 83-4°/0.04 mm Hg and melted at 51.0-.8° (111.34 m.p. 54°). From the intensity of the mass spectrum peaks at m/e 205 (0.9%), 206 (2.8%) and 207 (96.3%) we concluded that the product contained about 96% of the d3 isomer. The nmr spectrum in carbon tetrachloride showed only an aromatic multiplet at r = 1.95-2.63 and dd_methy1 peak at r = 7.3. The infrared spectrum in carbon tetrachloride showed strong absorption at 1775 cm-1. 32 Analytical Instruments The mass spectrometer was a Hitachi, Ltd. RMU-60. Infrared spectra were taken with a Perkin-Elmer 237 B model, Grating Infrared SpectrOphotometer. For the nmr spectra, either a Varian Associates A 56/60 D Analytical NMR Spectrometer or a Varian T-60 NMR spectrometer was used. Melting points were determined with an electro- thermal melting point apparatus by using a calibrated thermometer. RESULTS AND DISCUSSION MECHANISM OF ACID CHLORIDE SOLVOLYSIS The evidence for the borderline mechanism of acid chloride hydrolysis will be presented along with the suggestion that 8-methyl-l-naphthoyl chloride solvolyzes by a limiting mechanism in 65% and 70% acetone-water. Acid chloride solvolysis is typically borderline40 in character; namely, depending on reactant structure, solvent, temperature or other reaction conditions, acid chlorides react with nucleophiles differently, in a manner that we may describe either by a mixture of simultaneous Snl and AE reactions (dual mechanism),41 or by a unified mechanism42 in which the transition state changes as a function of reaction conditions. Since acid chlorides might show also typical limiting or nucleophilic behavior, they are good systems to correlate various aspects of isotope effects (AAGI, AAH* and AAST) and reaction mechanisms. The Effect of Solvent Changes on Mechanism The borderline character of some acid chloride hydrolyses has been demonstrated by Hammett op plots which are strongly solvent dependent (Figure 2). It was thought 33 34 50% Water 5% Water a 95% Acetonepdflfi 506 Acetone 1% Water in 2 Formic acid | 4* p = 4.4 0 log 7"- 04 -2 Figure 2. Plots of log @jko) against 0 for hydrolysis of substituted benzoyl chlorides in various solvent mixtures. 35 that the Snl mechanism takes over in an increasing number of cases as the solvent polarity increases.43 From rate-product studies of the hydrolysis of benzoyl chloride in aqueous solutions containing different 41C have amounts of d-nitroaniline, Gold and coworkers suggested that benzoyl chloride reacts by a nucleophilic mechanism in 80% acetone-20% water (w/w) and by about 50% limiting mechanism in 50% acetone-50% water (w/w) solution. As steric hinderance to AE reaction increases, acid chlorides tend to react 212 an Snl mechanism. Bender and Chen44 have studied the hydrolysis of p-substituted 2,6- dimethylbenzoyl chlorides in 99% acetonitrile-1% water solvents. The neutral (Hammett p = -3.85) and acid- catalyzed hydrolysis (Hammett p = -3.73) of these compounds proceed by a limiting mechanism, as supported by a common ion effect, salt effect, the large negative Hammett p values correlated with 0+ and by no carbonyl oxygen exchange. In 95% dioxane-S% water solvent, which has a lower dielectric 41a have found that the constant, Bunton and coworkers carbonyl oxygen of the mesitoyl chloride does exchange with water. This indicates that this sterically hindered acid chloride hydrolyzes, at least partially, by an AB mechanism. Thus, it is reasonable to assume that the sterically 45 hindered 8-methyl-l-naphthoyl chloride hydrolyzes mainly by a near limiting mechanism in polar media. 36 Correlation of Activation Parameters and Reaction Mechanism Long and coworkers46 have suggested that the entropy of activation (AS*) might serve as a convenient criterion of mechanism. Since in a nucleOphilic hydrolysis a water molecule participates in the bond formation with loss of its translational freedom, they predicted that reactions proceeding by nucleOphilic mechanisms will have lower entropies of activation relative to those of reactions which proceed by limiting mechanisms. Experimental results summarized in Table 11 are consistent with their prediction. However, Long suggested that the entrepy criterion must be used with caution and with other evidence. In view of the complex behavior of entropy of activation in solvolysis reactions48 Long's caution is fully justified. These cautious comments notwithstanding, we conclude that the entrepy of activation (ASI) data summarized in Table 12 are consistent with a borderline behavior of the solvolysis of the acid chlorides with decreasing solvent polarity, as AS* becomes always more negative. The small negative values (AS* = —2 e.u. in 65% acetone; AS* = -6 e.u. in 70% acetone) for 8—methyl-l-naphthoyl chloride (Table 13) are consistent with a limiting mechanism in these solvents, which becomes less limiting as the solvent polarity decreases. The large negative values (AS+ = ~26 e.u. in 48% MeOH; AS* = ~30 e.u. in 54% dioxane) for the nucleophilic base hydrolysis of methyl 8-methy1-l- naphthoate further support this view. 37 Table 11. EntrOpies of Activation for Limiting and Nucleoghilic Solvolyses in Aqueous-acetone at 50° 7 Substrate Solvent (% Acetone-Water) AS* (e.u.) (a) Limiting Mechanism t-BuCl 7O ~10.89 80 ~12.4l t-BuBr 7O ~10.20 80 —ll.23 MeOC6H4CH2Cl 70 -11.99 80 -l4.21 MeOC6H4CHZOTs 85 -10.29 bhzcnc1 70 ~10.26 80 ~12.96 NOZC6H4CHPhC1 70 ~ 9.96 NOZC6H4CHPhOTs 85 ~ 8.12 (b) Nucleophilic Mechanism PhCHZCl 70 ~23.96 PhCHzBr 70 ~23.75 PhCHZOTs 85 ~19.62 PhCH OTs 70 ~16.64 2 38 wumvcmpm may moefiu 03» 6pm whosho wouozc was m>Hhx< Emumoua Eoum woumasoamu .ADNV :oHumw>ov .D .smuma wom-6:mxofie 3 wen “new Amps: wom-Hoeme66e 2 ”OH "203 .H6umz www-maououm > smo mopAuaeea fiuu< mo moflQOHucm .NH ofinmb 39 Table 13. Activation Parametersa for the Hydrolysis of 8-Methyl-d0~ and ~d3~naphthoy1 Chlorides I.- 5 ‘m‘a: 5~=3== , = :1 mm‘n==~ "f-‘i’? -. m: 35”: “3'72"; . "3m $.74 olvent AH AS T%A/W) Is°t°Pe (cal/mole) (e.u.) 65 go 16,584 -1.3 $3 16,712 -1.3 70 $0 15,9821255 ~5.81l.0 g3 16,0941234 -5.41o.9 75 go 16,6011343 -5.211.3 g3 16,6261362 -5.111.4 80 do 16,6511519 -7.012.o 93 16,7041558 -6.812.1 85 go 16,9581805 -7.313.1 g3 16,9721798 -7.713.1 3Calculated from program AKTIVE and quoted errors are two times the standard deviation (20). 40 Isot0pe Effects as Criteria of Mechanism An excellent review on this subect was written recently 2’9 Schleyer9 and Scheppele.1d by Shiner, Since the first observation of larger isotope effects in near limiting solvolyses and smaller effects in near nucleOphilic diSplacements, intensive efforts have been made to use the observed isotOpe effects as criteria of reaction mechanisms. In general, for Sn2 reactions typical observed values are in the range 0.95-1.06 per 61deuterium at 25°, with variation in a systematic way as a function of the leaving group, nucleophile and substrate reactivity. For limiting mechanisms, the observed a-isotope effects are larger, with an upper limit that depends on the leaving group; about 1.15 for chloride and 1.22 for sulfonate groups,49 at 25°. The benzyl derivatives are very good examples by which to show this correlation, because substituent effects on solvolysis rates have shown that this series of compounds belong to the classical borderline region50 that might exhibit typical Snl and Sn2 behavior at the extremes of reactivity. Shiner and Rapp2 measured the a-d effects of p-methylbenzyl chloride in 70% and 97% trifluoroethanol- water mixtures (Table 14). These solvolyses were assumed to be largely limiting, because the observed a-isotope effects per D were reasonably constant [1.143(97% TFE) and 21J40(70% TFE)]. However, in 50% (V/V) ethanol-water 41 (Y-value = 1.655 at 25°) at 45°, p-methylbenzyl chloride showed an a-effect per D of 1.086. The authors explained it by a significant amount of nucleOphilic attack by solvent. It is interesting to compare these results with 51’52 from the that of Karabatsos and his coworkers solvolysis of 8-methyl-l-naphthylcarbinyl~a,a-d2~ chloride. In a much less polar solvent (67% acetone-water, Y = 0.2) and at a lower temperature (25°) this isotope effect already shows the upper limit of the limiting mechanism (kH/kD = 1.15 per D). The reason lies in the sterically hindered nature of the compound, a feature that is also present in 8-methyl-l-naphthoyl chloride. Furthermore, 8-methyl-1-naphthoyl chloride solvolyzes about 10 times faster than l-naphthoyl chloride,51 whereas methyl 8-methyl- l-naphthoate reacts 5000 times slower than methyl-l- naphthoate.61b All the above data provide good evidence to support a near limiting mechanism for the solvolysis of 8-methy1-1- naphthoyl chloride. THE ORIGIN OF THE ISOTOPE EFFECTS 51’53 have used two Karabatsos and his coworkers approaches in efforts to understand the origin of secondary deuterium isotOpe effects and to assess the relative contributions of hyperconjugation and nonbonded interactions to them. 42 bp-bromobenzenesulfonate Table 14. a-Deuterium Effects (per D)8 on Rates of Solvolysis of Some Benzyl Derivatives at 1:12:11: -‘ ————— F—r—- =[=-—==~==‘ =1 Leaving group Cl OBsb OBsb OBSF: Substituent p-CH3 none p-CF3 P-NO2 Solvents 971C 1.143 1.173 ---- 1.026d 80T ~~-- 1.161 ---- ---~ d 70T 1.140 ~~~~ 1.042 1.011 50E 1.086d ---- ---- ---- 70E ---- --~~ 1.019 1.006 80E ~--- 1.074 1.016 1.005 90E --~- 1.060 1.014 1.002 95E ---- 1.053 1.014 ~--~ a 15 (kn/knz) C971" indicates 97 wt% 2,2,2~trif1uoroethanol~3% water; 70E indicates 70 vol% ethanol-30% water etc. dEstimated from observed values at 45°, assuming normal temperature dependence. 43 They measured isotope effects as functions of temperature so as to calculate the relative contributions of AAH* and AASt. This is important, in View of the demonstration by previous studies that the contribution of AAS* is in some cases not only significant, but even dominant. Such a AASI contribution to isotope effects is not surprising in view of the importance of entrOpy in 54 One useful result equilibria and kinetics in general. of the temperature studies is to separate the better understood AAH* term, which can be interpreted in terms of hyperconjugation, inductive effects and nonbonded interactions, from the more complex AAST term. 2) They chose two types of systems in which to estimate the relative importance of hyperconjugation and nonbonded interactions: first, systems in which nonbonded interactions are predominant and hyperconjugation is negligible, and second, systems in which nonbonded interactions and hyperconjugation both contribute to the AAHT term. By comparing the experimental AAHT values obtained from both systems with the theoretically calculated AAE term due only to nonbonded interactions, they expected to be able to estimate the relative contributions of hyperconjugation and nonbonded interactions in systems in which both effects are operating. 44 Relative Importance of Enthalpy (AAH*)_and Entropy (AAS*) Term to IsotOpe Effects From the data summarized in Tables 15-16, it appears that both AAH* and AAS* contribute to the observed isotope effects and that both terms have the same sign. In the discussion that follows we will try to show that isotope effects based on AAH* may be better mechanistic criteria than those based on AAGI. We wish to make the following pertinent comments with respect to the data summarized in Tables 15-20. (a) The 8-23 isotOpe effects for the hydrolysis of acetyl chloride in 85% acetone-water (Table 15) are: from 423 Both AAHT, kH/kD = 1.21 and from AAG*, kH/kD = 1.11. isotope effects indicate some limiting character in the reaction mechanism and qualitatively agree with the theory of force constant changes. (b) In the nucleophilic hydrolyses of the acid chlorides and esters (Table 15), the isotope effects kH/kD due to AAG:t are near unity. However, in the activation process the force constants should increase, because the demand for hyperconjugation decreases and nonbonded interactions increase. Therefore, an inverse isotOpe effect (kH/kD < 1) is expected. Indeed, the AAH* term is reasonably large and positive leading to an inverse isotope effect. The experimental isotope effect is only near unity because of cancelling contributions from TAAS* and AAH* to 110*, .mufiefia oucopfimaou wmm wcfivaowx :ofiumfl>ow pudendum ecu mmsfiu 039 ohm wouosc mhouumn .UHQ HQHmZ meIOGOumum w>mw MOHNUHVEH H6m Squeegeu .eanfimmom mpumwmm m>fiuusvaH ecu mcofiuumhoucH popconcoz .cOMummsn:06th>m an“: mwcnomEou mo mfimxaonvx: .mH eHan 46 (c) The remote isotope effects due to AAGI are not only insensitive to reaction mechanism, but also misleading (Table 16). For example, Brown and his coworkersSS have suggested that the limiting activation process in the solvolysis of the 2,6,6-trimethyl~egd272~norbornyl system is sterically hindered by the interaction between the leaving group and the dddd~6-methyl group. On this basis, an inverse isotOpe effect is predicted. Contrary to expectation, the experimentally observed isotOpe effect62 (Table 16) is normal. However, the large experimental positive AAHT value (3223-6-CD AAHT = 204182 cal/mole; exo-6-CD AAHI = 3’ 3, 244196 cal/mole) gives a large inverse isotope effect (kH/kD = 0.70), as expected by the mechanism. The surprisingly small normal effect results from the slight domination of the TAAS+ term over the AAH* term. (d) From the results of the present study (Table 18), the negative AAH* [= -11416 cal/mole, (kn/kn)“ = 1.20 in 70% acetone-water] value is reasonable in view of the near limiting character of the solvolysis of the 8-methy1-l- naphthoyl chloride. The observed small isotope effect (kH/kD = 1.018) (Table 17) again results from the cancelling contribution of the TAAS* term (AAS* = ~0.41.0 e.u.) to the AAGI term. 47 .muHEHH oucowfimcou wmm wcflvfiofix :ofipmfi>ow unwwnmum ozu moeflp 03H ohm wouosv mhouumn .uuo Hope: wow-Ho:m:uo w>om mopmuwpafi mow "hope: wom-ocououm w>on moumuwpafi Hom canonso - IIIINHIIMIINNNNHR efieemfifimez mum muuommm UHCOHpuon paw shoe nm<< ou housnwhucou ucmcfiEowohm ezu ohm m:0wuumhmucH vovaoncoz :ufinz ca mwcsomEou mo mamkaOhwzz .oH manmh 48 Table 17. Temperature and Solvent Dependence of the IsotOpe Effect in the Hydrolysis of 8-hbthyl-do- and ~d3~l~naphthoyl Chlorides —— ” ~—. --_oa—«*—4—-_-— h..— H “.m---”- —‘— .— olvent o a (% A/W) Temp. kH/kD 65 ~15.4S 1.0181.005 -20.60 l.0231.002 70 ~10.39 l.0141.006 ~15.45 l.0181.010 -20.60 l.0231.006 75 - 0.46 l.0151.009 -10.39 l.0151.005 ~15.45 1.0181.002 -20.60 1.0181.008 80 ~ 0.46 l.0141.007 ~10.39 1.0081.009 ~15.45 1.0161.005 -20.60 1.0171.002 85 - 0.46 1.0121.004 ~10.39 1.0151.002 ~15.45 l.0131.002 -20.60 1.0141.003 3Calculated from program RATE and RRD. the standard deviations. Quoted errors are 49 .coflumummoua uco>H0m ou one muchno mpsaucfi newsz «amp Eouw vopmasuamu U .vonmfloz one: maOMumcHEuouou Qx\xx cw muouho ”Emhmoum HHmsz scum woumanuHmuo .Empmoum mozop pudendum exp moefip 03» ohm voposc muouhmm o.eH.o+ N.no.o- o.eo.o- NHNAH + omen - wHHOH - mm m.nfi.o- ~.nfl.o- H.nfl.o- eenme - oeewm - mmnAN - om H.HN.o- ~.HH.o- H.HH.o- efinam - wNHQN - eHHmN - me o.ee.o- o.we.o- o.ne.o- NHNOH- meeHH- eneHH- on e.o- e.o- e.o- mNH- mNH- mNH- me n.:.6v fi.s.ev h.s.ev fleHes\Heev fiefiee\fieev meHeE\Heev fia\< s>v eemee uemee .DMmee nemee eemee mwmee eee>flem I ol mam>~0hpxm eeeeefieu Heeeeeeee-fi- e- eee - e-Hs;eez-w me eeemem eeeeomH esp we ouaopcommm ounpmuomsoh on» scum woawfiuouom mhouoEmnmm :oflum>wuu< .wH manna w 50 Relative Contribution of Hyperconjugation and Nonbonded Interactions to the Enthalpy Term (AAHi) It is interesting to compare the nonbonded isotOpe effects due to AAHT with those from AAE values for the 8-methyl-l-naphthoyl chloride and its 8-methyl-d3 analog (Tables 19-20), which were calculated by using Bartell's 17 51,52 procedure and two different potential functions. From the fact that the dihedral angles between the planes of the ring and the nitro group of 1,8-dinitronaphtha1ene63 and l,5~dinitronaphthalene64 are 43° and 49°, it is reasonable to assume that the dihedral angle between the ring and COCl planes is between 45° and 60°. The calculated isotOpe effects for such angles are reasonably close to those observed experimentally. In Table 20 are summarized a few calculated and experimental isotOpe effects. Since the dihedral angles between the C-CO-C plane of the aromatic ring for l-acetylnaphthalene is 42°, it is reasonable to assume a dihedral angle for l-naphthoyl chloride of about 40°. Therefore, the calculated isotope effects are again reasonably close to the eXperimental ones. However, where hyperconjugation is possible, E1£° acetyl chloride and E-butyl chloride, the calculated nonbonded isotope effects (AAE) are much smaller than the experimental values. Therefore, within the limits of accuracy imposed by the assumptions of Bartell's procedure663 and the experimental values used (potential functions and geometry), our preliminary studies indicate that in ordinary systems with hyperconjugation possible, the dominant contributor to the AAHI term is hyperconjugation. 51 Table 19. Calculated Isotope Effects Due to Nonbonded Interactions for 8-Methyl-l—naphthoyl Chloride and 8~Methy1~d3 Analogs -—-_.— ---‘—- .—... .. . fin..-” Flu— .- -p.- deg C deg.d ca123:1e kéZED ca12fizle REZED 0 0 ~870 4.31 ~530 2.43 15 7.5 ~710 3.32 -520 2.03 30 14 ~460 2.17 -230 1.48 45 18 -250 1.52 -100 1.18 60 20 ~ll6 1.21 - 40 1.07 75 20 - 90 1.16 - 20 1.03 90 0 - 81 1.15 - 18 1.03 a AAE = AEH ~ AED, potential function from ref. 66a. bPotential function from ref. 66b. CDihedral angle between COCl and naphthalene ring. dDihedral angle between methyl group and naphthalene ring. 52 ounowfimcou wmm mcwvamwx :ofiumfi>ov ppmvcmum one mesa» 03p ohm wouozd muchum .mefiefifi v .n1< - 11¢ u wee .ame - awe u mean .mcflh ocoamcusmm: use Huou :ooZuon oamcm Hmnpogfion .woo .mop mm .200 .moh um Scam moEou newuwasuHmo wow wow: :oHuuqsm Hmfiucouomm eHH- . co m omm- me m .EfiH ow - on m so new: ma- H0m meo *m<< uwamu mqq p mmhv> vasomsmw momxao>aom mcwuwefiq oEom How ner¢d.aomomvmh~23 .z.#HH.AH.b&.A¢CN.vap~&3 ZANcmom.uh~a3 Am#o_uH.A_.1<.Acomcmvwhmaz AA~I2V\::2:mvbmcmu¥CV .CC}Z¢+.~VQCVHQCIZM om 2.~uH cm cc ouoczzm m¢¢A>< 5.0.22m4gmvhflflbafzm ®~ 2._H~ I~ CC .ouhmzrm zuzm AZ.~HH..~VPQVA¢ON.NVCwa obmrm~.xNH.m.o~u.uwhwc Ecmocau mzo XHszmm< 61 czm ochm co ed ch cc .z.~uH.AHcm..Hvaov.mom.mVMPua3 .z.~uH..H.cw..Hccac.Nom.m.m»na3 .z.~nH..H.rm..Hvra..com.mvm»~03 .om.~uH.Ava«.Aoom.m.whwa3 .m¢¢A.cha\.Hccmv+m¢¢..Hcrz\.Hch.vhqcm¢.~vaau.~cm ea 2.~u~ ed cc AHccm\.HcIan.H.aa md :.~uH ma cc znzu .z._nH.Achm..cha.AHvIu.Accra..eom.m.cmc chm Im~.xm_.\\\.m.o~i.uccppczcuz_ xc~.xm.~.oca.uu4<1~ immaI» 1N_.xm.~.c~u.nmuccluqqr I-.xm.m.m~m.nm»ao Ic.xm.\\\ch