AN ABSTRACT OF THE THESIS OF
WAYNE HARVIE STANLEY for the . M. S. . in Chemistry (Name) (Degree) (Major)
Date this thesis presented 40 ,sr /7 jc/64.
Title STUDIES ON THE MECHANISM OF THE ALKALINE HY-
DROLYSIS OF THIOLSULFINATES Redacted for Privacy Abstract approved c (Major professor) í-
The alkaline hydrolysis of phenylbenzenethiolsulfinate was studied at various pHs in an aqueous solution of 25% dioxane. A pH stat was constructed to follow the rate of disappearance of hydroxide ion under pseudo- first -order conditions. Comparative studies were made under identical conditions by following the disappearance of the thiolsulfin- ate in a buffered solution in the ultraviolet at 296 m,t(.
A product study was made which indicates that the reaction fol- lows the course indicated in Equation 1.
3(pSóS() + 20H 2COSS1) + 20S02 + H2O (Eq. 1)
The kinetic studies indicate that the reaction does not follow simple first -order kinetics but is quite complex. The first -order plots exhibit a marked curvature, being fast initially and slowing to a final lesser first -order rate. Possible schemes are discussed which can explain this curvature but no mechanistic conclusions are dr awn.
The extreme complexity of the data suggests that the glass elec- trode response is altered by the presence of the dioxane to such a degree as to render it unreliable as a mechanistic tool. For this reason, it is suggested that further studies on the mechanism of the reaction be carried out in purely aqueous media. To this end the synthesis of a sufficiently water - soluble thiolsulfinate was attempted.
The results of this work indicate that the alkaline hydrolysis of thiolsulfinates is a complex and very facile reaction, requiring study in several systems to fully elucidate its mechanism. STUDIES ON THE MECHANISM OF THE ALKALINE HYDROLYSIS OF THIOLSULFINATES
by
WAYNE HARVIE STANLEY
A THESIS
submitted to
OREGON STATE UNIVERSITY
in partial fulfillment of the requirements for the degree of
MASTER OF SCIENCE
June 1967 APPROVED:
Redacted for Privacy Professor of Chemistry in Ch(rge of Major
Redacted for Privacy Chairman of Department of Chemistry
Redacted for Privacy Dean of Graduate School
Date thesis is presented 4 < Subject Page INTRODUCTION 1 RESULTS 7 Product Study (I) 7 Kinetic Studies (II) 8 DISCUSSION 14 EXPERIMENTAL 19 pH Stat (I) 19 Preparation of p- Carboxyphenyl Disulfide (II) 24 Attempted Preparation of p- Carboxyphenyl p- Carboxybenzenethiolsulfinate (III) 24 Preparation of Phenylbenzenethiolsulfinate (IV) 25 Preparation of Benzenesulfinyl Chloride (V) 26 Thiophenol (VI) 26 Product Study (VII) 26 Preparation of Buffer Solutions (VIII) 27 Purification of Dioxane (IX) 27 Kinetic Runs in the pH Stat (X) 27 Kinetic Runs Via Ultraviolet Absorption (XI) 29 BIBLIOGRAPHY 30 APPENDIX 34 LIST OF FIGURES Figure Page 1 Pseudo -First -Order Kinetics of the Disappearance of Hydroxide Ion 34 2 Pseudo -First -Order Kinetics of the Disappearance of Hydroxide Ion in the Presence of Salts 35 3 Pseudo- First -Order Kinetics of the Rate of Disap- pearance of Thiolsulfinate. 36 4 The Dependence of Rate on the Apparent pH 37 5 Recorder Pen Assembly for Pipette Control 38 6 Pipette Control Module 39 7 pH Stat Reaction Vessel 40 LIST OF TABLES Table Page 1 Alkaline Hydrolysis of Phenylbenzenethiolsulfinate 7 2 Pseudo -First -Order Hydrolyses of Phenylbenzene- thiolsulfinate at Various Apparent pHs S 3 Alkaline Hydrolysis of Phenylbenzenethiolsulfinate in the Presence of Various Nucleophiles at Various Ionic Strengths 9 4 Kinetic Studies of the Rate of Disappearance of Phenylbenzenethiolsulfinate at Various Apparent pHs in Several Buffer Solutions 11 5 Comparison of pH Stat Data with Cary Model L -15 Absorption Data for the Alkaline Hydrolysis of Phenylbenzenethiolsulfinate 13 STUDIES ON THE MECHANISM OF THE ALKALINE HYDROLYSIS OF THIOLSULFINATES INTRODUCTION Thiolsulfinates, compounds of the formula RS(0)SR, appear in the literature as early as 1912 when they were reported by Zincke and co- workers (29, 30, 31, 32, 33). This group referred to them both as "sulfenic anhydrides" (31), when they were prepared by hydrolysis of an aromatic sulfenyl chloride, and as "sulfur oxides" (29, 30, 32, 33) when prepared by treatment of an aromatic sulfenyl chloride with sodium hydroxide in benzene. Zincke's structural as- signments were made on the basis of elemental composition, the fact that the compounds yielded the starting aromatic sulfenyl chlor- ides when treated with concentrated hydrochloric acid or phosphorus pentachloride in ether , and the fact that they "behaved as a n h y - drides" (33). By this Zincke meant that they dissolved in hot alkali or ammonia yielding strongly -blue solutions, although no salts were isolated. Recently, however, Oae and co- workers have raised doubts about at least two of these compounds (22) . In an attempt to repro- duce Zincke's synthesis of o- and p- nitrophenyl thiolsulfinate, Oae obtained a 1:1 mixture of the corresponding thiolsulfonate and disul- fide. 2 At about the same time as Zincke's work, Fries reported the synthesis of 2- anthraquinonesulfenic anhydride. He had previously prepared the free 1- anthraquinone sulfenic acid and its sodium salt by hydrolysis of the sulfenyl chloride (11), but attempts to make the 2 -acid resulted only in the formation of the anhydride (12). About 25 years later, Small and co- workers showed that the compound obtained from perbenzoic acid oxidation of allyl disulfide was identical to the natural antibacterial isolated from garlic (24). This group then prepared eight thiolsulfinates by peracid oxidation of their corresponding disulfides. All displayed antibacterial and anti - fungal action, which they surmised was due to reaction of the thiol- sulfinate with the -SH groups of the essential cell protein. Jacini and Lauria (15) also prepared several thiolsulfinates by oxidation of some unsymmetric disulfides with perbenzoic acid, which they called disulfide monoxides. In 1950, Burawoy and Turner (8) prepared several aromatic azothiolsulfinates. Then, in 1954, Backer and Kloosterziel noted that two different isomers of a thiol- sulfinate were obtained from the reaction (Equation 1) ArSOCI + Ar'SH Ar (Eq. 1) > it O if the aromatic groups in the starting compounds were reversed (1). In 1957, Vinkler and Klivenyi (25) prepared a series of compounds of the general formula RS(0)SR by the three methods reported in pre- vious literature (Equations 2, 3, 4). 3 ArSH + ArSOC1 --j ArrAr + HC1 (Eq. 2) 2ArSC1 + H20 - ArrAr + 2HC1 (Eq. 3) ArSSAr (0) ArSAr ö (Eq. 4) O A product study of the above reactions showed the compounds ob- tained to be the same in each case. Further evidence from infrared. spectra has borne out these conclusions (18, p. 248; 13, 14). Nearly all the workers that have reported on thiolsulfinates have noted their instability in aqueous media. Two reactions, closely related in that they involve the thiolsulfinate as the only or- ganic entity, have been reported. One is the disproportionation in acid solution (Equation 5). 2RSASR RSO2SR + RSSR (Eq. 5) This reaction has been studied both in this laboratory* and by sev- eral other groups (27) and will not be discussed further here. The other aqueous reaction =_s the alkaline hydrolysis (Equation 6). 2RSSR ----> 2RSO2 + RSSR (Eq. , O Zincke reported this reaction 50 years ago (33). He noted (31) that het alkaline solutions of nitro -substituted aromatic thiolsulfind ates developed intense colors which disappeared to yield the cor- responding disulfides and sulfinic acids, and he used this reaction as a qualitative test for the presence of thiolsulfinates (29, 31, 32, 33,. Cliff Venier; unpublished results. 4 Fries (11) also reported a colored solution occurred when 2- anthra- quinonethiolsulfinate was dissolved in a potassium hydroxide solution, but he did not attempt to isolate the resultant products. Small and co- workers (24) reported their alkyl thiolsulfinates were unstable to- ward alkalies, the products they isolated being the disulfides and sulfur dioxide. Savage and co- workers (23) have prepared and char- acterized the "S- monoxide" of cystine, cystamine and homocystine. They observed that the rate of hydrolysis of these thiolsulfinates in- creased with an increase in pH. Burawoy (8, 9) reported that al- kaline hydrolysis occurred with his aromatic -azo- thiolsulfinates, and he isolated the corresponding disulfide and sulfinic acid as pro- ducts. Burawoy stated that the alkaline hydrolysis of sulfenic acids and sulfenyl halides as well as thiolsulfinates produced essentially the same products in each case (Equation 6) (18, p. 283), which Kharasch postulated was due to the reaction sequence (Equation 7), RSC1. --> RSOH > RSSR 7PRSSR + RSO2 - (Eq. 7) in alkaline media (18, p. 392). Kharasch had previously reported that the brightly- colored solutions obtained in aqueous alkali were due to the presence of the sulfenic acid anion (6, 17), suggesting that the second step in Equation 7 above is reversible. Kharasch (18, p. 392) was led, therefore, to postulate the equilibrium (Equa- tion 8) 5 2RSOH RSSR + H2O RSH + RSO2H (Eq. 8) O as the mechanism for the so- called disproportionation cf sulfenic acids to mercaptans and sulfinic acids. This would imply that all reactions involving sulfenic acids will also involve thiolsulfinates as intermediates at one stage or another. Vinkler and Klivenyi (26, 27) do not agree that the sulfenic acid is in equilibrium with the thiolsulfinate, but they do postulate that the thiolsulfinate is the first hydrolysis product of sulfenic acid hy- drolysis regardless of the structure of the compounds involved. They found the disulfide and the sulfinic acid to be the major products of the reaction in all cases but found some sulfonate to be formed in the phenyl sulfenic acid hydrolysis. This product was believed to be produced via reduction of the thiolsulfinate by the initially- produced sulfinic acid (Equation 9). ArSSAr + ArSO2 - > ArSSAr + ArSO3- (Eq. 9) O Thiolsulfinates are involved in a variety of chemical and bio- chemical problems. They are known to exhibit both antibacterial and antifungal activity (24), to be effective retarders in free radical chain processes such as autoxidation and vinyl addition polymeriza- tion (2, 31) , and to be involved in both the oxidation and S -S bond scission of disulfides (2), to mention only a few. 6 This thesis, then, was undertaken in an effort to determine the mechanism of the basic hydrolysis of thiolsulfinates. Our approach was to utilize a pH stat to follow the hydrolysis kinetics at various pHs. Later, we attempted to follow the reaction at these same pHs via the thiolsulfinate ultraviolet absorption for comparison. Finally, a product study was undertaken to determine the reaction's stoi- chiometry. 7 RESULTS I) Product Study A sample of phenylbenzenethiolsulfinate was hydrolyzed in basic solution and the resulting products analyzed as described in the Ex- perimental. The amount of base required to hydrolyze a given a- mount of the phenylbenzenethiolsulfinate was determined from the infinity titers obtained from several hydrolyses conducted in the pH stat. The results of these experiments are listed in Table 1. Table 1. Alkaline Hydrolysis of Phenylbenzenethiolsulfinate. Initial OSSCD 3. 3 X 10-3 moles O Moles OH consumed /moles AO 0. 66 ± 0. 05 moles CßSO2 produced 2.19 ±0.02X 10 -3 moles OSSQ produced 2.23 + 0.02 X 10 -3 moles Percent O s recovered 99.8 ±0.11% These data establish that the overall stoichiometry for the reaction is 30 o + 2OH - ) 20SO2 + 20SSQJ + H2O (Eq. 1) Other products, previously reported in the literature (16, 24), were not observed in this case. 8 II) Kinetic Studies Initially, it was felt that the rate law foi the alkaline hydrolysis of thiolsulfinate would be first order in both hydroxide ion and thiol- sulfinate. The pH stat was constructed in order that the reaction could be studied under pseudo- first -order conditions. A series of these pseudo- first -order hydrolyses were conducted at various ap- parent pHs as listed in Table 2. Table 2. Pseudo -First -Order Hydrolyses of Phenylbenzenethiol- sulfinate at Various Apparent pHs. O Apparent -3 -1 -4 -1 -4 ((ÁSOSdJ)Ox 10 M kisec x 10-4 x 10 ki/kf pH 8.0 1.37 2.5 1,3 1.9 8.7 1.37 4.8 3.2 1.5 9.0 1.37 7.2 4.3 1.7 9.7 1.37 31..2 15.6 2.0 ID All runs conducted at 16. 6° C. The first -order plots of hydroxide consumption versus time for these reactions show that the hydrolysis does not follow strict first. - order kinetics (Figure 1, Appendix). Instead, the reaction plot curves and seems to have an initial first (k.) and a -order rate i slower final first -order rate (kf). The ratios of these rates vary consider- ably. Further, the dependence of either rate on the apparent pH does not follow a simple first -order dependence on hydroxide concentra- tion. 9 In other studies in this laboratory"'', some systems analogous to the alkaline hydrolysis of thiolsulfinates have shown significant catalysis of reaction rate by such nucleophiles as chloride ion. A study was undertaken to determine if any of the peculiar dependence on pH observed was actually the result of contamination of the rea- gents by small amounts of chloride ion and marked catalysis of the reaction by this chloride ion. Table 3 lists the results of these ex- periments. Table 3. Alkaline Hydrolysis of Phenylbenzenethiolsulfinate in the Presence of Various Nucleophiles at Various Ionic Strengths. Ionic Salt kisec -1 x 10-4 kfsec -1 x 10-4 ki /kf Strength Added 9.9 7.6 1.3 0.005 Na2SO4 9.2 5.1 1.8 0.005 NaNO3 9.3 6. 1 1.5 0.010 NaNO3 8.4 5.2 1.6 0.050 NaNO3 9.1 5.2 1.7 0.001 KC1 11.0 8.6 1.3 0.005 KC1 8.4 6.2 1.4 0.015 KC1 9.5 5.0 1.9 0.020 KC1 7.6 4.7 1.6 0.035 KC1 11.0 5.5 2.0 0.050 KC1 All runs conducted at pH 9.0 and temperature 16. 6° C. (4SS4)p= 1. 37 x 10 -3M in each run. Guaraldi, Giancarlo; unpublished results. 10 As the data in Table 3 indicate, the reaction is not significantly catalyzed by chloride ion, and the rate in the presence of chloride is larger than the rate at zero ionic strength by an amount which can be explained as being due solely to simple salt effects. Two other points are worth noting. First, in some of the reac- tions run in the presence of salts, the initial and final rates seem to coalesce, as reflected in the ratio ki /kf. Second, this lessening of the curvature of the first -order plots makes the determination of the initial and final rates even more difficult (Figure 2, Appendix). At this time, we felt that another method of following the reac- tion might provide us with more insight into the reaction mechanism. Since the hydroxide consumption shows such complexity, we decided to follow the rate of disappearance of the thiolsulfinate in a buffered solution. Advantage was taken of the large ultraviolet absorption peak for the phenylbenzenethiolsulfinate at 296 milk. A study was under- taken, utilizing the Cary recording spectrophotometer. The reac- tion was followed at various apparent pHs to determine the pH dependency of the reaction. Three different buffer systems were employed in their usable ranges so that specific anion effects could be detected if present. Examination of the results of this study (Table 4) indicates several things. First, the kinetics of the disap- pearance of thiolsulfinate, like those of the hydroxide ion, do not (4) Table U All S )0 4. runs 4.5 3.0 3.0 3.0 N2.0 2.0 N 2.0 N 3.0 W rA 2.0 N 2.0 N 2.0 N 2.0 N 2.0 N 3.0 3.0 3.0 2.0 3.0 W W W W W W N W x O O O O lTi O O O O O O O O O O O O Apparent Kinetic O 10-4M conducted I Studies pHs Apparent at pH 9.5 9.5 9.3 9.3 9.2 9.0 9.0 9.0 9.0 9.0 9.0 9.0 9.0 8.7 8.7 8.5 8.3 8.1 in 20. of Several 0°C. the Rate kisec-1 Buffer of 16 17 11.5 4.6 4.2 0.33 0.14 6.9 2.6 2.4 2.5 3.9 5.6 3.9 3.7 5.1 1.6 1.2 Disappearance x Solutions. 10 -3 kfsec 10 of 9.6 5.3 3.6 3.1 2.1 2.4 2.4 0.32 0.47 0.37 0.23 0.097 1.5 1.6 1.6 1.7 1.7 ,r; -1 Phenylbenzenethiolfulfinate x 10 -3 ki 1.6 I-. 1.8 I-" 2.2 N 1.9 r-' 1.5 r-' 1.7 r/.-.1.5 1.6 r-' 1.9 I-" 2.3 N 1.8 2.3 N N2.2 LO -' 3.4 W W)--'3.2 1.5 1.5 h-. - /kf ...... Q` Co EN) .0 Ui 1 ui as .D W OD W N) 0 1A N ln lTI 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 05M 10M 10M 15M 10M 10M 10M 10M 10M 10M 15M 15M 10M 10M 10M 10M 10M 10M Buffer Borate Borate Borate Borate Borate Borate Glycine Glycine Glycine Glycine Borate Glycine Glycine Borate Borate Carbonate Carbonate Carbonate at Various 12 follow strict first -order kinetics but seem also to exhibit initial and final rates (Figure 3, Appendix). Second, tae dependence of ~both these reaction rates on the apparent pH is not a simple one, being seemingly less than first order below an apparent pH 8. 7 and seem- ingly greater than first order above (Figure 4, Appendix). Third, the buffer used alters both reaction rates to only a small degree, probably due to ionic strength differences similar to those seen in the hydroxide consumption studies. A final study was then conducted to compare the hydrolysis kinetics observed in the pH stat with those observed spectrophoto- metrically. The determinations were conducted in the pH stat and the Cary L -15 under conditions as closely identical as possible. The results of these experiments are contained in Table 5. Table 5. Comparison of pH Stat Data with Cary Model L -15 Absorption Data for the Alkaline Hy- drolysis of Phenylbenzenethiolsulfinate. ** ' Apparent (CPSS)0 kiOH- sec-1sec kiAbs -1 kfOH sec-1 kfAbssec-1 kiOH _kiAbs kfOH kfAbs pH O 8.1 2.0 x 10 2,1 x10-4 1.4 x 10-4 1.5 x 10-4 1.3 x 10 0.7 x 10 0.2 x 10 -4 -4 8.3 3.0 x 10-4 3.3 x 10-4 4.2 x 10-4 2.5 x 10-4 1.4 x 10-4 -0.9 x 10 1.1 x 10 4 4 4 4 4 8.3 4.5 x 10-4 3.6 x 10-4 2.4 x 2.8 x10-4 1.6 x 10 1.2 x10-4 1.2 x 10-4 8.5 2.0 x 10-4 6.5 x 10-4 4.5 x 10-4 5.6 x 10-4 3.2 x 10-4 2.0 x 10 2.4x10-4 4 4 4 8.5 3.0x10-4 5.8 x 10 7.7 x 10 4.6x10-4 3.5x10-4 -1.9 x10-4 1.2 x 10 4 4 4 4 4 4 4 8.5 4.5 x 10 4.9 x 10 4.3 x 10 4.1 x 10 3.1 x 10 0.6 x 10 1.0 x10 4 4 8.7 2.0 x10-4 6.6 x 10 8.7x10-4 5.7x10-4 5.5x10-4 -2.1 x 10 0.2x10-4 8.7 3.0 x 10-4 5.1 x 10-4 6.9 x 10-4 4,.8 x 10-4 3.8 x 10-4 -1.8 x 10-4 1.0 x 10-4 8.7 4.5 x 10-4 6.1 x 10-4 6.2 x 10-4 5.5 x 10-4 4.2 x 10-4 -0.1 x 10-4 1.3 x 0-4 4 4 9.0 2.0 x10-4 1.2x10-3 3.6 x 10 3 7.7 x 10 2.4x10-4 -2.4 x 10 5.3 x 10-4 All runs conducted at 20 °C. All absorbancy runs made in 0.1 Borate Buffer /Dioxane at /1 '` k10H and kfOH = rates determined in the pH Stat. *** kiAbs and kfAbs = rates determined in the Cary Model L -15 w 14 DISCUSSION The results of the product study indicate that the alkaline hy- drolysis of thiolsulfinates takes the course indicated by Equation 1. 34 r + 2OH- ----> 20S02- + 21)SS4 + H2O (Eq. 1) b Any mechanism proposed for this reaction must account for this stoichiometry. In a reaction between hydroxide ion and thiolsulfinate, the hy- droxide ion may attack either the sulfinyl sulfur (Equation 2) or the sulfenyl sulfur (Equation 3). CpSSC + OH- -3 (PSO2H + CpSO- (Eq. 2) O OSS(1) + OH -II, (I:1SO + CI)SOH (Eq. 3) Our experimental work shows that neither the rate of hydroxide ion consumption nor the rate of disappearance of thiolsulfinate follow simple first -order kinetics. Instead, the first -order plots exhibit a curvature suggesting that both reactions are initially fast, slowing to a final lesser rate as the reaction proceeds (Figures 1, 2, and 3, Appendix). This behavior eliminates the possibility of a simple bi- molecular rate -determining step for this reaction, such as Equation 2 or 3 above. Some possible ways to account for the curvature in the first -order plots can be suggested. 15 First, if the major point of attack of the hydroxide ion is taken to be the sulfinyl sulfur of the thiolsulfinate, the intiai products would be sulfinic acid and mercaptide ion (Equation 2). The sulfinic acid would be immediately converted to the sulfinate (4SO2 -) in the alkaline media used for our studies (Equation 4). OSO2H + OH - > O-SO2 + H2O (Eq. 4) The mercaptide ion should react rapidly with another molecule of the thiolsulfinate, producing the disulfide and sulfenate ion (Equation 5). cl)S- + clScp --j cpSSp + CpSO (Eq. 5) Due to the stability of the disulfide, this reaction would not be ex- pected to be significantly reversible. If the sulfenate ion does not react further to yield products but disappears only by regenerating the thiolsulfinate (Equation 6), and 2OSO- + H2O > cSSO + 2OH- (Eq. 6) if this reaction is relatively slow, then, since Equation 6 regenerates both thiolsulfinate and hydroxide ion, the initial rates of disappear- ance of hydroxide and thiolsulfinate will be faster than those observed later in the reaction after the sulfenate ion concentration has built up to its steady-state value. The same type of kinetic behavior could also be obtained if attack of hydroxide ion occurs at both sulfinyl and at sulfenyl sulfur, but with the attack at sulfenyl sulfur being reversible (Equation. 7), and 16 with only the attack at sulfinyl sulfur leading to the final products. (1:0SCp + OH- CASO + 6SOH (Eq. 7) O Thus, if the attack at sulfenyl sulfur occurs somewhat more readily, and if the equilibrium concentration of sulfenate ion is not too small, one can also get a situation where the initial rate of disappearance of the thiolsulfinate and hydroxide ion is faster than their rate of disap- pearance after the sulfenate ion concentration has reached its equili- brium value. Although schemes of this type can account for the curvature of the first -order plots, they cannot account for the peculiar dependence of the observed rates on the apparent pH (Figure 4, Appendix). This peculiar dependence, however, may be an artifact resulting from failure of the glass electrode to measure meaningful pHs in our sys- tem. Kolthoff (20) has observed that glass electrode responses are unstable in solutions of sulfuric acid in acetonitrile, but stable in perchloric acid - acetonitrile solutions. Bates (4, p. 324) observed that partially aqueous solutions of ethanol, methanol and acetone all render glass electrodes unreliable at alkaline pHs when the organic material exceeds 500 of the total volume. Although we knew that the apparent pHs measured in our system were not exact, we initially felt they would differ from the true pH by a small, consistent factor that could be determined later. However, this peculiar dependence 17 of the observed rates on the apparent pH suggests that the presence of alkaline 25% dioxane has seriously altered the shape of the glass electrode's potentiometric curve, causing the apparent pHs to differ from the true in a complex manner over the range of values observed. A further peculiarity in the data which may possibly be attributed to faulty electrode response is the lack of any apparent trend in the differences between the initial rates of disappearance of the thiolsul- finate and hydroxide ion over the range of apparent pHs observed. The schemes above which account for the peculiar first -order plots predict definite trends in these initial rate differences. The direction of these trends depend primarily upon the existence and position of an equilibrium between sulfenic acid and its anion (Equation 8). (i)SOH + OH (1)SO- + H2O (Eq. 8) The observed differences, however, appear to be random. The various degrees of slowing of the observed rates of disappearance of hydroxide ion as reflected in the ratios ki /kf suggest that the glass electrode may be unable to follow rapidly changes in the pH of the solution being studied. This inability may alter the observed initial rates of hydroxide ion disappearance, thus varying the differences in initial rates of disappearance of hydroxide ion and thiolsulfinate in an unpredictable fashion. 18 The best solution to the many problems encountered in the course of this work would seem to be to study this reaction in purely aqueous media. This would permit the reliable use of the glass electrode in a system closely analogous to that utilized here. To this end, we have attempted the synthesis of the previously unreported p- carboxy- phenyl p- carboxybenzenethiolsulfinate which we feel would be soluble in water to the extent necessary for study. We first prepared the corresponding p- carboxyphenyl disulfide which we hoped to oxidize to the corresponding thiolsulfinate. Several attempts at this oxida- tion were unsuccessful, usually producing the thiolsulfonate in ap- proximately 50% yield. A few conclusions can be drawn from this work. First, the al- kaline hydrolysis of thiolsulfinate is an extremely facile reaction. Its rate is faster at one pH unit above neutrality than is the analogous acid disproportionation at eight pH units below neutral. Second, the reaction does not seem to be significantly catalyzed by nucleophiles such as chloride ion. Third, the reaction appears to be extremely complex, requiring study in several systems for elucidation of its mechanism. Finally, the characteristics of glass electrodes in semi - aqueous media require further study before they can be used to determine and follow rapidly - changing hydrogen ion concentration in mechanistic studies. 19 EXPERIMENTAL I) pH Stat The basic component of the pH stat is the Heath (Benton Harbor, Michigan) Model EUW -301 pH recording electrometer. This instru- ment indicates the pH of a solution over a preselected range of 10, 5, 2 or 1 pH units on ten -inch -wide recorder paper. The recorder pen is mounted in a carriage which slides on a steel rod and is actuated by a nylon cord attached to a servo motor drive system. A wedge - shaped aluminum block (Figure 5, Appendix) was machined to fit securely onto the top of the pen carriage. A one - inch -square, twelve- inch -long steel rod (Figure 5, Appendix) was mounted on brackets and clamped rigidly to the recorder housing directly above the path traveled by the block and pen carriage. A steel collar was machined which can be clamped onto the steel rod by means of a knurled bolt in the collar. A small, normally -open, s. p. s. t. micro - switch was mounted on the bottom of the collar such that the alu- minum block on the pen carriage actuates the switch without impeding the pen carriage movement. In this way, when the pH of the solution being measured by the electrometer reaches a predetermined value, the micro - switch is closed. A reaction which consumes acid or base can be held at a con- stant pH by allowing the micro - switch to control addition of acid or 20 base to the reaction solution. This was accomplished in our system (Figure 6, Appendix) as follows: When the switch S1 is turned to ON, one side of the 110 -120V, 60 -cycle line voltage is applied to the micro - switch S2 on the recorder. If the interrupter bypass switch S3 is in the bypass position, the line voltage is applied directly across a motor (pipette drive motor M1) which drives a Manostat Digi -Pet (Manostat Corporation, New York, New York) 1 ml digital pipette. This pipette is filled with acid or base solution, and its outlet tip is immersed below the surface of the reaction solution being followed. When the acid or base addition rate is too slow to allow optimal pH control, a stronger titrant solution is employed. When the addition rate is too fast, the interrupter bypass switch S3 is placed in the "interrupt" position. In this mode, the voltage to the pipette drive motor passes first through a cam -operated switch S4. The cam is driven by the interrupter motor M1 at 10 r. p. m. and has a 50 -100% adjustable lobe which actuates S4. By selecting either the normally -open or normally- closed side of S4 to control the motor, using switch S5 and properly adjusting the cam lobe, one can obtain any rate of titrant addition between 0 and 100% of the maximum pipetting rate. The pipette drive motor is also fitted with a protective clutch which is actuated by switch S6. The pipette can be refilled by re- versing the polarity of the pipette drive motor field windings with 21 the titrate - refill switch S7. The entire unit is isolated from the line by a five -amp fuse F1. With the electrometer's recorder span set at one pH unit (a change of 1 pH unit equals 10 inches of pen travel), and the control unit and titrant adjusted to optimal conditions, we have obtained con- trol of the pH of kinetic reaction solutions to ± 0.02 apparent pH units. This compares very well with results obtained with more elaborate units based on the Heathkit pH Recording Electrometer reported in the literature (21) . All reactions carried out with the pH stat were run in the re- action vessel illustrated in Figure 6 of the Appendix. Temperature control was maintained by circulating water from a regulated water bath through the vessel's jacket. The reaction solution was kept free of CO2 by bubbling dry nitrogen through it during the reaction and for 10 minutes before sample addition. CO2 was removed from the nitrogen by passing it through ascarite before admitting :.t to i.":.. reaction vessel. All samples were added as solutions of the thio- sulfina.te in dioxane through the sample addition part in the vessel. The port was kept stoppered before and after sample addition to prevent CO2 entry. A combination glass electrode- calomel reference electrode (Thomas Company, Philadelphia, Pennsylvania) was used in the work. Initially, the Thomas Model -60 electrode was used. Th 22 electrode has an essentially linear response over a pH range of 0 to 14. Two problems were encountered with this electrode. First, at indicated pHs greater than pH 9. 0, in both 5% and 25% dioxane solu- tions, the noise response of the system exceeded an indicated 0.15 pH units. Second, a slow drifting of the indicated pH of as much as 2.0 pH units was encountered over a 6 -hour period. A thorough check of the electrometer circuitry revealed the trouble to be with the electrode. Extensive grounding of all metal objects near the unit had no effect on either problem. Next, a Thomas L -15 electrode was substituted. This elec- trode has a linear response over the range of pH 0 -11 and bears a much heavier glass membrane. With this electrode, the noise re- sponse was eliminated, but the slow drifting, while reduced, was still present. Consultation with the A. R. Thomas Company'` sug- gested that the problem was due to the aging phenomena of the glass membrane in the partially dioxane solution. Also, Bates (4, p. 324) has shown that nonaqueous solvents alter the resistance of glass membranes in electrodes in varying degrees. The L -15 electrode was then rinsed in 2% hydrofluoric acid followed by washing first with distilled water and then with absolute ethanol. It was then Mr. Fred Allgrun; personal communication. 23 allowed to equilibrate 18 hours with 25% dioxane at a pH of 1.0. This was done to properly solvate the glass membrane with the me dia in which it would be used. Further testing showed the drift of the indi- cated pH to be eliminated. When not in use, the electrode was stored in 25% dioxane at a pH of 1.0 to 2.0. Before and after each run, the electrode was washed with absolute ethanol and air - dried. With these precautions, a stable reading was obtained in the kinetic apparatus after a 15- to 20-minute equilibration of the electrode in the reaction solution, and no further drifting was observed. However, examination of the data obtained with this electrode seems to imply that the pHs indicated by it, and the speed of its response to changes in the acidity or basicity of the solutions measured have been drastically altered. It is be- lieved that the dioxane affects the resistance of the glass membrane, impairing the electrode's ability to measure a meaningful pH, o:~ to respond quantitatively to changes in the hydrogen ion concentration of the media. Kolthoff (20) has observed very poor reproducability of electrode potentials in perchloric acid - acetonitrile solutions. When he substituted sulfuric acid, the reproducability improved to a usable degree. Bates (4, p. 326) has reported that the effect of methanol on glass electrodes, especially in alkaline solutions, renders their out- put unusable for pH determinations. We feel, therefore, that much 24 more research must be accomplished before this system can be used effectively in other than purely aqueous media. II) Preparation of p- Carboxyphenyl Disulfide In order to use the pH stat to examine the alkaline hydrolysis of a thiolsulfinate in purely aqueous media, an attempt was made to synthesize a water - soluble thiolsulfinate. We felt that the p- carboxv compound would be soluble in water to a large enough extent to be useful. Since many thiolsulfinates have been synthesized by oxida- tion of the corresponding disulfide (23, 24), we first prepared p- carboxyphenyl disulfide by the method of Allen and McKay (6, p. 580). This method involves the diazotization of p- aminobenzoic acid in acidic solution at 0 °C. The diazo solution was then carefully added to a cold, basic solution of sodium disulfide. After nitrogen evolu- tion had ceased, the p- carboxyphenyl disulfide was precipitated by the addition of hydrochloric acid. The precipitate was dissolved in dioxane, treated hot with decolorizing charcoal and recrystallized from a dioxane -water solution. The product, a brown powder, de- composed over 300 °C in agreement with the literature (5, p. 109). An infrared spectrum indicated no contaminants were present. Ili) Attempted Preparation of p- Carboxyphenyl p-Carboxybenzene- thiolsulfinate. Oxidation of the p- carboxyphenyl disulfide to the corresponding 25 thiolsulfinate was first attempted with peracetic acid. A sample, 1.2 g. (3.02 X 10 -3, moles) of the disulfide was dissolved in 120 ml. of glacial acetic acid, and 0.45 ml. (4.0 m moles) of 30% hydrogen per- oxide was added dropwise; the mixture was then stirred overnight at room temperature. Infrared spectra of the material isolated after the reaction showed it to be mainly the starting disulfide and some thiolsulfonate. A second attempt at oxidation was made by mixing 0.612 g. (2 X 10-3 moles) with 0.450 g. (2.1 X 10-3 moles) of sodium per- iodate in 50 ml. of distilled water. The mixture was stirred over- night at room temperature and the products isolated. Infrared spectra showed the material from the reaction to be a mixture of the original disulfide and the corresponding thiolsulfonate. Finally, an attempt to oxidize 0.918 g. (3.0 X 10-3 moles) of the disulfide was made by dissolving it in 5 ml. of acetone and adding 0,231 g. (3.1 X 10 -3 moles) of hydrogen peroxide in 5 ml. of ac ^ <.r., Isolation of the resulting precipitate showed it to be mainly the cc:: responding thiolsulfonate. IV) Preparation of horny 1 b enzenethiolsulfinate Phenyl benzenerhiolsulfinatc was prepared by reacting benzene - s .lfinyl chloride with thiophenol in the presence of pyridine in ether solution, by the method of Backer and Kloosterziel (1). The 26 compound obtained melted cleanly at 69.5 - 71 °C (Literature value: (1) 69- 70 °C). V) Preparation of Benzenesulfinyl Chloride Benzenesulfinyl chloride was prepared by the method of von Braun and Kaiser (7). This involved treatment of benzene sulfinic acid with thionyl chloride. The resultant yellow viscous liquid was used to prepare the phenylbenzenethiolsulfinate without further puri- fication. VI) Thiophenol Thiophenol (Aldrich Chemical Company) was used without puri- fication in the preparation of the phenylbenzenethiolsulfinate. VII) Product Study Phenylbenzenethiolsulfinate, 0.78 g. (3.33 X 10-3 moles) , was dissolved in 250 ml. of 0.1N borate buffer made 25% dioxane and ad- justed to pH 8.7. This solution was allowed to stand overnight at room temperature. Ffity milliliters of the solution was removed and titrated for sulfinic acid content with 0.2N NaNO2 (19). The re- maining 200 ml. solution was made up to 1 liter with water and ex- tracted three times with 200 ml. portions of ethyl ether. The ether was dried over anhydrous magnesium sulfate, then removed on a rotary evaporator from a tared flask. The resulting dry phenyl 27 disulfide was weighed. The product melted cleanly at 60- 61.5 °C (Literature value: (10, p. 1092) 66 °C). An infrared spectra. ino - cated no contaminants were obtained. VIII) Preparation of Buffer Solutions Solutions of either reagent grade glycine or reagent grade boric acid of 0.1M were made 25% in dioxane. The resulting solutions were then adjusted to the desired pH, as measured by the electro- meter, with 10 N NaOH. IX) Purification of Dioxane Reagent grade dioxane was purified by the method of Wiberg (28, p. 245). This consists of treating the dioxane first with aqueous hydrochloric acid and refluxing 12 hours. The solution was then made basic with potassium hydroxide and separated. The dioxane layer was then dried, refluxed over sodium metal for 12 hours, and distilled out of contact with air. X) Kinetic Runs in the pH Stat The alkaline hydrolysis of thiolsulfinates consumes hydroxide. One method of following the kinetics of this reaction is, then, to follow the rate of hydroxide consumption as a function of time. This was done in our system by conducting the reaction in the pH stat. The volume of a solution of known base concentration required to 28 maintain the reaction at a preset pH value was obtained from the Digi -Pet at timed intervals beginning with the addition of the the - - sulfinate to the reaction solution. Initially, the thiolsulfinate was added to the reaction solution as the solid crystalline material. However, it was discovered that the rate of its dissolution was slower than the subsequent hydrolysis. Therefore, all subsequent samples were added as solutions of the thiolsulfinate in 100µl. of purified dioxane, utilizing a 100,c,(1 pipette attached to a syringe. A stopwatch was started simultaneously with sample addition to time the Digi -Pet readings. The final infinity volume was obtained after not less than nine half -lives (obtained from the time needed for the solution to consume 50% of the volume of base required for com- pletion of the reaction as predicted by the reaction's stoichiometry) had elapsed. In all cases, the solvent was 25% dioxane in distilled water. Except in those cases where salt effects were being studied, the ionic strength of the solution was maintained at 0.10 by the addition of potassium chloride. The pH of the reaction solution was adjusted by addition of a small amount of 10 N NaOH after the desired tem- perature had been reached. Degassing the solution and equilibration of the electrode were accomplished during the time required for the 29 temperature to come to the desired value (usually 20 to 30 minutes). XI) Kinetic Runs Via Ultraviolet Absorption Advantage was taken of the large ultraviolet absorption peak characteristic of thiolsulfinates in the range 290 -300 mµ. Buffer solutions were prepared which were identical in pH and dioxane con- tent (see Preparation of Buffer Solutions) to the corresponding runs made in the pH stat. The buffer solution, 2.9 ml. , was added to a glass- stoppered ultraviolet cell and degassed with dry, CO2 -free nitrogen gas. The cell was then stoppered and allowed to come to the preselected temperature of the Cary Model L -15 recording spec- trophotometer. This machine utilizes a variable temperature cir- culating water bath to maintain the cell at a constant temperature. After sufficient time had elapsed for temperature equilibration, the thiolsulfinate was added as a solution in 100µ1. of dioxane. The course of the reaction was recorded and followed on the Cary Model L -15. 30 BIBLIOGRAPHY 1. Backer, H. J. and H. Kloosterziel. Thiolsulfinic esters. Re- cueil des Travaux Chimiques des Pays -Bas 73:129 -139. 1954.. (Abstracted in Chemical Abstracts 49:3877. 1955) 2. Barnard, D. and E. J. Percy. Oxidation of organic sulfides. Part XI. The synthesis of specifically -labeled phenyl (35S) benzenethiolsulfinate and phenylbenzenethiol. (35S) sulfinate. Journal of the Chemical Society, 1962, p. 1667 -1671. 3. Barnard, D. and E. J. Percy. Oxidation of thiolsulfinates to thiolsulfonates. Chemistry and Industry, 1960, p. 1332 -1333. 4. Bates, Roger G. Determination of pH. Theory and practice. New York, Wiley, 1964. 435 p. 5. Beilstein's Handbuch der Organischen Chemie. 4th ed. Vol- ume 10, Supplement 2. Berlin, Springer, 1949. 951 p. 6. Blatt, A. H. (ed.) Organic syntheses. Collective Volume II. London, Wiley, 1943. 653 p. 7. von Braun, Julius and Wilhelm Kaiser. Zür Kenntnis der Sul - finsäuren. Berichte der Deutschen Chemischen Gesellschaft 56:549 -553. 1923. 8. Burawoy, A. and C. Turner. O- mercaptoazo - compounds. Part 1. The coupling of tetrazotized 2:2' diaminodiphenyldisul fide with 2- naphthol. Journal of the Chemical Society, 1950, p. 469-477. 9. Burawoy, A. and C Turner. 0-mercaptoazo-compounds. Part II. 1(1 -mercapto -2 naphthylazo) -2- naphthol. Journal of the Chemical Society, 1952, p. 1286 -1292. 10. Chemical Rubber Company. Handbook of chemistry and physics. 36th ed. Cleveland, 1954. 3173 p. 11. Fries, K. and G. Schurman. Anthraquinonylsulfur chlorides. Berichte der Deutschen Chemischen Gesellschaft 52 :2170 -2195. 1919. (Abstracted in Chemical Abstracts 14:2182. 1920) 31 12. Fries, K. 1- anthraquinonesulfenic acid. Berichte der Deut- schen Chemischen Gesellschaft 45:2965 -2973. 1912. (Ab- stracted in Chemical Abstracts 7:1006. 19 3) 13. Ghersetti, S. and G. Modena. Vibrational behavior of the SO- group in alkyl and aryl thiolsulfoxides. Spectrochimica Acta 19:1809 -1814. 1963. 14. Ghersetti, S. and G. Modena. Behavior of the fundamental S --) O vibrational frequency in some sulfoxides. Annali de Chimica 5:1083 -1092. 1963. (Abstracted in Chemical Abstracts 60:1236 -1237. 1964) 15. Jacini, Giovanni and Franco Lauria. Organic asymmetric disul- fides and their monoxides. Gazzetta Chimica Italiana 80:762- 767. 1950. (Abstracted in Chemical Abstracts 46:4499. 1952) 16. Kharasch, Norman, Sylvester J. Potempa and Herbert L. Wehrmeister. The sulfenic acids and their derivatives. Chemical Reviews 39:269-332. 1946. 17. Kharasch, Norman, William King and Thomas C. Bruce. De- rivatives of sulfenic acids XVII. The hydrolysis of 2,4-dinitro- benzenesulfenyl chloride. Journal of the American Chemical Society 77:931 -934. 1955. 18. Kharasch, Norman (ed.) Organic sulfur compounds. Vol. I. New York, Pergamon Press, 1961. 624 p. 19. Kice, John L. and Kerry Bowers. Mechanism of the reaction of sulfinic acids II. The reaction of p -tolyl disulfide with p- toluene sulfinic acid. Journal of the American Chemical Society 84:2384 -2389. 1962. 20. Kolthoff, I. M. and M K. Chantooni, Jr. Calibration of the glass electrode in acetonitrile. Shape of the potentiometric curves. Dissociation constant of picric acid. Journal of the American Chemical Society 87:4428 -4433. 1965. 21. Malmstadt, H. V. and E. H. Piepmeir. pH stat with digital readout for quantitative chemical determinations. Analytical Chemistry 37(1):34 -44. 1965. 32 22. Oae, Shigeru and Shunichi Kawamura. The structure of Zincke's o- and p- nitrobenzenesulfenic anhydrides. Annual Report of the Radiation Center of Osaka Prefecture 2:120 -l?5. 1961. (Abstracted in Chemical Abstracts 58:5356. 196) 23. Savige, W. E. et al. The S- monoxides of cystine, cystamine, and homocystine. Tetrahedron Letters, 1964, 3289 -3293. 24. Small, Lavern D. , John Hays Bailey and Chester J. Cavallito. Alkyl thiolsulfinates. Journal of the American Chemical Society 69:1710-1713. 1947. 25. Vinckler, Elemer and Ferenc Klivenyi. Confirmation of the identity of the alleged sulfenic anhydrides and thiolsulfinates. Acta Chimica Academiae Scientiarum Hungaricae 11:15 -22. 1957. (Abstracted in Chemical Abstracts 52:6242. 1958) 26. Vinckler, Elemer and Ferenc Klivenyi. The mechanism of hydrolysis for aromatic sulfenyl chlorides. New data on the chemistry of esters of thiolsulfinic acids. Magyar Kemiai Folyoirat 65:451 -452. 1959. (Abstracted in Chemical Ab- stracts 54 :17304. 1960) 27. Vinckler, Elemer and Ferenc Klivenyi. Uber den Mechanismus der Hydrolyse Aromatischen Sulfenylchloride. Ein Neuer Bei- trag zur Chemie der Thiolsulfinsaureester. Acta Chimica Aca- demiae Scientiarum Hungaricae 22:345 -358. 1960. 28. Wiberg, Kenneth G. Laboratory technique in organic chemis- try. New York, Wiley, 1960. 262 p. 29. Zincke, Th. and Johanna Baeumer. Sulfuraryl chlorides IV. 4- chloro -2- nitrochlorothiolbenzene (p- chloro -o- nitro -phenyl- sulfur chloride). Conversion into thiazine derivatives. Justus Leibig's Annalen der Chemie 416:86 -99. 1918. (Abstracted in Chemical Abstracts 13:575. 1919) 30. Zincke, Th. and K. Eismayer. 2- naphthylsulfur chlorides. Berichte der Deutschen Chemischen Gesellschaft 51:751-767. 1918. (Abstracted in Chemical Abstracts 13:449. 1919) 31. Zincke, Th. and Fr. Farr. O- nitrophenylsulfur chloride and its transformation products. Justus Leibig's Annalen der Chemie 391:57 -88. 1912. (Abstracted in Chemical Abstracts 6 :2937. 1912) 33 32. Zincke, Th. and S. Lenhardt. P- nitrophenylsulfur chloride and its rearrangement products. Justus Leibig's Annalen der Chemie 400:2 -27. 1913. (Abstracted in Chemical Abstracts 7 :3746. 1913) 33. Zincke, Th. and H. Rose. 3- nitrotolyl -4- sulfur chloride. Justus Leibig's Annalen der Chemie 406:103 -126. 1914. (Ab- stracted in Chemical Abstracts 9:54. 1915) APPENDIX 1. S 0. ,1 4.3 10 sec sP 10 b 0 20 0 40 e Z tie51 E do ,, r i, e a a.,,, Osae Ptilc° 915 Zo' í°l' e 4.5 ati .e \-6 b G 11Li Cß . aioLaQ,31.1i11 <0 b© ,\o 10- 05 o o y{` o` , N1 KG Qov `., G 4s ' ; ,`' ; O 9 8 7 6 5 3 2 I I 100 200 500 1000 1500 2000 2500 Time (Sec.) Figure 3. Pseudo - First- -Order Kinetics of the Rate of Disappearance of Thiol- sulfin.ate. (pH = 8.5 in glycine buffer; Temperature = 20.0° C; Initial Concentration = 3 x 10-4M.) 37 1.80 O 1.90 2.00 2.10 2.20 2.30 2.40 2.50 2.60 2.70 2.80 -4 x 2.90 m o 3.00 3.10 3.20 3.30 3.40 3.50 3.60 3,70 3.80 I 1 1 1 1 1 1 1 1 i [ I I 8 1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 pH Figure 4. The Dependence of Rate on the Apparent pH. { - Collar 0 Leads to Bracket Control and Clamp Unit -e-----E -- Mach in e d Micro Switch -17I 81 Block I - I Ì ------I -- - I I I I -ii E Pen Carriagegr -- Recorder Housing -- r- Steel Rod for II - Pen Carriage II I I I I II - II I I 1 I I I I I I I II 11 1 I I I 1 I I I 11 I--I---7 --I---I-- - I--- I- ----I-- - - II II I II I I 1 I I I I I I I H II I- - -1 - -I H ---I -I -I -1 - -I------1- - I 11 I l I I 1 I 1 1 I I 1 11 II I i 2 ii /+/L i 2 -2 2. 7/LYL 11 ]/Y+ Recorder Chart Paper II U U Figure 5. Recorder Pen Assembly for Pipette Control Titrate 5a 0.47 M1 120 V Refill z 60^, / / z n 6 / Off < Clutch Interrupt Ad stable 0-50% S2 S3 On -. S5 S Cam S4 Off 1 / 50-100% Pen - Carriage Block Figure 6. Pipette Control Module. 40 .._Gas Inlet Sample Addition CI ,) ilii Electrode , Port I I Pipette i Ì --±-ii Gas Outlet --= , Water Outlet Water T Inlet Figure 7. pH Stat Reaction Vessel.