Oxidation of Organic Compounds

---

16 Liquid-Phase Oxidation of Thiols to Disulfides

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

J. D . H O P T O N , C. J. S W A N , and D . L. T R I M M Department of Chemical Engineering and Chemical Technology, Imperial College, London, S.W.7, England

The oxidation of a series of alkyl and aryl thiols in aqueous alkaline solution has been studied in the presence of various metal ions. Quantitative amounts of disulfide were produced in all cases. The oxidation rate of thiols has been found to be affected by the geometric size and electron-directing properties of substituent groups in the organic chains of the thiols. The best three catalysts, when added as simple salts, have been found to be copper, cobalt, and nickel. The de­ pendence of the rates of oxidation on the concentrations of reactants have been investigated in some detail.

>"phe oxidation of thiols to disulfides (3, 10) or to sulfonic acids (11, ^12) i n the absence and presence of catalysts has been the subject of several recent investigations. Attention has been focused primarily on the range of products which may be produced as a function of the reaction conditions. In general, the reaction seems to lead to the almost exclusive production of disulfides when oxidation occurs i n aqueous solution, and to a variety of disulfides and sulfenic, sulfinic, and sulfonic acids when carried out i n nonaqueous solutions (2, 12). In both cases the reaction has been suggested (10, 12) to involve the initial production of disulfide, followed by hydrolysis of this compound to produce sulfur-containing acids RSH + Β

= RS" + B H

RS- + 0

= RS- + 0 -

(2)

RS- + 0 -

= RS- +

(3)

2RS-

= RSSR

2

2

2

0 22

+

+H 0 2

= ±0

2

0 22

+ 20H

216 In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

(1)

(4) (5)

16.

HOPTON E T A L .

RSSR + O H " = RS- + RSOH

(6)

RSOH + O H " = RSO- + HoO

(6)

3RSO"

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

217

Thiols to Disulfides

= R S 0 - + RSSR 3

(6)

Alternatively, Berger (2) has suggested that disulfide, on the one hand, and sulfinic and sulfonic acids, on the other hand, are alternative products formed from the thiol molecule and anion, respectively. Although no unequivocal evidence for the mechanism involving attack on disulfides by hydroxyl ions has been reported, support for this concept has been obtained from experiments involving potassium hydroxide-hexamethyl phosphoramide-water solvents ( I I ) , where the yield of disulfide relative to higher acids increases with the water content of the solvent mixture. Owing to the lack of further supporting evidence, the conclusions of Berger must be considered somewhat doubtful. The importance of the electron transfer reaction between RS" and an electron acceptor (Reactions 2 and 3) has been amply confirmed by the observation that the least acidic thiols are least resistant to oxidation (2), and by the enormously enhanced rate of reaction i n the presence of redox catalysts, such as transition metal ions (13) or organic redox additives (14). In these latter cases, reactions of the type below become important, RS" + X

2 +

= RS- + X

+

(7)

the ion X being reoxidized by molecular oxygen. Although the oxidation of thiols to disulfides i n the presence of a catalyst is a reaction of commercial interest, it is only comparatively recently that the marked effects of impurities on the system has been realized. Wallace and co-workers (13, 14) have studied the metalcatalyzed oxidation of some thiols i n the presence of a few metal ions and complexes under comparable conditions, and they have suggested a general mechanism for the reaction, based on Reactions 1, 4, 5, 6, and 7. The rate of reaction was found to depend on the chemical nature and the physical state of the catalyst. The reaction was suggested to involve metal complexes in the solid state (13). +

As a preliminary to the detailed investigation of the kinetics and mechanism of the oxidation of thiols in the presence of metal-containing catalysts (8), the present paper describes a survey of the rates and end products of oxidation of a series of alkyl and aryl thiols under comparable conditions. The reaction in the absence and presence of various metalcontaining catalysts has been studied under conditions of minimal i m purity concentrations.

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

218

OXIDATION OF ORGANIC COMPOUNDS

1

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

Experimental Materials. Since the oxidation of thiols is strongly catalyzed by traces of metal ions, all experimental techniques were designed to prevent the introduction of extraneous metallic impurities. Preparation and storage of reagents were completed in acid-washed, steam-cleaned glassware. Deionized water was used to prepare all solutions. The purest available commercial samples of thiols were carefully distilled, and the middle fractions were stored under nitrogen. Solutions of sodium hydroxide and of simple metal salts were prepared from Analar grade chemicals. Oxygen and nitrogen, from cylinders, were purified by passage through traps cooled to —198°C, the oxygen being condensed and fractionally distilled before use. Apparatus. The course of reaction was followed by periodically measuring the volume of oygen absorbed at constant pressure, and by analysis. The apparatus and techniques for measuring oxygen uptakes have already been described (3). Spectrophotometric investigation of complexes i n solution was carried out using a Unicam SP800 spectrophotometer. Analysis. The titration of thiols and of disulfide by iodine and bromine has been described (3). Control experiments involving the titration of solutions containing known amounts of thiol and disulfide showed the methods to be accurate to within ±2%. Samples of metal complexes isolated from the final solutions were subjected to microanalysis (for carbon, hydrogen, oxygen, and sulfur). Metals were determined colorimetrically by the following methods— copper: as the complex formed with sodium diethyl dithiocarbamate (6); cobalt: as the nitroso-R salt complex (7); nickel: as the dimethylglyoxime complex (4). Results Although the investigation of the oxidation of thiols in the absence of added metal catalysts has been reported elsewhere (3), it is necessary to compare some results with experiments in the presence of catalysts. The oxidation of thiols to disulfides was studied under standard conditions in both cases which, unless otherwise stated, involve 50-ml. samples of solutions containing sodium hydroxide ( 2 M ) , metallic catalyst (copper, 1 0 " M ; other metals, 10" M) and thiol (0.5M) maintained at 30 °C. under a constant pressure of oxygen (750 mm. H g ) . Reaction rates determined by measuring oxygen uptake as a function of time were confirmed by the regular analyses of thiol and disulfide; the stoichiometry agreed with the over-all equation 5

3

4RSH + 0

2

= 2RSSR + 2 H 0 2

(8)

Experiments were carried out on the oxidation of ethanethiol using a wide variety of metal catalysts. Some typical results are shown in Table I from which it is evident that adding metal salts always results

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

16.

219

Thiols to Disulfides

HOPTON E T A L .

in the quantitative production of disulfides and that copper, cobalt, and nickel are by far the most effective of the simple metal salts added to the solutions. Table I.

Oxidation of Ethanethiol Catalyzed by Metal Ions at Standard Conditions RSSR in Final Solutions, %

Time for Completion of 90% Reaction, Hrs. 0

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

Metal Ce U V Cr Mo W Mn Fe Fe Co Ni Pd Pt Cu Ag Zn Cd Hg Al Tl Sn

Added as (NH ) Ce(N0 ) U0 (N0 ) 6H 0 4

2

3

2

3

2

h

voso Cr (S0 ) K S0 24H 0 4

2

4

3

2

4

6

2

7

2 4

4

2

2

4

2

4

2

7 1 4i 4 8 8 1 8 6 η 8 8 8 8

4

2

4

4

2

3

4

2

4

2

2

2

4

2

2

4

3

2

4

2

100 99 100 100 100 100 99

6*

2

4

—

8 8 7

2

(ΝΗ ) Μο 0 ·4Η 0 Na W0 · 2H 0 MnS0 · 4H 0 FeS0 · 7 H 0 Haemin CoS0 · 7 H 0 N i S 0 + aq PdCl PtCl CuS0 · 5H 0 AgN0 ZnS0 · 7 H 0 3CdS0 · 8 H 0 HgCl K S 0 · A1 (S0 ) · 24H 0 T1 S0 SnCl · 2 H 0 4

100 98 98

7 8

6

2

2

• Expressed in terms of the reaction 4 R S H + 0 To the nearest 30 minutes.

2

—

105 100

—

99 103

— — — —

= 2 RSSR + 2 H 0 . 2

b

The oxidation of various thiols by these three catalysts was investi­ gated under standard conditions. A t concentrations of copper of 1 0 ' M , the rate of reaction was controlled by the rate of shaking of the reaction vessel (Table II) and depended on the oxygen pressure above the solu­ tions. A t copper concentrations of 1 0 M the reaction rate was inde­ pendent of the rate of shaking. This difference was attributed to the rate of diffusion of oxygen into the solutions being rate determining at 10" M [ C u ] . Under such conditions, increased shake rates result in better mixing of the reactants in the liquid phase and in a greater surface area of contact between gas and liquid, thereby enhancing the diffusion of oxygen into and throughout the solutions. If the rate of oxygen diffusion is the slowest step in the sequence of reactions, then the over-all reaction rate should increase with shake rate, as observed at 10" M [ C u ] . 3

_5

3

3

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

220

OXIDATION OF ORGANIC COMPOUNDS

Table II.

1

Effect of Shake Rate on the Kinetics of Oxidation in the Copper-Catalyzed Reaction 6

[CM] , M

Shake Rate,

ΙΟ" ΙΟ"

360 380

3 3

10-4

10" a

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

6 c

Rate of Oxygen Uptake

a

6

0.80 1.23 0.74 0.84 1.57 0.60 0.60

3 1 0

360 400 310 400

5

Cycles per minute of the shaker. Initial rates, expressed as percentage of final uptake/min. Standard conditions, using ethanethiol. Table III. Thiol

Oxidation of Various Thiols in the Presence of Copper, Cobalt, and Nickel Salts" Copper (10' Μ)

Cobalt (10~ M)

5

3

Nickel (10~ M) S

Time to Time to Time to Final Final Final Complete Complete Complete 90% of RSSR, 90% of RSSR, 90% of RSSR, Reaction, Hrs. Reaction, Hrs. % % Reaction, Hrs. % EtSH Bu SH Bu'SH Bu SH Bu'SH Hex SH PhSH PhCH SH. n

8

n

2

a

1 1* 1* 2 >10 1* >10 3

100 101 102 100

—

104

— 98

>10 6 8 >10 5 >10 >10

101

—

4 15 12 >10 >10

— —

>10 >10

—

100 98 101

96 102 99

— —

101

— —

Standard conditions; metals added as simple salts.

Studies of the copper-catalyzed system were completed, then, at a copper concentration of 10" M. Results for all three systems, averaged over a large number of determinations, are presented in Table III. In all cases, the oxidation rate was smallest for experiments involving thiophenol and ferf-butanethiol. The oxygen uptake vs. time curves for cobalt-catalyzed reactions showed an initial high slope followed by a decrease in slope after ca. 30% reaction to a final steady value. Reaction systems containing cobalt and nickel were characterized by the production of flocculent precipitates of compounds other than hy­ droxides i n the presence of all thiols, except for thiophenol and tertbutanethiol. Samples of these complexes produced from the ethanethiol system were washed, dried, and subjected to microanalysis. For nickel, the precipitate could be separated into two fractions by extracting with 5

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

16.

HOPTON E T A L .

221

Thiols to Disulfides

chloroform. The results of analysis of the cobalt complex agreed with the empirical formula C o ( S E t ) . i , and of the nickel complexes with the formulae, N i ( S E t ) (insoluble fraction), N i ( S E t ) ( O H ) (soluble fraction). It would seem that these complexes contain coordinated thiyl entities, although coordinated disulfide may be involved in some cases (I ). 3

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

2

3

Some investigation was made of the oxidation rates of ethanethiol after filtering these precipitates from the reaction solution. The oxidation rate under standard conditions (Case 1) was compared with the rate after filtering the solutions before adding thiols, but after adding all other reagents (Case 2), with the oxidation rate of solutions filtered immediately after adding all reagents (Case 3), and with the rate where metals had been added to the systems as the complex produced in the reaction (Case 4). The results are reported in Table IV. Using the ethanethiol system as a model, we investigated the dependence of the oxidation rate on the concentration of thiol, of oxygen, and of hydroxide ion. The results for the copper- ( 1 0 M ) , cobalt(10~ M), and nickel- (10" M) catalyzed oxidations, together with the comparable system in the absence of added catalysts are recorded in Table V . _5

3

3

Discussion The contention that disulfides are the major products of reaction when thiols are oxidized in aqueous alkaline solution has been amply confirmed by the present investigation. Thus, in the absence and in the presence of various metal ions (Table I) and i n the oxidation of various simple alkyl and aryl thiols (Table III) disulfide has always been produced quantitatively. Under these circumstances, experiments have been designed to investigate the kinetics and mechanism of the reaction as a basis for further detailed studies. Although it is unrewarding to compare the kinetics of oxidation of various thiols (Table III) i n detail since the systems are complicated by such factors as the differing degrees of ionization of thiols and differing partition of various thiols between the alkaline solutions and the product disulfide, it is instructive to consider reaction trends. Comparing the oxidation rates of n-, iso-, sec-, and teri-butane thiols and noting the low oxidation rate of terf-butanethiol and of thiophenol (cf. benzylthiol) shows that the geometric configuration of the organic chain of the thiol must play an important part in controlling the rate of oxidation. O n the other hand, Wallace et al. ( 13 ) have suggested that the electron transfer reaction ( Reaction 7 ) may control the over-all reaction rate—a conclusion supported in general by the relative rates of oxidation of n-alkanethiols

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

222

OXIDATION O F ORGANIC COMPOUNDS

Table IV.

Oxidation of Ethanethiol by Soluble

Case 1 Metal Cu Co Ni α

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

h

Added Cone., Rate, M Mole/Liter/Sec. 10~ 10" ΙΟ"

5

3 3

13.2 Χ 10" 10.3 Χ 10" 15.1 Χ ΙΟ"

6

6 6

1

Case 2 Active Rate, Metal, M Moie/Liter/Sec.

Active^ Metal, M

1.0 Χ 10" 1.0 X 10" 1.0 Χ ΙΟ"

1.0 Χ 10" 8.9 Χ 1 0 1.3 Χ 10~

h

5

3 3

13.2 Χ 10" 7.6 Χ 10" 3.4 Χ ΙΟ"

6

6 6

δ δ 3

Case Γ, no filtration; Case 2, reactant solutions filtered before adding thiol; Case 3, Amount of metal present in reacting solutions, by analysis.

observed i n the present study (Table 3 ) . N o correlation could be ob­ served, however, between the catalytic activities of metal ions and their redox potentials, presumably as a result of the fact that metal ions must be present largely as insoluble metal hydroxides. Metal catalytic activity may be expected to be a function of the solubility of the active species and/or the ease of electron transfer to the catalyst. The results given i n Table I V show conclusively that the sug­ gestion that catalysis occurs at a gas-solid interface (13) does not hold in these systems. Preliminary experiments showed that copper ion- and haemin-catalyzed systems oxidized rapidly with no trace of solid precipi­ tation, and that cobalt and nickel catalysis were characterized by the production of colored solutions and precipitates. Filtration experiments showed these precipitates played only a small part in catalysis ( Table I V ). Filtration i n Case 2 removed insoluble hydroxides from solution, and the resulting rate is much lower than the standard experiment ( Case 1). The close agreement between results for Cases 1 and 3 shows that insoluble, colored precipitates are not involved i n the reaction, the effec­ tive catalysis depending on the amount of metal i n solution. O n the other hand, there seems to be a saturation value of complex i n solution since the results for Case 3 (metal originally present as hydroxide) and Case 4 (metal originally present as complex) are quite similar. The coordination atmosphere of the metal ion i n solution can also be expected to affect the reaction rate. Microanalytical results indicate that the active catalysts i n cobalt and nickel systems could well be metal thiolic species produced in situ. However, these complexes are appre­ ciably more soluble i n the. alkaline solutions than are metal hydroxides (see, for example, the analysis results reported i n Table I V ) , and it is not possible on the present evidence to differentiate between catalysis as a result of increased solubility (comparing metal hydroxides and metal thiolic complexes ), and catalysis as a result of differences i n the allowed ease of electron transfer. It is apparent, however, that most of the metals investigated (Table I) are poor catalysts because they form only the insoluble hydroxide complexes.

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

16.

HOPTON E T A L .

223

Thiols to Disulfides

Metal Salts at Standard Conditions

0

Case 3 Rate, Mole/Liter/Sec. 13.2 X 10" 9.9 Χ 10" 14.6 Χ 10-

6 6

β

Case 4 Active Metal, M

Rate, Moie/Liter/Sec.

b

1.0 Χ 10" 6.4 Χ 10" 5.3 X 10'

Active Metal, M b

— 10.2 X 10~ 14.8 Χ 10"

5 4

— 1.0 Χ 10" 0.99 Χ 10"

6

4

3

6

3

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

reactant solution filtered after adding thiol; Case 4, metal added as thiol complex. The dependence of the oxidation rate of ethanethiol on the concen­ trations of reactants other than metal has also been investigated (Table V ) . The dependence of rate upon the concentration of reactants was obtained from plots of oxygen uptake vs. time and of disulfide production vs. time (3). N o significant differences i n results between analytical methods were observed once oxygen-uptake figures had been corrected for the stoichiometry of the reaction (see Reaction 8 ) . The reaction rates increased with alkali concentration, but this was caused by the increasing solubility of oxygen i n the solutions of higher alkali concen­ tration. In all cases the reaction rate increased linearly with oxygen pressure i n the system, and i n most cases the reaction rate was inde­ pendent of the amount of thiol present. Some unusual results were found at very low alkali concentrations and high thiol concentrations, owing to the fact that thiol was not completely ionized under these conditions. Table V .

Kinetic Parameters for the Metal-Catalyzed Oxidation of Ethanethiol

Order of Dependence of Rate on Concentration of Metal [EtSH] Cu Co Ni 1

[ Ο / ] [ΝαΟΗ ] α

0 0 0 0

Activation Energy, Kcal./Mole

Initial Rate Constant (30°C.) 4.9 2.3 2.1 3.6

Χ Χ Χ Χ

10" 10" 10" 10"

2

1

1

1

liter mole sec." sec." sec. liter mole sec.' -1

1

-1

1

1

-1

16.4 4.3 7.5 8.0

Corrected for solubility or oxygen in NaOH solutions of varying concentration.

Oxidation i n the presence of copper ions may well be controlled b y the rate of reoxidation of cuprous ions since the reaction between cupric ions and thiols is known to be almost instantaneous ( 5 ). The over-all process is, however, rapid, and the rate of reoxidation of copper ions must also be fast. The kinetic results tend to support the

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.

Downloaded by COLUMBIA UNIV on April 17, 2013 | http://pubs.acs.org Publication Date: January 1, 1968 | doi: 10.1021/ba-1968-0075.ch016

224

OXIDATION OF ORGANIC COMPOUNDS

1

contention that this reaction is rate controlling since the oxidation of ethanethiol (Table V ) depends on oxygen concentration and is inde­ pendent of thiol concentration. Again, if the reoxidation of copper is rate controlling, the over-all oxidation rate of thiols in the presence of copper should be approximately the same except where the reaction between copper ions and thiols is slow—e.g., when the reaction is sterically hin­ dered by the geometry of the thiol molecule. The results in Table III bear out these suggestions. In summary, the oxidation of thiols to disulfides is quantitative i n aqueous alkaline solution and may best be effected at high oxygen pressures i n the presence of a catalyst. The catalyst should dissolve i n the alkaline solutions, and of the simple metal salts, the addition of copper, cobalt, and nickel results in the most effective catalysis. Acknowledgement The authors wish to express their gratitude to the British Petroleum Company for the award of Research Bursaries to two of them ( J . D . H . and C.J.S.). Literature Cited (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14)

Akerfeld, S., Lovgren, C., Anal. Biochem. 8, 223 (1964). Berger, H., Rec. Trav. Chim. 82, 773 (1963). Cullis, C. F., Hopton, J. D., Trimm, D. L., unpublished data. Mitchell, A. M . , Mellon, M . G., Ind. Eng. Chem., Anal. Ed. 17, 380 (1945). Reid, Ε. E., "Organic Chemistry of Bivalent Sulfur," Vol. 1, Chemical Publishing Co., New York, 1958. Sandell, Ε. B., "Colorimetric Determinations of Traces of Metals," p. 444, Interscience, New York, 1959. Shipham, W. H., Lai, J. R., Anal. Chem. 28, 1151 (1956). Swan, C. J., Trimm, D. L., A D V A N . C H E M . SER. 76, 182 (1968). Tarbell, D. S., "Organic Sulfur Compounds," Pergamon Press, New York, 1961. Wallace, T. J., Schriesheim, Α., J. Org. Chem. 27, 1514 (1962). Wallace, T. J., Schriesheim, Α., Tetrahedron 21, 2271 (1965). Wallace, T. J., Schriesheim, Α., Tetrahedron Letters 1967, 1131. Wallace, T. J., Schriesheim, Α., Hurwitz, H . , Glaser, M . B., Ind. Eng. Chem., Process Design Develop. 3, 237 (1964). Wallace, T. J., Miller, J. M . , Pobiner, H., Schriesheim, Α., Proc. Chem. Soc. 1962, 384.

RECEIVED October 23,

1967.

In Oxidation of Organic Compounds; Mayo, F.; Advances in Chemistry; American Chemical Society: Washington, DC, 1968.