Jiawei
Huang
,
Xiaoman
Li
,
Liang
Xu
* and
Yu
Wei
*
School of Chemistry and Chemical Engineering/State Key Laboratory Incubation Base for Green Processing of Chemical Engineering, Shihezi University, Shihezi, China. E-mail: xuliang4423@shzu.edu.cn; yuweichem@shzu.edu.cn
First published on 11th March 2025
The cleavage and utilization of C–S bonds is a critical challenge in synthetic chemistry, traditionally requiring metal catalysis and disposing of sulfur fragments as wastes. Herein, we report an electrochemical method for the regioselective cleavage of C(sp3)–S bonds of alkyl aryl thioethers under mild conditions. This electro-oxidative approach generates the corresponding cationic species of both C- and S-fragments after the cleavage of the C–S bonds. Subsequently, these species are captured by O-nucleophiles and converted into aldehydes/ketones and sulfinates. The simultaneous utilization of both C- and S-fragments not only significantly enhances the atom economy but also offers a sustainable alternative to traditional C–S bond cleavage strategies.
More specifically, to cleave unsymmetrical alkyl aryl thioethers, the well-established transition metal-catalyzed two-electron oxidative addition processes selectively target the C(sp2)–S bonds,9–11 and thus offer valuable cross-coupling methods for constructing C(sp2)–C and C(sp2)–heteroatom bonds.12–19 Although less explored, several one-electron protocols have recently been disclosed, inducing the cleavage of C(sp3)–S bonds regioselectively.12,20–23 No matter via which approach, such transformations usually suffer from the wasteful disposal of sulfur fragments (Scheme 1a), which not only diminishes the atom economy but also squanders potential synthetic opportunities.24–26 Therefore, the development of more atom-economical strategies for the C–S bond cleavage of thioethers, which allow for the full utilization of C- and S-fragments, is of paramount importance.
The burgeoning growth of electrochemical organic synthesis not only offers sustainable alternatives to conventional synthetic routes but also enables new reactivity and selectivity under mild conditions.27–31Via utilizing electrons as traceless reagents in redox reactions directly, electrochemical synthesis circumvents the need for stoichiometric chemical oxidants or reductants and thus proceeds with minimal waste.32–35 In this regard, electrochemical C–S bond activation has also demonstrated significant progress recently, converting thioethers into diverse products.36,37
For example, in 2023, the Malins and Connal group38,39 and the Lei group40 disclosed two elegant protocols for anodic oxidation induced C–S cleavage of thioethers, independently (Scheme 1b). In both cases, the transformations are believed to be induced by the initially oxidatively formed radical cation species. In the former case, starting from N,S-acetals, this species decomposes into thiyl radicals and iminium intermediates, which are trapped by O-nucleophiles to deliver N,O-acetals. In the latter case, this species is trapped by another thiyl radical, affording disulfides finally. In the same year, Lundberg, Ahlquist and co-workers disclosed a protocol for electrochemical reductive C–S cleavage of aryl alkyl thioethers (Scheme 1c).41,42 Alkyl radicals are generated efficiently, which can subsequently be reduced to alkyl anions, thereby participating in a series of desulfurization and functionalization processes. Despite the fruitful achievements, the abovementioned methods make use of either the C- or S-fragment after cleaving the C–S bonds. Therefore, challenges remain in the efficient utilization of both the cleaved C- and S-fragments, necessitating the development of other innovative strategies.
Meanwhile, it is well known that in the electrochemical reactions of thiols, the generated thiyl radicals can be further oxidized to deliver thiyl cations.43–46 Therefore, it was possible that the C- and S-fragments are both converted into the corresponding cation species, after the C–S cleavage. In this context, it was questioned whether C-cations and S-cations could be trapped by the nucleophiles simultaneously (Scheme 1b), thus affording two products in an atom-economical manner, without wasting any fragment. Despite mechanistic promise, some potential pitfalls should also be contemplated and circumvented. For example, the easily-oxidized thioethers have been demonstrated to undergo diverse electrochemical oxidation reactions with their C–S bonds intact, affording an array of products such as sulfones and sulfoxides (Scheme 1d).47–52
Exploiting this logic, we performed proof-of-principle experiments to validate the hypothesis. Herein, the corresponding results were disclosed to showcase the feasibility of this strategy for the first time. Under electrochemical oxidative conditions without sacrificing the electrode, chemical oxidants or metal catalysts, after the regioselective C(sp3)–S cleavage of aryl alkyl thioethers, the resulting alkyl and arylthiyl motifs would be trapped by O-nucleophiles and converted into aldehydes/ketones and sulfinates (Scheme 1e), respectively.
:
0.2 mL), with the additive TsOH·H2O (p-toluenesulfonic acid monohydrate, 0.3 mmol) and the electrolyte Et4NBF4 (0.15 mmol). In this regard, while it has been well documented in the literature regarding the electrochemical oxidation of S atoms to afford sulfoxides and sulfones, fortunately, the oxidative cleavage of the C(sp3)–S bond predominated in this reaction system. Without an electric current, the reaction did not occur (entry 2). When TsOH·H2O was absent, only a trace amount of the desired products was observed (entry 3). The reaction yield significantly decreased when no additional water was added (entry 4). Replacing TsOH·H2O with TFA resulted in a decreased yield (entry 5). Changing to other mixed solvents, such as MeCN/H2O (entry 6), decreased the yield of 2a and suppressed the formation of 3a, and generated another byproduct, thiosulfonates (see Scheme 5g). On using graphite electrodes instead of platinum electrodes, the yield of 2a and 3a decreased to 45% and 62%, respectively (entry 7). Regulating the current to 6.0 mA was also detrimental to the yields (entry 8). The yields of the two products were almost unaffected under an argon atmosphere (entry 9), thereby excluding the possibility of air as the oxygen source or oxidant.
| Entry | Variations from the ‘standard’ conditions | Yield 2a a (%) |
Yield 3a a (%) |
|---|---|---|---|
| Standard conditions: 1a (0.3 mmol), TsOH·H2O (0.3 mmol), Et4NBF4 (0.15 mmol), EtOH (2.0 mL), H2O (0.2 mL), constant current = 4 mA under air for 12 hours (6.0 F mol−1).a Isolated yields.b Not detected. TFA = trifluoroacetic acid. | |||
| 1 | None | 86 | 70 |
| 2 | Without current | N.D.b | N.D.b |
| 3 | Without TsOH·H2O | Trace | Trace |
| 4 | Without H2O | 39 | 45 |
| 5 | TFA instead of TsOH·H2O | 74 | 67 |
| 6 | MeCN instead of EtOH | 66 | N.D. |
| 7 | C/C instead of Pt/Pt | 45 | 62 |
| 8 | 6 mA instead of 4 mA | 65 | 53 |
| 9 | Ar atmosphere | 83 | 83 |
With the optimal reaction conditions in hand, we began to evaluate the scope of the electrochemical C(sp3)–S bond cleavage of various thioethers, as shown in Scheme 2. First, we examined aryl groups on sulfur and found that the presence of methyl or chlorine at the para-position (1a–1c) had a minimal effect on the reaction, with all achieving efficient C–S bond cleavage and high yields. Subsequently, the benzyl-substituted aryl groups were investigated, and it was found that a variety of electron-donating (such as methyl and tert-butyl) and electron-withdrawing atoms/groups (such as fluorine, chlorine, bromine, and cyano) at the benzyls’ para-position (1d–1h) were well tolerated, affording the corresponding para-substituted benzaldehydes and 3a in moderate to excellent yields. In general, the electronic properties of substituents in these molecules have a limited influence on the isolated yields. Moreover, under the optimal reaction conditions, the products derived from ortho- and meta-substituted benzylic thioethers were also successfully obtained. Substituents positioned at the ortho-positions exhibit good yields (2k, 77%; 2l, 61%) of the corresponding aldehydes. The yields of the corresponding sulfinates 3c (70%; 74%) were not significantly affected by steric hindrance, demonstrating the robustness of this procedure towards steric effects. Additionally, halides such as Cl and Br were well tolerated in this transformation (1a–1c, 1g, 1j, and 1l), leaving ample opportunities for further functionalization of the aldehyde products.
We subsequently evaluated the conversion of α-branched benzyl thioethers as substrates (1m–1t) and obtained moderate to good yields of ketone products (63%–80%) and sulfinates (63%–82%). Benzyl thioethers with α-aryls and α-alkyls showed good compatibility with the reaction conditions. The yields of diaryl ketones (2m–2r) and alkyl aryl ketones (2s and 2t) were also not significantly affected by the electronic properties of the substituents. It is worth noting that when employing ordinary alkyl phenyl thioethers as substrates, it is likewise possible to selectively cleave the C(sp3)–S bond, yielding alkyl aldehydes (2u–2y) and sulfinates (3c and 3b). Dialkyl sulfides are equally applicable to this system. When cyclohexyl(3-phenylpropyl)sulfane 1z is introduced into the reaction, the reaction proceeds smoothly. However, we only observe the cleavage of the primary carbon–S bond, resulting in the formation of phenylpropanal 2u and cyclohexyl sulfinate 3o, without any evidence of the cleavage of the secondary carbon–S bond.
To further explore the applicability of alkyl aryl thioethers in this transformation, a series of methyl aryl thioethers were investigated. As depicted in Scheme 3 (left), in these cases (4a–4k), only sulfinates were ultimately obtained. The substituents on the aryl ring of the thioethers had a notable impact on the yields of the products. Substrates bearing electron-withdrawing groups yielded better results compared to those with electron-donating groups (3a, 66%; 3c, 85%). When ortho-Cl or meta-Cl phenyl thioethers were employed, a noticeable decrease in yield was observed (3f, 55%; 3h, 48%) compared to their para-substituted counterparts (3c, 85%), indicating that steric hindrance significantly impacts the cleavage of C–S bonds. Additionally, various alcohols were investigated. Primary and secondary alcohols, such as MeOH, nPrOH, iPrOH and BnOH, were well tolerated under optimal conditions, affording sulfinates (3i–3l) in moderate to good yields. As the nucleophilicity of the alcohol decreased, the yield of the sulfinates gradually declined. Subsequently, we evaluated the applicability of this method to natural products, employing menthol and diacetyl-D-galactose as nucleophilic O-sources. The reactions progressed smoothly, affording the corresponding products 3m and 3n, respectively.
Afterward, we began to explore whether the presence of an aromatic group on sulphur was essential for cleaving the C(sp3)–S bond (Scheme 3, right). Starting from cyclohexyl benzyl thioethers as the initial raw material, the reaction can still proceed smoothly, generating the corresponding aryl aldehydes with moderate yields (38%–61%), indicating that there is regional selectivity in the cleavage of the C(sp3)–S bond, proving that the phenyl group on the sulfur is not essential. At the same time, we can also obtain cyclohexyl sulfinate 3o. Subsequently, under optimized conditions, regardless of whether the substituent's nature is electron-donating or electron-withdrawing, the C–S bond can be cleaved with moderate yields via benzyl thiols to afford the corresponding aryl aldehydes. ortho-Substituted benzyl thiols also performed well, affording aldehydes 2k and 2z in 63% and 43% yields, respectively.
To verify the practicality and scalability of the electrochemical cleavage of C(sp3)–S bonds, we conducted a gram-scale experiment. We attempted the electrochemical oxidative cleavage of 6.0 mmol (1.49 g) of thioether 1a. As shown in Scheme 4, by simply scaling up the amount of each reagent and maintaining a constant current of 20 mA for 48 hours, we were able to obtain 2a (0.62 g) with a yield of 72% and 3a (0.64 g) with a yield of 58%.
In order to further illustrate the reaction mechanism, the following control experiments (Scheme 5) were performed. First, radical trapping experiments were conducted to explore the reaction mechanism (Scheme 5a). Under standard conditions, radical scavengers such as 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO), butylated hydroxytoluene (BHT), and 1,1-diphenylethylene (1,1-DPE) all exerted inhibitory effects on the formation of 2a and 3a. Furthermore, HRMS analysis identified the capture products of sulfide radicals H and sulfone radicals F, suggesting that both types of radicals are likely generated and involved in the reaction mechanism. To further delve into the details of this transformation, intermediate experiments were also conducted (Scheme 5b). Under standard conditions, compound 7 was utilized to afford 2a in a yield of 73% and 3a in a yield of 81%, indicating that 7 might be a crucial intermediate in this chemical transformation. Then, when sulfone 8 (Scheme 5c), instead of thioethers, was treated under the optimized conditions, no desired products could be obtained, excluding the possibility of sulfone as the critical intermediate. Disulfide may be a key intermediate in the reaction, so we attempted to treat disulfide 9 under standard condition A. The reaction proceeded smoothly, affording sulfinate 3a in a yield of 74% (Scheme 5d). Alcohols might be another key intermediate in the reaction process. Therefore, we attempted to treat benzyl alcohol E under standard conditions (Scheme 5e). The reaction proceeded smoothly, affording 2a with a 79% yield. Next, conducting 18O labelling experiments with 18O-water resulted in the formation of 18O-containing 2a and 3a (Scheme 5f), indicating that the two merged oxygen atoms of these two products originated from H2O. This conclusion was also reinforced by the efficient reaction in the absence of air (O2), as shown in Table 1 (entry 9). Subsequently, to gain a deeper insight into the intermediates during the reaction process, we analysed the components of the reaction mixture after 4 hours of reaction (Scheme 5g). It was found that in addition to the reactants and products, sulfoxide 7 and disulfide 9 could also be isolated. Finally, we wanted to know what will happen when the reaction does not contain O-nucleophiles. When we changed the solvent from ethanol to acetonitrile, the reaction still proceeded, but no generation of compound 3a was observed. Instead, thiosulfonates 10 were obtained (Scheme 5h). We hypothesized that this might be due to the overoxidation of disulfides generated during the reaction under electrooxidative conditions.53,54
Based on the above mechanistic results and the previous report,40,55–58 two plausible mechanisms are proposed, as shown in Scheme 6. In the first pathway (path a), thioether 1 undergoes single electron oxidation at the anode, yielding a sulphur radical cation intermediate A. Then, in the presence of water, sulfoxide intermediate 7 is generated, which is further oxidized at the anode to form sulfoxide radical cation B. This intermediate B subsequently undergoes cleavage to form a carbocation C and a sulfoxide radical F. Afterwards, the carbocation combines with water to form D, which undergoes deprotonation to generate benzyl alcohol intermediate E. The following loss of electrons and protons at the anode will lead to carbonyl compound 2. Meanwhile, sulfoxide radical F undergoes anodic oxidation to generate sulfoxide cation G, followed by nucleophilic attack by alcohol to yield sulfinates 3. Alternatively, via another pathway (path b), sulphur radical cation intermediate A directly cleaves to form sulphur radical H and carbocation C. Subsequently, the homo-coupling of H generates disulfide 9, which subsequently undergoes electro-oxidation to form thiosulfinates I. Finally, the nucleophilic attack of I by ethanol yields sulfinates 3.
Footnote |
| † Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5qo00248f |
| This journal is © the Partner Organisations 2025 |