Diastereoselective Access to anti-β-Hydroxy Sulfoxides from ­Chiral Epoxides and Prochiral Sulfenate Anions: Mechanistic ­Insights, Scope, and Limitation

Abstract

Reported herein is a stereoselective route to anti-β-hydroxy sulfoxides through the reaction of epoxides with sulfenate anions. Extensive experimental/computational studies revealed the dual special roles of MgBr₂·OEt₂, serving to generate the bromohydrin alkoxide intermediate, which undergoes nucleophilic attack on the prochiral sulfenate in a diastereoselective manner. The present study has opened a general stereoselective synthetic route to anti-β-hydroxy sulfoxides.

Stereodefined β-hydroxy sulfoxides represent a structural motif that is useful in chiral reagents,[1] auxiliaries,[2] or building blocks for syntheses of natural/nonnatural products.[3] Several accesses have been devised, such as (a) diastereoselective reduction of β-keto sulfoxide,[4] and (b) nucleophilic 1,2-addition of α-sulfinyl carbanions to aldehydes (Scheme [1]),[5] both of which exploit the sulfoxide’s stereogenicity to induce a hydroxy-bearing carbon stereochemistry (S*→C* asymmetric induction).

In this communication, we disclose a reversed approach: C*→S* stereochemical induction through the reaction of a sulfenate anion (RS–O^–) with a chiral epoxide (Scheme [2a]).[6]

This study started from a finding in our synthetic study on natural flavonoids (Scheme [2b]).[7] The reaction of the enantiopure epoxide (–)-3 with lithium benzenesulfenate (2a), generated in situ by a standard protocol from sulfoxide 1a,[8] afforded the β-hydroxy sulfoxide (–)-4, which was successfully converted, through a Pummerer/Friedel–Crafts cascade, into flavan 5, a useful platform in oligocatechin synthesis.

Aside from this pleasing result, we became intrigued by an unexpected finding regarding two special roles for MgBr₂·OEt₂ in the conversion of (–)-3 into (–)-4: (1) as an efficient promoter, and (2) its complete stereoselectivity in giving anti-4. This reaction was sluggish and nonstereoselective in the absence of MgBr₂·OEt₂.

In view of the importance of stereodefined β-hydroxy sulfoxides (see above), we decided to study this particular conversion in a broader context, paying attention to (1) mechanistic aspects of the chemical/stereochemical properties associated with the role of magnesium salts, and (2) the generality of the substrate scope.

It should be noted that although we used racemic epoxides in our study, our results imply an enantioselective access to stereodefined β-hydroxy sulfoxides by starting with an enantiopure starting material.

Table 1 Oxidative Lactonization of ent-8 ^a

Entry	Promoter	Additive	Yieldb % of 7a	anti/syn c
1	MgBr2·OEt2	–	82	7.2:1
2d	–	–	30e	1:1.2
3	MgI2·OEt2	–	80	6.0:1
4	MgBr2·OEt2	LiIf	86	9.0:1
5	MgBr2·OEt2	LiIg	78	5.6:1
6	MgBr2·OEt2	NaIf	79	7.3:1
7	MgBr2·OEt2	Bu4NIf	78	7.2:1
8	MgBr2·OEt2	LiBrf	80	7.2:1

^a Reaction conditions: 6 (0.4 mmol), PhS(O)Li (0.8 mmol), promoter (2.0 equiv), THF (8 mL), RT, 9 h.

^b Combined yield of anti/syn-diastereomers.

^c Determined by ¹H NMR analysis.

^d RT, 24 h.

^e With 60% recovery of substrate 6.

^f 0.2 equiv.

^g 1.0 equiv.

As an initial study to grasp the general reaction profiles, we employed rac-epoxide 6 and we examined the conditions for its reaction with lithium benzenesulfenate (2a) (Table [1], entry 1). To the sulfenate anion 2a in THF, generated by treatment of sulfoxide 1a with lithium tetramethylpiperidide (LiTMP; –78 °C, 20 min), were added epoxide 6 and MgBr₂·OEt₂. When the temperature was allowed to increase, the reaction proceeded to furnish 7a in 82% yield. By ¹H NMR analysis, the diastereomeric ratio (dr) was determined to be 7.2:1 in favor of the anti-isomer.[9] In contrast, the reaction was sluggish in the absence of MgBr₂·OEt₂ even at room temperature (24 h), giving 7a in a low yield (30%) with poor diastereoselectivity with 60% recovery of epoxide 6 (entry 2). The use of MgI₂·OEt₂, instead of MgBr₂·OEt₂, led to a slight decrease in both the yield and diastereoselectivity (entry 3).[10] We further evaluated the efficacy of additives, finding that the addition of LiI improved the outcome. Specifically, the use of MgBr₂·OEt₂ (2.0 equiv) with LiI (0.2 equiv) furnished 7a in high yield and a 9:1 dr (entry 4).[11] In contrast, the use of LiI in a stoichiometric amount resulted in poorer yield and diastereoselectivity (entry 5). The addition of other additives (NaI, Bu₄NI, or LiBr) did not induce any noticeable effects (entries 6–8).

During these experiments, we noted the presence of reaction intermediate(s) by TLC monitoring.[12] Thus, under the conditions described in Table [1], entry 2, the reaction did not proceed at –78 °C (Figure [1]; TLC A). On the other hand, in the presence of MgBr₂ and LiI (Table [1], entry 4), epoxide 6 was quickly consumed within 30 min at room temperature (RT), and we observed a new spot with a small R_f value of 0.45 (Figure [1]; TLC B). Upon prolonged reaction (RT, 9 h), this spot gradually disappeared with the emergence of a more polar product (R_f = 0.30) (Figure [1]; TLC C).

Isolation and ¹H NMR analyses of these products clarified that the initial new spot (R_f = 0.45 in TLC B) was the bromohydrin 8a and/or the iodohydrin 8b, corresponding to a ring-opening product of epoxide 6. Subsequent nucleophilic attack by sulfenate anion 2a on 8a/8b gives the substituted products anti/syn-7a, suggesting the intermediacy of the alkoxides of halohydrin 8a and/or 8b.[13] [14]

For confirmation, control experiments were carried out (Scheme [3]). Upon exposure of epoxide 6 to MgBr₂·OEt₂ in the absence or presence of LiI, the reaction quantitively afforded the corresponding halohydrins 8a/8b at room temperature within 30 minutes.

We investigated the reactivity of ring-opened products 8a and 8b toward the sulfenate anion 2a (Table [2]). Interestingly, the reaction of bromide 8a proceeded even in the absence of MgBr₂·OEt₂, giving the substitution product 7a in 80% yield (Table [2], entry 1). However, the stereoselectivity was significantly decreased (anti/syn = 2:1). The reaction of 8a with LiI (0.2 equiv) also gave 7a in 81% yield with a slight increase in the diastereoselectivity (anti/syn = 3:1; entry 2), whereas the reaction with MgBr₂·OEt₂ (2.0 equiv) afforded 7a in 82% yield with 6:1 dr (entry 3).[15] Furthermore, the combination of MgBr₂·OEt₂ (2.0 equiv) and LiI (0.2 equiv) furnished 7a in 85% yield with 8:1 dr (entry 4). When iodohydrin 8b was employed as a substrate, 7a was obtained cleanly in 85% yield with 9:1 dr (entry 5). Even without adding LiI, the reactivity of iodide 8b and the stereoselectivity of the resulting 7a were not affected (entry 6). In sharp contrast, the stereoselectivity of the reaction of 8b decreased in the absence of MgBr₂·OEt₂ and LiI (entry 7), implying that the magnesium salt plays an important role in achieving a high diastereoselectivity. Notably, the reaction with the hydroxy-protected substrate 9 did not proceed, resulting in full recovery of the starting material (Equation 1).

Table 2 Reaction of Halohydrins 8a/8b with Sulfenate Anion 2a ^a

Entry	Promoter	Additive	X	Yieldb (%) of 7a	anti/syn c
1	–	–	Br	80	2.0:1
2	–	LiId	Br	81	3.0:1
3	MgBr2·OEt2	–	Br	82	6.0:1
4	MgBr2·OEt2	LiId	Br	85	7.2:1
5	MgBr2·OEt2	LiId	I	85	9.0:1
6	MgBr2·OEt2	–	I	85	9.0:1
7	–	–	I	82	4.0:1

^a Reaction conditions: 8a/8b (0.4 mmol), PhS(O)Li (2.0 equiv), MgBr₂·OEt₂ (2.0 equiv), THF, –78 °C to RT, 9 h.

^b Combined yield of anti/syn-diastereomers.

^c Determined by ¹H NMR analysis.

^d 2.0 equiv.

These results suggest that the formation of a magnesium alkoxide is essential, not only for promoting the reaction, but also for ensuring a high stereoselectivity (Figure [2a]), whereas the noncovalent interactions between the magnesium atom and the oxygen atom of the methoxy group cannot promote the reaction (Figure [2b]).

With the proposed stepwise mechanism in mind, we screened other solvents in an attempt to further optimize the reaction conditions. However, the chemical yield and stereoselectivity could not be improved (for details, see the Supporting Information).

We then used DFT calculations at the B3LYP/6-311+G(d, p) level to examine the proposed stepwise mechanism theoretically (Figure [3]).[16] Most importantly, the results of these calculations revealed that the sulfenate species and the magnesium alkoxide of the bromohydrin intermediate are precoordinated through the oxygen atom on the magnesium alkoxide and the lithium cation on the sulfenate species. In the transition state (TS), a crucial coordination is formed between the leaving bromine atom and the magnesium atom. It is exactly this coordination that enables the smooth capture of the bromide by the magnesium atom after the completion of the S_N2 reaction. In TS _anti , the phenyl group is positioned in a pseudo-equatorial orientation, which leads to an anti-product. Conversely, in TS _syn , the phenyl group is situated in the pseudo-axial position, which gives the syn counterpart. The difference in the relative free energy between TS _syn and TS _anti [ΔG ^‡(TS _syn )–ΔG ^‡(TS _anti )] is 1.4 kcal/mol. This diastereoselectivity could be ascribed to the difference in the degree of torsional strain between the two TSs. In TS _syn , a considerable torsional strain is generated between the S–Ar bond and the eclipsing C–H bond at the electrophilic reaction site. In contrast, such a strain is not present in TS _anti .

Next, we turned our attention to the applicability of the reaction to other sulfenate species (Scheme [4]). The reaction of 6 with the 2,6-dimethylphenyl derivative 2b proceeded sluggishly, taking 24 hours to reach completion, and giving 7b in a lower yield and a lower stereoselectivity (53%; 6.7:1 dr) than with 2a. In contrast, the ortho-bromophenyl derivative 2c reacted smoothly to give 7c in a high yield and a high stereoselectivity (9 h; 83% yield, 8:1 dr). The reactions of sulfenate species bearing an electron-rich phenyl group (2d and 2e) were slightly faster (6 h to reach completion), and afforded the corresponding β-hydroxy sulfoxides in excellent yields. In the case of the para-methoxyphenyl derivative 2d, the stereoselectivity was slightly improved (anti/syn = 11:1) compared with that of the nonsubstituted phenyl analogue (anti/syn = 9:1), whereas the reaction with the ortho-derivative 2e led to a lower stereoselectivity (anti/syn = 8:1).

The 2-naphthyl derivative 2f gave the best result, affording the corresponding product 7f in 96% yield with excellent diastereoselectivity (>20:1 dr). The simple alkyl-chain-substituted derivative 2g was also suitable for use in this reaction, giving anti-product 7g with 18:1 dr. The relative stereochemistry of the major diastereomers of 7b, 7c, 7e, and 7g were deduced to be the same as those of 7d and 7f, whose relative configurations were assigned as anti, based on single-crystal X-ray diffraction analyses.

Then, we investigated the scope of the reaction with respect to the epoxide substrate (Scheme [5]). We discovered that the naphthalene-2-sulfenate anion 2f could be installed onto various epoxides with generally excellent yields and diastereoselectivities. The presence of a bulky 2,6-dimethylphenyl group in epoxide 10 slightly influenced its reactivity. The yield and stereoselectivity of adduct 17a were slightly lower than those obtained with epoxide 6 (86%, 15:1 dr compared with 96%, >20/1 dr for 7f). In the case of 2-(2-methoxybenzyl)oxirane (11), adduct 17b [17] was obtained in a high yield (97%) with 15:1 dr. Phenethyl 1,2-oxirane (12) also tolerated this reaction, generating 17c in 96% yield with >20:1 dr. When styrene oxide (13) was employed, the regioisomeric products 17d and 17d′ were obtained in yields of 60 and 33%, with >20:1 and 15:1 dr,[18] respectively.

It should also be noted here that a halogen atom at the γ-position of the oxirane ring was tolerated. Thus, epichlorohydrin (14) and epibromohydrin (15) gave adducts 17e and 17f with excellent results. These products can be expected to serve as a useful chiral building block for further functional transformations. When the reaction of 15 with two equivalents of lithium naphthalen-2-sulfenate [2-NapS(O)Li] was allowed to proceed for longer, a double addition of the sulfenate species occurred to form the double adduct 17g in 94% yield, albeit with a poor dr (1.6:1). These results suggest that the initial addition reaction occurred with a high dr (see above), whereas the second nucleophilic attack resulted in a poor dr.[19]

The use of O-benzylglycidol (16), which contains an alkoxy group at the γ-position of the oxirane ring, also substantially affected the stereoselectivity of the reaction. Although the corresponding adduct 17h was obtained in excellent yield, no stereoselectivity was observed, suggesting the occurrence of relatively strong coordination between the internal magnesium atom and the proximal oxygen atom (I in Equation 2), which would interfere with the key coordination between the magnesium atom and the leaving halogen atom (as in II), thereby lowering the stereoselectivity.

To gain further insight into this reaction, we examined the diastereomeric γ-alkoxy substrates syn-18 and anti-18 (Scheme [6]). Although both reactions proceeded well to give the corresponding adducts in excellent yield, the reactions produced significantly different stereochemical results. Specifically, the reaction of syn-18 showed poor stereoselectivity (3:1 dr; Equation 3), whereas that involving anti-18 resulted in excellent diastereoselectivity (>20:1 dr, Equation 4),[20] demonstrating the importance of the Mg–Br coordination in achieving a high stereoselectivity.

Assuming a strong electron-donating nature for the γ-oxygen atom lone pair, the five-membered chelate III from syn-18 would interfere with the coordination between the magnesium atom and the leaving bromine atom in IV, resulting in poor stereoselectivity. In contrast, such a chelate structure V from anti-18 cannot be formed due to the severe steric repulsion between the phenyl and bromomethyl substituents. Therefore, the formation of the coordination structure VI ensures a high stereoselectivity. These observations provide a useful guideline for designing suitable substrates for this stereoselective transformation.

In summary, we have developed a highly efficient method for the synthesis of β-hydroxy sulfoxides. Various chiral sulfoxides were obtained in high yields (≤97%) and excellent stereoselectivities (>20:1 dr). Control experiments showed that the most important reaction intermediate is a halohydrin alkoxide derived from the corresponding epoxide. DFT calculations revealed that coordination between the sulfenate species and the halohydrin intermediate is crucial in regulating the conformation in the TS and, thereby, ensuring a high diastereoselectivity. If an enantiopure epoxide is used in this reaction, the asymmetric synthesis of a β-hydroxy sulfoxide becomes possible. Applications of this method to the synthesis of bioactive compounds are currently in progress in our laboratory.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

We gratefully thank Professor Hidehiro Uekusa (Tokyo Institute of Technology) for the X-ray diffraction analyses.

• References and Notes

• 1a Kosugi H, Hoshino K, Uda H. Phosphorus, Sulfur Silicon Relat. Elem. 1994; 95: 401
  CrossrefSearch in Google Scholar
• 1b Kosugi H, Hoshino K, Uda H. Tetrahedron Lett. 1997; 38: 6861
  CrossrefSearch in Google Scholar
• 1c Kosugi H, Abe M, Hatsuda R, Uda H, Kato M. Chem. Commun. 1997; 1857
  Crossref
• 2a Broutin P.-E, Colobert F. Org. Lett. 2003; 5: 3281
  CrossrefPubMedSearch in Google Scholar
• 2b Salom-Roig X, Bauder C. Synthesis 2020; 52: 964 ; and references therein
  Thieme ConnectSearch in Google Scholar
• 3a Carreño MC, Pérez-González M, Ribagorda M, Somoza Á, Urbano A. Chem. Commun. 2002; 3052
  CrossrefPubMed
• 3b Fernández de la Pradilla R, Viso A, Castro S, Fernández J, Manzano P, Tortosa M. Tetrahedron 2004; 60: 8171
  CrossrefSearch in Google Scholar
• 3c Bauder C, Martínez J, Salom-Roig X.-J. Curr. Org. Synth. 2013; 10: 885 ; and references therein
  CrossrefSearch in Google Scholar
• 4a Solladié G, Greck C, Demailly G, Solladié-Cavallo A. Tetrahedron Lett. 1982; 23: 5047
  CrossrefSearch in Google Scholar
• 4b Solladié G, Greck C, Demailly G. J. Org. Chem. 1985; 50: 1552
  CrossrefSearch in Google Scholar
• 4c Solladié G, Greck C, Demailly G. Tetrahedron Lett. 1985; 26: 435
  CrossrefSearch in Google Scholar
• 4d Kosugi H, Konta H, Uda H. J. Chem. Soc., Chem. Commun. 1985; 211
  Crossref
• 5a Miokowski C, Solladié G. J. Chem. Soc., Chem. Commun. 1977; 162
  Crossref
• 5b Miokowski C, Solladié G. Tetrahedron 1980; 36: 227
  CrossrefSearch in Google Scholar
• 5c Demailly G, Greek C, Solladié G. Tetrahedron Lett. 1984; 25: 4113
  CrossrefSearch in Google Scholar
• 5d Solladié G, Moine G. J. Am. Chem. Soc. 1984; 106: 6097
  CrossrefSearch in Google Scholar
• 6a Riddell AB, Smith MR. A, Schwan AL. J. Sulfur Chem. 2022; 43: 540
  CrossrefPubMedSearch in Google Scholar
• 6b Saito F. Synthesis 2023; in press
  Thieme Connect
• 7 Betkekar VV, Suzuki K, Ohmori K. Org. Biomol. Chem. 2022; 20: 7419
  CrossrefPubMedSearch in Google Scholar
• 8 Caupéne C, Boudou C, Perrio S, Metzner P. J. Org. Chem. 2005; 70: 2812
  CrossrefPubMedSearch in Google Scholar
• 9 The relative stereochemistry of the major diastereomer of 7a was determined by comparison with an authentic sample of anti-7a prepared by another method. For details, see the Supporting Information and Ref. 4d.
• 10 The use of other Lewis acids such as ZnBr₂, AlBr₃, InBr₃, ScBr₃, Sc(OTf)₃, In(OTf)₃, or Zn(OTf)₂ met with failure. For details, see the Supporting Information.
• 11 (R_S *,2S*)-1-Phenyl-3-(phenylsulfinyl)propan-2-ol (anti-7a); Typical Procedure A solution of sulfoxide 1a (0.800 mmol, 2.0 equiv) in THF (3 mL) was added to a solution of LiTMP (0.800 mmol, 2.0 equiv) in THF (3 mL) at –78 °C, and the mixture was stirred at –78 °C for 20 min to give the corresponding sulfenate anion PhS(O)Li (2a). MgBr₂.OEt₂ (207 mg, 0.800 mmol, 2.0 equiv) and LiI (11.0 mg, 0.0800 mmol, 0.2 equiv) were then added at –78 °C, and the mixture was stirred for 20 min. A solution of epoxide 6 (0.400 mmol, 1.0 equiv) in THF (2 mL) was then at –78 °C. The cooling bath was removed, the temperature was raised to RT, and the mixture was stirred for 9 h. The mixture was then poured into 1 M aq HCl (40 mL) at 0 °C, and the products were extracted with CH₂Cl₂ (3 × 50 mL). The combined organic layers were dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by column chromatography (silica gel, hexane–EtOAc) to give a diastereomeric mixture of 7a [yield: 90 mg (86%), dr = 9:1 (assessed by ¹H NMR peak integration of the mixture)]. Further purification by flash chromatography [silica gel, hexane–EtOAc (2:3)] and crystallization from CH₂Cl₂–hexane (1:5) gave the major diastereomer anti-7a as a white powder; mp 88–90 °C; R_f = 0.4 (hexane–EtOAc, 1:1). IR (neat): 3630, 3030, 2927, 2825, 1580, 1477, 1452, 1059, 1050, 765 cm^–1. ¹H NMR (600 MHz, CDCl₃): δ = 7.56–7.54 (m, 2 H), 7.53–7.50 (m, 3 H), 7.25 (t, J = 7.8 Hz, 2 H), 7.21–7.19 (m, 1 H), 7.09 (d, J = 7.2 Hz, 2 H), 4.44–4.40 (m, 1 H), 3.64 (d, J = 3.0 Hz, 1 H), 3.03 (dd, J = 13.8, 9.6 Hz, 1 H), 2.87 (dd, J = 13.8, 7.2 Hz, 1 H), 2.78 (dd, J = 13.8, 6.6 Hz, 1 H), 2.70 (dd, J = 13.2, 1.2 Hz, 1 H). ¹³C NMR (150 MHz, CDCl₃): δ = 142.9, 137.0, 131.1, 129.5, 129.4, 128.7, 126.9, 124.1, 67.8, 60.7, 43.4. HRMS (ESI): m/z [M + H]⁺ calcd for C₁₅H₁₇O₂S: 261.0944; found: 261.0945.
• 12 For details of the TLC analyses, see the Supporting Information.

The few examples that have so far been reported on the diastereoselective alkylation of prochiral sulfenate anions using a chiral electrophile are limited to the pioneering studies by Schwan and co-workers, who used enantiopure chiral β-aminoalkyl iodides to obtain the corresponding diastereomeric β-amino sulfoxides, for selected examples, see:
• 13a Schwan AL, Verdu MJ, Singh SP, O’Donnell JS, Ahmadi AN. J. Org. Chem. 2009; 74: 6851
  CrossrefPubMedSearch in Google Scholar
• 13b Söderman SC, Schwan AL. Org. Lett. 2011; 13: 4192
  CrossrefPubMedSearch in Google Scholar
• 13c Söderman SC, Schwan AL. J. Org. Chem. 2013; 78: 1638
  CrossrefPubMedSearch in Google Scholar

For selected recent examples of the enantioselective alkylation of prochiral sulfenate anions, see:
• 14a Zong L, Ban X, Kee CW, Tan C.-H. Angew. Chem. Int. Ed. 2014; 53: 11849
  CrossrefPubMedSearch in Google Scholar
• 14b Jia T, Zhang M, McCollom SP, Bellomo A, Montel S, Mao J, Dreher SD, Welch CJ, Regalado EL, Williamson RT, Manor BC, Tomson NC, Walsh PJ. J. Am. Chem. Soc. 2017; 139: 8337
  CrossrefPubMedSearch in Google Scholar
• 14c Wang L, Chen M, Zhang P, Li W, Zhang J. J. Am. Chem. Soc. 2018; 140: 3467
  CrossrefPubMedSearch in Google Scholar
• 15 We deduced that the deprotonation of the halohydrins 8a and 8b to form the corresponding magnesium alkoxide occurred due to the co-existence of 2,2,6,6-tetramethylpiperidine (HTMP), generated together with the formation of the sulfenate species 2a.
• 16 To simplify the computational calculation, we employed MgBr₂·OMe₂ instead of MgBr₂·OEt₂.
• 17 The relative configuration of 17b was determined based on single-crystal X-ray diffraction analyses. For details, see the Supporting Information.
• 18 The relative stereochemistry was not determined.
• 19 The poor diastereoselectivity might be attributable to the following reason. The key interactions between the leaving bromo atom and the magnesium atom were inhibited by the presence of the strongly coordinative sulfinyl group initially introduced by the reaction, which might coordinate with the magnesium atom. This could also account for the low diastereoselectivity in the reaction of compound 6 shown next.
• 20 The relative configuration of 19b was determined by single-crystal X-ray diffraction analyses. For details, see the Supporting Information. CCDC 2280543, 2280545, 2280547, and 2280548 contain the supplementary crystallographic data for compounds anti-7d, anti-7f, anti-17b and 19b. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures17b

Corresponding Author

Publication History

Received: 24 September 2023

Accepted after revision: 23 October 2023

Accepted Manuscript online:
23 October 2023

Article published online:
21 November 2023

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References and Notes

• 1a Kosugi H, Hoshino K, Uda H. Phosphorus, Sulfur Silicon Relat. Elem. 1994; 95: 401
  CrossrefSearch in Google Scholar
• 1b Kosugi H, Hoshino K, Uda H. Tetrahedron Lett. 1997; 38: 6861
  CrossrefSearch in Google Scholar
• 1c Kosugi H, Abe M, Hatsuda R, Uda H, Kato M. Chem. Commun. 1997; 1857
  Crossref
• 2a Broutin P.-E, Colobert F. Org. Lett. 2003; 5: 3281
  CrossrefPubMedSearch in Google Scholar
• 2b Salom-Roig X, Bauder C. Synthesis 2020; 52: 964 ; and references therein
  Thieme ConnectSearch in Google Scholar
• 3a Carreño MC, Pérez-González M, Ribagorda M, Somoza Á, Urbano A. Chem. Commun. 2002; 3052
  CrossrefPubMed
• 3b Fernández de la Pradilla R, Viso A, Castro S, Fernández J, Manzano P, Tortosa M. Tetrahedron 2004; 60: 8171
  CrossrefSearch in Google Scholar
• 3c Bauder C, Martínez J, Salom-Roig X.-J. Curr. Org. Synth. 2013; 10: 885 ; and references therein
  CrossrefSearch in Google Scholar
• 4a Solladié G, Greck C, Demailly G, Solladié-Cavallo A. Tetrahedron Lett. 1982; 23: 5047
  CrossrefSearch in Google Scholar
• 4b Solladié G, Greck C, Demailly G. J. Org. Chem. 1985; 50: 1552
  CrossrefSearch in Google Scholar
• 4c Solladié G, Greck C, Demailly G. Tetrahedron Lett. 1985; 26: 435
  CrossrefSearch in Google Scholar
• 4d Kosugi H, Konta H, Uda H. J. Chem. Soc., Chem. Commun. 1985; 211
  Crossref
• 5a Miokowski C, Solladié G. J. Chem. Soc., Chem. Commun. 1977; 162
  Crossref
• 5b Miokowski C, Solladié G. Tetrahedron 1980; 36: 227
  CrossrefSearch in Google Scholar
• 5c Demailly G, Greek C, Solladié G. Tetrahedron Lett. 1984; 25: 4113
  CrossrefSearch in Google Scholar
• 5d Solladié G, Moine G. J. Am. Chem. Soc. 1984; 106: 6097
  CrossrefSearch in Google Scholar
• 6a Riddell AB, Smith MR. A, Schwan AL. J. Sulfur Chem. 2022; 43: 540
  CrossrefPubMedSearch in Google Scholar
• 6b Saito F. Synthesis 2023; in press
  Thieme Connect
• 7 Betkekar VV, Suzuki K, Ohmori K. Org. Biomol. Chem. 2022; 20: 7419
  CrossrefPubMedSearch in Google Scholar
• 8 Caupéne C, Boudou C, Perrio S, Metzner P. J. Org. Chem. 2005; 70: 2812
  CrossrefPubMedSearch in Google Scholar
• 9 The relative stereochemistry of the major diastereomer of 7a was determined by comparison with an authentic sample of anti-7a prepared by another method. For details, see the Supporting Information and Ref. 4d.
• 10 The use of other Lewis acids such as ZnBr₂, AlBr₃, InBr₃, ScBr₃, Sc(OTf)₃, In(OTf)₃, or Zn(OTf)₂ met with failure. For details, see the Supporting Information.
• 11 (R_S *,2S*)-1-Phenyl-3-(phenylsulfinyl)propan-2-ol (anti-7a); Typical Procedure A solution of sulfoxide 1a (0.800 mmol, 2.0 equiv) in THF (3 mL) was added to a solution of LiTMP (0.800 mmol, 2.0 equiv) in THF (3 mL) at –78 °C, and the mixture was stirred at –78 °C for 20 min to give the corresponding sulfenate anion PhS(O)Li (2a). MgBr₂.OEt₂ (207 mg, 0.800 mmol, 2.0 equiv) and LiI (11.0 mg, 0.0800 mmol, 0.2 equiv) were then added at –78 °C, and the mixture was stirred for 20 min. A solution of epoxide 6 (0.400 mmol, 1.0 equiv) in THF (2 mL) was then at –78 °C. The cooling bath was removed, the temperature was raised to RT, and the mixture was stirred for 9 h. The mixture was then poured into 1 M aq HCl (40 mL) at 0 °C, and the products were extracted with CH₂Cl₂ (3 × 50 mL). The combined organic layers were dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by column chromatography (silica gel, hexane–EtOAc) to give a diastereomeric mixture of 7a [yield: 90 mg (86%), dr = 9:1 (assessed by ¹H NMR peak integration of the mixture)]. Further purification by flash chromatography [silica gel, hexane–EtOAc (2:3)] and crystallization from CH₂Cl₂–hexane (1:5) gave the major diastereomer anti-7a as a white powder; mp 88–90 °C; R_f = 0.4 (hexane–EtOAc, 1:1). IR (neat): 3630, 3030, 2927, 2825, 1580, 1477, 1452, 1059, 1050, 765 cm^–1. ¹H NMR (600 MHz, CDCl₃): δ = 7.56–7.54 (m, 2 H), 7.53–7.50 (m, 3 H), 7.25 (t, J = 7.8 Hz, 2 H), 7.21–7.19 (m, 1 H), 7.09 (d, J = 7.2 Hz, 2 H), 4.44–4.40 (m, 1 H), 3.64 (d, J = 3.0 Hz, 1 H), 3.03 (dd, J = 13.8, 9.6 Hz, 1 H), 2.87 (dd, J = 13.8, 7.2 Hz, 1 H), 2.78 (dd, J = 13.8, 6.6 Hz, 1 H), 2.70 (dd, J = 13.2, 1.2 Hz, 1 H). ¹³C NMR (150 MHz, CDCl₃): δ = 142.9, 137.0, 131.1, 129.5, 129.4, 128.7, 126.9, 124.1, 67.8, 60.7, 43.4. HRMS (ESI): m/z [M + H]⁺ calcd for C₁₅H₁₇O₂S: 261.0944; found: 261.0945.
• 12 For details of the TLC analyses, see the Supporting Information.

The few examples that have so far been reported on the diastereoselective alkylation of prochiral sulfenate anions using a chiral electrophile are limited to the pioneering studies by Schwan and co-workers, who used enantiopure chiral β-aminoalkyl iodides to obtain the corresponding diastereomeric β-amino sulfoxides, for selected examples, see:
• 13a Schwan AL, Verdu MJ, Singh SP, O’Donnell JS, Ahmadi AN. J. Org. Chem. 2009; 74: 6851
  CrossrefPubMedSearch in Google Scholar
• 13b Söderman SC, Schwan AL. Org. Lett. 2011; 13: 4192
  CrossrefPubMedSearch in Google Scholar
• 13c Söderman SC, Schwan AL. J. Org. Chem. 2013; 78: 1638
  CrossrefPubMedSearch in Google Scholar

For selected recent examples of the enantioselective alkylation of prochiral sulfenate anions, see:
• 14a Zong L, Ban X, Kee CW, Tan C.-H. Angew. Chem. Int. Ed. 2014; 53: 11849
  CrossrefPubMedSearch in Google Scholar
• 14b Jia T, Zhang M, McCollom SP, Bellomo A, Montel S, Mao J, Dreher SD, Welch CJ, Regalado EL, Williamson RT, Manor BC, Tomson NC, Walsh PJ. J. Am. Chem. Soc. 2017; 139: 8337
  CrossrefPubMedSearch in Google Scholar
• 14c Wang L, Chen M, Zhang P, Li W, Zhang J. J. Am. Chem. Soc. 2018; 140: 3467
  CrossrefPubMedSearch in Google Scholar
• 15 We deduced that the deprotonation of the halohydrins 8a and 8b to form the corresponding magnesium alkoxide occurred due to the co-existence of 2,2,6,6-tetramethylpiperidine (HTMP), generated together with the formation of the sulfenate species 2a.
• 16 To simplify the computational calculation, we employed MgBr₂·OMe₂ instead of MgBr₂·OEt₂.
• 17 The relative configuration of 17b was determined based on single-crystal X-ray diffraction analyses. For details, see the Supporting Information.
• 18 The relative stereochemistry was not determined.
• 19 The poor diastereoselectivity might be attributable to the following reason. The key interactions between the leaving bromo atom and the magnesium atom were inhibited by the presence of the strongly coordinative sulfinyl group initially introduced by the reaction, which might coordinate with the magnesium atom. This could also account for the low diastereoselectivity in the reaction of compound 6 shown next.
• 20 The relative configuration of 19b was determined by single-crystal X-ray diffraction analyses. For details, see the Supporting Information. CCDC 2280543, 2280545, 2280547, and 2280548 contain the supplementary crystallographic data for compounds anti-7d, anti-7f, anti-17b and 19b. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures17b

• Supplementary Material