Enantioconvergent Access to Chiral S(VI) Stereocenters by Kinetic Resolution of Sulfonimidoyl Chlorides

TL;DR

A new catalytic method enables the production of chiral sulfur(VI) compounds with high enantioselectivity.

Contribution

The study introduces a novel enantioconvergent kinetic resolution strategy for chiral sulfur(VI) stereocenters.

Findings

The method provides enantioconvergent access to sulfonimidoyl chlorides and sulfonimidates.

Noncovalent interactions and the catalyst's unfunctionalized backbone are crucial for enantioselectivity.

Abstract

Sulfur chirality has recently gained in importance across a wide range of chemical domains. However, the dearth of catalytic approaches for sulfur(VI) stereocenters continues to present a challenge for the broader utilization of chiral‐at‐sulfur motifs. We report herein the development of a previously unexplored approach for the kinetic resolution of chiral sulfur(VI) functionalities that leverages the electrophilic reactivity of sulfur(VI) stereocenters. This catalytic strategy provides enantioconvergent access to sulfonimidoyl chlorides and sulfonimidates, which are some of the least explored and catalytically accessible stereogenic S(VI) functionalities. This study also elucidates the key roles of noncovalent interactions and highlights the importance of the unfunctionalized backbone of the catalyst in imparting high enantioselectivity to the kinetic resolution of heteroatom stereocenters.

An enantioconvergent approach for the kinetic resolution of S(VI) stereocenters that leverages the electrophilic reactivity of the chiral sulfur atom is described. The catalytic process provides access to sulfonimidoyl chlorides and sulfonimidates, which are among the least studied S(VI) functionalities. Mechanistic studies reveal key roles of substrate interactions with the unfunctionalized catalyst backbone in the enantioselective process.

Figures

Kinetic resolution of sulfonimidoyl chlorides. a). Applications of chiral S(VI) functionalities. b). Sulfonimidoyl chlorides and sulfonimidate esters. c). Approaches for the kinetic resolution of S(VI) stereogenic centers. d). Kinetic resolution of sulfonimidoyl chlorides enabled by the copper‐catalyzed nucleophilic substitution at the S(VI) stereocenter.

Scope of the kinetic resolution of sulfonimidoyl chlorides. The yields and ee are reported for the isolated products. Reaction conditions: sulfonimidoyl chloride (0.1 mmol), diol 2 (0.17 mmol), Cu(OTf)2 (10 mol%), L1 (12 mol%), 3 Å molecular sieves (40 mg), Ag2CO3 (0.03–0.09 mmol), CH2Cl2 (1 mL), 9–23 h. a) The preparative scale synthesis was carried out with ent‐L1. a) 30 °C. b) 0 °C.

Synthetic applications of sulfonimidoyl chlorides and mechanistic studies of the kinetic resolution reaction. a). Conversion to synthetically and medicinally important chiral sulfur(VI) functionalities. Reaction conditions: a. KF (1 equiv.), dibenzo‐18‐crown‐6 (2 equiv.), PhCF3/cyclohexane (2: 1); b. KN3, THF; c. phenol (1.1 equiv.), Cs2CO3 (1.1 equiv.), 1,2‐dichloroethane; d. piperidine (1.1 equiv.), Na2CO3 (1.1 equiv.), CH2Cl2. b). Construction of sulfonimidate and sulfonimidamide derivatives of drugs and natural products from ent‐1a obtained with ent‐L1. c). Sulfonimidate tagging of drugs. Reaction conditions: carboxylic acid (1.5 equiv.), diisopropylethylamine (DIPEA, 2 equiv.), diisopropylcarbodiimide (DIC, 2 equiv.), 4‐pyrrolidinopyridine (PPY, 10 mol%), CH2Cl2. d). Construction of sulfoximines and sulfonimidamides from sulfonimidates 3. Reaction conditions: a. Me3OBF4, (2 equiv.), proton sponge (1.5 equiv.), CH2Cl2, 10 °C. b. RLi, Et2O, –78 °C (for 16–18, ent‐7, and 20) or –100 °C (for 19).

Mechanistic studies of the kinetic resolution reaction. a). Dependence of the ee of sulfonimidate 3a (●) and recovered sulfonimidoyl chloride 1a (■) (eeprod) on the ee of the catalyst (eecat). b). Dependence of the experimental (■) and theoretical (●) selectivity factors s on the ee of the catalyst. The experimental s values were calculated as shown in Table 1. The theoretical s values were calculated as a function of eecat: s = [s“(1 + eecat)–(1–eecat)]/[(1 + eecat) + s”(1–eecat)] for s' = 83 obtained with enantiomerically pure L1. c). Variable time normalization analysis (VTNA) for sulfonimidoyl chloride rac‐1a (■, 0.08 M; ●, 0.1 M). d). VTNA for diol 2 (▲, 0.17 M; ◆ 0.25 M). e). VTNA for the catalyst (▼, 7 mol%; ●, 10 mol%). f). X‐ray crystal structure of the catalyst–diol adduct Cu(ent‐L1)(κ 2O‐2)(κ 1O‐2)[OTf]2, 21. Hydrogen atoms and triflate anions are omitted for clarity.

Mechanistic computational studies of the kinetic resolution reaction of sulfonimidoyl chlorides. a). Catalytic cycle of kinetic resolution. b). Transition state structures TS‐R and TS‐S for the Cu‐catalyzed nucleophilic displacement process. c). Distortion–interaction analysis of transition state structures TS‐R and TS‐S, ΔE, kcal mol−1. d). Energy decomposition analysis of transition state structures TS‐R and TS‐S, ΔΔE‡  = ΔE‡ TS‐R –ΔE‡ TS‐S , kcal mol−1. e). Contribution of the individual L1 fragments to the dispersion component of the substrate–catalyst interaction in TS‐R and TS‐S, ΔΔE‡ disp = ΔE‡ disp( TS‐R )–ΔE‡ disp( TS‐S ), kcal mol−1. f). Complementary occupied–virtual pairs (COVPs) in the β space for the n O( ent ‐ 1a )→σ* 22 interaction in TS‐R.

Tables

Table 1. Kinetic resolution of sulfonimidoyl chlorides a)

		1a		3a
Entry	Variation from standard conditions	Yield, % b)	ee, % c)	Yield, % b)	ee, % c)	s
1 d)	None	52	96	43	91	83
2	With CuI	43	12	10	53	3
3	With CuCl2	94	3	3	84	12
4	With Fe(OTf)2	90	2	3	46	3
5	With Co(OTf)2	82	14	13	89	20
6	In EtOAc	58	32	38	65	6
7	In toluene	62	36	30	80	13
8	With K2CO3	68	24	17	94	41
9	With Cs2CO3	85	8	10	72	7
10	With ethylene glycol	73	2	23	2	1
11	With benzyl alcohol	98	1	0	0	–

Full text

Introduction

Sulfur(VI) stereocenters have recently emerged at the forefront of drug discovery, agrochemistry, asymmetric catalysis, and supramolecular and macromolecular applications.^[^ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 ^]^ S(VI) stereogenic functional groups, such as sulfonimidamides, sulfoximines, sulfonimidoyl halides, and sulfonimidates, exhibit a diverse range of physicochemical properties and reactivities, which can be systematically fine‐tuned by modifying the substitution pattern around the stereogenic sulfur atom (Figure 1a). Furthermore, chiral S(VI) functionalities, such as sulfonimidamides and sulfoximines, can offer distinct advantages over well‐established sulfonamides and sulfones, including increased polarity and solubility, and additional molecular vectors for noncovalent interactions.^[^ 4, 5, 7, 8 ^]^ However, the limited synthetic accessibility of the various chiral S(VI) functionalities remains a major obstacle to the broader adoption of S(VI) stereocenters.^[^ 3, 4, 7, 8, 9 ^]^ Furthermore, their divergent reactivities pose a challenge to the development of generalizable synthetic strategies that enable the construction of diverse chiral S(VI) functionalities from readily accessible S(VI) precursors.^[^ 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34 ^]^ As a result, most synthetic strategies to S(VI) stereocenters rely on the oxidative conversion of chiral S(IV) compounds. Although substantial progress in the development of new approaches to chiral S(IV) compounds has recently been achieved, the synthetic availability and structural diversity of chiral S(IV) precursors continue to pose significant challenges.^[^ 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53 ^]^ Additionally, the scope of methods for oxidative conversion of S(IV) to S(VI) functionalities also remains underdeveloped.^[^ 3, 7, 8, 9, 15 ^]^

Kinetic resolution of sulfonimidoyl chlorides. a). Applications of chiral S(VI) functionalities. b). Sulfonimidoyl chlorides and sulfonimidate esters. c). Approaches for the kinetic resolution of S(VI) stereogenic centers. d). Kinetic resolution of sulfonimidoyl chlorides enabled by the copper‐catalyzed nucleophilic substitution at the S(VI) stereocenter.

Sulfonimidoyl chlorides and sulfonimidates are among the least explored chiral S(VI) functionalities. First, synthesized in the enantioenriched form from sulfinamides by Johnson in 1971,^[^ 54 ^]^ sulfonimidoyl chlorides can be converted to other chiral S(VI) functionalities by reactions with nucleophiles (Figure 1b).^[^ 28, 54 ^]^ Sulfonimidoyl chlorides bearing electron‐withdrawing groups on the nitrogen atom are stable at ambient temperature and can be handled without any special precautions. Although this versatility can be advantageous for the late‐stage diversification of chemical libraries and the generation of neglected S(VI) pharmacophores, asymmetric synthesis of sulfonimidoyl chlorides continues to be challenging.

The synthetic strategies for the introduction of sulfonimidate groups into organic molecules have similarly been largely confined to the oxidative conversion of chiral S(IV) precursors.^[^ 55, 56, 57, 58, 59, 60 ^]^ Despite these challenges, recent applications of sulfonimidates as chiral S(VI) electrophiles and biological probes underscore their untapped potential in synthetic and medicinal chemistry.^[^ 55, 56, 57, 58, 59, 60, 61, 62 ^]^ Accordingly, the development of catalytic asymmetric approaches for sulfonimidoyl chlorides and sulfonimidates has recently gained attention in the chemistry of stereogenic sulfur functionalities,^[^ 28, 34 ^]^ yet persistent limitations in the structural diversity and synthetic accessibility remain largely unresolved.

To this end, we posited that the electrophilic reactivity of sulfonimidoyl chlorides could be leveraged for the development of a kinetic resolution reaction via a catalytic stereospecific S_N_2 substitution at the stereogenic sulfur atom with an appropriate alcohol nucleophile. This unprecedented approach would enable enantioconvergent access to chiral sulfonimidoyl chlorides and sulfonimidates, which could also be readily converted to other chiral S(VI) functionalities. Furthermore, it would provide insight into catalytic systems, which could promote kinetic resolution of electrophilic S(VI) stereocenters.

Although this strategy has not been experimentally validated, recent studies, facilitated by the nucleophilic reactivity of sulfoximines, sulfondiimides, and sulfonimidamides, have highlighted the synthetic potential of kinetic resolution reactions in chiral S(VI) settings (Figure 1c).^[^ 19, 23, 24, 25, 26, 27, 30 ^]^ For example, the organocatalytic acylation and alkylation of the nucleophilic nitrogen atom in sulfoximines and sulfonimidamides was employed by Bolm,^[^ 19 ^]^ Willis,^[^ 25 ^]^ and Biju.^[^ 30 ^]^ On the other hand, several approaches based on the transition metal‐catalyzed C─H functionalization of the ortho position in sulfoximines were developed by Cramer,^[^ 23 ^]^ Shi,^[^ 24, 27 ^]^ and Sahoo and Gandon.^[^ 26 ^]^ Additionally, elegant and mechanistically distinct stereoablative approaches based on sulfonimidoyl chlorides have recently been disclosed by Jiang^[^ 33 ^]^ and Wu.^[^ 34 ^]^ We report herein the development of a previously unexplored kinetic resolution reaction of sulfonimidoyl chlorides that enables synthetic access to a variety of chiral S(VI) functionalities, including sulfonimidates, sulfonimidamides, and sulfoximines (Figure 1d). This study provides a blueprint for the exploration of asymmetric catalytic approaches that are based on the electrophilic reactivity of S(VI) stereocenters. It also reveals the steering effect of stabilizing interactions between the substrate and the unfunctionalized backbone of the chiral catalyst, which can guide the development of other kinetic resolution approaches for heteroatom chirality.

Results and Discussion

Initial optimization studies revealed that the kinetic resolution of sulfonimidoyl chloride rac‐1a can be accomplished with diol 2 (Table 1). The selection of diol 2 was based on the premise that its chelating ability and the Thorpe–Ingold effect of the gem‐dimethyl moiety^[^ 63 ^]^ can facilitate complexation with a metal catalyst. Under optimal conditions, the reaction was catalyzed by copper(II) triflate and bisoxazoline ligand L1 (entry 1, Table 1) with silver carbonate as the base and dichloromethane as the solvent. Sulfonimidate ester 3a and sulfonimidoyl chloride 1a were produced with 91% and 96% ee, pointing to an efficient kinetic resolution process (s = 83). Catalysts based on other copper(I) and copper(II) salts (entries 2 and 3) and other metals (entries 4 and 5) demonstrated inferior catalytic activity. Other solvents and basic reagents also afforded products 3a and 1a with lower enantioselectivities and yields (entries 6–9). Notably, an evaluation of a wide range of alcohols and diols revealed that the use of diol 2 was crucial for achieving high selectivity (entries 10 and 11). Ligand selection was similarly important for achieving high enantioselectivity and efficiency. Structural adjustments in the optimal trans‐diphenyloxazoline fragment resulted in diminished selectivity (L2). Notably, the indane backbone was critical for achieving high enantioselectivity (L3), and more drastic changes in the ligand structure proved broadly detrimental to the reaction performance (L4‐L6). Furthermore, no formation of S(IV) by‐products was observed under the Lewis acid‐catalyzed reaction conditions. This result contrasts with the facile formation of S(IV) products in Lewis base‐catalyzed reactions with alcohol nucleophiles recently reported by Wu,^[^ 52 ^]^ underscoring the synthetic versatility of sulfonimidoyl chlorides.

Table 1: Kinetic resolution of sulfonimidoyl chlorides a)

With the optimal conditions established, we next explored the substrate scope of the kinetic resolution reaction (Figure 2). Substrates with various halogen substituents readily produced corresponding sulfonimidate esters 3b‐3e with 90%–99% ee and sulfonimidoyl chlorides 1b‐1e with 95%–99% ee (*s *= 79 – >200^[^ 64 ^]^). The nitro group was also well tolerated (90% ee for 3f and 98% ee for 1f; s = 87). High selectivity (s = 94 – >200) was also observed for the substrates with pendant methoxy and di‐ and trifluoromethoxy groups (3g‐3i, 93%–99% ee, and 1g‐1i, 93%–97% ee). Sulfone and cyano groups were likewise compatible with the developed method (3j and 3k, and 1j and 1k). Furthermore, substrates with diverse alkyl and aryl groups afforded sulfonimidate esters 3l‐3o and recovered sulfonimidoyl chlorides 1l–1o with high stereoinduction (s factors up to 180). The reaction also demonstrated excellent resolution performance, with s values in the range of 89 – >200 for sulfonimidoyl chlorides with various combinations of halogen, alkyl, and trifluoromethyl groups (3p‐3u and 1p‐1u). High enantioselectivity (s = 80) was also observed for naphthalene‐derived sulfonimidoyl chloride rac‐1v. Substrates with diverse substituents in the N‐sulfonyl residue were also amenable to efficient kinetic resolution. Halogen substituents in the N‐sulfonyl residue were well tolerated (3w‐3z and 1w‐1z) and provided s factors of up to 111. Sulfonimidoyl chlorides bearing trifluoromethyl and methoxy groups were equally compatible (3aa and 3ab, and 1aa and 1ab). Finally, the substrate rac‐1ac displaying an aliphatic N‐sulfonyl group was also tested, and sulfonimidate ester 3ac and recovered sulfonimidoyl chloride 1ac were afforded with 92% and 95% ee. X‐ray crystallographic analysis of sulfonimidoyl chloride 1d (Figure 2), as well as ent‐1b and ent‐1g and sulfonimidate ent‐3x obtained with ent‐L1 (see the Supporting Information) confirmed the absolute configurations of the kinetic resolution products. Furthermore, the developed method can be used to access gram quantities of the enantioenriched sulfonimidate and sulfonimidoyl chloride products (ent‐3d and ent‐1d) with concomitant ligand and silver(I) recovery. Lastly, in contrast to the high selectivity observed in the reactions of sulfonimidoyl chlorides 1a‐1ac, no kinetic resolution was observed when the aromatic group on the sulfur was replaced with an alkyl group or when the sulfonyl group on the nitrogen atom was replaced with the benzoyl group.

Scope of the kinetic resolution of sulfonimidoyl chlorides. The yields and ee are reported for the isolated products. Reaction conditions: sulfonimidoyl chloride (0.1 mmol), diol 2 (0.17 mmol), Cu(OTf)2 (10 mol%), L1 (12 mol%), 3 Å molecular sieves (40 mg), Ag2CO3 (0.03–0.09 mmol), CH2Cl2 (1 mL), 9–23 h. a) The preparative scale synthesis was carried out with ent‐L1. a) 30 °C. b) 0 °C.

We next determined whether the developed kinetic resolution method could facilitate synthetic access to other valuable chiral sulfur functionalities by stereospecific substitution reactions (Figure 3a). Sulfonimidoyl fluoride 4 could be readily produced from 1a with high enantiospecificity in a reaction with potassium fluoride in conjunction with a crown ether and in a nonpolar medium to suppress racemization. The substitution of chloride with azide likewise proceeded with excellent enantiospecificity (5). Furthermore, the sulfonimidoyl group could be efficiently transferred onto the oxygen atom in phenol, thereby producing aryl sulfonimidate 6. Similarly, high stereochemical fidelity was observed in a reaction with an amine that produced sulfonimidamide 7. Stereospecific sulfonimidoyl transfer could also be readily accomplished in the more complex structural settings of drugs and natural products following the synthetic procedures developed for compounds 6 and 7 (Figure 3b), thus yielding sulfonimidamide analogs of the antihistamine desloratadine (8) and the antipsychotic risperidone (9) and sulfonimidate analogs of tyrosine (10) and ezetimibe (11). Furthermore, the hydroxy group in products 3 can be used to introduce sulfonimidate residues into bioactive molecules (Figure 3c). To that end, carbodiimide‐mediated esterification enabled the construction of a sulfonimidate‐containing analog of the hepatitis C drug grazoprevir 12. Sulfonimidate analogs of the anti‐inflammatory and retinoid drugs acemetacin (13) and adapalene (14) were likewise readily synthesized. Lastly, an analog of the xanthine oxidase inhibitor febuxostat (15) was also accessed with high stereochemical fidelity.

Synthetic applications of sulfonimidoyl chlorides and mechanistic studies of the kinetic resolution reaction. a). Conversion to synthetically and medicinally important chiral sulfur(VI) functionalities. Reaction conditions: a. KF (1 equiv.), dibenzo‐18‐crown‐6 (2 equiv.), PhCF3/cyclohexane (2: 1); b. KN3, THF; c. phenol (1.1 equiv.), Cs2CO3 (1.1 equiv.), 1,2‐dichloroethane; d. piperidine (1.1 equiv.), Na2CO3 (1.1 equiv.), CH2Cl2. b). Construction of sulfonimidate and sulfonimidamide derivatives of drugs and natural products from ent‐1a obtained with ent‐L1. c). Sulfonimidate tagging of drugs. Reaction conditions: carboxylic acid (1.5 equiv.), diisopropylethylamine (DIPEA, 2 equiv.), diisopropylcarbodiimide (DIC, 2 equiv.), 4‐pyrrolidinopyridine (PPY, 10 mol%), CH2Cl2. d). Construction of sulfoximines and sulfonimidamides from sulfonimidates 3. Reaction conditions: a. Me3OBF4, (2 equiv.), proton sponge (1.5 equiv.), CH2Cl2, 10 °C. b. RLi, Et2O, –78 °C (for 16–18, ent‐7, and 20) or –100 °C (for 19).

Sulfonimidates 3 can also be readily converted to sulfoximines 16–19 (Figure 3d) via a trimethyloxonium salt‐mediated methylation of the hydroxy group, which is followed by a stereospecific substitution reaction with an organolithium reagent. Sulfonimidamides ** ent‐7** and 20 could also be prepared using the same procedure with the corresponding lithium amides as nucleophiles. Collectively, these results indicate that the developed kinetic resolution method can serve as a general synthetic platform for the construction of a variety of chiral sulfur(VI) functional groups via nucleophilic substitution at the sulfur stereocenter in both the sulfonimidate products and the recovered sulfonimidoyl chlorides with high stereochemical fidelity. Furthermore, the enantioconvergent approach enables the efficient generation of libraries of diverse stereogenic S(VI) functionalities from a single racemic precursor, such as sulfonimidates, sulfonimidamides, and sulfoximines 8–11, 16–20, from rac‐1a.

Prompted by the observed high enantioselectivity of the kinetic resolution reaction of sulfonimidoyl chlorides, we set out to elucidate the mechanistic details of the catalytic process.

To investigate the nature of the catalyst speciation, the influence of the enantiomeric purity of the catalyst on the selectivity of the kinetic resolution was first examined. In asymmetric catalysis, the dependence of the product ee on the enantiopurity of the catalyst is linear in the absence of higher order effects, whereas deviations from linearity (nonlinear effects, NLE) indicate catalyst aggregation or the involvement of multiple catalyst species.^[^ 65 ^]^ In contrast, normal catalytic behaviors in kinetic resolution exhibit intrinsically nonlinear dependence of the ee values of the product and recovered reactant on ee_cat_. Therefore, higher‐order effects are indicated by a deviation from the theoretically derived hyperbolic dependence of the experimental stereoselectivity factor s on the catalyst enantiopurity.^[^ 66 ^]^ In line with these considerations, both the ee of product 3a and that of recovered sulfonimidoyl chloride 1a were influenced by the catalyst enantiopurity (Figure 4a). Furthermore, the experimental stereoselectivity factor s demonstrated an excellent fit with the theoretical hyperbolic curve (Figure 4b) that was derived for the normal behavior of a catalytic system with the experimentally observed intrinsic stereoselectivity factor (s = 83) for the enantiomerically pure catalyst (See page S4 in the Supporting Information for further details). These results indicate that the catalytically active species contains only one molecule of the Cu/L1 complex. Furthermore, these observations are consistent with the results of the variable time normalization analysis (VTNA)^[^ 67 ^]^ (Figure 4c–e), which revealed that the reaction is first order for the Cu catalyst and sulfonimidoyl chloride rac‐1a. and zero‐order for diol 2.

Mechanistic studies of the kinetic resolution reaction. a). Dependence of the ee of sulfonimidate 3a (●) and recovered sulfonimidoyl chloride 1a (■) (eeprod) on the ee of the catalyst (eecat). b). Dependence of the experimental (■) and theoretical (●) selectivity factors s on the ee of the catalyst. The experimental s values were calculated as shown in Table 1. The theoretical s values were calculated as a function of eecat: s = [s“(1 + eecat)–(1–eecat)]/[(1 + eecat) + s”(1–eecat)] for s' = 83 obtained with enantiomerically pure L1. c). Variable time normalization analysis (VTNA) for sulfonimidoyl chloride rac‐1a (■, 0.08 M; ●, 0.1 M). d). VTNA for diol 2 (▲, 0.17 M; ◆ 0.25 M). e). VTNA for the catalyst (▼, 7 mol%; ●, 10 mol%). f). X‐ray crystal structure of the catalyst–diol adduct Cu(ent‐L1)(κ 2O‐2)(κ 1O‐2)[OTf]2, 21. Hydrogen atoms and triflate anions are omitted for clarity.

The nature of the catalyst in the resting state was next studied by means of diffusion‐ordered nuclear magnetic resonance spectroscopy (DOSY NMR).^[^ 68 ^]^ This method can reveal the complexation behavior of solutes by experimentally correlating their diffusion coefficients and molecular weights. The experimentally determined molecular weight of the Cu/L1‐containing species (MW = 851.8, ±2.3%) was consistent with that of a 1:2 (Cu/L1)–diol 2 complex (MW = 832.5).

The Structural Identity of the Catalytically Active Species was Further Probed by X‐Ray Crystallographic Analysis

Crystallographic studies revealed a (Cu/ent‐L1)–diol 2 complex with a 1: 2 Cu/diol 2 stoichiometry (21, Figure 4f), in accordance with the solution‐phase DOSY NMR data. Complex 21 features a copper center that is chelated by diol 2 and ent‐L1, with the second loosely bound (d Cu–O = 2.18 Å) monocoordinated diol 2 ligand occupying the apical position.

Taken together, the results of these experimental studies are consistent with a mechanism that involves the formation of complex 21 that, upon dissociation of the apical κ ^1^O‐diol 2 and subsequent deprotonation with the base, can produce Cu(II) dialkoxide intermediate 22 (Figure 5a). The subsequent stereoselective reaction favors ent‐1a and affords Cu‐bound sulfonimidate 23. Product 3a is then released via ligand exchange with diol 2. The byproduct chloride is sequestered by Ag^I^, which prevents deactivation of the dicationic Cu^II^ catalyst by the negatively charged halide ligand.

*Mechanistic computational studies of the kinetic resolution reaction of sulfonimidoyl chlorides. a). Catalytic cycle of kinetic resolution. b). Transition state structures TS‐R and TS‐S for the Cu‐catalyzed nucleophilic displacement process. c). Distortion–interaction analysis of transition state structures TS‐R and TS‐S, ΔE, kcal mol−1. d). Energy decomposition analysis of transition state structures TS‐R and TS‐S, ΔΔE‡  = ΔE‡

TS‐R –ΔE‡

TS‐S , kcal mol−1. e). Contribution of the individual L1 fragments to the dispersion component of the substrate–catalyst interaction in TS‐R and TS‐S, ΔΔE‡

disp = ΔE‡

disp(

TS‐R

)–ΔE‡

disp(

TS‐S

), kcal mol−1. f). Complementary occupied–virtual pairs (COVPs) in the β space for the n O(

ent

‐

1a

)→σ* 22 interaction in TS‐R.*

Density functional theory (DFT) analysis was carried out next to gain insight into the origin of the enantioselectivity in the Cu‐catalyzed nucleophilic displacement reaction (Figures 5b–f and S9–S19). The reaction of complex 22 with the racemic sulfonimidoyl chloride substrate proceeds over accessible barriers and favors TS‐R (ΔG^‡^ * = 10.5 kcal mol^−1^ for TS‐R and 13.5 kcal mol^−1^ for TS‐S; ΔΔG^‡^ * = 3.0 kcal mol^−1^; Figure 5b). The stereochemical outcome of the pathway that traverses the more favorable transition state structure TS‐R produces sulfonimidate 3a from sulfonimidoyl chloride ent‐1a, which can result in the accumulation of less reactive enantiomer 1a, in accordance with the experimentally observed outcome of the kinetic resolution process.

Furthermore, distortion–interaction activation strain model (ASM) analysis^[^ 69 ^]^ revealed that both TS‐R and TS‐S exhibit greater distortion in the substrate fragment caused by structural adjustment at the sulfur center, which is required to accommodate the incoming nucleophile (Figure 5c). However, the analysis also indicated that both transition state structures exhibit similar levels of distortion and that the enantioselectivity is largely driven by the substantially stronger interactions in TS‐R.

Energy decomposition analysis (EDA)^[^ 70 ^]^ confirmed that TS‐R benefits from stronger attractive interactions, particularly dispersion and charge transfer (orbital interactions between the fragments), which compensate for greater Pauli (steric) repulsion (Figure 5d).

Analysis of noncovalent interactions (NCIs) using the independent gradient model based on Hirshfeld partition (IGMH)^[^ 71 ^]^ was performed next to clarify the origin of these interactions and the roles of the L1 trans‐diphenyloxazoline and indane moieties in the preferential stabilization of TS‐R (Figure 5b). In broad agreement with the results of EDA, NCI analysis revealed that TS‐R benefits from more extended stabilizing dispersion interactions between the sulfonyl residue of the substrate and both the trans‐diphenyloxazoline and the indane fragments of the ligand. trans‐Diphenyloxazoline fragment is also engaged in stabilizing H···π interactions with the aryl group in the sulfonimidoyl residue of the substrate. These stronger interactions compensate for the Pauli repulsion between the α‐C─H in the oxazoline moiety and the arenesulfonimidoyl residue in TS‐R, as revealed by EDA. Additionally, an analysis of the contributions of individual fragments in ligand L1 (proximal and distal phenyl rings and indane and bisoxazoline residues) to the dispersion interactions with the substrate in TS‐R and TS‐S indicated that the indane moiety provided a decisive contribution to the stabilization of TS‐R (Figure 5e). Furthermore, in line with the results of EDA, an analysis of complementary occupied–virtual pairs (COVPs)^[^ 72 ^]^ also revealed that TS‐R is stabilized by stronger charge transfer interactions. In particular, TS‐R benefits from an n O( * ent * ‐ ** 1a ** )→σ*** 22 ** interaction between the oxygen lone pair in the sulfonyl group of the substrate and a σ* for the Cu─N/O_(_ ** 22) ** and the proximal C─H bond in the indane methylene group, which was not observed in TS‐S (Figure 5f).

Lastly, effective oxidation state (EOS)^[^ 73 ^]^ analysis indicated that both complex 22 and transition state TS‐R feature a Cu^II^ metal center, which rules out the involvement of metal‐mediated single electron transfer processes.

Collectively, the results of these studies point to an intricate synergy of dispersion and charge transfer (orbital) effects in imparting high enantioselectivity via the trans‐diphenyloxazoline and indane moieties of the Cu/L1 catalyst. Furthermore, they underscore the importance of substrate interactions with remote and unfunctionalized structural elements of a chiral catalyst for achieving efficient enantiocontrol in kinetic resolution reactions.

Conclusion

In summary, we developed a previously unexplored catalytic kinetic resolution approach for sulfonimidoyl chlorides that leverages the electrophilic reactivity of S(VI) stereocenters.^[^ 74 ^]^ The approach provides a blueprint for the expansion of kinetic resolution strategies to S(VI) chirality via mechanistically distinct transformations that complement the well‐established nucleophilic reactivity‐based approaches to sulfoximines and sulfonimidamides. Furthermore, the method provides catalytic access to synthetically challenging sulfonimidate esters that have been conventionally generated by the oxidative conversion of chiral S(IV) precursors. Additionally, experimental studies revealed the structural effects and mechanistic details of the copper‐catalyzed asymmetric process. Further mechanistic insight was gleaned from computational investigations of the key catalytic nucleophilic displacement step. These studies revealed the critical role of noncovalent interactions between the substrate and the unfunctionalized backbone of the chiral ligand in achieving the experimentally observed high enantioselectivity. Taken together, these studies suggest future avenues for the expansion of nascent kinetic resolution strategies to heteroatom chirality by harnessing mechanistically unexplored reactivity in well‐defined molecular settings.

Conflict of Interests

The authors declare no conflict of interest.

Supporting information

Supporting Information

Supporting Information