Efforts have been dedicated to the development of methodologies that enable efficient and general synthesis of valuable molecules using readily accessible substrates, while ensuring atom economy. Among these methodologies, sulfenylamination of alkenes [1–9] offers a direct and rapid route to assemble β-amino sulfides, which are prevalent in bioactive compounds (e.g., molecules I-IV, Scheme 1a) [10–13]. However, this transformation remains challenging, arising from the intrinsic reaction patterns of the two established strategies. The first one involves the utilization of thiiranium intermediates, where alkenes act as nucleophiles and subsequently undergo ring-opening with a separate nucleophile species (Scheme 1b, left) [3–6]. As an alternative, a few examples have explored the addition of thiyl radicals to alkenes, followed by single-electron oxidation and coupling with an additional nucleophile (Scheme 1b, right) [7–9]. These strategies rely on a three-component platform involving alkenes, sulfur (S), and nitrogen (N) sources. Consequently, such approaches suffer from inconvenient manipulations and poor atom economy. Furthermore, the generation of key intermediates like thiiranium or carbon cations, which are crucial for the difunctionalization process, restricts the use of electron-neutral alkenes. To the best of our knowledge, no reports have demonstrated the application of electron-deficient alkenes in sulfenylamination. Additionally, the sulfenylamination of α,α-disubstituted electron-neutral alkenes, which are valuable for constructing highly congested amines, is underdeveloped [9]. This can be attributed to the challenges posed by aza-nucleophilic coupling due to steric hindrance, as well as the propensity of in-situ-formed carbon cations for 1,2-elimination. In this context, there is a compelling need to develop a novel strategy that enables the generic sulfenylamination of alkenes in an atom-economical manner, accommodating a wide range of alkene substrates. Such a breakthrough represents a highly desirable pursuit for the scientific community.

In light of these challenges, we proposed a potential solution by investigating the feasibility of double radical-involved sulfenylamination, whereby the addition of thiyl radicals to the C=C bond of alkenes would lead to the formation of a carbon radical that could readily couple with nitrogen radicals (Scheme 1c). This tentative scenario is based on the understanding that radical couplings typically occur with minimal activation energy [14–18], thereby avoiding the formation of competing byproducts via carbon cation intermediates. Considering that the α-carbon radical adjacent to electron-withdrawing groups is electrophilic, it is crucial to functionalize the nitrogen radical with electron-donating groups to enhance its nucleophilic ability [19,20]. This approach aligns with the polar-matching rule, facilitating the efficient recombination of these electronically distinct radicals. Motivated by this concept, we became intrigued by the prospects of utilizing benzenesulfenamides as difunctional partners with alkenes to achieve the desired β-amino sulfides. Notably, since the 1970s [21–26], benzenesulfenamides have been sporadically employed in sulfenylamination of alkenes using Lewis acids as promoters. However, these approaches were limited by the use of thiiranium ions as key transition states, resulting in a severe restriction on the applicable alkene substrates and significantly limiting their synthetic utility. To address this limitation and ensure the generation of both nitrogen and sulfur radicals from benzenesulfenamides, we explored the prospect of harnessing sustainable photocatalytic energy transfer (EnT) for the homolytic cleavage of the N-S bond [27–37]. Notably, the direct irradiation conditions posed challenges due to the potential instability of the reaction products. Additionally, the low oxidation potential of benzenesulfenamides raised the possibility of heterolytic cleavage of the N-S bond via single-electron oxidation (SET) of the nitrogen atom by triplet photosensitizers. This SET pathway could lead to an intramolecular rearrangement on the benzene ring (Scheme 1c, bottom) [25,26], underscoring the need for addressing the chemoselectivity issue between EnT and SET pathways. Furthermore, considering the distinct preferences of electron-neutral and electron-deficient alkenes for specific nitrogen and sulfur radicals, achieving regioselectivity in the sulfenylamination process remains elusive [29–37]. These challenges highlight the complexity and importance of developing a novel methodology that addresses these issues, enabling efficient sulfenylamination with broad substrate scope and high regioselectivity.

Motivated by the potential of sulfenamides and the prospects of energy transfer (EnT) synthesis, we set out to test the aforementioned hypothesis. To explore the construction of challenging amines adjacent to congested carbons, we selected N-(4-methoxyphenyl)-S-phenylthiohydroxylamine (1a) and α,α-disubstituted prop–1-en-2-ylbenzene (2a) as our model substrates (Table 1). To begin, we evaluated the electrochemical properties of 1a, finding that it had an Ep value of +0.76 and +1.61 vs. SCE in CH₃CN, along with an E_T = 37.2 kcal/mol (See Supporting information for details). Additionally, we determined the bond dissociation energy (BDE) of the N-S bond in 1a to be 32.8 kcal/mol (see Supporting information). These results suggested the feasibility of N-S bond homolysis under excitation by energy transfer. Subsequently, we tested our self-developed photosensitizer, DPZ (E_t(S*/S^•‒) = +1.42 V vs. SCE in CH₃CN, E_T = 46.4 kcal/mol) [38–40], which has demonstrated widespread utility in visible light-driven photocatalytic reactions [41,42]. Upon irradiation with a 3 W blue LED (null = 456 nm), we obtained the desired product 3a in only 5% yield (entry 1, Table 1). In an effort to improve the yield, we screened various solvents (entries 2–5), ultimately finding that toluene led to a yield of 44% (entry 2). Notably, no byproducts such as 4, which could form through the intramolecular rearrangement of 1a, were detected. In addition, we tested several other commonly used photosensitizers, including Ir[dF(CF₃)ppy]₂(dtbbpy), EosinY, and 4CzIPN (entries 6–8). While the Ir complex resulted in a 65% yield of 3a (entry 6), we focused our attention on viable dicyanopyrazine-type organophotosensitizers for their clean and environmentally friendly attributes. It is worth mentioning that our ongoing research project involves the synthesis of a series of dicyanopyrazine derivatives by substituting 2-methoxythienyl with other aromatic groups to explore their feasibility as photosensitizers [38]. These derivatives can be conveniently prepared through a one-step synthesis. DPZ was chosen as our initial preference due to its higher catalytic efficiency compared to the others in these specific transformations. To further investigate the potential for catalytic ability, we carefully reexamined various derivatives of DPZ, including TDPZ, FDPZ, PDPZ, 4-FPDPZ, and 4-MOPDPZ (entries 9–13). Encouragingly, TDPZ (E_t(S*/S^•‒) = +1.90 V vs. SCE in CH₃CN, E_T = 51.1 kcal/mol) provided 3a in a 70% yield, while the other four derivatives were ineffective. Considering that null of TDPZ is 391 nm, we utilized a 3 W LED with null of 399 nm, which significantly improved the yield of 3a to 90% (entry 14). Higher energy photons were found to degrade the yield (entry 15). When the transformation was conducted in the absence of TDPZ, minimal product 3a was obtained (entry 16). No reaction was observed in the dark, underscoring the indispensability of light (entry 17). Finally, the transformation was performed in the presence of an ambient atmosphere, where 3a could not be detected while with only 1a completely depleted, most likely caused by the low energy of triplet oxygen leading to various side reactions (entry 18).

Following the establishment of optimized reaction conditions, we investigated the substrate scope of this two-component sulfenylamination protocol (Scheme 2). Initially, we tested the reaction of 2a with various benzenesulfenamides bearing S-aryls with diverse electron-withdrawing and electron-donating groups. This resulted in the formation of products 3b-3g with yields ranging from 53% to 71%. The protocol also tolerated sulfenamides with azaaryls instead of aryls as the S-substituent, as demonstrated by the formation of 3h in a 47% yield. Furthermore, the method proved to be compatible with different aliphatic sulfides, as indicated by the successful synthesis of 3i-3k. It is worth noting that the choice of the 4-methoxyphenyl (PMP) group as an N-protecting group serves two purposes: enhancing the nucleophilicity of the putative amino radical and providing an easily removable group. Nevertheless, we explored the substitution of PMP with other aryls, and products 3l-3m exhibited good compatibility with respect to substitutions on the N-aryl group. Next, we investigated various electron-neutral alkenes using 1a as the reaction partner. Notably, a wide range of α-methyl styrenes were well tolerated regardless of electronic properties or substitution patterns on the aromatic rings. This resulted in the formation of products 3n-3w in yields ranging from 55% to 89%. Furthermore, the high chemical yields observed for alkenes with fused aromatic (3x) and heteroaromatic (3y-3zb) ring substituents highlight the generality of this method. We also tested styrenes bearing distinct 2-alkyl substituents, leading to satisfactory yields of products 3zc-3zh. In the case of substrates containing two alkene moieties (3zf-3zg), sulfenylamination showed a preference for the more electron-rich one. Additionally, when 2-alkyls were replaced by 2-aryls, the reaction proceeded with higher reactivity and chemoselectivity (3zi). Simple styrene proved to be a valuable starting material for synthesizing biologically important aryl sulfides with a tertiary carbon atom adjacent to the amine group, as exemplified by the formation of products 3zj and 3zk in yields of 78% and 54%, respectively. Moreover, it is worth noting that a one-step deprotection of PMP in 3zk can provide access to a class of bioactive compounds, namely Na_v 1.4 channel blockers (molecule I, Scheme 1a). In addition to styrenes, two different 1,3-dienes reacted successfully with 1a, yielding adducts 3zl-zm with satisfactory yields. Notably, the reaction with terminal alkenes (3zm) displayed higher reactivity. Finally, exocyclic C=C bonds, including 5-, 6-, or hetero-rings, were also successfully transformed into sulfenylaminated products (3zn-3zp) with yields ranging from 43% to 55%. These results highlight the broad substrate scope of the developed sulfenylamination protocol, demonstrating its versatility in synthesizing a variety of valuable products.

Building on the promising results, we expanded the scope of this catalysis platform to investigate the efficiency with electron-deficient alkenes, represented by substrate 5 (Scheme 3). We initially focused on 2-arylacrylates due to their potential in directly synthesizing valuable α-amino acid derivatives that possess a structurally congested amine and a β-thioether motif (e.g., molecule II, Scheme 1a). Encouragingly, products 6a-c were obtained with satisfactory yields, regardless of the alkyl and aryl substituents on the esters. Considering the prevalence of imine-containing azaarenes in pharmaceutical compounds and their electron-withdrawing nature [43], we explored the sulfenylamination of 1a with various vinylazaarenes containing α-aryls or α-alkyls, as well as diverse valuable azaarenes. Consequently, products 6d-j were obtained in yields ranging from 48% to 95%. The broad applicability of electron-withdrawing groups to olefins inspired us to further investigate a specific amide-based olefin, as its corresponding product 6k could be conveniently deprotected to yield a modulator of the androgen receptor (molecule II, Scheme 1a). Furthermore, product 6l demonstrated the potential of this strategy to assemble alkyl sulfides on these valuable tertiary amines. These findings highlight the robustness and versatility of this catalytic system in facilitating the synthesis of intricate molecules containing electron-deficient alkenes. The successful formation of structurally diverse products further emphasizes the potential of this method in accessing biologically relevant compounds and expanding the synthetic toolbox for complex molecule synthesis.

To assess the generality and potential synthetic applications of this novel methodology, the ability for late-stage functionalization of complex bioactive molecules was investigated, as depicted in Scheme 4a. The objective was to evaluate its versatility in accessing a greater variety of valuable compounds that could be useful for drug development. In this investigation, we successfully assembled estrone and pregnenolone, chosen as representative molecules, onto products 7–9 by reacting 1a with the corresponding electron-neutral and electron-deficient alkenes. This demonstrated the capability of the method to perform late-stage functionalization on complex bioactive molecules. Furthermore, to further confirm the promising synthetic applications of this method, we selected product 5a and subjected it to oxidation using 3-chloroperoxybenzoic acid (m-CPBA) to yield the corresponding sulfone. The N-PMP group in the sulfone was then smoothly cleaved using cerium ammonium nitrate (CAN), resulting in the synthesis of sulfone-based α-amino ester 10 in 55% yield over two steps (Scheme 4b). This demonstrates the efficient access to pharmaceutically important molecules I-II (Scheme 1a) using this methodology. Additionally, this strategy has the potential to address the incompatibility issues observed with N-alkyl benzenesulfenamides in the photocatalytic difunctionalization approach. These results highlight the broad synthetic applications and potential for late-stage functionalization of complex bioactive molecules using this methodology. The ability to access pharmaceutically important compounds and overcome compatibility issues further underscores the value and utility of this novel catalytic system in drug development and synthetic chemistry.

In order to gain insight into the mechanistic aspects of this strategy, extensive mechanistic studies were conducted. Initially, control experiments (Table 1, entries 16 and 17) indicated that TDPZ serves as the photosensitizer in the photocatalytic sulfenylamination. However, considering the photochemical and photophysical properties of TDPZ, the photoactivated TDPZ (*TDPZ) could potentially initiate the reactions through either single-electron transfer (SET) or energy transfer (EnT) with benzenesulfenamides. This hypothesis is further supported by comparing the redox potentials and triplet energies of the species involved, as well as the results of the Stern-Volmer experiment. To elucidate the actual excitation mode, a series of elaborate studies were conducted. Firstly, when 1a was subjected to the established reaction conditions without the presence of alkenes, product 4 was obtained in an 84% yield. This result suggests the occurrence of a competing chemical transformation involving SET of benzenesulfenamides to *TDPZ (Scheme 5a). These findings provide important insights into the mechanistic aspects of this strategy. They indicate that the photoactivated TDPZ can engage in either SET or EnT processes with benzenesulfenamides, thereby initiating the desired transformations. Further investigation and characterization of the reaction intermediates and excited states are necessary to obtain a comprehensive understanding of the underlying mechanisms.

We then conducted several experiments to probe the involvement of radicals in the difunctional sulfenylamination process. Firstly, we attempted the transformation of 1a with 2a in the presence of TEMPO as a radical trapping reagent (Scheme 5b, in red). The detection of the trapped products 11 and 11′ by high-resolution mass spectrometry analysis confirmed the formation of thiyl radicals and their viability to undergo addition to olefins. Interestingly, the desired product 3a was not obtained, indicating that the presence of alkenes leads to a different interaction pattern between the photosensitizer and 1a, compared to the formation of 4. We further added 2.0 equiv. of BHT, a radical scavenger, to the reaction. While 1a was nearly consumed, only a small amount of product 4 was detected, and the yield of 3a was significantly reduced to 24% (Scheme 5b). This indicates that the difunctional sulfenylamination process involves radical intermediates. To explore the source of radicals, we examined the transformation using various polycyclic aromatic hydrocarbons as energy scavengers for the photoactivated sensitizers (Scheme 5c). Interestingly, it was found that naphthalene, which has a higher triplet energy than DTPZ, did not affect the transformation (entry 1), while 9,10-diphenylanthracene, with an energy similar to that of 1a, completely suppressed the generation of 3a while 1a was consumed (entry 4). We also investigated the transformation using other photosensitizers, and the results are summarized in Scheme 5d. Notably, two Ir complexes with poor oxidizing ability but sufficient triplet energy could yield product 3a (entries 1 and 2). In contrast, Eosin Y, with a lower E_T of 43.6 kcal/mol, failed to generate 3a (entry 4). These findings suggest that the reaction relies on the photoexcitation of the sensitizers to generate radicals. In a crossover experiment with two different benzenesulfenamides (1d and 1b) under the reaction conditions, after 30 h, two cross-coupling products (1e and 1f) were obtained (Scheme 5e). Additionally, a radical clock experiment was performed using molecule 12, which contains an alkene and a cyclopropane ring. As shown in Scheme 5f, a difunctionalized product (13) with the opened cyclopropane ring was obtained in 51% yield, suggesting the formation of both N- and S-radicals in the reaction system. It is worth mentioning that the different chemoselectivity between 12 and (1-cyclopropylvinyl)benzene (3zh, Scheme 2) probably originates from the considerable instability of the primary carbon radical produced from the latter. It thus readily undergoes intramolecular addition to yield the more stable benzylic radical, which just possesses a similar reactivity to the benzylic radical generated from 12. In addition, a light on/off experiment was performed, which showed that a radical chain process might be impracticable. Based on these results, a plausible mechanism involving homolysis of the N-S bond through EnT has been proposed (Scheme 5g). In addition to EnT-enabled homolysis, a sequential process of radical addition and radical coupling is responsible for the final formation of the sulfenylamination products. The excellent chemoselectivity observed in this reaction can be attributed to the kinetic preference of the radicals generated via EnT for addition to alkenes and/or cross-coupling, compared to the radical cations generated via SET that may undergo rearrangements. The efficiency of EnT between the photosensitizer and benzenesulfenamides is crucial for the overall reactivity of the system. These mechanistic insights provide a deeper understanding of the radical-involved difunctional sulfenylamination process and contribute to the development of this efficient and chemoselective synthetic methodology.

To further substantiate the proposed reaction pathways and elucidate the precise regioselectivity (Scheme 5g), density functional theory (DFT) calculations were then performed. As depicted in Fig. 1, the mechanism of the reaction can be reasonably postulated to commence with a triplet-triplet EnT event between triplet excited state of photosensitizer TDPZ and the singlet state of sulfenamide reagent 1a. Subsequently, a homolysis cleavage of the triplet excited state 1a* occurs via TS1 (0.8 kcal/mol), leading to the generation of S-radical and N-radical [44]. The ensuing step involves the active S-radical undergoing addition to the terminal carbon of styrene 2a via TS2, affording an addition intermediate II with a free energy barrier of 5.1 kcal/mol. In contrast, a parallel radical addition involving the N-radical and styrene 2a via TS3 necessitates surmounting a notably higher free energy barrier of 13.4 kcal/mol. Ultimately, a radical-radical cross coupling event between intermediate II and the N-radical via TS4 gave the final product 3a with a free energy barrier of 6.6 kcal/mol. On the other hand, another analogous radical coupling between intermediate III and the S-radical via TS5 necessitates overcoming a significantly higher free energy barrier of 19.6 kcal/mol. Considering the substantial disparity in free energy between TS2 and TS3, as well as between TS4 and TS5, it is apparent that the formation of the product 3a is kinetically favored. This rationalization effectively accounts for the remarkable regioselectivity that has been observed in this difunctional transformation.

In conclusion, we have developed a novel two-component sulfenylamination of alkenes [45–48] by precisely activating readily prepared sulfenamides through photocatalytic energy transfer. This approach offers several advantages, including high atom economy and exceptional chemoselectivity, making it compatible with a wide range of substrates. Notably, both electron-neutral and unprecedented electron-deficient alkenes can be efficiently incorporated using this catalytic platform. As a result, a diverse array of biologically important β-amino sulfides with secondary or tertiary carbon α to the amine moiety were obtained in high yields. The transition-metal-free nature of this approach holds great potential for the facile and efficient preparation of sulfur-containing α-amino acid derivatives and azaarene-based amines, which are crucial for drug development. The demonstration of sulfenamides’ ability to generate both N- and S-radicals through energy transfer opens up promising avenues for utilizing these valuable building blocks in radical-based chemical synthesis. Furthermore, this methodology can contribute to the synthesis of a wider range of important N- and/or S-containing molecules. Overall, our developed sulfenylamination strategy showcases a versatile and efficient approach for the incorporation of sulfenamides in synthetic chemistry, with broad applications in pharmaceutical and drug discovery research.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Er-Meng Wang: Investigation, Methodology. Ziyi Wang: Investigation, Methodology. Xu Ban: Writing – review & editing. Xiaowei Zhao: Data curation, Funding acquisition. Yanli Yin: Conceptualization, Funding acquisition, Supervision, Writing – original draft, Writing – review & editing. Zhiyong Jiang: Conceptualization, Funding acquisition, Supervision, Validation, Writing – original draft, Writing – review & editing.

We are grateful for financial support from the National Natural Science Foundation of China (Nos. 22171072, 21925103, and 22301061).

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.109843