Xiang-Xin
Zhang
ab,
Hao
Zheng
ab,
Yong-Kang
Mei
ab,
Yan
Liu
ab,
Ying-Ying
Liu
ab,
Ding-Wei
Ji
a,
Boshun
Wan
a and
Qing-An
Chen
*ab
aDalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, People's Republic of China
bUniversity of Chinese Academy of Sciences, Beijing 100049, People's Republic of China. E-mail: qachen@dicp.ac.cn; Web: http://www.lbcs.dicp.ac.cn
First published on 25th September 2023
A catalyst-free photosensitized strategy has been developed for regioselective imino functionalizations of alkenes via the formation of an EDA complex. This photo-induced protocol facilitates the construction of structurally diverse β-imino sulfones and vinyl sulfones in moderate to high yields. Mechanistic studies reveal that the reaction is initiated with an intermolecular charge transfer between oximes and sulfinates, followed by fragmentation to generate a persistent iminyl radical and transient sulfonyl radical. This catalyst-free protocol also features excellent regioselectivity, broad functional group tolerance and mild reaction conditions. The late stage functionalization of natural product derived compounds and total synthesis of some bioactive molecules have been demonstrated to highlight the utility of this protocol. Meanwhile, the compatibility of different donors has proved the generality of this strategy.
Considering the current problems, an alternative strategy is the direct photoexcitation of an electron donor–acceptor (EDA) complex. It is defined as a new molecular aggregate formed by an electron acceptor substrate A and a donor molecule D through non-bonding interactions.7 The two components A and D may not absorb visible light themselves, but their in situ formed EDA complex does. And subsequent intramolecular single-electron transfer (SET) is triggered to generate radical intermediates under the excitation of visible light.8 With regard to photochemical reactions involving EDA complexes, there are two reaction modes. On the one hand, the free radicals generated by the two substrates respectively could occur a direct radical–radical coupling reaction.9 On the other hand, the free radical intermediates produced by the EDA complex are reacted by free radical trapping reagents.5a,10 However, only a few protocols are able to incorporate the donor D and acceptor A motifs simultaneously onto the radical traps (Fig. 1C).4a,10g,11
Based on this point and our group's research on olefin functionalization,12 we wonder whether the two free radicals produced by the EDA complex can both be utilized with alkenes as trap reagents. The studies by Narasaka,13 Walton,14 Leonori,5 Yang6j and Glorius6a,d,6g,i have demonstrated oximes15 as good nitrogen radical precursors. The β-amino sulfones display unique pharmaceutical activities that render them as good candidates in drug discovery research. Then we speculated that the electron-poor O-aryl oximes and electron-rich sulfinates could form EDA complexes and enable access to sulfonyl radicals and iminyl radicals upon exposure to visible light. Herein, we report a catalyst-free intermolecular radical sulfonylimination of alkenes via photochemical EDA activation. Preliminary mechanistic studies suggest that the initial radical addition of the reactive transient sulfonyl radical with alkenes should be kinetically favored compared to that of the persistent iminyl radical. Meanwhile, this strategy can also be applied to other donors to construct structurally different β-amino compounds.
Entry | Deviation from standard conditions | 4a (%) |
---|---|---|
a Conditions: 1a (0.12 mmol), 2a (0.10 mmol), 3a (0.18 mmol), MeCN/DMSO (9:1, 1.0 mL), rt, 18 h. b Determined by HPLC with anthracene as the internal standard. c Isolated yield. | ||
1 | None | 88(82)c |
2 | Without light | — |
3 | R = Me or H | — |
4 | 390 nm | 76 |
5 | 427 nm | 78 |
6 | 456 nm | 65 |
7 | MeCN | 74 |
8 | DMSO | 46 |
9 | THF | 21 |
10 | DCM | 5 |
11 | MeCN/DMSO = 2:1 | 77 |
12 | MeCN/DCM = 9:1 | 73 |
With the optimized conditions in hand, we systematically explored the substrate scope of this sulfonylimination approach (Fig. 2). A wide range of styrene derivatives all underwent the sulfonylimination reaction smoothly, delivering the corresponding β-imino sulfones with excellent regioselectivity. Styrenes bearing electron-donating or electron-withdrawing substituents at different positions of the aromatic rings were tolerated (4a–4j). Halogen and chloromethyl groups that are susceptible in an EDA reaction system could be well compatible in this protocol. Other terminal aryl alkenes, such as 2-vinylnaphthalene and 2-vinylpyridine could also be applicable (4k and 4l). Excitingly, 1,3-enyne could be easily accommodated, leaving the alkyne moiety untouched (4m). Next, a series of disubstituted and multi-substituted alkenes were investigated under our sulfonylimination conditions. 1,1-Disubstituted alkenes with different electronic properties and steric hindrance could provide β-imino sulfones with a tertiary alkylamine centre (4n–4r). The desired products (4s–4w) could also be obtained from the coupling of different 1,2-disubstituted and multi-substituted alkenes. Notably, high regio- and diastereoselectivities were obtained for unsymmetrical internal alkenes. The anti-geometries between sulfonyl and iminyl groups were confirmed by single crystal analysis of 4s (CCDC: 2224155) and 4t (CCDC: 2224158). Gratifyingly, two consecutive distinct quaternary carbon centers could be constructed simultaneously (4w). Subsequently, a series of sulfinates and O-aryl oximes were evaluated with 4-tert-butylstyrene. Moderate to excellent yields were obtained for phenylsulfinates with different substituents at the para position (5a–5g). This protocol was also compatible with other aromatic groups, such as naphthyl (5h), thienyl (5i), and pyridyl (5j). Besides, alkyl sulfinates also worked in this protocol with high reactivities and regioselectivities (5k–5m). It is worth noting that the cyclopropyl group could remain intact in the target product (5m). O-Aryl oximes derived from other diaryl ketones were also well accommodated in this sulfonylimination (5n and 5o).
Fig. 2 Substrate scope of sulfonylimination of alkenes. Conditions: 1 (0.12 mmol), 2 (0.10 mmol), 3 (0.18 mmol), MeCN/DMSO (9:1, 1.0 mL), rt, 18 h. Isolated yields are given in all cases. |
The synthetic utilization of this protocol in late-stage functionalization was further evaluated through direct sulfonylimination of a range of complex alkene substrates (Fig. 3a). Natural products and drug-derived alkenes reacted smoothly to deliver the desired products in good yields (6a–6d). The ester groups, cyclic olefin and amides of complex olefins could be well maintained in this protocol. Ultimately, the synthesis of pharmaceuticals has been conducted to further demonstrate its synthetic utility.
Generally, de novo synthesis is used to synthesize drugs containing an amino-sulfone functionality. For changing key functional groups in drugs such as sulfonyl groups, it is often necessary to replace the corresponding substrate and re-synthesize from scratch. This greatly slows down the synthesis efficiency of drug molecules. Consequently, achieving modular and streamlined synthesis of drug molecules is critical for the rapid expansion of drug libraries. This sulfonylimination strategy was demonstrated efficaciously, which allowing different drugs to be synthesized in a concise manner (10, 15, and 21) with decent overall yields (35–53%, Fig. 3b). Only 3–4 steps were required to create these high functionalized molecules from simple starting materials. For drugs 15 and 21, the reaction conditions for the introduction of amino and sulfone groups in traditional synthesis are relatively harsh (strong acid and strong base).2e,i Meanwhile, these groups can't be introduced at the same time. In the synthesis of these three drugs, the construction of key alkenes, the functionalization of alkenes, and subsequent simple modifications could bypass the one-to-one correspondence between drugs and starting substrates, and could efficiently synthesize drugs in a modular and gentle manner. In addition, this sulfonylimination strategy provided a way to quickly enrich the drug library, by only simply changing different oximes and sulfinates in a desired step.
To gain insight into the sulfonylimination mechanism, we performed a series of UV-vis spectroscopic measurements. An obvious increase in absorption intensity (400–500 nm) was observed by the addition of colorless 3a into 1a, accompanied by a certain degree of yellow color deepening. To further investigate the interaction between 1a and 3a, a series of experiments were performed. The Job's plot measurement16 revealed a 1:1 stoichiometry of the corresponding EDA complex (Fig. S2 in the ESI†). In addition, NMR titration further revealed the presence of weak interaction between 1a an 3a (Fig. S3 in the ESI†). And the association constant KEDA between them was calculated to be 3.65 M−1 using the Benesi–Hildebrand method17 (Fig. S4 in the ESI†). This was proposed to result from the formation of an EDA complex (Fig. 4a). Next, free radical inhibition and capture experiments were performed. The addition of radical scavenger TEMPO (2,2,6,6-tetramethylpiperidinoxy) to the reaction mixture completely suppressed the reactivity. And in the presence of BHT, sulfonyletherification of olefin (7a:7b = 10:1) rather than the desired product 4a was formed mainly (Fig. 4b). Radical clock experiments with different cyclopropanes led to the corresponding ring-opening products 23 and 25 (Fig. 4c). And the products contained only sulfone functional groups, indicating that the order of addition of sulfone groups was preferential. These results hinted towards the radical nature of the reaction. Finally, different competitive experiments were carried out separately (Fig. 4d). An evaluation of the reactivities of sulfinate 3a and 3a′ gave the corresponding products 4a and 4a′ with 49% and 50% yield, respectively. This indicated that the overall rate of the generation of corresponding free radicals is probably comparable between aryl and alkyl sulfinates via EDA activation. When different alkenes reacted simultaneously with sulfinate 3a, more electron-rich product 4c was more inclined to be formed over 4f. This perhaps resulted from the stabilization of carbon radicals by increasing electron density.
Based on the aforementioned mechanistic studies and further DFT calculations, a plausible reaction mechanism is shown in Fig. 4e. First, the EDA complex 1A is formed via Coulomb interaction and π–π stacking between oxime 1a and sulfinate 3a. The existence of complex 1A is further supported by time-dependent density functional theory (TDDFT),9c,18 which indicates that the excitation results in a single electron transfer (SET) from 3a to 1a. Then under the irradiation of visible light, the complex 1A in the ground state is excited to give the first singlet excited state 1A*. The SET promotes the homolytic N–O bond cleavage of 1a. There are two possible mechanisms for N–O cleavage. It may occur in the first singlet excited state, directly giving persistent iminyl radical B (ref. 19) and ion pair intermediate C. Alternatively, the singlet 1A* undergoes intersystem crossing (ISC),20 reaching the triplet 3A, followed by N–O bond cleavage with a barrier of 4.2 kcal mol−1 (TS1). Although it is hard to determine which mechanism is actually involved, it is certain that iminyl radical B is generated under blue light irradiation. Subsequently, the resulting intermediate C undergoes intramolecular single electron transfer to deliver transient sulfonyl radical E. Through transition state TS2 (ΔG‡2 = 8.4 kcal mol−1), the active radical E attacks alkene 2a to produce C-centered radical F. In contrast, kinetically unfavorable radical additions between iminyl radical B, the sulphinic radical and alkene 2a need to overcome a higher free energy barrier, 17.2 kcal mol−1 (viaTS2N) and 24.4 kcal mol−1 (viaTS2O), respectively. Finally, this C-centered radical F and the long-lived iminyl radical B participate in a radical–radical cross-coupling process to yield the desired β-imino sulfone product 4a. Considering the large free energy difference between TS2 and TS2N, the formation of isomer 4aN can be kinetically suppressed, which theoretically explains the excellent regioselectivity of this transformation.
In order to test the generality of this EDA strategy, different donors were evaluated for the alkene difunctionalizations (Fig. 5a). Trifluoromethylimination of olefin could be achieved by using sodium trifluoromethanesulfonate as the donor (27a). As for potassium ethyl xanthate, the corresponding product could also be effectively obtained (27b). Organophosphorus compounds have a wide range of applications in the fields of organic synthesis, materials, coordination chemistry, medicine and agrochemicals due to their unique properties.21 Specifically, β-aminophosphine derivatives have been widely used to build a diversity of ligands or organocatalysts.2g,j,21f,g Therefore, the ability to rapidly construct this building block is crucial for expanding the diversity of organophosphorus libraries. With diarylphosphine oxides as donors, the corresponding β-iminophosphines could be obtained in moderate yields (27c–27e).
Vinyl sulfones are excellent Michael acceptors and they have widespread applications in biological research as covalent protease inhibitors.22 On the basis of the understanding of the reaction mechanism, this photo-induced intermolecular charge transfer strategy is further demonstrated in the construction of functionalized vinyl sulfones. After a quick reaction optimization (Tables S7–S10 in the ESI†), α-halogenated olefins bearing aromatic rings and an ester group all reacted smoothly (Fig. 5b), delivering the corresponding β-imino vinyl sulfones with acceptable yields (28a–28d). Interestingly, alpha carboxyl substituted olefin could be used to produce the desired product by decarboxylation (28a, X = CO2H). For different substituted sulfinates, the reaction system could also be well compatible (28e–28j). Moreover, this protocol worked in the late-stage functionalization of complex alkenes through direct vinylsulfonylimination (28k). Meanwhile, the β-imino vinylphosphines could be obtained by using α-bromostyrene as the capture reagent (28l and 28m).
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2223172, 2224155, 2224158 and 2248860. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3sc03667g |
This journal is © The Royal Society of Chemistry 2023 |