Ni/Photoredox Dual Catalysis Sulfone Compounds Synthesised with Carbon Nitride as the Semiheterogeneous Photocatalyst

An easily available heterogeneous semiconductor material, g-CN, proved to be feasible when combined with homogeneous nickel catalysts for light-mediated C(sp)-SO2Ar bond formation of aryl bromides with aryl sulfinates under mild conditions and base-free, unlocking a variety of cross-couplings. The metal-free heterogeneous semiconductor is totally recyclable from reaction system, and experimental results demonstrated a series of differently substituted substrates including electron donating groups and electron withdrawing groups can be tolerated with a satisfactory result. The method could even produce the classic drug Dapsone in large scale, showing strong practical application potential. Visible light dirven photoredox catalyted has gained numerous attentions as a powerful and energy-efficient method for chemical synthesis. The central of concept is rely on oxidation addition with specific radicals from precursors and trapped by transition metal complex such as nickel, palladium, cooper and so on. After elimination by reduction, the desired crosscoupling product is generated and the metal complex is reducted through photocatalytic cycle for next process. However, the whole process is highly depended on homogeneous photoresponse catalysts, such as ruthenium, iridium or organic dyes in numerous reports. It is a pity that these homogeneous photocatalysts are also accompanied by weakness such as the expensive expenses, separation from products, unable to recycle which limites the actual industrialized application to a certain extent. Based on the actual background and requirement, in recent years, some repersentive heterogeneous catalyst such as g-CN, TiO2, 10 CdS involved in photoredox process has been gradually developed. König’ groups reported ligand-free mpgCN/Ni dual photoredox catalytic protocols for C-N bond formation, and explored mpg-CN/Ni-dual catalysis for C(sp)C(sp) cross-coupling reactions yielding diarylmethanes analogously. Pieber and Seeberger reported mpg-CN in combination with nickel catalysis can induce selective C-O cross-couplings of carboxylic acids with aryl halides. Meanwhile, they also reported g-CN/Ni with light-mediated cross-couplings of aryl bromides with alcohols via C-O bond formation. Inspired by previous investigation, we hence developed mpg-CN/Ni dual catalysis sulfone compounds synthesised which are widely found in a number of drug-active molecules such as Dapsone, Vismodegib and Intepirdine (Scheme 1a) via C(sp)-SO2Ar bond formation. Compared with the traditional method (Scheme 1b), this kind of heterogeneous photocatalyst invovled in photoredox have the potential for industrial and large-scale development due to its advantages including mild conditions, visible light driven, base-free, high-yield outcomes and reutilization. In our consideration of this approach and previous report, we envisioned that aryl sulnate salts have low redox potentials (E1/2 = -0.37 V vs. SCE in CH3CN) 15 and could undergo single electron transfer oxidation by photogenerated hole form excited g-CN which is effective to generate sulfonyl radicals. Oxidative addition of Ni(0) species to an aryl halide delivers the Ni(II) intermediate which is tend to trap the sulfonyl radicals yielding a Ni(III) organometallic adduct. Subsequent reductive elimination produces the targeted C(sp)-SO2Ar cross-coupling product. Finally, the electron form semiconductor surface is utilized for the reduction of the Ni(I) species to Ni(0) species through another single electron transfer to complete the whole Ni catalytic cycle (Scheme 1c). Scheme 1. Significant of Sulfones and Experimental Design With this design in mind, we began to inveatigated the possibility and figure out the best conditions primarily. According to the Table 1, 4-bromotoluene 1a and sodium benzenesulphinate 2a was choosed to be the model substrate reseparately. To our delight, the product 1-methyl-4-(phenylsulfonyl) benzene 3a was obtained with the excellent yield up to 82% under 45W blue led (455 nm) irradiation and 10 wt% of g-CN, (Synthesis method was shown in Support information, SI), 5 mol% NiBr2 •DME and 10 mol% L1 added in DMF as the standard conditon (entry 1). Then, the reaction could not be happened when the absence of light, g-CN, or nickel catalyst respectively (entry 2). We chooesd the Ni(COD)2 as the nickel source but a modest yield was acquired (entry 3). Meanwhile, we also investigated the ligand effect to this reaciton system. Firsty, when no ligand was involved, the producted 3a did not observed at all (entry 4), and we changed similar ligands such as L2 or L3, but an unsatisfactory result (entry 5, 6) was shown. MeCN, MeOH or acrtone was selected to be solvent respectively, but trace producted was detected (entry 7). Shortening reaction time to 12h, the yiled was decreased to 68% accordingly (entry 8). The reaciton colud not be occurred without degasing (entry 9). Under the same conditions, chlorobenzene was introduced but unsatisfactory results were obtained, and we also chooed the iodobenzene as the substrate could also obtain 3a up to 78% yield (entry 10). Table 1. Screening Data for the Optimization of the Reaction Conditions towards phosphine oxide 3a. Entry Deviation from initial conditions Yield of 3a (%) 1 none 82 2 no light, no mpg-CN, no nickel catalyst N.D. 3 Ni(COD)2 instead of NiBr2· DME 37

Visible light dirven photoredox catalyted has gained numerous attentions as a powerful and energy-efficient method for chemical synthesis. 1 The central of concept is rely on oxidation addition with specific radicals from precursors and trapped by transition metal complex such as nickel, 2 palladium, 3 cooper 4 and so on. 5 After elimination by reduction, the desired crosscoupling product is generated and the metal complex is reducted through photocatalytic cycle for next process. However, the whole process is highly depended on homogeneous photoresponse catalysts, such as ruthenium, 6 iridium 7 or organic dyes 8 in numerous reports. It is a pity that these homogeneous photocatalysts are also accompanied by weakness such as the expensive expenses, separation from products, unable to recycle which limites the actual industrialized application to a certain extent.
Based on the actual background and requirement, in recent years, some repersentive heterogeneous catalyst such as g-CN, 9 TiO2, 10 CdS 11 involved in photoredox process has been gradually developed. König' groups 9a reported ligand-free mpg-CN/Ni dual photoredox catalytic protocols for C-N bond formation, and explored mpg-CN/Ni-dual catalysis for C(sp 2 )-C(sp 3 ) cross-coupling reactions yielding diarylmethanes analogously. 9b Pieber and Seeberger reported mpg-CN in combination with nickel catalysis can induce selective C-O cross-couplings of carboxylic acids with aryl halides. 9c Meanwhile, they also reported g-CN/Ni with light-mediated cross-couplings of aryl bromides with alcohols via C-O bond formation. 9d Inspired by previous investigation, 12 we hence developed mpg-CN/Ni dual catalysis sulfone compounds synthesised which are widely found in a number of drug-active molecules such as Dapsone, Vismodegib and Intepirdine (Scheme 1a) via C(sp 2 )-SO2Ar bond formation. Compared with the traditional method (Scheme 1b), 13 this kind of heterogeneous photocatalyst invovled in photoredox have the potential for industrial and large-scale development due to its advantages including mild conditions, visible light driven, base-free, high-yield outcomes and reutilization.
In our consideration of this approach and previous report, 14 we envisioned that aryl sulnate salts have low redox potentials (E1/2 = -0.37 V vs. SCE in CH3CN) 15 and could undergo single electron transfer oxidation by photogenerated hole form excited g-CN which is effective to generate sulfonyl radicals. Oxidative addition of Ni(0) species to an aryl halide delivers the Ni(II) intermediate 14 which is tend to trap the sulfonyl radicals yielding a Ni(III) organometallic adduct. Subsequent reductive elimination produces the targeted C(sp 2 )-SO2Ar cross-coupling product. Finally, the electron form semiconductor surface is utilized for the reduction of the Ni(I) species to Ni(0) species through another single electron transfer to complete the whole Ni catalytic cycle (Scheme 1c).

Scheme 1. Significant of Sulfones and Experimental Design
With this design in mind, we began to inveatigated the possibility and figure out the best conditions primarily. According to the Table 1, 4-bromotoluene 1a and sodium benzenesulphinate 2a was choosed to be the model substrate reseparately. To our delight, the product 1-methyl-4-(phenylsulfonyl) benzene 3a was obtained with the excellent yield up to 82% under 45W blue led (455 nm) irradiation and 10 wt% of g-CN, (Synthesis method was shown in Support information, SI), 5 mol% NiBr2 •DME and 10 mol% L1 added in DMF as the standard conditon (entry 1). Then, the reaction could not be happened when the absence of light, g-CN, or nickel catalyst respectively (entry 2). We chooesd the Ni(COD)2 as the nickel source but a modest yield was acquired (entry 3). Meanwhile, we also investigated the ligand effect to this reaciton system. Firsty, when no ligand was involved, the producted 3a did not observed at all (entry 4), and we changed similar ligands such as L2 or L3, but an unsatisfactory result (entry 5, 6) was shown. MeCN, MeOH or acrtone was selected to be solvent respectively, but trace producted was detected (entry 7). Shortening reaction time to 12h, the yiled was decreased to 68% accordingly (entry 8). The reaciton colud not be occurred without degasing (entry 9). Under the same conditions, chlorobenzene was introduced but unsatisfactory results were obtained, and we also chooed the iodobenzene as the substrate could also obtain 3a up to 78% yield (entry 10).  We then used the optimized reaction conditions to explore the scope of the reaction with respect to the variation of different substituted bromobenzene. As shown in Scheme 2. various aryl bromides substituted with electron donating group including methyl 3a, 3i, 3o, methoxy group 3n, tert-butyl 3f, or phenyl 3h, Naphthalene 3r or strong electron withdrawing group including trifluoromethyl 3c, 3j, 3p, fluorine atom 3d, 3k, 3q, cyan 3e, 3l at different positions (o, m, p) were reacted very well, generating the corresponding products with moderate to high yields. In addition, 1-methyl-5-(phenylsulfonyl)-1H-indole 3s also could be obtained and some heterocyclic compounds like pyridine 3t, thiazole 3u and furan 3v were also tolerated successfully via this method with bromide alternatives as the reagent. Scheme 2. Substrate Scope of Aryl Halide a a Reactions were performed under the standard conditions (Table 1, Entry 1) and isolated yields were reported. b Aryl iodide and 3.0 eq of sulfinate salt were used as the substrate.
In order to further investigate the comprehensive scope of the reaction method, various sodium benzenesulfonates were selected to participate in modle reaction condion, and the corresponding good results were obtained similarly. As shown in Scheme 3, some of the common substituent groups like methyl 4a, 4f, 4i, methoxy group 4d, tert-butyl 4e, Naphthalene 4k or trifluoromethyl 4b, fluorine atom 4c, 4g, 4h, chlorine 4j were proved to be compatible with suitable yield. In addition, pyridine compound 4l was suitable for this condition. Scheme 3. Substrate Scope of Sulfinate salt a a Reactions were performed under the standard conditions (Table 1, Entry 1) and isolated yields were reported. b Aryl iodide was used for the reaction partner.

Figure 3. Actual catalytic performance testing used in industrial manufacture
In view of dapsone is a famous bio-active molecule with the corresponding structure, we used the corresponding raw materials, through the standard explored conditions, to achieve the synthesis of this drug in large scale up to 74% yield successfully (Figure 1a). What's more, considering that the numbers of cycles from stabilization were the key to measure the practical application towards heterogeneous catalysts, we have studied the effect of g-CN though repeated to use. As shown in the Figure  1b, after each single reaction cycle, excellent conversion rate was obtained in spit of 5 times under the same conditions and recoverd g-CN still represent a thin layer appearance with TEM image attached.
In summary, a dual Ni/photocatalytic C(sp 2 )-SO2Ar coupling was developed using a carbon nitride semiconductor as recyclable photocatalyst with low toxicity. The semi-heterogeneous nickel/carbon nitride catalysis is an inexpensive, sustainable alternative to homogeneous protocols. The method selectively couples a broad range of aryl bromides with sodium benzenesulfonate in good to excellent isolated yields. Dapsone synthesis was also demonstrated on a gram scale and cyclic experiments also demonstrated the great potential of g-CN in such reactions.

Supporting Information
Experimental details, materials characterization and copies of relevant NMR spectra for all new products, this material is available free.

 ACKNOWLEDGMENT
We thank the Priority Academic Program Development of Shaanxi Higher Education Institutions and Postdoctoral research fund of Xi 'an Jiaotong University.