Reaction development
Our proof-of-concept work was commenced by testing the reaction of bench-stable oxazino-pyridine 1a’ with commercial sodium benzenesulfinate (PhSO2Na, 2a) as a nucleophile under electrochemical conditions. Extensive optimization revealed that the best results were obtained if a solution of 1a’, 2a, and nBu4NBF4 in a mixed solvent system of methanol and cyclopentyl methyl ether (CPME) was electrolyzed in an undivided cell equipped with graphite felt (GF) anode and Pt cathode with a constant current of 5 mA for 1.5 h at room temperature. Subsequent treatment with aqueous acid at 60°C gave the desired meta-sulfonated pyridine 3 with an overall yield of 73% (entry 2, Table 1). In the absence of the sulfinate, the oxazino-pyridine was fully decomposed under the electrolytic conditions (entry 1). This clearly illustrates the potential competitive oxidation of the dearomatized pyridine at the anode and thus demonstrates how challenging a selective reaction was. Mixtures of alcohols and ethers are most effective, in particular a 3:1 volume ratio of MeOH and CPME (entries 3–6). No desired product was observed in the absence of MeOH, the former may have dual roles: a) improving the solubility of sulfinate salt; the use of only CPME results in incomplete dissolution; b) functioning as a proton shuttle, which is supported by the detection of molecular dihydrogen through headspace gas chromatography (GC) analysis (see SI). Several electrode combinations, for example, (+)C|Pt(–) and (+)GF|GF(–) were further examined (entries 7 and 8), and the best result was obtained with (+)GF|Pt(–). Other sulfinate salts, such as PhSO2Li, PhSO2K, PhSO2Cs, and PhSO2NnBu4 as well as PhSO2NHNH2 were also effective, albeit in lower yields (entries 9 and 10). The in situ generated PhSO2Na from PhSO2H and Na2CO3 was much less efficient (entry 11). The current density is also crucial, with 5 mA being optimal. Unsatisfactory yields were obtained by either decreasing or increasing the density, which may be attributed to insufficient formation of the sulfonyl radical or unwanted oxidative degradation of the oxazino-pyridine at the anode, respectively (entries 12 and 13). The reaction yield dropped to 46% under air, likely resulting from the quenching of engaged radicals by molecular oxygen (entry 14). Removal of electrolyte occurred with moderate efficiency (entry 15). Control experiments validated that the electricity is indispensable and even uniquely efficient to spur this net-oxidation reaction; no or only traces of the product were delivered with conventional chemical oxidizing agents such as 1,4-benzoquinone (BQ), K2S2O8, Ag2CO3, and I2 (entries 16 − 18). This electrooxidative reaction has an excellent current efficiency of 78%, thus underlining the sustainability of the overall process.
Under the optimal electrochemical conditions, a broad range of sodium sulfinates reacted with various pyridines with perfect regioselectivity to afford meta-sulfonated products in moderate to good yields (Fig. 2). In all these examples, no double meta-sulfonation was detected. Scope with respect to the sulfinate components was tested with the simplest, unbiased pyridine as an acceptor. The established protocol is widely applicable to both electron-rich and electron-deficient aromatic sulfinates, bearing substituents in the ortho, meta, or para position. A wealth of common functional groups is compatible, including ether (6), fluoro (8), amide (7, 16), ester (15), keto (14), trifluoromethyl (17), and trifluoromethoxy (18). Oxidation-sensitive formyl and amino groups (13, 22) and easily reduced nitro groups (19) remain intact. Compounds bearing chloro (9 and 10), bromo (11), and iodo (12) groups reacted selectively, showcasing the orthogonality of this electrochemical process to traditional catalytic cross-couplings. Some fused (20 − 22) and heterocyclic sulfinates (23 − 25) are viable partners, as are alkenyl sulfinates. Furthermore, both primary (27 − 31) and secondary alkyl sulfinates (32 and 33) were smoothly converted meta-sulfonated pyridines without any detectable amount of unwanted desulfinative products, while tertiary series seemed to reach the performance limit of this process presumably due to steric hindrance (2ah in Figure S10 in SI).
The scope regarding pyridine components was then examined (Fig. 2b). A wide series of electronically and sterically varied pyridines underwent meta-selective sulfonation in moderate to good yields. Suitable substrates range from the parent pyridine as well as 2-, 3-, 4-monosubstituted to disubstituted pyridines. When two differently substituted pyridines are presented in one molecule, the redox-neutral dearomatization shows high selectivity toward less sterically hindered pyridine, allowing a chemoselective mono-meta-functionalization of polypyridine compounds (45). Along with pyridines, quinolines (48) and isoquinolines (47) engaged in this transformation by applying the same activation strategy to prepare the sulfonated heteroarenes with complete meta-selectivity. By this newly developed method, 8-chloro-3-(phenylsulfonyl)quinoline (48), a key intermediate of the potential Alzheimer’s disease drug intepirdine,42 could be facilely synthesized from inexpensive 8-chloroquinoline, instead of precedents based on transition metal-catalyzed cross-couplings of costly 8-fluoro/chloro-3-iodoquinoline.55–57 The mild electrochemical route also allows for late-stage modification of drug-like molecules such as valdecoxib (34), sildenafil (35), as well as nikethamide (40, 41) and (S)-cotinine (51), loratadine (52), vismodegib (53), febuxostat (54), abametapir (55), and metyrapone derivatives (56).
The meta-sulfonation reactions described above were performed as a two-pot procedure with the valuable sulfonyl group being introduced after dearomatization and isolation of the related oxazino-pyridine intermediates. This provides rapid access to structural analogs, as outlined above. For certain targeted meta-sulfonated compounds, we were pleased to find that conducting the entire sequence in one pot is feasible. For instance, one-pot meta-sulfonation of pyridine-containing drugs (49, 51 − 53, 56) as well as pyridoindoles (46) could be achieved with comparable yields.
Both two- and one-pot synthetic procedures could be easily scaled up, as exemplified by the synthesis of meta-sulfonated nikethamide 50 and 3-(phenylsulfonyl)pyridine (Fig. 3b), respectively. By taking advantage of the rich chemistry of its pyridyl and sulfonyl moieties, nikethamide-derived compound 50 can serve as a valuable synthetic hub. For instance, its skeletal oxidation led to N-oxide analogous 58 in good yield. The introduced sulfonyl handle could be further translated into ether (59),49 thioether (60),54 and sulfoxide (61) pharmacophores in one or two steps. Additionally, consecutive C − H bond functionalization of pyridines is appealing to construct polysubstituted pyridines. As illustrated in Fig. 3a, unsymmetric doubly meta-sulfonated pyridine 57, which otherwise is difficult to prepare, was obtained in an overall yield of 27% from pyridine via iterative dearomatization/e-sulfonation/aromatization tandem sequence. Furthermore, the current electrochemical protocol is applied to pyridine meta-C − H thiolation with thiophenols, as exemplified by the synthesis of 62, which can be further converted into sulfoximine 63 through oxidation and amination (Fig. 3c). These outcomes testified its robustness and practicability in medicinal and process chemistry.
To shed light on this electrochemical process, we performed some mechanistic studies (Fig. 4a). The reaction was completely terminated by 2,2,6,6-tetramethylpiperidinyloxy (TEMPO), which points toward a radical-involved process. Attempt to capture the radical with 1,1-diphenylethylene delivered the adduct 64 in 63% yield, which supports the engagement of the sulfonyl radical. In cyclic voltammetry, the oxidative peak of PhSO2Na was observed at + 0.94 V (curve b) while that of oxazino-pyridine 1a’ appeared at + 1.72 V (curve c, Fig. 4b). These results clearly reveal the preferential oxidation of sulfinates at the anode over oxazino-pyridines to form sulfonyl radicals. Based on these findings, our proposed mechanism was outlined in Fig. 4c. Initially, a nucleophilic sulfinate undergoes a preferential single electron oxidation over a oxazino-pyridine at the anode to form the sulfonyl radical, which is often deemed as electrophilic and thus might be matchable in polarity with the electron-rich oxazino-pyridine intermediate. This radical addition would occur exclusively at the β-nitrogen site in the dearomatized pyridine 1a’. The resulting allyl-type radical species, existing as two resonance forms II and III, could further lose an electron at the anode to form iminium-type species IV. Deprotonation and subsequent dearomatization lead to the targeted meta-sulfonated product. At the cathode, the dihydrogen byproduct is released.