Catalytic arylsulfonyl radical-triggered 1,5-enyne-bicyclizations and hydrosulfonylation of α,β-conjugates

Catalytic bicyclization of 1,5-enynes anchored by α,β-conjugates with arylsulfonyl radicals was established using TBAI and Cu(OAc)2 as co-catalysts.

The search for efficient cyclization reactions, particularly for those in radical cascade processes, has been actively pursued in the past several decades because they are extremely useful for the total synthesis of numerous important targets. 1,2 These reactions enable the rapid, reliable and straightforward creation of multicyclic ring systems using readily available starting materials with features such as unparalleled efficiencies, high functional tolerance and convenient conditions. 2 Among these cyclization reactions, the majority of efforts have been devoted to conducting radical ene-cyclization cascades, in which terminal alkenes were utilized in most cases via either metalfree or transition-metal-mediated radical processes (Scheme 1, eqn (1)). 3 However, the use of internal alkenes as radical acceptors has been highly challenging (Scheme 1, eqn (2)) 4 owing to their relatively low reactivity and larger steric hindrance as compared with their terminal counterparts.
1,5-Enynes endowed with extra unsaturated moieties are remarkable building blocks, and have been widely used for direct and selective tandem cyclizations via synergistic additions across C]C and C^C bonds in a one-step operation. 5 These cyclizations enhance both bond formation and annulation efficiencies with high levels of structural complexity and a reduced generation of waste. So far, two main methods for 1,5enyne cyclizations have been developed through metal catalysis 6 or electrophilic cyclization. 7 However, the radical bicyclization of 1,5-enynes for generating multi-substituted polycycles has not been well documented. The literature survey revealed that sulfonyl radicals can be generated from sulfonyl hydrazides and utilized in situ for the radical sulfonylation of alkenes. 8 Due to the importance of sulfonyl-containing compounds in photovoltaic materials, nonlinear optics and in general synthetic and medicinal areas, 9 we envisioned that under suitable catalytic radical conditions, the in situ generated sulfonyl radicals would be able to be involved in cascade bond-forming events with the internal C]C and C^C bonds of 1,5-enyne conjugate systems, resulting in 5-exo-dig/6-endo-trig bicyclizations and homolytic aromatic substitutions (HASs) (Scheme 2). Herein, we would like to report the preliminary results of this endeavour (Scheme 2).
At rst, 3-(2-(phenylethynyl)phenyl)-1-(p-tolyl)prop-2-en-1-one 1a was selected as a benchmark substrate to investigate the additions by sulfonyl radicals. With 20 mol% tetrabutylammoniumiodide (TBAI) as the catalyst, the reaction of substrate 1a with tosylhydrazide 2a was performed in CH 3 CN in the presence of benzoperoxide (BPO) (4.0 equiv.) as an oxidant at 70 C under air conditions, affording the expected benzo [b]uorens 3a, albeit with a low yield of 18% (Table 1, entry 1). Other solvents, such as dichloromethane (DCM), 1,4-dioxane and toluene, were also examined, with CH 3 CN showing the best performance (entries 2-4). Raising the reaction temperature to 100 C slightly ameliorates the yield of 3a (entry 5). A subsequent investigation of other catalysts was conducted in CH 3 CN. As illustrated in entries 6-8, different types of catalysts including I 2 , KI, and CuI were employed in the model reaction, and it turned out that I 2 and KI hardly facilitate the reaction (entries 6 and 7), while CuI as a catalyst only led to a poor yield of 16%. Next, we turned our attention to evaluating different additives (entries 9-11). We found that the addition of PivOH (1.0 equiv.) delivered 3a in a 35% yield (entry 11). Notably, the reaction of 1a and 2a in the presence of 2.0 equiv. of PivOH gave 3a in a 71% yield using a co-catalyst of TBAI (20 mol%) and Cu(OAc) 2 (5 mol%) with complete consumption of the starting material 1a (entry 15). Without PivOH, the yield of the expected product 3a decreased remarkably (entry 17). Further screening of other oxidants, such as TBHP (64% yield), DTBP (very poor yield) and H 2 O 2 (no product) for this transformation showed that BPO was the best choice (See ESI †).
With the optimized reaction conditions in hand, we examined the substrate scope of the sulfonyl hydrazides 2 by treating them with 1,5-enynes 1a (Scheme 3). As anticipated, the substituents on the phenyl ring of the arylsulfonyl hydrazides 2 did not hamper the catalytic process, but affected the reaction efficiency. Reactions of methyl-or bromo, t-butyl-substituted arylsulfonyl hydrazide 2 with 1a afforded the desired products in moderate to good yields. Additionally, benzenesulfonohydrazide exhibited a higher reactivity, allowing 1,5-enyne-bicyclization cascades toward the formation of benzo [b]uorens 3c in an 84% yield. 1,5-Enynes bearing an electron-donating or electron-withdrawing group (methoxy and chloro) at the para position of the aromatic ring (Ar 1 ) directly bound to the C^C bond gave the corresponding sulfonated products 3e and f in 55% and 68% yields, respectively. Alternatively, a naphthalen-1yl substituent linked to the C^C bond was also well-tolerated, affording the product 3g in a 60% chemical yield. Similarly, either electron-donating (methyl) or electron-withdrawing Scheme 2 Envisaged new reactivity of 1,5-enynes. (bromo) groups (R 1 ) at the para position of the phenyl ring tethered to the enone unit were well-suited for these radical 1,5enyne-bicyclizations (3a-3k). 1,5-Enynes 1 carrying electronneutral groups were also smoothly converted into the corresponding sulfonated benzo [b]uorens 3l-3n in 39-67% yields.
In view of our success with the synthesis of functional benzo [b]uorens 3, we reasoned that in the absence of alkyne moieties, chalcones 4 would be able to accept sulfonyl radicals via typical 1,4-additions, which would expand their utility for the synthesis of g-ketosulfones. We thus explored this possibility through a one-pot reaction of (4-chlorophenyl)-3-phenylprop-2en-1-one (4a) with 2a under the conditions described above. The expected g-ketosulfone 5a was obtained but with a lower yield (15%) initially. Aer careful optimizations were performed, we found that although Cu(OAc) 2 and PivOH did promote this catalytic process, the use of co-oxidants of BPO (2.0 equiv.) and TBHP (1.0 equiv., 70% in water) in 20 mol% of TBAI proved to be suitable for the current hydrosulfonylation, furnishing product 5a in a 77% yield. Subsequently, we further studied the reaction scope by reacting arylsulfonyl hydrazides 2 with various chalcones 4 under these conditions (Scheme 4). It turned out that the presence of various substituents, including methoxyl, methyl, chloro and bromo groups, on the aryl rings of the chalcones all worked well, giving access to a wide range of g-ketosulfones 5a-5o with yields ranging from 50% to 90%. Alternatively, arylsulfonyl hydrazides 2 carrying either electronically neutral or rich groups can be successfully engaged in this catalysis. Unfortunately, aliphatic sulfonyl hydrazide (5p) was proven not to be an adaptable substrate for this reaction, which may be ascribed to the relative instability of the sulfonyl radicals generated in situ from aliphatic sulfonyl hydrazides. Joining previously reported work, 10 this catalytic radical addition provided a new protocol for the formation of g-ketosulfones, which are important building blocks in organic synthesis.
To understand the mechanism, several control experiments were conducted. The treatment of 1,5-enyne 1h with tosylhydrazide 2a in the presence of radical scavenger TEMPO (4.0 equiv.) under standard conditions gave complex mixtures without the observation of the desired product 3l, conrming  the existence of a radical mechanism (Scheme 5, eqn (1)). In the absence of BPO, the reaction did not show the desired product (eqn (2)). To further conrm the sulfonylation sequence, subjecting 1,5-enyne 1h to the standard conditions in the absence of 2a failed to generate any desired benzo [b]uoren product 6 (eqn (3)). These control experiments suggest that BPO is essential for the catalytic cycles and the in situ generated sulfonyl radical triggers a 5-exo-dig/6-endo-trig bicyclization cascade.
On the basis of the above observations and those reported in literature, 8,11 a mechanism is proposed and shown in Scheme 6. The rst step is to form the sulfonyl radical A from the sulfonyl hydrazide using the benzoyloxy radical generated in situ from the I À anion-assisted decomposition of BPO. The intermolecular addition of the resulting sulfonyl radical A and the 1,5-conjugated enyne 1 followed by a 5-exo-dig cyclization gives intermediate B, in which the homolysis of carbon-copper(III) affords vinyl radical C. Intermediate C is converted into aryl radical D via a 6-endo-trig cyclization. Intermediate D undergoes SET (single electron transfer) oxidation and subsequent deprotonation to provide intermediate F. The tautomerization of F leads to the formation of benzo [b]uorens 3. Although the generation of sulfonyl radicals triggered by various oxidants has been achieved well, 8 the bicyclizations 12 towards fused carbocycles via sulfonyl radical initiated bifunctionalization of enynes is very rare in organic chemistry as mentioned earlier.

Conclusions
In summary, we have discovered new 1,5-enyne-bicyclization and hydrosulfonylation reactions of a,b-conjugates under convenient co-catalytic conditions. The addition of in situ generated sulfonyl radicals across the activated double bond is able to trigger a cascade 5-exo-dig/6-endo-trig bicyclization and HAS sequence, delivering tetracyclic sulfonylated benzo [b]uorens in a successive C-S and C-C bond-forming process. Using chalcones as replacements for 1,5-conjugated enynes, this reaction enables the hydrosulfonylation of alkenes to form gketosulfones with good to excellent yields. These two methods allow easy access to important functional sulfones for potential applications in organic and medicinal chemistry.