Hydrazones are a class of Schiff bases with a special molecular structure containing a substructure (-NHN = C-). Many studies have shown that hydrazones possess a wide range of physiological activities, including antioxidant, anti-inflammatory, antibacterial, insecticidal, antiviral, and antitumor activities. In recent years, hydrazones have been highly valued in the fields of medicine, pesticides, materials science, and testing reagents, and have broad development prospects (Yang et al., 1996; Khan, 2008; Özbek et al., 2009; Özdemir et al., 2009; Belkheiri et al., 2010; Özdemir et al., 2010; Özkay et al., 2010).

QuinShimizu’s group developed a method for the oxidation of N-sulfonyl hydrazide catalyzed by lead tetraacetate (Scheme 1A) (Shimizu et al., 1980). Subsequently, Ashok et al. established a new scheme for the synthesis of sulfinates (Scheme 1B) through the K₂CO₃-catalyzed rapid conversion of N-sulfonyl hydrazide (Korawat and Basak, 2020). Hossain et al. reported a synthetic route to 1,3-disubstituted allenes through the CuI-catalyzed cross-coupling of N-tosylhydrazones with terminal alkynes (Scheme 1C) (Hossain et al., 2013). Furthermore, palladium-catalyzed allylation is a reliable and widely used method (Trost et al., 2006; Lu and Ma, 2007; Mohr and Stoltz, 2007; Weaver et al., 2011)and has been extensively used in total synthesis (Trost and Crawley, 2003; Enquist and Stoltz, 2008; Huters et al., 2012). Therefore, metal-catalyzed cleavage of C-O bonds of ethers remains an intriguing topic. Herein, we report the Pd-catalyzed rearrangement of N-tosylhydrazones bearing allyl ethers to produce trans-olefin-substituted sulfonylhydrazones (Scheme 1D).

When the reaction was carried out with N-tosylhydrazones 1a in the presence of Pd(PPh₃)₄ in THF and K₂CO₃ as the base, the desired product (E)-N-allyl-N'-(2-hydroxybenzylidene)-4-methylbenzene sulfonohydrazide 2a was obtained. Different catalysts were screened for the reaction, such as Pd(OAc)₂, Pd(PPh₃)Cl₂, PdCl₂, Pd₂ (dba)₃, and Pd(PPh₃)₄. Among these, Pd(PPh₃)₄ proved to be the best catalyst, which led to 55% yield of the final compound (Table 1, entries 2–6). When the reaction was carried out in the absence of a catalyst, the target compound was not obtained (Table 1, entry 7).

Next, the reaction was carried out in different solvents such as toluene, EtOAc, dioxane, DMSO, DMF, and CH₃CN to determine the optimal solvent (Table 1, entries 8–13). Subsequently, the effects of different additives such as CuBr, NH₄Br, and TBAC, on the product yield were investigated. The product yield did not increase significantly in the presence of these additives (Table 1, entries 14 and 15). K₂CO₃ was the most effective in facilitating the reaction, while other bases such as TEA, Cs₂CO₃, NaH, NaOH, and t-BuOK led to significantly lower product yields. Increasing or decreasing the temperature had no significantly improve the reaction yield. (Table 1, entries 25, 26). Therefore, the optimal reaction conditions were 1a (0.25 mmol) as the substrate, Pd(PPh₃)₄ as the catalyst (5 mol%), and K₂CO₃ (0.5 mmol) as the base in dioxane (0.1 M) for 10 h at 80°C under N₂ conditions.

With the optimal conditions in hand, we explored the scope of the reaction. First, we investigated the effect of various substituted N-sulfonylhydrazones as substrates (Scheme 2) on the reaction. The results revealed that the reaction conditions showed good tolerance for the functional groups on these substrates. Not only halogen groups (3-Br, 4-Br, 5-Br, 3-Cl, 4-Cl, 5-Cl, 4-F, 5-F, 3,5-2F, and 3,5-2Cl) and electron-donating substituents (3-CH₃, 4-CH₃, and 5-CH₃) but also strongly electron-withdrawing (4-NO₂, 4-CF₃) groups could be tolerated under the optimized conditions, so that the reaction proceeded smoothly.

Halogen groups substituted at various positions on the benzene ring had different effects on the reaction. For example, halogen substitution at the 5-position of the benzene ring gave a higher yield (2k–2l) than did substitution at the 3- and 4-positions. In particular, 5-F substitution in the benzene ring generated the target compound in 95% isolated yield (2l). However, the reaction yields were significantly lower when double halogen substitution was present on the phenyl ring (2m, 2n). Moreover, the target compound (2o) was obtained smoothly when the substrate was charged with strong electron-withdrawing group (4-CF₃), with yields of 52%.

Subsequently, we focused our attention on the effect of different substituted sulfonylhydrazones on the reaction yields (Scheme 2). The results showed that this method has wide applicability (2p–2w). Electron-withdrawing groups increased the yield of the reaction, while electron-donating groups decreased the yield. For example, the yields obtained with halogen substitution were higher than those observed with methyl substitution (2p and 2q vs. 2r and 2s). Encouragingly, even with strong electron-withdrawing group substitution, the corresponding target compounds were furnished smoothly (2t, 2u). The reaction also proceeded smoothly when the p-toluenesulfonyl group was replaced by the benzoyl group (2v), giving the target product in 75% yield. Unfortunately, the reaction did not proceed smoothly when the p-toluenesulfonate group was displaced by the methyl formate group (2w).

The reaction catalyzed by Pd(0) afforded sulfonylhydrazones, mainly the trans-isomer. The structure of 2a was confirmed by X-ray single-crystal diffraction analysis, and the chemical structures of other examples were obtained by analogy (Figure 1, see Supplementary Material for details). Based on the above results, we performed a scale-up experiment to extend the adaptability of the reaction. When 7.0 mmol of 1a was reacted under palladium catalysis, the corresponding product 2a was obtained in 60% yield (Scheme 3).

Subsequently, the reaction mechanism was investigated. Upon introducing free radical inhibitors (TEMPO or BHT) into the system, the reaction proceeded smoothly to afford the corresponding products (Scheme 4, Eqs. 1 and 2). This result suggested that the reaction did not involve a free radical mechanism. Unfortunately, the reaction did not proceed smoothly when the allyl group was replaced by a 2-methylallyl group (Scheme 4, Eq. 3). When 1a was substituted by substrate 1b′, 2b′ was not obtained under standard conditions, but the compound 3 was afforded, indicating that the terminal double bond with substituent was easily removed in the reaction (Scheme 4, Eq. 4). When 1a was replaced by substrate 2a′, the target compound 2a could not be obtained under the standard conditions (Scheme 4, Eq. 5). In contrast, if N′-benzylidene-4-methylbenzenesulfonyl hydrazide was added to the reaction system, 2a and 3a were produced (Scheme 4, Eq. 6).

Based on these results and the literature reports, we propose a plausible reaction mechanism (Tang et al., 2021; He et al., 2019; Nakamura et al., 2007; Butt and Zhang, 2015; Huo et al., 2014; Nakamura et al., 2008; Liu et al., 2018; Ma and Jiao, 2002; Yamamoto and Radhak-rishnan, 1999; Sieber and Morken, 2006; Bates and Satcharoen, 2002; Hashmi et al., 2013; Kolundzič et al., 2014) (Scheme 5). Initially, 1 is added to Pd(0) via oxidation, followed by exchange with the ligand of 1 to give π-allylpalladium species B. Then, B undergoes reductive elimination to afford intermediate C, which reacts with Pd(0) to form intermediate D. Since there is no β-H atom, D is exchanged with molecule 1 to produce B and simultaneously generates the final product 2. In addition, we also propose a possible reaction mechanism when the reaction substrate is 1b′. Oxidative addition of 1b′ to Pd(0), isomerization and subsequent β-H elimination generate Pd-H species F. Reductive elimination of intermediate F to afford product 3 and release Pd(0) for the next cycle (Scheme 5).

In conclusion, we report the palladium-catalyzed rearrangement of N-tosylhydrazones bearing allyl ethers to generate trans-olefin-substituted sulfonylhydrazones. We also investigated the applicability of the reaction to furnish the corresponding products, regardless of the presence of strongly electron-donating or electron-withdrawing substituents. The reaction involves the breakage of C-O bonds and the formation of C-N bonds, which forms the basis for the study of rearrangement reactions. Further investigation into the application of this reaction is ongoing in our laboratory.