To explore optimal conditions, we first investigated the reaction of ɣ-functionalized enaminonitrile 1a with commercially available tosyl hydrazide 2a, TBHP as the oxidant and KI as the catalyst. The reaction in toluene at 80°C yielded an unexpected pyridazine product 3a (P1) in 46% (Table 1, Entry 1), which was unambiguously confirmed by single crystal X-ray diffraction (CCDC number: 2094834). Encouraged by these results, we attempted to improve the yield of 3a. Preliminary screening of various additives indicated that TBAI was the most effective for product formation (71% yield, entry 3). Among the oxidants studied, TBHP was found to be the most suitable for obtaining 3a in high yield (Entry 1), while the use of other peroxides significantly reduced the yield (entry 6–13). Organic oxidants produced moderate yields (entry 6–9). Inorganic peroxide oxidants resulted in higher yields compared to organic ones (Entry 10–13) and the reaction with oxone as the oxidant displayed a good yield of 3a (68%, entry 13). Additionally, the yield varied depending on the amount of sulfonyl hydrazide, with the best yield obtained at a stoichiometric ratio of 2.0 eq. of 2a (entry 14–16). When the temperature was above 80°C or below it, the yield was lower compared to the yield at 80°C (Entry 17–18). we identified suitable organic solvents for this reaction and discovered that nonpolar aprotic solvents were the best fit (entry 19–25).
Table 1
Optimization of reaction conditons[a]
<Optimization table> |
Entry | Oxidant (4.0 eq) | Additive (mol %) | Solvent (2.0 mL) | Time (h) | Yield [b] (%) | Entry | Oxidant (4.0 eq) | Additive (mol %) | Solvent (2.0 mL) | Time (h) | Yield [b] (%) |
1 | TBHP | KI | DCE | 2 | 46 | 14[d] | TBHP | TBAI | DCE | 2 | 61 |
2 | TBHP | NaI | DCE | 2 | 53 | 15[e] | TBHP | TBAI | DCE | 2 | 56 |
3 | TBHP | TBAI | DCE | 2 | 71 | 16[f] | TBHP | TBAI | DCE | 2 | 50 |
4 | TBHP | CuI | DCE | 2 | 50 | 17[g] | TBHP | TBAI | DCE | 2 | 58 |
5 | TBHP | I2 | DCE | 2 | 22 | 18[h] | TBHP | TBAI | DCE | 2 | 66 |
6[c] | TBHP | TBAI | DCE | 2 | 53 | 19 | TBHP | TBAI | CH3CN | 2 | 53 |
7 | DTBP | TBAI | DCE | 12 | < 5 | 20 | TBHP | TBAI | THF | 2 | 42 |
8 | DCP | TBAI | DCE | 12 | 18 | 21 | TBHP | TBAI | DMF | 2 | 20 |
9 | H2O2 | TBAI | DCE | 8 | 38 | 22 | TBHP | TBAI | DMSO | 2 | 20 |
10 | K2S2O8 | TBAI | DCE | 20 min | 50 | 23 | TBHP | TBAI | Toluene | 2 | 74 |
11 | Na2S2O8 | TBAI | DCE | 20 min | 53 | 24 | TBHP | TBAI | 1,4-dioxane | 2 | 47 |
12 | (NH4)2S2O8 | TBAI | DCE | 20 min | 52 | 25 | TBHP | TBAI | EtOH | 2 | NR |
13 | oxone | TBAI | DCE | 2 | 68 | | | | | | |
[a] Reaction conditions, unless stated otherwise: 1a (1.0 eq, 0.17 mmol), 2a (2.0 eq, 0.34 mmol), oxidant (0.68 mmol), additive (20 mol%) in 2.0 mL of solvent, stirred at 80°C. [b] isolated yield, [c] 10 mol% of TBAI was used, [d] 5.0 eq, [e] 3.0 eq, [f] 1.5 eq of tosyl hydrazide was used, [g] at 60°C, [h] at 90°C. NR no reaction. TBHP: tert-butyl hydroperoxide 70 wt. % in H2O, DTBP: di-tert-butyl peroxide, DCP: dicumyl peroxide, TBAI: tetra-n-butylammonium iodide. |
Having established the optimal reaction conditions, we expanded the substrate scope of the reaction, as summarized in Table 2. Initially, we examined reactions with various substituted sulfonyl hydrazides (Table 2A). Simple phenyl- or methoxyphenyl-substituted sulfonyl hydrazides afforded the desired products 3b-c in 75% and 71% yields, respectively. The presence of electron withdrawing groups, such as -CF3 and -NO2, did not hinder the reaction, yielding products 3d and 3e in 65% and 54% yields, respectively. Similarly, halogenated sulfonyl hydrazides provided the expected products in good yields (3f, 53% (F); 3g, 65% (Cl)). Notably, the reaction with dansyl sulfonyl hydrazide also provided the expected product 3h in 55% yield, demonstrating the suitability of this method for sterically bulky substituents. Sulfonyl hydrazide derived from thiophene also smoothly generated the anticipated product 3i in 63% yield. However, lower efficiency was observed when using alkyl sulfonyl hydrazide 3j.
[a] Reaction conditions, unless stated otherwise: 1a (1.0 eq, 0.17 mmol), 2a (2.0 eq, 0.34 mmol), oxidant (0.68 mmol), additive (20 mol%) in 2.0 mL of solvent, stirred at 80°C
Next, we investigated the scope of enaminonitriles under optimized conditions (Table 2B), which can be readily prepared.35 Almost all enaminonitriles responded positively to our standard condition and delivered the products in good yields. Enaminonitriles without any substitutions proceeded to generate the expected product 4a in 55% yield. Various electron donating groups, including Me (4b, 63%), n-propyl (4c, 63%), iso-propyl (4d, 63%), methoxy (4e, 54%), and dioxymethylene (4l, 67%), also provided the corresponding products in good yields. Halogen-substituted enaminonitriles were compatible with the conditions and afforded the desired products in good yields (4f, 51% (F); 4g, 63% (Br); 4h, 68% (I)). Additionally, various electron withdrawing groups introduced in the enamines, smoothly afforded the corresponding products in good yields without any difficulties (4i-k, 52–63%). This optimized condition could be successfully extended to sterically hindered substrates (4m, 51%), and heterocycle-derived enaminonitriles (4n, 50%) as well.
Encouraged by the above results, we further extended our protocol to different electron-withdrawing groups (EWGs) on enamines. When ester-substituted enamines were treated with sulfonyl hydrazides containing methyl, methoxy, chloro, and trifluoromethyl groups, the target products were obtained in moderate yields (Table 2C, 5a-f, 32–43%). Notably, when a nitro group was used as an EWG, the reaction was highly effective in producing denitrated pyridazines via cleavage of the C-NO2 bond. Substituted sulfonyl hydrazides readily reacted with nitro enamines to provide denitrated pyridazines in higher yields (Table 2C, 6a-d, 66–80%).
Next, we aimed to test this hypothesis with a variety of enaminonitriles, which exhibited different reactivity patterns (Scheme 1). Replacement of the α-position hydrogen with a nitrile group (e.g. 7) resulted in decomposition of the starting material under our optimized conditions. while the enaminonitrile 8, with a nitrogen atom replacing carbon at the ɣ-position failed to provide the desired product. Surprisingly, when the ɣ-position was blocked with a chloro group, there was no negative effect, and the substrate smoothly underwent the reaction, furnishing the pyridazine product 4g in 75% yield.
Scheme 1. Control experiment when the radical attack sites were blocked.
To gain a better understanding of the reaction pathways, we conducted controlled reactions (Scheme 2A). Initially, we carried out the reaction without both additive (TBAI) and oxidant (TBHP), which resulted in product 11a in 23% yield, instead of the expected pyridazine product (Entry 1). This result implicated that the tosylhydrazide initially underwent transamidation with dimethylamine, tautomerized to an α,β-unsaturated hydrazone, and finally underwent sulfonylation, resulting in product 11a via the ionic mechanism. In addition, adding TBHP resulted in a slight increase in 41% (Entry 2). Next, we aimed to investigate the correlation between sulfonyl hydrazide and oxidant TBHP. The reaction with a 1:4 ratio of these two reagents yielded both the pyridazine product (4c, 23%) and the tosyl α-substituted intermediate (12, 9%) (Entry 3). By conducting the reaction with 2.0 equiv. of TBHP at lower temperature of 40°C, we were able to isolate key intermediates, including the non-conjugated hydrazone 13 (mixture, 22%), the conjugated hydrazone 14 (8%), and the desired pyridazine 4c (15%). Under these conditions, we also unexpectedly obtained N-aminopyridinium ylide 15b in 34% yield (Entry 4). The formation of non-conjugated hydrazones 13 suggested that transamidation occurred in the first step. When conducting the above reaction at 80°C, we obtained the expected pyridazine 4c in 18% yield, along with N-aminopyridinium ylide 15b in 48% yield (Entry 5).
Scheme 2. Control experiment to elucidate the mechanism[a]
[a] Reaction conditions, 1 (1.0 eq), 2 (2.0 eq), TBHP (4.0 eq), in 2.0 mL of toluene, stirred at reflux. [b] Reaction conditions, 1 (1.0 eq), 2 (2.0 eq), TBHP (2.0 eq), TBAI (20 mol%) in 2.0 mL of toluene, stirred at 80°C.
To identify the reaction intermediates, we conducted various control reactions. Initally, we performed a model reaction with non-conjugated hydrazone 13 under our standard conditions without tosyl hydrazide, resulting in the formation of pyridazine 4c in only 15% yield. This observation indicated that sulfonyl migration partially occurred in the initial radical process.73 However, the yield of 4c increased upto 71% when 2.0 equivalents of external tosyl hydrazide (2a) was used (Scheme 2B). On the other hand, when
2.0 equivalents of benzene sulfonyl hydrazide (2b) instead of tosyl hydrazide (2a), were allowed to react with 13 under standard conditions, both the benzene sulfonyl-substituted pyridazine 4o and tosyl migrated
product 4c were obtained in a 6:1 ratio (Scheme 2C). These outcomes could imply that i) intra-molecular sulfonyl migration occurred from the non-conjugated hydrazone 13, ii) 13 was an actual intermediate for this reaction and iii) 2.0 equivalents of sulfonyl hydrazide were required to achieve higher yields
Based on the above control experiments, we also performed the standard reaction conditions without additive (TBAI), which afford the expected sulfonylation product albeit in low yields (Scheme 2D, 11a-c, 26–41%). Similarly, we achieved the expected N-aminopyridinium ylides in moderate yields when the standard reaction conditions were performed only with 2.0 equivalents of oxidant TBHP (Scheme 2E, 15a-d, 41–48%). The structure of compound 15d was confirmed by single crystal X-ray diffraction (CCDC number : 2216818).
To determine whether the reaction proceeds via a radical mechanism, we conducted radical trapping experiments (Scheme 3A). The addition of free radical scavengers such as TEMPO, BHT, and DPE to our standard reaction conditions suppressed pyridazine formation to 0%, 15%, and 21%, respectively. These results clearly indicate that the sulfonyl radical is likely involved in this reaction and that it proceeds through a radical process. We also evaluated the feasibility of this reaction by performing gram-scale reactions, which efficiently afforded product 3a in 65% yield without significant loss. Similarly, starting from the HWE adduct, treatment with DMF-DMA afforded the enaminonitrile 1b. 35 After workup, without purification, our optimized reaction conditions afforded pyridazine 4b in 63% yield in two consecutive steps (Scheme 3B).
To expand the scope of synthesized pyridazines, various modifications were carried out (Scheme 3C). Treatment of pyridazine 3a with Zn in acetic acid resulted in the removal of the sulfonyl group, leading to the formation of the pyrrole derivative 16 in 82% yield. 74 The synthesized pyridazine could act as a diene. For instance, 3a undergoes an inverse electron demand Diels-Alder reaction (IEDDA) with 4-ethynyl anisole in the presence of a catalytic amount of ZnBr2, resulting in the tetra-substituted benzene derivative 17 in 61% yield. 75 And then the sulfonyl group of pyridazine 3a was successfully utilized as a partner in the Suzuki-Miyaura-cross-coupling reaction with 3,4-dimethoxy phenyl boronic acid, providing the coupling product 18 in 60% yield. 76 Further functionalization of the cyano groups in 4g was achieved in the presence of potassium t-butoxide, resulting in the formation of an amide 19 in 57% yield, along with the replacement of the tosyl group by a t-butoxy group. Treatment of 4g with sodium methoxide or sodium azide resulted in the formation of the methoxy-substituted pyridazine 20 in 86% or azide-substituted pyridazine 21 in 93% yield, with the tosyl group being replaced by a methoxy or azide group, while the cyano group remained intact. 77
Based on our experimental results and DFT calculations (described later), a plausible reaction mechanism for the synthesis of pyridazine 3 was depicted in Scheme 4. Transamidation of tosyl hydrazide with enaminonitrile (1) gave intermediate A (13), which then undergoes attack by a tosyl radical, derived from TBAI and TBHP, to form another intermediate B, followed by proton transfer to afford intermediate C. The radical pathway of C finally leads to the formation of a six-membered pyridazine product. In detail, N-radical formation of C by a butoxy radical, followed by 6-endo-trig cyclization, afforded intermediate E. Deprotonation and consecutive elimination of tosyl radical and water generated pyridazine product P1 (i.e., 3a). On the other hand, if the reaction intermediate C proceeds via an ionic pathway, the N-aminopyridinium ylide 15 is obtained. Initially, deprotonation of C by a hydroxide ion gives intermediate G, which then undergoes protonation followed by 6-endo-trig cyclization to afford intermediate I. Finally, proton transfer of I efficiently delivers the N-aminopyridinium ylide P2 (i.e.,15a).
Scheme 4. Plausible reaction mechanisms
DFT Calculations:
To understand the factors controlling the selectivity in the synthesis of sulfonyl-substituted pyridazines, we performed DFT calculations for the 6-endo-trig cyclization of γ-functionalized enaminonitrile 1a and tosyl hydrazide 2a via radical and ionic pathways in toluene at 80°C. All DFT calculations were carried out at the M06-2X/def2-TZVP//M06-2X/6–31 + G(d) level of theory using the M06-2X functional 78 and the polarizable continuum model (PCM) 79 for solvation free energies in toluene, implemented in the Gaussian 09 program. 80 The enthalpic and entropic contributions were computed by frequency calculations at the M06-2X/6–31 + G(d) level of theory. Computational details are described in the Supporting Information.
The furnished intermediate enaminonitrile A can adopt several possible conformations depending on the orientations of nitrile, phenyl and enamine groups (see ESI‡). The preferred conformations of intermediates A and B in toluene are shown in Fig. 2. The structures of the transition states (ts12 and ts21) and the corresponding products (P1 and P2) for two pathways in toluene are also depicted in Fig. 2. The transition states ts12 and ts21 correspond to the 6-endo-trig cyclizations for D → E via a radical pathway and H0 → I via an ionic pathway, respectively.
Additionally, we compared the structure of the product P1 with the X-ray structure of 3a in Table 1. The torsion angle θ, which defines the orientation of the tosyl group, is illustrated in Fig. S5 of the Supporting Information. The X-ray structure 3a was found to be 2.93 kcal/mol less stable than P1 in toluene. Specifically, the optimized structure of 3a was isoenergetic with P1, but the value of the former was antisymmetric to that of the latter, as shown in Fig. S6 of the Supporting Information. Hence, the local conformation of the tosyl group in the X-ray structure of 3a may be attributed to the transition of θ = −49° (P1) → θ = 49° (the optimized X-ray structure) → θ = 85° (X-ray structure) driven by crystal packing, which was not considered for the isolated molecules in the present calculations. (P1) → θ = 49° (the optimized X-ray structure) → θ = 85° (X-ray structure) driven by crystal packing, which was not considered for the isolated molecules in the present calculations.
The free-energy profiles of the two competing 6-endo-trig cyclization pathways in toluene are shown in Fig. 3. The relative Gibbs free energies were calculated according to the balanced equations for the reactant complex, intermediates, transition states, and product complex listed in Table S1 of the Supporting Information.
The highest barriers (ΔG‡) were calculated to be 23.2 and 36.2 kcal/mol for transition states ts12 (i.e., D → E) and ts21 (i.e., H0 → I) of the radical and ionic pathways, respectively. This indicates that the radical pathway is kinetically favored by 13.0 kcal/mol over the ionic pathway. The second highest barriers were calculated to be 11.8 and 8.7 kcal/mol for transition states ts11 (i.e., the deprotonation of C to D) and ts22 (i.e., the proton transfer of I to the final product P2) of the radical and ionic pathways, respectively.
The relative free energies (ΔG) of the final product complexes of P1 and P2 compared to the starting complexes of the intermediate B were calculated as − 68.4 and − 19.5 kcal/mol, respectively. This indicates that the product P1, formed via the radical pathway is 48.9 kcal/mol more stable than the product P2, formed via the ionic pathway. Additionally, there were more stable intermediates than the final products, with intermediates F1 and G1 for the radical and ionic pathways being 76.3 and 37.9 kcal/mol more stable than the corresponding products P1 and P2, respectively. Hence, the DFT results suggest that the 6-endo-trig cyclization via the radical pathway is both kinetically and thermodynamically favored over the cyclization via the ionic pathway. This finding is consistent with our experimental results described above.