3-Dialkylamino-1,2,4-triazoles via ZnII-Catalyzed Acyl Hydrazide–Dialkylcyanamide Coupling

Zinc(II)-catalyzed (10 mol % ZnCl2) coupling of acyl hydrazides and dialkylcyanamides in ethanol leads to 3-dialkylamino-1,2,4-triazoles (76–99%; 17 examples). This reaction represents a novel, straightforward, and high-yielding approach to practically important 3-NR2-1,2,4-triazoles, which utilizes commercially available and/or easily generated substrates. Seventeen new 3-NR2-1,2,4-triazoles were characterized by HRESI+-MS and IR, 1H, and 13C{1H} NMR spectroscopies and five species additionally by single-crystal X-ray diffraction (XRD). The ZnII-catalyzed reaction proceeds via initial generation of the [Zn{RC(=O)NHNH2}3](ZnCl4) complexes (exemplified by isolation of the complex with R = Ph, 76%; characterized by HRESI+-MS, IR, CP-MAS TOSS 13C{1H} NMR, and XRD). Electronic effects of substituents at the acyl hydrazide moiety do not significantly affect the reaction rate and the yield of the target triazoles, whereas the steric hindrances reduce the reaction rate without affecting the yield of the heterocycles.


INTRODUCTION
1,2,4-Triazole and its derivatives represent an important class of five-membered heterocycles, and many aspects of their versatile organic 1,2 and coordination 3,4 chemistry have been repeatedly reviewed over the years. The increased number of publication on 1,2,4-triazoles relates to the extensive application of these heterocycles, their derivatives, and (1,2,4-triazole)-based metal complexes in materials chemistry 5−9 and also in medicinal chemistry insofar as these species display broad spectrum of biological activities (for recent reviews see refs 10−12 ).
The routes employing isothioureas give 3-R-1,2,4-triazoles with both donor and acceptor substituents R, but they require utilization of isothioureas that are typically available from separate multistep synthetic methodologies. The third and fourth methods considered above are characterized by moderate yields and were performed for cyanides featuring exclusively activating acceptor substituents. Thus, as can be inferred from the consideration of the reported methods, a convenient and high-yielding synthetic method leading to 3-amino-1,2,4triazoles is not yet developed.
In view of our general interest in reactions of metal-activated substrates featuring CN triple bonds (for our reviews see refs 50 and 51) and, in particular, synthetic approaches allowing facile transformations of rather unreactive cyanamides (reviews; 52,53 recent studies, see refs 54−56 ), we decided to study a possibility of Zn II -catalyzed synthesis of 3-NAlk 2 -1,2,4-triazoles from acyl hydrazides and dialkylcyanamides. Usage of zinc(II), apart from the low cost and rather environmentally friendly character of this metal center, was additionally stimulated by its recent applications in advanced organic synthesis 57−65 and by our own studies disclosing a detailed mechanism of Zn II /H +mediated generation of 5-amino-1,2,4-oxadiazoles via amidoxime−cyanamide coupling. 66 This work describes a new synthetic procedure based on zinc(II)-catalyzed acyl hydrazide−dialkylcyanamide coupling, which allows the utilization of cyanamides bearing donor alkyl substituents and gives 3-dialkylamino-1,2,4-triazoles under mild conditions and in high yields. We also established that the Zn IIcatalyzed reaction proceeds via initial generation of the [Zn{RC(O)NHNH 2 } 3 ](ZnCl 4 ) complexes and all our data are consistently disclosed in paragraphs that follow.
Complex 1 reacted with Me 2 NCN (3.6 equiv) in EtOH solution at 80°C for 6 h giving triazole 2 in 98% isolated yield (85% after 4 h; Scheme 2, b). All these initial experiments demonstrated that the coupling between acyl hydrazides and dialkylcyanamide can be conducted under Zn II -involving conditions to give practically important 3-NR 2 -1,2,4-triazoles. Owing to the known kinetic lability of zinc(II) centers, we further decided to study a possibility of generation of these heterocyclic systems via a more sustainable zinc(II)-catalyzed route, and the corresponding experiments are described in the next section.
2.2. Zn II -Catalyzed Acyl Hydrazide−Dialkylcyanamide Coupling. To begin optimization, benzoyl hydrazide and dimethylcyanamide were chosen as model substrates (Table 1). First, methanol, ethanol, and 1,4-dioxane were taken as reaction media because of good solubility of all reaction components in these solvents. In blank experiments conducted in the absence of any catalyst, only trace amounts of the triazole were detected by HRESI + -MS (entries 1−3). We started catalyst screening from the cheap and broadly available ZnCl 2 . In the presence of ZnCl 2 (10 mol %; for the catalyst optimization see later), the reaction was performed in ethanol and 1,4-dioxane for 18 h upon reflux to give the triazole in nearly quantitative 1 H NMR yields (entries 6 and 7), whereas in methanol the yield was 51% (entry 4). This moderate yield of the product in MeOH is probably due to lower reflux temperature of the reaction mixture and, indeed, the yield increased by 35% on heating the reaction mixture for additional 18 h (entry 5). Considering these data, ethanol was chosen as a solvent for the reaction as more suitable (less toxic and dangerous and inexpensive) than 1,4-dioxane.

ACS Omega
Article and FeCl 3 , gave triazoles in substantially lower yields than those with ZnCl 2 because of unselective generation of various unidentified species (entries 10−13). Thus, ZnCl 2 was chosen for further optimization. This included variation of reaction time and also change of relative quantity of ZnCl 2 and dimethylcyanamide.
Monitoring of the reaction indicated that the reaction in refluxing ethanol takes 6 h (entries 14−16) to give the target compound in 96% isolated yield. Because of that, further optimization was performed by heating for 6 h. Variation of the relative quantity of ZnCl 2 indicated that the decrease of the catalyst amount results in decreased yield of the triazole (entries 17−18), whereas the increase of the catalyst amount to 15 mol % does not affect conversion and preparative yield of the triazole within experimental error (entry 19). Hence, 10 mol % of ZnCl 2 was applied in further experiments. Variation of the relative quantity of dimethylcyanamide (from 1:1 to 1:2 molar ratios) indicated that 1.2 equiv of the cyanamide is an optimal quantity; increase of the relative quantity of Me 2 NCN complicates the isolation and leads to lower isolated yield of the triazole (entries 20−22). On the basis of our experimental data, the reaction of acyl hydrazide with 1.2 equiv of dialkylcyanamide in the presence of ZnCl 2 (10 mol %) in ethanol for 6 h upon reflux was chosen as basic reaction conditions for further syntheses.
After the optimization, the substrate scope of acyl hydrazides was studied (Scheme 3). In general, this reaction proceeds smoothly for 6 h to furnish 2−14 in good-to-excellent yields. Slightly lower isolated yield (76%; compare with 91% 1 H NMR yield) for 4 is probably because of rather good solubility of 4 in Et 2 O, which causes a loss of the triazole on purification. Steric hindrances have a negligible effect on reaction yields but significantly affect the reaction time; it takes 24 h to reach full conversion of the acyl hydrazide for ortho-substituted triazoles 11 and 12. Longer reaction time was also required for the synthesis of 7, 8, and 13, and it is probably because of the heterogeneity of the reaction mixture.
Next, we varied dialkylcyanamides as the reaction partners (Scheme 3). The reaction proceeds for dialkylcyanamides giving 15−18 in good-to-excellent yields (76−99%). The developed procedure was applied for the synthesis of 17 new 3-NAlk 2 -1,2,4triazoles that were characterized by HRESI + -MS and IR, 1 H, and 13 C{ 1 H} NMR spectroscopies and five compounds also by X-ray crystallography.

ACS Omega
Article ing 3-NPh 2 -1,2,4-triazole (ca. 50% 1 H NMR yield) along with a mixture of unidentified products; all our attempts of their separation failed. Variation of the catalyst amount from 5 to 20 mol % and reaction time from 1 to 24 h did not improve the selectivity. Thus, preparation of the diarylamino triazoles requires another synthetic methodology. The reported protocol is not applicable for the reaction of acyl hydrazides with conventional aromatic and aliphatic nitriles, because the application of these RCN's give only trace amounts of the corresponding triazoles (detected by HRESI-MS) even after 2 d. 2.3. Analytical and Spectroscopy Data. Compound 1 was characterized by inductively coupled plasma atomic emission spectroscopy (ICP-AES)-based Zn elemental analysis, molar conductivity, HRESI + -MS, IR, and CP-MAS TOSS 13 C{ 1 H} NMR and additionally by single-crystal X-ray diffraction ( Figure 1). It gives satisfactory ICP-AES-based Zn elemental analysis for the proposed formula. Molar conductivity is 15.7 S cm 2 mol −1 in EtOH, which is lower than that expected for 1:1 electrolytes (35−45 S cm 2 mol −1 ), 69 most likely because of a dynamic equilibrium with nonionic complexes, for example, [ZnCl 2 {PhC(O)NHNH 2 } n ] (n = 1 or 2). The HRESI + massspectrum of 1 exhibits a peak corresponding to the quasi-ions [M + Cl − L] + . The IR spectrum displays strong bands at 3240− 3155 cm −1 from ν(N−H). Medium intensity bands at 3056− 2855 cm −1 were assigned to the ν(C−H) vibrations. The spectra exhibit very strong bands at 1637 and 1609 cm −1 from ν(CO) and ν(CN). The CP-MAS TOSS 13 C{ 1 H} NMR spectrum displays two sets of signals. One signal in the region of δ 170.56 corresponds to the carbonyl group C atom, whereas another group of signals is from the aromatic system (δ 139.18−122.41).
In the molecular structure of 1·H 2 O, the coordination polyhedron of the zinc(II) center in the anionic species displays tetrahedral geometry, and Zn−Cl distances [2.232(2)− 2.329(3) Å] are typical for the Zn−Cl bonds. 70 In the cation, the coordination polyhedron exhibits a typical octahedral geometry with fac-configuration of the ligands ( bonds have a transitive order between single and double bonds, which indicates an extensive electron delocalization. 71 Triazoles 2−18 were unknown before this work and they were characterized by HRESI + -MS, IR, 1 H and 13 C{ 1 H} NMR spectroscopies. In addition, 2, 6, 9, 13, and 16 were studied by single-crystal X-ray diffraction (Figures 2 and 74S−78S).
Compound 3 was isolated and characterized as a stable solvate 3·2/3EtOAc. The HRESI + mass-spectra of 2−18 exhibit a set of peaks corresponding to the quasi-ions [M + H] + , and 2, 3, 5, 6, and 18 in addition display peaks corresponding to the quasi-ions [M + H + H 2 O] + . The IR spectra of 2−18 display one weak-tomedium band in the range of 3463−3137 cm −1 , which was attributed to the N−H stretches. Medium intensity bands at 3163−2712 cm −1 were assigned to the ν(C−H) vibrations. The spectra exhibit very strong band at 1632−1568 cm −1 from ν(CN). The spectrum of 3·2/3EtOAc additionally exhibits one strong band at 1610 cm −1 from ν(CO) of ethyl acetate. In addition, the IR spectrums of 8 and 14 display two strong bands at 1152−2519 and 1345−1334 cm −1 assignable to asymmetric and symmetric valence stretches of NO 2 , 25 respectively. The 1 H NMR spectra of 2, 3·2/3EtOAc, and 4−18 recorded in (CD 3 ) 2 SO display two sets of signals corresponding to the two tautomeric forms. In particular, two broad singlets from the NH were observed at δ 13.59−12.81 and 12.76−12.13. The spectra of dimethylamino triazoles 3·2/3EtOAc, 2, and 4−18 display one singlet signal at δ 2.99−2.86 from NMe 2 , whereas the spectrum of 15 exhibit quartet and triplet at δ 3.41 and 1.13, respectively, which were attributed to the NEt 2 resonances. In accord with the 1 H NMR data, the 13 C{ 1 H} NMR spectra of 2, 3·2/3EtOAc, and 4−18 also display two sets of signals. Two signals in the region of δ 160.79−157.01 correspond to the quaternary C atoms of the 1,2,4-triazole ring. The spectra of 2, 9, 13, 14, 17, and 18 also exhibit additional two signals corresponding to the minor tautomer, which were observed at δ 167.49−166.72 and 154.64−153.18. For 2, 3·2/3EtOAc, and 4−18 the signals of aromatic systems are located in the typical region of δ 159.21−115.15. The spectra of 2, 3·2/3EtOAc, and 4−18 also exhibit a signal in the δ 38.88−38.72 range, which is attributed to NMe 2 . Only one tautomeric form was observed in the 1 H NMR spectra of 5 and 12 recorded in less polar (CD 3 ) 2 CO, but poor solubility of 2, 3·2/3EtOAc and 4−18 in (CD 3 ) 2 CO does not allow to obtain high quality spectra even at high acquisition time. In addition, availability of two set of signals in the spectra due to a partial dimerization of the compounds was ruled out, because of relative integral intensities of the two forms during 10-fold dilution of the compound were    (Figures 2 and S74−S78). 71 2.4. Kinetic Study. In order to estimate electronic effects of the substituents at the acyl hydrazide moiety and in attempt to establish a correlation with one of sigma-constants (σ, σ 0 , σ + , or σ − ), we performed a kinetic study with a series of parasubstituted benzoyl hydrazides. The results of the 1 H NMR monitoring of the reaction rate for compounds 2, 3, and 6 dissolved in (CD 3 ) 2 CO (top) and 2−4, 6−8 dissolved in (CD 3 ) 2 SO are shown in Figure 3; the obtained rate constant values are collected in the table on Figure 3.
We found that neither the nature of the solvent, nor the substituent electronic effects significantly influence the observed pseudo-first order rate constant k, which is ca. 10 −3 s −1 in all cases. Owing the similarity of all the calculated reaction rate constants (major part of them are the same within 3σ), we were unable to find a correlation with any of the sigma-constants. As can be inferred from consideration of Figure 3, there are some deviations from the linearity at times longer than ca. 1000 s, which are probably caused by the catalyst degradation. Based upon the kinetic study it can be concluded that electronic effect of substituents in acyl hydrazide does not significantly affect the reaction rate and these data are useful for further control of the coupling.
2.5. Plausible Mechanism of the Reaction. Taking into account our previous results and the literature data, 72−74 we suggest a plausible mechanism of the Zn II -catalyzed coupling. Firstly, acyl hydrazide reacts with ZnCl 2 to give [Zn{RC( O)NHNH 2 } 3 ] 2+ species (Scheme 4, a; these species are known from ref 72 and one complex with R = Ph was isolated and characterized in this work) followed by coordination of NCNR 2 2 to the zinc(II) center (b); coordination numbers 5 and 6 are typical for the zinc(II) complexes bearing sterically unhindered ligands. 73,74 Coordination of the cyanamide to the zinc(II) center electrophilically activates the CN group toward addition of nucleophiles. 66,73 Reversible decoordination of the NH 2 moiety from the kinetically labile zinc(II) center followed by nucleophilic addition of this group to the electrophilically activated CN moiety gives the Zn II -bound N-acylamino guanidine (c). Most likely effect of R 1 on the addition is substantially lower that the activating ability of the zinc(II) center and, therefore, in the kinetic study the R 1 effect was not observed. The slope of the dependence was used to estimate the rate of pseudofirst order reaction. a No data due to partial heterogeneity of the reaction mixture.

ACS Omega
Article Next, solvent-supported decoordination of the guanidine NH group proceeds, followed by nucleophilic attack of this group to the electrophilically activated (by coordination to the Zn II ) carbonyl C atom to give the five-membered heterocyclic ligand (d), which undergoes further elimination of H 2 O accomplishing the final 3-amino-1,2,4-triazole (e) and regenerating the catalyst (f).

CONCLUSIONS
This work describes a novel highly efficient synthetic methodology for the synthesis of 3-dialkylamino-1,2,4-triazoles derived from Zn II -catalyzed acyl hydrazide and dialkylcyanamide coupling. The optimized protocol utilizes simple and readily available reagents, and the reaction proceeds under mild conditions to give the triazoles in good-to-excellent yields (76−99%). The Zn II -catalyzed reaction proceeds via initial generation of the [Zn{RC(O)NHNH 2 } 3 ](ZnCl 4 ) complexes. Electronic effects of substituents at the acyl hydrazide moiety do not significantly affect the reaction rate and the yield of the target triazoles, whereas the steric hindrances reduce the reaction rate without affecting the yield of the heterocycles.

EXPERIMENTAL SECTION
4.1. Materials and Instrumentation. Solvents, metal salts, and cyanamides were obtained from commercial sources and used as received. Acyl hydrazides were synthesized from hydrazine hydrate and carboxylic esters by the known method. 75 All syntheses were conducted in air. Chromatographic separation was carried out on Macherey-Nagel silica gel 60 M (0.063−0.2 mm). Analytical thin-layer chromatography (TLC) was performed on unmodified Merck ready-to-use plates (TLC silica gel 60 F254) with UV detection. Melting points were measured on a Stuart SMP30 apparatus in capillaries and are not corrected. ICP-AES-based elemental analysis for Zn was performed on a Shimadzu ICPE-9000 instrument (cooling flow 10 L/min, plasma flow 0.6 L/min, carrier flow 0.7 L/min) with utilization of monoelemental standard for Zn (Merck). For the AES, the sample and the standard were dissolved in 0.1 M HNO 3 . The calibration curves were plotted in the range 0.001− 10.0 mg/L. The mass concentration value according to the calibration characteristic of each element and the margin of its absolute error were calculated on the basis of three parallel measurements of the sample. Molar conductivity of 7 × 10 −4 M solution in EtOH was measured on a Mettler Toledo FE30 conductometer using an Inlab710 sensor. Electrospray ionization (ESI) mass-spectra were obtained on a Bruker maXis spectrometer equipped with an ESI source. The instrument was operated in positive ion mode using an m/z range 50−1200. The nebulizer gas flow was 1.0 bar and the drying gas flow was 4.0 L/ min. For HRESI + , the studied compounds were dissolved in MeOH. Infrared spectra (4000−400 cm −1 ) were recorded on a Shimadzu IR Prestige-21 instrument in KBr pellets. For the characterization, 1 H and 13 C{ 1 H} NMR spectra were measured on a Bruker Avance 400 in (CD 3 ) 2 SO at ambient temperature; the residual solvent signal was used as the internal standard.
4.2. X-ray Structure Determinations. Crystals of 1·H 2 O, 2, 6·H 2 O, 9, 13, and 16 suitable for X-ray diffraction (XRD) were obtained by slow evaporation in air of a solution of appropriate triazole in Et 2 O at RT. Single-crystal X-ray diffraction experiments were carried out using Agilent Technologies "SuperNova" diffractometer with monochromated Mo Kα and Cu Kα radiation. The crystals were kept at 100(2) K during data collection, except for 9, which was destroyed at this temperature and it was measured at 200(2) K. The structures had been solved by the Superflip 76,77 and ShelXS/ShelXT 78 structure solution programs using charge flipping and direct methods, respectively, refined by means of the ShelXL 79 program, and incorporated in the OLEX2 program package. 80  Powder of any one of the acyl hydrazides RC(O)NHNH 2 (0.21 mmol) was added to a solution of ZnCl 2 (0.021 mmol) in (CD 3 ) 2 SO (0.4 mL) or (CD 3 ) 2 CO (0.4 mL) placed in an NMR tube, whereupon 10-fold excess R 2 ′NCN (2.1 mmol) was added to the mixture. The NMR tube was closed, and the obtained homogeneous solution was kept at 60°C for 2 h in the NMR spectrometer. Assuming that the reaction is pseudo first-order, the reaction rate constant k could be estimated from the slope of the initial time dependence of the concentration of the product. In this work, the reaction kinetics was monitored by measuring 1 H NMR spectra every 48 s (4 scans, repetition time 4 s), following the initial equilibration period of 5 min. The 1 H NMR spectra were measured on a Bruker Avance III 500 spectrometer (operating frequency 500.13 MHz for 1 H). The rate constant k was estimated by measuring the logarithms of the relative (normalized) integrated intensities of the product's ortho-CH proton signal, fitting their initial time dependence (300−800 s) by a straight line, and calculating the slope.   For 14 and 16, small amounts of some solid byproducts were formed during the reaction and they were removed by centrifugation before the next step. Poorly soluble in EtOAc 8 was dissolved in 100 mL of EtOAc prior to purification by column chromatography. For 2−18, the crude product was subjected to column chromatography on silica gel (eluted with ethyl acetate), and the solvent was evaporated in vacuo at 50°C. A colorless or pale beige (for 2, 3·2/3EtOAc, 4−7, 9−13 and 15−18) or yellow (8,14) crystalline residue was washed by two 1.5 mL portions of diethyl ether under ultrasound treatment and dried at 50°C in air.     ■ REFERENCES