Polyfunctional Nitriles in Organic Syntheses: A Novel Route to Aminopyrroles, Pyridazines and Pyrazolo[3,4-c]pyridazines

Phenacylmalononitrile 1 reacts with dimethylformamide dimethyl acetal to yield an enaminone which could be readily converted into a pyrrole or an aminopyridazine by treating with ammonium acetate and hydrazine hydrate, respectively. Compound 1 reacted with hydrazine hydrate in ethanol at room temperature to yield the dihydropyridazine 9 as a single product. In refluxing ethanol this product further reacted with hydrazine hydrate to yield the novel dihydropyrazolopyridazinamine 10.


Introduction
Malononitrile and malononitrile derivatives are versatile reagents and their chemistry has been studied in the past [1][2][3] and still attracts considerable interest [4][5]. In the past thirty years we have reported several new approaches to a variety of polyfunctional heterocycles utilizing malononitrile or substituted malononitriles as precursors [6][7][8][9][10] and several of these products as been established to act as anti-profiler agents. Very recently we have reported on the utility of benzylmalononitrile as precursor to diaminopyrazoles, diaminoisoxazoles, thiazoles and condensed azoles [11]. In the present paper we report the results of our exploration of the synthetic potential of 2-(2-oxo-2-phenylethyl) OPEN ACCESS malononitrile (1) as a heterocycle precursor. This work has allowed us to develop a new route to aminopyrroles and aminopyridazines. The amines formed are of potential utility in the dye industry and as precursors for pharmaceuticals. In addition to that, the reported structures of the reaction products of (2-oxo-2-phenylethyl) malononitrile (1) with hydrazine hydrate have been reexamined and corrected in the light of our findings.

Results and Discussion
Phenacylmalononitrile has been prepared by treating malononitrile with phenacyl bromide. While the literature procedure [12] afforded the desired product in 79 % yield, reaction of phenacyl bromide with malononitrile in ethanolic potassium hydroxide solution gave 2-(2-oxo-2-phenylethyl) malononitrile (1) in 85% yield. Heating phenacyl bromide and malononitrile in the presence of potassium hydroxide solution in a microwave oven at 85 °C gave an 80% yield of the target product. Reaction of 1 with dimethylformamide dimethyl acetal afforded 2-(3-(dimethylamino)-1-oxo-1phenylprop-2-en-2-yl) malononitrile (2) in 75% yield. Although 2 may also exist in the Z form, only 2 (the E form) was isolated according to the NMR NOE difference which indicated that the olefinic proton at δ = 8.44 ppm is not sterically proximal to the methylene proton at δ = 7.10 ppm. Refluxing 2 in acetic acid in presence of ammonium acetate afforded the aminopyrrole carbonitrile 3 in 60% yield. On the other hand, reaction of 2 with hydrazine hydrate afforded the aminopyridazine 4 that was readily oxidized to 5 upon treatment with H 2 O 2 in acetic acid (Scheme 1). It is assumed that initially ammonia adds across the double bond in 2 to yield an intermediate that then looses dimethylamine or alternatively, initially losses dimethylamine and then cyclizes to form 4 (Scheme 1).
It has been previously reported by Abdelrazek et al [12] that (2-oxo-2-phenylethyl) malononitrile (1) reacts with hydrazine hydrate to yield 4-phenacylpyrazole-3,5-diamine (8). Subsequently Elnagdi et al [13] have shown that the major product of this reaction was in fact the 3-oxo-6-phenyl-2,3,4,5tetrahydropyridazine-4-carbonitrile (9). Very recently Abdelrazek [14] claimed that in ethanolic solution a pyridazineimine was isolated. These should not be possible as water formed during the reaction would readily hydrolyze readily any imine possibly formed. However, they have also indicated that in refluxing ethanol other product was formed in less than 35% yield and assumed it to be 8, reported earlier by Abdelrazek et al. [12]. Now we have found that in ethanol at room temperature 1 reacts with hydrazine hydrate to yield 9 as sole product in 96 % yield. When 9 was refluxed in ethanol with hydrazine hydrate a product of molecular formula C 11 H 11 N 5 (213.2) was formed. This proved identical with the product obtained by Abdelrazek [12] or the one identified by Elnagdi et al. [13]. It thus became clear that 8 had never been isolated and that product believed earlier to be 8 really must have another structure. After inspection of the spectral and analytical data now wish to assign this product as 5-phenyl-4,7-dihydro-1H-pyrazolo[3,4-c]pyridazin-3-ylamine (10) obtained by further condensation of dihydropyridazine carbonitrile 9 with hydrazine hydrate (Scheme 2). The NMR data of 10 suggest it is in a fast tautomeric equilibrium with the corresponding 2H compound. The dihydropyridazine 9 was readily oxidized to 11 on attempted coupling with benzenediazonium chloride. Compound 11 was also obtained upon oxidizing 9 in an AcOH-H 2 O 2 mixture (Scheme 2). Reduction of 1 with sodium borohydride in propanol solution afforded 2-amino-5-phenyl-4,5dihydrofuran-3-carbonitrile (12) in 60% yield (Scheme 2). This compound has been synthesized earlier [15] in almost the same way, but no spectral data had been reported. 1 H-NMR and 13 C-NMR of now reported for the first time.  We next shifted our interest to an exploration of the chemistry of 2-cyano-5-phenyl-3,5dioxopentanonitrile, reported by Abdelrazek and Salah El-Din [16] to be formed upon refluxing ethyl benzoyl acetate and malononitrile in ethanolic piperidine solution. Unfortunately, in our hands ethyl benzoylacetate did not react with malononitrile under the reported reaction conditions or even more vigorous ones. Thus it is concluded that 2-cyano-5-phenyl-3,5-dioxopentanonitrile could not have been obtained as reported. It is of interest that in the paper 2-cyano-5-phenyl-3,5-dioxopentanonitrile was claimed to be oil and the authors have not reported analytical data.

General
All melting points are uncorrected and were determined on a Sanyo (Gallaenkamp) instrument. Infrared spectra were recorded in KBr on a Perkin-Elmer 2000 FT-IR system. 1 H-NMR and 13 C-NMR spectra were determined on a Bruker DPX spectrometer operating at 400 MHz for 1 H-NMR and 100 MHz for 13 C-NMR using in CDCl 3 or DMSO as solvents and TMS as internal standard; chemical shifts are reported in δ (ppm). Mass spectra were measured on VG Autospec Q MS 30 and MS 9 (AEI) spectrometers, with EI 70 EV. Elemental analyses were measured by means of LEOCHNS-932 Elemental Analyzer. General purpose silica gel on polyester 20 x 20 cm TLC plates with UV indicator were used in TLC experiments.
Procedure 2: A mixture of compound 1 (1.84 g, 0.01 mol) and hydrazine hydrate (0.50 g, 0.01 mol) in ethanol (10 mL) was refluxed for 5 h (followed by TLC until completion using ethyl acetatepetroleum ether 1:1 as eluent). The reaction mixture was cooled and poured onto ice-water. The solid product thus formed was collected by filtration and washed with hot ethanol to extract the white product 9. The residue was crystallized from N,N-dimethylformamide (DMF) to yield purple crystals of 10.