Synthesis of unnatural α-amino esters using ethyl nitroacetate and condensation or cycloaddition reactions

We report here on the use of ethyl nitroacetate as a glycine template to produce α-amino esters. This started with a study of its condensation with various arylacetals to give ethyl 3-aryl-2-nitroacrylates followed by a reduction (NaBH4 and then zinc/HCl) into α-amino esters. The scope of this method was explored as well as an alternative with arylacylals instead. We also focused on various [2 + 3] cycloadditions, one leading to a spiroacetal, which led to the undesired ethyl 5-(benzamidomethyl)isoxazole-3-carboxylate. The addition of ethyl nitroacetate on a 5-methylene-4,5-dihydrooxazole using cerium(IV) ammonium nitrate was also explored and the synthesis of other oxazole-bearing α-amino esters was achieved using gold(I) chemistry.

S3 toluene (120 mL) and this was extensively degassed by bubbling a stream of argon into the solution. The considered furan-bearing aldehydes (0.0416 mol) and ethyl nitroacetate (5.54 g, 0.0416 mol) were added followed by piperidine (0.35 g, 0.0041 mol) and this was heated to reflux under an inert atmosphere for the times described below. The suspension was cooled, filtered, the solid washed with toluene and concentrated to dryness (see below for the specific purification of compound 2n) before undertaking a reduction using sodium borohydride in refluxing isopropanol as described above. Alternatively for the reduction step, the suspension was cooled, filtered and the solid washed with dry ethanol (150 mL). To this solution, sodium cyanoborohydride (2.51 g, 0.0416 mol) was added and this was stirred overnight under a calcium chloride-protected atmosphere. This was diluted in water made acid with 1 N hydrochloric acid (50 mL) and extracted with ethyl acetate. The organic layer was washed with water brine, dried over magnesium sulfate, concentrated to dryness and the residue purified as described below. In the specific case of compound 2n, a chromatography over silica gel (cyclohexane/dichloromethane 1:1) led to the isolation of the pure compound as colorless oil (which slowly darkened over weeks) containing a 2:3 mixture of isomers (11.7 g, 53%).   [2,3].
The mixture was left to stir overnight, part of the ethanol was removed under vacuum and the residue was dispersed in water and ethyl acetate. Then, the suspension was made basic with 22% ammonia, extracted with ethyl acetate and the organic layer was washed with water, brine, dried over sodium carbonate and concentrated to dryness to yield the -amino esters which were in some cases further purified as described below. Note: these oily -amino esters are not very stable, and unless the corresponding hydrochloride is made (by adding 4 N HCl in dioxane and concentration to dryness), many of them will slowly dimerize into the corresponding solid piperazine-2,5-diones even when stored at −20 °C.
The solution was concentrated to dryness, this was dispersed in water and ethyl acetate, the organic layer was washed with water, brine, dried over magnesium sulfate and concentrated to dryness and the residue purified by a chromatography over silica gel (cyclohexanedichloromethane 2/1) to give compound 9 as a 94% pure oil (1.89 g, 46%) as seen from the NMR data which were identical with the reported one [4]. Note: the impurity was tentatively identified as (2-nitroethyl)benzene.

Preparation of ethyl 2-amino-2-methyl-3-phenylpropanoate (10):
By using the reduction protocol described above for the preparation of compound 1, compound 10 was obtained from compound 9 as a white solid (1.66 g, 90%) after the addition of a 4 N solution of hydrogen chloride in dioxane and a thorough concentration to dryness. The NMR data were identical with the reported one [4].

Ethyl 2-amino-3-(furan-2-yl)-2-methylpropanoate (12):
By using the reduction protocol described above for the preparation of compound 1, compound 12 was obtained from compound 11 as an oil  with extra precautions (see note below). The solution was concentrated to dryness and purified by a chromatography over silica gel (cyclohexanedichloromethane 1/2). The corresponding fraction was heated at 95 °C under a 4 mbar pressure for a while to completely remove traces of unreacted ethyl nitroacetate to yield compound 16 as an oil (5.24 g, 72%) with NMR spectra similar to the previously described one [6]. Note: Upon opening the tube a serious overpressure was observed along with the evolution of plenty of bubbles. The evolving gas was identified by bubbling it into a filtered solution of calcium hydroxide which clouded immediately thus proving it was carbon dioxide.
This phenomenon is very much in accord with a previous report focusing on the thermolysis of ethyl nitroacetate to give formonitrile oxide [7]. Since two equivalents of ethyl nitroacetate are used in our preparation of 16, we suggest that the evolving formonitrile oxide is actually the species responsible for the dehydration of the other equivalent of ethyl nitroacetate into the ethoxy-carbonylformonitrile oxide required to obtain a [2 + 3] cycloaddition. Moreover, we stopped using a closed reactor when running this reaction on a larger scale and it functioned as well in an open flask.

Preparation of ethyl 2-amino-4-phenylbutanoate (18):
Compound 16 (0.47 g, 2.14 mmol), ammonium formate (2.7 g, 21.4 mmol) and 10% palladium over charcoal (0.11 g, 0.10 mmol) were heated to reflux in ethanol (50 mL) for one hour. This was filtered, the insoluble washed with a small amount of ethanol and 37% hydrochloric acid (1.6 mL, 13.6 mmol) was added before adding powdered zinc portion wise (0.42 g, 6.3 mmol). The mixture was stirred for 90 minutes, then the resulting solution was diluted in water, made basic with 22% ammonia and extracted with ethyl acetate. The organic layer was washed with water, brine, dried over sodium carbonate and concentrated to dryness to yield the amino ester 18 as an oil (0.32 g, 72%). 1 Compound 16 (0.86 g, 3.93 mmol) and sodium hydroxide (2.11 g, 58.98 mmol) were dissolved in cold ethanol (5 mL) and water (5 mL). The solution was stirred at 0 °C for 2 hours, diluted in water and made acidic with 1 N hydrochloric acid.

Ethyl 5-(benzamidomethyl)isoxazole-3-carboxylate (
The mixture was diluted in water, extracted with ethyl acetate, the organic layer washed with water, brine, dried over magnesium sulfate and concentrated to dryness. The resulting oil was dissolved in ethanol (100 mL), hydroxylamine hydrochloride (2.76 g, 39.7 mmol) and dry pyridine (3.3 mL, 40.8 mmol) were added and the mixture heated to reflux for 6 hours. The resulting suspension was diluted in water, extracted with ethyl acetate, the organic layer washed with water, brine, dried over magnesium sulfate and concentrated to dryness. The residue was purified by a chromatography over silica gel (cyclohexaneethyl acetate 1/1) and the corresponding fraction was dispersed in boiling cyclohexane and filtered, after cooling, to yield the target -hydroximino ester 35 as white