Surprising orientation in ring synthesis of 3,5-dimethylpyrazin-2(1H)-one

One of the standard methods to prepare pyrazin-2-ones is by the condensation of a 1,2-dicarbonyl compound with an amino acid amide (Garg et al., 2002; Jones, 1949; Karmas & Spoerri, 1952). For example, pyruvaldehyde (1) (see scheme) reacts with glycinamide (2a) to give 6-methylpyrazin-2-one (3a) (Yates et al., 1995). The orientation in this ring synthesis represents the combination of the amide N atom with the ketone carbonyl and of the amine group with the aldehyde carbonyl group. In the course of our studies on the dipolar cycloaddition reactions of 3-oxidopyraziniums (Kiss et al., 1987; Allway et al., 1990; Yates et al., 1995), we required 3,6dimethylpyrazin-2-one (3b) and assumed that, by analogy, it would result from a reaction of pyruvaldehyde with alaninamide (2b).

The reaction of pyruvaldehyde with alaninamide gave the title compound, C 6 H 8 N 2 O, and not the anticipated 3,6-dimethylpyrazin-2-one.

Comment
One of the standard methods to prepare pyrazin-2-ones is by the condensation of a 1,2-dicarbonyl compound with anamino acid amide (Garg et al., 2002;Jones, 1949;Karmas & Spoerri, 1952). For example, pyruvaldehyde (1) (see scheme) reacts with glycinamide (2a) to give 6-methylpyrazin-2-one (3a) (Yates et al., 1995). The orientation in this ring synthesis represents the combination of the amide N atom with the ketone carbonyl and of the amine group with the aldehyde carbonyl group. In the course of our studies on the dipolar cycloaddition reactions of 3-oxidopyraziniums (Kiss et al., 1987;Allway et al., 1990;Yates et al., 1995), we required 3,6dimethylpyrazin-2-one (3b) and assumed that, by analogy, it would result from a reaction of pyruvaldehyde with alaninamide (2b).
Reaction of pyruvaldehyde with alaninamide produced a pyrazinone, as anticipated, but standard spectroscopic analysis could not unambiguously confirm the structure of the product. For example, 1 H NMR spectroscopy revealed two threehydrogen singlet signals corresponding to the two methyl groups at 2.22 and 2.41 and a one-hydrogen singlet signal for the ring C-hydrogen at 6.88, but these data are consistent both with the anticipated structure (3b) and also with its isomer, 3,5-dimethylpyrazin-2-one (3c).
Suitable crystals were grown from ethyl acetate and an X-ray analysis carried out. This showed the product to be 3,5dimethylpyrazin-2(1H)-one (3c) (Fig. 1). Currently, we have no explanation for this unexpected regioselectivity; however, the moral from this result is that, for each pyrazinone synthesized by this method, unambiguous proof of structure must be sought. of pyruvaldehyde (40%, 0.36 g, 2 mmol) in methanol (0.5 ml) also precooled to 243 K. Next, with stirring, aqueous sodium hydroxide solution (12.5 M, 0.50 ml, 2.5 mmol) was added dropwise while the temperature was maintained below 263 K. The mixture was allowed to stand at 268 K for 2 h, then at r.t. for 3 h. To the mixture was added hydrochloric acid (12 M, 0.5 ml) followed by solid NaHCO 3 (0.25 g) to neutralize excess acid, and the whole was evaporated to dryness in a vacuum at 363 K. The residue was extracted with three portions (2 ml) of boiling chloroform. Evaporation of the extract left a yellow solid (205 mg, 83%). This was recrystallized from ethyl acetate (2 ml) to give colourless crystals (58 mg, 24%; m.p. 417-419 K).

Data collection
Rigaku R-AXIS diffractometer ' scans Absorption correction: none 12419 measured reflections 839 independent reflections 724 reflections with I > 2(I)

Refinement
Refinement on F 2 R[F 2 > 2(F 2 )] = 0.041 wR(F 2 ) = 0.110 S = 1.10 839 reflections 88 parameters H atoms treated by a mixture of independent and constrained refinement H atoms bonded to C were included in calculated positions using the riding model, with C-H distances of 0.93 and 0.96 Å and with U iso (H) = 1.5U eq (C) for methyl H atoms and 1.2U eq (C) for the other H atoms; atom H1, attached to N1, was found by difference Fourier methods and refined isotropically.  The molecular structure of (3c), with displacement ellipsoids drawn at the 50% probability level. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.