l-Serine methyl ester hydrochloride

In the enantiopure crystal of the title compound, C4H10NO3 +·Cl−, intermolecular O—H⋯Cl and N—H⋯Cl hydrogen bonds link the molecules into layers parallel to (001).

Financial assistance for this project was provided by the Dutch Organization for Scientific Research, Chemical Sciences (NWO-CW).
Supplementary data and figures for this paper are available from the IUCr electronic archives (Reference: VM2012).

Comment
Esterification of the carboxyl group of amino acids plays an important role in the synthesis of peptides, especially due to the increased solubility in non-aquous organic solvents (Bodanszky, 1993). The synthesis of methyl esters is straightforward and can be performed by the reaction of HCl gas with a suspension of the amino acid in methanol. In this reaction the hydrochloride of the amino acid methyl ester is obtained, which is the subject of the present study.
A molecular plot of the title compound (I) is shown in Fig. 1. As a consequence of the protection of the carboxyl group, only the amino group is available as protonation site. The O1-C1-C2-N1 and C4-O1-C1-C2 torsion angles of -175.99 (7) and 179.72 (7) ° indicate an extended structure of the backbone. Nevertheless, we do not see a stabilization of this extended structure by an intramolecular hydrogen bond between the ammonium group and O2. Such a stabilization would require a N-H bond in the plane of O2-C1-C2-N1, which is not the case here. Similar extended structures are also found in the methyl ester hydrochlorides of L-cysteine (Görbitz, 1989) and L-tyrosine (Bryndal et al., 2006). Bond lengths and angles in (I) are as expected (Table 1).
With three H-atoms at N1 and one H-atom at O3 the molecule has four hydrogen bond donors. Three oxygen atoms and the chloride anion could act as hydrogen bond acceptors. In fact, only the chloride is used as a hydrogen bond acceptor, here (Table 2). This results in an infinite two-dimensional hydrogen bonding network parallel to the ab plane, as shown in Fig. 2. In the c direction the hydrogen-bonded planes are separated by hydrophobic OCH 3 groups. Interestingly, the two-dimensional motif is also reflected in the morphology of the crystal, where (001) has the smallest dimension.
Despite a β-angle of 90.090 (1)° there is no orthorhombic symmetry in this crystal structure. The R int value for Lauesymmetry mmm is 31% compared to 2% for 2/m. We also did not find indications for pseudo-orthorhombic twinning. The reflections in the diffraction images were not split, and an analysis of the F o /F c listing with the TWINROTMAT routine of PLATON (Spek, 2009) did not suggest the presence of twinning.
Because (I) is derived from enantiopure L-serine, the absolute configuration was known in advance. Nevertheless, the presence of chloride provides enough enantiomorph distinguishing power (Friedif = 123, Flack & Shmueli, 2007) to allow a reliable experimental confirmation of the absolute structure. This was done using the Flack parameter (Flack, 1983), which resulted in x = 0.00 (3), and the Hooft parameter (Hooft et al., 2008), which resulted in y = 0.005 (15). As expected, the standard uncertainty of the Hooft parameter is significantly lower than in the Flack parameter, but both parameters confirm the correct absolute structure of (I).
Experimental 0.4 g of L-serine methyl ester hydrochloride (obtained commercially from Aldrich) was dissolved in 10 ml absolute ethanol followed by slow evaporation at room temperature. Single crystals suitable for X-ray diffraction were obtained after a final seeding step by adding a tiny amount of solid starting material.
supplementary materials sup-2 Refinement All H atoms were located in difference Fourier maps. H atoms bonded to N and O atoms were refined freely with isotropic displacement parameters. H(C) atoms were refined using a riding model (including free rotation of the methyl substituents), with C-H = 0.95-1.00 Å and with U iso (H) = 1.2 (1.5 for methyl groups) times U eq (C).
Friedel pairs were kept separate during the refinement and the Flack parameter was included in the least-squares matrix using TWIN/BASF instructions in SHELXL97. This has been shown to reduce the uncertainty of the Flack parameter compared to the hole-in-one algorithm (Flack & Bernardinelli, 2000). Fig. 1. Molecular structure of (I), with atom labels and 50% probability displacement ellipsoids for non-H atoms. 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 Rfactors(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.