Glycine zinc sulfate pentahydrate: redetermination at 10 K from time-of-flight neutron Laue diffraction

We report a redetermination based on single-crystal neutron diffraction data and Raman spectra for glycine zinc sulfate pentahydrate.

Single crystals of glycine zinc sulfate pentahydrate [systematic name: hexaaquazinc tetraaquadiglycinezinc bis(sulfate)], [Zn(H 2 O) 6 ][Zn(C 2 H 5 NO 2 ) 2 -(H 2 O) 4 ](SO 4 ) 2 , have been grown by isothermal evaporation from aqueous solution at room temperature and characterized by single-crystal neutron diffraction. The unit cell contains two unique ZnO 6 octahedra on sites of symmetry 1 and two SO 4 tetrahedra with site symmetry 1; the octahedra comprise one [tetraaqua-diglycine zinc] 2+ ion (centred on one Zn atom) and one [hexaaquazinc] 2+ ion (centred on the other Zn atom); the glycine zwitterion, NH 3 + CH 2 COO À , adopts a monodentate coordination to the first Zn atom. All other atoms sit on general positions of site symmetry 1. Glycine forms centrosymmetric closed cyclic dimers due to N-HÁ Á ÁO hydrogen bonds between the amine and carboxylate groups of adjacent zwitterions and exhibits torsion angles varying from ideal planarity by no more than 1.2 , the smallest values for any known glycine zwitterion not otherwise constrained by a mirror plane. This work confirms the H-atom locations estimated in three earlier singlecrystal X-ray diffraction studies with the addition of independently refined fractional coordinates and U ij parameters, which provide accurate internuclear X-H (X = N, O) bond lengths and consequently a more accurate and precise depiction of the hydrogen-bond framework.

Chemical context
Numerous coordination compounds of glycine (Glyc) with divalent metal sulfates are known. For the case of zinc, there is an anhydrous species, 2GlycÁZnSO 4  and two hydrates, GlycÁZnSO 4 Á3H 2 O and GlycÁ-ZnSO 4 Á5H 2 O. The trihydrate is dimorphic, occurring either as an orthorhombic crystal (space group Pca2 1 ) or as a monoclinic crystal (P2 1 /n) depending on the synthesis route (Fleck & Bohatý, 2004). The monoclinic form is isotypic with compounds of general formula GlycÁM(II)SO 4 Á3H 2 O where M(II) = Mg, Co or Fe (Oguey et al., 2013a(Oguey et al., ,b, 2014. Compounds with the general formula GlycÁM(II)SO 4 Á5H 2 O are known only as isotypic triclinic crystals (P1) for M(II) = Mg, Mn, Co, Fe and Zn (Lindqvist & Rosenstein, 1960;Elayaraja et al., 2007;Fleck & Bohatý, 2006;Tepavitcharova et al., 2012). Solubility data have been published for a purported GlycÁNiSO 4 Á5H 2 O Alymkulova & Salyeva, 1987). We have collected as-yet unpublished X-ray powder-diffraction data from this species, showing that it is isotypic with the other known members of the series. The existence of GlycÁCuSO 4 Á5H 2 O has been reported by Thilagavathi et al. (2012) but their work is in error, and quite unambiguously describes the well-known material CuSO 4 Á5H 2 O. ISSN 2056-9890 We recently carried out the first neutron diffraction study of GlycÁMgSO 4 Á3H 2 O and GlycÁMgSO 4 Á5H 2 O using perdeuterated powder specimens (Howard et al., 2016). GlycÁ-MgSO 4 Á5H 2 O tends to form masses of crystals that are both of poor quality and are too small for single-crystal neutron diffraction study; however, this is not the case for GlycÁ-MnSO 4 Á5H 2 O and GlycÁZnSO 4 Á5H 2 O, where fine tabular to blocky single crystals with volumes substantially in excess of 10 mm 3 are formed with ease ( Fig. 1). The objective of this work was to carry out the first single-crystal neutron diffraction study of any GlycÁM(II)SO 4 Á5H 2 O compound, specifically using a specimen with M(II) = Zn.
Optical and mechanical properties of the title compound were reported by Balakrishnan & Ramamurthy (2007), although they incorrectly give the composition as GlycÁ-ZnSO 4 Á7H 2 O. The effect of doping GlycÁZnSO 4 Á5H 2 O with cobalt is described by El-Fadl & Abdulwahab (2010). Three prior structure refinements from single-crystal X-ray diffraction data have been reported (Balamurugan et al., 2011;Tepavitcharova et al., 2012;Oguey et al., 2013c); comparisons with this work are detailed in Section 2.

Structural commentary
Although the stoichiometry of the material is accurately reflected in its common name, glycine zinc sulfate pentahydrate, the presence of two symmetry-inequivalent Zn sites means that the crystallographically proper structural composition is the 'double' formula [GlyÁZnSO 4 Á5H 2 O] 2 , or more precisely [Zn(H 2 O) 6 ][Zn(H 2 O) 4 (C 2 H 5 NO 2 ) 2 ](SO 4 ) 2 ; the unit cell contains one of these units.
The Zn1 coordination octahedron consists of tetraaquadiglycine zinc(II) with the glycine zwitterion (NH 3 + CH 2 COO À ) coordinating to Zn by one of the carboxylate oxygen atoms (Fig. 2); the inversion centre results in an all-trans configuration for these units. The Zn2 octahedron has the form hexaaquazinc(II); the sulfate tetrahedra are isolated, accepting hydrogen bonds primarily (but not exclusively) from Zn-coordinating water molecules (Fig. 3). The S-O bond lengths (Table 1) reflect the number of hydrogen bonds accepted by each apical oxygen atom with a statistical significance which was not apparent from the powder refinement of Howard et al. (2016) but which are in excellent agreement with the single-crystal X-ray study of Tepavitcharova et al. (2012).
Bond lengths and angles of the glycine zwitterion agree very well with other determinations of related compounds made by X-ray single-crystal diffraction at higher temperatures and extremely well with the determinations inglycine at room temperature by neutron single-crystal diffraction (Jö nsson & Kvick, 1972;Power et al., 1976), particularly in respect of their mean N-H bond lengths (1.039 Å ) and mean C-H bond lengths (1.090 Å ). The glycine zwitterion is remarkably planar, with torsion angles O7-C1-C2-N1 = À1.18 (3) and O8-C1-C2-N1 = 179.23 (2) , even by comparison with, for example, glycine nickel bromide tetrahydrate (Fleck & Bohatý, 2005), glycine lithium chromate monohydrate and glycine lithium molybdate (Fleck et al., 2006) Microphotograph of a representative GlycÁZnSO 4 Á5H 2 O single crystal viewed along the a axis; insets show details of less well-developed facets (e.g., 111 and 102). Drawings with each face labelled by the Miller index are shown on the right and a quantitative representation of the model is included in the CIF data. Figure produced and CIF code exported using WinXMorph (Kaminsky, 2005(Kaminsky, , 2007. research communications Table 1 Comparison of bond lengths (Å ), polyhedral volumes (Å 3 ) and various distortion metrics (cf., Robinson et al., 1971) in GlycÁZnSO 4 Á5H 2 O from this work and the three preceding single-crystal X-ray diffraction studies.
The distortion index and quadratic elongation are dimensionless, whereas the bond-angle variance is in units of degrees squared.

Figure 2
Local coordination environment of the Zn1 atom. Displacement ellipsoids are drawn at the 50% probability level for H and 90% for all other atoms. Dashed lines indicate N-HÁ Á ÁO hydrogen bonds. [Symmetry code: (i) 1 À x, Ày, 1 À z.]

Figure 3
Local coordination environment of the Zn2 atom.
where the glycine zwitterion lies on a mirror plane, are values of 180 realized (Fleck & Bohatý, 2005). In -glycine, the equivalent torsion angles are À19.60 (3) and 161.28 (2) . In respect of the heavy atoms, agreement in the interatomic distances and angles between the single-crystal X-ray and single-crystal neutron refinements (Table 1) is excellent, with some differences emerging in respect of the room-temperature refinement by Balamurugan et al. (2011). However, the neutron data provide a substantial improvement in accuracy with respect to the X-ray data in the hydrogen atom's frac-tional coordinates and U ij parameters. This is not surprising since neutrons are sensitive to the nuclear positions and X-rays to the electron density; in the covalent X-H bond the centroid of the H-atom's electron-density distribution is displaced towards the heavy atom by 0.1 Å , yielding X-H distances from 10-15% shorter than the true internuclear separation (Coppens, 1997). Table 2 compares X-H bond lengths from a range of GlyÁM(II)SO 4 Á5H 2 O crystals obtained by single-crystal X-ray diffraction and by neutron powder diffraction. In the work of Elayaraja et al. (2007), methyl research communications Table 2 Comparison of X-H(D) bond lengths (Å ) from earlier work (a-e) with our own (f).
Element symbols indicate the cation in each compound. 'X-ray' denotes single-crystal X-ray diffraction; 'NPD' denotes a neutron powder diffraction experiment on a deuterated analogue carried out at 10 K; 'neutron' indicates single-crystal neutron diffraction on a protonated analogue carried out at 10 K. Note that the atom symbols employed in our work are the same as those used by Elayaraja et al. (2007) and by Howard et al. (2016). Although other authors have used different atom labels -and indeed use them inconsistently in their own reports -we list equivalent contacts in this table.

Figure 4
Packing of polyhedra in the structure of GlycÁZnSO 4 Á5H 2 O viewed along a (left) and along c (right). ZnO 6 octahedra are green, SO 4 tetrahedra are yellow.
hydrogens were positioned geometrically and allowed to ride with U iso (H) = 1.2U eq (C); water hydrogen atoms were refined with restraints; the N-H bond lengths were all restrained to be equal. Balamurugan et al. (2011) placed the majority of their hydrogen atoms geometrically, although failed to identify the third amine hydrogen atom; water and the two amine H atoms were refined isotropically whereas the two methyl hydrogen atoms were riding on the C atom. Tepavitcharova et al. (2012) placed the hydrogen atoms in GlyÁZnSO 4 Á5H 2 O geometrically and treated all of them as riding on their associated heavy atom during refinement. Oguey et al. (2013c) allowed all water hydrogen atoms to refine isotropically but fixed the methyl and amine hydrogens to ride on C and N, respectively. The coordinates of hydrogen atoms in this work were unrestrained and allowed to refine anisotropically.

Glycine dimers
A hitherto unrecognized aspect of the supramolecular structure of GlycÁM(II)SO 4 Á5H 2 O compounds is the presence of glycine dimers (Fig. 5). These occur as closed cyclic structures formed by N-HÁ Á ÁO hydrogen bonds between the amine group of one glycine zwitterion and the Zn-coordinating carboxylate oxygen (O7) of another zwitterion, related to the first by an inversion centre. A similar cyclic dimer occurs in the structure of -glycine. A direct comparison between the dimers in GlycÁZnSO 4 Á5H 2 O and in -glycine is shown in Fig. 6; clearly, the main difference between these two dimers is the orientation of the carboxylate group, which is presumably due to the influence of a divalent metal being coordinated by  Table 3 Hydrogen-bond geometry (Å , ). (7) 1.793 (7) 2.755 (4) 169.0 (7) O5-H5BÁ Á ÁO8 ii 0.997 (7) 1.656 (8) (3) 127.9 (6) C2-H2BÁ Á ÁO10

Figure 5
Connectivity between adjacent Zn1 octahedra is via a closed cyclic glycine dimer. As before, displacement ellipsoids are drawn at the 50% probability level for H and 90% for all other atoms. Dashed lines indicate N-HÁ Á ÁO hydrogen bonds. [Symmetry codes: (i) 1 À x, 1 À y, 1 À z; (ii) 1 À x, Ày, 1 À z; (iii) x, 1 + y, z.] the O7 carboxylate oxygen. Experimental studies of aqueous solutions indicate that only glycine monomers exist in the liquid phase (Huang et al., 2008). However, there has been widespread disagreement on this matter from computational studies, which indicate either that there are no dimers (Hamad & Catlow, 2011), substantial quantities of closed zwitterionic dimers (Friant-Michel & Ruiz-Ló pez, 2010), or a small fraction of open dimers (Yani et al., 2012) present in saturated solutions. The presence or absence of glycine polymerization in coordination compounds such as these may be useful in understanding the association of glycine in saturated aqueous solutions during nucleation and the role of solvated metal ions in polymerizing amino acids in Earth's Hadean oceans (Kitadai et al., 2011(Kitadai et al., , 2016 or in extraterrestrial oceans elsewhere in our solar system (Kimura & Kitadai, 2015). Comparison of the closed cyclic dimers involving zwitterionic glycine that occur in the crystal structures of -glycine (top) and in GlycÁZnSO 4 Á5H 2 O (bottom).

Vibrational spectroscopy
Laser-stimulated Raman spectra were measured using a portable B&WTek i-Raman Plus spectrometer equipped with a 532 nm laser (P max = 37 mW at the probe tip) that records spectra over the range 171-4002 cm À1 with an optimal resolution of 3 cm À1 . Measurements were carried out on powdered specimens of -glycine and GlycÁZnSO 4 Á5H 2 O. Samples were measured in thin-walled glass vials using the BC100 fibre-optic coupled Raman probe; the total integration time and laser power for each sample is provided with the tabulated results (see supplementary material). The Raman spectrum of GlycÁZnSO 4 Á5H 2 O (Fig. 6) is virtually identical with that of GlycÁMgSO 4 Á5H 2 O reported in Howard et al. (2016) and is in excellent agreement with the spectrum shown in Tepavitcharova et al. (2012). Numerical data of the Raman spectrum are provided as an electronic supplement; peak positions and vibrational mode assignments are given in Table 4. The main differences between the two divalent-metal-substituted compounds include the blueshifting of octahedral deformation modes and blue-shifting of both symmetric and asymmetric COO À stretching modes. A large blue-shift of (A) and (S) COO À occurs when glycine coordinates to Mg 2+ and the shift increases when glycine coordinates to Zn 2+ . Raman spectra of -glycine and GlycÁZnSO 4 Á5H 2 O are shown in Fig. 7.

Synthesis and crystallization
GlycÁZnSO 4 Á5H 2 O was crystallized by evaporation at room temperature of an equimolar aqueous solution of -glycine (Alfa Aesar A13816) and ZnSO 4 Á7H 2 O (Sigma Aldrich Z4750) in deionized water (Alfa Aesar 36645). Unlike the MgSO 4 -bearing analogue, GlycÁZnSO 4 Á5H 2 O forms large well-faceted crystals that are both amenable to morphological study and suitably large for single-crystal neutron diffraction analysis. Fig. 1 shows photographs of a representative crystal viewed along its a axis and series of drawings with indexed crystal faces.

Data collection and refinement
Crystal data, data collection and structure refinement details are summarized in Table 5. Data were collected from a pair of single crystals at a series of four discrete rotational positions about the vertical axis, each frame being counted for 5 h, equivalent to 800 mAhr of ISIS proton beam current per frame. The structure of Glyc(d 5 )ÁMgSO 4 Á5D 2 O at 10 K reported by Howard et al. (2016) was used as a starting point for the refinement. A total of eleven peaks, with the largest (F obs -F calc) values were omitted from the refinement; such outliers are fairly common in SXD measurement when peaks occur close to the edges of detectors. A mild restraint on the U ij parameters of the sulfur atom was imposed (SHELX ISOR command) in order to avoid a slightly non-positive-definite displacement ellipsoid. Since sulfur has the smallest neutron scattering cross section of any atom in the structure, and since it is both comparatively heavy and the temperature is very low, it is not surprising that -within errors -the effective U iso parameter should refine to a small negative value. Raman spectra of -glycine (top) and GlycÁZnSO 4 Á5H 2 O (bottom). Selected vibrational modes are labelled and a complete quantitative listing is given in Table 4.

Special details
Experimental. For peak integration a local UB matrix refined for each frame, using approximately 50 reflections from each of the 11 detectors. Hence _cell_measurement_reflns_used 550 For final cell dimensions a weighted average of all local cells was calculated Because of the nature of the experiment, it is not possible to give values of theta_min and theta_max for the cell determination. The same applies for the wavelength used for the experiment. The range of wavelengths used was 0.48-7.0 Angstroms, BUT the bulk of the diffraction information is obtained from wavelengths in the range 0.7-2.5 Angstroms. The data collection procedures on the SXD instrument used for the single-crystal neutron data collection are most recently summarized in the Appendix to the following paper Wilson, C.C. (1997). J. Mol. Struct. 405,[207][208][209][210][211][212][213][214][215][216][217] Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. The variable wavelength nature of the data collection procedure means that sensible values of _diffrn_reflns_theta_min & _diffrn_reflns_theta_max cannot be given instead the following limits are given _diffrn_reflns_sin(theta)/lambda_min 0.06 _diffrn_reflns_sin(theta)/lambda_max 1.38 _refine_diff_density_max/min is given in Fermi per angstrom cubed not electons per angstrom cubed. Another way to consider the _refine_diff_density_ is as a percentage of the scattering density of a given atom: _refine_diff_density_max = 5.7 % of hydrogen _refine_diff_density_min = -6.1 % of hydrogen 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.