Triaqua­manganese sulfite revisited

Ortho­rhom­bic Mn(SO₃)(H₂O)₃ has been reinvestigated by single-crystal X-ray diffraction in two possible space groups, viz. P2₁2₁2₁ (with all atoms in general positions) and Pnma (with the mol­ecule bisected by a mirror plane). The results confirm the lower symmetry assigned in a previous single-crystal neutron diffraction study. However, the refinement of the P2₁2₁2₁ model requires the introduction of racemic twinning and soft positional and displacement restraints for the H atoms. The importance of a scrupulous report on symmetry absence violations as standard policy in crystallographic work is discussed.

S1. Comment

Mn(SO₃)(H₂O)₃ is known to crystallize in two polymorphic varieties, viz. a monoclinic form (Engelen & Freiburg, 1979; Johansson & Lindqvist,1980) and an orthorhombic form (Baggio and Baggio, 1976; Engelen, 1983; Basso et al., 1991; Gonschorek et al., 1996). Both structures present the same local coordination scheme (SO₃ bound to three different cationic units, via its three O atoms) but with a different three-dimensional connectivity (to be discussed below). While the structure of the monoclinic form is well established, that of the orthorhombic form still presents an apparent contradiction between the Pnma symmetry originally determined and the P2₁2₁2₁ symmetry associated with an ordered distribution of its aqua H atoms.

X-ray studies could not distinguish definitively between the options `Pnma + splitting' and `P2₁2₁2₁ + ordering'. As a result of `real symmetry' or `strong pseudo-symmetry', both Pnma and P2₁2₁2₁ treatments would result in similar difference Fourier maps, from which it was almost impossible to differentiate genuine H atoms from eventual ghost images. A single-crystal neutron diffraction study (Gonschorek et al., 1996) determined P2₁2₁2₁ as the correct space group, at least for the deuterated samples used. Unfortunately, this seemingly accurate work ended up in a report impaired by a number of numerical inconsistencies, some of which affected relevant atomic coordinates involved in the `hot' part of the structure [i.e. the part of the structure most relevant to the present discussion?], and rendered much of the reported geometry meaningless [EDITOR: Please insert footnote mark here].

###################################################### Footnote: The reported coordinates for atoms O3 and D2 in Gonschorek et al. (1996) are clearly in error, as confirmed by their calculated distances to the Mn cation (not presented in the paper), Mn—O3 = 1.84 Å (expected ca 2.20 Å) and Mn···D2 = 1.58 Å (expected 2.65–2.85 Å), respectively. In fact, only a few of the reported bonds and angles can be derived precisely from the published coordinates. ######################################################

We present here the results of the structure refinement of a good X-ray data set in the two controversial space groups, namely (i) P2₁2₁2₁ and (ii) Pnma. Fig. 1 shows the corresponding ellipsoid plots and numbering schemes, while Tables 1 and 3, present relevant distances and angles. Some redundant values have been included in Table 3 in order to facilitate comparison with corresponding values in Table 1.

A simplified view of the H-pruned structure as obtainable in a difference Fourier synthesis is shown in Fig. 2(a). The `H-like' atoms drawn are the maxima clearly appearing in both centro- and non-centrosymmetric difference Fourier maps; the two maps look equivalent except for the presence of a mirror in Pnma (Fig. 1b) versus a pseudo-mirror in P2₁2₁2₁ (Fig. 1a).

First, there is a clear distinction between atom O1W with its two well behaved H atoms and the remaining atoms O2W and O3W, each showing three plausible maxima in the Fourier maps. These `proto-H atoms', in turn, fall into two categories, viz. (i) H2WA and H3WA (one for each aqua ligand), which do not present any possible `steric collision' with other atoms in the structure and are in principle good candidates to be real H atoms, and (ii) other H atoms that do present steric hindrance problems [in P2₁2₁2₁: H2WB···H3WC(2 - x, 1/2 + y, 1/2 - z] = 1.24 Å and H2WC···H3WB(x, 1 + y, z) = 0.93 Å; in Pnma: H2WB···H2WB(2 - x, 2 - y, 1 - z) = 1.34 Å and H2WC···H2WC(x, 2.5 - y, z) = 1.04 Å]. In P2₁2₁2₁, this problem could be overcome by the judicious choice of one H atom from each set (H2WB and H3WC, or H2WC and H3WB) as the correct partners of H2WA and H3WA. In Pnma, the existence of H2WB(C) would necessarily imply the existence of H3WB(C); in this case, only a split model is possible in the form of a 50% mixture of the two possible non-centrosymmetric solutions.

The ab-initio analysis of systematic absences (and their violations) favoured P2₁2₁2₁ rather than Pnma (Table 1) The strongest violation occurs for the multiply determined (0,4,1) reflection with an intensity I ≈ 60σ in seven different measurements in the whole data set. Overall, the strong pseudo-symmetry is apparent from the very low mean intensities of the offending reflections.

In order to test both hypotheses, three unrestrained refinements in the two possible space groups were performed, including two ordered models in P2₁2₁2₁ (one for each possible choice of the controversial H atoms, viz. H2WA, H2WB/H3WA, H3WC or H2WA, H2WC/H3WA, H3WB) and a disordered structure in Pnma, with a fully occupied H2WA and two half occupied H2WB and H2WC sites. The results of both P2₁2₁2₁ refinements yielded significantly lower R factors (0.0199 and 0.0215 versus 0.0238), though at the cost of nearly twice as many parameters and with the need of a (nearly equi-populated) racemic twin model (0.45/0.55%). In addition, the behaviour of the controversial H atoms during the completely free refinement was rather erratic and a smooth convergence could only be achieved when soft restraints were applied. In contrast, the Pnma model proved to be quite robust and, in spite of its slightly larger R factor, provided a sensible and stable solution even when severely over-refined.

Our conclusion is that Mn(SO₃)(H₂O)₃ crystallizes in the non-centrosymmetric P2₁2₁2₁ space group, as confirmed by the lower R indices and the violations of the expected systematic absences in Pnma. In ordinary circumstances, without previous information or with only a medium-quality X-ray data set, it would be impossible to discriminate between the P2₁2₁2₁ and Pnma space groups and the latter would be normally preferred. This result alerts us to the fact that it might sometimes be easy to overlook annoying violations of systematic absences.

For the sake of completeness, a very brief description of the P2₁2₁2₁ structure of Mn(SO₃)(H₂O)₃ follows: The basic structural unit consists of an Mn^II cation octahedrally coordinated to three water molecules and atoms O1, O2ⁱ and O3ⁱⁱ from three symmetry-related sulfite ions (Fig. 1a). The anion is bonded through each one of its O atoms to a different cation in a µ₃-mode to generate a double chain running along b, where both the cations and anios are surrounded by aqua ligands (Fig. 2b). These good hydrogen-bond donors find an adequate number of acceptors to build up an extremely complex hydrogen-bonding network. The interactions are either inter-chain (entries 1–3 in Table 2) or intra-chain (entries 4–6) and connect each chain to six others (Fig. 3). This chain structure is the most conspicuous difference between the orthorhombic and monoclinic forms of Mn(SO₃)(H₂O)₃. In fact, the latter presents a completely analogous cation coordination scheme but with a slightly different interconnectivity between polyhedra, which leads to a tightly woven three-dimensional covalent structure. These packing differences, however, do not affect the compactness of the material, as the two structures differ in their crystal densities by less than 1%.

S2. Experimental

The compound was prepared as described by Baggio & Baggio (1976) by allowing equimolar aqueous solutions of MnSO₄ and Na₂SO₃ to diffuse slowly in a U-shaped tube through a much denser Na₂SO₄ solution, which acted both as a separation medium (avoiding initial mixing of the original solutions) and as a deterring agent (slowing down the diffusion rate). After a few weeks, colourless prisms adequate for X-ray diffraction could be observed at the glass walls.

S3. Refinement

Owing to the real or strong pseudo-symmetry presented by non-H atoms in both possible space groups, three plausible candidates for the two H atoms corresponding to each one of the two controversial aqua molecules (O2W and O3W) clearly appeared in the difference Fourier map (see Fig. 2a). In P2₁2₁2₁, the correct set was chosen through a combination of steric arguments and least-squares refinement, which was extremely sensitive to the correct choice (R/wR factors: 0.0199/0.0520, 0.0215/0.0579). The model with lower R effectively seems to correspond to the set reported by Gonschorek et al. (1996). However, and in spite of the extremely accurate data set (R_int = 0.020), refinement of the H atoms in the non-centrosymmetric space group was not possible without the use of restraints [O—H = 0.80 (5) Å and U_iso(H) = 0.06 (4) Å^-1]. The refinement also required an allowance for racemic twinning (0.45/0.55%). In the Pnma refinement, the split H-atom positions and their U_iso values converged quite smoothly to reasonable values without the need of any external restraint.

Computing details

For both compounds, data collection: SMART-NT (Bruker, 2001); cell refinement: SAINT-NT (Bruker, 2000). Data reduction: SAINT-NT for I_P212121; SAINT-NT) for I_Pnma. For both compounds, program(s) used to solve structure: SHELXS97 (Sheldrick, 1997); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: SHELXTL-NT (Bruker, 2000); software used to prepare material for publication: SHELXTL-NT and PLATON (Spek, 2005).

Figures

	Fig. 1. Displacement ellipsoid plots (50% probability level) of the Mn polyhedron in the P212121(a) and Pnma (b) models. Symmetry independent atoms are drawn as filled ellipsoids. Symmetry codes as in Tables 1 and 3, respectively.
	Fig. 2. Schematic diagrams presenting the possible H-atom positions (a) and their correct assignment in P212121 (b). Note the formation of chains along the horizontal b direction.
	Fig. 3. Packing diagram in P212121 viewed down the chain direction, showing a central chain (bold lines) connected to six different neighbours (thin lines). The inter-chain hydrogen-bonding interactions are shown by broken lines.

(I_P212121) Triaquamanganese sulfite

Crystal data

Mn(SO3)(H2O)3	F(000) = 380
Mr = 189.05	Dx = 2.391 Mg m^−3
OrthorhombicP212121	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: P 2ac 2ab	Cell parameters from 2841 reflections
a = 9.7577 (10) Å	θ = 3.1–26.1°
b = 5.6319 (6) Å	µ = 2.86 mm^−1
c = 9.5579 (10) Å	T = 295 K
V = 525.25 (10) Å^3	Prism, colourless
Z = 4	0.16 × 0.10 × 0.08 mm

Data collection

Bruker SMART CCD area-detector diffractometer	1190 independent reflections
Radiation source: fine-focus sealed tube	1152 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.021
φ and ω scans	θmax = 27.9°, θmin = 3.0°
Absorption correction: multi-scan (SADABS; Sheldrick, 2001)	h = −12→12
Tmin = 0.658, Tmax = 0.804	k = −7→7
4391 measured reflections	l = −12→12

Refinement

Refinement on F^2	Secondary atom site location: difference Fourier map
Least-squares matrix: full	Hydrogen site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.020	All H-atom parameters refined
wR(F^2) = 0.052	w = 1/[σ^2(Fo^2) + (0.033P)^2 + 0.26P] where P = (Fo^2 + 2Fc^2)/3
S = 0.97	(Δ/σ)max < 0.001
1190 reflections	Δρmax = 0.28 e Å^−3
99 parameters	Δρmin = −0.29 e Å^−3
15 restraints	Absolute structure: Flack (1983), 424 Friedel pairs
Primary atom site location: structure-invariant direct methods	Absolute structure parameter: 0.45 (4)

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq
Mn	0.72347 (2)	0.74583 (7)	0.31243 (3)	0.01620 (10)
S	0.43471 (4)	0.74645 (13)	0.15987 (4)	0.01545 (11)
O1	0.58817 (13)	0.7523 (4)	0.13265 (14)	0.0283 (3)
O2	0.3783 (3)	0.9583 (3)	0.0795 (3)	0.0278 (6)
O3	0.3843 (3)	0.5299 (3)	0.0774 (3)	0.0257 (6)
O1W	0.86732 (16)	0.7570 (6)	0.48472 (15)	0.0337 (3)
O2W	0.8640 (3)	0.9976 (3)	0.2024 (4)	0.0225 (6)
O3W	0.8615 (3)	0.4888 (3)	0.2019 (3)	0.0206 (5)
H1WA	0.886 (5)	0.631 (6)	0.517 (5)	0.060 (13)*
H1WB	0.902 (5)	0.859 (7)	0.524 (6)	0.068 (14)*
H2WA	0.861 (6)	0.992 (7)	0.120 (4)	0.073 (14)*
H2WB	0.942 (3)	0.994 (5)	0.226 (3)	0.032 (8)*
H3WA	0.870 (4)	0.518 (5)	0.120 (3)	0.021 (8)*
H3WC	0.840 (3)	0.349 (5)	0.206 (3)	0.033 (8)*

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Mn	0.01598 (15)	0.01500 (15)	0.01762 (15)	0.0002 (2)	−0.00080 (9)	−0.0003 (2)
S	0.0173 (2)	0.0145 (2)	0.01450 (19)	0.0002 (4)	−0.00028 (14)	−0.0001 (3)
O1	0.0169 (6)	0.0451 (8)	0.0228 (7)	−0.0008 (12)	−0.0021 (5)	0.0005 (12)
O2	0.0404 (14)	0.0223 (9)	0.0207 (14)	0.0120 (10)	−0.0023 (17)	0.0008 (9)
O3	0.0374 (14)	0.0202 (9)	0.0195 (13)	−0.0117 (9)	−0.0002 (17)	−0.0013 (8)
O1W	0.0320 (8)	0.0433 (10)	0.0260 (7)	−0.0011 (15)	−0.0105 (6)	−0.0024 (15)
O2W	0.0206 (12)	0.0227 (10)	0.0244 (14)	−0.0013 (9)	0.0000 (16)	0.0002 (8)
O3W	0.0272 (13)	0.0177 (10)	0.0170 (12)	0.0007 (9)	0.0017 (15)	−0.0017 (8)

Geometric parameters (Å, º)

Mn—O1	2.1673 (13)	S—O3	1.533 (2)
Mn—O2^i	2.162 (2)	O1W—H1WA	0.79 (3)
Mn—O3^ii	2.185 (2)	O1W—H1WB	0.77 (3)
Mn—O1W	2.1647 (14)	O2W—H2WA	0.79 (3)
Mn—O2W	2.236 (2)	O2W—H2WB	0.79 (2)
Mn—O3W	2.242 (2)	O3W—H3WA	0.80 (3)
S—O1	1.5202 (13)	O3W—H3WC	0.81 (3)
S—O2	1.522 (2)
O1^i—Mn—O2	96.40 (10)	O1W—Mn—O2W	86.66 (11)
O1—Mn—O3^ii	94.40 (10)	O1W—Mn—O3W	89.28 (10)
O1—Mn—O1W	176.08 (8)	O2W—Mn—O3W	79.58 (5)
O1—Mn—O2W	89.43 (10)	O1—S—O2	104.65 (14)
O1—Mn—O3W	90.19 (10)	O1—S—O3	104.21 (13)
O2^i—Mn—O3^ii	95.58 (6)	O2—S—O3	104.36 (8)
O2^i—Mn—O1W	87.49 (11)	S—O1—Mn	117.64 (7)
O2^i—Mn—O2W	168.96 (9)	S—O2—Mn^ii	120.82 (14)
O2^i—Mn—O3W	91.00 (9)	S—O3—Mn^i	119.27 (14)
O3^ii—Mn—O1W	85.64 (11)	H1WA—O1W—H1WB	112 (3)
O3^ii—Mn—O2W	93.30 (9)	H2WA—O2W—H2WB	109 (4)
O3^ii—Mn—O3W	171.50 (9)	H3WA—O3W—H3WC	106 (3)

Symmetry codes: (i) −x+1, y−1/2, −z+1/2; (ii) −x+1, y+1/2, −z+1/2.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
REMOVE—*···*	The following lines beginning with REMOVE are dummy	?	?	?
REMOVE—*···*	lines included so as to systematize the symmetry codes	?	?	?
REMOVE—*···*	They are expected to be removed by the Editor at the	?	?	?
REMOVE—*···*	final edition stage. Thanks. RB.	?	?	?
REMOVE—X···X^i	Dummy line to be removed!!!!	?	?	?
REMOVE—X···X^ii	Dummy line to be removed!!!!	?	?	?
REMOVE—*···*	Please also remove from the footnote the (irrelevant)	?	?	?
REMOVE—*···*	symmetry codes i to ii herein introduced as artifacts	?	?	?
REMOVE—*···*	******************************************************	?	?	?
O1W—H1WA···O1^iii	0.79 (3)	2.44 (3)	3.227 (4)	172 (5)
O2W—H2WA···O3^iv	0.79 (3)	1.90 (4)	2.686 (5)	171 (6)
O3W—H3WA···O2^iv	0.80 (3)	1.92 (3)	2.711 (4)	171 (3)
O1W—H1WB···O1^v	0.77 (3)	2.42 (3)	3.134 (4)	155 (5)
O2W—H2WB···O3W^vi	0.79 (2)	2.04 (3)	2.830 (2)	176 (3)
O3W—H3WC···O2W^vii	0.81 (3)	1.99 (3)	2.766 (2)	158 (3)

Symmetry codes: (i) −x+1, y−1/2, −z+1/2; (ii) −x+1, y+1/2, −z+1/2; (iii) −x+3/2, −y+1, z+1/2; (iv) x+1/2, −y+3/2, −z; (v) −x+3/2, −y+2, z+1/2; (vi) −x+2, y+1/2, −z+1/2; (vii) x, y−1, z.

(I_Pnma) Triaquamanganese sulfite

Crystal data

Mn(SO3)(H2O)3	F(000) = 380
Mr = 189.05	Dx = 2.391 Mg m^−3
OrthorhombicPnma	Mo Kα radiation, λ = 0.71073 Å
Hall symbol: -P 2ac 2n	Cell parameters from 2841 reflections
a = 9.7577 (10) Å	θ = 3.1–26.1°
b = 5.6319 (6) Å	µ = 2.86 mm^−1
c = 9.5579 (10) Å	T = 295 K
V = 525.25 (10) Å^3	Prism, colourless
Z = 4	0.16 × 0.10 × 0.08 mm

Data collection

Bruker SMART CCD area-detector diffractometer	670 independent reflections
Radiation source: fine-focus sealed tube	661 reflections with I > 2σ(I)
Graphite monochromator	Rint = 0.022
φ and ω scans	θmax = 27.9°, θmin = 3.0°
Absorption correction: multi-scan (SADABS; Sheldrick, 2001)	h = −12→12
Tmin = 0.658, Tmax = 0.804	k = −7→7
4138 measured reflections	l = −12→12

Refinement

Refinement on F^2	Primary atom site location: structure-invariant direct methods
Least-squares matrix: full	Secondary atom site location: difference Fourier map
R[F^2 > 2σ(F^2)] = 0.024	Hydrogen site location: difference Fourier map
wR(F^2) = 0.068	All H-atom parameters refined
S = 1.35	w = 1/[σ^2(Fo^2) + (0.0226P)^2 + 0.7961P] where P = (Fo^2 + 2Fc^2)/3
670 reflections	(Δ/σ)max < 0.001
59 parameters	Δρmax = 0.30 e Å^−3
0 restraints	Δρmin = −0.33 e Å^−3

Special details

Geometry. All e.s.d.'s (except the e.s.d. in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell e.s.d.'s are taken into account individually in the estimation of e.s.d.'s in distances, angles and torsion angles; correlations between e.s.d.'s in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell e.s.d.'s is used for estimating e.s.d.'s involving l.s. planes.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Ų)

	x	y	z	Uiso*/Ueq	Occ. (<1)
Mn	0.72347 (5)	0.7500	0.56239 (5)	0.01633 (17)
S	0.43469 (8)	0.7500	0.40981 (8)	0.0155 (2)
O1	0.5881 (3)	0.7500	0.3828 (3)	0.0285 (6)
O2	0.3815 (2)	0.9648 (3)	0.32849 (18)	0.0270 (4)
O1W	0.8675 (3)	0.7500	0.7350 (3)	0.0345 (7)
H1W	0.896 (4)	0.850 (8)	0.772 (5)	0.074 (16)*
O2W	0.8626 (2)	1.0050 (3)	0.4522 (2)	0.0220 (4)
H2WA	0.870 (4)	0.971 (7)	0.371 (4)	0.047 (13)*
H2WB	0.934 (6)	0.999 (11)	0.476 (6)	0.027 (12)*	0.50
H2WC	0.834 (6)	1.163 (11)	0.456 (6)	0.029 (11)*	0.50

Atomic displacement parameters (Ų)

	U^11	U^22	U^33	U^12	U^13	U^23
Mn	0.0161 (3)	0.0152 (3)	0.0177 (3)	0.000	−0.00080 (18)	0.000
S	0.0173 (4)	0.0146 (4)	0.0145 (4)	0.000	−0.0003 (3)	0.000
O1	0.0169 (12)	0.0452 (17)	0.0233 (13)	0.000	−0.0021 (10)	0.000
O2	0.0394 (10)	0.0214 (9)	0.0202 (8)	0.0134 (8)	−0.0018 (7)	0.0001 (7)
O1W	0.0317 (15)	0.046 (2)	0.0260 (15)	0.000	−0.0116 (12)	0.000
O2W	0.0237 (10)	0.0215 (9)	0.0210 (9)	−0.0017 (8)	0.0010 (7)	0.0011 (8)

Geometric parameters (Å, º)

Mn—O1	2.166 (3)	S—O2^iii	1.5289 (18)
Mn—O2^i	2.1716 (18)	S—O2	1.5289 (18)
Mn—O2^ii	2.1716 (18)	O1W—H1W	0.72 (4)
Mn—O1W	2.167 (3)	O2W—H2WA	0.80 (4)
Mn—O2W	2.2396 (19)	O2W—H2WB	0.73 (6)
Mn—O2W^iii	2.2396 (19)	O2W—H2WC	0.93 (6)
S—O1	1.519 (3)
O1—Mn—O2^i	95.34 (7)	O2^ii—Mn—O2W^iii	170.43 (8)
O1—Mn—O2^ii	95.34 (7)	O1W—Mn—O2W	87.99 (9)
O1—Mn—O1W	177.14 (11)	O1W—Mn—O2W^iii	87.99 (9)
O1—Mn—O2W	89.82 (8)	O2W—Mn—O2W^iii	79.78 (10)
O1—Mn—O2W^iii	89.81 (8)	O1—S—O2^iii	104.38 (10)
O2^i—Mn—O2^ii	95.39 (11)	O1—S—O2	104.38 (10)
O2^i—Mn—O1W	86.58 (8)	O2^iii—S—O2	104.63 (15)
O2^i—Mn—O2W	170.43 (8)	H1W—O1W—H1W^iii	103 (5)
O2^ii—Mn—O2W	92.15 (7)	H2WA—O2W—H2WB	102 (5)
O2^ii—Mn—O1W	86.58 (8)	H2WA—O2W—H2WC	107 (4)
O2^i—Mn—O2W^iii	92.15 (7)	H2WB—O2W—H2WC	109 (6)

Symmetry codes: (i) −x+1, y−1/2, −z+1; (ii) −x+1, −y+2, −z+1; (iii) x, −y+3/2, z.

Hydrogen-bond geometry (Å, º)

D—H···A	D—H	H···A	D···A	D—H···A
REMOVE—*···*	The following lines beginning with REMOVE are dummy	?	?	?
REMOVE—*···*	lines included so as to systematize the symmetry codes	?	?	?
REMOVE—*···*	They are expected to be removed by the Editor at the	?	?	?
REMOVE—*···*	final edition stage. Thanks. RB.	?	?	?
REMOVE—X···X^i	Dummy line to be removed!!!!	?	?	?
REMOVE—X···X^ii	Dummy line to be removed!!!!	?	?	?
REMOVE—X···X^iii	Dummy line to be removed!!!!	?	?	?
REMOVE—*···*	Please also remove from the footnote the (irrelevant)	?	?	?
REMOVE—*···*	symmetry codes i to iii herein introduced as artifacts	?	?	?
REMOVE—*···*	******************************************************	?	?	?
O1W—H1W···O2^iv	0.72 (4)	2.47 (4)	3.062 (3)	141 (4)
O2W—H2WA···O2^v	0.80 (4)	1.91 (4)	2.698 (3)	167 (4)
O2W—H2WB···O2W^vi	0.73 (6)	2.10 (6)	2.834 (4)	178 (7)
O2W—H2WC···O2W^vii	0.93 (6)	1.89 (6)	2.759 (4)	154 (5)

Symmetry codes: (i) −x+1, y−1/2, −z+1; (ii) −x+1, −y+2, −z+1; (iii) x, −y+3/2, z; (iv) −x+3/2, −y+2, z+1/2; (v) x+1/2, y, −z+1/2; (vi) −x+2, −y+2, −z+1; (vii) x, −y+5/2, z.