Supporting information
Crystallographic Information File (CIF) https://doi.org/10.1107/S1600536807036355/br2051sup1.cif | |
Structure factor file (CIF format) https://doi.org/10.1107/S1600536807036355/br2051Isup2.hkl |
MgSO4.7H2O (Merck, p.A.) was dehydrated at 973 K for 12 h in an open porcelain crucible. X-ray powder diffraction (XRPD) showed a single phase product. 0.5 g of the polycrystalline material was mixed with 50 mg PtCl2 and heated in sealed and evacuated silica ampoules in a temperature gradient 1173 → 1073 K for one week. Under these conditions PtCl2 is decomposed and the released Cl2 acts as the transport agent. After the reaction time, the ampoule was taken out of the two-zone furnace and was quenched in a cold water bath. Only few single crystals of β-MgSO4 with an unspecific habit and maximal edge lengths up to 0.4 mm were obtained in the colder zone of the ampoule, indicating rather small transport rates which has also been observed in previous studies (Gruehn & Glaum, 2000).
In contrast to the previous refinement from powder data (Coing-Boyat, 1962) where the non-standard setting Pbnm of space group No 64 was used, the structure was refined in the standard setting Pnma. Atomic coordinates were taken from the isotypic compound ZnSO4 (Wildner & Giester, 1988) as starting parameters and finally standardized with the program STRUCTURETIDY (Gelato & Parthé, 1987). The present study confirms the basic structural features determined from the previous investigation by Coing-Boyat (1962), but with a much higher precesion and more reliable interatomic distances.
Magnesium sulfate is dimorphic and crystallizes in a low-temperature modification (α-MgSO4; space group Cmcm, CrVO4 structure type) and a high-temperature modification (β-MgSO4, space group Pnma, CuSO4 structure type). The corresponding crystal structures have already been determined from intensity data using the Debye–Scherrer method (α-MgSO4: Rentzeperis & Soldatos, 1958; β-MgSO4: Coing-Boyat, 1962). The previous structure refinement of β-MgSO4 converged at a reliability factor R[F] = 0.144 without inclusion of temperature factors and indication of standard uncertainties. In order to obtain more precise results for comparative studies with other MIISO4 phases, where M is a first row transition metal, the crystal structure of β-MgSO4 was re-determined by means of single-crystal data.
The crystal structure of β-MgSO4 contains one Mg, one S and three O atoms in the asymmetric unit. The basic structural features are [MgO4/2O2/1]∞ chains made up of edge-sharing [MgO6] octahedra and SO4 tetrahedra. The chains run parallel to [010] and are interconnected by corner-sharing with the SO4 tetrahedra into a framework structure (Fig. 1, 2).
The [MgO6] octahedron (1 point symmetry) is considerably distorted and shows a [2 + 2+2] coordination, with two short Mg—O distances to the terminal O atoms and two medium and two long distances to the bridging O atoms of the [MgO4/2O2/1]∞ chains. However, the average Mg—O distance of 2.114 Å is in good agreement with the sum of the ionic radii (2.08 Å; Shannon, 1976).
The SO4 tetrahedron (m point symmetry) is slightly distorted, with an average S—O bond length of 1.471 Å which is likewise in very good agreement with the value of 1.473 Å given by Baur (1981) for more than 100 S—O bonds in various sulfates(VI).
The O atoms have coordination numbers of 2 (O1) and 3 (O2, O3). O1 has one Mg and one S as neighbours, both with the shortest observed Mg– and S—O distances. O2 and O3 act as the bridging atoms in the [MgO4/2O2/1]∞ chains and thus have two Mg and one S as coordination partner.
Results from the bond valence sum (BVS) calculations (Brown, 2002), using the parameters of Brese & O'Keeffe (1991), are in accordance with the expected values (in valence units) of 2 for Mg, 6 for S and 2 for O: Mg 2.02, S 6.05, O1 2.04, O2 2.09, O3 1.90.
During the submission process of the present article the author was informed that the crystal structures of α- and β-MgSO4 have also been re-determined on the basis of neutron powder diffraction data. The results (lattice parameters, atomic coordinates, standard uncertainties, distances) published very recently (Fortes et al., 2007), are very similar to those of the single-crystal study of β-MgSO4 presented here. However, in contrast to the neutron powder study of β-MgSO4 at 300 K (Fortes et al., 2007), no constraints had to be applied during refinement. Therefore the present study is considered as an accurate supplement of the neutron powder study of β-MgSO4 at room temperature.
For standardization of structure data, see Gelato & Parthé (1987). The structure of the second polymorph of MgSO4 is described by Rentzeperis & Soldatos (1958). Very recently, redeterminations of the structures of both α- and β-MgSO4 by neutron powder diffraction were published by Fortes et al. (2007). [Please check added text and rephrase as necessary. Should this also be mentioned in the Abstract?] For a review of chemical transport reactions for preparative purposes, including metal sulfates, see Gruehn & Glaum (2000). An overview of isotypic sulfates of the CuSO4 and the CrVO4 structure types is given by Wildner & Giester (1988). For the bond-valence model, see Brown (2002) and Brese & O'Keeffe (1991). Average S—O distances were calculated by Baur (1981) and ionic radii were taken from Shannon (1976).
Data collection: CAD-4 Software (Enraf–Nonius, 1989); cell refinement: CAD-4 Software; data reduction: HELENA implemented in PLATON (Spek, 2003); program(s) used to solve structure: coordinates taken from an isotypic structure (Wildner & Giester, 1988); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: ATOMS (Dowty, 2000); software used to prepare material for publication: SHELXL97.
MgSO4 | F(000) = 240 |
Mr = 120.37 | Dx = 2.934 Mg m−3 |
Orthorhombic, Pnma | Mo Kα radiation, λ = 0.71073 Å |
Hall symbol: -P 2ac 2n | Cell parameters from 25 reflections |
a = 8.5787 (8) Å | θ = 11.3–17.4° |
b = 6.6953 (6) Å | µ = 1.21 mm−1 |
c = 4.7438 (5) Å | T = 293 K |
V = 272.47 (5) Å3 | Plate, colourless |
Z = 4 | 0.18 × 0.14 × 0.09 mm |
Enraf–Nonius CAD-4 diffractometer | 598 reflections with I > 2σ(I) |
Radiation source: fine-focus sealed tube | Rint = 0.032 |
Graphite monochromator | θmax = 36.0°, θmin = 4.8° |
ω/2θ scans | h = −14→14 |
Absorption correction: ψ scan (PLATON; Spek, 2003) | k = −11→11 |
Tmin = 0.780, Tmax = 0.864 | l = −7→7 |
4736 measured reflections | 3 standard reflections every 120 min |
683 independent reflections | intensity decay: none |
Refinement on F2 | Primary atom site location: isomorphous structure methods |
Least-squares matrix: full | w = 1/[σ2(Fo2) + (0.0233P)2 + 0.0698P] where P = (Fo2 + 2Fc2)/3 |
R[F2 > 2σ(F2)] = 0.017 | (Δ/σ)max < 0.001 |
wR(F2) = 0.046 | Δρmax = 0.48 e Å−3 |
S = 1.13 | Δρmin = −0.41 e Å−3 |
683 reflections | Extinction correction: SHELXL97 (Sheldrick, 1997), Fc*=kFc[1+0.001xFc2λ3/sin(2θ)]-1/4 |
35 parameters | Extinction coefficient: 0.144 (8) |
0 restraints |
MgSO4 | V = 272.47 (5) Å3 |
Mr = 120.37 | Z = 4 |
Orthorhombic, Pnma | Mo Kα radiation |
a = 8.5787 (8) Å | µ = 1.21 mm−1 |
b = 6.6953 (6) Å | T = 293 K |
c = 4.7438 (5) Å | 0.18 × 0.14 × 0.09 mm |
Enraf–Nonius CAD-4 diffractometer | 598 reflections with I > 2σ(I) |
Absorption correction: ψ scan (PLATON; Spek, 2003) | Rint = 0.032 |
Tmin = 0.780, Tmax = 0.864 | 3 standard reflections every 120 min |
4736 measured reflections | intensity decay: none |
683 independent reflections |
R[F2 > 2σ(F2)] = 0.017 | 35 parameters |
wR(F2) = 0.046 | 0 restraints |
S = 1.13 | Δρmax = 0.48 e Å−3 |
683 reflections | Δρmin = −0.41 e Å−3 |
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. |
Refinement. Refinement of F2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F2, conventional R-factors R are based on F, with F set to zero for negative F2. The threshold expression of F2 > σ(F2) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F2 are statistically about twice as large as those based on F, and R- factors based on ALL data will be even larger. |
x | y | z | Uiso*/Ueq | ||
Mg | 0.0000 | 0.0000 | 0.0000 | 0.01046 (10) | |
S | 0.32046 (3) | 0.2500 | 0.02598 (5) | 0.00547 (8) | |
O1 | 0.37429 (7) | 0.06895 (9) | 0.16526 (13) | 0.01011 (12) | |
O2 | 0.14650 (10) | 0.2500 | 0.03881 (18) | 0.00824 (14) | |
O3 | 0.37517 (10) | 0.2500 | 0.73002 (17) | 0.00961 (15) |
U11 | U22 | U33 | U12 | U13 | U23 | |
Mg | 0.01031 (18) | 0.00964 (18) | 0.01143 (19) | −0.00266 (14) | 0.00449 (13) | −0.00363 (13) |
S | 0.00485 (11) | 0.00647 (11) | 0.00508 (11) | 0.000 | −0.00029 (7) | 0.000 |
O1 | 0.0113 (3) | 0.0090 (2) | 0.0101 (2) | 0.00216 (19) | −0.00221 (19) | 0.00243 (18) |
O2 | 0.0047 (3) | 0.0088 (3) | 0.0112 (3) | 0.000 | 0.0004 (2) | 0.000 |
O3 | 0.0108 (4) | 0.0121 (3) | 0.0059 (3) | 0.000 | 0.0020 (3) | 0.000 |
Mg—O1i | 1.9743 (6) | Mg—Mgiv | 3.3477 (3) |
Mg—O1ii | 1.9743 (6) | Mg—Mgv | 3.3477 (3) |
Mg—O2iii | 2.1012 (5) | S—O1vi | 1.4558 (6) |
Mg—O2 | 2.1012 (5) | S—O1 | 1.4558 (6) |
Mg—O3ii | 2.2670 (6) | S—O3vii | 1.4803 (9) |
Mg—O3i | 2.2670 (6) | S—O2 | 1.4936 (9) |
O1i—Mg—O1ii | 180.00 (5) | O3ii—Mg—O3i | 180.00 (4) |
O1i—Mg—O2iii | 94.01 (3) | O1vi—S—O1 | 112.75 (5) |
O1ii—Mg—O2iii | 85.99 (3) | O1vi—S—O3vii | 109.26 (3) |
O1i—Mg—O2 | 85.99 (3) | O1—S—O3vii | 109.26 (3) |
O1ii—Mg—O2 | 94.01 (3) | O1vi—S—O2 | 107.37 (3) |
O2iii—Mg—O2 | 180.00 (4) | O1—S—O2 | 107.37 (3) |
O1i—Mg—O3ii | 92.50 (3) | O3vii—S—O2 | 110.82 (5) |
O1ii—Mg—O3ii | 87.50 (3) | S—O1—Mgviii | 137.14 (4) |
O2iii—Mg—O3ii | 105.28 (3) | S—O2—Mg | 126.44 (2) |
O2—Mg—O3ii | 74.72 (3) | S—O2—Mgv | 126.44 (2) |
O1i—Mg—O3i | 87.50 (3) | Mg—O2—Mgv | 105.61 (4) |
O1ii—Mg—O3i | 92.50 (3) | Six—O3—Mgviii | 127.32 (2) |
O2iii—Mg—O3i | 74.72 (3) | Six—O3—Mgx | 127.32 (2) |
O2—Mg—O3i | 105.28 (3) | Mgviii—O3—Mgx | 95.18 (3) |
Symmetry codes: (i) −x+1/2, −y, z−1/2; (ii) x−1/2, y, −z+1/2; (iii) −x, −y, −z; (iv) −x, y−1/2, −z; (v) −x, y+1/2, −z; (vi) x, −y+1/2, z; (vii) x, y, z−1; (viii) −x+1/2, −y, z+1/2; (ix) x, y, z+1; (x) x+1/2, −y+1/2, −z+1/2. |
Experimental details
Crystal data | |
Chemical formula | MgSO4 |
Mr | 120.37 |
Crystal system, space group | Orthorhombic, Pnma |
Temperature (K) | 293 |
a, b, c (Å) | 8.5787 (8), 6.6953 (6), 4.7438 (5) |
V (Å3) | 272.47 (5) |
Z | 4 |
Radiation type | Mo Kα |
µ (mm−1) | 1.21 |
Crystal size (mm) | 0.18 × 0.14 × 0.09 |
Data collection | |
Diffractometer | Enraf–Nonius CAD-4 |
Absorption correction | ψ scan (PLATON; Spek, 2003) |
Tmin, Tmax | 0.780, 0.864 |
No. of measured, independent and observed [I > 2σ(I)] reflections | 4736, 683, 598 |
Rint | 0.032 |
(sin θ/λ)max (Å−1) | 0.826 |
Refinement | |
R[F2 > 2σ(F2)], wR(F2), S | 0.017, 0.046, 1.13 |
No. of reflections | 683 |
No. of parameters | 35 |
Δρmax, Δρmin (e Å−3) | 0.48, −0.41 |
Computer programs: CAD-4 Software (Enraf–Nonius, 1989), CAD-4 Software, HELENA implemented in PLATON (Spek, 2003), coordinates taken from an isotypic structure (Wildner & Giester, 1988), SHELXL97 (Sheldrick, 1997), ATOMS (Dowty, 2000), SHELXL97.
Mg—O1i | 1.9743 (6) | S—O1 | 1.4558 (6) |
Mg—O2 | 2.1012 (5) | S—O3iii | 1.4803 (9) |
Mg—O3ii | 2.2670 (6) | S—O2 | 1.4936 (9) |
Symmetry codes: (i) x−1/2, y, −z+1/2; (ii) −x+1/2, −y, z−1/2; (iii) x, y, z−1. |
Magnesium sulfate is dimorphic and crystallizes in a low-temperature modification (α-MgSO4; space group Cmcm, CrVO4 structure type) and a high-temperature modification (β-MgSO4, space group Pnma, CuSO4 structure type). The corresponding crystal structures have already been determined from intensity data using the Debye–Scherrer method (α-MgSO4: Rentzeperis & Soldatos, 1958; β-MgSO4: Coing-Boyat, 1962). The previous structure refinement of β-MgSO4 converged at a reliability factor R[F] = 0.144 without inclusion of temperature factors and indication of standard uncertainties. In order to obtain more precise results for comparative studies with other MIISO4 phases, where M is a first row transition metal, the crystal structure of β-MgSO4 was re-determined by means of single-crystal data.
The crystal structure of β-MgSO4 contains one Mg, one S and three O atoms in the asymmetric unit. The basic structural features are [MgO4/2O2/1]∞ chains made up of edge-sharing [MgO6] octahedra and SO4 tetrahedra. The chains run parallel to [010] and are interconnected by corner-sharing with the SO4 tetrahedra into a framework structure (Fig. 1, 2).
The [MgO6] octahedron (1 point symmetry) is considerably distorted and shows a [2 + 2+2] coordination, with two short Mg—O distances to the terminal O atoms and two medium and two long distances to the bridging O atoms of the [MgO4/2O2/1]∞ chains. However, the average Mg—O distance of 2.114 Å is in good agreement with the sum of the ionic radii (2.08 Å; Shannon, 1976).
The SO4 tetrahedron (m point symmetry) is slightly distorted, with an average S—O bond length of 1.471 Å which is likewise in very good agreement with the value of 1.473 Å given by Baur (1981) for more than 100 S—O bonds in various sulfates(VI).
The O atoms have coordination numbers of 2 (O1) and 3 (O2, O3). O1 has one Mg and one S as neighbours, both with the shortest observed Mg– and S—O distances. O2 and O3 act as the bridging atoms in the [MgO4/2O2/1]∞ chains and thus have two Mg and one S as coordination partner.
Results from the bond valence sum (BVS) calculations (Brown, 2002), using the parameters of Brese & O'Keeffe (1991), are in accordance with the expected values (in valence units) of 2 for Mg, 6 for S and 2 for O: Mg 2.02, S 6.05, O1 2.04, O2 2.09, O3 1.90.
During the submission process of the present article the author was informed that the crystal structures of α- and β-MgSO4 have also been re-determined on the basis of neutron powder diffraction data. The results (lattice parameters, atomic coordinates, standard uncertainties, distances) published very recently (Fortes et al., 2007), are very similar to those of the single-crystal study of β-MgSO4 presented here. However, in contrast to the neutron powder study of β-MgSO4 at 300 K (Fortes et al., 2007), no constraints had to be applied during refinement. Therefore the present study is considered as an accurate supplement of the neutron powder study of β-MgSO4 at room temperature.