Download citation
Download citation
link to html
Single crystals of magnesium diiodide have been grown and the structure solved for the first time from single-crystal X-ray diffraction data. This study confirms that MgI2 is isostructural with CdI2 (C6 or 2H structure type). The space group is P\bar3m1 with the Mg atom on a site with 3m symmetry (Wyckoff site 1a) and the I atom on a site with m symmetry (Wyckoff site 2d). Trends in the 2H structures of dihalides are discussed briefly.

Supporting information

cif

Crystallographic Information File (CIF) https://doi.org/10.1107/S0108270103025769/bc1028sup1.cif
Contains datablocks global, I

hkl

Structure factor file (CIF format) https://doi.org/10.1107/S0108270103025769/bc1028Isup2.hkl
Contains datablock I

Comment top

MgI2 belongs to the series of known dihalides (AX2, where X = F, Cl, Br). Surprisingly, single-crystal growth of MgI2 has never been reported and no refined structure exists. Seventy years ago, Blum et al. (1933) derived a structural model from powder X-ray diffraction data. The conclusion reached was that MgI2 crystallizes in a layered structure isotypic with CdI2 (defined as C6 or 2H type). Here, the growth of good quality single crystals has allowed the structure of MgI2 to be determined for the first time.

The single-crystal diffraction data set obtained confirms that MgI2 crystallizes in the trigonal space group P3m1, adopting the same hexagonal close-packed layered structure as CdI2, with unit-cell parameters a = 4.1537 (7) Å, c = 6.862 (2) Å, V = 102.53 (4) Å3, Z = 2, and c/a = 1.652. Within the structure, the I anions are hexagonally close packed, whilst the Mg2+ cations occupy all of the octahedral holes between alternate layers of the close-packed anions. Hence, one-half of the available octahedral holes throughout the structure are occupied. The structure can also be thought of as being built up from layers of MgI6 octahedra. Each such octahedron shares edges with six adjacent identical units within each MgI2 layer (Fig. 1).

The six equal Mg—I bond distances of 2.9183 (5) Å and the I—Mg—I bond angles (ϕ) of 90.739 (17)° describe a slightly distorted MgI6 octahedron. Interestingly, the only other example of a magnesium iodide in the literature with a refined crystal structure is the ternary compound CsMgI3 (McPherson et al., 1980), in which the Mg—I distance is 2.899 (4) Å, in good agreement with the value obtained in this work. Apart from the original work of Blum et al. (1933), the only other measurement of the Mg—I distance in MgI2 was performed by Akishin & Spiridonov (1958), who used electron diffraction to determine the bond length in the gas phase. Unsurprisingly, the value of 2.52 Å in the triatomic is significantly shorter than the value we observe in the solid state. The M—X distances differ slightly from the corresponding values in CdI2 (2.989 Å; Palosz & Salje, 1989), in which the octahedral CdI6 unit is slightly more distorted than the MgI6 octahedron, with I—Cd—I bond angles of 90.49°. Iodide anions in MgI2 are separated by intraplanar distances of 4.1009 (11) and 4.1537 (7) Å across the thickness and along the length of the Mg–I layer, respectively. Magnesium cations are similarly separated within layers by 4.1537 (7) Å. Iodide ions in adjacent layers are separated by a longer interplanar distance of 4.2722 (12) Å.

The notable feature of this diiodide and the structure type in general is the van der Waals interlayer gap between hexagonally close packed I ions. The interlayer gap (d) in MgI2 [3.535 (2) Å] is significantly wider than the Mg–I layer thickness (t) [3.326 (2) Å]. The same trend is observed in CdI2, but the difference between these two distances is far smaller. Furthermore, by contrast, the absolute value of d in CdI2 is smaller than in MgI2, whereas t is larger. The relevant structural information for both MgI2 and CdI2 appears in Table 2. The values for other isostructural dihalides are included for comparison. Note here that the data for FeCl2 represent the high-pressure polymorph at 6.4 K bar (1 bar = 10 5 Pa) (Vettier & Yelon, 1975).

The c/a ratio observed here for MgI2 is in close agreement with that previously reported (1.66; Blum, 1933). The t/d ratio obtained previously for MgI2 can only be inferred as 1 assuming the z position for I is 0.25 as reported. Table 2 reveals that the vast majority of dihalides with the CdI2 structure form with 1.6 c/a 1.65. PbI2 is perhaps the most notable exception in this respect, with the largest a parameter and A—X bond length, and consequently the smallest c/a ratio reported (Palosz et al., 1990). Interestingly, it is the only dihalide with t/d > 1. In fact, only a select number of the dihalide 2H structures reported in the literature and depicted in Table 2 have been refined. In the remainder of cases, the halide z coordinate, z(X), is assumed to be 0.25. A number of trends are nonetheless evident from the pool of refined data. As might be expected, the ratio t/d increases with increasing A—X bond length and decreasing X—A—X bond angle (ϕ). Bonding in the essentially ionic layers (manifested in t) generally plays a more significant role here than the weaker anion replusion between layers (influencing d). Naturally, the halide z coordinate (2 d site; 1/3, 2/3, z) follows the same trends with A—X and ϕ. Interestingly, again here, PbI2 also possesses the most covalent A—X bond, thus minimizing interlayer repulsion. It is, therefore, not so unexpected that d < t in this compound. In fact, there is a good correlation between ionic character of the A—X bond, as defined by Pauling (1960), and the t/d ratio (Fig. 2). A similar correspondance obviously exists between z(X) and the bond ionicity. The c/a ratio is broadly constant across the structure type (and invariant with A—X and ϕ) but, within the observed range, increases slightly with decreasing t/d [z(X)].

As the number of accurately refined metal dihalide structures increases, the relationships between the various structural variables should become clearer to the point where it should become possible to predict (interpolate) z(X), and hence bond length, on the basis of the A—X bond ionicity or the c/a ratio. The continuing advances made in structure determination from powder data should facilitate and expediate this process.

Experimental top

The MgI2 crystals were produced during the attempted reaction of MgI2 (Aldrich, anhydrous beads 99.9%, ground into a polycrystalline powder) with Mg3N2. Mg3N2 powder was prepared by the reaction of pure magnesium (Strem Chemicals, 99.9% purity) with dry nitrogen gas at 973 K. Due to the air sensitivity of the reactants and products involved, all manipulations were carried out in glove-boxes (either recirculating nitrogen-filled or evacuable argon-filled). Stoichiometric ratios of the reactants were thoroughly mixed and ground together, then pressed to form a pellet (ca 1 g). The pellet was placed in a molybdenum foil liner and transferred to a stainless steel crucible, which was subsequently welded shut in an argon atmosphere. The sealed crucibles were heated in a tube furnace (1323 K, 5 d) under flowing argon to prevent oxidation of the steel crucibles. The furnace was cooled slowly (20 K h−1). The crucibles were opened in an N2-filled glove-box. Opaque irregular crystals were observed on the molybdenum foil. Crystals were selected in a recirculating N2 filled glove-box under an optical microscope and placed under RS3000 perfluoropolyether (Riedel de Haën) on a microscope slide prior to mounting on the diffractometer. The moisture-free viscous perfluoropolyether protects the crystals from atmospheric oxygen and moisture without interfering with the diffraction experiment.

Computing details top

Data collection: SMART (Bruker, 2001); cell refinement: SAINT (Bruker, 2002); data reduction: SAINT and SHELXTL (Bruker, 2001); program(s) used to solve structure: SIR97 (Altomare et al., 1999); program(s) used to refine structure: SHELXL97 (Sheldrick, 1997); molecular graphics: ATOMS (Dowty, 1998); software used to prepare material for publication: enCIFer (CCDC, 2003), PLATON (Spek, 1990, 2002) and WinGX (Farrugia, 1998).

Figures top
[Figure 1] Fig. 1. The coordination environment of Mg2+ (dark-blue ellipsoids) and I (cyan ellipsoids) within octahedral layers. Displacement ellipsoids are drawn at the 50% probability level.
[Figure 2] Fig. 2. Plot of layer thickness/interlayer gap ratio (t/d) versus ionic character (ionicity) of the A—X bond. Data are taken from Table 2.
(I) top
Crystal data top
MgI2Dx = 4.504 Mg m3
Mr = 278.11Mo Kα radiation, λ = 0.71073 Å
Trigonal, P3m1Cell parameters from 529 reflections
a = 4.1537 (7) Åθ = 3.0–27.3°
c = 6.862 (2) ŵ = 15.24 mm1
V = 102.53 (4) Å3T = 150 K
Z = 1Plate, colourless
F(000) = 1180.20 × 0.20 × 0.08 mm
Data collection top
Bruker SMART1000 CCD area-detector
diffractometer
115 independent reflections
Radiation source: normal-focus sealed tube112 reflections with I > 2σ(I)
Graphite monochromatorRint = 0.081
ω scansθmax = 27.3°, θmin = 3.0°
Absorption correction: integration
(SHELXTL; Bruker, 2001)
h = 54
Tmin = 0.093, Tmax = 0.505k = 54
623 measured reflectionsl = 85
Refinement top
Refinement on F20 restraints
Least-squares matrix: fullPrimary atom site location: structure-invariant direct methods
R[F2 > 2σ(F2)] = 0.021 w = 1/[σ2(Fo2) + (0.037)2 + 0.102P]
where P = (Fo2 + 2Fc2)/3
wR(F2) = 0.063(Δ/σ)max = 0.001
S = 1.13Δρmax = 0.86 e Å3
115 reflectionsΔρmin = 0.91 e Å3
7 parameters
Crystal data top
MgI2Z = 1
Mr = 278.11Mo Kα radiation
Trigonal, P3m1µ = 15.24 mm1
a = 4.1537 (7) ÅT = 150 K
c = 6.862 (2) Å0.20 × 0.20 × 0.08 mm
V = 102.53 (4) Å3
Data collection top
Bruker SMART1000 CCD area-detector
diffractometer
115 independent reflections
Absorption correction: integration
(SHELXTL; Bruker, 2001)
112 reflections with I > 2σ(I)
Tmin = 0.093, Tmax = 0.505Rint = 0.081
623 measured reflections
Refinement top
R[F2 > 2σ(F2)] = 0.0217 parameters
wR(F2) = 0.0630 restraints
S = 1.13Δρmax = 0.86 e Å3
115 reflectionsΔρmin = 0.91 e Å3
Special details top

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 (Å2) top
xyzUiso*/Ueq
Mg0.00001.00001.00000.0142 (9)
I0.33330.66670.75763 (6)0.0120 (3)
Atomic displacement parameters (Å2) top
U11U22U33U12U13U23
Mg0.0091 (11)0.0091 (11)0.024 (2)0.0045 (6)0.0000.000
I0.0105 (4)0.0105 (4)0.0150 (5)0.00525 (18)0.0000.000
Geometric parameters (Å, º) top
Mg—I2.9183 (5)
Mgi—I—Mg90.739 (17)
Symmetry code: (i) x+1, y, z.

Experimental details

Crystal data
Chemical formulaMgI2
Mr278.11
Crystal system, space groupTrigonal, P3m1
Temperature (K)150
a, c (Å)4.1537 (7), 6.862 (2)
V3)102.53 (4)
Z1
Radiation typeMo Kα
µ (mm1)15.24
Crystal size (mm)0.20 × 0.20 × 0.08
Data collection
DiffractometerBruker SMART1000 CCD area-detector
diffractometer
Absorption correctionIntegration
(SHELXTL; Bruker, 2001)
Tmin, Tmax0.093, 0.505
No. of measured, independent and
observed [I > 2σ(I)] reflections
623, 115, 112
Rint0.081
(sin θ/λ)max1)0.646
Refinement
R[F2 > 2σ(F2)], wR(F2), S 0.021, 0.063, 1.13
No. of reflections115
No. of parameters7
Δρmax, Δρmin (e Å3)0.86, 0.91

Computer programs: SMART (Bruker, 2001), SAINT (Bruker, 2002), SAINT and SHELXTL (Bruker, 2001), SIR97 (Altomare et al., 1999), SHELXL97 (Sheldrick, 1997), ATOMS (Dowty, 1998), enCIFer (CCDC, 2003), PLATON (Spek, 1990, 2002) and WinGX (Farrugia, 1998).

Selected geometric parameters (Å, º) top
Mg—I2.9183 (5)
Mgi—I—Mg90.739 (17)
Symmetry code: (i) x+1, y, z.
Selected structural data (Å, °) for MX2 dihalides with the CdI2 (2H) structure top
MX2M—Xaϕac/atdt/dbz(X)
CdI2i2.98990.491.623.42113.4430.990.2492
PbI2ii3.22790.161.533.7383.2481.150.2675 (2)
MgI22.9183 (5)90.739 (17)1.653.32673.53530.940.24237 (6)
MnI2iii2.9290.441.653.3463.4820.960.245 (2)
MnBr2iv2.72990.271.623.1363.1361.000.25
TiBr2v2.6593.591.793.2463.2461.000.25
FeBr2vi2.63691.471.652.9653.2610.910.238 (5)
VCl2vii2.53890.321.622.9152.9151.000.25
TiCl2viii2.52790.401.652.9372.9371.000.25
MgCl2ix2.50593.211.632.7263.20.850.23
TiCl2x2.49993.351.783.053.051.000.25
FeCl2xi2.48392.411.62.7442.990.920.2393 (2)
CaI2xii3.1293.3131.553.483.4810.25
Notes: (a) in some cases, s.u.values are not reported in the literature; (b) where t/d = 1, literature assumes that z(X) =1/4, t = thickness of layer, d = interlayer gap. References: (i) Palosz & Salje (1989); (ii) Palosz, Steurer & Schulz (1990); (iii) Cable, Wilkinson & Wollan (1962); (iv) Wollan, Koehler & Wilkinson (1958); (v) Ehrlich, Gutsche & Seifert (1961); (vi) Haberecht, Borrmann & Kniep (2001); (vii) Ehrlich & Seifert (1959); (viii) Baenziger & Rundle (1948); (ix) Bassi, Polato, Calcaterra & Bart (1982); (x) Gal'perin & Sandler (1962); (xi) Vettier & Yelon (1975); (xii) Blum (1933).
 

Follow Acta Cryst. C
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds