Hexaaquaaluminium(III) tris(methanesulfonate)

The title compound, [Al(H2O)6](CH3SO3)3 (common name: aluminium methanesulfonate hexahydrate), was crystallized from an aqueous solution prepared by the precipitation reaction of aluminium sulfate and barium methanesulfonate. Its crystal structure is the first of the boron group methanesulfonates to be determined. The characteristic building block is a centrosymmetric unit containing two hexaaquaaluminium cations that are connected to each other by two O atoms of the –SO3 groups in an O—H⋯O⋯H—O sequence. Further O—H⋯O hydrogen bonding links these blocks in orthogonal directions – along [010] forming a double chain array, along [10-1] forming a layered arrangement of parallel chains and along [101] forming a three-dimensional network. As indicated by the O⋯O distances of 2.600 (3)–2.715 (3) Å, the hydrogen bonds are from medium–strong to strong. A further structural feature is the arrangement of two and four methyl groups, respectively, establishing ‘hydrophobic islands’ of different size, all positioned in a layer-like region perpendicular to [101]. The only other building block within this region is one of the –SO3 groups giving a local connection between the hydrophilic structural regions on both sides of the ‘hydrophobic’ one. Thermal analysis indicates that a stepwise dehydration process starts at about 413 K and proceeds via the respective penta- and dihydrate until the compound completely decomposes at about 688 K.

The title compound, [Al(H 2 O) 6 ](CH 3 SO 3 ) 3 (common name: aluminium methanesulfonate hexahydrate), was crystallized from an aqueous solution prepared by the precipitation reaction of aluminium sulfate and barium methanesulfonate. Its crystal structure is the first of the boron group methanesulfonates to be determined. The characteristic building block is a centrosymmetric unit containing two hexaaquaaluminium cations that are connected to each other by two O atoms of the -SO 3 groups in an O-HÁ Á ÁOÁ Á ÁH-O sequence. Further O-HÁ Á ÁO hydrogen bonding links these blocks in orthogonal directions -along [010] forming a double chain array, along [101] forming a layered arrangement of parallel chains and along [101] forming a three-dimensional network. As indicated by the OÁ Á ÁO distances of 2.600 (3)-2.715 (3) Å , the hydrogen bonds are from medium-strong to strong. A further structural feature is the arrangement of two and four methyl groups, respectively, establishing 'hydrophobic islands' of different size, all positioned in a layer-like region perpendicular to [101]. The only other building block within this region is one of the -SO 3 groups giving a local connection between the hydrophilic structural regions on both sides of the 'hydrophobic' one. Thermal analysis indicates that a stepwise dehydration process starts at about 413 K and proceeds via the respective penta-and dihydrate until the compound completely decomposes at about 688 K.

Experimental
According to some few initial studies that show aluminium methanesulfonate to have a high catalytic activity for certain syntheses of esters and acetals, the compound is supposed to be an attractive halogen free substitute for the frequently used aluminium trifluoromethanesulfonate, for instance Zhang, 2007).
Because water is a better coordinating ligand as compared to methanesulfonate (Paul et al., 1974), aluminium methanesulfonate not unexpectedly crystallized as the hexaaqua complex compound I from the aqueous solution we prepared by the precipitation reaction of aluminium sulfate and barium methanesulfonate. In contrast, recent work of other groups stated the tetrahydrate to be the only product crystallizing from aqueous solution Zhang, 2007). With chelating anions like oxalate (Taylor, 1978) and acetylacetonate (Hon & Pfluger, 1973;McClelland, 1975) no aqua but tris(chelate) complexes with these counterions directly engaged in the sixfold coordination about the aluminium cation are formed. sulfonate (Cameron et al., 1990) and the nitrate (Herpin & Sudarsanan, 1965;Lazar et al., 1991).
Despite the intrinsic threefold symmetry of all the ionic components as well as the quasi-hexagonal metric of the unit cell, there is no obvious relation of the solid state structure of I to any kind of close packing of ionic components. As In terms of graph-set analysis (Etter et al., 1990), there are entirely twelve motifs to be considered for a complete description of the hydrogen bond pattern. As no bifurcated bonds are present, these motifs can easily be denominated using the labels of the H atoms involved. The results of the graph set analysis presented here are mainly restricted to first order aspects. Main paths of O-H···O hydrogen bonding along the orthogonal directions [0 1 0], [1 0 -1] and [1 0 1] can be characterized by the graph sets C 1 2 (6) (H8, H10), C 3 4 (14) (H1, H9, H4, H12) and C 2 2 (8) (H6, H7), respectively. Some parts of these paths coincide with components of cyclic graphs (R 2 4 (12) (H4, H12), R 4 4 (14) (H1, H9)). From the 'molecular′ point of view, the characteristic building block of the solid is a centrosymmetric unit composed of two hexaaquaaluminium cations that are connected by single O atoms of two -SO 3 groups (R 2 4 (12) (H4, H12)) ( Fig. 3). By translation along [0 1 0] a double chain structure results. As mentioned above, the main path of hydrogen bonding in this direction is characterized by the graph set C 1 2 (6) with one acceptor O atom being part of the -SO 3 group including S2 (Fig. 4). Connecting elements perpendicular to the chain propagation direction are the methanesulfonate anions with atom S3 ([1 0 -1]) and atom S1 ([1 0 1]) (Fig. 5).
Usually, methanesulfonates tend to build layer-like structures with strictly separated hydrophilic and hydrophobic areas, the latter consisting of methyl groups connected by van der Waals forces (Trella et al., 2012). Although the structure of I is best described as a three-dimensional network, there is an obvious relationship to this principle of construction: Two and four methyl groups, respectively, establish 'hydrophobic islands′ of different size, all positioned in a layer-like region perpendicular to [1 0 1] (Fig. 5). The only other building block within this region is the -SO 3 group including S1, giving local connection between the hydrophilic structural regions at both sides of the 'hydrophobic′ one.

Refinement
A single-crystal suitable for structure determination was harvested from the mother liquor, directly transferred into the cooling stream of a Stoe IPDS diffractometer and investigated at -100 (2) °C. Thirteen reflections were excluded from the experiment, with one effected by the beam stop and twelve from the Lorentz zone of the one circle diffraction experiment, including two strong ones.
All H atom positions were located in difference Fourier maps. Positional parameters of hydrogen atoms belonging to water molecules were refined. In case of H5 and H7 the distance to the respective parent atoms O12 and O13 was restrained to 0.83 Å with a standard uncertainty of 0.03 Å. H atoms of the -CH 3 groups were treated applying angle constraints (H-C-H 109.5°; S-C-H 109.5°). They were free to rotate about the S-C bond and additionally the C-H distances were allowed to vary, with the same shifts being applied along all the C-H bonds of a group. Anisotropic displacement parameters of all non-hydrogen atoms and individual isotropic displacement parameters for all H atoms were refined.   The hexaaquaaluminium cation with its highly asymmetric coordination environment of ten sulfonate anions. [Symmetry codes: (i) x, 1 + y, z; (ii) 1.5-x, 0.5 + y, 1.5-z; (iii) -0.5 + y, 1.5-y, -0.5 + z; (iv) -0.5 + x, 0.5-y, -0.5 + z; (v) 1-x, -y, 2-z;

Hexaaquaaluminium(III) tris(methanesulfonate)
Crystal data 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 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.