Chemistry-structure-simulation or chemistry-simulation-structure sequences?: The case of MIL-34, a new porous aluminophosphate

A new aluminophosphate, MIL-34, is investigated from its as-synthesized structure to its calcined microporous form. Single-crystal X-ray diffraction measurements on the as-synthesized MIL-34 (Al-4(PO4)(4)OH .C4H10N, space group P-1, a = 8.701(3) Angstrom, b = 9.210(3) Angstrom, c = 12.385(3) Angstrom, alpha = 111. 11(2)degrees, beta = 101.42(2)degrees, gamma = 102.08(2)degrees, V = 863.8(4) Angstrom (3), Z = 2, R = 3.8%) reveal a 3-D open framework where Al atoms are in both tetrahedral and trigonal bipyramidal coordinations. It contains a 2-D pore system defined by eight rings where channels along [100] cross channels running along [010] and [110]. CBuA molecules are trapped at their intersection. Al-27, P-31, and H-1 MAS NMR spectroscopies corroborate these structural features. Calcination treatments of a powder sample of the as-synthesized MIL-34 indicate its transformation into the related template-free structure that is stable up to 1000 degreesC. Lattice energy minimizations are then used in order to anticipate the crystal structure of the calcined MIL-34, starting with the knowledge of the as-synthesized structure exclusively. Energy minimizations predict a new regular zeotype structure (AlPO4, space group P-1, a = 8.706 Angstrom, b = 8.749 Angstrom, c = 12.768 Angstrom, alpha = 111. 17 degrees, beta = 97.70 degrees, gamma = 105.14 degrees, V = 846.75 Angstrom (3), Z = 2) together with a thermodynamic stability similar to that of existing zeotype AlPOs. Excellent agreement is observed between the diffraction pattern calculated from the predicted calcined MIL-34 and the experimental X-ray powder diffraction pattern of the calcined sample. Finally, the atomic coordinates and cell parameters of the calcined MIL-34 predicted from the simulations are used to perform the Rietveld refinement of the calcined sample powder pattern, further corroborated by Al-27 and P-31 NMR measurements. This unique combination of experiment and simulation approaches is an interesting and innovative strategy in materials sciences, where simulations articulate the prediction of a possible template-free framework from its as-synthesized templated form. This is especially valuable when straightforward characterizations of the solid of interest with conventional techniques are not easy to carry out.