A simulated annealing approach for solving zeolite crystal structures from two-dimensional NMR correlation spectra.

An improved NMR crystallography strategy is presented for determining the structures of network materials such as zeolites from just a single two-dimensional (2D) NMR correlation spectrum that probes nearest-neighbor interactions, combined with the unit cell parameters and space group information measured in a diffraction experiment. The correlation information contained within a 2D spectrum is converted into a "connectivity matrix" which is incorporated into a cost function whose minimum is searched for using a simulated annealing algorithm. The algorithm was extensively tested on over 150 zeolite frameworks from the International Zeolite Association database of zeolite structures and shown to be very robust and efficient in reconstructing the structures from connectivity information. The structure determination of the pure silica zeolites ITQ-4, Ferrierite, and Sigma-2 from experimental 2D (29)Si double-quantum NMR spectra and powder X-ray diffraction data using this improved approach is also presented.