Is the ³¹P chemical shift anisotropy of aluminophosphates a useful parameter for NMR crystallography?

The ³¹P chemical shift anisotropy (CSA) offers a potential source of new information to help determine the structures of aluminophosphate framework materials (AlPOs). We investigate how to measure the CSAs, which are small (span of ~20-30 ppm) for AlPOs, demonstrating the need for CSA-amplification experiments (often in conjunction with ²⁷Al and/or ¹H decoupling) at high magnetic field (20.0 T) to obtain accurate values. We show that the most shielded component of the chemical shift tensor, δ₃₃, is related to the length of the shortest P-O bond, whereas the more deshielded components, δ₁₁ and δ₂₂ can be related more readily to the mean P-O bond lengths and P-O-Al angles. Using the case of Mg-doped STA-2 as an example, the CSA is shown to be much larger for P(OAl)_4–n(OMg)_n environments, primarily owing to a much shorter P-O(Mg) bond affecting δ₃₃, however, since the mean P-O bond lengths and P-O-T (T = Al, Mg) bond angles do not change significantly between P(OAl)₄ and P(OAl)_4–n(OMg)_n sites, the isotropic chemical shifts for these species are similar, leading to overlapped spectral lines. When the CSA information is included, spectral assignment becomes unambiguous, therefore, while the specialist conditions required might preclude the routine measurement of ³¹P CSAs in AlPOs, in some cases (particularly doped materials), the experiments can still provide valuable additional information for spectral assignment.