Labile degree of disorder in bismuth-oxophosphate compounds: illustration through three new structural types.

Here, we analyze the crystal structures of three new Bi/M oxophosphates, focusing on the ambiguity between order and disorder in different structural subunits. The three structures are original but systematically built on the assembly of O(Bi,M)4 tetrahedra into various 1D-oxocenterd units, separated by PO4 groups that create cationic channels. Two main subunits show versatile degrees of disorder, i.e., the cationic channels and some of the terminal O(Bi,M)4 entities. (a) In the compound [Bi2(Bi1.56K0.44)(dis)O3]K0.88(dis)(PO4)2, the K/K and K/Bi disorder is total on both nano- and micro-sized domains. (b) In the incommensurately modulated [Bi10(Bi∼0.5Cd∼0.5)8(dis)O16](Bi0.6Cd0.8)2(ord)(PO4)8, only the cationic channels show an ordered Bi/Cd arrangement which can be modified by minor stoichiometric changes between domains. (c) In [Bi18Zn10O21](ord)Zn5(ord)(PO4)14, both subunits are almost perfectly ordered (complex Bi/Zn sequence) into a 7-fold supercell, but this order strongly depends on the observation scale and is mainly lost in micronic-grains also due to slight compositional changes. However, the refined noncentrosymmetric organization is maintained (SHG tests) in the bulk. The relative stability of ordered versus disordered sites is discussed on the basis of the existence of two possible mixed sites and probably depends on the M chemical nature. Disorder was characterized by use of solid-state (31)P NMR probing for the first two cases. Finally, the observed disordered or long periodicities along the infinite dimension suggest the sketch of a periodic/rigid skeleton of O(Bi,M)4 units with counterions filling the interspace in more or less disordered arrangements.

Inorganic polar blocks into controlled acentric assemblies.

We show here a strategy to predict the crystal structure, formulate, and prepare new noncentrosymmetric (NCS) bismuth-phosphate based compounds. It is based on the cooperative-arrangement of polar building units (BUs) which can be created at particular stoichiometric conditions. The arrangement of such BUs into NCS compounds arise from the shortest-periodicity of repartition of the cationic charges in NCS structures than in the plausible, but never observed centrosymetric polytypes. This work validates the possibilities for the prediction of an extended series of novel compounds, tuning the size of BUs within a variety of controlled edifices. Despite their closed chemical composition, all the predicted terms appeared strikingly stable at precise stoichiometries.