This is an archive copy of the IUCr web site dating from 2008. For current content please visit https://www.iucr.org.

5. Discussion

The recommended nomenclature for each phase formed as the result of one or more structural phase transitions is the six-field notation described in §3. However, following the first use in a paper of the full six-field notation for a given phase, it is recommended that, if the phase is commonly associated with a label, then the first two fields only be used later in the paper to identify that phase. If the phase is not commonly associated with a label, then the second two fields should be used.

Let us now consider possible extensions of this nomenclature. The mutually contradictory nomenclature systems currently used in the literature do not take into account a number of frequently encountered complex situations such as the formation of incommensurate phases. The new nomenclature can be readily and simply extended to such phases. In the third field, which normally specifies the space group, the space group of the average structure can instead be indicated while in the fourth field, which normally contains the number of formula units in the cell, the incommensurate character can be indicated as well as the number of independent modulation wave vectors, the direction(s) of modulation and the approximate modulation period in the appropriate crystallographic direction. The intermediate phase of quartz can hence be denoted:

- $\vert$846-844 K $\vert$P622 (177) $\vert$incomm.; k = 1, a, 30a$\vert$- $\vert$-.

Within the framework of this supplementary convention, the succession of crystalline phases and an incommensurate phase in the case of potassium selenate (K₂SeO₄), for example, can be denoted as follows:

I $\vert \gt \,$630 K $\vert$P6₃/mmc (194) $\vert$Z = 1 $\vert$non-ferroic $\vert$-.

II $\vert$630-130 K $\vert$Pnam (62) $\vert$Z = 2 $\vert$ferroelastic $\vert$3 variants.

III $\vert$130-93 K $\vert$Pnam (62) $\vert$incomm.; k = 1, a, 3a $\vert$ferroelastic $\vert$3 variants.

IV $\vert$<$\,$93 K $\vert$Pna2₁ (33) $\vert$Z = 6 $\vert$ferroelastic, ferroelectric $\vert$3 ferroelastic variants; 6 ferroelectric variants.

Classifying certain long-period or polytype structures (such as SiC or ZnS) as incommensurate, however, is not widely accepted by the scientific community. A similar difficulty concerns the occurrence of quasicrystalline phases which, possibly, could be specified by their orientational symmetry (point group). Consequently, extension of the new nomenclature to situations such as these, as well as to magnetic phase transitions, has been postponed to a later Report.

Another type of question that is presently unresolved concerns the possible inclusion, in the proposed nomenclature, of physical classifications such as the `martensitic character' of a phase since these differ from the present ferroic classification. The difficulty in using both types of classification lies in the fact that, while the two concepts definitely overlap (martensitic and ferroelastic twinning is similar), their connection has not yet been fully clarified.

Note added in proof. Professor J. Kobayashi, Waseda University, has kindly pointed out that two new BaTiO₃ phases have been reported (Sawaguchi et al., 1985, 1990; Akishige et al., 1990) to form on cooling the 6H $\vert \gt \,$1733 K $\vert$phase. The first entry in §4.1 should hence be replaced by the following three entries:

6H $\vert \gt \,$1733 K $\vert$P6₃/mmc (194) $\vert$Z = 6 $\vert$non-ferroic $\vert$metastable to 222 K if cooled rapidly from the stable phase above 1733 K.

- $\vert$222-74 K $\vert$C222₁ (20)$\vert$ Z = 12 $\vert$non-ferroic $\vert$6H-related.

- $\vert$<$\,$74 K $\vert$P2₁ (4) $\vert$Z = 6 $\vert$ferroelectric, ferroelastic $\vert$6H-related.

It is a pleasure to thank Professor P. Tolédano, first Chairman of the Working Group for his guidance, and Professor G. Chapuis for serving as advisor to the Working Group.

Copyright © 1998 International Union of Crystallography