Applications of the lattice-complex concept

Fischer, W.; Koch, E.

doi:10.1107/97809553602060000532

International
Tables for
Crystallography
Volume A
Space-group symmetry
Edited by Th. Hahn

pdf | chapter contents | chapter index | related articles

International Tables for Crystallography (2006). Vol. A. ch. 14.3, pp. 873-876
https://doi.org/10.1107/97809553602060000532

Chapter 14.3. Applications of the lattice-complex concept

W. Fischer^a and E. Koch^a ^*

^a Institut für Mineralogie, Petrologie und Kristallographie, Philipps-Universität, D-35032 Marburg, Germany
Correspondence e-mail: kochelke@mailer.uni-marburg.de

Chapter 14.3 gives a short introduction to some applications of lattice complexes: (i) The knowledge of the assignment of the Wyckoff positions to lattice complexes considerably facilitates the study of geometrical properties of point configurations. (ii) Relations between crystal structures with different symmetries are often discernible because the corresponding Wyckoff positions either belong to the same lattice complex or because a limiting-complex relationship exists. (iii) Wyckoff positions belonging to the same lattice complex show analogous reflection conditions. (iv) If a phase transition of a crystal is connected with a group–subgroup degradation, comparison of the lattice complexes corresponding to the Wyckoff positions of the original space group on the one hand and of its various subgroups on the other hand very often shows which of these subgroups are suitable for the low-symmetry modification. (v) Many incorrect space-group assignments to crystal structures could be avoided by simply looking at the lattice complexes (and their descriptive symbols) that correspond to the Wyckoff positions occupied by the different kinds of atoms.

Keywords: lattice complexes; point configurations; reflection conditions; phase transitions; relations between crystal structures; space-group assignment.

14.3.1. Geometrical properties of point configurations

| top | pdf |

To study the geometrical properties of all point configurations in three-dimensional space, it is not necessary to consider all Wyckoff positions of the space groups or all 1128 types of Wyckoff set . Instead, one may restrict the investigations to the characteristic Wyckoff positions of the 402 lattice complexes. The results can then be transferred to all noncharacteristic Wyckoff positions of the lattice complexes, as listed in Tables 14.2.3.1 and 14.2.3.2 .

The determination of all types of sphere packings with cubic or tetragonal symmetry forms an example for this kind of procedure (Fischer, 1973 , 1974 , 1991a ,b , 1993). The cubic lattice complex I4xxx, for example, allows two types of sphere packings within its characteristic Wyckoff position $[I\bar{4}3m]$ 8c xxx .3m, namely [9/3/c2] for [x = 3/16] and [6/4/c1] for $[{3/16} \lt x \lt {1 \over 4}]$ (cf. Fischer, 1973). Ag₃PO₄ crystallizes with symmetry $[P\bar{4}3n]$ (Deschizeaux-Cheruy et al., 1982 ) and the oxygen atoms occupy Wyckoff position 8e xxx .3., which also belongs to I4xxx. Comparison of the coordinate parameter [x = 0.1491] for the oxygen atoms with the sphere-packing parameters listed for $[I\bar{4}3]$ m c shows directly that the oxygen arrangement in this crystal structure does not form a sphere packing.

Other examples for this approach are the derivation of crystal potentials (Naor, 1958 ), of coordinate restrictions in crystal structures (Smirnova, 1962 ), of Patterson diagrams (Koch & Hellner, 1971 ), of Dirichlet domains (Koch, 1973 , 1984 ) and of sphere packings for subperiodic groups (Koch & Fischer, 1978 ).

The 30 lattice complexes in two-dimensional space correspond uniquely to the `henomeric types of dot pattern' introduced by Grünbaum and Shephard (cf. e.g. Grünbaum & Shephard, 1981 ; Grünbaum, 1983 ).

14.3.2. Relations between crystal structures

| top | pdf |

Frequently, different crystal structures show the same geometrical arrangement for some of their atoms, even though their space groups do not belong to the same type. In these cases, the corresponding Wyckoff positions either belong to the same lattice complex or there exist close relationships between them, e.g. limiting-complex relations.

Examples

(1) The Fe atoms in pyrite FeS₂ occupy Wyckoff position $[4a\ 000\ .\bar{3}]$ . of $[Pa\bar{3}]$ (descriptive symbol F) that belongs to the invariant lattice complex $[Fm\bar{3}m\ a]$ . Accordingly, the Fe atoms in pyrite form a face-centred cubic lattice as do the Cu atoms in the element structure of copper.
(2) Cuprite Cu₂O crystallizes with symmetry $[Pn\bar{3}m]$ . The oxygen atoms occupy Wyckoff position $[2a\ 000\ \bar{4}3m]$ (descriptive symbol I) and the copper atoms position $[4b\ {1 \over 4} {1 \over 4} {1 \over 4}\ .\bar{3}m]$ (descriptive symbol $[{1 \over 4} {1 \over 4} {1 \over 4}\; F]$ ). Position 2a belongs to lattice complex $[Im\bar{3}m\ a]$ and position 4b to $[Fm\bar{3}m\ a]$ . Therefore, the O atoms form a body-centred cubic lattice like the W atoms in the structure of tungsten, and the copper atoms form a face-centred cubic lattice. The tungsten configuration is shifted by $[({1 \over 4} {1 \over 4} {1 \over 4})]$ with respect to the copper configuration.
(3) K₂NaAlF₆ (elpasolite, cf. Morss, 1974 ) and K₂PbNi(NO₂)₆ (cf. Takagi et al., 1975) crystallize with symmetry $[Fm\bar{3}m]$ and $[Fm\bar{3}]$ , respectively. $[\displaylines{&\hbox{K}_{2}\hbox{NaAlF}_{6}\hfill\cr &\quad\! \matrix{\hbox{Al}\hfill &4a \hfill &m\bar{3}m \hfill &000 \hfill &F\hfill\cr \hbox{Na} \hfill &4b \hfill &m\bar{3}m \hfill &{1 \over 2}{1 \over 2}{1 \over 2} \hfill &{1 \over 2}{1 \over 2}{1 \over 2} F\cr \hbox{K} \hfill &8c \hfill &\bar{4}3m \hfill &{1 \over 4}{1 \over 4}{1 \over 4} \hfill &{1 \over 4}{1 \over 4}{1 \over 4} P_{2}\cr \hbox{F} \hfill &24e \hfill &4m.m \hfill &x00 \hfill &F6z\hfill\cr & & &x = 0.219\hfill \cr}}]$ $[\displaylines{&\hbox{K}_{2}\hbox{PbNi}(\hbox{NO}_{2})_{6}\hfill\cr &\quad\! \matrix{\hbox{Ni} \hfill &4a \hfill &m\bar{3}. \hfill &000 \hfill &F\hfill\cr \hbox{Pb} \hfill &4b \hfill &m\bar{3}. \hfill &{1 \over 2}{1 \over 2}{1 \over 2} \hfill &{1 \over 2}{1 \over 2}{1 \over 2} F\hfill\cr \hbox{K} \hfill &8c \hfill &23. \hfill &{1 \over 4}{1 \over 4}{1 \over 4} \hfill &{1 \over 4}{1 \over 4}{1 \over 4} P_{2} \hfill\cr \hbox{N} \hfill &24e \hfill &mm2.. \hfill &x00 \hfill &F6z\hfill\cr & & &x = 0.1966\cr \hbox{O} \hfill &48h \hfill &m.. \hfill &0yz \hfill &F6z2x \hfill \cr}}]$ As the descriptive lattice-complex symbols for the various atomic positions show immediately, the two crystal structures are very similar. The only difference originates from the replacement of the fluorine atoms in elpasolite by NO₂ groups in K₂PbNi(NO₂)₆, which causes the symmetry reduction from $[Fm\bar{3}m]$ to $[Fm\bar{3}]$ .
(4) The crystal structure of CoU (Baenziger et al., 1950 ) may be interpreted as a slightly distorted CsCl (or β-brass, CuZn)-type structure. CsCl corresponds to Wyckoff positions 1a and 1b of $[Pm\bar{3}m]$ with descriptive symbol P and $[{1 \over 2}{1 \over 2}{1 \over 2}\ P]$ , respectively; Co and U both occupy Wyckoff position 8a .3. xxx of $[I2_{1}3]$ with for U and for Co. As the descriptive symbol $[2_{1}2_{1}..\ P_{2}Y^{*}1xxx]$ shows, this Wyckoff position belongs to a Weissenberg complex with two invariant limiting complexes, namely P ( $[Pm\bar{3}m\ a]$ ) and $[Y^{*}\ (I4_{1}32\ a)]$ . corresponds to $[P_{2}]$ , $[x = {1 \over 4}]$ to $[{1 \over 4}{1 \over 4}{1 \over 4}\ P_{2}]$ , $[x = {1 \over 8}]$ to $[^{+}Y^{*}]$ and $[x = {7 \over 8}]$ to $[^{-}Y^{*}]$ . Consequently, the uranium and cobalt atoms form approximately a $[P_{2}]$ and a $[{1 \over 4}{1 \over 4}{1 \over 4}\ P_{2}]$ configuration, respectively.

Publications by Hellner (1965, 1976a ,b ,c , 1977 , 1979 ), Loeb (1970), Smirnova & Vasserman (1971), Sakamoto & Takahasi (1971), Niggli (1971), Fischer & Koch (1974), Hellner et al. (1981) and Hellner & Sowa (1985) refer to this aspect.

14.3.3. Reflection conditions

| top | pdf |

Wyckoff positions belonging to the same lattice complex show analogous reflection conditions. Therefore, lattice complexes have also been used to check the reflection conditions for all Wyckoff positions in the space-group tables of this volume.

The descriptive symbols may supply information on the reflection conditions. If the symbol does not contain any distribution-symmetry part, the reflection conditions of the Wyckoff position are indicated by the symbol of the invariant lattice complex in the central part (e.g. $[P4/nmm\ g]$ : C4xx shows that the reflection condition is that of a C lattice, $[hkl\!:\! h+k = 2n]$ ). In the case that the site set consists of only one point, i.e. the Wyckoff position belongs to a Weissenberg complex, all conditions for general reflections hkl that may arise from special choices of the coordinates can be read from the central part of the symbol (e.g. $[P4/nmm\ c\hbox{:}\ 0{1 \over 2}0\; .. 2\ CI1z]$ indicates that, by special choice of z, either $[hkl\hbox{:}\ h+k = 2n]$ or $[hkl\hbox{:}\ h+k+l = 2n]$ may be produced).

14.3.4. Phase transitions

| top | pdf |

If a crystal undergoes a phase transition from a high- to a low-symmetry modification, the transition may be connected with a group–subgroup degradation. In such a case, the comparison of the lattice complexes corresponding to the Wyckoff positions of the original space group on the one hand and of its various subgroups on the other hand very often shows which of these subgroups are suitable for the low-symmetry modification.

This kind of procedure will be demonstrated with the aid of a space group $[R\bar{3}m]$ and its three translation-equivalent subgroups with index 2, namely R32, $[R\bar{3}]$ and R3m. In the course of the subgroup degradation, the Wyckoff positions of $[R\bar{3}m]$ behave differently:

The descriptive symbols R and $[00{1 \over 2}\ R]$ refer to Wyckoff positions $[R\bar{3}m\ 3a]$ and 3b as well as to Wyckoff positions R32 3a and 3b and $[R\bar{3}\ 3a]$ and 3b. Therefore, all corresponding point configurations and atomic arrangements remain unchanged in these subgroups. In subgroup R3m, however, the respective Wyckoff position is 3a with descriptive symbol R[z], i.e. a shift parallel to [001] of the entire point configuration is allowed.

The descriptive symbol R2z for $[R\bar{3}m\ 6c]$ occurs also for R32 6c and $[R\bar{3}\ 6c]$ . Again both subgroups do not allow any deformations of the corresponding point configurations or atomic arrangements. Symmetry reduction to R3m, however, yields a splitting of each R2z configuration into two R[z] configurations. The two z parameters may be chosen independently.

As M and $[00{1 \over 2}\ M]$ are the descriptive symbols not only of $[R\bar{3}m\ 9e]$ and 9d but also of $[R\bar{3}\ 9e]$ and 9d, $[R\bar{3}]$ does not enable any deformation of the corresponding atomic arrangements. In R32 and in R3m, however, the respective point configurations may be deformed differently, as the descriptive symbols show: R3x and $[00{1 \over 2}\; R3x]$ (R32 9d and 9e), $[R3x\bar{x}\hbox{[}z\hbox{]}]$ (R3m 9b).

Wyckoff positions $[R\bar{3}m\ 18f]$ and 18g (R6x and $[00{1 \over 2}\ R6x]$ ) correspond to R32 9d and 9e (R3x and $[00{1 \over 2}\ R3x]$ ), to $[R\bar{3}\ 18f\ (R6xyz)]$ , and to R3m 18c $[(R3x\bar{x}2y\hbox{[}z\hbox{]})]$ . In R32, the hexagons 6x around the points of the R lattice are split into two oppositely oriented triangles 3x, which may have different size. In $[R\bar{3}]$ and in R3m, the hexagons may be deformed differently.

Wyckoff position $[R\bar{3}m\ 18h\ (R6x\bar{x}z)]$ corresponds to sets of trigonal antiprisms around the points of an R lattice. These antiprisms may be distorted in R32 18f (R3x2yz) or rotated in $[R\bar{3}]$ 18f (R6xyz). In R3m 9b $[(R3x\bar{x}\hbox{[}z\hbox{]})]$ , each antiprism is split into two parallel triangles that may differ in size.

In each of the three subgroups, any point configuration belonging to the general position $[R\bar{3}m]$ 36i splits into two parts. Each of these parts may be deformed differently.

14.3.5. Incorrect space-group assignment

| top | pdf |

In the literature, some crystal structures are still described within space groups that are only subgroups of the correct symmetry groups. Many such mistakes (but not all of them) could be avoided by simply looking at the lattice complexes (and their descriptive symbols) that correspond to the Wyckoff positions of the different kinds of atoms. Whenever the same (or an analogous) lattice-complex description of a crystal structure is also possible within a supergroup, then the crystal structure has at least that symmetry.

Examples

(1) The crystal structure of β-LiRhO₂ has been refined in space group $[F4_{1}32]$ (cf. Hobbie & Hoppe, 1986 ). $[\openup3pt\matrix{\hbox{Rh} \hfill &16c \hfill &.32 \hfill &{1 \over 8}{1 \over 8}{1 \over 8} \hfill &T\hfill \cr \hbox{Li} \hfill &16d \hfill &.32 \hfill &{5 \over 8}{5 \over 8}{5 \over 8} \hfill &{1 \over 2}{1 \over 2}{1 \over 2}\ T\hfill\cr \hbox{O} \hfill &32e \hfill &.3. \hfill &xxx \hfill &..2\; D4xxx \hfill\cr}]$ The same atomic arrangement is possible in the supergroup $[Fd\bar{3}m]$ of $[F4_{1}32]$ , as can easily be read from Table 14.2.3.2 : $[\openup3pt\matrix{\hbox{Rh} \hfill &16c \hfill &.\bar{3}m \hfill &{1 \over 8}{1 \over 8}{1 \over 8} \hfill &T\hfill\cr \hbox{Li} \hfill &16d \hfill &.\bar{3}m \hfill &{5 \over 8}{5 \over 8}{5 \over 8} \hfill &{1 \over 2}{1 \over 2}{1 \over 2}\ T\hfill\cr \hbox{O} \hfill &32e \hfill &.3m \hfill &xxx \hfill &..2\; D4xxx\hfill\cr}]$ Therefore, β-LiRhO₂ should be described in $[Fd\bar{3}m]$ .
(2) KIAs₄O₆ (Pertlik, 1988 ) has been described with symmetry P622. $[\openup3pt\matrix{\hbox{I} \hfill &1a \hfill &622 \hfill &000 \hfill &P\hfill\cr \hbox{K} \hfill &1b \hfill &622 \hfill &00{1 \over 2} \hfill &00{1 \over 2}\ P\hfill \cr \hbox{As} \hfill &4h \hfill &3.. \hfill &{1 \over 3}{2 \over 3}z \hfill &G2z\hfill \cr \hbox{O} \hfill &6i \hfill &2.. \hfill &{1 \over 2}0z \hfill &N2z\hfill \cr}]$ Space group allows the same atomic arrangement: $[\openup3pt\matrix{\hbox{I} \hfill &1a \hfill &6/mmm \hfill &000 \hfill &P\hfill\cr \hbox{K} \hfill &1b \hfill &6/mmm \hfill &00{1 \over 2} \hfill &00{1 \over 2}\ P\hfill\cr \hbox{As} \hfill &4h \hfill &3m. \hfill &{1 \over 3}{2 \over 3}z \hfill &G2z\hfill\cr \hbox{O} \hfill &6i \hfill &2mm \hfill &{1 \over 2}0z \hfill &N2z\hfill \cr}]$ Therefore, KIAs₄O₆ should be described in .

14.3.6. Application of descriptive lattice-complex symbols

| top | pdf |

Descriptive symbols of lattice complexes – at least those of the invariant lattice complexes – have been used for the description of crystal structures (cf. Section 14.3.2 and the literature cited there), for the nomenclature of three-periodic surfaces (von Schnering & Nesper, 1987 ) and in connection with orbifolds of space groups (Johnson et al., 2001 ).

References

Baenziger, N. C., Rundle, R. E., Snow, A. T. & Wilson, A. S. (1950). Compounds of uranium with transition metals of the first long period. Acta Cryst. 3, 34–40.Google Scholar

Deschizeaux-Cheruy, M. N., Aubert, J. J., Joubert, J. C., Capponi, J. J. & Vincent, H. (1982). Relation entre structure et conductivité ionique basse temperature de Ag₃PO₄. Solid State Ionics, 7, 171–176.Google Scholar

Fischer, W. (1973). Existenzbedingungen homogener Kugelpackungen zu kubischen Gitterkomplexen mit weniger als drei Freiheitsgraden. Z. Kristallogr. 138, 129–146.Google Scholar

Fischer, W. (1974). Existenzbedingungen homogener Kugelpackungen zu kubischen Gitterkomplexen mit drei Freiheitsgraden. Z. Kristallogr. 140, 50–74.Google Scholar

Fischer, W. (1991a). Tetragonal sphere packings. I. Lattice complexes with zero or one degree of freedom. Z. Kristallogr. 194, 67–85.Google Scholar

Fischer, W. (1991b). Tetragonal sphere packings. II. Lattice complexes with two degrees of freedom. Z. Kristallogr. 194, 87–110.Google Scholar

Fischer, W. (1993). Tetragonal sphere packings. III. Lattice complexes with three degrees of freedom. Z. Kristallogr. 205, 9–26.Google Scholar

Fischer, W. & Koch, E. (1974). Kubische Strukturtypen mit festen Koordinaten. Z. Kristallogr. 140, 324–330.Google Scholar

Grünbaum, B. (1983). Tilings, patterns, fabrics and related topics in discrete geometry. Jber. Dtsch. Math.-Verein. 85, 1–32.Google Scholar

Grünbaum, B. & Shephard, G. C. (1981). A hierarchy of classification methods for patterns. Z. Kristallogr. 154, 163–187.Google Scholar

Hellner, E. (1965). Descriptive symbols for crystal-structure types and homeotypes based on lattice complexes. Acta Cryst. 19, 703–712.Google Scholar

Hellner, E. (1976a). Verwandtschaftskriterien von Kristallstrukturtypen. I. Z. Anorg. Allg. Chem. 421, 37–40.Google Scholar

Hellner, E. (1976b). Verwandtschaftskriterien von Kristallstrukturtypen. II. Die Einführung der Gitterkomplexe P, J und F. Z. Anorg. Allg. Chem. 421, 41–48.Google Scholar

Hellner, E. (1976c). Verwandtschaftskriterien von Kristallstrukturtypen. III. Die kubischen Überstrukturen des ReO₃-, Perowskit- und CaF₂-Typs. Z. Anorg. Allg. Chem. 421, 49–60.Google Scholar

Hellner, E. (1977). Verwandtschaftskriterien von Kristallstrukturtypen. IV. Ableitung von Strukturtypen der I-, P- und F-Familien. Z. Anorg. Allg. Chem. 437, 60–72.Google Scholar

Hellner, E. (1979). The frameworks (Bauverbände) of the cubic structure types. Struct. Bonding (Berlin), 37, 61–140.Google Scholar

Hellner, E., Koch, E. & Reinhardt, A. (1981). The homogeneous frameworks of the cubic crystal structures. Phys. Daten-Phys. Data, 16–2, 1–67.Google Scholar

Hellner, E. & Sowa, H. (1985). The cubic structure types described in their space groups with the aid of frameworks. Phys. Daten-Phys. Data, 16–3, 1–141.Google Scholar

Hobbie, K. & Hoppe, R. (1986). Über Oxorhodate der Alkalimetalle: β-LiRhO₂. Z. Anorg. Allg. Chem. 535, 20–30.Google Scholar

Johnson, C. K., Burnett, M. N. & Dunbar, W. D. (2001). Crystallographic topology and its applications. Crystallographic Computing 7. Proceedings from the Macromolecular Crystallography Computing School, edited by P. E. Bourne & K. Watenpaugh. IUCr/Oxford University Press. In the press.Google Scholar

Koch, E. (1973). Wirkungsbereichspolyeder und Wirkungsbereichsteilungen zu kubischen Gitterkomplexen mit weniger als drei Freiheitsgraden. Z. Kristallogr. 138, 196–215.Google Scholar

Koch, E. (1984). A geometrical classification of cubic point configurations. Z. Kristallogr. 166, 23–52.Google Scholar

Koch, E. & Fischer, W. (1978). Types of sphere packings for crystallographic point groups, rod groups and layer groups. Z. Kristallogr. 148, 107–152.Google Scholar

Koch, E. & Hellner, E. (1971). Die Pattersonkomplexe der Gitterkomplexe. Z. Kristallogr. 133, 242–259.Google Scholar

Loeb, A. L. (1970). A systematic survey of cubic crystal structures. J. Solid State Chem. 1, 237–267.Google Scholar

Morss, L. R. (1974). Crystal structure of dipotassium sodium fluoroaluminate (elpasolite). J. Inorg. Nucl. Chem. 36, 3876–3878.Google Scholar

Naor, P. (1958). Linear dependence of lattice sums. Z. Kristallogr. 110, 112–126.Google Scholar

Niggli, A. (1971). Parameterfreie kubische Strukturtypen. Z. Kristallogr. 133, 473–490.Google Scholar

Pertlik, F. (1988). The compounds KAs₄O₆X (X = Cl, Br, I) and NH₄As₄O₆X (X = Br, I): Hydrothermal syntheses and structure determinations. Monatsh. Chem. Verw. Teile Anderer Wiss. 119, 451–456.Google Scholar

Sakamoto, Y. & Takahasi, U. (1971). Invariant and quasi-invariant lattice complexes. J. Sci. Hiroshima Univ. Ser. A, 35, 1–51.Google Scholar

Schnering, H. G. von & Nesper, R. (1987). How nature adapts chemical structures to curved surfaces. Angew. Chem. Int. Ed. Engl. 26, 1059–1080.Google Scholar

Smirnova, N. L. (1962). Possible values of the x coordinates in single-parameter lattice complexes of the cubic system. Sov. Phys. Crystallogr. 7, 5–8.Google Scholar

Smirnova, N. L. & Vasserman, E. I. (1971). The line diagrams of crystalline substances. Structural types of the cubic system from invariant lattice complexes. Sov. Phys. Crystallogr. 15, 791–794.Google Scholar

Takagi, S., Joesten, M. D. & Lenhert, P. G. (1975). Potassium lead hexanitronickelate(II). Acta Cryst. B31, 1968–1970.Google Scholar

International Tables for Crystallography (2006). Vol. A. ch. 14.3, pp. 873-876
https://doi.org/10.1107/97809553602060000532