Representations of the symmetry groups of infinite crystals
Bachir Bekka, Christian Brouder · 2025
Read Representations of the symmetry groups of infinite crystals by Bachir Bekka, Christian Brouder (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-22265.txt.
Introduction
Bachir Bekka and Christian Brouder examine symmetry in infinite crystals. In that setting the usual finite-crystal shortcuts break down. Textbook treatments use a finite crystal with periodic boundary conditions, making the Brillouin zone a finite point set. Covering continuous crystal momentum, and whether band energy levels vary smoothly, requires truly infinite crystals whose symmetry groups are infinite and non-compact.
For such groups, character theory cannot be used. The authors instead build the tools from harmonic analysis. It includes the group algebra, the regular representation, and induction of irreducible forms from projective forms of point groups. They cover magnetic and non-magnetic groups in any dimension. Later sections treat restriction to subgroups and tensor products of induced forms.
Catalog record
| CREATOR | Bachir Bekka, Christian Brouder |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 122 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-22265
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Representations of the symmetry groups of infinite crystals Preprint: AIP/123-QED Bachir Bekka, Christian Brouder Abstract We investigate the representations of the symmetry groups of infinite crystals. Crystal symmetries are usually described as the finite symmetry group of a finite crystal with periodic boundary conditions, for which the Brillouin zone is a finite set of points. However, to deal with the continuous crystal momentum 𝐤 {\mathbf{k}} required to discuss the continuity, singularity or analyticity of band energies ϵ n ( 𝐤 ) \epsilon_{n}({\mathbf{k}}) and Bloch states ψ 𝐤 \psi_{\mathbf{k}} , we need to consider infinite crystals. The symmetry groups of infinite crystals belong to the category of infinite non-compact groups, for which many standard tools of group theory break down. For example, character theory is no longer available for these groups and we use harmonic analysis to build the group algebra, the regular representation, the induction of irreducible representations of the crystallographic group from projective representations of the point groups and the decomposition of a representation into its irreducible parts. We deal with magnetic and non-magnetic groups in arbitrary dimensions. In the last part of the paper, we discuss Mackey’s restriction of an induced representation to a subgroup, the tensor product of induced representations and the symmetric and antisymmetric squares of induced representations. I Introduction Space groups and their representations are essential tools for the theory of electrons, phonons and magnons in solids, as well as for the description of the crystallographic and magnetic structures of materials. Space groups are thoroughly studied and widely used in condensed matter calculations, many textbooks are devoted to them 84 ; 74 ; 27 ; 28 ; 119 ; 113 ; 13 ; 87 ; 107 ; 68 ; 80 ; 47 ; 20 ; 105 ; 127 , computer packages and web sites (e.g. www.cryst.ehu.es ) describe their properties. Still, space groups face a fundamental open problem that we want to solve in the present paper. The properties of a crystalline sample, which is a finite object, are most often investigated by representing it either as an infinite crystal or as a finite crystal with Born-von Kàrmàn boundary conditions 17 linking opposite sides of the crystal (that we call a toric crystal because it is topologically equivalent to a torus). In solid-state physics, both infinite and toric crystals are used, in spite of the differences between the two models 51 . On the one hand, an infinite crystal has to be used when the wavevector {\mathbf{k}} varies continuously, for example to study the continuity of an energy band E_{n}({\mathbf{k}}) with respect to {\mathbf{k}} or to know whether the Wannier functions can be exponentially localized 25 . On the other hand, a (finite) toric crystal is used to investigate the symmetry properties of a crystal, because the symmetry group of a toric crystal is finite and powerful techniques are available to work with finite groups. However, the Brillouin zone of a toric crystal is a finite set of points and no continuity or analyticity questions can be addressed. As a concrete consequence, Bloch functions computed at a finite number of points of the Brillouin zone pick up random phases from one point to the other and making these phases “consistent” is an additional work that has not to be done in an infinite crystal. Moreover, it is necessary to address the problem of the influence of the size of the torus (i.e. the number of unit cells in each direction for which the toric crystal is periodic) and to prove the validity of the thermodynamic limit (i.e. the limit where the size of the torus tends to infinity) for the observables of interest. In many papers, even by the most distinguised authors, both points of views are intermingled without proper justification. For instance the landmark paper of solid-state physics by Bouckaert, Smoluchowski and Wigner 18 uses both results on irreducible representations of space groups derived for a finite toric crystal 121 and a continuity argument requiring an infinite crystal. The present paper aims at providing tools and proofs to deal consistently with crystal symmetries and continuous wavevectors, i.e. to work with infinite crystals. The main difficulty is that many problems break loose when we go from finite to infinite groups: an infinite group can have infinite-dimensional irreducible representations (as for the familiar Lorentz group), its collection of irreducible representations can be so wild that it is not classifiable modulo equivalence ( 10 , Thm. 8.F.3) , the regular representation might not be a (generalized) sum of irreducible representations in a unique way ( 10 , Thm. 1.G.11) and a finite-dimensional representation is is not necessarily equivalent to a unitary representation 120 , to name just a few of them. A further difference specific to the case of crystallographic groups has to be considered: if {\mathbf{k}} is a vector of the Brillouin zone and {\mathbf{t}}_{\mathbf{m}} a lattice vector, we can define f(n)=e^{in{\mathbf{k}}\cdot{\mathbf{t}}_{\mathbf{m}}} for any (signed) integer n . We can see that f(n)=e^{2\pi i\alpha n} , where \alpha is a rational number when the crystallographic group is finite while \alpha can be an irrational number when the group is infinite. In the first case f(n) is periodic, it takes a finite number of values and each value is reached an infinite number of times. In the second case f(n) takes an infinite number of different values and each value is reached only once. This ergodic behavior considerably complicates the discussion of the characters of irreducible representations, which become almost-period functions in the sense of von Neumann 131 . As a consequence of these differences between finite and infinite groups, it is not clear whether results derived for finite toric crystals are still valid for infinite crystals. As we shall see, the main technical difference between finite and infinite groups is the necessity to use tools from analysis (topology, integrals, distributions, etc.) instead of finite sums. This beautiful mixture of group theory and functional analysis is called harmonic analysis (see Ref. 102 for a history of the subject). We not only want to provide a mathematically correct description of the irreducible representations of infinite crystallographic groups in arbitrary dimension, but we also wish to provide safe tools that solid-state physicists can use to investigate infinite crystals. We deal with crystallographic groups in arbitrary dimension because crystallographic groups of dimensions higher than 3 (i.e. superspace crystallography) are used to describe quasicrystals, incommensurately modulated crystals 73 as well as incommensurate magnetic structures 116 . The extension of results on 3-dimensional space groups to arbitrary dimensions is not always trivial because some papers use the fact that space groups are solvable 121 and this is no longer true for dimensions n\geq 4 . Crystallographic groups were thoroughly studied by mathematicians because they are related to Hilbert’s 18th problem and the solution of any Hilbert’s problem brings eternal fame to its solver. However, irreducible representations of crystallographic groups are not well known in the mathematical literature and we have recourse to the more general (but highly complex) theory of locally compact groups, for which approachable textbooks are now available 78 ; 10 ; 58 ; 59 . Still, the gap between the terminology used in physics and that used in