1.5 Symmetry operations

Symmetry operations are transformations that leave a crystal structure looking identical to how it started. They form the foundation for classifying crystals and predicting their physical properties. Every restriction on a material's behavior, from which electronic transitions are allowed to whether it can be piezoelectric, traces back to its symmetry.

This topic connects symmetry types (translation, rotation, reflection, inversion) to the classification schemes you'll use throughout solid-state physics: Bravais lattices, point groups, and space groups. It also covers how symmetry constrains physical properties and what happens when symmetry breaks.

Types of symmetry operations

A symmetry operation is any transformation you can apply to a crystal that maps it onto itself. The four fundamental types are translation, rotation, reflection, and inversion. Combining these generates more complex symmetries and ultimately determines how we classify crystal structures.

Translation symmetry

Translation symmetry means the crystal looks the same after being shifted by a specific displacement vector. This is the defining feature of a crystal: atoms repeat periodically in space.

• The lattice vectors $\mathbf{a}_1$, $\mathbf{a}_2$, $\mathbf{a}_3$ define the repeating unit cell. Any translation $\mathbf{T} = n_1\mathbf{a}_1 + n_2\mathbf{a}_2 + n_3\mathbf{a}_3$ (where $n_1, n_2, n_3$ are integers) maps the lattice onto itself.
• This periodicity is what makes Bloch's theorem possible: electronic wave functions in a periodic potential can be written as a plane wave modulated by a function with the same periodicity as the lattice.
• Translation symmetry is unique among symmetry operations because it involves no fixed point. Every other symmetry operation (rotation, reflection, inversion) leaves at least one point unmoved.

Rotation symmetry

A crystal has rotation symmetry when it looks identical after rotation by a specific angle about an axis. Only certain rotation angles are compatible with translational periodicity.

• Crystals are restricted to $n$-fold rotational axes where $n = 1, 2, 3, 4,$ or $6$. A 5-fold or 8-fold rotation axis, for instance, cannot tile space periodically, so these are forbidden in conventional crystals.
• An $n$-fold axis means the minimum rotation angle that maps the crystal onto itself is $\frac{360°}{n}$.
• Proper rotations involve rotation only. Improper rotations (also called rotoinversions) combine a rotation with an inversion through a point on the axis.
• A square lattice has 4-fold rotational symmetry (invariant under 90° rotations), while a hexagonal lattice has 6-fold symmetry (invariant under 60° rotations).

Reflection symmetry

Reflection symmetry means the crystal is unchanged when reflected across a mirror plane.

• Mirror planes in crystals are labeled based on their orientation relative to rotation axes. A vertical mirror plane ($\sigma_v$) contains the principal rotation axis, while a horizontal mirror plane ($\sigma_h$) is perpendicular to it.
• A cubic crystal has nine mirror planes: three that bisect opposite faces (perpendicular to each cube axis) and six diagonal planes that pass through opposite edges.
• Reflection changes the handedness of a coordinate system. If you reflect a right-handed structure, you get a left-handed one. When the crystal is invariant under this, the two handednesses are equivalent.

Inversion symmetry

Inversion maps every point $(x, y, z)$ to $(-x, -y, -z)$ through a center of symmetry (the inversion center).

• A crystal has inversion symmetry if its structure is unchanged when you replace every atom's position with its negative relative to the inversion center.
• Both the BCC and FCC lattices possess inversion symmetry. The simple cubic lattice does as well.
• Inversion symmetry has deep physical consequences. Materials with inversion symmetry cannot be piezoelectric or ferroelectric. It also forbids electric dipole transitions between states of the same parity.

Symmetry operations in crystals

The full set of symmetry operations present in a crystal determines its classification. Three hierarchical schemes organize this information: Bravais lattices describe the translational symmetry, point groups capture the rotational/reflective symmetry at a site, and space groups combine both.

Bravais lattices

There are exactly 14 Bravais lattices in three dimensions. These are the only distinct ways to arrange points periodically in space such that every point has an identical environment.

• The 14 lattices are grouped into 7 crystal systems based on the relationships between lattice parameters: triclinic, monoclinic, orthorhombic, tetragonal, cubic, trigonal, and hexagonal.
• Within a crystal system, different Bravais lattices arise from different centering types. The cubic system, for example, contains three: simple cubic (P), body-centered cubic (BCC, or I), and face-centered cubic (FCC, or F).
• Each Bravais lattice is defined by its lattice vectors and the symmetry operations that map it onto itself.

Point groups

Point groups describe the set of symmetry operations that leave at least one point fixed. They capture everything except translational symmetry.

• There are 32 crystallographic point groups, each a unique combination of rotations, reflections, and inversions compatible with a 3D lattice.
• Point groups are labeled using Hermann-Mauguin (international) notation. For example, $4/mmm$ indicates a 4-fold axis with a perpendicular mirror plane and two additional mirror plane types.
• Schoenflies notation is also common, especially in molecular physics. The group $C_{2v}$ has a 2-fold rotation axis and two vertical mirror planes. The group $O_h$ has the full symmetry of an octahedron (48 operations total).
• The point group of a crystal directly determines which physical properties (like optical activity or piezoelectricity) are allowed.

Space groups

Space groups are the most complete symmetry description of a crystal. They combine point group operations with translational symmetry, including two additional symmetry elements that arise from this combination.

• There are 230 distinct space groups in three dimensions.
• Screw axes combine rotation with a fractional translation along the rotation axis. A $2_1$ screw axis, for instance, rotates by 180° and translates by half a lattice vector.
• Glide planes combine reflection with a fractional translation parallel to the mirror plane.
• Space groups use Hermann-Mauguin notation with a leading letter for the lattice centering type. For example, $P2_1/c$ is a primitive monoclinic lattice with a $2_1$ screw axis and a $c$-glide plane. $Fm\bar{3}m$ is the space group of the FCC lattice with full octahedral symmetry.

Mathematical representation of symmetry operations

Symmetry operations can be expressed as matrices acting on coordinates or as operators acting on functions. This mathematical framework is what makes symmetry practically useful for calculations and deriving selection rules.

Transformation matrices

Each symmetry operation corresponds to a matrix that transforms the position vector of any point in the crystal.

• For 3D crystals, these are $3 \times 3$ matrices. A symmetry operation acting on a point $\mathbf{r} = (x, y, z)$ gives a new point $\mathbf{r'} = \mathbf{M}\mathbf{r}$, where $\mathbf{M}$ is the transformation matrix.
• The identity operation is simply the $3 \times 3$ identity matrix. Inversion is $-\mathbf{I}$ (negative identity).
• A 4-fold rotation about the $z$-axis (90° counterclockwise) has the matrix:

$\begin{pmatrix} 0 & -1 & 0 \\ 1 & 0 & 0 \\ 0 & 0 & 1 \end{pmatrix}$

You can verify this: the point $(1, 0, 0)$ maps to $(0, 1, 0)$, which is indeed a 90° counterclockwise rotation in the $xy$-plane.

• For operations that include translation (relevant for space groups), you use a $4 \times 4$ augmented matrix or write the operation as $\mathbf{r'} = \mathbf{M}\mathbf{r} + \mathbf{t}$, where $\mathbf{t}$ is the translation vector.

Symmetry operators

While transformation matrices act on coordinates, symmetry operators act on functions like wave functions or charge densities.

• A symmetry operator $\hat{R}$ acts on a function $f(\mathbf{r})$ to produce a new function: $\hat{R}f(\mathbf{r}) = f(\mathbf{R}^{-1}\mathbf{r})$. The inverse appears because we're transforming the argument of the function.
• The eigenfunctions of a symmetry operator are functions that remain unchanged (up to a multiplicative phase or constant) under the operation. These eigenfunctions form the basis functions of the irreducible representations of the symmetry group.
• The reflection operator for the $xy$-plane acts on coordinates as:

$\hat{\sigma}_{xy}: (x, y, z) \rightarrow (x, y, -z)$

which can be represented by the matrix $\begin{pmatrix} 1 & 0 & 0 \\ 0 & 1 & 0 \\ 0 & 0 & -1 \end{pmatrix}$

Consequences of symmetry

Symmetry doesn't just help you classify crystals. It places hard constraints on physical behavior: which energy levels are degenerate, which transitions can occur, and how complex your calculations need to be.

Degeneracy of energy levels

When a Hamiltonian has a particular symmetry, distinct quantum states related by that symmetry must share the same energy. This is symmetry-enforced degeneracy.

• The degree of degeneracy equals the dimension of the irreducible representation of the symmetry group to which those states belong. A 2D irreducible representation means a doubly degenerate level.
• Perturbations that lower the symmetry can lift (split) these degeneracies. An external magnetic field, for example, breaks time-reversal symmetry and splits previously degenerate spin states.
• Classic example: in a cubic crystal field, the five $d$ orbitals of a transition metal ion split into a doubly degenerate $e_g$ level and a triply degenerate $t_{2g}$ level. The cubic symmetry forces this specific splitting pattern.

Selection rules for transitions

Symmetry determines which quantum transitions are allowed. A transition between states $|i\rangle$ and $|f\rangle$ via operator $\hat{O}$ is allowed only if the matrix element $\langle f | \hat{O} | i \rangle \neq 0$.

Group theory provides a shortcut: the direct product of the irreducible representations of $|i\rangle$, $\hat{O}$, and $|f\rangle$ must contain the totally symmetric representation. If it doesn't, the matrix element is exactly zero and the transition is forbidden.

• In a centrosymmetric crystal (one with inversion symmetry), electric dipole transitions between states of the same parity are forbidden. This is because the electric dipole operator is odd under inversion, so the product of two even (or two odd) states with an odd operator cannot be totally symmetric.
• These rules save enormous effort. Instead of computing integrals, you can rule out transitions by symmetry alone.

Simplification of calculations

Symmetry reduces the computational work needed to solve quantum mechanical problems in crystals.

• Group theory lets you classify quantum states by their irreducible representations. States belonging to different representations don't mix, so the Hamiltonian matrix becomes block-diagonal.
• You can construct symmetry-adapted basis functions that automatically respect the crystal's symmetry. Each block of the Hamiltonian then corresponds to one irreducible representation and can be diagonalized independently.
• For a molecule or cluster with $D_{4h}$ symmetry, this might reduce a single large matrix eigenvalue problem into several much smaller ones, dramatically cutting computation time.

Symmetry breaking

Symmetry breaking occurs when a system transitions to a state with lower symmetry than the governing Hamiltonian. This is central to understanding phase transitions and the emergence of ordered states in solids.

Spontaneous symmetry breaking

In spontaneous symmetry breaking, the Hamiltonian retains its full symmetry, but the system's ground state does not.

• The system "chooses" one of several equivalent lower-symmetry states. Which state it picks may depend on fluctuations or boundary conditions.
• A non-zero order parameter characterizes the broken symmetry. In a ferromagnet, the order parameter is the spontaneous magnetization vector $\mathbf{M}$.
• Above the Curie temperature, a ferromagnet is paramagnetic with full rotational symmetry. Below it, the spins align along a particular direction, breaking that rotational symmetry spontaneously. The Hamiltonian doesn't prefer any direction, but the ground state does.
• Spontaneous symmetry breaking is also responsible for structural phase transitions, superconductivity, and charge density waves.

Explicit symmetry breaking

Explicit symmetry breaking happens when an external field or perturbation directly removes a symmetry from the Hamiltonian itself.

• Applying an electric field to a centrosymmetric crystal breaks inversion symmetry. The field defines a preferred direction, so $(x, y, z)$ and $(-x, -y, -z)$ are no longer equivalent.
• This can lift degeneracies that were protected by the original symmetry, creating new energy splittings.
• Explicit breaking is often used experimentally to probe a material's response: applying strain, electric fields, or magnetic fields and measuring how properties change reveals information about the underlying symmetry.

Applications of symmetry operations

Symmetry arguments appear throughout solid-state physics. Three major applications illustrate how symmetry shapes the physical behavior of real materials.

Electronic band structure

The symmetry of a crystal directly determines the structure of its electronic energy bands.

• High-symmetry points in the Brillouin zone (like $\Gamma$, $X$, $K$, $L$) are locations where the wave vector $\mathbf{k}$ is invariant under a subgroup of the point group. Electronic states at these points can be labeled by irreducible representations.
• Symmetry can enforce band degeneracies and band crossings at specific $\mathbf{k}$-points. These crossings are "protected" by symmetry and cannot be removed without breaking that symmetry.
• Graphene's famous linear (Dirac cone) dispersion near the $K$ points arises from the hexagonal lattice symmetry. The two sublattices of the honeycomb structure and the $C_{3v}$ symmetry at $K$ force the bands to touch linearly.

Phonon dispersion relations

Symmetry also governs the vibrational modes (phonons) of a crystal.

• The number and symmetry of phonon modes at any $\mathbf{k}$-point are determined by the space group. Group theory predicts how many distinct branches exist and their symmetry labels.
• Raman-active and infrared-active modes are determined by selection rules. In a centrosymmetric crystal, a mode that is Raman-active is infrared-inactive, and vice versa. This is the mutual exclusion rule.
• A diatomic linear chain has one acoustic branch and one optical branch. The optical branch involves the two atoms in the basis moving out of phase, and its activity depends on whether the mode changes the dipole moment (IR-active) or the polarizability (Raman-active).

Ferroelectricity and piezoelectricity

These phenomena are directly tied to the presence or absence of inversion symmetry.

• Piezoelectricity requires the absence of a center of symmetry. Of the 32 point groups, 21 lack inversion symmetry, and 20 of those are piezoelectric (the lone exception is the cubic group $432$, which has other symmetry constraints that cancel the effect).
• Ferroelectricity additionally requires a unique polar axis, meaning the point group must be one of the 10 polar point groups. Ferroelectric materials have a spontaneous electric polarization that can be reversed by an applied field.
• Barium titanate ($\mathrm{BaTiO_3}$) is a classic ferroelectric with a perovskite structure. Above 120°C it's cubic and centrosymmetric (paraelectric). Below 120°C, the Ti atom shifts off-center, breaking inversion symmetry and producing a spontaneous polarization. It undergoes further phase transitions (tetragonal → orthorhombic → rhombohedral) at lower temperatures, each involving additional symmetry lowering.