1.2 Miller indices

Definition of Miller indices

Miller indices are a compact notation for labeling planes (and directions) in a crystal lattice. They let you take any plane slicing through a periodic array of atoms and assign it a unique set of three integers (hkl). Once you have those integers, you can calculate interplanar spacings, predict diffraction peaks, and communicate crystal orientations without ambiguity.

Notation and conventions

• A plane's orientation is written as three integers in parentheses: (hkl).
• Negative indices are shown with a bar over the number: $(\ \bar{1}10)$ means $h = -1,\ k = 1,\ l = 0$.
• Curly braces {hkl} denote a family of symmetry-equivalent planes (e.g., {100} in a cubic crystal includes (100), (010), (001), and their negatives).
• Square brackets [hkl] label a direction in the lattice, while angle brackets ⟨hkl⟩ label a family of equivalent directions.
• All planes that are parallel to each other and have the same orientation share the same Miller indices. Shifting a plane by any number of lattice translations doesn't change its (hkl) label.

Significance in crystallography

• Miller indices give every diffraction spot a name. Each peak in an X-ray diffraction pattern corresponds to a specific (hkl) reflection.
• They let you compute interplanar spacing $d_{hkl}$, which links directly to diffraction angles through Bragg's law.
• They encode symmetry information: the set of allowed and forbidden (hkl) reflections tells you the Bravais lattice type and, ultimately, the space group.

Determining Miller indices

The standard procedure converts a plane's axis intercepts into a set of small integers. Getting comfortable with this recipe is essential because you'll use it repeatedly when reading diffraction data or sketching crystal planes.

Steps for calculation

1. Set up axes. Choose the three crystallographic axes $a$, $b$, $c$ of the unit cell as your coordinate system.
2. Find the intercepts. Determine where the plane crosses each axis, expressed in units of the lattice parameters. For example, a plane might cross at $2a$, $3b$, and $\infty c$ (meaning it runs parallel to the $c$-axis).
3. Take reciprocals. Invert each intercept: $\frac{1}{2},\ \frac{1}{3},\ \frac{1}{\infty} = 0$.
4. Clear fractions. Multiply through by the least common denominator to get integers: $\frac{1}{2} \times 6 = 3,\ \frac{1}{3} \times 6 = 2,\ 0 \times 6 = 0$.
5. Write the result. The Miller indices are (3 2 0).

A plane parallel to an axis has an intercept of $\infty$ on that axis, so its reciprocal is 0. A plane that intercepts on the negative side of the origin gives a negative index.

Worked example

Suppose a plane intercepts the axes at $1a$, $\infty b$, $\frac{1}{2}c$.

1. Intercepts: $1,\ \infty,\ \frac{1}{2}$
2. Reciprocals: $1,\ 0,\ 2$
3. These are already integers, so the Miller indices are (1 0 2).

Reciprocal lattice perspective

There's an elegant alternative viewpoint. Each set of Miller indices (hkl) corresponds to a reciprocal lattice vector:

$\mathbf{G}_{hkl} = h\,\mathbf{b}_1 + k\,\mathbf{b}_2 + l\,\mathbf{b}_3$

where $\mathbf{b}_1, \mathbf{b}_2, \mathbf{b}_3$ are the reciprocal lattice basis vectors. This vector $\mathbf{G}_{hkl}$ is perpendicular to the (hkl) planes, and its magnitude is $2\pi / d_{hkl}$. The reciprocal lattice picture is what makes diffraction theory work: each diffraction spot sits at a reciprocal lattice point.

Properties of Miller planes

Symmetry considerations

The symmetry of the crystal determines which planes are physically equivalent. In a cubic crystal, (100), (010), and (001) are all equivalent by symmetry and belong to the family {100}. In a lower-symmetry system like monoclinic, those three planes are generally not equivalent.

Certain reflections can be systematically absent because of the space group. For example, in a body-centered cubic (BCC) lattice, reflections with $h + k + l$ = odd are forbidden. These "extinction rules" are a direct consequence of how the basis atoms interfere destructively for specific (hkl) values.

Spacing between planes

The perpendicular distance between adjacent (hkl) planes is the d-spacing, $d_{hkl}$. For a cubic crystal with lattice parameter $a$:

$d_{hkl} = \frac{a}{\sqrt{h^2 + k^2 + l^2}}$

Higher Miller indices mean smaller d-spacing, which means more closely packed planes. For a tetragonal crystal ($a = b \neq c$):

$\frac{1}{d_{hkl}^2} = \frac{h^2 + k^2}{a^2} + \frac{l^2}{c^2}$

Orthorhombic, hexagonal, and lower-symmetry systems each have their own formula, but the principle is the same: d-spacing depends on both the indices and the lattice parameters.

Interplanar angles

The angle $\phi$ between two planes $(h_1 k_1 l_1)$ and $(h_2 k_2 l_2)$ in a cubic system is:

$\cos\phi = \frac{h_1 h_2 + k_1 k_2 + l_1 l_2}{\sqrt{h_1^2 + k_1^2 + l_1^2}\;\sqrt{h_2^2 + k_2^2 + l_2^2}}$

This is just the dot product of the two plane-normal vectors, normalized. For non-cubic systems the metric tensor of the lattice must be included, which makes the formula longer but follows the same logic.

Applications in crystal systems

Cubic structures

Cubic crystals are the simplest case because all three axes are orthogonal and equal in length. The plane normal direction [hkl] is parallel to the reciprocal lattice vector $\mathbf{G}_{hkl}$, so geometry is straightforward.

Key extinction rules to remember:

• Simple cubic (SC): All (hkl) reflections are allowed.
• BCC: Only reflections with $h + k + l$ = even are allowed.
• FCC: Only reflections where h, k, l are all odd or all even are allowed.

These rules come from the structure factor and are one of the first things you check when indexing a powder diffraction pattern.

Hexagonal structures

Hexagonal lattices have two equal axes ($a_1, a_2$) at 120° and a third axis ($c$) perpendicular to them. Standard three-index Miller notation works, but it can obscure the six-fold symmetry. That's why the four-index Miller-Bravais system is often preferred (see below).

Materials with hexagonal close-packed (HCP) structures, such as Ti, Mg, Zn, and graphite, are routinely described using this system.

Monoclinic structures

In monoclinic crystals, one interaxial angle ($\beta$) differs from 90°, so the relationship between (hkl) and the plane normal is less direct. You need the full metric tensor to compute d-spacings and angles. Many organic molecular crystals and some important inorganic compounds (e.g., certain phases of $\text{ZrO}_2$) crystallize in monoclinic space groups.

Miller-Bravais indices

Comparison with Miller indices

The Miller-Bravais system uses four indices (hkil) instead of three. The extra index $i$ is defined by the constraint:

$i = -(h + k)$

This redundancy is intentional. It makes symmetry-equivalent planes in the hexagonal system look obviously related. For instance, the three prismatic planes that are equivalent by the 6-fold axis are $(10\bar{1}0)$, $(0\bar{1}10)$, and $(\bar{1}100)$. In standard three-index notation, these would be (100), (010), and $(\bar{1}\bar{1}0)$, which don't look as clearly related.

Converting between systems

To go from three-index (hkl) to four-index (hkil):

1. Keep $h$ and $k$ the same.
2. Compute $i = -(h+k)$.
3. Keep $l$ the same.

To go back, just drop the third index $i$.

Directions use a different conversion rule. A three-index direction $[UVW]$ converts to four-index $[uvtw]$ via:

$u = \frac{1}{3}(2U - V),\quad v = \frac{1}{3}(2V - U),\quad t = -(u+v),\quad w = W$

Diffraction and Miller indices

Bragg's law

Bragg's law is the central equation connecting Miller indices to experiment:

$n\lambda = 2d_{hkl}\sin\theta$

Here $\lambda$ is the X-ray wavelength, $\theta$ is the angle between the incident beam and the lattice planes, $d_{hkl}$ is the interplanar spacing for the (hkl) planes, and $n$ is the order of diffraction (usually folded into the indices so you work with first-order reflections only).

In practice, you measure $\theta$ for each peak, compute $d_{hkl}$, and then figure out which (hkl) values are consistent with the crystal system. This is how lattice parameters are determined experimentally.

Structure factor

The structure factor $F_{hkl}$ determines the intensity of each diffraction peak:

$F_{hkl} = \sum_{j} f_j \, e^{2\pi i(hx_j + ky_j + lz_j)}$

where $f_j$ is the atomic scattering factor of atom $j$ and $(x_j, y_j, z_j)$ are its fractional coordinates in the unit cell. If $F_{hkl} = 0$, that reflection is extinct (has zero intensity). The pattern of which reflections vanish is how you identify the Bravais lattice and the presence of screw axes or glide planes.

For example, in an FCC lattice with one atom at each lattice point, $F_{hkl}$ vanishes whenever the indices are a mix of odd and even, confirming the extinction rule mentioned earlier.

Importance in material science

Texture analysis

Polycrystalline materials rarely have randomly oriented grains. Texture refers to the statistical distribution of grain orientations. You describe texture using Miller indices: a rolled aluminum sheet, for instance, might develop a {110}⟨112⟩ texture, meaning the {110} planes tend to lie parallel to the rolling plane and the ⟨112⟩ directions align with the rolling direction.

Texture is measured with pole figures, which plot the distribution of a chosen (hkl) plane normal across all sample orientations. Orientation distribution functions (ODFs) give the full three-dimensional texture description.

Grain orientation

Individual grain orientations are specified by stating which crystallographic direction points along each sample axis. Grain boundaries, where two differently oriented grains meet, strongly influence mechanical strength, corrosion resistance, and electrical conductivity. Techniques like electron backscatter diffraction (EBSD) map grain orientations across a sample surface, producing color-coded maps indexed by Miller indices.

Computational methods

Software for index determination

Several widely used packages handle Miller index calculations and visualization:

• VESTA and CrystalMaker for 3D visualization of planes and directions.
• CrystalDiffract and FullProf for simulating and refining diffraction patterns.
• GSAS-II for Rietveld refinement, which fits an entire diffraction pattern to extract lattice parameters and atomic positions.

These tools take raw diffraction data and automate the process of assigning (hkl) labels to each peak.

Automated indexing techniques

Modern diffraction experiments can generate thousands of peaks. Automated indexing algorithms (e.g., TREOR, DICVOL, or the algorithms built into GSAS-II) search for a unit cell that accounts for all observed d-spacings. Machine learning approaches are increasingly used to handle noisy data or complex multi-phase samples, which is especially relevant in high-throughput materials discovery.

Advanced concepts

Negative indices

A negative Miller index simply means the plane intercepts the corresponding axis on the negative side of the chosen origin. The plane $(\bar{1}00)$ is parallel to (100) but intercepts the $-a$ axis. Physically, $(\bar{1}00)$ and $(100)$ are the same family of parallel planes, so they have identical d-spacings and diffraction behavior. The distinction matters when you're specifying a particular crystal face (e.g., in crystal growth or surface science).

Rational vs. irrational indices

Real crystal planes always have rational (integer) Miller indices. This follows directly from the periodicity of the lattice: a plane that intersects lattice points periodically must make rational intercepts with the axes. Irrational indices would describe a plane incommensurate with the lattice, which doesn't correspond to a real set of lattice planes. Such planes can appear in theoretical treatments of incommensurate structures or quasicrystals, but they fall outside the standard Miller index framework.

Limitations and alternatives

High-index planes

Planes like (7 11 3) are perfectly valid but hard to visualize. They correspond to surfaces with high densities of atomic steps and kinks. In surface science and catalysis, these high-index surfaces can be highly reactive, so they matter practically even if they're geometrically complex. Stereographic projections and computer visualization are the main tools for working with them.

Non-crystalline materials

Miller indices require translational periodicity, so they don't apply to amorphous materials (glasses, many polymers) or quasicrystals. For amorphous solids, structural information comes from the pair distribution function $g(r)$, which describes the probability of finding an atom at distance $r$ from another atom. Quasicrystals, which have long-range order without periodicity, require higher-dimensional indexing schemes (e.g., six integer indices for icosahedral quasicrystals). These are active research areas in condensed matter physics.