Home / Learn / Exam prep / GATE Physics / Lesson 7.1
All lessons

Module 7: Solid state physics · Lesson 7.1 · about 50 min

Bravais lattices and the reciprocal lattice

Primitive vectors and cells, the reciprocal basis and why every reciprocal vector stands for a family of planes, the reciprocal lattices of sc, bcc, fcc, hexagonal and oblique lattices, Wigner-Seitz cells and Brillouin zones with their high-symmetry points, higher zones, and Fermi spheres inside the zone.

Key points, meanings and formulas on one page: choose "Save as PDF" in the print box.

Module 7 · about 50 minutes. This lesson takes the vocabulary of the JAM crystal structure lesson (lattice, basis, Miller indices) and builds the one idea every later solid-state lesson runs on: the reciprocal lattice and its Brillouin zones. Diffraction, phonons, electron bands and Fermi surfaces are all drawn in this space.

Anything that repeats with the lattice (an electron density, a potential, a vibration pattern) can be written as a sum of plane waves. Only certain wave vectors fit the repeat. Those wave vectors form a second lattice, the reciprocal lattice, and the cell around its origin, the first Brillouin zone, is the natural home of every wave in the crystal.

1. Bravais lattices and primitive cells, quickly

A Bravais lattice is the set of points R→=n1a→1+n2a→2+n3a→3 with integer ni, and its defining property is that every point has identical surroundings, in the same orientation. There are 5 in 2D and 14 in 3D. Arrays that fail the orientation test are lattices with a basis: the honeycomb (graphene), hcp, diamond and the kagome net.

A primitive cell holds exactly one lattice point; its volume Vc=|a→1⋅(a→2×a→3)| is the same for every choice of primitive vectors. Centred lattices are described with a bigger conventional cell, so keep the two volumes apart:

Lattice Conventional cube holds Primitive vectors (one choice) Vc
sc 1 point axˆ, ayˆ, azˆ a3
bcc 2 points a2(−1,1,1), a2(1,−1,1), a2(1,1,−1) a3/2
fcc 4 points a2(0,1,1), a2(1,0,1), a2(1,1,0) a3/4
hexagonal — axˆ, a2(xˆ+3yˆ), czˆ 32a2c

The symmetric primitive cell is the Wigner-Seitz cell: all points closer to one lattice point than to any other. It is built by drawing the perpendicular bisector planes to the neighbours and keeping the innermost region, so it has the full symmetry of the lattice. For sc it is a cube, for bcc a truncated octahedron, for fcc a rhombic dodecahedron.

2. The reciprocal lattice

Ask which plane waves eiG→⋅r→ have the period of the lattice. We need eiG→⋅(r→+R→)=eiG→⋅r→ for every R→, so

eiG→⋅R→=1for all R→⟺G→⋅R→=2π×integer

The solutions form a lattice of their own, generated by vectors b→j with a→i⋅b→j=2πδij:

b→1=2πa→2×a→3a→1⋅(a→2×a→3),b→2=2πa→3×a→1a→1⋅(a→2×a→3),b→3=2πa→1×a→2a→1⋅(a→2×a→3)

Then G→=hb→1+kb→2+lb→3 gives G→⋅R→=2π(hn1+kn2+ln3), as required. A lattice-periodic function therefore has the Fourier series f(r→)=∑G→fG→eiG→⋅r→ with fG→=1Vc∫cellfe−iG→⋅r→d3r.

`$\vec b_j$`
reciprocal primitive vectors; units of inverse length (Å⁻¹); `$\vec b_1$` is perpendicular to `$\vec a_2$` and `$\vec a_3$`
`$\vec G$`
a reciprocal lattice vector: the wave vector of a plane wave with the lattice period
Reciprocal cell volume
`$(2\pi)^3/V_c$`: a small real cell means a big reciprocal cell
Physical meaning
The direct lattice says where the atoms repeat; the reciprocal lattice says which spatial frequencies the crystal can contain. A diffraction pattern is a picture of the reciprocal lattice.

Three facts follow at once and are worth memorising:

  1. The reciprocal of the reciprocal is the direct lattice (the definition is symmetric in a→ and b→).
  2. The reciprocal lattice depends only on the Bravais lattice, never on the basis. The basis decides how strongly each G→ scatters (the structure factor, lesson 7.2).
  3. G→hkl is normal to the planes (hkl) and |G→hkl|=2π/dhkl, for any lattice, when h,k,l have no common factor.

Why G is normal to (hkl). The plane nearest the origin cuts the axes at a₁/h, a₂/k, a₃/l. Two vectors lying in it are a₁/h − a₂/k and a₂/k − a₃/l.

Dot them with G = h b₁ + k b₂ + l b₃. G · (a₁/h − a₂/k) = 2π − 2π = 0, and the same for the other. So G is perpendicular to the plane.

The spacing. The distance from the origin to this plane is the projection of a₁/h on the unit normal: d = (G/|G|) · a₁/h = 2π/|G|.

Consequence. Every family of lattice planes is one point of the reciprocal lattice (its shortest G in that direction). Multiples nG are the higher orders of Bragg reflection.

Worked example (non-cubic spacing). A tetragonal crystal has a=4.0 Å and c=6.0 Å. Its reciprocal axes are orthogonal with lengths 2π/a,2π/a,2π/c, so |G→101|2=(2π)2(1/a2+1/c2) and 1/d2=1/16+1/36=0.0903 Å⁻², d101=3.33 Å. For hexagonal axes the same method gives 1/d2=43(h2+hk+k2)/a2+l2/c2; the angle between two plane normals is the angle between their G→s, never between the bare index triples (unless the crystal is cubic).

3. Reciprocal lattices you must know

Work b→1=2πa→2×a→3/Vc for the fcc vectors above: a→2×a→3=a24(−1,1,1) and Vc=a3/4, so b→1=2πa(−1,1,1). These are bcc primitive vectors.

Direct lattice Reciprocal lattice Shortest G→ First Brillouin zone
sc, edge a sc, edge 2π/a 2π/a cube
bcc, edge a fcc, cube edge 4π/a 22π/a (the {110} planes) rhombic dodecahedron (12 faces)
fcc, edge a bcc, cube edge 4π/a 23π/a (the {111} planes) truncated octahedron (8 hexagons + 6 squares)
2D hexagonal, side a hexagonal, side 4π/(3a), turned 30∘ 4π/(3a) regular hexagon
2D oblique, a1,a2,γ oblique, angle 180∘−γ `$ \vec b_i

In fcc-indexing language: 2πa(h,k,l) is a reciprocal vector of fcc only when h,k,l are all even or all odd, and of bcc only when h+k+l is even. These are exactly the X-ray selection rules for the two cubic lattices: missing reflections are just points that are not in the true reciprocal lattice.

Change the lattice in the figure and check the claims: b→1 stays perpendicular to a→2, the hexagonal reciprocal lattice is turned by 30°, and the readout shows that every zone has the same area.

Left: a 2D lattice and its primitive cell. Right: its reciprocal lattice with the first four Brillouin zones, found by counting how many Bragg lines (perpendicular bisectors of reciprocal vectors) lie between each point and the origin.

4. Brillouin zones

The first Brillouin zone is the Wigner-Seitz cell of the reciprocal lattice. Its faces are Bragg planes: the perpendicular bisectors of G→. They matter because a wave with wave vector k→ is Bragg-reflected exactly when k→ ends on such a plane. For elastic scattering |k→+G→|=|k→|, i.e.

2k→⋅G→+G2=0⟺k→⋅Gˆ′=12|G→′|(G→′=−G→)

So electrons with k→ on a zone face are reflected and band gaps open there; phonons on a zone face are standing waves with zero group velocity.

The nth zone is the set of points reached from the origin by crossing exactly n−1 Bragg planes. Higher zones are broken into pieces, but each can be moved by reciprocal vectors to fill the first zone exactly once, so every zone has the same volume (2π)3/Vc. For the square lattice of side a, the second zone lies between the first square (side 2π/a) and the square of side 22π/a turned by 45°: area 8π2/a2−4π2/a2=4π2/a2, as promised.

High-symmetry points have standard names. For the fcc lattice (zone points in units of 2π/a):

Point Where Coordinates Distance from Γ
X centre of a square face (1,0,0) 2π/a
L centre of a hexagonal face (12,12,12) 3π/a (nearest face)
K middle of an edge between two hexagons (34,34,0) 32π/(2a)
W corner (square meets two hexagons) (1,12,0) 5π/a (farthest)

For bcc: H=2πa(1,0,0) (corner, farthest), N=2πa(12,12,0) (face centre, nearest, 2π/a), P=2πa(12,12,12). For the 2D hexagonal lattice: M at 2π/(3a) and K at 4π/(3a), where graphene’s Dirac cones sit. A point is inside the fcc zone when |kx|+|ky|+|kz|≤32 and every |ki|≤1 (units 2π/a); on the surface when one of these is an equality.

Worked example (the zone in numbers). Copper is fcc with a=3.61 Å. Vc=a3/4=11.76 ų, so VBZ=248.05/11.76=21.1 Å⁻³. ΓX=2π/a=1.74 Å⁻¹ and ΓL=3π/a=1.51 Å⁻¹.

“The Brillouin zone is the conventional cell of the reciprocal lattice, and higher zones are shells of growing radius.” Neither. The first zone is the Wigner-Seitz cell (it has the full symmetry, and its volume is the primitive one, (2π)3/Vc, not (2π)3/a3). Higher zones are not spherical shells: a point at distance 1.2π/a from Γ in a square lattice can be in the first zone (along a diagonal) or the second (along an axis). The rule is “how many Bragg planes have I crossed”, never “how far am I”.

5. Counting states and the Fermi sphere in the zone

With periodic boundary conditions on a crystal of N cells, the allowed k→ form a fine grid with one point per (2π)3/(NVc). The first zone therefore holds exactly N wave vectors: one per primitive cell. Each electron band holds 2N electrons (spin), and each phonon branch has N modes. Because k→ and k→+G→ describe the same Bloch state or phonon, ℏk→ is a crystal momentum, conserved only up to ℏG→.

Worked example (does the Fermi sphere fit?). A monovalent fcc metal has n=4/a3 electrons, so kF=(3π2n)1/3=(12π2)1/3/a=4.91/a. The nearest zone faces (L) are at 3π/a=5.44/a. Ratio 0.90: the sphere sits inside the zone, which is why Na, K, Cu, Ag and Au have nearly spherical Fermi surfaces. A divalent simple cubic metal has kF=(6π2)1/3/a=1.24π/a: its sphere pokes through the faces (into zone 2) while the corners of zone 1 stay empty, so it conducts although “two electrons per cell fill one band”.

To see 3D zones from every angle, use the Brillouin zone viewer; the Miller index calculator works spacings and angles for every crystal system, and the band structure explorer shows what happens to electrons at the zone faces. The Ewald construction in the next lesson uses all of this.

  • a→i⋅b→j=2πδij; G→ satisfies eiG→⋅R→=1. The reciprocal lattice depends only on the Bravais lattice.
  • G→hkl⟂(hkl) and |G→hkl|=2π/dhkl in every crystal; plane normals and angles come from G→, not from [hkl].
  • sc ↔ sc (2π/a); bcc ↔ fcc with cube edge 4π/a; 2D hexagonal → hexagonal of side 4π/(3a), turned 30°; 2D angle γ→180∘−γ.
  • Zone volume (2π)3/Vc with the primitive volume: sc 8π3/a3, bcc 16π3/a3, fcc 32π3/a3. All zones have equal volume.
  • fcc zone: X at 2π/a, L at 3π/a, K at 32π/2a, W at 5π/a. bcc zone: N at 2π/a, H at 2π/a. Hexagonal 2D: K at 4π/3a.
  • The zone holds N wave vectors; ℏk→ is crystal momentum, conserved modulo ℏG→.

Practise: exam questions on this lesson

Next: lesson 7.2 puts the reciprocal lattice to work in diffraction: the Laue condition, the Ewald sphere and the structure factor.

Check yourself

1The reciprocal lattice of an fcc lattice with cube edge a is
2What is the volume of the first Brillouin zone of a bcc lattice with cube edge a?
3In any crystal, the normal to the planes (hkl) is parallel to

Sources

  • N. W. Ashcroft and N. D. Mermin, Solid State Physics (1976), chapters 4, 5 and 9
  • C. Kittel, Introduction to Solid State Physics, 8th ed. (2005), chapter 2
  • J. M. Ziman, Principles of the Theory of Solids, 2nd ed. (1972), chapter 1
  • GATE Physics (PH) syllabus, Solid State Physics section, published by the organising IIT each year

Your ticks stay in this browser. Save them as a file on the course home.

Updated 2026-11-22. Not affiliated with the IITs, IISc or the GATE organising institute. The general aptitude part of GATE is not covered here. Check the official GATE website for the current pattern. All questions are original.