Crystal Lattice

About Crystal Lattice

On April 8, 1982, in a laboratory at the National Bureau of Standards (now NIST) in Gaithersburg, Maryland, an Israeli materials scientist named Dan Shechtman pointed an electron microscope at a rapidly cooled aluminum-manganese alloy and saw a diffraction pattern that should not have existed. The bright spots on the pattern arranged themselves in concentric rings of ten — a pattern with ten-fold rotational symmetry, which immediately implied an underlying five-fold symmetry in the crystal structure. According to every textbook of crystallography from Bravais (1850) to Schoenflies and Fedorov (1891) to the Internationale Tabellen of 1935 (Hermann, ed.) and the International Tables for X-ray Crystallography of 1952 onward, this was impossible. Crystals could have 1-fold, 2-fold, 3-fold, 4-fold, or 6-fold symmetry. Not 5-fold. Never 5-fold. The rule was a mathematical consequence of how repeating unit cells fill space, and it had stood for 130 years.

Shechtman wrote "10 Fold???" in his lab notebook and spent the next two years trying to convince colleagues that what he had seen was real. Linus Pauling, the two-time Nobel laureate, famously dismissed Shechtman in remarks that became part of the published history of the discovery, calling him a "quasi-scientist" and suggesting the diffraction pattern was an artifact of multiple twinning rather than a real underlying structure. Shechtman was asked to leave his research group. The original paper (Shechtman, Blech, Gratias, and Cahn, 1984, Physical Review Letters 53:1951) appeared more than two years after the observation, and only after Ilan Blech worked out a structural model consistent with the data. In 2011, the Nobel Prize in Chemistry was awarded to Shechtman alone "for the discovery of quasicrystals." Pauling had died in 1994. The textbook had to be rewritten.

The Shechtman story is a useful entry point to crystal lattices because it illustrates what is and is not negotiable about crystalline order. The 14 Bravais lattices — the complete set of 3D arrangements of points that fill space with translational periodicity — are real. Bravais's 1850 enumeration is correct. The 230 space groups identified independently by Evgraf Fedorov in Russia and Arthur Schoenflies in Germany around 1890-1891 are correct. The crystallographic restriction theorem, which says that periodic crystals cannot have 5-fold, 7-fold, or higher symmetries, is mathematically true for periodic structures. What Shechtman showed is that physical solids are not always periodic. Quasicrystals are aperiodic — they have long-range order but no translational repetition — and the crystallographic restriction does not apply to them. The 5-fold symmetry that is forbidden to periodic crystals is permitted, even natural, in quasicrystals.

The 14 Bravais lattices arise from combining the seven crystal systems (triclinic, monoclinic, orthorhombic, tetragonal, cubic, rhombohedral/trigonal, and hexagonal) with the four possible centering types (primitive, base-centered, body-centered, face-centered). Not all combinations are independent; some duplicate others under symmetry equivalence. The result is exactly 14 distinct space lattices in three dimensions. In two dimensions there are 5 Bravais lattices and 17 wallpaper groups (the latter being a famous result Escher exploited extensively in his tilings).

Each Bravais lattice is described by a unit cell — the smallest parallelepiped that, when repeated by translations along three vectors, generates the entire lattice. The cubic system has unit cells with all three edges equal and all angles 90°; minerals crystallizing in this system include halite (NaCl, the cubic form of common salt), pyrite (FeS₂, "fool's gold," forming spectacular cubic crystals in many mineral deposits), galena (PbS), and diamond (a face-centered cubic form of carbon). The hexagonal system has two equal axes 120° apart and a third perpendicular axis; quartz (low quartz, the most common form of SiO₂) is technically trigonal but its unit cell is hexagonal, beryl (the family that includes emerald and aquamarine) is hexagonal, and ice (the common phase Ih) is hexagonal — which is why snowflakes have six-fold symmetry.

The full classification — 230 space groups — adds rotational, mirror, and inversion symmetries within each lattice to the translational symmetries that define the lattice itself. Fedorov's 1891 paper in the Russian Mineralogical Society's journal contained an initial enumeration of 229 groups; Schoenflies's 1891 book Krystallsysteme und Krystallstructur reached 227 independently. Through correspondence between 1889 and 1891, the two corrected each other's lists to arrive at the now-canonical 230. Barlow's 1894 independent derivation confirmed the result. All three approaches agreed because the result is a theorem of group theory: 230 is what you get when you enumerate all the ways periodic translations can combine with point symmetries in three-dimensional space.

The crystallographic restriction theorem says that a periodic crystal can have rotational symmetries of order 1, 2, 3, 4, or 6 — and no others. The reason is geometric: a rotation by 360°/5 = 72° applied to the shortest non-zero lattice vector generates a new lattice vector shorter than the original — which contradicts the assumption that the shortest non-zero vector exists. The same construction with 360°/3, 360°/4, and 360°/6 does not produce a shorter vector, so those orders are compatible with a discrete lattice. The result is that hexagonal and triangular tilings cover the plane but pentagonal ones do not. Until 1982, this seemed to settle the question of solid-state structure: every crystal had to be one of the 230 space groups.

Shechtman's discovery showed that the assumption "every solid is periodic" was wrong. Quasicrystals like Al₆Mn (the alloy Shechtman studied), Al₆₃Cu₂₄Fe₁₃, and many others have long-range orientational order (their diffraction patterns are sharp, indicating ordered atomic positions) but no translational periodicity. The atomic positions are determined by an aperiodic tiling — a tiling that never exactly repeats no matter how far you extend it. The mathematical prototype of such a tiling was discovered by Roger Penrose in 1974: the Penrose tiling uses two rhombic tiles (a "fat" rhombus with 72° and 108° angles, and a "thin" rhombus with 36° and 144° angles) to cover the plane in patterns with local 5-fold symmetry but no exact periodicity.

The Penrose tiling preceded Shechtman's discovery by eight years, and its connection was made retrospectively: Alan Mackay had calculated in 1981 that a 3D Penrose-like quasiperiodic structure would produce diffraction patterns with sharp peaks and 5-fold symmetry, and Levine and Steinhardt's 1984 paper coined the term "quasicrystal" and showed mathematically that the Penrose-tiling structure matched the diffraction patterns Shechtman had measured. Even earlier, Islamic architectural tilings from the 13th-15th centuries — particularly the Darb-i-Imam shrine in Isfahan (1453) and other Iranian and Central Asian sites — contained tiling patterns that Lu and Steinhardt (2007, Science 315:1106) identified as locally Penrose-like, though Makovicky's 2007 comment (Science 318:1383) argued that the patterns are periodic at the building scale and only locally five-fold symmetric. The dispute remains live; what is uncontested is that medieval Islamic craftsmen produced patterns with local five-fold symmetry centuries before Penrose's mathematical derivation.

Natural quasicrystals exist. The first one identified in nature, icosahedrite (Al₆₃Cu₂₄Fe₁₃), was reported in 2009 from a Koryak Mountains sample (Bindi et al. 2009, Science 324:1306); its extraterrestrial origin in the Khatyrka meteorite was confirmed in 2012 (Bindi et al., PNAS 109:1396). It had formed in space, possibly at the high pressures of an asteroid collision, billions of years before reaching Earth. A second natural quasicrystal, decagonite, was found in the same meteorite in 2015. Their existence in nature settles a lingering question: quasicrystals are not just laboratory curiosities. They are a real, stable phase of matter that the universe produces on its own.

The crystal lattice as a sacred-geometry topic is therefore richer than the usual "geometric purity" framing. The 14 Bravais lattices are real and the 230 space groups are real, but so is the aperiodic tiling that breaks the 130-year-old rule. The classification of order has had to expand. The lesson is not that the textbook was wrong; it is that the textbook needed more names for the kinds of order matter can have.

Mathematical Properties

A Bravais lattice is the set of all points generated by integer linear combinations of three linearly independent translation vectors a₁, a₂, a₃: L = {n₁a₁ + n₂a₂ + n₃a₃ : n_i ∈ ℤ}. The unit cell is the parallelepiped spanned by the three vectors. The seven crystal systems classify the cells by the relations among the edge lengths (a, b, c) and the angles (α, β, γ):

The crystallographic restriction theorem states that the rotational symmetries of any periodic Bravais lattice in 2D or 3D must be of order n where 2cos(2π/n) is an integer. This is satisfied by n = 1, 2, 3, 4, 6 (giving 2, 2, 1, 0, −1) and no others. In particular, n = 5 gives 2cos(72°) ≈ 0.618, not an integer, so 5-fold symmetry is forbidden in periodic structures.

The 230 space groups arise from combining the 14 Bravais lattices with the 32 crystallographic point groups (the allowed rotational and reflective symmetries that map the lattice to itself) and the additional symmetries of screws (combinations of rotation and translation along the axis) and glides (combinations of reflection and translation along the mirror plane). Fedorov 1891 and Schoenflies 1891 derived this list independently; Barlow 1894 confirmed it. The result is one of the cleaner theorems of geometric group theory.

Quasicrystals break the periodic assumption while retaining long-range orientational order. The mathematical structure is the "cut-and-project" construction: a quasicrystal in n dimensions is the projection of a slice through a periodic lattice in a higher dimension. For icosahedral quasicrystals, the higher dimension is 6; for the Penrose tiling, it is 5. The Fourier transform of a quasicrystal is a dense set of sharp peaks, indicating order, but the peaks are at positions involving irrational ratios (often involving the golden ratio φ = 1.618), so the structure never repeats by any single translation.

Occurrences in Nature

Crystal lattices are the underlying structure of essentially every mineral. Geologists and mineralogists classify minerals by their crystal system, chemistry, and habit (the external shape that growth typically produces).

The cubic system contains many of the most familiar minerals:

The hexagonal and trigonal systems contain quartz (low quartz, the trigonal form of SiO₂ with hexagonal unit cell — by far the most common silicate), beryl (Be₃Al₂Si₆O₁₈, the family of emerald, aquamarine, and morganite), tourmaline (a complex borosilicate), and graphite (a hexagonal lattice of carbon with weak interlayer bonding). Ice in its common form Ih is hexagonal, which is why snowflakes have six-fold symmetry (see Snowflake Symmetry).

The other systems contain less-familiar minerals: gypsum (CaSO₄·2H₂O) is monoclinic; sulfur is orthorhombic in its common form; cassiterite (SnO₂) is tetragonal. Most minerals fall into one of the higher-symmetry systems (cubic or hexagonal); fewer crystallize in the lower-symmetry triclinic or monoclinic systems, though many silicates do.

Beyond minerals, crystal lattices appear in every solid metal, in salts and ionic compounds, in semiconductor materials (silicon is face-centered cubic with the diamond structure; gallium arsenide is face-centered cubic with the zincblende structure), and in many organic crystals (sucrose, urea, aspirin). Frozen water vapor produces hexagonal ice crystals on cold surfaces; the structure of bone is calcium phosphate (hydroxyapatite) in hexagonal crystals embedded in a collagen matrix.

Quasicrystals were synthesized starting in 1982 (Shechtman et al.); natural quasicrystals are now known from two sites in the Khatyrka meteorite, with icosahedrite (Al₆₃Cu₂₄Fe₁₃, 2009) and decagonite (Al₇₁Ni₂₄Fe₅, 2015) being the first two formally named.

Architectural Use

Crystal lattices entered architecture as a self-conscious design vocabulary in the 20th century with the rise of geodesic, modular, and crystalline geometric forms in modernist and postmodernist building. Buckminster Fuller's geodesic domes (developed from the 1940s onward) drew on the geometry of the regular polyhedra and the close-packed sphere arrangements that underlie cubic crystal lattices.

Specific buildings:

Islamic geometric ornament from the 13th-15th centuries contains tile patterns with Penrose-like quasiperiodic structure. The Darb-i-Imam shrine in Isfahan, Iran (completed 1453) has tile patterns that Lu and Steinhardt (2007, Science 315:1106) identified as quasicrystalline, with local 5-fold and 10-fold symmetries and no exact periodicity. Other Iranian and Central Asian sites with similar patterns include the Friday Mosque of Isfahan and various madrasas in Samarkand. Medieval Islamic mathematicians and tile-cutters developed techniques to construct these patterns by hand, six centuries before quasicrystals were understood mathematically.

Construction Method

To draw a unit cell of a cubic crystal, mark eight points at the corners of a cube. For face-centered cubic, add a point at the center of each of the six faces; for body-centered cubic, add a single point at the center of the cube. For diamond structure, take face-centered cubic and add a copy of the structure offset by (¼, ¼, ¼) of the cube edge.

To draw a unit cell of a hexagonal crystal (such as graphite, ice, or beryl), mark points at the corners of a regular hexagonal prism: six corners on the top hexagonal face, six on the bottom, and (depending on the lattice) interior points at specific fractional positions.

To construct a Penrose tiling, the most common method uses two rhombic tiles: a 'fat' rhombus with 72° acute angles and a 'thin' rhombus with 36° acute angles, both with the same edge length. Matching rules — usually given as marked arcs or arrows on the edges — determine which tile can be placed adjacent to which. Following the matching rules forces aperiodicity; violating them allows periodic patterns. An inflation/deflation construction (each tile is broken into smaller tiles of the same two types) generates arbitrarily large patches of the tiling from a small starting region.

For 3D quasicrystal construction, the cut-and-project method projects a six-dimensional hypercube lattice through a 3D irrational slice. This is implemented in mathematical software (Mathematica, Quasitiler) and gives Penrose-like 3D structures with icosahedral symmetry, matching the diffraction patterns of physical icosahedral quasicrystals like Al₆₃Cu₂₄Fe₁₃.

Spiritual Meaning

Crystals occupy an unusual position in contemplative traditions. Their material reality is mineral — they are inorganic, often inert, durable across thousands of years — but their visual order (clear faces, sharp angles, internal regularity) has been read across many cultures as evidence of an underlying intelligibility in matter.

In Tibetan and Indian Vajrayana Buddhism, crystals (particularly quartz and rock crystal) are used in ritual implements (the vajra is sometimes constructed with crystal components) and as offerings on altars. The clear quartz crystal in particular is associated with the clear light of mind in Mahamudra and Dzogchen traditions — the analogy is between the transparency of the crystal and the unobstructed quality of awareness. Specific crystals (turquoise, coral, lapis lazuli) have associations with particular protective or auspicious functions in Tibetan tradition, often layered onto pre-Buddhist Bön mineral lore.

Indigenous Australian, Native American, and Mesoamerican traditions used crystals in healing, divination, and ceremony. The pre-Columbian Maya cut and polished obsidian (volcanic glass, technically amorphous rather than crystalline) into mirrors and ritual blades; the Aztec god Tezcatlipoca was associated with obsidian. The Australian Aboriginal traditions of quartz crystals as sources of healing power for medicine men ("clever men") are documented in 19th and 20th century ethnographies (A.P. Elkin's 1945 Aboriginal Men of High Degree covers this in detail).

In the Hermetic, Alchemical, and Renaissance Western traditions, crystals were associated with the perfection of matter — the idea that mineral growth, given enough time and the right conditions, would approach geometric perfection. The 17th-century mineralogist Niels Stensen (Nicolas Steno) made the foundational discovery in 1669 that crystal faces of a given mineral species always meet at the same angles regardless of size or perfection of the specimen — the "constancy of interfacial angles" that later became one of the foundations of crystallography. Steno's biography combined scientific work in geology and anatomy with a religious vocation (he eventually became a Catholic bishop and was beatified in 1988); for Steno, the geometric regularity of crystals was a direct demonstration of divine order in nature, and the science was a form of theology.

For contemporary practitioners, the crystal lattice can be approached as an object lesson in the relationship between local rule and global order: every atom in a crystal follows simple local bonding rules with its neighbors, and the macroscopic geometric form emerges from the systematic repetition of those rules. The form is not imposed from outside; it is the natural result of local order propagating through space. Whether one finds this metaphysically suggestive or simply geometrically interesting depends on one's framework. The fact that the universe produces minerals at all — that under certain conditions, atoms self-organize into lattices that can persist for billions of years — is, at minimum, a striking feature of physical reality.