Crystallography and Mineralogy: Symmetry, Form and Twinning, Projections, Mineral Classification and Optics, Nucleation and Mineral Chemistry

This chapter answers the Part B1 heading “Crystallography and Mineralogy”: “Elements of crystal symmetry, form, and twinning; Crystallographic projections; Classification of minerals, physical and optical properties of rock-forming minerals; Mineral nucleation; Mineral chemistry (major, trace, and rare earth elements), cation re-calculations.” It is the one chapter of this subject in the Geology and Geophysics (GG) paper. It teaches symmetry and the crystal systems, Miller indices, zones, interplanar spacing and Bragg's law, projections, twinning, silicate classification with physical and optical properties, nucleation theory, and mineral chemistry with cation recalculation. The numericals are Bragg angles, d-spacings, birefringence, critical nucleus radius, cation recalculation on an oxygen basis and end-member content.

1. Crystal symmetry and the seven systems

A crystal's form reflects an internal periodic lattice. Its symmetry elements are rotation axes, mirror planes, a centre of inversion and rotoinversion axes. Only 1-, 2-, 3-, 4- and 6-fold axes can occur, since only these fit a repeating lattice. Their combinations give 32 crystal classes (point groups) in 7 crystal systems; adding translations gives 14 Bravais lattices and 230 space groups.

The seven crystal systems
SystemAxes and anglesExamples
Cubica = b = c; all 90°Halite, galena, fluorite, garnet, pyrite
Tetragonala = b ≠ c; all 90°Zircon, rutile
Orthorhombica ≠ b ≠ c; all 90°Olivine, orthopyroxene, barite
Hexagonala = b ≠ c; γ = 120°, α = β = 90°Beryl, apatite
TrigonalHexagonal axes, or rhombohedral a = b = c with α = β = γ ≠ 90°Quartz, calcite, tourmaline
Monoclinica ≠ b ≠ c; α = γ = 90° ≠ βOrthoclase, micas, gypsum, augite
Triclinica ≠ b ≠ c; no angle fixedPlagioclase, microcline, kyanite

The highest cubic class, 4/m 3̄ 2/m (m3m), has three 4-fold, four 3-fold and six 2-fold axes, nine mirror planes and a centre, 48 operations in all; halite, galena, fluorite and garnet belong to it. Pyrite (2/m 3̄) has only three mirror planes. Quartz is trigonal at ordinary temperature, not hexagonal.

2. Form, Miller indices, zones, spacing and Bragg's law

Steno's law: angles between corresponding faces of a mineral are constant. A face is labelled by Miller indices (hkl): take its intercepts on the axes in units of the cell edges, invert them and clear fractions. Intercepts 2a, 3b and 6c give 1/2, 1/3, 1/6, which times 6 give (321). A face parallel to an axis has intercept ∞ and index 0. (hkl) is one plane, {hkl} a form (all symmetry-equivalent faces: in the cubic system {100} is the cube with 6 faces, {111} the octahedron with 8, {110} the rhombic dodecahedron with 12), and [uvw] a direction. Hexagonal and trigonal crystals use four-index Miller–Bravais (hkil), with i = −(h + k).

Zone law: face (hkl) lies in the zone of axis [uvw] when hu + kv + lw = 0, and the zone axis common to two faces is the cross product [k1l2 − l1k2, l1h2 − h1l2, h1k2 − k1h2]. For (110) and (001) it is [1 −1 0]. Interplanar spacing d is d = a/√(h² + k² + l²) for cubic, 1/d² = h²/a² + k²/b² + l²/c² for orthorhombic and 1/d² = (4/3)(h² + hk + k²)/a² + l²/c² for hexagonal crystals. Bragg's law, nλ = 2 d sin θ, gives the diffraction angle. Worked example: halite, a = 5.64 Å, plane (200): d = 2.82 Å; for λ = 1.54 Å and n = 1, sin θ = 1.54/(2 × 2.82) = 0.273, θ = 15.8° and the detector angle is 2θ = 31.7°.

⚠️ 2θ and higher orders
The detector sits at 2θ, not θ. A second-order reflection (n = 2) from (hkl) is the same as a first-order reflection from (2h 2k 2l), whose spacing is d/2, so it appears at a larger angle.

3. Crystallographic projections

A projection reduces face orientations to a plane diagram. Each face is represented by its pole, where the face normal meets a reference sphere, located by the azimuth φ around the c axis and the polar angle ρ from it. Poles of faces in one zone lie on one great circle. Clinographic and orthographic projections serve for drawing crystal shapes.

Projections compared
ProjectionDistance of pole from centreProperty and use
Stereographicr = R tan(ρ/2), R the primitive radiusPreserves angles; circles stay circles; zones are great circles; standard for crystal and structural orientation
Gnomonicr = d tan ρ, d the plane's distance from the sphere's centreZones are straight lines, good for indexing; faces at ρ = 90° go to infinity
Equal-areaArea-preserving radial scalePoles keep their density, so used to contour orientation data

Worked example: on a stereographic net with primitive radius R = 10 cm, a pole at ρ = 60° plots at r = 10 tan 30° = 5.77 cm and one at ρ = 90° on the primitive, r = 10 cm; on a gnomonic plane at d = 10 cm the same ρ = 60° pole is at 10 tan 60° = 17.3 cm. The interfacial angle between two faces is the arc between their poles measured along the great circle through both.

4. Twinning

A twin is an oriented intergrowth of crystals related by a symmetry operation that is not part of the crystal's own point group: a twin plane (reflection), a twin axis (usually a 180° rotation) or a centre. The boundary is the composition plane. Contact twins share a planar boundary (spinel-law twins on (111)); penetration twins interpenetrate (Carlsbad twins of orthoclase, the pyrite iron cross, staurolite crosses at about 60° and 90°); polysynthetic twins repeat many times (albite twinning of plagioclase on (010)); and cyclic twins repeat around an axis (aragonite and chrysoberyl trillings). Microcline's tartan (cross-hatched) twinning combines the albite and pericline laws.

By origin, growth twins form during crystallisation; transformation twins form when a high-symmetry phase inverts to a lower-symmetry one on cooling, as in Dauphiné twins of quartz at the high–low inversion near 573 °C (at atmospheric pressure) and the tartan twins commonly attributed to the monoclinic to triclinic inversion of K-feldspar; and deformation (gliding) twins form under stress, as in calcite, whose twin lamellae serve as a palaeostress and strain gauge. In plagioclase, adjacent albite lamellae extinguish alternately under crossed polars, which separates twinning from exsolution lamellae, which differ in composition rather than in orientation.

5. Classification and physical properties of rock-forming minerals

Minerals are classified by the dominant anion or anionic group (Dana and Strunz): native elements, sulphides, oxides, halides, carbonates, sulphates, phosphates and silicates. Silicates, the bulk of the crust, are classified by how the SiO₄ tetrahedra link.

Silicate structural classes
ClassLinkageSi : OExamples
NesosilicateIsolated tetrahedra1 : 4Olivine, garnet, zircon, kyanite
SorosilicatePaired tetrahedra2 : 7Epidote group
CyclosilicateRings1 : 3Beryl, tourmaline, cordierite
Inosilicate, single chainSingle chains1 : 3Pyroxenes
Inosilicate, double chainDouble chains4 : 11Amphiboles
PhyllosilicateSheets2 : 5Micas, chlorite, clays, talc
TectosilicateFrameworks1 : 2Quartz, feldspars, feldspathoids, zeolites

Cleavage mirrors structure: micas have one perfect cleavage; pyroxenes two near 90° (87° and 93°) and amphiboles two at about 56° and 124°; feldspars two near 90°; calcite three rhombohedral (about 75° and 105°); halite three at 90°; fluorite four (octahedral); quartz none, with conchoidal fracture. Mohs hardness: talc 1, gypsum 2, calcite 3, fluorite 4, apatite 5, orthoclase 6, quartz 7, topaz 8, corundum 9, diamond 10. Specific gravity, lustre, streak and magnetism complete the hand-specimen description.

6. Optical properties in thin section

In plane-polarised light a mineral shows colour, pleochroism (colour change on rotation, as in biotite, hornblende and tourmaline), relief (contrast with the mounting medium, set by the difference in refractive index; the Becke line moves into the higher-index medium when the stage is lowered), cleavage and form. Isotropic minerals (cubic crystals such as garnet and fluorite, and glass) stay dark in every position between crossed polars. Anisotropic minerals split light into two rays. Uniaxial minerals (tetragonal, hexagonal, trigonal) have indices ω and ε, with optic sign positive if ε > ω (quartz) and negative if ε < ω (calcite). Biaxial minerals (orthorhombic, monoclinic, triclinic) have nα < nβ < nγ and an optic angle 2V, positive when the acute bisectrix is γ.

Birefringence δ = n(max) − n(min), that is |ε − ω| for uniaxial and nγ − nα for biaxial minerals. Retardation Γ = t δ, with t the thickness in the same units as Γ, fixes the interference colour: in a standard 30 μm (30 000 nm) section quartz (δ = 0.009) gives Γ = 270 nm, first-order grey to white, and calcite (δ = 0.172) gives 5160 nm, pale high-order pastel colours. First-order red-violet is at about 550 nm, so the order number is roughly Γ/550.

Identifying common minerals in thin section
MineralRelief and birefringenceDiagnostic features
QuartzLow; grey, δ about 0.009Colourless, no cleavage, undulose extinction; uniaxial positive
PlagioclaseLow; grey, δ about 0.007 to 0.013Polysynthetic albite twins
OlivineHigh; upper second- to third-order colours, δ about 0.035 to 0.05Colourless, no cleavage, irregular cracks
AugiteHigh; second-order, δ about 0.02 to 0.03Two cleavages near 90°, inclined extinction
HornblendeModerate to high; δ about 0.02Green-brown pleochroism, cleavages at 56° and 124°
GarnetVery high; isotropicDark between crossed polars, rounded, no cleavage

7. Mineral nucleation

Crystals nucleate when a melt or solution is undercooled or supersaturated. Classical nucleation theory balances the volume free energy released against the energy of the new interface: ΔG(r) = (4/3)π r³ ΔGv + 4π r² γ, where ΔGv < 0 per unit volume and γ is the interfacial energy. The barrier peaks at the critical radius r* = −2γ/ΔGv, with ΔG* = 16π γ³/(3 ΔGv²); smaller clusters redissolve and larger ones grow. For γ = 0.3 J/m² and ΔGv = −1.5 × 10⁸ J/m³, r* = 2 × 0.3/(1.5 × 10⁸) = 4 nm. Since |ΔGv| rises roughly in proportion to the undercooling ΔT, r* falls as 1/ΔT and ΔG* as 1/ΔT².

Heterogeneous nucleation on existing surfaces (crystals, walls, xenocrysts) lowers the barrier by a factor f(θ) = (2 + cos θ)(1 − cos θ)²/4, below 1, and so occurs at smaller undercooling. Nucleation and growth rates each rise and then fall with undercooling, nucleation peaking at the larger undercooling. Slow cooling gives few nuclei that grow large (coarse, equigranular), faster cooling many small crystals, and quenching gives glass, or dendritic and spherulitic forms when diffusion limits growth. A change in cooling history makes porphyritic texture. Ostwald ripening grows large grains at the expense of small ones, to reduce interfacial energy, and metamorphic minerals often need overstepping of the reaction boundary to nucleate.

8. Mineral chemistry: major, trace and rare earth elements, and cation recalculation

Major-element analyses are given as oxide weight per cent. To get a formula on a stated oxygen basis (olivine 4, pyroxene 6, feldspar 8, garnet 12, anhydrous amphibole 23, anhydrous mica 11 or 22):

  • Divide each oxide wt% by its molecular weight to get moles of oxide, then multiply by the number of cations per oxide (SiO₂ 1, Al₂O₃ 2) and the oxygens per oxide (SiO₂ 2, Al₂O₃ 3, FeO and MgO 1).
  • Sum the oxygen proportions, divide the oxygen basis by this sum to get the normalising factor, and multiply every cation proportion by it. Check the cation total against the ideal (olivine 3 on 4 O, pyroxene 4 on 6 O, feldspar 5 on 8 O, garnet 8 on 12 O).
  • If all iron is analysed as FeO, charge balance estimates Fe³⁺ by Droop's formula, Fe³⁺ = 2X(1 − T/S), with X the ideal oxygens, T the ideal cation total and S the calculated total with all Fe as Fe²⁺.

Worked example, olivine on 4 O (molecular weights SiO₂ 60.08, FeO 71.85, MgO 40.30): SiO₂ 39.0, FeO 15.0, MgO 45.0 wt% give 0.649, 0.209 and 1.117 mol, oxygen sum 2.624, factor 4/2.624 = 1.525, and cations Si 0.990, Fe 0.318, Mg 1.702, total 3.01, so (Mg₁.₇₀Fe₀.₃₂)Si₀.₉₉O₄ and Fo = 100 × Mg/(Mg + Fe) = 84. Solid-solution content is always a mole ratio of cations, not a weight ratio.

Trace elements (below about 0.1 wt%) are described by the partition coefficient D = C(mineral)/C(melt). Compatible elements (D > 1), such as Ni, Cr and Co in olivine and pyroxene, stay in the solid; incompatible ones (D < 1), such as Rb, Ba, Th, U and the light REE, concentrate in the melt. By Goldschmidt's rules ions of the same charge and radius within about 15% substitute freely, and where charges differ the higher-charged ion is preferred by the crystal. The rare earth elements (La to Lu) are trivalent with radius shrinking along the series (lanthanide contraction), so light REE are more incompatible than heavy REE; they are plotted normalised to chondrite. Eu also occurs as Eu²⁺ and substitutes for Ca and Sr in plagioclase, so plagioclase removal leaves a negative Eu anomaly in the melt and accumulation a positive one. Garnet takes heavy REE strongly, so garnet in the source gives HREE-depleted melts, and Ce shows an anomaly under oxidising conditions (Ce⁴⁺), as in seawater.

Key takeaways

  • 32 classes fall into 7 systems; only 1-, 2-, 3-, 4- and 6-fold axes exist; class m3m has 9 mirror planes and 48 operations; quartz and calcite are trigonal. Miller indices are inverted intercepts cleared of fractions, and a face lies in a zone when hu + kv + lw = 0.
  • Bragg's law nλ = 2 d sin θ with the detector at 2θ; cubic d = a/√(h² + k² + l²). Stereographic projection (r = R tan(ρ/2)) preserves angles, gnomonic (r = d tan ρ) turns zones into straight lines, equal-area is for contouring.
  • Twins are related by an operation outside the point group: contact, penetration, polysynthetic and cyclic; growth, transformation (Dauphiné, microcline) and deformation (calcite) in origin. Albite twinning is on (010).
  • Silicates are classified by Si : O (olivine 1 : 4, pyroxene 1 : 3, amphibole 4 : 11, micas 2 : 5, framework 1 : 2). Birefringence is δ = nmax − nmin and retardation Γ = t δ. Uniaxial quartz is positive and calcite negative; isotropic minerals stay dark between crossed polars.
  • Nucleation has a barrier peaking at r* = −2γ/ΔGv, with ΔG* ∝ γ³/ΔGv². More undercooling means smaller r* and a lower barrier; heterogeneous nucleation lowers it further. Cooling rate controls texture, from coarse to glassy to porphyritic.
  • Cation recalculation: moles of oxide, oxygen proportions, factor = oxygen basis / sum, cations = moles × cations per oxide × factor. Solid-solution end members are mole ratios. D > 1 is compatible, D < 1 incompatible; plagioclase removal makes a negative Eu anomaly and garnet retains heavy REE.

Practice questions (18)

Attempt each one before opening the answer. Every explanation names the tempting wrong option as well as the right one, because that is where marks are lost.

  1. How many mirror planes does the highest-symmetry cubic crystal class, 4/m 3̄ 2/m (m3m), possess?

    1. 3, the three coordinate planes only, as in pyrite
    2. 6, the six diagonal planes through opposite edges only
    3. 9, three coordinate planes plus six diagonal planes
    4. 12, one plane for each of the twelve edges of the cube
    Show answer

    Answer: C — 9, three coordinate planes plus six diagonal planes

    The class m3m has three mirror planes parallel to the cube faces and six diagonal planes, nine in all, with 48 symmetry operations in total. Three planes is the number for pyrite (2/m 3̄), which is the tempting answer, and the other counts do not correspond to any combination of the cube's symmetry planes.
  2. A crystal face cuts the a, b and c axes at 2a, 4b and 1c respectively. Its Miller indices are:

    1. (214)
    2. (241)
    3. (124)
    4. (421)
    Show answer

    Answer: A — (214)

    Miller indices are the reciprocals of the intercepts cleared of fractions: 1/2, 1/4, 1 multiplied by 4 gives 2, 1, 4, so (214). The set (241) simply lists the intercepts without inverting them, which is the usual slip, and (124) and (421) reverse the order of the axes.
  3. Which of the following faces lies in the zone whose axis is [1 −1 0]?

    1. (121)
    2. (211)
    3. (101)
    4. (112)
    Show answer

    Answer: D — (112)

    A face (hkl) lies in the zone [uvw] when hu + kv + lw = 0. With [u v w] = [1 −1 0] this reduces to h − k = 0, so h = k. Only (112) satisfies it: 1 − 1 + 0 = 0. (121) gives −1, (211) gives +1 and (101) gives +1.
  4. Fluorite is cubic with a = 5.46 Å. X-rays of wavelength 1.54 Å are diffracted from the (111) planes. The detector angle 2θ for the second-order reflection (n = 2), in degrees to one decimal place, is ____.

    Numerical answer — type the value.

    Show answer

    Answer: 58.5

    For cubic crystals d(111) = a/√3 = 5.46/1.732 = 3.152 Å. Bragg's law nλ = 2 d sin θ gives sin θ = 2 × 1.54/(2 × 3.152) = 0.4885, so θ = 29.24° and 2θ = 58.5°. Using n = 1 gives 2θ = 28.3°, and reporting θ instead of 2θ gives 29.2°; both are the usual slips.
  5. Quartz can be indexed on hexagonal axes with a = 4.91 Å and c = 5.41 Å. For hexagonal crystals 1/d² = (4/3)(h² + hk + k²)/a² + l²/c². The interplanar spacing d of the (101) planes, in Å to two decimal places, is ____.

    Numerical answer — type the value.

    Show answer

    Answer: 3.34

    For (101), h = 1, k = 0, l = 1, so h² + hk + k² = 1. Then 1/d² = (4/3)(1)/24.108 + 1/29.268 = 0.05531 + 0.03417 = 0.08948 Å⁻², and d = 3.34 Å. Dropping the factor 4/3, which belongs to the hexagonal metric, would give d = 3.64 Å.
  6. In which projection do the poles of all the faces of one zone plot on a straight line?

    1. Stereographic projection, because every zone appears as a straight diameter of the net for any orientation
    2. Gnomonic projection, because a plane through the sphere's centre cuts the tangent plane in a straight line
    3. Equal-area projection, because preserving the area enclosed by the poles of a zone makes the zone straight
    4. All projections alike, because the poles of a zone always lie in one plane through the sphere's centre
    Show answer

    Answer: B — Gnomonic projection, because a plane through the sphere's centre cuts the tangent plane in a straight line

    In gnomonic projection each pole is projected from the sphere's centre onto a tangent plane, so the zone, being a plane through the centre, appears as a straight line. In stereographic projection the zone is a great circle, a straight diameter only for the special case of a plane through the net's centre, which is the tempting claim. The poles of a zone do lie in one plane through the centre, but other projections turn that plane into a curve.
  7. Which of the following statements about twinning are correct?

    1. Polysynthetic albite twinning in plagioclase is on the plane (010)
    2. The tartan (cross-hatched) twinning of microcline combines the albite and pericline laws
    3. The operation relating twin individuals always belongs to the point group of the untwinned crystal
    4. Staurolite cruciform twins are penetration twins whose two individuals cross at about 60° or 90°
    Show answer

    Answer: A — Polysynthetic albite twinning in plagioclase is on the plane (010); B — The tartan (cross-hatched) twinning of microcline combines the albite and pericline laws; D — Staurolite cruciform twins are penetration twins whose two individuals cross at about 60° or 90°

    Albite twinning is on (010), microcline's tartan pattern is the albite plus pericline combination, and staurolite crosses form at about 60° or 90°, so those three are correct. A twin operation is by definition not part of the crystal's own point group; if it were, it would be a symmetry of the single crystal and produce no twin.
  8. A rock-forming silicate has double chains of tetrahedra with Si : O = 4 : 11 and two cleavages meeting at about 56° and 124°. It belongs to the:

    1. pyroxene group, whose single chains of tetrahedra give Si : O = 1 : 3 and two cleavages near 90°
    2. mica group, whose tetrahedral sheets give Si : O = 2 : 5 and a single perfect cleavage
    3. amphibole group, whose double chains give Si : O = 4 : 11 and cleavages at about 56° and 124°
    4. olivine group, whose isolated tetrahedra give Si : O = 1 : 4 and no good cleavage
    Show answer

    Answer: C — amphibole group, whose double chains give Si : O = 4 : 11 and cleavages at about 56° and 124°

    Double-chain silicates with Si : O = 4 : 11 are the amphiboles, whose cleavage angles are about 56° and 124°. Pyroxenes are the tempting alternative since both are chain silicates, but their single chains give 1 : 3 and cleavages near 90°. Micas are sheet silicates and olivine has isolated tetrahedra.
  9. In thin section a colourless mineral has high relief, no cleavage, irregular cracks, parallel extinction and interference colours of upper second to third order. It is most likely:

    1. olivine, which combines high relief and strong birefringence with irregular cracks instead of cleavage
    2. augite, which has high relief and second-order colours but is distinguished by two cleavages near 90° and inclined extinction
    3. garnet, which has very high relief but is isotropic and so stays dark between crossed polars
    4. quartz, which has no cleavage but has low relief and first-order grey colours
    Show answer

    Answer: A — olivine, which combines high relief and strong birefringence with irregular cracks instead of cleavage

    High relief with no cleavage, irregular cracks, parallel extinction and upper second- to third-order colours is the standard set for olivine. Augite would show two cleavages near 90° and inclined extinction, garnet would be isotropic and stay dark, and quartz has low relief and first-order grey, so each of those fails on at least one stated property.
  10. A uniaxial mineral has ω = 1.658 and ε = 1.486. Its optic sign and its birefringence are:

    1. positive, with δ = 0.172
    2. negative, with δ = 3.144, the sum of the two indices
    3. biaxial negative, with δ = 0.172
    4. negative, with δ = 0.172
    Show answer

    Answer: D — negative, with δ = 0.172

    The sign is negative because ε < ω, and the birefringence is the difference |ε − ω| = 1.658 − 1.486 = 0.172, the very high value of calcite. A positive sign would need ε > ω, the birefringence is a difference rather than a sum, and the stem already states the mineral is uniaxial.
  11. A mineral grain in a thin section 25 μm thick shows a retardation of 600 nm. Its birefringence, to three decimal places, is ____.

    Numerical answer — type the value.

    Show answer

    Answer: 0.024

    Retardation Γ = t δ, so δ = Γ/t = 600 nm / 25 000 nm = 0.024. Forgetting to convert 25 μm to 25 000 nm gives 24, which is the usual slip.
  12. Which of the following statements about optical properties are correct?

    1. Isotropic minerals such as garnet remain dark in every position between crossed polars
    2. Tetragonal, hexagonal and trigonal minerals are optically uniaxial
    3. Quartz is optically negative because its extraordinary index is lower than its ordinary index
    4. Pleochroism is observed in plane-polarised light, with the analyser out of the beam
    Show answer

    Answer: A — Isotropic minerals such as garnet remain dark in every position between crossed polars; B — Tetragonal, hexagonal and trigonal minerals are optically uniaxial; D — Pleochroism is observed in plane-polarised light, with the analyser out of the beam

    Isotropic minerals stay extinct, the three lower-symmetry non-cubic systems named are uniaxial, and pleochroism needs only polarised light with no analyser, so those three are correct. Quartz has ε (1.553) greater than ω (1.544), so it is optically positive, not negative.
  13. In classical nucleation theory a crystal nucleus has interfacial energy γ = 250 mJ/m² and the volume free-energy change on crystallising is ΔGv = −4 × 10⁸ J/m³. The critical radius r* = −2γ/ΔGv, in nm to two decimal places, is ____.

    Numerical answer — type the value.

    Show answer

    Answer: 1.25

    Convert γ to SI: 250 mJ/m² = 0.25 J/m². Then r* = 2 × 0.25 / (4 × 10⁸) = 1.25 × 10⁻⁹ m = 1.25 nm. Using γ = 250 without the conversion gives 1250 nm, a thousand times too large, which is the usual slip.
  14. In classical nucleation theory the barrier is ΔG* = 16π γ³ / (3 ΔGv²). Take γ as constant and |ΔGv| as proportional to the undercooling ΔT. If the undercooling doubles, the homogeneous nucleation barrier changes by a factor of:

    1. 2, because a larger undercooling raises the driving force and so raises the barrier in the same proportion
    2. 1/4, because ΔG* varies as 1/ΔGv², and ΔGv doubles
    3. 1/2, because the barrier varies inversely with the driving force ΔGv
    4. 1/8, because the critical radius falls by half and the barrier scales with the cube of the radius
    Show answer

    Answer: B — 1/4, because ΔG* varies as 1/ΔGv², and ΔGv doubles

    Since ΔG* ∝ 1/ΔGv² and ΔGv doubles, the barrier falls to 1/4. The factor 1/2 forgets the square, which is the tempting answer. The 1/8 argument is wrong because ΔG* = (4π/3) γ r², which varies as the square of r, not the cube (halving r* gives 1/4), and a larger undercooling lowers the barrier, it does not raise it.
  15. An orthopyroxene analysis gives SiO₂ 57.0, MgO 40.0 and FeO 3.0 wt% (take molecular weights SiO₂ 60.08, MgO 40.30, FeO 71.85). Recalculated on a basis of 6 oxygens, the number of Si atoms per formula unit, to two decimal places, is ____.

    Numerical answer — type the value.

    Show answer

    Answer: 1.94

    Moles: SiO₂ 57.0/60.08 = 0.9487, MgO 40.0/40.30 = 0.9926, FeO 3.0/71.85 = 0.0418. Oxygen proportions: 2 × 0.9487 + 0.9926 + 0.0418 = 2.9318. Factor = 6/2.9318 = 2.0465, so Si = 0.9487 × 2.0465 = 1.94 (with Mg 2.03 and Fe 0.09, cation total 4.06). Forgetting the factor of 2 oxygens per SiO₂ gives a factor of about 3.03 and Si near 2.87.
  16. An olivine contains 38.0 wt% MgO and 20.0 wt% FeO (molecular weights MgO 40.30, FeO 71.85; ignore any other Mg- or Fe-bearing component). Its forsterite content Fo = 100 × Mg/(Mg + Fe) in mole terms, to the nearest whole number, is ____.

    Numerical answer — type the value.

    Show answer

    Answer: 77

    Moles: MgO 38.0/40.30 = 0.9429 and FeO 20.0/71.85 = 0.2784. Fo = 100 × 0.9429 / (0.9429 + 0.2784) = 77.2, so 77. Using the weight ratio 38/(38 + 20) = 65.5 is the usual slip, because end-member content is a mole ratio.
  17. A basaltic magma crystallises and removes a large amount of plagioclase. The chondrite-normalised REE pattern of the residual melt will show:

    1. a negative Eu anomaly, because Eu²⁺ substitutes for Ca and Sr in plagioclase and is removed with it
    2. a positive Eu anomaly, because plagioclase rejects Eu²⁺, which therefore stays behind in the melt
    3. no Eu anomaly, because all the REE are trivalent and so behave identically during crystallisation
    4. a depletion of the heavy REE, because plagioclase has very high partition coefficients for Yb and Lu
    Show answer

    Answer: A — a negative Eu anomaly, because Eu²⁺ substitutes for Ca and Sr in plagioclase and is removed with it

    In reduced magmas part of the europium is Eu²⁺, which has a radius and charge close to Ca²⁺ and Sr²⁺ and enters plagioclase, so removing plagioclase depletes the residual melt in Eu relative to its neighbours: a negative Eu anomaly. Europium does not behave like the other, strictly trivalent, REE, which rules out the no-anomaly answer. Plagioclase does not strongly retain the heavy REE; garnet does.
  18. Which of the following statements about mineral chemistry and cation recalculation are correct?

    1. A compatible element has a partition coefficient D greater than 1 and is retained in the crystallising solid
    2. The forsterite content of olivine is the mole ratio 100 × Mg/(Mg + Fe), not a weight ratio
    3. Olivine analyses are normally recalculated on a basis of 6 oxygens per formula unit
    4. A cation total far from the ideal for the chosen oxygen basis signals a poor analysis, an unanalysed element or ferric iron
    Show answer

    Answer: A — A compatible element has a partition coefficient D greater than 1 and is retained in the crystallising solid; B — The forsterite content of olivine is the mole ratio 100 × Mg/(Mg + Fe), not a weight ratio; D — A cation total far from the ideal for the chosen oxygen basis signals a poor analysis, an unanalysed element or ferric iron

    Compatible elements have D above 1, end-member contents are mole ratios, and a cation total that misses the ideal value is a standard quality check, so those three are correct. Olivine, (Mg,Fe)₂SiO₄, is recalculated on 4 oxygens; 6 oxygens is the pyroxene basis.