Crystal Field Theory, Mineral Spectroscopy and Lattice Defects
CSIR NET Earth Sciences Part C geology: crystal field splitting, OSPE and site preference, mineral colour, Mössbauer, IR, Raman, point defects, twins.
- 20 min
On this page · 13 sections
Prerequisites
- d-orbital splitting, CFSE and high-spin and low-spin configurations as chemists learn them: Crystal Field, Ligand Field and MO Theory.
- Site symmetry and lattice geometry: Crystal Symmetry, Reciprocal Lattice and Diffraction.
- Compatible and incompatible elements and the composition of the mantle: Meteorites, Earth’s Composition and Differentiation.
The Hook: Why Ruby Is Red and Emerald Is Green
Crystal Field Theory in Minerals
Splitting of the d orbitals
In a free ion the five 3d orbitals are degenerate. Surrounded by anions (O\(^{2-}\) in silicates and oxides), orbitals pointing at the anions are raised in energy and those pointing between them are lowered.
| Site | Lower set | Upper set | Splitting | Barycentre rule |
|---|---|---|---|---|
| Octahedral (6-fold) | \( t_{2g} \) (\( d_{xy}, d_{yz}, d_{xz} \)) | \( e_g \) (\( d_{z^2}, d_{x^2-y^2} \)) | \( \Delta_o = 10Dq \) | \( t_{2g} \) at \( -0.4\Delta_o \), \( e_g \) at \( +0.6\Delta_o \) |
| Tetrahedral (4-fold) | \( e \) | \( t_2 \) | \( \Delta_t \approx \tfrac{4}{9}\Delta_o \) | \( e \) at \( -0.6\Delta_t \), \( t_2 \) at \( +0.4\Delta_t \) |
| Cubic (8-fold) | \( e \) | \( t_2 \) | \( \Delta_c \approx \tfrac{8}{9}\Delta_o \) | as tetrahedral |
\( \Delta \) grows with cation charge (\( \Delta \) for M\(^{3+}\) is larger than for M\(^{2+}\)), with shorter metal–oxygen distance (roughly as \( R^{-5} \) in the point-charge model), and with pressure, which shortens bonds.
Mineral sites are rarely perfect octahedra. The M1 and M2 sites of olivine and pyroxene are distorted, which splits the levels further and produces several absorption bands where an ideal octahedron would give one.
Crystal field stabilisation energy
CFSE is the net lowering of energy of the d electrons relative to the barycentre. In oxygen-coordinated minerals, Fe, Mn, Co and Ni are almost always high-spin at crustal and upper-mantle conditions (oxygen is a weak-field ligand). Pressure increases \( \Delta \), and spin transitions of Fe in lower-mantle minerals (ferropericlase, bridgmanite) are an active research area.
| Ion | d\(^n\) | Octahedral (HS) configuration | CFSE (oct) | CFSE (tet) |
|---|---|---|---|---|
| Ti\(^{3+}\) | d\(^1\) | \( t_{2g}^1 \) | \( -0.4\Delta_o \) | \( -0.6\Delta_t \) |
| V\(^{3+}\) | d\(^2\) | \( t_{2g}^2 \) | \( -0.8\Delta_o \) | \( -1.2\Delta_t \) |
| Cr\(^{3+}\) | d\(^3\) | \( t_{2g}^3 \) | \( -1.2\Delta_o \) | \( -0.8\Delta_t \) |
| Mn\(^{3+}\), Cr\(^{2+}\) | d\(^4\) | \( t_{2g}^3 e_g^1 \) | \( -0.6\Delta_o \) | \( -0.4\Delta_t \) |
| Mn\(^{2+}\), Fe\(^{3+}\) | d\(^5\) | \( t_{2g}^3 e_g^2 \) | 0 | 0 |
| Fe\(^{2+}\) | d\(^6\) | \( t_{2g}^4 e_g^2 \) | \( -0.4\Delta_o \) | \( -0.6\Delta_t \) |
| Co\(^{2+}\) | d\(^7\) | \( t_{2g}^5 e_g^2 \) | \( -0.8\Delta_o \) | \( -1.2\Delta_t \) |
| Ni\(^{2+}\) | d\(^8\) | \( t_{2g}^6 e_g^2 \) | \( -1.2\Delta_o \) | \( -0.8\Delta_t \) |
| Cu\(^{2+}\) | d\(^9\) | \( t_{2g}^6 e_g^3 \) | \( -0.6\Delta_o \) | \( -0.4\Delta_t \) |
| Zn\(^{2+}\) | d\(^{10}\) | full | 0 | 0 |
Octahedral site preference energy
$$\boxed{\text{OSPE} = |\text{CFSE}_{\text{oct}}| - |\text{CFSE}_{\text{tet}}|}$$Using \( \Delta_t = \tfrac{4}{9}\Delta_o \) and the same \( \Delta_o \) for each ion (a simplification for comparing patterns):
| d\(^n\) | CFSE oct | CFSE tet | OSPE |
|---|---|---|---|
| d\(^3\) (Cr\(^{3+}\)) | \( 1.2\Delta_o \) | \( 0.356\Delta_o \) | \( 0.844\Delta_o \) |
| d\(^8\) (Ni\(^{2+}\)) | \( 1.2\Delta_o \) | \( 0.356\Delta_o \) | \( 0.844\Delta_o \) |
| d\(^4\), d\(^9\) | \( 0.6\Delta_o \) | \( 0.178\Delta_o \) | \( 0.422\Delta_o \) |
| d\(^7\) (Co\(^{2+}\)) | \( 0.8\Delta_o \) | \( 0.533\Delta_o \) | \( 0.267\Delta_o \) |
| d\(^6\) (Fe\(^{2+}\)) | \( 0.4\Delta_o \) | \( 0.267\Delta_o \) | \( 0.133\Delta_o \) |
| d\(^5\), d\(^{10}\) | 0 | 0 | 0 |
Real values also scale with each ion’s own \( \Delta_o \), and Cr\(^{3+}\), being trivalent, has a much larger \( \Delta_o \) than Ni\(^{2+}\). The pattern still holds: Cr\(^{3+}\) and Ni\(^{2+}\) have the strongest preference for octahedral sites; Mn\(^{2+}\), Fe\(^{3+}\) and Zn\(^{2+}\) have none.
Geological consequences
| Observation | Crystal field explanation |
|---|---|
| Ni is strongly compatible in olivine; Ni falls steeply in a fractionating basalt | high OSPE of Ni\(^{2+}\); olivine’s M sites are octahedral, while silicate melt offers many lower-coordinated sites |
| Cr is concentrated in spinel (chromite), clinopyroxene and garnet; chromitite layers form early in layered intrusions | very high OSPE of Cr\(^{3+}\) |
| Magnetite is an inverse spinel, Fe\(^{3+}\)[Fe\(^{2+}\)Fe\(^{3+}\)]O\(_4\) | Fe\(^{3+}\) (d\(^5\)) has no preference and fills the tetrahedral site; Fe\(^{2+}\) gains some CFSE in the octahedral site |
| Chromite FeCr\(_2\)O\(_4\) is a normal spinel | Cr\(^{3+}\) insists on octahedral sites |
| Mn\(^{2+}\) and Fe\(^{3+}\) enter tetrahedral or octahedral sites freely | zero CFSE for d\(^5\) |
| Fe\(^{2+}\) prefers the more distorted M2 site of orthopyroxene over M1 | extra stabilisation in the distorted site |
| Mn\(^{3+}\) and Cu\(^{2+}\) prefer distorted sites | Jahn–Teller effect for d\(^4\) (HS) and d\(^9\) |
Colour and Optical Absorption
| Cause | Mechanism | Examples |
|---|---|---|
| Crystal field (d–d) transitions | electron promoted between split d levels; spin-allowed bands are moderately intense | ruby and emerald (Cr\(^{3+}\)); peridot (Fe\(^{2+}\) in olivine); green chrome diopside |
| Metal-to-metal (intervalence) charge transfer | electron hops between neighbouring cations; intense, broad bands | blue sapphire (Fe\(^{2+}\) \( \to \) Ti\(^{4+}\)); Fe\(^{2+}\) \( \to \) Fe\(^{3+}\) in blue amphiboles, cordierite, vivianite |
| Oxygen-to-metal charge transfer | O\(^{2-}\) \( \to \) Fe\(^{3+}\) in the UV, tailing into the visible | yellow to red-brown in Fe\(^{3+}\)-bearing minerals |
| Colour centres | electron or hole trapped at a defect, often after natural irradiation | purple fluorite; smoky quartz (Al-associated hole centre); amethyst (Fe plus irradiation) |
| Band gap | semiconductor absorbs photons above its gap | cinnabar (red), sphalerite (yellow to colourless when Fe-poor) |
| Physical (optics) | interference, diffraction, scattering | labradorescence, precious opal, moonstone adularescence |
Ruby versus emerald. In ruby the Cr–O distances in corundum are short, \( \Delta \) is larger, and the main spin-allowed band sits in the yellow-green; red light is transmitted. In beryl the Cr site has a weaker field, the band shifts to lower energy (towards yellow-orange) and the transmission window moves to green. Alexandrite (Cr-bearing chrysoberyl) sits in between, so it looks green in daylight and red in incandescent light.
Idiochromatic minerals owe their colour to an essential constituent (malachite, azurite, rhodochrosite); allochromatic minerals to a trace impurity or defect (ruby, emerald, amethyst). Pleochroism in thin section (biotite, hornblende, tourmaline) arises because the absorbing transitions depend on the polarisation direction relative to distorted, oriented sites.
Remote sensing. Fe\(^{2+}\) in octahedral sites gives broad crystal field bands in the near infrared: olivine a composite band near 1 μm; pyroxenes two bands near 1 μm and 2 μm whose positions shift with Ca and Fe content. These bands are how olivine and pyroxene are mapped on the Moon, Mars and asteroids.
To convert band position: \( \lambda\,(\text{nm}) = 10^7/\tilde{\nu}\,(\text{cm}^{-1}) \). A band at 18,000 cm\(^{-1}\) is at 556 nm; at 10,000 cm\(^{-1}\) it is at 1,000 nm (1 μm). And 1 cm\(^{-1}\) \( \equiv \) 0.01196 kJ mol\(^{-1}\).
Mineral Spectroscopy: What Each Method Sees
| Technique | Probes | Mineralogical use |
|---|---|---|
| Optical (UV–Vis–NIR) absorption | electronic transitions: d–d, charge transfer | colour, oxidation state, site of transition metals, remote sensing |
| Infrared (IR) | vibrations that change the dipole moment | OH and H\(_2\)O (O–H stretch about 3,000–3,700 cm\(^{-1}\)); water in nominally anhydrous minerals; Si–O stretching about 800–1,200 cm\(^{-1}\); carbonate and sulphate groups |
| Raman | vibrations that change polarisability | fast identification of polymorphs and inclusions: quartz main band near 464 cm\(^{-1}\), coesite near 521, diamond 1,332, graphite G band near 1,580 cm\(^{-1}\) |
| Mössbauer (\( ^{57}\)Fe) | nuclear transitions of \( ^{57}\)Fe (14.4 keV gamma ray) | Fe\(^{2+}\)/Fe\(^{3+}\) ratio, coordination, site occupancy, magnetic ordering |
| EPR (ESR) | unpaired electrons | Mn\(^{2+}\), Fe\(^{3+}\), radiation-induced centres; ESR dating |
| NMR (MAS) | nuclear spin environment (\( ^{29}\)Si, \( ^{27}\)Al, \( ^{1}\)H) | Si polymerisation (Q\(^n\) species), Al coordination (4, 5, 6), Al–Si order in feldspars, glass structure |
| XANES / EXAFS (synchrotron) | X-ray absorption edges | oxidation state and local coordination of a chosen element, even at low concentration |
| Cathodoluminescence | emission under an electron beam | growth zoning in zircon, quartz, carbonates; guides spot dating |
Mössbauer in more detail
A \( ^{57}\)Fe Mössbauer spectrum of a paramagnetic silicate is a set of doublets, one for each kind of Fe site.
- Isomer shift (centre of the doublet) reflects s-electron density at the nucleus: larger for high-spin Fe\(^{2+}\) (roughly 1.0–1.3 mm s\(^{-1}\) relative to α-Fe) than for Fe\(^{3+}\) (roughly 0.2–0.5 mm s\(^{-1}\)). It identifies oxidation state, and within a valence it shifts with coordination.
- Quadrupole splitting (separation of the two lines) reflects the asymmetry of the electric field at the nucleus. High-spin Fe\(^{2+}\) usually shows large splitting (one extra electron outside a half-filled shell); high-spin Fe\(^{3+}\) (spherical d\(^5\)) shows small splitting.
- Area of each doublet is roughly proportional to the amount of Fe in that site (assuming equal recoil-free fractions), so Fe\(^{3+}\)/ΣFe can be read from area ratios.
- Magnetically ordered phases (magnetite, hematite) give sextets from hyperfine magnetic splitting.
\( ^{29}\)Si MAS NMR and polymerisation
Silicate tetrahedra are labelled \( Q^n \), where \( n \) is the number of bridging oxygens (0 to 4). The \( ^{29}\)Si chemical shift becomes more negative as \( n \) increases, roughly from about −60 to −80 ppm for \( Q^0 \) (olivine) to about −100 to −120 ppm for \( Q^4 \) (quartz). Substituting Al for Si in neighbouring tetrahedra shifts the signal back to less negative values, which is how Al–Si ordering is studied.
graph TD
Q[What do you need to know?] --> A[Fe2+ vs Fe3+ in a silicate]
Q --> B[Which SiO2 polymorph is in an inclusion?]
Q --> C[How much OH is in olivine?]
Q --> D[Al coordination or Al-Si order]
Q --> E[Growth zoning in zircon]
A --> A1[Mossbauer; XANES; EPMA flank method]
B --> B1[Raman micro-spectroscopy]
C --> C1[FTIR, polarised]
D --> D1[27Al and 29Si MAS NMR]
E --> E1[Cathodoluminescence imaging]Lattice Defects
Real crystals depart from perfect periodicity. Defects are classified by dimension.
Point defects (zero-dimensional)
| Defect | Description | Charge balance | Example |
|---|---|---|---|
| Schottky | a cation vacancy and an anion vacancy, in stoichiometric ratio | vacancies in the ratio of the formula | NaCl, KCl, MgO |
| Frenkel | an ion leaves its site for an interstitial position | vacancy and interstitial of the same ion | AgCl (cation); CaF\(_2\) (anion Frenkel) |
| Substitutional impurity | foreign ion on a regular site | may need a coupled substitution or vacancy | Cr\(^{3+}\) for Al\(^{3+}\) in corundum |
| Interstitial impurity | foreign atom in an interstice | H\(^+\) in olivine, Na in beryl channels | |
| Electronic defects | electrons and holes, small polarons (Fe\(^{3+}\) on an Fe\(^{2+}\) site) | conduction in olivine and the upper mantle | |
| Colour centres | electron trapped at an anion vacancy (F-centre) or hole trapped near an impurity | purple fluorite, smoky quartz |
Point defects exist at equilibrium because they raise configurational entropy. For a Schottky pair with formation energy \( E_s \),
$$\boxed{\frac{n_s}{N} \approx \exp\!\left(-\frac{E_s}{2kT}\right)}$$and for Frenkel defects \( n_f \approx \sqrt{N N_i}\,\exp(-E_f/2kT) \), where \( N_i \) is the number of interstitial sites. The factor 2 appears because each event creates two defects. Concentrations rise steeply with temperature.
Why they matter:
- Diffusion in minerals proceeds mainly by vacancy and interstitial mechanisms, so diffusion rates (and hence closure temperatures in geochronology and the zoning of garnet) depend on defect concentration.
- Electrical conductivity of the mantle reflects defects in olivine: small polarons, Mg vacancies and hydrogen. Magnetotelluric surveys use this to infer temperature and water content.
- Non-stoichiometry: pyrrhotite Fe\(_{1-x}\)S has ordered Fe vacancies; wüstite is always Fe\(_{1-x}\)O.
- Trace-element uptake often needs a vacancy or a coupled substitution to keep charge balance.
Line defects: dislocations
A dislocation is a line that bounds a region of the crystal that has slipped by one lattice vector, the Burgers vector \( \mathbf{b} \).
| Type | Relation of \( \mathbf{b} \) to the line | Motion |
|---|---|---|
| Edge | \( \mathbf{b} \perp \) line (extra half-plane) | glides on its slip plane; climbs out of it by vacancy diffusion |
| Screw | \( \mathbf{b} \parallel \) line | can cross-slip onto another plane |
| Mixed | in between |
Dislocation glide and climb are the basis of dislocation creep, the main mechanism of high-stress mantle flow, which produces lattice-preferred orientation of olivine and hence seismic anisotropy. In dry olivine at high temperature the dominant slip system is
\[100\](010). Dislocations can be seen by TEM, or by decoration (oxidising olivine makes them visible in the optical microscope). Low-angle (subgrain) boundaries are walls of dislocations formed by recovery.
Planar defects
| Defect | Description | Mineral example |
|---|---|---|
| Grain boundary | interface between crystals of different orientation | all polycrystalline rocks; 120° triple junctions after annealing |
| Twin boundary | interface between two related orientations of the same crystal | see twinning below |
| Stacking fault | error in a layer sequence | micas, chlorite, ZnS; source of polytypes |
| Antiphase boundary | interface where the order of Al and Si (or cations) is out of step | pigeonite, omphacite, anorthite (antiphase domains) |
| Exsolution lamella interface | coherent or incoherent boundary between host and lamella | perthite, pyroxene lamellae |
Twinning
A twin is a symmetrical intergrowth of two or more individuals of the same mineral related by a symmetry operation (the twin law) that is not a symmetry of the crystal itself: reflection across a twin plane, rotation about a twin axis, or inversion.
| Type | Character | Example |
|---|---|---|
| Contact twin | two individuals meet on a composition plane | spinel-law twins on \( \{111\} \); gypsum swallow-tail |
| Penetration twin | individuals interpenetrate | Carlsbad twin in orthoclase; staurolite crosses; fluorite |
| Polysynthetic (repeated) twin | many parallel lamellae on the same law | albite law in plagioclase (composition plane (010)) |
| Cyclic twin | several individuals around a common axis | chrysoberyl trillings; rutile |
By origin: growth twins (formed during growth: Carlsbad, spinel law), transformation twins (formed when a high-symmetry phase inverts to lower symmetry: the cross-hatched “tartan” twinning of microcline, combining albite and pericline laws, formed from monoclinic K-feldspar; Dauphiné twins in quartz formed on cooling through the β-to-α transition), and deformation (mechanical) twins (formed by stress: calcite e-twins, used as a palaeo-piezometer, and some plagioclase lamellae).
Quartz shows two important laws: Dauphiné twins (rotation of 180° about c, both parts of the same handedness) and Brazil twins (left- and right-handed parts related by reflection).
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| d–d (crystal field) band: weaker, on one ion | Charge-transfer band: intense, between ions or O–metal | |
| IR: dipole change; OH, H\(_2\)O, carbonate | Raman: polarisability change; polymorphs, inclusions | |
| Isomer shift: oxidation state | Quadrupole splitting: site distortion and electron asymmetry | |
| Schottky: paired vacancies, density falls | Frenkel: vacancy plus interstitial, density nearly unchanged | |
| Edge dislocation: \( \mathbf{b} \perp \) line | Screw dislocation: \( \mathbf{b} \parallel \) line | |
| Growth twin: formed during crystallisation | Transformation twin: formed on inversion to lower symmetry | |
| Dauphiné twin: same handedness | Brazil twin: opposite handedness | |
| Idiochromatic: colour from essential element | Allochromatic: colour from trace impurity or defect |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- Which two divalent or trivalent first-row transition-metal ions have the largest octahedral site preference in oxygen-coordinated minerals?
- What does the isomer shift in a \( ^{57}\)Fe Mössbauer spectrum mainly indicate?
- Name the defect in which an ion moves from its regular site into an interstitial site.
Answers
- Cr\(^{3+}\) (d\(^3\)) and Ni\(^{2+}\) (d\(^8\)).
- The oxidation state (s-electron density at the nucleus), with a smaller dependence on coordination.
- Frenkel defect.
Level 2: Part B Style
Practice Q1. During fractional crystallisation of a basaltic magma, Ni decreases much faster than Mn in the residual melt. The best explanation is:
(a) Ni\(^{2+}\) is larger than Mn\(^{2+}\) and is excluded from olivine (b) Ni\(^{2+}\) has a large octahedral site preference energy, whereas Mn\(^{2+}\) has none (c) Ni is chalcophile and is lost to a sulphide liquid, while Mn is siderophile (d) Mn\(^{2+}\) has a Jahn–Teller distortion that favours melt
Answer: (b)
Ni\(^{2+}\) (d\(^8\)) gains \( 1.2\Delta_o \) in an octahedral site and so is strongly taken up by olivine’s octahedral M sites. Mn\(^{2+}\) (d\(^5\) high-spin) gains nothing, so it partitions by size and charge only. (c) has some truth for sulphide-saturated magmas, but Mn is not siderophile and the question describes ordinary silicate fractionation. (d) Jahn–Teller applies to d\(^4\) and d\(^9\), not d\(^5\).
Practice Q2. Fe\(^{2+}\) occupies an octahedral site in a silicate with \( \Delta_o = 10{,}000 \) cm\(^{-1}\). Its CFSE is closest to:
(a) 24 kJ mol\(^{-1}\) (b) 48 kJ mol\(^{-1}\) (c) 72 kJ mol\(^{-1}\) (d) 120 kJ mol\(^{-1}\)
Answer: (b) about 48 kJ/mol
High-spin Fe\(^{2+}\), \( t_{2g}^4 e_g^2 \): CFSE \( = 4(-0.4) + 2(0.6) = -0.4\Delta_o = -4{,}000 \) cm\(^{-1}\). Converting: \( 4{,}000 \times 0.011963 = 47.9 \) kJ mol\(^{-1}\). Option (d) uses \( 1.0\Delta_o \); option (c) uses \( 0.6\Delta_o \).
Practice Q3. Which technique is best suited to distinguish coesite from quartz in a 10 μm inclusion inside a garnet?
(a) Powder X-ray diffraction (b) Mössbauer spectroscopy (c) Raman micro-spectroscopy (d) \( ^{29}\)Si MAS NMR
Answer: (c)
Raman with a focused laser can analyse a micrometre-sized inclusion in place through the host, and the two polymorphs have different main bands (quartz near 464, coesite near 521 cm\(^{-1}\)). Powder XRD and NMR need the phase separated in bulk; Mössbauer needs iron.
Practice Q4. In a crystal, the fraction of Schottky defects is \( \exp(-E_s/2kT) \). If \( E_s = 2.0 \) eV, by about what factor does the defect fraction increase between 1,000 K and 1,200 K? (\( k = 8.617\times 10^{-5} \) eV K\(^{-1}\))
(a) 1.2 (b) 2.6 (c) 6.9 (d) 48
Answer: (c) about 6.9
\( n/N \) at 1,000 K \( = \exp(-2/(2\times 8.617\times10^{-5}\times 1000)) = \exp(-11.60) = 9.1\times 10^{-6} \). At 1,200 K: \( \exp(-9.67) = 6.3\times 10^{-5} \). Ratio \( = 6.9 \). Option (d) is the result of dropping the factor 2 (\( 6.9^2 \approx 48 \)).
Level 3: Part C Style
Practice Q5. Using \( \Delta_t = \tfrac{4}{9}\Delta_o \) and assuming all ions are high-spin with the same \( \Delta_o \), rank the octahedral site preference of Fe\(^{2+}\), Co\(^{2+}\), Ni\(^{2+}\) and Mn\(^{2+}\).
(a) Ni > Co > Fe > Mn (b) Co > Ni > Fe > Mn (c) Ni > Fe > Co > Mn (d) Mn > Fe > Co > Ni
Answer: (a) Ni > Co > Fe > Mn
| Ion | oct CFSE | tet CFSE (\( \times\tfrac49 \)) | OSPE |
|---|---|---|---|
| Ni\(^{2+}\) d\(^8\) | \( 1.2\Delta_o \) | \( 0.8\Delta_t = 0.356\Delta_o \) | \( 0.844\Delta_o \) |
| Co\(^{2+}\) d\(^7\) | \( 0.8\Delta_o \) | \( 1.2\Delta_t = 0.533\Delta_o \) | \( 0.267\Delta_o \) |
| Fe\(^{2+}\) d\(^6\) | \( 0.4\Delta_o \) | \( 0.6\Delta_t = 0.267\Delta_o \) | \( 0.133\Delta_o \) |
| Mn\(^{2+}\) d\(^5\) | 0 | 0 | 0 |
Note that Co\(^{2+}\) gains more CFSE in a tetrahedral site (\( 1.2\Delta_t \)) than any other divalent ion in this list, which is why Co\(^{2+}\) is common in tetrahedral sites (blue cobalt glasses and spinels), yet its OSPE is still positive.
Practice Q6. A Mössbauer spectrum of a clinopyroxene shows two doublets:
| Doublet | Isomer shift (mm s\(^{-1}\), rel. α-Fe) | Quadrupole splitting (mm s\(^{-1}\)) | Area (%) |
|---|---|---|---|
| A | 1.15 | 2.10 | 62 |
| B | 0.38 | 0.75 | 38 |
Which statements are correct?
- Doublet A is high-spin Fe\(^{2+}\) and doublet B is Fe\(^{3+}\).
- Fe\(^{3+}\)/ΣFe is about 0.38 if recoil-free fractions are equal.
- If total FeO\(_\text{T}\) by microprobe is 10.0 wt%, the Fe\(_2\)O\(_3\) content is about 4.2 wt%.
- The absence of a sextet shows that no magnetite inclusions large enough to be magnetically ordered are present in significant amounts.
(a) 1 and 2 only (b) 1, 2 and 3 only (c) 1, 2 and 4 only (d) 1, 2, 3 and 4
Answer: (d) all four
- 1 true: large shift and large splitting are typical of Fe\(^{2+}\); small shift and splitting of Fe\(^{3+}\).
- 2 true: area ratio \( 38/(62 + 38) = 0.38 \).
- 3 true: Fe\(^{3+}\) as FeO \( = 0.38 \times 10.0 = 3.80 \) wt%. Converting FeO to Fe\(_2\)O\(_3\) multiplies by \( M(\text{Fe}_2\text{O}_3)/(2M(\text{FeO})) = 159.69/143.69 = 1.111 \): \( 3.80 \times 1.111 = 4.22 \) wt%. (FeO then is \( 6.20 \) wt%.)
- 4 true: magnetically ordered magnetite at room temperature gives six-line patterns; none are seen, so any magnetite is below detection.
Practice Q7. A colourless quartz crystal turns smoky brown after gamma irradiation, and the colour disappears on heating to a few hundred degrees Celsius. A second, nearly Al-free synthetic quartz shows no colour after the same dose. What is the best interpretation?
(a) Irradiation oxidised Fe\(^{2+}\) to Fe\(^{3+}\), creating an O \( \to \) Fe\(^{3+}\) charge-transfer band (b) Irradiation created a hole trapped at an oxygen next to Al\(^{3+}\) substituting for Si\(^{4+}\) (a colour centre), which heating anneals (c) Irradiation produced Schottky defects that scatter light (d) Irradiation converted some quartz into coesite
Answer: (b)
The colour needs Al (absent in the second sample), forms on irradiation and bleaches on mild heating: the signature of a colour centre. The Al\(^{3+}\) for Si\(^{4+}\) substitution, charge-balanced by an alkali or H\(^+\), is the precursor; irradiation removes an electron from a neighbouring oxygen and the compensating ion moves away. (d) is impossible at low pressure; (c) would not give a specific absorption band; (a) cannot explain the Al dependence.
Quick Revision Table
| Idea | Fact |
|---|---|
| Octahedral split | \( t_{2g} \) at \( -0.4\Delta_o \), \( e_g \) at \( +0.6\Delta_o \) |
| Tetrahedral split | \( e \) at \( -0.6\Delta_t \), \( t_2 \) at \( +0.4\Delta_t \); \( \Delta_t = \tfrac49\Delta_o \) |
| Largest OSPE | Cr\(^{3+}\), Ni\(^{2+}\) (\( 0.844\Delta_o \)); zero for d\(^5\), d\(^{10}\) |
| Spinels | magnetite inverse; chromite normal |
| Jahn–Teller | d\(^4\) HS (Mn\(^{3+}\)), d\(^9\) (Cu\(^{2+}\)) |
| Ruby / emerald | Cr\(^{3+}\), strong / weaker field |
| Blue sapphire | Fe\(^{2+}\)–Ti\(^{4+}\) charge transfer |
| Fe\(^{2+}\) NIR bands | olivine about 1 μm; pyroxene about 1 and 2 μm |
| Mössbauer | Fe\(^{2+}\): larger isomer shift and quadrupole splitting; areas give Fe\(^{3+}\)/ΣFe |
| Raman | quartz 464, coesite 521, diamond 1,332, graphite 1,580 cm\(^{-1}\) |
| \( ^{29}\)Si NMR | more negative shift with higher \( Q^n \) |
| Schottky | \( n/N = e^{-E_s/2kT} \); paired vacancies (NaCl, MgO) |
| Frenkel | vacancy plus interstitial (AgCl; anion in CaF\(_2\)) |
| Dislocations | edge \( \mathbf{b}\perp \) line; screw \( \mathbf{b}\parallel \) line; olivine \[100\](010) |
| Twins | growth (Carlsbad, spinel), transformation (microcline tartan, Dauphiné), deformation (calcite e-twins) |
Key Takeaways
What’s Next
- Mineral Structures, Polymorphism and Solid Solution: the structures in which these ions sit, and how phases transform and unmix.
- Magma Genesis, Phase Equilibria and Magma Mixing: Ni and Cr partitioning at work in fractionating magmas.
- Electronic Spectra of Complexes in Chemical Sciences, for term symbols and Tanabe–Sugano diagrams.
- Mineralogy and Petrology: Quick Revision.