Mössbauer and NQR Spectroscopy
CSIR NET chemistry: recoil-free γ resonance, isomer shift, quadrupole and magnetic splitting in ⁵⁷Fe and ¹¹⁹Sn; NQR frequencies, η and ionic character.
- 16 min
On this page · 18 sections
- Prerequisites
- The Hook: Which Iron Is Which in Prussian Blue?
- The Mössbauer Effect
- The Experiment
- Isomer Shift (δ)
- Quadrupole Splitting (ΔE_Q)
- Magnetic Hyperfine Splitting
- NQR: Principle
- NQR Energy Levels and Frequencies
- What NQR Frequencies Tell You
- Mössbauer and NQR Side by Side
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- High-spin and low-spin configurations of Fe(II) and Fe(III): Crystal Field, Ligand Field and MO Theory.
- Nuclear spin, quadrupolar nuclei and their effect on NMR: NMR Spectroscopy of Inorganic Compounds.
- Penetration and shielding of s electrons: Electronic Configuration and Effective Nuclear Charge.
- Tin(II) lone pair and the inert-pair effect: Group 14 and 15 Compounds.
The Hook: Which Iron Is Which in Prussian Blue?
The Mössbauer Effect
A nucleus emitting a γ ray of energy \( E_\gamma \) recoils with energy
$$E_R = \frac{E_\gamma^2}{2Mc^2}$$For a free ⁵⁷Fe atom (\( E_\gamma = 14.4 \) keV) this is about \( 2 \times 10^{-3} \) eV, roughly \( 10^5 \) times the natural line width (\( \Gamma \approx 4.7 \times 10^{-9} \) eV). A γ ray emitted by a free atom therefore cannot be absorbed by another free atom.
In a solid, a fraction f of events occurs with the whole lattice taking up the recoil momentum (and no phonon created). These recoil-free events emit and absorb at exactly the nuclear transition energy. f (the Lamb–Mössbauer factor) is larger for low γ energies, stiff lattices and low temperatures. That is why Mössbauer spectra are recorded on solids or frozen solutions, often at low temperature.
The Experiment
- Source: for ⁵⁷Fe, ⁵⁷Co in a metal matrix (⁵⁷Co decays by electron capture to excited ⁵⁷Fe, which emits the 14.4 keV γ ray). For ¹¹⁹Sn, a ¹¹⁹ᵐSn source (23.9 keV).
- Energy scan: the source moves at a velocity v relative to the absorber; the Doppler shift \( \Delta E = E_\gamma v/c \) sweeps the energy. Spectra are plotted against velocity in mm s⁻¹. For ⁵⁷Fe, 1 mm s⁻¹ ≡ 4.8 × 10⁻⁸ eV.
- Detector behind the absorber records transmission: resonance appears as a dip.
- Nuclear spins (⁵⁷Fe and ¹¹⁹Sn): ground state I = 1/2, excited state I = 3/2.
| Isotope | \( E_\gamma \) (keV) | Typical use |
|---|---|---|
| ⁵⁷Fe | 14.4 | iron oxidation and spin states, minerals, proteins |
| ¹¹⁹Sn | 23.9 | Sn(II) versus Sn(IV), organotins |
| ¹²⁹I, ¹²¹Sb, ¹⁹⁷Au, ¹⁵¹Eu, ⁹⁹Ru | various | specialised studies |
Isomer Shift (δ)
The nucleus has a slightly different radius in its excited and ground states. Its energy therefore depends on the electron density at the nucleus, \( |\psi(0)|^2 \), which comes only from s electrons (and, slightly, \( p_{1/2} \)). If source and absorber differ in that density, the resonance is shifted:
$$\delta = \text{const} \times \left(|\psi(0)|^2_{\text{absorber}} - |\psi(0)|^2_{\text{source}}\right)\left(R_e^2 - R_g^2\right)$$δ is quoted relative to a standard (for ⁵⁷Fe usually α-iron at room temperature; for ¹¹⁹Sn, SnO₂ or BaSnO₃).
For ⁵⁷Fe, \( R_e < R_g \), so the factor \( (R_e^2 - R_g^2) \) is negative: more s-electron density at the nucleus → more negative δ.
How d electrons enter: d electrons shield the s electrons from the nucleus, expanding the s orbitals and reducing \( |\psi(0)|^2 \). Removing a d electron (Fe²⁺ → Fe³⁺) increases s density and lowers δ. Covalency and back-bonding (low-spin, π-acceptor ligands such as CN⁻) also delocalise d density away and increase s population, lowering δ further.
| ⁵⁷Fe species | Approximate δ (mm s⁻¹, vs α-Fe) |
|---|---|
| High-spin Fe(II) | about +0.9 to +1.3 |
| High-spin Fe(III) | about +0.2 to +0.5 |
| Low-spin Fe(II) | about −0.1 to +0.4 |
| Low-spin Fe(III) | about −0.1 to +0.3 |
| Fe(IV), Fe(VI) | lower still (Fe(VI) in ferrate clearly negative) |
The clean, exam-proof result: high-spin Fe(II) has the largest isomer shift.
For ¹¹⁹Sn, \( R_e > R_g \), so the sign is reversed: more 5s density → more positive δ. Sn(II) compounds, with a 5s² lone pair, have large positive δ (relative to SnO₂); Sn(IV) compounds have δ near zero to moderately positive. Tin’s isomer shift therefore distinguishes Sn(II) from Sn(IV) at a glance.
Quadrupole Splitting (ΔE_Q)
A nucleus with I ≥ 1 has an electric quadrupole moment Q (non-spherical charge distribution). In an electric field gradient (EFG) its levels split. For ⁵⁷Fe the excited I = 3/2 level splits into \( m_I = \pm 3/2 \) and \( \pm 1/2 \); the ground I = 1/2 level does not split. The spectrum becomes a doublet, separation ΔE_Q, with the isomer shift at its midpoint.
The EFG has two sources:
- Valence (electronic) contribution: an asymmetric distribution of the metal’s own d electrons. Dominant when present.
- Lattice (ligand) contribution: an unsymmetrical arrangement of different ligands. Smaller.
| Fe species | d configuration | Valence EFG | Typical ΔE_Q |
|---|---|---|---|
| HS Fe(III) | \( t_{2g}^3 e_g^2 \), spherical \( ^6A_1 \) | none | small (from ligands only) |
| LS Fe(II) | \( t_{2g}^6 \), spherical in \( O_h \) | none | small (zero for \( [\text{Fe(CN)}_6]^{4-} \); ligand-induced in mixed complexes) |
| HS Fe(II) | \( t_{2g}^4 e_g^2 \): one extra \( t_{2g} \) electron | large | large, about 2–3.5 mm s⁻¹, temperature-dependent |
| LS Fe(III) | \( t_{2g}^5 \): one \( t_{2g} \) hole | moderate | moderate |
So the two parameters together nail the state:
graph TD
A["57Fe spectrum: read δ and ΔEQ"] --> B{"δ about 0.9–1.3 mm/s?"}
B -->|"yes, ΔEQ large"| C["high-spin Fe(II)"]
B -->|"no"| D{"ΔEQ small?"}
D -->|"yes, δ about 0.2–0.5"| E["high-spin Fe(III)"]
D -->|"yes, δ near 0"| F["low-spin Fe(II), symmetric"]
D -->|"no, moderate ΔEQ, δ near 0"| G["low-spin Fe(III) or LS Fe(II) with unlike ligands"]Magnetic Hyperfine Splitting
If the nucleus feels a magnetic field (an external field, or the internal field of a magnetically ordered solid), both nuclear levels split into their \( m_I \) sublevels: I = 3/2 into four, I = 1/2 into two. The selection rule \( \Delta m_I = 0, \pm 1 \) allows six transitions: a sextet. For a randomly oriented sample the line intensities are 3:2:1:1:2:3.
- α-Fe at room temperature: sextet corresponding to an internal field of about 33 T (used for velocity calibration).
- Magnetite \( \text{Fe}_3\text{O}_4 \) shows two overlapping sextets (tetrahedral and octahedral sites); above its low-temperature (Verwey) transition the octahedral site shows an averaged Fe(2.5+) because electrons hop faster than the Mössbauer time scale.
- Paramagnets with slow electron relaxation can also show magnetic splitting at low temperature.
Line width and time scale
The natural width of the 14.4 keV level (lifetime about 0.14 µs) is about 0.1 mm s⁻¹; the narrowest observed lines (source plus absorber) are about 0.2 mm s⁻¹. Processes faster than about \( 10^{-7} \) s (electron hopping, spin-crossover interconversion if fast) are averaged; slower ones show separate sub-spectra. Spin-crossover iron(II) complexes typically show separate HS and LS doublets whose areas change with temperature.
NQR: Principle
NQR is “NMR without a magnet”. A quadrupolar nucleus (I ≥ 1) in a solid sits in an EFG from its bonding electrons and neighbouring charges. The interaction splits the \( m_I \) levels even with no external field, and radio-frequency radiation (typically 1–1000 MHz) induces transitions between them.
- Only in solids (in liquids, tumbling averages the EFG to zero).
- No signal from nuclei in cubic environments (zero EFG): Cl⁻ in rock-salt NaCl gives no NQR line.
- Each crystallographically distinct site gives its own line(s), so the number of lines counts inequivalent sites.
- Frequencies usually fall as temperature rises, because thermal librations partly average the EFG.
Common NQR nuclei: ¹⁴N (I = 1), ³⁵Cl and ³⁷Cl (3/2), ⁷⁹Br and ⁸¹Br (3/2), ⁶³Cu (3/2), ⁷⁵As (3/2), ¹²⁷I (5/2), ²⁷Al (5/2), ¹²¹Sb (5/2), ²⁰⁹Bi (9/2).
NQR Energy Levels and Frequencies
The EFG is described by its largest component \( eq = V_{zz} \) and the asymmetry parameter \( \eta = (V_{xx} - V_{yy})/V_{zz} \) (0 ≤ η ≤ 1; η = 0 for axial symmetry). The quadrupole coupling constant is \( e^2Qq/h \) (in MHz). For an axial EFG:
$$E_m = \frac{e^2Qq\,[3m^2 - I(I+1)]}{4I(2I-1)}$$Levels depend on \( m^2 \), so ±m stay degenerate in zero field.
| I | Levels | Transitions (η = 0) | Note |
|---|---|---|---|
| 1 (¹⁴N) | m = 0, ±1 | \( \nu = \frac{3}{4}\frac{e^2Qq}{h} \) | for η ≠ 0, three lines: \( \nu_\pm = \frac{3}{4}\frac{e^2Qq}{h}\left(1 \pm \frac{\eta}{3}\right) \), \( \nu_0 = \frac{1}{2}\frac{e^2Qq}{h}\eta \) |
| 3/2 (³⁵Cl, ⁷⁹Br) | ±1/2, ±3/2 | one line, \( \nu = \frac{1}{2}\frac{e^2Qq}{h} \) | with η ≠ 0: \( \nu = \frac{1}{2}\frac{e^2Qq}{h}\sqrt{1 + \eta^2/3} \); one line cannot give both \( e^2Qq \) and η |
| 5/2 (¹²⁷I) | ±1/2, ±3/2, ±5/2 | two lines, \( \nu_1 = \frac{3}{20}\frac{e^2Qq}{h} \), \( \nu_2 = \frac{3}{10}\frac{e^2Qq}{h} \) | ratio exactly 2 when η = 0; departure from 2 gives η |
For I = 3/2 nuclei, η is obtained by adding a small magnetic field (Zeeman-perturbed NQR) or from single-crystal work.
Isotope check
The two chlorine isotopes have quadrupole moments in the ratio \( Q(^{35}\text{Cl})/Q(^{37}\text{Cl}) \approx 1.27 \). Two NQR lines in this frequency ratio, with intensities near 3:1, identify a chlorine site unambiguously; the same trick works for ⁷⁹Br/⁸¹Br.
What NQR Frequencies Tell You
Bond ionicity (Townes–Dailey picture). For a halogen, the EFG comes mainly from an unbalanced p shell. In a pure covalent σ bond (one p electron short in the bonding direction), the coupling constant approaches the free-atom value; as the bond becomes ionic, the p shell fills towards a spherical closed shell and the EFG falls. In the simplest form (neglecting hybridisation and π bonding):
$$\frac{(e^2Qq)_{\text{molecule}}}{(e^2Qq)_{\text{atom}}} \approx 1 - i$$where i is the ionic character of the bond. For ³⁵Cl, the free-atom value is about 110 MHz.
Other uses:
- Number and symmetry of sites in a crystal (e.g. distinguishing terminal and bridging halogens: bridging Cl in \( \text{Al}_2\text{Cl}_6 \)-type dimers gives lines at different frequencies from terminal Cl).
- Phase transitions (lines split or merge as symmetry changes).
- Detection of nitrogen-rich compounds (¹⁴N NQR) in solids without a magnet, which has been explored for screening explosives and drugs.
Mössbauer and NQR Side by Side
| Mössbauer | NQR | |
|---|---|---|
| Radiation | γ rays (keV), Doppler-scanned | radio frequency (MHz) |
| Nuclei | few (⁵⁷Fe, ¹¹⁹Sn, …) | any quadrupolar nucleus with I ≥ 1 |
| Sample | solid / frozen solution | solid only |
| Measures | δ (s density), ΔE_Q (EFG), magnetic field | EFG: \( e^2Qq \), η |
| Key output | oxidation and spin state | bond character, site count |
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Isomer shift: s-electron density (monopole) | quadrupole splitting: EFG (quadrupole) | |
| ⁵⁷Fe: larger ( | \psi(0) | ^2 ) → lower δ |
| Doublet (ΔE_Q) | sextet (magnetic hyperfine) | |
| Mössbauer: γ rays, Doppler velocities | NQR: radio frequency, no field | |
| NQR: zero field | NMR: needs a field; quadrupole only perturbs | |
| Recoil-free fraction rises on cooling | NQR frequency rises on cooling (librations reduced) |
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
- How many lines does a ⁵⁷Fe spectrum show with (a) quadrupole splitting only, (b) magnetic splitting only?
- Why is no NQR signal observed for chlorine in crystalline NaCl?
- For ⁵⁷Fe (14.4 keV), what energy does a Doppler velocity of 1 mm s⁻¹ correspond to?
Answers
- (a) 2 (doublet); (b) 6 (sextet).
- Cl⁻ sits in a cubic (octahedral) site: the EFG is zero, so there is no quadrupole splitting.
- \( \Delta E = E_\gamma v/c = 14\,400 \times (10^{-3}/3.00 \times 10^8) = 4.8 \times 10^{-8} \) eV.
Level 2: Part B Style
Practice Q1. Which iron compound is expected to show the largest ⁵⁷Fe isomer shift?
(a) \( \text{K}_4[\text{Fe(CN)}_6] \) (b) \( \text{K}_3[\text{Fe(CN)}_6] \) (c) \( \text{FeSO}_4 \cdot 7\text{H}_2\text{O} \) (d) \( \text{FeCl}_3 \)
Answer: (c)
High-spin Fe(II) has the most d-electron shielding, hence the least s density at the nucleus and the most positive δ (about +1.2 mm s⁻¹ for the aqua ion). The cyanides are low spin with strong π back-bonding (δ near zero); FeCl₃ is high-spin Fe(III) (about +0.4–0.5).
Practice Q2. Which of these gives the largest ⁵⁷Fe quadrupole splitting?
(a) \( [\text{Fe(H}_2\text{O)}_6]^{3+} \) (b) \( [\text{Fe(H}_2\text{O)}_6]^{2+} \) (c) \( [\text{Fe(CN)}_6]^{4-} \) (d) \( [\text{FeF}_6]^{3-} \)
Answer: (b)
HS Fe(II) (\( t_{2g}^4 e_g^2 \)) has an asymmetric valence-electron distribution and a large EFG. HS Fe(III) (a, d) and LS Fe(II) (c) have spherical d shells.
Practice Q3. For the ¹¹⁹Sn Mössbauer spectra of SnCl₂ and SnCl₄, which statement is correct?
(a) SnCl₄ has the larger positive isomer shift because Sn(IV) has the higher charge (b) SnCl₂ has the larger positive isomer shift because of its 5s² electron density (c) both have the same isomer shift because both are chlorides (d) isomer shifts cannot be measured for tin
Answer: (b)
For ¹¹⁹Sn, \( R_e > R_g \), so greater 5s density at the nucleus gives a more positive δ. Sn(II) retains a 5s² pair.
Practice Q4. A ¹²⁷I (I = 5/2) site with an axially symmetric EFG shows its lower-frequency NQR line at 300 MHz. The second line and the quadrupole coupling constant are:
(a) 600 MHz; 2000 MHz (b) 450 MHz; 1000 MHz (c) 600 MHz; 1000 MHz (d) 150 MHz; 2000 MHz
Answer: (a)
For η = 0, \( \nu_1 = \tfrac{3}{20}e^2Qq/h \) and \( \nu_2 = 2\nu_1 \). Taking 300 MHz as \( \nu_1 \): \( e^2Qq/h = 300 \times 20/3 = 2000 \) MHz and \( \nu_2 = 600 \) MHz.
Level 3: Part C Style
Practice Q5. An iron compound shows two quadrupole doublets of equal area at 80 K: doublet A (δ = −0.05, ΔE_Q ≈ 0.1 mm s⁻¹) and doublet B (δ = +0.42, ΔE_Q ≈ 0.4 mm s⁻¹), δ relative to α-Fe. Which conclusions are justified?
- A is low-spin Fe(II) in a nearly cubic environment.
- B is high-spin Fe(III).
- A and B could be the C-bound and N-bound iron sites of a cyanide-bridged compound.
- B is high-spin Fe(II).
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2 and 3 only (d) 3 and 4 only
Answer: (a) 1, 2 and 3 only
- δ near zero and almost no splitting: low-spin \( t_{2g}^6 \) Fe(II) with six equivalent strong-field (C-bound) ligands. Correct.
- δ ≈ 0.4 and small ΔE_Q: spherical \( ^6A_1 \) high-spin Fe(III). Correct.
- That is exactly the Prussian-blue pattern (Fe(II)–C, Fe(III)–N). Correct.
- Wrong. HS Fe(II) would show δ about 1 and ΔE_Q of 2–3 mm s⁻¹.
Practice Q6. A solid organochlorine compound shows a ³⁵Cl NQR line at 34.0 MHz and a weaker line at about 26.8 MHz. Take \( Q(^{35}\text{Cl})/Q(^{37}\text{Cl}) = 1.27 \), η ≈ 0. Which statements are correct?
- The quadrupole coupling constant \( e^2Qq/h \) for ³⁵Cl is 68.0 MHz.
- The 26.8 MHz line is from ³⁷Cl at the same site.
- The intensity ratio of the two lines should be close to 3 : 1.
- Two lines show that there are two inequivalent chlorine sites.
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2 and 3 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
- For I = 3/2, \( \nu = \tfrac12 e^2Qq/h \): \( e^2Qq/h = 68.0 \) MHz. Correct.
- 34.0/1.27 = 26.8 MHz. Correct.
- Natural abundances about 76 : 24. Correct.
- Wrong. The frequency ratio identifies an isotope pair at one site.
Practice Q7. The ³⁵Cl quadrupole coupling constant in a metal chloride is 60.0 MHz; for the free Cl atom it is about 109.7 MHz. Using the simple relation \( (e^2Qq)_{\text{mol}}/(e^2Qq)_{\text{atom}} = 1 - i \), which statements are correct?
- The ionic character i of the M–Cl bond is about 0.45.
- The pure NQR frequency of this site is 30.0 MHz.
- A more ionic M–Cl bond would give a higher NQR frequency.
- Warming the sample would be expected to lower the observed frequency slightly.
(a) 1, 2 and 4 only (b) 1 and 3 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 4 only
- \( 1 - i = 60.0/109.7 = 0.547 \), so i ≈ 0.45. Correct.
- \( \nu = 60.0/2 = 30.0 \) MHz. Correct.
- Wrong. More ionic means a more nearly closed, spherical p shell, a smaller EFG and a lower frequency.
- Correct. Thermal librations average the EFG (Bayer behaviour).
Quick Revision Table
| Idea | Fact |
|---|---|
| Recoil | \( E_R = E_\gamma^2/2Mc^2 \) ≈ 2 × 10⁻³ eV for free ⁵⁷Fe; solids give recoil-free fraction f |
| ⁵⁷Fe | 14.4 keV from ⁵⁷Co; I = 1/2 → 3/2; 1 mm s⁻¹ = 4.8 × 10⁻⁸ eV; Γ ≈ 0.1 mm s⁻¹ |
| δ (Fe) | more s density → lower δ; HS Fe(II) ≈ 0.9–1.3 > HS Fe(III) ≈ 0.2–0.5 > LS ≈ 0 |
| δ (Sn) | more s density → higher δ; Sn(II) > Sn(IV) |
| ΔE_Q | large for HS Fe(II), moderate LS Fe(III), small HS Fe(III) and LS Fe(II) |
| Magnetic | sextet 3:2:1:1:2:3; α-Fe ≈ 33 T |
| Prussian blue | LS Fe(II)–C + HS Fe(III)–N (same as Turnbull’s blue) |
| NQR | solids, zero field, I ≥ 1, no line in cubic sites |
| Frequencies | I = 1: \( \tfrac34 \); I = 3/2: \( \tfrac12 \); I = 5/2: \( \tfrac{3}{20} \), \( \tfrac{3}{10} \) (× \( e^2Qq/h \)) |
| Cl isotopes | ν(³⁵Cl)/ν(³⁷Cl) ≈ 1.27, intensity ≈ 3:1 |
| Townes–Dailey | \( e^2Qq_{\text{mol}}/e^2Qq_{\text{atom}} \approx 1 - i \); ³⁵Cl atom ≈ 110 MHz |
Key Takeaways
What’s Next
- Mass Spectrometry, Electron Spectroscopy and Microscopy: composition, surface oxidation states and images.
- Previous topic: EPR Spectroscopy.
- Spin crossover and mixed valence context: Magnetic Properties of Complexes.
- Chapter quick revision.