Electron Paramagnetic Resonance: Principles

Unpaired electron spins and microwave resonance

Lesson 3675 of 4,500 · Advanced Spectroscopy

Learning objectives

Introduction

Electron paramagnetic resonance, also called electron spin resonance, detects species with unpaired electrons. Organic radicals, many transition-metal ions and certain defects in solids are examples. The method does not count all electrons: a diamagnetic substance with all electrons paired normally gives no ordinary EPR signal. Its sensitivity to a single unpaired spin makes EPR particularly useful for short-lived reaction intermediates, redox centres and the local environment of paramagnetic ions.

Core explanation

An electron carries spin angular momentum and a magnetic moment. In a static magnetic field B, the magnetic interaction splits the spin states. For an isolated electron-like spin S = 1/2, two spin projections form a lower- and a higher-energy level. Radiation is absorbed when its photon energy matches the separation: hν = g μB B for the simple isotropic case. Here h is Planck's constant, ν the microwave frequency, μB the Bohr magneton and g an effective dimensionless factor. In a typical continuous-wave experiment, a fixed microwave frequency is applied while the magnetic field is swept until resonance occurs.

For X-band instruments, ν is often near 9–10 GHz and the resonance of a spin with g near 2 occurs around 0.34 T. This is a scale, not a universal fixed field: changing frequency or g changes the field. The free-electron magnitude of g is approximately 2.0023; interactions with orbitals and the surrounding atoms shift the effective value. In solids, g may differ with orientation, so a powder can give a broad pattern rather than one narrow line. Nearby magnetic nuclei further split lines through hyperfine interactions; those patterns can show where spin density resides.

Most teaching diagrams show absorption versus magnetic field, but continuous-wave EPR commonly displays the first derivative of absorption. A derivative line rises and falls through a zero crossing at the approximate resonance centre. Two lobes of a derivative line are not automatically two separate paramagnetic species. The spectrometer uses field modulation and phase-sensitive detection to improve sensitivity. Signal intensity depends on concentration, transition probability, temperature, saturation, linewidth and acquisition settings. Quantifying spin number needs calibration and conditions that keep the response in the linear regime.

EPR differs from NMR both in the particle observed and in the frequency scale. Electron magnetic moments are much larger than nuclear moments, so electron spin transitions occur at microwave frequencies for routine laboratory fields, whereas NMR uses radiofrequencies. The greater magnetic moment can give high sensitivity but often also rapid relaxation and broad lines. One may freeze a solution to stabilise an intermediate, yet freezing changes molecular motion and can reveal anisotropy that was averaged in the liquid. The source Bruker's EPR 101 guide describes field-swept microwave resonance and the central role of g in practical instruments.

An EPR signal alone seldom identifies an exact chemical structure. A radical near g = 2 could be one of many organic species. Use g, hyperfine splitting, temperature dependence and perhaps isotopic substitution or reaction controls together. Conversely, no signal need not prove no unpaired electrons: a sample may be too dilute, the line too broad, relaxation too rapid or the selected microwave mode inappropriate. The experiment interrogates transitions allowed and detectable under its particular conditions.

Step-by-step reasoning

1. Ask whether the sample can contain an unpaired electron. 2. Choose frequency and magnetic-field sweep range that can satisfy hν = gμB B. 3. Locate derivative zero crossings or fit an absorption lineshape to estimate resonance fields. 4. Compare g and hyperfine pattern with plausible spin-bearing centres. 5. Check concentration, temperature, power and controls before assigning intensity or absence.

Visual explanation

Draw two horizontal energy levels for spin projections in a magnetic field. Their gap widens as B grows. Draw a fixed-height microwave photon beside them: at one field, the level gap equals that photon energy and absorption occurs. On a field-swept trace, this appears as a derivative-shaped feature centred at that field.

Real-world analogy

Think of two floors in a building whose separation grows as a control dial is turned. A person can jump between floors only when given exactly the required energy. The microwave photon supplies a fixed jump energy, while the magnetic-field dial tunes the floor spacing into resonance.

Real-world example

During an oxidation reaction, an initially diamagnetic molecule may briefly form a radical. A rapid-freeze sample can trap that intermediate. Its EPR resonance confirms a detectable unpaired spin, while splitting by nearby nuclei can help locate the spin density. A matched unreacted sample establishes that the feature was generated by the reaction rather than a contaminant or instrumental background.

Why?

Why do closed-shell molecules usually give no ordinary EPR signal? Their paired electron spins cancel the net spin magnetic moment of the ground state. Without accessible transitions between unpaired-spin sublevels at the microwave energy used, the resonance experiment has nothing to detect.

Common misconception

"One derivative peak means two absorptions because it has two lobes." The lobes are positive and negative slopes of one absorption line. Read the derivative's zero crossing as a first estimate of the centre; inspect additional lines before counting species.

Worked example

Question: An X-band EPR experiment uses ν = 9.50 GHz. Estimate the field for g = 2.00 using h = 6.626 × 10⁻³⁴ J s and μB = 9.274 × 10⁻²⁴ J T⁻¹.

Reasoning: Rearrange the resonance equation to B = hν/(gμB). The numerator is 6.626 × 10⁻³⁴ × 9.50 × 10⁹ = 6.295 × 10⁻²⁴ J. The denominator is 2.00 × 9.274 × 10⁻²⁴ = 1.855 × 10⁻²³ J T⁻¹. Dividing gives about 0.339 T, a realistic X-band field. This estimates the centre before hyperfine or anisotropic splitting.

Answer: Approximately 0.339 T.

Quick check

1. Would pure water give ordinary EPR under normal conditions? Answer: No; its electronic ground state has paired electrons. At fixed g, raising microwave frequency would instead raise the resonance field in proportion by hν = gμB B.

Exam focus

Use the resonance equation with SI units and distinguish field-swept EPR from frequency-swept NMR presentations. State explicitly that EPR requires an accessible paramagnetic spin transition. Explain why derivative detection does not double the number of lines.

Advanced insight

The simple two-level picture assumes an effective S = 1/2 spin with an approximately isotropic g. Transition-metal systems can have multiple spin states, anisotropic tensors, zero-field splitting and exchange coupling. Pulsed EPR uses spin echoes rather than a continuously modulated field and can measure relaxation or spin–spin distances. The free-electron reference scale and related constants are maintained in NIST CODATA data.

Summary

EPR detects unpaired electronic spins by matching microwave energy to magnetic-field-induced level separations. For a simple S = 1/2 centre, hν = gμB B. Field position, splitting and linewidth carry information about electronic surroundings, but assignment needs multiple features and suitable controls. Ordinary diamagnetic samples are generally silent.

Practice questions

1. Which has an ordinary EPR signal more plausibly, a neutral closed-shell molecule or its radical ion? Answer: The radical ion, if its unpaired electron gives a detectable transition. 2. What is swept in a standard continuous-wave field-swept EPR experiment? Answer: The static magnetic field while microwave frequency is approximately fixed. 3. What produces two lobes around one derivative EPR line? Answer: Positive and negative slopes of one absorption feature. 4. Why might an EPR-silent sample still contain paramagnetic species? Answer: Concentration, linewidth, saturation, rapid relaxation or experimental sweep range may prevent detection.