Electron–Nuclear Double Resonance¶

Detect nuclear-spin transitions through their effect on a coupled electron-resonance signal to resolve local hyperfine interactions.

Version: v1 · 2026-10-03 · History
Domain-specific #: 13183
Domain group: Natural Sciences
Origin domain: Physics
Subdomains: Electron Paramagnetic Resonance, Spin Spectroscopy → Physics
Aliases: ENDOR, Electron nuclear double resonance spectroscopy, Electron-nuclear double resonance

Core Idea¶

Electron–nuclear double resonance (ENDOR) is a magnetic-resonance method for detecting transitions of nuclei coupled to a paramagnetic electron by watching how nuclear-frequency excitation changes an electron-resonance response. It pairs the local nuclear-spin information with sensitive electron-spin detection. Where hyperfine splittings crowd or broaden an ordinary electron paramagnetic resonance (EPR) spectrum, a nuclear-frequency ENDOR spectrum can separate interactions that EPR alone does not readily resolve.^[1]^[2]

The frozen seed described one familiar continuous-wave (CW) implementation: partially saturate an EPR line, sweep radio-frequency (RF) nuclear excitation, and detect changes in EPR intensity. That is not the whole abstraction. Pulsed ENDOR uses microwave and RF pulse sequences and can observe changes in an electron-spin echo. What survives across variants is the coupling-mediated detection of driven nuclear transitions through an electron response, not one universal desaturation pathway.^[2]^[3]^[4]

Structural Signature¶

Sig role-phrases:

Paramagnetic spin system — An unpaired electron has a detectable resonance and couples to one or more magnetic nuclei. A sample with no relevant electron–nuclear coupling cannot yield the sought contrast.^[2]
Electron-resonance preparation and readout — Microwave interrogation establishes an electron-spin-dependent observable: conventional EPR intensity, a pulsed EPR echo, or in an optically assisted silicon experiment an Auger-electron/photoconductive signal. These are protocol alternatives, not equivalent detector names.^[3]^[4]^[5]
Nuclear-frequency drive — RF irradiation addresses coupled nuclear transitions. This second resonance distinguishes ENDOR from ordinary EPR.^[1]^[2]
Coupling-mediated signal change — Resonant nuclear excitation alters spin populations or coherences enough to change the electron-detected response; its mechanism depends on the pulse or CW protocol.^[2]
Spectral interpretation — RF-dependent line positions and patterns constrain hyperfine and sometimes quadrupole couplings; deriving nuclear assignments or geometry needs a spin model and independent evidence.^[2]

Condensed: coupled electron–nuclear spins → electron readout → nuclear RF perturbation → electron-detected nuclear spectrum → model-qualified coupling inference.

What It Is Not¶

Not EPR alone. EPR may show electron-spin transitions and hyperfine structure, but without driven nuclear-frequency excitation it is not ENDOR.
Not direct NMR. NMR detects a nuclear response directly; ENDOR detects a nuclear transition through its effect on the electron resonance.^[1]
Not necessarily CW desaturation. The seed's partial-saturation mechanism applies to a CW implementation; pulsed ENDOR has different sequence and contrast formation.^[2]^[4]
Not automatically a unique structural solution. A line constrains a coupling; assignments, electron-spin distribution and geometry can remain underdetermined.
Not identical to ESEEM. Electron-spin echo envelope modulation can reveal nuclear couplings without the applied nuclear RF transition constitutive of ENDOR.^[2]

Scope of Application¶

ENDOR applies to paramagnetic centers with coupled nuclear spins: crystal defects, trapped electrons, donor spins in semiconductors, radicals and metal-containing sites. The same double-resonance architecture can be used in solids and molecular systems, but different line broadening, relaxation and accessible RF ranges change which variant works. ETH Zürich's EPR materials explicitly distinguish CW and pulse ENDOR, and an original silicon-donor study shows a nonbiological semiconductor application.^[2]^[5]

The method is most informative when a nuclear-frequency spectrum adds resolution or selectivity unavailable from the EPR line alone. It requires an electron-resonance signal and a detectable nuclear coupling. A missing ENDOR line does not prove a nucleus absent: RF excitation bandwidth, coupling strength and relaxation can make a transition invisible in the chosen sequence.^[2]^[4]

Clarity¶

The word Double (manifold) identifies two linked resonances rather than two independent spectra. One addresses an electron transition and observes its response; another drives a nuclear transition. Hyperfine interaction couples the two, so RF resonance is visible indirectly in an EPR-derived signal. This explains why nuclear information can be read with electron-spin sensitivity without treating the experiment as ordinary NMR.^[1]^[2]

It also separates observable from interpretation. A frequency-dependent EPR, echo or electron-spin-dependent electrical change is evidence of a coupled nuclear transition. A claim about which atom produced it or how far that atom lies from the electron is a model-dependent structural inference. Those are not the same level of fact.

Manages Complexity¶

An EPR line may contain overlapping or unresolved hyperfine contributions. ENDOR shifts part of the question into nuclear-frequency space, where narrower and fewer lines can make individual couplings easier to distinguish. The paired excitation/readout strategy uses the electron signal as an amplifier of otherwise hard-to-observe local nuclear information.^[2]

This is selective compression. ENDOR simplifies a crowded electron spectrum only for nuclei and couplings visible under the chosen conditions. Adding RF sweeps or pulse sequences introduces its own sensitivity windows and artifacts. More spectral resolution does not eliminate the need to assign lines or to compare candidate spin models.^[2]^[4]

Abstract Reasoning¶

First establish a paramagnetic signal and a candidate set of coupled nuclei. Then choose a CW or pulsed electron readout and apply RF excitation while monitoring the electron response. If the response changes at a nuclear transition frequency, use the line pattern, magnetic-field dependence and a spin Hamiltonian to constrain hyperfine or quadrupole parameters. Compare alternative nuclei and structural models before inferring geometry; no single peak automatically specifies a bond length or coordination arrangement.^[2]

Counterfactuals test the identity. Without RF excitation, an EPR spectrum is not ENDOR. Without electron-detected contrast, a direct nuclear spectrum is NMR. If changing the pulse sequence reveals a coupling absent in CW, the missing CW line reflects protocol visibility rather than proof that the nucleus was not present.^[3]^[4]

Knowledge Transfer¶

The same electron-detected nuclear-resonance relation transfers from a defect in a crystal to a donor in silicon or a molecular radical. The application-specific parts are the spin Hamiltonian, material, nuclei, relaxation regime and measurement sequence. A line-assignment rule from one sample is not portable without those conditions.^[5]^[2]

At a broader level ENDOR uses measurement through coupled variables, but that general idea belongs to a wider Measurement abstraction, not to ENDOR as a prime. The reviewed strict DAG parent is Resonance as a presupposed operation: ENDOR requires coupled nuclear and electron resonant responses, although the two-resonance protocol is not a subtype of one response.

Examples¶

Auger-electron-detected phosphorus ENDOR in silicon¶

Franke and colleagues measured phosphorus donors in isotopically enriched ²⁸Si using selective optical creation of donor-bound excitons and an Auger-electron/photoconductive readout of donor electron-spin state. The ENDOR sequence combines microwave manipulation with an RF nuclear pulse; the final optical readout changes when RF is resonant, rather than monitoring an electron-spin echo as the ENDOR signal. At B=351 mT and an electron transition at 9760 MHz, the paper reports two ³¹P nuclear-resonance peaks, fitted by hyperfine coupling A=117.52(2) MHz and nuclear g_n=2.259(2). It separately uses donor-bound-exciton line 2 to detect one resonance at 65.162 MHz. Its later nuclear spin-echo measurement is not evidence that the ENDOR readout was an EPR echo. The natural-silicon sample in the same paper has AEDMR proof of principle, not the reported full ENDOR spectrum; this case refers to enriched ²⁸Si.^[5]

Mapped back: system = phosphorus donor electron coupled to its ³¹P nucleus in ²⁸Si; electron readout = optically enabled Auger-electron/photoconductive donor-state signal; nuclear drive = RF pulse; mediated change = two RF-dependent ENDOR peaks; interpretation = fitted 117.52(2) MHz hyperfine coupling under the donor spin model.

Pulsed Davies ENDOR of nitrogen inside a fullerene¶

Tyryshkin and colleagues report a genuinely different electron readout: pulsed Davies ENDOR on ¹⁴N@C₆₀ dissolved in CS₂, detected through a two-pulse electron-spin echo after microwave preparation and an RF nuclear pulse. The endohedral nitrogen has S=3/2, I=1, and the paper gives isotropic hyperfine coupling a=15.7 MHz; experiments address the central EPR-triplet line M_I=0. Their observed finding is that conventional Davies-ENDOR visibility can be severely suppressed when nuclear relaxation T₁n is long; a modified sequence adds a post-detection RF pulse to improve repetition and signal-to-noise. Those are source-reported properties and results, not a fabricated defect spectrum. Neither 15.7 MHz nor a visibility change by itself specifies the cage geometry uniquely.^[4]

Mapped back: system = endohedral ¹⁴N electron–nuclear spins in N@C₆₀; electron readout = pulsed EPR echo; nuclear drive = RF pulse within Davies ENDOR; mediated change = RF-dependent echo visibility limited by T₁n; interpretation = 15.7 MHz coupling and protocol-dependent visibility, not an automatic structure determination.

Near miss: ordinary EPR¶

An EPR line shows hyperfine splitting, but no nuclear-frequency drive is applied. The information may still be useful, yet only one resonance has been driven. This is EPR, not ENDOR.

Structural Tensions¶

Nuclear resolution versus signal visibility. Nuclear-frequency spectra may resolve weak couplings better than EPR, while RF bandwidth, relaxation and coupling selection can suppress ENDOR contrast. Assuming every absent line means absent coupling is unsafe; broadening settings indiscriminately can sacrifice sensitivity. Diagnostic: does the chosen CW or pulsed sequence cover the coupling and relaxation regime of the candidate nucleus?^[2]^[4]

Coupling constraints versus structural uniqueness. Precise line positions narrow possible nuclei and environments, but several assignments or geometries can share compatible hyperfine parameters. Conservative interpretation leaves alternatives open; premature assignment creates false certainty. Diagnostic: which independent field, isotope or complementary measurements distinguish the candidate spin models?^[2]

Structural–Framed Character¶

ENDOR has a precise two-resonance perturbation-and-readout structure but is physics-instrument-framed. Its vocabulary travels among EPR laboratories and paramagnetic materials; applying the name to any generic two-channel detector would import magnetic-resonance assumptions rather than recognize the same method. Human practice selects CW or pulsed sequences, RF windows and spectral models; the method emerged from experimental physics and is standardized through specialist instrumentation, though not defined by one institution. Its evaluative weight is low: resolution and sensitivity are technical performance criteria, not moral rankings. The more portable idea of indirect measurement belongs to a broader Measurement prime, not to ENDOR itself. Its character: strongly structural inside magnetic-resonance practice, but domain-specific because its carrier and readout are electron–nuclear spins.

Structural Core vs. Domain Accent¶

The abstract skeleton is to perturb one coupled degree of freedom and detect its response through another with better readout. The domain mechanism is electron–nuclear hyperfine coupling under magnetic resonance, with microwave electron interrogation and RF nuclear excitation. Remove that mechanism and one has a generic indirect assay, not ENDOR; remove RF drive and one has EPR or echo-modulation methods. The named abstraction does not clear the prime bar because its identity requires paramagnetic spins and specific frequency domains. Resonance is a strict presupposed parent of the protocol, not a claim that ENDOR is a subtype of one resonant response.

This entry presupposes Resonance.

Resonance is a strict composition/presupposes parent: ENDOR needs frequency-selective nuclear and electron resonant responses coupled by the spin system; resonance can occur without ENDOR. Measurement is a broad conceptual neighbor but not an asserted strict parent here. Analytical Technique may classify some chemical ENDOR uses but is not a verified genus of all solid-state ENDOR. An EPR-method intermediate may eventually support a separate precise parent.

Relationships to Other Abstractions¶

Current abstraction Electron–Nuclear Double Resonance Domain-specific

Parents (1) — more general patterns this builds on

Electron–Nuclear Double Resonance presupposes Resonance Prime

ENDOR requires frequency-selective nuclear and electron resonance responses coupled through the spin system.

Hierarchy paths (10) — routes to 8 parentless roots

Electron–Nuclear Double Resonance → Resonance → Amplification → Founder Effect → Path Dependence → Dependency

Show alternative paths (9)

Neighborhood in Abstraction Space¶

Electron–Nuclear Double Resonance sits in a sparse region of the domain-specific corpus (82^nd percentile for distinctiveness): few abstractions share its structure, so a faithful description tends to retrieve it precisely.

Family — Unclustered & Miscellaneous (2551 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-10-08

Not to Be Confused With¶

Do not collapse ENDOR into EPR, NMR, ESEEM or one continuous-wave implementation. “Double resonance” means electron and nuclear transitions are coupled through the measured response, not merely that two unrelated instruments were used. Hyperfine and quadrupole parameters are inferential targets, not automatic outputs of every spectrum.