Fact-checked by Grok 2 weeks ago

Extended X-ray absorption fine structure

Extended X-ray absorption fine structure (EXAFS) is a spectroscopic technique within X-ray absorption spectroscopy (XAS) that analyzes the oscillatory modulations in the X-ray absorption coefficient (μ(E)) at energies typically 30–1000 eV above a core-level absorption edge of a selected atom, providing detailed information on the local atomic structure surrounding that atom.^[1] These oscillations arise from the interference between the outgoing photoelectron wave ejected from the absorbing atom and the waves backscattered by neighboring atoms, enabling the determination of interatomic distances (with precision ~0.01–0.02 Å), coordination numbers, and the identity of scatterer types up to distances of about 5–6 Å.^[2] The EXAFS signal is extracted as the function χ(k) = (μ(k) - μ₀(k))/μ₀(k), where k is the photoelectron wavevector, and is theoretically modeled using equations such as χ(k) ≈ Σ [N_j f_j(k) e^{-2k²σ_j²} sin(2kR_j + δ_j(k))] / (k R_j²), incorporating amplitude reduction factors, phase shifts, and disorder parameters.^[3] The EXAFS phenomenon was first observed in the 1920s through early X-ray absorption experiments, but its theoretical foundation as a structural probe was established in the 1970s, driven by advancements in synchrotron radiation sources that provided the necessary high-intensity, tunable X-rays.^[4] Key theoretical developments, including multiple-scattering treatments and ab initio calculations via codes like FEFF, emerged in the 1980s and 1990s, transforming EXAFS from a qualitative tool into a quantitative method for analyzing disordered and amorphous materials.^[2] Today, data analysis relies on Fourier transforms to convert k-space spectra to radial distribution functions in R-space, allowing separation of contributions from different coordination shells.^[5] EXAFS is element-specific, requiring no long-range order, and is particularly valuable for studying dilute systems (down to parts-per-million concentrations), making it indispensable in fields such as catalysis, environmental science, biology, and materials characterization.^[3] Applications include determining active site geometries in enzymes, local structures in battery materials, and speciation of pollutants in soils, often complementing techniques like X-ray diffraction for crystalline samples.^[5] Despite limitations in distinguishing multiple scatterers of similar distances or handling strong multiple scattering, ongoing improvements in time-resolved and operando measurements continue to expand its utility.^[1]

Fundamentals

Definition and Principles

Extended X-ray absorption fine structure (EXAFS) is a spectroscopic technique that probes the local atomic environment around a specific absorbing atom by measuring the oscillatory fine structure in the X-ray absorption coefficient, denoted as μ(E), at energies typically ranging from 50 to 1000 eV above the absorption edge.^[6] These oscillations provide quantitative information on interatomic distances, coordination numbers, and structural disorder through the analysis of interference effects in the photoelectron wave. The method is element-specific, as it targets core-level electrons of a chosen atom via its characteristic absorption edge, allowing determination of the local structure in materials without requiring long-range crystalline order.^[6] The physical origin of EXAFS oscillations lies in the interference between the outgoing photoelectron wave excited by the incident X-ray and the backscattered waves from surrounding atoms. When an X-ray photon with energy exceeding the binding energy of a core electron is absorbed, it ejects the electron into the continuum, creating a spherical photoelectron wave that scatters off neighboring atoms; the interference modulates the absorption probability, producing the characteristic fine structure.^[6] This backscattering process encodes details about the distances and types of neighboring atoms, enabling reconstruction of the radial distribution function around the absorber. EXAFS measurements rely on the fundamental principles of X-ray absorption, beginning with absorption edges such as K-edges (for 1s core electrons) or L-edges (for 2p core electrons), which mark the threshold energies for core-level excitations and ranging from approximately 1 keV to over 100 keV for elements from sodium to uranium, depending on the specific absorption edge (K or L).^[7] The absorption process follows Beer's law, expressed as I = I_0 e^{-\mu x}, where I is the transmitted intensity, I_0 is the incident intensity, \mu is the linear absorption coefficient, and x is the sample thickness; variations in \mu(E) are detected to isolate the fine structure. These high energies are necessary to penetrate samples and excite deeply bound core electrons, ensuring the ejected photoelectrons have sufficient kinetic energy (on the order of hundreds of eV) for probing several coordination shells.^[6] Unlike X-ray diffraction, which requires periodic crystalline structures to produce interpretable patterns, EXAFS is particularly advantageous for investigating disordered, amorphous, or dilute systems, as it focuses solely on the short-range (up to ~5-6 Å) environment around the absorbing atom and is insensitive to overall crystallinity or phase.^[6] This makes it ideal for studying catalysts, biomolecules, and nanomaterials where local coordination geometry is key. Synchrotron radiation serves as the primary source due to its high intensity and tunability, enabling high signal-to-noise data collection.^[6] Complementarily, X-ray absorption near-edge structure (XANES) provides information on electronic structure and oxidation states closer to the edge.^[6]

Relation to Broader X-ray Absorption Techniques

X-ray absorption spectroscopy (XAS) is a broad technique that examines the absorption of X-rays by matter, capturing features from pre-edge regions through the absorption edge to the extended fine structure beyond, providing element-specific insights into local atomic and electronic environments.^[8] Within this framework, extended X-ray absorption fine structure (EXAFS) specifically analyzes the oscillatory modulations in the absorption coefficient at energies greater than approximately 30-50 eV above the absorption edge, emphasizing quantitative determination of interatomic distances, coordination numbers, and disorder in the local structure around the absorbing atom.^[9] This high-energy region arises from interference effects in the photoelectron wave scattered by neighboring atoms, distinguishing EXAFS from lower-energy features in XAS. In contrast, X-ray absorption near-edge structure (XANES), often considered the near-edge portion of XAS spanning roughly -20 eV below to +50 eV above the edge, primarily probes the electronic structure, including oxidation states, bond covalency, and coordination geometry through transitions to bound states and multiple scattering within a few angstroms.^[8] While XANES offers qualitative fingerprints of chemical speciation and symmetry, EXAFS extends further to yield precise radial distribution functions, enabling the extraction of structural parameters like bond lengths with sub-angstrom accuracy.^[9] These complementary roles highlight how XANES captures short-range electronic influences, whereas EXAFS focuses on longer-range structural details, often requiring distinct theoretical treatments for analysis. The integration of XANES and EXAFS forms the complete X-ray absorption fine structure (XAFS) spectrum, typically spanning 150 eV below to over 1000 eV above the edge, allowing a holistic view of both electronic and structural properties around the target atom.^[8] Advanced multiple-scattering theories bridge the two regions by accounting for complex photoelectron paths in denser coordination shells, enhancing the accuracy of combined XAFS interpretations.^[9]

Theoretical Basis

Photoelectron Scattering Mechanism

In extended X-ray absorption fine structure (EXAFS) spectroscopy, the process begins with the absorption of an X-ray photon by a core electron of the absorbing atom, leading to the ejection of a photoelectron if the photon energy exceeds the binding energy of that electron. This ejected photoelectron propagates as a spherical wave outward from the absorber, with a de Broglie wavelength given by \lambda = \frac{h}{\sqrt{2m(E - V)}}, where h is Planck's constant, m is the electron mass, E is the photon energy, and V is the effective potential experienced by the electron. The wavelength decreases as the kinetic energy of the photoelectron increases, typically ranging from about 1 to 3 Å in the EXAFS region, enabling sensitivity to atomic-scale structural details around the absorber.^[2] The dominant mechanism underlying EXAFS signals is the single-scattering approximation, in which the photoelectron wave scatters primarily off a single neighboring atom (or shell of equivalent atoms) before returning to the absorber. In this process, the scattering is treated using muffin-tin potentials, which approximate the atomic potential as a spherical region of constant charge density around each atom, separated by interstitial regions. The interaction introduces energy-dependent phase shifts \delta(k), which account for the distortion of the outgoing and backscattered waves due to the local potential; these phase shifts are crucial for determining interatomic distances and are computed from the scattering properties of the neighboring atoms. This approximation holds well for photoelectron energies above approximately 20-30 eV beyond the absorption edge, where the mean free path of the electron is sufficiently long to probe multiple coordination shells without significant multiple interactions.^[2]^[10] The EXAFS oscillations arise from the quantum mechanical interference between the outgoing photoelectron wave and the wave backscattered from neighboring atoms, resulting in constructive and destructive patterns that modulate the absorption coefficient. The photoelectron wavevector is defined as k = \frac{\sqrt{2m(E - E_0)}}{\hbar}, where E_0 is the absorption edge energy and \hbar is the reduced Planck's constant; the modulation function \chi(k) thus varies sinusoidally with k, with its amplitude governed by the backscattering amplitude f(k)—which depends on the scattering angle, atomic number, and k—and the mean free path \lambda(k), which typically spans 5-20 Å and decreases with increasing energy due to inelastic scattering losses. These interference effects encode information about the local atomic environment, such as bond lengths and coordination numbers, through the periodicity and envelope of \chi(k).^[2]^[10] While the single-scattering approximation captures the primary physics of EXAFS for most interatomic distances greater than about 2 Å, limitations emerge in cases of closely spaced atoms (below 2 Å), where multiple-scattering events— involving the photoelectron bouncing between two or more neighboring atoms—become significant and can distort the simple oscillatory pattern. These effects are more pronounced at lower energies near the absorption edge but are generally secondary in the extended region, with detailed treatment deferred to advanced analysis methods.^[2]

Mathematical Formulation

The mathematical formulation of extended X-ray absorption fine structure (EXAFS) is grounded in the single-scattering approximation, which describes the oscillatory modulation of the absorption coefficient due to interference between the outgoing and backscattered photoelectron waves. The core EXAFS signal, denoted as \chi(k), is expressed as

\chi(k) = \sum_j \frac{N_j |f_j(k)|}{k R_j^2} e^{-2k^2 \sigma_j^2} e^{-2 R_j / \lambda(k)} \sin \left[ 2 k R_j + \delta_j(k) \right],

where the sum is over coordination shells j, k is the photoelectron wavenumber, N_j is the coordination number, R_j is the absorber-scatterer distance, |f_j(k)| is the backscattering amplitude, \sigma_j^2 is the mean-square disorder, \lambda(k) is the mean free path, and \delta_j(k) is the total phase shift.^[11]^[12] This equation provides the quantitative basis for extracting structural parameters from experimental spectra, with the sine term capturing the interference oscillations and the exponential factors accounting for damping effects.^[13] The derivation begins with the plane-wave approximation for the photoelectron wavefunction, \psi(k, r) \approx e^{i k r}, which simplifies the scattering but neglects wave curvature. To incorporate the spherical nature of the emitted photoelectron, the wavefunction is refined to \psi(k, r) = e^{i k r} / (k r), introducing the $1/r radial dependence and enabling backscattering from neighboring atoms at distance R_j. The interference term arises from the sum of the outgoing wave and the backscattered wave, f_j(k) e^{i \delta_j(k)} e^{i k (2 R_j)} / (k R_j)^2, modulated by damping due to inelastic scattering (e^{-2 R_j / \lambda(k)}) and disorder (e^{-2 k^2 \sigma_j^2}). The spectrum is transformed from energy E to k-space via k = \sqrt{2 m (E - E_0)/\hbar^2}, where E_0 is the absorption threshold energy, emphasizing the oscillatory behavior in k.^[11]^[12] Key parameters in the EXAFS equation encode structural and dynamic information: the phase of the sine term, $2 k R_j + \delta_j(k), yields the interatomic distance R_j through Fourier analysis; the amplitude scales with the coordination number N_j, providing the number of scatterers; the Debye-Waller-like factor \sigma_j^2 quantifies thermal and static disorder, broadening oscillations at higher k; and E_0 aligns the energy scale, often refined during fitting. The scattering amplitude |f_j(k)| and phase shift \delta_j(k) depend on the atomic number Z of the scatterer and are typically computed ab initio.^[13]^[12] While the plane-wave approximation suffices for many cases, extensions to curved-wave theory improve accuracy, particularly for low-Z elements or short interatomic distances (R_j < 2 Å), by fully accounting for the spherical wave propagation and angular momentum effects in multiple-scattering paths. This approach, formalized in rapid computational implementations, replaces the simple $1/R^2 term with more precise radial integrals, enhancing reliability for near-edge regions.

Experimental Techniques

Instrumentation and Beam Sources

The primary instrumentation for extended X-ray absorption fine structure (EXAFS) experiments relies on synchrotron radiation sources, which provide tunable, high-brilliance X-ray beams essential for capturing the weak oscillatory signals in the absorption spectrum. These sources, such as those at the European Synchrotron Radiation Facility (ESRF) and the Advanced Photon Source (APS), deliver beams with brilliance on the order of 10^{13}–10^{15} photons/s/mm²/mrad²/0.1% bandwidth (as of 2025, following upgrades at facilities like ESRF-EBS and APS-U).^[14]^[15] Recent upgrades, such as the ESRF Extremely Brilliant Source (EBS) in 2020 and the APS Upgrade (APS-U) completed in 2024, have further increased beam brilliance and enabled advanced operando and time-resolved EXAFS studies. These high intensities are particularly crucial given the EXAFS modulation amplitude of approximately 1-10% above the absorption edge. Beamlines at these facilities often employ bending magnets or insertion devices like undulators to generate the continuum spectrum, with the former providing broader but lower-brilliance emission and the latter offering peaked, higher-flux output tunable via gap adjustments.^[16] Energy selection in synchrotron EXAFS setups is achieved using double-crystal monochromators, commonly fabricated from silicon (Si(111) or Si(220) reflections), which diffract the polychromatic beam according to the Bragg equation to isolate energies near atomic absorption edges. For instance, K-edges of heavier elements like iron occur around 7 keV, while L-edges of transition metals typically occur in the range of 0.4–1.0 keV, requiring precise tuning over 40-1000 eV above the edge for extended fine structure data.^[16] Harmonic contamination from higher-order reflections is mitigated by detuning the second crystal, reducing unwanted higher-energy components by up to 100-fold while maintaining throughput.^[16] These monochromators often incorporate sagittal bending for horizontal focusing, optimizing beam size to 0.5-1 mm at the sample position. Detection of EXAFS signals typically involves ionization chambers for transmission mode measurements, where gas-filled chambers (e.g., filled with N_2 or Ar) quantify incident (I_0) and transmitted (I_t) beam intensities to compute the absorption coefficient μ(E) = \ln(I_0 / I_t).^[16] For dilute samples, fluorescence mode employs detectors like the Lytle detector, a specialized ion chamber with a large solid angle (up to 0.3 sr) positioned at 90° to the beam path to capture emitted fluorescence photons efficiently, particularly for K-edges above 4 keV.^[16] High-throughput applications utilize energy-dispersive multi-element detectors, such as those with 13-100 germanium elements, which enable parallel energy binning and reduce acquisition times for time-resolved studies.^[17] Laboratory-based alternatives to synchrotrons have emerged, though they face flux limitations compared to central facilities, typically achieving 10^5-10^6 photons/s versus synchrotron levels of 10^{11}-10^{12}. Traditional X-ray tubes with rotating anodes provide continuous spectra in the 5-30 keV range but suffer from higher noise and lower brilliance, restricting use to simpler samples. Recent 2020s advances in laser-plasma X-ray technology, such as laser wakefield accelerators, generate ultrashort pulses with average fluxes of ~2.4 \times 10^5 photons/eV at 9 keV per shot, enabling single-shot EXAFS over 250 eV bandwidths for ultrafast dynamics, though scalability to higher repetition rates remains a challenge.^[18] Bending magnet-like sources in compact setups offer modest tunability but are limited to fluxes significantly below those of synchrotrons (typically 10^6–10^7 times lower for conventional lab sources).^[19]

Sample Preparation and Detection Modes

Sample preparation for extended X-ray absorption fine structure (EXAFS) experiments requires careful consideration of the material form to ensure homogeneity and optimal absorption. Solid samples, such as powders, are typically ground to fine particles smaller than the absorption length (often <1 μm) and diluted with low-Z binders like boron nitride (BN) or polyethylene to minimize scattering and achieve uniform thickness without pinholes. Liquid samples are prepared in thin cells or cryo-trapped in frozen states to maintain stability during measurement, while gaseous samples utilize flow cells with appropriate detectors like Lytle detectors filled with gases such as argon or krypton. Sample thickness is optimized so that the edge jump Δμx ≈ 1, where Δμ is the change in absorption coefficient across the absorption edge and x is the thickness, ensuring a balance between signal strength and avoidance of excessive attenuation or harmonic distortions. Transmission mode is the most straightforward detection method, directly measuring the X-ray absorption coefficient μ(E) by recording the incident beam intensity I₀ in an upstream ionization chamber and the transmitted intensity I_t in a downstream chamber, yielding μ(E) = \ln(I_0 / I_t) / x. This approach is ideal for concentrated samples with absorber concentrations exceeding 1 wt%, as it provides high signal-to-noise ratios for homogeneous, bulk materials like foils or pellets. Uniform sample preparation is critical to prevent artifacts from thickness gradients. Fluorescence detection mode captures the characteristic X-rays (e.g., Kα or Kβ lines) emitted from the absorbing atoms following photoionization, making it suitable for dilute or trace-level absorbers below 0.1 wt% where transmission signals would be weak. It employs energy-dispersive or multi-element detectors positioned at 90° to the incident beam to maximize collection efficiency while minimizing elastic scattering. For thicker or more concentrated samples, self-absorption effects—where emitted fluorescence is reabsorbed—necessitate corrections, often using models that account for sample geometry and composition. Additional detection modes expand EXAFS applicability to specialized conditions. Total electron yield mode measures the sample's secondary electron current, offering surface sensitivity (typically probing depths of 2–5 nm) ideal for thin films, catalysts, or vacuum-compatible surfaces without self-absorption issues. Quick-EXAFS utilizes rapid monochromator scanning (e.g., 1–10 minutes per spectrum) for time-resolved studies of dynamic processes. In-situ setups, such as electrochemical cells with windows transparent to X-rays, enable measurements under operational conditions like varying electrode potentials or temperatures, often combining transmission or fluorescence detection. Artifacts in EXAFS data arise from preparation flaws and must be mitigated for reliable results. Pinholes or voids in pressed pellets cause beam leakage, inflating transmission intensities and distorting edge shapes; these are avoided through fine grinding, sieving, and uniform pressing, verified by beam scans. Radiation damage, prevalent in organic or biological samples due to beam-induced photolysis or structural changes, is countered by cryogenic cooling (e.g., liquid nitrogen temperatures), helium purging, or minimizing exposure time via quick-EXAFS protocols.

Data Analysis Methods

Preprocessing and Extraction

The preprocessing and extraction phase in EXAFS analysis involves transforming raw X-ray absorption spectra, typically measured as absorption coefficient μ(E) versus energy E, into a clean oscillatory signal χ(k) suitable for further interpretation. This critical step isolates the fine structure oscillations arising from photoelectron scattering while removing instrumental artifacts, background contributions, and noise, ensuring the data reflects atomic-scale structural information accurately.^[5] Normalization begins with pre-edge subtraction to eliminate low-energy instrumental backgrounds and atomic absorption trends below the absorption edge. A linear fit, polynomial (often of order 0-2), or Victoreen function—empirically derived to approximate atomic absorption—is fitted to the pre-edge region, typically 100-200 eV below the edge, and subtracted from the entire spectrum. The resulting pre-edge-subtracted data is then divided by the edge jump Δμ, defined as the difference between the post-edge absorption (50-150 eV above the edge) and the extrapolated pre-edge baseline at the edge energy E₀, yielding a normalized spectrum μ(E)/μ₀(E) that approaches 0 below the edge and 1 above it. This process accounts for variations in sample thickness, absorber concentration, and beam intensity, with errors in edge step estimation potentially biasing coordination numbers by up to 10-20% in subsequent fits.^[6]^[20]^[21] Background removal follows to extract the EXAFS oscillations by subtracting a smooth approximation of the non-oscillatory absorption μ₀(E). This is commonly achieved using a multi-segment polynomial spline function fitted over the post-edge region (e.g., 150 eV to the data endpoint), with the spline's curvature controlled by a parameter R_bkg (typically 0.9-1.5 Å) to avoid over-subtraction of true EXAFS signals from the first coordination shell. Early methods relied on tabulated atomic absorption coefficients from McMaster et al. for μ₀(E) estimation, while modern approaches include Bayesian inference to model background and oscillations simultaneously, minimizing artifacts in low-k regions by incorporating prior structural knowledge and reducing mean-squared error in χ(k) by factors of 2-5 compared to traditional splines. The isolated EXAFS function is then expressed as χ(E) = [μ(E) - μ₀(E)] / Δμ, which is converted to wavenumber space via k = √[2m(E - E₀)/ℏ²], where m is the electron mass and ℏ is the reduced Planck's constant, yielding χ(k) for analysis. Typical k-ranges span 3-15 Å⁻¹, windowed to exclude low-k noise and high-k attenuation.^[6]^[22]^[23] Edge energy E₀ is determined to anchor the energy-to-k conversion and define the onset of the EXAFS region (≈30-50 eV above E₀). Standard methods identify E₀ as the energy of the maximum in the first derivative dμ/dE, corresponding to the inflection point of the absorption edge, or alternatively as the zeroth crossing of the second derivative, providing reproducibility within 0.1-0.5 eV for K-edges. This choice influences phase shifts in subsequent Fourier transforms but is often refined empirically during modeling to optimize fits.^[6]^[24] Error handling during preprocessing mitigates noise and artifacts that degrade signal quality, particularly in dilute or disordered samples where signal-to-noise ratios can drop below 100:1 at high k. Multiple scans are averaged after alignment—using reference foils or edge derivatives to correct energy drifts—to reduce statistical noise by √N, where N is the number of scans, often achieving 20-50% improvements in χ(k) amplitude. Glitches from monochromator harmonics or detector saturation are identified via sharp discontinuities in μ(E) or its derivatives and excised or interpolated using cubic splines over minimal intervals (1-5 eV), preserving overall spectral integrity. For fluorescence detection, dead-time corrections (τ ≈ 1-10 μs) are applied to counteract pulse pile-up, ensuring accurate μ(E) recovery up to count rates of 10⁵-10⁶ s⁻¹. These steps prepare χ(k) for windowing, typically with Hanning or Kaiser functions over the selected k-range, to suppress sidelobes in the subsequent Fourier transform without introducing significant distortion.^[6]^[22]^[25]

Modeling and Interpretation

The modeling and interpretation of EXAFS data begin with the application of a Fourier transform to the preprocessed oscillatory function χ(k), typically weighted by k² or k³ to enhance the contribution of higher-frequency components corresponding to more distant shells.^[5] This transformation converts the data from k-space (wavenumber) to R-space, yielding a pseudo-radial distribution function that peaks near the absorber-scatterer distances, though phase shifts in the scattering introduce an offset of approximately 0.5 Å, requiring careful interpretation to align peaks with actual bond lengths.^[26] The resulting magnitude and phase of the R-space transform provide visual insights into coordination shells, with the imaginary part aiding in phase alignment during subsequent fitting. Fitting procedures employ least-squares refinement to compare experimental data against theoretical models, often using software packages like FEFF for ab initio calculation of scattering potentials and paths, and IFEFFIT or its graphical interface Artemis for iterative optimization.^[27]^[28] Single-shell fits suffice for nearest-neighbor analysis in simple systems, assuming isolated contributions from one type of scatterer, while multiple-shell fits account for overlapping signals from successive coordination spheres, improving accuracy for disordered or complex structures.^[29] Theoretical scattering amplitudes and phase shifts, derived from the absorber and scatterer atomic potentials, serve as inputs to these models, enabling quantitative matching in either k-space or R-space.^[30] During refinement, key structural parameters are extracted by minimizing the difference between observed and calculated χ(k) or its Fourier transform: interatomic distance R, coordination number N, mean-square disorder σ², and energy threshold shift E₀.^[31] Error estimation relies on correlation matrices from the least-squares process, which quantify parameter uncertainties and covariances, such as the strong coupling between E₀ and R that can inflate apparent errors if not constrained.^[32] Inclusion of multiple-scattering paths, particularly for low-Z scatterers or angles near 180°, enhances model fidelity by capturing higher-order photoelectron trajectories, reducing residuals in fits to data from crystalline or oriented samples.^[27] Recent advances incorporate machine learning to predict phase and amplitude functions, bypassing some ab initio computations for faster analysis of large datasets; for instance, neural networks trained on first-shell EXAFS have achieved rapid mapping of spectra to structural parameters in nanocatalysts with errors below 0.02 Å for bond lengths.^[33] Post-2020 developments, such as deep operator networks, further enable end-to-end prediction of EXAFS from related spectra like XANES, supporting high-throughput interpretations.^[34] Polarized EXAFS extends interpretation to anisotropic systems by varying the electric field vector relative to the sample orientation, revealing directional differences in bond lengths and disorders, as demonstrated in thin films where in-plane and out-of-plane scattering paths differ by up to 0.1 Å.^[35]

Applications

Structural Analysis in Materials

Extended X-ray absorption fine structure (EXAFS) spectroscopy is widely employed in materials science to elucidate the local atomic environment in solids lacking long-range crystalline order, such as catalysts, amorphous alloys, and nanomaterials. By analyzing the interference patterns in the photoelectron wave scattered from neighboring atoms, EXAFS provides element-specific information on interatomic distances, coordination numbers, and disorder within a radial distance of approximately 5 Å around the absorbing atom. This capability is particularly valuable for heterogeneous systems where traditional diffraction methods falter due to structural variability or polydispersity.^[36] In catalytic materials, EXAFS enables detailed probing of active sites in supported nanoparticles, revealing coordination geometries and dynamic changes under operational conditions. For instance, in platinum on carbon (Pt/C) catalysts used in polymer electrolyte fuel cells, in situ EXAFS measurements have identified variations in Pt-Pt coordination numbers and bond lengths during oxygen reduction reactions, correlating these structural shifts with enhanced electrocatalytic activity in Pt alloys like Pt-Co. Such studies demonstrate how EXAFS captures the evolution of nanoparticle surfaces, where coordination reductions from 12 in bulk Pt to as low as 7-9 in small clusters influence reaction kinetics and stability. For amorphous and disordered materials, EXAFS quantifies local atomic order by extracting pair distribution functions that describe the probability of finding atoms at specific distances from the central absorber. In metallic glasses, such as Ni-Zr alloys, EXAFS reveals a dense random packing model with average coordination numbers close to 12, but with increased disorder parameters compared to crystalline counterparts, enabling the distinction between topological and chemical short-range order. Similarly, in oxide glasses like those in the Li₂O-V₂O₅-P₂O₅ system, EXAFS identifies coordination shells around vanadium atoms, showing octahedral environments with V-O bond lengths around 1.9 Å, which inform conductivity mechanisms in these ionically conducting materials.^[37]^[38] In nanomaterials, EXAFS detects size-dependent structural modifications, such as bond length contractions due to surface effects and quantum confinement. For gold nanoparticles, EXAFS analysis of samples with average diameters of 1.7 nm shows an Au-Au bond length contraction of about 2 pm relative to bulk gold (2.884 Å), attributed to undercoordination at the surface and validated through ab initio simulations. In situ EXAFS further tracks phase transitions in nanostructures, like the melting or sintering of metal clusters, by monitoring changes in Debye-Waller factors and coordination numbers during thermal cycling. EXAFS complements X-ray diffraction (XRD) by focusing on short-range order (<5 Å) in multi-component systems, where its element selectivity allows isolation of specific atomic pairs without requiring long-range periodicity. While XRD excels at lattice parameters in crystalline phases, EXAFS provides critical local insights in amorphous or nanocrystalline materials, such as quantifying solute clustering in alloys that XRD alone cannot resolve. This synergy has been demonstrated in combined quick-EXAFS/XRD setups for real-time monitoring of structural dynamics in catalysts and thin films.^[8]^[36]

Investigations in Biological and Environmental Systems

EXAFS has been instrumental in elucidating the coordination environments of metal centers in metalloproteins, providing precise measurements of metal-ligand bond distances and coordination numbers. In hemoglobin, EXAFS analysis of the iron site in both oxy and deoxy forms revealed Fe-N distances of approximately 2.06 Å in the deoxy state and 1.98 Å in the oxy state, highlighting the structural changes associated with oxygen binding. Similarly, in carbonic anhydrase, EXAFS studies of the zinc active site demonstrated Zn-N and Zn-O distances around 2.0 Å and 1.95 Å, respectively, confirming tetrahedral coordination by histidine and water ligands essential for catalytic activity. These structural parameters, such as coordination number (N) and interatomic distance (R), enable detailed modeling of active sites without requiring crystalline samples.^[39]^[40]^[41] Temperature-dependent EXAFS measurements further reveal dynamic aspects of metalloprotein function, such as vibrational disorder quantified by the mean-square displacement parameter (σ²), which increases with temperature and reflects ligand flexibility in enzymes like hemoglobin and carbonic anhydrase. In environmental remediation, EXAFS facilitates speciation of heavy metals, identifying sorption mechanisms on mineral surfaces; for instance, arsenate (As(V)) sorption on manganese oxides shows bidentate binuclear complexes with As-O distances of 1.69 Å and As-Mn distances of 3.19 Å, informing strategies for contaminant immobilization. EXAFS also determines oxidation states and coordination in soils and waters, as seen in lead and zinc speciation where Pb forms inner-sphere complexes on iron oxides, reducing mobility in contaminated aquifers.^[41]^[42]^[43] In biogeochemistry, EXAFS elucidates microbial uptake mechanisms by probing metal coordination during bioaccumulation; studies on uranium in wetland rhizospheres show U(VI) reduction to U(IV) phosphates within bacterial cells, with U-O distances shifting from 1.80 Å to 2.30 Å post-uptake. In-situ speciation in natural samples, often using fluorescence detection for trace concentrations, has revealed arsenic speciation in microbial mats, where As(III) dominates in reduced zones bound to sulfhydryl groups. Challenges in these applications include radiation sensitivity of biological samples, which can alter metal oxidation states during prolonged exposure, and weak signals from dilute systems requiring high-flux synchrotrons. Synergies with electron paramagnetic resonance (EPR) and nuclear magnetic resonance (NMR) enhance speciation by combining EXAFS structural data with electronic and magnetic insights, as demonstrated in copper-containing proteins.^[44]^[42]^[45]^[46]

Historical Development

Early Discoveries and Theoretical Foundations

The discovery of X-rays by Wilhelm Conrad Röntgen in 1895 laid the groundwork for subsequent investigations into their interaction with matter, including absorption phenomena. In 1913, Maurice de Broglie and J. Herweg independently reported observations of X-ray absorption edges, noting sharp discontinuities in the absorption of X-rays by elements such as silver and bromine when the photon energy matched the binding energy of inner-shell electrons. These edges marked the onset of detailed spectroscopic studies, though the weak intensity of laboratory X-ray sources limited early precision. By the 1920s, finer details emerged in absorption spectra. In 1920, Hugo Fricke, collaborating with William Duane, observed oscillatory variations in X-ray absorption above the edges for high atomic number elements like tungsten and gold, attributing them to structural influences near the absorbing atom but without a full theoretical framework. Concurrently, Walther Kossel described the near-edge fine structure—now known as Kossel structures—as arising from transitions of photoelectrons to unoccupied levels in neighboring atoms, providing an early interpretation of these features as evidence of atomic coordination. However, these observations were constrained by the low flux and resolution of available X-ray sources, preventing reliable detection of extended oscillations beyond about 50 eV above the edge. In 1931–1932, Ralph de Laer Kronig advanced a foundational theoretical model for the extended fine structure, proposing that the oscillations result from interference between the outgoing photoelectron wave and those backscattered by surrounding atoms, akin to standing waves confined within the potential wells of the solid's short-range order. This quantum mechanical approach shifted focus from long-range crystalline periodicity to local atomic environments, predicting periodic modulations in absorption that depend on interatomic distances and scattering phases, though experimental verification remained challenging due to signal-to-noise limitations. Post-World War II advancements in the 1960s and 1970s revitalized the field through refined instrumentation and reinterpretations. Farrel W. Lytle adapted commercial X-ray diffractometers with fluorescence detection to capture clearer extended fine structure spectra from dilute samples, coining the term "EXAFS" in 1965. Collaborating with Dale E. Sayers and Edward A. Stern, they introduced Fourier transform analysis in 1971 to extract radial distribution functions directly from the oscillatory data, confirming Kronig's interference mechanism and enabling quantitative structural determinations without single crystals. These efforts, initially using laboratory sources, established EXAFS as a probe for disordered systems, setting the stage for synchrotron-enhanced precision.

Key Milestones and Modern Evolution

The advent of synchrotron radiation in the 1970s marked a pivotal shift for EXAFS, enabling high-intensity, tunable X-ray beams essential for detailed spectroscopic measurements. The Stanford Positron Electron Accelerating Ring (SPEAR) at the Stanford Synchrotron Radiation Laboratory (SSRL) became operational in 1972, facilitating the first dedicated EXAFS experiments that overcame the limitations of laboratory X-ray sources. These early measurements, including the initial XAS spectrum recorded in 1974, demonstrated the technique's potential for probing local atomic structures with unprecedented resolution.^[47] Concurrently, Edward A. Stern's 1974 single-scattering theory provided a foundational framework for interpreting EXAFS oscillations as backscattering from neighboring atoms, simplifying data analysis and establishing the core methodology still used today. Theoretical advancements in the 1980s and 1990s enhanced the accuracy of EXAFS modeling by incorporating ab initio calculations and multiple-scattering effects. The development of the FEFF code in the mid-1980s by John J. Rehr and colleagues introduced parameter-free multiple-scattering computations for EXAFS phase shifts and amplitudes, allowing reliable predictions without empirical fitting for a wide range of materials.^[48] Building on this, the GNXAS package in the 1990s, pioneered by Augusto Filipponi and Antonio Di Cicco, extended the approach to full multiple-scattering analysis, improving structural refinements for disordered and complex systems by accounting for higher-order scattering paths. Instrumentation milestones in the 1990s and 2000s expanded EXAFS capabilities for dynamic studies. The commissioning of third-generation synchrotrons, such as the European Synchrotron Radiation Facility (ESRF) in 1994, delivered X-ray beams with brilliance orders of magnitude higher than previous sources, enabling routine high-throughput EXAFS on dilute samples and in operando conditions.^[49] Quick-EXAFS (Q-EXAFS) techniques, first demonstrated in the late 1980s, advanced in the 1990s and 2000s to achieve sub-second acquisition rates through rapid monochromator scanning, which facilitated investigations of transient structural changes in catalytic and photochemical processes.^[50] From the 2010s to 2025, innovations have democratized EXAFS access and refined data processing. Laboratory-based systems using laser-plasma X-ray sources emerged around 2015, providing compact, ultrafast alternatives to synchrotrons for time-resolved EXAFS with pulse durations below 100 fs, suitable for studying femtosecond dynamics in materials. Integration of artificial intelligence and machine learning in data analysis, particularly since 2020, has automated spectral fitting and structural prediction, reducing processing time from hours to minutes while handling noisy datasets from in situ experiments; as of 2025, frameworks like XASDAML have further enhanced predictive capabilities for coordination numbers and radial distances.^[51] The ESRF's Extremely Brilliant Source (EBS) upgrade, completed in 2020, further boosted beam coherence and stability by a factor of 100, enhancing EXAFS sensitivity for nanoscale and low-concentration analyses, with ongoing operando applications in battery research.^[52]

References

[1]
Extended x-ray absorption fine structure---its strengths and ...
Oct 1, 1981 · The authors review the development of extended x-ray absorption fine structure (EXAFS) within the last decade.Missing: paper | Show results with:paper
[2]
Theoretical approaches to x-ray absorption fine structure
Jul 1, 2000 · The authors focus on extended x-ray absorption fine structure (EXAFS) well above an x-ray edge, and, to a lesser extent, on x-ray absorption ...
[3]
[PDF] Fundamentals of XAFS - Lehigh University
X-ray absorption fine structure (XAFS) refers to the details of how x-rays are absorbed by an atom at energies near and above the core-level binding ...Missing: aspects | Show results with:aspects
[4]
EXAFS: New Horizons in Structure Determinations - Science
Abstract. Although the phenomena of extended x-ray absorption fine structure (EXAFS) were observed as early as the 1930's, EXAFS has only recently, with ...
[5]
[PDF] Fundamentals of XAFS - CARS
XAFS is an element-specific spectroscopy in which measurements are made by tuning the X-ray energy at and above a selected core-level binding energy of a ...
[6]
https://doi.org/10.2138/rmg.2014.78.2
[7]
X‐ray absorption spectroscopy—XAS, XANES, EXAFS - Iglesias‐Juez
Aug 15, 2021 · The main advantage of XAFS is that it is element specific and sensitive to the oxidation state and local (fine) structure (i.e., it reveals ...
[8]
https://onlinelibrary.wiley.com/doi/10.1002/cjce.24291
[9]
The Historical Development of X-ray Absorption Fine Spectroscopy and of Its Applications to Materials Science
### Summary of Historical Evolution of EXAFS
[10]
Theory of the extended x-ray-absorption fine structure | Phys. Rev. B
Oct 15, 1974 · Theory of the extended x-ray-absorption fine structure. Edward A. Stern. Department of Physics, University of Washington, Seattle, Washington ...Missing: paper | Show results with:paper<|control11|><|separator|>
[11]
[PDF] X-ray Absorption Fine-Structure Spectroscopy
Jul 7, 2019 · Development of the EXAFS Equation χ(k) = f (k). kR2sin[2kR + δ(k)]. The EXAFS Equation for 1 scattering atom. Theory / EXAFS Equation. M ...
[12]
[PDF] Lecture Notes on Extended X-ray Absorption Fine Structure (EXAFS)
Mar 16, 2023 · In the seminal paper of 1976 [Beni and Platzman, 1976], Beni and Platzman introduce the correlated Debye model and evaluate the ratio DCF ...
[13]
None
### Summary of EXAFS Mathematical Formulation from https://www.esrf.fr/files/live/sites/www/files/events/conferences/2017/User_Meeting_2017/Part%20I%20.pdf
[14]
[PDF] Basics of Synchrotron Radiation Beamlines and Detectors
Basics of Synchrotron Radiation Beamlines and Detectors. •Basics of synchrotron radiation. •X-ray optics as they apply to EXAFS experiments. •Detectors. Page 2 ...
[15]
X-Ray Absorption Fine Structure (XAFS) Spectroscopy – A Review
This review describes the basic phenomenon of XAFS, theory of EXAFS and the method of extracting structural parameters by EXAFS which relate to the local ...
[16]
None
### Summary of Lab-Based Laser-Plasma X-ray Sources for EXAFS (5-30 keV)
[17]
An improved laboratory-based x-ray absorption fine structure and x ...
This element-specific technique encompasses both the x-ray absorption near edge structure (XANES), an acutely sensitive probe of a compound's oxidation state ...
[18]
14.2. XAFS: Pre-edge Subtraction, Normalization, and data treatment
After reading in data and constructing μ ( E ) , the principle pre-processing steps for XAFS analysis are pre-edge subtraction and normalization. Reading data ...Missing: background | Show results with:background
[19]
[PDF] A method for normalization of X-ray absorption spectra
The paper describes MBACK, a method to normalize X-ray absorption data to mass absorption coefficients, which is not sensitive to background shape.Missing: preprocessing removal
[20]
[PDF] XAS Data Processing - The International X-ray Absorption Society
XAS data processing includes aligning/merging EXAFS signals, normalizing spectra, converting to wavenumber, Fourier transforms, and using background functions.Missing: preprocessing | Show results with:preprocessing
[21]
SIXPack Documentation
SIXPack is a software package for XAS analysis, unifying SamXAS and SamView, guiding data averaging, calibration, background removal, and fitting.
[22]
X-ray Absorption Near-Edge Structure (XANES) Spectroscopy
Jan 1, 2014 · In addition to the XANES region, at higher energies the extended X-ray absorption fine structure (EXAFS) region is found. The spectral shape ...
[23]
[PDF] Sources of Noise in EXAFS Experiments
Aug 11, 1988 · Noise in EXAFS experiments has a number of sources, among them unavoidable ran- dom fluctuations in the number of photons absorbed by the ...
[24]
[PDF] Basics of EXAFS Processing
Fourier Transform of a function that is: De-localized in k-space ⇒ localized in R-space. Localized in k-space ⇒ de-localized in R-space. Page 14. 14. File ...
[25]
(PDF) EXAFS analysis using FEFF and FEFFIT - ResearchGate
Some of the advanced EXAFS analysis features of FEFF and FEFFIT are described. The scattering path formalism from FEFF and cumulant expansion are used as the ...
[26]
[PDF] Basics of EXAFS Data Analysis
EXAFS data analysis involves data processing, modeling, background subtraction, and using Artemis as an interface to Ifeffit for theoretical calculations.Missing: procedures | Show results with:procedures
[27]
[PDF] Scott Calvin
Jun 30, 2011 · – Change the k-range of data being used. – Change the R-range of the Fourier transform being fit. – Change the k-weight. – Remove a constraint.
[28]
[PDF] Toward Quantitative Calculation and Analysis of X-Ray Absorption ...
The oscillatory EXAFS signal χ(k) is then Fourier transformed to real (R) space which gives an amplitude that is akin to a radial distribution function, with ...
[29]
[PDF] EXAFS Data Analysis with FEFF - CARS
EXAFS data analysis with FEFF uses the EXAFS equation to model data, refining parameters like R, N, and σ2 to match data. The best fit is determined by the ...Missing: matrices | Show results with:matrices
[30]
[PDF] EXAFSPAK - Stanford Synchrotron Radiation Lightsource
Gaussian pre-edge function can be used to remove this. The polynomial pre-edge subtraction works by fitting a polynomial of order. -1 in our example through ...
[31]
First shell EXAFS data analysis of nanocatalysts via neural networks
We introduce a neural-network-based framework for rapidly mapping the extended XAFS (EXAFS) spectra onto structural parameters.
[32]
[PDF] Towards a Machine Learning Approach to Interpreting X-ray Spectra ...
Machine learning uses DeepONets to predict EXAFS from XANES, showing the near-edge region contains EXAFS structural information. This approach is suited for ...
[33]
Structural anisotropy in amorphous SnO2 film probed by X-ray ...
Jul 17, 2013 · Polarization-dependent X-ray absorption measurements reveal the existence of structural anisotropy in amorphous (a-) SnO2 film.
[34]
Combined (Q)EXAFS/XRD: Technique and applications
XRD and EXAFS can be regarded as complimentary techniques since XRD probes structures with long-range order whereas EXAFS is sensitive mainly to the local ...
[35]
EXAFS studies of the local order in amorphous and crystalline nickel ...
EXAFS studies of the local order in amorphous and crystalline nickel-zirconium alloys. I. Principles and evaluation of the crystalline alloys.
[36]
X-ray Absorption Fine Structure (XAFS) Studies of Oxide Glasses—A ...
X-ray Absorption Fine Structure (XAFS) spectroscopy has been widely used to characterize the short-range order of glassy materials since the theoretical basis ...
[37]
Extended X-ray absorption fine structure determination of iron ...
Jul 1, 1978 · EXAFS spectra have been obtained of oxy and deoxy complexes of haemoglobin and of the 'picket fence' porphyrin, using synchrotron radiation ...
[38]
Spinach Carbonic Anhydrase: Investigation of the Zinc-Binding ...
Spinach Carbonic Anhydrase: Investigation of the Zinc-Binding Ligands by Site-Directed Mutagenesis, Elemental Analysis, and Exafs. Click to copy article link ...
[39]
X-ray Spectroscopies Revealing the Structure and Dynamics of ...
Feb 2, 2011 · Reactions catalyzed by metalloproteins occur at the active centers, making them the focus of many spectroscopic investigations in an attempt ...
[40]
X-ray absorption fine structure study of As(V) and Se(IV) sorption ...
X-ray absorption fine structure (XAFS) spectroscopic analysis at the As, Se, and Mn K-edges was used to study arsenate [As(V)O43−] and selenite [Se(IV)O32−] ...
[41]
X-ray Absorption Spectroscopy of Lead and Zinc Speciation in a ...
The formation of insoluble metal sulfides in the environment may reduce the mobility and bioavailability of heavy metal contaminants and potentially ...
[42]
Uranium Biogeochemistry in the Rhizosphere of a Contaminated ...
The objective of this study was to determine if U sediment concentrations in a U-contaminated wetland located within the Savannah River Site, South Carolina ...
[43]
(IUCr) Using XAS to monitor radiation damage in real time and post ...
However, X-ray absorption spectroscopy (XAS) applications with biological systems are challenging due to the sensitivity of the sample. Achieving high ...
[44]
Metalloprotein active site structure determination: Synergy between ...
Here we review the potential of X-ray absorption spectroscopy to provide information and perspective which could aid in improving the accuracy of metalloprotein ...
[45]
The Historical Development of X-ray Absorption Fine Spectroscopy ...
This essay sketches the development of X-ray Absorption Fine Spectroscopy (XAFS) ever since the second half of 20th century.
[46]
The FEFF9 code - The FEFF Project - University of Washington
The code yields scattering amplitudes and phases used in many modern XAFS analysis codes, as well as various other properties. FEFF 9 (Version 9.6): FEFF 9 has ...Missing: 1980s | Show results with:1980s
[47]
History - European Synchrotron Radiation Facility (ESRF)
In 1988, the ESRF made history as the world's first third-generation synchrotron light source. The ESRF served as a model for many synchrotrons worlwide.
[48]
Quick scanning exafs: First experiments - ScienceDirect.com
By speeding up a conventional X-ray monochromator by a factor of 100 using the quick scanning EXAFS (QEXAFS) mode the data collection time required for a ...
[49]
ESRF-EBS - European Synchrotron Radiation Facility (ESRF)
ESRF–EBS (Extremely Brilliant Source) is the ESRF's 150M€ facility upgrade, spread over the period 2015-2022, bringing its scientific users a first-of-a ...Missing: EXAFS | Show results with:EXAFS