X-ray photoelectron spectroscopy (XPS) measures the kinetic energies of electrons emitted from a specimen under X-ray irradiation. It is widely used to investigate which elements are present near a surface and how their chemical environments affect observed electron energies.
The gold standard links a calibrated electron spectrum to a defensible surface-chemistry claim. The analyst must distinguish elemental peaks, energy referencing, charging, background subtraction and sensitivity corrections before reporting oxidation states or atomic proportions.
eduKateSG · X-ray Photoelectron Spectroscopy
Find your next learning step
Use photoelectric energy accounting, correct illustrative peak areas, then test depth, charging and chemical-state assumptions.
Full chapter index · Practice and transfer · Education root · How X Works
Full chapter index
Define the surface evidence · Chapters 1–2
Calculate the spectrum · Chapters 3–4
Check calibration and depth · Chapters 5–7
Interpret and transfer · Chapters 8–10
Read across the ecosystem · Chapters 11–12
CHAPTER 1 OF 12 · Define the surface evidence
1. XPS measures escaping electrons, not an entire bulk object
An incident X-ray photon can transfer energy to an electron, which may escape the sample if it reaches the surface without losing too much energy. XPS instruments analyse the electron kinetic-energy distribution under controlled conditions.
The resulting spectrum contains peaks associated with electron levels of elements and sometimes distinguishable chemical states. The measurement is surface sensitive because electrons travelling through a solid can undergo inelastic scattering before reaching vacuum.
Routine XPS often probes information from the outer few nanometres to roughly several nanometres, depending on energy, material and geometry. The phrase surface sensitive does not mean exactly one atomic layer or one universal depth.
A surface measurement is not automatically representative of the bulk. Oxide, adventitious contamination, adsorbates or a thin coating may dominate the interpretation while the underlying material has different composition.
State the sample condition, whether it was exposed to air, the analysed area, angle and instrument configuration. Without that context the spectrum’s scientific question is underspecified.
Core-level peaks provide characteristic binding-energy regions associated with elements. The NIST XPS database compiles many reference energies and chemical shifts, but its records must be used with an awareness of specimen conditions and spectral referencing.
A peak near a familiar textbook position is evidence worth testing, not conclusive proof of a precise compound. Overlapping lines, Auger features, energy loss structures, instrumental effects and chemical shifts can produce ambiguity.
Survey spectra help identify likely elements; high-resolution regions may then help discriminate chemical states under a physically justified fit. A peak decomposition is a model with parameters, not a direct observation of separate chemical species.
XPS detects emitted electrons, which distinguishes it from X-ray fluorescence, where emitted X-ray photons are recorded. X-ray diffraction asks about crystalline order rather than simply the binding energies of near-surface electrons.
Quantification and chemical-state assignment require their own evidence chain. An analyst should report raw region, corrected energy reference, fit assumptions and alternative explanations.
CHAPTER 3 OF 12 · Calculate the spectrum
3. Worked example: convert kinetic energy into binding energy
In a simple calibrated photoelectron-energy account, the binding energy relative to the chosen spectrometer reference is EB = hν − EK − Φspec, with photon energy hν, measured electron kinetic energy EK, and an appropriately defined spectrometer work-function term Φspec.
Take illustrative photon energy hν = 1486.6 eV, kinetic energy EK = 950.0 eV and spectrometer term Φspec = 4.5 eV. Then EB = 1486.6 − 950.0 − 4.5 = 532.1 eV.
All three quantities are energies expressed in electronvolts. The number is chosen for teaching and does not certify that an observed feature is oxygen, a particular oxide or an exact oxidation state.
The equation assumes the stated energy reference and a properly calibrated analyser. On an insulating surface, charging can shift apparent peak positions; different referencing strategies can also change reported energies.
To transfer the method, first draw a labelled energy diagram. Then test whether the assumptions needed to interpret a peak are supported by reference measurements rather than concealed inside the arithmetic.
CHAPTER 4 OF 12 · Calculate the spectrum
4. Worked example: relative sensitivity-factor correction
Suppose two well-separated peaks from a hypothetical homogeneous surface give fitted background-corrected areas A₁ = 1,200 and A₂ = 800 in the same arbitrary area units. Their relative sensitivity factors for the stated instrument method are S₁ = 2 and S₂ = 1.
The simple normalized intensity terms are A₁/S₁ = 600 and A₂/S₂ = 800. Their sum is 1,400, so estimated relative atomic fractions are x₁ = 600/1,400 ≈ 42.9% and x₂ = 800/1,400 ≈ 57.1%. They add to 100% by construction.
These are not mass percentages, not a bulk assay and not generally absolute atomic concentrations. The result assumes suitable sensitivity factors, an adequately fitted background and matrix or geometry conditions compatible with the model.
Using uncorrected peak areas would instead give 1,200/(1,200 + 800) = 60% for the first element. This difference illustrates why raw peak height or area is not a sufficient composition estimate.
NIST’s discussion of ISO 18118:2024 emphasises standardised sensitivity factors and matrix effects. Report the chosen RSF source, corrections, uncertainty and excluded elements rather than merely the final percentages.
CHAPTER 5 OF 12 · Check calibration and depth
5. Calibration and energy referencing govern chemical-state claims
Binding-energy calibration must be distinguished from correcting an unknown sample for charging. Calibration checks the instrument energy scale against suitable references; charge correction attempts to account for shifts associated with the analysed sample.
ISO 15472:2010 addresses calibration of XPS binding-energy scales for specified instrument conditions. A standard has a defined applicability range: it is not enough to mention its number without checking whether the instrument and preparation meet the stated scope.
NIST work on energy-scale checks highlights the importance of measuring and reporting drift, linearity and calibrated reference points. A precisely printed peak position is not necessarily a precise measurement.
Surface charging can be spatially or chemically nonuniform. A single uniform shift may not correct differential charging across insulating domains, and a conventional reference peak can itself vary with contamination or chemical context.
If an oxidation-state conclusion hinges on a small shift, test alternative line assignments, reference choices, background models and charge effects. Uncertainty and plausible alternatives belong next to the peak table.
CHAPTER 6 OF 12 · Check calibration and depth
6. Depth sensitivity depends on electron transport and geometry
The probability that an electron escapes without inelastic energy loss decreases with the distance travelled in the solid. The inelastic mean free path, emission angle and material all affect which depths contribute strongly.
An ideal flat, homogeneous overlayer model has substrate attenuation approximately I/I₀ = exp[−d/(λ cos θ)], where d is film thickness, λ is an appropriate effective attenuation length and θ is the emission angle relative to the surface normal.
For an illustrative normal-emission case θ = 0°, let d = 2 nm and λ = 2 nm. Then I/I₀ = exp(−1) ≈ 0.368. At θ = 60°, cos θ = 0.5 gives exp(−2) ≈ 0.135 under the same simple assumptions.
These values describe a stylized model, not a thickness result for any real coating. Surface roughness, elastic scattering, nonuniform thickness, chemical variation and the correct attenuation-length definition can change the inference.
Angle-resolved XPS and sputter-depth profiling pose additional inverse and preparation questions. Sputtering can preferentially remove elements or alter chemical states, so the resulting depth profile is not always the untouched original material.
For an explicit reconstruction limit, work through two angle-resolved signals that admit different depth profiles. The example checks physical bounds, a third angle and the assumptions behind a selected profile.
For a worked lateral-coverage model, see why a uniform-film fit can misread a patchy overlayer. It distinguishes local thickness, area-average coating and an apparent uniform equivalent, then checks a conditional two-angle inverse.
CHAPTER 7 OF 12 · Check calibration and depth
7. Peak areas are constrained by instrument settings and fitting
The spectrum depends on pass energy, analyser transmission, lens settings, X-ray source, detector response and scan conditions. A comparison of intensities across instruments requires compatible sensitivity and calibration treatment.
Peak fitting usually involves an energy window, background model, line shape and constraints on shared doublet parameters where relevant. A plausible mathematical fit can still be chemically wrong if the model permits impossible components.
Separate peak detection from peak attribution and quantification. Good practice preserves the measured spectrum, displays the fit and residuals, and documents decisions about overlapping transitions.
ISO 5861:2024 addresses intensity calibration for defined quartz-crystal-monochromated Al Kα instruments and conditions. It should not be casually generalized beyond its scope.
Revisit reference specimens and background choices when apparent chemical shifts or composition changes are close to method uncertainty. A figure with smooth curves cannot replace a calibration record.
CHAPTER 8 OF 12 · Interpret and transfer
8. Thin films, oxides and contamination demand competing explanations
Consider a metallic component measured after exposure to room air. A surface oxide peak may be genuine, but it does not by itself establish oxide thickness, continuity, bulk corrosion extent or the failure mechanism of the component.
A coating can attenuate substrate signals and create an apparent composition dominated by its outer layer. If the question concerns an interface, one surface spectrum may be underdetermined without an angular series, sputter history or complementary method.
Hydrocarbon contamination can affect both apparent composition and a chosen referencing procedure. Handling, packaging, humidity and premeasurement cleaning should be documented without assuming that every cleaning method is chemically harmless.
Explain what sample area is actually illuminated and what lateral heterogeneity is unresolved. A local spectrum taken over one patch may not represent a whole manufactured batch.
Independent microscopy, elemental mapping, diffraction or controlled chemical preparation may help discriminate alternatives, but each method brings different sampling depths and limitations.
CHAPTER 9 OF 12 · Interpret and transfer
9. Surface chemistry and bulk structure require different evidence
XPS is well suited to questions about surface elemental composition and some chemical states. It is generally not a direct, standalone measure of bulk crystal structure, three-dimensional pore geometry or full sample stoichiometry.
X-ray fluorescence can probe characteristic photon emission and often different sampling depths. X-ray absorption spectroscopy interrogates energy-dependent absorption, while X-ray diffraction uses interference to constrain ordered structures.
Computed tomography reconstructs spatial attenuation contrast. Combining it with XPS may be useful when the question concerns both geometry and surface state, but spatial co-registration and specimen compatibility must be considered.
A rigorous interpretation says which part of the mechanism is directly supported by XPS and which is a hypothesis transferred from other evidence. Avoid the phrase proves the material is X when multiple phases or states can match the recorded lines.
Reproducibility includes the instrument condition and sample history. An apparently small difference between two batches can be due to storage, handling or acquisition settings rather than manufacture.
CHAPTER 10 OF 12 · Interpret and transfer
10. Practice, FAQs and a changed-sample test
Practice A: If the illustrative kinetic energy increases from 950.0 eV to 952.0 eV with hν and Φspec unchanged, the calculated binding energy decreases from 532.1 eV to 530.1 eV. Explain why a charged sample might break a naive chemical interpretation of this difference.
Practice B: Recalculate relative atomic fractions when the first corrected peak area is halved. Write down exactly which normalization and matrix assumptions you retain.
FAQ — Is XPS a bulk chemical assay? Usually not. Conventional XPS is surface sensitive, so the analysed region may differ from the interior.
FAQ — Does one energy peak specify one oxidation state? Not automatically. Charging, overlaps, fitting and closely spaced chemical shifts must be evaluated with references and context.
FAQ — Can an XPS peak area be read directly as atomic percentage? No. Sensitivity factors, background correction, sampling and matrix effects enter the estimate.
The transfer task is to move from a clean metal specimen to a polymer-coated semiconductor. List which referencing, charging, depth and beam-effects assumptions require new evidence.
CHAPTER 11 OF 12 · Read across the ecosystem
11. Connected chemistry, X-ray and materials-reading routes
To interpret surface spectra well, connect the chemistry root to X-ray absorption and diffraction as separate measurement branches. Those fields provide complementary evidence but cannot retroactively validate an inadequately calibrated XPS peak.
The next reading route is decided by the physical question: near-surface chemistry, crystalline arrangement, internal geometry or energy-dependent absorption. Use the right instrument evidence for each.
Follow the connected eduKateSG articles according to the next question you need to answer. Links lead to distinct measurements or methods, not interchangeable evidence.
- The Gold Standard of Spectroscopy — spectral interpretation across methods
- The Gold Standard of Chemistry — chemical bonds, states and reactions
- The Gold Standard of X-ray Absorption — different mechanisms of energy-dependent absorption
- The Gold Standard of X-ray Fluorescence — distinguishing emitted X-ray photons from photoelectrons
- The Gold Standard of X-ray Diffraction — crystalline order and structural evidence
- The Gold Standard of X-ray Computed Tomography — interior attenuation geometry, distinct from surface chemistry
- The Gold Standard of Physics — the photoelectric effect, electrons and energy accounting
CHAPTER 12 OF 12 · Read across the ecosystem
12. Primary databases, standards and next learning step
All energies, intensities, sensitivity factors and film parameters in the worked examples are invented as explicit teaching cases. They are not reference peak assignments, calibration constants or chemical analyses of real samples.
Choose one published spectrum and assemble an evidence table: specimen handling, source energy, charge compensation, calibration, selected peaks, fitting approach, sensitivity factors, angle, sampling depth, uncertainty and plausible alternative assignments.
Primary sources and technical references support the instrument physics, uncertainty and interpretation boundaries:
- NIST XPS Database, SRD 20 and surface-data resources — reference binding energies, chemical shifts and source conditions
- Powell (2019), Calibrations and Checks of XPS Binding-Energy Scales, NIST — calibration and drift of reported energies
- ISO 15472:2010, XPS Calibration of Energy Scales — defined procedure and applicability for binding-energy calibration
- NIST, Standard Approaches to XPS Quantification and ISO 18118:2024 — relative sensitivity factors and matrix-effect concerns
- AIP Journal of Applied Physics, A Step-by-Step Guide to Perform XPS — surface sensitivity, acquisition and interpretation methodology
- ISO 5861:2024, Method of XPS Intensity Calibration — intensity-scale calibration within specified instrument conditions
None of the hypothetical peak energies should be used as a chemical identification. Instrument-specific referencing, charging and quantitative protocols determine which conclusions an actual measurement can support.
The instructional standard is to define a measurable question, show a worked method and challenge it with a changed context. The immutable Clementi Secondary 1 Mathematics teaching reference illustrates that educational discipline but is not a specialist technical reference for X-ray Photoelectron Spectroscopy.