X-Ray Photoelectron Spectroscopy (XPS; aka Electron Spectroscopy for Chemical Analysis, ESCA)

David Mogk, Imaging and Chemical Analysis Laboratory, Montana State University

What is X-Ray Photoelectron Spectroscopy?

X-ray photoelectron spectroscopy (XPS) is a surface sensitive, non-destructive technique used routinely to analyze the outermost ~10 nm (~30 atomic layers) of natural and engineered materials. XPS is routinely used to determine a) the composition of material surfaces (elemental identification), the relative abundances of these components on surfaces (semi-quantitative analysis), and c) the chemical state of polyvalent ions by measuring the binding energies of elements, which is related to the nature and strength of their chemical bonds. XPS is used to characterize the surfaces of diverse materials such as inorganic compounds (minerals), semiconductors, organic compounds, and thin films and coatings on natural and engineered materials. XPS is used to support research on surface-mediated processes such as sorption, catalysis, redox, dissolution/precipitation, corrosion, and evaporation/deposition type reactions. It is almost always the case that the surface composition and chemistry of materials, measured on the order of a few atomic layers (~10 nm), is different from the "bulk" composition determined by methods such as energy dispersive spectrometry (EDS) with excitation volumes that can extend as much as 3 microns into the material.

Fundamental Principles of X-Ray Photoelectron Spectroscopy

XPS is an application of the photoelectric effect (Acrobat (PDF) 184kB Jul29 21) described by Einstein (1905, and was awarded the Nobel Prize in 1921), in which electrons are emitted from atoms in response to impinging electromagnetic radiation. Einstein predicted that photoelectrons would be produced from a material when the energy of impinging photons exceed the binding energy of electrons in that material; the energy is proportional to the frequency (ℎν ) not the intensity or duration of exposure to the incident electromagnetic radiation. The kinetic energy of an emitted electron is related to the binding energy of each electron, and because atoms have multiple orbitals at different energy states, the resulting response will be a range of emitted electrons with different binding energies (and kinetic energies) thus producing an XPS spectrum.

These relations are represented by the equation:

E_kinetic = E_{photon (ℎν)} - E_{binding -} φ

where E_kinetic is the kinetic energy of the photoelectron measured by the instrument, E_photon is the energy of the incident photon (X-ray in this case, which is a known and fixed value), E_binding is the binding energy of a given electron, and φ is the work function, the energy difference between the vacuum energy (Ev) level and the Fermi (Ef) level of a solid.

Dr. Kai Siegbahn and colleagues from Uppsala University in Sweden recognized the potential of using photoelectrons for chemical analysis (thus, Electron Spectroscopy for Chemical Analysis) and was awarded the Nobel Prize for Physics in 1981 for these contributions.

X-Ray Photoelectron Spectroscopy Instrumentation--How Does it Work?

Instrumentation

×

Schematic diagram of the components of an XPS Instrument.
Provenance: Diagram from Malgorzata Kot, PhD dissertation.
Reuse: This item is offered under a Creative Commons Attribution-NonCommercial-ShareAlike license http://creativecommons.org/licenses/by-nc-sa/3.0/ You may reuse this item for non-commercial purposes as long as you provide attribution and offer any derivative works under a similar license.

XPS instruments have the following components:

Ultrahigh vacuum system; typically operating conditions are at <10^-9 Torr. This is required because the emitted photoelectrons have a relatively low energy and are readily absorbed by ambient atmosphere. In other words, photoelectrons have a relatively small mean free path between sample and detector as they are readily sorbed by gas molecules in the chamber.
X-ray source; Al Kα of Mgα X-rays are typically used to excite the sample; a monochrometer is used to permit only X-rays of this fixed energy to impinge on the sample.
An electron energy analyzer is used to discriminate among the energies of the photoelectrons that are produced. This is typically a Concentric Hemispherical Analyzer (CHA).
Ar ion gun is used a) to sputter material surfaces to "dust off" environmental contaminants (e.g., sorbed adventitious carbon from the atmosphere) off of material surfaces, and b) to obtain depth profiles across surface layers on the material surface.
Charge neutralization capability using an electron "flood gun" is used to minimize surface charging under the X-ray beam.
Computer and data reduction software. XPS data are compared to inventories or archives of experimentally determined XPS data of standard reference materials such as the Handbook of X-ray Photoelectron Spectroscopy, edited by J Chastain, with authors J.F. moulder, W. F. STickle, P. E. Sobol and K. D. Bomben (PHI Perkin-Elmer Corporation, 1992) or the NIST X-ray Photoelectron Spectroscopy Database.

Data Representation

XPS data are presented as spectra that plot Binding Energy (eV) on the X-axis vs. measured photoelectron counts on the Y-axis. Data are typically collected in a) Survey mode to obtain the complete inventory of elements on a material surface, and b) high resolution scans of peaks of interest to reveal the bound state (chemical bonds) involving elements of interest. The binding energies of the numerous photoelectrons emitted from a surface sample are used as a "fingerprint" to identify elements present. Chemical shifts in XPS spectra are observed when an element enters a different bound state, which results in changes in the binding energy of core electrons. In general, increased oxidation state (removal of valence electrons) increases the Binding Energy and addition of valence electrons decreases the Binding Energy.

For electrons in p, d, or f orbitals, two peaks are observed and the separation of these peaks is known as spin orbit splitting. The energy difference (in eV) and the ratio of areas under these split peaks is used to confirm the identity of the elements. Nomenclature for identifying peaks follows this form, nlj, where n is the period or principle quantum number, l is the type of orbital (with values l =0 for S, l=1 for P, l=2 for D, and l=3 for F orbitals), and j represents the spin angular momentum number where j = l + s and s is ± 1/2. Thus, XPS peaks for As will be represented as 3s (a single peak)., 3p_1/2 or 3p_3/2, or 3d_3/2 or 3d_5/2. The spin-orbit splitting ratio is 1:2 for p levels, 2:3 for d levels and 3:4 for f levels. Note that Auger electron peaks are also generated, and these peaks are labeled as the LMM transition.

Reference XPS spectra for arsenic are presented below from the PHI Electronics Handbook of X-ray Photoelectron Spectroscopy.The spectrum on the left shows the survey of elemental arsenic. The binding energies of the As photoelectrons and Auger electrons are shown in the table in the lower right. The table at the bottom of the figure on the right shows the changes in Binding Energies for As occurring in different chemical species. The resolution of the Binding Energies is typically ±0.1 eV, so shifts of an eV or more are significant in interpreting the bound state of the element of interest. In general, reduced states of elements have a lower binding energy than their oxidized states.

For complex materials, such as organic compounds that have a variety of bond types (e.g., C-C, C=C, C-O, C-H, C-N, etc.), it is often necessary to obtain a high resolution scan and then deconvolute the spectrum to determine the binding energies that correspond to the various bond types, as numerous peaks will commonly contribute to a composite asymmetrical peak with numerous shoulders. (See videos on Curve Fitting prepared by Surface Science Western for more examples).

An example of use of XPS to characterize the organic compound polyethylene tetrephthalate (PET) is provided by PHI Electronics, Technical Application, Complementary XPS and TOF-SIMS for Organic Analysis . The figure below shows two important features: 1) the presence of surface contaminants on the PET as it has been analyzed "as received" with no special treatment used to remove surface contaminants, and b) peak shifts of the C 1s peak related to different bonds that are present in the PET and surface C compounds (C-O, C-H, C-C).

Charge Compensation

Insulating samples may provide challenges to XPS analysis due to build up of surface charge. Metals and other conducting materials will be able to continuously transfer electrons to the surface to replace electrons loss due to photoelectron and Auger processes. However, the loss of these electrons on the surfaces of insulators results in positive charge build up as photoelectrons are emitted, and these photoelectrons may exhibit a shift to slightly higher binding energies as a result. This is generally compensated by using an electron flood gun to replace the lost photoelectrons. In addition, it is also advantageous to correct for slight shifts in the photoelectron spectrum due to charging by calibrating to the C 1s peak (from adventitious, surface C) at 284.8 eV.

Sample Depth

The intensity of photoelectrons emitted at the surface (I_s) is determined by the Beer-Lambert Law: I_s = I_oe^-d/λ where I_ois the intensity of the photoelectrons emitted at depth d below the surface and λ is the inelastic mean free path of the electron in the material. Most λ's are on the order of 1-3.5 nm for AlKα X-rays, so the sampling depth is typically3-10 nm.

Sputtering and Depth Profiling

Materials exposed to the atmosphere almost universally will have atmospheric contaminants such as adventitious carbon on their surfaces. It is common practice to use an argon ion gun to gently "dust off" these surfaces to expose the actual materials of interest.

Chemical depth profiles can be obtained by sputtering with the Ar beam to monitor changes in the surface composition and chemical state of elements of interest. The sputter rate can be calibrated with standards of known thickness (e.g., Si wafer with SiO thin film). Sputtering can either be done continuously or sequentially. Typically specific peaks of elements of interest are monitored to demonstrate relative changes in abundance or peak shifts in the binding energies that indicate changing bound states as a function of depth.

A representative depth profile is displayed in the image on the right, produced by Nathaniel Rieders, ICAL Montana State University. This sample has a thin film of B deposited on carbon steel. A number of interesting features are displayed: 1) There is a very thin layer of sorbed oxygen on the outermost surface of this sample, and also at the interface between the B thin film and the iron substrate; 2) analysis of B and O demonstrate the capability of analyzing light elements; 3) the sputter rate was calibrated at ~2 nm/cycle, so the thickness of the B layer (15 cycles) has been measured as 30 nm. The data are presented as atomic percent (Y-axis) and all represented elements are normalized to sum to 100 atomic %. This depth profile was produced by selecting only the peaks of interest (referred to as a multiplex routine), and then sequentially sputtering with the Ar gun for a fixed duration followed by analysis of the elements of interest.

Additional factors that may affect the quality of depth profiles include: a) surface roughness of the sample, b) redeposition of sputtered species, c) "knock-on" effect that results in implantation or atomic mixing of surface components into the bulk, material; d) differential sputtering rates of different chemical components; and e) presence of crystal defects (dislocations) and underlying crystal structure.

Quantitative Analysis

Surveys of all elements present on a material surface can produce semi-quantitative analyses by measuring the area under each peak and applying appropriate elemental sensitivity factors (from published tables, determined experimentally or theoretically; e.g., Seah et al., 2001; Wagner, 1983; Battistoni et al., 1985; Handbook of X-ray Photoelectron Spectroscopy). A general expression for quantitative analysis is:

C_x=(I_x/S_x)/(ΣI_i/S_i)

where C_xis the concentration of the element of interest, I_xis the measured intensity of the element of interest, S_x is the elemental sensitivity factor, and ΣI_i/S_iis the sum of the ratios of intensities divided by the sensitivity factors of all other elements measured in the analysis.

Typically, atomic concentrations can be determined to ±10%.

For elements in numerous bound states on a material, high resolution scans that display deconvoluted spectra for a given element can be used to determine the relative abundances of each chemical state by integrating the area under the curve for each peak.

Applications of XPS

XPS is widely used in characterization of both natural and engineered materials.

Studies of surface-mediated reactions such as sorption, catalysis, REDOX, dissolution/preciptitation
Material Science
Polymers
Medical Devices
Thin Films and Coatings
Microelectronic Devices
Medical and Biological Samples
Geologic materials (see references below)

Strengths and Limitations of XPS

Strengths of XPS include:

Non-destructive analysis of materials.
Ability to detect all elements except for H and He. XPS surveys will obtain inventories of all elements present on material surfaces.
Small shifts in binding energies can be measured (~0.1 eV) that provides information about the bound state of elements present. these data are obtained by collecting spectra over limited energy range to reveal the fine structure of XPS spectra for a given element.
Surface-sensitive analysis to determine composition of material surfaces a few atomic layers thick (~ 10 nm).
Surface contaminants are usually easily removed using ion (Ar) beam sputtering methods.
Semi-quantitative analyses can be obtained within ± 10% atomic concentration.
Depth profiles may be obtained to demonstrate chemical stratigraphy on the nm scale on material surfaces.
Little or no sample preparation is required.
X-ray beams produce relatively little charging effects compared with electron beam methods; charging can easily be addressed with charge compensation methods using an electron flood gun.
Some modern XPS instruments have imaging capabilities.
Mature XPS databases exist to rapidly identify elements and their chemical state.

Limitations of XPS include:

XPS uses an ultrahigh vacuum chamber (< 10^-9 Torr) and some samples are either not stable or volatilize under UHV conditions.
X-ray beams cannot be focused in the same manner as electron beams; so the analyzed surface areas are significantly larger. Typically the analyzed area will be mm x mm or 10's or 100's of microns across at best--producing an averaged signal over these areas. Modern XPS instruments may have "small spot" capabilities, but this may be achieved by physically stepping down the beam size which reduces the count rate. Small spot XPS can be done via X-rays produced in a synchrotron source.
Although charge compensation is often effective, some samples may produce severe charging problems that compromise the quality of the analysis.
As a surface sensitive method, XPS is not an appropriate method for identifying bulk material substrates.

User's Guide--Sample Preparation, Data Acquisition, Data Representation, Reporting and Interpretation

Sample Preparation

Typically, samples used for XPS analysis are analyzed "as received" because any chemical treatments will leave a contaminating residue. Here are some practical tips:

Small samples are preferred, as desorption of surface gases will impact the ultrahigh vacuum. Similarly, introduction of volatile species (e.g., some types of epoxy mounts) is discouraged into the UHV chamber to help prevent contamination of the sample chamber and detectors.
If samples must be washed to remove surface contaminants, a light hydroacrbon solvent (e.g., hexane) is recommended.
Some XPS instruments are equipped with a LN2 "cold finger" and analysis at low temperatures can minimize release of volatile compounds,if these are surface components of interest.
Sputtering with an ion beam (e.g., Ar) is routinely used to remove surface contaminants such as adventitious carbon.
Some XPS instruments are equipped with a fracture stage in the sample chamber which produces fresh surfaces that have not been exposed to atmosphere.
Powders may be analyzed by XPS. Powders can be mounted for analysis on a) double-stick C tape, b) pressing the powder onto a thin Indium foil (which is both malleable and conducting), or c) suspending the powder in a liquid suspension and allowing this to dry on a conducting substrate such as a Si wafer. Be careful to make sure the powder is well-affixed to the sample holder because loose powder in the sample chamber can contaminate the chamber and other samples, and can ultimately degrade detectors.
Some types of samples (metals, alloys, minerals) may be mechanically polished. which can reduce complications due to surface charging and roughness.

Data Acquisition

Data acquisition is conducted in two modes, Survey or Detailed (high spectral resolution). The analytical mode is dictated by the type of information that is desired. Survey mode provides a complete inventory of elements present on a material surface. Detail analysis seeks to reveal small but discrete shifts in binding energy that reveal information on the bound state of the element of interest. Calibration of the XPS instrument can be done using metallic Cu, Ag, and Au standards as described by Seah et al., (1993).

Survey Mode: Scans of the full XPS spectrum can be done in the range from 0-1100 eV binding energies. If specific elements are of interest, and particularly if they are present in only trace amounts, the key reference peaks for these elements should be identified from existing data tables and the related energy levels should be scrutinized to see if the elements can be detected.
Detailed (Narrow) Scans are done to look at the fine structure of specific peaks of interest. The acquired peaks can then be deconvoluted using computer modeling software routines to determine the chemical state of elements of interest by comparing to published databases.

In addition:

Quantitative analysis can be done
- In Survey mode to determine the relative atomic percentages of elements present on the sample surface, and
- In Detailed mode to determine the percentages of an element in different bound states.
Depth Profiles--using specific peaks of interest in Detail mode (either to compare elemental atomic percent, or to monitor changes in the bound state of a given element), the sample is sequentially sputtered with an ion (Ar) beam with calibrated sputter rate and then XPS data are acquired. In aggregate, this series of sputter/analysis intervals will produce a chemical stratigraphy across the near surface of the sample into the material substrate.

Data Representation, Reporting, and Interpretation

XPS spectra, including the multiple photoelectron lines and Auger electron lines for all elements except for H and He, are compared to published databases to identify elements and their chemical states. The PHI Electronics Handbook of X-ray Photoelectron Spectroscopy or NIST X-ray Photoelectron Spectroscopy Database are good places to start. C 1s or O 1s lines are almost always present on material surfaces and can be used to calibrate the XPS spectra if shifts in Binding Energies are expected.

Curve Fitting in XPS: Resources for Good Practices and Tools for Avoiding Mistakes; a Technical Report from PHI Electronics

References for Further Exploration

Theory and Practice of XPS

Baer, D.R., Artyushkova, K., Richard Brundle, C., Castle, J.E., Engelhard, M.H., Gaskell, K.J., Grant, J.T., Haasch, R.T., Linford, M.R., Powell, C.J. and Shard, A.G., 2019. Practical guides for x-ray photoelectron spectroscopy: First steps in planning, conducting, and reporting XPS measurements. Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films, 37(3), p.031401. Practical Guides for X-Ray Photoelectron Spectroscopy (XPS): First Steps in planning, conducting and reporting XPS measurements
Battistoni, C., Mattogno, G. and Paparazzo, E., 1985. Quantitative surface analysis by XPS: a comparison among different quantitative approaches. Surface and interface analysis, 7(3), pp.117-121.
Seah, M.P., 1993. XPS reference procedure for the accurate intensity calibration of electron spectrometers—results of a BCR intercomparison co‐sponsored by the VAMAS SCA TWA. Surface and interface analysis, 20(3), pp.243-266.
Seah, M.P., Gilmore, I.S. and Spencer, S.J., 2001. Quantitative XPS: I. Analysis of X-ray photoelectron intensities from elemental data in a digital photoelectron database. Journal of Electron Spectroscopy and Related Phenomena, 120(1-3), pp.93-11
Stevie, F.A. and Donley, C.L., 2020. Introduction to x-ray photoelectron spectroscopy. Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films, 38(6), p.063204. Introduction to X-ray Photoelectron Spectroscopy
Wagner, C.D., 1983. Sensitivity factors for XPS analysis of surface atoms. Journal of electron spectroscopy and related phenomena, 32(2), pp.99-102.

Applications of XPS in the Earth and Environmental Sciences (EES)

Here are some representative articles showing a range of applications of XPS to EES

Hochella, M.F. Jr. 1988, Auger electron and X-ray Photoelectron Spectroscopies, in F.C. hawthorne (ed.), Spectroscopic Methods in Mineralogy and Geology, Reviews in Mineralogy, Volume 18, Mineralogical Society of America, pp. 573-638.
Hyland, M.M. and Bancroft, G.M., 1990. Palladium sorption and reduction on sulphide mineral surfaces: An XPS and AES study. Geochimica et Cosmochimica Acta, 54(1), pp.117-130.
Ilton, E.S., Post, J.E., Heaney, P.J., Ling, F.T. and Kerisit, S.N., 2016. XPS determination of Mn oxidation states in Mn (hydr) oxides. Applied Surface Science, 366, pp.475-485.
Jean, G.E. and Bancroft, G.M., 1986. Heavy metal adsorption by sulphide mineral surfaces. Geochimica et Cosmochimica Acta, 50(7), pp.1455-1463.
Jean, G.E. and Michael, B.G., 1985. An XPS and SEM study of gold deposition at low temperatures on sulphide mineral surfaces: Concentration of gold by adsorption/reduction. Geochimica et Cosmochimica Acta, 49(4), pp.979-987.
Lu, H.B., Campbell, C.T., Graham, D.J. and Ratner, B.D., 2000. Surface characterization of hydroxyapatite and related calcium phosphates by XPS and TOF-SIMS. Analytical chemistry, 72(13), pp.2886-2894.
Muir, I.J., Bancroft, G.M. and Nesbitt, H.W., 1989. Characteristics of altered labradorite surfaces by SIMS and XPS. Geochimica et Cosmochimica Acta, 53(6), pp.1235-1241.
Muir, I.J., Bancroft, G.M., Shotyk, W. and Nesbitt, H.W., 1990. A SIMS and XPS study of dissolving plagioclase. Geochimica et Cosmochimica Acta, 54(8), pp.2247-2256.
Nesbitt, H.W., and Bancroft, G.M., 2014, High Resolution Core- and Valence-Level XPS, in G.S.. Henderson, D.R. Neuville, and R.T. Downs (eds.), Spectroscopic Methods in Mineralogy and Material Sciences, Reviews in Mineralogy and Geochemistry Volume 78, Mineralogical Society of America, pp.271-329.
Nesbitt, H.W., Bancroft, G.M., Pratt, A.R. and Scaini, M.J., 1998. Sulfur and iron surface states on fractured pyrite surfaces. American Mineralogist, 83(9-10), pp.1067-1076.
Perry, D.L. and Grint, A., 1983. Application of XPS to coal characterization. Fuel, 62(9), pp.1024-1033.
Pratt, A.R., Muir, I.J. and Nesbitt, H.W., 1994. X-ray photoelectron and Auger electron spectroscopic studies of pyrrhotite and mechanism of air oxidation. Geochimica et Cosmochimica Acta, 58(2), pp.827-841.
Rosso, J.J. and Hochella Jr, M.F., 1996. Natural Iron and Manganese Oxide Samples by XPS. Surface Science Spectra, 4(3), pp.253-265.
Stipp, S.L. and Hochella Jr, M.F., 1991. Structure and bonding environments at the calcite surface as observed with X-ray photoelectron spectroscopy (XPS) and low energy electron diffraction (LEED). Geochimica et Cosmochimica Acta, 55(6), pp.1723-1736.