X-ray photoelectron spectroscopy

from Wikipedia, the free encyclopedia
Typical XPES system with hemisphere analyzer, X-ray tubes and various preparation methods
The inside of an XP spectrometer

X-ray photoelectron spectroscopy (English: X-ray photoelectron spectroscopy , XPS, often electron spectroscopy for chemical analysis , ESCA) is an established method from the group of photoelectron spectroscopies (PES) to the chemical composition, especially of solids or its surface to determine nondestructively . You first get an answer to the question of qualitative element analysis, i.e. which chemical elements the solid consists of. Only hydrogen and helium can not be directly detected due to the small cross- sections. The information depth corresponds to the penetration depth of the unscattered or elastically scattered electrons and is usually up to three nanometers. The electronic structure of a solid is examined with angle-resolved photoelectron spectroscopy.

Measuring principle

Photoelectron spectroscopy (PES) is based on the external photo effect , in which photoelectrons are released from a solid body by electromagnetic radiation (in this case X-rays ) . The electrons are split off from the inner atomic orbitals (core electrons). In a simplified model, the photoemission process takes place in three steps. First, the electron is excited by the incident photon, then the excited electron is transported to the surface and, as a third step, the photoelectron exits. The binding energy E B , which can be determined from the kinetic energy of the photoelectrons, is characteristic of the atom (more precisely even for the orbital) from which the electron originates. The analyzer used for the measurement (usually a hemispherical analyzer ) is set using electrostatic lenses and counter voltages so that only electrons of a certain energy can pass through it. For the XPS measurement, the electrons that still arrive at the end of the analyzer are detected by a secondary electron multiplier, so that a spectrum is created that is usually shown in a graph by plotting the intensity ( counting rate ) against the kinetic energy of the photoelectrons.

XP spectrum of magnetite (Fe 3 O 4 )

The picture on the right shows a typical spectrum in which the kinetic energy of the electrons has already been converted into the binding energy using the photoelectric equation . The nomenclature of the labeling means here: Fe2p for electrons that come from an iron atom (Fe), more precisely from the p-orbital of the L-shell. Similarly, the shortened notation O1s corresponds to the electrons from the s orbital of the K shell of oxygen .

The spectrum shows the clear spin-orbit splitting between the 2p 1/2 and 2p 3/2 levels in the iron atom using the two Fe2p emission lines . The additional indices 1/2 and 3/2 indicate the total angular momentum of the electron in these orbitals. A comparable split can be found for emissions from all orbitals (with the exception of the s orbitals ) of all elements.

Quantitative evaluation of the measurement

The intensity, i.e. the count rate of these measurements, is proportional to the frequency of occurrence of the various elements in the sample. In order to determine the chemical composition of a solid, one has to evaluate the area below the observed lines that are characteristic of the elements. However, some measurement-specific features must be observed. For example, before a photoelectron leaves the solid, it can excite other electrons and transfer part of its kinetic energy to them. These so-called secondary electrons have practically no discrete energy distribution and therefore contribute evenly to the growth of the background in an XP spectrum. In the adjacent illustration of an XP spectrum of magnetite, this behavior can be seen from the step-like increase (in the direction of greater negative binding energy) of the counting rate after each line. This background must be deducted before evaluating the areas using suitable methods, for example by subtracting a linear background. More precisely, the underground deduction is based on a method that goes back to DA Shirley and is called Shirley underground correction. The most precise (and most complex) method consists in precisely determining the course of the inelastic energy loss cross-section K (E'-E) using electron energy loss spectroscopy and thus modeling the subsurface; The Tougaard underground is based on the fact that this curve can be approximated well using three parameters (strictly speaking, the product of K (E'-E) and the mean free path).

In addition, it should be noted with XPS measurements that the probability of triggering a photoelectron is energy-dependent, element-specific and orbital-dependent. In order to take this fact into account, the values ​​for the areas that are determined under the respective lines must be corrected by so-called sensitivity factors or cross- sections, which can be found in different tables.

Furthermore, the probability that photoelectrons generated by the X-ray radiation actually leave the solid and can be detected depends on how often they are scattered or reabsorbed in the solid. This loss rate depends on the kinetic energy of the photoelectrons and the composition of the solid. This effect can be taken into account by considering the mean free path of the electrons in the solid. The corresponding data are tabulated for at least most of the elements and simple connections.

Taking all the effects mentioned into account, the evaluation of the spectrum shown on the right results in an iron-to-oxygen ratio of 3: 4, the empirical formula of the material being investigated is Fe 3 O 4 , it is magnetite . The detection limits of the various chemical elements fluctuate considerably due to the widely varying cross sections. The different electron scattering losses have a less strong influence. The light elements lithium , beryllium and boron (1s lines ) have poor detection limits (with the widespread use of Al-K α radiation) with about 1 at% (i.e. at least 1% proportion of the atoms). For example, the elements copper , tin and gallium (2p lines), tellurium and neighboring elements (3d lines) as well as the heavy elements gold to uranium (4f lines) with values ​​mostly well below 0.1 at have very good detection limits -%. The noise of the background contributes to the level of the detection limit in addition to the signal that originates from scattered electrons.

Advanced analysis

Another essential piece of information about the chemical bonding in the sample is based on Kai Siegbahn's discovery that the spectra of the core electrons depend on the chemical environment of the system under investigation. In the XPS spectra of the same element to differences depending on the chemical form it is present, in the binding energy of a core electron of up to several electron volts, this is called chemical shift (engl. Chemical shift hereinafter). For example, the determination of the charge distributions in a chemical compound can also determine which oxidation state an atom or an atom group has. In many cases, the shape of the spectra can also provide information about the valence state of an element.

Photoemission is accompanied by other physical processes, such as photoluminescence , or the appearance of Auger electrons and much more, which in turn have been used as a separate measurement method.

An elucidation of the constitution of chemical compounds is not always possible with the XPS method, because the selectivity is much lower than z. B. in nuclear magnetic resonance spectroscopy . In addition to other spectroscopic methods, the procedure can be helpful in certain cases.

Radiation sources

The most common X-ray sources used in XPS are Al-K α or Mg-K α sources, although more exotic X-ray sources also generate silicon , titanium or zirconium X-ray lines. In the last 20 years the use of synchrotron radiation , which is ideally suited as a source of excitation due to its almost unlimited tunability of the photon energy and monochrome, has become more and more popular. Thus, the range of accessible stimulating photon energy is from a few discrete values ​​(e.g. Al K α , hν  = 1486.6 eV and Mg K α , hν  = 1253.6 eV) to a continuum ranging from a few electron volts up to 20 keV is enough, has been expanded.

Technical details

Since it means a very high technical (and therefore financial) effort to carry out measurements with the help of synchrotron radiation, the aforementioned X-ray tubes are widely used in standard XPS analyzes. The energy of the photoelectrons generated with these X-ray sources is in the range between 0 eV and 1500 eV, which means for PES measurements that the emitted electrons come from a maximum depth of the examined sample, which is between 0 and 100 Å. The limiting factor here is the mean free path of electrons in the solid. This is the reason why the XPS is mainly used for the analysis of solid surfaces.

For such measurements, a base pressure of the analysis chamber in the ultra-high vacuum (UHV) range is usually necessary, which is between 5 × 10 −9  Pa and 5 × 10 −8  Pa (for comparison: the air pressure is approx. 10 5  Pa) . This requirement for good vacuum conditions results from the inevitable contamination of the sample with adsorbates from the ambient air, such as. B. water or carbon , which can be several micrometers and thus would not allow a measurement on the solid surface of interest. In order to circumvent this problem, samples are placed in the UHV chamber and suitable methods, such as e.g. B. prepared highly pure by sputtering , tempering, filing or splitting of single crystals.

Angle-dependent XPS measurement for improved identification of ultra-thin layers

However, this inherent weakness in the measurement method can be turned into a powerful advantage. The information depth of the PES is limited by the mean free path of the electrons in a solid. For a metal e.g. B. this is just 1 to 2 nm, which is three orders of magnitude smaller compared to the penetration depth of the X-rays (depending on the material 1 to 10 µm). It is clear that only those photoelectrons can be detected by the detector that can also leave the solid body. The intensity contribution to the spectrum thus decreases exponentially with increasing depth. In the angle-resolved X-ray photoelectron spectroscopy (engl. Angle-resolved X-ray photoelectron spectroscopy , ARXPS) can be obtained by varying the angle of the detector with respect to the extreme sensitivity of the measurement surface can be achieved sample to be measured. In the area of ​​surface physics in particular, the first monolayer of a sample can be examined in this way.

literature

  • JM Hollas: Modern methods in spectroscopy. Vieweg, Braunschweig / Wiesbaden 1995, ISBN 3-540-67008-4 .
  • D. Briggs, MP Seah (Ed.): Practical Surface Analysis, Volume I - Auger and X-ray photoelectron spectroscopy. John Wiley & Sons, Chichester 1990, ISBN 0-471-92081-9 .
  • M. Henzler, W. Göpel: Surface physics of the solid. Teubner, Stuttgart 1991, ISBN 3-519-13047-5 .
  • Gerhard Ertl , J. Küppers: Low Energy Electrons and Surface Chemistry. VCH, Weinheim 1985, ISBN 0-89573-065-0 .
  • Stefan Hüfner : Photoelectron spectroscopy, principles and applications. (= Springer Series in Solid State Sciences. Vol. 82). Springer, Berlin / Heidelberg / New York 1996, ISBN 3-540-41802-4 .
  • K.-M. Schindler: Photoelectron Diffraction. In: Chemistry in Our Time. 30, No. 1, 1996, pp. 32-38, doi: 10.1002 / ciuz.19960300106 .
  • Karsten Levsen: Physical methods in chemistry: ESCA. In: Chemistry in Our Time . 10, No. 2, 1976, ISSN  0009-2851 , pp. 48-53.
  • T. Cremer, M. Stark, A. Deyko, H.-P. Steinrück, F. Maier: Liquid / Solid Interface of Ultrathin Ionic Liquid Films: [C 1 C 1 Im] [Tf 2 N] and [C 8 C 1 Im] [Tf 2 N] on Au (111) . In: Langmuir . tape 27 , no. 7 , 2011, p. 3662-3671 , doi : 10.1021 / la105007c .

Web links

Commons : X-ray Photoelectron Spectroscopy  - collection of images, videos and audio files

Footnotes and individual references

  1. ↑ a b J. J. Yeh, I. Lindau: Atomic subshell photoionization cross sections and asymmetry parameters: 1 ≤ Z ≤103 . In: Atomic data and nuclear data tables . tape 32 , no. 1 , 1985, pp. 1–155 , doi : 10.1016 / 0092-640X (85) 90016-6 .
  2. ^ DA Shirley: High-Resolution X-Ray Photoemission Spectrum of the Valence Bands of Gold . In: Physical Review B . tape 5 , no. 12 , 1972, p. 4709-4714 .
  3. ↑ S. Tougaard, P. Sigmund: Influence of elastic and inelastic scattering on energy spectra of electrons emitted from solids . In: Physical Review B . tape 25 , no. 7 , 1982, pp. 4452-4466 .
  4. ↑ Electron Inelastic-Mean-Free-Path Database ( Eng. ) NIST . Archived from the original on May 27, 2010. Retrieved December 18, 2011.