Quantitative surface analysis by Auger and x-ray photoelectron spectroscopy

Recent developments in quantitative surface analysis by Auger (AES) and x-ray photoelectron (XPS) spectroscopies are reviewed and problems relating to a more accurate quantitative interpretation of AES/XPS experimental data are discussed. Special attention is paid to consideration of elementary physical processes involved and influence of multiple scattering effects on signal line intensities. In particular, the major features of core-shell ionization by electron impact, Auger transitions and photoionization are considered qualitatively and rigorous approaches used to calculate the respective transition probabilities are analysed. It is shown that, in amorphous and polycrystalline targets, incoherent scattering of primary and signal Auger and photoelectrons can be described by solving analytically a kinetic equation with appropriate boundary conditions. The analytical results for the angular and energy distribution, the mean escape depth, and the escape probability as a function of depth of origin of signal electrons as well as that for the backscattering factor in AES are in good agreement with the corresponding Mote Carlo simulation data. Methods for inelastic background subtraction, surface composition determination and depth-profile reconstructions by angle-resolved AES/XPS are discussed. Examples of novel techniques based on x-ray induced photoemission are considered.     

Error surface calculations on the parameters defining oxygen depth profile models fitted to ARXPS data obtained from polystyrene exposed to an oxygen/helium plasma

Error surfaces are calculated for the fitting of concentration–depth profiles to angle‐resolved x‐ray photoelectron spectroscopy (ARXPS) data. The shapes of the error surfaces indicate that model parameters related to composition (especially at the very surface of the sample) are well constrained by the data, whereas parameters related to depth have a less significant impact on the fit. It is then shown that certain parameters in the different depth profile models employed are highly correlated and that the different models convey essentially the same information in different ways. Finally, a compromise profile definition is proposed for the fitting of constrained but flexible depth profiles to ARXPS data. Copyright © 2003 John Wiley & Sons, Ltd.     

Nondestructive in‐depth composition profile of oxy‐hydroxide nanolayers on iron surfaces from ARXPS measurement

In this work the maximum entropy method (MEM) is applied, for the first time, to angle‐resolved X‐ray photoelectron spectroscopy (ARXPS) data from oxy‐hydroxide films on iron surfaces. This nondestructively derives information on the in‐depth distribution of the composition and chemical state. An MEM algorithm was created and first tested on the simulated data. The reconstructed composition depth profiles agreed very well with the theoretical ones up to 5% Gaussian noise added to the data. The same algorithm was then applied to ARXPS data from iron samples to investigate the in‐depth variations in the composition and chemical state of the nanosized oxy‐hydroxide film naturally grown on the iron surface. The resulting surface film presents a complex multilayer structure with concentration gradients. The effect of air exposure on the structure was also investigated. Copyright © 2006 John Wiley & Sons, Ltd.     

On the choice of the regularization parameter for the interpretation of ARXPS data using a multilayer model

We report our investigation of the dependence of the profile extracted from ARXPS data on the value of the regularization parameter α. We argue that a choice based upon the L-curve criterion, which does not require knowledge of the variances in the data, is less satisfactory than an approach based on choosing α such that χ²/N = 1.     

Studies on the migration in PA6‐OMMT nanocomposites: effect of annealing on migration as evidenced by ARXPS (angle resolved X‐ray photoelectron spectroscopy)

The present paper deals with experiments in which the angle resolved X‐ray photoelectron spectroscopy (ARXPS) technique was applied to investigate the phenomenon of migration in polyamide 6 with organically modified montmorillonite (PA6‐OMMT) nanocomposites. This is the first time ARXPS was used, to gain a more detailed insight into the migration process. Curve‐fitting analyses are reported which enable the differentiation between a manual mixture of PA6 with OMMT at room temperature from a nanocomposite structure. The ARXPS technique was applied to annealed samples at 250, 275, and 285°C, for 2 hr, and with three take of angles, α, of 90, 60, and 30°. The depth of the layer investigated is 9 nm in case the sample surface is well defined. By tilting the take‐off angle of the beam‐out electrons one can determine the intensity of the signals at various distances from the surface within the investigated layer. The concentration gradient of the Si signals is observed. The rate of migration is found to be more rapid in the layer closest to the surface. The intensity of the Si signals in the investigated layer is found to be 1.8 to 9.1 times that of the same composite sample at room temperature. This ratio determines the extent of migration. At temperatures higher than 250°C and times of annealing greater than 30 min a decrease in the extent of migration is observed. This decrease is explained by the gradual decomposition of the surfactant and the consequent removal of the polymeric matrix molecules from the migrating exfoliated units, culminating in denuded alumino‐silicate layers. These layers aggregate to noncolloidal microcomposite particles which do not migrate. The concentration of the exfoliated units decreases with consequent decrease in migration. A gradient of decreasing concentration of Si2p signals was observed after various times of annealing, where the gradient becomes more uniform with increase in time. Copyright © 2008 John Wiley & Sons, Ltd.     

Studies of the oxidation of iron by air after being exposed to water vapour using angle‐resolved x‐ray photoelectron spectroscopy and QUASES™

Air oxidation of Fe was compared with and without a pre‐exposure to water vapour. Angle‐resolved x‐ray photoelectron spectroscopy (ARXPS) and QUASES^? were used to determine the thickness of the oxide layer formed and its composition. The extent of oxidation was found to be much less if the surface was pre‐exposed to water rather than air alone. Studies performed using ARXPS were able to show that the hydroxyl‐containing layer located at the surface after Fe was exposed to water vapour was located below the surface after exposure to air. This observation suggested that the oxidation of Fe in air is mediated by cation diffusion. Copyright © 2004 John Wiley & Sons, Ltd.     

Quantitative ARXPS investigation of systems with ultrathin aluminium oxide layers

Angle‐resolved x‐ray photoelectron spectroscopy (ARXPS) is a non‐destructive method to investigate the near‐surface structure of specimens with a flat surface. For interpretation of the electron intensities emitted from different depth regions, model calculations are necessary. Based on an earlier algorithm we have developed a program for ARXPS studies of thin multilayers. In our model calculation the sample structure is treated as consisting of several layers (one to three) on the substrate, whereas the top layer can be incomplete. Emitted electrons are assumed to be attenuated exponentially in the layers. Different atomic volumes, electron attenuation lengths (including consideration of elastic scattering) and assumptions on stoichiometry are taken into account for the particular layers. As an application of our model calculations we present a study of a set of Al samples that were oxidized by different methods, i.e. natural and plasma oxidation (plasma obtained by electron cyclotron resonance). The oxide layers produced by plasma oxidation were protected by a 2 nm thick Co film, before exposing the samples to the air. Additionally, in order to check our results of the ARXPS model calculation, x‐ray reflectometry (XRR) analysis was used. Copyright © 2004 John Wiley & Sons, Ltd.     

Photoelectron angular distributions of Cu,Ag, Pt and Au samples: experiments and simulations

Angular distributions of photoelectrons emitted from semi‐infinite Cu, Ag, Pt and Au specimens have been measured for off‐normal emission angles in the range between 20 and 70° with a Thermo Theta Probe electron spectrometer. Experimental peak intensities for peaks of atomic subshells observable in the spectra were compared with results of simulations using the NIST Database for the Simulation of Electron Spectra for Surface Analysis (SESSA) that takes into account the effects of (i) anisotropy of the photoelectric cross‐section; (ii) elastic scattering of the photoelectrons; and (iii) the finite solid angle of the detector. In addition, a separate correction was made to the simulated intensities for the effects of surface excitations. The combined influence of these effects was found to significantly affect the angular distributions. Furthermore, it was found that ratios of the calculated peak intensities of the observed subshells for a particular material to the measured intensities deviate from unity by typically less than 1% after corrections for multiple inelastic scattering and surface excitations. Copyright © 2010 John Wiley & Sons, Ltd.     

Fick's law of diffusion depth profiles applied to the degradation of oxidized polystyrene during ARXPS analysis

Angle‐resolved x‐ray photoelectron spectroscopy (ARXPS) measurements were made, using Al Kα and Mg Kα radiation alternately, on a polystyrene sample that had been exposed to a helium plasma. It was observed that oxygen was introduced into the sample surface by the plasma treatment, and that some of it was lost over a period of 5 h under x‐ray irradiation in the vacuum of the spectrometer. Laplace transforms of Fick's law of diffusion profiles were derived and applied to the data. The ARXPS results obtained in this study are consistent with a sample history in which the oxidation of the polymer surface resulting from exposure to plasma is controlled by a diffusion process, whereas the loss of oxygen during exposure to x‐rays is principally controlled by a first‐order reaction such as the liberation of oxygen (presumably as CO₂) from carbon–oxygen groups by the action of radicals created by the ionizing radiation. Copyright © 2005 John Wiley & Sons, Ltd.