Search

Total reflection X-ray photoelectron spectroscopy (TRXPS) is applied in the analysis of Ti and TiO2 nanolayers deposited on silicon and silicon dioxide substrates. The idea of application of total-reflection phenomenon for exciting X-ray used in the XPS technique is briefly discussed. The experimental setup and measurement conditions for the studied Ti and TiO2 layers are presented. The XPS spectra were registered both for the non-total and total reflection regimes. The survey spectra and C1s, N1s, Ti2p and O1s photoelectron peaks are shown. For energy calibration, the position of C1s photoelectron peak was applied (C-C component, binding energy 284.8 eV). The peak to background ratios are discussed as regards the dependence of the excitation angle. An increase of this ratio for the glancing angle 1°, being below critical angle of the X-ray beam and sample material, results in an improvement of XPS detection limit by factor up to 2. In the case of the Ti nanolayer, additionally, the thickness of the overlayer TiO2 is determined. As an example of applying the TRXPS technique, the analysis of Ti nanolayers implanted by highly charged Xe35+ ions of 280 keV energy is discussed. The Xe3d and O1s photoelectron peaks are presented and discussed.

The measurements of the M- and N-shell ionization probabilities in gold bombarded by highly charged S q+ (q = 4–14) ions of energy 0.4–3.8 MeV/amu are reported. The ionization probabilities were extracted from the measured L-X-ray spectra which were substantially modified by the multiple ionization effects. The observed L-X-ray line shifts and broadening allowed us (Banaś et al., 2002) to extract both the M- and N-shell ionization probabilities for the zero impact parameter. These ionization probabilities, which correspond to the moment of X-ray emission , were correlated with the ionization probabilities for the moment of ion impact ionization by using the vacancy rearrangement procedure developed here. In this way the ionization probabilities for the moment of ion impact multiple ionization were obtained. By comparing to the theoretical predictions of the semiclassical approximation (SCA) (Trautmann and Kauer, 1989) and the “geometrical model” (GM) (Sulik et al., 1987) based on the binary encounter approximation (BEA), the role of the electronic wave functions and the binding effect is discussed. Generally, the calculations using the hydrogenic wave functions reproduce the measured ionization probabilities.

Particle-induced X-ray emission (PIXE) technique is usually applied using typically 1 MeV to 3 MeV protons or helium ions, for which the ion-atom interaction is dominated by the single ionization process. For heavier ions the multiple ionization plays an increasingly important role and this process can influence substantially both the X-ray spectra and atomic decay rates. Additionally, the subshell coupling effects are important for the L- and M-shells ionized by heavy ions. Here we discuss the main features of the X-ray emission induced by heavy ions which are important for PIXE applications, namely, the effects of X-ray line shifts and broadening, vacancy rearrangement and change of the fluorescence and Coster–Kronig yields in multiple ionized atoms. These effects are illustrated here by the results of the measurements of L X-ray emission from heavy atoms bombarded by 6 MeV to 36 MeV Si ions, which were reported earlier. The strong L-subshell coupling effects are observed, in particular L 2 -subshell, which can be accounted for within the coupling subshell model (CSM) developed within the semiclassical approximation. Finally, the prospects to use heavy ions in PIXE analysis are discussed.

The L-shell X-ray production cross sections have been measured for sulphur ions in the energy range of 12.8–120 MeV for Au and Bi elements. The experimental L X-ray spectra were analyzed using the method that takes into account the multiple ionization in outer shells. The L-subshell ionization cross sections have been obtained from measured X-ray production cross sections for resolved Lα 1,2 Lγ 1 and Lγ 2,3 transitions using the L-shell fluorescence and Coster–Kronig yields modified by the multiple ionization effects in the M and N shells. The results are compared with the predictions of ECUSAR theory, which is the modified ECPSSR approach describing both direct ionization and electron-capture processes and the semiclassical approximation (SCA) calculations for direct ionization. These approaches were modified by the L-subshell coupling effects within the “coupled-subshell model” (CSM). Both modified approaches are in good agreement with the data. Remaining discrepancies are discussed in terms of the L-shell decay rates modified for the multiple ionization effects.

The coupling and binding effects in ionization of L-shell electrons by heavy ions have been studied by measuring the L γ (L–N,O) X-ray transitions in gold excited by an impact of O, Si and S ions in the energy range 0.3–5 MeV/amu. The L-subshell ionization cross sections were extracted from measured X-ray spectra taking into account their substantial modifications due to the multiple ionization effects, which also change the fluorescence and Coster–Kronig yields. In order to interpret the data the L-subshell couplings were incorporated in the theoretical calculations based on the semiclassical approximation (SCA) within the “coupled subshell model” (CSM). Resulting SCA–CSM calculations, which describe the data for oxygen ions quite satisfactory, are further modified to include the saturation of the binding effect in order to describe the data for heavier Si and S ions. We find that the semiclassical SCA–CSM calculations of L-shell ionization by heavy ions ( Z 1 ≫ 1), which include corrections both for coupling and binding saturation effects, agree reasonably well with the measured data.

The effect of the projectile K X-rays line shifts with respect to the corresponding diagram lines has been used for evaluation of a mean equilibrium charge, q ¯ , for sulphur ions passing with energies of 9.6, 16.0, 22.4 and 32.0 MeV through Al, Ti and Fe foils. The q ¯ values for sulphur ions passing through Al and reference C foils are approximately equal for the all discussed energies. In the case of Ti and Fe foils the q ¯ values are greater for the smallest energy and the difference between q ¯ values in these two groups of targets is decreasing with increasing sulphur ion energy. For sulphur ions with energy of 32.0 MeV there is no difference in q ¯ values in all targets.

The method of analysis of the multiple ionization effects in low-resolution X-ray spectra induced by heavy ions is described here. It is shown that, by fitting the X-ray spectra measured by a semiconductor detector with the proposed model, which accounts for the multiple ionization effects, the ionization probabilities can be determined as free fitting parameters. This approach is based on the assumption that the intensity distribution of emitted X-ray satellites is approximately described by the binomial distribution parameterized by the ionization probability at the moment of X-ray emission. In particular, we demonstrate that the excited X-ray satellites, when measured by a low-resolution semiconductor detector, appear as the Gaussian profile, which is shifted and broadened with respect to the diagram line. Moreover, we find that both X-ray line shift and width are expressed in terms of the multiple ionization probabilities, as well as the X-ray energy shifts per vacancy. These observations allowed to develop a novel method of X-ray spectra fitting. Detailed discussion of the approximations used in the present approach is given. The method developed, using the calculated Dirac–Fock X-ray energy shifts per vacancy, was applied to determine the multiple ionization probabilities in M- and N-shell by fitting the measured X-ray spectra for L γ (L→N,O) transitions induced by heavy ions in selected high-Z atoms.

Multiple ionization in outer M- and N-shells in Ta, Os, Au, Bi, Th and U targets was studied for O, Si and S ion impact at energies of 0.3-2.2 MeV amu-1. Excited Lγ(L-N, O) x-rays measured with a semiconductor Si(Li) detector were resolved by using a newly developed method accounting for simultaneous x-ray energy shift and line broadening due to the multiple ionization. The measured ionization probabilities, corrected for the effect of the vacancy rearrangement up to the moment of L x-ray emission, are interpreted as the ionization probabilities for M- and N-shells for the zero impact parameter. The ionization probabilities for the N-shell are found to be substantially increased by the super-Coster-Kronig transitions. Derived ionization probabilities for M- and N-shells are compared with the predictions of the geometrical model (GM) and the semiclassical (SCA) calculations in the united-atom limit. The universal scaling of the measured ionization probabilities for the M- and N-shells is discussed within the GM and SCA approaches.

Solid state effects for selected L–O(5p, 5d) X-ray transitions excited by O, Si and S ions of energies 0.4–2.0 MeV/amu are studied in metallic Os, Au, Bi, Th and U targets by measuring the L-X-rays with a semiconductor Si(Li) detector. Using the developed method of analysis of X-ray energy shifts and line broadening, combined with a change of the intensity-ratios, the ionization probabilities in M-, N- and O-shells during L γ X-ray emission were estimated. We found that the ionization probabilities measured in this way for M-, N- and O-shells are generally in good agreement with the theoretical predictions of the “geometrical model”, with the exception of L–O(5d) transition for the lighter (Os, Au and Bi) of the studied metals. We attribute this effect to the fast refilling of the 5d vacancies created in the collision by the delocalized valence band electrons. Such an effect is expected to be observed only when many electrons are removed from outer shells as a result of the multiple ionization.

Multiple ionization in M- and N-shells in solid Au, Bi, Th and U targets was studied for O, Si and S ions of energies 0.4–2.0 MeV/amu. L γ X-rays measured with semiconductor Si(Li) detector were analysed by using a newly developed method of the simultaneous determination of X-ray energy shifts and line broadening caused by the multiple ionization in outer shells. In this approach both X-ray energy shifts and widths are expressed in terms of ionization probabilities and calculated energy shifts per vacancy, allowing thus the unique fitting of the L γ X-rays and determination of ionization probabilities. Derived ionization probabilities for M- and N-shells exhibit universal scaling predicted by the geometrical model (GM) for the ionization probabilities at the zero impact parameter. The influence of time evolution of the vacancies formed in M- and N-shells by ion impact is discussed.

In the present work, an alternative approach for the evaluation of the equilibrium K-, L-, and M-shell ionizations and the mean charge state q for projectiles passing through various targets has been proposed. The approach is based on measured K x-ray energy shifts and line intensity ratios and utilizes the theoretical analysis of projectile spectra using multiconfiguration Dirac-Fock calculations. It was applied for the satellite and hypersatellite K lines in the x-ray spectra emitted by sulfur projectiles passing with energies of 9.6-122 MeV through carbon, aluminium, titanium, and iron targets, recorded by a Si(Li) detector. It was found that only in the high projectile energy region there was significant dependence of mean equilibrium K-shell ionization on the target atomic number. The equilibrium L-shell ionization rises with the increase of sulfur energy until 32 MeV, but for higher energies the changes are very weak. The equilibrium M-shell ionization changes very weakly for low projectile energy while for higher energies this ionization is practically constant. For each target, the estimated value of q rises with the increase of the sulfur energy value. The dependence of the sulfur charge state on the target atomic number was discussed by taking into account the crossmore » sections for ionization, decay, and electron capture processes. The data were compared with the experimental data measured by other authors and with the predictions of Shima's and Schiwietz and Grande's semiempirical formulas. The presented good agreement points out that this alternative approach delivers quantitative results.« less

The coupling and binding effects have been studied in $L$-subshell ionization of heavy Au, Bi, Th, and U atoms by an impact of ${^{28}\text{S}\text{i}}^{q+}$ ions in the energy range of 8.5--36.0 MeV. The measured $\mathit{L}$ x-ray spectra were analyzed taking into account the multiple ionization effects in outer $M$ and $N$ shells. The $L$-subshell ionization cross sections have been obtained from measured x-ray production cross sections using the $L$-shell fluorescence and Coster-Kronig yields which were modified for a reduced number of electrons and closed Coster-Kronig transitions in the multiply ionized atoms. The results are compared with the available calculations, which are based on the semiclassical approximation (SCA) as well as the plane-wave Born approximation (PWBA). We demonstrate that for silicon ion impact these theoretical approaches have to be modified to include the $L$-subshell coupling effect using the ``coupled subshell model'' (CSM) as well as the saturation of the binding effect at the united atom limit. The calculations modified for both effects are in much better agreement with the data. In particular, an order-of-magnitude improvement of agreement between the data and the SCA-CSM calculations including the binding-saturation effect is reported for low-energy Si ions for the ${L}_{2}$-subshell. The results are also compared with the predictions of the PWBA based ECPSSR and ECUSAR theories accounting for the energy-loss (E), Coulomb-deflection (C), and relativistic (R) effects treating the binding effect within the perturbed stationary state (PSS) approximation with correction for the binding-saturation effect introduced to describe the united-atom and separated-atom (USA) limits.

The K{alpha} and K{beta} satellite and hypersatellite x-ray lines emitted by highly ionized sulfur projectiles passing with energies from 65 MeV up to 122 MeV through carbon foils of thickness of 15-210 {mu}g cm{sup -2} have been recorded using a Si(Li) detector. The additional hypersatellite Ky{sup h} peak proves that for such high energies of the sulfur ions very high subshells (4p and 5p) could be occupied. In order to study the dynamics of formation of K-shell vacancy fractions of sulfur projectiles passing through a carbon foil the dependence of sulfur K x-ray production cross sections on foil thickness has been examined separately for each recorded line using the three component model. For each projectile energy the values of K-shell hole production cross sections and K-shell electron capture cross sections (both common for all recorded x-ray lines in the case of each projectile energy) have been fitted, as well as the specific values (for each recorded x-ray line) of K-shell hole filling cross sections, which are directly connected with average lifetimes of appropriate states of sulfur ions. The obtained ''experimental'' values of K-shell vacancy production cross sections are much higher than the theoretical predictions. This suggests that apart from themore » ionization process the excitation from K shell into higher shells is responsible for a production of K-shell vacancies, which has been confirmed by recent classical trajectory Monte Carlo calculations.« less

Multiple ionization in M-, N- and O-shells in solid Au, Bi, Th and U targets was studied for Oq+, Siq+ and Sq+ ions of energies 0.4–2 MeV/amu under charge equilibration condition. From L-x-rays, detected by Si(Li) detector, the ionization probabilities for M- and N-shells were derived. This was done by fitting the measured spectra using a developed model, which describes x-ray energy shifts and line broadening in terms of the ionization probabilities. Derived ionization probabilities are compared with the predictions of the geometrical model (GM). For M- and N-shell the data agree quite well with calculations. The ionization probabilities for O-shell, estimated from line intensity ratios, are in strong disagreement with theoretical predictions, indicating an influence of solid state effects. An importance of multiple ionization effects on derivation of L-subshell ionization cross sections is evidenced and discussed.