Comment on relative ion sputtering yield measurements by integration of secondary ion energy distribution using a retarding-dispersive ion energy analyzer

The location and width of the peak in the sputtered ion energy distributions of W⁺ from the clean surface can be explained by a simple power dependence on energy for the ionization probability. Research sponsored by NASA under Contract/Grant Number NGR 21-002-096     

Angular-resolved energy distribution of secondary ions emitted from a silicon target sputtered by a xenon ion beam

This paper reports preliminary results obtained on an experimental apparatus dedicated to the study of angular resolved energy distribution of particles emitted from a sputtered target. Secondary ions emitted during the bombardment of a silicon target by xenon ions at a primary energy of 10keV have been studied. In its low energy part the distribution reaches a maximum around 8eV, and then decreases according to an E ^–1 law. In the range 200eV to 1000eV, a second maximum appears whose height depends on the emission angle. Apart from this range, the angular distributions have a cosine square-like shape. On the contrary, the angular distribution of ions with energy between 200eV and 1000eV is pointed in a forward direction near the specular reflection direction of the ion beam. It is assumed that the measured ions correspond to two ionic populations: secondary ions sputtered according to the linear cascade theory and recoil silicon target ions.     

On the influence of reactive gases on sputtering and secondary ion emission

As a continuation of earlier sputtering yield measurements in an ion microprobe, the influence of oxygen and nitrogen on sputtering yield, ionisation efficiency and depth resolutions has been studied. For inert gas bombardment the yield of Ti and V falls sharply at a certain oxygen exposure. While this decrease in yield can be correlated with an increase in surface binding energy in the case of titanium, cone formation causes the yield to drop for oxygen exposed vanadium. In contrast, during nitrogen bombardment the only effect of oxygen exposure is a drastic increase of the ionisation efficiency; the sputtering yield or the depth resolution Δz/z is hardly influenced by oxygen coverage. As was observed earlier in the case of Cu?Ni layers, Δz is essentially constant for erosion depthsz?800 Å, thus yielding better resolution at large depths than is to beexpected from a sequential layer removal model. The extent of the transition zone Δz, is determined by surface topography and thus depends on the target composition as well as its structure.     

On the velocity distribution of excited Fe-atoms by sputtering of iron

From the velocity distribution of excited sputtered particles detailed information on the excitation process can be obtained. In the present paper the first direct measurement of velocity distribution of excited atoms sputtered from a metal target is presented. The irradiation of the Fe-target was performed with 10keV Ar⁺-ions. The sputtered atoms were detected using pulsed laser induced fluorescence (LIF). The sputtered Fe atoms in the metastable statea ⁵ F ₅ at 0.86 eV shows a much broader distribution, than found for the ground-state atoms, but no energy threshold, implied in the statistical excitation models, was found.     

Energy transfer to a copper surface by low energy noble gas ion bombardment

The energy transferred to a copper surface by bombardment with Xe⁺, Ar⁺, and He⁺ ions with kinetic energies in the range 100–4000 eV has been studied by our group in previous experiments. There were significant experimental uncertainties for that data at energies below about 200 eV. The present investigation overlaps the previous work, extends the energy range to 10 eV, and includes data for Ne⁺. Particular emphasis is placed on the energy range below 200eV. A specially designed ion source was employed in these experiments. A polycrystalline copper film deposited onto a highly sensitive calorimeter was used as the target material. The results show that the Xe⁺ ion deposits more than 97% of its energy over the entire range investigated whereas the lighter ions deposit a decreasing fraction of their energy below about 1 keV. The decrease is largest for the lightest ion (He⁺). In all cases the deposited energy is about or more than 70% of the incident energy. It will be shown that the present results are in agreement with previous measurements for copper and are qualitatively in good agreement with computer calculations using the TRIM.SP code.On leave from: Institut für Schicht- und Ionentechnik, Forschungszentrum Jülich GmbH, W-5170 Jülich, Fed. Rep. Germany     

A comparison of ion-induced electron emission and secondary ion yields

Ion-induced electron emission from solid surfaces is studied using a beam of caesium ions. Features of the spectra obtained during depth profiling of layered structures suggest a novel technique for investigating ion-induced Auger processes. Depth profiles are presented in terms of measured secondary ion signals, electron-induced Auger emission, and the intensities of features in the ion-induced electron spectra. It is shown that changes in features of the ion-induced electron spectra can be related to changes of chemical composition and sputtering probability. These help in the interpretation of variations in secondary-ion yields with matrix composition during depth profiling.     

Ion energies at the cathode of the DC planar magnetron sputtering discharge

The energy distribution of the ions striking the cathode of the dc planar magnetron system was measured over a range of the typical sputtering conditions (magnetic field of 0.07–0.13 T, argon pressure of 0.01–10 Pa, discharge voltage of 250–600 V). The results obtained allow us to conclude that the major part of the incident ion flux originates in the cathode fall region. The theoretical model developed in terms of mobility theory makes it possible to evaluate the cathode fall voltage and its dependence on the sputtering conditions. It was found that the normalized integral form of the incident ions, energy spectrum is practically independent of the sputtering discharge parameters.     

On the fluence dependence of the sputtering yield for low-energy noble gas ions

The fluence dependence of the sputtering yield has been studied on amorphous silicon under uhv conditions with 0.5 to 5keV Ar⁺ using the KARMA technique (Kombinierte Auger/Röntgen Mikro-Analyse). It allows to measure simultaneously the surface composition, the differential sputtering yield, and the total amount of implanted gas. For all energies, the yield increases initially and reaches saturation after the removal of a layer the thickness of which is closely correlated to the ion range. Gas implantation as a cause for these fluence effects can be ruled out by quantitative analysis. The relative yield increase is found to be larger for low energies than for higher ones. Both these findings can be qualitatively explained by a simple damage collection model.     

Analysis of solids by secondary ion and sputtered neutral mass spectrometry

A mass spectrometer is described, which allows the analysis of sputtered neutral and charged particles as well as of residual gas composition. This combined SIMS, SNMS, and RGA instrument consists of a scanning primary ion beam column, an electron impact ionizer, an electrostatic energy filter and an rf quadrupole mass analyzer.Various examples of surface and bulk analysis are presented which demonstrate the beneficial complementary features of these techniques. These are, in particular: a substantial reduction of the matrix effect and fewer complications with samples of low electrical conductivity in SNMS, and the possibility of measuring the depth distribution of gases included in small cavities in the solid in the SNMS/RGA mode. SIMS, on the other hand, allows in many cases higher detection sensitivities.EURATOM Association     

Angular distribution and differential sputtering yields for low-energy light-ion irradiation of polycrystalline nickel and tungsten

Angular differential and total sputtering yields of polycrystalline nickel and tungsten have been measured for 1 and 4 keV H⁺ and 4 keV He⁺ ion bombardment at incidence angles between 0° and 80°. The differential sputtering yields (dY/dΩ) were determined with the aid of the collector technique, whereas the total yieldY was determined from the weight loss of the target during irradiation. Asymmetric angular distributions are observed at oblique angles of incidence, the emission maximum being shifted in forward direction (with respect to the incident ions). Even more pronounced than the change in shape of the emission distribution is an increase in the differential yield:dY/dΩ rises with increasing incidence angle over the whole range of ejection angles, the increase being most prominent in the direction of primary recoil emission. This effect is therefore ascribed to emission of surface atoms in direct projectile-surface atom collisions. On short term leave from Institut für Plasmaphysik der Kernforschungsanlage Jülich GmbH, D-5170, Jülich, Fed. Rep. Germany.     

A contribution to the investigation of ion impact desorption by ion scattering

Sputtering or ion impact desorption of adsorbed layers can be directly investigated by low-energy ion scattering. Because of its specific sensitivity to surface atoms, this method provides the possibility of monitoring the decrease of the signal from the adsorbate or the increase of the signal from the substrate. Both signals were studied with He⁺ backscattering from the system O on Ni (110). The measured desorption cross-sections and the principal implications for both ways of observation are discussed. The use of the substrate signal for the desorption study can be of major advantage, particularly in the case of light adsorbates.     

Enhancement of the secondary ion emission induced by fast clusters

The emission yields of the secondary ions are measured by using a conventional time of flight (TOF) technique under bombardments of Mg and C₂, Ni and Si₂ with different energies, and C_n, Si_n and Ni_n (n = 1-3) with the different charge states and with energy of 1.5 MeV per atom, respectively. For the bombardments of C_n, Si_n and Ni_n, the enhancements of the secondary ion emissions increase with increasing cluster sizes and charge states. For the bombardments of Mg and C₂, Ni and Si₂, although the mass and the nuclear charges of C₂ and Si₂ are the same as or equivalent to Mg or Ni, respectively, the enhancements of the secondary ion emissions induced by the clusters of C₂ and Si₂ in a wide energy range are also clearly indicated. The instantaneous collective interaction of the cluster constituents plays an important role in the secondary ion emissions.     

AES investigations of Ar+ ion retention in Si during Ar sputtering

Argon retention in silicon has been studied by AES in the energy range between 1 and 15 keV at bombardment fluences up to ∼10¹⁸ ions/cm². AES data of implanted argon in silicon near the surface region, as obtained during sputtering, can be interpreted qualitatively by a simple model of ion collection. Discrepancies between calculated and measured saturation values of collected argon ions indicate that during implantation at high fluences addition surface effects become important and that the simple model of ion collection has to account for this. Quantitative AES correlated with RBS indicates pronounced concentration gradients of argon in silicon near surface regions.     

The measurement of energy spectra of neutral particles in low energy ion scattering

An apparatus is described for low energy (0.1–10 keV) ion scattering (LEIS) experiments. A time of flight (TOF) spectrometer is incorporated in the system to be able to measure the energy of particles in the neutral state after scattering. The energy resolution ΔE/E of the TOF spectrometer is discussed and found to be 0.5% (FWHM). This is sufficient for our scattering experiments. An electrostatic analyzer (ESA) is used to measure the energy of scattered ions [ΔE/E=0.5% (FWHM)]. Experiments show that in general the ion dose needed to obtain a TOF spectrum (2×10¹⁰ ions/cm²) is much smaller than the dose needed for an ESA-spectrum (6×10¹³ ions/cm²). The ion spectra measured with the TOF spectrometer, by subtracting the neutral yield from the total yield, as well as with the ESA are found to agree quite well. This provides a way to calibrate the TOF spectrometer. The determination of the ion fraction of scattered particles is discussed [10 keV⁴⁰Ar⁺ on Cu(100), scattering angle 30°]. It is shown that the TOF spectrometer is able to measure light recoil particles (e.g. hydrogen) from a heavy substrate. In the analysing system is, in addition to the TOF spectrometer, also incorporated a stripping cell to measure the energy of neutral scattered particles. An energy spectrum of neutral scattered particles measured with both methods is shown.     

The dimer sputtering mechanism of Cu(001) at low bombarding energy

The emission of dimers during bombardment of a Cu(001) surface by Cu atoms of 300 and 1000eV energy is studied. A molecular dynamics simulation method based on many-body potentials is employed. At 300eV bombarding energy, around 81% of the sputtered dimers originate from second-nearest neighbor sites. 58% of these are ejected in a collision sequence correlated by the projectile. These dimers lead to a second maximum in the kinetic energy distribution of emitted dimers at around 8eV, besides a maximum at 4eV. Only the latter is found at 1000eV bombarding energy. As in this case mostly next-neighbor surface atoms are sputtered, the specific emission mechanism found at 300eV is irrelevant. Finally, we show that the direction of the angular momentum of sputtered dimers is correlated with the original surface site of the dimers.     

Evaluation of secondary ion yield enhancement from polymer material by using TOF-SIMS equipped with a gold cluster ion source

We investigated the enhancement of the secondary ion intensity in the TOF-SIMS spectra obtained by Au⁺ and Au₃⁺ bombardment in comparison with Ga⁺ excitation using polymer samples with different molecular weight distributions. Since the polymer samples used in this experiment have a wide molecular weight distribution, the advantages of the gold cluster primary ion source over monoatomic ion could accurately be evaluated. It was observed that the degree of fragmentation decreased by the usage of cluster primary ion beam compared with monoatomic ion beam, which was observed as a shift of the intensity distribution in the spectra. It was also found out that the mass effect of Au⁺ and Ga⁺ as monoatomic primary ion, resulted in about 10-60 times of enhancement for both samples with different molecular distributions. On the other hand, the Au₃⁺ bombardment caused intensity enhancement about 100-2600 compared with Ga⁺ bombardment, depending on the mass range of the detected secondary ion species. The cluster primary ion effect of Au₃⁺, compared with Au⁺, therefore, was estimated to be about 10-45.     

Studies of CdTe surfaces with secondary ion mass spectrometry,rutherford backscattering and ellipsometry

The composition and the stability of chemically etched, mechanically polished and oxidized surfaces of single crystals of cadmium-telluride were studied by secondary ion mass spectroscopy (SIMS), Rutherford backscattering (RBS) and ellipsometry. CdTe surfaces etched using a solution of bromine in methanol were found to be enriched in cadmium but a film, identified to be an oxide of tellurium, was observed to grow on it at room temperature and in air. The thickness of this film increased over long periods of timet linearly versus lnt. Mechanically polished samples and also chemically etched surfaces which were oxidized in a solution of hydrogen peroxide in amoniac were found to be stable.     

Surface analysis of insulating materials by Secondary Ion Mass Spectrometry (SIMS)

For a non-conducting solid sample (TiO₂ powder) the conditions for compensation of the charge build-up at the surface caused by the ion impact in SIMS are experimentally investigated. The compensation is achieved by an additional auxiliary electron beam of low energy. The resolution and the intensities of the secondary ions were measured as a function of the ratio of the current densities of the electron and the ion beams. The compensation for negative secondary ions, and especially for those with higher masses, is more critical than for positive ones. The intensities are influenced by the different values of the mean emission energies and the form of the energy distributions. Examples of mass spectra by SIMS for some insulators are given.     

Ion beam sputtering - the effect of incident ion energy on atomic mixing in subsurface layers

Secondary ion mass spectrometric methods were used to study the effect of incident ion energy on atomic mixing in subsurface layers. A Ta₂O₅ film containing a 50 A ^3lP-rich layer 230 A below the surface was depth profiled for phosphorus using normal incidence 160 primary particles of various energies (1.75 to 18.5 keV). A pronounced energy effect was observed in the widths of 31^P profiles generated by >4 keV ¹⁶O. For 18.5 keV ¹⁶O, the observed profile contained two distinct components-the 31^P-rich layer and the 31^P recoil distribution. This later condition, prevails when the peak of the incident ion damage distribution occurs at a depth which equal or exceeds the distance from the surface to the 31^P-rich layer. The desirable primary ion beam energy for characterizing the true elemental distribution is dictated by the shape and location of the layer to be profited. In most instances, a low energy beam is preferred.     

The peculiarities of the low-energy ion scattering by polycrystal targets

In the present work, experimental and computer simulation studies of low-energy (E₀ = 80-500 eV) Cs⁺ ions scattering on Ta, W, Re target surfaces and K⁺ ions scattering on Ti, V, Cr target surfaces have been performed for more accurate definition of mechanism of scattering, with a purpose of evaluation of an opportunity of use of slow ions scattering as a tool of surface layers analysis. The choice of the targets was based on the fact that the ratios of atomic masses of target atoms and ions μ = m₂/m₁ were almost the same for all cases considered and greater than 1 (direct mass ratio) however, the difference of binding energies of target atoms in the cases of Cs⁺ and K⁺ scattering was almost twice as much. It has been noticed that the dependencies of the relative energy retained by scattering ions at the maximum of energy distribution versus the initial energy E_m/E₀ (E₀) have a similar shape in all cases. The relative energy retained by scattering ions increases while the initial energy of incidence ions decreases. The curves are placed above each other relative to the binding energies of target atoms, to show what this says about the influence of binding energy on a process of scattering of low-energy ions. The correlation between value of energy change maintained by an ion for different values of E₀ in the case of scattering by targets with different masses of atoms and its binding energies is experimentally established. The contrary behavior of the E_m/E₀ (E₀) dependencies concerning the target atom binding energy quantity E_b for cases with direct (μ > 1) and inverse (μ < 1) mass ratio of colliding particles is established. The comparison of experimental energy distributions with calculated histograms shows that the binary collision approximation cannot elucidate the abnormally great shift in the maxima of relative energy distributions towards greater energy retained by scattering ions.