Rainer P. H. Garten

Purification of reagents and separation of element traces by electrodeposition onto a graphite tube cathode

SummaryThe use of a commercially available AAS graphite tube atomizer as an electrode for the cathodic electrodeposition of element traces at constant-current in a hydrodynamic flow-through-system is described. After deposition the graphite tube is transferred into an atomizer unit for subsequent sensitive determination by atomic absorption spectrometry. The deposition yield of ≥99% (2 h; 1.5 A) for Co was verified, using60Co as a radioactive tracer, by monitoring the concentration fall in the electrolyte solution down to ≤5 pg/ml, and the deposited material. The efficiency of the system for the purification of electrolyte solutions was found to be very high. For the determination of the residual impurities, e.g. in very concentrated NH4F-solutions, a simple ion-exchange procedure was worked out for Cu, Mn, Ni, Pb, Zn. To get back the high purity solid NH4F, the solvent was evaporated partly and the precipitated crystals were sucked off in a PTFE-funnel. To avoid contamination, crystallization was carried out in a closed container system using the principle of subboiling distillation. The purity of the recrystallized NH4F-salt increased by a factor of 2 to 3 (crystallization yield 50–70%) so that a solid product with an impurity level below 0.5 ng/g was obtained.

Trends in applications and strategies in the analysis of thin films, interfaces and surfaces

Abstract Prominent aims and general aspects of the analysis of thin films, interfaces and surfaces are reviewed. The terminology is presented, and a limited sketch displays the major techniques important in this field. Some topical subjects in thin film analysis are identified, starting from its application for support of technology. This leads to reflections upon some trends concerning both optimization and application.

Matrix factors affecting quantitative analysis of AES for binary alloys

Abstract To improve the accuracy of quantitative AES analysis, matrix effects disturbing the linearity between the intensity of an Auger signal and the atomic concentrations have to be considered. Matrix factors accounting for these effects refer to (1) electron back-scattering, (2) attenuation length, (3) Auger line shape, (4) initial surface roughness, (5) sputter-induced surface roughness during depth profiling, (6) preferential sputtering, (7) elemental sensitivity and (8) atomic density. Measurements are presented on three systems of binary alloys (Ag/Cu, Ni/Cr and Pb/Sb), whose elemental concentrations range from 0 to 100%. They indicate that the relation between the measured and the true concentration can be expressed by a single formula with two parameters describing summarily the various matrix effects mentioned above. We derive the separation of the contributing factors (1) to (8) from the equation of the Auger electron intensities for a binary system, and their inclusion into the two-parameter approach mentioned. Values for these parameters are empirically estimated for the systems under investigation. The significance of individual factors (1) to (8) in these systems is discussed. The matrix factors mentioned do not fully describe all experimental data obtained from these alloys.

Calibration of sputtering yields for AES depth profiling of oxide layers on aluminium by means of carrier-gas heat extraction analysis

SummaryFor the purpose of obtaining suitable reference materials for technical coatings on aluminium, the feasibility of calibration of oxide layers by means of heat extraction was studied. Oxide layers were prepared on aluminium sheets by immersion in water of 50 or 80 ° C and annealing at 500 or 600 °C in argon atmosphere. The layer thickness was calculated from the oxygen content of the sample as measured by carrier-gas heat extraction analysis.The total sputtering yield of aluminium oxide layers was obtained via the correlation of AES depth profiling with heat extraction analysis results. This was demonstrated for high purity (99.9%) and technical purity (98.5%) aluminium with its original roughness from the rolling process, on which 20 to 1,000 nm thick oxide layers had been grown.The sputtering yields for the oxide layers prepared were found to be 3.9±0.8 atoms/ion, i.e., about four times higher than that for α-Al2O3. Calibration of depth profiling on such technical quality oxide layers on aluminium was found to be feasible with a relative precision of 10 to 20%.

Vergrabene Nitrid-Schichten in Silicium für Kalibrierproben zur quantitativen Auger-Elektronenspektrometrie (AES)

SummaryBuried layers of silicon nitride in silicon are produced by high-dose ion implantation and are checked for their suitability as calibration samples for quantitative thin film analysis. For this purpose, N+ ions (150 and 300 keV; 0.35 to 1×1018 N+ cm−2) are implanted into silicon single crystals and the samples annealed at 1,200°C for up to 15 h. The signal intensities and the sputter time obtained by AES/ sputtering can be converted into nitrogen content and sample depth by means of independent calibration measurements. The absolute depth scale is obtained by AES microanalysis at angle lapped surfaces (angle <1°) and by comparison with Monte Carlo simulation. The accuracy obtained is about 30 nm at a profile depth of 0.3 μm. The nitrogen content is determined quantitatively by means of the measured implantation dose.Additional methods of calibration are discussed. It is shown that the samples used are suitable as calibration samples for the silicon/nitrogen system.ZusammenfassungVergrabene Nitridschichten in Silicium werden durch Hochdosis-Ionenimplantation hergestellt und auf ihre Eignung als Kalibriermaterial für die quantitative Dünnschichtanalyse geprüft. Dafür werden N+-Ionen (150 und 300 keV; 0,35–1·1018 N+ cm−2) in Si-Einkristallen implantiert und durch Temperung (1200°C, bis 15 h) formiert. Die mit AES/Sputtering erhaltenen Signalintensitäten als Funktion der Sputterzeit können mit Hilfe unabhängiger Kalibriermessungen in die Stickstoffgehalte als Funktion der Probentiefe transformiert werden. Die Tiefenzuordnung wird durch AES-Mikroanalyse an Schrägschliffen (<1°) und durch Vergleich mit Monte-Carlo-Simulation mit einer Genauigkeit von ca. 30 nm bei 0,3 μm Profiltiefe erhalten. Der Stickstoffgehalt wird mittels der gemessenen Implantationsdosis bestimmt. Weitere zusätzliche Bestimmungsmethoden werden diskutiert.Es zeigt sich, daß die beschriebenen Proben als Kalibriermaterial für das Stoffsystem Silicium/Stickstoff geeignet sind.

Accurate determination of element traces in high purity reference materials by instrumental neutron activation analysis INAA

SummaryInstrumental neutron activation analysis INAA was applied for the trace element determination in some high purity candidate reference materials. By following an extensive calibration procedure, accuracy at the 1.2 to 6% level was obtained for trace determinations of Ta in Nb, Ir in Rh, and Co in Si. Pt-determination in Rh was considerably impaired by a concomitant 8-fold Ir-excess. Results agreed very well with results from other laboratories using neutron activation analysis and inductively coupled plasma-mass spectrometry (Ta), and X-ray fluorescence analysis, Rutherford-backscattering spectrometry and flame atomic absorption spectrometry (Co). The predominant advantages of INAA in this application are its high analytical sensitivity and its robustness against matrix effects and problems due to sample contamination.

Mass spectrometric analysis of ceramics after decomposition with elemental fluorine

The features of a combustion with elementary fluorine for the case of compact SiC ceramics and model substances for boron containing ceramics (H3BO3 and Na2B4O7) were investigated with the aim of their decomposition and analysis. On-line detection of the gaseous decomposition products by quadrupole mass spectrometry using electron impact ionisation was studied. Limitations by blanks and transport interferences were investigated. Standard addition as well as the isotope dilution technique were used for calibration in the case of B, C and W at the trace and major component level.

Application of factor analysis in electron spectroscopic depth profiling on copper oxide

Depth profiles were taken by x-ray photoelectron spectrometry/Ar-ion sputtering from copper sheets oxidized during 30 min in air at 200 or 300°C, respectively. The data of the depth profiles were subjected to factor analysis in order to determine the relevant components of the copper oxide layers. Factor analysis shows the existence of a cuprous oxide layer (Cu2O) on both specimens under bombardment with 2 keV argon ions during depth profiling. Rutherford backscattering measurements and carrier gas fusion analysis were successfully applied to determine the oxygen content of the oxide layers. Results corroborate the identification of Cu2O in both layers.

Elemental depth profile analysis of hard coatings of tungsten carbide by auger electron (AES-) microprobe/sputtering

ZusammenfassungTechnische CVD-Hartstoffbeschichtungen zum Verschleißschutz aus 8 bis 15 μm dicken Wolframcarbidschichten auf Stahl wurden mit Auger-Elektronen-Spektrometrie (AES)/Ionenätzen untersucht. Dem Problem der Analyse des Carbid-Signals, das allgemein in der Dünnschichtanalyse durch zufällige, allgegenwärtige Kohlenstoffbelegungen gestört wird, wurde besondere Aufmerksamkeit gewidmet. Halbquantitative AES-Tiefenprofile von W, C, Fe, Cr, O zeigten reproduzierbar charakteristische Unterschiede zwischen Proben, die unter verschiedenen Beschichtungsbedingungen hergestellt worden waren, sowie solchen mit guter bzw. schlechter Haftung auf dem Grundmaterial. Durch den Einsatz der Auger-Mikrosonde wurden trotz starker Probenaufrauhung auch nach Abtrag dicker Schichten > 10 μm durch Ionenzerstäuben gute Tiefenauflösungen von 0,5 bis 1 μm in der Größe der ursprünglichen Probenrauheit erhalten.Der Einsatz von Standard-Dünnschichtmaterialien, die zuvor durch eine Kombination verschiedener dünnschichtanalytischer Methoden umfassend charakterisiert wurden, wird als eine Möglichkeit diskutiert, die Richtigkeit der quantitativen Dünnschichtanalyse derartiger technischer Beschichtungen zu verbessern.SummaryTechnological wear-resistant hard coatings prepared by chemical vapour deposition (CVD) of 8 to 15 μm tungsten carbide on steel, were studied by Auger electron spectrometry (AES)/sputtering. Special regard was given to the problem of the analysis of the carbide signal, which is interfered in surface thin-film analysis, in general, by adventitious, ubiquitous carbon overlayers. Semiquantitative AES in-depth profiles of W, C, Fe, Cr, O reproducibly showed characteristic differences between samples from different coating processings and those with good and bad coating/base adhesion, respectively. Although strong sputter-induced roughening occurred, quite acceptable depth resolution of the order of initial sample roughness of 0.5 to 1 μm was obtained by means of the Auger microprobe even after sputter removal of thick layers > 10 μm.Standard thin-film materials, well-characterized by a combined approach using different complementary thinfilm analytical methods, are discussed as a means to improve accuracy of quantitative in-depth analysis of such rough technological coatings.

Characterization of SiOxNyHz and SiNyHz using RBS, ERDA and NRA In memoriam Priv.-Doz. Dr. Rainer P.H. Garten

Abstract Rutherford backscattering (RBS) measurements, elastic recoil detection analysis (ERDA) and nuclear reaction analysis (NRA) were applied to determine the composition, the thickness and the density of plasma-deposited SiOxNyHz and SiNyHz layers on silicon substrate. RBS-measurements were carried out with 2.3 MeV 4He-ions in random and channeling geometry to determine the atomic ratios of N Si in the surface layers, and the layer thicknesses. Depth profiling for hydrogen in the samples was performed by means of α-ERDA using 2.3 or 2.9 MeV 4He-ions. In order to determine the atomic ratios of H N , heavy ion elastic recoil detection (HIERDA) using 32 MeV 28Si7+-ions was applied. The oxygen contents were measured using the nuclear reaction 16O(d,p)17O. Depth profiling for nitrogen was also performed by means of NRA using the reaction 14N(d,α)12C. In addition, ellipsometric measurements of the layer thicknesses were utilized to determine the densities in absolute units (atoms/m3 or kg/m3).