Quantitative Trace Analysis by Wavelength-Dispersive EPMA

 “Trace” elements may be defined as elements whose concentrations are of a similar order to the detection limit. In WD analysis the detection limit is a function of the ‘figure of merit’ P²/B, where P is the pure-element peak intensity and B the background intensity. With normal analysis conditions detection limits of ∼100 ppm are typical, but substantial improvements can be achieved by using higher values of accelerating voltage and beam current. Long counting times are also advantageous, but should preferably be divided into relatively short alternating peak and background measurements to minimise the effect of instrumental drift. Using separate routines for trace and major element analysis is desirable owing to their different requirements. As the statistically defined detection limit is reduced, errors due to background nonlinearity and interferences (overlaps) from other elemental peaks become more probable. Spectrum simulation is useful for optimising background offsets and choice of crystal to minimise interferences, and estimating interference corrections when these are necessary. ‘Blank’ standards containing none of the trace elements of interest are also useful for quantifying background nonlinearity.     

Light Element Analysis of Individual Microparticles Using Thin-Window EPMA

 The determination of the concentration of light elements, such as carbon, nitrogen and oxygen, in e.g. atmospheric aerosol particles is important to study the chemical behaviour of atmospheric pollution. The knowledge of low-Z element concentrations gives us information on the speciation of nutrients (species having nutritional value for plants) and toxic heavy metals in the particles. The capability of the conventional energy-dispersive EPMA is strongly limited for the analysis of low-Z elements, mainly because the Be window in the EDX detector hinders the detection of characteristic X-rays of light elements such as C, N, O and Na. WDS is suitable for analysis of light elements, but the measurement of beam sensitive microparticles requires the minimisation of the beam current and the measurement time. A semi-quantitative analytical method based on EPMA using an ultra-thin window EDX detector was developed. It was found that the matrix and geometric effects that are important for low-energy X-rays can be reliably evaluated by Monte Carlo calculations. Therefore, the quantification part of the method contains reverse Monte Carlo calculation done by iterative simulations. The method was standardised and tested by measurements on single particles with known chemical compositions. Beam-sensitive particles such as ammonium-sulphate and ammonium-nitrate were analysed using a liquid nitrogen cooled sample stage. The shape and size of the particles, which are important for the simulations, were determined using a high-magnification secondary electron image. Individual marine aerosol particles collected over the North Sea by a nine-stage Berner cascade impactor were analysed using this new method. Preliminary results on five samples and 4500 particles show that the method can be used to study the modification of sea-salt particles in the troposphere.     

EPMA Major and Trace Element Analysis in Garnet and its Petrological Application

 A comparison between major and trace element concentrations in garnet performed by electron microprobe (EPMA) technique is reported. Quantitative spot analyses and X-ray maps of major elements (Fe, Mg, Mn, Ca) and the trace element yttrium in garnets from metamorphic rocks are presented. The selected garnet samples come from meta-pelitic and meta-basic specimens belonging to the tectonic unit of the Monte Rosa Nappe (Western Alps). In the metapelites, the quantitative Y distribution maps display a prominent increase at the core, the Y abundance varying by over two orders of magnitude, from about 80 ppm (rim) to over 2100 ppm (core). The Y profiles show well defined patterns with sharp features that do not correlate with major element distributions. A roughly comparable pattern can be supposed only with Mn. The Y distribution suggests that the diffusion of Y through the garnet is very slow compared to the major elements, thus the Y results are suitable for geothermometric estimates. In the metabasites, the Y spatial distribution is characterised by an increasing content from the core to the rim, displaying a zoning pattern opposite to the metapelite garnet. Quantitative EPMA analyses range from 1100 ppm at the rim to values lower than the detection limit at the core. Therefore, the Y content in the garnet can be related to several chemical and physical variables such as the bulk rock composition and the phase assemblage. In particular, in the xenotime-bearing metapelitic system the Y distribution seems to be correlated with metamorphic peak temperature.     

Application of Thin-Window EPMA to Environmental Problems in Hungary

 The combination of single-particle analysis using thin-window EPMA and a reverse Monte Carlo quantification procedure has been proven to provide semi-quantitative elemental concentrations, including light elements. The capabilities of the method are demonstrated through three different environmental applications in Hungary. Lake Balaton is the largest lake in central Europe. It is suspected that the atmosphere is an important source of environmental deterioration of the Lake, relative to the pollutant supply by rivers and direct discharges. Thin-window EPMA results of around 25,000 individual particles indicated that the composition of the aerosol did not show characteristic seasonal variation, it was more correlated to the daily meteorological circumstances. A serious heavy metal pollution of the river Tisza occurred on 10 March 2000, from a mine-dumping site in Romania. Sediment samples were taken from the main riverbed at six sites, on 16 March 2000. In some of the samples, pyrite type particles were observed in high abundance, revealing their dumping-site origin. Biomass burning for energy production has been increased recently, mostly in households and for space heating. Fly-ash and bottom-ash samples were collected at a 600 kW heating plant at Sződliget during standard winter operation. Most particles contained over 65% of unburned organic substance. The concentrations of major elements were obtained for each individual particle using thin-window EPMA, providing accurate matrix composition for further μ-XRF investigations of the trace element composition.     

EPMA and Quantitative MCs+-SIMS of Metal-DLC Coating Materials

 Compositional characterization of metal-DLC (metal-containing diamond-like carbon) hard coatings is carried out by (WDS)-EPMA and MCs⁺-SIMS. EPMA enables accurate (± 5% relative) quantitative analysis including minor concentrations (0.1–10 at%) of N, O and Ar. Under conditions of “near-surface” EPMA (E₀ < 10 keV) the influence of surface oxide films on “pure” metal standards may be a limiting factor in respect of accuracy. Depth profiling of sufficiently “thick” layered structures (film thickness ≥ 2 μm) is carried out by EPMA-line scans along mechanically prepared bevels. The depth resolution is about 0.2 μm. SIMS in the MCs⁺-mode enables high resolution (< 20 nm) depth profiling of metal-DLC layered structures including the determination of H (1–20 at%). MCs⁺-SIMS, i.e. employing Cs⁺ primary ions and monitoring MCs⁺ molecular secondary ions (M is the element of interest) is presented as a promising route towards sufficiently accurate (10–20%) SIMS-quantification. Matrix-independent relative sensitivity factors for MCs⁺-SIMS are derived from homogeneous coating materials defined by EPMA. EPMA proves to be also useful to detect problems related to SIMS of Ar in metal-DLC materials. The combination EPMA-SIMS is demonstrated as an effective analytical strategy for quality control in industrial production and to support the development of metal DLC layered structures with optimum tribological properties.     

Quantitative Analysis of Thin Aluminium-Oxynitride Films by EPMA

 Thin films of aluminium oxynitride with diverse composition were prepared by dc-magnetron sputtering of aluminium, utilising sputtering power as well as argon, oxygen and nitrogen gas flows to vary the composition. Since film properties depend mainly on the content of incorporated oxygen and nitrogen, a method for quantitative analysis of the main constituents based on electron probe micro analysis with energy dispersive detection was developed. The excellent precision of the quantitative results for aluminium as well as oxygen and nitrogen are shown. Furthermore, a film layer analysis program was applied for the quantification of several films deposited under the same deposition parameters on silicon wafers, from 520 nm down to 40 nm thickness, showing that electron probe micro analysis with energy dispersive detection is a reliable method for quantitative compositional analysis of thin aluminium oxynitride films down to approximately 20 nm thickness. Since this method of analysis provides only bulk information, expected inhomogeneities of the depth distribution of the film components were checked by secondary ion mass spectrometry depth profiles of two thin films and correlated to the EPMA results. The thickness of the films was determined by ellipsometry. Received September 1, 1998     

Detection Limits of Grazing-Exit EPMA for Particle Analysis

 In conventional electron probe microanalysis (EPMA), the electron-induced X-rays are measured at large take-off angles of about 45°. In the grazing exit EPMA (GE-EPMA) method, they are measured at small angles (< 1°). X-rays emitted from deep positions can not be detected at grazing exit angles due to refraction effects at sample-vacuum interface; therefore, it is possible to measure X-rays emitted only from near the surface with a low background. GE-EPMA is especially suitable for the analysis of particles deposited on a flat sample carrier. The detection limits of GE-EPMA were investigated for artificial particles (Al₂O₃, Fe₂O₃ and PbO₂ , particle sizes: 1 ∼ 18 μm) deposited on flat sample carriers of Au thin films–Si wafers. The detection limits improved with decreasing exit angle. The detection limits for characteristic X-rays at an exit angle of approximately 1.1° were 2–4 times lower than at 45°. A minimum detection limit of ca. 0.1% was obtained for Al in small particles.     

Comparative Analysis of a Solar Control Coating on Glass by AES, EPMA, SNMS and SIMS

 A solar control coating was analysed by different methods of surface analysis with respect to the layer sequence and the composition and thickness of each sublayer. The methods used for depth profiling were Auger electron spectroscopy, electron probe microanalysis, secondary neutral mass spectroscopy and secondary ion mass spectroscopy based on MCs⁺. The structure of the coating was unknown at first. All methods found a system of two metallic Ag layers, embedded between dielectric SnO_X layers. Additionally, thin Ni-Cr layers of 1–2 nm were detected on top of the Ag layers. Thus the detected layer sequence is SnO_X/Ni-Cr/Ag/SnO_X/Ni-Cr/Ag/SnO_X/glass. The Ni:Cr ratio in the nm-thin layers could be quantified by every method, the Cr fraction corresponding to less than one monolayer. We compare the capabilities and limitations of each method in routinely investigating this solar control coating. Importance was attached to an effective investigation. Nevertheless, by combining all methods, measuring artefacts could be uncovered and a comprehensive characterisation of the system was obtained.     

EPMA of Porous Media: A Monte Carlo Approach

 We examine the influence of the sample porosity on the X-ray emission from mesoporous alumina bombarded with kilovolt electrons. Experimental results show that there is a loss of X-rays (Al Kα and O Kα) from those samples when compared to a fully dense mono-crystalline alumina (sapphire), which depends on the X-ray line, the measurement time and the embedding medium. Both geometrical and charging effects may be responsibl e for this signal loss. Monte Carlo simulations of the X-ray intensity emitted from porous alumina, using different models to describe sample porosity, show that the geometrical effect of porosity itself cannot account for the X-ray loss. Charge trapping effect and/or its combination with porosity is therefore expected to be the major cause of the signal loss.     

Determination of Nb, Ta, Zr and Hf in Micro-Phases at Low Concentrations by EPMA

 Depletion of high field strength elements (HFSE: Nb, Ta, Zr, Hf ) relative to other lithophile trace elements in arc magmas and variations of Nb/Ta and Zr/Hf ratios in mantle-derived rocks can be addressed through studies of minerals, which concentrate and fractionate these elements. The presence of rutile, a common accessory Ti-oxide phase in various mantle rocks, has often been invoked to explain the Nb and Ta depletion in arc lavas because it has the highest HFSE abundances among the known mantle minerals. In this study, we measure the concentrations of Nb, Ta, Zr and Hf (at > 200 ppm) in rutile of two metasomatized mantle lherzolites using a Cameca SX-100 electron microprobe and obtain Nb/Ta ratios with an accuracy of about ± 5%. Mass balance calculations indicate that ≤ 1−5% of Nb and Ta in the rocks reside in major minerals and that the balance is hosted by accessory Ti-oxides. The Nb/Ta ratios vary significantly in nearby rutile grains in both peridotites (17–33, average 23; 12–37, average 21). Therefore, individual rutile grains may not be representative of the total grain population. However, Nb/Ta ratios measured in the bulk rock lherzolites by solution ICP-MS (21 ± 0.3) are within the analytical error of the average Nb/Ta values calculated for 5–7 rutile grains in both samples. These results emphasise that a representative grain selection must be analysed in order to determine trace elements contents of bulk rocks from data on accessory phases.     

EPMA and GDOES in Functional-Gradient Hardmetal Systems

 WDS-EPMA and GDOES measurements were applied for the characterization of the in-situ surface modification process of (Ti, W)C-based hardmetals by in-diffusion of carbon and nitrogen leading to complicated microstructures composed by layers and/or gradients. The calibration of the analytical techniques was performed with chemically characterized specimens prepared by diffusional techniques or hot-pressing. In addition, for the GDOES method, certified reference materials (CRMs) have been used. The work was started from a re-investigation of the W-C and Mo-C systems in order to measure homogeneity regions of the various phases prepared by use of diffusion couples. Subsequently, the metal and non-metal contents of various sequences of layers and/or gradients in functional-gradient hardmetals (FGHs) were measured. Both methods were employed as complementary tools to get information about the element-concentration distribution in FGHs so that a better understanding of the influence of process parameters for fabrication of special hardmetals is possible.     

Depth Profile Analysis on the Nanometer Scale by a Combination of Electron Probe Microanalysis (EPMA) and Focused Ion Beam Specimen Preparation (FIB)

Electron probe microanalysis (EPMA) offers high sensitivity and high accuracy in quantitative measurements of chemical compositions and mass coverages. Owing to the low detection limits of the wavelength-dispersive technique, monolayers with mass coverages of about 0.05µgcm^–2 can be detected. Assuming a density of 5gcm^–3 this corresponds to a thickness of 0.1nm. With these advantages in mind, EPMA was extended to depth profile analysis in the sub-micron range using a surface removal technique.The present paper shows how depth profile analysis can be improved by combining EPMA and the focused ion beam (FIB) technique. The focused ion beam system uses a Ga⁺ ion beam. The ion beam allows the milling of defined geometries on the nanometer scale, so that very shallow bevels with exactly defined angles in relation to the surface can be obtained. Low surface damage is expected due to low sputtering effects. Calibrated WDX measurements along the bevel deliver quantitative concentration depth profiles. First results obtained with this new combination of methods will be presented for a multilayered sample used in optical data storage.     

XPS Depth Profile Analysis of a Thin Non-Conducting Titanate Superlattice

 New BaTiO₃-SrTiO₃ (BTO-STO)-superlattices which may be interesting for future electronic applications have been investigated by X-ray photoelectron spectroscopy (XPS) depth profiling. At first XPS measuring conditions were optimized for that non-conducting and thin layer systems (21 nm double layer thickness) considering the practical instrumental limitations. Second a simulation of the sputtering process for the concrete experimental conditions were done by a dynamic T-DYN code. By comparison of experimental and simulated depth profiles the maximum sample roughness could be estimated to be in the range of 2 nm.     

Volcanic Quartz Growth Zoning Identified by Cathodoluminescence and EPMA Studies

 Quartz is a common phenocryst in acidic volcanic rocks but rarely contributes to the understanding of the magmatic or eruptive processes due to its apparent chemical homogeneity. Cathodoluminescence studies indicate that volcanic quartz is strongly zoned with respect to the trace elements. The determination of this zoning can help to elucidate the crystallisation history of the magmatic rock. Polished thin sections from rhyolitic and dacitic volcanic rocks were examined by cathodoluminescence mounted on an optical microscope and analysed by electron probe microanalysis. X-ray mapping of the trace elements incorporated in the zoned quartz was also performed. The interpretation of the data obtained from these different analytical methods provides useful information on the chemical zoning of single crystals and therefore on the chemico-physical conditions of the melt from which the quartz crystallised. Quartz from dacitic rock shows a complex cathodoluminescence oscillatory zoning, consisting of large red-brown and blue bands concentric to a resorbed core. Quartz from rhyolite shows simpler cathodoluminescence zoning, consisting of dark to bright blue luminescent bands. The EMPA analyses and the X-ray maps combined with the cathodoluminescence images suggest a clear correlation between the concentrations of some trace element (mainly Al and Ti, acting as activators of luminescence) and the cathodoluminescence colours pattern zoning. The incorporation of these trace elements during the growth of the crystals is linked to small variations in the thermo-chemical properties of the magma such as temperature and chemical composition.     

EPMA Measurements of Diffusion Profiles at the Submicrometre Scale

 Concentration profiles due to (inter)diffusion in materials may require high spatial resolution. These profiles may be measured by electron probe microanalysis, which allows one to determine the elemental composition with a good accuracy provided measurement ‘artefacts’ can be accounted for. Standard phenomena are usually corrected by commercial softwares that assume a homogeneous elemental composition in the analysed area. However, in the case of a diffusion process on a small scale, the composition is no longer homogeneous and the effect of the hemispherical volume of the X-ray emission on the spatial resolution of the concentration profiles, and consequently on the diffusion coefficients, has to be considered. Moreover, (secondary) fluorescence across interfaces or interphases has to be evaluated. A radial X-ray distribution associated with the characteristic depth distribution, φ(ρz), allows for the definition of a 2D X-ray emission function that enables the computation of the entire process for a given concentration profile.     

Determination of Nitrogen in Duplex Stainless Steels by EPMA

 We describe the quantitative determination of low nitrogen contents (< 1 %wt) in highly-alloyed, duplex stainless steels by EPMA using a calibration curve. Five monophasic stainless steels with various nitrogen concentrations were used for the calibration curve, namely, three highly alloyed steels prepared in the laboratory, with nitrogen contents 0.066, 0.565 and 1.016 %wt and two industrial alloys with nitrogen concentrations 0.174 %wt and 0.023 %wt. For these samples, the plot of nitrogen X-ray intensities versus nitrogen contents exhibited an acceptable linear relationship. Average nitrogen contents of the analysed duplex alloys were found to vary in the range < 0.045–0.07 %wt in ferrite and 0.52–0.77 %wt in austenite. These results suggest that the nitrogen solubility limit in ferrite has been reached; on the other hand, nitrogen solubility in austenite increases due to the high manganese and low nickel contents.     

An Expert System for EPMA

 Experimental parameters for an electron microprobe are numerous and their choice influence the quality and accuracy of the analysis results. The expert system intends to optimise the choice of every parameter as well as to automate the stages of a quantitative analysis for all kinds of materials. In short, the expert system aims to master the accuracy of results and to control the time of analysis. The starting point of the expert system is an interactive questionnaire about the sample and about the expectations of the analysis (the accuracy of the result and/or the duration of the analysis). Then, the expert system makes a semi-quantitative analysis on the sample. It gives the sample composition in a first approximation. This is necessary for the expert system to optimise all the parameters for an accurate quantitative analysis. Each parameter is modelled by a specific algorithm. The expert system selects the parameters by minimising, at every stage, the statistical error generated by the algorithm. In this way, it operates in a sufficiently independent way to create a quantitative analysis configuration adapted to an unknown sample. The expert system employs the usual methods of quantification but the results are obtained in half the time. Moreover, the accuracy of the result is guaranted by the statistical error calculation as the expert system is running.     

金刚石复合界面扩散系数的电子探针研究

结合金刚石与硬质合金复合材料的研究，对含轻元素的界面扩散问题提出了能谱仪与波谱仪原位互补分析研究的方法，从扩散深度和扩散浓度分布两种角度探讨了扩散系数的电子探针测定；同时讨论了该方法在金刚石与硬质合金复合钻头上的应用。     

EPMA Present and Future

 In electron probe microanalysis the objective has always been to gain the maximum information from a sample. To this end efforts have been directed at increasing detection efficiency, reducing the size of the excitation volume, analysing more difficult specimens and improving the accuracy of quantitative measurements whilst, at the same time, attempting to minimise the complexity of the analysis. The present paper comments on the progress that has been made in these areas, focuses on the latest hardware and software developments taking place and looks forward to how these may impact on the technique of EPMA.     

Data Preprocessing in Peak Shape Analysis of Auger Electron Spectra

 Factor analysis is an established method of peak shape analysis in Auger electron spectrometry. The influence of different commonly used data preprocessing tools onto the results of factor analysis is demonstrated on AES depth profiles of multilayers and implantation profiles. For the analysis of Auger electron spectra it has been traditional to differentiate spectra by Savitzky and Golay’s method to remove background and to elucidate changes in peak shape. For phosphorus implanted in titanium it is shown that background removal works not ideal so that inelastic losses of the Ti(LMM) Auger peak can affect the result of factor analysis for the P(LVV) peak located at ca. 250 eV lower in kinetic energy. The contribution of such losses to the background can be corrected by shifting the spectra so that the high energy side above the peak equals zero. Numerical differentiation can introduce correlated error into the data set. To diminish edge effects the reduction of filter width at the edges and cutting off the outermost data points is recommended. The precision of spectrum reproduction is considered as a crucial test for the number of principal components. The reliability factor is investigated as a measure for the goodness of spectrum reproduction.