Comparative study of organic dyes with time-of-flight static secondary ion mass spectrometry and related techniques.

A series of cationic, zwitterionic and anionic fluorinated carbocyanine dyes, spin-coated on Si substrates, were measured with time-of-flight static secondary ion mass spectrometry (TOF-S-SIMS) under Ga(+) primary ion bombardment. Detailed fragmentation patterns were developed for all dyes measured. In the positive mode, the resulting spectra showed very intense signals for the precursor ions of the cationic dyes, whereas the protonated signals of the anionic dyes were hardly detected. Differences of three orders of magnitude were repeatedly observed for the secondary ion signal intensities of cationic and anionic dyes, respectively. All measured dyes yielded mass spectra containing several characteristic fragment ions. Although the secondary ion yields were still higher for the cationic than the anionic dye fragments, the difference was reduced to a factor of < or =10. This result and the fact that M(+), [M + H](+) or [M + 2H](+) are even-electron species make it very likely that the recorded fragments were not formed directly out of the (protonated) parent ions M(+), [M + H](+) or [M + 2H](+). In the negative mode, none of the recorded spectra contained molecular information. Only signals originating from some characteristic elements of the molecules (F, Cl), the anionic counter ion signal and some low-mass organic ions were detected. A comparative study was made between TOF-S-SIMS, using Ga(+) primary ions, and other mass spectrometric techniques, namely fast atom bombardment (FAB), electrospray ionization (ESI) and matrix-assisted laser desorption/ionization (MALDI). The measurements showed that MALDI, ESI and FAB all give rise to spectra containing molecular ion signals. ESI and FAB produced M(+) and [M + H](+) signals, originating from the cationic and zwitterionic dyes, in the positive mode and M(-) and [M - H](-) signals of the anionic and zwitterionic dyes in the negative mode. With MALDI, molecular ion signals were measured in both modes for all the dyes. Structural fragment ions were detected for FAB, ESI and MALDI in both the positive and negative modes. Compared with the other techniques, TOF-S-SIMS induced a higher degree of fragmentation.

Laser microprobe mass spectrometry of quaternary phosphonium salts: Direct versus matrix-assisted laser desorption.

The use of laser microprobe mass spectrometry (LMMS) for the structural characterization of thermolabile quaternary phosphonium salts has been evaluated. A comparison has been made between LM mass spectra obtained by direct analysis of "neat" organic salts and the corresponding "matrix-assisted" LM mass spectra. Main limitations of LMMS for the direct analysis of neat organic salts (i.e., no matrix) result from (1) formation of artifact ions that originate from thermal degradation and surface recombination reactions and (2) poor shot-to-shot reproducibility of the spectra. Dilution of the organic salts in a suitable, UV-absorbing matrix (e.g., nicotinic acid) significantly enhances the quality of the LM mass spectra. Improvements are: (1) an increase of the ion yield of preformed cations, (2) reduction or elimination of thermal decomposition and other deleterious surface reactions, and (3) a much better shot-to-shot spectral reproducibility. An interesting analytical feature is that these LM mass spectra, which contain only a few matrix peaks, can be obtained for subnanogram amounts of sample.The results also show that triphenylphosphonium salts with polycyclic aromatic substituents can be used as "molecular thermometers" to probe both the temperatures experienced by the sample molecules during the laser-induced desorption ionization process and the internal energies of the desorbed ion species. In this way, quaternary phosphonium salts can be used for evaluating whether improvements have been achieved by applying different sample treatments. Comparison of four different matrices (i.e., nicotinic acid, ammonium chloride, glycerol, and 3-nitrobenzylalcohol) indicates that the effectiveness of a matrix to reduce thermal degradation and to decrease the internal energies of the ions depends on the UV-absorption characteristics and the volatilization/sublimation temperature of the matrix material.

Combined characterization of composite tabular silver halide microcrystals by cryo-EFTEM/EELS and cryo-STEM/EDX techniques.

The combination of cryo-energy filtering transmission electron microscopy (EFTEM)/electron spectroscopic diffraction (ESD)/electron energy-loss spectroscopy (EELS) and cryo-energy-dispersive X-ray (EDX) analysis in the scanning transmission (STEM) and scanning (SEM) modes was applied for the characterization of composite tabular Ag(Br,I) microcrystals. A low-loss fine structure in EEL spectra between 4 and 26 eV was attributed to excitons and plasmons possibly superimposed with interband transitions and many-electron effects. The contrast tuning under the energy-filtering in the low-loss region was used to image the crystal morphology, defect structure (random dislocations and ¿111¿ stacking faults) and bend and edge contours as well as electron excitations in the microcrystals. Sharp extra reflections at commensurate positions in between the main Bragg reflections and diffuse honeycomb contours in ESD patterns of the microcrystals taken near the [111] zone were assigned to the number of defects in the shell region parallel to the grain edges and polyhedral clusters of interstitial silver cations, respectively. The imaginary part of the energy-loss function, Im (-1/epsilon), and the real and imaginary parts, epsilon1 and epsilon2, of the dielectric permittivity were determined by means of a Kramers-Kronig analysis. An assignment of exciton peaks based on calculations of electronic band structure of silver bromide is proposed. Inner-shell excitation bands of silver halide were detected in line with EDX-analyses. The energy-loss near-edge structure (ELNES) of the AgM4,5-edge governed by spin-orbital splitting between the 3d3/2- and 3d5/2-states has been evaluated. Combined silver and halide distributions were obtained by a three-window method (EFTEM) and by EDX/STEM including area mapping and line profiling of iodide.

Mathematical description of a direct current glow discharge in argon.

In order to achieve a better understanding of the glow discharge, different models have been developed for the different species present in the plasma. An overview of the models is given and some typical results are presented. These results include, among others, the densities and energy distributions of the plasma species, the electric field and potential distribution, the contribution of different ionization mechanisms to the ionization of argon and sputtered atoms, the relative contribution of different plasma species to the sputtering process, and the variation of the cathode dark space length and the electrical current as functions of voltage and pressure. The validity of the present models is supported by the good agreement of the calculated current-voltage curves with experiment.

Determination of noble metals by neutron-activation analysis.

The scope and limitations of neutron-activation analysis in the field of the platinum metals are discussed. Special attention is given to interferences that occur in determination of noble metals in platinum metals and in natural samples. The problem of standards, carriers and chemical yield determination is emphasised. A recommendation is given of modern techniques for counting radionuclides formed by neutron activation of the platinum metals.

Determination of gold in germanium by neutron-activation analysis.

Gold was determined in gold-doped germanium by neutron-activation analysis. The irradiated samples were dissolved in aqua regia, in the presence of gold carrier. The matrix activity was eliminated by distillation, and gold precipitated in the residue with hydroquinone. The gold fraction was essentially radiochemically pure. For a neutron dose of 5 x 10(15)n/ cm(2), a sensitivity of 1 ppM Au can be obtained, with a 2.4-g sample. The samples investigated contained from 24 to 36 ppM Au.

Elemental analysis of high-purity solids by mass spectrometry.

The applications of mass spectrometry in the determination of trace elements in some of the high-purity solid materials used in modern technology are reviewed.

Determination of sulphate content of anionic surfactants: sodium di-2-ethylhexyl sulphosuccinate.

A simple nephelometric method for the determination of sulphates in the anionic surfactant Aerosol-OT is described. The method is based on the formation and stabilization of a barium sulphate dispersion in a water-alcohol mixture and measurement of the turbidity at 650 nm. Beer's law is followed for sodium sulphate concentrations in Aerosol-OT ranging from 0 to 7000 microg/g. The lower determination limit is about 100-200 microg/g and the reproducibility about 5% for typical commercial Aerosol-OT samples. The method can also be used for other anionic surfactants by adjusting the water-ethanol ratio of the solvent.

Determination of oxygen and silicon in rocks by 14-MeV neutron-activation analysis and its precision.

Oxygen and silicon have been determined in the six new U.S. Geological Survey standard rocks and two standard refractory materials by 14-MeV neutron-activation analysis, followed by both single and multichannel analyser counting of the induced (16)N (4.5-8 MeV gamma radiation) and (28)Al activities (1.78 MeV photopeak). Owing to the long analysing time per pulse, dead-time corrections are necessary in multichannel analyser counting. Four methods were investigated in this work: counting in the live-time mode without additional correction, short counting corrected by an external live-time scaler, the method of Bartoseck, and the method of Schonfeld. Each measurement was controlled by a simultaneous measurement with a single-channel analyser. Silica ignited at 1000 degrees was used as a reference. Correction was made for the interfering elements, F, P, Al, Fe and Mg, using literature data. Attention was paid to neutron, gamma and beta attenuation. For oxygen the mean coefficient of variation for a single determination with a single-channel analyser was 1.7 % for silicon 1.1 %. The mean results for single-channel counting were compared with literature values.

Mass-spectrometric multielement analysis of copper and copper-base alloys, with electrical detection.

Relative sensitivity coefficients have been determined for 21 elements in copper and its alloys, with iron as an internal standard, by spark-source mass-spectrometry with electrical detection and magnetic peak-switching. Twenty calibration standards ranging from pure copper to 60%-copper alloys were used. The sensitivity coefficients measured appear to be independent of the elemental concentration and are obtained with a mean precision of 15%.

Relative sensitivity coefficients for the analysis of steel by spark-source mass-spectrometry.

Quantitative analysis by spark-source mass-spectrometry requires the knowledge of socalled sensitivity coefficients for the elements being determined. Five series of analyses have been carried out on five different steel standard reference materials (NBS-SRM 661-665), using photoplate detection. The relative sensitivity coefficients (S(R)) of Ti, V, Cr, Mn, Co, Ni, Cu, As, Zr, Nb, Mo, Sn, Sb, La, Ta and W were determined vs. iron as an internal standard. The S(R) values were independent of the elemental concentration. A relative standard deviation of about 15% was obtained. The accuracy as confirmed by comparing the results for a pure iron sample with those obtained by neutron-activation analysis was within the same limits.

Accuracy of analysis of steels by use of a spark-source mass spectrometer with electrical detection.

The precision and accuracy of spark-source mass spectrometry with electrical detection has been studied, with five steel standard reference materials (NBS-SRM 661-665). Two different modes of analysis have been evaluated, magnetic scanning with electrical detection of the individual ions in sequence, using the total ion-current as reference, and magnetic switching between masses, with current integration. Measurements of isotope abundances have been used to evaluate the precision. The relative sensitivity coefficients of Ti, V, Cr, Mn, Co, Ni, Cu, As, Zr, Nb, Mo, Sb, La, Ta and W have been determined vs. iron as an internal standard. The accuracy of analyses based on these experimentally measured relative sensitivity coefficients was confirmed by comparing the results for a pure iron sample with those obtained by neutron-activation analysis.

Comparison of experimental and calculated relative sensitivity coefficients for spark-source mass-spectrometry of metals.

Sensitivity calibration has been performed for the spark-source mass-spectrometric analysis of iron, copper and aluminium matrices, with standard reference materials. The experimental relative sensitivity coefficients, corrected for discrimination effects in the mass spectrometer, are compared with values obtained with various empirical approaches to calculate relative sensitivity coefficients for an r.f. spark. The best correlation found is only of the order of 50%.

Sensitivity calibration for the spark-source mass-spectrometric analysis of aluminium and its alloys.

The analysis of aluminium and its alloys has been studied with use of five standards from Johnson Matthey and ten aluminium alloys from Aluminium Pechiney. The relative sensitivity coefficients for Mg, Al, Si, Ti, Cr, Mn, Ni, Cu, Zn, Sn and Pb were determined vs. iron as an internal standard. They were obtained, by using electrical detection, with a mean precision of 10% relative standard deviation. The sensitivity coefficients measured appear to be independent of the elemental concentration, but for some elements, especially the more volatile ones, remarkable changes were noticed when sparking conditions were altered and when the electrode temperature was changed by cooling with liquid nitrogen.

A radiochemical study of the fire assay of rhodium with copper and the subsequent rhodium-copper separation.

(102,102m)Rh has been used as a tracer to study the fire-assay collection of rhodium in copper. The rhodium and copper are separated on a Dowex 50 cation-exchanger in hydrochloric acid of pH 1.5.

A comparison between secondary-ion mass-spectrometry and spark-source mass-spectrometry for the quantitative analysis of steel wire.

A comparison has been made between the results of the matrix-ion species ratio (MISR) method for quantification of secondary-ion mass-spectrometry data and spark-source mass-spectrometry analysis using photoplate detection for analysis of the steel basis of AlZn coated wire products. For SIMS quantification a suitable set of sensitivity factors, corrected for the actual surface sampling condition, was used. The results of both methods compare well. The SIMS results were, for most elements, within 25% of the concentration determined by SSMS. This could indicate that reasonably accurate results can be obtained by using the matrix-ion species ratio method for SIMS.

Study of high-voltage breakdown and material consumption in spark-source mass spectrometry and their significance in analytical applications.

The breakdown voltage has been found to be dependent on the gap width between the electrodes and on the melting point of the sample elements in spark-source mass-spectrometry (SSMS). The number of discharges per pulse train and the time required to reach the first discharge depend only on the chosen breakdown voltage. The spark gap is proportional to the "radius" of the volume sampled (for a given element) and this radius is linearly related to the reciprocal of the melting point of the elements (23 different elements, metals or semiconductors), when fixed spark-parameters are used. The effect of electrode temperature on material consumption can be qualitatively explained by a fictive increase or decrease in melting point of the element. Knowledge of the relations between the different spark and instrumental parameters and the volume or weight of sample consumed can be applied to the study of the homogeneity of samples, to in-depth analysis by SSMS and to the analysis of microsamples.

Numerical investigation of ion-energy-distribution functions in single and dual frequency capacitively coupled plasma reactors.

Ion-energy-distribution functions (IEDFs) are numerically investigated in capacitively coupled (cc) radio frequency (rf) Ar/CF(4)/N(2) discharges by a one-dimensional particle-in-cell/Monte Carlo model. The simulation considers electron-neutral collisions, various kinds of collisions of ions (Ar+, CF+3, N+2, F-, and CF-3) with neutral, positive-negative ion, and electron-ion recombination. The influence of pressure, applied voltage amplitude, and applied frequency on the Ar+, CF+3, and N+2 IEDFs is presented. The dependence on the frequency regime is investigated by simulations of the Ar/CF(4)/N(2) mixture in single (13.56 MHz) and dual frequency (2+27 MHz or 1+27 MHz) cc reactors. A comparison of the simulation results with analytical calculations in a collisionless rf sheath is discussed. The results show that the IEDFs shift toward the low energies with increasing pressure or decreasing applied voltage amplitude. The Ar+ and N+2 IEDFs exhibit secondary maxima due to the charge transfer collisions. The CF+3 IEDF has a peak at high energies in consistency with the average sheath potential drop. The IEDFs in the dual frequency regime are broad and bimodal.

Electron anisotropic scattering in gases: a formula for Monte Carlo simulations.

The purpose of this Brief Report is to point out the mistake in a formula for anisotropic electron scattering, previously published in Phys. Rev. A 41, 1112 (1990), which is widely used in Monte Carlo models of gas discharges. Anisotropic electron scattering is investigated based on the screened Coulomb potential between electrons and neutral atoms. The approach is also applied for electron scattering by nonpolar neutral molecules. Differential cross sections for electron scattering by Ar, N2, and CH4 are constructed on the basis of momentum and integrated cross sections. The formula derived in this paper is useful for Monte Carlo simulations of gas discharges.

Local and fast relaxation phenomena after laser-induced photodetachment in a strongly electronegative rf discharge.

A one-dimensional self-consistent particle in cell/Monte Carlo method is used to study the local (at the laser impact region) and fast relaxation phenomena after laser-induced photodetachment in a strongly electronegative SiH(4)/H(2) rf discharge. The relaxation process of the local densities of the charged plasma species has been studied in association with the time evolution of the local electric field. The phenomena predicted theoretically about the relaxation processes, such as the potential well, the electrostatic oscillation, the long lasting potential structure, the distortion of the early potential perturbation on the measurement of negative ion temperature, and the depression in the positive ion density profile at the edges of the laser impact region, have been confirmed by our simulation results. Compared to the relaxation in weakly electronegative discharges, the local and even the global electric field in strongly electronegative discharges, has been weakened strongly after photodetachment. The relaxation of the local electric field lasts 100 rf cycles with the recovery of the local electron density and the local electron energy. The electrostatic oscillation exhibited as the deviation in quasineutrality, is very strong and continues over several rf cycles in our case. The large dip in the center of the positive ion density profile, observed in the experiment, is also reproduced by our model.