Alessandro D'Ulivo

Thermally stabilized iridium on an integrated, carbide-coated platform as a permanent modifier for hydride-forming elements in electrothermal atomic absorption spectrometry. Part 3. Effect of L-cysteine

Iridium (2 µg) was deposited on a carbide-coated platform, pre-treated with about 1.2–1.3 µmol of Zr or W, and was evaluated as a permanent modifier in electrothermal atomic absorption spectrometry. The noble metal is not vaporized from the integrated platform of the transverse-heated graphite atomizer, provided that the atomization and clean-out temperatures do not exceed 2050–2100 and 2100–2200 °C, respectively. Under comparable conditions, Pd exhibits much worse thermal behaviour, being volatilized above 1300 and 1500 °C from Zr- and W-treated platforms, respectively. Pyrolysis–atomization curves were plotted for numerous hydride-forming and volatile analytes: Sb, As, Bi, Cd, Pb, Te, Tl, Sn and Se. The best characteristic masses for integrated absorbance measurements with the Ir–Zr-treated platforms are 92, 30, 176, 2.4, 35, 50, 65, 71 and 45 pg, respectively. Vaporization temperatures are generally above 1000 °C, except for Cd. The effect of atomization temperature on sensitivity in peak height and integrated absorbance measurements is discussed. Double peaks were observed for Bi and Te with Ir–W-treated platforms.

Thermally stabilized iridium on an integrated, carbide-coated platform as a permanent modifier for hydride-forming elements in electrothermal atomic absorption spectrometry. Part 2. Hydride generation and collection, and behaviour of some organoelement species

The in situ collection of volatile hydrides in an electrothermal atomizer with an integrated platform pre-treated with 110 µg of Zr or 240 µg of W and 2 µg of Ir for permanent modification was studied. An optimization study of the performance characteristics of an automated FI-HG–ETAAS system based on an FI hydride generator interfaced with a transverse-heated graphite atomizer and longitudinal Zeeman-effect background correction was elaborated. The HG step for AsIII, AsV, BiIII, SbIII, SbV, SeIV, SnIV and TeIV, as well as for several alkylated species of As and Sn, was optimized by means of a full factorial 32 design, the factors being the concentrations of acid and tetrahydroborate (or their supply rates in µmol s–1).The corresponding regression equations are tabulated, and representative response surfaces and contour diagrams are plotted. All inorganic hydrides except for SnH4, are generated and collected with high efficiency at tetrahydroborate concentrations of 0.25–0.4% m/v, sample acidity of 1.5–3 mol l–1 HCl, trapping temperatures of 400 °C and a purge gas flow of argon of 100–130 ml min–1. The optimum conditions for stannane and alkyltin hydrides are: pH 1–4, tetrahydroborate concentrations of 0.2–0.4% m/v, trapping temperatures between 400 and 600 °C and argon flow rates of 60–120 ml min–1. Arsine, monomethylarsine and dimethylarsine are effectively collected on both coatings at temperatures between 400 and 500 °C and purge gas flow rates of 70–120 ml min–1. Optimum HG conditions differ strongly for AsIII, AsV, monomethylarsonate and dimethylarsinate species with this FI system, unless L-cysteine is added. Organoelement species of As, Sn and Se are thermally stabilized in a similar manner on both Ir–Zr- and Ir–W- treated platforms, the least stable species being selenornethionine and trimethylselenonium. The best levelling-off effect on the integrated absorbance for different analyte species (isoformation) is observed for As and the worst for organotins, particularly for trialkylated species such as tributyltin, trimethyltin and trimethylselenonium. Relatively better isoformation is achieved for organotins on Ir–W- and for organoselenium on Ir–Zr-treated platforms. The long-term stability of the Ir–Zr and Ir–W modifier coatings during at least 600–700 thermal cycles is demonstrated. The Ir–Zr treatment is preferred to Ir–W for hydride trapping, owing to lower atomization temperatures, longer lifetime of the atomizer and an absence of double peaks. Such peaks persist for Bi and Te on Ir–W-treated platforms. The best characteristic masses in integrated absorbance measurements with Ir–Zr-treated platforms are close to those for the direct injection mode, viz., 35, 107, 83, 43, 104, 48, 31, 32, 153, 146, 148, 145 and 152 pg for ASIII, BiIII, SbIII, SeIV, SnIV, TeIV, monomethylarsonate, dimethylarsinate, monomethyltin, dimethyltin, trimethyltin, diethyltin and monobutyltin, respectively. Analytical results for As, Sb and Se in certified reference materials (water and autoclave-decomposed sediments) are in good agreement with the certified contents.

Mercury speciation by liquid chromatography coupled with on-line chemical vapour generation and atomic fluorescence spectrometric detection (LC-CVGAFS)

Reverse phase chromatography (RPC) coupled on-line with UV-vis diode array detector (DAD) and cold vapour generation atomic fluorescence spectrometry (CVGAFS) is proposed for the speciation and determination of inorganic and organic mercury (methylmercury, ethylmercury and phenylmercury) in the form of cysteine, penicillamine and glutathione complexes. The mercury-thiol complexes are separated on a C(18) Reverse Phase column and oxidized on-line with bromine, generated in situ by KBr/KBrO(3) in HCl medium, in order to fast convert organic mercury species to inorganic mercury in less than 2.5s, at room temperature, in a 30cm knitted coil. Hg(II) is selectively detected by AFS in a Ar/H(2) miniaturized flame after sodium borohydride reduction to Hg(0). Under optimized conditions, on-line bromine treatment gives recoveries of thiol-complexed methylmercury, ethylmercury and phenylmercury with respect to inorganic mercury ranging between 79 and 85%, 80 and 85%, 63 and 76%, respectively, depending on the complexing thiol employed. Optimized elution conditions were provided in the three complexing agents. The detection limits (LODc) for inorganic mercury, methylmercury, ethylmercury and phenylmercury, in the optimized conditions complexed with thiols were about 16, 18, 18 and 20pg (as mercury), respectively, a relative standard deviation (R.S.D) ranging between 1.5 and 2.0%, and a linear dynamic range between 0.1 and 100ng injected. LC-DAD-CVGAFS method has been validated by analysing two certificate reference material, DORM-2 and NIES CRM 13, obtaining 98+/-6 and 97+/-5% of methylmercury recovered, respectively.

Cold vapour atomic fluorescence studies on the behaviour of mercury(II) and mercury(II)-thiol complexes. An alternative route for characterization of –SH binding groups

The behaviour of Hg II and Hg II -thiol complexes (RSH=L-cysteine, DL-penicillamine, propane-2-thiol, glutathione, thiosalicylic acid) following their reduction with alkaline sodium tetrahydroborate to give Hg 0 has been studied by using a continuous flow reaction system coupled with atomic fluorescence spectrometric (AFS) detection. The quantitative reduction of Hg II to Hg 0 takes place with a specific amount of sodium tetrahydroborate according to the stoichiometric reaction of mercury with alkaline NaBH 4 . The complete reduction of Hg II -thiol complexes to Hg 0 requires a molar excess of NaBH 4 of up to six orders of magnitude, depending on the type of complex. Under an appropriate excess of reductant, Hg II and its thiol complexes are not distinguishable giving the same AF molar response. The method allows the discrimination of Hg II from Hg II -thiol complexes without any preliminary separation. Applications to the indirect titration of thiols and to the determination of the number of accessible }}n1SH groups in pure ovalbumin samples are reported.

Masking agents in the determination of selenium by hydride generation technique

Abstract The effects of several masking agents in the determination of selenium by hydride generation was studied using a continuous flow hydride generator coupled with atomic absorption spectrometry. A miniature argon hydrogen diffusion flame was employed as the atomizer. The effects of masking agents (KI, NaSCN, thiourea, l -cysteine, 1,1,3,3 tetramethyl-2-thiourea) were studied both in the absence and in the presence of selected interfering species (Cu, Ag, Au, Ni, Co, Pd, Pt and Fe) and using different addition strategies of the masking agents to the reaction system: in batch mode, either to sample or NaBH4 reducing solution; in on-line mode, to the sample either before or after the HG step). The combined effect of some masking agents was also investigated. The addition mode of the masking agent to the reaction system could play a decisive role in the control of interfering processes both in the absence and in the presence of concomitants. The addition of NaSCN to the reducing solution of NaBH4 produced a moderate catalytic effect, similar to the one obtained in the presence of KI, and improved the tolerance limits for Cu, Ni, Co and Pd. Linearity of calibration graphs was unaffected by the on-line addition of 1,1,3,3 tetramethyl-2-thiourea to sample solution, while under similar conditions thiourea produced dramatic curvature of calibration graphs. The combined use of KI (added to reducing solution) and 1,1,3,3 tetramethyl -2-thiourea (added on-line to the sample) exhibited masking properties comparable but not superior to those of thiourea, except for Pt and Pd for which good tolerance limits were achieved. In the absence of KI in the reductant solution the masking efficiency of 1,1,3,3 tetramethyl-2-thiourea was considerably lowered. The addition of some masking agents such as thiourea, l -cysteine and 1,1,3,3 tetramethyl thiourea on-line to reaction solution after the NaBH4+KI reduction step, could be highly effective in the control of Cu and Ag interferences. The method was applied to determination of trace of selenium in pure copper standard reference materials.

Selenium hydride atomization, fate of free atoms and spectroscopic temperature in miniature diffusion flame atomizer studied by atomic absorption spectrometry

The mechanism of hydride atomization and the fate of free atoms was investigated in the miniature diffusion flame. Selenium hydride was used as a model for the other hydrides. Mercury vapor was employed as an analyte to study physical processes, such as macroscopic movements and free atom diffusion, controlling the distribution of free analyte atoms in the observation volume, separately from chemical reactions of the free atoms. Free atoms were detected by atomic absorption spectrometry. Spectroscopic temperature measurements based on atomic absorption at 196.1 and 204.0 nm Se lines were used to determine the temperature distribution. The spatial temperature distribution was highly inhomogeneous ranging from 150°C to 1300°C. The whole flame volume is actually a cloud of hydrogen radicals maintaining analyte in the free atom state since hydrogen radicals formed in outer zone of the flame diffuse to its cooler inner parts.

Determination of cadmium in aqueous samples by vapour generation with sodium tetraethylborate(III) reagent

A volatile cadmium compound can be generated from aqueous solutions containing Cd2+ at trace levels, by reaction with sodium tetraethylborate(III) at room temperature. Preliminary experiments failed to identify this compound but it appears to be sufficiently stable to be used for analytical purposes. A detection limit for the compound of 0.2 ng ml–1(3σ of a blank determination, 1-ml sample volume) can be achieved using non-dispersive atomic fluorescence spectrometry with argon-hydrogen mini-flame atomisation whereas with conventional atomic absorption spectrometric measurements in a flame-heated silica tube the detection limit is 1 ng ml–1(3δ of the base-line noise, 1-ml sample volume).

Interferences in hydride atomization studied by atomic absorption and atomic fluorescence spectrometry

Abstract A hydride atomizer able to operate in the flame-in-tube mode and in the miniature diffusion flame mode was used to investigate interferences of arsenic in selenium atomization. A twin-channel continuous flow hydride generator was utilized to eliminate liquid phase interferences. Both atomic absorption and atomic fluorescence detectors (EDL sources) were employed. The miniature diffusion flame can tolerate interferent concentrations up to 70 μg ml −1 . The magnitude of interferences in the flame-in-tube atomizer is controlled by the distance between the atomization and detection zones. The best tolerance to interferents, comparable with that in the miniature diffusion flame, was obtained for the minimum distance of the zones. The figures were deteriorated by two orders of magnitude when increasing the distance between the observation and the atomization zones to 50 mm. Also a curvature and rollover of calibration graphs was observed when increasing the distance. The presence of the interferent enhanced substantially the curvature and rollover, so that the magnitude of observed interferences was dependent on the analyte concentration. All the observed interferences and the calibration graph curvature are due to the decay of free analyte atoms by reactions in the free space. The nature of the species formed is discussed. No significant depletion of hydrogen radicals was observed. As demonstrated by measurements in the miniature diffusion flame, the species formed can be reatomized by interaction with hydrogen radicals with an efficiency better than 90%.

Determination of antimony by continuous hydride generation coupled with non-dispersive atomic fluorescence detection

A sensitive method for the determination of Sb at ultratrace levels was developed by coupling continuous hydride generation with non-dispersive atomic fluorescence detection. A miniature argon–hydrogen diffusion flame was employed as the atomizer and a commercially available electrodeless discharge lamp as the light source. One of the main problems was the scattering signal generated by small droplets of solution which markedly deteriorated the signal-to-noise ratio. A simple way to remove the scattering signal was to operate under mild reaction conditions in order to minimize droplet formation. Under the optimized conditions, a limit of detection of 22 pg cm–3 of Sb (3s of the blank) was achieved, with a precision of 1.2% at the 5 ng cm–3 level and the calibration graphs were linear over more than 4 decades of concentration. L-Cysteine was employed both in the pre-reduction step and in the control of the interference effects arising from concomitant elements and acid mixtures. The analytical procedure was applied to the determination of Sb in certified reference materials of sediments, metallic copper and riverine water.

Improving the analytical performance of hydride generation non-dispersive atomic fluorescence spectrometry. Combined effect of additives and optical filters ☆

Abstract The effects of tetrahydroborate and acid concentration and the presence of l -cysteine and thiourea were investigated in the determination of As, Bi and Sn using continuous flow hydride generation atomic fluorescence spectrometry (HG AFS). The aim was to find conditions allowing the control of those effects exerting negative influence on the analytical performance of the HG AFS apparatus. The effects taken into account were: (i) the radiation scattering generated by carryover of solution from the gas–liquid separator to the atomizer; (ii) the introduction of molecular species generated by tetrahydroborate decomposition into the atomizer; and (iii) interference effects arising from other elements in the sample matrix and from different acids. The effects (i) and (ii) could be controlled using mild reaction conditions in the HG stage. The effect of HG conditions on carryover was studied by radiation scattering experiments without hydride atomization. Compromised HG conditions were found by studying the effects of tetrahydroborate (0.1–20 g l−1) and acid (0.01–7 mol l−1) concentration, and the addition of l -cysteine (10 g l−1) and thiourea (0.1 mol l−1) on the HG AFS signals. The effect of optical filters was investigated with the aim of improving the signal-to-noise ratio. Optical filters with peak wavelengths of 190 and 220 nm provided an improvement of detection limits by factors of approximately 4 and 2 for As and Te, respectively. Under optimized conditions the detection limits were 6, 5, 3, 2, 2 and 9 ng l−1 for As, Sb, Bi, Sn, Se and Te, respectively. Good tolerance to various acid compositions and sample matrices was obtained by using l -cysteine or thiourea as masking agents. Determination of arsenic in sediment and copper certified reference materials, and of bismuth in steel, sediment, soil and ore certified reference material is reported.