David E. Davey

Arsenic Speciation and Toxicity in Biological Systems

Although it is now commonly accepted that toxicity and bioavailability varies with As species, extensive research has been carried out on biological and environmental samples to assess toxicity and risk associated with As exposure based on total concentrations that may be in error. The health investigation guideline for the Australian environmental protection measure is 100 mg/kg (As(tot)), which would cause potential risk to human health if all the As present in a sample were bioavailable (ANZECC 1992). Similarly, the MPC for As in food is 1 mg/ kg (fresh weight), but this concentration may include contributions from As(III), AsV, and all organic species. Thus, a food substance, such as seafood, could have a high total concentration exceeding the guidelines, but most of the As would be in forms that are nontoxic to humans; i.e., the bioavailability is low, and the food would therefore be perfectly safe to eat. On the other hand, a food that has high bioavailability of As consequently is more toxic. Overall, it appears that contamination of water by As is probably more harmful to humans than As in food grains or vegetables, because As bioavailability in water is generally higher than its bioavailability in food. Nevertheless, As in food crops could make significant contribution toward total daily intake. Therefore, failure to consider the contribution of As species on their bioavailability could introduce a substantial bias into the estimation of risks associated with exposure as well as evaluation of As toxicity. In conclusion, As must be regarded as an important environmental toxicant because of its acute and chronic toxic properties and extensive presence in the environment. Much remains to be learned about its toxicology and biochemistry for better understanding of this important contaminant.

Speciation of arsenic in ground water samples: A comparative study of CE-UV, HG-AAS and LC-ICP-MS

The performance of capillary electrophoresis-ultraviolet detector (CE-UV), hydride generation-atomic absorption spectrometry (HG-AAS) and liquid chromatography-inductively coupled plasma mass spectrometry (LC-ICP-MS) have been compared for the speciation of arsenic (As) in groundwater samples. Two inorganic As species, arsenite (As(III)), arsenate (As(V)) and one organo species dimethyl arsenic acid (DMA) were mainly considered for this study as these are known to be predominant in water. Under optimal analytical conditions, limits of detection (LD) ranging from 0.10 (As(III), AsT) to 0.19 (DMA) mug/l for HG-AAS, 100 (As(III), DMA) to 500 (As(V)) mug/l for CE-UV and 0.1 (DMA, MMA) to 0.2 (As(III), As(V)) mug/l for LC-ICP-MS, allowed the determination of the above three species present in these samples. Results obtained by all the three methods are well correlated (r(2)=0.996*** for total As) with the precision of <5% R.S.D. except CE-UV. The effect of interfering ions (e.g. Fe(2+), Fe(3+), SO(4)(2-) and Cl(-)) commonly found in ground water on separation and estimation of As species were studied and corrected for. Spike recovery was tested and found to be 80-110% at 0.5mug/l As standard except CE-UV where only 50% of the analyte was recovered. Comparison of these results shows that LC-ICP-MS is the best choice for routine analysis of As species in ground water samples.

On-line preconcentration and separation of palladium, platinum and iridium using α-amino pyridine resin with flame atomic absorption spectrometry

A flow injection on-line microcolumn method has been studied for the separation and preconcentration of noble metals, Pd, Pt and Ir, followed by flame AAS determination. An alpha-amino pyridine resin was used as preconcentration reagent. Non-noble metals studied could be eliminated by 2M HCl solution in the rinsing stage. A group-eluent for all Pd, Pt and Ir was studied. Separation of individual noble metals could also be accomplished using selective eluents. This approach can be used both for the separation and determination of Pd, Pt and Ir in samples. The sensitivities and the sample frequency have been improved. The method has been applied to certified samples.

Determination of copper in natural waters by batch and oscillating flow injection stripping potentiometry

Copper trace analysis in natural waters rich in chloride ions and organic matter was performed by means of a copper-selective electrode (Cu-ISE), inductively coupled plasma-atomic emission spectrometry (ICP-AES), anodic stripping voltammetry (ASV), batch potentiometric stripping analysis (PSA) and oscillating flow injection stripping potentiometry (OFISP). The results obtained by the first two analytical methods showed that copper was present mainly in the form of various complexes with inorganic (mainly chloride) ions and organic ligands. The organic constituents impeded the application of ASV in the quantitative determination of labile copper concentration in the water samples mainly due to adsorption of organic matter onto the working electrode. Consistent analytical data concerning the labile fraction of copper were obtained by both batch PSA and OFISP in hydrochloric acid medium when standard addition or direct calibration were used. However, discrepancies between the batch PSA results obtained by the two calibration methods mentioned above in nitric acid medium were observed. Unlike batch PSA, OFISP appears to be versatile with respect to the sample matrix in both standard addition and direct calibration approaches. On the basis of this and the other well-known advantages of OFISP it was concluded that this analytical technique is suitable for fast and inexpensive trace heavy metal analysis in natural waters with unknown matrices.

Potentiometric flow-injection determination of iodide and iodine.

A potentiometric flow-injection system is described, in which iodide in the concentration range 10(-6)-10(-1)M may be determined at rates of up to 360 samples/hr at a flow-rate of 17.8 ml/min. Iodine, after on-line reduction to iodide with 0.1M sodium metabisulphite, can also be determined, with a throughput of 60 samples/hr for 10(-5)-10(-3)M iodine. Analyses of two pharmaceutical preparations for iodide and iodine are reported, and the results are in reasonable agreement with titrimetric values.

Effect of the pH value of the precipitation solution on the CO sensitivity of α-Fe2O3

Abstract Using Fe(NO3)3·9H2O as raw material, α-Fe2O3 was prepared by the precipitation method. The CO sensitivities of sensors made from the product both in air and N2 backgrounds were tested. It was found that the sensitivities are significantly affected by the pH value of the precipitation solution. The CO gas sensitivity of sensors prepared from materials with different precipitation pH values varies in a manner indicating predominance of different gas sensing mechanisms. The response is also dependent on the operating temperature and the nature of the background stream gas. Based on the experimental results, combined with studies using Fourier transform infrared spectroscopy, X-ray photoelectron spectroscopy, X-ray diffraction and scanning electron microscopy, the possible sensing mechanisms are discussed. It seems that the chemically bonded free cluster OH plays a key role in the improved CO gas sensitivity. The BET method was employed for specific surface area determination.

Trace gold determination by on-line preconcentration with flow injection atomic absorption spectrometry.

A reverse-phase extraction column method has been employed as an on-line preconcentration technique for trace gold analysis by flame atomic absorption spectrometry. Di(methylheptyl)methyl phosphonate (DMHMP) loaded onto a macroporous resin was used as the immobilized phase. A thiourea-HCl solution was found successful in eluting the gold. The experimental parameters were optimized by Simplex Optimization, with 17 tests needed to obtain optimal conditions. Sensitivities of 5.2 mug/l. and 2.3 mug/l., with sample frequencies of 45/hr and 48/hr, were obtained by using single and dual-columns respectively. The recoveries for mixed composition samples were 93-110%.

Rapid determination of fluoride in potable waters by potentiometric flow injection analysis

Abstract A potentiometric flow injection analysis system is described, enabling tap water and other fluoride-bearing matrices of low interferent level to be determined at the rate of 360 samples per hour using an electrode polished with slurried alumina. Important parameters, such as carrier stream composition, sample volume and detector cell design are discussed with respect to this system, Fluoride electrodes regenerated with silver fluoride and silver epoxy are evaluated in flow injection mode, both before and after polishing.

Flow-injection determination of phosphate with a cadmium ion-selective electrode

A flow-injection method is described, in which phosphate standards are introduced into a reagent stream containing Cd(2+) ,resulting in the formation of Cd(3)(PO(4))(2). The associated reduction in free metal concentration is sensed by a cadmium-selective electrode. With the exception of major interference from iodide and moderate interference from bromide and thiocyanate, the system exhibits excellent response to phosphate and selectivity over several common anions in solutions buffered at pH 8.4. A maximum sampling rate of 160/hr is possible for phosphate standards in the concentration range 10(-1)-10(-1)M with a 10(-4)M Cd(2+) reagent stream at a total flow-rate (carrier and reagent stream combined) of 8.4 ml/min.

An optimized on-line preconcentration system for analysis of trace gold in ore samples

A flow injection on-line preconcentration technique, combined with graphite furnace atomic absorption spectrometry for the determination of gold in ore samples has been presented. An alpha-amino pyridine resin (AP) was loaded onto a microcolumn as the preconcentration reagent and a solution of 90% acetone-5% HCl-5% H(2)O was used for the gold elution. The Scaled Simplex Method was employed to optimize the flow injection manifold. The selected factors were optimized by the simultaneous consideration of two responses, absorbance and column retention efficiency, with only 12 tests. A detection limit of 0.065 mug Au/l (3sigma) with a relative standard deviation of 4.3%, was achieved. Ore samples containing gold have been analysed successfully using the proposed method.