On-line coupling of an ultraviolet titanium dioxide film reactor with a liquid chromatography/hydride generation/inductively coupled plasma mass spectrometry system for continuous determination of dynamic variation of hydride- and nonhydride-forming arsenic species in very small microdialysate samples

We describe a method for continuously monitoring both hydride- and nonhydride-forming arsenic species in 10-microL microdialysate samples by coupling together on-line high-performance liquid chromatography (HPLC), a post-column UV/TiO2 film reactor, and hydride generation (HG) inductively coupled plasma mass spectrometry (ICP-MS). To maximize the signal intensities of the desired arsenic species, we optimized the photocatalytic oxidation efficiency of the analyte species and used a rapid on-line pre-reduction process to convert the oxidized species into As(III) prior to HG-ICP-MS determination. The UV/nano-TiO2 film reactor was manufactured by coating nano-TiO2 onto the interior of a glass tube. Impregnation and sol-gel methods were employed to deposit the TiO2 films, and their effectiveness for the oxidation of organic arsenicals was compared. To enhance the decomposition efficiency of organic arsenicals, we investigated the effects of the acidity and the composition of the column effluent. Because of the improved HG efficiency toward the tested arsenicals and the adoption of a segmented flow technique to retain the peak resolution in our on-line LC-UV/nano-TiO2 film reactor-HG-ICP-MS instrument, the detection limits for arseneous acid [As(III)], monomethylarsonic acid (MMA), dimethylarsinic acid (DMA), arsenic acid [As(V)], and arsenobetaine (AsB) were all in the submicrogram-per-liter range (based on 3 sigma) for 10-microL injections. A series of validation experiments--analyses of certified reference urine and rabbit serum samples--indicated that these methods can be applied satisfactorily to the continuous determination of As(III), MMA, DMA, As(V), and AsB in blood and in the extracellular space of target organs.     

Application of sample pre-oxidation of arsenite in human urine prior to speciation via on-line photo-oxidation with membrane hydride generation and ICP-MS detection

A pre-oxidation procedure which converts arsenite [As(III)] into arsenate [As(v)] was investigated in urinary arsenic speciation prior to on-line photo-oxidation hydride generation with ICP-MS detection. This sample pre-oxidation method eliminates As(III) and As(v) preservation concerns and simplifies the chromatographic separation. Four oxidants, Cl2, MnO2, H2O2 and I3-, were investigated. Chlorine (ClO-aq) and MnO2 selectively converted As(III) into As(v) in pure water samples, but the conversion was inefficient in the complex urine matrix. Oxidation of As(III) by H2O2 was least affected by the urine matrix, but the removal of excess H2O2 at pH 10 proved difficult. The most appropriate oxidant for the selective conversion of As(III) into As(v) with minimal interference from the urine matrix is I3- at pH 7. Unlike H2O2, excess oxidant can be easily removed by the addition of S2O3(2-). The I3-(-)S2O3(2-) treatment on a fortified sample of reconstituted NIST SRM 2670 freeze dried urine indicated that arsenobetaine (AsB), dimethlyarsinic acid (DMA), monomethylarsonic acid (MMA) and As(v) were not chemically degraded with recoveries ranging from 95 to 102% for all arsenicals. Sample clean-up involved pH adjustment prior to C18 filtration in order to achieve efficient As(III) conversion and quantitative recoveries of AsB and DMA. The concentrations determined in NIST SRM 2670 freeze dried urine were AsB 17.2 +/- 0.5, DMA 56 +/- 4 and MMA 10.3 +/- 0.3 with a combined total of 83 +/- 5 micrograms L-1 (+/- 2 sigma).     

Photolytic oxidation of As species for determination of total As (including the ‘hidden’ As fraction) in coastal seawater by hydride generation-atomic fluorescence spectrometry

Ultraviolet irradiation (photolysis) in alkaline medium was applied for pretreatment of seawater samples so as to accurately determine total As by continuous-flow hydride generation-atomic fluorescence spectrometry. This sample pretreatment is meant to convert non-reducible As forms into inorganic As, which easily forms arsine. The optimised parameters were the treatment time and the pH of the medium. The behaviour of four hydride-reactive As species [As(III), As(V), MMA, DMA], and AsB, i.e. a typical non-hydride-reactive As species, when subjected to UV irradiation was studied. UV irradiation at pH 1 lead to conversion of all species into As(V) with the exception of AsB and DMA. Conversions of DMA and AsB into As(V) at pH 11 in less than 30 min were observed under UV irradiation. The limit of detection of As (measured as As(V)) by hydride generation-atomic fluorescence spectrometry was 0.1 μg/L and the repeatability of the oxidation procedure was about 10%. The method was applied to determination of total and directly reducible As at 11 sampling points of the Galician Coast (Atlantic Ocean, Spain). Total As concentrations were in the range 1.4-4.8 μg/L. A significant As fraction, between 20 and 44%, depending on the sampling point, corresponded to non-reducible As which was converted by UV irradiation into hydride-reactive As. This fraction should represent the sum of DMA, which yields a low sensitivity in the continuous flow-AFS system, and the hidden As fraction.     

Complementary chromatography separation combined with hydride generation-inductively coupled plasma mass spectrometry for arsenic speciation in human urine

This study aimed to establish complementary high performance liquid chromatography (HPLC) methods including three modes of separation: ion pairing, cation exchange, and anion exchange chromatography, with detection by inductively coupled plasma mass spectrometry (ICPMS). The ion pairing mode enabled the separation of inorganic arsenate (As(V)), monomethylarsonic acid (MMA(V)), and dimethylarsinic acid (DMA(V)). However, the ion pair mode was unable to differentiate inorganic arsenite (As(III)) from arsenobetaine (AsB); instead, cation exchange chromatography was used to isolate and quantify AsB. Anion exchange chromatography was able to speciate all of the aforementioned arsenic species. Potential inaccurate quantification problem with urine sample containing elevated concentration of AsB, which eluted immediately after As(III) in anion exchange or ion pairing mode, was overcame by introducing a post-column hydride generation (HG) derivatization step. Incorporating HG between HPLC and ICPMS improved sensitivity and specificity by differentiating AsB from hydride-forming arsenic species. This paper emphasizes the usefulness of complementary chromatographic separations in combination with HG-ICPMS to quantitatively determine concentrations of As(III), DMA(V), MMA(V), As(V), and AsB in the sub-microgram per liter range in human urine.     

Simultaneous pressurized enzymatic hydrolysis extraction and clean up for arsenic speciation in seafood samples before high performance liquid chromatography-inductively coupled plasma-mass spectrometry determination

The feasibility of pressurized conditions to assist enzymatic hydrolysis of seafood tissues for arsenic speciation was novelty studied. A simultaneous in situ (in cell) clean-up procedure was also optimized, which speeds up the whole sample treatment. Arsenic species (As(III), MMA, DMA, As(V), AsB and AsC) were released from dried seafood tissues using pepsin as a protease, and the arsenic species were separated/quantified by anion exchange high performance liquid chromatography (HPLC) coupled to inductively coupled plasma-mass spectrometry (ICP-MS). Variables inherent to the enzymatic activity (pH, temperature and ionic strength), the amount of enzyme (pepsin), and factors affecting pressurization (pressure, static time, number of cycles and amount of dispersing agent, C-18) were fully evaluated. Pressurized assisted enzymatic hydrolysis (PAEH) with pepsin can be finished after few minutes (two cycles of 2 min each one plus 3 min to reach the hydrolysis temperature of 50 °C). A total sample solubilisation is not achieved after the procedure, however it is efficient enough for breaking down certain bonds of bio-molecules and for releasing arsenic species. The developed method has been found to be precise (RSDs lower than 6% for As(III), DMA and As(V); and 3% for AsB) and sensitive (LOQs of 18.1, 36.2, 35.7, 28.6, 20.6 and 22.5 ng/g for As(III), MMA, DMA, As(V), AsB and AsC, respectively). The optimized methodology was successfully applied to different certified reference materials (DORM-2 and BCR 627) which offer certified AsB and DMA contents, and also to different seafood products (mollusks, white fishes and cold water fishes).     

Determination of urinary arsenic species using an on-line nano-TiO2 photooxidation device coupled with microbore LC and hydride generation-ICP-MS system

We have developed an on-line digestion device-based on the nano-TiO₂-catalyzed photooxidation of arsenic species—for coupling between microbore anion-exchange chromatography (μ-LC) and hydride generation (HG)-inductively coupled plasma mass spectrometry (ICP-MS) systems that can be used for the determination of urinary arsenic species. To maximize the signal intensities of the desired arsenic species, we optimized the photocatalytic oxidation efficiency of the analyte species and developed a rapid on-line pre-reduction process for converting the oxidized species into As(III) prior to HG-ICP-MS determination. Under the optimized conditions for the nano-TiO₂-catalyzed photooxidation-i.e., using 1 g of nano-TiO₂ per-liter, at pH 5.2, and illuminating for 3 min- As(III), monomethylarsenoic acid (MMA), and dimethylarseinic acid (DMA) can be converted quantitatively into As(V). To attain maximal hydride generation efficiency, 0.5% Na₂S₂O₄ solution, which can reduce As(V) to As(III) virtually instantaneously upon on-line mixing, was added as a pre-reductant prior to performing the HG step. In light of all the HG efficiency of tested arsenicals were improved and a segmented-flow technique was employed to avoid the loss of peak resolution when using our proposed on-line μ-LC-UV/nano-TiO₂/HG-ICP-MS, the detection limits for As(III), MMA, DMA, and As(V) were all in the range of sub-microgram-per-liter (based on 3 sigma). A series of validation experiments-analysis of neat and spiked urine samples-indicated that our proposed methods can be applied satisfactorily to the determination of As(III), MMA, DMA, and As(V) in urine samples.     

Application of fast ultrasound water-bath assisted enzymatic hydrolysis--high performance liquid chromatography-inductively coupled plasma-mass spectrometry procedures for arsenic speciation in seafood materials

Enzymatic hydrolysis of seafood materials for isolating arsenic species (As(III), As(V), DMA and AsB) has been successfully performed by assisting the procedure with ultrasound energy (35 kHz) supplied by an ultrasound water-bath. The use of pepsin, as a proteolytic enzyme, under optimized operating conditions (pH 3.0, temperature 40 °C, enzyme to sample ratio of 0.3) led to an efficient assistance of the enzymatic process in a short period of time (from 4.0 to 30 min). The enzymatic extract was then subjected to a clean-up procedure based on ENVI-Carb™ solid phase extraction (SPE). An optimized anion exchange high performance liquid chromatography (HPLC) coupled to inductively coupled plasma-mass spectrometry (ICP-MS) permitted the fast separation (less than 15 min) of six different arsenic species (arsenite, As(III); arsenate, As(V); dimethylarsinic acid, DMA; and arsenobetaine, AsB; as well as monomethylarsonic acid, MMA; and arsenocholine, AsC) in a single run. Relative standard deviations (n = 11) of the over-all procedure were 7% for AsB and DMA, 11% for As(III) and 9% for MMA. HPLC–ICP-MS determinations were performed using aqueous calibrations covering arsenic concentrations of 0, 5, 10, 25, 100 and 200 μg L⁻¹ (expressed as arsenic) for As(III), As(V), MMA, DMA and AsC; and 0, 125, 250, 500, 750, 1000 and 2000 μg L⁻¹ (expressed as arsenic) for AsB. Germanium (5 μg L⁻¹) was used as an internal standard. Analytical recoveries from the anion exchange column varied from 96 to 105% (enzymatic digests spiked with low target concentrations), from 97 to 104% (enzymatic digests spiked with intermediate target concentrations), and from 98 to 103% (enzymatic digests spiked with high target concentrations). The developed method was successfully applied to two certified reference materials (CRMs), DORM-2 and BCR 627, which offer certified AsB and DMA contents, and also to different seafood samples (mollusks, white fish and cold water fish). Good agreement between certified and found AsB concentrations was achieved when analyzing both CRMs; and also, between certified and found DMA concentrations in BCR 627. In addition, the sum of the different arsenic species concentrations found in most of the analyzed samples was statistically similar to the assessed total arsenic concentrations after a total sample matrix decomposition treatment.     

Arsenic speciation patterns in freshwater fish

Muscle of 16 freshwater fish (9 different species belonging to 4 different families) was analysed for arsenic species using HPLC separation (anion and cation exchange) followed by on-line UV-decomposition, hydride generation and AFS detection. The main arsenic compounds found in the extracts were arsenobetaine (AsB), which accounted for 92–100% of extractable arsenic in species of salmonids (Salmo marmoratus, Oncorhynchus mykiss, Salmo trutta m. fario), and dimethylarsinic acid (DMAA), which accounted for 75% of extractable arsenic in burbot (Lota lota). AsB was also found in lower concentrations in almost all other fish species analysed (Silurus glanis, L. lota, Barbus barbus, Rutilus pigus virgo, Chondrostoma nasus). Arsenite (As(III)) and trimethylarsine oxide (TMAO) were detected in low concentrations in some representatives of Cyprinidae only (R. pigus virgo, C. nasus). Except in salmonids, an unknown cationic compound was present in most of the samples in relatively low concentrations. Cluster analysis of the generated data seems to indicate that there is a correlation between fish family and the arsenic speciation pattern. This is especially clear for the salmonids which show a completely separate cluster and thus a very distinct arsenic speciation pattern.     

Evaluation of atomic fluorescence spectrometry as a sensitive detection technique for arsenic speciation

The potential of coupling anion-exchange high-performance liquid chromatography, hydride generation and atomic fluorescence spectrometry (HPLC–HG–AFS) for arsenic speciation is considered. The effects of hydrochloric acid and sodium tetrahydroborate concentrations on signal-to-background ratio, as well as argon and hydrogen flow rates, were investigated. Detection limits for arsenite, dimethylarsinic acid (DMA), monomethylarsonic acid (MMA) and arsenate were 0.17, 0.45, 0.30 and 0.38 μg l⁻¹, respectively, using a 20-μl loop. Linearity ranges were 0.1–500 ng for As(III) and MMA (as arsenic), and 0.1–800 ng for DMA and As(V) (as arsenic). Arsenobetaine (AsB) was also determined by introducing an on-line photo-oxidation step after the chromatographic separation. In this case the limits of detection and linear ranges for the different species studied were similar to the values obtained previously for As(V). The technique was tested with a human urine reference material and a volunteer's sample. © 1998 John Wiley & Sons, Ltd.     

Speciation measurements by HPLC-HGAAS of dimethylarsinic acid and arsenobetaine in three candidate lyophilized urine reference materials.

Speciation measurements of dimethylarsinic acid (DMA) and arsenobetaine (AsB) in three candidate lyophilized urine reference materials are described. The measurements were based on cation-exchange liquid chromatography coupled to hydride generation atomic absorption spectrometry with on-line digestion of the organic. As species by alkaline persulfate solution aided by ultraviolet radiation. Arsenic concentrations as DMA were significantly different in the three samples. The mean values for the three samples were 4.1 +/- 0.3, 55.3 +/- 1.2 and 134.1 +/- 1.5 micrograms l-1, respectively. No significant differences in AsB concentrations were observed among the three samples. The mean As concentrations as AsB in the three samples were 17.4 +/- 0.4, 17.7 +/- 0.2 and 17.5 +/- 0.3 micrograms l-1, respectively. By off-line digestion of the urine samples, total As concentrations in the three materials were also obtained. The mean values were 23.4 +/- 0.3, 76.6 +/- 1.6 and 151.3 +/- 1.8 micrograms l-1, respectively. These results correlated well with the results obtained by neutron activation analysis in our laboratory (r = 0.999; p < 0.0001).