Arsenic in marine tissues — The challenging problems to electrothermal and hydride generation atomic absorption spectrometry

Abstract

Analytical problems in determination of arsenic in marine tissues are addressed. Procedures for the determination of total As in solubilized or extracted tissues with tetramethylammonium hydroxide and methanol have been elaborated. Several typical lyophilized tissues were used: NIST SRM 1566a ‘Oyster Tissue’, BCR-60 CRM ‘Trace Elements in an Aquatic Plant ( Lagarosiphon major)’, BCR-627 ‘Forms of As in Tuna Fish Tissue’, IAEA-140/TM ‘Sea Plant Homogenate’, NRCC DOLT-1 ‘Dogfish Liver’ and two representatives of the Black Sea biota, Mediterranean mussel ( Mytilus galloprovincialis) and Brown algae ( Cystoseira barbata). Tissues (nominal 0.3 g) were extracted in tetramethylammonium hydroxide (TMAH) 1 ml of 25% m/v TMAH and 2 ml of water) or 5 ml of aqueous 80% v/v methanol (MeOH) in closed vessels in a microwave oven at 50 °C for 30 min. Arsenic in solubilized or extracted tissues was determined by electrothermal atomic absorption spectrometry (ETAAS) after appropriate dilution (nominally to 25 ml, with further dilution as required) under optimal instrumental parameters (pyrolysis temperature 900 °C and atomization temperature 2100 °C) with 1.5 μg Pd as modifier on Zr–Ir treated platform. Platforms have been pre-treated with 2.7 μmol of zirconium and then with 0.10 μmol of iridium which served as a permanent chemical modifier in direct ETAAS measurements and as an efficient hydride sequestration medium in flow injection hydride generation (FI-HG)–ETAAS. TMAH and methanol extract 96–108% and 51–100% of As from CRMs. Various calibration approaches have been considered and critically evaluated. The effect of species-dependent slope of calibration graph or standard additions plot for total As determination in a sample comprising of several individual As species with different ETAAS behavior has been considered as a kind of ‘intrinsic element speciation interference’ that cannot be completely overcome by standard additions technique. Calibration by means of CRMs has given only semi-quantitative results. The limits of detection (3 σ) were in the range 0.5–1.2 mg kg − 1 As dry weight (wt.) for direct ETAAS analysis of extracts in both TMAH and MeOH. Within-run precision (RSD%) was 5–15% and 7–20% for TMAH and MeOH extracts at As levels 4–50 mg kg − 1 dry wt., respectively. The hydride active fraction of As species in extracts, i.e. the sum of toxicologically-relevant arsenic species (inorganic As(III), inorganic As(V), monomethylarsonate (MMA) and dimethylarsinate (DMA)) was determined by FI-HG–ETAAS in diluted tissue extracts. Arsine, monomethylarsine and dimethylarsine were generated from diluted TMAH and MeOH extracts in the presence of 0.06–0.09 mol l − 1 hydrochloric acid and 0.075 mol l − 1 l-cysteine. Collection, pyrolysis and atomization temperatures were 450, 500, 2100 and 2150 °C, respectively. The LODs for the determination of hydride forming fraction (arsenite + arsenate + MMA + DMA) in TMAH and MeOH extracts were in the range 0.003–0.02 mg kg − 1 As dry wt. Within-run precision (RSD%) was 3–12% and 3–7% for TMAH and methanol extracts at As levels 0.15–2.4 mg kg − 1 dry wt., respectively. Results for the hydride forming fraction of As in TMAH and MeOH extract as % from the certified value for total As (for CRMs) or vs. the total As in TMAH extract (for real marine samples) are generally in agreement.