Rapid screening of water soluble arsenic species in edible oils using dispersive liquid–liquid microextraction

Highlights

• Water-soluble arsenic compounds are extracted by using DLLME.

• Arsenic is measured by electrothermal atomization atomic absorption spectrometry.

• A simple procedure to check the presence of the most toxic arsenic species is given.

Abstract

A methodology for the non-chromatographic screening of the main arsenic species present in edible oils is discussed. Reverse dispersive liquid–liquid microextraction was used to extract water soluble arsenic compounds (inorganic arsenic, methylarsonate, dimethylarsinate and arsenobetaine) from the edible oils into a slightly acidic aqueous medium. The total arsenic content was measured in the extracts by electrothermal atomic absorption spectrometry using palladium as the chemical modifier. By repeating the measurement using cerium instead of palladium, the sum of inorganic arsenic and methylarsonate was obtained. The detection limit was 0.03 ng As per gram of oil. Data for the total and water-soluble arsenic levels of 29 samples of different origin are presented. Inorganic arsenic was not found in any of the samples marketed as edible oils.

Introduction

The need to know the exact nature and behaviour of the different arsenic compounds present in the environment and in foods does not need to be emphasised (Cornelis, Caruso, Crews, & Heumann, 2005). However, this is not an easy task since the element is ubiquitous, its chemistry complex (giving arise to a large number of compounds, especially in foods), and it is often found in low concentrations that hinder interpretation of any analytical data obtained (Francesconi and Kuehnelt, 2002, Rumpler et al., 2008). When dealing with foods, measurement of the total arsenic content is necessary but not necessarily sufficient, for risk assessment since the toxicity of the different species varies widely. The usual way of measuring the total content of this element in foods involves mineralisation of the sample before using a sensitive analytical technique such as hydride generation atomic absorption spectrometry (HG-AAS), atomic fluorescence spectrometry (HG-AFS) or inductively coupled plasma mass spectrometry (ICP-MS) (Chen et al., 2001, Delgado-Andrade et al., 2003, Kohlmeyer et al., 2005). However, even these sensitive analytical techniques are not sufficiently sensitive for some purposes since the dissolution stage involves dilution. Alternative choices to total dissolution include the use of emulsions or direct introduction into the electrothermal atomizer of an atomic absorption spectrometer (ETAAS), although sensitivity still remains a drawback (Chen et al., 2003, Karadjova and Venelinov, 2002).

The difficulties involved in measuring low levels of arsenic are further increased if speciation is to be carried out since the total amount is distributed across a wide variety of arsenic compounds. This means that, in practice, ICP-MS is required, usually in conjunction with a chromatographic technique (Chu and Jiang, 2011, Schmeisser et al., 2005). The analytical superiority in terms of sensitivity of ICP-MS over other analytical methods is unquestionable, but it is expensive both as regards acquisition and maintenance.

Edible oils obtained from fish are recommended as nutritional supplements because some components have benefits in human health, although they represent a possible source of arsenic intake since the element accumulates in the marine environment (Francesconi, 2010). There are abundant studies reporting levels of arsenic in these food/supplements, due to the complex chemistry involved (Devallay and Feldmann, 2003, Rumpler et al., 2008, Taleshi et al., 2008, Taleshi et al., 2010). The relevance of the subject, therefore, is beyond doubt and the approach here presented tackles it from a particular point of view.

This contribution describes our studies using dispersive liquid–liquid microextraction (DLLME) for the isolation of water-soluble arsenic species present in edible oils of animal or plant origin. DLLME, which was introduced several years ago (Rezaee et al., 2006), represents an environmentally friendly methodology since it uses minimal amounts of organic reagents. Usually DLLME involves the extraction of analytes from an aqueous phase (the sample) into a small volume of organic phase in the presence of a third solvent that aids rapid and efficient dispersion (Escudero et al., 2013, Pena-Pereira et al., 2010, Rivas et al., 2009). In this work the technique was used in the opposite way; analytes were extracted from an organic phase (oil samples) into the aqueous phase. In this way, pre-concentration was achieved and the reliability of the analytical data improved. In addition ETAAS, an analytical technique that is available worldwide, was used for the final analysis. The pre-concentration effect inherent in DLLME increased the sensitivity of the final measurement. Appropriate selection of the experimental conditions allows discrimination between the most toxic inorganic arsenic species and the less toxic compounds (mainly, but not exclusively, arsenobetaine) without the need for a chromatographic stage (López-García, Briceño, & Hernández-Córdoba, 2011). Complete speciation was not possible but a reliable and relatively easy-to-perform assessment of the toxicity of the water-soluble arsenic compounds present in edible oils was achieved.