Talanta 94 (2012) 123–132 Contents lists available at SciVerse ScienceDirect Talanta j ourna l ho me page: www.elsevier.com/locate/talanta Characterization of disposable optical sensors for heavy metal determination Jadranka Vukovi ´ c a,∗ , María Ariza Avidad b , Luis Fermín Capitán-Vallvey b a Department of Analytics and Control of Medicines, Faculty of Pharmacy and Biochemistry, University of Zagreb, A. Kovaˇ ci´ ca 1, HR-10000 Zagreb, Croatia b ECsens, Department of Analytical Chemistry, Campus Fuentenueva, Faculty of Sciences, University of Granada, E-18071 Granada, Spain a r t i c l e i n f o Article history: Received 16 November 2011 Received in revised form 27 February 2012 Accepted 5 March 2012 Available online 8 March 2012 Keywords: Optical membranes Heavy metal determination Disposable sensors Prevalidation strategy Multicomponent analysis a b s t r a c t This paper presents the development, characterization and quality control of analytical methods based on the use of disposable optical sensors for determination of heavy metals. Chromogenic reagents such as 1-(2-pyridylazo)-2-naphthol, (2-pyridylazo)resorcinol, Zincon, Ferrozine, and Chromazurol S were used to develop optical sensors of heavy metal ions found as contaminants in pharmaceutical substances and products, such as Zn(II), Cu(II), Ni(II), Fe(II), and Fe(III). The chromogenic reagents were immobilized in polymeric membranes by spin-coating from cocktails containing all reagents needed. The methods were prevalidated using a comprehensive quality control strategy based on a system of mathemati- cal/statistical testing and diagnosis of each prevalidation step. This system involved characterization of analytical groups; checking of two limiting groups; testing of data homogeneity; recognition of outliers; and determination of analytical functions, limiting values, precision and accuracy. The prevalidation strat- egy demonstrated the reliability of the proposed method and pointed out some limitations. Combining the optical sensors with multicomponent linear regression allowed simultaneous determination of mul- tiple metals in synthetic mixtures with different compositions. Good agreement between experimental and theoretical amounts of heavy metals in the mixtures was obtained for the majority of sensors and metals. Even better agreement was obtained between the experimental and theoretical total amounts of metals in the mixtures. The proposed analytical methods were successfully applied to the determination of zinc in pharmaceutical preparations of insulin and the determination of metal mixtures in a commer- cial nasal spray of isotonic seawater. The reliable and sensitive individual optical sensors developed in this study may be useful for designing a multimembrane optical tongue that with appropriate further optimization can be used for screening heavy metals in various matrices. © 2012 Elsevier B.V. All rights reserved. 1. Introduction Heavy metals are ubiquitous and monitoring them is important in various fields because of their effects on ecosystems and living organisms [1–5]. For example, an important part of drug quality control is detecting and determining heavy metals. Bulk drug sub- stances and their intermediates can be contaminated by metals in many ways, such as from raw materials, reagents, and solvents; from electrodes, reaction vessels, storage containers, plumbing and other equipments used in synthesis; or from exposure to air-borne particles. Metal catalysts and metal reagents, are substances used in the synthesis of the drug substance or an excipient used in a medicinal product, and, therefore, can also introduce metals into drug preparations [6,7]. Although evaluated for their potential risk to human health and placed into one of three classes (metals of ∗ Corresponding author. Tel.: +385 1 4920089; fax: +385 1 4920089. E-mail addresses: jadranka@pharma.hr (J. Vukovi ´ c), mariavidad@hotmail.com (M.A. Avidad), lcapitan@ugr.es (L.F. Capitán-Vallvey). significant, low, and minimal safety concern), some of them, such as Fe, Zn, and Cu are also important in human nutrition. For over a century [7], drug quality control departments have relied on the heavy metal limit test recommended by most phar- macopoeias [8–11]. Besides the heavy metal limit test, European Pharmacopoeia [8] prescribes determination of specific metals in drug substances, such as copper in ascorbic acid, zinc in insulin, and nickel in polyols. The heavy metal limit test is based on sul- fide precipitation in a weakly acidic medium and comparison with a lead solution. The test can be easily transferred from one labo- ratory to another and does not require expensive instrumentation or highly trained laboratory personnel. However, it suffers from several disadvantages. It requires subjective visual interpretation, large amounts of sample, and usually a heating or ashing step that causes losses of volatile elements. In addition, it does not provide any qualitative or element-specific information. Several attempts have been made to overcome these limitations [12,13], but no major improvements have been achieved. Thus, this stan- dard method remains suitable for only a few elements, and it cannot specify the content of any particular metal ion but only the overall content of ions [12]. 0039-9140/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2012.03.006