SciELO - Scientific Electronic Library Online

vol.44 issue3A chemically defined medium for production of actinomycin D by Streptomyces parvulusPurification and properties of asparaginase from the testa of immature seeds of pea (Pisum sativum L.) author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand




Related links


Brazilian Archives of Biology and Technology

Print version ISSN 1516-8913On-line version ISSN 1678-4324

Braz. arch. biol. technol. vol.44 no.3 Curitiba Sept. 2001 

Evaluation of Arsenic and Selenium in Brazilian Soluble Coffee by Inductively Coupled Plasma Atomic Emission Spectrometry with Hydride Generation


Éder José dos Santos1* and Elisabeth de Oliveira2
1Instituto de Tecnologia do Paraná (TECPAR), CP 357, CEP 81310-020, Curitiba, PR, Brazil; 2Universidade de São Paulo (USP), Instituto de Química, CP 26077,CEP 05508-900, SP, Brazil




A method for the evaluation of arsenic and selenium in soluble coffee by inductively coupled plasma atomic emission spectrometry with continuous hydride generation to attend the Brazilian food legislation is described. Samples were digested with nitric acid and hydrogen peroxide in a focused microwave system. Slow heating eliminated nitric acid and selenium (VI) was reduced to selenium (IV) by addition of 6 mol/L hydrochloric acid and heating at 90° C under a reflux system. The influence of sample acidity on sensitivity was investigated. Hydrochloric acid 6 mol/L was the most suitable reaction medium. Practical detection limits of 2.0mg/L for As and 1.0m g/L for Se were achieved and attended the Brazilian food legislation. The results of recoveries on spiked samples demonstrate the reliability and accuracy of the procedure.

Key words: ICP-AES, As, Se, soluble coffee




The production of Brazilian soluble coffee is almost entirely destined for exportation that was estimated to be 62 thousand tons in 1995 (Silva, 1996). The determination of macronutrients, micronutrients and toxic elements in soluble coffee is of great general interest, particularly because of the importance of coffee in nutrition to millions of people around the world (Krivan et al., 1982). Previous work on the composition of coffee and its products is mainly concerned with the mineral contents and studies on toxic elements are scarcely reported (Clarke et al., 1974; Gillies et al., 1983; Horwitz et al., 1974; Kuennen et al., 1982; Kapur et al., 1974; Koch et al., 1989; Lara et al., 1975/76). No information exists on the content of arsenic and selenium in this product although a number of publications (Goulden et al., 1981; Nakahara et al., 1985; Nakahara, 1981; Nygaard et al., 1982; Oliveira et al., 1983; Pahlavanpour et al., 1980; Ybanez et al., 1992) demonstrated that hydride generation, followed by introduction of gaseous hydrides into an ICP was a suitable method for the determination of several trace elements, including arsenic and selenium, for which detection limits were inadequate for foods and related matrices when conventional pneumatic nebulization was used. The objective of this work was to develop a method for the evaluation of arsenic and selenium in soluble coffee by inductively coupled plasma atomic emission spectrometry with hydride generation in order to attend the Brazilian food legislation. The influence of sample acidity on sensitivity of the method was also studied.



Reagents: Distilled and deionized water (deionizing Permution); arsenic (V) and selenium (IV) stock standard solutions (1000 mg/mL) (Titrisol Merck); nitric acid, (Merck), 65%; hydrogen peroxide, (Merck), 30%; hydrochloric acid, (Merck), 37%, use of a 6 mol/L solution; Sodium borohydride, (Merck), 99.9%, use of a 1 % w/v solution in 0.1 mol/L NaOH.

Apparatus : Focused microwave system model 7400 Spex with 250 mL glass tubes; hydride generation was carried out in a continuous mode by using a two-channels peristaltic pump (Gilson Instrument Co., Minipuls) to introduce sample and borohydride reagent to a phase separator from Oliveira et al., 1983 (Figure 1). The operating conditions for hydride generation system are summarized in Table 1. A Baird PSX sequencial ICP spectrometer with a focal lengh Czerny-Turner monochromator, a holografic grating with a linear dispersion of 0.74 nm/mm (first order), a spectral resolution of 0.014 nm and a RF generator of 2.3 kW at 40.68 MHz were employed. The operating conditions are also summarized in Table 1.





Standard solutions preparation and calibration curves : Calibration curves for the analytes arsenic (V) and selenium (IV) were made by using four different standard solutions: 0.0 mg/L, 5.0 mg/L, 10.0 mg/L and 20.0 m g/L in 6 mol/L HCl.

Sample preparation: Twenty-one samples of soluble coffee were collected from the Brazilian market and were used in the present work. Part one of sample preparation was developed in our laboratory (Santos et al.,1997), and part two was adapted from Brimmer et al.,1987.

Part one: Five gram of sample was introduced into a 250 mL glass tube (from the focused microwave system model 7400 SPEX) and 30 mL of HNO3 was added. After 30 minutes of initial reaction at ambient temperature, the glass tube was placed on the focused microwave system, heated at 105 W power for 5 minutes, and allowed to cool for 5 minutes. This heating and cooling procedures were repeated. Then 2.5 mL of H2O2 were added and the heating and cooling procedures were repeated once more.

Part two: The solution prepared above were transferred to a 100 mL boiling flask and evaporated at 90 ± 5 ° C to almost dryness. Twenty-five mL of 6 mol/L HCl were added to this and the solution was heated at 90 ± 5 ° C under reflux system for 30 minutes. Three replicate analyses were carried out.

Spiked samples: Volumes of the standard solutions containing 0.25 mg of arsenic (V) and 0.25 mg of selenium (IV) were added to 1g of three samples (equivalent to 10.0 mg/L in the final solutions on which those analytes were determined). Two replicate analyses were carried out and these spiked samples were treated as described on sample preparation item (part one and two).



Optimization of operating parameters

The phase separator constructed according to the specifications from Oliveira et al.,1983, (Figure 1), was used with few problems in the operating conditions of the PSX-Baird spectrometer. Flow rates of 0.6 L/min., used at conventional solution nebulization by pneumatic nebulizer, were also suitable for determination of arsenic and selenium by hydride generation. At flow rates below 0.6 L/min., the formation of hydrogen bubbles was high, hindering the perfect operation of the drain. On the other hand, at flow rates higher than 0.6 L/min., the back pressure created tended to force all the waste liquid out of the U-tube in the phase separator and usually extinguished the plasma. A concentration of NaBH4 more than 1 % w/v also extinguished the plasma and at less than this value the signal was very poor (Oliveira et al.,1983). The effect of the hydrochloric acid concentration on sensitivity was studied. The following solutions were prepared:

- 100 mL of 1 to 7 mol/L HCl (solutions of the blank);
- 100 mL of standard solutions of 50.0 m g/L of As (V) and Se (IV) in 1 to 7 mol/L HCl;
- 1 % w/v NaBH4 solution in 0.1 mol/L NaOH.

Figure 2 shows a plot of the different HCl concentrations versus the relative intensity of the analytes arsenic (V) and selenium (IV).



The base line represented by blank solutions (blk) showed invariable values at different

concentrations of hydrochloric acid. The response had the best sensibility for selenium and arsenic with 5 and 6 mol/L HCl.

Linearity, instrumental detection limits and practical detection limits

The parameters of analytical curve and detection limits were investigated under the conditions detailed in Table 1 previously presented and are shown in Table 2. Typical calibrations were linear up to 20.0 mg/L. The instrumental detection limit is defined as the concentration that produces a net line intensity equivalent to three times the standard deviation of the blank or background signal (Oliveira et al.,1983; Thompson, 1987). The blank was the ordinate intercept calculated from the linear regression analysis of the calibration data and showed in Table 2. However, under the conditions of this analysis, the background signal of the samples was different from that produced by the standard solutions and the blank, so the practical detection limits used in this work were greater than instrumental detection limits by a factor of 2 since these results were better and a more realistic estimate for quantitative analysis.



Determination of arsenic and selenium

The results obtained for the arsenic and selenium in twenty-one samples of Brazilian soluble coffee are presented in Table 3. The values were below the practical detection limits and were the following, in mg/g of soluble coffee: As not detectable up to 0.010 mg/g and Se not detectable up to 0.005 mg/g.



The results for three spiked samples treated as described on sample preparation are shown in the Table 4.




The application of the hydride generation technique with inductively coupled plasma atomic emission showed to be suitable to determine arsenic and selenium in soluble coffee with practical detection limits giving values below those specified in Brazilian food legislation (Compêndio da Legislação de Alimentos-Atos do Ministério da Saúde, 1988), that fixed the maximum arsenic content in 1.0 mg/g and selenium in 0.3 mg/g. The spiked samples treated as described on sample preparation showed acceptable results, indicating accuracy and precision of the present method of analysis.



The financial support from CNPq and FAPESP is greatefully acknowledged.




O presente trabalho descreve um método para determinação de arsênio e selênio em café solúvel por espectrometria de emissão atômica com plasma acoplado a geração de hidretos, com o objetivo de atender aos limites estabelecidos pela legislação Brasileira de alimentos. Digestões das amostras foram realizadas com ácido nítrico e peróxido de hidrogênio em sistema de microondas focalizadas. O excesso de ácido nítrico foi eliminado através de aquecimento lento e o selênio (VI) foi reduzido a selênio (IV) através de tratamento com ácido clorídrico 6 mol/L e aquecimento a 90° C em sistema de refluxo. A influência da acidez na sensibilidade foi investigada, sendo o meio mais estável a concentração de ácido clorídrico de 6 mol/L. Os limites de determinação foram de 2.0 mg/L para o As e 1.0 mg/L para o Se. Os resultados das amostras enriquecidas demonstram a precisão e exatidão do método proposto.




Brimmer S. P.; Fawcett W. R. and Kulhavy K. A. (1987), Quantitative Reduction of Selenate Ion to Selenite in Aqueous Samples. Anal. Chem., 59, 1470-1471        [ Links ]

Compêndio da Legislação de Alimentos - Atos do Ministério da Saúde (1988), In: Aditivos Incidentais tabela II, v.1, cap.3, resolução n° 4.        [ Links ]

Clarke R. J. and Walker L. J. (1974), Potassium and and Other Mineral Contents of Green, Roasted and Instant Coffees. J. Sci. Fd Agric., 25, 1389-1404        [ Links ]

Gillies M. E. and Birkbeck J. A. (1983), Tea and Coffee as Sources of Some Minerals in the New Zealand Diet. Am. J. Clin. Nutr., 38, 936-942        [ Links ]

Goulden P. D.; Anthony D. H. J. and Austen K. D. (1981), Determination of Arsenic and Selenium in Water, Fish, and Sediments by Inductively Coupled Argon Plasma Emission Spectrometry. Anal. Chem., 53, 2027-2029        [ Links ]

Horwitz C. and Van Der Linden S. E. (1974), Cadmium and Cobalt in Tea and Coffee and their Relationship to Cardiovascular Disease. S. Afr. Med. J., 48, 230-233        [ Links ]

Krivan V.; Barth P. and Morales A. (1993), Multielement Analysis of Green Coffee and its Possible use for the Determination of Origin. Mikrochim. Acta, 110, 217-236        [ Links ]

Kuennen R. W.; Hahn M. H.; Fricke F. L. and Wolnik K. A. (1982), Hydride Generation and Condensation Flame Atomic Absorption Spectroscopic Determination of Antimony in Raw Coffee Beans and Processed Coffee. J.Assoc. Off. Anal. Chem., 65 (5), 1146-1149        [ Links ]

Kapur J. K. and West T. S. (1974), Determination of Lead in Instant Coffee and Tea Powders by Carbon Filament Atomic Absorption Spectrometry. Anal. Chim. Acta, 73, 180-184        [ Links ]

Koch K. R.; Bruno Pougnet M. A. and De Villiers S. (1989), Determination of Aluminium Leves in Tea and Coffee by Inductively Coupled Plasma Optical Emission Spectrometry and Graphite Furnace Atomic Absorption Spectrometry. Analyst, 114, 911-913        [ Links ]

Lara W. H.; Toledo M. and Takahashi M. Y. (1975/76), Teores de Cobre em Café Torrado e Moído e emCafé Bebida. Rev. Inst. Adolfo Lutz, 35/36, 17-22        [ Links ]

Nakahara T. (1981), Application of Hydride Generation to the Determination of Trace Concentrations of Arsenic by Inductively - Coupled Plasma Atomic Emission Spectrometry. Anal. Chim. Acta, 131, 73-82        [ Links ]

Nygaard D. D. and Lowry J. H. (1982), Sample Digestion Procedures for Simultaneous Determination of Arsenic, Antimony, and Selenium by Inductively Coupled Argon Plasma Emission Spectrometry with Hydride Generation. Anal. Chem., 54, 803- 807        [ Links ]

Nakahara T. and Kikui N. (1985), Determination of trace Concentrations of Selenium by Continuous Hydride Generation– Inductively Coupled Plasma Atomic Emission Spectrometry. Spectrochim Acta, 40B (1/2), 21-28        [ Links ]

Oliveira E.; McLaren J. W. and Berman S. S. (1983), Simultaneous Determination of Arsenic, Antimony, and Selenium in Marine Samples by Inductively Coupled Plasma Atomic Emission Spectrometry. Anal. Chem., 55 (13), 2047-2050        [ Links ]

Pahlavanpour B.; Thompson M. and Thorne L. (1980), Simultaneous Determination of Trace Concentrations of Arsenic, Antimony and Bismuth in Soils and Sediments by Volatile Hydride Generation and Inductively Coupled Plasma Emission Spectrometry. Analyst, 105, 756-761        [ Links ]

Santos E. J. and Oliveira E. (1997), Avaliação e Definição de Procedimento de Mineralização de Amostras de Café Solúvel para Determinação de Nutrientes e Contaminantes Inorgânicos por ICP-AES. Arq. Biol. Tecnol., 40 (3), 632-641        [ Links ]

Silva P. (1996), O Mercado de Café no Brasil. Alimentos & Tecnologia, 64, 26-30        [ Links ]

Thompson M. (1987), Inductively Coupled Plasmas in in Analytical Atomic Spectrometry. In: Montaser Montaser A., Golightly D. W. ed. Analytical Performance of Inductively Coupled Plasma - Atomic Emission Spectrometry. New York. VHC Publisher , 163-198        [ Links ]

Ybanez N.; Cervera M. L. and Montoro R. (1992), Determination of Arsenic in Dry Ashed Seafood Products by Hydride Generation Atomic Absorption Spectrometry and a Critical Comparative Study with Platform Furnace Zeeman - effect Atomic Absorption Spectrometry and Inductively Coupled Plasma Atomic Emission Spectrometry. Anal. Chim. Acta, 258, 61-71        [ Links ]



Received: April 04, 2000;
Revised: June 27, 2000;
Accepted: November 23, 2000.



* Author for correspondence

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License