Evaluation of acid matrix effects in the determination of major elements in biomass by atomic absorption spectrometry from an environmentally friendly point of view

• Autor(a) principal: Teixeira, P. Alexandra
• Data de Publicação: 2015
• Outros Autores: Calisto, Sandra C., Lopes, Helena, Trancoso, Maria Ascensão
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.9/2591
• topic: Biomass; Solid biofuels; Atomic absorption spectrometry; Evaluation; Measurement uncertainty; Quality control

The estimation of major element content in solid biofuels is required for prediction and prevention of eventual ash-related problems during combustion. These analyses have to be achieved with minimum impact on the environment. The quantitation of Al, Ca, Mg, Na, K, Fe, Si and Mn in biofuels was carried out according to EN 15290 using acid decomposition of solid samples followed by atomic absorption spectrometry (AAS). A microwave-assisted acid digestion with a HNO3/H2O2/HF mixture was used, followed by HF complexation using H3BO3. Due to the presence of tetrafluoroboric acid complex in the digestion solution, matrix effects were noticed during elemental quantification by AAS. Standard addition calibration methods did not compensate for this matrix effect. Matrix effects that constrain an analytical response may be overcome by applying the procedure used for samples to the calibration standards using the same reagents. However, this entails using large amounts of toxic reagents. In this work, the fluoric–boric acid matrix matching was assessed statistically using one-way ANOVA tests. For the seven groups of nitric acid and reagent blank (HNO3/H2O2/HF/H3BO3) mixtures used, ranging from 0 to 1 volume ratios, no significant differences were observed for Si, Al, Fe and Mn. The calculated F values were lower than the critical value, F 6,14 = 2.85 (p = 0.05). However, for Ca, Mg, Na and K, significant differences were observed. Tenfold dilution was used for samples where the mass fraction exceeded the analytical dynamic range of the AAS instrument. The calibration solutions were prepared using the reagent blanks in the same proportion thus decreasing the amount of acids used. The procedure was validated using SRM 1573a—tomato leaves—purchased from the National Institute of Standards and Technology. Target recoveries of (1 ± 0.1) were achieved.