Alternative methodologies for the determination of aldehydes by capillary electrophoresis

• Autor(a) principal: Pereira, Elisabete Alves
• Data de Publicação: 1999
• Outros Autores: Tavares, Marina Franco Maggi, Cardoso, Arnaldo Alves [UNESP]
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1093/jaoac/82.6.1562; http://hdl.handle.net/11449/231655

This paper describes 2 alternative methodologies for the determination of selected aldehydes (formaldehyde, acetaldehyde, propionaldehyde, acrolein, and benzaldehyde) by capillary electrophoresis (CE). The first approach is based on the formation of aldehyde-bisulfite adducts and employs free solution CE with reversed electroosmotic flow and indirect detection, using 10 mmol/L 3,5-dinitrobenzoic acid (pH 4.5) containing 0.2 mmol/L cetyltrimethylammonium bromide as the electrolyte. This novel methodology showed a fairly good sensitivity to concentration, with detection limits with respect to a single aldehyde on the order of 10-40 μg/L, a reasonable analysis time (separation was achieved in <8 min), and no need for sample manipulation. A second approach was proposed in which 2,4-dinitrophenylhydrazine derivatives of the aldehydes were detected in a micellar electrolyte medium (20 mmol/L borate buffer containing 50 mmol/L sodium dodecyl sulfate and 15 mmol/L β-cyclodextrin). This latter methodology included a laborious sample preconcentration step and showed much poorer sensitivity (0.5-2 mg/L detection limit, with respect to a single aldehyde), despite the use of sodium chloride to promote sample stacking. Both methodologies proved adequate to evaluate aldehyde levels in vehicular emissions. Samples from the tailpipe exhaust of a passenger car vehicle without a catalytic converter and operated with an ethanol-based fuel were collected and analyzed; the results showed high levels of formaldehyde and acetaldehyde (0.41-6.1 ppm, v/v). The concentrations estimated by the 2 methodologies, which were not in good agreement, suggest the possibility of striking differences in sample collection efficiency, which was not the concern of this work.