Determination of Al, Cd, and Pb in Brazilian sugar cane spirit, cachaca by ETAAS using matrix-matched calibration

In this work, a method is described for the determination of aluminium, cadmium, and lead in the Brazilian sugar cane spirit, cachaca, by electrothermal atomic absorption spectrometry (ETAAS) using matrix-matched calibration, i.e., calibration curves obtained at 40% (v/v) in ethanol-water media (the average alcohol content in cachaea). The methodology was applied to the determination of these three metals in 53 cachaca samples obtained from 5 3 different Jequitinhonha High Valley producers, in the Minas Gerais State, Brazil. All metals were determined with or without permanent chemical modifiers such as Zr+Rh and Ir+Rh. The selection of pyrolysis and atomization temperatures was determined in the presence of the matrix. Best results were obtained without the use of chemical modifiers for Al and Cd. For Pb, the results without a modifier were similar to those obtained using Zr + Rh (500 and 250 mug, respectively) or with Ir + Rh (500 and 250 mug) as the permanent modifier. For aluminum, without the use of a chemical modifier, the best pyrolysis and atomization temperatures were 1000 and 2500degreesC, respectively, with a characteristic mass (in.) of 9.0 pg (recommended 10 pg). It can therefore be assumed that cachaca itself in some way acts as modifier because it allows the use of higher temperatures for the three studied analytes. Two samples of cachaca, spiked with different amounts of Al, showed recoveries between 87.5% and 103.0%, with a relative standard deviation (RSD) (n=3) of less than 5%. The detection limit (LOD) (k=3, n=10) was 2.0 mug L-1. The At results for the 53 samples varied from non-detectable to 22.4 mug L-1. For cadmium, without the use of a modifier, the best pyrolysis and atomization temperatures were 1000 and 1400degreesC, respectively, with m(o) 0.5 pg (equal to recommended). Two samples spiked with cadmium presented recoveries from 103.1-114.6%, with an RSD of less than 4%; the LOD was 0.07 mug L-1. In the 53 samples, the Cd values varied from non-detectable to 0.7 mug L-1. For lead, without use of a modifier, the best pyrolysis and atomization temperatures were 900 and 1900degreesC, respectively, with in. 11.0 pg (recommended 10 pg). Two samples spiked with different amounts of the metal showed recoveries from 95.1-118.8%, with an RSD (n=3) less than 6%; the LOD was 0.6 mug L-1. For 53 samples, the Pb results varied from non-detectable to 526.0 mug L-1. For the three studied analytes, Al, Cd, and Pb, the calibration curves for matrix matching had an r (lineal regression coefficient) higher than 0.99.