DESENVOLVIMENTO DE MÉTODOS ANALÍTICOS CROMATOGRÁFICOS, ESPECTROFLUORIMÉTRICOS E ELETROFORÉTICOS PARA A DETERMINAÇÃO DE AZAARENOS BÁSICOS EM QUEROSENE DE AVIAÇÃO / DEVELOPMENT OF ANALYTICAL METHODS BASED ON CHROMATOGRAPHY, FLUORESCENCE AND ELECTROPHORESIS FOR THE DETERMINATION OF BASIC AZAARENES IN JET FUEL SAMPLES

ELAINE ROCHA DA LUZ

In this work, analytical methods were developed for the selective determination of basic azaarenes. Solid phase extraction (SPE) in combination with high performance liquid chromatography with fluorescence detection (HPLCFD) has been used for the sensitive method determination of six basic azaarenes (7,8-benzoquinoline - 78BQ, 7,9-dimethylbenz[c]acridine - 79DMBA, 9-amino- 1,2,3,4-tetrahydroacridine - 9ATHA, 9-methylacridine - 9MA, acridine - A, and dibenz[a,j]acridine - DBA) in jet fuel samples. The extraction process was performed in a single step using a propyl sulfonic acid cartridge (PRS). The HPLC system consisted of C18 column with a selected detection program of optimum excitation ((lambda) exc) and emission ((lambda)em) wavelengths. A gradient elution with ACN and phosphate buffer (pH 6.5) allowed an efficient and fast separation of the azaarenes within 15 min. The LOD and LOQ values, based on signal-to-noise ratio 3:1 and 10:1, respectively, were between 0.0013 and 0.021 and from 0.0044 to 0.072 ng per injection. The calibration curves showed linear behavior in range from LOQ to 250 mg L(-1) (r(2) >0.99). For the spiked concentration of 6.0 (u)g L(-1), the recoveries were from 92 to 107% for jet fuel samples, except for 9ATHA, which presented a lower recovery value (68%). Finally, the proposed method was applied to the quantification of those six basic azaarenes in a commercial kerosene sample and in three jet fuel samples. The presence of 78BQ and DBA was confirmed in the jet fuel samples. Capillary electrophoresis methods were evaluated for the quantification of the basic azaarenes in jet fuel samples. In MECC, the best conditions for the separation of the analytes, in the preliminary phase of study, have been: 20 mmol L(-1) borate buffer with 25% of methanol and 40 mmol L(-1) SDS (pH 9.5) as the background solution; instrumental conditions: 30 kV, 30 °C and hydrodynamic injection (50 mbar) for 10 s. In CZE, the best conditions for the separation of the analytes have been: 50 mmol L(-1) phosphate buffer with 25 % of acetonitrile (pH 2.65) as the background solution; instrumental conditions: 25 kV, 25 °C and hydrodynamic injection (50 mbar) for 150 s; sample solvent: 1 mmol L(-1) phosphate buffer with 20 % of acetonitrile. A fused-silica capillary of 50 mm ID x 64.5 cm (56 cm effective length) x 150 mm path length has been used. The LOD and LOQ values were between 0.68 and 3.2 and from 1.1 to 7.7 (u)g L(-1), respectively. The previous results showed positive perspectives to apply CZE with on-line pre-concentration for the determination of the basic azaarenes in jet fuel. However, a more detailed study to improve the resolution and sensitivity and to reduce the analysis time is still necessary. Moreover, the procedure of the analytes extraction from the kerosene sample must be adjusted for providing good results of recovery and to be compatible with the developed method of analysis. Finally, an alternative method was developed, using fluorescence spectra acquisition by synchronous scan and partial least square regression (PLS), for determination of the same basic azaarenes. 40 synthetic mixtures were used in the range of 10 to 100 (u)g L(-1). The synchronous scan spectra were obtained using (lambda)(delta) = 30,120 and 150 nm. In the PLS models, 20 samples were used for calibration and 20 for test. The clean-up procedure was adapted from the HPLC-FD method employing Methanol:HCl 0,01 mol L(-1) as the final solvent of the extraction. The extracts were analyzed by HPLC and the results were used as reference values. The LOD and LOQ values, calculated from the net analytical signal (NAS), were between 0.045 and 2.0 and from 0.15 to 6.6 (u)g L(-1), respectively. The relative standard errors of prediction (%RSEP) were from 3.0 to 10% for the sample test set. For the spiked jet fuel sample the %RSEP were between 4.9 and 11% for four of the six studied analytes, since it

AUTOR(ES)

DATA DE PUBLICAÇÃO

RESUMO

ASSUNTO(S)

ACESSO AO ARTIGO

Documentos Relacionados