ABSTRACT
This study introduces a novel method for the quantification of malachite green (MG), a pervasive cationic dye, in surface water by synergizing multiphase electroextraction (MPEE) with digital image analysis (DIA) and partial least square discriminant analysis. Aimed at addressing the limitations of conventional DIA methods in terms of quantitation limits and selectivity, this study achieves a significant breakthrough in the preconcentration of MG using magnesium silicate as a novel sorbent. Demonstrating exceptional processing efficiency, the method allows for the analysis of 10 samples within 20 min, exhibiting remarkable sensitivity and specificity (over 0.95 and 0.90, respectively) across 156 samples in both training and test sets. Notably, the method detects MG at low concentrations (0.2 µg L-1) in complex matrices, highlighting its potential for broader application in environmental monitoring. This approach not only underscores the method's cost-effectiveness and simplicity but also its precision, making it a valuable tool for the preliminary testing of MG in surface waters. This study underscores the synergy among MPEE, DIA, and chemometric tools, presenting a cost-efficient and reliable alternative for the sensitive detection of water contaminants.
ABSTRACT
In this work, a new large-volume multiphase, multi-sample electroextraction device was developed and applied to selectively extract malachite green (MG) from water samples. This device was easily constructed with ordinary materials and capable of extracting ten samples simultaneously, obtaining MG preconcentrated on a solid support, to fit into a pipette tip. A multi-well plate was applied to extract MG from aquaculture water samples, and the extracts containing the desorbed MG were analysed by LC-DAD and LC-MS/MS. The signals from both detectors were used in two independent validation procedures. Linearity, matrix effect, selectivity, precision, trueness, and limits of detection and quantification were all evaluated. For both detectors, linearity was demonstrated in the range of 0.5-5 µg L-1 (R2 > 0.98). Matrix effect was insignificant for LC-DAD only, and the average preconcentration factor was about 60 times. Recoveries ranged from 94 to 113% for LC-DAD and 95-115% for LC-MS/MS analysis. ANOVA was applied to estimate the standard deviation under repeatability (6.96-8.61% for LC-DAD and 5.98-7.41% for LC-MS/MS) and within-reproducibility (6.96-8.61% for LC-DAD and 6.56-7.41% for LC-MS/MS) conditions. The limits of detection and quantification for LC-MS/MS analysis were 4.29 and 28.74 ng L-1, respectively, while, for LC-DAD, these limits were 14.29 and 95.81 ng L-1, respectively. The results demonstrated that the developed method was suitable for determining MG in water samples, and the large-volume multiphase, multi-sample electroextraction device proved to be a powerful sample preparation technique to obtain high clean-up and large preconcentration levels, which are of paramount importance for environmental applications.