On-disc electromembrane extraction-dispersive liquid-liquid microextraction: A fast and effective method for extraction and determination of ionic target analytes from complex biofluids by GC/MS.

In this study, an electromembrane extraction-dispersive liquid-liquid microextraction (EME-DLLME) was performed using a lab-on-a-disc device. It was used for sample microextraction, preconcentration, and quantitative determination of tricyclic antidepressants as model analytes in biofluids. The disc consisted of six extraction units for six parallel extractions. First, 100 μL of a biofluid was used to extract the analytes by the drop-to-drop EME to clean-up the sample. The extraction then was followed by applying the DLLME method to preconcentrate the analytes and make them ready for being analyzed by gas chromatography (GC). Implementing the EME-DLLME method on a chip device brought some significant advantages over the conventional methods, including saving space, cost, and materials as well as low sample and energy consumption. In the designed device, centrifugal force was used to move the fluids in the disc. Both sample preparation methods were performed on the same disc without manual transference of the donor phases for doing the two methods. Scalable centrifugal force made it possible to adjust the injection speed of the organic solvent into the aqueous solution in the DLLME step by changing the spin speed. Spin speed of 100 rpm was used in dispersion step and spin speed of 3500 rpm was used to sediment organic phase in DLLME step. The proposed device provides effective and reproducible extraction using a low volume of the sample solution. After optimization of the effective parameters, an EME-DLLME followed by GC-MS was performed for determination of amitriptyline and imipramine in saliva, urine, and blood plasma samples. The method provides extraction recoveries and preconcentration factors in the range of 43%-70.8% and 21.5-35.5 respectively. The detection limits less than 0.5 μg L with the relative standard deviations of the analysis which were found in the range of 1.9%-3.5% (n = 5). The method is suitable for drug monitoring and analyzing biofluids containing low levels of the model analytes.