Micro-solid phase extraction methods based on molecularly imprinted polymer for determination of toluene metabolites in urine samples

Abstract

Toluene is one of the most widely used solvents in various industries. Workers exposure to toluene can be monitored using hippuric acid and o-cresol as urine biomarkers for biological exposure index. However, extraction of biological fluids often results in sample emulsification in liquid-liquid extraction (LLE) while also reducing the efficiency of sorbent materials in solid phase extraction (SPE). Therefore, it is imperative to solve the emulsification problem and enhance the reusability and selectivity of SPE sorbent. In this study, a new SPE method was developed using molecularly imprinted polymers (MIPs) based on hippuric acid (MISPE-HA) and o-cresol (MISPE-Cre). The MIPs were prepared by sol-gel polymerization of hippuric acid (0.3 mmol) or o-cresol (0.5 mmol), 3-propyltrimethoxymethacrylate (800 µL), ethanol (12 mL), tetraethoxysilane (TEOS) (12 mL) and catalyst (400 µL of hydrochloric acid). BET analysis for MISPE-HA and MISPE-Cre demonstrated surface areas of 388.64 m2/g and 574.90 m2/g, with pore volumes of 0.4638 cm3/g and 0.5314 cm3/g, respectively. Furthermore, both MISPE samples displayed pore diameters of 4.70 nm and 2.20 nm, respectively, which proved the presence of mesoporosity in their structures. MISPE-HA showed a higher adsorption capacity (9.67 mg/g) compared to MISPE-Cre (0.99 mg/g). The results indicated that the kinetics of adsorption obeyed well the pseudo-second order model and Scatchard plot for adsorption isotherm. In order to improve analyte mobility and reusability, new dispersive solid phase extraction (DISPE) methods were developed based on surface imprinted hippuric acid grafted onto silica gel (MIP@SiO2-HA) and surface imprinted o-cresol grafted onto silica gel (MIP@SiO2-Cre). The MIP@SiO2 materials were prepared by hippuric acid (0.3 mmol) or o-cresol (0.5 mmol), 3-propyltrimethoxymethacrylate (800 µL), tetraethoxysilane (10 mL), activated silica gel (0.1 g) and acetic acid (0.01 mol/L). Nitrogen adsorption showed that both MIP@SiO2-HA and MIP@SiO2-Cre were mesoporous materials with pore diameters of 4.70 nm and 4.76 nm, respectively. BET study showed that, MIP@SiO2-HA had a surface area of 122.10 m2/g and pore volume of 0.4638 cm3/g, while MIP@SiO2-Cre exhibited surface area of 106.90 m2/g and pore volume of 0.1991 cm3/g. MIP@SiO2-HA indicated a good adsorption capacity (9.68 mg/g) compared to that of MIP@SiO2-Cre, which had a low adsorption capacity of 0.99 mg/g. MIP@SiO2 was also found to be fitted well with pseudo-second order model for kinetic adsorption and Scatchard plot for adsorption isotherm. The MIP@SiO2 and MISPE method validation indicated good limit of detection (LOD) and good limit of quantification (LOQ) with relative standard deviation (RSD) less than 5% and high extraction recovery more than 90%, which complied with ISO 17025:2005. The results demonstrated that these adsorbents have potential applications in the determination of toluene metabolites in urine samples due to the improved selectivity, better elution ability for target molecule and high reusability.