Optimisation of HPLC-based methods for the separation and detection of herbicide glyphosate and its major metabolite in water

Abstract

Water storage dams play an important part in the collection and purification of water destined for human consumption. However, the nutrient rich silt in these dams promotes rapid growth of aquatic plants which tend to block out light and air. Glyphosate is universally used as the effective non-selective herbicide for the control of aquatic plants in rivers and dams. Invariably there is residual glyphosate present in water after spraying of dams and rivers with glyphosate herbicide. The amount of residual glyphosate is difficult to determine on account of high solubility of glyphosate in water. Thus a method of sample preparation and a sensitive HPLC method for the detection of trace amounts of glyphosate and its major metabolite aminomethylphosphonic acid (AMPA) in water is required. A crucial step in sample preparation is pre-column derivitization of glyphosate with 9-fluorenylmethyl chloroformate (FMOC-Cl). For sample pretreatment, water samples were derivatized with FMOC-Cl at pH 9, extracted with ethyl acetate and sample clean-up was carried out by passing a sample through the SPE cartridge. For SPE, recovery studies were done to choose a suitable cartridge for glyphosate and AMPA analysis. The following cartridges were compared, namely, C18, Oasis HLB and Oasis MAX SPE cartridges. Best recoveries (101% for glyphosate and 90% for AMPA) were obtained using 500 mg of C18 solid-phase extraction cartridge. The eluent from SPE cartridge was injected into HPLC column. Three types of separation columns (namely; C18 column, silica based amino column and polymeric amino column) were compared for the separation of glyphosate and AMPA. The best separation of glyphosate and AMPA in water samples was achieved using a polymeric amino column and a mobile phase at pH 10 which contained a mixture of acetonitrile and 0.05 M phosphate buffer (pH 10) 55:45, (v/v) respectively. The method was validated by spiking tap water , deionized water and river water at a level of 100 μg/l. Recoveries were in the range of 77% -111% for both analytes. The method was also used in determining the levels of glyphosate and AMPA in environmental samples. This method gave detection limits of 3.2 μg/l and 0.23 μg/l for glyphosate and AMPA respectively. The limits of quantification obtained for this method were 10.5 μg/l and 3.2 μg/l for glyphosate and AMPA respectively.