Degree Name

Doctor of Philosophy (PhD)


School of Chemistry - Faculty of Science


Phytoestrogens are a subclass of endocrine disrupting chemicals (EDCs). Like other EDCs, research has demonstrated that phytoestrogens also have adverse effects on wildlife in the aquatic environment. In the past few years, many analytical methods have been developed. However, most of them were for natural and synthetic estrogens or pharmaceuticals in the environment, and very few involved phytoestrogens. So far as we know, there are no literature reports about the occurrence of enterolignans in the environment and wastewater samples. So, the occurrence and distribution of the three important type phytoestrogens (isoflavone, enterolignans and coumestrol) in the aquatic environmental samples and wastewater need to be fully investigated in order to assess their potential impact. In this work, a simultaneous LC-ESI-MSn analytical method for the analysis of the important phytoestrogen isoflavones, enterolinans, and coumestrol at nanogram / L levels in the aquatic environment and wastewater samples has been developed and optimised. This method was successfully used in the analysis of different aqueous samples with high selectivity and sensitivity. This LC-ESI-MSn analytical method has been validated by its sensitivity, accuracy, and precision. The method detection limit (MDL) was found to be from 0.1 to 3 ng/L in surface water and STP effluent; and ranged from 0.5 to 5 ng/L in STP influent. Accuracy was determined by the recovery experiments and evaluated by relative standard deviation (RSD %). The average recoveries of the all analytes in the different matrix samples were ranged from 85% to 95%. The RSD (%) values for all analytes in the surface water and STP effluent were found <6.5%, and in STP influent sample were <8.1%. Signal suppression is a common problem in the LC-ESI-MS quantitative analysis. In this work matrix effects were tackled by two ways: one way was to reduce the matrix interference by optimisation of sample preparation and improvement of chromatographic separation, another way was to select an appropriate calibration approach to further compensate the remaining matrix effects. Although approaches applied in sample preparation have effectively reduced the matrix interference, the matrix interference could not be totally eliminated from the sample extracts. Internal standard has shown to be the best way to compensate for matrix effects, but in the multi-component analysis one internal standard was deficient in compensating for the signal suppression on all analytes. Finding internal standards for each analytes is often difficult or impossible. To solve this problem and further compensate for remaining matrix effects, a matrix-matched calibration approach was selected. It has been clearly demonstrated in a systematic study that matrix-matched calibration with one internal standard is thus a practical alternative option to compensate for residual matrix effects in multi-component analysis of environmental samples. With the increasing use of mass spectrometric techniques for identification and quantitation, a clear understanding of the fragmentation behaviour of the analysed phytoestrogen isoflavones, enterolignans and coumestrol is required. This enhanced understanding will, in addition, aid the structural identification of other flavonoids by MS. To our knowledge, no systematic study has been reported of the fragmentation and structures of isoflavones in negative ion electrospray ionization mode by multi-stage mass spectrometric (MSn) fragmentation using ion trap mass spectrometry. Furthermore, enterolignans and coumestrol are two important types of phytoestrogens, but so far as we know, there is no literature has been reported on their MS fragmentation pathways. The fragmentation studies of isoflavones have elucidated the fragmentation pathway for deprotonated isoflavones in electrospray ionization using MSn ion trap mass spectrometry and triple quadrupole mass spectrometry. The fragmentation pathway of enterodiol, enterolactone and coumestrol were also proposed for the first time. Analytical results in this work showed that all analysed phytoestrogens were found to be present at trace levels in the aquatic environment, whilst some of the analytes were present at comparative high concentrations such as daidzein and enterolactone. Comparatively high levels of daidzein, enterolactone, genistein and formononetin were also detected in dairy farm runoff. Enterolactone were analysed in aquatic environment samples for the first time and was found at comparatively high levels. In wastewater samples, high levels of analytes were found in the STP influent. But in the STP effluent samples, concentrations of the analysed phytoestrogens were found at very low concentrations, especially in the final treated effluent from advanced tertiary treatment plants. These analytical results suggest that the source of phytoestrogen contamination in the environment samples seems not limited to the discharges of STP effluent, but was more likely attributed to the waste from livestock in the nearby farmyards. Comparatively higher levels of the analytes detected in the dairy farm runoff water samples supported the supposition. Another possible contamination of the analysed phytoestrogens is the dried and spoiled grass on the ground and in the farm grass silage, but more research on the livestock waste and spoiled grass are needed to make this conclusion, and would be interesting.

02Whole.pdf (2102 kB)