REPOSITORY > RESULTS

Abstract

The presence of pharmaceutical residues in surface waters is an emerging environmental issue that provides a new challenge to treatment systems for potable water, wastewater and water reuse. Their principle pathways into the sewerage system are twofold: first, after being administered to patients they are normally excreted as various metabolites together with the unchanged parent compound; secondly, it is common practice to dispose outdated or unused medicines either down the drain or flush down the toilet. In either case, significant quantities of pharmaceutical residues eventually find their way to wastewater treatment plants and the occurrence of these residues in surface waters and groundwaters is due to the inefficient removal processes of conventional water treatment and emphasises the need for enhanced or new treatment technologies. The removal of pharmaceuticals and other organic micro-pollutants still only provides a partial indication of the efficiency of the various treatment methods because of the possible generation of nontargeted toxic intermediates more resistant to degradation. When it comes to pharmaceutical residues many gaps exist in our knowledge regarding degradation mechanisms, the identity of transformation products, and their impact on and their fate in the environment. Insights into this relatively unexplored field are hindered by the quantity and the variety of products present at various degrees of transformation. To resolve this will require the application of sophisticated instruments capable of providing complementary information for structural assignment and robust analytical methods, which require meticulous method development and validation.
This thesis describes an analytical method for determining key pharmaceutical representatives of nonsteroidal antiinflammatory drugs: ibuprofen, ketoprofen, naproxen and diclofenac, in aquatic matrices, and its validation at the intralaboratory and interlaboratory level. The analytical method is based on solid phase extraction and derivatisation followed by GC-MS. An intralaboratory validation revealed the method to be satisfactory in terms of linearity (r2 ≥0.990), sensitivity, limits of detection (2 – 6 ng L-1), and extraction efficiency (>84 %), thus proving its suitability for determining NSAIDs in aqueous environmental samples. The method was then applied to various Slovenian surface water, groundwater and tap water samples. The concentrations of pharmaceuticals (range 10 – 300 ng L-1) determined were comparable to those found elsewhere in Europe. Furthermore, naproxen, when compared to the other selected pharmaceuticals is dispensed in the highest amounts, was detected in the majority of samples also in the highest concentrations. Interlaboratory validation involving GC-MS and LC-MS was also performed in two separate round robin studies. In summary, it was shown that the GC-MS method was superior to LC-MS method when determining ibuprofen, naproxen and ketoprofen in complex matrices, while neither of the two methods was satisfactory for determining diclofenac. In addition, the process of filtration and the filter material had no affect on the determination of NSAIDs.
Experiments involving the biological removal of pharmaceuticals with activated sludge revealed two groups: readily biodegradable compounds (removal ≥87 %) including ibuprofen, ketoprofen and naproxen, and recalcitrant compounds diclofenac, clofibric acid and carbamazepine, showing ≤59 %, ≤30 % and 16 % elimination, respectively. The study was performed in laboratory scale bioreactors, and the results were in agreement with actual data from municipal wastewater treatment plants. Because the activated sludge in the bioreactors was continuously exposed to the selected pharmaceuticals the effects on the makeup of the microbial community in the activated sludge were monitored. Overall, the pharmaceuticals in concentrations ≥50 μg L-1 resulted in a change in the microbial composition in the activated sludge, which became exaggerated with increasing concentration. The pharmaceuticals reduced the diversity in the activated sludge and especially worrying was the notable absence of Nitrospira, the bacteria responsible for 2nd stage nitrification.
This thesis also examines the biotransformation of certain pharmaceuticals. Interestingly, only the persistent pharmaceuticals diclofenac, clofibric acid and carbamazepine yielded any discernable biotransformation products. Among a number of diclofenac‟s biotransformation products identified was 1-(2,6-dichlorophenyl)-1,3-dihydro-2H-indol-2-one, hydroxy-diclofenac, a benzoquinone imine derivative and a nitro-analogue of diclofenac. Further, 4-chlorophenol was identified in the bioreactor feed along with clofibric acid, and acridine and 9-acridone were formed during the biotransformation of carbamazepine. From the literature and according to this study certain transformation products are also formed via different mechanisms involving the abiotic breakdown or human and animal metabolism. Among the identified biotransformation products most notable is the incorporation of a nitro-group into the diclofenac molecule. This study also focused on the complementary use of mass spectrometric techniques including GC-IT, GC-Q and LC-QqTOF to detect as many transformation products as possible, and to apply different mass spectrometric techniques for cross-confirmation of their chemical structures. Among the available mass spectrometric techniques the LC-QqTOF was the most powerful in resolving the chemical structures of the transformation products because of its unique ability to perform both tandem mass fragmentation and accurate mass measurement.
Similar experiments under abiotic conditions, primarily UV irradiation and chlorine dioxide disinfection, were made. Out of the eight metabolites of carbamazepine detected, the structures of seven transformation products, formed during at least one of the treatment methods, were determined. In addition, one compound was generated by thermal decomposition during sample analysis. This study also compared treatment technologies according to the removal of carbamazepine and the production and decay of its transformation products. The most efficient at removing carbamazepine was UV treatment (93 %), while 76 % of acridine and <10 % of acridone were removed. Alternatively, acridine (≤92 %) and acridone (≤40 %) were more susceptible to biological treatment than carbamazepine (16 %). Therefore, based on the enhanced biodegradability of carbamazepine residues achieved by UV irradiation, a coupled treatment technology is proposed involving an initial UV treatment step followed by biological treatment. Similarly, chlorine dioxide degrades carbamazepine more efficiently (54 %) than acridine (38 %), and it may be applied prior to biological treatment.
Finally, at least three identified transformation products: 4-chlorophenol from clofibric acid; acridine and 9-acridone from carbamazepine, were found to be more toxic and hazardous than the parent compounds, a fact which supports the original thesis that considering only a parent pharmaceutical, more harm can be done by insufficient treatment than by completely avoiding it. Thus, with an aim of achieving complete mineralisation of both parent compounds and their transformation products, further development of treatment technologies, possibly involving additional and/or sequenced treatments, is needed. The efficiency of newly-developed treatment technologies, however, will require scale-up and evaluation, both from a scientific and economic perspective.