Coupling ozone with GAC, AIX and biochar: Removal of pharmaceuticals from the biologically treated wastewater and fate of their transformation products

Abstract

In order to prevent water pollution in the water bodies, there has been a rising interest in improving cost-efficient quaternary treatment technologies to efficiently remove pharmaceuticals from the effluents of wastewater treatment plants. Various methods, including physical, biological, and chemical processes, are being utilized to eliminate organic micropollutants (OMPs) which include pharmaceuticals and persistent pollutants such as per- and polyfluoroalkyl substances (PFAS). Among these advanced techniques, ozonation and activated carbon adsorption are currently suggested as the most feasible options for substantially decreasing pharmaceutical concentrations in the wastewaters. Carbon-based materials such as activated carbon are notably effective adsorbents used for removing pharmaceuticals. Likewise, ozone is a highly potent oxidizing agent capable of oxidizing micropollutants directly via O3 itself or indirectly through the generation of hydroxyl radicals. However, after ozonation, instead of being mineralized, compounds can be converted into other substances known as transformation products, which may pose greater toxicity than the original compound. Additionally, the reaction between bromide and ozone produces bromate which is a toxic and carcinogenic by-product. To mitigate potential adverse effects from ozonation, it is often recommended to implement post-treatments such as biological or adsorptive systems like granular activated carbon (GAC) to eliminate potential transformation products and by-products. When combined with activated carbon adsorption, ozonation acts as an additional method for removing compounds that are resistant to adsorption. Crucially, activated carbon, with its extensive specific surface area and high concentration of functional groups, has demonstrated its ability to eliminate transformation products and by-products that may be generated during ozonation. While the effectiveness of the O3-GAC pairing is well-documented in literature, less attention has been given to combinations like ozonation with anion exchange (AIX) or other potentially more sustainable sorption materials such as biochar. The O3-AIX combination is particularly intriguing for this study, as many wastewater treatment plants, including the WWTP that provided wastewater for this research, are grappling with the challenge of removing PFAS. The treated wastewater from the aforementioned WWTP is released into the Fyrisån River, which flows into Ekoln Lake, Mälaren Lake, and eventually to the Baltic Sea. Fyrisån River also contributes to replenishing a groundwater source used for the city's drinking water. Consequently, the conventionally treated wastewater, containing untreated micropollutants, is discharged into a river that ultimately serves as a drinking water source. Thus, this study aims to simulate a potential combined advanced treatment step for the wastewater treatment plant. This research is particularly noteworthy as it not only provides guidance for implementing advanced techniques to remove micropollutants in full-scale wastewater treatment plants, but also investigates the fate of transformation products within three different combined systems: ozone and granular activated carbon filtration, ozone and ion exchange, and ozone and biochar. The objective of this thesis was both to examine the effectiveness of removing 24 selected pharmaceuticals and to monitor the fate of 7 of their metabolites, including oxidation transformation products. This investigation utilized a combination of processes, namely O3-GAC, O3-biochar (with two different types of biochar), and O3-AIX, in laboratory-scale experiments using actual effluent from a full-scale WWTP. The entire system was operated with three different O3 dosages, each maintained continuously for two weeks. Various sorption filters, including two types of biochar (one derived from forest biomass and the other from sewage sludge), reactivated GAC, and an AIX resin, were assessed. The evaluation of results focused not only on micropollutant removal but also on the generation of transformation products and by-products. 23 out of the 24 pharmaceuticals examined were detected in the effluent wastewater collected from the Kungsängsverket WWTP in total. The findings revealed that concentrations of sertraline, trimethoprim, fluconazole, atenolol, and sulfamethoxazole were below 500 ng/L, whereas the average concentrations of venlafaxine, desvenlafaxine, fexofenadine, bicalutamide, and lamotrigine were above 5,000 ng/L. According to the study findings, the average removal of selected pharmaceuticals varied between 8.8% and 97% with an O3 dosage of 0.28 g O3/g DOC, while it ranged from 86% to 99% for higher O3 dosages (0.96 and 2.17 g O3/g DOC). Pharmaceuticals such as fluconazole, atenolol, metoprolol, and tramadol exhibited relatively lower removal rates (9-15%) with the specific O3 dosage of 0.28 g O3/g DOC compared to furosemide, propranolol, clindamycin, and clarithromycin, which showed high removal rates (>90%). Tertiary amines like cetirizine and fexofenadine, known for their high reactivity with ozone, achieved removal rates of 79% and 89%, respectively, via 0.28 g O3/g DOC in this study. Furthermore, highly reactive compounds such as carbamazepine, diclofenac, sulfamethoxazole, and trimethoprim were removed by 70%, 85%, 70%, and 88%, respectively, with 0.28 g O3/g DOC, consistent with existing literature. Conversely, fluconazole exhibited a removal rate of 9% with an O3 dosage of 0.28 g O3/g DOC, while atenolol had an average removal rate of 15%. Among all materials tested, GAC emerged as the top-performing sorbent, effectively removing nearly all compounds below the limit of quantification (LOQ) even after continuous operation for two weeks (BV=864). The potential efficacy of biochar 2 for pharmaceutical removal, which was derived from sewage sludge, was particularly significant for the overall sustainability of the WWTP. Although biochar 1 exhibited better performance than biochar 2, both sorption materials showed decreased sorption capacity over the two-week period (BV=864) for most target compounds, including carbamazepine, fexofenadine, tramadol, fluconazole, sulfamethoxazole, and erythromycin. By the end of the continuous two-week operation, biochar 1 achieved removal rates ranging between 30% and 89% (mean 68%), while biochar 2 removed selected compounds at rates of 8.5% to 82% (mean 38%). Conversely, AIX that has been included for PFAS removal, demonstrated lower removal rates as expected after two weeks compared to biochars 1 and 2, ranging between 2% and 55% (average: 20%) for positive removal rates (BV= 3,264). Based on the findings, GAC exhibited the highest performance when paired with ozone (>99%), followed by biochar 1. Generally, the combination of ozone with biochar 1 proved to be more effective (mean=91%, range: 42-99%) than with biochar 2 (mean=79%, range: 29-99%). As anticipated, the combination of ozone with AIX yielded the lowest removal rates for pharmaceuticals (mean=58%, range: 6-98%). Based on the findings, six out of seven metabolites were identified in samples both pre- and post-ozonation. The results suggested that while the concentrations of certain metabolites decreased during ozonation, some metabolites, including oxidation transformation products like citalopram N-oxide, exhibited an increase over the two weeks of continuous operation. On average, citalopram concentration decreased by 81%, whereas the concentration of citalopram N-oxide increased by 19% with an O3 dosage of 0.28 g O3/g DOC. With the system operating at 2.17 g O3/g DOC, citalopram's average removal reached the LOQ, while the increase in citalopram N-oxide exceeded to 33%. Furthermore, all detected metabolites were eliminated to below the LOQ using GAC after two weeks of operation. Concentrations of most metabolites exhibited a linear decrease over time for biochar 1 and biochar 2, while for AIX, concentrations of certain metabolites increased over time. During all three O3 dosages (0.28, 0.96, and 2.17 g O3/g DOC), bromate concentrations remained below 5 µg/L. At the lower O3 dosage of 0.28 g O3/g DOC, the bromide concentration in the utilized WWTP effluent was 1.03 mg/L, whereas at the higher O3 dosages of 0.96 and 2.17 g O3/g DOC, the bromide concentrations were 0.52 and <0.50 mg/L, respectively. This variation resulted in an inability to assess the potential formation of bromate. Removal of DOC via different O3 dosages ranged from 19% to 26%, while GAC removed over 90% of DOC under all operational conditions. Conversely, AIX only removed less than 10% of initial DOC across all operational conditions, while in all cases, biochar 1 and biochar 2 removed within the range of 18-23% and 5-10%, respectively. In summary, ozonation exhibited high removal efficiency of pharmaceuticals and their metabolites at higher O3 dosages (>0.96 g O3/g DOC), while at lower O3 dosages (0.28 g O3/g DOC), a post-treatment became necessary for effective pharmaceutical removal. However, higher O3 dosages entail increased operational costs and pose a risk of transformation product formation. Therefore, employing combined systems for pharmaceutical and metabolites elimination is suggested as a preferable alternative to sole reliance on ozonation as the advanced treatment method. Comparative analysis of different post-treatment filter sorbents indicated that GAC yielded the most favorable results for pharmaceutical and metabolite removal. Conversely, the adsorption capacities of two distinct biochar types diminished over the continuous two-week operation, whereas GAC's performance remained consistent throughout. Biochar 1 outperformed biochar 2 in terms of pharmaceutical removal. AIX exhibited the lowest removal efficiencies, suggesting it may not suffice as a polishing step for ozonation when simultaneous removal of pharmaceuticals and PFAS is targeted. Overall, the combination of O3 with GAC demonstrated the most effective performance for pharmaceutical removal. Biochar holds promise as a more sustainable substitute for GAC, as it can be sourced from renewable materials like wood. However, there is a need for ongoing development to better understand the efficacy of combined O3-filter systems, with a focus on considering long-term operation. Before scaling up to a full-scale WWTP, conducting a life cycle assessment and feasibility analysis would be prudent steps to take.