Insights into emerging organic contaminants behavior in different full-scale sludge treatment systems

1. Background Sludge treatment processes typically incorporate physical, chemical, and biological unit operations to reduce sludge volume, odour, vector attraction, and pathogens in the product biosolids [1]. Biosolids are regulated with respect to pathogen, nutrient, and heavy metal content if land applied [2]. However, biosolids may also carry a wide range of anthropogenic chemicals such as pharmaceuticals, personal car e products, surfactants, plasticizers, flame retardants, and water and stain resistant coatings [3], collectively called emerging organic contaminants (EOC). The presence of trace levels of EOCs has been identified as a concern associated with their land application as they could potentially mobilize into soil, infiltrate into ground water, or be uptaken by vegetation [4]. While the long-term risks of exposure to humans are relatively unknown, some of the EOCs are known to be toxic, persistent and bioaccumulative [3,4]. Hence, an understanding of the occurrence and fate of EOCs in sludge handling systems that produce biosolids will provide a basis on which risks can be quantified. Further, most studies pertaining to EOCs and wastewater solids only document the contaminant presence in the product biosolids and there is little information on their removal through sludge processing systems. Further, the relative performance of different physical, chemical and biological sludge treatment processes is also largely unknown The objective of this study was to study the behavior of five major classes of EOCs: per- and poly-fluoroalkyl substances (PFAS), polybrominated diphenyl ethers (PBDE), alkyl phenols and ethoxylate derivatives, bisphenol A (BPA), and triclosan in common sludge treatment systems. The final paper will highlight the key findings from this multi-plant study and discuss the practical implications for biosolids management practices. 2. Study Design Samples of raw sludge and the final stabilized biosolids were collected from nine sludge treatment systems that included pelletization (P), alkaline stabilization (AS), aerobic digestion (AE1 and AE2) and anaerobic digestion (AN1 to AN5) processes. Grab samples were collected in glass jars on three consecutive days within a one-month period in each of 2015 and 2016 and shipped overnight on ice to the lab for contaminant analysis. The target analytes included 13 PFAS, 21 PBDEs, 4 alkyl phenols and ethoxylate derivatives, BPA, and triclosan. All the EOC analyses were conducted by SGS AXYS International (Sidney, BC) following stringent quality control procedures. 3. Major Findings The measured concentrations were used to estimate contaminant mass flows (mg/day) in and out of the system (contaminant concentration ng/g x solids flow kg/day) to assemble contaminant mass balance through the system. Percent removal was then calculated to evaluate the relative performance of the treatment systems. An overview of five major findings from this study are subsequently presented. A detailed discussion on the influence of the contaminant and treatment plant related factors will be included in the final paper. 3.1 All types of sludge treatment lead to EOC removal: While sludge treatment systems are primarily designed to prepare biosolids for subsequent disposition, the results of this study suggest that some removal of EOCs also occurs during the process. Up to 87% removal of triclosan was observed in AE2 during aerobic digestion. Nonylphenol monoethoxylate and diethoxylates were removed in both aerobic (Mono: 59 to 86%; Di: 84 to 95%) and anaerobic digestion (Mono: 35 to 79%; Di: 69 to 96%). Alkaline stabilization, where the raw sludge was stabilized at high pH using lime and cement kiln dust, was effective in reducing ΣPBDE mass flows by 53 ±12%. During pelletization, where the raw sludge was thermally dried and compacted, there were low removals of perfluorodecanoate (PFDA) (17%) and ΣPBDE (6%) mass flows. Hence, all treatments resulted in some degree of contaminant removal but removals differed between technologies. 3.2 Contaminant formation is possible during sludge treatment: The conditions established in sludge treatment can also lead to transformation of contaminants resulting in an increase in mass flows through the systems. Such an effect was commonly observed for selected PFAS where the average mass flow of the measured compounds increased by 59%, 19%, 2%, 264%, and 13% in AS, P, AE1, AN1, and AN5, respectively (Figure 1). Further, a net increase in mass flow of 4-nonylphenol was consistently observed in all five anaerobic digestion systems. Evidence of reductive debromination of fully brominated PBDE congeners into partially brominated PBDE congeners was also seen in anaerobic digestion systems. Thus, sludge treatment systems can transform precursors to increase contaminant mass flows through treatment. 3.3 Removal is dependent on sludge feed characteristics: The raw sludge fed to sludge treatment processes can differ because of differences in the inputs of contaminants to the sewershed and in the types of wastewater treatment processes generating the sludge. EOC use and release in a given sewershed will be reflected in the contaminant loading from the sludge treatment process. For example, BPA was measured in the raw sludge in all treatment systems except AE2. Since BPA was not detected in the raw sludge at AE2, this substance was also not detected in the biosolids at this treatment system. Differences in upstream wastewater treatment processes design can also impact the EOC content in the final biosolids. For instance, the biosolids where the raw sludge was predominantly composed of primary sludge had less diversity in overall PFAS composition compared to blended raw sludge feeds (primary and waste activated sludge) where a wide variety in PFAS composition was observed. Hence, differences in raw sludge properties were found to impact the biosolids quality following treatment. 3.4 Removal is dependent on the type of sludge treatment: The target compounds behaved differently in physical, chemical and biological treatments leading to different final biosolids quality. For example, while alkaline stabilization was effective in removing PBDEs, anaerobic digestion led to an increase in PBDE mass flows. In addition, the behavior of perfluorodecanoate (PFDA) differed between treatment systems with responses ranging from moderate removal (41%) to substantial formation (-264%). Even in the studied biological processes, there were distinctions between aerobic and anaerobic sludge treatments. For example, PFOA had low levels of net removal (< 25%) during anaerobic digestion (AN3 and AN5) and contrarily a net increase in mass flow by -68% and -282% in aerobic digestion systems (AE1 and AE2 respectively). The differences in transformations was found to impact the final biosolids quality. 3.5 Removal is dependent on the plant operations: Site specific factors, especially in biological treatment systems, such as the solids retention time, staging of the treatment processes and the active microbial consortia appeared to impact contaminant fate through treatment and hence biosolids quality. For instance, BPA was removed in AN2 with an average efficiency of 75% whereas formation of BPA was seen in other anaerobic digestion processes (AN1, AN4 and AN5). In aerobic digestion, there were low levels of average removals of perfluorohexanoate (PFHxA) (28%) in plant AE1, whereas AE2 reported low average levels of formation (-11%). A detailed discussion on the site-specific factors and operating conditions that may be responsible for the observed removals and biosolids quality will be presented in the full paper. 4. Significance of the findings This study evaluated contaminant loadings and removal of five major classes of EOCs namely PFAS, PBDEs, alkyl phenol, BPA, and triclosan in different full scale sludge treatment systems. The lessons learned will benefit all stakeholders of the biosolids industry in understanding the responses of EOCs thorough sludge treatment. The removal efficiencies presented in this study will enable the treatment plant managers to appraise the performance of sludge handling systems with respect to EOCs. The contaminant loading estimates could be used to communicate the significance of biosolids quality to end users. Further, there is little information describing contaminant mass balances and overall removal through sludge treatment. This study bridged this critical knowledge gap and addressed major questions related to PFAS, PBDEs and phenolics in physical, chemical and biological sludge treatment processes.