Elsevier

Chemosphere

Volume 271, May 2021, 129530
Chemosphere

Linking PFAS partitioning behavior in sewage solids to the solid characteristics, solution chemistry, and treatment processes

https://doi.org/10.1016/j.chemosphere.2020.129530Get rights and content

Highlights

  • Sludge stabilization significantly affected PFAS Kd in biosolid.

  • Secondary treatment did not significantly affect PFAS Kd in sludge.

  • Biosolids generally had significantly lower PFAS Kd compared to secondary sludge.

  • Protein was stronger predictor of PFAS Kd than organic matter or lipid fraction.

  • Elevated mono- and divalent cations increased PFAS Kd, while high pH decreased it.

Abstract

Per- and polyfluoroalkyl substances (PFAS) have gained increasing attention due to the potential health risks that they present. Secondary sludge and biosolids are known as notable PFAS emission routes to the environment. In this study, partitioning behavior of 14 PFAS were investigated across four secondary wastewater treatment types (activated sludge, trickling filter, biological nutrient removal, and rotating biological contactor; n = 10) and three sludge stabilization methods (composting, aerobic digestion, and anaerobic digestion; n = 6). Batch experiments were conducted to evaluate how PFAS sorption to secondary sludge and biosolid was affected by various treatment methods, solid properties, and solution chemistry parameters. Insignificant differences in compound-specific partitioning coefficients (Kd) were observed among the four secondary treatment methods. However, sludge stabilization resulted in significantly different partitioning behavior among biosolid samples, in which anaerobically digested biosolids generally had significantly higher Kd values compared to aerobically digested and composted biosolids (anaerobic digestion > aerobic digestion > composting). Multiple linear regression models were developed to explain analyte-specific Kd values across the biosolid samples and identified that solid-specific property significance was as follows: protein fraction > organic matter fraction > lipid fraction. Stabilization generally decreased the PFAS sorption capacity relative to the secondary sludge samples. Furthermore, PFAS Kd increased with elevated calcium concentrations and ionic strengths and decreased with increasing pH values in sludge and biosolid samples. These findings could inform the decision-making process to reduce the release of PFAS to the environment.

Introduction

Per- and polyfluoroalkyl substances (PFAS) are a class of diverse man-made chemicals which include a non-polymeric subgroup that is often characterized by a hydrophobic fluorinated carbon chain attached to an ionic headgroup (Kumar, 2005). The strong C–F bonds in PFAS make them thermally and biologically stable while the surfactant characteristics result in partitioning to interfaces and comparatively high mobility; more detailed discussion of PFAS sources, physio-chemical properties, fate, transport, and effects can be found elsewhere (Arvaniti and Stasinakis, 2015; Giesy and Kannan, 2002; Kannan, 2011; Kumar, 2005). These unique characteristics justify their extensive use in many consumer products such as cookware, personal-care products, and food packaging (Giesy and Kannan, 2002). PFAS have increasingly gained attention from the public, scientific, and regulatory sectors due to the carcinogenic, reproductive, and endocrine disruptive effects that they present (Ahrens, 2011; Gorrochategui et al., 2014).

Previous research has illustrated that water resources recovery facilities (WRRF1; also known as wastewater treatment plants) are notable PFAS emission routes to the environment, reflecting both liquid and solid discharge routes. According to Northeast Biosolid and Residual Association, more than seven million dry tons of sewage solids were produced at WRRF in 2004, of which 55% were land applied as soil quality enhancing additives (biosolids) and the remaining 45% were discarded through incineration and landfill disposal options. These sewage solids management practices have resulted in the release of approximately 3000 kg PFAS per year to agricultural lands and landfills (Venkatesan and Halden, 2013). PFAS may be taken up by plants and find their way into the agricultural products and foods, constituting a noteworthy human exposure route. PFAS compounds may accumulate within shallow depths of soil or leach into the soil and make it to groundwater (Blaine et al., 2013; Ghisi et al., 2019; Wen et al., 2014). PFAS pollution in groundwater can be mostly correlated to the flux of short chain PFAS compounds on land. One study found short chain PFAS (<C8) at depth of 1.2 m and more following a land application of biosolids as a soil amendment (Washington et al., 2010). Considering the significant quantities of PFAS-contaminated sewage solids (i.e., sludge and biosolid) generated by WRRF and the overall resistance of PFAS to degradation in conventional treatment trains (Deng et al., 2010; Lewis et al., 2020; Rayne and Forest, 2009; Ross et al., 2018), PFAS partitioning to sewage solids is considered a major removal pathway in WRRF and, consequently, a significant contributor to the release of such chemicals to the environment.

WRRF unit processes and treatment trains, including secondary treatment processes and sludge stabilization methods, can affect the sewage solids physical and chemical characteristics (Guerra et al., 2014; Stasinakis, 2012). For example, anaerobic digestion was found to reduce the fraction of volatile solids and increase sorption capacity of hydrophobic contaminants (Stasinakis, 2012). Another study reported that considerably higher concentrations of longer chain length perfluoroalkyl acids were found in digested sludge samples compared to non-digested samples (Guerra et al., 2014). Additionally, previous research has illustrated that PFAS concentrations were affected by type of treatment process (e.g., primary vs. secondary treatment) and type of unit process (e.g., aerobic biological vs. advanced biological nutrient removal) (Guerra et al., 2014; Schultz et al., 2006). Furthermore, operational parameters associated with the secondary treatment methods such as hydraulic retention time (HRT) and temperature have been found to be correlated with the conversion of PFAS precursors, with higher precursor conversion levels reported at higher HRT and temperatures (Guerra et al., 2014). Perfluoroalkyl acid (PFAAs; the stable terminal transformation products) formation through biotransformation of fluorinated precursors is a notable mechanism affecting the PFAA load in the WRRFs effluent, sometimes resulting in higher post-treatment concentrations relative to the influent concentrations (Arvaniti et al., 2012; Schultz et al., 2006; Stasinakis, 2012). These findings suggest that the treatment trains adopted in WRRF could impact the mass flow and PFAS concentration profiles as well as their sorption capacity to sewage solids.

A review of past studies illustrates a wide range of PFAS partitioning coefficients in various environmental media. Thus, it is not yet clear how solid properties and solution chemistry parameters affect PFAS sorption behavior in sewage solids (secondary sludge and biosolid) resulting from various treatment processes. Additionally, while previous research have investigated the effects of various treatment processes on the fate of PFAS in WRRF (Chen et al., 2012; Guerra et al., 2014; Lazcano et al., 2019; Wang et al., 2016), to our knowledge, no previous studies have linked PFAS partitioning behavior in secondary sludge and biosolids to treatment methods and few studies have compared PFAS sorption capacity of secondary sludge with their associated biosolids. Therefore, the objectives of this study were to investigate the effects of a variety of solution-specific characteristics (pH, mono- and polyvalent cation concentration), solid characteristics (organic matter, protein, and lipid content), PFAS characteristics (chain length, head group), and treatment methods (secondary treatment and sludge stabilization methods) on the sorption of PFAS to sewage solids.

Section snippets

Chemicals and materials

The evaluated PFAS are anions under environmentally relevant conditions (circum-neutral pH values), so the anionic names are employed throughout the current research. The complete list of PFAS products and manufactures can be found in the supplementary information (Table S1 and section 1.1, respectively), which included eight perfluorocarboxylate (PFCA)2, four perfluorosulfonate (PFSA)3, and two fluorotelomer sulfonate (FtS)4

Partitioning behavior in secondary sludge

Evaluation of isotherm fitting efforts for 14 PFAS across the secondary sludge samples revealed mixed results on whether the linear or Freundlich model better fit the data (full fitting details in Tables S4–S19). Generally, these samples exhibited both isotherm forms, as exemplified by RBC_J in Fig. 1, though the distribution of best-fit was sample-dependent (Fig. 1b). The linear model generally fit all isotherm data moderately to very well (R2>0.9) and generally prevailed as the best fit for

Conclusions

The results of the current study can lead the efforts on minimizing the release of PFAS through the solid effluents of WRRF. Even though PFAS-specific parameters such as their total mass, profile, and composition are not expected to vary significantly through the treatment processes, solution chemistry, solid characteristics, and stabilization methods can change the leaching potential of the sewage solids and hence can impact the release of PFAS to the environment. Since controlling the

Authors contributions

Farshad Ebrahimi: Investigation, Writing – original draft, Visualization, Formal analysis, Methodology. Asa Lewis: Investigation, Writing – review & editing. Christopher M. Sales: Conceptualization, Resources, Writing – review & editing, Funding acquisition. Rominder Suri: Conceptualization, Writing – review & editing, Funding acquisition. Erica R. McKenzie: Supervision, Funding acquisition, Resources, Writing – review & editing, Conceptualization, Methodology

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This material is based upon work supported by the National Science Foundation under Grant No. CBET-1805588, the Water Research Foundation under Award No. 5002, and fellowship support from Temple University to Farshad Ebrahimi. Additionally, the Sciex X500R purchase was partially supported by an Army Research Office DURIP award #W911NF1910131 and the Temple University College of Engineering. The opinions expressed in this article are those of the authors and do not necessarily reflect the views

References (36)

Cited by (38)

View all citing articles on Scopus
View full text