Reliable Wastewater Disinfection with Peracids: Integrating Batch Kinetics and Hydrodynamics

INTRODUCTION Concerns related to the formation of disinfection byproducts (DBPs) during chlorine disinfection of wastewater and other safety concerns regarding chlorine operation initiated a great interest in alternative chemical disinfectants. For example, peracetic acid (PAA) received significant attention due to its minimal aquatic toxicity and efficacy in reducing regulated fecal indicator bacteria (FIB) (Antonelli et al., 2006; Manoli et al., 2019; WEF, 2020). The generally low efficacy of PAA in inactivating viruses in wastewater along with the ongoing discussions about revision of the United States Environmental Protection Agency Ambient Water Quality Criteria to include a viral indicator (Dunkin et al., 2017), led to a strong interest in another peracid, performic acid (PFA), which its bacterial and viral inactivation capacity usually outperforms PAA (Ragazzo et al., 2020). The objective of this study was to establish efficacy of PFA as a disinfectant compared to PAA and sodium hypochlorite for secondary effluent wastewater applications. The inactivation of indigenous FIB by PFA has been evaluated and compared with PAA and sodium hypochlorite under identical conditions. The disinfection data were interpreted using the concept of integral CT (ICT) and modeled by an ICT-based double-exponential microbial inactivation kinetic model to develop ICT-response data. This abstract presents the initial results of the Water Research Foundation Project #5219 titled 'Performic Acid Disinfection in Wastewater Effluent and CSOs In Context of Disinfection By-Products and Future Ambient Water Quality Criteria for Viruses', with the participation of more than 10 wastewater treatment plants (WWTPs) across the United States and Canada. Similar experiments with the ones presented herein for the first WWTP will be performed for all participating utilities, thus the results of this project may have a significant impact on disinfection practices in North America. Additional to FIB experiments, disinfection tests on viral indicators inactivation and DBPs formation will also be performed in the framework of this project. METHODS Secondary effluent wastewater was collected from a WWTP in ON, Canada (activated sludge-based processes). Its basic water quality characteristics were: pH=7.6, TSS=7.7 mg/L, BOD=10.8 mg/L, COD=13.5 mg/L, TN=18.6 mg/L, and TP=0.4 mg/L. Disinfection bench tests were performed in 2 L of wastewater at 2 disinfectant concentrations (1 and 2 mg/L for PFA and chlorine; 2 and 3 mg/L for PAA). Samples were collected at 6 contact times (0, 1, 3, 5, 10 and 30 min) and analyzed for disinfectant concentration (Chemetrics and Hach methods) and E. coli, fecal coliforms, total coliforms and enterococci (standard membrane filtration methods) (GAP EnviroMicrobial Services, London, ON, Canada). RESULTS The residual disinfectant concentration data were used to calculate exposure to chemical disinfectant (i.e., integral CT) by fitting initial demand (D, mg/L) and decay (k, 1/min) (Eq. 1): ICT=((C0-D)/k)∙(1-e^(-k∙t))(1) where ICT is the integral estimate of the time-dependant residual disinfectant concentration (mg min/L), C0 is the initial concentration of disinfectant (mg/L) and t is the contact time (min). The FIB inactivation data were plotted as a function of integral CT (Figures 1-4) and modeled by a double-exponential inactivation kinetic model in the domain of ICT (Eq. 2): N=N0∙(1-β)∙e^(-kd∙ICT^m )+N0∙(β)∙e^(-kp∙ICT)(2) where N0 and N are the initial and after treatment concentration of bacteria (CFU/100 mL), β is the fraction of particle-associated bacteria, m is a parameter that describes initial resistance of bacteria to disinfectant, and kd and kp are the ICT-based inactivation rate constants for dispersed (free) and particle-associated bacteria, in (L/(mg min))m and L/(mg min), respectively. PFA exhibited the highest inactivation efficiency for all studied indigenous FIB, followed by chlorine and PAA (Figures 1-4). For example, model-predicted integral CT requirements for 1-log reduction of total coliforms and enterococci were determined as 3-4 mg min/L, 8-10 mg min/L and 40-45 mg min/L for PFA, chlorine and PAA respectively (Figures 3 and 4). Based on model predictions, to achieve a 2-log reduction of both commonly regulated E. coli and fecal coliforms, PFA required ~10 times lower ICT than both PAA and chlorine (Figures 1 and 2). We also present an advanced disinfectant dosing control strategy that accounts for changes in wastewater quality and quantity, together with the hydrodynamics of the contact chamber. Bench test results of PFA (Eqs. 1 and 2) were integrated into a process model to evaluate four dosing strategies, namely, flow pacing, constant ICT, constant ICT advanced demand, and constant ICT advanced demand/decay. Figure 5 and Table 1 show that the advanced control strategy achieves consistent performance and optimizes disinfectant dosage. CONCLUSIONS Results show that PFA was superior to both PAA and chlorine. Although the PFA pricing estimates are currently higher than that of hypochlorite, the lower ICT requirements of PFA may result in a competitive cost compared to chlorine. This indicates the need for site-specific testing to better understand disinfection efficiency and cost benefits of PFA as a wastewater disinfectant.