The Impact of Brine on Primary Settling of Wastewater

Potable reuse is expanding in many regions. Operation of potable reuse schemes will lead to a variety of system-wide impacts on existing water and wastewater infrastructure, and understanding these impacts will be an important element of integrated water management in the future. Potable reuse often includes reverse osmosis (RO) and produces a waste stream of brine, which in some cases will be disposed of in the existing downstream sewershed and augment the influent of existing wastewater treatment plants (WWTPs).This study examined the impact of brine on the performance of chemically enhanced primary treatment (CEPT) at a WWTP. The changes in salinity and organics makeup because of brine addition could affect the performance of flocculation and settling in CEPT by shifting the major ion profile and primary particle properties. However, because the system-level impacts of water reuse is an emerging field of study, the existing body of literature is relatively small and few studies have examined the impacts of brine on flocculation and settling. One of the few studies on the coagulation and flocculation of RO brine (Ho et al., 2015) found highly variable performance with ferric chloride outperforming other coagulants, but the mechanisms were unclear.In this study, a series of jar tests were conducted on the CEPT process at varying brine share and coagulant dose. The influent was sampled for total suspended solids (TSS) and the initial pH and turbidity of the jars were measured. Next, ferric chloride coagulant was dosed, and rapidly mixed into the solution at the maximum paddle speed for 30 seconds. The mixing speed was then lowered for flocculation for 2.5 minutes and anionic polymer was dosed. During flocculation, the shear rate was 26 s-1 and the energy dissipation rate, which has been shown in previous studies to be the relevant scaling parameter for flocculation (Tse et al., 2013), was 0.6 mW/kg on average and 1.2 mW/kg at the mixer tips. Finally, the mixing paddles were turned off for 15 minutes to simulate full-scale sedimentation, giving an effective capture velocity of 0.2 mm/s. The pH and turbidity of the final supernatant were measured, and samples were also taken for TSS.The primary influent TDS was generally around 1,700 mg/L, and it was augmented with RO brine with 5,000-6,000 mg/L TDS. The RO brine also contained a higher share of recalcitrant dissolved organics, with a much lower ratio of chemical oxygen demand to dissolved organic carbon compared to the primary influent. Brine augmentation, at fractions ranging from 2.7% to 54% of the primary influent, was observed to affect the performance of the CEPT process. Removal of turbidity and TSS decreased in response to increasing brine share, when holding coagulation and flocculation conditions constant (Figure 1). In addition, removal rates increased in response to increasing coagulant dose at constant brine fraction (Figure 2).Figure 1. CEPT performance vs. blended TDS at constant coagulant doseFigure 2. Removal efficiency at 27% brine with variable coagulant dose This effect was observed to be roughly proportional to the level of brine addition, rather than acting as a threshold effect. This effect appears to be driven by changes to the properties of the fluid and primary particles as a result of brine addition: adding brine both increases the density and viscosity of the fluid through higher TDS; and decreases the density of the primary particles that make up the floc by potentially contributing recalcitrant and low-molecular-weight organics rejected by RO. These effects combine to make the flocs settle more slowly by decreasing the difference between floc and fluid density. A sedimentation model, in the form presented by Adachi and Tanaka (1997), confirmed that this hypothesized mechanism is consistent with the measured effects of brine share and coagulant dose, and with the proportional (rather than threshold) nature of the effect.Additional testing found that CEPT performance could be restored to its current baseline performance by increasing coagulant dose, which appears to be because the coagulant increases the density of the flocs. Coagulant dose optimization trials run in a series of pseudo-randomized block experiments found a roughly linear relationship between influent brine proportion and additional coagulant required to bring CEPT performance back up to brine-free levels (Figure 3).Figure 3. Change in coagulant dose to match baseline performance