Dynamic Membranes for Enhancing Resources Recovery from Municipal Wastewater

Abstract

This paper studied the feasibility of using dynamic membranes (DMs) to treat municipal wastewater (MWW). Effluent from the primary settler of a full-scale wastewater treatment plant was treated using a flat 1 µm pore size open monofilament polyamide woven mesh as supporting material. Two supporting material layers were required to self-form a DM in the short-term (17 days of operation). Different strategies (increasing the filtration flux, increasing the concentration of operating solids and coagulant dosing) were used to enhance the required forming time and pollutant capture efficiency. Higher permeate flux and increased solids were shown to be ineffective while coagulant dosing showed improvements in both the required DM forming time and permeate quality. When coagulant was dosed (10 mg L−1) a DM forming time of 7 days and a permeate quality of total suspended solids, chemical oxygen demand, total nitrogen, total phosphorous and turbidity of 24 mg L−1, 58 mg L−1, 38.1 mg L−1, 1.2 mg L−1 and 22 NTU, respectively, was achieved. Preliminary energy and economic balances determined that energy recoveries from 0.032 to 0.121 kWh per m3 of treated water at a cost between €0.002 to €0.003 per m3 of treated water can be obtained from the particulate material recovered in the DM.

Evaluating the Feasibility of Employing Dynamic Membranes for the Direct Filtration of Municipal Wastewater

“…Important differences in performance were achieved between the results obtained in this study using raw MWW as influent and previous studies that used primary settler effluent [ 15 ]. Much shorter DM self-forming periods were obtained with raw MWW, allowing the larger supporting material pore size.…”

“…Indeed, the particles between 100 and 1000 µm increased significantly when raw MWW was used (see Figure 3 ), which favoured the development of a DM by promoting the formation of a preliminary cake layer on the supporting material. However, as the self-forming time was reduced, the filtration fouling growth rate increased, achieving permeability losses in the DM of between 10.03 and 5.21 LMH bar −1 day −1 (see Table 2 ), depending on the supporting material used in this study with the 2.27 LMH bar −1 day −1 reached when filtering primary settler effluent [ 15 ]. Then, although using raw MWW could be considered as a more suitable influent for applying DMs when filtering MWW, more energy may be required to control the DM thickness.…”

“…Figure 1 shows a schematic diagram and picture of the pilot plant. Further information on this system can be found in Sanchis-Perucho et al [ 15 ].…”

“…DM performance was evaluated based on its permeability evolution. 20 °C-standardized permeability ( K 20 ) was calculated according to the following expression, which can be deduced from [ 15 ]: where J T represents the recorded permeate flux, T is the temperature and TMP ave is the average transmembrane pressure recorded during each filtration cycle.…”

“… where P M is the mixing pump energy requirements (W), Q M is the mixing volumetric flow rate (m 3 s −1 ), ρ is the mixed liquor density (Kg m −3 ), g is the acceleration of gravity (m s −2 ), L and L eq are the pipe length and equivalent pipe length (m), respectively, v is the liquor velocity (m s −1 ), f is the friction factor, D is the pipe diameter, and ( z 1 − z 2 ) is the height difference (m). On the other hand, the following expression was used to estimate the energy production when transforming the recovered organic matter into biogas [ 15 ]: where E R is the energy recovery (kWh m −3 ), COD R is the recovered COD concentration in the membrane module (kg m −3 ), Y CH 4 is the theoretical anaerobic methane yield of MWW sludge (3.5·10 −4 m 3 of methane per kg of COD), CV CH 4 is the methane calorific power (9.13 kWh per m 3 of methane), and η CHP is the CHP system methane electricity generation efficiency. A η CHP of 35% was used considering the different CHP technologies currently available [ 18 ].…”