High rejection reverse osmosis membrane for removal of N-nitrosamines and their precursors

Abstract

Direct potable reuse is becoming a feasible option to cope with water shortages. It requires more stringent water quality assurance than indirect potable reuse. Thus, the development of a high-rejection reverse osmosis (RO) membrane for the removal of one of the most challenging chemicals in potable reuse - N-nitrosodimethylamine (NDMA) - ensures further system confidence in reclaimed water quality. This study aimed to achieve over 90% removal of NDMA by modifying three commercial and one prototype RO membrane using heat treatment. Application of heat treatment to a prototype membrane resulted in a record high removal of 92% (1.1-log) of NDMA. Heat treatment reduced conductivity rejection and permeability, while secondary amines, selected as N-nitrosamine precursors, were still well rejected (>98%) regardless of RO membrane type. This study also demonstrated the highly stable separation performance of the heat-treated prototype membrane under conditions of varying feed temperature and permeate flux. Fouling propensity of the prototype membrane was lower than a commercial RO membrane. This study identified a need to develop highly selective RO membranes with high permeability to ensure the feasibility of using these membranes at full scale.

Modelling and optimisation of a multistage Reverse Osmosis processes with permeate reprocessing and recycling for the removal of N-nitrosodimethylamine from wastewater using Species Conserving Genetic Algorithms

“…This higher recovery is achieved not only due to permeate recycle by also at the expense of higher energy consumption. Also note, the NDMA rejection rate 93.1% is higher than that reported by Fujioka et al[14] which was 92%. Although the recovery rate and energy consumption were not explicitly mentioned by Fujioka et al[14], the 92% NDMA rejection was achieved at the expense of reduced membrane permeability between 20 -35% for several types of membranes at the same selected feed concentration of 1000 ng/L.…”

“…Due to its inherent property, the SCGA method has provided several optimum solutions for each selected configuration A to E. The nondominated solution is shown in bold in Table 5. A close look at the results shows that the permeate reprocessing design has significantly improved the removal of NDMA compared with the findings of the literature, which approximately ranged between 40 -70% [9,11,12] except that of Fujioka et al [14] (92% rejection was obtained but under heat treatment with a substantial reduction of water transport parameter and recovery rate). Interestingly, solution A1 is close to that obtained by Fujioka et al [14] in terms of rejection.…”

“…2 using the SCGA technique. The inlet feed concentration of NDMA for all configurations is 1x10 -6 kg/m³ (1000 ng/L) which is same as that of Fujioka et al [14]. The final NDMA rejection ( ( ) ) and energy consumption (EC) are chosen as the two objective functions for the optimisation problem where NDMA rejection is maximised and EC is minimised simultaneously for each configuration (A to E), while optimising the operating conditions of plant feed pressure, flow rate and temperature.…”

“…Fujioka et al [13] used a full-scale retentate reprocessing design of three spiral wound RO membranes connected in a series configuration and confirmed a range of 40 to 61% NDMA removal. More recently, Fujioka et al [14] have improved the NDMA removal to 92% based on an immersion of RO membranes in high-temperature ultrapure water. One possible explanation of this is the fact that heat treatment may tighten the membrane structure and therefore improve the removal of NDMA.…”

Membrane processes

“…A high-rejection RO membrane was used to remove NDMA using a heat treatment on the membrane that resulted in high ANNUAL LITERATURE REVIEW levels of rejection with less fouling than that of the membrane that was not heat-treated (Fujioka, Ishida, Ishida, Shintani, & Kodamatani, 2018). Two synthesized zeolite nano-adsorbents were used in an MBR to reduce fouling by Hazrati, Jahanbakhshi, and Rostamizadeh (2018).…”