Scaling in reverse osmosis (RO) process

• Main Author: Tang, Wei Guang
• Other Authors: Chong Tzyy Haur
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Environmental engineering::Water treatment
• Online Access: http://hdl.handle.net/10356/71591
• Institution: Nanyang Technological University

As water scarcity and demand increases, seawater desalination has become increasingly important as a stable and reliable water source. Nanofiltration (NF) has shown great effectiveness as a pretreatment for seawater reverse osmosis (SWRO). High operating pressures of 17 bars – 40 bars are usually required. For this project, novel low pressure NF hollow fibres membranes and synthetic seawater will be used. These novel membranes were developed by the Singapore Membrane Technology Centre (SMTC) and were used for other processes such as forward osmosis or brackish water softening. The focus of this project will be on scaling. The overall aim of this project is to investigate whether it is suitable to use the novel NF membranes for pretreatment for the SWRO process. The objective of this experiment is to measure the changes in permeate flux and ion rejection across a membrane module due to the increase in feedwater concentration caused by permeate recovery. Two types of NF membranes were used - one with high rejection and one with moderate rejection. However, as experimental hollow fibres are used, they are not long enough to simulate the rise in concentration across the membrane modules. To replicate the increase in concentration of the feed solution, retentate recirculation was used instead. Across a membrane module, the recovery rate changes. In this experiment, the effect of recovery on permeate flux and solute rejection rate is investigated. Permeate flux decreases somewhat proportionally to an increase in recovery rate. For solute rejection rate, a threshold seems to exist. Before a certain recovery rate, solute rejection rate decreases fair proportionally. Beyond the recovery rate, the decline in rejection rate becomes more exponential. Actual seawater was also used for a run. The trends and values obtained are similar, suggesting that using synthetic seawater provides us with a sufficient estimate of the membrane performance with actual seawater.