Performance benchmarking of polyamide composite Sea water reverse osmosis membranes

Reverse osmosis (RO) plant owners, operators, designers and membrane manufacturers have benefitted greatly due to adoption of standardization in element design, material and supporting equipments (Fritzmann et al. 2007). Standardization allows transparency and accountability in performance benchmarking of different membrane elements, especially when a new product is introduced to the market and a plant owner is willing to yardstick against established players. In the current study, we compared performance of three polyamide composite RO membranes (one new membrane, the other two are established manufacturers) for seawater desalination.

The study was conducted at a desalination plant in Egypt, from 06^th June 2015 to 30^th April 2016. The 21,000 m³/day capacity plant receives seawater from Red Sea and the feed water conditions are summarized in Table 1.

The plant operators spared three pressure vessels (PVs) for the study, each containing six elements. The location of the test PVs in membrane train is shown below (Figure 1).

The general information and application data for each membrane is summarized in Table 2. SWRO1 is the new membrane entrant to market while the other two are well established and have numerous references globally. The data in Table 2 indicate that except for larger membrane area in SWRO2 other elements characteristics are comparable.

The plant operators maintained train feed pressure at 60 bar for the duration of the study. Feed pressure of the three test PVs could not be measured due to system constraints. Nevertheless, the objective was to profile performance, i.e. permeate conductivity, salt rejection and permeate flow, in varying feed seawater conditions.

Permeate conductivity and flow was measured in the first 24 hours of start-up operation of the three PVs. Their comparative values are shown in Figure 2 and Figure 3. SWRO1, the new entrant, had higher permeate conductivity and lower permeate flow than SWRO2 and SWRO3. The two established membranes had comparable conductivity values while SWRO3 had better permeate flow values.

Figure 4 and Figure 5 show permeate conductivity and salt rejection results of the three SWRO membranes, respectively, for the whole duration of the study. As expected, it took a couple of days before performance became stabilized as well as approaching manufacturer specification for all the membranes. The average salt rejection in SWRO1, SWRO2 and SWRO3 were 98.95%, 99.64% and 99.62%, respectively. As observed during start-up, performance was very similar in SWRO2 and SWRO3 while SWRO1 had higher permeate conductivity and lower salt rejection values consistently.

Permeate flow data, as illustrated in Figure 6, stabilized quickly for SWRO1. SWRO2 stabilized after approximately 50 days while steady decline was observed in SWRO3. The average flow values for SWRO1, SWRO2 and SWRO3 were 3.60 m³/hour, 4.10 m³/hour and 3.82 m³/hour, respectively. SWRO2 permeate flow, while considering higher membrane area, was better than the other two during the whole trial. Flow recovered in all the membranes after chemical cleanings, only to decline immediately.

To get better understanding of the performance of the individual elements, conductivity profiling was done periodically during the study. The comparative values for SWRO1, SWRO2 and SWRO3 are summarized in Figure 7. The observation was consistent as above with SWRO1 having consistently higher permeate conductivity than the established SWROs in all six elements. SWRO2 and SWRO3 had similar profile across the elements. Another observation from Figure 7 was that the conductivity increased from lead element (Element 1) to last element (Element 6). This was expected as elements treat progressively higher saline water as seawater passes through the elements in a pressure vessel.

The new membrane (SWRO1) had higher permeate conductivity and lower salt rejection values than the other two established products. Similar trend was observed as far as permeate flow was concerned. However, SWRO3 had the sharpest permeate flow decline in the first 170 days. When all three parameters were considered, SWRO2 performed the best during the pilot trial.