Design and operating characteristics of pilot scale reverse osmosis plants

Design and operating characteristics are analyzed for two pilot scale reverse osmosis plants. The first plant is installed in Sharjah to desalinate brackish water for a small community. The second plant is installed in Qatar to test the feasibility of desalting high-salinity seawater for irrigation purposes. Analysis is performed using a semiempirical model, which requires knowledge of membrane salt reject and permeate recovery. Further analysis is made by using a permeability model, which is capable of predicting the required membrane area. Predictions of the two models are in good agreement with the available field data. Both models can provide the plant engineers with a useful tool to assess plant performance and analyze deviations from original design conditions.

Table 7 Simulation parameters for the permeability model for the Qatar SWRO plant Simulation parameter Value Membrane area (m2) Salt permeability (kg/s ppm m2) Water permeability (m3/s kPa m2) Permeate pressure (kPa) System temperature (°C) Average molecular weight of salt (kg/kmole) Pressure dropper module (kPa) Analysis of the design and operating characteristics is performed for two types of RO plants. The first plant is located in Sharjah and operates on low-salinity brackish water. The second plant is located in Qatar and is designed to desalinate high-salinity seawater for irrigation purposes. The analysis is performed using two models. The first is semi-empirical, which requires knowledge of the membrane recovery ratio and the membrane salt rejection. The second is the permeability model, which requires definition of the water and salt permeability coefficients. Predictions of 0.05 1 × 10−8 7 × 10−7 Table 8 Profiles of the Qatar SWRO plant predicted by the permeability model Module number Qf (m /s) Qb (m3/s) Qp (m3/s) Xf (ppm) Xp (ppm) Xb (ppm) X (ppm) pf (kPa) pb (kPa) Δpm (kPa) Δpm (kPa) 1.42E–03 1.30E–03 1.15E–04 58,000 230.9 63,101.2 60,442.7 4523.0 1.30E–03 1.20E–03 9.84E–05 63,101.2 292.4 68,225.7 65,563.0 4903.0 1.20E–03 1.12E–03 8.25E–05 68,225.7 376.6 73,216.3 70,632.5 5277.5 1.12E–03 1.05E–03 6.73E–05 73,216.3 490.7 77,860.0 75,466.3 5632.1 1.05E–03 9.99E–04 5.35E–05 77,860.0 655.2 82,060.8 79,904.8 5953.1 9.99E–04 9.58E–04 4.16E–05 82,060.8 882.0 85,596.3 83,790.9 6228.0 RR SR 0.0809 0.9960 0.0756 0.9954 0.0685 0.9945 0.0600 0.9933 0.0508 0.9916 0.0416 0.9893 the two models are in good agreement with the collected field data for the two plants. Agreement is found for predictions of the total production rate, product salinity, overall recovery ratio, overall salt rejection, and the required number of membrane modules and pressure vessels. The semi-empirical model is rather simple and does not require iterative solution. It provides flow rate, concentration, and pressure profiles. However, its predictions do not include an important design parameter, which is the module membrane area. This important design parameter is hidden in the module recovery ratio and salt rejection ratios. On the other hand, the permeability model includes explicitly the module permeation area, which appears in the flux equations for water and salt. This knowledge requires simultaneous definition of the permeability coefficients for water and salt. Both models provide useful tools to analyze plant data and to understand variations in system performance.