Effect of operating parameters on boron removal from seawater using membrane distillation process

Theoretical and experimental studies of direct contact membrane distillation (DCMD) using commercially flat sheet hydrophobic polyvinylidene fluoride (PVDF) membrane were investigated to remove boron from simulated and natural seawater. Mass transfer mechanism and theoretical flux were determined using the dusty-gas model (DGM). Mass transfer analysis showed that the combined Knudsen-molecular model is considered to be the dominant mechanism which describes the transport of water vapor through the membrane pores. Then, the effect of various operating parameters such as feed temperature, boron concentration, salt concentration and feed solution pH on DCMD performance was investigated. Experimental runs showed that the permeate flux was enhanced exponentially with feed temperature, and maximum permeate flux of 27.5 kg/m2∙h was obtained at 74 °C. The permeate flux decreases slightly with feed boron and salt concentration. Feed solution pH had no significant effect on DCMD performance. DCMD process can produce water with high boron rejection of more than 90% even at feed concentration as high as 200 mg/L. Natural seawater, containing 5.37 mg/L of boron, was treated by DCMD process. During 18 h of working period, the permeate flux had showed a reduction of about 15.75%, and permeate boron was kept below 0.47 mg/L. The decreasing of contact angle value can explain the performance decline of DCMD process. © 2015 Elsevier B.V. All rights reserved.

DCMD process using PVDF membrane is a promising alternative technology able to remove boron contaminant from synthetic water and seawater. In this study, the applied methodology consists to investigate the effect of operating parameters such as temperature, boron concentration, salt concentration and pH solution on DCMD performance (permeate flux and boron rejection). The evaluation of the transfer model and their comparison with the experimental values has been realized. Mass transfer analysis showed that the Knudsen-molecular diffusion is the dominant mechanism able to describe the transport of water-vapor through the PVDF membrane. Feed temperature is considered to be the most significant parameter on DCMD permeate flux. At 74 °C, the permeate flux reached 27.5 kg/m2∙h and boron rejection efficiency closed 90.5%. Feed salt and boron concentration have slight negative effect on permeate flux and no marked influence on boron rejection efficiency. In the range of feed pH solution (3–11), the permeate flux and boron rejection had no marked variation, which present a great advantage compared to RO process. DCMD application of seawater, containing 5.37 mg/L of boron, showed slightly decrease of permeate flux from 11.87 to 10 kg/m2∙ h and boron concentration in the permeate was kept below 0.47 mg/L. The decline of the performance of DCMD process has been explained by the reduction of the surface CA measurement. For long working period, it is recommended to add a simple pretreatment step in order to get a stable DCMD performance for boron removal and seawater desalination. Acknowledgments The authors wish to thank Nesrine Hmidi for the constant support during this work.