MODIFIED REVERSE OSMOSIS SYSTEM FOR TREATMENT OF PRODUCED WATERS
This report describes work performed during the second year of the project ''Modified reverse osmosis system for treatment of produced waters.'' We performed two series of reverse osmosis experiments using very thin bentonite clay membranes compacted to differing degrees. The first series of 10 experiments used NaCl solutions with membranes that ranged between 0.041 and 0.064mm in thickness. Our results showed compaction of such ultra-thin clay membranes to be problematic. The thickness of the membranes was exceeded by the dimensional variation in the machined experimental cell and this is believed to have resulted in local bypassing of the membrane with a resultant decrease in solute rejection efficiency. In two of the experiments, permeate flow was varied as a percentage of the total flow to investigate results of changing permeate flow on solute rejection. In one experiment, the permeate flow was varied between 2.4 and 10.3% of the total flow with no change in solute rejection. In another experiment, the permeate flow was varied between 24.6 and 52.5% of the total flow. In this experiment, the solute rejection rate decreased as the permeate occupied greater fractions of the total flow. This suggests a maximum solute rejection efficiency for these clay membranes for a permeate flow of between 10.3 and 24.6% of the total; flow. Solute rejection was found to decrease with increasing salt concentration and ranged between 62.9% and 19.7% for chloride and between 61.5 and 16.8% for sodium. Due to problems with the compaction procedure and potential membrane bypassing, these rejection rates are probably not the upper limit for NaCl rejection by bentonite membranes. The second series of four reverse osmosis experiments was conducted with a 0.057mm-thick bentonite membrane and dilutions of a produced water sample with an original TDS of 196,250 mg/l obtained from a facility near Loco Hill, New Mexico, operated by an independent. These experiments tested the separation efficiency of the bentonite membrane for each of the dilutions. We found that membrane efficiency decreased with increasing solute concentration and with increasing TDS. The rejection of SO{sub 4}{sup 2-} was greater than Cl{sup -}. This may be because the SO{sub 4}{sup 2-} concentration was much lower than the Cl{sup -} concentration in the waters tested. The cation rejection sequence varied with solute concentration and TDS. The solute rejection sequence for multi-component solutions is difficult to predict for synthetic membranes; it may not be simple for clay membranes either. The permeate flows in our experiments were 4.1 to 5.4% of the total flow. This suggests that very thin clay membranes may be useful for some separations. Work on development of a spiral-wound clay membrane module found that it is difficult to maintain compaction of the membrane if the membrane is rolled and then inserted in the outer tube. A different design was tried using a cylindrical clay membrane and this also proved difficult to assemble with adequate membrane compaction. The next step is to form the membrane in place using hydraulic pressure on a thin slurry of clay in either water or a nonpolar organic solvent such as ethanol. Technology transfer efforts included four manuscripts submitted to peer-reviewed journals, two abstracts, and chairing a session on clays as membranes at the Clay Minerals Society annual meeting.
- Research Organization:
- New Mexico Institute of Mining and Technology (US)
- Sponsoring Organization:
- (US)
- DOE Contract Number:
- FC26-00BC15326
- OSTI ID:
- 816388
- Resource Relation:
- Other Information: PBD: 15 Sep 2002
- Country of Publication:
- United States
- Language:
- English
Similar Records
MODIFIED REVERSE OSMOSIS SYSTEM FOR TREATMENT OF PRODUCED WATERS
Removal of cadmium and chromium from a pretreated wastewater with reverse osmosis. Master's thesis