Search

An NH3/CO2 forward osmosis (FO) membrane brine concentrator (MBC) pilot was tested in the desalination of frac flowback and produced waters from natural gas extraction operations in the Marcellus shale region. The average concentration of these waters was 73,000 ± 4200 mg/L total dissolved solids (TDS), with an average hardness of 17,000 ± 3000 mg/L as CaCO3. Pretreatment included chemical softening, media filtration, activated carbon, and cartridge filtration. Average pilot performance characteristics were: system recovery of 64 ± 2.2%, nominal water flux of 2.6 ± 0.12 L/m2-h, concentrated brine concentration of 180,000 ± 19,000 mg/L TDS, and product water with 300 ± 115 mg/L TDS. The thermal energy required by the FO MBC pilot, when operated within the efficient flow specification of the draw solution recycling system, averaged 275 ± 12 kWhth/m3 of product water, approximately 57% less thermal energy input than that estimated for a conventional evaporator operated in a comparable single stage, non-mechanical vapor compression (MVC) configuration. In an MVC configuration, which uses electrical rather than thermal energy, modeling indicates that the FO MBC process will require 42% less electrical energy than a conventional forced circulation MVC evaporator.

The energy requirements of ammonia–carbon dioxide forward osmosis (FO) desalination are predicted by the use of chemical process modeling software (HYSYS). The FO process is modeled using single or multiple distillation columns to separate draw solution solutes from the product water for solute recycling within the FO system. Thermal and electrical energy requirements of the process are calculated, as well as a combined term for equivalent electrical work. The results of the simulations are compared to the energy requirements of current desalination technologies. Energy savings of FO compared to current technologies, on an equivalent work basis, are projected to range from 72% to 85%. Forward osmosis desalination is in an early stage of its development, and several areas of future work promise opportunities to improve its energy utilization and cost.

A novel forward (direct) osmosis (FO) desalination process is presented. The process uses an ammonium bicarbonate draw solution to extract water from a saline feed water across a semi-permeable polymeric membrane. Very large osmotic pressures generated by the highly soluble ammonium bicarbonate draw solution yield high water fluxes and can result in very high feed water recoveries. Upon moderate heating, ammonium bicarbonate decomposes into ammonia and carbon dioxide gases that can be separated and recycled as draw solutes, leaving the fresh product water. Experiments with a laboratory-scale FO unit utilizing a flat sheet cellulose tri-acetate membrane demonstrated high product water flux and relatively high salt rejection. The results further revealed that reverse osmosis (RO) membranes are not suitable for the FO process because of relatively low product water fluxes attributed to severe internal concentration polarization in the porous support and fabric layers of the RO membrane.

article i nfo Osmotically driven membrane processes (ODMPs) such as forward osmosis (FO) and pressure retarded osmo- sis (PRO) are extensively investigated for utilization in a broad range of applications. In ODMPs, the operating conditions and membrane properties play more critical roles in mass transport and process performance than in pressure-driven membrane processes. Search of the literature reveals that ODMP membranes, especially newly developed ones, are tested under different temperatures, draw solution compositions and concentra- tions, flow rates, and pressures. In order to compare different membranes, it is important to develop standard protocols for testing of membranes for ODMPs. In this article we present a standard methodology for testing of ODMP membranes based on experience gained and operating conditions used in FO and PRO studies in recent years. A round-robin testing of two commercial membranes in seven independent laboratories revealed that water flux and membrane permeability coefficients were similar when participants performed the experi- ments and calculations using the same protocols. The thin film composite polyamide membrane exhibited higher water and salt permeability than the asymmetric cellulose-based membrane, but results with the high permeability thin-film composite membrane were more scattered. While salt rejection results in RO mode were relatively similar, salt permeability coefficients for both membranes in FO mode were more varied. Results suggest that high permeability ODMP membranes should be tested at lower hydraulic pressure in RO mode and that RO testing be conducted with the same membrane sample used for testing in FO mode.