(486d) Forward Osmosis Using Draw Solutions Manifesting Liquid-Liquid Phase Separation

Over the past decade, Forward Osmosis (FO) technology has emerged as a promising alternative to conventional Reverse Osmosis (RO) to produce fresh water through desalination of seawater. In FO, water is extracted from a saline water source (such as seawater) by a draw solution having a higher osmotic pressure than the saline water; the diluted draw solution is then split into a stream of concentrated draw solution and a stream of essentially pure water in process steps collectively referred to as regeneration. Draw agents whose regeneration is aided via liquid-liquid phase separation have gained much attention in recent years [1-4]. Specifically, aqueous solutions of these draw agents manifest either a lower critical solution temperature (LCST) phase transition or upper critical solution temperature or both. With draw solutions manifesting LCST, the FO step is typically performed at a temperature (T) below (or near) the LCST; for medium to high concentration of the draw agents, the osmotic pressure of the draw solution decreases with increasing T. Modest heating of the diluted draw solution facilitates regeneration by causing liquid-liquid phase separation yielding a water-rich phase and a second phase rich in the draw agent.

In this presentation, we describe the results from an experimental study where mixtures of two different glycol ethers (GE), tripropylene glycol methyl ether (TPM; MW=206 Da) and tripropylene glycol n-butyl ether (TPnB; MW=248 Da), were used as draw agents. These are larger in size than the GEs used in previous studies [2].

Cloud point temperatures of aqueous solutions of these GE mixtures were determined. By adjusting the relative amounts of TPM and TPnB, one can tune the cloud point temperature and the LCST. We also determined the activity of water (in the 288-323K range) at various draw solution compositions, using an Aqualab TDL water activity meter. At high GE wt% and temperature modestly below the cloud point T, the osmotic pressure of the draw solution is much higher than that of seawater, enabling their use as draw agents in FO.

The kinematic viscosities of these draw solution mixtures at different temperatures were found using a micro-Ubbelohde viscometer. Through NMR measurements, the self-diffusion coefficients of the GEs in the draw solutions at different temperatures were determined.

FO experiments were conducted using commercial CTA embedded support FO membrane (Fluid Technology Solutions, Inc., USA) to assess water flux, reverse flux of the draw agent, and the effect of orientation of the membrane (support layer facing the draw solution â?? the so-called FO mode; and, its inverse â?? the so-called Pressure Retarded Osmosis (PRO) mode). These experiments were performed with fresh water on one side and draw solution on the other side of the membrane. As expected [5], the water flux was higher in the PRO mode than in the FO mode. While the initial water flux was high (~7 lit/m²h ), it dropped rapidly to ~ 2 lit/m²h indicating appreciable membrane fouling. Even after thorough washing, the original high flux state could not be recovered.

In the PRO mode, increasing the circulation rate of the draw solution led to measurably large increase in the water flux, clearly revealing the effect of concentration polarization. FO experiments were done at three different temperatures; water flux increased when the operating temperature was increased from 295K to 305K, but decreased when the operating temperature was increased to 315K. This can be rationalized qualitatively as a consequence of the temperature dependence of the osmotic pressure, kinematic viscosity and diffusivity.

Appreciable reverse flux was recorded in the FO experiments (4.91 gr/m²h in FO mode and 2.90 gr/m²h in PRO mode), indicating that molecular weight in the range of 200-250 Da is not sufficiently large to suppress reverse flux. FO experiments performed with aqueous solution of polypropylene glycol (PPG 425; average MW=425 Da) showed lower reverse fluxes (2.53 gr/m²h in FO mode and 1.74 gr/m²h in PRO mode), suggesting that even higher molecular weights are required to suppress reverse flux of draw agent.

In summary, as a result of reverse flux and membrane fouling, using aqueous solutions of glycol ethers directly as draw solutions does not appear promising. Their indirect use, as in the process described by [6], which uses an ionic salt as a draw agent and employs an aqueous polymer solution to recover water from the diluted draw solution, appears to be more realistic.

Keywords: forward osmosis, lower critical solution temperature, glycol ethers, liquid-liquid phase separation

References:

1. Mok, Y., et al., Circulatory osmotic desalination driven by a mild temperature gradient based on lower critical solution temperature (LCST) phase transition materials. Physical Chemistry Chemical Physics, 2013. 15(44): p. 19510-19517.

2. Nakayama, D., et al., Lower critical solution temperature (LCST) phase separation of glycol ethers for forward osmotic control. Physical Chemistry Chemical Physics, 2014. 16(11): p. 5319-5325.

3. Noh, M., et al., Novel lower critical solution temperature phase transition materials effectively control osmosis by mild temperature changes. Chemical Communications, 2012. 48(32): p. 3845-3847.

4. Zhao, D., et al., Thermoresponsive copolymer-based draw solution for seawater desalination in a combined process of forward osmosis and membrane distillation. Desalination, 2014. 348: p. 26-32.

5. Cath, T.Y., et al., Standard methodology for evaluating membrane performance in osmotically driven membrane processes. Desalination, 2013. 312: p. 31-38.

6. Rajagopalan, N. and V.A. Patel, Solvent removal and recovery from inorganic and organic solutions, 2014, Patent US8852436.