Session: 09-18-01: Innovations in Storage, Recovery and Upgrade of Thermal Energy

Paper Number: 165970

Harnessing Thermal Gradients to Drive Reverse Osmosis Desalination via an Expanding Working Fluid 

Clean water is essential to life, however access to freshwater remains limited in many parts of the world. Moreover, demand is rapidly increasing due to population growth, increased per capita consumption, environmental pollution, and climate change. For these reasons, engineering solutions are needed to produce freshwater by desalinating salt water and treating wastewater for reuse. Low grade heat is abundant and available from a variety of renewable sources, and can be an important energy source for sustainable water desalination, treatment, and reuse.

Using heat to separate water and solutes is an old concept, and can be done as simply as by boiling water to collect condensate. In recent years, there has been renewed interest in developing thermally-driven water separation technologies because of their compatibility with a number of sustainability metrics. Namely, low grade heat can be plentiful and cheap, and is available from renewable sources such as solar radiation, ocean thermal gradients, geothermal, and waste heat sources. A number of processes have been developed to make use of thermal energy, including multi-effect distillation, multi-stage flash distillation, membrane distillation, solar stills, and others. While each have their own merits, all suffer from the fact that thermal separation is inherently energy intensive. This is because in all cases, separation is achieved by the phase change of water from liquid to vapor, which occurs preferentially relative to solutes and impurities in the feed. The heat of vaporization of seawater is approximately 667 kWh per m³ of desalinated water. Although some of this heat can be recovered during condensation, even highly efficient thermal systems consume on the order of 10-100 kWh/m³. 

In contrast, the energy requirements for reverse osmosis (RO) are much lower. This is one of the reasons that RO has grown to dominate the desalination industry, with global capacity now approaching 100 million m³/day. The basic RO process uses an applied pressure to force clean water across a highly selective semi-permeable membrane that rejects solutes and other impurities. As such, RO avoids the energy intensive step of vaporization. At its thermodynamic limit, approximately 0.8 kWh of energy is needed to produce 1 m³ of clean water permeate from seawater. Commercial RO desalination plants consume on the order of 1-10 kWh/m³. This is significantly less energy than used for thermal separation, but it should be noted that RO uses energy in the form of work, usually electricity. This is more costly than low-grade heat and therefore energy use remains one of the principal operating costs of RO.

Considering the respective advantages of both thermal separation (cheap, low-grade heat) and RO (inherent energy efficiency) raises the question of whether a process could be designed to take advantage of both. In this study, we investigate the thermodynamic limits of a novel thermally-driven reverse osmosis (TDRO) process and compare its performance to other water separation technologies. The proposed TDRO system consists of a piston set that uses the thermal expansion of a saturated working fluid to act on a saltwater feed water source to drive clean water permeate across a RO membrane. We show that the amount of heat needed is highly sensitive to feed concentration, recovery ratio, selection of the working fluid, operating temperature, and size of the working piston. A minimum thermal load of 20 kWh/m³ is achieved for 50 % recovery of a seawater feed source, when (1) water is selected as the working fluid, (2) the working fluid is operated at 247 °C, and (3) the piston is sized properly. This translates to a first law efficiency of 7.9%, a second law efficiency of 18.1%, and a gain output ratio above 32. At lower working temperatures of 110 °C, specific heat increases slightly to 24 kWh/m³ and the gain output ratio drops to 26. These metrics compare favorably with other thermal separation technologies, suggesting that TDRO can be an important means of harnessing low-grade heat for water purification.

Presenting Author: Jonathan Maisonneuve Oakland University

Presenting Author Biography: Prof. Maisonneuve is an Associate Professor in the Department of Mechanical Engineering at Oakland University (Rochester, MI, USA) and an Adjunct Professor in the Department of Bioresource Engineering at McGill University (Montreal, QC, Canada). He obtained his Ph.D. degree in electrical engineering from Concordia University. Prior to that he obtained his M.Eng. degree in building engineering also from Concordia University, and his B.Sc. degree in environmental science from McGill University. His work focuses on membrane processes for energy, water, and environment applications.

Authors:

Saber Khanmohammadi Oakland University
Sanjana Yagnambhatt Oakland University
Dan Delvescovo Oakland University
Jonathan Maisonneuve Oakland University