It is not easy to store liquid water in Arctic regions. The cold temperatures require the installation of heaters to prevent it from freezing, permafrost makes burying supply pipelines difficult, and freshwater sources may be scarce. As a result, and perhaps somewhat counterintuitively, freshwater is scarce in many settlements above the Arctic Circle, which leads to a host of different health problems. In addition, even in places where freshwater is available, it is often stored in tanks, sometimes centralised, and the required heating relies on the combustion of fossil fuels, making this a source of CO2 emissions.
There are hence quite a few problems that need to be solved to improve the water supply in those regions. We started out by wishing to address only one, albeit an important one: Preventing the water from freezing without the use of fossil fuels, or indeed any heating at all. Earlier research at ETH Zurich and the University of Zurich showed that this is possible if the water is mixed with a commercially available lipid. This produces a mesophase that remains in a liquid-like state down to temperatures as low as -120°C. Later, upon an innovative initiative by Sirin Orbital Systems AG, a Zurich-based company for space applications, we explored the concept to design water storage and supply systems for a manned lunar base. The idea to also use this for terrestrial applications was born out of this work, bringing potential space tech to a very earthly environment.
Unfortunately, the situation was not as easy as outlined. While producing the mesophase is a rather straightforward endeavour, extracting the water from the mixture again for use is rather challenging. We successfully used distillation to achieve the extraction in earlier work. As this is a rather energy-hungry procedure, we started looking for alternatives. This led us to testing two additional methods, pervaporation and nanofiltration, in this project. We fought with clogging membranes, malfunctioning pumps, faulty electric motors, leakages, and procurement issues. But finally, we got our results. We could also measure energy use, on which data is very scarce. In essence, nanofiltration did not work, no matter how hard we tried. Pervaporation worked well, i.e. we successfully extracted water from the mixture. However, precise process design and hardware selection had a major impact on energy use, and there were challenges with the membrane lifetimes. Distillation as a well-established baseline also worked well and required less energy than pervaporation, but it generally leads to bulkier installations and is less suited to highly viscous fluids like ours.
Having the energy consumption in hand, we then set out to see how our technology may compare with existing solutions. This turned out to be more difficult than anticipated, mainly due to the lack of detailed data on energy use of water storage and supply systems in cold regions. Finally, we found North Pole, a small city in Alaska, US. Their water distribution network was significantly expanded after the local groundwater became contaminated by a nearby industrial facility and the new system is rather well documented. A top-level representation is shown in Figure 1. A system based on using the mesophase is shown in Figure 2. The main difference is that our system does not require heating for storage or transport in the distribution network, while the reference system from Figure 1 needs to compensate for these. However, the mesophase is significantly more viscous than water itself, and so the electric power consumption required to drive the pumps is higher. As a result, system performance naturally depends on environmental parameters. In our scenario, the central storage tank was replenished at regular intervals, as it would be in off-the-grid settlements, and we assessed the situation between these intervals.
A good measure of the efficiency of a system performance is the energy use per household per day. This is plotted in Figure 3 for the system from Figure 1 for different daily tap-offs. For example, 0.5% indicates that 0.5% of the water from the storage tank is extracted daily. For low tap-offs, more energy is required to maintain the water in the storage tank at sufficiently high temperatures to avoid freezing. In addition, the energy consumption for the newly proposed process is shown in the Figure 3. Since no heating is required, energy needs are driven by the energy requirements of mesophase transport in the hydraulic networks and the energy required to extract water from the mesophase either by distillation or pervaporation. The data shows that for reasonably low ambient temperatures and low tap-off rates, the newly proposed system leads to improvements in energy efficiency.
While these results are encouraging, further work is required to select better membranes, improve membrane lifetimes, further reduce energy use, and demonstrate the concept on a small-scale prototype. So, there’s still a long way to go, but we’re on it.
The team consisted of Dr. Marius Banica, Dr. David Rütti, Dr. Reinhard Berger, Ms. Monika Blaser, Mr. Gianluca Manfioletti, and Mr. Marco Randon, all at the Zurich University of Applied Sciences in Winterthur.
Marius Banica is Principal Lecturer and Head of Research at ZHAW Zurich University of Applied Sciences in Switzerland. His project took place in 2025 and 2026 with financial support from the SPI Technogrants funding instrument.
Header illustration: Microsoft Copilot-generated image by grantee, all rights reserved