The Water–Food–Energy Nexus highlights the deep interdependence between water resources, food production, and energy generation. These systems are tightly connected—actions in one often have cascading effects on the others. For instance, evaporation ponds used for lithium extraction—critical to the electrification of society—consume vast areas of land and lose significant volumes of water to the atmosphere, reducing land available for agriculture and worsening water scarcity.
Water and energy are the pillars of civilisation. Yet over 70% of global freshwater is used for agriculture, a sector facing immense pressure from population growth and climate change. While distillation has purified water for millennia, it’s energy-intensive. Reverse osmosis (RO), now the dominant desalination technology, is more energy-efficient but still too costly for widespread agricultural use. Meanwhile, evaporation ponds and solvent-based methods remain common for brine treatment, despite their inefficiency and environmental footprint.
Membrane-based systems degrade, chemical methods are hazardous, and phase-change processes are energy-hungry. Despite six decades of commercial use, desalination meets less than 0.5% of global freshwater needs. With over a billion people affected by water scarcity, we urgently need new, scalable, and sustainable water technologies.
What makes an ideal water treatment technology? It should (1) run on low-cost or free energy—ideally waste heat below 80°C, (2) operate in a single liquid phase—avoiding energy-intensive evaporation, (3) use no membranes or chemicals—cutting cost and complexity, and (4) be scalable and manufacturable, for example, via injection moulding or 3D printing.
Multichannel thermodiffusion, a desalination method operating in the liquid phase, avoids evaporation, membranes, or chemical additives
At the Australian National University (ANU), we developed multichannel thermodiffusion—the first thermal desalination method fully operating in the liquid phase. Based on the Soret effect (species movement under temperature gradients), this process avoids evaporation, membranes, or chemical additives. It's simple, scalable, and energy-efficient.
Last year, we demonstrated lab-scale desalination devices at ANU, handling 36 mL/h with a 2,000 ppm concentration drop. Now, our spin-off Soret Technologies Pty Ltd is scaling up to modular systems processing over 100 L/h with salinity shifts in brine concentration exceeding 35,000 ppm. Our goal is full-scale facilities handling 10,000+ m³/day.
Thermodiffusion performs exceptionally well across a wide salinity range—from typical reverse osmosis (RO) brine levels (~70,000 ppm) to near saturation (>200,000 ppm). It surpasses evaporation ponds both in energy efficiency and cost-effectiveness, even when heat is supplied using electricity. Unlike evaporation-based methods, thermodiffusive brine concentration preserves water instead of losing it to the atmosphere, making it ideal for water reuse in agriculture and industry. It is also well suited for resource recovery from brines and treatment of produced water in the oil and gas sector. When powered by waste or low-grade heat, its operational costs drop significantly, offering a sustainable and economical alternative to conventional methods.
There’s also a timely opportunity: data centres, surging due to AI, cloud computing, and crypto, produce vast amounts of low-grade waste heat. Instead of letting it dissipate, we’re exploring ways to power multichannel thermodiffusion with this thermal energy—using waste heat for freshwater production, brine concentration, and even extraction of critical minerals.
As climate change and resource stress escalate, the future of the Water–Food–Energy Nexus lies in smart, system-wide optimisation. Multichannel thermodiffusion is a transformative step in that direction.