By @Ocean · October 20, 2026
Mix two harmless solids — a salt cousin of a B vitamin and the urea found in fertilizer — and they melt into a liquid at room temperature. That liquid can pull carbon dioxide out of a gas stream, release it again with gentle heat, and do it over and over. This is a deep eutectic solvent (DES), and it is one of the most promising "green" ways to scrub air.
Air scrubbing matters at every scale. Power plants and factories need to strip CO₂, sulfur dioxide and volatile organic compounds (VOCs) from their exhaust. Greenhouses, buildings and, eventually, sealed sea and space habitats need to keep the air inside breathable. Today most heavy-duty gas scrubbing uses water-based amine solutions, which work but are corrosive, volatile and energy-hungry to regenerate. DESs offer a cheaper, gentler, more designable alternative — with real trade-offs this article walks through.
What a deep eutectic solvent is
A DES is a liquid made by mixing a hydrogen-bond acceptor (HBA) with a hydrogen-bond donor (HBD) in the right ratio. The two partners hydrogen-bond so strongly to each other that neither can lock into its own crystal, so the mixture melts far below either ingredient.
The founding example came from Andrew Abbott's group at Leicester in 2003. Choline chloride melts at 302 °C and urea at 133 °C, but a 1:2 mixture freezes at only 12 °C, making it usable as a room-temperature solvent. That mixture is now nicknamed "reline."
The recipe space is huge. Common ingredients include:
- Acceptors (HBAs): choline chloride, betaine, quaternary ammonium or phosphonium salts, and "superbases" such as DBN or DBU
- Donors (HBDs): urea, glycerol, ethylene glycol, levulinic acid, amino acids, sugars, organic acids
When every ingredient comes from nature — sugars, organic acids, amino acids, choline — the result is called a natural deep eutectic solvent (NADES). Chemists sometimes reserve the word "deep" for mixtures that melt below what an ideal mixture would predict, but in practice most researchers use DES for any such eutectic liquid.
Why this matters for air: DESs behave much like ionic liquids — practically non-volatile and non-flammable — yet they are cheaper, easier to make, generally less toxic and often biodegradable. A liquid that won't evaporate into the air you're cleaning is a big deal for a scrubber.
How a DES grabs a gas
A DES captures a gas in one of two ways, and the choice decides almost everything about the scrubber.
- Physical absorption. The gas dissolves into the hydrogen-bond network, held by weak forces. Plain DESs like reline or choline chloride with levulinic acid work this way. They release the gas easily, so regeneration is cheap, but they need high pressure to load up — fine for a pressurized industrial stream, weak for ordinary air.
- Chemical absorption. One ingredient reacts with the gas. Amine-based DESs form carbamates with CO₂, and superbase DESs (built on molecules like DBN) activate an alcohol partner that locks CO₂ in as a carbonate. Amine-functionalized DESs capture far more at low CO₂ concentrations, but tend to be very viscous.
Because the two ingredients can be swapped, a DES can be tuned for a specific target gas. Reported results span the major air pollutants:
| Target gas | Example DES | Reported result |
|---|---|---|
| CO₂ from open air (~400 ppm) | TBN superbase + benzyl alcohol | 0.154 g CO₂ per g from >90% humid air; releases CO₂ at 70 °C |
| CO₂ (concentrated) | DBN + 2-imidazolidone, 2:1 | 1.75 mol CO₂ per mol DES at 45 °C, 1 bar |
| CO₂ (concentrated) | Choline chloride + levulinic acid | Regenerates best at 80 °C; moderate CO₂/N₂ selectivity |
| SO₂ | EmimCl + ethylene glycol, 2:1 | 1.15 g SO₂ per g at 20 °C, 1 atm; reversible |
| VOCs (e.g. toluene, 1-decene) | Tetrabutylammonium bromide + decanoic acid | Up to 60,000× more 1-decene than water; no capacity loss over cycles |
| NH₃, H₂S, NOₓ | Various tailored DESs | Reviewed as promising; selective NH₃ and SO₂ capture from mixed streams reported |
Numbers come from different labs under different conditions, so read them as "what's possible," not a head-to-head ranking.
How a DES scrubber works
A DES scrubber is a loop: the same liquid soaks up gas in one vessel and gives it back in another, then returns to soak up more.
- Absorb. Dirty air or exhaust bubbles up through the DES in an absorber column, or flows past it across a membrane. The target gas dissolves or reacts; scrubbed air leaves the top.
- Move. The gas-loaded liquid is pumped to a regenerator.
- Release. Warming the liquid, or dropping the pressure, breaks the weak bonds and drives the gas out as a concentrated stream. For several DESs that takes only about 70–80 °C, well within reach of waste heat or solar thermal.
- Return. The clean DES is cooled and sent back to the absorber.
This is the same architecture industry already uses for amine scrubbing, which means DESs could, in principle, slot into existing equipment rather than requiring a new kind of plant.
DES versus the incumbent: amine scrubbing
The benchmark is 30 wt% monoethanolamine (MEA) in water, the workhorse of industrial CO₂ capture. DESs beat it on safety and corrosion, and can beat it on regeneration energy, but lose on viscosity and maturity.
| Property | Aqueous MEA (30 wt%) | Deep eutectic solvents |
|---|---|---|
| Volatility | Solvent and water evaporate; amine losses | Practically non-volatile |
| Corrosion | Strongly corrosive to steel | ChCl + levulinic acid: up to 92% lower corrosion rate than 30 wt% MEA |
| Regeneration heat | High; water's heat capacity and vaporization dominate | Amine DESs have lower specific heat than 30 wt% MEA; some release CO₂ at 70–80 °C |
| Viscosity | Low, water-like | Often high, slowing gas uptake and pumping |
| Toxicity and sourcing | Petrochemical; degrades into harmful by-products | Often food-grade or bio-based, but some superbase DESs proved unexpectedly toxic |
| Maturity | Decades of commercial plants | Lab and pilot scale |
There's a catch in the energy story. One process model found MEA-based DESs need much less desorption energy than classic MEA, but at low CO₂ concentrations they need large solvent flows and costly compression. A solvent is only as good as the whole process wrapped around it.
What still stands in the way
DESs are promising, not proven at scale. Recent reviews converge on the same short list of problems, calling for better stability, lower viscosity and improved regeneration.
- Viscosity. Thick liquids absorb gas slowly and cost energy to pump. Workarounds include adding a little water, heating the absorber, or soaking the DES into a porous solid.
- Water is double-edged. Small amounts thin the liquid and can help, but water acts as an anti-solvent for CO₂ in reline. Real air is humid, so moisture-tolerant formulas matter.
- Durability. The best direct-air-capture DES so far kept 74% of its capacity after 50 cycles and 50% after 100 — impressive for a lab, short of what a plant running for years needs.
- Toxicity is not automatic. "Green" depends on the ingredients. Choline-and-urea is benign; some superbase blends are not.
- Understanding. Capture mechanisms are still incompletely understood, so much design remains trial and error, though machine-learning screening is speeding it up.
- Hybrid forms. One promising fix is immobilizing a DES inside a solid framework. A 2026 study loaded a DES into the metal-organic framework ZIF-8 and saw stable capture under humid conditions over 100 cycles, combining liquid chemistry with solid handling.
Why stewards should care
A scrubber liquid made from food-grade ingredients, that doesn't evaporate and can be reused with low-grade heat, fits a stewardship mindset: clean the air without poisoning it in the process.
- Industry and cities. The nearest-term role is cleaning smokestacks and process gases — CO₂, SO₂ and VOCs — where today's solvents are corrosive or toxic.
- Closing the carbon loop. Captured CO₂ doesn't have to be waste. Researchers have paired DES capture with releasing the CO₂ into microalgae cultures that grow on it, and others use DESs to convert CO₂ into useful chemicals.
- Greenhouses and buildings. Low-temperature regeneration means waste heat, or even solar heat, could recycle the solvent — a natural fit for indoor farms that want CO₂ and homes that want less of it.
- Sea and space habitats. Sealed habitats must strip the CO₂ their crews exhale, continuously, for years. Submarines have long relied on amine scrubbing. A non-volatile, low-toxicity liquid that regenerates at 70–80 °C is exactly the kind of property set a future habitat designer would want — though no DES has yet been flight- or sea-qualified, and the cycling-lifetime problem above would need solving first.
The bigger lesson is design philosophy. A DES is built from parts you choose, so you can ask not just "will it capture?" but "is every ingredient safe, renewable and recoverable?" That's the question stewards should ask of every technology.
Sources and further reading
- Abbott et al. (2003), Novel solvent properties of choline chloride/urea mixtures, Chem. Commun.
- Deep Eutectic Solvents: Properties and Applications in CO₂ Separation (review)
- Advances in Deep Eutectic Solvents for CO₂ Capture (2025 review)
- CO₂ capture and conversion using DESs: recent progress and outlooks (2025)
- Eshaghi Gorji et al. (2025), Direct Air Capture with a superbase DES, Environ. Sci. Technol.
- Jiang et al. (2019), Superbase/acylamido-based DESs for CO₂ absorption
- Aboshatta & Magueijo (2021), CO₂ absorption and desorption by ChCl/levulinic acid DESs
- Aboshatta & Magueijo (2022), corrosion and carbon capture of ChCl/levulinic acid DESs
- Yang et al. (2017), Efficient absorption of SO₂ by EmimCl-EG DESs
- DESs as absorbents for VOC mixtures in static and dynamic processes
- Utilization of DESs to reduce release of hazardous gases: a critical review (2021)
- Analysis of operating conditions for CO₂ capture using DESs
- AIChE 2019, regeneration energy of amine-based DESs
- Immobilized DES–ZIF-8 composites for CO₂ capture (2026)
- Enhancing CO₂ Capture Utilizing DESs (2025 preprint)
