Staged Thermal Desorption for DAC Water Management
OverviewAnalysisSolutions
Complete
·Feb 2, 2026The Core Insight
The thermodynamics of desorption differ more than adsorption — water releases at lower temperature
- Industry has focused on adsorption selectivity, but Clausius-Clapeyron says desorption is where the real opportunity lies. ΔH_ads(H₂O) ≈ 44 kJ/mol vs ΔH_ads(CO₂) ≈ 60-80 kJ/mol on amines.
- This means water's equilibrium pressure increases ~8x from 25°C to 65°C, while CO₂'s increases only ~3x.
- Temperature staging gives you free selectivity.
Viability
Solvable
- The thermodynamic basis for selectivity improvement is solid; the question is magnitude of improvement, not feasibility.
Key Decision
If you prioritize speed and low risk, start with temperature staging (concept-1) + methylsilane treatment (concept-3) on existing sorbents. If you have access to industrial waste heat and can accept 2-8 hour cycle times, humidity-swing (concept-7) offers the lowest energy pathway.
Solution Paths
01READY NOW
Temperature-Staged Desorption Protocol
Exploit ΔH difference via two-step heating (65°C then 110°C) — no material change needed, validation in weeks
02READY NOW
Methylsilane Surface Treatment
Post-synthesis hydrophobic coating on existing sorbents — proven chemistry, $2-5/kg cost increment
Recommendation
- If this were my project, I'd run temperature staging experiments next week.
- The math is unambiguous — ΔH differences of 20-35 kJ/mol translate to clear separation windows.
- You can get TGA-MS results in 2-4 weeks for $5-15K, and you'll know if you've got a 1.5x or 3x improvement with zero material changes.
- That's the cheapest, fastest path to meaningful progress.
- In parallel, I'd prepare a methylsilane treatment optimization matrix.
- Five loading levels, your current best sorbent, TGA-MS characterization.
- Another $5-10K, another 2-4 weeks.
- If temperature staging gives you 2x and methylsilane gives you 1.5x, they might stack to 3x — and you've solved your problem for under $30K with no new materials at scale.
- I would NOT start with the hierarchical silica/AMP system unless the simpler approaches plateau at <1.5x.
- That's a 6-12 month, $200K+ development program.
- It's the right path if you need systematic, defensible improvement — but only after you've exhausted the cheap wins.
- The humidity-swing angle is interesting but only if you have a specific co-location opportunity.
- If you're sitting next to a data center with cheap waste heat, model it out.
- If not, deprioritize.
- The biomimetic buried active site concept is intellectually beautiful but 18-36 months from anything deployable — file it under 'watch' unless you have R&D budget to burn.
- The one thing I'd do before any of this: make a phone call.
- Someone at Climeworks or Global Thermostat knows exactly what they do for selectivity and hasn't published it.
- A 30-minute conversation might save you 6 months of rediscovering known approaches.