water-purification

Water purification (perchlorate remediation)

Process Semi-native water
TRL Mars
Energy intensity
Required by
0
Requires
4

Cleans raw Martian water — from ice, atmosphere, or brine — to the purity each use demands: removing perchlorate (toxic, and a catalyst poison), chlorides, sulfates, and particulates. The defining Mars step is perchlorate destruction, by biological reduction (perchlorate-respiring bacteria) or catalytic/UV methods, converting ClO₄⁻ to harmless chloride — which doubles as the chlor-alkali feed. Almost every water-using node in the tree depends on this one; it is the keystone of the water economy.

Last reviewed: 2026-06-14

Governing equations

Perchlorate reduction — the keystone reaction. Perchlorate-respiring bacteria (or a catalyst) strip the oxygens, yielding chloride and water. The toxic contaminant becomes the chlor-alkali brine feed; nothing is wasted. [1]

The stepwise enzymatic pathway (perchlorate → chlorate → chlorite → chloride + O₂) — the chlorite-dismutase step releasing oxygen is what makes microbial perchlorate respiration energetically favorable. [1]

Purification targets are use-specific: potable water has a strict health limit, but electrolyzer/synthesis feed must be even cleaner because trace perchlorate irreversibly poisons catalysts over weeks. [2]

\text{train: filtration} \to \text{ClO_4^- destruction} \to \text{ion exchange / RO} \to \text{polish}

The multi-barrier treatment train: remove solids, destroy perchlorate, strip remaining ions (ion exchange or reverse osmosis), then polish to the required grade — the standard defense-in-depth of water treatment. [3]

Key constants & quantities

Symbol Value Units Conditions Description
Regolith perchlorate 0.4–0.6 wt% ClO₄⁻ Perchlorate abundance in Martian regolith (Phoenix/Curiosity) — the contaminant load that every regolith-sourced water carries.[4]
Potable ClO₄⁻ limit 0.001–0.015 mg/L (sub-ppm) Health-based perchlorate limit for drinking water — the strict spec the potable train must hit (orders below the regolith load).[2]
Feedwater purity (electrolysis) 0–1 µS/cm (ASTM Type I/II) Electrolyzer/synthesis feed must be ultrapure — even trace perchlorate poisons catalysts, so this grade is stricter than potable.[5]
Bioreactor perchlorate removal 95–99.9 % Perchlorate destruction efficiency of an established perchlorate-reducing bioreactor — high and reliable once the culture is mature.[1]
Purification energy 0.5–4 kWh / m³ Energy to treat water to use-grade — RO/ion exchange dominate; far below the electrolysis or thermal energy downstream.[3]
O₂ co-product 0.45 stoichiometric kg O₂ / kg ClO₄⁻ reduced Oxygen released by perchlorate reduction (4 O per ClO₄⁻) — a minor but free oxygen bonus from detoxification.[1]

Operating envelope

ParameterRangeUnitsSource
Bioreactor temperature 20 – 40 °C [1]
Feed perchlorate 100 – 6000 mg/L (raw regolith water) [4]
Potable output ClO₄⁻ 0 – 0.015 mg/L [2]
Electrolysis-feed conductivity 0 – 1 µS/cm [5]
Treatment energy 0.5 – 4 kWh/m³ [3]

Mass balance

Basis: 1 m³ raw regolith-sourced water purified (illustrative)

Inputs

Raw water (perchlorate-laden) 1,000 kg [4]
Hydrogen (bioreduction) 0.1 kg [1]
Electrical energy 2 kWh [3]
Ion-exchange resin / RO membrane 1 consumable [3]
  • Raw water (perchlorate-laden): From ice mining, atmospheric capture, or brine; carries perchlorate, chloride, sulfate, fines.
  • Hydrogen (bioreduction): Electron donor for microbial/catalytic perchlorate reduction; from the H₂ bus.
  • Electrical energy: Pumping, RO/ion exchange, UV — use-grade dependent.
  • Ion-exchange resin / RO membrane: Periodic replacement; import until local polymer chemistry supplies media.

Outputs

Purified water 990 kg [3]
Recovered brine (chloride) 10 kg [6]
Oxygen (from ClO₄⁻ reduction) 0.2 kg [1]
  • Purified water: To electrolysis, potable, hydroponics, leaching — graded per use.
  • Recovered brine (chloride): Reduced-perchlorate chloride brine → chlor-alkali feed. Contaminant becomes feedstock.
  • Oxygen (from ClO₄⁻ reduction): Minor free O₂ bonus.
TRL · Earth
9/ 9
TRL · Mars
4/ 9
Water treatment is mature municipal/industrial engineering, and perchlorate bioremediation is established practice at terrestrial contaminated sites (groundwater, rocket-fuel cleanup). Spacecraft water processing (ISS WRS) handles crew water at TRL 9. The Mars gap is the integrated regolith-water train at the high perchlorate loads Mars presents — studied, not yet operated off-Earth.[1]
Energy budget
2 kWhe / m³ water purified to use-grade (perchlorate destruction + ion removal) [3]

Purification energy is small against what the clean water then enables — electrolysis alone costs ~50 kWh/kg H₂. But its reliability is disproportionately critical: a purification failure silently poisons every downstream catalyst and crop.

Variants & trade-offs

Biological perchlorate reduction (the Mars baseline)

[1]

Perchlorate-respiring bacteria in a bioreactor reduce ClO₄⁻ to chloride at ambient temperature, using H₂ or organics as electron donor.

Materials: Perchlorate-reducing bacterial culture · Bioreactor + H₂/organic donor · Downstream filtration
  • Ambient temperature, low energy; self-replicating catalyst
  • Highly selective and efficient at high perchlorate loads
  • Couples to bioregenerative systems and the chlor-alkali feed
  • Biological — slow to start, sensitive to upsets and toxins, needs careful control
  • Produces biomass and chloride brine to manage

When preferred: Bulk perchlorate destruction once a stable culture is established — the energy-cheap default.

Catalytic / electrochemical reduction

[1]

Heterogeneous catalysts or electrochemical cells reduce perchlorate without biology — faster startup, no living system to maintain.

Materials: Reduction catalyst (e.g. Re/Pd) · Electrochemical cell or H₂ feed
  • No biological lag; robust to toxins; controllable
  • Compact; pairs with the H₂/electricity the colony already has
  • Catalyst is an import and can be fouled; higher energy than biology
  • Lower TRL for the high Mars perchlorate loads

When preferred: Fast startup, polishing, or where a bioreactor is impractical.

Membrane / ion exchange (ion removal + polish)

[3]

Reverse osmosis and ion-exchange resins strip perchlorate and other ions physically — the polishing and ultrapure-grade step after destruction.

Materials: RO membranes · Perchlorate-selective and mixed-bed resins · High-pressure pump
  • Reaches ultrapure (electrolysis/synthesis) grade; removes all ions, not just perchlorate
  • Mature, reliable, well-characterized
  • Membranes/resins are import consumables that foul and exhaust
  • Concentrates perchlorate into a reject stream that still needs destruction

When preferred: Final polish to electrolysis/synthesis grade and potable standards.

Thermal / UV destruction

[7]

UV photolysis or thermal decomposition breaks perchlorate without biology or catalysts — simple, robust, energy-intensive.

Materials: UV-C source or thermal reactor
  • No consumable catalyst/culture; robust and simple
  • Useful where biological systems can't be maintained
  • Higher energy; UV lamps are consumables; slower for high loads

When preferred: Small-scale, emergency, or pre-treatment polishing.

Failure modes

Mode Cause Detection Mitigation
Perchlorate breakthrough to crew/crops (safety-critical)[2] Treatment failure or breakthrough lets perchlorate reach potable or hydroponic water; crew thyroid toxicity and crop bioaccumulation follow. Online ion chromatography on treated water; periodic crew/crop monitoring. Multi-barrier train (destruction + ion exchange), continuous monitoring with interlocks, conservative margins; never single-barrier for potable.
Catalyst-poison breakthrough to electrolysis (availability-critical)[5] Trace perchlorate or chloride passes to the electrolyzer/synthesis feed, irreversibly poisoning catalysts over weeks. Feedwater conductivity and ion monitoring; electrolyzer voltage drift as a lagging indicator. Ultrapure polish (RO/mixed-bed) on electrolysis feed, tighter spec than potable, guard monitoring.
Bioreactor culture crash[1] Temperature excursion, toxin (heavy metals, biocide), or starvation kills the perchlorate-reducing culture; destruction stops. Perchlorate removal efficiency drop; culture health monitoring. Stable conditions, toxin exclusion (clean feed), backup catalytic/UV train, culture reseeding capability.
Membrane / resin fouling and exhaustion[3] Particulates, scaling, and ion loading foul/exhaust RO membranes and resins, collapsing flux or breakthrough. Transmembrane pressure rise, conductivity breakthrough, resin capacity tracking. Upstream filtration, antiscalant, clean-in-place, scheduled replacement, lead-lag beds.
Reject/brine mishandling[6] The concentrated perchlorate reject from RO/ion exchange is more toxic than the feed; mishandling re-contaminates the loop. Reject-stream tracking; loop perchlorate trending. Route reject to destruction (bioreactor) and the chloride product to chlor-alkali; closed, audited handling.

Mars adjustments

Perchlorate is the defining Mars water problem[4]

Impact: Earth water treatment rarely faces perchlorate; on Mars it is in everything at percent levels — toxic to crew, poisonous to catalysts, and bioaccumulated by crops. Purification on Mars IS, first and foremost, perchlorate management.

Mitigation: Perchlorate destruction (bio/catalytic) as the core unit operation, sized to the regolith load, with ion-exchange polish.

The poison is also a feedstock[6]

Impact: Reducing perchlorate yields chloride brine — exactly the chlor-alkali feed — plus a little oxygen. The detox step is a deliberate producer for the chlorine economy, not a waste-disposal afterthought.

Mitigation: Co-locate purification with chlor-alkali; route the chloride product as feed, closing a chemistry-pillar loop.

Use-specific grades, one shared resource[3]

Impact: Electrolysis feed must be ultrapure, potable water strictly health-limited, hydroponics nutrient-balanced, industrial water tolerant — yet all draw from one purification system feeding a shared loop where contamination propagates.

Mitigation: Graded treatment trains (common destruction + use-specific polish), loop segregation, point-of-use final treatment.

It is the single highest-leverage water node[5]

Impact: Roughly every water-using node in the tree — electrolysis, propellant, life support, agriculture, chemistry, mining — cites this step as a precondition. A purification failure is a settlement-wide failure.

Mitigation: Redundant trains, deep monitoring, conservative margins; treat purification uptime as life- and mission-critical.

Consumables must go local eventually[1]

Impact: RO membranes, ion-exchange resins, and UV lamps are imports; a self-sufficient colony must regenerate resins, make membranes (polymer chain), and lean on the regenerable biological route.

Mitigation: Bias toward biological destruction (no consumable catalyst), regenerable resins, local membrane production as a polymer-chain goal.

Alternatives & substitutes

Import clean water (no local purification)[8]

  • Guaranteed purity; no treatment plant
  • Water is heavy, bulk-consumed cargo — importing it defeats the purpose of a self-sufficient colony

When preferred: Tiny early outpost only; the thing ISRU exists to eliminate.

Use only the cleanest source (atmospheric water)[9]

  • Atmospheric/condensed water carries far less perchlorate than regolith brine — less treatment needed
  • Atmospheric capture is low-yield and energy-intensive; can't supply bulk demand alone

When preferred: Supplementary clean-water source reducing (not eliminating) treatment load.

Tolerate impurity (use-specific low grade)[4]

  • Some uses (dust suppression, some construction) tolerate untreated water
  • Most uses (crew, crops, electrolysis) cannot — and perchlorate spreads through any shared loop

When preferred: Non-contact industrial uses only, in a strictly segregated loop.

Requires

References

  1. Coates, J. D., & Achenbach, L. A. (2004). Microbial perchlorate reduction: rocket-fuelled metabolism. Nature Reviews Microbiology, 2(7), 569–580. doi:10.1038/nrmicro926 — The canonical reference on perchlorate-reducing bacteria — the biology behind reducing ClO₄⁻ to harmless chloride, central to Mars water detox and the chlor-alkali feed loop.
  2. U.S. Environmental Protection Agency (2020). Perchlorate in Drinking Water. EPA Office of Water. EPA 815-F-20-002. — EPA Reference Dose 0.7 µg/kg/day; advisory concentration 15 µg/L for drinking water.
  3. Crittenden, J. C., Trussell, R. R., Hand, D. W., Howe, K. J., & Tchobanoglous, G. (2012). MWH's Water Treatment: Principles and Design, 3rd Edition. Wiley. ISBN 978-0-470-40539-0. — The definitive water-treatment engineering reference: coagulation, filtration, adsorption, ion exchange, membranes, disinfection, and process-train design.
  4. Hecht, M. H., Kounaves, S. P., Quinn, R. C., West, S. J., et al. (2009). Detection of Perchlorate and the Soluble Chemistry of Martian Soil at the Phoenix Lander Site. Science, 325(5936), 64-67. doi:10.1126/science.1172466 — First in-situ measurement of perchlorate in Mars regolith — 0.4–0.6 wt%.
  5. ASTM International (2018). Standard Specification for Reagent Water. ASTM D1193-06(2018). ASTM D1193-06(2018). doi:10.1520/D1193-06R18 — Type I/II reagent water purity standards (conductivity <1 µS/cm).
  6. Davila, A. F., Willson, D., Coates, J. D., & McKay, C. P. (2013). Perchlorate on Mars: a chemical hazard and a resource for humans. International Journal of Astrobiology, 12(4), 321-325. doi:10.1017/S1473550413000164 — Biological reduction of perchlorate as pre-treatment for ISRU water.
  7. Catling, D. C., Claire, M. W., Zahnle, K. J., Quinn, R. C., Clark, B. C., Hecht, M. H., & Kounaves, S. (2010). Atmospheric origins of perchlorate on Mars and in the Atacama. Journal of Geophysical Research: Planets, 115(E1), E00E11. doi:10.1029/2009JE003425 — Perchlorate concentration in Mars regolith (0.4-0.6 wt%) — catalyst poisoning hazard.
  8. Anderson, M. S., Ewert, M. K., & Keener, J. F. (2018). Life Support Baseline Values and Assumptions Document (BVAD). NASA Johnson Space Center. NASA/TP-2015-218570/REV1. — The authoritative ECLSS reference: crew metabolic rates, consumable mass balances, atmosphere/water/waste loop sizing, and life-support technology trades.
  9. Grover, M. R., & Bruckner, A. P. (1998). Water Vapor Extraction from the Martian Atmosphere by Adsorption in Molecular Sieves. AIAA 98-3301, 34th AIAA/ASME/SAE/ASEE Joint Propulsion Conference. doi:10.2514/6.1998-3301 — The WAVAR concept: capturing Mars atmospheric water vapor (~210 ppm) by molecular-sieve adsorption and thermal regeneration — a water source independent of ice deposits.