vacuum-radiator

Vacuum radiator

Subsystem Semi-native thermal
TRL Mars
—
Energy intensity
—
Required by
0
Requires
2

Rejects waste heat from any thermal source on Mars surface to the cold sky. Three architectures span the trade space: ISS-style pumped-fluid panels with redundant coolant loops, Kilopower-class heat-pipe direct-radiating fins, and deployable lightweight composite panels. The dominant Mars-specific failure mode is dust accumulation — same hazard as PV arrays, manifesting as emissivity drop and rising solar absorption.

Last reviewed: 2026-06-09

Governing equations

Stefan-Boltzmann net radiation. σ = 5.67 × 10⁻⁸ W/(m²·K⁴). Hot-side T⁴ scaling means radiator area drops as the fourth power of operating temperature. [1]

Effective Mars sink temperature combines deep-space sky radiation with absorbed direct + diffuse solar. The α_s / ε ratio matters more than either alone. [2]

Radiator area for a given heat-rejection load. At 400 K rad / 200 K sink: ~ 720 W/m². At 800 K rad: ~ 22 000 W/m² — 30× compaction by raising T. [2]

Net heat rejected accounts for solar absorption derating the radiator. Mars dust raises α_s, drops ε, and shifts the balance unfavorably. [2]

Key constants & quantities

Symbol Value Units Conditions Description
σ 5.670374e-8 W / (m² · K⁴) — Stefan-Boltzmann constant. The proportionality buried in every spacecraft thermal calculation.[3]
ε_clean,Al 0.85 ±0.05 (emissivity) — Emissivity of anodized aluminum, the most common spacecraft radiator surface. Z93 white paint achieves 0.92.[2]
α_s,clean 0.15 (solar absorptivity) — Solar absorptivity of anodized aluminum. The α_s / ε ratio of 0.18 means modest solar heating relative to radiation cooling.[2]
q_400K 580–850 W / m² (clean, Mars sky) — Heat rejection per m² for a 400 K radiator surface (typical cryocooler hot-side or habitat thermal loop). Range covers diurnal sky variation.[2]
q_800K 20000–25000 W / m² (clean, KRUSTY-class) — Heat rejection per m² for an 800 K radiator (high-T reactor hot-side). 30× the 400 K case — why Brayton/Stirling architectures dominate space fission.[4]
T_sky,Mars 150–220 K (effective) — Effective radiation sink temperature for an upward-facing Mars radiator. Mostly deep space (3 K) modulated by Mars atmospheric IR emission.[5]
d_dust,radiator 0.05–0.2 %/sol Δε — Emissivity degradation rate from dust accumulation. Lower than PV soiling because dust on radiator only changes emissivity; PV loses transmittance.[6]
m_panel 4–10 ±25 % kg / m² (deployable) — Deployable lightweight radiator panel mass density including coolant lines, structure, MLI back-coating. ISS PV thermal panels at ~ 9 kg/m².[2]

Operating envelope

ParameterRangeUnitsSource
Hot-side temperature 250 – 1100 K [2]
Working-fluid pressure 1 – 50 bar [1]
Coolant flow rate 0.01 – 1 kg/s per kW rejected [1]
Effective emissivity (clean) 0.78 – 0.92 [2]
α_s / ε ratio (target) 0.15 – 0.25 [2]

Mass balance

Basis: Reject 10 kW thermal at 400 K hot-side, Mars equatorial siting

Inputs

Aluminum panel + coolant lines 130 kg (deployable, ~ 17 m² at 7.5 kg/m²) [2]
Pumped coolant (PAO / NH₃ / water-glycol) 15 kg circulating inventory [1]
Pump + control electronics 12 kg [1]
Pump electrical power 50 W (continuous) [1]
  • Pump electrical power: ~ 0.5 % of rejected thermal load — small but non-zero parasitic.

Outputs

Heat radiated to Mars sky 10 kW [2]
  • Heat radiated to Mars sky: At clean ε = 0.85, T_rad = 400 K, T_sink = 180 K: ~ 720 W/m² × 17 m² ≈ 12.2 kW gross; 10 kW net after pump + structure losses.
TRL · Earth
9/ 9
TRL · Mars
7/ 9
ISS Active Thermal Control System (ATCS) has operated since 2000 — TRL 9 for low-Earth orbit. Mars surface radiators are TRL 7: KRUSTY radiator fin tested 2018 at full power, MOXIE thermal management tested in situ since 2021. No purpose-built Mars-surface kilowatt-scale radiator system has flown yet, but the design transfer from KRUSTY + ISS is straightforward.[4]
Energy budget
0.005 kWhe / kWh thermal rejected (pumped-loop variant) [1]

Pump parasitic typically < 1 % of heat rejected. Heat-pipe radiators are purely passive — zero electrical demand. Active retraction/deployment for storm protection adds small one-time loads.

Variants & trade-offs

Heat-pipe radiator fins (Kilopower / KRUSTY)

[4]

Sodium-filled heat pipes carry reactor heat from core to fin array; titanium or stainless fins radiate to sky. Purely passive — no pump, no working-fluid loop. The KRUSTY architecture proven in 2018.

Hot-side T
600–1100 K
Fin material
0–0 Ti or 316SS
Stack lifetime
100000–150000 h
Materials: Sodium-filled heat pipes (Hastelloy or stainless) · Titanium or 316SS fins · White ceramic emissive coating · Structural standoffs (Inconel)
  • Purely passive — no pump, no fluid loop, no electrical demand
  • Compact: 30× area reduction vs 400 K pumped loop at 800 K hot-side
  • KRUSTY flight-proof at 1 kWe scale
  • Inherently redundant — single heat-pipe failure does not stop rejection
  • Hot-side T requirement constrains coupled system architecture
  • Materials cost is high — titanium fins not cheap
  • Heat-pipe orientation-sensitive; Mars surface mount required at design
  • Cannot be retracted in storms; dust impact direct

Pumped-fluid panel (ISS ATCS heritage)

[2]

Liquid coolant (ammonia, PAO, or propylene glycol) flows through panel-embedded tubes; panel radiates to sky. ISS has operated this architecture at 70 kW class since 2000.

Hot-side T
275–450 K
Working-fluid pressure
1–30 bar
Stack lifetime
80000–150000 h
Materials: Aluminum 6061 panel face-sheet · Embedded stainless or titanium coolant tubes · Coolant: NH₃ (ISS), PAO, or 50/50 propylene-glycol water · Pump + accumulator + bypass valve set
  • Mature ISS heritage at multi-kW scale
  • Adaptable to wide range of hot-side T
  • Retractable / deployable architectures protect against storms
  • Single-point throughput easily monitored + controlled
  • Pump is single-point failure (redundant pumps standard)
  • Coolant leak = system-wide failure (ammonia, especially, is hazardous)
  • Higher mass per kW than heat-pipe variant
  • Working fluid freeze on cold Mars night is a real risk

Lightweight composite deployable

[2]

Carbon-fiber substrate with thin-film emissive coating + integrated heat pipes or fluid lines. Roll-out or fold-out architecture. Used on Solar Probe + Lunar Gateway power systems.

Hot-side T
275–400 K
Stowed volume ratio
5–20 × area reduction
Stack lifetime
50000–100000 h
Materials: Carbon-fiber-reinforced polymer face-sheet · Aluminum heat-pipe condensers · Silver-Teflon thermal-control coating · Deployable hinge / boom mechanism
  • Lowest mass per kW (< 4 kg/m²)
  • Compact stowed volume — frees launch fairing
  • Deploys post-landing — terrain-tolerant
  • Mars UV ages composites and adhesives
  • Deployment mechanism is single-use, high-consequence
  • Lower hot-side T capability than KRUSTY fins

Failure modes

Mode Cause Detection Mitigation
Dust accumulation (emissivity degradation)[6] Mars atmospheric dust deposits on radiator surface; alters emissivity (drops 5–15 %) and solar absorption (rises 30–60 %). Hot-side T climbs at constant load; nominal vs actual rejection ratio. Vertical or near-vertical panel orientation halves accumulation rate; electrostatic dust shedding; periodic mechanical wipe for accessible installations; design margin for 30 % derate.
Micrometeorite puncture (working-fluid loop)[7] Hypervelocity impact penetrates coolant tube; fluid leaks, loop pressure drops. Pump pressure drops; coolant inventory falls. Whipple shield over coolant tubes; multi-loop architecture so one puncture does not zero rejection; isolation valves at loop boundaries; chemical leak detector.
Coolant freezing on cold side[2] Reduced thermal load causes radiator outlet T to drop below freezing point of coolant (ammonia 195 K, PAO 233 K). Coolant solidifies, fractures panel tubes. Flow rate drops; temperature distribution non-uniform. Bypass valve to recirculate warm coolant during low-load; choice of low-freeze-point fluid (NH₃ down to 195 K); active heater on radiator outlet in cold dormant operations.
Pump bearing failure[2] Continuous-duty pump bearings wear; vibration climbs. Accelerometer signature; flow at constant ΔP drops. Redundant pumps with automatic switchover; magnetically suspended bearings; programmed replacement every 40 000 h.
Heat-pipe dry-out (passive variant)[4] Working fluid (Na, K, water) exceeds wick return capacity at high heat flux; pipe interior dries; thermal conductance collapses. Fin T rises while pipe inlet T drops — characteristic signature. Conservative heat-flux design (50 % of dry-out limit); high-conductivity wick designs; gravity-aided orientation on Mars surface.
Coating degradation under Mars UV[6] Solar UV degrades white paint (Z93, S13G/LO) and silvered-Teflon thermal coatings, raising α_s. Solar absorptivity measured by panel T excursion at peak sun. UV-stable coatings (silver-FEP outperforms standard paints); periodic visual inspection; sacrificial outer layer where mission-critical.
Micrometeorite swarm during dust storm[7] Mars Year 18 (1971) global storm coincided with seismic anomalies; literature suggests dust storms may correlate with elevated impactor flux. Multiple simultaneous coolant-loop punctures. Robust Whipple shielding on multi-loop systems; loop-level isolation valves; n+1 loop architecture.

Mars adjustments

No convective heat transfer[1]

Impact: Mars 600 Pa atmosphere is too thin for meaningful convection. Heat rejection is essentially purely radiative — unlike Earth where convection dominates below 60 °C.

Mitigation: Design entirely around Stefan-Boltzmann radiation. No fan or blower options for serious heat loads. Hot-side T elevation (Brayton, Stirling, high-T heat pipes) is the only path to compact radiators.

Dust accumulation on radiator surface[6]

Impact: Same hazard as PV arrays. Dust raises α_s (solar absorption goes up) and drops ε (emissivity goes down) — both unfavorable. Net rejection capacity can drop 30 % over months of dust deposition.

Mitigation: Vertical or near-vertical mounting halves deposition rate; periodic mechanical or electrostatic cleaning; oversizing for 30 % derate; multi-panel architecture so one fouling does not collapse the system.

Effective sink T is Mars sky, not deep space[5]

Impact: On Mars surface, the radiator sees a sky at 150–220 K effective T (deep space + atmospheric IR + scattered solar). This is warmer than the 3 K of deep space, reducing rejection capacity by ~ 5–10 % vs orbital systems.

Mitigation: Hot-side T elevation; horizon-aware geometry (point at coldest part of sky); diurnal load shifting (peak heat-reject at night).

Diurnal radiator T cycling[2]

Impact: Mars sun heats unshielded radiator panels to ~ +20 °C peak; Mars night cools to ~ −90 °C. Diurnal ΔT ~ 110 °C × 365 sols/Mars year = 365 cycles/year — 3× the fatigue cycles of LEO.

Mitigation: Sun-shield outer layer; design for thermal cycling fatigue (compliant interconnects, CTE matching); avoid CTE mismatches at coolant-tube-to-panel interface.

Working fluid freeze risk[2]

Impact: Mars overnight T below −90 °C exceeds freezing point of most common coolants. PAO freezes at −40 °C; propylene glycol/water at −37 °C; only ammonia handles full Mars night without solid phase.

Mitigation: Choose ammonia for full-night-survival systems (ISS heritage); active heaters on cold-side coolant lines; bypass valves to maintain minimum flow.

Alternatives & substitutes

Regolith conductive heat sink[1]

  • Massive thermal mass available for transient loads
  • Decoupled from sky / dust storm conditions
  • No moving parts
  • Steady-state heat rejection requires very large buried heat-exchanger area (regolith thermal conductivity ~ 0.05 W/m·K)
  • Only works as transient buffer, not continuous rejection
  • Saturates over multi-sol operation

When preferred: Transient heat dumps (mining surge loads), thermal buffer for habitat between night-cold and day-warm.

Sublimation cooling (water or ammonia)[1]

  • No fixed infrastructure — consume coolant, vent to atmosphere
  • Very high specific energy (2.84 MJ/kg water)
  • Used on early Mercury / Gemini spacecraft
  • Consumable — drives resupply mass linearly with thermal load
  • Vented water lost to ISRU loop
  • Only suitable for transient or backup

When preferred: Emergency thermal dump; backup during radiator maintenance; integrated with venting requirements for closed-loop ISRU.

Requires

Inputs

References

  1. Bergman, T. L., Lavine, A. S., Incropera, F. P., & DeWitt, D. P. (2017). Fundamentals of Heat and Mass Transfer, 8th Edition. John Wiley & Sons. ISBN 978-1-119-32042-5. — Standard undergraduate / engineering reference for heat transfer: Stefan-Boltzmann radiation, conduction, convection.
  2. Gilmore, D. G. (Ed.) (2002). Spacecraft Thermal Control Handbook, Volume 1: Fundamental Technologies. The Aerospace Press / AIAA. ISBN 978-1-884989-11-4. — Canonical spacecraft thermal-control reference: radiator design, materials, coatings, MLI, heat pipes.
  3. Tiesinga, E., Mohr, P. J., Newell, D. B., & Taylor, B. N. (2021). CODATA Recommended Values of the Fundamental Physical Constants: 2018. National Institute of Standards and Technology. doi:10.1103/RevModPhys.93.025010 — Faraday constant, gas constant, fundamental physical constants.
  4. Mason, L., Gibson, M., Poston, D., Briggs, M., Sanzi, J., & Bell, J. (2018). A Small Fission Power System for NASA Exploration: KRUSTY Test Results. Nuclear and Emerging Technologies for Space (NETS) Conference, Las Vegas. NASA/TM-2018-219782. — KRUSTY full-power test 2018; Mars surface fission TRL 6 demonstration.
  5. Savijärvi, H. (1999). A model study of the atmospheric boundary layer in the Mars Pathfinder lander conditions. Quarterly Journal of the Royal Meteorological Society, 125(553), 483-493. doi:10.1002/qj.49712555310 — Mars boundary layer + effective sky temperature modeling for radiative heat-transfer applications.
  6. Gaier, J. R., Ellis, S., & Hanks, N. C. (2002). Aeolian removal of dust types from photovoltaic surfaces on Mars. NASA Glenn Research Center, NASA/TM-2002-211837. NASA/TM-2002-211837. — Mars dust deposition + removal mechanisms on optical / radiator surfaces; α_s and ε degradation rates.
  7. Christiansen, E. L. (2003). Meteoroid/Debris Shielding. NASA Johnson Space Center, TP-2003-210788. NASA/TP-2003-210788. — Whipple shielding theory and ISS design; ballistic-limit equations for hypervelocity impact.