How a liquid droplet radiator works.

Replace the panel with the fluid.

A conventional radiator pumps coolant through a large solid panel, and the panel does the radiating. A liquid droplet radiator removes the panel. The hot coolant is sent across open space as a sheet of fine droplets. Small droplets have an enormous surface area for their mass, so they radiate heat efficiently as they travel. On the far side, a collector recaptures the cooled droplets and returns them to the loop.

The idea trades a structural problem for a fluid-handling one. Instead of building a large, light, stiff surface, you have to form droplets precisely, keep them on course across open space, and catch nearly all of them again. The open problems below set out what has to be true for that trade to pay off.

It also changes what happens when the heat load falls. A solid panel radiates all the time, so when the payload idles or the spacecraft passes into shadow, the panel and its coolant keep cooling, and many designs carry heaters to stop the coolant freezing. A droplet radiator can be turned down or switched off: the generator stops, the fluid stays inside the loop, and there is no exposed panel to keep warm.

Area is set by temperature. Mass is set by how you make the area.

Every surface radiates heat at a rate set by its temperature. The Stefan-Boltzmann law gives the heat radiated per unit area:

$q=\epsilon \sigma T^4$

where ε is emissivity (how close the surface is to a perfect radiator, at most 1), σ is the Stefan-Boltzmann constant and T is absolute temperature. Good radiator coatings already have emissivity close to 1, so little is left to gain there. Temperature is capped: waste heat has to flow downhill from the electronics to the radiator, so the radiator runs cooler than the chips it cools. With both terms pinned, the only free variable is area. Double the heat and you double the area.

Under the assumptions of Fig. 1 on our homepage (ε = 0.9, T = 325 K, double-sided, no environmental load), each square metre of double-sided panel rejects about 1,140 W, so a megawatt needs about 878 m².

If physics fixes the area, the lever left is the mass it takes to make that area. A panel has to carry coolant channels, a structure stiff enough to deploy, and protection against micrometeoroids, and all of it grows with area. Droplets change that arithmetic. A sphere of radius r and density ρ has surface area 4πr² and mass ⁴⁄₃πr³ρ, so its area per unit mass is

$\frac{A}{m}=\frac{3}{\rho r}$

Halving the droplet radius doubles the radiating area per kilogram of fluid. Fine droplets turn a small mass of liquid into a very large radiating surface. The difficulty, and the reason the idea is hard, is that the droplets have to be made, aimed and caught again.

Five blocks in a loop.

Heat exchanger

Moves heat from the payload's own cooling loop into the radiator fluid.

Pump

Circulates the fluid around the loop and supplies the pressure the generator needs.

Generator

Turns the pressurised liquid into streams of droplets, aimed across open space at the collector.

Droplet sheet

The radiating surface. Heat leaves the droplets as infrared radiation while they are in flight.

Collector

Catches the cooled droplets and returns the liquid to the loop.

Mass per kilowatt, as a range.

In our modelling, a droplet radiator is 52% to 74% lighter per kilowatt than a two-sided solid radiator. Compared with a two-sided solid radiator panel at 4 kg/m², 325 K and emissivity 0.9, rejecting to a 200 K sink with 200 W/m² absorbed. The range runs from a bare panel to a panel with loop, deployment and micrometeoroid allowances.

Pluvio cost model v3, 1 October 2026

The principle behind it will not change: every assumption is chosen to favour the incumbent panel, so the comparison errs against us. We have not yet shown any of this in hardware; that is what the test programme is for.

Studied decades ago, set aside, now needed.

The references below record the earlier research. No mission of that era needed to reject enough heat to justify a new kind of radiator. To our knowledge, an LDR has never flown operationally.

Later work in Japan tested droplet generators and collectors in drop-shaft microgravity facilities.

References

1. Mattick, A.T. and Hertzberg, A. (1981). Liquid droplet radiators for heat rejection in space. Journal of Energy, 5(6), 387–393. doi.org/10.2514/3.62557
2. Presler, A.F., Coles, C.E., Diem-Kirsop, P.S. and White, K.A. III (1985). Liquid droplet radiator program at the NASA Lewis Research Center. NASA TM-87139. Prepared for the Fourth AIAA/ASME Thermophysics and Heat Transfer Conference, Boston, June 1986. ntrs.nasa.gov/citations/19860002779
3. White, K.A. III (1987). Liquid droplet radiator development status. NASA TM-89852, July 1987. ntrs.nasa.gov/api/citations/19870010920/downloads/19870010920.pdf
4. Pfeiffer, S. and White, K.A. (1989). Conceptual design of a liquid droplet radiator space flight experiment. SAE Technical Paper 891565. doi.org/10.4271/891565
5. Totani, T., Itami, M., Nagata, H., Kudo, I., Iwasaki, A. and Hosokawa, S. (2002). Performance of droplet generator and droplet collector in liquid droplet radiator under microgravity. Microgravity Science and Technology, 13(2), 42–45. doi.org/10.1007/BF02872070
6. Chong, D., Zhu, M., Zhao, Q., Chen, W. and Yan, J. (2021). A review on thermal design of liquid droplet radiator system. Journal of Thermal Science, 30(2), 394–417. doi.org/10.1007/s11630-021-1419-2

What still has to be proven.

A droplet radiator is simple to describe and difficult to build. These are the questions that decide whether it works. Each one maps to a stage of our test programme.

1. Droplet generation. Bench

   Can we produce a uniform, well-aimed sheet of fine droplets, reliably and for long durations?

   A radiator needs many streams of droplets, uniform in size and spacing, each aimed well enough to reach a collector across open space. Small errors in aim grow with distance, and a generator has to hold its performance for years without clogging or wearing.

2. Recapture. Bench · Vacuum

   What fraction of droplets do we recover, and does fluid loss over a mission lifetime stay within budget?

   Every droplet that misses the collector is fluid lost for good. Over a multi-year mission even a small miss fraction adds up, so recapture efficiency sets how much spare fluid a spacecraft has to carry. Collection also has to avoid splashing, which creates new, uncontrolled droplets.

3. Vacuum behaviour. Vacuum

   How does the fluid behave, including evaporative loss, across the operating temperature range in vacuum?

   In vacuum, any liquid evaporates at a rate set by its vapour pressure, which rises steeply with temperature. The fluid has to stay liquid across the operating range, lose as little mass as possible to evaporation, and keep its properties over a long mission.

4. Microgravity. Microgravity

   Does the sheet stay stable, and does collection still work, without gravity and while the spacecraft manoeuvres?

   On the ground, gravity bends every droplet's path and helps pull liquid into a collector. In orbit neither is true, and spacecraft manoeuvres add accelerations the sheet has to tolerate. Drop towers and parabolic flights give seconds of microgravity; only flight gives more.

5. Contamination. Vacuum · Orbit

   Can we keep fluid off optics, solar arrays and other sensitive surfaces?

   Droplets that escape the sheet, and vapour from evaporation, can land on optics, solar arrays and sensors. A droplet radiator has to show that whatever it loses goes somewhere harmless.

6. Integration. Flight

   Can the droplet loop couple cleanly to a payload's thermal system and operate for years?

   A radiator is one part of a spacecraft's thermal system. The droplet loop has to take heat from the payload's loop, start and stop safely, follow changing heat loads, and run for years with little or no maintenance.

Open problems copy on /technology

Four stages, each with a job.

1. Bench BUILDING

   Measures

   Whether we can form a stable sheet of droplets and recapture it, and how much heat the sheet rejects, at laboratory pressure.

   Pass means

   A sheet that forms repeatably, a recapture fraction we can measure and improve, and heat rejection that agrees with our model.

2. Vacuum chamber NEXT

   Measures

   The same quantities under vacuum, plus how the fluid behaves across the operating temperature range, including evaporative loss.

   Pass means

   Radiated heat and fluid loss in vacuum close enough to the model that we trust it for the next stage.

3. Microgravity PLANNED

   Measures

   Sheet formation and collection without gravity, during the short microgravity periods of parabolic flight.

   Pass means

   Collection that still works without gravity, and a sheet whose behaviour we can predict.

4. Flight demonstration PLANNED

   Measures

   Heat rejection, fluid loss and contamination in orbit, over a meaningful operating period.

   Pass means

   Flight data an operator can use to specify a droplet radiator for their spacecraft.

Test programme copy on /technology