The assumption most engineers carry into this comparison is that GaAs wins on performance, so it must win overall. That is only true if performance is the only variable on the scorecard.
A 2025 study published in Applied Sciences (MDPI) evaluated solar cell degradation across silicon, GaAs, triple-junction, and CIGS technologies using telemetry data from small satellite missions launched after 2020. The study found that GaAs cells provide high efficiency in LEO due to a direct bandgap and superior radiation hardness, and that is a real advantage. Nobody is disputing the physics. MDPI
But for most LEO constellations below 1,000 km, the radiation environment is manageable. Silicon degrades in that environment, but so does GaAs, and the degradation curves are closer than the marketing materials suggest. A NASA technical analysis concluded that for reference mission parameters, the cost of solar power from silicon planar arrays is less than from GaAs arrays, though the outcome is sensitive to array purchase cost, solar cell efficiency, and specific mass. NASA Technical Reports Server.
That last point is the one most programs miss. The trade is not silicon vs. GaAs in a vacuum. It is silicon vs. GaAs across your specific orbit, mission duration, mass budget, and program cost target. For a three-to-five-year LEO mission where mass is not the binding constraint, the cost advantage of silicon does not disappear at end of life.
SOURCE is not the right choice for every mission. Above 1,000 km, silicon radiation tolerance becomes a real issue. But below that altitude, for programs optimizing on cost and scale, the trade study does not point where most people assume it does.
Run the numbers for your mission. We are happy to help. Visit source.space.
