Thermophotovoltaics: A Solar Cell Aimed at a Hot Object
The idea sounds almost too simple. Heat something until it glows, point a photovoltaic cell at the glow, collect electricity. No moving parts, no working fluid, no turbine. For decades the efficiency was too poor to be interesting, and then in the last few years it was not: laboratory cells passed 40 percent, which puts a device with no moving parts in the same range as industrial turbomachinery. The reason it works now, and the reason it still needs something hot, are both worth understanding.
How It Differs From a Solar Cell
A conventional solar cell is optimised for the Sun's spectrum, which peaks in visible light. A thermophotovoltaic cell faces an emitter at one to two and a half thousand kelvin, which radiates mostly in the infrared, so its semiconductor is chosen with a much lower band gap - typically in the range of half an electronvolt to three quarters, against about 1.1 for silicon.
The physics of absorption is the same photoelectric process: a photon with more energy than the band gap promotes an electron across it, and the junction separates the resulting charges. What changes is the matching between the source spectrum and the absorber.
There is one structural advantage a solar cell cannot have. The emitter is a few millimetres away and it belongs to the system, so photons that the cell cannot use need not be lost. A mirror behind the cell reflects those low-energy photons straight back into the emitter, where they are reabsorbed and keep it hot.
That photon recycling is the advance that changed the field. Sub-band-gap photons are the dominant loss channel in any TPV device, and returning them to the source rather than converting them to waste heat is what took efficiencies from the twenties into the forties.
What It Is Useful For
The most discussed application is thermal energy storage for the grid. Surplus electricity heats a cheap medium such as graphite or molten silicon to very high temperature; when power is needed, TPV cells convert the stored heat back to electricity. The attraction is that the storage medium is inexpensive and the converter has no moving parts, so it can be built at scales where mechanical turbines are impractical.
Waste heat recovery is a second area, though a harder one: industrial exhaust is usually far cooler than the optimum emitter temperature, and TPV efficiency drops steeply as the emitter cools.
Portable and remote power is a third. A TPV generator burning fuel is quiet, has no moving parts and needs no maintenance, which matters in settings where a small engine would be unacceptable.
In space, TPV has been considered as a replacement for the thermoelectric conversion used in radioisotope generators, where its higher efficiency would mean more power from the same quantity of isotope - a direct comparison with the thermal approach discussed on our page about betavoltaics, which skips heat altogether at much smaller scales.
The Limit It Cannot Cross
TPV is a heat engine. It takes heat from a hot reservoir, converts part of it to work, and rejects the rest to a cold one - the cell itself, which must be kept cool. Its efficiency is therefore bounded by the Carnot limit set by the two temperatures, and no engineering improvement reaches past it.
This matters because thermal radiation is sometimes described loosely as an ambient energy source, as though a cell could simply be pointed at a room and produce power. It cannot. A cell at room temperature facing a room at the same temperature absorbs and emits radiation at equal rates, and the net work is exactly zero.
That is not a limitation of current technology. It is the second law, and our page on free energy sets out why a body in equilibrium with its surroundings yields no work regardless of how much radiation is passing through it. TPV works precisely because someone heated the emitter first.
The distinction is worth holding onto because it separates two claims that sound similar. Converting radiation from a hot source is established engineering with efficiency records in the peer-reviewed literature. Extracting work from equilibrium radiation is not a harder version of the same problem; it is a different claim, and it is excluded. Neutrinovoltaic research is careful on exactly this point: its question concerns thermal fluctuations and ambient radiation driving current in an engineered material, and any honest version of it has to respect the same boundary.
Frequently asked questions
How is a TPV cell different from a solar panel?
It uses a much lower band gap semiconductor, matched to infrared radiation from an emitter at one to two and a half thousand kelvin rather than to the Sun's visible spectrum. It also sits close to its emitter, which allows unusable photons to be reflected back.
How efficient is it?
Laboratory devices have recently exceeded 40 percent, which is comparable to or better than steam turbines. The improvement came largely from reflecting sub-band-gap photons back into the emitter instead of wasting them.
What is it used for?
Thermal energy storage for the grid is the most discussed application, along with waste heat recovery, portable power without moving parts, and potentially as a replacement for thermoelectric conversion in space power systems.
Does it need something hot?
Yes, necessarily. TPV is a heat engine and requires a temperature difference between emitter and cell. Its efficiency is bounded by the Carnot limit set by those two temperatures.
Can it harvest ambient thermal radiation?
No. A cell at room temperature facing surroundings at the same temperature absorbs and emits at equal rates, for exactly zero net work. That is the second law, not a limitation of current technology.
