Another vacuum tube technology with current applications and substantial advantages over semiconductor approaches to the same problems is the Thermionic Converter. This is a vacuum-tube technology heat engine that turns temperature differences into electric power - by boiling electrons off a hot electrode and collecting them, at a somewhat more negative voltage (like 0.5 to 1 volt), at a cooler electrode.
Semiconductor approaches such as the Peltier Cell tend to be limited in operating temperature due to the materials involved, and lose a major fraction of the available power to non-power-producing heat conduction from the hot to the cold side of the device. Thermionic converters, by contrast are vacuum devices, and inherently insulating (with the heat conducted almost entirely by the working electrons, where it is doing the generation, or parasitic infrared radiation, which can be reflected rater than absorbed at the cold side.) They work very well at temperatures of a couple thousand degrees, a good match to combustion, point-focused solar, and nuclear thermal sources.
Thermionic converters have been the subject to recent improvements, such as graphine electrodes. The power density limitation of space charge has been solved, by using a "control grid" to encourage to charge to move along from the emitter to the collector and magnetic fields to guide it (so it doesn't discharge the control grid and waste the power used to charge it).
Current thermionic technology can convert better than 30% of the available thermal energy to electrical power and achieves power densities in the ballpark of a kilowatt per 100 square cm (i.e. a disk about 4 1/2 inches in diameter). That's a reasonably respectable carnot engine. This makes it very useful for things like topping cycles in steam plants: You run it with the flame against the hot side so it is at the combustion temperature, and the "cold" side at the temperature of the superheated steam for your steam cycle. Rather than wasting the energy of that temperature drop (as you would with a pure steam cycle) you collect about a third of it as electricity.
It also beats the efficiency of currently available solar cell technology (and the 33.4% Shockleyâ"Queisser theoretical limit for single-junction cells), if you don't mind mounting it on a sun-tracker. Not only that, but you can capture the "waste heat" at a useful temperature without substantial impairment to the electrical generation or heat collection, and thus use the same surface area for both generation and solar heating. (Doing this with semiconductor solar cells doesn't work well, because they become far less efficient when running a couple tens of degrees above room temparature.)