Thermodynamics Of Heat Rejection
The second law sets a floor on how much heat any thermal engine must dump; efficiency reduces it but never to zero, so thermal plants always reject heat.
The second law, in one line
No heat engine converts all of its input heat to work. A fraction, set by the ratio of the hot and cold temperatures, must be rejected to a cold reservoir. This is the Carnot limit, and every real engine falls short of it, rejecting even more.
# Carnot efficiency and rejected-heat fraction
T_hot = 833 # K, ~560 C steam
T_cold = 308 # K, ~35 C sink
carnot = 1 - T_cold/T_hot
print(round(carnot,3)) # ideal ceiling ~0.63
real_eff = 0.36 # typical plant
rejected_fraction = 1 - real_eff # heat dumped per unit input
print(round(rejected_fraction,2)) # ~0.64 rejectedA real plant at 36% efficiency rejects about 64% of its input heat. That rejected heat is the cooling load, and evaporation is the cheapest way to carry it, which is why steam plants consume water.
Does the burner escape the second law?
No. The second law applies to any conversion. But it applies to heat engines, machines that extract work from a temperature difference. A direct energy converter is not a heat engine; it decelerates charged particles against an electric field. Its losses are resistive and thermal, but the bulk of the particles' energy is captured without ever being thermalized, so there is no large Carnot-mandated rejection.
The honest residual
Some energy still becomes heat: converter inefficiency, the 5.44% neutron fraction, and magnet losses. That residual obeys ordinary heat-transfer rules and must be cooled. It is simply far smaller than a steam plant's second-law-mandated rejection, which is why the burner's water need is small.