Cool may be a bit of a misnomer. you want to transfer energy (as heat) from the reactor to the generator. To do this you pump a fluid in a cycle. It picks up heat at the reactor and drops it off at the turbine/generator. You want the reactor to get hot and stay hot but if you didn't move the heat it would just heat up until it melted and your turbine would sit there motionless and sad.
But doesn’t the normal cycle of water to the reactor and then to the turbine move heat? Why do you want to cool separately to taking useful heat energy away?
Nuclear reactors have a heat efficiency of ~33% so you need to dump the remaining somewhere. That is 66% of the heat energy remain after the turbine has removed the possible mechanical energy. Just an annoying factor of Carnot's Law.
If you can do district heating or similar it's all good, but you still need a permanent source of cooling capacity with enough redundancy to cool the reactor in case of an unplanned shutdown.
Increasing temperatures increase steam pressure until a rupture occurs, and the water boils away. The steel reactor vessel and structural elements will then inevitably melt, and the reaction will continue until the molten fuel can spread out enough to lose criticality. This is called a meltdown.
This is why many new reactor designs do not use water as coolant but instead molten salt[0] or inert gas[1], to avoid the risk of steam explosion. This alone does not make those reactors inherently safe but it can help keep radioactive material from being violently spread outside the reactor in the event of catastrophic failure.
> Increasing temperatures increase steam pressure until a rupture occurs, and the water boils away.
Generally, no. Chernobyl is the biggest example of this happening, but it only happened because the reactor itself was basically exploding anyway.
Steam pressure doesn't build up gradually, generally. Relief valves -even just burst points- are just too simple and robust. The much more common issue is that hot reactor fuel causes water to break down and release hydrogen which accumulates at the top of the reactor and eventually explodes, and then the water boils off.
> Relief valves -even just burst points- are just too simple and robust
I'm not sure you've heard about this[0] accident:
"The accident began with failures in the non-nuclear secondary system, followed by a stuck-open pilot-operated relief valve in the primary system, which allowed large amounts of nuclear reactor coolant to escape."
As one of the other sibling comments noted, a turbine can extract maybe 33% of the thermal energy if it’s operating at the theoretical maximum Carnot efficiency. The other 66% must be dumped somewhere to avoid accidents.
If you want a temperature gradient then you want the reactor as hot as you can and the condenser as cool as you can, so I still don’t understand why you want to cool the reactor?
The cooling system is also how you extract energy from the reactor. Anything you do to extract energy will have this effect; if it didn’t you’d have a perpetual motion machine.
The cooling sounds like a side-effect then. The goal isn't to cool for the sake of it. You want the energy out of it. If you just want a cooling system you could spray water onto it and let the steam escape. That'd be an effective cooling system. What people really mean is that they want to usefully extract the energy.
Well, about 10% of the heat a reactor creates is from the radioactive decay of fission products, so even when a reactor is “scrammed” (shutdown, no fission occurring) you need to be able to extract megawatts of heat to keep it stable - hence you need powered pumps, access to water, condensation towers, etc.
