What’s amazing to me is that the kind of graphite used in nuclear reactors doesn’t burn until it’s white hot, somewhere around 1650°C.

Graphite doesn't burn well, but it does erode away in a stream of hot oxygen or CO2. https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/196600... The CO2-based Advanced Gas-cooled Reactor runs at 650C to keep core lifespan reasonable. https://en.wikipedia.org/wiki/Advanced_Gas-cooled_Reactor

That is not a property of the graphite, its a property of the atmosphere (CO₂, He) or lack thereof (H₂O, molten salts) around the graphite. The Chernobyl core started burning as soon as external atmosphere hit the graphite after an explosion.

Actually, it is a property of graphite. You may be surprised.

http://nucleargreen.blogspot.com/2011/03/does-nuclear-grade-...

Translations for SI units:

1000°F = 538°C

2000°F = 1093°C

The SI units are 811 K and 1366 K (or 800 K and 1400 K, using a reasonable number of sig figs.)

Celsius is also an SI derived unit

Oh thanks to both comments - I learned something !

Using liquid water as your moderator really constrains how hot you can go.

St. Vrain was helium cooled and graphite moderated and it only ran at 1430F, which still caused some hair-raising operational problems: https://www.nrc.gov/docs/ML0403/ML040340070.pdf

High temps are very good for efficiency, but are difficult for safety. Only China is building VHTRs right now. If anyone ever builds a supercritical water reactor then it would give you decent efficiency without the problems of exotic coolants.