Chemical Forums
Specialty Chemistry Forums => Other Sciences Question Forum => Topic started by: Enthalpy on August 26, 2020, 09:16:00 AM
-
Hello everyone and everybody!
As ITER recently celebrated some landmark with the usual spiel of "unlimited clean energy for free", I want to share my disappointment about tokamaks and nuclear fusion.
========== Tritium from external sources ==========
The only conceivable reaction in a tokamak in foreseeable future is D-T fusion, but our environment contains no tritium in useful amount, because the production events are rare and T decays quickly.
T is presently produced in uranium reactors, where excess neutrons irradiate Li, or in Candu reactors, D. This can't be upscaled:
- 1 U or Pu fission provides less than 1 extra neutron, but 200MeV energy
- 1 neutron hitting Li or D produces less than 1 T
- Consuming the T in a fusion reactor produces <20MeV energy
- The fusion reactor would add <10% power to the necessary fission reactors. Useless.
========== Tritium regeneration ==========
Alternately, a fusion reactor would breed its own tritium. Neutron irradiation of Li is considered. The public hears about this, but nothing more.
Alas, T breeding is difficult and has drawbacks:
- Consuming 1 T in a fusion produces 1 neutron
- 1 neutron hitting an Li produces at best 1 T
- There are losses
========== Neutron multiplication ==========
All considered (all possible?) processes seek to multiply the neutrons to regenerate T. A 14MeV fusion neutron breaks a heavy atom like Pb to release several neutrons of lesser energy. Though:
- Papers report a computed 1.15 breeding factor only. Other design constraints, which abund at tokamaks, may very well drop the factor below 1.
- Neutron multiplication is polluting! At identical produced energy, the waste from Pb destruction is as much radioactive as fission is.
While the public may discover these drawbacks, (a part of) the fusion community knows them well and investigated two decades ago the bombardment of heavy atoms by fusion neutrons and the resulting pollution.
I should come back with figures about the pollution.
Marc Schaefer, aka Enthalpy
-
Examples here of pollutant production at a Pb-Li coolant meant to regenerate tritium. Doc in the Handbook of Chemistry and Physics and in Peter Reimer 's PhD thesis:
https://kups.ub.uni-koeln.de/420/1/11v4360.pdf
204Pb makes 1.4% of natural Pb. The neutron doubling process is efficient (2.1b over 5.3b at 14MeV) and leaves 203Pb, decaying in 2.2 days by electron capture of 0.97MeV with γ emission.
206Pb makes 24% of natural Pb. When hit by a 14MeV neutron, it can emit an α to leave 203Hg, decaying in 47 days by β- with a 0.28MeV γ. Section for this production is only 0.7mb over 5.3b at 14MeV but 206Pb is abundant.
Investigating more would bring more cases.
1.4% abundance or 0.7mb reaction section may look rather small, but:
- 235U produces 131I in 2.8% and 137Cs in 6.1% of the fission events;
- Fission of 235U brings 200MeV. It takes 8× more D-T and n-Li reactions to produce as much heat. Combine both, you get as much 203Pb as 131I per MW.
Now, one may argue that isotopes 204 and 206 could be removed from Pb...
- Well, no. Never completely. Changing a concentration by a factor of 10 is already a big effort. But 10× less pollutants is still far too much.
- I'm confident other pollutants are produced by the isotopes 207 and 208, like 204Tl.
- I only checked neutrons with 14MeV as they're emitted. As they thermalise before being used by 6Li, more reactions occur.
- Such reactions look inherent to tritium regeneration.
In a leak of hot coolant, I imagine the 16% lithium ignite in air (or don't they?), with the fire releasing in the atmosphere the contained pollutants.
Marc Schaefer, aka Enthalpy
-
Is there any neutron multiplier better then Pb?
More efficient ones, sure there are: Pu and 235U. They even provide 200MeV in addition to the neutrons. But then you don't need D-T any more.
100Mo and elements near Pb were investigated too. I didn't check their waste nor efficiency, but all usual descriptions prefer Pb.
Poitevin, who conceived tritium breeding blankets, suggests Be. I assert there isn't enough Be on Earth to feed the D-T reaction.
We have >100 000 t of beryllium according to Usgs
https://www.usgs.gov/centers/nmic/beryllium-statistics-and-information (https://www.usgs.gov/centers/nmic/beryllium-statistics-and-information)
https://pubs.usgs.gov/periodicals/mcs2020/mcs2020-beryllium.pdf (https://pubs.usgs.gov/periodicals/mcs2020/mcs2020-beryllium.pdf)
that's the "resources" or exploitable ore. The "reserves", profitable at present economic conditions, are smaller.
If 9g Be produce ~1.4 mole of T that provides 25MeV/atom heat converted to 35% in electricity, the output is 130 TJe/kg of Be or 13 000 EJe.
From Wiki, in 2017 Mankind consumed
https://en.wikipedia.org/wiki/World_energy_consumption (https://en.wikipedia.org/wiki/World_energy_consumption)
580EJ of which 92EJ were electricity.
If other uses continue to need 270 t/year of beryllium, resources cover :
- 100 years of electricity consumption. That's less than coal or gas.
- 21 years of total energy consumption - and D-T fusion claims to replace all hydrocarbons.
Unless someone sees a better neutron multiplier, we have only Pb and its radioactive waste.
Marc Schaefer, aka Enthalpy
-
Earth offers 20 000 000 t molybdenum resource according to Usgs
https://pubs.usgs.gov/periodicals/mcs2020/mcs2020-molybdenum.pdf
of which 9.6% are 100Mo, only 1.7× as many moles as beryllium.
Same conclusion: there isn't enough molybdenum on Earth, we have only Pb and its radioactive waste.
Marc Schaefer, aka Enthalpy