4 ms·

These are certainly superb scientific results. The problem is that even 1 DPA is significant for structural materials, and is basically a nightmare for electri

These are certainly superb scientific results. The problem is that even 1 DPA is significant for structural materials, and is basically a nightmare for electrically or mechanically sensitive ones. For example light-water fission reactors achieve about 50 DPA to their reactor pressure vessels over their 40-50 year lifetime. Its hard to imagine that anything called a super-conductor could maintain Cooper-pairs after getting the average atom displaced from its original position. Additionally CFS ran at 20 Kelvin... which again is hard to imagine happening at scales other than CERN-like efforts (ie one-of-a-kind installs): "In September 2021 Commonwealth Fusion Systems (CFS) created a test magnet with ReBCO wires in which flowed a current of 40,000 amperes, with a magnetic field of 20 tesla at 20 K." [1] Edit: Also magnets have 1/R^3 field die-away, in direct competition with 1/R^2 particle dispersion die-away.... meaning your magnets control better as you make cavities smaller, however it also means the magnets get much more DPA :(. [1] https://en.wikipedia.org/wiki/Rare-earth_barium_copper_oxide https://en.wikipedia.org/wiki/Rare-earth_barium_copper_oxide

pfdietz 5y ago
Magnets are highly shielded from neutrons in a fusion reactor. The real problem is the first wall, which necessarily is not shielded. Materials are an ongoing problem, and it's not just dpa. Abdou at UCLA pretty convincingly argues that increasing dpa of wall materials isn't even the most pressing problem; reliability of the wall/blanket is. http://www.fusion.ucla.edu/abdou/abdou%20presentations/2018/Abdou%20CFETR%20Presentation%2011-27-2018%20Final%20V50.pdf http://www.fusion.ucla.edu/abdou/abdou%20presentations/2018/... I should add that fusion neutrons add a problem beyond that experienced in fission reactors: the neutrons are sufficiently energetic that production of helium by (n,alpha) reactions in materials is a serious issue. The helium migrates to small very high pressure bubbles that break materials from the inside.
- omgJustTest 5y ago
  Temperature & DPA have to be talked about together. I think Abdou's point is that you can replace the blanket more often and target ~20 DPA to reduce the mean-time-between-failures (MTBF). From the PDF: "Results show: anticipated MTBF is hours/days (required is years), and MTTR is 3-4 months (required is days), and availability is very low < 5%" I don't have a problem with talking about MTBF and replacing blankets etc but we are talking about research, not a reactor. Also of note look at the MTBF of magnets and cryogenics in this system! (1.14years and 0.57 years)
  - pfdietz 5y ago
    Abdou's point is that the RAMI problem has to be solved to continue to do research, and solving it has a higher priority than getting higher dpa tolerant materials.
    - DennisP 5y ago
      In ARC the blanket is liquid and the wall is replaceable. Seems like that's a practical solution to these issues. What does seem like a real problem for them is the one you've brought up before: the limited supply of beryllium.
      - pfdietz 5y ago
        See Abdou's comment on molten salt in that slide deck. In particular, the large amount of Be needed is a showstopper for large scale adoption.
- ncmncm 5y ago
  > Magnets are highly shielded from neutrons in a fusion reactor. How? The lithium/beryllium "blanket" (i.e. thousand tons of molten metal in a maze of pipes) is inside the coils?
  - pfdietz 5y ago
    Of course. Look, did you really think superconducting magnets would stay superconducting if you hit them with megawatts/m^2 of neutrons? The heating alone would cause them to cease to be superconducting. If you look at the ARC design, the blanket tank interposes between the coils and the plasma. There's also another neutron shield there (made of titanium dihydride) to further reduce the neutron dose on the magnets. Fun fact: the temperature of the salt in the blanket tank is > the decomposition temperature of TiH2, so in a serious accident large amounts of hydrogen gas could be produced. There's a reason the ARC reactor is 20 meters tall and weighs as much as several WW2 US destroyers. In your typical fusion reactor design, the blanket is not just inside the coils, it's inside the vacuum vessel (shielding the vacuum vessel from radiation damage and direct plasma heating too). This means any leak from the blanket is a leak into the fusion chamber itself, a criticality 1 issue for which redundancy is not possible.

pfdietz 5y ago

Magnets are highly shielded from neutrons in a fusion reactor. The real problem is the first wall, which necessarily is not shielded. Materials are an ongoing problem, and it's not just dpa. Abdou at UCLA pretty convincingly argues that increasing dpa of wall materials isn't even the most pressing problem; reliability of the wall/blanket is. http://www.fusion.ucla.edu/abdou/abdou%20presentations/2018/Abdou%20CFETR%20Presentation%2011-27-2018%20Final%20V50.pdf http://www.fusion.ucla.edu/abdou/abdou%20presentations/2018/... I should add that fusion neutrons add a problem beyond that experienced in fission reactors: the neutrons are sufficiently energetic that production of helium by (n,alpha) reactions in materials is a serious issue. The helium migrates to small very high pressure bubbles that break materials from the inside.

omgJustTest 5y ago
Temperature & DPA have to be talked about together. I think Abdou's point is that you can replace the blanket more often and target ~20 DPA to reduce the mean-time-between-failures (MTBF). From the PDF: "Results show: anticipated MTBF is hours/days (required is years), and MTTR is 3-4 months (required is days), and availability is very low < 5%" I don't have a problem with talking about MTBF and replacing blankets etc but we are talking about research, not a reactor. Also of note look at the MTBF of magnets and cryogenics in this system! (1.14years and 0.57 years)
- pfdietz 5y ago
  Abdou's point is that the RAMI problem has to be solved to continue to do research, and solving it has a higher priority than getting higher dpa tolerant materials.
  - DennisP 5y ago
    In ARC the blanket is liquid and the wall is replaceable. Seems like that's a practical solution to these issues. What does seem like a real problem for them is the one you've brought up before: the limited supply of beryllium.
    - pfdietz 5y ago
      See Abdou's comment on molten salt in that slide deck. In particular, the large amount of Be needed is a showstopper for large scale adoption.
ncmncm 5y ago
> Magnets are highly shielded from neutrons in a fusion reactor. How? The lithium/beryllium "blanket" (i.e. thousand tons of molten metal in a maze of pipes) is inside the coils?
- pfdietz 5y ago
  Of course. Look, did you really think superconducting magnets would stay superconducting if you hit them with megawatts/m^2 of neutrons? The heating alone would cause them to cease to be superconducting. If you look at the ARC design, the blanket tank interposes between the coils and the plasma. There's also another neutron shield there (made of titanium dihydride) to further reduce the neutron dose on the magnets. Fun fact: the temperature of the salt in the blanket tank is > the decomposition temperature of TiH2, so in a serious accident large amounts of hydrogen gas could be produced. There's a reason the ARC reactor is 20 meters tall and weighs as much as several WW2 US destroyers. In your typical fusion reactor design, the blanket is not just inside the coils, it's inside the vacuum vessel (shielding the vacuum vessel from radiation damage and direct plasma heating too). This means any leak from the blanket is a leak into the fusion chamber itself, a criticality 1 issue for which redundancy is not possible.