passive_fist comments on Global catastrophic risks connected with nuclear weapons and nuclear energy - Less Wrong

1 Post author: turchin 21 December 2015 12:20PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (16)

Comments (16)

You are viewing a single comment's thread. Show more comments above.

Comment author: turchin 22 December 2015 10:55:41PM 0 points [-]
  1. Ups, it was a typo in my last comment "it was shown that small pure fission (read fusion) weapons are possible, but their yield is like 5 tons" . it is in wiki:https://en.wikipedia.org/wiki/Pure_fusion_weapon. No plutonium. I think that we can't exclude possibility that such device may be upgraded (for example, using U-238 blanket) and many other tricks. I have some more links in the map and if one is interested he may follow, but I hope that the best results in pure fusion are kept tightly in secret.
  2. Ok.
  3. Yes, one probably may get more dangerous isotopes using U-238 blanket, if we consider chemistry of isotopes and their ability to accumulate in human body. Strontium, plutonium, Iodine and cesium may result from U-238 fission and all on them could accumulate in human body because of their similarity to calcium or other elements. Internal radiation is also much more dangerous.

But the main idea of salted bomb is not about which type of blanket to use - it is very technical question which require large calculations, and it may be found that some kind of blanket is even more effective in killing all humanity than either cobalt or U-238 (think about C-14, polonium, tritium, combinations etc). The main idea of salted bomb is that it is specially design to produce long term atmospheric contamination and that it is very large and stationary. And it is defence or blackmail weapon, not offensive. By large I mean like 20 000 tons dry weight. And 10-100 gigaton of explosive power. It is much more than all ICBM nukes combined. And most of them do not use U-238 blanket as they try to make them more clean.

By the way one may use ICBM to create something like U-238 salted gigabomb, if he use them to attack existing nuclear power stations. I heard that it was a fear North Korea may attack Japan's nuclear power stations with their nukes. I can't and don't want make exact calculation about it.

Comment author: passive_fist 23 December 2015 12:08:47AM 0 points [-]

  1. I see what you are talking about now. Flux-compression driven fusion is most likely not going to work (which explains why there has been no serious effort to pursue it). It's useful to compare it to the Sandia capacitor-powered Z-machine. To achieve fusion you need (a) a lot of energy, delivered in (b) a short amount of time (preferably nanoseconds as the fuel will tear itself apart at timescales much longer than that), in (c) a very small space. The best EPFCG so far has achieved about 100 MJ and 256 MA, but the killer is the time scale, which is on the excruciatingly slow millisecond level. By contrast, the Sandia machine can deliver about 10 MJ and 27 MA on a nanosecond timescale, and is still far from achieving fusion ignition (currently at least two orders of magnitude away). A planned tripling of energy output via a future upgrade is not expected to produce any fusion ignition either. All of this is evidence against the feasibility of EPFCG fusion. To me, it's damning evidence. A quote in that same page you linked says, "the U.S. Is not known to have and is not developing a pure fusion weapon and no credible design for a pure fusion weapon resulted from the DOE investment".

  2. It's not true that ICBMs do not have U-238 blankets. Virtually all modern warhead designs use a U-238 pusher/tamper and some also use a U-238 hohlraum (some omit this in favor of other metals, but the U-238 tamper is still there). I see what you are saying about a 'blackmail weapon' but I don't see how this is any different from the existing MAD doctrine (via nuclear ICBMs, which are more dangerous and more cost-effective).

Comment author: turchin 23 December 2015 09:17:18AM *  0 points [-]
  1. The only way to really prove that pure fusion weapons are possible would be demonstrate them.... But I will answer more generally. We known that tech progress exist and it tends to make things more effective, cheap and widespread. It do it in many domains. In the middle of 20 century we saw quick progress in nuclear weapons. Could we assume that no any progress is possible in this domain? I think we can't take it for granted. The quote from DOE can't prove that pure weapons are impossible, as I would expect that would say the same if they find the way to create them but want to keep in secret. This claim was made the same year - 1998 - as was published an article which describes 3 ton pure bomb (upgradable to 10 t neutron bomb with U-238 blanket) http://scienceandglobalsecurity.org/archive/sgs07jones.pdf It was almost 20 years ago and we don't know if any progress was made on the topic. Personally I think that pure weapons may be created by some unexpected combination, like cold fusion device compressed by Z-pinch.
  2. MAD doctrine kills only two superpowers, but nuclear blackmail kills all human population of the world as collateral damage. The difference is human extinction.
Comment author: passive_fist 28 December 2015 02:30:17AM 0 points [-]

  1. Strictly speaking, the only major barrier to development of fission weapons (once the possibility of prompt criticality was realized) was enrichment. Even a simple gun-type bomb design suffices if you want to build a fission weapon, but you have to get the nuclear material first, and that's where the bulk of the scientific and technological effort in the Manhattan project was focused. Even today, enrichment is still only the major barrier to aspiring nuclear states/groups. Once it was identified that this was the problem that needed to be solved, the scientists quickly came up with a plan on how to tackle it. But there is no plan or pathway to pure fusion weapons. As far as we know, they could be physically impossible. I'm not discounting the possibility of some incredibly secret pure fusion weapon, but if such a weapon existed it would be exceedingly silly to spend billions of dollars on facilities like NIF or the Z machine - and keep in mind that these projects were funded by and do research for the government agencies responsible for nuclear weapons development. What's the point? (Also, cold fusion does not exist.)

  2. Wrong. A country with a sizeable stockpile of nuclear ICBMs can target and kill anyone it wishes. It's not restricted to just bombing the other superpower.

Comment author: turchin 28 December 2015 10:31:29AM 0 points [-]
  1. This is a map of possible risks, not a map of claims. All it says is that if pure fusion (or other simple nukes) will be created it will make situation with proliferation much more difficult. For example laser enrichment is much simpler than traditional and it was recognised as proliferation risk. We can't say how, but tech progress is making nukes cheaper and simpler and it is a problem. https://en.wikipedia.org/wiki/Separation_of_isotopes_by_laser_excitation
  2. It can kill anyone, but not everyone. The world have around 5 million villages and small towns, and you need at least one bomb for each one to kill. On the peak of cold war the world had less than 100 000 bombs. If you really want to kill everyone, you should try something special like artificial nuclear winter or summer.
Comment author: passive_fist 28 December 2015 11:52:51AM *  0 points [-]

I realize that it's a map of risks, I'm just saying the possibilities don't even remotely fall into comparable levels of risk. "Death from nuclear ICBM" is quite imaginable and possible. Not only that, there was a time when it almost seemed imminent and inevitable. And it could easily become that way again. Whereas "death from cold fusion" is essentially of zero meaningful concern.

Maybe it would be useful if you could attach some kind of crude probabilities to your estimates. I can fill a pdf with items like "death from massive leprechaun attack" but it wouldn't be a very useful guide.

Comment author: turchin 28 December 2015 12:37:25PM *  0 points [-]

While I do not appreciate your wording "death from cold fusion" when we speak about risks of proliferation connected with new technologies, I already added some kind of probability estimation to the map and painted boxes in one of three colors. But instead of probability I used "Importance of risks", which more clearly connected with what we should do to prevent them.

"Importance (or urgency) of risks is subjectively estimated based on their probability, timing, magnitude of expected effect and scientific basis for the risk. Importance here means how much attention and efforts we should put to control the risk.

Green – just keep it in mind, do nothing Yellow – pay attention, do reasonable efforts to prevent Red – pay immediate attention to prevent" The pdf is here: http://immortality-roadmap.com/nukerisk2.pdf

In it only two risks are red: nuclear war and nuclear-biological war.

The risks of large scale proliferation connected with new technologies is yellow.

and the risk of Jupiter detonation is green.

Powered by Reddit Powered by Reddit