Jeffrey LewisFire Resistant Pits

Scott Sagan observes that the W80 warhead — you know the type accidentally transported from Minot to Barksdale — doesn’t have a fire-resistant pit (FRP).

Why does that matter?

Well, this is an old discussion, but when an aircraft carrying nuclear weapons have crashed, they’ve spread plutonium creating a nasty, if very localized, clean-up issue. (Hint to local first responders: be very careful using water.)

In 1990, the Drell Panel on Nuclear Weapons Safety recommended that all US nuclear weapons on aircraft have FRPs:

[A fire-resistant pit] would further reduce the liklihood of plutonium dispersal in fire accidents involving warheads equipped with IHE. In particular, current FRPs are designed to provide molten plutonium containment against the (~ 1000° C) temperatures of an aircraft fuel fire that lasts for several hours.


[The Panel recommends that the United States] build all nuclear bombs loaded onto aircraft — both bombs and cruise missiles — with IHE and fire-resistant pits. These are the two most critical safety features currently available for avoiding plutonium dispersal in the event of aircraft fires or crashes.

[Drell’s 1992 testimony has more.]

The recommendation — which, frankly, would have been a major undertaking to implement — was rejected in favor of a decision simply not to transport such warheads by air in peacetime. Ray Kidder explains the alternative:

An alternative to providing bomb and cruise missile nuclear
warheads with FRPs would be to use the following safety measures to reduce the need for them. In peacetime, these measures would prohibit:

(a) The transport of nuclear weapons by air or their deployment aboard aircraft in close proximity to operating runways.

(b) The refueling or engine startup of aircraft with weapons onboard or nearby.

A summary of accidents involving U.S. nuclear weapons systems is provided in Appendix E. The large majority (84%) of these accidents involved aircraft

The flight from Minot to Barksdale, obviously, egregiously violates that arrangement. The Air Force is right to be reassuring — an accident would have not resulted in a detonation or widespread plutonium dispersal.

However, had the B-52 crashed, and the plutonium burned, I don’t think I’d want to be the Air Force spokesperson explaining that the plutonium contamination was limited to the immediate area of the crash site.

Update: A colleague suggests that DOD may still routinely transport weapons like the W80 by air. If so, that is a policy that should be changed.


  1. Bob Massman (History)

    The probability of the Minot event times the probability of the 52 crashing during their transport times a pu fire? Come on – sounds like someone is trying to invent a crisis.

  2. Lao Tao Ren (History)

    In all fairness to Dr Lewis, is that the proper probability calculation?

    Should it really be # of Warheads multiplied by the Sum of the probability of each and every type of incident that would expose it to fire and other risks.

    That might get a bigger number considering how many of these weapons are ‘out there’ scattered in storage, on vehicles / vessels, etc. at any given moment.

    It would be fair to index the anticipated impact of incidents against events that are likely to occur in places where the contamination would have limited impact vs. high potential impact areas.

    ie suppose a vessel bearing nuclear munitions caught fire at a port beside a heavily populated area?

  3. Alex W. (History)

    Of course, if you had asked slightly more than a month ago what the probability of a B-52 accidentally transporting six live warheads would be, anyone who said anything more than nearly zero would have been trying to “invent a crisis.”

    (Of course, it wouldn’t have been the first time that a B-52 crashed loaded with nukes, if it happened — re: the 1968 Thule Air Base Crash.)

    (I’m not saying it is probable, likely, or even possible, but I’m just saying that it’s easy to mistake something having not happened with something never having the possibility of happening.)

  4. woot

    I think the point here is that an accident of the type explained here, however remote, can, like most of the problems involving nuclear weapons, be rather easily avoided if the necessary resources are supplied.

  5. W George

    A good argument for RRW across the shelf….huh?

  6. Bob Massman (History)

    What’s the value of a FRP involved in an aircraft crash? If the pit(s) is breached the vanadium offers no protection. A fire on the aircraft before it’s Airborne is a different story.

  7. Jeffrey Lewis (History)

    W. George:

    I can imagine other ways to solve the safety issues raised here, but I take your point. So maybe, although RRW series won’t get around to the the W80 for a while.

    RRW-1 would replace the W-76 on the Trident system, while RRW-2 would replace warheads on ICBMS. Eventually, though, the W80 could also be replaced. Here is what Jon Medalia at CRS said:

    According to lab staff, one approach is to look at warheads that could be used on SLBMs and intercontinental ballistic missiles (ICBMs). For example, in this approach, the first warhead produced under RRW (here called RRW-1) would be an SLBM warhead but would potentially be compatible with ICBMs as well, and RRW-2 would be designed as an ICBM warhead but would be compatible with SLBMs. In this way, the two warheads would be each other’s backup: if one failed, the other could be deployed on both missile types. This approach would permit two warhead designs to replace four current warhead types (W76 and W88 for SLBMs, W78 and W87 for ICBMs) now in the stockpile… Further in the future, according to lab staff, this approach might be applied to bomber-carried weapons (bombs and air-launched cruise missiles), possibly permitting the replacement of four types of such nuclear weapons (B61 and B83 gravity bombs, and W80 and W84 cruise missile warheads) with two RRW types.

    So maybe RRW-3 or -4.

    Bob asks about the value of the FRP. In the event that the FPR is breached, he’s right — it is of no value. But a crash wouldn’t necessarily breach the FRP. In that event, the pit would allow the wreckage to burn on top of the warhead without dispersing the plutonium.

  8. Bob Massman (History)

    “But a crash wouldn’t necessarily breach the FRP.”

    I concur, but here we go again with the likelihood of the event. I think SNL has done some in-depth studies of this and I think rupture of the pit is very credible.

    Regarding benefits of the RRW, imo, providing a FRP for the W80 is a much lower priority than replacing HMX based HE in the W76 and W88, etc. No, we don’t transport these WH’s via air but they do represent much greater safety/security issues than the W80 and other IHE systems.

  9. Jeffrey Lewis (History)


    I haven’t seen the studies, though that seems very plausible to me.

    NNSA agrees with you about the HE issue. John Harvey, when he was at CISAC, wrote a very interesting paper with Stefan Michalowski arguing that the D-5’s protruding stage put rocket propellant in close proximity to the W76 without HE. He proposed using a modified W89 as an “enhanced safety warhead.” This is, I would argue, the germination of the RRW idea.

    I have other concerns about RRW, but replacing the W76 with something with IHE seems like a good idea to me.

  10. Bob Massman (History)

    For years, the Navy’s argument against using IHE in their SLBM WH’s was that it would affect the throw weight and “why worry about a few more pounds of HMX in the WH when it’s setting on top of a missile loaded with tons of HMX based propellant” (not a direct quote..Just my version of their words).
    This position was always near-sighted because it totally ignored the serious safety and security issues related to sensitive HE WH’s including manufacturing, transportation and stockpile. The Navy’s owns a small part of the total issue.

  11. TrustButVerify (History)

    I think it’s worth pointing out that there are different levels of risk at work here. If you have to move warheads over a long distance, it is statistically safer to move them by airlift than by road. It’s also quicker, and more secure; there’s no need for all the hugger-muggery of a weapons convoy. And then there’s the inevitable question of terrorist activity- it may be unlikely, but someone will inevitably bring it up in the planning session.

    No, I’d rather they moved these things by air. Imagine the trouble of having to find routes around protesters carrying NO NUKES CONVOYS IN MY STATE signs.