Redactions

Cleansing thermonuclear fire

by Alex Wellerstein, published June 29th, 2018

What would it take to turn the world into one big fusion reaction, wiping it clean of life and turning it into a barren rock? Asking for a friend.

Graphic from the 1946 film, “One World Or None,” created by the National Committee on Atomic Information, advocating for the importance of the international control of atomic energy.

One might wonder whether that kind of question presented itself while I was reading the news these days, and one would be entirely correct. But the reason people typically ask this question is in reference to the story that scientists at Los Alamos thought there was a non-zero chance that the Trinity test might ignite the atmosphere during the first wartime test.

The basic idea is a simple one: if you heat up very light atoms (like hydrogen) to very high temperatures, they’ll race around like mad, and the chances that they’ll collide into each other and undergo nuclear fusion become much greater. If that happens, they’ll release more energy. What if the first burst of an atomic bomb started fusion reactions in the air around it, say between the atoms of oxygen or nitrogen, and those fusion reactions generated enough energy to start more reactions, and so on, across the entire atmosphere?

It’s hard to say how seriously this was taken. It is clear that at one point, Arthur Compton worried about it, and that just the same, several scientists came up with persuasive reasoning to the effect that this could not happen. James Conant, upon feeling the searing heat of the Trinity test, briefly reflected that maybe this rumored thing had, indeed, come to pass:

Then came a burst of white light that seemed to fill the sky and seemed to last for seconds. I had expected a relatively quick and bright flash. The enormity of the light and its length quite stunned me. My instantaneous reaction was that something had gone wrong and that the thermal nuclear [sic] transformational of the atmosphere, once discussed as a possibility and jokingly referred to a few minutes earlier, had actually occurred.

Which does at least tell us that some of those at the test were still joking about it, even up to the last few minutes. Fermi reportedly took bets on whether the bomb would destroy just New Mexico or in fact the entire world, but it was understood as a joke. 1

The introduction of the Konopinski, Marvin, and Teller paper of 1946. Filed under: “SCIENCE!

In the fall of 1946, Emil Konopinski, Cloyd Marvin, and Edward Teller (who else?) wrote up a paper explaining why no detonation on Earth was likely to start an uncontrolled fusion reaction in the atmosphere. It is not clear to me whether this is exactly the logic they used prior to the Trinity detonation, but it is probably of a similar character to it. 2 In short, there is only one fusion reaction based on the constituents of the oxygen that had any probability at all (the nitrogen-nitrogen reaction), and the scientists were able to show that it was not very likely to happen or spread. Even if one makes assumptions that the reaction was much easier to initiate than anyone thought it was likely to be, it wasn’t going to be sustained. The reaction would cool (through a variety of physical mechanisms) faster than it would spread.

This is all a common part of Manhattan Project lore. But I suspect most who have read of this before have not actually read the Konopinski-Marvin-Teller paper to its end, where they end on a less sure-of-themselves note:

There remains the distant possibility that some other less simple mode of burning may maintain itself in the atmosphere.

Even if the reaction is stopped within a sphere of a few hundred meters radius, the resultant earth-shock and the radioactive contamination of the atmosphere might become catastrophic on a world-wide scale.

One may conclude that the arguments of this paper make it unreasonable to expect that the N+N reaction could propagate. An unlimited propagation is even less likely. However, the complexity of the argument and the absence of satisfactory experimental foundations makes further work on the subject highly desirable.

That’s not quite as secure as one might desire, considering these scientists were in fact working on developing weapons many thousands of times more powerful than the Trinity device. 3

The relevant section of the Manhattan District History (cited below) interestingly links the research into the “Super” hydrogen bomb with the research into whether the atmosphere might be incinerated, which makes sense, though it would be interesting to know how closely linked these questions where.

There is an interesting section in the recently-declassified Manhattan District History‘s that discusses the ignition of the atmosphere problem. They repeat essentially the Konopinski-Marvin-Teller results, and then conclude:

The impossibility of igniting the atmosphere was thus assured by science and common sense. The essential factors in these calculations, the Coulomb forces of the nucleus, are among the best understood phenomena of modern physics. The philosophic possibility of destroying the earth, associated with the theoretical convertibility of mass into energy, remains. The thermonuclear reaction, which is the only method now known by which such a catastrophe could occur, is evidently ruled out. The general stability of matter in the observable universe argues against it. Further knowledge of the nature of the great stellar explosions, novae and supernovae, will throw light on these questions. In the almost complete absence of real knowledge, it is generally believed that the tremendous energy of these explosions is of gravitational rather than nuclear origin. 4

Which again is simultaneously reassuring and not reassuring. The footing on which this knowledge was based was… pretty good? But like good scientists they were happy, at least in secret reports, to acknowledge that there might in fact be ways for the planet to be destroyed through nuclear testing that they hadn’t considered. Intellectually honest, but also terrifying.

The ever relevant XKCD.

This issue came up again prior to the Operation Crossroads nuclear tests in early 1946, which was to include at least one underwater shot. None other than Nobel Prize-winning physicist Percy Bridgman worried that detonating an atomic bomb under water might ignite a fusion reaction in the water. Bridgman admitted his own ignorance into nuclear physics (his area of expertise was high-pressure physics), but warned that:

Even the best human intellect has not imagination enough to envisage what might happen when we push far into new territory. … To an outsider the tactics of the argument which would justify running even the slightest risk of such a colossal catastrophe appears exceedingly weak. 5

Bridgman’s fears weren’t really that the world would be destroyed. He worried more that if the scientists appeared to be cavalier about these things, and it was later made public that their argument for the safety of the tests was based on flimsy evidence, that it would lead to a strong public backlash: “There might be a reaction against science in general which would result in suppression of all scientific freedom and the destruction of science itself.” Bridgman’s views were strong enough that they were forwarded to General Groves, but it isn’t clear whether they resulted in any significant changes (though I wonder if they were the impetus for the write-up of the Konopinski-Marvin-Teller paper; the timing kind of works out, but I don’t know).

There isn’t a lot of evidence that this problem concerned the scientists too much going forward. They had other things on their mind, like building thermonuclear weapons, and it quickly became clear that starting a large fusion reaction with a fission bomb is hard. Which is, in its own way, an answer to the original question: if starting a runaway fusion reaction on purpose is difficult, and requires very specific kinds of arrangements and considerations to get working even on a (relatively) small scale, then starting one in the entire atmosphere, is likely to be impossible.

Operation Fishbowl, Shot Checkmate (1962) — a low yield weapon, but something about its perfect symmetry and the trail of the rocket that put it into the air invokes the idea of a planet turning into a star for me. Source: Los Alamos National Laboratory.

Great — cross that one off the list of possibilities. But it wouldn’t really be science unless they also, eventually, re-framed the question: what conditions would be required if we were to try and turn the entire planet into a thermonuclear bomb? In 1975, a radiation physicist at the University of Chicago, H.C. Dudley, published an article in the Bulletin of the Atomic Scientists warning of the “ultimate catastrophe” of setting the atmosphere on fire. This received several rebuttals and a lot of scorn, including one in the pages of the Bulletin by Hans Bethe, who had previously addressed this question in the Bulletin in 1946. Interestingly, though, Dudley’s main desire — that someone re-run these calculations on a modern computer simulation — did seem to generate a study along these lines at the Lawrence Livermore National Laboratory. 6

In 1979, Livermore scientists Thomas A. Weaver and Lowell Wood (the latter appropriately a well-known Edward Teller protege) published a paper on “Necessary conditions for the initiation and propagation of nuclear-detonation waves in plane atmospheres,” which is a jargony way to ask the question in the title of this blog post. Here’s the abstract:

The basic conditions for the initiation of a nuclear-detonation wave in an atmosphere having plane symmetry (e.g., a thin, layered fluid envelope on a planet or star) are developed. Two classes of such a detonation are identified: those in which the temperature of the plasma is comparable to that of the electromagnetic radiation permeating it, and those in which the temperature of the plasma is much higher. Necessary conditions are developed for the propagation of such detonation waves for an arbitrarily great distance. The contribution of fusion chain reactions to these processes is evaluated. By means of these considerations, it is shown that neither the atmosphere nor oceans of the Earth may be made to undergo propagating nuclear detonation under any circumstances. 7

Now if you just read the abstract, you might think it was just another version (with fancier calculations) of the Konopinski-Marvin-Teller paper. And they do rule out conclusively that N+N reactions would ever be energetic enough to be self-propagating. But it is far more! Because unlike Konopinski-Marvin-Teller, it actually focuses on those “necessary conditions”: what would need to be different, if you did want to have a self-propagating reaction?

The answer they found: if the Earth’s oceans had twenty times more deuterium than they actually contain, they could be ignited by a 20 million megaton bomb (which is to say, a bomb with the yield equivalent to 200 teratons of TNT, or a bomb 2 million times more powerful than the Tsar Bomba’s full yield). If we assumed that such a weapon had even a fantastically efficient yield-to-weight ratio like 50 kt/kg, that’s still a device that would weigh around a billion metric tons. To put that into perspective, that’s about ten times more mass than all of the concrete of the Three Gorges Dam. 8

So there you have it — it can be done! You just need to totally change the composition of the oceans and need a nuclear weapon many orders of magnitude more powerful than the gigaton bombs dreamed of by Edward Teller, and then, maybe, you can pull off the cleansing thermonuclear fire experience.

Which is to say, this won’t be how our planet dies. But don’t worry, there are plenty other plausible alternatives for human self-extinction out there. They just probably won’t be as quick.


I am in the process of finishing my book manuscript, which is the real job of this summer, so most other writing, including blogging, is taking a back seat for a few months while I focus on that. The irreverent title of this post is taken from a recurring theme in the Twitter feed of anthropology grad student Martin “Lick the Bomb” Pfeiffer, whose work you should check out if you haven’t already.

  1. This undergraduate paper by Stanford student Dongwoo Chung, “(The Impossibility of) Lighting Atmospheric Fire,” does a really nice job of reviewing some of the wartime discussions and the scientific issues.[]
  2. Emil Konopinski, Cloyd Marvin, and Edward Teller, “Ignition of the Atmosphere with Nuclear Bombs,” (14 August 1946), LA-602, Los Alamos National Laboratory. Konopinski and Teller also apparently wrote an unpublished report on the subject in 1943. I have only seen reference to it, as report LA-001 (suspiciously similar to the LA-1 that is the Los Alamos Primer), but have not seen it.[]
  3. Teller, in October 1945, wrote the following to Enrico Fermi about the possibility of a “Super” detonating the atmosphere, as part of what was essentially a “Frequently Asked Questions” about the H-bomb: “Careful considerations and calculations have shown that there is not the remotest possibility of such an event [ignition of the atmosphere]. The concentration of energy encountered in the super bomb is not greater than that of the atomic bomb. In my opinion the risks were greater when the first atomic bomb was tested, because our conclusions were based at that time on longer extrapolations from known facts. The danger of the super bomb does not lie in physical nature but in human behavior.” What I find most interesting about this is his comment about Trinity, though Teller’s rhetorical point is an obvious one (overstate the Trinity uncertainty after the fact in order to emphasize his certainty at the present). Edward Teller to Enrico Fermi (31 October 1945), Harrison-Bundy Files Relating to the Development of the Atomic Bomb, 1942-1946, microfilm publication M1108 (Washington, D.C.: National Archives and Records Administration, 1980), Roll 6, Target 5, Folder 76, “Interim Committee — Scientific Panel.”[]
  4. Manhattan District History, Book 8, Volume 2 (“Los Alamos – Technical”), paragraph 1.50.[]
  5. Percy W. Bridgman to Hans Bethe, forwarded by Norris Bradbury to Leslie Groves via TWX (13 March 1946), copy in the Nuclear Testing Archive, Las Vegas, NV, document NV0128609.[]
  6. H.C. Dudley, “The Ultimate Catastrophe,” Bulletin of the Atomic Scientists (November 1975), 21; Hans Bethe, “Can Air or Water Be Exploded?,” Bulletin of the Atomic Scientists 1, no. 7 (15 March 1946), 2; Hans Bethe, “Ultimate Catastrophe?,” Bulletin of the Atomic Scientists 32, no. 6 (1976), 36-37; Frank von Hippel, “Taxes Credulity (Letter to the Editor),” Bulletin of the Atomic Scientists (January 1946), 2.[]
  7. Thomas A. Weaver and Lowell Wood, “Necessary conditions for the initiation and propagation of nuclear-detonation waves in plane atmospheres,” Physical Review A 20, no. 1 (1 July 1979), 316-328. DOI: https://doi.org/10.1103/PhysRevA.20.316.[]
  8. Specifically, they conclude it would take a 2 x 107 Mt energy release, which they call “fantastic,” to ignite an ocean of 1:300 (instead of the actual 1:6,000) concentration of deuterium. As an aside, however, the collision event that created the Chicxulub Crater (and killed the dinosaurs, etc.) is estimated to have released around 5 x 1023 J, which translates into about 120 million megatons of TNT. So that’s not a totally unreasonable energy release for a planet to encounter over the course of its existence — just not from nuclear weapons.[]

5 Responses to “Cleansing thermonuclear fire”

  1. Years later, there’s an echo of this in the concerns that particle accelerators–Fermilab’s Tevatron, Brookhaven’s RHIC, or CERN’s LHC–might create black holes, or something more exotic, and bring about Doomsday.

    There was a guy who filed lawsuits. And sometimes picketed the labs.

    For the Relativistic Heavy Ion Collider, Brookhaven commissioned a report: 1999 press release.
    Review of Speculative \Disaster Scenarios” at RHIC

    Years later, CERN also explained the risks of the Large Hadron Collider to the press.

    These machines have been running for a while now. I’m happy to report that planet Earth still exists.

  2. Jonah Speaks says:

    Ellsberg discusses this issue in Chapter 17 of his recent book, “Doomsday Machine.” He indicates the theoreticians (supposedly) whittled the odds of atmospheric ignition down to 3 chances in a million before exploding the bomb. However, the experimentalists “knew” that the odds of doom because of error or major omission in the calculations could be significantly higher. It was never shown to be impossible.

    He also says there is no historical evidence that the Manhattan Project scientists ever brought this issue to the attention of U.S. civilian authorities, presumably deciding among themselves that this was “acceptable risk.” By contrast, the German scientists discussed this issue with Hitler, and Hitler said no to the bomb.

  3. Spencer Weart says:

    Don’t have the reference handy, but I recall reading that some people witnessing the Mike test, when the fireball just kept growing and growing, momentarily thought that now they had really screwed up…

    • “Something I will never forget was the heat. Not the blast. It was a little scary, the cloud — there is an illusion that if something is very high you think it is on top of you. And although they were at least 20 or 25 miles away, I had the feeling that the cloud was on top of me. But the heat just kept coming, just kept coming on and on. And it was really scary… It is really quite a terrifying experience because the heat doesn’t go off. Now on kiloton shots it’s a flash and it’s over, but on those big shots its really terrifying the heat that comes from those.” — Harold Agnew

      “‘You would swear,’ one sailor wrote, that the whole world was on fire.” — From a collection of eyewitness accounts of the Mike test

      (Both as quoted in Parsons and Zaballa, Bombing the Marshall Islands, on 45-46.