How nuclear bombs can be made from the fuel of nuclear power stations
An extract from Who’s Watching the Nuclear Watchdog? A Critique of the Australian Safeguards and Non-Proliferation Office (2007) by Richard Broinowski , Tilman Ruff, AlanRoberts and Jim Green.
In criticising the Australian ConservationFoundation/ Medical Association for the Prevention of War publication ‘An Illusion of Protection’, ASNO objects in particular to the passage where the writers ‘assert that reactor grade plutonium has been used in nuclear tests’; ASNO maintains that they have ‘ignored evidence to the contrary‘ which ASNO presented to a parliamentary enquiry.
The main source for the assertion criticised is, of course, the declaration to that effect by the United States Department of Energy (DOE),which needs examining in some detail. It is available on the internet at:<www.osti.gov/opennet/document/press/pc29.html>.
The page is headed ‘U.S. Department of Energy, Office of the Press Secretary, Washington DC, 20585’. There follows immediately the document’s title: ‘Additional Information Concerning Underground Nuclear Weapon Test of Reactor-Grade Plutonium’.
The phrase can be found throughout the document:
“The test confirmed that reactor-gradeplutonium could be used to make a nuclear explosive … weapons can be constructed with reactor-grade plutonium … erroneous statements made elsewhere about the potential use of reactor-grade fuel for nuclear weapons…”
And so on. Altogether, in this document of a little over three printed A4 pages, ‘reactor grade’ occurs no less than 22 times. ASNO apparently sees this as simply an outdated usage of words, and quotes from its own testimony at a parliamentary enquiry:
“There is some confusion over [this test, because] at that time “reactor-grade” was much closer to weapons-grade than is currently the case. While the US has never revealed the quality of the plutonium used in that test, there are indications that it was of “fuel-grade”,an intermediate category between weapons-grade and reactor-grade, which has been recognised as a separate category since the 1970s”.
Why this grading? It is because two quantities of plutonium will behave differently as explosive material, if their percentages of plutonium-240 differ. It has thus been convenient to classify or ‘grade’ plutonium according to this percentage.
But if the material in this particular test was falsely classified, as ASNO is suggesting, this cannot possibly be explained by a ‘confusion’ about grading systems. For in this same document giving ‘additional information’ on the reactor-grade test, the DOE also states:
“Prior to the 1970’s, there were only two terms in use to define plutonium grades: weapon-grade (no more than 7 percent Pu-240) and reactor-grade (greater than 7 percent Pu-240). In the early 1970’s, the term fuel-grade (approximately 7 percent to 19 percent Pu-240) came into use, which shifted the reactor-grade definition [to] 19 percent or greater Pu-240.”
The DOE is thus noting that the new definition had been in force for some three decades. (All these quotations come from the text as downloaded on July 20, 2007.) When it repeats the phrase ‘reactor-grade’, as it does a score of times, we must assume this means what the DOE gives elsewhere in that same document as the current definition namely, having 19% or more of plutonium-240.
If in fact the test material had less Pu-240 than this, the DOE was asserting an untruth and continues to do so. Certainly we should consider this possibility, and look carefully at the evidence against the agency’s truthfulness evidence which, we are told, includes what ASNO presented to the enquiry, and now quotes in its critique.
On the test material it quotes itself as saying:
“[T]here are indications that it was of “fuel-grade”, an intermediate category between weapons-grade and reactor-grade…’ More generally, it stated also that ‘ASNO is not aware of any successful test explosion using reactor-grade plutonium, typical of light water reactor fuel.”
While ASNO’s opinions should receive all due respect, it is hard to see ‘indications’ and being ‘not aware’ as falling into the category of evidence, which ASNO upbraids us for ‘ignoring’.
Its reference to De Volpi (1996) could be more germane, since De Volpi raised many queries about the DOE report that deserve study. It should be noted, however, that his queries were mainly concerned with the report’s omission of certain details about the circumstances of the test explosion, rather than with the grade of the test material. His final, somewhat ambiguous ‘guess’ (the word he uses) is that this grade was ‘closer to the low end (81%) of the definition ‘ – that is, to the ‘19% Pu-240’ mark.
No convincing case appears in all this to reject the DOE’s report that ‘A successful test was conducted in 1962, which used reactor-grade plutonium in the nuclear explosive in place of weapon-grade plutonium.’ Yet, by writing that it ‘is not aware of any successful test explosion using reactor-grade plutonium …’ , ASNO is clearly declaring, with no ifs or buts, that this key statement in the report is untrue.
It is difficult to understand how ASNO can be so certain of the DOE’s dishonesty. Perhaps a clue is given in its concluding paragraphs, on the problems arising with such test material, and its earlier remarks on ‘technical difficulties’ and the need for ‘experienced weapon designers’. The emphasis here on the barriers against weapon development is in marked contrast to the estimates from various people experienced in the weapons field, of whom Carson Mark (Director, Theoretical Division, Los Alamos National Laboratory, 1947-1972) is altogether typical:
“… all of the plutonium isotopes are fissionable. Indeed, a bare critical assembly could be made with plutonium metal no matter what its isotopic composition might be … Reactor-grade plutonium with any level of irradiation is a potentially explosive material …The difficulties of developing an effective design of the most straightforward type are not appreciably greater with reactor-grade plutonium than those that have to be met for the use of weapons-grade plutonium. The hazards of handling reactor-grade plutonium, though somewhat greater than those associated with weapons-grade plutonium, are of the same type and can be met by applying the same precautions.”
ASNO comments, in this concluding section, on the fact that, for the purpose of applying its safeguards measures, the International AtomicEnergy Agency (IAEA) defines the great bulk of plutonium samples as ‘direct-use’ material, that is, ‘nuclear material that can be used for the manufacture of nuclear explosives components without transmutation or further enrichment’.
The general public may not be as impressed as ASNO seems to be, with the distinction which (as it writes) ‘might seem a fine one’, between ‘nuclear explosives’ and ‘nuclear weapons’.
We doubt whether North Korea could have been dissuaded from its proliferative work, if persuaded that the best it could do was impress the world with only a ‘nuclear explosive’. It is not evident, either, that its leadership was greatly concerned with the reliability, storage life and so on of the devices. What they wanted was a big explosion, and they got it. We can be sure also that Osama bin Laden will not be passing up the chance to seize a ‘contaminated’ bundle of fuel rods, out of dissatisfaction with the final product’s shelf life or exact explosive power.
There has been so prolonged a discussion about this particular test that we might wonder what on earth the fuss is about. Is there some bigger question hiding behind it all?
Indeed there is, one that the MAPW/ACF publication states quite clearly. Referring to the Non-Proliferation Treaty (NPT) and the role of a Non-Nuclear Weapon State (NNWS), the document’s Executive Summary (page 7) asserts:
“Article IV enables a NNWS to acquire nuclear materials, technology and infrastructure. However, once such a nuclear capacity is realised the potential for NNWS to acquire nuclear weapons is inescapable, as evidenced by the concern expressed about Iran’s nuclear programme. There are clear examples demonstrating that NNWS can become nuclear weapons capable relatively quickly. By legitimising and encouraging the expansion of nuclear fuel cycle capabilities around the world the NPT has the perverse effect of promoting the means for a cascade of proliferation.”
Why does civilian nuclear power have this dangerous military consequence? Because, inside the fuel rods of the most widespread type of nuclear power station, based on a Light Water Reactor (LWR), there are created large quantities of plutonium-239 – the explosive used in the atomic bomb that destroyed Nagasaki. ‘Large quantities’ here means: enough each year for dozens of Nagasaki bombs.
Two other Energy Science Briefing Papers (numbers 9 and 17) explain how nuclear power stations have been connected, historically and inescapably, with nuclear weapons. Here we indicate why the explosive properties of their fuel itself must be of concern.
There is a useful book describing in some detail the ways in which reactor fuel supplies the nuclear material for nuclear bombs. The publication, Taming the next set of strategic weapons threats, was published in June 2006 by the Strategic Studies Institute (SSI) of the US Army War College jointly with the Nonproliferation Policy Education Center. It can be downloaded free from the site <www. StrategicStudiesInstitute.army.mil>.
The most relevant section is Chapter 5, by Victor Gilinsky: ‘A Fresh Examination of the Proliferation Dangers Of Light Water Reactors’.
The fuel rods of course need some treatment before the nuclear material for a bomb is obtained. Two major points are:
* First, the plutonium must be separated from the other elements (the rods must be ‘reprocessed’); although only chemical methods are needed to effect the separation, the radioactivity of the rods can make special handling apparatus necessary.
* Second, some of the plutonium-239 will have been further converted, after the absorption of a second neutron, into the heavier form (‘isotope’) of plutonium-240; but it so happens that this substance reduces the explosive power of the bomb – thus, from a weapons point of view, it must be regarded as a ‘contaminant’.
Neither of these presents an insuperable barrier to a government seeking nuclear weapons. If it has a nuclear power station, it automatically acquires plutonium and, to process it, the trained work force the station will need anyway.
As long as the fuel rod remains in the reactor, the two kinds of plutonium – the explosive 239, and the ‘contaminant’ 240 – will each be continuously added to. To keep down the level of contaminant, the fuel rod should be removed from the reactor early – that is, before its usual replacement time of, say, five years.
Is it possible to choose this withdrawal time so that the explosive plutonium-239 has been created in quantities usable for weapons, while the percentage of contaminant plutonium-240 is still small enough to leave it with plenty of explosive power? Only too possible, unfortunately.
The simple way to get the nuclear material for bombs from a power station emerges clearly from Gilinsky’s discussion: don’t leave the fuel rods in the reactor for the full five years or so – withdraw them early. There is a range of time in which the plutonium thus obtained will be both plentiful enough for a weapon, and not too severely contaminated.
In one method that he treats in detail, withdrawal is done at the scheduled (and therefore expected) time. But the rods withdrawn (and reprocessed) are not those that have done their appointed service of five years or so, but ones that have spent only about a third of this time, perhaps 20 months, in the reactor.The plutonium in such rods would contain, according to Gilinsky, about 14% of the contaminant plutonium-240. He asked Harmon W. Hubbard to assess the explosive power of the resulting bombs, using publicly available information only. (Hubbard has performed such work for the US Government. The details of his calculations are given in Hubbard(2003).)
The result: a couple of dozen bombs which, although more contaminated, would each give a bomb yield averaging around a quarter or so of full Nagasaki strength. The appropriate unit for measuring this explosive power is tons-of TNT-equivalent. Even the more contaminated case will yield thousands of these units. As Gilinsky comments, the trouble with the idea that the contaminant will make such bomb material ‘fizzle’ is that, with nuclear weapons, “the fizzle yield is still pretty large”.
It should be further noted that it is Hubbard’s lowest estimates which are quoted above. They increase significantly – up to a 12,000-ton-TNT average – if the proliferating country could ‘take advantage of the wide availability of declassified nuclear weapons information and the enormous increases in computing and other technological aids since the 1945 Trinity shot’ (p.83).
All of this means that the common LWR reactor must be regarded as a potent source of material for nuclear bombs. But, to make a bomb, this material would need to be separated from the unwanted material in the fuel rods; does this step of ‘reprocessing’ constitute a formidable – or at least, easily detectable – stage in the progress towards a bomb?
Unfortunately, no. As Gilinsky notes, ‘[T]he feasibility of small-scale, and possibly “quick and dirty” reprocessing of LWR fuel has been known for 30 years.’ He gives no less than four (published) references to designs for the building of ‘quick and dirty’ reprocessing plants, one of them as small as 65 feet square – just under 20metres a side.
The overall conclusion forced on us is a disturbing one: if a country possessing nuclear power stations, of this common type at least, is not manufacturing nuclear weapons, it is only because it has chosen not to do so.
Returning now, in conclusion, to the earlier debate, we see that, ironically enough, the whole argument about whether the 1962 test was reactor-grade or fuel-grade becomes of limited interest anyway. To a significant extent, the grade can be fixed simply by removing the rod from the reactor earlier or later, as Gilinsky’s approach makes clear.
This means that the history of a fuel rod inside a nuclear power station can be sketched as follows:
When the reactor starts up, the rod contains no plutonium to speak of. After a short time – it might be a few weeks or a few months – a significant amount of the weapon material Pu-239 has collected in it, and a small percentage of the contaminant Pu-240. This is the weapon-grade stage inits history, and a few kilograms of it are capable of destruction measured in tens of thousands of tons of TNT. After more months, the fraction of the contaminant has grown, and the explosive power will drop to an average of ‘only’ a thousand tons or so of TNT.
With bad luck or incompetent procedure, it can fall to even less than that. The North Korea test, for example, was reported at around 600 tons of TNT. That was still enough to command ‘respect’, and vastly improved treatment by the US.
It seems very likely that lessons from the North Korea case have been absorbed by other states, both ‘rogue’ and not so rogue. They will have learnt that, to get the iron fist threatening them replaced by a velvet glove, even crude nuclear devices will do. And, if nuclear power is spread more widely through the world, the material for those weapons will often be inside their electrical power station, waiting to be extracted.
When the bomb material Pu-239 is accompanied by a ‘contaminant’ more ready to fission (like the Pu-240), the latter provides a source of neutrons that can ‘pre-initiate’ the chain reaction at an earlier moment than that designed for maximum release of the Pu-239 fission energy. Thus the yield depends on the precise moment when the initiating neutrons enter. Instead of a fixed yield, the probabilities of the various emission times and their corresponding bomb yields have to be calculated.
If you wish to follow and assess procedures of this detailed type, the references to consult below are Hubbard 2003, Lovins 1980 and Mark 1993. These studies have found the kind of results mentioned above: bomb yields from various plutonium isotopic mixtures that have a minimum of around a thousand tons-TNT and usually much more. It would be interesting to have similar calculations from workers who have arrived at happier conclusions based on more than a qualitative belief. They seem difficult to find, however, and we would be glad to learn of any that exist in order to peruse them.
Australian Conservation Foundation and Medical Association for the Prevention of War, October 2006, “An Illusion of Protection”, www.acfonline.org.au/articles/news.asp?news_id=1012&c=254457
ASNO (Australian Safeguards and Non-Proliferation Office), 2006, Reactor-Grade Plutonium: Use In Nuclear Weapon Tests, www.asno.dfat.gov.au/infosheets/rgp_dec06.pdf
De Volpi, Alex, 1996, “A Coverup of Nuclear Test Information?”,<http://units.aps.org/units/fps/newsletters/1996/october/aoct96.cfm#a2>
DOE (US Department of Energy), 2007, “Additional Information Concerning Underground Nuclear Weapon Test of Reactor-Grade Plutonium”, www.osti.gov/opennet/document/press/pc29.html
Gilinsky, V. “A Fresh Examination of the Proliferation Dangers Of Light Water Reactors”, Chapter 5 in Sokolski (ed.), <www.StrategicStudiesInstitute.army.mil>.
Hubbard, Harmon W., 2003. “Plutonium from Light Water Reactors as Nuclear Weapon Material”,<www.npec-web.org/projects/hubbard.pdf>
Lovins, Amory, 1980, “Nuclear weapons and power-reactor plutonium”. Nature, 283, pp. 817-823, (No. 5750, February 28 1980), www.rmi.org/images/PDFs/Security/S80-01_NucWeaponsAndPluto.pdf
Mark, J. Carson, “Explosive Properties of Reactor-Grade Plutonium”, Science & Global Security, 1993, Volume 4, pp.11l-128, www.fissilematerials.org/ipfm/site_down/sgs04mark.pdf
Sokolski, H. (ed.), June 2006, “Taming the next set of strategic weapons threats”, Strategic Studies Institute (SSI) of the US Army War College and Nonproliferation Policy Education Center,