ML20064K404
| ML20064K404 | |
| Person / Time | |
|---|---|
| Site: | 05000142 |
| Issue date: | 12/16/1982 |
| From: | Thomas Taylor COMMITTEE TO BRIDGE THE GAP |
| To: | |
| Shared Package | |
| ML20064K001 | List: |
| References | |
| NUDOCS 8301180387 | |
| Download: ML20064K404 (63) | |
Text
O NUCLEAR RECULATCRY COMMISSION f[y UNITED STAIES CF AFERICA 3EFCRETHEATCMICSAFETYANDLICENSING30ARh 1 ' *,,g/
/
4 e
s,, h In the Matter of Docket No. 50-12 THE RECENTS CF THE UNIVERSITY m
CF CALIFCHNIA (Proposed Reneva Facility License)
(UCLA Research Reactor)
DECLARATION OF DR. THEODORE B. TAYLOR I, Theodore 3. Taylor, declare as follows:
1.
F;om 1949 to 1956 I worked on the design of nuclear explosives at the 'Los Alamos Scientific Iaboratory. From 1956 to 1964 I worked at the General Atomic Division of General Dynamics Corporation, during which period I helped design the TRICA research reactor. From 1964 to 1966 I was deputy director (scientific) of the Defense Atomic Support Agency in Washington.
The following years have been spent as an independent consultant to the U.S. Atomic Energy Commission and a number of other organizations, working on nuclear safeguards issues and other energy-related matters.
I served on the Presidential Commission on the Accident at Three Mile Island.
I am co-author of Nuclear Theft: Risks and Safeguards, as well as a number of other books and articles dealing with nuclear safeguards and related subjects.
A more detailed statement of my professional qualifications is attached hereto.
1 1
2.
It is my understanding that the University of California, as part of l
its license renewal application for the UCLA Argonaut-type research reactor, l
has requested a Special Nuclear Paterials license to possess Highly Enriched Uranium (HEU).
I understand that the initial request was for 9400 grams of uranium-235 at 93% enrichment, but that it has since been amended to 4999 grams at the same enrichment.
l 8301100387 830112 PDR ADOCK 05000 C
l
. 3.
Barring some extraordinary circumstance of which I am unaware, it is my opinion that UCLA's request for kilogram quantities of F3U should be denied.
I know of no reason that could justify the unique safeguards risks associated with grant of a license to UCLA for materially su internationally dangerous.
I come to this conclusion as a former designer of both nuclear wespons and research reactors, and from extensive involvement in the field of* nuclear safeguards.
The basis for this conclusion is as follows.
235 4
Uranium enriched to 93% in the isotope U is weapons-grade.
It is weapons-grade because it is directly usable in nuclear weapons without any further isotopic enrichment.
5 Because it is weapons-grade material, 93% enriched uranium is a potentially attractive target for theft or diversion by terrorist groups or nations intent on acquiring nuclear weapons.
The potential consequences of such theft er diversion are very grave.
6.
A group or nation capable of making nuclear weapons from 93% enriched uranium would not have any significant difficulty in separating the uranium from the uranium-aluminum eutectic in which it is found in flat-plate tffR-type research reactor fuel.
The methods f or doing that are widely published and fairly straightforward.
7.
Such a national or subnational group would also not have substantial difficulty in removing the uranium from MTR-type research reactor fuel that had been irradiated sporadically in a low-power reactor (e.g., a few hours per week at a maximum power of 100 kwth).
It is more difficult than separating the uranium from unirradiated fuel plates, but it is'something that could be done with techniques that are widely published. If the safety rules are compatible with a high degree of military urgency to obtain the uranium for weapons purposes, then the fact that the fuel contained some fissien products would not get in the way very much.
M One such simple method is described on page 97 of John McPhee's The Curve of Bindirs Energy, Random House, NY, 1973, 1974, attached.
_3_
8.
Furthermore, the fact that some of the HEU at UCLA might be in the form of irradiated fuel, either in the core or in storage holes, while other material was in the form of fresh fuel in separate storage, would be little if any deterrent to theft of all the material. Dose rates far in excess of what the UCLA reactor appears capable of routinely producing in its fuel bundles would be necessary, in my opinion, for the fuel to be "self-protecting."
If the calculations in Dr. Roger Kohn's September 4,1982, declaration are correct, fuel bundles in the UCLA reactor drop below 100 rea/ hour (at three feet from an, accessible surface without intervening shielding) after just 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> of shutdown, below 34 rea/ hour after one day, and apparently down to a few rem per hour after a shutdown or storage period of a few weeks.
If correct, and assnaing that the reactor operates only a few hours per week, neither irradiated fuel in storage nor in the core would be self-protecting and would be little deterrent to dedicated individuals or groups intent on acquiring nuclear weapons material.
9 To assert that the nearly 5000 grams of uranium-235 at 93% enrichment in the form of uranium-aluminum flat plate fuel, as requested by UCI.A would be of no interest to someone intent on manufacturing a nuclear weapon would be -
simply incorrect. It is, in my opinion, a credible threat that people might break into a research reactor facility such as UCLA's in order to acquire HEU. Particularly if one supposes a blackmarket and people selling stolen HEU at, say, a. hundred.'thousand-dollars per kilo.
i i
10.
It is my opinion that the original request for 9400 grams of uranium-235 was excessive.
It is my opinien that the current request for 4999 grams is also excessive. Anything more than a kilogram, in my opinion, would be clearly excessive. From a proliferation standpoint there should be a significant burden upon the University to demonstrate that it could not perform the functions for which the reactor was intended without the requested highly enriched uranium. I would say that even more broadly:
highly enriched uranium should not be used in quantities more than a few hundred grams under any circumstances unless there l's an absolutely compelling reason to do so.
I know of no such reason that would be relevant to the UCLA reactor.
I l
l l
. l 11.
It is my opinion that the enrichment requested by UCLA (93%) is i
excessive. For general use, for research, the only advantage in going from 20% enriched uranium to 93% enriched uranium is that the neutron flux may increase somewhat, at the same power level.
Idm't know what that i
increase in flux is for this specific reactor, but'it is very hard to imagine that it would change significantly the nature of the experimental program they would carry out with it if the enrichment were reduced.
Everybody wants to have all the neutrons they can get, but UCLA has already limited itself, by limiting the reactor to 100 kilowatts, ani if they wanted
.more neutron flux they might go to 120 kilowatts, for example. One is l
talking about not anywhere near doubling the flux lar increasing the enrichment.
12.
k.9 kilograms of uranium-235 in 93% enriched form would have, by far, greater potential consequences if stolen than an equal or even somewhat larger amount of uranium-235 in 20% enriched fuel. The critical mass of fully enriched uranium in metallic form with an orrHnav reflector is roughly 20 kg.
It depends on what,the reflector is.
With a more efficient reflector, the critical mass is considerably liss.
For example, critical mass with a thick reflector made of beryllium is approximately 11 kg cf uranium-235 For 20% enriched uranium in metallic form with an ordinary reflector, the normal density critical mass of contained uranium-235 rises to over 50 kg.
It is Possible to make.a nuclear explosive with 20% enriched uranium but it would be very heavy, very inefficient ami difficult to bantile.
You would need approximately three times as much uranium-235 in the 20%
enriched form as you would in the 93% enriched form in order to make a critical mass, but more important than that is the' fact that you would be dealing with three times five or fifteen times as much uranium, which is harder to move, harder to compress, much harder to move, much harder to compress.
2/ These figures are for the critical mass of uranium-235 at normal density.
When compressed, the critical mass is considerably less. As I wrote in Nuclear Theft: Risks and Safeguards (Ballinger, Cambridge, Fass.,1974) at
- p. 20 (attached):
"If, on the other hand, the material is to be used in an implosion type of fission bomb, the amount required may be significantly lower than these quantities.
Faterials that are compressed above their normal densities have a lower critical mass than when they are uncompressed.
In the special case when both the core ani the reflector are compressed by the same factor, the critical mass is reduced by the square of that factor. Thus, when a spherical core and reflector assembly that is initially close to one critical mass is compressed to twice its initial density, it will correspond to About four critical masses."
The maximum compression achievable in an implosion type fission weapon, and thus the minimum amount of uranium necessary to make such a supon, depends upon the knowledge, skill, equipment and facilities of the bomb maker.
+--- -
. 13. The potential consequences of theft of 4.9 kilograms of 93% enriched 18 =
uranium would thus,, in general, Lw w.bly greate,r than thos arising be considena W %.9 5 -
ets em %,
from the theft of a sii)gle kilogra3 at 20% enrichment. A iipwever,
.::M Sefe,S --$
n some
.f =
circumstances the consequences might not be very different, those circumstances being if someone simply needed one kilogram to supplement an amount already obtained or if they needed or wanted some highly enriched uranium, for example, to send to the threatened authority in some kind of blackmail threat to make them take them more seriously. To send a few grams would be so relatively easy to do that it might not have much effect on the credibility of the threat.
To send a kilogram begins to be quite significant and they can simply say we haven't sent you more tocause, as you know, that is a significant fraction of the total amount we would need. We just want you to know we have kilogran quantities of this material and here is a kilogram.- So, in that restricted arena of types of threats, the difference between one kilogram of 20% enriched and several kilograms of 93% enriched is not.very i
much.
So far as what is necessary to make a bomb there is a great deal of difference between those numbers.
14 Therefore, in my view, a request by an institution such as UCIA for nearly five kilograms of highly enriched uranium would be excessive unless there is some overriding reason that I cannot imagine, some reason involvir4 national defense research, and even then I know of no such research which would require HEU fuel for such a reactor. There is no crucial research at university reactors of which I amasare that would require weapons-grade uranium.
I would say, in fact, that for uses of any. kind, any quantity of HEU would be excessive.
I say that because'even in gram quantities there should be a special reason why people must have it because that material in gram quantities could be extremely helpful to people making a threat.
Which might not necessarily be a hoax.
It just reduces the amount of material they need in order to make a bomb, and also establishes credibility for a threat.
15 I have been informed that the original Argonaut reactor operated on 20%
enriched fuel and that a number of other Argonauts have likewise operated on 20% enriched fuel. Assuming that that is correct, I know of no reason-and find it hard to imagine one-why UCLA's Argonaut reactor should not likewise operate on 20% enriched uranium. I have also been informed that
f
_f.
University of Florida's Argonaut reactor was to be involved in testing some 4.8% enriched uranium oxide fuel. Assuming that that also is correct, that kind of fuel would likewise seem to me as something worth pursuing for other Argonaut reactors such as UGLA's because it is even further away from weapons material.
16.
I have also been informed that General Atomic currently has available low enriched TRIGA fuel for conversion of research reactors presently utilizing highly enriched NTR-type plate fuel. I would view conversion of the UCIA reactor to low enriched TRIGA fuel as a favorable alternative to the granting of the request for highly enriched uranium because it is much more difficult to make weapons of any kini out of lower enrichment uranium and the compromises in neutron flux, which I think is the main concern, are not large, if present at all.
17 Besides significant safeguards advantages in making a conversion to low-enriched TRIGA fuel, there would also be significant safety benefits.
There certainly was a major safety benefit in the TRIGA fuel when we designed it and the TRIGA reactor.
'Ihe main feature is a very strong prompt negative temperature coefficient, much higher than other research reactors, which meant.that even if it went prompt critical the rise in temperature would extremely rapidly stop the chain reaction.
That still stands. There is, of course, a level of excess reactivity above which that safety feature of not being able to damage any of the fuel with an accidental excursion is no longer true, ani you can always make it over-critical by design, to make it not have that self-limiting feature, but I don't see any excuse for doing it with any medium power research reactor, up to a megawatt at least.
i I
- 18. The level of excess reactivity at which the prompt negative temperature coefficient ceases to be an effective self-limiting feature would be much, much higher for a TRIGA reactor or a reactor wi*.h TRIGA fuel than for the Argonaut with MTR-type fuel. At comparable levels of excess reactivity, J/ A good description of the origins, intent, and basic operatirs principles of the TRIGA and its fuel can be found in Freeman Dyson's Disturbing the Universe (Harper & Row, 17,1979), pgs.94-102, attached.
i b
l l
the TRIGA fuel would definitely have significant safety advantages.
- 19. The difference in the TRIGA fuel is due to the fact that when the ratio of captures (in the water ani other materials) to fissions (in the fuel) goes up, the reactivity goes down. That effect is produced by a change in temperature in the fuel itself, relative to the cooling water, ani thus requires no heat conduction. It happens instantaneously because the heat is liberated by the fission reaction right in the fuel.
In MTR-type research reactors the heat has to be transferred to the water, which takes a while, to make it expand.
It is the expansion of that water plus some other off acts that have to do with the water haring-to heat up that makes j
the reactivity go down. Because of this time-delay involved with the transfer of heat from the fuel meat to the water in MTR-type reactors, the shutdown mechanism is slower, permitting greater energy release before shutdown for an excursion of the same exponential period and a greater opportunity for fuel melting to occur before the excursion terminates than l
1s true with the TRIGA.
i 20.
That shutdown mechanism in the MTR-type reactors, requiring transfer j
of the heat to the water to cancel the reactivity, can produce effects in the water like boiling or a sudden expansion of the coolant which can, I
in effect, do some damage, even if fuel melting does not occur.
The likelihood of there being changes in the fuel arrangement or in the grid supports or whatever is less for the TRIGA than for the Argonaut, according to the comparative designs when I knew about them. I was f=mmme with the Argonaut design at the time we designed the TRIGA.
21.
In addition to the TRIGA fuel having a negative temperature coefficient that is far more prompt (i.e., it comes into play much faster), the size of the negative coefficient is also far larger, so the excursion is terminated much sooner, providing substantial additional protection against fuel melting.
22.
The conversion to lower enriched uranium fuel, be it TRIGA fuel or flat plate fuel, would likely increase the negative temperature coefficient because of another factor, the Doppler effect.
I can't be absolutely sure there aren't some compensating factors, but the Doppler broadening coefficient contribution to the temperature coefficient will go up and I don't know of any raason why other things would happen to :take it go down.
~
l 23 A reduction in enrichment of the fuel can thus decrease the possibility or consequences of a destructive power excursion cr criticality accident.
Low enriched fuel has a contribution to limiting an excursion, an abating effect, that is the cancelling out of the reactivity because of the increased capturing by uranium-238.
I can't make the blanket statement that for any i
reactor design that it would be safers it would be dependent upon a given change in th: reactivity.
- 24. I understand that UCLA has, in addition to requesting hTU, requested a license for 32 grams of plutonium-239 in a plutonium-beryllium neutron source, and that when the reactor was first constructed the Pu Be source was requestett as a start-up source but was soon replaced with a millicurie radium-toryllium source because the Pu-Be was too strong a source for the intended use. If this is true, I think it would be irresponsible not to ask UCLA whether they still have a significant naed for that plutonium and to have a fairly detailed set of criteria on which to base a judgment as to whether' to grant the requested license for the material.
i 25 I would have security concerns about a 2 curie plutonium source, not in terms of anything like the security that one would argue fcr regarding kilogram quantities of plutonium or highly enriched uranium, but I would want to restrict access to it fairly severely. Two curies is a large source.
It can be used in a threatening way.
Under certain circumstances it might be used to distribute radioactivity.
j
- 26. In Nuclear Theft: Risks anti Safeguards. relevant portions of which are attached hereto as an exhibit, Mason Willrich and I discussed the potential consequences of use of a few grams of plutonium-239 as a radiological weapon We have already stated that plutonium, in the form of extremely 1
small particles suspended in air, is exceedir4 y toxic. 'Ihe total l
weight of plutonium-339 which, if inhaled, would be very likely to cause death by lung cancer is not well knowng but is probably between i
ten and 100 micrograms (millionths of a gram). Even lower internal l
l doses, perhaps below one micro 6 ram, might cause significant shortening of a person's life.
The total retained dose of plutonium that would be likely to cause death from fibrosis of the lung)within a few days is about a dozen nilligrans (thousandths of a gran.
j Y chapters 1, 2, 6. and 7, which address both the issue of plutonium as a potential radiological dispersal weapon and HEU as potential materitl~for a clandestine fission explosive.
g quoted fron pages 24-26
i
) In terms of the total weight of material that represents a lethal dose, plutonium-239 is at least 20,000 times more toxic than cobra venom or potassium cyanide, and 1,000 times more toxic than heroin or modern nerve gases.
It is probably less toxic, in these same terms, than the toxins of some especially virulent biological i
crganisms, such as anthrax germs.
The amounts of plutonium that could pose a threat to society are accordingly very =n=11 One hundred grams (three and one half ounces) of this material could be a deadly risk to everyone working in a large office biilding or factory, if it were effectively dispersed.
In open air, the effects would be more diluted by wind and weather, but they would still be serious and long-lasting.
- 27. In Nuclear Theft, we calculated the area in square meters that could experience lethal inhalation doses and the area that could experience.
significant contamination requiring some evacuation and cleanup, given release of different quantities of plutonium-239 Based on these calculations, release of 32 grams of plutonium-239 could create lethal doses throughout 16,000 square meters of building and significant contamination requiring some evacuation and cleanup of 1,600;000 square meters.
(This assumes release of the material in the form of an aerosol of finely divided particles uniformly in air throughout the bs41 ding and one hour exposure.) By comparison, we indicated in Nuclear Theft that 500 square meters corresponds to the area of one floor of many typical office buildings and 50,000 square meters is comparable to the entire floor area of a large skyscraper.
We concluded: "Even a few grams of dispersed plutonium could pose a serious danger to the occupants of a rather large office building or enclosed industrial facility."
~
28.
The situation for dispersal out-of-doors would be somewhat different.
I We said in Nuclear Theft: "The dispersal in large open areas of plutonium with lethal concentrations of radioactivity is likely to be much more difficult to carry out effectively than dispersal indoors.
. With a few dozens of grams of plutonium, however, it would be relatively easy to contaminate several square kilometers sufficiently to require the evacuation of people in the area and necessitate a very difficult and expensive decontamination operation." We said further:
"After the plutonium-bearing particles settled in an area, they would remain a potential hazard until they were leached below the surface of the ground or were carried off by wind or surface water
_ drainage.
. Thus, in an urlan area witu very little rainfall, a few grams of plutonium optimally aispersed out of doors might seriously contaminate a few square kilcmeters, but only over a very much smaller area would it pose a lethal threat." And we determined that " [a ] variety of ways to disperse plutonium with timed devices are conceivable.
These would 211ow the threatener to leave the area before the material is dispersed. Any plutonium contained inside such a device would not be a hazard until it was released."
- 29. Thus, 32 grams of plutonium-239 can be used in a threatening way, with significant radiological consequences.
On the other hand there are lots of other things around universities that can be used in threatening ways too. And I have never come to a clear set of decia. ions about what to do about gram quantities of plutonium. 2 curies of plutonium-239 is a lot of radioactive material, with a half life that is long enough to be important, and it should be looked after. carefully. Whether security for it should be better or not as good as security for some cultured viruses in a bacteriological laboratory, that I.can't answer. I would say, however, that because there is a radiological hazard involved with the potential use of such a plutonium source as a radiological weapon, unless there is a strong reason for CCI.A to have it, it would be an unnecessary hazazd.
- 30. In terms of what kind of security would be sufficient to satisfactorily minimize the risk of theft or diversion of 4.9 kg of hTU, consistent with the potential consequences of its theft, I would say that it should not be possible for a group of people that are quite knowledgeable about security and how to defeat it to describe to a group of experts a credible scenario for stealing the material.
In other words, I'm thinking in terms of a sort of jury situation, with people who are professionals in one way or another in knowledge of security, thefts, and how they get carried out.
I would say that security would be adequate if, ani only if, such a group of people could not be presented with a theft plan that they thought had a reasonable chance of working.
$ 31.
I would say that an almost exclusive reliance on intrusion alarms tied into a campus police statien would absolutely not be adequate in my view, because that does not say anything about the ability of the campus police to effectively intercept people carrying out the theft. Now, if it happens, if there were an intrusion alarm, between the point of entry and the material, and such physical inrriers in between that they couldn't be credibly penetrated until the campus police and a little force had:gotten there, then I would say that would be effective.
But by definition I'm saying that there is something more than an intrusion alarm, that is, barriers, very heavy containers or equivalent, something that would create a physical delay of some significance, greater than the time that there would be an assurance of getting protective people there. When I say protective' people, I do not nean simply one or two watchaen checking in but a force capable of doing somithing to 'successfully prevent the theft. To argue that detection of the theft or post-theft reporting would be sufficient would be, in my opinion, highly irresponsible and not at all consistent with the potential consequences of theft of 4.9 kg of mi:U, consequences which could be very, very significant.
Failure to adequately protect a6ainst theft, rather than merely being able to detect and report theft,'would be unconscionable.
32.
If 4900 grams of uranium-235 at.93% enrichment in uranium-aluminum fuel plates were stolen, I would view that as having extremely significant potential consequences, simply because of the fact that it was being stolen.
What I maan by this is that it is no longer a hypothetical question. When i
it has actually happened, it means that some people have gone to the trouble t
l to steal that material. Having done so, it could be entering the black market, it could be joined with other materials which are on the black market, it could go in the direction of being the start of the construction of weapons.
I would say that if I heard that there are 4.9 kilograms of highly enriched uranium that had been -stolen, I'd be extremely concerned. That would have international implications of great importance.
. Conclusion 29 I wculd say in conclusion that an application for a license to use highly enriched uranium in quantities more than a few grams should put the burden of proof on the applicant to make it very clear that it is truly necessary to use IEU as opposed to using 20% or less enriched uranium.
I have not seen that case made at UCIA.
I suspect that case has not been made by a number of other installations that have the Jame material.
The reason why this is so important is that.this material can be used for making nuclear weapons, and the hazards of nuclear weapons being produced do not necessarily disappear if the quantities themselves are not sufficient to rake one nuclear weapon, because these sources could be pooled.
30.
I do not know whether the Licensing Board knows what the minimum quantities of uranium-235 necessary to make a nuclear weapon are.
I suspect they dcm't, and it is very important to know these numbers in making rational decisions about the consequences of the theft of those material'.
I would say flatly that I would be very concerned about theft, s
clear evidence of theft, certainly of a kilogram or more of highly enriched uranium, and I think everyone should be.
If you ask how much less than a kilog:am, I really couldn't go into that, and I want to make sure I am not being taken to say that one kilogram.of highly enriched uranium is the minimum quantity necessary to make a bomb. That minimum quantity is not a well defined number at all. It depends on the talents, experience, reouirements and so on of the designers.
31.
From the point of view of a person who has designed nuclear weapons, 4.9 kilograms of 93% enriched uranium is certainly a significant quantity.
There is no question about that.
It sould not be around in situations where theft could occur unless there is some vast overriding reason such as national defense, and I know of no such reason which could be remotely applicable to UCIA.
. 32.
It is my opinion that HEU should be prohibited internationally, and we have the mechanisms set up to do that in the United States. HEU should be prohibited except under conditions that I would say are extraordinary.
The prohibition should come first and the exception should come later.
No one should have that quantity of HEU unier any circumstances for any purpose.
I note that the NRC has recently (August 24,1982) issued a statement of policy committing itself to exercising its licensing authority to reduce, "to the==v4=n= extent possible," the use of HEU in domestic and foreign research reactors. (47 FR 37007). It appears to me very important that the Atomic Safety and Licensing Board carry out that NRC policy in the case of the UCLA application for HEU new pending before it.
- 33. Therefore, it is my opinion that UCLA's application for kilogram quantities of 93% enriched uranium, given what I know about what it can be used for in terms of construction of a nuclear weapon or making of a nuclear threat, and also insed on what I know about the needs of research reactors, should not be granted. If it is important that the reactor centinue to operate because of its contribution to research or education, UCLA should be directed to use fuel of a lower enrichment.
34 I would say further that in every case the institutions that are using research reactors should look very carefully at the risks in using them versus the benefits otn.ained-and the risks are significant-and treat them as potentially very dangerous pieces of equipment.
The danger is especially worrisome when there are kilogram or more quantities of nuclear materials from which nuclear weapons can be.made. The whole institution, not merely the nuclear engineering department, should make that assessment. My advice wou'ai be to avoid having reactors unless you absolutely have to.
35 If you must have a reactor, always use the lower enrichment and keep the lowest quantity of uranium on hand. I doubt that there is any crucial research at a university requiring weapons-grade uranium.
I declare und penalty of perjury that the above is true and correct to the best of my knowledge and belief.
'Iheodore 3. Taylor Executed at Damscas, Yaryland, this Ib day of December, 1982.
e l
l l
l l
^---'P" r
,3-
Professional qualifications DR. TlfEODORE 3. TAYIOR My name is Theodore 3. Taylor.
I am Chairman of the Board of Taylor, Kirkpatrick, Inc., a technical consulting firm.
I received my 3S de6ree in physics in 1945 from the California Institute of Technology and my PhD degree in theoretical physics from Cornell University in,1954.
From 1946 to 1949 I worked as a theoretical physicist at the University of California Radiation Laboratory in Berkeley.
From 1949 to 1956 I was on the staff of the los Alamos Scientific Iaboratcry, working en the design of nuclear explosives.
One of my primary assignments at Los Alamos was to design fission bombs of very==all physical dimensions and mass.
In 1956 I joined the General Atomic Division of General Dynamics Corporation where, along with Biward Teller, Freeman Dyson, and others, I helped design the TRIGA research rt, actor. 'While at General Atomic I was also technical director of the Nuclear Space Propulsion Project (Project Orion).
From 1964 to 1966 I was deputy director (scientific) of the Defense Atomic Support A ency, U.S. Department of Defense.
6 I spent the following two years in Vienna, Austria, as an independent consultant to the U.S. Atomic Energy Commission and several other organizations, working on the subject of international safeguards for nuclear materials.
In 1967 I founded the International Research and Technology Corporation, a ccmpany primarily concerned with studies of the impact of technology on society, which I served as Chairman of the Board until 1976.
i From 1976 to 1980 I was Visiting Lecturer with rank of Professor in the Mechanical and Aerospace Engineering Department at Princeton University.
l In the aftermath of the accident at the Three Mile Island nuclear power plant, I served as a Commissioner on the President's Commission on the Accident at Three Mile Island (the "Kameny Commission.")
I am a member of the American Association for the Advancement of j
Science and the American Physical Society, and have served as a consultant to the Air Force Science Advisory Board, 1955-58, Los Alamos Scientific Iaboratory, 1956-64, Aerospace Corporation, 1960-61, Atomic Energy Commission, 1966-70, Defense Atomic Support A6ency, 1966-69, Rockefeller Foundation, 1977-79, l
and was chairman of the Los Alamos Study Group, Air Force Space Study Commission, 1961.
In 1965 I was one of the recipients of the Ernest O. Lawrence Memorial Award of the AEC for work on the development of nuclear explosives and the TRIGA research reactor.
O
I am co-author of The Restoration of the Earth (1973),
Risks uxi Safeguards Nuclear Proliferation (1977),
The Next Twenty Years (1979)(1974),
Nuclear Theft:
, and author of numerous articles on Energy:
nuclear safe 6uards and proliferation in technical journals and popular media.
A mors detailed account of my activities in the safe 6uards field can be j
found in John McPhee's book, The Curve of 31nding Energy (1973,1974).
I maintain an active "Q" clearance.
S 0
O
9 e g %~
s-19
~
s g,'A p s
my t
I THE C[RVE J
OF BBDING E3ERGY
.c j
John McPhee l
~
l It
'I
.!i 3
i 1
l
(
h i
i BALLANTINE BOOKS
- NEW YORK i
l
)
l --- --. - -
~
s 6
t t
t i
l C
i i
.t'I Copyright @ 1973,1974 by John McPhee AII rights reserved. Pub!Ished in the United States by Ba!!antine I
Books, a division of Random House, Inc., New York, and simul-l3 taneously in Canada by Random House of Canada, Limited, e
Toronto, Canada
'!}t The text of this book oJginally appeared in The New Yort r,
! f.
and was developed with the editorial counsel of William Shawn
- 3 and Robert Bingham.
f Ilbrary of Congress Cate og Card Number: 74-1226 i:;
ISBN 0-345 28000-8
)f This edition published by arrangement with Farrar, Straus and Giroux I,
Af anufactured in the United States of America j
First Ba!!antine Books Edition: June 1975 Fourth Printing: December 1979 First Canadian Printing: July 1975 i
e
. ~,
~ -. - ~. ~
- ~ ~ ~ -
i u
I and say, 'O.K.'-or they file a statement of lack of compliance, and this has to be good enough to stand up much simpler to use " broken buttons"-chunks of me-in court."'
tallic uranium-235-than uranium oxide, im example, "The present safeguards system is not a system for It would be easier to begin with the right form of uranium axide than with uranium hexafluoride. Con-i the future," Suda said. "In order to prove suitable, it l
comitantly, though, a clardestine bombmaker would has to dape the developing industry, rather than play catch-up all the time."
have to settle for what he could get, on a scale of ava;1-What, then, if someone did have a few flasks of ability, and he could use uraniur:,-235 in almost any uraaium hexafluoride? Fully enriched. Took it off a form. There is no absolute need to have uranium metat truck outside a Mcdonald's in Wheeling, West Vir-If the oxide were med, the sacrifice in yield would not d,
ginia. Took it out of a freight room in a New York air-be prohibitive. He odde is a powder, easy,to handle, 1
port. It does not matter where or how the material was easy to pour. It could be packed into a box.
obtained, whether the theft was a hit or an inside job.
I asked if there would not be a density problem in it it is hardly arguable that the material is there for the using a material so relatively fluffy compared to metal.
taking. If Ted Taylor, imagining himself to be the thief, He said, "Any high explosive that you have in the had enough uranium safely sequestered, what would he thing will see to it that the density problem dis-
- ,I do with it to convert it to a form that could be used in appears." %c more he thought about it, he said, the i
a bomb 7 more convinced he had become that the oxide would be
' [; '
In rural Maryland, no asore than thirty miles from particularly serviceable for a crude bomb, and conve-br Washington, a friend of mine has about a hundred l
nient as well, for great amounts of U-235 in oxide forut
~i acres of land with a cabin in the center of it. A stream move around the country.
runs past the cabin, and hillsides that are covered with I asked him what someone would do who wanted to d
deep deciduous forests rise away on every side. De change the oxide into metal.
3 i
)
cabin has a big fireplace, no electricity, kerosene laa-Taylor said he wou d put about four and a half kilo-terns, and a roof that projects six or eight feet over a l
grams of the powder en a sibraag tray in a laboratory front porch, on which there is a table and some chairs.
furnace, and then heat up some hydrofluoric acid in a
.s 3
8 y
Taylor and I went there almost every day for a week or I
stoppered flask. Through a tube ia the t'p of the flask,
't so, sat on the porch, and loohd across the stream and hydrogen-fluoride gas would move into the furnace.
)..
meadow into the woods. The place was convenient. It Ilest the furnace up to five hundred degrees centigrade.
'i was near his home. Ile had a pair of binoculars, with He hydrogen fluoride and the uranium-oxide powder which he followed birds, and a slide rule, with which form water and -canium tetrafluoride, also a powder.
4j he created imaginary weapons. I had notebooks and l
In a ratio of six to one, put uranium tetrafluoride and peneds, the table to wnte on, and a lot of Ictsure time, powdered magnesium into a graphite crucible. Add ro-
' d, because he spoke slowly, if at all, making rure that ev-tassium chlorate as a chemical heat generator. Put the ja, crything he sed was in a context as available to the crucible into a strong steel container. Using electrical i
O world in public print as it was to him from memory.
ignition wire-like the wire in a toaster-get the Nothing he said there crossed barriers of secrecy that temperature of the material in the crucible up to six had not already been taken down He was pursuing, in hundred degrees. At that temperature, the uranium f
8 1
its many possible forms, the unclassified atomic bomb.
tetrafluoride and the powdered magnesium ignite. In I
Here would be a scale of convenience. It would be combustion, they become uranium metal and magnesium y
94 flooride. Iet cool to a hundred degrees. Now spray 95 b
i 2kA ak*Me h iMh. ieEth hM p3MbM1 M&MN' 4b NM th @** U f
M 4M 44 O
-m-__
i n
water on the crucible to bring it to room temperature.
ide. Add powdered magnesium to the UF., burn it, He metal inside is known as a derby. Four kilograms of and vou get a derby of uranium metal."
.i U-235. Repeat the process.
The fuel plates. hat run certain research and test
,1 For the various procedures involved in convertin 8 reactors are thin strips of metal only about two feet d
uranium from one compound to another, or for long and four inches wide. What could someene do bringing it ultimately to metallic form, the necessa Y with a stack of those?
j equipment can be made at home, or can be sought in I l Put them in an aqueous solution of lye and fertilizer.
the Yellow Pages of the telephone directory, under I ne lye would have to be quite pure, though-good so-
" Laboratory Equipment & Supplies," or, for that mat-dium hydroxide.ne fertilizer would be sodium mtrate.
d ter, under " Hardware--Retail." Mest useful of all The fuel plates consist of an aluminum-uranium alloy would be the catalogue of a large chemical-supply sandwiched between layers of uncomplica.ed alumi-num. After five hours in the lye and fertsizer, the house, such as Fisher Scientific, which has branches in most cities. De furnace costs less than a hundred dol-aluminum has dirolved and the uracium is in suspen-y lars. A graphite crucible costs three do!!ars. Hydroflu-
. sion. Add barium nitrate to keep the uranium from dis-y l solving. Den put the whole business into a centn-y oric acids costs seven dollars a quart, and magnesium oxide costs twenty-one dollars a pound. A vibrating i
fuge-say, a six-hundred-dollar centrifuge from Fisher tray is simple to make. n works with a little motor and 1
Scientific. Whirl it there for twenty minutes at eight vibrates like a bed in a Holiday Ica. Uranium oxide on hundred Gs. Pour off the aluminum solution. In the 1
a vibrating tray will mix more readily with hydrogen i" bottom of the centrifuge tubes is solid uranium-235.
i j.
fluoride to form uranium tetra 8uoride and water.
Uranium-zirconium hydride is the fuel for at>out
-},
in 1969, Vincent D'Amico, a safeguards specialist at half the research reactors," Ted continued. "And uranium-zirconium alloy is a step along the, way to
.t the Atomic Energy Commission, got word that an air shipment of fifteen kilograms of uranium hexafluoride, making it. The alloy is stockpiled m ugmficant in a steel cylinder, was missing. He went out to search amounts, depending on business. Fuel for TRIGA 1
the country for it, and eventually found it in a freight reactcts, for example, is made in San Diego and then room at Lopa Airport, in Boston. UP. is the most l
shipped all over the world. If you wanted the pure al-L.
abundant form in svbich fully enriched uranium travels.
loy, you would have to steal at in San Diego. If you h
It comes out of Portsmouth, Ohio, in steel bottles and is want the hydride, go to Bandug, or wherever, or get it i
distributed to conversion plants that change it to oxide in transit. The core of a standard TRIGA contains only i
or metal.
two to six kilos of uranium, so a bombmaker would
- Ilow would you-if you had stolen some+ turn probably have to collect the stuff. If you were stearing i
UF. into metal?"
fuel being shipped, you would have to perform at least i.
" Mix it with carbon tetrachloride in an evacuated four thefts. He reactors use cylindrical rods of hy-nickel container. Four parts of carbon tetrachloride to dride, clad in aluminum or stainless steel. You burn off one part UF.. IIeat the mixture-a stove will do-to a the hydrogen at a thousand degrees, and dissolve the hundred and fifty degrees centigrade. The contents I
zirconium in sodium hydroxide."
react and form uranium tetrafluoride and fluorinated i
"llow much uranium is the least you might need?" I
[i -
carbon chloride. De UF. is a loose cake of solid mate-4 asked him.
kilograms of fully enriched uranium," he sa,s twenty "He classical figure for the critical mass t rial Wash it with weak acid or alcohol. From there, it's i
id. "The the same as it was with the conversion of uranium ox-96 9
97
<4 a e#
_,m O M d d 5 N O.a. 4 4 Mi $ fs#sg-M Q,,, h q g,n ' g g g,
a e
4
- o O
i t:
I 1
{
classical statement is that it takes that much to make a faced e. ht pnncipal requirements in 1942. % ey ig 1
needed nucler and neutrome data-energy estimates, t.
bomb. Hat statement isn't true. It takes much less-and how much less depends on how good you are at l
and so forth. They needed equation-of-state data to es-timate assembhes or explosions. They needed to know making bombs."
l the probability of uuttating a neutron chain. %ey 3
"How much less?"
I "All I can say is it's not a nit-pick. It isn't a matter of needed a way to estimate the dependence of efliciency e
saying twenty and meaning eighteen. It matters a lot on varms parameters-such as the mass of material, a
how much less, but that is classified, and there is noth-ener87 generation, and features of disassembly -or it t i ing we can do about that, I guess. But if someone gets would be impossible to decide if, say, five critical L '
I masses were needed for an effective bomb, or one and hold of the IAs Alamos critical-mass summaries, he
+
J can see how much material is critical in various one-tenth, or whatever. Dey needed to develop numer-
' i-jI '
forms-various ways of shaping the metal, various re-ical techniques for making neutron raultiplications.
I flectors wrapped around it. You write to the National Hey needed hydrodynamic calculations. %ey needed ll Technical Information Service, in Washington, for the e n:Puting equipment. "And, finally," Mark said, "they critical-mass summaries. Hey cost three dollars. In one needed people who could ask the right questions and of them it says that the critical mass varies inversely suggest the significance of the answers when they found i
with the square of the density of the metal and reflec-them--call them physicists, if you want. When the 1
tor. If both the reflector and the core are compressed United States began work, it was well equipped on by the same amount-remember, this is an implosion Item Eight-tha people who could ask the questions- -
system-the critical mass is reduced by the square of and on nothing else. Those people were the constella-d that amount. This is as close as you can sidle up to this tion of Los Alamos. %st is what is assumed is needed now-but it is not so. You now have Items One to class fied point."
t; Dusk had long since come down. We quit for the Seven. You don't need to ask the questions. You need
[ ll l day. The corners of Taylots mouth turned down for a an ingenious fellow, perhaps, but not really all that
~ 1 moment, and be said, "A small group has not had the much so. Ile,is hitchhiking on the talents of others.
d opportunity before to rearrange people and buildings You don't need a lab anymore to measure cross-sec-this way."
tions.,They're all measured and published. If you need
'I ;
Not long thereafter, when we were in Los Alamos, equation-of-state data, you can go over to the high M
Carson Mark talked about clandestine bombmaking, school and find out what it is. Everything is unclassified C;
and be said, "Everybody has it in mind that it would except plutonium. But equations of state for heavy be impossible to do. They say you would need your elements tend to be identical. See the Rare Afetals own Manhattan Project. They speak of the scale of Handbook. Any reactor-theory textbook now wiM te!!
ingenuity, of the required gemust they think of it as a you the probability of initiating a neutron chain. The n..
i tremendous operation. But the context has changed. It work that has been done on maximum-credible reactor i
would not be impossible now. It does not take a fleet ot' accidents wdl tell you what you need to know about ef-Einsteins to accomplish, or even Ted Taylors, for that ficiency. You can get neutron calcu!ations by mail. For matter. It is not beyond reach. It is much within reach.
. hydrodynamic calculations, reaa Richtmyer and von I
There's a great di1Terence between 1942 and now.=
Neumann on how to avoid the discontinuity of shocks.
Mark went on to explain that the people in Project Y As for adequate computers, most airline offices have i
(the Los Alamos part of the Manhattan Project) had them. When people first began work here at Los Ala-9g 99 it l
~ 3 kLrwr r-
%. 1, 6 4 M 4 O. M._.d M t"b @ d W nr hAatwa*** ** ^ 9 ***** ^ *"******, _7 rwi 4
~
---iv
-r w -
a s
I
NuclearTheft:
-.-,...,.e_gy,,,,,,,..........,d,_ -a,_
Risks and Safeguards l
l l
Mas.n Willri.h The.d.re B. Tayl.r ie-l u
,\\
cc
[~' kb c'
Q ":a C E. 3,
{
k 7"> Ub 3 fo5 FP cn
/
u N'
l l
4 C
Ballinger Publishing C.mpany o Cambridge, Mass.
A Subsidiary ofJ.B. Lippiscott Company
GSMarra Lilwory Table of Contents Mb YO ' ?
A2 K'd v i
Published in the United States of Amelica by llallinger Publidiing Company l
Cambndge, Mass.
List of Tables
.l is List of Figuees y
Copyright O 1974 by The Ford Foundation. All rights reserved. No part of this Forewood McGeorge Burkly XI pubhcation may be reproduced, stored in a retrieval system,or transmitted in any form or by any means, electronic, mechanical, photocopy, recording or Preface S. Inngig 77,,,,,,,,
otherwise, without the prior written consent of the publisher. Firsi Printing xm 1974.
AcknowleAynents xv Ubrary of Congress Catalog Card Number: 73-19861 g
International Standard Book Number: 0 88410.07 6i B Intrebn l
l Printed in the United States of America Chapw Two Nuclose Weapons 5
library of Congress Cataloging in Publication Data Overview 5
Resources Required To Make Fission Explosives 10 willrich, htason.
Effects of Nuclear Explosions 21 Nuclear thef t: risks and safeguards-Radiokigical Weapons 24 Pure Fusion Explosives 28 Bibliography: p.
- l. Atomic energy industries-Secunty measures.
j
- l. Taylor.1heodore B.,1925-joint author.
Chapter Three
- ,1:ord Foundation. Energy Policy Project.
Nuclem Fuel Cycles: 1973-30 29 til. Tide.
73-19861 Introduction g g j,7 IID9698. A2W54 29 e
isBN 0 884 0-207-6 lih Power Reactor Fuel Cycles 30
_ 9g _4 pg Research and Development Uses of Nuclear Materials 56 v
I
vi Nuclear Theft: Risks and Safessards Contents vn
- Chapter Four Nuclear Power Scenarios: 1980-2000 59 g
59 introduction Costs of Effective Safegusrds 163 60 Alle. native Mixes of LWR,IITGR,and LMFBR Power Plants Allocation of Safeguards Costs 165 71 Other Types of Fission Power Plants Costs of ineffectisc Safeguard 5 166 74 Contsolled Thermonudear Power (Fusiont Chapter Ten Chapter Five as an Reconenendation*
167 77 U.S. Saf eguards Against Nuclear Theft 167 77 Introduction cconiniendations
,7, 78 Ilistorical Development: Mid-1960s to Present 82 Existing Requirements 100 Appendices Regulatory Approach Appendix A llistorical Background 375 Chapter Six l
Appendix B Foreign Nudear Power: Reactor Types and 107 Forecasts
- 93 Risks of Nuclear Theft Ppendix C U.S. AEC Reg datory Guide 5.7-Cont tot of 107 Introduction
"" "I Access :o Protected Areas Vital Areas, 109 Thel't by One Person Acting Alone and Material Access Areas (June 1973) 203 112 Theft by a Criminal Group APPendia D Risks of Governmental Diversion in Nca-Nuclear-I14 TheIt by a Terrorist Group 88Pon Countries 215 i16,
Diversi m by a Nuclear Enterprise gP
""'"*"'8 237 i17 Diversion by a Political Faction Within a Nation 119 Nuclear Black Market l
Bibliogre$y 24g gg 245 Chapter Seven Nuclear Safeguards: Basic Considerations 121 121 About the Authors 251 The Context 122 Purposes of a Nuclear Safeguards System 125 Functious of a Nuclear Safeguards System 127 Framework Chapter Eight 135 Safeguard Measures 135 Measures to Prevent Theft 148 Detection of Completed Nuclear Thefts 152 Recovery of Stolen Nuclear Materials 155 Responses to Nuclear Threats 159 The Principle of Containment
Tables
^
l TaWes 1
2-1 Tsamage Radii for Various Effects of Nuclear Explosions as 7 unctions of Yield 23 2-2 Lettui and Significant ;'ontarnination Areas for Release of Air Suspensions of Plutoni.Inside Buildings 25 3-I Characteristics of Typical US. Reactors 50 3-2 Forecasts of US. Nuclear Power anu Associated Nuclear Material Flows-l972-1980 52 3-3 Total Masscs of Material that Contain One Fast Critical Mass of Material. if Separated from other Materials and Converted to Metal (Total Mass in Kilograms) 54 1-4 Fuel Cycle Materials Rank Ordered by increasing Weights Required to Yield One Critical Mass of Fission Explosive Material in Metallie Form without Uranium Ervichment 54 3-5 Fuel Cycle Materials Rank Ordered by increasing Weights Required to Provide One Critical Mass of Fission Explosive Material without either Chemical Processing or Isotope Enrichment 55 4-I U.S. Nuclear Power Forecast 60 4-2 Nuclear Power Options.1980- 2(X10 63 4-3 Annual Production Rates of Nuclear Weapon Materials 65 7-1 Possible Nuclear Material Categories for a System of " Graded" Safeguards 130 B-1 Characteristics of Typical Foreign Reactors 195 B-2 Forecasts of Foreign Nuclear Power and Associated Nuclear Material l' lows-1972-1980 1%
B-3 Foreign Nuclear Power, by Country 197 B-4 Foreign Forecast of installed Nuclear Power Capacity (Timmsands of Megawatts) Excluding Conmmnist Countries 198 ix
Foreword Nuclear Theft: Risks andSafeguards x
Is 5 Plutonium ain! Ihghly Enriched Uranium inputs and Outputs for Foreign Reactors, Excluding Communist Countries for the Years 199 1981 -2000 (Tonnes)
Total Plutunium Output arut t1-235 Input (in Tonnes) at the end 10 6 200 of the Year Figures 33 3-l 1.ight Water Reactor (LWR) Fuel Cycle 44 liigh-Temperature Gasamled Reactor (llTGR) Fuel Cyde 3-2 48 3 l.iquid Metal-Cooled Fast Breeder (IAlF11R) Fuel Cycle 208 Secure Access Passageway into hotected Area (Unattended)
C-I 208 Secure Access Passageway at Entrance to Protected Area (Attended)
C2 210 C-3 Secure Access Passageway between Change Rooms 211 C4 Secure Access Passageway at Exit from Material Access Area In December 1973 the Trusices of the Ford Foundation authorized the organization of the Energy Policy Prtiect. In su'rsequent decisions the Trustees have approved supnorting appropriations to a total of 54 million, which is being spent over a three year period for a series of studies and reports by respuasible authorities in a wide range of fields. The hisect Director is S. David Freenun, and the hiiect lus lud tiw continuing advice of a distinguished Advisory Board chaired by Gilbert White.
This analysis of "Nudear Heft" is an early result of the Pniect. As Mr. Freenun explains in his Preface, neither the Foundation not the hoject presumes to judge the authors' specific conclusions and recommendations. We (ki commend tids report to the public as a serious and responsible analysis which has been subjected to review by a number of qualified readers.
This study,like many others in the Project, deals with a sensitive and difficult question of public policy. Not all ofit is easy reading, and not all those we have consulted have agreed with all of it. Nor does it exluust a subject which is complex, rapidly noving, and partly liidden under classificatius both reasonable and unreasonable.The nutters it addresses are of great and legitinute interest not mdy t; those who are investing heavily in nuclear power but also, by their very nature, to every citizen and community in the country, and the perspectives of Ihese interested parties are not likely to be identical.
In this last respect the present study reflects tensions which are intrinsic to the whole of the Energy Policy hoject-tensions between one set of objectives and another. As the worldwide energy crisis has become evident to us all, we have had nuny graphic illustrations of such tensions,and there are umre ahead. Tids is what usually happens when a society faces hard choices, all of them carrying costs that are Imth hunun and material.
But it is important to understand that there is a fundanental difference between present tension and permanent conflict.he thesis accepted by our Board of Trustees when it authorized the Energy Policy Project was tlut v,
4
Chapter One Introduction i
This book is about a narrow but important energy p>licy issue. We analyze the possibility of nuclear violence using fissionable snaterial that inight be stolen from the U.S. nucleas power irmlustry, and we discuss what ca i armi should be done to prevent that from happening.
Nuclear energy is rapidly becoming a nujor source of electric p>wer in alw United States and a growing number of other countries. Nuclear power requires the prmluction, processing, and use as fuel of very large anmunts of plutonium and high-enriched uraniunt. Ilowever, only a few kilograms of these fissionable nuterials are,enough for a nuclear explosive capable of mass destruw m, and tens of grams of plutonium aie enough for a device capable of causing widespead radioactive contamination. Moreover, the design and nunufacture of a crmie nui.; ear explosive is no kmger a thflicult task technically, and a plutonium dispersal device is much simpler to nuke than an explosive.
Therefore, nwasures are necessary to ensure slut the materials intended for use as nuclear fuelare not diverled for use in actsinvolving nuclear stucats or violence. These nwasures, or safeguards, must te effective, because a successful nuclear ifwit cou'd enable a snull group to threaten the lives of many people, alw social order within a nation, and the security of the international comnmnity of nations.
Experts in govermnent and industry have known of IIe security risks inherent in nuclear power for many years. Tiwy have worked kmg and hard to develop safeguards against the dangers of nuclear theft. Ilowever, nuny governnwntal policyr=ukers and industrial leaders in the energy field are only vaguely aware of tim problem, and umst of the general public does not know that it exists.
This study is intended, therefore, to contribute to public under-standing of the technical facts and policy issues involved. We believe that these facts and issues affect substantially Imth the development of nuclear power and I
=
2 Nuclear Theit: Arsks andSafeguards introduction 3 the security of the American peopic. Of course, we hope slut our study will also crininals are no nmie perceptive llun the general public about opporturdties to pursue crininal purposes.
stimulate thought and action anumg exper15.
Obviously, there is no pe fect solution to the poblem of nuclear But flee basic flaw in the avgunwnt against informing tle public is thef t any n ne ilun there is a fisul solutson Io the poblem of coime in society.
that it ignores the nature of the security risks in nuclear power. If the risks were liut there are safeguards which,afimt emented, will seduce sie risk of nuclear temporary, the danger of inspiring nuclear theft might well justify withholdmg V
theft to a very low level--a level which, in our opinion,is acceptable. Moreover, information from the public, llowever, when security risks are inherent in a we are convinced that the costs of effedive ufcgturds will be small compared to long term activity, which is clearly the case with nuclear theft, the public in a the total costs of nuclear power.
denocratic society has a right to know, and timse with knowledge luve a duty A swist of contsoversy has engulfed the nuclear power industry from to inform. Indeed,infornwd public opinion is essential to effective safeguards.
the beginning. power reactor safety, emergency core cooling systems, radioactive A wW ge h M m Hg h q w mm u k effluents, tlwrnul pollution, and radimetive waste disposal have each been the dangers of theft are serious and simuld be brought ta the attentkm of tie public, subject of intesminable legal poceedmss, potracted political maneuvering' Yet,it nuy be argued, to do so at this time is unforturute,since the respmsible costly delays in constructism schedules,and sensational newspaper headlines, agencies of governnwnt and the nuclear power industry itself are already hard We have attempted. to the best of our ability, to make this book an Fessed to deal with reactor safety and environnwntal poblems. The nuclear i
objective statenwn of the is. sues arising in om particular area of risk selated a theft issue is easy to distort and sennaionalize. Groups unalterably opposed to e
the development and use o~ nuclear power-2n area in which we feel qualified to nuclear power could inject the issue into nuclear plant licensing hearings and cunmient. Our purpose is to povide a awans whereby the risk of nuclear theft other regulatory poceedings, and thereby cause further costly and dangerous and the cost of ef fective safegu-;ds can be weiched, along wills other risks and delays in nweting future demands for electric power.
costs against the very large benefits of nuclear power. We hope out wosk will be We find this Ene of argunwn to be w,thout uwait,and chief anums i
useful to all pasticipants in the decisionmaking pocess concerning the role of er reasms is tining. T1w years just ahead povide the last chance to deveksp nuclear
>wer in meeting the needs of our natkm and the world for energy in kmg-term safeguards that will deal effectively with the risks of nuclear theft.
the futu. The pronmters of nuclear power may deplore it-and its critics may Once the omtersal-flows in the nuclear power industry are as eixxnmus as welcome it-as an attack on the U.S. nuclear industry and governnwnt po"cy eycted a few years fran now, W win be = late.Mwem, the faHme to deal affecting the indust y. That is not our intention.
with a poblem of such entical importance to the futwe success of nuclear Our study contains no classified informatkm. Drawing from the Power cannot be justified on the ground that sie industry simultaneously faces wealth of unclassified data available, however,it does describe in general terms several other difficult poblenu.
im>w nuclear explosives and radiokigical devices can be made, where in the nuclear power industry the nulevials for nuking such weapons are pesent,and A third basic argument is that the pusibility of nuclear violence as a why and how vasious groups within society might attempt to obtain such result of nuterial being diverted from imiustry is not a real poblem. Various nutesials and to use them to threaten or cause catastrophic destruction.
reasons are put forth to su'pport this argunwnt.
1his information amt the analysis derived from it are necessary in Souw crperts assert that alune who are alarmed tend to underesti-j order to understand tie security risks inherent in the developnent and nute the d?fficulties of manufacturing nuclear emplosives.or to overestimate she widespread use of nuclear power, and to povide a basis for consideration of willingness of groups witida society to resort to threats of nuss tiestructi<m.
various safeguaids against nuclear theft.
Although we recognize that these are debatable issues,we are camvinced that the But imw much does the public need to know about these matters?
risks are real and serious. This book sets furth the reasons for our camviction i:,
1his question luunts us, and we klieve it merits discussion before proceeding detail.
fur t her.
In addition, it is sometimes asserted that if a crindnal or terrorist To us, the nmst compelling argunwnt against informing the pubh.
EBoup really wanted nuclear weapms, tien the group would be nmre likely to c
almut Ihe sisks of nuclear theft is that such an effort night inspire warped or evil attempt to steal a fm' ished and sophisticated device from tie U.S. ndlitary minds. Scenarios of nuclear hijackings or bomb threats might become self-stockpile than to steal nuterials from which it could make its own crude l
fulfilling prophecies. The argument is especially forceful when acts of terrorism explosives. Our reply is that if military stockpiles are not adequately potected i
are widespread and organized crime appears to be llourishing. Ilowever, it against theft at pesent (and we expess no opinion on this piint), then existing i
igrmres the fact that a large anmunt of infornution in nmch greater detail than Potective measures should he strengthened. Ilut tids diws amt peclude we present here is already in the public skmiain. Moreover, it assumes that potecting basic materials as well.
-. ~
.-_~
- ~ -
4 Nuclew Theft: Risks and Safeguads
%tu %
Finally, it nay be argued that if some group were intent on extrenw Nuh Ww I
violence to achieve its ends, alwre are nuny powerful chemical and biological agents llut can be obtained mee easily and used more effectively than nuclear weapms. We agree that tlwre n many non-nuclear ways to inHict enornmus I
hann on larne numbers of innocent human beings, and we deplore the fact.
Specine threats of nuclear violence should be compared with other lethat threats winen deciding upon an acceptable level of effectiveness for safeguards against nuclear theft. Tiwre nuy well be certain biological agents whose violent uses could create as much damage as plutonium thspersal devices. The risks of nuclear i
explosives are, however, incomparable and unprecedented.
Our hope is that, you who read this book will tlwreafter have a better idea of imw effective yam want safeguards against nuclear theft to be, For in the linal analysis, the level of risk accepted will affect us all and future generatisms as the nuclear age unfolds.The choice is ours to nuke as a nation, and we believe 4
it should he nude on broad ecormmic and social grounds after full public OVt.RVIEW discussion.
The first question we must explore is whether a successful theft of nuclear nulerials from tlw nuclear power industry would puse a genuine llucat. Could i
simw of the nuteriafs used as nuclear fuel in the power industry be used in l
weapons? Are these nulerials present in the industry in forna and quantities that are practical for the illicit manufacture of bombs? If a thief succeeds in neking a nuclear weapon from these materials, how much danuse miglit he cause?
Every educated persam already knows the single nmst essential fact about inw to make nuclear explosives: they wmk. Before the fuss atomic lxmib exploiled in the Trinity test near Alamogordo, New Mexico, in 1945, no one knew for cersain that it would wiwk. There was a possibility siist alie kind of fission chain reaction which had been sustained in the Chicago pile could rmt be accelerated to produce a large exploskm. Indeed, s(mw of the los Alanms weapon design group strongly suspected that Trinity would rmt explode. A
" pool" of yield estinutes nude before the test ranged from little nwwe than the yield of the high exp:osive used to trigger the nur' ear expknion to several tens of kilotons. (A kiloton is a unit of energy equal to the energy released by the explosion of one thousand tons of TNT. A negaton corresponds to the energy released by exploding one million tons of TNT.) The actual yield, ek,se to twenty kilotons, was significantly greater titan nmst of the estinutes nude before the test.
The certainty that an idea will work in principle is a large step toward finding mys to carry it out. Ihwing the twenty.eight years since the Trinity test nmch has happened to make it easier to elesign and fatwicate a nuclear expknive,and to provide a high degree of confiilence that the design will he successful. The first fission explosives built in the USSR, the United 5
~
Nuclear Wrmons 7 6 Nuclear Theft: Risks andSafeguards Kingdom, France,and China apparently worked quite well. A number of nuclear weapon technology and formerly thrector of tie Lawrence Radiation 12 hora.
explosives with design features very different from the Trinity device, including tory in California and ihrector of Defense Research and Engincesing in the the bomb exploded over Iliroshinu, wosked well the Grst time they were ined or ikpartment of Defense:
tested.
. it must be appreciated that the only.hrficult part of mating a Ever since the successful test of tie " Mike" device at Eniwetok in fission bomb of some sort is the preparation of a supply of 1952, it has been krmwn that fission exph>sions can be used to initiate
- ** "*I "I'd*'8"'le purity; the design of the lmmb itself thermonuclear explosions with yields in the megaton range. All govern-ments that have developed fission explosives have also successfully developed Nm M
M W @m h m I
high yield thermimuclear explosive devices. Less ll.an three years elapsed fissionable material itself. The two elements commonly used are between China's first detonation of an A-bomb and its first test of a uranium and plutonium. Each of these elements can exist as isotopes thermonuclear exph>sive device-compared to seven years for the U.S., four for of several different atornic weights according to the number of the USSR, five for the United Kingdimi, and eight for France.
neutrons included in correspond.ng nuclei, as in U-232, U-233, Until 1951, most of the info mation required for the design and U-234, U -235.. U-238, Pu-239, and Pu-240. Not all of the construction of fission chain reacting systems, both reactors and fission isotopes of these elenwnts are suitable for use in a nuclear explosive, in particular, it is important to une a material with nuclei that are explosives (A-bombs), was classified. A large body of this information was c8Pable of undergoing fission by neutrons of all energies, and that declassified in conjunction with President Eisenhower's " Atoms for Peace" r & ase, n the average, more than one neutron upon fissioning. The speech before the United Nations on Ikcember 8,1953, the enactment of the nutmals which possess these properties and can be made avadable Atomic Ene:gy Act of 1954, and the first international Conference on the i
$ d Peaceful Uses of Atomic Energy at Geneva in 1955. Subsequent further
"".h*f, "di "h
4 w-declassification and public dissemination of new infosmation of this type has U-235 or Pu-239 + neutron + 2 fission products + 2 or more been extensive.
neutrons in the initial draft of this book that was circulated to reviewers,we (average) + 2 samma rays (average) included in this chapter a rather extensive set of references to unclassified technical publications that would be available to a fission explosive design effort, The total prompt energy release per fission is about 180 million pasticularly one with the objective of nuking a compact, efficient explosive with electron volts. This means that the complete fissioning of I kikigram (2.2 lb) of U-235 or Pu-239 releases an energy equivalent to about a reasonably predictable yield. Tiw entire draft, including the references, was 17,000 tons of chemical explosive, also submitted to the U.S. Atomic Energy Commission (AEC) for formal Crisical Mu-Ilowever, I kilogram of U-235 or Pu-239 metal, classification review and was determined to contain no classified information.
which as about the size of a golf ball, will not explode by itself.The Nevertheless, a number of the reviewers recommended that the set of references
'" d i$
for tiiis chapter and sonw of the text not be included in the published form of p
d g
this book. They believed tids infornution, though obtainable by a systematic causing a second fission. If, however, the sphere contained about 16 literature search, would provide more assistance to an illicit fission explosive kilograms (35.2 lb) of Pu-239 (della phase) or fifty kikigrams (110 design team than would be prudent to collect together in one publication. We Ib) of U-235, the mass would be critical. That is to say, for each have made appropriate deletions in the published version.We believe, howeser.
fission which occurs, one of the neutrons produced would on the slut the concern about the republication in a book such as this of certain' average cause a further fission to occur. If more material were added, unclassified information and references supports the central point we will the number oi neutrons in tlic assembly would multiply.
The mass of fissionable material needed to achieve a critical mass develop: if the essential nuclear materials are at hand, it is possible to make an is also determined by the type and amount of material placed atomic bomb using infornation that is available in the open literature.
ar und it. This external material, called a tamper, serves to reflect To give the reader some idea of the detailin which fission explosive back into the fismonable matnial some of the neutrons which woulJ design Principles are described in widely distributed publications, and ako to otherwise leave. For example, the presence of a tamper made of provide a point of departure for other parts of th.is chapter, we psesent below a U-238 one inch thick around a sphere of plutonium reduces' the rather extensive quotation from the article about nuclear weapons in the mass required to produce criticality from 16 kilograms to 10 Encyclopatir Americans' by.lohn S. Foster, a well-known expert on nuclear kilograms (221b;.
v e
Nuclex H% pons 9 8 Nuclear Theft: Risks and Safeguards To produce a nuclear explosion, one must bring together an originally classified but now in the public dom'ain includes; the measured and assembly which is substantially above critical, or supercritical. For calculated critical masses ol various lission explosive nulerialsa in various types
~
example, suppose that by some means a nuss of nuterial equal to of tampers or reflectors; the nuclear properties of materials used in fission two critical trusses is assembled, and a neutron is injected which explosives, and practically all information concerning the chemistry and stasis a chain reaction. Within two millionths of a second or less,the metallurgy of plutonium and uranium.
energy developed within the fissionable material will cause it to A fission explosive design team wosking in 19U thus lus available to explode and release a nuclear yield cquivalent to several hundred it, in the unclassified technical literalme, considerably nere on the televant tons of high exphisive. ihe actual y, eld depends on the particular g
i characteristics of the masses and types of nuterialsinvolved.
ngners den the Nnity device was tested.The exception is experimental and initiation of the Explosion-Because a supercritical assembly calculated data related to the actual perfornunce of the non-nuclear components naturally tends to explode, a major aspect of the design is related to of specific bomb assemblies. The mathematical and experimental tools one needs the way in which the material is brought together. The simplest form involves a procedure by which two or more pieces, which by to acquire sucli data, however, are extensively described in the tecimical themselves are subcritical, are brought together. Om can imagine, literature on nuclear reactor engineering, on high explosive teclumlogy, and on for example, a hollow cylinder inside of which two cylindrical slugs the behavior of materials at very high pressures and temgnratures.
of fissionable material are pushed together by chemical propellant.
It is generally known that fission explosi(ms can serve as a trigger to Wlule such an approach can he used to provide a nuclear explosion, a ignite thermonuclear fuels such as deuterium or tritium (which are variant forms considerable mass of fissionable material is required. Nudar of hydrogen, the lightest element). When the atomic nuclei of these light explosives involving considerably less fissionable material use a elements fu'se together, huge amounts of energy are released. A considerable technique by which the nuclear maternal is compressed, or imploded.
A simple picture of this so-called implosion technique can be thernonuclear explosives (liknybs) has been made public,especially the results gained by imagining a sphere of fissionable material and tamper of iensive unclass:fied work m the United States and other countries on m
which is slighth below critical. Under these conditions, a neutron controlled thermonuclear (fusion) reactor systems. 'Ihe basic design pt.nciples born in the central region of the fissionable material has almost an even chance of producing a fission before it leaves the metal. If the for thermonuclear explosives, however, remain classified.Ilow long the " secret" assembly is now compressed to twice the original density, the radius of the ll-bomb will be kept cut of the public donnin is speculative. There are is then reduced to about 8/10 of its initial value. A neutron leaving thousands of people who know and understand the basic principles from the central region under the compressed conditions must pass personal expcoence working within the security classification systems of the five through atoms which are more closely spaced by a factor of two.
nations that have tested li-bombs, and their number co:it!nues to increase.
although the total distance is reduced only 20 percent. Conse-Further unclassified development of controlled 6ermonuclear power concepts is quently, the chance of causing a fission is actuaIly increased by also bound to nuke access to classified information kss important to an likmb approximately 2 X 0.8, or 1.6 times.The assembly,s now obviously de@ tem as tim pt h a mN R m mmW uMd h &
i very supercritical,although only one critical mass was used*
11 bomb " secret" will not be kept from public view through the end of this The trick, of course, is to compress to several times normal
- "I"'I' density the mass of fissionable material and tamper. This requires Since, however, it is impossible to discuss fission-fusite explosives in pressures above 10 million pounds per square inch. Such pressures any delad in an unclassified publication, we have concentrated our attention on can be developed through the use of high explosive. The nuclear core fission explosives in this luok. Furthernere, as long as some kind of fission could be placed in the center of a large sphere of high explosive.
Compression of the fisaionable material :s httained.by lighting the explosion is required to ignite the thermonuclear fuel in an Il-bomb, the controls I
outer surface of the high explosive simultaneously at somethinglike 100 points spaced roughly evenly over tLe surface. This procedure
- d'0"*"II"" explaive meterials" to mean those materials that, without produces a roughly spherical, in-going detonation wave which, on fuather chemical processing or tantope separation, can be disectly urd as the core neaterial l
strik.ing the metal core, provides the necessary compression to lead for fission explosives. We define " nuclear weapon ninterials" to mies. those neaterials that to a nuclear explosion.
can be used as the cose matesial for fission emplosives after chesnical conversions involving processes snuch sisapter than chemical reprocessing ofirradiated nuclear materials or inoto,e 8eparahon. llence, fission explosive materials is a narrower teeni than nuclear weap,n This eneyclopedia article presents a description of the general materials. As we stealt see, einese two casesones of nesclear materials are the prianary concern principles for the design of nuclear explosives. In addition, information of this study.
r 0
u 10 Nuclear Theft: Risks andSafeguards Nuclear ninpons 11 1
to prevent illicit use of fission explosive nuterials have a direct bearing on the tor any of a variety of purposes and under diverse circumstances. In view of this control of illicit production of Il bombs. Finally, the danuge slut could be situation, we concentrate in the folkming paris of this clupter on a discussion of inflicted by lission explosions provides, we beheve, sufficient,iustification for the mininam time ard resources acquired to make a lission exphnise with a ellective safeguards designe I to pevent theit or illicit prmhsetion of fission
)ictd that could be expected to be equal to at least a few tens ut tous ut high explosive nutenials. The possibility of pure fusion explosives is discussed beiefly ex plosive.
at tiw end of this chapter.
Fission Exploswe Meeerials Nuclear noterials do not necessarily have to expkale to cause severe danuge over large areas. Some radioactive materials, includmg nuny that are A material umst have certain charscleristics to be usable disectly in poduced in nuclear power reactors, are anxmg tie most toxic substances the core of a fission bomb. First of all.it must be capable of sustaining a fission known. Radiologi at weapons that would disperse fission poducts or other chain reaction. This nwans the nulerial nmst contain isotopesh that can be split j
fissioned by neutrons, releasing in stun nere than one neutson as a
[
I radioactive materials have been seriously considered for nnlitary use. We have no or evidence, however, that any government has found such weapons to be casequence of fissioning. Second, alw average time between the " birth" of a i
sufficiently effective, compared to chemical or biological waifare agents and neutron by fission and the time it produces another fission, called the neutron i
other weapons lincluding nuclear explosives), to include them in military
" generation time," must be shmt compared to the time it takes for pressure to arsenals. Nevertheless, we have considered several types of radiological devices build up in the core.Too inuch pessure early in the chain reaction can cause the i
tlut might be used by terrorists or ollwr non-governmental groups-or perhaps core to expand sufficiently to become sub-critical,i.e., to lose so nuny neutrims even by individuals-to expose large numbers of people to radiation or to cause by leakage from ahe surface that the chain reaction cannot be maintained. Third, the evacuation of urban areas or major industsial facilities. We have given ile critical nuss and volume of the fission explosive nulerial nmst be particular attention to possibilities for dispersing plutonium since that material sufficiently small so that the siie and weight of the mechanism for assembling 6
1 is present in large quantities in nuclear power fuel cycles and is exceedingly toxic note Ihan one caitical mass-whether based on the " gun" or "imphasion" if breathed into the lungs in the form of very snull particles, design-will be suull enougli to suit the purposes of those who want to use the fission bomb.
RESOURCES REQUIRED TO MAKE The quantities of lission explosive materials that would be required FISSION EXPLOSIVES to make nuclear explosives, and the poblems an illicit bomb net er would face t
r f
in using them, depend on which lission explosive nuterials are involved. The Objectives distinctive characteristics of each lission explosive material must be understood ne tinw and resources required to design and malie nuclear m order to deternnne wh'ere in the nuclear puwer industry the key materials are explosives depend strongly on the type of exploaive wanted. it is much more to be found, to assess the specific risks of nuclear theft, and to decide whicli i
difficult to make large numbers of rehable, efficient, and lightweight nuclear safeguards measures are appropi.,te in particular circumstances. We shall briefly I
warheads for a national military program than to make several crude, inefficient summarire sonw of the nest impxtant characteristics of plutonium, uranium nuclear explosive devices with unpredictable yields in the range of, say, that is inghly enriched in the isotope uranium-215, and uranium-233 fall Duee i
one hundred to several thousand tons of ordinary high explosive. His is one of these materials are or wdl he used in inge quantities as nuclear fuel to reason why experts in the design and construction of nuclear explosives often poduce electric power, and all sluce can te used, separately or in combinatiims' disagree with each other about how difficult it is to make them.Those who have with each other, to make fission expkisives.
wo ked many years on the developnwnt of nuclear warheads for ever more sophisticated nuclear-tipped nsssile systems often base their opinions on their b n element sisch as maniune or hydrogen occurs in a nuneber of detierent A
I own experience, without having thmight specifically about nuclear explosive
'i= tores." His sneans that different asumic nuclei uf she etenient niay contain different devices flut are designed to be as easy to make as possible. Unkke nmst national
"""6"' d "'"ons, although the ammher M procons in the nuclei and the nuniber and I
anangesnent of the electroies revolving attend the neoclei wi11 he the same. The numbers of l
governments, a clamlestine nuclear bomb nuker may care little whether gs Inotons and ciectrons bonind toereher interly determine stic cammcal properases of the bombs are heavy, inefilcient, and unpredictable. They may serve his purposes so esement, which weit be bene 6assy the sa.ne regardiess of she isotope invos,ed, sto,e,c,,
l kmE as theY are transportable by autonmbile and are very hkely to explode with ent intopes d the sam element may have very shfferent nuclear properhes. llence.
n isotopes of uraniuni, a very heavy ciennent, are lekely so split or fisskm mhrn struck r
a yield equivalent to at least 100 tons of chenu. cal explosive.
by a neutron, whise une of the isotopes of hydrogen, a very hght element, are hkely to Thus, aside from the essential fisshm explosive maierials, there is a
=nbine or fuse sugestier under eensain cienhties. Both lission and fuskun reacesons omvers
[
""""8" * '8F*
wide range of resources required to nuke different types of nuclear explosives I
a t
P i
i I
Nuclear Weapons 13 12 Nuclear Theft: Risks and Safeguards Another cluracteristic of plutonium that lus considerable im-At the outset, it is useful to bear in mind that, of the tiuce basic constituents of the nudear age, neither plutonium nor uranium-233 occur in pntance in :he camstruction of a nuclear explosive is tlut, with poper
~~:e contains less precautions,it can be handled safely. The products of plutonium -239 and 240 ruture in significant anmunts, eid uranium as,
radioactive decay are painutill hehum nuclei called "alplu particles? These tiun one percent uranium-235.
puticles luve very suull penetrating p>wer,a untlimeter or less in hunun tissue, Plutoninm. Plutonium is pimiuced in nuclear reactors that contain compared to the very high enerity x-rays, or "gamr a says," ilut are emitted in uraniunr238, the nut abundant isotope of natural uranium. Neutrons released large numbers by many other radioactive isotopes. Plutonium-241 prinurdy m the hssion process are ccptured in uranium-238, forming uranium-239. This emirs electrons, or " beta rays." whkh also have very little penetrating power.
radmaciively decays, with a lulf-life' of alume twenty minutes, to nep.
And the spimtaneous lission r.cutrons produced in plutoniu n-240 are too few sumum-239, whith subsequently also decays, with a lulf-life of a little nane to constitute a radiological hazard. As a censequence of these characteristics, than two days, to form plutimium-239.1his isotope of plutonium is relatively plutonNm can be a severe radiological hazard only if it is retained inside the very : table, with a half-ble of nuire ilun 24,000 years it is the plutonium hunun body, especially in the lungs.
isotope ol greatest interest for use as the cme nuterial in tission explosives.
Airborne plutonium partkles, snull enough to be barely visible, are Another isotope that is nude in nuclear icactors, plutonium-240,is among the nmst toxic substances known. Inlulation of particles the site of discussion. A plutonium-239 nucleus occasionally specLs of dust and weighing a total of some ten nnitionths of a gram is hLely to also i:npm tant to our captmes a neutron without lissioning, to piinluce plutoniu n, -240. Plu-cause lung cancer. A few thousandths of a gram of snull particles of plutonium tonimn -240 cannot be tissioned by neutrons of all energies. This isotope, (taken together, aboue the size of a pinhead),if inluted, can cause death from uistead of tissioning, is none likely to capture another neutron, resulting in fibrosis of the lungs within a few weeks or less. As long as it is not breathed in or plutonium-241. Thus, plutonimm-240 tends to act as a " poison" in a chain otherwise injected into the bloodstream or critical organs, however, large reacting system and it cannot be used,by itself,as the core material for a fission quantities-many kilograms-of plutonium can be safely lundled for hours without any significant radioh>gical hatards. Therefore, plutonium tlut is being explosive.
Plutonium-240 lus another propesty that is important to a bomb processed must be always kept inside some kind of airtight container such as a designer seeking to use plutonium nude in power reactors. It occasionally plastic bag or one of the increasingly familiar " glove boxes" that are standard tissions spontaneously, witinut being struck by a neutron, and in so doing, emaipment in laboratories that handle highly toxic nuterials. In simit, plutoniunt releases several neotwns. The neutron production rate resulting from sponta-must be handled with considerable respect.
neous irssmn nuy oc sulficient to innuence the cluin reaction. Under sonne The optinul chemical form of plutonium to use in :lission bomb is conditions, one of these neutrons might start a fission chain reaction in the core generally the pure metal. Metallic plutonium occuss in several different " pluses" nutenial of a fission bomb before the core is assembled into a highly compressed, with different densities. So-called alpha pluse" plutonium (which las nothing superciitical state. This might cause the tmmh to "predetonate" and selease to do with alpha particles) has a density about nineteen times greater ilun water cimsiderably less energy tion it would if the start of the cluin reaction lud been at nornul pressure, while delta-phase plutonium is about sixteen times nmie dense than water. T he critical nuss of a sphere of dense alplu-pluse fm ther delayed.
't he relative annunt of plutonium-240, compared to plu.
plutonium-239 inside several inches of berylhum metal (an especially gimd tommn-239, ineseases with the length of time the plutonium is exposed to neutron renector)is about four kilograms and atmut the size of a baseball. The neutrons m a nuclear reactor. In typical ower reactors now in operation m the critical nuss of a sphere of delta pluse p.utonium tlut contains percentages of l
timted States, plutonium-240 accounts for 10 to 20 percent of the plutonium plutonium-239, 240, and 241 typical of plutonium nude in today's nuclear in the fuel assemblies when they are renoved from a reactor for reprocessing.
luwer reactors is about eight kilograms when it is insMe a several-inch thisk this concentiation is sulticient to nuke the piesence of plutonium-240 an reflector of steel or copper (neither of which is as good a neutron reflector as heryllium).
impoitant consideration in the design of a fission immb. But it does not preveat Plutonium imide, which is used as fuel nutesial in some types of the plutonimn produced in nuclear power reactors hum being usable in fission bombs llut wouhl be very hkely to pinduce explosions in the kiloton range.
nuclear power reactors, could also be used directly in a nudear explosive. The oxygen in plutonium oxide, which has the chemical formula pug, affects the 2
ability of the plutonium to sustain a rapid chain scaction in several ways. The cIhe half hfe of a sadioactive notope is the average time required for half of a "Xygen takes up space, thereby reducing ihe number of atoms or piuionium per pven etuanmy of the isotope to decay and fosm some other isotope.
=
i i
Nuclear Wemons IS 14 Nuclear Theft: Risks andSafeguards culuc centimeter. This tends to increase the critical nuss, since a neutron must intinutely mixed with an oxide of uranium tlut is shghtly emiched with travel further slan it would in plutamium metal before making a fission. But l
uranium-235. Whether or not such an oxitle mixtme could he uwd, even in oxygen atoms are much mme effective tisan the much Iwavier plutonium atoms
{
pinciple as the core nuscrial for a fission bomb depends im the relative in slowing down neutrams by billiard ball typc collisions. In the language of concentrations of plutonium and uranium. Mixed uranium-plutonium oxide fuel nuclear engineers, oxygen is a neutron " moderator." Since tSe probabihty that a suitable liv use in the kinds of p>wer reactors now operatirg in the United neutron will cause a fisshm in plutomum-239 tends to increase as the neutron States lus much toe low a cimeentration of plutonium (in the range of I to 5 percent) to make the fuel material directly usable in a fisshe txnnh 'the j
slows down, this effect of the pesence of oxygen (or some other moderator) pucesses necessary to extract the plutonium from such a mixture,in the form i
tends to decrease the critical nuss. But the increase in fission probability resulting from slower neutron vehicities cannot compensate for the effect of the of reasonably pme plutonium oxide,are less complicated than tbsc required to i
decreased concentration of plutonium atoms contained in plutonium oxide, so reduce plutonium oxide to metallic f or m. and they are also tiitwouglily described slut the net effect is that the critical nuss of the oxide is somewhat greater than in unclassified publications. Once luving separa ed the plutonium oxide from the uraaium oxide, an illicit Inimb nuker would fe,e the sa,a chice we slut of plutonium metal. When well compacted, plutonium oxide tus a critical nuss that is about one and a half times as large as the critical nuss of metallic peviously described.
Mixtures of plutonium and uranium oxides suitable for use in the plutonium.
A particul.-e number classembled critical masses of plutonium oxide kind of " fast breetier" reactor m>w under intensive development could, in pinci ie, be used without furtlwr c!smical separation as etwe nulerial fm a will also expkale less efficiently than the same number of critied masses of P
metallic plutonium. The reason is that the neutum generation tinw is kmger in fission tmmb. In order to poduce the same explosive yield, Wwever, the l
plutonium oxide than in the metal, since the average distance between annunt of plutonium required would be at least severallines greater slun if the dutonium atoms is greater and the neuertms are generally moving note slowly.
plutcaium oxide were separated. Thus, the additional effor t required Io separate Consequently, if plutonium oxide is used instead of the metal,less energy would the plutonium, at least as the oxide from the plutoniunturanium mixture used be released by tiie time the buildup of pessure in the core caused it to expand in breeder reactor fuel, would generally be worthwhile.
to the point where increased leakage of neutums from the core would cause tne After plutonium has been produced from the uranium-238 in a i
reactor,it is extracted from spent fuel at a fuel repocessing plant. It is then in thain reaction to stop.
An illicit bomb nuker wb possessed plutonium oxide would have the form of a liquid plutonium nitrate solution. Plutonium nitrate solution can two options. Either he could use it directly as bomb material and settle for a sustain a fission clain reaction; in fact, the mininium critical nuss of plutonium tmmb that was smnewhat inefficient, or he could go to the trouble of removing in solution is considerably snuller than the critical mass of metallic plutonium.
the oxygen so that he would need to use only about two thirds as much This is because hydrogen atoms in the solutiori are very effective in slowing plutonium and would achieve a legher explosive yield. Whichever way he chose, down the neutrons, thereby increasing the chances they will cause fission. Under Wwever, the bomb nuke % would have to be extremely careful always to keep sonw condiskms, the critical mass can be as snull as a few hundred grams.
the plutonium inside airtight enclosures, and to monitor all steps in the process llowever, unlike the oxide, plutonium nitrate solution cannot be used directly in with some kind of radiation detector to nuke sure he never accidentally the core of a nuclear bomb. The reason is that the neutron generation time of the plutonium in solution is much too kmg. The solution would form steam I
assembled a critical mass.
The processes for converting plutonium oxide to metalik plutonium bubbles that would disassemble the bomb before the nuclear energy lud built up are described in detailin widely distributed, unclassified publications. Moreover, to explosive proporskms.
Plutonium nitrate soluthm is not difficult to convert to usable form.
[
all the scyaired equipment and chemicals can be purchased from commercial firms for a few thousand dollars or less. We find it credible that a perssm with it is easier to nuke plutonium oxide from plutonium nitrate solution than it is experience in laboratory chenustry and netallurgy could assemble all the to separate nexed oxides in order to reduce plutonium oxide to metal. A I
soluuon of sodium oxalate, a conumm chemical, added to plutonium nitrate requiied iafornution, equipment, and chemicals, ami safely carry out all the operations needed to reduce plutonium oxide to nwtal in a clandestine solution, will form a precipitate of plutonium oxalate wisch is insoluble in water. The plutonium oxalate can be separated from the soluthm by simple laboratmy in a few nonths.
The peceding discussion is based on the assumpthm that a bomb filtration and then heated in an oven to form p!utonium oxide powder. As kmg naker would love acquired plutonium oxide before it had been mixed with as the steps are carried out with small batches of plutonium-a few hundred other oxides. When plutonium oxide is used in nuclear power reactors,it is often grams at a time-there is no danger of accidenially forndng a supercritical nuss.
-... _ ~ -.._
4 Nuclear skapone gy 16 Nuclear Thef t: Risks and Safeguards At enrichment levels above 10 pescent, the situation becomes he pessim pe forming these operathms would, of course, have to take the cannplicated. The critical nuss of nwtallic uranium at to pescent emichnwns peccautiims mentioned 2'mvc in osder to keep from getting significant internal a good nmIstm teHector,is about 1,000 Lilograms, including 100 kihgrams I
w l
doses of plutunium_
- ontained uranium-235. Though very heavy, this would still be a sphere of o
miy abou! a foot and a half in diameter. At 20 percent enrichment, the b
High-enriched Uennsens. Natural uranium contains 99.3 per cent crutcal inass drops to 250 kikigrams (fifty kilograms of contained ura.
uranium-238 and about 0.7 percent unanium-235. Uranium-238 cannot, by muru-235), and at 50 percent ensiehneent it is fifty kik. grams includh itself, sustam a fission chain teaction under any conditions. Nearly pure twenty five of uranium-235. At 100 percent enrichment, the critical mass a uranium-235 (mose than 90 percent U-235).on the other hand,is very suitable wanium-235 is about fifteen kilograms,and stuit the size of a softball for nuking fission explosives. A given number of critical masses of uranium-235 it 's probable that sonw kind of rasskm explosive with a yield wdl explode with lower elficiency and, generilly, a sonewhat lower explosive eilunralent to at least a few tens of tons of high explosive could he made with yield than the sanw number of critical nusses of plutonium-239 nwtailie uranium at any enrichnent level significantly above 10 percent but alw The spherical critical nuss of uranium-235 at emanul density, which
'9"ed anmunt of uranium-235 and the overall weight of the bomb is seduced i
is close to twenty times tlie density of water,is between almut eleven kilograms amatically as the enrichnwns is increased to almut 50 peicent. Since nmst and twenty-live kilograms, dependmg on tlw type of neutron reHector that nuclear power reactors use uranium fuel that is either enriched below 10 percent smrounds it. This is about thece times the cultical mass of alpha.phasr w above 90 percent, we are pinutily concerned with uranium enriched above
?
plutonium-239. Without any reflectos at all, alw critical mass of uranium-235 is Perrent. Unless otherwise noted,we use the term" low ensiched uranium" to j
shghtly,xire than fifty kilortams.
Unlike plutonium, utanium-235 is not particularly toxic. No nwan uranium emiched above its natural concentratiim, but below 10 percent-intermediate ennched uranium" to nwan uranium enriched between 10 percer 4
sadiation shiciding or protective coverings are necessary to handle it safely in anti 90 percent; and "hisleenriched uranium' to mean usarium enriched above f
quantities less than a critical mass. Uranium-235 does not fisskm spontaneously percent.
at a significant rate, thus releasing neutsons that might prematurely mitiate a Natural or low-enricled uranium in alw form of a gas, uranium 4
nucleas chain reactiim before a weapim assendily has become highly super.
lexaquoride (UF.), can be furtler emiched in an isotope emichment plant in I
critical >8 The critical mass of uranium-235 in the form of oxide (U0 ) or order to oblam high-enriched uranium. After enrichment, the gas can be 2
carbide (UC ), which are forms used as fuel in sonw types of nuclear reactors, is liquified under pressure for storage and shipment. Uranium hexanuoride is 2
about 50 percent greater than the critical mass of alw nwtal. Either alw oxide or relatively easy to convert to uranium oxide or nwtal.
i the carbide can be used directly as the core material for a bomb. The steps Two nwthods for emiching uranium that have been highly develo d seguired for converting uranium oxide to metal are similar to those for the are gaseous diffusion and gas centnfugation. As far as we know gase us conveision of plutonium oxide, except that tiw safety precautions are much less diffuskm is the ondy method that has been used thus far for large scale sisingent. Generally speaking, uranium is easier to convert from one chemical or separation of uranium tsotopes. blany important detaiis of Ilw gaseous diffusion physical.'orm to another than is plutonium.
Usanium-235 must be "ensicheJ" above its concentrathm in natural iso 6pe separation process remain classirwd. It is well known, however, that it wquires very large annunts of electric power (enough to meet the needs of a uranimn in order to make it usable as the core material in a fission bomb. 'lheI c8ty wille a populalism of several hundred thousand) and las j
degree of enrichment requised is difficult to define with any precision. Helow an ca tal swestnwnts (of the order of hundieds of millions of skillars,at least)it emichment level of about to pe: cent (i.e., the fraction of all uranium atoms tliat om x
.I equipment and huge facilitws.
aie uranium-235 in a mixture of U-235 and U-238 atoms is equal to 10 As far as we have been able to deterirme, the performance percent), uranium cannot be used to make a practical fission homb, even though j
tacten,st,cs of gas centrifuge techniques for uranium isotope separation have
]
i it can be used with a neutum nederator to sustain a " slow" fisskm chain in a t been discussed in detail in the unclassified literature. It is generally clained reactor. This is basically for the same reas<ms that a solution of plutonium the electric power and capital investments required for a trif nitrate cannot be used to make a nuclear exploskm.
l "5 " P an Bestsa dUranium 238, however, dues spontaneously fission at a rate that, thaugh n rif ge sys n s re e e w y complex. Tliey require very many individual roughly 8,nue tinws slower than plutonium.24n, can under sonne circunistances affect the ges which nmst be designed to exceedingly chise physical tolerances.
course of a ihain reas tion in a fission twnti-
=
18 Nuclear Theft: Risks andSafeguards Nucliur Weapons 19 A thial metind for uranium enrichment would make use of laser potect workers in plants slut routinely fundle large quantities of the nutesial.
beams to stimulate atomic or unlecular transitions in U-235 (but not in These ganuna rays do not necessarily present a dangerous lurard to an illicit U-238). l.aws techniques have recently received considerable attention, and bomb nuker who is working, without any shielding.close to kilogram quantities nuy timceivably lead to large reductions in the cost and complexity of uranium of uranium-233. Ilowever, the total time of direct, close up expisure to the isotope separation in the futuse. At the present time arJ for at least a few more nuterial must be limited to several asien Imurs in order slut the cumulative years, however, isotope enrichment facilities for cimverting either natural or
&>se of gamma rays receised annunts to no note than aNmt a dozen chest k,w-ennched uranium to high-emiched uranium will be extremely costly and x-rays. Although such exposure within a few rinntlis or less is considtrably ennplex, and probably beyond the reach of any but the highly imiustrialized greater tlan that permitted wuskers at nuclear facilities, it might be of little na tions.
concern Io an illicit bomb maker.
liigh-enriched uranium hexalluoride is too dilute to use directly in Uranium-233 is much less dangerous to breathe or ingest than any practical type of fission bomb. It is casier to convert the llrmride to uranium plutonium, but it is more danycrous in this respect than uranium-235. People oxide than to metal, but both conversions could be carried out, conceivably in a working with unconfined uranium-233 could simply take the peccaution of clandestine laNnatory, using chemicals and equipment that can easily be wearing nusks tiesigned to filter out suull particles,and of nuking sure they do muchased conuncicially. liigh+priched uranium lexanuoride is hkely to be less rmt work with the material when they have any open wounds Alternatively, attractive to a nuclear thief than the oxkle or metal, but it is likely to be they could take the same pecautions as those required for lundling plutonium.
considesably unre attractive than low. enriched or natural uranium.
Since the chemistry and metallurgy of uranium-233 are pactically identical to those of uranium-235, its camversion from one finm to another Uranium-233. This isotope is prmluced in nuclear reactors that requires the same pucesses. As is the case for plutonium or high-enriched contain thorium. When a neutram is captured in thorium-232, the isotope of uranium, the oxide or carbide forms of uranium-233 could be used as cose shotium that occurs in nature, it ferms thoniuin -233. This radioactively decays, nutesial for fission bimibs. Similaily, this would require about 50 percent more with a half-life of about twenty minutes, to protactinium-233, which nuterial, and poduce a somewhat lower yield than if metal were used in the i
subsequently also decays, with a half-life of about a numth, to uranium-233.
same ype of bomb.
'Ihis isotope is relatively very stable, with a half life of about 160,000 years. The critical nuss of manium-233 is imly aN>ut 10 percent greater than the critical "Strategscally Sagsnirecant" Qinantities of Fission Explosive Mate-nuss of plutonium, and its explosive cfficiency, under comparable conditions, as rials. Our discussi m so far nuy lave suggested to some readers that the about the same as plutonium, it is much less dangerous to woik with tiun mininuun quantity of a Hssion explosive material required to make ome kind of plutonium.
Gssion immb, sometints called the " strategically significant" quantity, is in ways that are anahigous to the poduction of variant forms of roughly equal to the spherical csitical nuss of that nulerial,in metallic form, plutonium in a uranium-fueled reactor, several other isotopes cf uranium, inside a gmd neutron reHector, or tamper. This is not alie case. Tlie amount hesides uranium-2'3, are formed in a reactor that contains thorium. Some of required depends on the particular type of fission explosive in which it is used.
these, such as uranium-234, act as a dilutant, theseby increasing the critica' nuss if the material is to he used in a gun-type of fission explosive, which of uranium-233 almut ten to twenty peicent. None of these isotopes. Lowever, becomes supercritical when umre flun one critical mass is assembled at nornul Gssion spmtaneously at a rate high enough to affect the course of a chain density, the additional amount depends on air desired explosive yield. In his reaction during assembly of nore than one critical mass in a fission bomb. In Emyclopedis emericans article, Foster states that a nuclear yield equivalent to this iespect, uranium-233 is similar to uranium-235.
several hundred Ions of high explosive will he released if a nuss of nuterial equal one of the uranium isotopes formed in reactors that conta,n to two critical masses is assembled and a neutron is injected to start the chain i
thorium is uranium-232. This decays siirough a rather complicated radioactive reaction. The actual yield depends on the particular characterisbes of the masses chain to form severalisotopes that emit gamna rays, a pasticulaily penetratmg and types of noterials involved. On this basis one might argue that, to be on the form of radiation. Uranium-232 is not separated from uranium-233 at a nucleaf safe side with regasd to potecting maclear materials from IheIt, the "stra.
fuel repnicessing plant, the chemical popesties of different isotopes of the same legically significant quantity" of a nuterial should be its critical mass, as a element being practically identical. Uranium-233, as used in the nuclear sphere of the nulerial in metallic form, inside a liiick tamper of beryllium. We industry, will therefore contain enough uranium-232 (typically several hundred love chosen this arrangement because it cinnespmds to the lowest critical masses parts per million) to require concrete or other types of gamma ray shielding to of Ossion explosive nuterials that are given in publislied reputs. For piutonimn,
P I
i 20 Nuclear Theft: Risks and Safeguards Nuclear Wewons 21 s
high-emiched uranium, and uranium-2.13 these nusses are, resyctively, atmut and a substantial amount of chemical high explosive could,within several weeks, four, eleven, and four and one half kilograms.
design and build a crude fission bomb. By a " crude fission bomb" we mean one if, on the other hand, the nuterial is to be used in an implosion type that would have an excellent chance of exploding and would probably explode of lission bomb, the anmunt required may be significantly lower than these with the power of at least 100 tons of chemical high explosive. This could be quantities. Materials that are compessed above their mumal densitivs have a done using materials and equipment that coukt be purchased at a hardware store l
lower critical nuss than when they are uncompressed. In the special case when ami frmn conunercial suppliers of scientific equipinent for student laboratories.
l both the core and the renector are compressed by the same factor, the critical lhe key persons or person would have to be reasonably inventive nuss is reduced by the square of that factor.Thus, when a splwrical core and and adept at using laboratory equipment and tools of about the same complex.
reflector assembly that is initially close to one critical mass is compressed to ity as those used by students in chemistry and, physics laboratories arm! madiine i
twice its initial density, it will correspmd to about four critical masses. The shops. They or he would have to be able to understand some of the essential dependence of the densities of heavy elements on their pressures and concepts and procedures that are described in widely distributed technical pub.
temperatures (their " equations of state"), and tlw pressures tlut can be achieved lications concerning nuclear explosives, nuclear reactor technohigy,and chemical in various types of chemical explosive assemblics are described in unclassified explosives, ami would have to know where to find these publications. Whoever pubhcations. But this infornution alone does smt tell one Imw high are the was principally involved would also have to be willing to take moderate risks of compressions that can actually be acideved in practical implosion systems.1he serious injury or death, reason is that the compressions achieved in an actual device depen *,in detail,on Statements similar to those made above about a plutonium oxide low the device is designed. In particular, the compression achieved depends on bomb could also be made atmut fission bombs nude with higleemiched uranium low close the implosion is to being perfectly symmetrical.
or uranium-233. Ilowever, the ways these nuterials night be assembled in a
'Iherefore, the minimum anmu of fission explosive material fission bomb couhl differ in certain important respects.
required Io nuke a reasonably powerfut impi.an type fission bomb depends on We have reason to believe that nuny people, including some who low much the bomb nuker knows, on his ability to predict the detailed have extensive knowledge of nuclear weapm technology, will strongly disagree behavior of implosion systems during the implosion and the chain reacting with our conclusion. We also know that some experts will not. Why is this a nhases, and on the skills, equipmeet, and facilities at his disposal for building the sutsect of wide disagreement anmng experts? We suspect tlut at least part of the hvice.
reastm is that very few of the experts have actually spent much time pmdering One might argue slut, to be on tiw safe side again, a strategically this question: "What is the easiest way I can think of to make a fission bomb, significant quantity of plutonium, higleenriched uranium, or uranium-233 given enough fission explosive nuterial to assendte nere than one munul should be defined as the snullest amount that could reasonably be expected to density critical nuss?" The answer to this questios nuy have little to do with le used in a fission bomb designed by the best experts in nuclear explosive the kinds of questions that nuclear weapon designers in the United States, tie technology. liven if such quantitics were defined, they would be higidy Soviet Union, the United Kingdom. France, or Peoples Republic of China ask classificit. Nevertheless, the issue of wlut simuld te corsidered as a strategically themselves when they are trying to devise a better smclear weapim for military significant quantity of fission. explosive nuterial for purposes of developing an purposes. Ilut the question is likely to be forennst in time mind of an illicit bomb a
effective system of safeguards against nuclear llwit is one that recurs at various nuker.
points tiuoughout this study. Suffice it to say at this point that it is an Whatever opinions anyone nuy have about the hkelilumd that an imentant policy question for which there can be no purely technical answer, individual or very small group of people would actually steal nuclear nulerials and use them to make fission bombs, timse opinions simuld not be based on a l
Skills and NonrNuc4eae Rescueces Pesumption that all types of fissism bombs are very difficult to make.
I Hequired to Make Fission Bombs As a result of extensive reviews of publications that are available to the general public and that relate to the tecimology of nuclear explosives, EFFECTS OF NUCLEAR EXPLOSIONS unclassified conversations with nuny experis in nuclear physics and engineering.
and a considerable amount of thought on the sutiece,we conclude:
Even a "small" nuclear explosion could cause enormous havoc. A crude fission Under conceivable circumstances, a few persims, possibly even one bomb, as we have described it,might yield as much as twenty kilotons of explo-person working alone, v ho possessed about ten kilograms of plutonium oxide sive power-the equal of the Nagasaki Atomb. But even much less powerful i
Nuclear SVeapons 23 22 Nuclear Thett: Rish s and Safeguards l
l gy devices, with yields ranging down to the equivalent of one ton of chemical high 9g Eg explosve, could cause terrible destruction.
A nuclear explosion would generally produce considerably more hE Q22,E ;
danuge than a chemical explosion of the same yield. A nuclear explosion smt b$
~ " * '
only releases energy in the form of a blast wave and heat, but also large quantities of potentially lettut penetrating radiations (gamma rays and neutrons)
J and radioactive nuterials that nuy settle over a large area and thereafter lethally j,
irradiate unsheltered people in the "falhiut" area. The selative importance of g
these different forms and elfects of nuclear energy in producing damage depends j
g 4-iggge j on the size of the explosion, the way the explosive is designed, and the
~"
II, I
E characteristics of the target sica. Radiation released within a ndnute after the explosion iso <alled " prompt" radiation) tends to be nore important in snull j
,y j
explosions than large ones. The total amounts of prompt radiation released in g
1{
}
two thfferent nuclear explosions with the same overall explosive yield nuy 8og
~***gg
=
0 y
a t
Et
- g2 j differ, by a factor of ten or note, depending on low the bombs are designed.
I g
'the relative importance of the effects of fallout, compared to other effects,
)
g local weather conditions, the nature of the immediate
.9
=
j the depends on environnent of the explosion, and tie availability of striter for people in the f
M y
2 vicinity of the explosion. A miclear explosion in the air generally producesless 5 f Ey 2;- o o o g o j local fallout than a comparable explosion on the ground.The danuge produced g
j di t
~"7Q l by the blast wave from an explosion also depends on the tapography of the
}
g h
immedute e moundings, and on the structural ciuracteristics of buildingsin the ggk k 3 2
4 o
taiget sica.
We can illustrate such differences by a few examples. A nuclear j
4
.;..p
~
R88 g f explosion with a one-ton yield in the open in a sparsely populated area might y
,Jjk produce slight danuge. But the same explosion on a busy street might deliver a WJ
-mod Ji
=
lethal dose.of radution to most of the occupants of buildings, as well as to
.l 5
~~
people along the streets, within about 100 meters of the detonation. A nuclear j
E h j,,
5, exphision with a yield of ten tons in the central courtyard of a large office y
,207%Q; g
luniding might expose to letlut radiation as nuny as 1,000 people in the l
"8 J
=
buihhng. A comparable exphision in the center of a football stadiurn during a j
f
~ ~M 100,000 spectators. A nuclear u
nujor game could lethally irradiate as nuny as explosion with a 100-ton yield in a typical suburban residential area might kill f
l j
peilups as nuny as 2,000 people, prinurily by exposure to fallout. The same f
3
$y,
explosion in a paiking lot beneath a very large skyscrapea might kill as many as d
g 50.000 people and destmy the entire building.
p 4 E$k@S(( {
E
- Jd To give Ine reader some idea of the distances witidn which various y
types of danuge might be produced by nudear explosions of different ). elds, we
]
love prepaied the estir utes presented in Table 2-1. These estinutes are only e-t f
E rough approxinutions for the reasons given above.
h
.3 Prompt radiation released during or very soon after the explosion Q5
.gg!$$
E w
o8.jjE$ !
can be in two forms, gamnu rays and neutrons, both of which can easily f,j gj2 penetiate at least several inches of nost noterials. Gamnu ray and neutron dose
= o 8 E! Q w
r 24 Nuclear Theft: Risks and Safeguards Nuclear Weapons 25 levels can be stated in terms of the Rill, which is related to the Roentgen, a few days is about a duen milhgranu (thousandths of a gram). All these unit of ten used for measuring x-ray dosages. A sadution exposure of about live estinutes, particularly those sebted to simwtening of hfe from lung cancer, are huralsed REM of either gamnu rays or neutrons abstubed over a person's entire uncertain, parfly because the respmses of different individuals to the same deses hinly (a so-called "wtuile txxty" anse) would kill fulf the people so exposed of plutonium are hkely to vary considerably. For pnposes of this discussitm, within a few weeks or less. A radiation aise of about 1,000 REM would kill particularly for comparisons with other toxic substances, we assume that fifty ainose all the people exp> sed. De pompt va&ation is released so rapidly that nicrogrann of plutonium-239 represent a " lethal" dose, i e., the amount slut these would not be time for people in the vicinity of alie explosion to take cover would be very hkely to cause eventual death if it weie internally atnorbed.
in shelters or behind buildings.
i In terna of the total weight of material slut repesents a letful &>se, Delayed radiation from the fallout of a nuclear explosiam could platonium-239 is at Icast 20,000 times nere toxic tlun col'ra venom or deliver letlul dises to people who renuin in the open wiiere radioactive delnis pitassium cyanide, and 1,000 times nere toxic than heroin or :malern nerve has settled long enough for them to acceive a total dose of roughly 500 REM-gases. It is pobably less toxie, in these same ternu, than the toxins of some
%e ranges of distances indicated in Table 2-1 for radioactive fallout are based especially virulent biological oeganisms, such as anthrax gerna.
to the assumptions that the wind vehicity in the area is about five miles per he annunts of plutonium that could pose a threat to society are lumr, and tlut expised people renuin within the area for tne hour, for yields accordingly very snull. One hundred grann (three and one half ounces) of this less than one kiloton, increasing to twelve luiuss for a yield of one megaton.
nuterial could be a deadly sisk to everyone working in a large office buil&ng or
%cse distances are the nost uncertain of any simwn in the table, since they factory,if it were effectively dispersed. In open air, the effects would be nuwe depend strongly on the hical weather conditions, the amount and characteristics diluted by wind and weather, but they would still be seek >us and lorig-lasting.
of the surface nuterial tint would be picked up in an explosion's fireball and The quantities of plutonium that might poduce severe luiards in later deposited on the ground, the extent to which people would be able to take large areas are sumnurised in the very crude estinutes gesented in Table 2-2.
cover or leave the area quickly after an explosion, and nuny other factors.
To estinute flie areas within whicli people night be exposed to letful anses
%e distances indicated in Table 2-1 for severe and nuxlerate blast inside a building, we assunse that dispersed plutonium is pimarily plu.
damage and cratering aie considerably nore pedictable than the distances for tonium-239 in the form of an aerosol of finely divided particles distributed severe danage by radiation. A peak ovespessure of ten pounds per square inch uniformly in air llwougiumt the building, We also assunw tlut exposure of would be likely to cause very severe danuge to ainest all residential and office people to the ccmtaminated air is for one hour, that ten percent of the inhaled buildings, and nederate danuge to heavily seinforced concrete buildings.Tluce particles are retained in their lungs,and that,as stated cadier, the lethal setained pounds per square inch would cause severe danuge to wood frame residential dose of plutonium is lifty micrograms. These conditions might be achieved by j
huildings.
carefully introducing the plutonium acrosol into the intake of a building's air To sumnurire, the human casualties and poperty danuge that could conditioning system. This might be quite difficult to do in many cases.
he caused by nuclear explosions vary widely for different tyres of explosions detonated in different places. Nevertheless, it is clear that under a variety or Table 2-2.
Lethel and Significant Contemination Areas for Release ciremmtances, even a nuclear explosion one hundred times snuller than the one of Air ', _
n of Plutoniuniinside Buildings slut destroyed Iliroshinu could luve a terrible impact on society.
s,g j;c,,, cy,,,,,j,,,j,,
inhektrium f.erhal Dose Requiring Some Eracuarium Amount of of SuspendalMeterial and Ocarmp RADIOLOGlCAL WEAPONS p,,,ong,,gega,,,4
(,,,,g,,4,,,,,,,,,,
g,,,,g,,,,,,,,,,,,,,
I sram
~500
-50.000 Plutonium Dispersal Devices 00 srams
~50.000
~5,000,000 We luve already stated that plutonium, in the form of extremely snull particles suspended in air, is exceedingly toxic. The total weight of An ares of 500 square nzters(about 5,000 square feet)corresemds r utonium-239 which, if inhaled, would he very hkely to cause death by lung to the area of one ik>or of many typical office buildinge An area of 50,000 f
cancer is not well known, but is pohably between ten and 100 micrograna square meters (atmut 500,000 square feet)is comparable to the entire ik>or area (millionths of a gram). Even lower intenial sk>ses, perlups below one microgram, of a large skyscraper. Even a few grann of dispersed plutonium coul.1 pise a nught cause significant simetening of a person's life. The total retained dose of sesious danger to the occupants of a rather large office building or enclosed plutanium slut would be hkely to cause death from fibrosis of the lung within a industrial facility.
l i
Nuclear niewms 27 26 Nuclear Thef t: Risks and Safeguards People who absorb Icthal but not massive doses of plutonium would I
The areas in which plutonium contandnation would be significant not sense any of its effects for weeks, ax perhaps years. The pesence of rmely j
enesh to require evacuation and subsequent decontamination are roughly divided plutomum in an area could be detected amly wish sensitise radiatkm 4
estinuted to be about 100 times the areas subjected to a lethal dose. Almut a
""*il"'ing equipment. Such espipment is snow ostly used tii anonitor alw l
dozen grann of plutonium dispersed througiout the largest enchised building in Pesence of plutonium os other dangerously radkwctive materials in nuclear l
the world night nuke the entire building unusable for the many weeks thag installatkms. Except in such installatkms, therefore, people would not know would be required to complete costly decontamination operatkms.
they were exposed until they were told, either by tinise responsible far sie De dispersal in large open areas of plutonium with lethal con.
slucat, or by sim cone iri authority wlio happened to detect the plutonium with t.entrations of rathoactivity is hkely to be much umre difficult to carry out "5huments.
effectively than cSspersal inikxws. The height of the affected zone would be We are not aware of any successful rum military attempts to use difficult to hold down to a few feet. Even a very gentle, two-nsle.per-hour chenncal, bacteriological, or radiohngical piinms to contannnate lange areas.
becere would disperse ihe suspended malerial several kilonwters slownwind in an Whether any such means will be used in the future for criminal or terrosist hour This would make it extrenely difficult to use less than about one kikigram Purposes is, we beheve, an even armue speculative apestion alian winether nuclear i
of plutonium to produce screre radiation hazards. With a few dozens of grams of explcsives will be to used. Many types of potentially lethal poisons are run nuwe i
plutonium, however, it would be relatively easy to contaminate several square difficult to acquire than chemical high exphnives. llowever, high exphisives are kiloneters sufficiently to sequire the evacuation of people in flee area and heing used with greater frequency ami in increasing anm>unts by terrosists and i
necessitate a very difficult and expensive decontandnation operation.
extortionists, while we have found no evidence that they have ever used i
After the plutoniunebearing particles settled in an area,they would P"somms agents. The pactically instasitaneous, quite obvious destructhm that senuin a potential luzard until they were leached below tie surface of the is Prmiucesi by an expk>sion apparently better suits the purposes of terrorists ground or were canied off by wind or surface water drainage. As kmg as the and tortiomsts than poismas that act nuwe skiwly and subtly, but slut are at prticles venuined on tie suiface, sometidng night happen to draw them back least as deadly. Unlike other poisons, however, plutonium can be used either as a into the air. Contamination levels of about a nderogram of plutonium per square Poism or as explosive material. Accordingly,a threat using a plutonium dispersal meter would be likely to be deemed unacceptable for public health. Thus,in an device could conceivably lie folhiwed by a threat involving plutonium used in a usban area witti little rainfall, a few grams of plutonium optinully dispersed out macleas expkmave.
of doors night sesiously contaminate a few square kilometers,but only over a very much snuller area would it p>se a lettui threat.
Oeine Types of Reiselegical Weapons So far in our discussion, we have considered only plutonium-239, As part of mar research for this study, we consklered,in some sletait' the isotope of plutonium that is poduced in the largest quantities in nuclear lie effects that might be produced by dispersing radioactive materials ollwr than scactors. Plutonium-238, which is also made in significant quantities in saw i "'"i"'". w by purgumely -palsing vasioun types of unshiehled nuckar reactors d
reactors, is camsiderably nuise toxic tiun plutonium-239. Its halfIsfe for to destructs,on without acideving a real nuclear exphskm..We conclude alist endtting alpha pasticles is only about eiglity.seven years,instead of about 25,000 neither type of weapon would be as effective as a plutonium dispersal device or years; one gram of plutonium 238 tlerefore emits alpha particles at approxi.
a low-yield fission bonib.
nutely 300 lines the ate that plutonium-239 skies. As a result,Ihe lethal close Spent nuclear reactor fuel and the fission poducts separated from of plutonium-238 is almut 1/300 of what it is for plutonium-239. We mentiori reactw fuels at a chemical reprocessius plant are, potentially, extrenwly this because plutonium-238 has been used in radioisotope-powered nuclear liazardous if dispersed in a populated area. Ilut alwy would also be very I)
"battesies," and is being seriously considered for use in power supplies for heart dangerous h> handle in sufficient quantities to pise a tiveal to a large area pumps in people suffering from certain types of heart disorders. As much as because they endt highly penetrating genmu rays, thus requiring heavy sidelding sixty granis of plutonium-238, the equivalent in toxicity of almost twenty to Protect thieves or weapm makeis.In shint, pluhmium wtmid he easier to use Aitograms of plutonium-239, nuy be in each such leart-pump battery. This is for destructive purposes than radioactive fission poducts.
I enough nuterial to produce serious contaminution of hundreds of square miles, If a nuclear reactor core were pulsed Io destructism,it would release if dispersed in tle form of snull particle 3.
a cugiaratively suull amount of energy equigalent to, at nm>st, a few laundseil A variety of ways to disperse plutonium with timed devices are Poumis of Ingh expkuive from a device weighing several tons. It would also conceivable. These would alksw tie threatener to leave the area before the release anumnts of radiation and radioactive materials slut would be vesy snull noterial is dispersed. Any plutonium contained inside such a device would not cmnpased to a k>w-yield nuclear explosion unless the reactor had heesi mM f
be a hazard until it was released.
28 Nuclear Thef t: Risks and Safeguards
%ter bree high p>wer levels for sone time before use as a weapon. Under sucli Nuclear Fuel Cycles: 1973-1980 it conditions, it would have to be transp>rted in heavy sideldmg and would pose to even greater handling problenu than stolen spent nuclear reactor fuel. Generally d
speaking, therefore,it would be easier to nuke and use a fission tunnb tiun to
,\\
nuke and pulse a nuclear reactor core in a way slut would produce danuge on p1 tie scale of a fission immb.
9{
g ff K
- ?
y PORE FUSION EXPLOStVES g
A pure fusion explosive would be a device that would not require any fission g
g
' trigger" to initiate explosive thernmnuclear (fusion) reactions in very light y
hydrogen isotopes such as deuterium and tritium. There is considerable discussion inShe unclassified literature concerning the p>ssibility of developing
=,
this type of exph>sive. No successful development has yet been announced,and we have no reason to beheve it lus taken place.
Recent papers suggest that it nuy be p>ssible to use intense lase' INTROOUCTION pdses to imphale snull " pellets" of deuterium and tritium (and possibly pure deuterium) in such a way as to cause the pellets to expk>de. The concept is in Chapter 2 we considered the nuclear materials and other resources required to desenbed in the context of its possible use for sie generation of electric power.
nuke fission bmnbs and described the damage that ctadd result from nuclear Very small thernmnuclear explosions would be connned, possibly with magnetic explosions or the dispersal of plutonium. In order to appreciate the risk tlut liehls,and the explosion energy would be extracted to produce electricity-nuclear weapon materials might be stolen from the nuclear power industry,our intensive research and development on such systems is under way,n next step is to describe the facilities and operations that, taken together, i
AEC laboratories end at least one industrial laboratory. Scae people working on comprise the " nuclear fuel cycles" required to suppmt each major type of laser-induced fusion suggest that the scientific feasibility of tie concept may be reactor used to generate electric power, A typical nuclear fuel cycle includes successfully demonstrated within a year or two. There is etmsiderable contn>.
facH ties for mining, converting, emiching, fabricating, using, repsocessing, and versy, however, about when the practicality of laser. induced fusion may be recycling nuclear fuels. lt also includes all the transportatiim links between these denonstrated. Whettwr or not laser-triggered fusion could be developed into facilities.
pactical and transpn table nuclear explosives with yields equivalent to or greater We want to know which points in each fuel cycle need. safeguards than tons of chemical high explosives is not revealed in the unclassified against theft. Where can materials be found, both now and in the future, that are hieratuse,and the answer nuy well be unknown.
usable for making nuclear weapons? What quantities of these materiats,in what in any case, we do not believe tlut pure fusion explosives could be physical and chemical forms,could thieves expect to Gnd at different stages of a nude clandestinely in the foreseeable futme without highly soplasticated fuel cycle? Ilow heavy and how large are the units that containJhese nuterials equipment and exceptionally highly skilled and experienced specialists.
hkely to be? In short, we intend to provide in this chapter a factual basis for deciding which parts of nuclear fuel cycles are inherendy most vulnerable to attempted thefts of nuclear weaptm materials. We will then be prepared to consider various measures to safeguard against nuclear : hefts in schsequent NOTES TO CHAPTER TWO chapters.
We also briefly discuss in this chapter certain research applicatiims of
- 1. John S. Foster, " Nuclear weapons", Encyclopedia Americana, nuclear energy because they now involve considerable quantities of nuclear Volume 20, pp. 520-522, Americana Corporation, New York,1973. Reprinted weapon materials, sometimes in finms that are especially suscepiible to theft. We with permission of the Encyclopedia Americana, copyright 1973,'the Americana do not mean to imply, however, that these are the only civilian applications of rmclear energy where a risk of nuclear theft nuy exist. Other serious possibdities Cmpora tion.
might arise beyond 1980. We restrict ourselves to the risks of theft primarily 29 I
P I
106 Nuclear Theft: Risks and Safeguards Chapter Sin Risks of Nuclear Theft
$14-13,749; pt 4. p. 243141972). The AlC BuJget for 1incat Year 1974 does not break est the cosnparable figures for safeguards attsvitecs. 7hc safeguards research and Jewetopment tudget was increased daphtly to $4.4 unihsm for att ogeraanins. Meport by the joint Conunittee on Atomic t'ncrgy, Auth<wrzmg Approgwrarkwr for the Atomic Energy (hmmsssem JW l' anal Year I974,9 3J Cams,1st Sess., p. I 2 I e 973 t INTRODUCTION It is all too easy to imagine innumerable possibilities for nuclear theft-a parade of horrors. It is extremely difficult, however, to determine where alw line should be drawn between credible and incredible risks, between risks that sliould be safeguarded against and those that can be safely ignored. An assessnwnt of the risks of nuclear theft is even more speculative than an analysis of the risis of major accidents in Ilve operation of nuclear power reactors. With resswet to reactor operation, eisks to public safety arise primarily from the possibilities of malfunctioning nuchines. In regard to nuclear llwit, however, the risks to national and individual security arise primarily from nulfunctioning people.
Nevertheless, the safety risk analysis applicable to reactor accidents and the analysis of security risks applicable to nuclear theft have two difficulties in comnam. In sie first place, both types of analysis deal with very low probability risks of very great damage. It is noteworthy, however, that the damage which might result froan a nuclear theft is potentially much greater than the danuge that could result from the nuximum cre:lible accident in the operation of a nuclear power reactor. Second, as to both areas of risk, there is, and hopefully will continue to be, a lack of actual experience involving substantial danuge o the public on which to base predictions.
As fuel for power reactors, nuclear weapon material' will range in commercial value from 53,000 to $15,000 per kilogram-roughly comparable to the value of black market herois:.The sanw material might be hundreds of eines more valuable to some group wanting a powerful means of destruction.
aThroughout chapter 6 and the rennainder of this took, we use " nuclear weapon niaterial" to mean neaternal that can be used in fission explosives or,in she case of dI utonium, in dispersal devices either directly or with shemical cusiversions that are much smpter processes than those involved in reprocessing irradiated nuclear fuels or in isotope ensichment.
107
n.
IM Nuclear Theft: Risks and Safeguards Risks of Nuclear Theft IM Furthermme, the costs to society per kilogram of nuclear material used for the risks woidd seem to be greater in other countries than in the US., while destructive purposes would be immense. The dispersal of very small anmunts of others may be geester in the US. than elsewhere. Moreover.nuterial stolen hom Hnely divided plutonium could necessitate evacuation and decontamination the U.S. nuclear power industry might be used to threaten the security of people operations covering several square kilometers for long periods of time and in foreign countries and their governments.Similarly,materialdiverted from the costing tens or hundreds of nullions of dollars. The danuge could run to many nuclear industry in a foreign country might form the basis for a nuclear threat nillions of dollars per gram of plutonium used. A nuclear explosion with a yield within the U.S. (The related risks of governmental diversion in non-nuclear.
of one kiluron could desiroy a nujor industrialinstallation or several large office weapon count:ies are considered in Appendix D).
bu.ldings costing hundreds of millions to billions of dollars. The hunoreds or thousands of people whose health miglit be severely damaged by dispersal of THEFT SY ONE PERSON ACTING ALONE plutonium, or the tens of thousands of thousands of people who might be kdled by a low. yield nuclear explosion in a densely populated area represent Reasons incalculable but inunense costs to society. These intainsic values and potential The possible reasons for one person to attempt to steal nuclear costs should be borne in mind throughout our analysis f the risks of theft of weapon material from the nuclear power industry cover a broad spectrum. On nuclear weap,n material from the nuclear pewer industry.
one end of the motivation spectrum is financial gain, and on the other is a sick The analysis which follows focuses exclusively on the potential expression of extreme alienation from society as a whole. In between lie such security risks involved in the development and use of nuclear power. We have motives as settling a grudge against the managenwnt of a nuclear plant, or a avoided analogies to a multitude of other security risks, some of which appear strong conviction tliat nuclear weapon prohferation is a good thing. Money equally sieserving of study and concem. For example, biological or chemical would seem to be the most likely general motive for an individual to steal agents might be diverted from their intended medical or industrial uses for use in nuclear nuterial, assuming a buyer were available. (The terrorist would normally very powerful weapons, or they might be produced in clandestine laboratories be operating as past of a group rather than alone.)
operated by criminal groups. Chemical high explosive: have been frequently used More specifically, the hme person who contemplates theft of nuclear for ciiminal and terrorist purposes, often with devastating effects. Thus, it is weapon nuterial may do so with any of a large number of particular uses fcr the important to view the security risks implicit in nuclear power as a cost to be nuterial in mind. Possible uses include the following:
weighed agamst the benefits of nucleas energy as a source of electric power,and also as an integral part of the general problem of violence that afflicts society.
pe k Market Sale. The entire amount of stolen material might be ac With these cautionary thoughts in mind, we may explore the sold in one transaction, if a large quantity of nuclear material would bring a pissibilities for and consequences of diversion of nuclear material from the psemium price. Alternatively, snull amounts might be sold over long periods of nuclear power industry to illicit use. Our analysis is mainly intended to provile time in separate transactions, if the thicIviewed his ill.gotten gains as something readers with a nmre infornwd basis for nuking their own judgments concernmg hke a veny precious metal to be liquidated in installnwnts as income is needed.
the ciedibility of the risks involved-judgments which can be expected to differ widely since tFey will be necessarily based on individual views of human nature.
Ransom of Stolen Material. If tarefully worked out, the thicf We consider the risks of nuclear theft by different types of potential might be able to obtain at least as high a ransom for the stolen mateiial as he thieves: one unstable or criminal person acting alone; a profit-oriented criminal would be able to get by sale in a black market. The nuclear enterprise stolen group; a terrorist group; a nuclear enterprise; and a political faction within a from would be one possible target of such a blacknuil scheme;another might be nation. For each type of potential risk, we outline the reasons for theft, the the U.S. governnwnt. The nuclear enterpise, the governnwnt, and -depending scope of the risk,and various methods of thievery. Finally,we examme the main on his tactics-ilw thief himself, might have a strong interest in keeping from the poblems associated with nuclear black nurket operations. The nature and public any infornution about a nuclear theft. This possibility raises two extent of such a market, if any, generally affects the specific risks of theft questions: In what circumstances do the American people have a right or a need peviously considered.
to know about a theft of material from the U.S. nuclear power industry? And, Although our study concems piimarily the theft of nuclear material furthermore, do other governments have a right or a need to be informed about imm the U.S. nuclear power industry, the risk analysis is also applicable to such a theft, if circumstances indicate that the stolen material has likely been p>ssibilities in other countries with m, clear power inductiles. Indeed, some of taken out of the country?
~.
110 Nuclear Theft: Risks andSafepwds Risks of Nuclex Theft 111 Fabrication of a Weapon and Actual Nuclese Dreat. As indicated ahihty of govenmwntal institutims to cope with the secusity problem. On the in Chapter 2, the manuf acture of a fission explosive or plutonium dispersal other hand, if a pohey of not paying off on any nuclear bomb threat we e e
device nuy be within the capabdities of one person wosking ahme, assuming he adopted, it might have to be accompanied by strict and enforceable urban pissesses the requisite technical competence. But what would the individual do evacuation plans which could be carrieil out inunediately upm receipt of a with his nuclear iveapon? As with stolen nuterial,he might sell the device in the credible thicat. II credible nuclear thicats occurred of ten, an urtun conununity black nurket or ransom it. Any sevel of gtwernment-municipal, state, or would be paralyzed at enormous costs to society as a whole. The alternative federal-might be a target for blackmail of this type, and a governmental would be to assume the risk and ignose any nuclear bomb threat.
authosity might be prepared to pay a very high price to gain possession of the if the government adopted a policy of trying as best it could to device. The blacknuiler would, of course, have to establish ths credibility of the distingmsh between the actual imclear threat and the hoax, the consequenees of nuclear threat, but this would not seem difficult. One easy way to do so would a wrona choice would again be nuclean catastrophe. Therefore,ll a acceptability he to send the authorities a design drawir g of alie device, pe: haps together with a of such a policy would depend on a foolproof method of discriminating between sample of tiee nuclear nuterial used and photographs of the actual device.
the real thicas and the hoax. It is difficult to inusine such a methixl.
As with the rarsom of stolen nuclear material, the blacknuiler could nuke his denunds and conduct the entire transaction in secret,or he might from Scope of the Risk the outset or at some stage in the negotiation. make his demands known to the Fortunately, not everyone is a potential thier of nuclear manesial.
public, the governmental authorities wouhl probably wish to keep the matter The greatest risk of nuclear theft by one individual acting alone is posed by secret, at least until an energency evacuatum became necessary. If the nuclear persons authosized access to nuclear nulerial at facilities (mainly nuclear threat were dischised to the public, serious panic couhl result. The aluentener industry employees), and to persons authorized control over nuclear material wuuld have to be sure that, whatever his demands, they were satisfied prior to or sluring shipnwnt between facilities in the various fuel cycles. This considerably simultaneoudy with the government's gaining possession of the device. This narrows the scope of the risk ofindividual theft. But it also means that someone might be very difficult to arrange, especially for a kme individual.
who is in a position to steal nuclear nulesial by himself may well possess the technical knowledge required to handle it safely ar.4 use it destructively.
Nuclear Hoax. If a design description plus a sample of nuclear llowever, anyone can nuke a nuclear threat simply by lifting a nuterial would establish the credibility of a nuclear threat, why would the telephone. A very large number of people could make a nuclear tiucat that is threatener have to actually fabricate and emplace a fission explosive or credible-at least up to a point-but still be a hoax. At least one such threat has plutonium dispersal tievice in order to obtain satisfaction for his denunds? If already occurred.(This was the extensively reported Orlando nuclear lawnb hoax government authorities were willing to pay oil a nuclear bluff or hoax, the stescribed in Chapter 5.)
potential profit or political utility of a small anwunt of nuclear weapon material would be increased substantially. One or a series of such hoaxes would greatly Options. The lone thief who is an employee in a nuclear facility or complicate the problem facing a governnwnt. Even the appearance of suc-smwwhere in the transportation system for nuclear material has two basic cumbing to a nuclear threat, whether genuine or not, might be an aJded options for acqui:mg nuterial for fission expkisives or radiological weapons:
incentive to potential shieves.
(1) he can attempt to steal a large anmunt of material at one time;or (2) he can i
If a person perpetrates a nuclear imax on a government that has take a small anmunt each time in a series of thefts. One possible scenario for a previously expeiienced one or more bomb threats nude payoffs,and recovered large theft by an indivislual from a nuclear facility would be to fake an acendent the devices, the hoax will probably be successful. lf, however, a government has involving the risk of employees being exposed to high radiation levels, or some made payoffs a*. a result of credible hoaxes, but not recovered any devices,it other emergency condition which sequises the inmediate evacuation of all nuy establish a policy of no nxwe payoffs. This could create a situation of persons from the facility. TI,s thief might then be able to make off with a extreme danger. The next credible bomb threat might be the real thing, and a significant quantity of material tinough the emergency safety exits. Individual nuclear catastrophe would he the probable result.
acts of theft of nuclear material he transit or in storage during transit could also On the one hand, a government policy of aying off all credible result.if successful,in the loss oflarge amounts of material.
P nuclear immb threats would probably increase the frequency of such threats to The possibility and significance of a series of thefts of snull anmunts intderable levels. The results could be a large drain on Im' ancial resources, great l
of nuclear material would depend on the detection threshold and the elapsed anxiety in people living in urban areas,and widespread loss of confidence in the l
time between the events and discove y of their occurrence. It seems that
i i
112 Nucles Theft: Risks andSafegurds Risks of Nucles Theft 113 nuterials accountancy alone would provide insulGcient protection against snull Options l
theils hy a plant employee given the limit of enor of nuterial unaccounted for 11 seems very hkely slut a criminal group would be able to develop a (I EMUF) in any such system, as discussed in Chapters 5 and 7, and the ipability to apply sophisticated means, incluJing substantial force if necessary, knowledge the employee would normally have of what the LEMUF was.
- in order to carry out a nuclear theft. Therefore, the analysis which follows focuses em the technical capabilities a group might have for dealing with nuclean nuterial, not its capabilities to use force or stealth to obtain it.
THEFT BY A CRIMINAL GROUP Misd i Nh @iliar MaM m a m e @ %
nuclear theft would have to be able to recognize precisely the nuterialit wanted Reasons and to understand the procedures required for its safe handling. Regarding the tact es of nuclear theft, a crinnnat group with such a minimal nuclear capabihty weapon nu er als.
ne s i s ey, 1 In t
a e
>l would have two basic options, in the first place,it could attempt to mGl: rate black market or ransom dealings in the malesials tiemselves,in fabricated fission nuclear mdustry or transportation facilities through which nuclear material explosive devices, or m. fabricated plutonium dispersal devices. The corollary passes, and then attempt to steal very small quantities of nutenal without being i
]
reason is that the possession of a few fission explosives or radiological weapons detected. Secondly,.t could attempt to burglarize a nuclear facihty or h.. yack a i
might place a crinnnal group rather effectively beyond the reach of law w
canying a nuckar Mpnwnt and take a large amount at one tinw.
enforcement authorities. A criminal organization nnght use the threat of nuclear violence against an urban population to deter police action directed against its if successful with either a series of small nuclear thefts or a single nuclear theft operations. The organization might also use nuclear threats to large one, a crindnal group with mimmel tecluncal competence would possess extort from the government a tacit or explicit relaxation of law enforcement material that it could sell to others or use to blackmail the enterprise stolen from Dese are basicaHy the sanw options available to one person acting ahme.
activities directed against a broad range of other lucrative criminal operations.
llowever, an organized group would have much greater capabilities than one person to make arrangenwnts for either time black nusket sale or ele ransom of Scope of the Risk stolen nuterial to a nuclear enterprise or a govemnwntal authority.
To what extent would crimiaal groups beconw interested in the potential for financial gains in illicit trade in nuclear material? It may be argued Capability to Manesfacture Neoclese Weapons. A criminal group that the potential gains are so large that a wide variety of criminal organizations could acquire the technical competence to fabricate nuclear weapons in a would attempt to exploit the possibilities of nuclear theft. To tie contrary, number of ways. A group nwmber with a well developed scientific and however, it may be argued that crindnal groups prinurity interested in money mathematical talent could develop the required competence on lus own without are likely to be politically conservative, and that they would not devehip a black g
formal training; or a group member with sonw aptitude and a college education nurket in a commodity such as nuclear material which coul J have revolutionary might be sent to a year or two of graduate school;or the group might recruit,or plitical implications. Moreover, a large nuclear theft might prompt a massive kidnap and coerce someone already possessing the requisite techmcal skiHs.
govemmental crackdown and lead to a widespread public outcry, whereas the Altern;tively, sonwone with the requisite skill ndsht decide to pursue a career in continued existence of organiicd crime on a large scale might depend on the crime rather than lawful industry and take the initiative to form his own susceptibility of some government officials to corruption and on a degiee of crininal group in order to profit from nuclear theft.
1 public indifference.
A favorable hication could be selected for the weapon manufac-The possession of a few nuclear weapons as a detenent against law turing facilities. This nnght be in the midst of an intensively industrialized area enforcement nuy be viewed by a criminal group as more of a risk than a benefit, or it night be in a rert.,te and inaccessible region. Some foreign government in order to obtain the advantage of a deterrent effect, the crindnal group might be willins e nost a clandestine manufacturing operation outside the U.S.
pssessing such weapons would have to be willing to inflict large scale, j
Any govermoent opposed to nuclear weapon proliferation might find it indiscriminate harm on society. Moreover, hke nuclear war between nations,if extrenwly difficult to deal with a criminal group which had the capability no the deterrent failed and a crindnal group either used nuclear weapons or failed to nunufacture nuclear weapon devices if the group's manufacturing facilities were use them, the group itself would probably not survive the crisis as an located on territory under the jurisdiction of a government that was amenable or organization.
indifferent to such prohferation.
i
'p Risks of Nuclear Theft I15 I14 Nuclear Theft: Risks andSafeguards Scope of the Risk Tlw capabilities and preferences of Potential buyers-terrorist The scope of the risk of theft by terrorist groups would seem t:o poups, national governments, or political factions within national govern-depend largely on how widespread terrorist behavior becomes in the future.
nents could well be the decisive factor determining *vhether a profit. oriented Although any assessment in this regard is highly speculative, present trends criminal group wouhl develop its own capability to manufacture nuclear dimmig h 'hce d vihe initid h mie min weapons. For example, national governments interested in the clandestine groups seems to be increasing in nuny paris of tie world. Terrorist organizations acquisition of nuclear weapons might prefer to pur:hase the requisite materialin are increasing their technical sophistication, as evidenced by the armaments and order to manufacture weapons tailored to their particular requirements.
w cs they use. Such groups are also rapidly developing transnational knks with flowever, terrorist groups might provide a ready market for fabricated nuclear each other in order to facilitate the flow anong countries of arms and explosive devices.
ammunition and even of terrorist persimnel.Whatever works as a terrinist tactic Capability to Manufacture Nuclese Weapon Material. It seems very in one part of el,e world appears likely to be picked up and possiNy emulated unhkely that a criminal group could develop its own capability to produce elsewhere. One wonders Imw in the long run nuclear power industries can significant amounts of plutonium or uranium-233. The operations required are devek>p and prosper in a world where terrorist activities are widespread and numerous and complicated, and on too large a scale. There are a number of persistent. For if present trends continue, it seems only a question of time reasons why it is also unlikely that a criminal group would be capable of before some terrorist organization exploits the possibihties for coercion which enriching uranium, at least in the near future. The technology to separate are latent in nuclear fuel.
uranium isotopes by means of centrifugation, one alternative nwthod to diffusion (which requires huge facilities), is being developed in various countries Opeions under conditions of governmental or commercial secrecy. The operation of Terrorist groups might become a large source of Nack market centrifuges would be a demanding task technically The crininal group would demand for nuclear weapons. Ilawever, such a group may prefer, for various have to steal a number of centrifuges in order to acquire a capability to produce reasons, to develop its own capabilities of stealing and using nuclear materials. A significant quantities of high-enrichest uranium from stolen low-enriched or terrorist group may wish to be independent of any ordinary criminal enterprise; natural uranium. Given the cost of one centrifuge, inventory controls capable of the group may believe that a spectacular nuclear theft would serve its purposes; detecting the theft of one or more centrifuges would seem justified. If a theft or the group may be able to obtain the material it wants more cheaply by were promptly detec'ted, it would seem that the government would have a stealing it than by buying it on the black market. It is difncult to imagine that a relatively long time to recover the stolen centrifuges. Ilowever, the successful determined terrmist group could sms acquire a nuclear weapon manufacturing development and widespread application of laser techniques for isotope capability once it had the required nuclear weapon materials. In this regard, a separation would seem to have substantial implications for the spread of terrorist's willingness to take chances with his own health or safety, and to use uranium enrichment capabilities, possibly to criminal groups as well as to many coercion to obtain information or services from others, simuld be cimtrasted commercial enterprises.
with the probably more conservative approach of persons engaged in crime for numey.
THEFT BY A TERRORIST GROUP
'The theft options of a terrorist group would not differ substantially i
from those available to a profit-oriented criminal group. But whereas there may Reasons be incentives working on all sides to keep the fact of theft by a profit-oriented Although financial gain should not be excluded as a possibility, the criminal group secret from the public, there may be reasons why a terrorist dominant nmtive of a terrorist group attempting to obtain nuclear material group would want a sucressful nuclear theft to be well publicized. Theft of a would probably be to enhance its capabilities to use or threaten violence. An large amount of nuclear material would not only acquire for the terrorist group a important, though secondary purpose might well be to provide itself with an significant capacity for violence or the elucat of whdence, but also the process of effective deterrent against police action. In these respects, a terrorist group execut ng a succenful theft could itself generate widespread anxiety. People possessing a few nuclear weapons would be in a qualitatively different position would become concerned not only in the country where the theft occurred, but offensively and defensively from such a group possessing only conventional also in a country or countries against which the group's activities might be arms. Ilence, theft of fuel from the nuclear power industry might place nuclea' ultimately aimed. Ilowever, one important reason why a terrorist group may weapons in the hands of groups that were quite willing to resort to unlimited prefer to keep its nuclear theft operations a secret,if possible, would be its own vii>lence.
116 Nuclear Theft: Risks and Safeguards Risks of Nuclear Theft 117,
vulnerability to swift and forceful governnwns action during the period between accountancy anomalies as they arise-an easy way to balance the htmks.
nuclear thef t and completion of Ilw fantication of fission explosive e*evices or Furthernure, the managers of a nuclear facility nuy view nunipu?ation of radiological weapons.
noterial balances as a way to increase sl.ghtly the profitability of the entespiise.
The abihty of a :pvernnwnt, whether U.S. or foreign, to deal with an (The possibility of collusion hetween the managers of civdian nuclear operations emergent terrorist nuclear threat would depend on the location of the group's and government authorides in the clandestire dive:sion of nuclear material for base of operations, particularly the location of its weapon manufacturing use in a broad renge of government military programs, which is a concern j
facilities. This nuy be unknown and hard to determine,or it nuy be located on peinurity with respect to non-nuclear-weapim countsics, is camsidered in territory subject to the jurisdiction of a government that is for s mic season Appendis D.)
not prepared to take decisive action against tlw group involved.
Once a terrorist group possesses fission explosives or radiological Scope of the Risk weapons, the group's options for their coercive use, both aggressively and to
'the sisk that rmclear enterprise nunagers might nunipulate material deter enforcement action against it, cover the complete range of options balances to dieir own advantage seems to be inherent in the nuclear power discussed previously for an indivit ual acting alone and for profit-oriented industry because of the high intrinsic value of the materials involved and the fact t
criminal groups. Ilowever,if a terrorist group were involved, doubts concerning that no one will know exactly how much is actually Howing through a major the credibihty of many options previously considered would be substantially facility. In addisian to the presumed himesty of miclear plant nunagers, senmved, and the inner h>gic of the possibilities for nuclear coercion would however, there are limitations on the scope of this particular diversiim risk. If an control. These possibilities would be exploited by a group of people who might
- arms length ** commercial relationship exists between the operitors of distinct tw quite free of the practical, intellectual, or emotional restraints that tend to steps in the fuel cycle, dw possibilities for diversion by nulerish balance inhibit the use of violence by other groups.
manipulations would be lessened. In addition, since one person could probably r
rmt get very far in a complicated nunipulation process, a conspiracy within the DIVERSION BY A NUCLEAR ENTERPHISE plant would be necessary. This would substantially increase both the d:Hiculty of diversion and she risk of detection.
Government nulerials accountancy requirenwnts could arguably Options We consider here <mly the risk that the managers of a nuclear have the effect of either increasing or reducing incentives wi'hin industry to enterprise might divert to an illicit use some of the material Howing through nunipulate nxicar nuterials balances. Vigoious government enforcement of facdities under their operational control. The most hkely diversion option would stringent materials accountsney requirements might increase the incentives for be for the nunagers of processing facilities to manipulate materal balances plant nunagers to cheat the system in order to be sure they could balance the within the margins of uncertainty in the accountancy system. The nuclear txmks and keep their facilities operating efHeiently. Ilowever, a tax govern-noterial input of a fuel reprocessing or fabrication plant is not known to anyone mental attitude towards materials accountancy might reduce incentives for exactly. Therefore, the input could be stated to be at the lower limit of the discipline within industrial opesatiims, open up opportunities for much larger range of uncertainty, or in oder words at the lower limit of the limit of error of nuniptdations of nulerials balances, and peihaps create conditions in which nutesial unaccounted for (LEMUF) The output cordd then be stated to be large scale diversions by criminal or terrorist groups could occur without tinwly either at the lower or at the ur.per limit of de LEMUF. If the material output detection.
were stated to be at the lower limit, the excess nuterial,if any,could be diverted and secsetty kept or disposed of. If, however, tie output were stated at the i
DIVERSION BY A POLITICAL FACTION upper limit, the plant management might be able to charge its customers for l
WI fHIN A NATION more nuterial than was actually present.
Scope of the Risk Reasons The governnwnt of a nation is normally not of one mind. The The nunagers of a nuclear enterprise may want ro divert material m possession by a faction or interest group within the government of enough order to cover up previous nuterial kisses known to the nunagement but not yet nuclear nuterial in a suitable form to make a few weapons might significantly discovered by the AEC authorities. The nunagers nuy want to have some affect stie internal balance of political forces within a nation. This particular sisk clandestine material on hand simply as a convenient way to remove material of nuclear diversion would seem negligible in sie U.S. Ilowever, it could be
118 Nucifar Thett: Risks and Safeguards Risks of Nuclear Theft 119 substantial in a nation where force was commonly used as a means of Finally, it 'nuy be noted that in a cmmt y where siolence is transfening governmental power and authority. It should be noted that in considered to be a necessary catalyst for p>htical change, a W itical faction nuy l
countries where force is frequently used as an instrument for political change, decide to dsop out of the govemment, take to the hills, and begin a civil war. A the hne between pditical faction and aiminal group would sometimes be group which ca<ried with it a significant quantity of nuclear weanm nuterial ddficult to draw. This diversion risk is considered briefly here because of its would be in a far different piditical positism than one wisch took'along imly potential bearing on U.S. foreign relations and its relevance to the possible conventional arms and deemical esplosves.
development of a nuclear black market.
NUCLEAR BLACK MARKET Reasons For Diversion The overriding reason why a political faction within a governnwnt The existence m lack of a nurket for stolen nuclear nutesial, and the might want to divent nuclear weapon material would be to enhance its power to characteristics of such a market, would substantially affect the divessior. risks aclaieve its own immediate or future political objectives. The specific objectives previously considered. In general, the profit incentives for nuclear diversion might he either domestic or international.
would be increased greatly if stolen miclear nuterial were easy to thspose ofin in terms 01 domestic politics, preemptive diversion by a political transactions on a black market. Although the obstacles in the way of black faction in order to shore up ;ts power base is one possibility protective diversion nmLet development appear quite large, the potential for profits by the by a faction fearing it was about to be suppressed or outlawed is another. In middlemen in the market could also be very great.
cither of these circunutances, the reason for nuclear diversion would be to assure Sellers in a nuclear black market might be any of the potential st.ibihty or to deter the use of violence against themselves.The credibility of the thieves previously discussed. A ready nurkct could increase not only the threat or use of nuclear force in a coup #crat would seem difficult to establish, incentives for thefts, but also the probability that stolen material could be however.
successfully ransomed as an alternative to nurketing it. The existence of a In terms of international policy, whether or not to acquire nuclear well-developed black masket would peihaps he especially pernicious, because it weapons is an issue that is likely to be on the governmental agenda of many would ease the problems an individual acting alone would othe. wise face in non-nuclear weapons nations from tine to time in tie future. Adherence to the disposing of any nuclear material he might steal.
nuclear non-proliferation trcaty and acceptance of'nternational Atomic Energy Terrorist groupe and national governments are the more likely Agency safeguards cannot be expected to set.le the issue permanently, although customers in a black market. There would also seem to he possibiliths for the such governmental action should substantially strengthen the position of those operators of a nuclear black market to stimulate demand. Termrist groups often within a government who are opposed to the acquisition of nuclear weapons.
appear to emulate each other's tactics. Moreover, an initial sale or two of nuclear Those who favor the development of such weapons may view diversion of weap ms to petty dictators with dreams of glory might thereafter enable the noterial from nuclear industry as a convenient and effective way to confront the operators in a nuclear black market to play on the fears of more respmsible government with a fait accompli, and to reverse in f act the non-nuclea;. weapon leaders, who would then have no way of knowing which nations had secret decision.
nuclear weapon stuckpiles. A nuclear black market could offer the governments of nations without any previous civilian or mili:ary nuclear capabilities Options opportunities for acquiring nuclear weapims. Such a development could, A political faction planning a nuclear diversion might have two ways therefore, greatly lacrease the dangers of nuclear weapon proliferation through.
to accomp'ish the result that would not be available to criminal or terrorist out tie world.
groups. First, the owners m nunagers of an industrial facility with an inventory A black market in nuclear material would seem to require a subtle of nuclear weapm nutesials might actively support one faction against another and complex structure, possibly composed of several loosely affiliated groups.
in an internal power struggle. Therefore, they might be quite willing to transfer The nmket would probably become transnationai in scope since demands for some of the nuterial under their control to tie faction they were supporting, stolen nuclear material or fabricated weapons would not necessarily cone from a and peihaps to provide assistance in weapons manufacture. Second, the armed country that has the sources of supply. Weapm fabrication or material forces, or particular units of the armed forces, might be persuaded to participate processmg services may or nuy not be part of the market operations. If eley in the plot and to seiie the nuclear material that the governmental faction Scre, these activities might take place in remote areas or wlwre a government was wining to kmk tie other way.
wanted.
t i
0 120 Nuclear Theft: Risks andSaferiards Ctsapter Seven A criminal or terrorist group might thus target its efforts on Nuclear Safeguards: Basic Consideration'-
especially vulnerable nuclear fuel or facilities anywhere in the world. The stolen noterial might then be passed through various middlemen and processing steps and sold ultinutely to purchasers in other countries far away from the scene of original theft.
The evolution of a nuclear black market would be a hazardous and uncertain affair.11 may be doubted whether such a market could ever achieve the institutmnat stability or hmg term viability that would pose a nujor threat.
If one or more major nuclear abeits occur, governments everywlere nuy be prompted to act swiftly and decisively to foreclose any possibilities for disposition of stolen material.1; rom the preceding analysis it would seem, however, that a few successful thefts could increase incentives for black market fornution, and that an incipient nuclear black nurket would incre.:.se the hkchhood of nuclear theft or other types of diversion attempts. It should be rmted that no national gir..rnment actiny, unilaterally could prevent a nuclear black market from developing if the conditions were ripe. Like the risks of nuclear thef t, the dinensions of a nuclear black market are potentially global.
Thus far in this study we have examined the nugnitude of the U.S.
rmclear power ladustry and tim potential risks of nuclear ineft. We have also discussed the present AEC regulatary requirements designed to protect and account for nuclear materials, and observed that a safeguards system is not yet fully developed. Clearly, umch remains to be done-and urgently-if an effective system of safeguards against nucicar theft is to be fully operational before very large anmunts of fission explosive nulerials begin to flow through the U.S.
nuclear power industry, in this chapter we explore a number of basic issues related to the developnwnt of a nuclear safeguards system, including how effective such a system should be, and we also suggest a framework for the development of a ivarleiy of safegaard options. In chapters 8 and 9 we analyze specific safeguard nwasures and ccmsider the costs of a safeguards system.
THE CONTEXT We are concerned in this study with safeguards to ensure that nuclear nuterialis not diverted fro n civilian industry to an illicit use. This particular objective should be viewed as part of regulating and controlling the civilian nuclear power l
industry in order to achieve several important purposes that are in the public interest. Aside from safeguards to prevent or detect theft, the control of nuclear nuterial is necessary for two nujor reasons: to ensure thai valuable materials are imed efficiently as fuel for the generation of electric power or heat; and to i
ensure that radioactive noterials that could endanger human health are used safely and are not inadvertently released to the environnwnt in dangerous
(
uuantities or willfully dispersed by acts of sabotage. Controls desigswd to avert inefficient or unsafe use of nuclear material nmy either complement or conflict with safeguards to ensure against theft.
121 l
q7 l
t 122 Nuclear Theft: Risks andSafeguards For r. tample, governmental material accountancy requirements may pohtical ugecavals, and, perhaps above all, shifts in public attitudes towards largely build upon inventory controls adopted by the plant management in the interest of efficient processing operations. As ancther example, both public Fu,s much having been said,in order to move further in our awlysis health and safety and safeguards agamst theft point toward the useof specially nmst grapple with the term " acts of nuclear violence," and with the term developed heavy containers for tie shipment of nudear nuterials. Ilowever,
,"eff.
ective assurance." in attempting to give nure concrete meanings to such plutonium that is shipped in the form of an oxide powder is less hazardous to terns, we must distinguish the practical from the impractical, the obtainable public health, but slightly more rI a bomb ri-k in tiw event of theft, than frmn t!w unobtainable.
nintonium that is shipped in the form of a liquid nitrate solution.
" Acta of nuckar violence" might encompass an infinite variety of circumstances ranging from hoaxes, to threats involving actual nuclear weapms, to actual fission explosions or intentional plutonium dispersal. At one extreme, PURPOSES OF A NUCLEAR g
st is impossible to provide assurance against the occurrence of nuclear imeats SAF EGU ARDS SYSTEM that are hoaxes. As we saw in Chapter 6, all sorts of people could make a nuclear threat that is credible-at least up to some point-and still be a hoa t. It is Perhaps the nmst difficult task of all in developing and implementing a nuclear doubtful that any responsible governnwnt would comi elely ignore a n uclear d
safeguards system is the formulation of nwaningful objectives. It was relatively lxnnb threat. much less publicly declare it to be a hoax simply because the easy to develop an objective for the US. space program in the 1960s. President threat was not substantiated by receipt of a nuclear exphsive design or a snull Kennedy did this in 1961 when he said: "We shall place man on the nmon and anmunt of fission explosive nuterial. nreats using radiological weapons could bring him back to earth before the end of this decade." It is also possible to be even more credible with mininut amounts of substantiatinginformation.ne develop " full employment" as a continuing national goal and then to define a 4 real question, therefore, is whether a nuclear safeguards system can provide or 5 percent level of unemployment as unsatisfactory performance.Though it is assurance that hoaxes can be distinguished from real threats, and that seal noch nere difficult for the United States to nuintain full employment than to threats would be nest unlikely.
place a man on the nmon, both objectives are nuaningful to government, to At the other extreme in the range 'of acts of nuclear violence.ne i
industiy,and to the nun in the street.
unannounced fission explosions in urban areas. Ilere agair. we must conclude, When it comes to nuclear safeguards, what should be the objective of regrettably, that regardless of its effectiveness, a nuclear safeguards system US. policy? We may initially and tentatively state the purpose of a nuclear safeguards systems as follows: to proride effective assurance against acts of applicable to the nuclear power industry in this country cannot provide nuckar ri<dence using meterial un&rwfrdIf obtained from the nuc/cor powr complete assurance tint unannounced fission explosions will not occur in the industry. When words are strung together in this way, the result is an opaque and United States in the future. Apari from the fact that a foreign government might abstract statement of the problem. Ilowever, it should be noted that many accidentally or intentionally explode a nuclear weapon in tlw United States, a statements of purpose in legislation and administrative regulations are even more fission explosive might be smuggled from a foreign country by a terrorist group and then detonated.
vague and less meaningful. For example, the legislative standard in the Atomic Energy Act for evaluating US. nuclear nuterials safeguards is that the controls Furtiermore, the possibility llat a past, undetected theft is the must provnte assurance against activities " inimical to the common defense and source of a real nuclear threat cannot even now be discounted entirely. ne seemity or to the health and safety of the public." Neverthebss, our tentative anmunt of firsion explosive nulerial unaccounted for in the US. nuclear power formulation of purpose set forth above is useful as a point of departure.
mdustry and industrial enterprises performing work under contract for the AEC We have avoided use of the word " goal" in our statenwnt regarding has already eneeded the point where complete assurance against immh threats safeguards because this word seems to imply the existence of some milestone using diverted nuterial is possible. Moreover, no future safeguards system that which, if reached, signals the completion of a task. The risks of nuclear theft will will be practical can offer 100 percent assurance against theft.
Een who is to decide, and on what basis, what level or risk of persist in the foreseeable future, though it will be possible to reduce their hkelihood and impact considerably. Consequently, the development and nuclear vMience can or should be acceptable as a social cost of the use of nuclear maintenance of effective safeguards will require continuing effort. Specific goals energy to neet future needs for electric power? Is the explosive destruction or h
P tonium contamination of a large than area somewhere in the world to he and objectives w,Il probably have to be revised often in the light of advances in nuclear technology, growth of the nuclear industry, changes in the level and tolerated if it does not occur nmre than once a year? Once every fifty years?
3 Never? Dese question *. need to be addressed by political leaders, not character of acts of violence (not necessarily nuclear), national and international I,
professional experts.We do not presume to answer them in this study.
it 124 Nuclear Theft: Risks and Safeguards Nucleadafegua@: M &nkak m Very difficult issues 2.150 arise when we try to define " effective when the nun in the street renuins unaware of the nature and scope of the risk assurance" in our tentative statement of purpose of nuclear safeguards. Even if to which he will be exposed. Nor can general public consent be inferred from agreenent could be reached ccmcerning some nuximum acceptable level of risk broad legislative delegations of relevant authority to the AEC and the Joint of nuclear violence using nuterial stolen from the nuclear power industry,how Conunittee on Atomic Energy of the Congress, when public hearings on this effecrire should the assurance be that the specified thresholds of violence will specific nuclear risk have never been held, and when umst members of Congress not be exceeded? Who should decide, on what basis, what level of effectiveness remain as uninformed in this respect as the people (lut elected them, is sufficient, given the fact that 100 percent assurance is impossible, no matter Finally, the problem of nuclear theft exists wherever nuclear inwer what we do? Perhaps a hmk at nmre famdiar hazardous hunun activities can industries exist. A seccessful nuclear theft in une country nuy result in shed some light on these questions, widespread destruction in another, far distant country a few weeks or several Are present highway, vehicle, and operator licensing safeguards years later. Attitudes toward levels of risk and effective tess of nuclear against serious autonmbile accidents in the U.S. " effective?" The American safeguards can be expected to cover at least as wide a range between countries as people are apparently willing to tolerate umre alun 50,000 deaths per year as a between groups within one country, such as the United States.
. result of autonmbile accidents, and many drivers still abject to the cost or Given the difficulties discussed above,it seems that all attempts to inconvenience of rudimentary safeguards, such as seat belts and simulder L: raps.
develop a meaningful statement of overall goals for a nuclear safeguards system Measures that would reduce the highway death rate to,let us say,500 people per nny well end in frustration. Ilowever, our discussion thus far does Icad us to year would probably be called highly effective, yet they would not lessen the conclude 3s follows: In view of the seriousness of the risks arising out of a giicf of someone whose wife or husband or child was one of the 500 fatalities.
successful nuclear theft, the safeguards system applicable to the nuclear power We accept a low commercial aircraft accident rate, and an even lower industry should employ the best available technology and institutional mech-train accident rate, and delegate to experts the decision as to how safe our anisms. The safeguards system should be developed and implemented with a commercial aircraft or railroads should he. In these and other matters of public view to keeping the risks of nuclear theftaslowaspracriarble. We believe these safety, the level of risk denunded by society as a whole, and even by individuals, statements can serve as a useful guide to the development and implementation of is never iero. A combination of attention by safety experts, promotion by a nuclear safeguards system that will function effectively in a dynamic world in people who nuke a living from the hazardous activity, and public outcries when which technological, economic, social, and political factors are changing rapidly.
the risks begin to twm tem large, tends to produce a level of risk that is generally accepted. Perhaps the acceptable level of effectiveness of nuclear safeguards FUNCTIONS OF A NUCLEAR SAFEGUARDS SYSTEM could evolve in a similar way over time.
liefore adopting this approach, however, several distinguishing factors should be taken into account. In transportation accidents the number of in order to provide effective assurance against acts of nuclear violence using human casualties per crash,a few in the case of autonmbiles and one hundred or nuterial stolen from the nuclear power industry, a nuclear safeguards system as a nure in airplanes, is comprehensible. The frequency of accidents involving whole simuld perform four interrelated functions:
fatalities in relation to passenger. miles traveled can be determined. 'lhese statistics provide a basis whereby persons can make individual judgments 1.
prevention of theft; conctrning levels of risk involved in travel by a particular nmde,and voluntarily 2.
detection of theft; decid< whether the I enefits to them are worth sne risk.
3.
recovery of stolen material; When it comes to the risks associated with various levels of 4.
response to threats of nuclear violence.
effectiveness of a nuclear safeguards system, tens of thcusands of human beings nuy be killed in a single act of nuclear violence,and such acts will occur seidom "An ounce of prevention is worth a pound of cure." The relevance at nmst and hopefully never. This leaves us with no basis for weighing the of this old saw to a nuclear safeguards system is apparent from the risk analysis probabilities involved.
in Chapter 6. Nevertheless, by far the most effort to date has been devoted to This seems to nuke it even note important for the people in a the development of means to detect unlawful diversson after it has happened.
democratic society to have an o[imitunity to consent,in some way, to the risk lhe detection method that has received the most aticntion until very recently of nuclear violence implicit in a particular level of effectiveness of safeguards, has been accountancy-record keeping, inventory controls, reports, and indepen-Such consent cannot be presumed from an absence of broad public concern dent audits. It should be noted that accountancy, unlike other possible methmls
- _ - ~.
126 Nuclear Theft: Risks andSafeguards Nuclear Safeguards: Basic Considerations 127 of detecting diversion, such as continuous surveillance, makes little if any FRAMEWORK contribution to the related function of preventing theft.
Fortunately, the developnwnt of means to prevent theft is now in developing a nuclear safeguards system,it is useful to think from a conceptual receiving much greater attention. Such well known and widely used means as framework provided by altree basic questions: What nuy be controlled? Who physical barriers, locks, alarms, etc., are being required. Ilowever, relatively little nuy do the controlling? And what are the means of ctmtrol? We will discuss the effort has been devoted either to the use of substantial manpower or to the fust two questions here. Specific measures to pever,t nuclear diversion, to development of more advanced tecimological methods for achieving physical detect completed nuclear thefts, to recover stolen nuclear nuterials, and to security. His is in nurked contrast to the efforts that have been devoted to respond to nuclear threats will be explored in Chapter 8.
various sophisticated techniques for the assay of nuclear materials,especially the non-destructive nwasurement of the nuterial content of fuel elements, scrap What May lie Controlled?
storage drums, etc.-eIforts which are related to the detection of theft after it NucIcar natterial flows through and between a variety offacilities, nas happened.
from mines to radioactive warte storage. Special information is necessary in The need for means to recorer material after it has been stolen is order to build and operate the facilities and produce, process, and use the now officially recognized. Very httle has been disclosed, about what, if nutriials flowing through the nuclear fuel cycles. And of course, nuclear anything, las been actually (kme to provide for recovery of stolen material or industry would not happen witimut propic. Thus, material, facilities,informa.
noterial that is simply hist. Similarly, little has been said about what happens if tion, and people may be alw subjects of control under a nuclear safeguards I
nuterial is unaccounted for and the anmunts involved exceed the allowable system.
l limits of error. Furthernure, the government has not ye publicly recognized the e
need for contingency plans for responses to nuclear threats. Perhaps government Materials. Nuclear safeguards systems are based primarily on o!tieials will continue to respond on an ad hoc basis, as in the past, or perhaps controls over materials which flow through the various nuclear fuel cycles.
I plans have been developed but not disck> sed.
'Iherefore, detailed discussion of tids aspect of safeguards is necessary at this While efforts to impove na-destructive assay tnd other accoun-point.
tancy techniques designed to detect maeerial theft should be continued, there All nuclear material may be subject to safeguards. Control measures are compelling reasons why nwier efforts should be devoted to development of nr w be initially applied to every shovelful of ore containing uranium or thorium the best practical measures to pevent theft. For one, detection will merely be that is removed from the ground, or they may even apply to deposits of uranium the event which triggers recovery operations, and these operations might well and timrium ore in tiw earth's crust. Safeguards may extend to nuclear material
?
fail. For another, some prevention measures are also effective means for as it flows throughout the fuel cycle and continue to apply to material that is detection befbre the successful completion of a theft. The sianaling of an alarm recycled efter chemical reprocessing. Measures to ensure against theft quy apply nuy also autonatically close exits from a facility, as well as summon on-site to the fissionable material that is produced in a rmclear reactor and to each security forces promptly to the scene of an attempted theft. Some preventive successive generation of fissionable material as it is produced. Safeguards nur neasures should be plainly visible,both as a deterrent to the potential thief who even extend to radioactive waste material that is stored permanently. Such a is only casually investigating a possibility and as a way of building public comprehensive control scheme is now unrealistic and unworkable. Ilowever,.the confidence. The psycholigical atnunphere created by the nuclear safeguards original proposals for nuclear disarmanwnt proposed by tie U.S.gmernment at system nuy be as important as tlw technical capability of the system.
the eral of World War ll-the socalled Baruch plan-called for just sucle a
The public must be as fully informed as possible about the comprehensive. scheme. At that time, the government believed such a scheme to prevention and detection functio,ns of the safeguards system in order to build be necessary as a precondition for the destruction of its own.tockpile of nuclear confidence in its effectiveness. Ik>weves, the extent to which the recovery and bombs, which was then very small, and as a worldwide regulai sry framework for response pl ases should be revealed to the public is a difficult question. Revealing the development ofindustrial uses of nuclear energy.
the details of these paris of the system in order to produce public confidence in Alternatively, a variety of exemptions from a particular safeguards their ef fectiveness could in itself substantially reduce their effectiveness.
system are possible. Nuclear material nuy be exempt when it is present only in llowever, the general public and, even more importantly, any potential thieves small quantities or in certain forms. Thus, if the total quantity of plutonium or must believe that the government has planned carefully about what will be done high-enriched uranium in a country is less than one kilogram, that quantity is to recover any stolen material and to respond to any nuclear threat.
exempt from international safeguards under lie NI'I'. Nuclear material may also I
128 Nuclear Theft: Risks andSa0 guards Nuclear Safeguards: Basic Considerations 129 be exempt from a certain safeguards system, or timse safeguaids nuy be setting exemption lindts, how are these limits to be chosen, and by whom?
suspended and another system imp > sed if and when the nuterial is being used Simuld the limit on plutonium be set at one gram? Ten grams? One hundred fo, certain purposes. Thus, nu'crial nay be e empt from the safeguaids grams? In any event, the present exemption limits of one kilogram (I AEA and appicable to civilian industry when it is used under governmental autho ity m AFC limits for nuterials accountancy) or two kilogranu (AEC limits for physical the manufacture of nuclear weapons or in military propulsion reactors.
protection) are too high if radiological threats with plutonium dispersal devices The establishnwns of safeguards exemption linuis for snull quan-are taken seriou ly.
tilies of nuterial raises a number of difficult questions. Should the exemption Tir : are also difficult questions concerning the exemption of linuts be related directly to the minimum amount of a particular fission certain nuclear nuterials from safeguards requirements for physical protection, explosive material that is required to nuke one nuclear explosive? f so, how based on she extent of the dilution of fission explosive nuterials by other should this amount be determined, and by whom? It some of our previous nuterials. Dilution of uranium-235 with uranium-238 is a special case of this.
discussions, we have used the well retlected, sphenical critical nusses of Below what enrichment level, if any, should an exemption for uranium he plutonium, high-emiched uranium, and uranium-233 at normal densities as established? The 20 percent enrichment threshold which is presently used by the points of reference. As discussed in Chapter 2, however, it has been widely AEC for physical protection is rather a bitrary, since fast critical assemblics can published slut an implosion system car, be used to significantly compress the be nude with uranium enriched somewhat below 20 percent, though not as low the critical nuss decreases as the as the 3 to 5 percent emichment level used for LWR fuel. But how about other core of fission explosive nuterial, an(dd seem possible that significantly compression increases. Therefore, it we dilutants? It nuy be argued that the dilution of high-enriched uranium with snuller quantities of these nuterials than their critical nusses at nornal densities graphite and silicon carbide as in IITGR fuel assemblies, where the dilutants are can be used to nuke fission explosives. But how much smaller, and how very difficult to extract, should be used as the4 asis for safeguards exemptions dependent is the minimum amount on the knowledge and skills of the weapon similar to the existing exemption for low-enriched uraniwn.
designers and fabricators? Are thresholds for exemption related to the types of Alternatively, it nuy be argued that the present AEC exemption fission exphisives that could reasonably be expected to be designed and built by from physical protection requirements of low-enriched uranium should be one individual in a basenwnt type operation? By a highly competent,but snull narrowed. More than half the separative wmk required to produce 90 percent non governmental organization? By the participants in an intensive,long term emiched uranium lus been done in enriching uranium to 3 to 5 percent for LWR effort sponsored by an industrially advanced nation? Given the fact that answers fuel. It is true that further enrichment of 3 to 5 percent fuel is required in order to these questions require access to classified information, how can the public be to use it in a workable nuclear explosive,and the risk that a criminal or terrorist assured that the limits established are reasonable?
group might possess its own enrichment capability is now very snull. But this veihaps even nore difficult questions, which are largely nutters of possibility nuy become more likely in the future.
subjective judgment, involve the possible eventual pooling of stolen nuterials,as Regardless of their quantity, ores containing nuclear nuterial may be in a black nurket. Even if it were possible to specify the minimum amounts of entirely exempt from control or exempt up to a large limit because the difficulty various nuterials that it is reasonable to expect could be used in an illegal fission and cost of processing the materials to usable form nuke the risks arising fmm explosive nunufacturing effort, what portion of these amounts should be theft negli ible. For shnitar reasons, quantities of refined U 0 (yellowcake)
F 3
exempt from physical protection? Simuld the specifie,i portions be time-prior to enrichment may be exempt, although the concentration of uranium in dependent, allowing for the possible buildup of black nurket stockpiles and yctiow cake is very much greater than its concentration in o e. Ilere again the rates of flow?
justification for the exemption of even these nuterials frm safeguards nuy be it nuy also be argued slut the establishment of exemption quantities undercut by future developments in enrichment technology.
should take into account the potentia 1 hazards represented by different materials On balance, given the projected size of the flows of plutonium and if they were used for non-explosive radiological threats. As we have seen, very high-enriched uranium through the nuclear power industry in the U.S., it is small quantities (grams) of plutonium could be used in radiological weapons, reasonable to expect that potential thieves would prefer these fission explosive whereas nuny kilograms of uranium-233 would be required to produce nubrials to low-enriched uranium. Therefore, less stringent controls on comparable hazards. Ilowever, high-enriched manium, even in very large law. enriched unanium, as part of the " graded" safeguards system advocated by quantities,is categorically unsuitable for non-explosive radiological weapims.
nuny govermnents and industrial officials, seems reasonable for the present and if risks of theft of plutonium-hearing fuel nuterials or concen-near future. If a graded safeguards approach is adopted,it is also arguable, as trated fission products for use in radiological weapons are taken into account in indicated above, that fission explosive materials higidy diluted by other materials
1 Nuclear Safeguards: Basic Considera tions 131 130 Nuclear Theft: Risks and Safeguards found it difikult to decide whether the overall risk of theft of nutetials for use that are very ddficult to extract should be subject to less stringent requirements than undduled fission explosive nuterials.
in fission exphisives is greater or less llun the risk of theft of snull amourits of Based on the peceding discussion, Table 7-1 is a tentative listing of plutonium which could cause radiological danuge over a large area.
nuclear noterial categories. The table is presented to illustrate the concept of
" graded" safeguards and is not a specific poposal for a classification system.
Facilities. Safeguard measures may be applied to all or simie of the The various classes are hsted in decreasing order by the degree to which facilities in the fuel cycles through which nuclear nuterials llow. Controls safeguards might be applied to a " strategic" quantity of each nuterial.
applied to nuclear facilitics could be intended to pevent or pomptly detect an illicit use of the facility itself-for example, if the facility were used for the Table 7-1.
Possible Nuclear Material Categories for a System of secret processing of nuclear nuterial which was not subject to safeguards.Such
" Graded" Saf eguards controls are important at the international level, although they are effected mdirectly through nuterial accountancy requirements. To the extent there is a g
danger that nuclear enterprises might engage in nuterial diversion, such I.
Unddured fission explosive materials.
Metallic plutonium.high<ruiched uranium
- acceuntancy rerIuirements would also be important to national governments.
or uranium-233. Oxides or carbides of above materials.
By far the nmst important controls applied to facilities, however,are 11.
Materials suitable for radiologicat de-I% onium.
those intended to reduce the vulnerability of nuterML in them to diversion. In vices whether or not dduled with this respect,the physical protection measures used at different facilities may also
- E#
"E #""
"E#
- 111, i
explosive materials dduted by Plutonium nitrate solution. Mixed pluto-high-ennched uranium or plutom.um oxides vs. storage facilities for fuel oil.er materials that can be separated nium and low <nriched uranium oxides.
4 without isotope separation or " hot Mixed thorium and high-enriched uraniuln assemblics containing highly diluted nuclear materiai; transportation systems for tab" processing facil ties.
oxides. IITGR silicon coated fuet particles.
shipment of high. enriched uranium hexafluoride vs. those for shipment of IV. t ow-ernrithed uranium,whether dduted LWR fuel (without pluonium recycle).
low-enriched uranium LWR fuel assemblies; LMFBR fuel fabrication plants and LWR plants with plutonium recycle vs. LWR fud fabrication plants witimut or not.
V.
Naturat uranium UF, (enrichment plant feed). Up, (yel-lowcaker. Uranium ore, plutonium recycle.
VI. 1horium All forms of thorium.
Information. Control measures may also be applied to nuclear A " strategic" quantity is related to the minimum amount required infornution Governmental classification, industrial secrecy and legal protection to pmduce one fission explosive or radiological device that could be used of trade secrets and patents are examples of kinds ofinformation controls.'Ihe effectively to attack an area of some specified dinwnsions. The term " strategic U.S. government has adopted a classification scheme and stringent contml nwasures to restrict the flow of certain nuclear information within the United l
quantity" may be very difficult to define for any particular material. We States and from the United States to other countiles.
nevertheless see no way to avoid the problem of defining such quantities; otherwise the concept of graded safeguards becomes meaningless and the fact At present, the only renuining areas of classified information that some nuclear nuterials are more dangerous than others is ignored.
relevant to the nuclear iheft problem involve uranium enrichnent pmcesses and Therefore,one of the first steps in the detailed design of a safeguards the design and manufacture of explosives. Other information is widely available, system, must be to establish values for strategic quantities of the various nuclear including detailed information concerning reactors, fuel fabrication, and nuterials used in all fuel cycles, quantities below which safeguards are not to be repmcessing.
applied. For fission explosive materials, values should be established in a llowever, just because information is available does not mean that dassified analysis, although the values determined should he made public. For anyone can use or misuse it. Some nuclear processes are complex and are not ratioactive unterials that could be used in dangerous radiological devices, this understood without considerable scientific or tecimical education. But, in cuidd pmbably be done without reference to c!assified information, general, the most that should be expected fmm controls on access to nuclear Table 71 shows some materials, such as plutonium in various forms, information is a delay before the " cat gets out of the bag."It is noteworthy that as falling into two categories (CI' ass 11 and Class 111). One might argue that the I AEA is specifically authorized under its statute to establish and administer Class 11 should be considered separately from the others, or that such materials safeguards applicable to nuclear information, but,tids authority has remained.
should be at a different level from the one we have chosen. We have simply unused and is generally considered to be unworkable as a basis for safeguards.
l l 132 Nucles Theft: Risks and Safeguads Nucles Safeguads: Basic Considerations 133 People. Finally, controls nuy be applied to people.These con'trols Governnwnes. Governmental authorities nuy not only impose nuy be applied to all perstms, or to all tinise possessing certain knowledge duties and fix respmsibilities on others regarding implenwntation of nuclear concerning nuclear science or technology, or nuy be limited to employees of safeguards, but nuy thennelves perform control functions and assume safeguard.
nuclear enterprises and others with access to nuclear activities. Of course, such ing responsibilities. Moreover, various levels-national, state and local-and controls would be a restriction on hunun freedom,t>ut they maybe justified in kinds -licensing and enforcement -of governmental authouty nuy be brought to view of the risks of theft. For example,as mentioned in Chapter 5, the AEC has bear in different ways. What a government an do depends on the jurisdictional requested Congress to enact legislation permitting the governnwnt to administer and constitutional limits of its authority, and by the technology available. What an approval program for persons who have access to significant quantities of a government willdo in regard to safeguards is,of course, a political nutter.
nuclear weapm nuterial.The legislation s intended to provide a regulatory basis i
for improving the assurance of the tush or thiness of such persons.
International Agencies. International organizations nuy also have duties and resptmsibilities regarding nuclear safeguards. It is imputant to recall Who May Contros?
that the safeguarding duties and respmsibilities of an internatiimal organization One or more of a variety of persons and instituthus nuy act as such as the IAEA have been delegated to it by member states. In general, these controllers of nuterial as it flows through the nuclear fuel cycles.
functions nuy be exercised on the territory of a member state e.ly with the consent of its government. The United States governnwnt has offered to permit Employees. Individual employees of enterpcises that produce, the application of I AEA safeguards to all nuclear activities in the United States, pocess, use, store, or transent nuclear material may be assigned safeguarding except those of direct national security significance. This offer, which is still duties. Respmsibility for safeguards at the employee level may be fixed on one outstanding and renuins to be implemented, is hitended mainly for political or a few employees at a nuclear facihty or spread among all employees. Control effect; namely, to play down the discrimination inherent in the Non Prolifer.
respmsibilities nuy be imposed on individual employees by nunagenwnt as an ation Treaty between nuclear. weapon and non. nuclear.weapm countries operational procedure,or by a governnwntal authority as a legal duty.
Neither the offer nor its implementalism bears importantly on the problem of Given the amount of the nuclear nuterial flows involved, the theft of nuterhl from the nuclear power industry in the U.S. It is intended to incentives for theft, and the uncertainties inherent in any teclu.ical measures,it help reduce the risk of governmental diversion in other countries through seems clear that no nuclear safeguards system can be effective without the widespread adherence to the treaty.
participation of the employees of nuclear operations. Each employee should be fully informed about the risks and consequences of nuclear theft where he is employed lie must clearly understand his duties related to its prevention and detection. And finally, he must comprehend that the penalties for failure to perfmm his safeguarding duties, or for engaging in any unlawful diversion attempt hinuelf, will be swift and severe.
i' Enterprises. Nuclear enterpises nuy have safeguarding duties related to the prevention and detection of theft. The managers of an enterprise i
will, of course, have a vital fina.el interest in efficient operations. Up to a p> int, a manager can be expected to take measures to ensure that nuclear 1
^
noterial is not lost or stolen. That point will be reacled when alw marginal cost of safeguards to the enterprise exceeds alw prospective cost of nuterial hut, wasted, or stolen, regardless of the potential cost to society as a result of a successful nuclear theIt. It is clear that certain nwasures will be necessary at the enterprise level which would not be developed or implemented by managenwnt on its own initiative. A governnwntal authority must impose these nwasures as legal duties on the enterprise. At the enterprise level, respmsibility for failure to perform, or negligence in perfornunce of safeguarding duties nuy be imposed on the enterpise as a whole or on the nunagers of the enterprise,or both.
&m.;c.Sc 4;;.;:. 3 gl c...*.. s: ;yn y g& ; w; ;p y.,g iA Qg j gh, g
.g
- ,.;p f
y.<
/ -
)
b F R E E M A N DYSON a*.'"*k 3
3,'.
? ge ni i
3R i,?
?le,,,..', ' r
,f v.
f)
'rnm; a
s D::S UR31 ~G
,r
,m, l
H H
H
.Ja J
.J
.w su s
a su l
l l
l l
l ll ARPER & ROW, PUBLISilERS NEW YORK ru n t is u r n A s P A nT Cambridge
' London Mexico City l' N D A T I O N PROCHAM
}{agerstown i
Philadelphia i g Sao Paulo San Francisco LL__lt Sydney 1817 1
N g
"** N N ** eig,
--e=
- g.
,g p
-,wq-
.n 4.e -
~y d~'
iP 386
. '
- f *. *
)
w l
p
Russi:n scisntists, who hid be:n building nuctrar rectcrs in isola-tion rnd secrecy, wcra Ebl3 fcr the Srst tims to meet ena another and discuss their werk with censidsrable freed:m. Messes of hitherto secret documents were presented openly to the conference, making available to scientists of all countries almost all the basic scientiSc facts about the Sssion of uranium and plutonium and a large fraction of the engineering information that was needed for the building of commercial reactors. A spirit of general euphoria prevailed. Innu-merable speeches proclaimed the birth of a new era ofinternational Little Red Schoolhouse cooperation, the conversion or intellectual and material resources
.away from weapon building into the beneScent pursuit of peaceful
- l nuclear power, and so on and so on. Some part of what was said in i
l these speeches was true. The conference opened channels of com-i Eddington the astronomer,in the book New Pothmays in Science, munication between the technical communities in all countries, and which I read as a boy in Winchester, not only warned us against the personal contacts which were established in 1955 have been nuclear bombs but promised us nuclear power stations. Here is the succeafully maintained ever since. To some small extent, the habit happier side of his vision of the futura:
of openness in international discussions of peaceful nuclear technol-ogy has spread into the more delicate areas of weaponry and politics.
We build a great generating station of, say, a hundred thousand kilowatts The high hopes raised in Geneva in 1955 have not proved entirely cepacity, and surround it with wharves and sidings where load after load of illusory.
I fuel is brought to feed the monster. My vision is that some day these fuel The technical preparations for the Geneva meeting were made
)
arrangements will no longer be needed; instead of pampering the appetite by an international group of seventeen scientiac secretaries. The j
of the engine with delicacies like coal and oil, we shall induce it to work on scientiSc secretaries worked in New York for several months, driving j
a plain diet (of subatomic energy. If that day ever arrives, the berges, the hard bargains on behalf of their governmeMs, making sure that each trucks, the cranes will disappear, and th a year's supply of fuel for the power-station will be carried in in a tea-cup.
participating ccantry would reveal a fair share of its secrets and I
receive a fair share of the limelight. They worked in obscurity and This visior. had always remained vivid in my mind, together with waded through vast quantities of paper. The success of the confer-thz warning against the military use of subatomic energy which ence was' entirely d ie to their efforts. One of the two Americans in appears a few pages later in the book. Eddington used the word the group of seventeen was Frederic de Hoffmann, a thirty-year-old subatomic" to describe what we now call nuclear or atomic energy.
physicist then employed as a nuclear expert by the Convair Division We all knew even in 1937 that the world would soon run out of coal l
of the General Dynamics Corporation in San Diego, California.
and oil. The possible availability of nuclear energy to satisfy the As soon as the Geneva meeting was over, Freddy de Hoffmann perceful needs of mankind was one of the few hopeful prospects in decided the time had come to give the commercial development of a dark period of history.
1 nuclear energy a serious push. For the Erst time it would be possible In August 1955, while I was quietly working on spin waves in l
to build reactors and sell them on the open market, free from the Brrkeley, a mammoth international conference on the peaceful uses bureaucratic miseries of secrecy. He persuaded the top management cf atomic energy was held in Geneva under the auspices of the of the General Dynamics Corporation to set up a new division called i
United Wations. This was a decisive moment in the development of General Atomic, with himself as president. Central Atomic began its
~
i nuclear energy. American and British and French and Canadian and life at the beginning of 1956 with no buildings, no equipment and no 4
94 i
.t
~
u,, a a. w a.n~,..~ s -
stIrif. Freddy rented a httia red schoolhouse thzt had been aban-tors. Our prim:ryjob was to End out whrth:r there was any speciSc doned as cbsolzta by the San Diego public school system. Ha pro-
)
type cf reretor thst looked promising as a commircial vintura for posed to move into the schoolhouse and begin designing reactors General Atomic to build and sell.
there in June.
He lectures were excellent. ney were especially good for me, Freddy had been at Los Alamos with Edwnd Teller in 1951 and coming into the reactor business from a position of total ignorance.
had made some of the crucial calculations leading to the invention But even the established experts learned a lot from each other. The of the hydrogen bomb. He invited Teller to join him in the school-physicists who knew everything that was to be known about the house for the summer of 1956. Teller accepted with enthusiasm. He physics of reactors learned about the details of the chemistry and knew that he and Freddy could work well together, and he shared engineering. The chemists and engineers learned about the physics.
Freddy's strong desire to get away from bombs for a while and do Within a few weeks we were all able to understand each other's something constructive with nuclear energy.
problems.
Freudy also invited thirty or forty other people to spend the l
The afternoon sessions quickly crystallized into three working summer in the schoolhouse, most of them people who had been groups, with the titles " Safe Reactor,"
- Test Reactor" and " Ship i
involved with nuclear energy in one way or another, as physicists, Reactor." These were considered to be the three main areas where l
chemists or engineers. Robert Charpie, even younger than Freddy, aa immediate market for civilian reactors might exist. In retrospect j
had been the other American in the group of scientific secretaries of it seems strange that electricity-producing power reactors were not I
a the Geneva meeting. Ted Taylor came directly from les Alamos.
on our list. Freddy knew that General Atomic must ultimately get where he had been the pioneer of a new art form, the design of small into the power reactor busmess, but he wanted the company to begin clEcient bombs that could be squeezed into tight spaces. For some with something smaller and simpler to gain experience. The ship 6
reason, although I had never had anything to do with nuclear energy reactor was intended to be a nuclear engine for a merchant ship, and l
l cnd was not even an American citizen, I was also on Freddy's list.
the test reactor was intended to be a small reactor with a very high j
i Probably this was a result of my encounter with Teller the previous neutron Sux which could be used for the testing of component parts summer. Freddy promised me a chance to work with Teller. I ac-of power reactors. Both these reactors wosld be competing directly capted the invitation gladly. I had no idea whether I would be suc-
'I' and the Atomic Energy Commission. Both of them were designed with existing reactors that had already been developed for the Navy cessful as a reactor designer, but at least I would give it a try. For nineteen years I had been waiting for this opportunity to make Ed-during the summer and then abandoned when Freddy concluded l
dington's dream come true.
that thet had no contmercial future. The safe reactor was the only Freddy de Hoffmann was my Srst encounter with the world of Big product of our little red schoolhouse which actually got built.
Business. I had never before met anybody with the authority to make The safe reactor was Teller's idea, and he took charge of it from dreisions sn quickly and with so little fuss. I found it remarkable that
[
the beginning. He saw clearly that the problem of safety would be this authority was given to somebody so young. Freddy handled his j
power lightly. He was good-humored, and willing to listen and learn.
' l decisive for the long-range future of civilian reactors. If reactors were unsafe, npbody in the long run would want to use them. He told He always seemed to have time to spare.
l Freddy that the best way for General Atomic to break quickly into We assembled in June in the schoolhouse, and Freddy told us his the reactor market was to build a reactor that was demonstrably safer plan of work. Every morning there would be three hours oflectures.
I than anybody else's. He denned the task of the safe reactor group in i
The people who were already expert in some area of reactor technol-I the following way: The group was to design a reactor so safe that it ogy would lecture and the others would ' earn. So at the end of the could be given to a bunch of high school children to play with,.
summer we would all be experts. Meanwhile we would spend the without any fear that they would get hurt. This objective seemed to i
afternoons divided into working groups to invent new kinds of reac-i me to make a great deal of sense. Ijoined the safe reactor group and 4
]
e m.
l steady level of operation without melting any of its suel.
spent the nzst two months with Tell:r Eghting our wsy through to One of ths Erst steps toward ths d: sign of the safs ructor was to l
a satisitetory solution of his probirm.
introduce an idsa call:d ths " warm nrutten principle," which says
{
werking with Telizr was as exciting as I had imaginzd it would that warm neutrons are less easily captured than cold neutrons and i
be. Almost every day he came to the schoolhouse with some hare-are less effective in causing uranium atoms to Ession. The neutrons brained new idea. Some of his ideas were brilliant, some were practi-in a water-cooled reactor are slowed down by collisions with hydro-ett, and a few were brilliant and practical. I used his ideas as starting gen atoms and end up with roughly the same temperature as the points for a more systematic analysis of the problem. His intuition hydrogen in whatever place they happen to be. In an ordinary water-and my mathematics 6tted together in the design of the safe reactor Just as Dick Feynman's intuition and my mathematics had Stted
/
cooled reactor, after the postulated idiot has blown out the control rods, the fuel will be growing rapidly hot but the water will still be togzther in the understanding of the electron. I fought with Teller cold, with the result that the neutrons remain cold and their effec-4 as I h d fought with Feynman, demolishing his wilder schemes and tiveness in causing Ession is undiminished, an.d therefore the fuel j
squeezing his intuitions down into equations. Out of our Serce dis-l egreements the shape of the safe reactor gradually emerged. Of continues to grow hotter untilit Snally melts or vaporizes. But sup-l pose instead that the reactor was designed with only half of the course I was not alone with Teller as I had been with Feynman.The 4
- i safn reactor group was a team of ten people. Teller and I did most hydrogen in the cooling water and the other half of the hydrogen mixed into the solid structure of the fuel rods. In this case, when the l
of the shouting, while the chemists and engineers in the group did idiot yanks out the control rods, the fuel will grow hot and with it the l l most of the real work.
hydrogen in the fuel rods, while the hydrogen in the water remains l
Rmetors are controlled by long metal rods containing substances cold. The result is then that the neutrons inside the fuel rods are l
~
you want to make the reactor run faster, you pull the control rods a warmer than the neutrons in the water. The warm neutrons cause less Sssion and escape more easily into the water to be cooled and littis way out of the reactor core. When you want tu shut the reactor ~
captured, and the reactor automatically stabi!!zes itself within a few drwn, you push the control rods all the way in. The Erst rule in thousandths of a second, much faster than any mechanical safety l
operating a reactor is that you do not suddenly yank the control rods switch could hope to operate. So the reactor carrymg half of its l
out of a shut-down reactor. The result of suddenly pulling out the hydrogen in its fuel rods is inherently safe.
control rods would in most cases be a catastrophic accident, including There were many practical disculties to be overcome before as cna of its minor consequences the death of the idiot who pulled the rods. Alllarge reactors are therefore built with automatic control these ideas could be embodied in functioning hardware. The great-est contribution to overcoming the practical disculties was made by systrms which make it impossible to pull the rods out suddenly.
Massoud Simnad, an Iranian metallurgist who discovered how to These reactors possess " engineered safety," which means that a make fuel rods containing high concentrations of hydrogen. He j
l c:tastrophic accident is theoretically possible but is prevented by the made the rods out of an alloy of uranium hydride with zirconium wry the control system is designed. For Teller engineered safety was hydride. He found the right proportions of these ingredients to mix i
n2t good enough. lie asked us to design a reactor with " inherent i
saf2ty," meaning that its safety must be guaranteed by the laws together and the right way to cook them. When the fuel rods emerged from Massoud's oven, they looked like black, bord, shiny of n:ture and not merely by the details of its engineering. It must metal, as tough and as corrosion-resistant as good stainless steel.
6 be rife even in the hands of an idiot clever enough to by-pass the entira control system and blow out the control rods bith dynamite.
After we had understood the physics of the safe reactor and the J
St:ted more precisely, Teller's ground rule for the safe reactor ~
chemistry of its fuel rods, many questions still remained to be an-l was that if it was started from its shut-down ccadition and all its F
swered. Who would want to buy such a reactor? What would they use contrcl rods instantaneously removed, it would settle down to a j it for? How powerful should it'be? How much should it cost? Teller 4
!l i
I I
i u... u l - l insisted from the beginning that it should not be just a tay ist reactor
.t
~
j Berksisy I hid decided n:t ts consider him an enemy. In San Diegs j. cxperts to play with. It must be not caly safa, but also powzrful he became a lifslong friend.
i ' enough to do something useful. What could it do?
After Teller and I and the rest of the summer visitors departed, h most plausible ce for a reactor of this kind would be to the few people who remained at General Atomic undertook thejob i
produce short-lived radioactive isotopes for medical research and of turning our preliminary sketches of the Triga into a working reac-tor.ne Snal design was worked out by Ted Taylor Stan Koutz and j
diagnosis. When radioactive isotopes are used as biochemical tracers t3 study malfunctions in living people, it is much better to use iso.
Andrew McReynolds. it took less than three years from Teller's origi-
]
topes that decay within a few minutes or hours so that they are gone nal proposal in the summer of 1956 for the first batch of Trigas to be as soon as the observation i. over. The disadvantage of short-lived built, licensed and sold.h basic price v as a hundred and forty-four isotopes is that they cannot be shipped from one place 'to another.
thousand dollars, not including the building. %e Trigas sold well and hy must be made where they are used. So our safe reactor might have continued to sell ever s:nce. ne last time I checked the total, 4
come in handy for a big research hospital or medical center that sixty had been sold. It is one of the very few reactors that made 4
i wanted to produce its own isotopes. We calculated that for this pur.
- money for the company which built it.
l pose a power level of one megawatt would be generally adequate.
In June 1959, all the people who had worked in the schoolhouse
% other uses that we envisaged for our reactor were for training to get General Atomic started were invited back to attend the ollicial students in nuclear engineering departments of universities, and for i
dedication ceremonies of the General Atomic Laboratories. The i
! doing research in metallurgy and solid-state physics using beam. of change in three years was startling. Instead of a rented schoolhouse, Freddy now had a magni 6 cent set of permanent buildings con-l j j neutrons to explore the structure of matter. If the reactor was used structed in a modernistic style on a mesa on the northern edge of San l l f:r nrutron beam research, a power of one megawatt would be
- i r
- ther low, and so we also designed a high-powered version that Diego. He had well-equipped laboratories and machine shops, with l
{ I could be run at ten megawatts. Freddy named the safe reactor a staff already growing into the hundreds. In one of the buildings was i
x TRICA, the letters standing for Training, Research and Isotopes, the prototype Triga, fully licensed and ready to perform for prospec-I tive customers. Freddy had persuaded Niels Bohr himself, by com-l General Atomic-In September the summer's work in San Diego was coming to an mon consent the greatest living physicist after the death of Einstein, j
cnd and I took a bus ride to Tijuana in Mexico to buy presents for my to come from Copenhagen to preside over the dedication.
I family. As I was walking through Tijuana after dark, a small dog ran The climax of the dedication ceremony was a demonstration of up to me from behind and bit me in the leg. Tijuana was so' overrun the capabilities of the Triga. Freddy had attached to the speaker's l
. with sickly and mangy dogs that there was no chance whatever of podium a switch and a large illuminated dial. At the end of his I
i i
c:tching and identifying the animal that bit me. So I went to a clinic speech, Niels Bohr pressed the switch and a muilled hiss was heard in b J:da every day for fourteen days to take the Pasteur treatment from the direction of the Triga building. W noise came from the i
, against rabies. The doctor who gave me the irdections impressed on l
sudden release of compressed air that was used to pull the control l'
rods at high speed out of the Triga core.h pointer on the large dial, me f:rcefully the fact that the treatment itself was risky, causing in g
which was graduated to show the power output of the Triga in mega-I j one case out of six hundred an allergie encephalitis which was almost watts, swung over instantaneously to 1500 megawatts and then I
- !as f
- tal as rabies. He told me to figure the odds carefully before l i eginning the treatment. I decided to take the shots, and I was l
quickly subsided to half a megawatt. The demonstration was over. it b
{ iconsequently under some emotional strain for the last two weeks of had been rehearsed many times before, to make sure there would be j
- the summer. Edward Teller was extremely helpful. He had in his no unpleasant surprises. The little reactor did in fact run at a rate of l youth in Budapest lost a font in a streetcar accident, and he knew 1500 megawatts for a few thousandths of a second before its warm how ts give effective moral support in a situation of this kind. In neutrons brought it under control. After the ceremony we went and l
4 l
I l
l
102 l AMERICA Little Red Schoolhouse l 103 r
saw it sitting quirtly et the botton. ofits pool of cooling water. IIere as he was with ths Trigs, it would htve paid cff handsomsly (2r it was. It was hard to believe. Ilow could one believe that nature General Atomic, and the whole nuclear industry of the United States would pay attention to all the theoretical arguments and calculations would be in much better shape than it now is. it is impossible to make that we had fought over in the schoolhouse three years earlier? But res.1 progress in technology without gambling. And the trouble with hers was the proof. Warm neutrons really worked.
gambling is that you do not always win.
In the evening there was a picnic supper on the beach, with l
The HTCR was competing directly with the light-water power Freddy and Niels Bohr and various other dignitaries. After eating.
l reactors which have from the beginning monopolized the United Bohr became restless. It was his habit to walk and talk. All his life he t
States nuclear power industry. Neither HTGR nor light-water reac-had been walking and talking, usually with a single listener who f
tors are inherently safe in the sense that the Triga is safe. Both could concentrate his full attention upon Bohr's convoluted sent-depend on engineered safety systems to push in the control rods and i
ences cnd indistinct voice.That evening he wanted to talk about the shut down the nuclear reaction in case of any trouble. Both have i
futurs of atomic energy. He signaled to me to come with him, and enough residual radioactivity to vaporize the core and cause a major wa walked together up and down the beach. I was delighted to be L
accident if the cooling of the core is not continued after shutdown.
1 so honored. I thought of the abbot in the monastery at the foot of F6, I
The main difference between the two reactors is that the HTCR has j
and I wondered whether it would now be my turn to look into the a much bigger core for the same output of heat.The HTGR core has crystal ball. Bohr told me that we now had another great opportunity such a great capacity for soaking up heat that it will take many hours
! to gain the conSdence of the Russians by talking with them openly to reach the melting point after a shutdown, even if there is a com-i l cbout all aspects of nuclear eaergy. The first opportunity to do this l
p!cte failure of emergency cooling systems. A light-water power had been missed in 1944, when Bohr spoke with both Churchill and reactcv core will melt in a few minutes under the same conditions.
l Roosevelt and failed to peuusde them that the only way to avoid a I
The wo st conceivable HTGR accident would be an exceedingly l
disastreus nuclear arms race was to deal with the Russians openly
[..
. messy aBair, but it would be deEnitely less violent and less unman-i beftro the war ended. Bohr talked on and on about his conversations 1
ageable than a comparable accident in a light-water reactor. In this with Churchill and Roosevelt, conversations of the highest historical
' sense the HTGR is a fundamentally safer system.
importance which were, alas, never recorded. I clutched at every The HTGR is not only safer than a light-water reactor but also w:rd as best I could. But Bohr's voice was at the best of times barely more eEcient in its use of fuel. These are its two great advantages.
cudibla. There on the beach, each time he came to a p,articularly It has two gres3t disadvantages: It is more expensive to build, and it i
crucial point of his confrontations with Churchill and Roosevelt, his 1
has more disculty with controlling the leakage of small quantities of voies seemed to sink lower and lower until it was utterly lost in the l
radioactive assion products during normal operation. Freddy gam-I sbb and How of the waves. That night the abbot's crystal ball was bled on the expectation that superior safety and eEciency would in s
cloudy.
[
the long run cause the world to turn to the HTGR for electric power.
Fcr Freddy, the Triga was only a beginning. He knew that Gen-He may well turn out to have been right, but the long run was too ertl At:mic's survival would in the end depend on its ability to build long for his company. In the short run, the disadvantages of capital and sell full-scale power reactors. Already in 1959 the major part of g
cost and of complexity of the leakage containment system stopped the laboratory's efforts were devoted to the development of a power him from breaking into the market. He sold only two HTGRs and rrect:r. Freddy had decided to stake his future on a particular type
- b never went into production with a full. scale model. Finally, in the
- , e f powrr reactor, the High Temperature Graphite Reactor or HTGR.
f late 1970s the political uncertainties surrounding nuclear power
', All of us who were involved with General Ator.:e sapported this made the outlook for the HTCR seem commercially hopelen. Gen-decist:n. It was a big gamble, and it ultimately failed. But I still think eral Atomic canceled its c'ontracts with its few remaining HTGR Freddy's decision was right. If he had been as lucky with the HTCR customers and announced that it was no longer in the Ession power I
j I
e i
I l
a 1
n i
Little Red Schoolhouse l 105 tractor business. Several yeers cerlizr, Freddy htd mcved across the strest fr:m Generel Atomic to becoma presidrnt of the Salk Institute and mtn gtrs took contrel. N t only in priv:ta industry but also in f'
for Biological Studies. General Atomic still continues to build and sell the government laboratories, at L~s Alamos, uvermore, Oak Ridge i,
Trigas and to support an active program of research in controlled and Argonne, the groups of bright young people who used to build
{
fusion. No longer is nuclear Ession power a promising new frontier and invent and experiment with a great variety of reactors were for young scientists and forward.looking businessmen.
disbanded. The accountants ahd managers decided that it was not What went wrong with nuclear power? When Freddy invited me -
cost effective to let bright people play with weird reactors. So the i
weird reactors disappeared and with them the chance of any radical to work on reactors in 1956,1 jumped at the opportunity to apply my T
talents to this great enterprise of bringing cheap and unlimited en-improvement beyond our existing systems. We are left with a very small number cf reactor types in operation, each of them frozen into i
trgy to mankind. Edward Teller and the other inhabitants of the i
schoolhouse all felt the same way about it. Finally we were learning a huge bureaucratic organization that makes any substantia: change j
j how to put nuclear energy to better use than building bombs. Finally imposGble, each of them in various ways technically unsatisfactory,
{
we were going to do some good with nuclear energy. Finally we were each of them less safe than many possible alternative designs which
{
going to supply the world with so much energy that human drudgery have been discarded. Nobody builds reactors for fun any more. ne i
i spirit of the little red schoolhouse is dead. That, in my opinion, is and poverty would be abolished. What went wrong with our dreams?
There is no simple answer to this question. Many historical forces what went wrong with nuclear power.
+
j conspired to make the development ofnuclear energy more trouble-When my father was a young man, he used to travel around some and more costly than we had expected. If we had been wiser, Europe on a motorcycle. Sixty years before Robert Pirsig, he learned we might have foreseen that after thirty years of unful611ed promises to appreciate the art of motorcycle maintenance and the virtue of' j
a new generation of young people and of political leaders would arise technology based upon respect for quality. He sometimes came to a
who regard nuclear energy as a trap from which it is their minion villages where no motorcycle had been before. In those days every i
to liberate us. It is only natural that the dreams of thirty years ago rider was his own repairman. Riders and manufacturers were to-should not appeal to the young people of today. They need new gether engaged in trying out a huge variety"of different models, visions to keep them moving ahead. It is easy to understand in a learning by trial and error which designs were rugged and practical i
g neral way why the political atmosphere surrounding nuclear en-and which were not. It took thousands of attempts, most of which l
ergy has changed so markedly for the worse since the days of the ended in failure, to evolve the few types of motorcycle that are now little red schoolhouse. But I believe there is a more specific explana-on the roads.:The evolution of motorcycles was a Darwinian process ti:n for many of the troubles which now beset the nuclear power '[
of the survival of the Ettest. That is why the modern motorcycle is ellicient and reliable, industry. This is the fact that within the indutry itself,the spirit of 1 the schoolhouse did not prevail.
Contrast this story of the motorcycle with the history of commer-i The fundamental problem of the nuclear power industry is not cial nuclear power. In the worldwide effort to develop an economical ratetor safety not was te disposal, not the dangers of nuclear prolifer-nuclear power station, less than a hundred different types of reactor sti:n, real though all these problems sue '"he fundarnental problem have been operated. The number of different types under develop-i j
of the industry is that nobody any longer has any fun building reac-ment grows constantly smaller, as the political authorities in various ters. It is inconceivable under present conditions that a group of countries eliminate the riskier ventures for reasons of economy.
enthusiasts could assemble in a schoolhouse and design, build, test, here now exist only about ten types of nuclear power station that i
license and sell a reactor within three years. Sometime between 1960 have any hope of survival, and it is impossible under present condi-i and 1970, the fun went out of the business. The adventurers, the
}
i tions for any radically new type to receive a fair trial. This is the i
Experimenters, the inventors, were driven cut, and the accountants fundamental reason why nuclear power plants are not as successful i
as motorcycles. We did not have the patience to try out a thousand I
i I
i I
i
I S..,g k
, u'
. ~
v b
+ -. -.
3.. p.
..a. N..
e.
y d 4l
'N.b l l\\
1
~'7' i 106 l A M ERIC A Gl-e'I'i.d; 4 different designs, and so the really good reactors were never in.
vented. Perhaps it is true in technology as it is in biological evolution g" 4 i
that wastage is the key to efficiency. In both domains, small creatures
. lj.
7 evolve more easily than big ones. Birds evolved while their cousins 54 h the dinosaurs died.
l f -C$i.]
i 6
is there any hope for the future of nuclear power? Of ecurse there l
"Q I
is. The future is unpredictable. Political moods and fashions change
- ./
/
fast. One fact that will not change is that mankind will need enor-mous quantities of energy after the oil runs out. Mankind will see to it that the energy is produced, one way or another. When that day 80 tutu b' i
', 1 a
- .',l comes, people will need nuclear power reactors cheaper and safer than those we are now building. Perhaps our managers and account-ants will then have the wisdom to assemble a group of enthusiasts in e
d a little red schoolhouse and give them some freedom to tinker l'
June 5,1927,-
The begint around.
l 4,
I Cerman town s
3'L
}
the Verein fU f,'
Travel Society
(
for six years b l
l carried throu3
,, j rockets witho-I O !!
the first roma-
,tj j
{
organization s LM j the initiative
....O..
Braun joined t
^"'
.;c ;
j
}
played an acti gft,7 ]
{
a strange way 3.!
duced at the 2
!;.T, Germany and
""..y Cermans of t
- -.3 4
)
efforts by the a
lT, ~.J.l 9 them. The Vf
-.}j I
a historian, w neer. By his -
Chlivion, as C a
~,., - [ !'.
him to Cante n.
s
,4 Willy Ley
-m
,' !, j h twer.ty-seven t'
- g [
Sn.ce Travel
-],
' !j f.
-:. s
}
I Z
xn.-~ e,==n.r.,y., -mq gw.~9:a =:; -
W _.
.,,,g #
=
- k. ;...
N-
-