Letter to Bulletin of the Atomic Scientists

Published March/April 1997 as "Two tracks are better than one"
(Published in lightly edited form, with inexplicable
deletion of the final sentence)

Richard L. Garwin
IBM Fellow Emeritus
Thomas J. Watson Research Center
P.O. Box 218
Yorktown Heights, NY 10598-0218
(914) 945-2555
FAX: (914) 945-4419
INTERNET: RLG2 at watson.ibm.com

November 22, 1996
Revised December 18, 1996

Mr. Mike Moore Editor Bulletin of the Atomic Scientists 6042 South Kimbark Avenue Chicago, IL 60637

Dear Mike,

The Nov./Dec. 1996 issue was a feast, with excellent articles by Rebecca Johnson, by Khariton, Adamskii, and Smirnov, by Frank von Hippel and Suzanne Jones, and by Ed Lyman, not to mention the usual valuable sections such as the NRDC Nuclear Notebook and the Bill Arkin piece.

I limit myself to comments on "Take a Hard Look at Subcritical Tests" by von Hippel and Jones, and "Weapons Plutonium: Just Can It" by Lyman. I agree with von Hippel and Jones that if the U.S. conducts any sub-critical weapon tests underground, then we should take the lead in "transparency measures" so that it is perfectly clear that these are not hydronuclear tests or super-critical tests producing even a gram of fission yield.

I am confident that a Department of Energy Stockpile Stewardship Program can maintain the U.S. stockpile of nuclear weapons of existing type safe and reliable for decades or even centuries, but I would close the Nevada Test Site and save several hundred million dollars per year. Sub-critical tests, as detailed by von Hippel and Jones, would then take place above ground at Los Alamos, as is planned for DAHRT.

Whether such tests are worth the money is another question, that has as much to do with the maintenance and sharpening of skills as it does with the data obtained. Reasonable people differ on this. In addition, the performers of work tend to argue that more such work is required, while those concerned about expenditures may be on the other side. But even the performers sometimes underestimate what needs to be done.

As an author of the two National Academy of Sciences CISAC volumes on "Management and Disposal of Excess Weapons Plutonium", 1994 and 1995, I was particularly interested to read Edwin S. Lyman's endorsement of the "can-in-canister" immobilization scheme, in which plutonium oxide would be vitrified without fission products and canned in small containers, which would then be cast in the large canisters in the Savannah River vitrification line for high-level waste.

The CISAC report strongly advocated that the U.S. proceed in parallel with implementation of two options for disposal of the anticipated 50 tons of excess weapon plutonium-- immobilization in glass with high-level fission product waste, and burning as MOX in existing reactors. While it is important to dispose of the excess U.S. weapon plutonium, it is even more urgent to dispose of excess Russian weapon plutonium, given the current economic and political disarray in Russia. DOE has been working with its Russian counterpart, MINATOM and others in Russia to develop the options, and the interim report (1) of a 10-member U.S.-Russian Commission on which I serve stated this recommendation after a further examination of the prospects and problems. This dual-track approach was chosen by President Clinton for the disposition of excess U.S. weapon plutonium and announced December 9 at a press conference by Secretary Hazel O'Leary.

DOE agrees that disposition of excess weapon plutonium will cost money, estimated by CISAC as $ 0.5-2.0 B for either immobilization or MOX option, or for a combination of the two. To avoid delay from impediments in a single approach, it is desirable for the U.S. to pursue both options, but to encourage Russia to dispose of its weapon Pu expeditiously it is essential for the U.S. to include reactor burning of Pu in its approach.

Lyman introduces the concern that U.S. utilities will not be willing to pay the same price for MOX fuel as for LEU fuel (perhaps with a small subsidy) but will hold the Department of Energy at ransom for additional billions. It is certainly true that past governmental programs have sometimes encountered such problems, as in the helium conservation program, and the usual remedy of competition is sometimes perverted by the actions of a powerful member of the House or Senate. Alerted to this potential problem, DOE needs explicitly to consider how to arrange the bidding or other contractual arrangement to accomplish the goal of burning W-Pu as MOX, while limiting the cost incurred. DOE must also take into account potential bankruptcy of the reactor operators.

As for the immobilization option, "can-in-canister" is not my favorite. I don't believe that the adequate consideration has been given to the simpler approach of single-step addition of a stream of plutonium compound to the fission-product stream entering the melter.

CISAC considered a concentration of about 1% Pu in the glass, which would dispose of the 50 tons of excess W-Pu by incorporation into about 1/3 of the glass logs to be made at Savannah River. If the "cans" in the can-in-canister approach occupy 10% of the volume, then another vitrification line needs to be set up (could be done in a glove box) to handle the filling of the cans, and that line has to have 10% of the capacity of the Savannah River line. Furthermore, the glass or ceramic needs to accommodate 10 % Pu instead of 1 %. Finally, there has always been the concern that the plutonium cans could be obtained free of contamination by penetrating radiation by the use of strip charges of explosives against a canister, that would strip the outer shell and shatter the glass, allowing the cans to be removed more readily than by dissolving the entire highly radioactive glass log and chemically separating the plutonium.

It is important to distinguish the burning of MOX as a means of disposition of excess W-Pu from reprocessing and recycle of new plutonium. The fabrication of MOX alone with free plutonium costs more than the purchase of fully-paid LEU fuel, with uranium at anywhere near present prices, and obtaining the plutonium by reprocessing would greatly increase that cost. Fortunately, the adverse economics of plutonium recyle in commercial reactors is aligned with the strong U.S. effort to prevent the proliferation of nuclear weapons.

I believe that both the MOX option and the immobilization option should be pursued by the U.S., and quickly, for the disposition of weapon plutonium.

Sincerely yours,

Richard L. Garwin


(1) J.P. Holdren (Co-Chair), J. Ahearne, R.L. Garwin, W.K.H. Panofsky, J.J. Taylor, and E.P. Velikhov (Co-Chair), A.A. Makarov, F.M. Mitenkov, N.N. Ponomarev-Stepnoi, and F.G. Reshetnikov "Interim Report of the U.S.-Russian Independent Scientific Commission on Disposition of Excess Weapons Plutonium", 16 September 1996.