A Report from Martin Kalinowski 
INESAP: a Working Group of the International Network of Engineers and 
                               Scientists (INES)

 SUBCRITICAL TESTS                                         (48:SUB-CRI.TXT)
The U.S. Department of Energy plans to conduct 6 subcritical tests in 1996
and 1997. The explosions will be carried out in two series named `REBOUND'
and `HOLOG' 300 meters underground in the Low Yield Nuclear Experiment
Research (LYNER) tunnel at the Nevada Test Site which was originally
constructed for hydronuclear experiments. The first subcritical test is
scheduled for June 18, but the date may change. The second is expected on
September 12 and further four tests will follow in 1997. The U.S. DOE
announced the award of 1.5 billionUS $ for a five year performance-based
contract starting in January 1996 to theBechtel Nevada Corporation for the
management and operation of the Nevada Test Site. This information goes
back to a press release of the Department of Energy dated October 25, 1995. 
A short note on these tests and a personal assessment of their implication for
nuclear disarmament and non-proliferation was sent out by me amongst others to
the INESAP discussion list on internet. This started a lively discussion on
the list and I received further personal messages. I learned a lot through
this discussion and would like to present the most important points here.

Inputs to the following information was provided by J. R. Russell, Annette
Schaper, Peter Zimmerman, and others. Some more information I drew from an
article written by Tom Zamora Collina in the January/February edition of The
Bulletin of the Atomic Scientist. These people deserve acknowledgement for the
information and assessment they provided but are of course not  responsible for
the following presentation which may be biased by my perception.
What is a subcritical experiment?
There is still some confusion about what subcritical tests are and what they
are not. It is clear that subcritical tests involve fissionable material and
increase the multiplication of neutrons within the material above the natural
background due to spontaneous fissions. A nuclear chain reaction stops after
two or three `generations' of induced fissions.  As a result the rate at which
nuclear energy is released increases for a short time, but remains very tiny
compared to the chemical explosive. The power released by the fissions
stimulated by the compression of the plutonium remains even below the power
generated by the alpha decay of the plutonium which is due to its natural half
life. Thus, subcritical experiments comply with the US policy to go for a true
zero-yield test ban, if one accepts that zero-yield in fact means almost-zero

Apparently it is not correct that these experiments are always hydronuclear
experiments. In hydrodynamic experiments solid matter behaves like a liquid
due to a shock wave which may for example be caused by a chemical
explosion. If nuclear material is involved this is called hydronuclear

Such tests could be conducted with nearly complete nuclear weapons except
that the fissile pit is replaced by a mechanically equivalent part which is
made for example out of depleted or natural uranium including a very small
amount of fissile material. The fissile material content is kept low enough
to ensure that the material does not become critical upon explosion of the
surrounding chemical explosive. 
In principle subcritical tests could be of that type but not necessarily,
especially not in the case of the six subcritical experiments announced by
the U.S. Department of Energy. The planned experiments involve assemblies
that do not resemble a nuclear weapon. However, the DoE does not rule out
the possibility of conducting `bomb configuration' tests in the future. The
planned experiments will use explosively driven flyer plates to study the
reaction of plutonium under extremely high pressures. The two tests planned
for this year will use 50 kg or less of high explosives. The purpose is to
get data to better understand basic physical properties of fissile
materials. For example the equation of state for plutonium can be
determined under realistic conditions. One particular concern is how the
age of fissile material affects its equation of state. 
 Unfortunately there is one publication that gives the impression that
 subcritical experiments are a class of hydronuclear experiments.
 Robert N. Thorn, Donald R. Westervelt write in the abstract of their paper
 `Hydronuclear Experiments' (Report LA-10902-MS, Los Alamos February 1987):
 `Hydronuclear experiments, a method for assessing some aspects of nuclear
 weapon safety, were conducted at Los Alamos during the 1958-61 moratorium on
 nuclear testing. The experiments resulted in subcritical multiplying
 assemblies or had a very slight degree of supercriticality and, in some
 cases, involved a sligth, but insignificant, fission energy release.'
Why underground?
In my view the main problem of these tests is that they are conducted
underground and are a reason for and a demonstration of maintaining the
U.S. test site and the readiness to resume testing at any time. In
contrary, France has committed to close its test site. Also, verification
of the test ban and transparancy at the testing sites will be complicated
by conducting undergrond subcritical tests. Since the LYNER tunnel could be
accessed by inspectors it is conceivable that measures could be taken to
increase the overall transparency of these experiments, and possibly as
well as of other activities on the test site. Cooperative, on-site
verification would be necessary without being too intrusive with respect to
classified information on bomb design. 
 The currently favored approach for a CTBT does not ban preparations of
underground tests and and does not foresee any "pre-test" verification
 Subcritical tests disperse toxic and radioactive material including the
 relatively small amount of fission products they produce. Therefore, it is of
 advantage to conduct them at traditional test sites in shafts in order to
 protect against radiation releases to the environment. There might be an
 alternative. Los Alamos National Laboratory purchased large submarine like
 vessels that can be used to test the characteristics of explosives with up to
 10 kg of TNT-like explosives. The vessel is completely sealed. The 
 products of the explosions are trapped inside where they can be analyzed 
 or filtered before being trapped inside where they can be analyzed or 
 filtered before being released into the atmosphere. However, it can be 
 anticipated that no licence would be granted to conduct subcritical 
 tests in such a container. It has to be admitted that the isolation and 
 containment capabilities at LYNER are far better and have a large
 safety margin because it was constructed for low-yield nuclear explosions.
 The other type of "hydrodynamic" tests that does not involve fissionable
 nuclear material pose no significant radiation hazard and can be 
 conducted in a laboratory. For this purpose the Dual Axis Hydrodynamic 
 Radiographic Test Facility is under construction in Los Alamos. But this 
 kind of experiments is less suitable for "stockpile maintenance" activities.
 What about proliferation?
 A dangerous door for nuclear proliferation would be opened, if the USA
would be successful in keeping subcritical tests unbanned under the CTBT.
Any other country could consider using subcritical tests as a contribution for
 development and testing of nuclear weapons as long as it is not a non-nuclear
 weapon states party to the NPT. Under current interpretation of the NPT for
 these countries such tests are not allowed, because their only purpose is to
 gather data useful for nuclear weapons. Care should be taken that the CTBT
does not open an ambiguity in this question.
 Can subcritical tests be banned by the CTBT?
 Though they are subcritical, they release a tiny amount of nuclear energy.
 Therefore the Chinese and Indian proposal for treaty language to ban any
 nuclear weapon tests which release nuclear energy would be able to
 cover the planned subcritical tests.
 China in 1996: Ban all "nuclear weapons tests which release nuclear
energy."  The problem with this wording is that it still leaves the loophole of
so-called peaceful nuclear explosions. Recently China has withdrawn this demand
 apparently due to pressure from other nuclear weapons states. However, it may
 be proposed by other countries in the future.
 India in 1995: `Each State Party undertakes to prohibit and to prevent, and
 not to carry out, any nuclear weapon explosion, or any other nuclear test
 explosion, or any release of nuclear energy caused by the assembly or
 compression of fissile or fusion material by chemical explosive or other
 means, at any place under or beyond its jurisdiction or control.'
 The difficulty with this language is that it would cover many experiments in
 many countries which are partly clearly civilian and partly have
 civil-military ambivalence like inertial confinement fusion (ICF). 
 Therefore the proposed Indian and Chinese language is unacceptable without
 further clarification. It seems that India and China may not be
 willing to sign a CTBT that freezes in their view the technological gap
between themselves and other nuclear weapon states. It may well be that subcritical
 tests do jeopardize the conclusion of the CTBT. But some people even suspect
 that these countries want to delay progress or are even not willing to
sign a  CTBT at all. It may be that in order to avoid to be blamed for the
failure of the negotiations India and China put forward far reaching demands
that will clearly be refused by other countries.
 The big progress in the CTBT negotiation was that President Clinton announced
 in early August 1995 that the US is committed to a "true zero-yield" which
 would ban any nuclear weapons explosion without setting a threshold for the
 yield. The exact definition of what would be covered by the `true zero-yield
 ban' is not clear yet, because it the yield is not really absolutely zero in
 subcritical experiments. Even plutonium that lies on the shelf generates
 nuclear energy due to spontaneous fission. Hydronuclear experiments are for
 sure covered by the zero-yield proposal, but even their definition is not
 clear. Criticality may be the deviding line and supercritical 
 experiments would  be banned, but not subcritical ones. They would be
included among so called treaty consistent activities.
 France and Britain soon agreed to the `true zero' ban. Within the USA this
 progress marks a compromise which was reached with a great deal of effort by
 the disarmament proponents. The nuclear weapon establishment feels it gave
away as much as possible. Public objections against hydronuclear testing
did not really put the negotiations in danger and seems eventually to prove
successful. But this may be different for the case of subcritical tests.
 From a realistic point of view it should be clear that no more contraints
can be put to the recognized nuclear weapon states without seriously
putting the conclusion and acceptance of the CTBT at risk.  It may as well
be that the subcritical tests do jeopardize the conclusion of the CTBT and
there are several other sensitive issues still open. Therefore, a
substantial quarrel about subcritical tests might contribute significantly
to put the conclusion of the CTBT in 1996 in jeopardy. Saying this, it should be
 made clear that the reason behind this is the real problem.
 The real problem is of course that nuclear weapon states want to continue
 weapons research and though they argue that this would be required to
 maintain the confidence in their aging arsenal it is apparent that all the
 technologies involved are more or less well suited to continue research and
 development of new nuclear weapons. However, the danger is that with a success
 to conclude the CTBT this may remain below the limit of public awareness and
 We need the CTBT to be concluded this year to bring the world a very important
 step forward in non-proliferation and nuclear disarmament towards a
 nuclear-weapon-free world. We need further steps of nuclear disarmament and
 means to cut back nuclear weapons research and development following soon.