RE: Two questions on Japan accident

["Minnema, Douglas" <Douglas.Minnema@ns.doe.gov>]

Steve,

Let me try to answer some of the questions here.  (For your info, I have
both health physics and criticality safety knowledge and experience.)

See below:

> -----Original Message-----
> From:	Steve Cima [SMTP:cima@bellsouth.net]
> 
> Can a knowledgeable radsafer educate me a bit by clarifying two points
> from the IAEA news release on the accident:
> 
> 1. Is 18% normal for power reactor fuel enrichment?
> 
	Typically, US power plants will operate at around 5% or less
enrichment.  However, if the fuel is for a fast reactor, such as has been
suggested here, then 18.8% enrichment is appropriate.

> 2. Is 35 lbs @ 18% really likely to go critical?
> 
	Depends on the configuration, but if the material is in solution it
does not take much.  For example, fully enriched uranium, under ideal
conditions, can go with as little as 700 grams.  It depends heavily on the
physical geometry, concentration, presence of reflectors, ratio of fissile
material to hydrogen in the solution, etc.  The simple answer obviously is
"yes".

	Also, a few other comments to be added based on other messages sent
out:

	1.	Criticality accidents such as these generally do not have
significant "explosive yields", because while it may seem fast, the energy
release is still fairly slow.  The physical damage is usually due to vapor
generation and rapid expansion in the solution -  "steam explosions".  There
is also the potential for hydrogen generation and explosion under the right
conditions.  Even in Chernobyl, which was a criticality accident, the
physical damage, other than fuel melt, was due to these mechanisms.

	2.	Radiation doses from criticality accidents can be very
severe, and potentially lethal, for anybody within the first 25 feet or so
from the event.  This is not a hard and fast rule, but an observation based
upon previous events.  Also, since we are looking at acute effects, the
doses should be evaluated as absorbed dose (rads or Grays) rather than dose
equivalent (rems or Sieverts).  Its the difference between stochastic and
non-stochastic effects that counts.

	3.	Depending on the system, there will be some material
released, dominated by the noble gases and halogens, but these will
typically be minor compared to the direct radiation levels, which can extend
to a great distance, albeit not at lethal levels.  Localized contamination
could be significant, wide-spread contamination not so much so.  DOE's
Release Fractions Handbook has a good discussion of the potential quantities
involved.  Generally there is not a large fission product inventory from
such events.

	4.	While criticality accidents can and obviously do occur, each
one is unique to the point that it is difficult to find generic solutions to
the safety issues.  Therefore, the normal criticality safety evaluations
depend heavily on the analysis of each individual process and its potential
upset conditions.  One thing almost universally applied is the
"double-contingency" principle, which states that no single failure in
system or operation should be able to lead to a criticality event.  This is
taken to include the concept of "double-batching", which states that the
design should be able to take twice the planned quantity of material and
still stay subcritical.  In this case, that is what was violated, since it
appears to be more like a "sextuple-batch" (6x).

	5.	While the number may appear to be small, a k-effective of
1.044, as stated in one message, is significantly above critical.  It would
take fairly large changes to the system to take it subcritical.

	6.	A system like this will behave in a quite predictable
manner, actually.  It will progress down one of two possible paths,
depending on the size of the initial pulse, which depends on several
factors.  If the initial pulse is big enough to disrupt the system
('belching' out enough liquid, disrupting the geometry, etc), then the
system will go subcritical and stay that way.  If smaller, then the system
will oscillate through several more smaller pulses until it reaches a
thermal equilibrium at some power level.  It could then stay operating at
that level until it loses enough liquid to go subcritical, or something else
disrupts the system.  In this case, that apparently was done by the removal
of the water coolant blanket, which acted as a neutron reflector.  (Note
that the natural reactors of Africa operated for millions of years once an
equilibrium condition was established.)

	My personal view is that these should be treated as an industrial
accident, probably on a scale of a small chemical plant explosion - very
serious to the workers, mostly an irritation and minor long term risk to the
public.  It is unfortunate that the fear of radiation has blown it out of
perspective.

	Sorry for the abuse of the bandwidth, but I hope this all helps.

	Doug Minnema, Ph.D., CHP
	Radiological Control Advisor for Defense Programs
	Department of Energy
	<Douglas.Minnema@ns.doe.gov>

	what few thoughts i have are truly my own


************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html