RE: Japanese nuclear accident: Did the reaction oscillate?

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

Bjorn and others,

Your are quite correct.

Depending on the initial conditions, a criticality accident can indeed
oscillate for some time period until the system thermally equilibriates, and
then it can achieve a quasi- steady state mode of operation, which is
exactly what seems to have happened in Japan.

When a super-prompt critical system first pulses, the reactivity of the
system reflects about the prompt critical point, and the energy is deposited
adiabatically.  The pulse is typically shutdown by either thermal expansion
or doppler broadening of resonance capture peaks in U-238, if present.  This
will leave the system below prompt critical, but still above delayed
critical (and still generating fissions).  If the first pulse did not
physically upset the system too much (e.g., splashing out enough liquid to
go sub-critical), then the energy is slowly lost through cooling, and the
system will return to a prompt-critical state and re-pulse, although
typically at a lower level.  This cycle will repeat several times over the
next minutes to hours, depending on the system, until the system reaches
thermal equilibrium between fission heating and heat losses to the
environment.  At this point, the system will reach delayed critical, and
will continue to operate until something changes, such as further loss of
liquid due to boiling.  If the system has an effective heat sink, then it
could operate indefinitely, since there may not be any boiling, only slow
evaporation.  

Note that this is also what happened in the Russian criticality accident of
1997 with the metal system.  There was no liquid, and the heat losses were
sufficient to not allow the fuel to heat up to melting; consequently the
system operated until it was physically dismantled by a robot after a couple
of days.

A very simple but effective model for this type of system is the
point-reactor kinetics equations, which can be found in most nuclear
engineering text books. 

Hope this helps,

Douglas M. Minnema, PhD, CHP
Defense Programs
US Department of Energy
<Douglas.Minnema@ns.doe.gov>

"what few thoughts i have are truly my own"

> -----Original Message-----
> From:	Bjorn Cedervall [SMTP:bcradsafers@hotmail.com]
> Subject:	Re: Japanese nuclear accident: Did the reaction oscillate?
> 
> Response to my own question:
> >weeks ago). My question is: How can a neutron dose rate be sustained >at 
> >approximately the same level for hours after a criticality >accident?
> 
> As I understand it - the reaction _must_ have been stabilized 
> mechanistically (the reflector effect of the cooling water being one of 
> those factors, other ones: temperature, concentration, geometry...) to run
> a 
> continuous fission reaction. In other words: First a sharp peak in power
> and 
> neutron flux and then a rapid stabilization phenomenon. Right or wrong?
> 
> Bjorn Cedervall   bcradsafers@hotmail.com
> 
> 

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