Chornobyl on the Internet, Part 2

by
William W. Zuzak, Ph.D., P.Eng.
April 1996

Chornobyl on the Internet, Part 2

Wayne Cooney joecooney@aol.com notes that a similar experiment was carried out successfuly at the Peach Bottom #1 (Philidelphia Electric) HTGR in 1967. A previous test at Chornobyl had been inconclusive.

3.2 Comments and questions of Will Zuzak

(a) Between 00:28 - 00:32 hrs the power dropped from 500 MW(t) to 30 MW(t). I am surprised that the power could drop that low so quickly. I vaguely recall a figure of 7% (of 3200 or 500 MW(t)?) associated with delayed neutrons.

- Could operators of the CANDU provide comments and comparable figures for the CANDU?

Wally Cichowlas thermofx@idirect.com states If time of reactor trip is zero, then:

(b) At 00:32 hrs, the operator retracted a number of control rods to attempt to bring the power back up to 800-1000 MW(t) at which the experiment was to have been carried out.

To me, this is the crucial point in the sequence at which the experiment should have been abandoned. I am surprised that any knowledgeable operator would have attempted to override the xenon poisoning that had obviously set in.

- Would operators of the CANDU attempt such an exercise?

Wally Cichowlas thermofx@idirect.com states:

In principle, there is nothing wrong with attempting to overcome a xenon transient - provided you have a reasonable expectation of catching it in time with the analyzed/engineered means at your disposal. Individual CANDUs will differ, owing to design differences. I would suppose that most designs would allow their RRS (reactor regulating system) to operate automatically to insert positive reactivity which may or may not be timely enough to overcome the transient (typically, there are anywhere from 20-40 minutes available to do this, depending on the available reactivity margin). RRS will do what it has been programmed to do, within allowable safety limits. If it fails, the unit shuts down for thitysixish hours.

(c) By 01:00 hrs, the power could not be increased above 200 MW(t).

- Why not?

[W.C.: insufficient excess reactivity; steady state not reached; poison still building up]

(d) At 01:23:40 hrs, the AZ-5 button was pushed. Somewhere, I read that insertion of the control rods was a 20 second operation. - How far had the control rods descended before the explosion occurred? Did they descend smoothly or were they obstructed? [W.Z.: On page 71 of his book (see below) Grigori Medvedev states that the 7 m long control rods got stuck at 2.5 m probably because the reactor channels had been twisted out of shape and the rods had jammed.]

It seems to me that it is necessary (and possible) to have shutdown of prompt criticality within milliseconds. - What is the comparable time for CANDU reactors?

(e) It has always been my understanding that the explosion occurred at 01:23:44 hrs (not 01:24:00 hrs as stated above). - What evidence is there that the events described after 01:23:44 hrs actually occurred as stated?

[W.C.: INSAG-7 confirms 01:24:00; NUREG-1250 had explosions at 01:23:48 and 01:24:00]

4. The Physics of the Explosion

4.1 Chernobyl: Ten Years On

Excerpt from the OECD Nuclear Energy Agency report titled "Chernobyl: Ten Years On",

http://www.nea.fr/html/rp/chernobyl/chernobyl.html

.... the operators could not prevent an overwhelming power surge, estimated to be 100 times the nominal power output.

The sudden increase in heat production ruptured part of the fuel and small hot fuel particles, reacting with water, caused a steam explosion, which destroyed the reactor core. A second explosion added to the destruction two to three seconds later. While it is not known for certain what caused the explosions, it is postulated that the first was a steam/hot fuel explosion, and that hydrogen may have played a role in the second.

4.2 Comments and questions of Will Zuzak

(a) Note that the term "prompt criticality" does not appear in any of the descriptions above. The neutron flux in a nuclear reactor (which is linearly related to the power) consists of "prompt" neutrons (emitted immediately during the fissioning of a U235 nucleus) and "delayed" neutrons (emitted from the fission products -- seconds, minutes, hours(?) -- after they are created by the fission process). During the steady state operation of a nuclear reactor, the criticality is exactly 1.0 (i.e. as many neutrons are created as destroyed per unit time). Ability to control the power of a reactor during a reasonable time frame is thanks to the delayed neutrons. Should a reactor ever go "prompt critical", then its neutron flux and power would rise exponentially (as occurred at Chornobyl).

- What are the exponential time constants for the Chornobyl reactor (2.0% enrichment), the CANDU reactor (0.7% U235/U238 in natural uranium), and the American PWR (2.4% enrichment), as compared to a nuclear bomb (93% enrichment)?

- What were the fluxes and/or the relative fractions of delayed and prompt neutrons during the accident sequence described in section 3.1 above?

(b) What is a steam explosion? How is it defined?

(c) Did the following reaction (which is presumably exothermic above 1800 Celsius) occur during the Chornobyl explosion?

2(H2O) + Zr --> ZrO2 + 2(H2)

The exothermic Zircaloy-steam reaction is exothermic for all temperatures. The reaction rate becomes of some importance in typical CANDU accident analysis for temperatures >~1200 C. There is a significant increase in the reaction rate at ~1500 C, above which the reaction can "take off" under conditions in which the supply of oxygen is sufficient. Such takeoff will lead to temperature escalation which is not terminated until either the Zircaloy is totally consumed, significant heat transfer is provided to the Zircaloy, or the reaction becomes oxygen starved.

Patrick Reid e-mail: pjreid@mi.net
ALARA Research, Inc. Saint John, Nb. Canada
Voice: (506) 674-9099 Fax: (506) 674-9197

(d) What is fuel vapour? What was the chemical reaction of its explosion?

Wally Cichowlas thermofx@idirect.com states

At high flux, it is no stretch to reach a point of "fuel centreline melting" - literally a point where the fuel (still in a solid piece, encapsulated in its sheath) will begin to melt near the center of the element. At sufficiently high power, there is no reason to exclude the possibility of fuel actually vaporizing. However, the reference literature indicates that (based on Japanese lab tests of fuel elements having similar designs) "fuel elements begin to disintegrate at an enthalpy of 220 cal/g UO2 (T=3300 K). At an enthalpy of 285 cal/g UO2, the fuel elements rupture, and at 320 cal/g UO2 they explode and disperse in small pieces." It is therefore unlikely that much if any fuel reached a point where it was still contained, but vaporized.

(e) Did "dryout" occur in the fuel channels either preceeding or after 01:23:40 hrs when the AZ-5 button was pushed?

(f) Have any detailed simulations of the explosion been carried out? Wally Cichowlas thermofx@idirect.com states Plenty. INSAG-7 lists some of these, and reviews some of the major ones.

5. Names of Deceased

From the English language book of Grigori Medvedev, "The Truth About Chornobyl", Basic Books 1991, ISBN 0-465-08775-2 and from the "Chornobyl Commission Report", April 1987 archived at http://www.mwukr.ca/, I have compiled the following list of fatalities:

Komar, Stepan -- April 26, 1991 TV interview of Boris Alesheev (fireman) Legasov, Valerii -- nuclear spokesman who committed suicide April 27, 1988 after writing a stinging condemnation of the bureaucracy.

In my opinion, it would be appropriate to archive the full names and biographies of all the casualties of the explosion and its aftermath at some website.

6. Criminal Responsibility

In July 1987, the Soviets held a closed trial of six of the plant personnel, who were accused of responsibility for the accident.

Although they were at home asleep at the time of the accident, Bryukanov and Fomin were convicted and sentenced to imprisonment, as was Dyatlov who was present and does carry major responsibility for the accident. If I recall correctly, the other three were acquitted.

To my knowledge, the transcripts of the trial were never declassified. Also unavailable are the debriefing interviews of all the reactor personnel who were on site at the time of the accident.

Surely, it is time that all information associated with the Chornobyl accident be declassified and made available on the Internet.

7. Documents to be archived at http://www.mwukr.ca/chlin.htm

Over the years, I have written several documents and letters concerning the Chornobyl explosion. These are (will be, I hope) archived at the Project Selo (now defunct but superceded by the InfoUkes site http://www.infoukes.com/ website. Here, I will summarize the main points:

(a) "Steam explosion" is inappropriate terminology to describe the Chornobyl accident. Its connotation is that of a rather benign rupture of an over-pressurized vessel. It is misleading to suggest that "massive steam pressure ... blew the reactor open". Simple over-pressization would not have destroyed the inner core of the reactor. The presence of large chunks of graphite moderator and reactor fuel scattered for several kilometres around the reactor indicates the existence of highly supersonic shock waves emanating from the reactor core. Only a "prompt critical" nuclear reaction or a runaway Zr+2(H2O)=ZrO2+2(H2) reaction could produce the required pressure increases on such a short time scale.

Jim Dukelow js_dukelow@pnl.gov states

When I think of "steam explosion", it's not an overpressurization failure of a pressure vessel, but rather the reaction that can occur when molten metal drops into a pool of water or onto a wet surface. The initial shock wave fragments the metal, and the large surface area of the fragmented metal reacts quickly and violently with the water. Under the right conditions (or, perhaps, the wrong conditions), a steam explosion can very efficiently convert thermal energy to mechanical energy.

I have read plausible descriptions of the Chernobyl accident sequence that involved an initial prompt criticality that fragmented the fuel; a steam explosion reaction of the fragmented fuel with the remaining cooling water, which began disassembling the core; followed by a second, larger prompt criticality spike due to the positive Doppler reactivity insertion caused by the rapid cooling of the fragmented fuel (by the steam explosion). In this scenario, the mechanical energy of the explosion derived from the conversion of thermal energy to mechanical by the steam explosion, and the physical expansion caused by the two prompt criticality spikes.

I have also read other explanations of the sequence which assign greater importance to a steam/zirconium reaction as a source of the energetics (compounded, of course by the subsequent burning of the hydrogen generated in the steam/zirconium reaction).

Michael Baker re vapor/steam explosions:

(7) Corradini, Kim, and Oh, "Vapor Explosions in Light Water Reactors: A Review of Theory and Modelling," Progress in Nuclear Energy, Vol. 22, No. 1, pp. 1-117, (1988)

(8) Corradini, "Vapor Explosions: A Review of Experiments for Accident Analysis," Nuclear Safety, Vol. 32, No. 3, pp. 337-362, (1991)

(2) Light water (H2O) cooled, graphite moderated reactors have a positive void coefficient because the hydrogen isotope absorbs a significant fraction of neutrons (i.e. the reactivity increases markedly when the light water turns to steam). Because heavy water (D2O) absorbs far fewer neutrons, heavy water cooled reactors such as the CANDU alleviate this problem to a great extent.

Is there any reason why heavy water could not be used to replace the light water in the RBMK reactors?

Jeremy Whitlock whitlock@cu32.crl.aecl.ca does not see any problems from the neutronics point of view, but points out the possible complications from the technological, economic and tritium points of view.

(3) Viktor Sheludko, who ran the computer program SKALA at Unit #4, arrived at the scene about 6 hours after the explosion and took the magnetic tapes which contained the reactor parameter readouts over to Unit #1 to get the printouts. He also talked to an eyewitness who was fishing nearby who told the following story:

"First, there was a bright plume of light (burning like a candle) which rose to about half the height of the 150 meter high stack and slowly subsided. About a minute later, another plume erupted to the full height of the stack which was so bright that you could clearly see the red horizontal bands painted on the stack. The explosions occurred several seconds later."

Could anyone corroborate this eyewitness testimony? What are the ramifications of these plumes?

Will Zuzak
CHORN_96.D26 = Chornobyl on the Internet
1996-04-26

Copyright © 1996 Dr. W. Zuzak

since March 1st 1997