The Chernobyl Accident: The Course of Events

Fredrik Arnerup

March 2001


Table of Contents
1. Introduction
1.1. Introduction to theory of fission reactors
1.1.1. The core
1.1.2. Reactor operation
2. The accident
2.1. The RBMK reactor
2.2. The course of events
2.2.1. 25 April
2.2.2. 26 April
2.2.3. 6 May
3. Causes
3.1. Design flaws
3.1.1. Positive void coefficient
3.1.2. Control rod insufficiency
3.1.3. Lack of containment
3.2. Operator errors
4. Conclusions
5. Remaining RBMK reactors
6. Implications for Western reactors
Bibliography

This report describes the chain of events that led up to the explosive failure of the fourth nuclear reactor at Chernobyl, near Kiev, Ukraine. Unit 4 was destroyed in a combined steam and chemical explosion in the early morning of 26 April 1986. The explosion was the result of an uncontrolled power increase in the reactor core. The accident can be blamed on dangerous construction properties of the reactor and insufficiently trained personnel. It has led to modifications of remaining Chernobyl-type reactors, of which several are still in operation in the former Soviet Union.


1. Introduction

On the morning of the 28th of April, 1986, an employee of the Forsmark nuclear power station in Sweden set off the radiation monitors in a routine check. His shoes were found to be radioactive, due to dust picked up from the outside. After some initial concern about leaks in the Forsmark facility, it was concluded that the contamination had originated elsewhere.

Eventually, the USSR government admitted that Unit 4 of the Chernobyl power plant in Ukraine near Kiev had exploded, releasing a cloud of radioactive dust into the atmosphere.

The accident at Chernobyl was neither the first nor the last serious accident that has occurred in the nuclear industry, but it is by far the worst in almost every aspect. The Chernobyl accident is a worst-case scenario of a poorly designed reactor. The purpose of this report is to give a brief description of the course of events that led up to the explosive failure of Unit 4 and what has been done to prevent this from ever happening again.


1.1. Introduction to theory of fission reactors

In order to understand why the Chernobyl accident happened, it is necessary to have some knowledge about the operation of nuclear reactors. If you are already versed in the theory of nuclear fission, you may want to skip this section.

A nuclear reactor consists of a core that produces heat and a cooling system that carries the heat away and turns it (usually) into electric power. The coolant is usually water or CO2 gas.


1.1.1. The core

The core has three important parts: the fuel, the moderator and the control rods.


1.1.1.1. Fuel

The heat is generated in the core by a chain reaction of nuclear fission. A few heavy nuclei are fissile, i.e. they may split into two fragments when hit by slow-moving or thermal neutrons. When a nucleus fissions, large amounts of energy are released, as well as new neutrons. These neutrons may go on and induce fission in new nuclei. Thus, a chain reaction can be achieved.

The fuel most commonly used in reactors is uranium. The uranium found in nature is mostly 238U, which is not fissile, but there are also small amounts of the fissile isotope 235U. The fuel usually has to be enriched to contain a few per cent of 235U for a reactor to operate. The fuel is put into fuel rods as uranium dioxide.


1.1.1.2. Moderator

The neutrons emanating from fission have very high velocities. Before they can induce new fissions, they have to be slowed down. This is achieved by letting the neutrons collide with light nuclei. After a few collisions the neutrons will have lost almost all their kinetic energy and are ready to fission new fuel nuclei. The substance containing the light nuclei is called the moderator. The moderators used in practice are water, heavy water and carbon (in the form of graphite.)


1.1.1.3. Control rods

The control rods are the main system for controlling the nuclear reaction in the core. As more neutrons are emitted from the fissions than needed to sustain a chain reaction, some neutrons have to be removed to prevent the core from running amok. The control rods do that by absorbing neutrons. The rods are made of boron carbide, or, in the case of older reactors, cadmium. The operator can slowly push the rods in and out of the core for fine control of the chain reaction.


1.1.2. Reactor operation


1.1.2.1. Neutron flux and reactor power

As mentioned above, the number of neutrons emitted from fission are more than necessary. Some neutrons leave the core, some are absorbed in the moderator or the structural support of the core and some are absorbed in the fuel without inducing fission. Obviously, for the reactor to operate at a steady state, exactly one neutron from every fission needs to induce fission in a new nucleus.

The power output of the core is closely related to the neutron flux. The neutron flux is a measure of the amount of neutrons in the core passing through a unit area per second. An increase in the flux means an increase in the number of fissions per second and thus an increase in reactor power.

Let us define the multiplication factor k as the ratio of the number of neutrons in the present generation to the number of neutrons in the previous generation. If we have steady state, then k=1 and the reactor is critical. If we start pulling out the control rods, k>1 and the power will slowly increase exponentially. The core is then supercritical. If we start pushing the control rods into the core, k<1 and the core is subcritical and the power will decrease exponentially. In case of an emergency the control rods can be pushed into the core quickly, effectively stopping the chain reaction. This is called an emergency shutdown, or SCRAM[1].


1.1.2.2. Prompt criticality

If k exceeds a certain limit (1.0064 for 235U), the power will not increase slowly, but very rapidly. The core has then gone prompt critical and is virtually unstoppable. The security systems are designed to SCRAM the reactor long before this happens.


1.1.2.3. Reactor dynamics

The multiplication factor varies during the operation of the reactor, due to changes in temperature, fuel burnup and poisoning.

Almost all reactors are self-regulating, i.e. when power and temperature increase, k decreases and the power increase is halted. The reactor is then said to have a negative temperature coefficient.

After running the reactor for a long time, the number of fissile nuclei in the fuel decreases and k decreases. This is fuel burnup.

"Poisons" are elements which "steal" neutrons from the fuel, as the control rods do. Some of the fission products, that is, the nuclei resulting from fission of the fuel, are poisons. The most important is xenon 135, which absorbs neutrons to form 136Xe. It has a half-life of 9.21 hours, so it decays when the reactor is turned off, but is quickly built up again when the reactor is turned on. Both the production and destruction of xenon increase with neutron flux.

The operator can compensate for all these effects by moving the control rods.


2. The accident


2.1. The RBMK reactor

The reactors in Chernobyl are of type called RBMK[2], a design almost exclusively used in the former Soviet Union. The RBMK are graphite-moderated boiling water reactors. The reasons for choosing the RBMK were the plutonium production capability and the possibility of building large-scale high-power reactors.

Unit 4 was a 1000 MW electrical power (thermal power was 3200 MW) RBMK installed in 1983.


2.2. The course of events

Because of the speed of the disaster and the total destruction of the reactor it is difficult to tell what really caused the accident in Unit 4, but this is what is generally believed to have happened:


2.2.1. 25 April


2.2.1.1. About 1:00 am

As part of a demonstration of the safety features (in retrospect, this may seem slightly ironic) of the reactor, the operators are beginning to reduce power from a nominal 3200 MW. The test is supposed to show that, in the case of a combined emergency shutdown and power grid failure, the momentum of a single turbine is enough to power the coolant pumps and other vital systems until the emergency diesel generators come on line. There is nothing extraordinary about this test; it is a common test in Western plants as well.


2.2.1.2. 1:06 am

One of the turbines is switched off as a part of the test.


2.2.1.3. 3:47 am

Thermal power reaches 1600 MW.


2.2.1.4. 2:00 pm

The emergency core cooling system is disconnected as a part of the test. The test is now scheduled to continue with a further reduction of power. However, because of a power shortage on the grid, Unit 4 is ordered to keep operating until further notice. The test should be aborted by now. It is not.


2.2.1.5. 11:10 pm

The test is resumed with the intention of reducing power to 700-1000 MW, but the power drops more rapidly than expected, and stops at 30 MW.


2.2.2. 26 April


2.2.2.1. 1:00 am

Power is raised to 200 MW. Running the core at low power has caused a buildup of xenon. This is compensated for by retracting the control rods more than regulations allow. An emergency shutdown will at this point take about 20 seconds. The test is still not aborted.


2.2.2.2. 1:23:04 am

The test is started by reducing the flow of steam to the turbine. As a result, water flow through the core is reduced and boiling is increasing. Reactor power is rising.


2.2.2.3. 1:23:40 am

An emergency shutdown is attempted, by inserting the control rods. It has the opposite effect.


2.2.2.4. 1:23:43 am

Unit 4 goes prompt critical.


2.2.2.5. 1:24 am approx.

A steam explosion followed by a chemical explosion destroys the reactor building and starts a fire.


2.2.2.6. About 5:00

External fires have been put out by firefighters. Although the chain reaction was stopped when the reactor was destroyed there are still considerable amounts of heat generated from radioactive decay and chemical fires.


2.2.2.7. About 6:00

Unit 3 (which resides in the adjacent building) is shut down.


2.2.3. 6 May

Internal fires have been put out. Five thousand tons of boron carbide (inhibiting further chain reactions), limestone, lead, sand and clay has been dumped on what remains of Unit 4.


3. Causes


3.1. Design flaws

The Chernobyl reactors had several dangerous properties that contributed to the accident. The following are deemed the most important.


3.1.1. Positive void coefficient

In section "Reactor dynamics" we stated that in almost all reactors, the multiplication factor is decreased when temperature increases. This also holds for the RBMK reactors, with one exception. When running the reactor at a low power level, increasing boiling of water in the core, which means less water around the core, (steam is virtually transparent to neutrons, so we call it "void," hence "void coefficient") leads to increasing power.

Although light water (as opposed to heavy water) is often used as moderator, it also works as a poison, absorbing neutrons. The coolant in Unit 4 worked as a poison and when boiling increased, k increased, leading to higher power, leading to more boiling ... etc.


3.1.2. Control rod insufficiency

When the operators noticed the sharp increase in power they attempted to insert the control rods into the core. This did not help because:

  • The rods could not move fast enough.

  • The lower part of each rod was made not of boron carbide, but graphite. This was because when the rods were retracted the empty space was filled with water which acts as a poison, thus decreasing the effect of retracting the rods. This was not desirable, so a graphite rod was attached to the bottom of each control rod to keep the water out.

When the operators started to push the control rods back in, the boron carbide parts were completely clear of the core. Below the graphite part was a column of water. Inserting the rods initially had the effect of pushing the water away which meant decreasing the amount of poison which meant increasing k.

The intense heat deformed the core and the control rods stuck before they could be completely inserted.


3.1.3. Lack of containment

Strong concrete buildings surround most Western reactors. Unit 4 didn't have anything like that. If it did, radiation might not have leaked into the environment.


3.2. Operator errors

The accident would not have happened unless the operators had made several serious errors. First of all the test should have been aborted when things were not going as planned. The test was initiated with a number of safety systems turned off. The power level was lower than planned, increasing the importance of the positive void coefficient. The core was suffering from severe xenon poisoning, so the control rods had to be almost fully retracted, leaving too small a margin of safety. The operators seemed completely unaware of the fact that the effect of xenon poisoning would decrease rapidly, should the power level rise.


4. Conclusions

The Chernobyl accident resulted from a combination of external circumstances, engineering design flaws and errors made by badly trained operators. The test was started at extreme operating conditions. Closing the valve to the turbines increased boiling of the coolant. The positive void coefficient started a power excursion which accelerated when the poisoning of the core decreased as the flux increased. This could have been stopped by the control rods, had they not been too far out of the core, as well as badly designed. Instead, the control rods delivered the final blow. The fuel rods went white-hot and shattered. The hot fuel made the water dissociate into hydrogen and oxygen. The cooling system exploded from the pressure of the steam, then the hydrogen could react with the air outside and there was a chemical explosion.


5. Remaining RBMK reactors

In light of the Chernobyl accident, modifications have been made to the other RBMK reactors. These include redesigned control rods, faster control rod mechanisms and changes of fuel enrichment to reduce the effect of positive void coefficient.

All reactors in Chernobyl are now shut down. Around 17 Chernobyl-type reactors are still in operation, the closest of which are two large reactors in Ignalina, Lithuania.


6. Implications for Western reactors

As the RBMK reactors are very much different from anything in the West, we have very little to learn from the Chernobyl accident. An uncontrolled power increase like in Unit 4 is almost impossible in most other reactors. Also, almost all Western reactors have strong containment buildings. The accident at Three Mile Island has shown that even a partial meltdown of the core, although a financial disaster, need not be an environmental one.


Bibliography

[Elements] D J Bennett, J R Thomson, 1989, The Elements of Nuclear Power, .

[SKI] SKI, Swedish Nuclear Power Inspectorate (Statens K√§rnkraftsinspektion).

[IAEA] IAEA, International Atomic Energy Agency.

Notes

[1]Short for Security Cadmium Rod Axe Man. The first reactor, Chicago Pile 1, had a security system consisting of a cadmium rod hanging above the core in a rope. A man would always stand on guard by the rope and in case of emergency the operators would shout at him and he would cut the rope with his axe.
[2]Reaktory Bolshoi Moshchnosti Kanalynye (high-power pressure-tube reactors.) Another acronym often used is LWGR (Light-Water Graphite moderated boiling water Reactor.)