By RICHARD JENSEN.
A LITTLE MORE THAN thirty years ago, just after eleven o’clock, on Friday night, April 26, 1986, the people of Kiev were going to bed. As the city slowed, demand for electricity decreased to the point that the grid operator in Kiev, who was responsible for balancing electrical production with demand, felt that one of the generators at the Chernobyl power plant could be taken offline without affecting service.
This generator, number four, was scheduled for a safety test. Actually, the test was to have taken place hours beforehand, but the grid operator requested a delay to meet evening electrical demand after another power plant in the region went offline.1
Part of the USSR’s ‘Atom Mirny’ or ‘peaceful atom’ program, this RBMK-1000 reactor installed at Chernobyl was three times larger than the average nuclear reactor in the United States2. It had been running at a fraction of its normal output for much of the day3 as part of a plan to test the backup systems that would continue to circulate cooling water around the reactor core in the event that the reactor itself had to be shut down during an emergency. Reactor four had been online since 1982 and still had not passed that safety test. This may be a primary reason why plant engineers pushed ahead with the test despite increasingly unfavorable circumstances. The day shift was expected to carry out the test, but it was not begun until nearly the end of the evening shift, and it would be carried out primarily by night shift workers.4
The test parameters required the reactor output to be about 700 megawatts, less than a quarter of the nameplate capacity of 3200 megawatts. At about midnight, reactor output hit 700 megawatts and continued to drop. For unknown reasons, Leonid Toptunov, the inexperienced reactor operator, inserted the control rods farther into the reactor, which reduced the power output even more.5 A reactor’s control rods are effectively its thermostat. The chain reaction which produces heat in a reactor is caused by neutrons that have been ejected from one atomic nucleus hitting the nucleus of other atoms causing them to split, thus releasing more neutrons. Control rods are made of material that absorbs those free neutrons without splitting and releasing more neutrons, thus keeping the chain reaction that heats the water in the reactor core below a certain level, depending on how many of them are inserted in the core and how far they are inserted.6
Power output from the reactor continued to drop. In fact, it was dropping too fast to be accounted for by the control rods alone. In a matter of minutes, output had fallen to 30MW. Effectively, the reactor had shut down.7
Because this was classified as simply a test of the electrical systems in the plant, as opposed to a test of the overall plant itself, the regulatory agency that oversaw the USSR’s nuclear plants was not informed of the test, even though there was a representative of the agency present at the Chernobyl complex.8
Had they been informed, they might have had staff on scene that would have understood the reactor’s sudden shutdown.
The Soviet government trained nuclear plant personnel on a need-to-know basis and no one working on that reactor on April 26 knew what had just happened.9 However, the engineers and technicians did not stop the test, nor did they call in expert assistance. Possibly, they feared discipline or perhaps even the loss of their jobs for accidentally shutting down the reactor. What they got was far worse.
ONE OF THE ELEMENTS produced in a nuclear reactor is iodine-135. This is an unstable element that quickly decays into xenon-135 (one of the neutrons turns into a proton). Xenon-135 readily absorbs a neutron and turns into xenon-136, a stable element. When a reactor is functioning above a certain threshold, iodine-135 is always turning into xenon-135 and xenon-135 is turning into xenon-136 almost as soon as it becomes xenon-135. Below that threshold, iodine-135 turns into xenon-135 faster than xenon-135 turns into xenon-136. This means that xenon-135 is accumulating in the reactor core.
And since xenon-135 readily absorbs neutrons without splitting and releasing more neutrons, as it accumulates it reduces the amount of energy produced by the reactor. This has the effect of further increasing the amount of xenon-135 in the core in a vicious circle that will quickly shut down a reactor in a process known as ‘reactor poisoning.’ This is what was happening in reactor 4 a few minutes after midnight on the 26th.
The safest way of dealing with reactor poisoning is to let the xenon-135 decay naturally over time. This can take a day or longer.10
Instead, the staff began to remove control rods from the reactor pool.
This is the other way of dealing with reactor poisoning, and it is very, very dangerous.11
Although several errors in the planning and execution of the safety test had already occurred, there had been no irreversible mistakes made. But, by 12:3012 in the morning on April 26th, the fate of reactor 4 had been all but sealed.
The control room staff, who had already disabled the emergency backup cooling system, disconnected the automatic control rod extraction system and manually extracted many of the core’s control rods to their maximum heights.
Extraction of these control rods increased the amount of power produced by the reactor for a while before xenon-135 began to build up again, cooling down the reactor, which caused the reactor staff to remove even more control rods to keep the power output within a 160-200 megawatt window in order for the safety test to be carried out.
During this misguided effort to maintain power, a number of alarms were sounding. The core was oscillating rapidly between bursts of power, when the control rods were removed, and rapid decreases, as xenon-135 built up. These alarms were ignored.
At 1:05 in the morning, the control room crew turned on auxiliary pumps that increased the rate at which water circulated between the core and the turbines and cooling towers. At this point the water was flowing too quickly to be adequately cooled in the cooling towers and was returning to the reactor core dangerously close to the boiling point. At the same time, this higher rate of flow reduced the overall temperature of the water in the core, which resulted in fewer steam bubbles in the water.
This reduction in steam bubbles caused a further loss of power, as water absorbs neutrons more effectively than steam. That loss of power led to the removal of additional control rods from the reactor core.
By 1:23, when the operators began the test, 193 of the reactors’ 211 control rods had been removed. Of these 193 rods, 181 had been manually retracted and locked into place. In their effort to maintain a steady power output, the crew removed ten control rods that were never, under any circumstances, to be removed from the core.13
WHEN THE TEST BEGAN, the steam output to the power turbines was shut off and the backup diesel generators were turned on. The idea was that the turbines would power the pumps for the cooling system while they coasted down, after the steam was cut off, but before the diesel generators had reached full operation.
However, as the turbines coasted down, the pumps slowed down. When the pumps slowed down, the water in the core began to heat up. As it began to heat up, more and more steam bubbles formed. These steam bubbles absorbed fewer neutrons than the water, causing the core to get even hotter, creating even more steam bubbles in a chain reaction.14
The automated control system for the reactor responded to this increase in core temperature by inserting control rods into the reactor pool. However, the system could only reinsert 12 control rods. All of the rest had been locked into place outside of the core.
Thirty-six seconds after the test began, someone in the control room hit the SCRAM button, which inserts every control rod into the reactor at once. On reactors like Chernobyl 4, this process took around 18 seconds, a dangerously long time.15
Further complicating the control rod insertion process were the large graphite rods that were inserted into the control rod tubes when the control rods were removed. Graphite is a nuclear moderator. When a uranium atom splits and releases a neutron, that neutron is moving too fast to readily split another uranium nucleus. Nuclear moderators slow down neutrons, thus making the neutrons more likely to split other uranium nuclei. Crucially, these graphite rods did not extend the full height of the reactor core. Instead, they had a roughly five-foot gap at the top and bottom of the core that was filled with water.
The idea was that the control rods would be more effective if they not only absorbed neutrons, thus reducing the reaction rate, but also displaced a nuclear moderator that enhanced the reaction rate.
In theory, this approach was quite clever. In practice, it resulted in an intricate control rod insertion process that could not be performed quickly. It also created a situation where a SCRAM could cause a sudden and uncontrolled power spike.
ABOUT THREE YEARS EARLIER, during a test of the SCRAM system at the Ignalina power plant in Lithuania, operators had noticed an unexpected power spike immediately before the reactor shut down. Because the reactor shut down normally after that, this power spike was not widely known, nor were reactor operating protocols adjusted to prevent it.16
At Chernobyl, the power generation was occurring mainly near the top and bottom of the reactor.17 As the control rod was reinserted from the top of the reactor, that graphite section which enhances nuclear reactions, is pushed through the bottom of the core. This replacement of neutron absorbing water with neutron moderating graphite caused a rapid power spike at Chernobyl that brought the power output to 530 megawatts only a few seconds after the SCRAM button had been pushed.
That rapid increase in power output apparently caused several fuel rod casings to break, which in turn damaged the control rod channels preventing the full insertion of the control rods into the reactor. The power spike also quickly converted the xenon-135 into xenon-136 thus eliminating the ‘poison’ effect the xenon-135 was having on the reactor, causing the power output to escalate even more quickly.
The reactor, which had been struggling to maintain 200 megawatts of power less than a minute earlier, registered 33,000 megawatts in a runaway chain reaction18 as the fuel rods disintegrated and the cooling water flashed into steam. Pressure from the steam buildup blew the thousand-ton lid off of the core and through the roof of the reactor building less than twenty seconds after the emergency shutdown had been tripped19.
A few seconds later, there was another explosion. To this date, there is no consensus on what caused that second explosion.
Valery Illych Khodemchuk, a pump operator who had just turned 35 was probably killed instantly when the lid of the core landed on and destroyed the main circulation pumps. His body was never recovered.
Under the reactor, Vladimir Nikolaevich Shashenok was found pinned beneath a beam with a broken spine and ribs, and burns over most of his body. Like Valery, Vladimir, a father of two, had just turned 35. Shashenok was not scheduled to work that weekend. He had been called in, apparently to assist with this safety test. Shashenok was alive when he was removed from the power plant, dying in a Kiev hospital a few hours later.
These men were the first two victims of the Chernobyl disaster.
When the reactor core exploded, thousands of pieces of the graphite moderator embedded in the core were scattered around the reactor building and the turbine building immediately next to it. This material was intensely hot, and it started fires throughout the reactor complex. Within minutes of the explosions, plant fire crews were on the scene.
The firefighters who responded received almost nine years’ worth of radiation exposure every second they spent on the roof of the turbine building. Those who worked in or near the reactor received over thirteen years’ worth of radiation in the same time.20 Seven of these firefighters died from acute radiation sickness. An unknown number died from the long-term effects of their exposure. Anatoly Zakharov, a firefighter who survived the night later gave this chilling description of the experience, “we knew we were dying there. A metal taste in my mouth and it felt like someone was touching my body all over from inside, muscles, bones, everything.”21
Surprisingly, reactor 3 remained online throughout this entire ordeal. Personnel at that reactor were simply given respirators and potassium iodide tablets and told to keep working. It wasn’t until 5AM that Yuri Bagdasarov, head of the night shift, shut down the reactor against the wishes of the chief engineer, Nikolai Fomin. Nonessential staff were then sent home.22
IN THE REMAINS of the reactor 4 building, Alexander Akimov was relying on poorly designed radiation dosimeters that couldn’t register radiation at the levels that they were currently exposed to. He believed that the reactor itself was still intact, and that a steam tank had exploded. Throughout the night he sent plant personnel on a vain effort to restore water circulation to the reactor core. Akimov and most of his staff would be dead of acute radiation sickness within three weeks.23
The explosions also released an unknown quantity of radioactive material into the atmosphere. Within hours of the disaster, residents of Pripyat, the nearby town built for the employees of the Chernobyl plant and their families, reported tasting metal, along with other symptoms of radiation sickness including headaches, vomiting and uncontrollable coughing.
On April 27, the Soviet government, realizing the extent of a disaster that had been characterized, in the early hours of the 26th, as nothing more than a fire, began to evacuate Pripyat. A large cloud of radioactive material was working its way northwest over Ukraine, Belarus and Europe, but it had not yet been detected. The evacuation of Pripyat was carried out in near total secrecy.
It wasn’t until Monday, April 28, that the wider world—or even the rest of the Soviet Union—knew that something serious had happened in Chernobyl. That morning, workers at the Forsmark nuclear power plant in Sweden were found with radioactive particles on their clothing. After ruling out a leak at their own facility, the Swedish government rapidly identified the source as fallout from somewhere in the Soviet Union.24
Remediation efforts began almost immediately. Reportedly, Russian military pilots seeded a large cloud of radioactive dust over Belarus that caused a black rain to fall on the town of Gomel25. Robots were brought in to scavenge debris from the wreckage of the reactor building and the turbine hall, but these devices failed almost immediately when the radiation from the debris destroyed the electronics in the robots. In their place, soldiers and volunteers were brought in, given assurances that the state would care for their families, and then put in protective gear and sent on in shifts that gave them a lifetime’s dose of radiation in 40 seconds. Thousands of these ‘bio-robots’ were used to clean up the site.26
Below the floor of the reactor, two large pools of water remained. These pools, one on top of the other, were designed to provide backup cooling water for the circulation pumps, and to absorb heat from minor steam leaks. If the reactor core, which was still over 2,000 degrees Fahrenheit, melted through the concrete into one of these pools, it would likely cause a second steam explosion that would rival the first explosion in magnitude. The only way to drain the pools was to dive into the outlet channel outside the plant and swim up to the sluice gates and open them manually. Alexei Ananenko, an engineer at the plant, knew where the gates were and how to open them. With the assistance of Valeri Bezpalov and Boris Baranov, the sluice gates were opened on May 2, draining the reservoirs and preventing a second steam explosion.
Although Boris had been equipped with a light, it failed shortly after they entered the tunnel, and the three of them had to feel their way along in darkness. All of them were exposed to a lethal dose of radiation during the dive and were already suffering from radiation sickness when they climbed out of the water27.
At the same time, the Soviet government was scrambling to assemble a commission that would determine the causes of the meltdown. The preliminary report, released less than four months after the accident, blamed the injured and the dead. Decisions by plant staff to disable safety protocols and ignore operational guidelines were highlighted and the multiple flaws in the design of the RBMK-style reactor at Chernobyl were minimized—if they were mentioned at all.28
The Soviets made heroes of the firefighters and the “liquidators” and the implicit condemnation of anyone associated with the plant itself carried right on into survivor’s pensions. The widows of plant engineers are given less than the widows of firefighters.29
THE NOTION THAT a reckless crew of plant personnel pushed a well-designed reactor beyond its limits in the early morning hours of April 26th was the standard narrative for years. This failure to candidly discuss the shortcomings of the RBMK design did not sit well with Valery Legasov. Legasov was a deputy director of the Kurchatov Institute of Atomic Energy, which had been assigned the task of reporting on the causes of the Chernobyl meltdown. He was outraged by the way the institute’s report ignored the flaws of the RBMK reactors and ashamed of his own failure to speak candidly when given the opportunity in Vienna where the official Soviet report was presented. Legasov spent two years fighting behind the scenes to have some of the more glaring problems with the reactor addressed. When these efforts failed, in frustration and despair, he hung himself in the stairwell of his office building on April 27, 1988.30 TASS, the Soviet news agency, was careful not to mention the cause of death in his obituary and grossly distorted his view of the accident.31
It wasn’t until 1989, when Grigori Medvedev published The Chernobyl Notebook, that the rest of the world got a glimpse of the dysfunctional bureaucracy and the disregard for safety that permeated the Soviet nuclear energy industry. In an amended report published in 1993, the International Atomic Energy Agency revised its 1986 conclusions, and finally discussed the serious design flaws of the RBMK reactors and the failure of the Soviet government to adequately train plant engineers on reactor physics and to provide clear and unambiguous operating instructions for the reactors.32
Assessing the long term consequences of the Chernobyl accident have proven almost impossible. Not only because the fallout impacted millions of square miles, but also because access to the limited available statistics has been difficult to obtain in areas of the former Soviet Union where the effects are greatest.
The Chernobyl accident, which followed close on the heels of the Three Mile Island accident in 1979 effectively ended nuclear plant construction worldwide. At least one estimate projects a total of 985,000 premature deaths, worldwide from the Chernobyl disaster. By contrast the Global Burden of Disease Study estimated that 1.2 million people died from air pollution in China in 2010.33
Richard Jensen is a writer and historic preservation consultant in Sioux Falls, South Dakota. He has articles published or forthcoming in Aviation History, American History and South Dakota Magazine. He has also written on Walker Evans and Ted Jung.
More Chronicle & Notices.
- Medvedev, Zhores; The Legacy of Chernobyl; W.W. Norton, 1990 paperback edition; pp. 36-38 (Medvedev 1) ↩
- Operational reactor maximum output calculated at 1,060MW based on data compiled by the IAEA and reported here; multiple sources list the Chernobyl reactor capacity at 3,200MW. ↩
- IAEA Report INSAG-7 Chernobyl Accident: Updating of INSAG-1 Safety Series, No.75-INSAG-7; IAEA, 1992; p. 112 (INSAG-7) ↩
- Medvedev 1, pp. 36-38 ↩
- ibid. ↩
- Primarily taken from the US Nuclear Regulatory Commission. ↩
- Chernobyl Accident Appendix 1; World Nuclear Association, 2009. ↩
- INSAG-7; pp.51, 52 ↩
- This can be inferred from the inappropriate steps taken to remedy the situation, and the extreme improbability that a lower-level employee at Chernobyl would know more reactor physics than his superiors, and the lack of involvement by nuclear administrative agency personnel in the test. ↩
- A practice referred to as an “iodine pit”. ↩
- The IAEA stresses the need to closely monitor output and insert control rods, as opposed to the course undertaken at Chernobyl. ↩
- The timeline of the disaster is well documented; see for instance INSAG-7 pp. 53-55 ↩
- Medvedev, Grigori; The Truth About Chernobyl; Basic Books, 1991 ↩
- Shlyakhter, Alexander and Richard Wilson; “Chernobyl: the inevitable results of secrecy”; Public Understanding of Science; Vol 1, No. 3, 1992; p. 252 ↩
- INSAG-7 p. 4 ↩
- Ibid., p. 13 ↩
- Ibid., p. 6 ↩
- The output the reactor reached before it exploded cannot be known with certainty. This was the last reading on the control panel: Kessler, Gunter, et. al; The Risks of Nuclear Energy Technology: Safety Concepts of Light Water Reactors; Springer, 1992; p. 181 ↩
- Reactors are designed to operate with what is called ‘delay criticality’, this means that most neutrons released when a uranium atom splits interact with some other atomic nucleus before they split another uranium atom, with all of this occurring at a steady rate. This delay makes the nuclear reactions slow enough that they can be governed by control rods and other safety mechanisms. When a reactor goes into ‘prompt criticality’, neutrons from split uranium atoms are splitting other uranium atoms almost immediately, but the reaction is still stable (e.g. ‘critical’), that is, power output is not escalating. ‘Supercriticality’ occurs when the reactor’s power output is increasing because the splitting of each individual uranium atom is, on average leading to the splitting of more than one additional uranium atom, thus causing the power to increase at an exponential rate. In a matter of seconds, the Chernobyl reactor went from delay criticality to prompt supercriticality. See DOE Fundamentals Handbook: Nuclear Physics and Reactor Theory Vol. 1 (US Dep’t of Energy, 1992; p. 32), and also DOE Fundamentals Handbook: Nuclear Physics and Reactor Theory Vol. 2 (US Dep’t of Energy, 1993; p. 8) ↩
- Medvedev, Grigori; “Chernobyl Notebook”, translated and published in Soviet Union Economic Affairs, October 23, 1989; p. 24: 20,000 roentgens per hour radiation levels were recorded on the roof; 30,000 near the reactor. That rate was equal to 5.5 and 8.3 roentgens per second respectively; the NRC estimates average exposure in the US is .62 roentgens (expressed here as millirems) ↩
- taken from and interview reported by Fox News. ↩
- Medvedev, Z. 44 ↩
- Medvedev, G. 247, 48 ↩
- Taken from Radio Sweden. ↩
- Taken from The Telegraph. There are many reports of black rain falling in and around Gomel, but the whole episode is poorly documented—or at least the documentation available is not very reliable. ↩
- Taken from the National Geographic. ↩
- Chernobyl: The End of the Nuclear Dream, 1986, p.178 ↩
- Taken from LA Times. ↩
- Taken from The Moscow Times. ↩
- reported by Spiegel Online International ↩
- Taken from the New York Times. ↩
- INSAG-7 pp. 17-25 ↩
- Reported by the New York Times. ↩