The history and science of nuclear energy and the climate problem, part 4: Reactor designs, the brilliant and the stupid

  1. The history and science of nuclear energy and the climate problem, part 1: Why solar and wind energy is insufficient
  2. The history and science of nuclear energy and the climate problem, part 2: Historic context, and the development of nuclear fission energy
  3. The history and science of nuclear energy and the climate problem, part 3: The current state and potential of nuclear energy
  4. The history and science of nuclear energy and the climate problem, part 4: Reactor designs, the brilliant and the stupid
  5. The history and science of nuclear energy and the climate problem, part 5: Stigmas, populism, opportunism, and capitalism

Part 4: Reactor designs, the brilliant and the stupid

In these early days [of nuclear energy] we explored all sorts of power reactors, comparing the advantages and disadvantages of each type. The number of possibilities was enormous, since there are many possibilities for each component of a reactor—fuel, coolant, moderator. The fissile material may be 233U[ranium], 235U[ranium], or 239P[l]u[tonium]; the coolant may be: water, heavy water, gas, or liquid metal; the moderator may be: water, heavy water, beryllium, graphite—or, in a fast- neutron reactor, no moderator. I have calculated that, if one counted all the combinations of fuel, coolant, and moderator, one could identify about a thousand distinct reactors. Thus, at the very beginning of nuclear power, we had to choose which possibilities to pursue, which to ignore.

In this quote from Weinberg’s book The First Nuclear Era: The Life and Times of a Technological Fixer he explains that nuclear energy is not a monolithic thing, it is instead a whole field of research[1] It is of course to be expected that the historic context has had an enormous effect on, as Weinberg says, which possibilities were pursued and which were ignored. Some authors who oppose nuclear energy, treat nuclear physics as if it were exclusively the science of pressurized light water reactors (the most common type of nuclear reactor). They would proceed to describe, in their eyes, the disadvantages of ‘nuclear energy’ when in fact they are describing only pressurized light water reactors. In doing so, these authors prove that they have not done a single bit of research, since a simple scroll through Wikipedia would have revealed that this point of view is wrong and that there are in fact dozens of types of nuclear reactors, each with several subtypes and each with their own distinct advantages and disadvantages[2] Such authors dogmatically oppose anything that even contains the word ‘nuclear’, and will yell buzzwords like ‘nuclear weapons’, ‘radioactive waste’ and ‘Chernobyl’ at you if you dare defend any kind nuclear reactor.

We have already seen that the link between nuclear weapons and nuclear energy is purely historical, and not fundamental, technological or physical. It was the political choices and historic context which has pushed technology towards optimization for the creation of nuclear weapons, and not some fundamental law of physics. Some designs for nuclear reactors have already been described: the (pressurized) light water reactor, the (pressurized) heavy water reactor, and the liquid fluoride thorium reactor. A complete comparative analysis of all types of nuclear reactors would be outside of the scope of this article (the interested reader can find more reading material in the sources list), the take away point is that nuclear energy is not a monolithic thing, and should not be treated as such by authors who wish to engage in a serious scientific discussion. The liquid fluoride thorium reactor was highlighted because it shows that Uranium is not the only possible fuel, and because it is my personal favourite. However, it is not the only modern (generation IV) design, nor is it the only inherently safe design[3]

Figure 10: INES[4]

Since nuclear accidents are always part of the usual critique against nuclear energy, we will analyse the major nuclear accidents in this fourth part, using the knowledge obtained in the first three parts. The International Atomic Energy Agency (IAEA) has defined the International Nuclear Event Scale (INES) to categorize nuclear accidents. It is a scale from 0 (Deviation) to 7 (Major Accident), throughout history there has been one level 6 (Serious Accident) event, and two level 7 events. In my opinion the scale is somewhat vague (and according to Wikipedia I am not the only one that thinks the INES is confusing), but let’s use it here anyway[5]

Kyshtym: haste makes waste

When the USSR started to seriously invest in nuclear weapons, after the bombing of Hiroshima and Nagasaki, they were already behind in this race. The Soviet physicists had almost no experience with nuclear weapons, and were under large pressure to catch up with the Americans. In the period from 1945 to 1948, the Mayak Production Association was constructed with extreme haste and with very little concern for other factors such as safety or environmental impact. The aim was to create as much Plutonium as quickly as possible, all other concerns were secondary.

This light water reactor used the water from lake Kyzyltash directly for cooling, only to dump the contaminated water directly back into the lake. Additionally, the nearby lake Karachay was used as a dumping ground for the waste products. It did not take long for this area to become the most contaminated place on earth.

And as if that wasn’t enough, on the 29th of September 1957 the cooling of one of the storage tanks failed and was not repaired. The temperature steadily rose, eventually resulting in an explosion and a radioactive cloud contaminating areas hundreds of kilometres away. Because the Plutonium production reactor was considered a state secret, the evacuation of 10.000 people only took place one week later and proceeded without any explanation. It would take another 18 years before the details became publicly known.

The IAEA categories this as a level 6 accident. Though it is unclear how many people died as a result of the explosion and the prolonged contamination of the area, estimates are of the order of hundreds. The entire area was declared to be a nature reserve, and with that excuse the Soviet regime forbade anyone from entering[6]

The root cause of the accident can obviously be found in the haste and pressure to produce Plutonium and catch up with the Americans in the arms race. This caused carelessness and the cutting of all the corners that could possibly be cut, which is of course bound to go wrong at some point. They simply did not care about a possible accident, or about the environmental damage they caused, the eyes were only on the prize of winning the arms race.

Chernobyl: a disaster that started on the drawing board

The worst nuclear accident in history is that of Chernobyl, categorized as a level 7 by the IAEA. This disaster shows like no other that some reactor designs are just a very bad idea. The Chernobyl power plant consisted of four high-power channel-type reactors, a design that, first of all, utilizes an exceptional amount of water. Every channel requires about one and a half bathtubs of water per minute, reactor number 4 had 1661 of such channels.

Additionally, the control rods, which are supposed to reduce the amount of neutrons in the reactor by capturing them, were made out of graphite at the bottom and neutron absorbing boron carbide at the top. Graphite is a neutron moderator meaning that it slows neutrons down which increases the chance of the neutron causing a Uranium-235 to fission, therefore the graphite part does the opposite of what the control rods are supposed to do, it increases the reactivity of the core instead of decreases it. The graphite bottom was added to prevent neutron absorbing water from filling the space left vacant when the control rod was in the extracted position. So the idea was that in the down position the boron carbide part of the control rod would be in the core, absorbing the neutrons and therefore slowing down, and eventually stopping, the reaction. And in the up position the graphite part of the control rod would be in the reactor, moderating the neutrons and increasing the reactivity of the core. A good idea in principle, it does however have an unintended side effect: If the control rod is completely extracted (graphite and boron carbide part) then the space left vacant will fill up with water anyway. Which means that when the control rod is inserted back in again, it is the neutron moderating graphite part that will first displace the neutron absorbing water before the neutron absorbing boron carbide part enters the reactor. This means that putting the control rods back in from the completely extracted position will first increase the reactivity of the core before decreasing it, a strange and unexpected property for a control rod to have.

Furthermore, the reactor had a dangerously high positive void coefficient. This coefficient indicates how the reactor responds to bubbles of boiling cooling fluid. A positive void coefficient indicates that the reactivity increases when there is a void (bubble) where the cooling fluid is supposed to be, a negative void coefficient on the other hand indicates that a void decreases the reactivity of the core. So a negative void coefficient is good, because when the reactor becomes too hot and therefore the amount of bubbles increases, the reactivity will automatically decrease. The high-power channel-type reactors had a very positive void coefficient, which means that when the reactor starts to overheat, more bubbles will form, causing more fission reactions to occur, causing more heat, causing more bubbles, etc. In other words, a positive feedback loopback, something you don’t really want in a nuclear reactor[7]

One might wonder, if the design has these critical flaws, why build it anyway? Well first of all, the design was an adaptation of the existing military Plutonium production reactors, so the technique was semi-familiar. It was adapted in such a way that it could perform a hybrid function and produce both Plutonium and electricity. As we have seen before these goals don’t really align well, if you want to create electricity you want continued uninterrupted operation, and you want to burn up as much as possible, including the Plutonium, you therefore keep refuelling to a minimum. On the other hand, if you want to create Plutonium for weapons you’ll need to refuel often to prevent the Plutonium-239 from already fissioning in the reactor, and to prevent too much Plutonium-240 from building up. So how was this contradiction resolved? Well the reactor was designed such that it could be refuelled without turning it off. And that was one of the major selling points of the high-power channel-type reactors, it allowed for individual channels to be replaced while the reactor was operational. Furthermore, the use of the neutron moderator graphite, which absorbs very little neutrons, allowed for the use of very low enriched Uranium as fuel[8]

For this design a guaranteed supply of cooling water was essential. Therefore, Chernobyl had several back-up diesel generators to supply the water pumps with electricity in the case of an emergency situation. However, it would take 60 to 75 seconds for the generators to start up and take over the electricity generation for the water pumps. That does not sound like a lot, but an inadequate cooling of about a minute is more than enough for things to become problematic for such a reactor with a strongly positive void coefficient that is in normal operation already notoriously difficult to handle. To resolve the situation, it was hoped that the main steam turbines could be used to bridge the gap. Since it would take some time for the turbine to spin down when the reactor switched off, it might just be able to produce enough electricity to power the water pumps while the diesel generators were starting up. This of course needed testing, and after three failed tests a fourth test was planned for the 25th of April 1986.

Due to unexpected downtime of a different power plant, the test was postponed by 11 hours, which meant that now the night shift had to perform the test instead of the day shift that was prepared for it. As the test procedure described, the reactor was slowly powered down. However, at some point the reactor power suddenly dropped completely, why this happened is unknown. To restore the power to the level required to complete the test most of the control rods were manually raised. As part of the plan extra water pumps were turned on, which increased the water flow, but also the inlet temperature of the water since the water now spent less time releasing its heat in the cooling towers. Therefore, the safety margin was reduced, and the increased water flow caused the temperature in the reactor to drop which reduced the amount of bubbles and thus decreased the power output of the reactor again. To restore the power level once again, two circulation pumps were turned off, and even more control rods were extracted. The result was an extremely unstable reactor, nearly all of the 211 control rods had been extracted manually, including all but 18 of the fail-safe manually operated rods of the minimum 28 that were supposed to remain fully inserted.

Despite this, the test was continued anyway. The steam supply to the turbines was cut off, causing the turbines to slowly spin down, and the diesel generators started to turn on and take over. As the power supply to the water pumps dropped, so did the water flow, increasing the temperature in the reactor, causing more bubbles, causing the temperature to rise even more. The idea was that 39 seconds into the tests the diesel generators would be able to supply enough of the water pumps with electricity to sustain a sufficient water flow[9]

However, 36 seconds into the test an emergency full shutdown was manually initiated, who did this and why exactly is unclear because those involved died soon after. Such an emergency stop is called a SCRAM in English, which is either derived from the verb, or refers back to that the first nuclear reactor (Chicago Pile-1) actually had an emergency control rod tied to a rope hanging above the reactor and a man with an axe standing next to it to cut the rope in case of an emergency: “Safety Control Rod Axe Man[10] The Russian name (AZ-5) is way less interesting, and translates to “Emergency Protection of the 5th Category”. In any case, a SCRAM immediately causes all control rods to be completely inserted into the reactor automatically, with the intent of completely stopping the nuclear reaction as fast as possible. However as we recall, the control rods of the high-power channel-type reactors have a rather special feature which causes the reactivity of the reactor to first increase before it decreases.

The graphite part of the control rods got inserted into the reactor, replacing the neutron absorbing water, and increasing the reactivity of the reactor. This contra-intuitive feature of the reactor had been observed before in this reactor design, but had never been communicated with the personnel of the Chernobyl power plant (in fact, earlier partial meltdowns and other (nuclear) accidents in Chernobyl unit 1 and the Leningrad power plant, which was of the same design, were treated as state secrets and never communicated with the workers at the plant[11] Combined with the unstable state the reactor already was in, this proved to be the final straw. The reactor overheated, some of the fuel rods fractured, blocking the further insertion of some of the control rods. The reactor output rose to 10 times its nominal output. The steam pressure built up and caused an explosion, more channels ruptured, the coolant supply lines got severed, and the fuel got released into the coolant. A second more powerful explosion occurred two to three seconds later, ejecting most of the reactor core, the exposed hot graphite caught fire when it came into contact with oxygen[12] Engineer Alexander Yuvchenko was in his office when disaster struck, and survived, in a fascinating interview he describes his experience:

To get a clearer idea of what had happened we walked outside. What we saw was terrifying. Everything that could be destroyed had been. The entire water coolant system was gone. The right-hand side of the reactor hall had been completely destroyed, and on the left the pipes were just hanging. That was when I realised that Khodemchuk was definitely dead. The place where I was told he’d been standing was in ruins. The huge turbines were still standing, but everything around them was rubble. He must have been buried under that. From where I stood I could see a huge beam of projected light flooding up into infinity from the reactor. It was like a laser light, caused by the ionisation of the air. It was light-bluish, and it was very beautiful. I watched it for several seconds. If I’d stood there for just a few minutes I would probably have died on the spot because of gamma rays and neutrons and everything else that was spewing out. But Tregub yanked me around the corner to get me out the way. He was older and more experienced.[13]

It took 36 hours before the evacuation of the nearby Pripyat began. Residents were told this was temporary, but soon after, the evacuation of the area became permanent. To this day 2600 km2 of Ukraine is uninhabited due to the large amount of radioactivity in the area[14] .In the first three months after the disaster 31 people died of acute radiation syndrome (=radiation sickness). Furthermore, it is estimated that from the five million inhabitants of the contaminated area, about 4000 died as a result of radiation induced cancer[15] (The exact number is heavily debated, some estimate it to be lower, some estimate it to be way higher. In this article we will stick to the IAEA number of 4000, which was reaffirmed in 2008, based on 20 years of research and on the actual increase of cancer cases. We will not indulge in any extravagantly ridiculous numbers pushed by opponents of nuclear energy based on models which are not supported by peer-reviewed science, models which assume that even the tiniest amount of radiation will increase the risk of developing cancer.)

Fukushima: how a natural disaster became a nuclear disaster

The nuclear reactors at Fukushima were of the boiling light water type, meaning that in this reactor the water boils in the reactor and the resulting steam drives a turbine. In contrast the water in a pressurized light water reactor does not boil, instead the water stays in liquid form, and exchanges its heat with a secondary water circuit which is not pressurized and does boil, the resulting steam of this secondary circuit drives a turbine[16] Compared to Chernobyl’s high-power channel-type reactors, the reactors at Fukushima did not use graphite, instead the light water was used both as the coolant and as the neutron moderator[17]

On the 11th of March 2011 a massive earthquake occurred near Japan’s largest island. The resulting forces applied to the Fukushima reactor were well above the design limits for continued operation, and therefore reactors 1,2 and 3 switched off automatically, reactors 4,5, and 6 had already been powered down for routine inspection and refuelling. As a result of the earthquake one of the two connections to the electricity grid of reactors 1,2 and 3 broke, therefore 13 emergency diesel generators kicked in to supply the cooling of the reactors with enough power.

So far nothing serious had happened. However, about 50 minutes after the earthquake a huge tsunami with waves of up to 14 meters splashed over the 10 meter high seawall (the 35 meter high natural seawall had been removed to ease construction). This caused a flooding of the basement, which for some reason was where the reactors’ emergency diesel generators were (engineers had actually pointed out way earlier that the location of the generators left them vulnerable to flooding). The switching station that provided a connection with three different backup generators located higher up the hill also flooded and was no longer functional. The cooling power was now entirely dependent on a set of batteries, which quickly ran out. Additional portable generators and batteries came late due to poor road conditions as a result of the earthquake and the tsunami, and when they did arrive the flooding made it impossible to connect them to the cooling pumps. Eventually the cooling pumps stopped and reactors 1,2 and 3 began to overheat (even when such a reactor is off it still generates some heat due to radioactive decay occurring in the fuel rods). This caused a meltdown in reactors 1,2 and 3.

Furthermore, the high temperatures in the reactors caused the Zirconium coating of the fuel rods to react with steam to form Hydrogen. The concentration of Hydrogen kept increasing and eventually caused an explosion in reactor 1 on the 12th of March, a similar explosion occurred in reactor 3 on the 14th and in reactor 4 on the 15th because reactor 4 happened to be connected to reactor 3 via a ventilation shaft. Reactor 2 did not explode.

Despite being a level 7 event on the INES scale nobody actually died of radiation sickness directly, though many people were injured by the explosions or received serious radiation burns. Moreover, tens of people did die as a result of the evacuation, mostly the elderly and patients from local hospitals. Additionally, large amounts of radioactive material were released into the environment, and in 2015 the tap water in Tokyo was still significantly more radioactive compared to other Japanese cities. TEPCO, the company owning the power plant, had been warned several times both internally and externally about the risk of flooding due to a tsunami. On the 12th of October 2012 TEPCO admitted that they had been negligent, and that they had not taken the necessary precautions out of fear that doing so would trigger lawsuits and protests. They also acknowledged that they could have reduced the impact of the disaster, or even prevented it entirely, if they had adhered to international norms and recommendations. Several other similar nuclear power plants were able to achieve a cold shutdown, despite that some of those had also been flooded by the tsunami, the Onagawa Nuclear Power Plant was closer to the epicentre of the earthquake but survived just fine thanks to its 14 meter high sea wall[18]

In the same interview we saw earlier, when asked about what he thinks of nuclear power, Yuvchenko hits the nail on the head when he answers: “I’m fine about it, as long as safety is put head and shoulders above any other concern, financial or whatever. If you keep safety as your number one priority at all stages of planning and running a plant, it should be OK.[19] We see time and time again that when things go wrong we can trace the cause to that safety was not the primary concern. In Kyshtym it was the thirst for Plutonium that led to safety and environmental concerns to be completely cast aside. In Chernobyl the desire to create both Plutonium and electricity led to an unstable and unsafe hybrid design, whose oddities and flaws were never communicated with the workers at the plant because of a hierarchical and bureaucratic culture of secrecy. In Fukushima it was financial and reputational concerns that led to that inadequate precautionary measures were taken.

We have already seen that the history of nuclear physics has been riddled with ‘other concerns’ besides safety and the prosperity of humankind, the consequences of which have given the entire field a bad reputation. We have also seen that there are many different reactor designs, some safer than others, some even inherently safe. The conclusion is that there is no reason at all to declare that all of nuclear energy is ‘inherently dangerous and unsafe’, as some opponents of nuclear energy tend to do. To end this section as we began, with a quote from Weinberg:

We nuclear people have made a Faustian bargain with society. On the one hand we offer—in the catalytic nuclear burner (i.e., the breeder [reactor])—an inexhaustible source of energy. Even in the short range, when we use ordinary reactors, we offer energy that is cheaper than energy from fossil fuel. Moreover, this source of energy when properly handled is almost nonpolluting. Whereas fossil-fuel burners emit oxides of carbon, nitrogen, and sulfur… there is no intrinsic reason why nuclear systems must emit any pollutant except heat and traces of radioactivity. But the price that we demand of society for this magical source is both a vigilance from and longevity of our social institutions that we are quite unaccustomed to.[20]

But…. what about the waste?

There will always be a bit of waste, because every fission of a fissile element results in waste products which usually are unstable and will therefore be radioactive to some extent before they eventually decay to a stable isotope. How much waste depends on the design of the reactor, we have already seen that the liquid fluoride thorium reactor produces less waste compared to the more conventional light water reactors, and the radioactivity of the waste that it produces goes down to background levels way faster.

Spent nuclear fuel will contain the products of the fission reaction, and some actinides consisting of the remaining fuel and some heavier actinides that were formed due to neutron capture of the fuel elements. The exact composition of the spent fuel depends on the original fuel, but also on the amount and speed of the neutrons in the reactor during operation. It is mostly those heavier actinides which are problematic because of their long life times and high amounts of radioactivity[21] Luckily, those heavier actinides can in principle be ‘burned up’ by forcing them to fission, creating lighter more manageable isotopes.

The reprocessing of nuclear fuel separates the light fission products from the remaining fuel and other heavy elements, the reprocessed fuel can then be used as fuel once more. This recycling increases the amount of energy that can be extracted from the fuel by as much as 60 times, and also has the benefit of removing the problematic heavy actinides from the waste that eventually needs storing[22] Using this method 96% of spent fuel can be recycled, the remaining 4% are the fission products which should be stored until their radioactivity decreases to background levels[23]

How long this waste should be stored depends on how well it was reprocessed and on the original composition of the spent fuel before reprocessing (and thus on the used fuel, and on the type of reactor). The easiest way to store this waste is to simply bury it in a dry earthquake-free place. In 2015 Finland has started the construction of such a deep geological repository, the aim is that it will be open for about 100 years during which it can be filled with nuclear waste. After this period the tunnel will be sealed, providing a long-term storage for the waste 520 meters under the ground[24] Problem solved!

Luckily the energy density of nuclear energy is extremely high, and therefore the amount of waste is very low. A typical 1000 megawatt nuclear power plant produces merely 27 tonnes of unprocessed spent fuel in one year[25] That is the equivalent of 20 m3, and can be reduced to as little as 3 m3 by reprocessing. That is a miniscule amount for a whole gigawatt over a whole year.

To not saddle future generations with nuclear wasteis an argument one often hears against nuclear energy. In the above we have seen that this argument is complete rubbish, first of all because the amount of radioactive waste is not a lot, and secondly because good long-term solutions exist. Contrary to carbon dioxide and other gaseous waste, which is a problem for which there still is no good solution at all.