To understand what happened at Chernobyl, we first have to understand how the RBMK (Soviet reactor) was designed - it’s a unique design not found in the US.

And to understand the RBMK design, we first have to understand how nuclear energy works in general - so we’ll go over a local Pressurized Water Reactor (Shearon Harris) as a good baseline.

Nuclear reactors get their energy from the nuclear fission process.

Nuclear fission occurs when a neutron hits a Uranium-235 atom, which splits into multiple smaller atoms and releases 2-3 neutrons on average. These neutrons have the potential to hit more Uranium-235 atoms and sustain a chain reaction.

The smaller atoms left over from the fission process are called daughter isotopes and are naturally unstable. They naturally decay (split up) into even smaller atoms while releasing a few neutrons of their own. Since it takes these neutrons a bit longer to appear after the initial fission reaction, they are referred to as delayed neutrons.

The fact that the fission reaction produces energy just means that the product particles travel at very high speeds (i.e., they carry a large amount of energy).

Natural Uranium ore (which is over 99% Uranium-238 and only 0.7% Uranium-235) is enriched to ~5% Uranium-235 and shaped into ceramic pellets. Each pellet contains the same amount of energy as roughly 1700 pounds of coal.

These pellets are placed into metal tubes, or fuel rods. The cladding (metal portion) of the fuel rods is composed of a Zirconium alloy, which does an excellent job of conducting heat (energy) while also letting neutrons pass through (i.e., the neutrons are more likely to hit Uranium-235 atoms instead of being absorbed or blocked by the cladding itself).

The fuel rods are bundled into fuel assemblies.

The fuel assemblies form the reactor core. Shearon Harris’ core is composed of 157 fuel assemblies.

Water is pumped from the bottom of the core, through the assemblies (around the fuel rods), and out the top.

Recall that the fission process produces a lot of energy (i.e., very fast moving particles) within the core. Most of these particles are deposited within the fuel pellets themselves (by way neutrons and fission products bumping into things), which raises the temperature of the fuel to ~1000 F. The fuel rods act as heaters for the water as it flows through the core. By the time the water exits the top of the core, it has been heated to over 600 F.

The water doesn’t boil, since it’s pressurized to ~2200 psia. Instead, it flows through a steam generator, where it transfers its heat to a different loop of cooler water as the latter flows through thousands of small tubes (which maximize the surface area for heat transfer). It is then pumped back into the bottom of the core, where it repeats the cycle. This loop of water that flows through the core is known as the primary loop.

The “different loop” of water that receives heat from the primary loop within the steam generator is called the secondary loop. Cold water in the secondary loop flows around the hot tubes and boils into steam. Steam expands and takes up more space than liquid water, so the pressure inside the steam generator increases as more steam is created.

This pressurized steam is directed through a turbine, which spins a big magnet and generates electricity - essentially a "reverse motor," which uses electricity to spin a shaft.

Most of this electricity is directed to the grid, but some of this electricity is diverted back into the plant to power components like pumps and chillers. In this way, the plant actually powers itself.

Once the steam has passed through the turbine, it has lost its pressure, as its energy has been removed. It is directed to the condenser which uses water from the lake - again, flowing through bundles of tubes - to condense the secondary loop steam back into liquid form so that it can be pumped back into the steam generator and repeat the cycle.

The lake water pumped through the condenser tubes draws heat from the secondary loop, and is directed to the lower-mid section of the cooling tower, where it is sprayed out via thousands of nozzles. Some of this water is cooled by the air and drains down into the cooling tower basin where it is pumped back into the condenser to complete the cycle. The rest of the water evaporates into steam and flows out the top.

In addition to the recirculated flow from the condenser, the cooling tower basin also gets its water from the Ultimate Heat Sink, which is Harris Lake for the Shearon Harris Nuclear Plant.

To recap - the reactor uses the fission process to generate heat, which is transferred into primary loop water. Heat from the primary loop is used to boil secondary loop water in the steam generator to create steam. This high-pressure steam is directed through the turbine to generate electricity, and is ultimately cooled and condensed by transferring its heat to water drawn from the lake.

Back to reactor physics - what is criticality?

We want the reactor to be critical with zero reactivity at steady-state, full power operation.

It’s a balancing act - recall that each fission reaction produces 2-3 neutrons on average. If we produce too many neutrons, we become supercritical. If we absorb too many neutrons, we become subcritical. Interestingly enough, delayed neutrons play a big role in controlling the reactor since they give us time to adjust reactivity as needed before continuing on with the chain reaction.

Control rods are made of materials that absorb neutrons, and can be inserted into the reactor to reduce the number of neutrons flying around - thereby reducing the number of fissions in the core.

Boron is also a great neutron absorber. The fission process can be enabled or hampered as needed by changing the concentration of dissolved boron within the primary loop.

So we want the reactor to be critical, but each fission reaction produces 2-3 neutrons, each of which has the potential to cause another fission reaction - on the surface, it seems like the main struggle in controlling the reactor would be in limiting the number of neutrons that end up inducing fission. But in reality, the opposite is true!

Neutrons are produced with an energy on the order of 10e6 eV (electron volts) - this means that they’re moving very, very fast. But due to quantum mechanics, the fission reaction is actually more likely to happen for slower neutrons - so these high-energy “fast” neutrons actually don’t end up doing anything for us unless we slow them down, or moderate them.

How do we slow neutrons down? With moderators, or particles that can “steal” the energy of a neutron away if the neutron hits it. Water, which is composed of one oxygen atom and two hydrogen atoms, is a great candidate - hydrogen atoms have roughly the same mass as neutrons, so they do a great job of slowing them down a la Newton’s Cradle.

Conveniently for us, water is already being pumped through the core to transfer heat to the secondary loop! In US reactors, water is both the moderator and the coolant. It slows the neutrons down (facilitating the nuclear reaction) and transfers the energy (heat) from the core and into the steam generator.

What does this mean for reactivity if the water heats up and expands? Less water in the same amount of space -> less moderation -> less fissions -> less power! This is what’s called a negative moderator coefficient. 

The reactor at Chernobyl was an RBMK style reactor, which is a very different reactor design to the US ones. It was designed to be as cheap as possible:

• Single-loop design - water does boil in the core, and the steam produced in this process is used to turn the turbine.

  • A liquid-vapor mixture leaves the top of the core, and the vapor (steam) is separated out and directed to the turbine via steam drums

  • The remaining liquid collects at the bottom of the drum, where it mixes with the condensate (the condensed steam, after it has passed through the turbine) and is pumped back into the bottom of the core

• Low enrichment/bigger core - the uranium in the core is only enriched to ~2% Uranium-235, which means that the reactor had to be much larger than US designs

  • In larger reactors, neutrons are more likely to hit uranium atoms and cause more fissions before they fly out of the core

• Graphite moderator - enables the neutrons to be used more efficiently, considering the lower-enriched core

  • Graphite does a good job of moderating neutrons without absorbing them

  • Water does a good job of moderating neutrons, but absorbs some as well

  • This means that when a graphite moderator is employed and the neutrons are already being moderated by it, the main effect of water is that of a neutron absorber

    • What does this mean for reactivity now if the water heats up and expands? Less water in the same amount of space -> less absorption -> more fissions -> more power! This effect is even more extreme when the water boils and steam bubbles - or voids - are produced. This is what’s referred to as a positive void coefficient.

The Soviets wanted to maximize power output while still being able to control the reactor.

When we insert our control rods all the way in, our intent is to reduce the number of fissions in the core (i.e., reduce core power).

When we withdraw our control rods, our intent is to increase the number of fissions in the core (i.e., increase core power).

The Soviets realized that they could get a further increase in power output by attaching graphite followers to the ends of their control rods, which were positioned in the middle of the core when the control rods were fully withdrawn.

Since the RBMK reactor core is so big, the control rods need to be manipulated to control local reactivity in different regions of the core (spatial control), as well as overall reactor power (power control).

The RBMK had procedural limits on how many rods could were allowed to be fully withdrawn at all times - control rod positions needed to be configured in such a way that a certain amount of immediate negative reactivity would be guaranteed in the event of a SCRAM, which is when all control rods are quickly inserted into the core. Positioning the control rods outside of those limits could result in a delay of the negative reactivity injection, since it would take longer for the absorbiest parts of the rods to reach the regions of the core with the highest amounts of neutrons flying around.