🧑🎓
What exactly is the "keff" value that the simulator calculates? I see it changes when I move the sliders.
🎓
Basically, keff is the "effective multiplication factor." It's the most important number in reactor physics. If keff = 1, each fission generation produces exactly enough neutrons to trigger the next one, so the reaction is stable—this is called "criticality." In the simulator, when keff hits 1, you'll see the neutron population stabilize.
🧑🎓
Wait, really? So what do the control rods and the moderator type actually do to change that number?
🎓
Great question! They affect different parts of the neutron lifecycle. For instance, control rods (like boron) are strong neutron absorbers. When you increase the "Control Rod Absorption" slider, you're removing neutrons from the system, which directly lowers keff. The moderator, like water or graphite, slows down fast neutrons so they're more likely to cause fission. Try switching from "None" to "Heavy Water" and watch keff increase!
🧑🎓
So the "Uranium Enrichment" must be about the fuel itself. How does a higher percentage of U-235 change the reaction?
🎓
Exactly. Natural uranium is mostly U-238, which doesn't fission easily with slow neutrons. Enrichment increases the concentration of the fissile U-235. In practice, a higher enrichment means each fission event releases more neutrons on average, boosting the $\eta$ factor in our equation. Slide the enrichment up to 5% (typical for power reactors) and you'll see it's much easier to reach keff = 1 than with natural uranium at 0.7%.
The core physics of a sustained chain reaction is captured by the Six-Factor Formula, which accounts for all the ways neutrons are produced, lost, or utilized in a reactor core.
$$k_{eff}= \eta \cdot f \cdot p \cdot \varepsilon \cdot P_{NL}$$
$\eta$ (neutrons per fission): Average number of neutrons released per fission event. Primarily depends on fuel enrichment.
$f$ (thermal utilization factor): Fraction of thermal neutrons absorbed by the fuel (not by control rods or moderator).
$p$ (resonance escape probability): Chance a neutron slows down (moderates) without being captured by U-238 at intermediate energies.
$\varepsilon$ (fast fission factor): Small increase in neutrons from fissions caused by fast neutrons before they slow down.
$P_{NL}$ (non-leakage probability): Chance that neutrons do not escape the reactor core entirely. Our simulator assumes an infinite core, so $P_{NL}= 1$.
The simulator uses this formula to calculate keff based on your settings. The resulting keff then determines the neutron population growth or decay over time, following a simple exponential model.
$$N(t) = N_0 \cdot (k_{eff})^n$$
$N(t)$: Neutron population at time $t$.
$N_0$: Initial neutron count you set.
$n$: Number of fission generations over time $t$.
When $k_{eff}> 1$, $N(t)$ grows exponentially (supercritical). When $k_{eff}< 1$, it decays (subcritical). At $k_{eff} = 1$, it remains constant (critical).
Common Misunderstandings and Points to Note
First, you might wonder why neutrons don't immediately drop to zero even when you set the "Control Rod Absorption Rate" to its maximum in this simulator. This is because control rods absorb neutrons "probabilistically." For example, even with a 90% absorption rate, a neutron might, with some luck, slip through and contribute to the next generation. In a real reactor as well, even when control rods are inserted fully, neutrons decay exponentially and do not stop instantaneously.
Next, you might think that increasing the "Uranium Enrichment" will always lead to criticality, but this is not the case if the moderator or core size is inappropriate. For instance, even if you set enrichment to over 90% (weapons-grade), if you set the moderator to "None," fast neutrons are less likely to sustain a fission chain, and the system may remain subcritical. Conversely, a natural uranium reactor (0.7% enrichment) using "Light Water" as a moderator is theoretically not feasible. This is because light water absorbs too many neutrons; actual "Light Water Reactors" only become viable when enrichment is increased to 3–5%.
Finally, note that the "Neutrons per Generation" graph in the simulator shows a smooth curve. This is the result of averaging a stochastic process. In a real reactor, especially when neutron counts are low (e.g., during startup), the birth and death of neutrons involve probabilistic "fluctuations" and can sometimes show unexpected variations. For more detailed analysis using CAE, methods like the "Monte Carlo method," which tracks this stochastic behavior, become necessary.
Related Engineering Fields
The core concept of this tool, the "effective multiplication factor (keff)," is actually applied in various engineering fields beyond reactor physics. One is radiation shielding design. For example, when determining the thickness of concrete walls in nuclear facilities, calculations are made for how much neutrons or gamma rays attenuate as they pass through material. This can also be viewed as a type of "non-leakage probability" calculation, sharing the same physics as considering neutron leakage outside the core in the simulator.
Another is amplification phenomena in lasers and semiconductors. The process where photons undergo stimulated emission and are amplified within a laser medium is mathematically very similar to the process where neutrons induce fission and multiply. In both cases, whether the amplification factor (gain/keff) exceeds 1 as a particle (photon/neutron) passes through the medium determines the condition for steady-state oscillation (criticality).
Furthermore, epidemic models for infectious diseases share the same structure. The basic reproduction number R0 is the average number of secondary infections produced by one infected individual, which corresponds precisely to keff. An epidemic occurs (supercritical) when R0 > 1 and subsides (subcritical) when R0 < 1. In terms of adjusting parameters to control system behavior, reactor control and public health policy share a common thought framework.
For Further Learning
Once you are comfortable with this simulator, the next step is to be able to estimate for yourself, using mathematical formulas, why changing a parameter affects keff in a certain way. To do this, learn how each factor in the six-factor formula is specifically calculated. For example, the thermal utilization factor $f$ can be calculated from "microscopic data" such as the fission cross-section of the fuel and the absorption cross-section of the moderator. In formula terms, it looks like $f = \frac{\Sigma_a^{fuel}}{\Sigma_a^{fuel} + \Sigma_a^{moderator} + \Sigma_a^{other}}$, where $\Sigma_a$ is the macroscopic absorption cross-section.
As a learning sequence, we recommend first learning the basics of the "neutron diffusion equation". This equation treats neutron flow as a continuum and is the most fundamental method for calculating neutron distribution within a reactor core and keff. A simplified form of diffusion theory is likely used behind this simulator as well. In textbooks, it's common to start with the one-group diffusion equation and progress to multi-group diffusion theory which considers energy dependence.
Ultimately, it's good to be aware of industry-standard CAE simulation codes (like SRAC, MVP, MCNP, etc.). Unlike this educational tool which uses a single formula, these codes perform high-precision keff calculations directly relevant to real reactor design, taking into account complex 3D geometries and continuous energy distributions. The experience of gaining an intuitive feel for parameter sensitivity with this tool will serve as a powerful foundation for understanding the meaning of input parameters in those professional codes.