Theory: How Neutrons Interact with Matter
Why Neutrons Matter in a Reactor
A nuclear reactor runs on neutrons. Each fission releases two to three fast neutrons with energies near 2 MeV, and the chain reaction continues only if enough of them cause new fissions. Because neutrons have no charge, they pass through electron clouds untouched and interact only with nuclei. Whether a neutron scatters, is captured, or causes fission depends on the nucleus it meets and on its energy. Reactor engineers describe this with cross sections, and every design decision, from the choice of moderator to the material of the control rods, starts with them.
Types of Neutron Interaction
| Interaction | What happens | Reactor importance |
|---|---|---|
| Elastic scattering (n,n) | The neutron bounces off; kinetic energy is shared, the nucleus is not excited | Main slowing-down process in moderators |
| Inelastic scattering (n,n′) | The nucleus is left excited and emits a gamma ray | Slows fast neutrons in heavy materials such as U-238 and iron |
| Radiative capture (n,γ) | The neutron is absorbed and a gamma ray is emitted | Loss of neutrons; produces Pu-239 from U-238 and activates structures |
| Particle emission (n,α), (n,p) | A charged particle is emitted | B-10(n,α) in control materials and BF₃ detectors |
| Fission (n,f) | The nucleus splits, releasing about 200 MeV and 2 to 3 neutrons | The source of reactor power |
Microscopic and Macroscopic Cross Sections
The microscopic cross section σ is the effective target area one nucleus presents to a neutron for a given reaction, measured in barns (1 b = 10⁻²⁴ cm²). It is a probability, not a physical size: cadmium presents about 2500 b to a thermal neutron, while carbon presents only 0.0035 b for absorption. The macroscopic cross section Σ combines σ with the number of nuclei per cubic centimetre, N, and is the probability of interaction per centimetre of travel. Its inverse is the mean free path, the average distance a neutron travels between interactions. For a compound, N counts molecules and σ is the sum over the atoms in one molecule.
Σ = N σ (cm⁻¹, with σ in cm²) Σ_t = Σ_s + Σ_a
λ = 1 / Σ_t (mean free path, cm)
Beam Attenuation and Reaction Rates
In a narrow, collimated beam, any interaction removes a neutron from the beam, so the uncollided intensity falls exponentially with the total cross section. In a reactor the neutrons move in all directions, and we describe them by the neutron flux φ = nv (neutrons per cm² per second). The rate of any reaction per cubic centimetre is the product of the flux and the macroscopic cross section for that reaction.
R = Σ φ (reactions per cm³ per second)
Moderation: Slowing Neutrons Down
Fission in U-235 is far more likely for slow (thermal) neutrons near 0.025 eV than for fast ones, so thermal reactors use a moderator to slow neutrons through elastic collisions. A neutron loses the most energy when it hits a nucleus of similar mass: in a head-on collision with hydrogen it can lose all of its energy, but against carbon it keeps at least 72%. The average logarithmic energy loss per collision, ξ, captures this. A good moderator must also scatter often (large Σ_s) and absorb little (small Σ_a); the moderating ratio combines all three properties.
Number of collisions: n = ln(E₀ / E) / ξ
Slowing-down power = ξ Σ_s Moderating ratio = ξ Σ_s / Σ_a
Reading Cross-Section and Slowing-Down Graphs
Engineers read nuclear data from graphs. For a 1/v absorber, σ(E) = σ₀ √(0.0253/E), so on log–log axes the cross section is a straight line of slope −0.5. Boron-10 and lithium-6 follow this law, which makes them ideal for thermal-neutron detectors and control materials. Heavy nuclides such as U-238 also show sharp resonances, where the capture cross section jumps by several orders of magnitude at particular energies (the first U-238 resonance is at 6.67 eV). On a slowing-down graph with a logarithmic energy axis, the average energy after n collisions is E₀e^(−nξ), a straight staircase whose steepness is ξ.
Thermal-Neutron Data Used in This Lab (0.0253 eV)
| Material | ρ (g/cm³) | M (g/mol) | σ_s (b) | σ_a (b) | ξ |
|---|---|---|---|---|---|
| Light water H₂O (per molecule) | 1.00 | 18.015 | 44.8 | 0.664 | 0.927 |
| Heavy water D₂O (per molecule) | 1.105 | 20.03 | 10.54 | 0.0012 | 0.510 |
| Graphite C | 1.70 | 12.011 | 4.75 | 0.0035 | 0.158 |
| Iron Fe | 7.874 | 55.845 | 11.6 | 2.56 | 0.0353 |
| Cadmium Cd | 8.65 | 112.41 | 5.6 | 2520 | 0.0177 |
Values are rounded thermal-neutron data of the kind tabulated in the DOE Fundamentals Handbook (Nuclear Physics and Reactor Theory) and on the Chart of the Nuclides. Molecular values for H₂O and D₂O include both hydrogen atoms and the oxygen atom; for single-atom moderators, ξ for hydrogen is 1.000, deuterium 0.725, beryllium 0.209, and carbon 0.158.
Shielding and control
Cadmium and boron have huge absorption cross sections, which is why they are used in control rods, shutdown systems, and neutron shields.
Moderator choice
Light water slows neutrons quickly but absorbs some; heavy water and graphite absorb so little that they can sustain a reactor on natural uranium.
Industry
Neutron moisture gauges, oil-well logging tools, and neutron radiography all rely on hydrogen slowing neutrons and on materials absorbing them.
Apparatus
The equipment used on a neutron beam line and in moderator studies. In the simulation these are modelled for you, and the readings correspond to what each instrument would show.
Instructions: Running the Virtual Experiment
You are qualifying materials and moderators for a research reactor. The simulation has two immersive 3D work areas and two graph work areas. The parts build from undergraduate work in Parts A to C to graduate-level analysis in the last step of Parts D and E; the level of each part is shown beside its title. In each one you set up the experiment, take readings, work out the answer yourself, enter it, and press Check. Record every reading and prediction in your lab report tables, with a labelled screenshot of each setup. Drag in any 3D view to orbit, and scroll to zoom. Use the data table in the Theory section.
Simulation: Neutron Beam Line, Moderator Tank, and Neutron Graphs
A collimated beam of thermal (0.025 eV) neutrons leaves the shielded source, passes through the sample slab, and reaches the BF₃ detector. Neutrons that interact in the slab are removed from the beam: cyan tracks reach the detector, and the ones that stop in the slab were scattered or absorbed. Parts A and B use this bench.
Drag to orbit · scroll to zoom. 3D beam line: a shielded neutron source and collimator, a sample slab on a stand, and a BF₃ detector with its counter; neutrons that pass through are shown in cyan, absorbed ones stop in the slab.
Setup
Readings
Detector background is about 2 counts per second and is already subtracted from the counts shown.
Work it out, then check
| Material | x (cm) | N (per cm³) | Σ_t (cm⁻¹) | λ (cm) | Predicted I/I₀ | Measured I/I₀ |
|---|
A 2 MeV fission neutron enters the moderator tank and slows down through elastic collisions. Its colour changes from red (fast) through orange and green to blue (thermal) as it loses energy, and the counter records each collision. Part C uses this tank.
Drag to orbit · scroll to zoom. 3D moderator tank: a fast neutron enters the tank and zig-zags through collisions with moderator nuclei, changing colour from red (fast) to blue (thermal); the collision counter and energy readout update as it slows.
Setup
Readings
For the bulk moderators, Σ_s and Σ_a are calculated from the data table in the Theory section. The ideal hydrogen and deuterium nuclei are for collision counting only.
Work it out, then check
| Moderator | ξ | E₀ → E | n (collisions) | ξΣ_s (cm⁻¹) | ξΣ_s/Σ_a |
|---|
Most absorbers used in detectors and control rods follow the 1/v law at low energy: the slower the neutron, the longer it spends near a nucleus and the larger the cross section. On log–log axes this law is a straight line. Part D uses this work area.
Absorption cross section against neutron energy on log–log axes: the straight 1/v line of the absorber you chose, with the U-238 capture curve and its 6.67 eV resonance for comparison. The dot marks the energy you set.
Show the plotted values
Setup
Readings
σ(E) = σ₀ √(0.0253/E) and v = 2200 √(E/0.0253) m/s. The U-238 curve is a simplified training shape around its first resonance.
Work it out, then check
| Absorber | E (eV) | Speed (m/s) | σ (b) |
|---|
Each elastic collision removes, on average, the same fraction of a neutron's logarithmic energy, so on a logarithmic energy axis the average energy falls in equal steps of ξ. The steeper the ladder, the better the moderator slows neutrons. Part E uses this work area.
Average neutron energy against collision number (logarithmic energy axis) for hydrogen, deuterium, beryllium, and carbon, with the target energy shown as a dashed line. The dot marks your chosen moderator after k collisions.
Show the plotted values
Setup
Readings
Average energy after n collisions: E_n = E₀ e^(−nξ). Collisions to reach E: n = ln(E₀/E)/ξ.
Work it out, then check
| Moderator | ξ | k | E after k (eV) | Collisions to target | Relative to hydrogen |
|---|
These immersive 3D work areas and interactive graph work areas are the main version; the accessible copy delivers the same set-up, work-out, and check exercises with native controls and screen-reader announcements.
Team Questions
Example Lab Report
Sample report showing the expected format and level of detail. Use it as a guide for your own submission, and include labelled screenshots of each setup. Measured counts will differ slightly from these because neutron emission is random.
Neutron Interactions, Cross Sections, and Moderation
Nuclear Engineering Technology | Section: [Your Section] | Date: [Date]
Lab Members: [Names of all members present]
Purpose
To calculate atom densities, macroscopic cross sections, and mean free paths for reactor materials; to verify exponential attenuation of a thermal-neutron beam; to calculate absorption reaction rates; to compare moderators by the number of collisions needed for thermalization and by their moderating ratio; and to read cross-section and slowing-down graphs the way reactor and detector engineers use them.
Theory
The atom density is N = ρN_A/M, the macroscopic cross section is Σ = Nσ, and the mean free path is λ = 1/Σ_t. A narrow beam is attenuated as I = I₀e^(−Σ_t x), and the reaction rate density is R = Σφ. A neutron needs n = ln(E₀/E)/ξ collisions to slow down, and moderators are compared by ξΣ_s/Σ_a.
Table A: Atom Densities and Cross Sections Undergraduate
| Material | N (per cm³) | Σ_t (cm⁻¹) | Σ_a (cm⁻¹) | λ (cm) |
|---|---|---|---|---|
| Light water | 3.343 × 10²² | 1.520 | 0.0222 | 0.658 |
| Heavy water | 3.322 × 10²² | 0.350 | 3.99 × 10⁻⁵ | 2.86 |
| Graphite | 8.523 × 10²² | 0.405 | 2.98 × 10⁻⁴ | 2.47 |
| Iron | 8.491 × 10²² | 1.202 | 0.217 | 0.832 |
| Cadmium | 4.634 × 10²² | 117.0 | 116.8 | 0.0085 |
Sample calculation (light water): N = 1.00 × 6.022×10²³ / 18.015 = 3.343×10²² molecules/cm³; Σ_t = 3.343×10²² × 45.46×10⁻²⁴ = 1.520 cm⁻¹; λ = 1/1.520 = 0.658 cm.
Table B: Beam Transmission (I₀ ≈ 300 000 counts in 60 s) Undergraduate
| Material | x (cm) | Predicted I/I₀ | Measured I/I₀ |
|---|---|---|---|
| Light water | 1.00 | 0.2188 | 0.2185 |
| Heavy water | 2.00 | 0.4964 | 0.4971 |
| Graphite | 5.00 | 0.1319 | 0.1315 |
| Iron | 1.00 | 0.3005 | 0.3009 |
| Cadmium | 0.05 | 0.00288 | 0.00285 |
Absorption rate density in light water at φ = 1.0×10¹² n/cm²·s: R = 0.0222 × 10¹² = 2.22×10¹⁰ per cm³·s.
Table C: Moderator Comparison (2 MeV → 0.025 eV) Undergraduate
| Moderator | ξ | n (collisions) | ξΣ_s (cm⁻¹) | ξΣ_s/Σ_a |
|---|---|---|---|---|
| Hydrogen nucleus | 1.000 | 18.2 | — | — |
| Deuterium nucleus | 0.725 | 25.1 | — | — |
| Light water | 0.927 | 19.6 | 1.39 | 62.5 |
| Heavy water | 0.510 | 35.7 | 0.179 | 4480 |
| Beryllium | 0.209 | 87.1 | 0.159 | 169 |
| Graphite | 0.158 | 115.2 | 0.0640 | 214 |
Table D: Cross Section against Energy (1/v law) Upper undergraduateGraduate step
| Absorber | E (eV) | Speed (m/s) | σ (b) |
|---|---|---|---|
| Boron-10 | 1 | 13 830 | 610.8 |
| Boron-10 | 0.0251 | 2191 | 3855 |
| Lithium-6 | 0.01 | 1383 | 1495 |
| Uranium-235 fission | 0.1 | 4374 | 294.2 |
| Uranium-235 fission | 10 | 43 740 | 29.42 |
Slope of every 1/v line on log–log axes: −0.5. Slider −1.6 gives E = 0.0251 eV, so σ is within 0.4% of σ₀. Include a screenshot of the graph with the U-238 resonance visible.
Graduate step: Maxwellian-averaged cross section σ̄ = (√π/2)σ₀√(293.6/T): boron-10 3403 b at 293.6 K and 2436 b at 573 K; uranium-235 fission 371.1 b at 573 K.
Table E: Slowing-Down Ladder (2 MeV → 0.025 eV, k = 10) Upper undergraduateGraduate step
| Moderator | ξ | k | E after k (eV) | Collisions to target | Relative to hydrogen |
|---|---|---|---|---|---|
| Hydrogen | 1.000 | 10 | 90.8 | 18.2 | 1.00 |
| Deuterium | 0.725 | 10 | 1 420 | 25.1 | 1.38 |
| Beryllium | 0.209 | 10 | 247 000 | 87.1 | 4.78 |
| Carbon | 0.158 | 10 | 412 000 | 115.2 | 6.33 |
Graduate step: exact ξ(A) and collisions from 2 MeV to 0.025 eV: A = 12, ξ = 0.1578, 115.3 collisions; A = 16, ξ = 0.1200, 151.7; A = 56, ξ = 0.0353, 515.6; A = 238, ξ = 0.00838, 2172.
Discussion
The calculated cross sections show a striking range: one mean free path is 0.66 cm in water but only 0.085 mm in cadmium, so half a millimetre of cadmium removed more than 99.7% of the thermal beam. Measured transmissions agreed with e^(−Σ_t x) to within counting statistics in every case. This huge absorption is why cadmium, boron, and hafnium are used in control rods and neutron shields, while iron, with a moderate Σ_a, is acceptable as a structural material.
Light water slows neutrons in the fewest collisions and has the largest slowing-down power, but its absorption gives it the lowest moderating ratio. Heavy water absorbs so little that its moderating ratio is about 70 times larger, which is why heavy-water reactors can operate on natural uranium, while light-water reactors need uranium enriched to about 3 to 5% U-235. Graphite needs many more collisions, so graphite-moderated cores must be physically large.
The graphs showed the same physics at a glance. On log–log axes every 1/v absorber fell on a straight line of slope −0.5, so boron-10 at 1 eV still absorbs about 611 b; this is why boron is effective in detectors and control rods across the whole thermal range. The U-238 curve rose by about four orders of magnitude at 6.67 eV, which explains why neutrons slowing through the resonance region can be lost to capture. On the slowing-down ladder, ten collisions brought a neutron in hydrogen from 2 MeV to about 91 eV, but left it at about 0.41 MeV in carbon; carbon needed 6.3 times as many collisions as hydrogen, matching the moderator ranking in Table C.
The graduate steps removed two simplifications. Averaging the 1/v cross section over a Maxwellian spectrum gave a value 11% below σ₀ at room temperature, and heating the moderator to 573 K lowered boron-10 absorption to 2436 b, 37% below the tabulated thermal value; reactor codes therefore use spectrum- and temperature-averaged cross sections, and this softening of absorption with temperature feeds into the reactivity coefficients. Deriving ξ from collision kinematics reproduced carbon (115 collisions) and showed why oxygen adds little moderation in water and why heavy fuel nuclei such as U-238 (over 2000 collisions) play no part in slowing neutrons down.
Conclusion
Neutron behaviour in each material followed Σ = Nσ and I = I₀e^(−Σ_t x). Moderators were ranked by slowing-down power (water highest) and by moderating ratio (heavy water highest), linking these cross sections directly to reactor design choices.
Practice Questions
Show all work and include units. Use the data table in the Theory section.