MCNP Example: PWR Fuel Pin Model

What you'll learn

First pin cell · 10 / 1018 min read
  • Build a four-region pin cell — fuel, gap, cladding, moderator — from four concentric cylinders and two axial planes.
  • Apply boundary conditions with the surface prefixes: + for a white boundary and * for a reflecting plane.
  • Define the four materials the model needs, including mt4 lwtr.10t for thermal scattering in water.
  • Score a flux spectrum with f4:n plus an e4 energy grid, and fission heating with f7:n.
  • Say why these boundary conditions give k rather than keff, and what k to expect for fresh 4.5% fuel.

Before you start

Problem Description

This example models a typical PWR fuel pin cell: a UO₂ fuel pellet, helium-filled gap, Zircaloy-4 cladding, and water moderator — enclosed by a white cylindrical boundary and two reflecting planes, which together stand in for an infinite lattice.

The four radii below are the ones every card on this page refers to, and they are the dimensions of a real 17×17 pressurized water reactor pin rather than round numbers. The equivalent cell radius is the one that is not a physical measurement: it is chosen to give a circle of the same area as the square 1.26 cm pitch cell, which is what makes a cylindrical outer boundary a fair stand-in for a square lattice.

RegionOuter radius (cm)MaterialDensity (g/cm³)
Fuel pellet0.4096UO₂, 4.5%10.4
Gap0.4178Helium0.0001
Cladding0.4750Zircaloy-46.56
Moderator0.7108Light water0.998

Complete Input File

Hover over any highlighted section to see what it does and why. Tap on mobile.

mcnp — hover sections to explore
PWR Fuel Pin Cell Model
c ---- Cell Cards ----
c cell mat density surfaces params
1 1 -10.4 -1 5 -6 imp:n=1 $ UO2 Fuel
2 2 -0.0001 1 -2 5 -6 imp:n=1 $ Helium Gap
3 3 -6.56 2 -3 5 -6 imp:n=1 $ Zircaloy-4 Cladding
4 4 -0.998 3 -4 5 -6 imp:n=1 $ Water Moderator
5 0 (4 : -5 : 6) imp:n=0 $ Outside the cell
 
c ---- Surface Cards ----
1 cz 0.4096 $ Fuel Radius
2 cz 0.4178 $ Gap Outer Radius
3 cz 0.4750 $ Clad Outer Radius
+4 cz 0.7108 $ Cell boundary (+ = white, see notes)
*5 pz 0.0 $ Bottom of the slice (* = reflecting)
*6 pz 1.0 $ Top of the slice, 1 cm tall
 
c ---- Data Cards ----
c Materials (positive fractions = atom, negative = weight)
m1 92235.70c 0.045 $ 4.5% enriched UO2 (atom fractions)
92238.70c 0.955
8016.70c 2.0
m2 2004.70c 1.0 $ Helium fill gas
m3 40090.70c 0.5145 $ Zirconium, natural (atom fractions)
40091.70c 0.1122
40092.70c 0.1715
40094.70c 0.1738
40096.70c 0.0280
m4 1001.70c 2.0 $ Light water (atom ratios)
8016.70c 1.0
mt4 lwtr.10t $ S(a,b) thermal scattering
 
c kcode: neutrons/cycle k-guess skip total
kcode 5000 1.0 50 250
ksrc 0 0 0.5 $ Mid-slice, not on a boundary
 
c Tallies
f4:n 1 $ Track-length flux in fuel cell
e4 1e-9 1e-8 1e-7 1e-6 1e-5 1e-4 1e-3 1e-2
0.1 1 2 3 4 5 6 7 8 9 10 $ Energy bin upper bounds (MeV)
f7:n 1 $ Fission energy deposition in fuel
 
c Output control
prdmp j 300 1 2

Annotated MCNP Input

Hover over any highlighted section to see a detailed explanation. Tap on mobile.

Cell cards (4 regions + graveyard)
Cylindrical surfaces (1–3)
Cell boundaries (+4, *5, *6)
UO₂ fuel material
Helium fill gas
Zircaloy-4 cladding
Water + S(α,β) scattering
KCODE settings
KSRC initial source
F4/E4 flux spectrum tally
F7 fission power tally
Output control

What to expect

Fresh 4.5% fuel at this hydrogen-to-heavy-metal ratio gives a k in the range 1.3 to 1.5. If your answer lands there, the deck is behaving. If it lands near 1.0, look first at the thermal scattering card, because a pin cell without mt4 loses a large part of the moderation that makes the lattice work — and it runs without complaint.

The flux spectrum from f4:n should show the thermal peak highest in the water and depressed inside the pellet. That dip is self-shielding, and it is physical rather than a modeling artifact: the outer layers of the pellet absorb the thermal neutrons before they reach the center, which is why a fuel pin burns from the outside in. The f7:n fission heating should be confined almost entirely to the fuel, since the fissions are there and F7 scores the recoverable energy at the fission site rather than following the gammas.

With 200 active cycles the standard deviation on k lands around 0.0003 to 0.0006, and all ten statistical checks should pass. The thermal and epithermal flux bins converge easily here because the pin cell is small and every history visits every region; that comfortable situation does not survive the move to a full core.

Three approximations in this deck are worth naming, because each is a deliberate trade rather than an oversight. The cylindrical outer boundary replaces the true square cell — a Wigner–Seitz equivalence that preserves the moderator volume but not the corner geometry, and which four reflecting planes at the real pitch would avoid at the cost of three more surfaces. The boundary conditions make the model infinite, so what comes out is k, and a finite reactor built from this lattice would sit lower by its leakage.

And the cross sections are room-temperature .70c data, which describes a cold shutdown core rather than an operating one. Moving to hot conditions means both a higher fuel temperature and a lower water density — and the two push k in the same direction, downward, by several thousand pcm together.

Where to take it next

This deck is a good platform for reactivity coefficients precisely because it is small enough to run repeatedly. Sweeping enrichment from 2 to 5 % traces out how k saturates — the returns diminish, because at some point you are adding U-235 to a lattice that has already thermalized everything it can. Changing the pitch varies the hydrogen-to-heavy-metal ratio and finds the moderation optimum, which is the calculation that shows why a pressurized water reactor is deliberately built on the under-moderated side of that peak. And reducing the water density models voiding, giving the sign and magnitude of the void coefficient directly.

Beyond that, the interesting additions each bring in new physics rather than new geometry. A burnable absorber layer on the pellet surface changes the reactivity history. The tmp card sets a temperature per cell so that Doppler broadening can be resolved. And the burn card turns a single k into a depletion history, at which point the run stops being a few minutes of work.

Card semantics on this page follow MCNP6.3.1 Theory & User Manual (LA-UR-24-24602 Rev. 1), §5.5.1 VOL: Cell Volume, §5.8.10 KCODE: Criticality Source and §5.8.11 KSRC: Criticality Source Points.

Full reference list on the attribution page.

Check yourself

  • Build a four-region pin cell from concentric cylinders and two axial planes?
  • Say why the curved outer boundary takes + and the axial planes take *?
  • Define the four materials, including mt4 for thermal scattering in water?
  • Score a flux spectrum with f4:n and fission heating with f7:n?
  • Explain why this model reports k rather than keff?