Serpent Manual

User Manual:

Open the PDF directly: View PDF .
Page Count: 164

Download
Open PDF In Browser	View PDF

Serpent – a Continuous-energy Monte Carlo
Reactor Physics Burnup Calculation Code
June 18, 2015

User’s Manual
Jaakko Leppänen

Preface
This documentation is a User’s Manual for the Serpent continuous-energy Monte Carlo reactor physics burnup calculation code.1 Code development started at the VTT Technical Research Centre of Finland in 2004, under the working title “Probabilistic Scattering Game”,
or PSG. This name is used in all publications dated before the pre-release of Serpent 1.0.0
in October 2008. The name was changed to due to the various ambiguities related to the
acronym. The code is still under development and this manual covers only the main functionality available in June 18.
The official Serpent website is found at http://montecarlo.vtt.fi. Support and minor updates
in the source code are currently handled via the Serpent mailing list, in which all users are
encouraged to join by sending e-mail to: Jaakko.Leppanen@vtt.fi. Any feedback is appreciated, including comments, bug reports, interesting results and ideas and suggestions for future development. A discussion forum for Serpent users is found at http://ttuki.vtt.fi/serpent.
For a quick start, experienced Monte Carlo code users are instructed to view the lattice input
examples in Chapter 11 starting on page 133.

For referencing the code, use either the website: “http://montecarlo.vtt.fi” or this report: “J. Leppänen. Serpent – a Continuous-energy Monte Carlo Reactor Physics Burnup Calculation Code. VTT Technical Research
Centre of Finland. (June 18, 2015)”

Contents
Preface

1 Installing and Running Serpent

Compiling Serpent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Running the Code . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Parallel Calculation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Nuclear Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Data Types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Directory File . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Radioactive Decay and Fission Yield Data . . . . . . . . . . . . . .

General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Input format . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Input cards . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Comment lines and sections . . . . . . . . . . . . . . . . . . . . .

Dividing the input into several files . . . . . . . . . . . . . . . . . .

Input errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

The Universe-based Geometry Model in Serpent . . . . . . . . . . . . . . .

Surface Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Surface types . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Positive and negative surface sides . . . . . . . . . . . . . . . . . .

Surface examples . . . . . . . . . . . . . . . . . . . . . . . . . . .

Cell Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Cell types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Cell examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Fuel pin definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Nests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Universes and Lattices . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Universe transformations and rotations . . . . . . . . . . . . . . . .

Lattices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Universe and lattice examples . . . . . . . . . . . . . . . . . . . .

Repeated Boundary Conditions . . . . . . . . . . . . . . . . . . . . . . . .

HTGR geometry types . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Implicit particle fuel model . . . . . . . . . . . . . . . . . . . . . .

Explicit particle / pebble bed fuel model . . . . . . . . . . . . . . .

HTGR geometry examples . . . . . . . . . . . . . . . . . . . . . .

Geometry plotter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4 Materials
4.1

Material definitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Nuclides . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Material cards . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Thermal scattering libraries . . . . . . . . . . . . . . . . . . . . . . . . . .

Doppler broadening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Material examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Neutron Population and Criticality Cycles . . . . . . . . . . . . . . . . . .

Energy grid reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . .

Library File Paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Unresolved resonance data . . . . . . . . . . . . . . . . . . . . . . . . . .

Doppler-Broadening Rejection Correction (DBRC) . . . . . . . . . . . . .

Boundary conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Source rate normalization . . . . . . . . . . . . . . . . . . . . . . . . . . .

Group constant generation . . . . . . . . . . . . . . . . . . . . . . . . . .

5.10 Full-core power distributions . . . . . . . . . . . . . . . . . . . . . . . . .

5.11 Delta-tracking options . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.12 Cross section data plotter . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.13 Fission source entropy . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.14 Soluble absorber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.15 Iteration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5.16 Fundamental mode calculation . . . . . . . . . . . . . . . . . . . . . . . .

5.17 Equilibrium xenon calculation . . . . . . . . . . . . . . . . . . . . . . . .

5.18 Miscellaneous parameters . . . . . . . . . . . . . . . . . . . . . . . . . . .

Main output file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Version, title and date . . . . . . . . . . . . . . . . . . . . . . . . .

Run parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . .

File paths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Delta-tracking parameters . . . . . . . . . . . . . . . . . . . . . .

Run statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Energy grid parameters . . . . . . . . . . . . . . . . . . . . . . . .

Unresolved resonance data . . . . . . . . . . . . . . . . . . . . . .

Nuclides and reaction channels . . . . . . . . . . . . . . . . . . . .

Reaction mode counters . . . . . . . . . . . . . . . . . . . . . . .

6.1.10 Slowing-down and thermalization . . . . . . . . . . . . . . . . . .

6.1.11 Parameters for burnup calculation . . . . . . . . . . . . . . . . . .

6.1.12 Fission source entropies . . . . . . . . . . . . . . . . . . . . . . .

6.1.13 Fission source center . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.14 Soluble absorber . . . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.15 Iteration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.16 Equilibrium Xe-135 calculation . . . . . . . . . . . . . . . . . . .

6.1.17 Criticality eigenvalues . . . . . . . . . . . . . . . . . . . . . . . .

6.1.18 Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.19 Point-kinetic parameters . . . . . . . . . . . . . . . . . . . . . . .

6.1.20 Six-factor formula . . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.21 Delayed neutron parameters . . . . . . . . . . . . . . . . . . . . .

6.1.22 Parameters for group constant generation . . . . . . . . . . . . . .

6.1.23 Few-group cross sections . . . . . . . . . . . . . . . . . . . . . . .

6.1.24 Fission product poison cross sections . . . . . . . . . . . . . . . . .

6.1.25 Fission spectra . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.26 Group-transfer probabilities and cross sections . . . . . . . . . . .

6.1.27 Diffusion parameters . . . . . . . . . . . . . . . . . . . . . . . . .

6.1.28 Pn scattering cross sections . . . . . . . . . . . . . . . . . . . . . .

6.1.29 P1 diffusion parameters . . . . . . . . . . . . . . . . . . . . . . . .

6.1.30 B1 fundamental mode calculation . . . . . . . . . . . . . . . . . .

6.1.31 Assembly discontinuity factors . . . . . . . . . . . . . . . . . . . .

6.1.32 Power distributions in lattices . . . . . . . . . . . . . . . . . . . . .

History output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7 Detectors
7.1

Detector Input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting the Response Function . . . . . . . . . . . . . . . . . . . .

Setting the Energy Domain . . . . . . . . . . . . . . . . . . . . . .

Setting the Spatial Domain . . . . . . . . . . . . . . . . . . . . . .

Surface Current Detectors . . . . . . . . . . . . . . . . . . . . . .

Detector output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Detectors in Burnup Calculation . . . . . . . . . . . . . . . . . . . . . . .

8 Burnup calculation

General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Depleted materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Irradiation history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Options for Burnup Calculation . . . . . . . . . . . . . . . . . . . . . . . .

Library File Paths . . . . . . . . . . . . . . . . . . . . . . . . . . .

Normalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Solution of Depletion Equations . . . . . . . . . . . . . . . . . . .

Calculation of Transmutation Cross Sections . . . . . . . . . . . .

Cut-offs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Nuclide Inventory . . . . . . . . . . . . . . . . . . . . . . . . . . .

Additional Output . . . . . . . . . . . . . . . . . . . . . . . . . . .

Decay heat production in multiple precursor groups . . . . . . . . .

Output in independent mode . . . . . . . . . . . . . . . . . . . . . . . . .

Output in coupled mode . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Burnup calculation examples . . . . . . . . . . . . . . . . . . . . . . . . .

Material and lattice examples . . . . . . . . . . . . . . . . . . . . .

Irradiation history examples . . . . . . . . . . . . . . . . . . . . .

9 External Source Mode

General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Source definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Setting the Spatial Distribution . . . . . . . . . . . . . . . . . . . .

Setting the Directional Distribution . . . . . . . . . . . . . . . . . .

Setting the Energy Distribution . . . . . . . . . . . . . . . . . . . .

Source files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Source Examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10 Reaction rate mesh plotter

10.1 Mesh input . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10.2 Mesh output . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11 Complete Input Examples

11.1 Quick start . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11.1.1 VVER-440 lattice calculation . . . . . . . . . . . . . . . . . . . . .

11.1.2 BWR lattice calculation . . . . . . . . . . . . . . . . . . . . . . . .

11.1.3 CANDU lattice calculation . . . . . . . . . . . . . . . . . . . . . .

11.1.4 Mixed UOX/MOX PWR lattice calculation . . . . . . . . . . . . .

11.2 Burnup calculation examples . . . . . . . . . . . . . . . . . . . . . . . . .

11.2.1 Pin-cell burnup calculation . . . . . . . . . . . . . . . . . . . . . .

11.2.2 PWR assembly burnup calculation . . . . . . . . . . . . . . . . . .

Chapter 1
Installing and Running Serpent
1.1

Compiling Serpent

The Serpent code is written in standard ANSI-C language. The code is mainly developed
in the Linux operating system, but it has also been compiled and tested in MAC OS X and
some UNIX machines.1 The Monte Carlo method is a computing-intensive calculation technique and raw computing power has a direct impact on the overall calculation time. It should
be taken into account that the unionized energy grid format used in Serpent requires more
computer memory compared to other continuous-energy Monte Carlo codes. One gigabyte
of RAM should be sufficient for steady-state calculations, but a minimum of 3 Gb is recommended for burnup calculation.
The source code is compiled simply by running the GNU Make utility.2 The Makefile provides for detailed instructions and various options for different platforms. Serpent uses the
GD open source graphics library [1] for producing some graphical output. If this library is
not installed in the system, the source code must be compiled with the “NO_GFX_MODE”
option. The compilation should not result in any errors or warning messages and it should
produce an executable named “sss”. Any problems in installation should be reported by
e-mail to: Jaakko.Leppanen@vtt.fi.
Code updates are provided to registered users by distributing the updated source files by
e-mail. New files replace old ones and the code must be re-compiled for the changes to take
effect.

The main platforms in PSG/Serpent development have been a 2.6 GHz dual-core AMD Opteron PC with
5 Gb RAM running Fedora Core 4 and an iBook G4 with 1.2 GHz PowerPC processor and 768 Mb RAM
running OS X v10.4.
2
For a detailed description of Makefiles, see: http://www.gnu.org/software/make.

1.2 Running the Code

Running the Code

All interaction between the code and the user is handled through one or several input files
and various output files, as described in the following chapters. The code is run from the
command line interface. The general syntax is:
sss []
where

is the name of the main input file
are the options

The input file is a standard text file containing the input description. The input can also be
divided into several files which are referred to in the main file.
The available options are:
print version information and exit
run the simulation using random number seed from
previous calculation
-plot
terminate run after geometry plot
-testgeom
test the geometry using randomly sampled
neutron tracks
-checkvolumes calculate Monte Carlo estimates for material
volumes by sampling random points
-mpi
run simulation in parallel mode (see Sec. 1.3)
-disperse
generate random particle or pebble distribution
files for HTGR calculations

-version
-replay

The replay option forces the code to use the same random number seed as in a previous run.
Without this option, the seed is taken from system time and written in a separate seed file
(named .seed) for later use. The seed can also be set manually in the input
using the “set seed” option.3
The geometry test option can be used for debugging the geometry in addition to the geometry
plotter. The code randomly samples neutron tracks across the geometry and checks that the
cells are correctly defined. Some input errors can spotted using this option.
The volume checking option can be used to verify that the volumes used in the calculation
are correct. The code is able to calculate cell volumes for simple lattice geometries, but
some more complicated geometry types require the values to be set by the user. The volumes
3

The results of a Monte Carlo calculation depend on the sequence of pseudo random numbers used during
the simulation. This sequence is fixed by the random number seed and the calculation can be repeated using the
same seed. The capability to reproduce the same simulation is important, for example, for debugging purposes.
Some codes, such as MCNP [2], use a fixed seed value, which results in the same results every time the code is
run. The Serpent code uses by default a different seed for each run and hence the results are different as well.
This behavior can be overridden by the replay command line option or by setting the seed manually in the input
file.

1.3 Parallel Calculation

are used for normalizing reaction rates for detectors and burnup calculation. The number of
random points should be large (at least 1,000,000) for good statistical accuracy.
The random particle / pebble distribution generator works by prompting the user information
on the volume type and dimensions, particle data and packing fractions. The code then
generates a distribution inside the desired volume without overlapping any particles. The
data is written in a file using format that can be directly read into the explicit HTGR geometry
model (See Sec. 3.8.2 on page 40). The option is available from code version 1.1.5 on.
IMPORTANT NOTES ON RUNNING THE CODE:
1. The seed file is overwritten by a new value each time the code is run without the replay
option and the old seed is lost.
SEE ALSO:
1. Dividing the input into several files (Sec.2.2.3 on page 16)
2. Setting the random number seed manually (Sec. 5.18 on page 73)
3. Geometry plotter (Sec. 3.9 on page 42)
4. Setting material volumes manually (Sec. 4.1.2 on page 49)

Parallel Calculation

Serpent uses the Message Passing Interface (MPI) [3] for parallel calculation. To activate this
capability the code must be compiled with the “PARALLEL_MPI” option (see the Makefile
for details) and the MPI libraries must be installed on the system.
There are two options for running the code in the parallel calculation mode. The first option
is to use the standard MPI tools, such as mpirun:
[user@host mpitest]$ mpirun -np 10 sss input

This command executes the calculation in 10 hosts as defined in the parallel environment.
The second option is to use the built-in MPI runner and define the number of tasks in the
command line:
[user@host mpitest]$ sss -mpi 10 input

In this calculation mode, the code attempts to run mpirun on its own. This may require small
modifications in the source code or may not work at all in some systems. The file path for
mpirun is defined by the “MPIRUN_PATH” precompiler variable in the “header.h” source
file.

1.4 Nuclear Data

IMPORTANT NOTES ON PARALLEL CALCULATION:
1. Parallel calculation is available from version 1.0.3 on.
2. When multiple tasks are sharing the same memory space, the size of allocated memory
is also multiplied. This should be taken into account when setting the memory size in
the compilation.
3. The methodology is still under development. The calculation lacks error tolerance and
load sharing and the mode should be used only in systems consisting of identical hosts.
Most of the MPI routines were directly adopted from PSG and features exclusively
available in Serpent (including burnup calculation) are not thoroughly tested.
SEE ALSO:
1. The MPI standard: http://www-unix.mcs.anl.gov/mpi/
2. The mpirun script:
http://www-unix.mcs.anl.gov/mpi/www/www1/mpirun.html

The Serpent code reads continuous-energy interaction data from ACE format cross section
libraries. The current installation package contains libraries based on JEF-2.2, JEFF-3.1,
ENDF/B-VI.8 and ENDF/B-VII evaluated data files. Since the data format is shared with
MCNP, alternative data for various isotopes should be readily available to most users. There
are also several ACE format data libraries based on different evaluations publicly available
through the OECD/NEA Data Bank [4]. New libraries can be produced from raw ENDF
format data using the NJOY nuclear data processing system [5].

Three types of cross sections are available in the data files. Continuous-energy neutron cross
sections (type 1) are used for the actual transport simulation. The data contains all necessary
reaction cross sections, together with energy and angular distributions, fission neutron yields
and delayed neutron parameters.
The second data type is the dosimetry cross section (type 2). Dosimetry cross sections exist for a large variety of materials and may include derived reaction modes not commonly
encountered in transport calculation. The data may consist of one or several partial cross
sections, but all energy and angular distributions are omitted. The data can be used with
detectors but not in physical materials included in the transport calculation.

1.4 Nuclear Data

Thermal scattering cross sections (type 3) are used to replace the low-energy free-gas elastic
scattering reactions for some important bound moderator nuclides, such as hydrogen in water
or carbon in graphite. Thermal systems cannot be modelled using free-atom cross sections
without introducing significant errors in the spectrum and the results.

The cross section data is accessed by using a separate directory file, which differs from the
“xsdir” file commonly used with ACE format data. A conversion between the two formats
can be made by running the “xsdirconvert” utility script, included in the installation package:
[user@host xsdata]$ xsdirconvert.pl data.xsdir >> data.xsdata

The Serpent directory file contains the data necessary for the code for locating the cross
section libraries and forming the material compositions. Each line in the directory file has
the following format:

is the name identifying the nuclide in the input file
is the actual nuclide name in the data
is the type of the data
is the isotope identifier (1000*Z + A)
is the isomeric state number (0 = ground state)
is the atomic weight
is the nuclide temperature (in K)
is the binary format flag (0 = ASCII, 1 = binary)
is the data path for the library

EXAMPLES:
1001.06c
H-1.06c
8016.06c
O-16.06c
40000.06c
Zr-nat.06c
92235.09c
U-235.09c
92238.09c
U-238.09c
95342.09c
Am-242m.09c
lwtr.03t
Np-237.30y

1001.06c
1001.06c
8016.06c
8016.06c
40000.06c
40000.06c
92235.09c
92235.09c
92238.09c
92238.09c
95342.09c
95342.09c
lwtr.03t
93237.30y

1
1
1
1
1
1
1
1
1
1
1
1
3
2

1001
1001
8016
8016
40000
40000
92235
92235
92238
92238
95242
95242
0
93237

0
0
0
0
0
0
0
0
0
0
1
1
0
0

1.00783
1.00783
15.99492
15.99492
91.21963
91.21963
235.04415
235.04415
238.05078
238.05078
242.05942
242.05942
0.00000
239.10201

600.0
600.0
600.0
600.0
600.0
600.0
900.0
900.0
900.0
900.0
900.0
900.0
0.0
0.0

0
0
0
0
0
0
0
0
0
0
0
0
0
0

/xs/1001_06.ace
/xs/1001_06.ace
/xs/8016_06.ace
/xs/8016_06.ace
/xs/40000_06.ace
/xs/40000_06.ace
/xs/92235_09.ace
/xs/92235_09.ace
/xs/92238_09.ace
/xs/92238_09.ace
/xs/95342_09.ace
/xs/95342_09.ace
/xs/tmccs1
/xs/llldos1

1.4 Nuclear Data

93237.30y 93237.30y 2 93237 0 239.10201

0.0 0 /xs/llldos1

The alias is the nuclide name used in the input file and it may or may not be the same as
the actual isotope name. The xsdirconvert tool writes two entries for each nuclide, one using
the original name and another one using the element symbol and the isotope number. The
data types are: 1 = continuous-energy, 2 = ACE dosimetry. 3 = thermal scattering, The
temperature entry is used with transport data only and the atomic mass with transport and
dosimetry cross sections.
Isomeric states are identified from the state number4 (see Am-242m in the example). There
is no standard convention on how to name these isotopes in the ACE format data, but the
xsdirconvert-tool assumes that the mass number of isomeric state nuclides is increased above
300. If another convention is used, the state number must be set manually in the directory
file. It is recommended that the isomeric state entries are always carefully checked after
running xsdirconvert.

Radioactive Decay and Fission Yield Data

Radioactive decay and fission yield data is needed for running the Serpent code in the independent burnup calculation mode. It is recommended that the libraries are included in the
coupled mode as well, since it enables the data to be reproduced in the output file, making it
directly available to the coupled calculation.
The decay constants and fission product distributions are read from standardized ENDF format data files [6]. The format is directly accessible and the data requires no preprocessing.
JEF-2.2, JEFF-3.1, ENDF/B-VI.8 and ENDF/B-VII data libraries are included in the installation package. More data can be downloaded from various Internet sources:
– OECD/NEA Data Bank: http://www.nea.fr/html/dbdata/
– Los Alamos T2 Nuclear Information Service: http://t2.lanl.gov
– US National Nuclear Data Center: http://www.nndc.bnl.gov
– US Radiation Safety Information Computational Center:
http://www-rsicc.ornl.gov
– IAEA Nuclear Data Centre: http://www-nds.iaea.org
– JAEA Nuclear Data Center: http://wwwndc.tokai-sc.jaea.go.jp
IMPORTANT NOTES ON INTERACTION DATA:
4

The information on isomeric states is needed for burnup calculation only. All nuclides are treated similarly
in the transport simulation.

1.4 Nuclear Data

1. The weight in the directory file is given as the atomic weight, not the atomic weight
ratio as in MCNP xsdir files.
2. The temperature in the directory file is given in Kelvin, not in MeV as in the MCNP
xsdir files.
3. Binary data is not supported in the current code version.
4. The data path in the directory file must refer to the absolute, not the relative location
of the library file.
5. The code always uses the first matching entry in the directory file. The use of duplicate
isotope names may lead to unexpected results.
SEE ALSO:
1. Setting up the file paths (Sec.5.4 on page 57)
2. Material definitions (Chapter 4 on page 47)

Chapter 2
Input
2.1

The Serpent code has no interactive user interface. All communication between the code and
the user is handled through one or several input files and various output files discussed in
Chapter 6. User-defined detectors are discussed as a separate item in Chapter 7 and burnup
calculation in Chapter 8.

The format of the input file is unrestricted. The file consists of white-space (space, tab or
newline) separated words, containing alphanumeric characters(’a-z’, ’A-Z’, ’0-9’, ’.’, ’-’).
If special characters or white spaces need to be used within the word (file names, etc.), the
entire string must be enclosed within quotation marks.

The input file is divided into separate data blocks, denoted as cards. The file is processed
one card at a time and there are no restrictions in what order the cards should be organized.
The input cards are listed in Table 2.1 and detailed descriptions are provided in the following
chapters. All input cards and special command words are case-insensitive. Each input card is
delimited by the beginning of the next card. It is hence important that none of the parameter
strings used within the card coincide with the card identifiers in Table 2.1.

2.2 Input format

Table 2.1: List of commands and input cards
Card
cell
dep
det
disp
ene
include
lat
mat
mesh
nest
particle
pbed
pin
plot
set
src
surf
therm
trans

Description
cell definition
irradiation history
detector definition
implicit HTGR particle fuel model
detector energy binning
read a new input file
lattice definition
material definition
reaction rate mesh plotter
nest definition
particle definition
explicit HTGR particle / pebble bed fuel model
pin definition
geometry plotter
misc. parameter definition
external source definition
surface definition
thermal scattering data definition
universe transformation

Chapter / Section
3.3
8.3
7.1
3.8.1
7.1.2
2.2.3
3.6.2
4.1.2
10.1
3.5
3.8
3.8.2
3.4
3.9
5.1
9.2
3.2
4.2
3.6.1

Page
24
110
95
39
99
16
30
48
131
27
39
40
27
42
53
126
20
49
29

Comment lines and sections

The Serpent code provides two types of comments for the input files. The percent-sign (%)
or hash (#) are used to define a comment line. Anything from this character to the end of the
line is omitted when the input file is read. The alternative is to use C-style comment sections
beginning with “/*” and ending with “*/”. Everything within these delimiters is omitted,
regardless of the number of newlines or special characters between them.

Dividing the input into several files

Complicated input descriptions can be simplified by dividing the cards into separate files.
This capability may also be useful if different calculation cases share some partial data.
Additional input files are recursively read from the main file using the include-command:
include ""
where

is the file path for the input file

When this command is encountered, the program will first read the included file before

2.2 Input format

continuing with the main file. The number of nested input files is unrestricted. Since file
names and paths often include non-alphanumeric characters, it is good practice to always
enclose the strings within quotation marks.

The Serpent code performs some error checking on the input file before proceeding with the
calculation. These checks include:
– Checking that there are an even number of quotation marks.
– Checking the correct number of parameters for some input cards.
– Checking the type (string, integer, real) of some parameters.
– Checking that the values of some parameters are within a reasonable range.
– Checking that all cards that are referred to in other cards are defined.
– Checking that all referred files exist.
– Checking that the input contains sufficient data for running the simulation.
– Various checks related to specific input cards.
Failure in any of the checks results in an error message and the termination of the calculation.
Most common input errors are caused by missing parameters or mistyped command words.
In the former case, the result is often an error message related to parameter type or number.
The program does not recognize card names with typing errors, but rather processes the entire
card as if was a set of parameters belonging to the previous card. Such errors may stop the
calculation later on for entirely different reasons, or in the worst case, run the simulation with
a set of parameters totally different from what the user intended. In case of any unexpected
behavior, the typing of the card names should the first thing to be checked.
IMPORTANT NOTES ON INPUT FORMAT:
1. The input file consists of white-space separated words containing alphanumeric characters. If special characters or white spaces need to be used (file names, etc.), the entire
string must be enclosed within quotation marks.
2. Each card is delimited by the beginning of the next card and it is hence important
that the card names are not used in for other purposes, for example as cell or material
names. If the name of an input card is spelled incorrectly, the previous card is not
delimited, which may result in a completely unexpected behavior.

3. Running the Serpent code should never result in crash or termination without an error
message. In such case, please report the problem by e-mail to
Jaakko.Leppanen@vtt.fi.

Table 2.2 summaries the most essential units used in the code.
Table 2.2: Units used in the Serpent code.
Quantity
Distance
Area
Volume
Time
Energy
Microscopic cross section
Macroscopic cross section
Mass
Mass density
Atomic density
Power
Power density
Neutron flux
Reaction rate
Burnup
Burn time

Unit
cm
cm2
cm3
s
MeV
b
1/cm
g
g/cm3
1024 /cm3
W
kW/g
1/cm2 s
1/cm3 s
MWd/kgU
days

(depends on the case)
(barn = 10−24 cm2 )

( = 1/barn×cm)

(reaction rate density)
(per total initial heavy metal)

IMPORTANT NOTES ON UNITS:
1. Power, neutron flux, reaction rate and all related quantities depend on how the neutron
source rate is normalized.
SEE ALSO:
1. Source rate normalization (Sec. 5.8 on page 61)

Chapter 3
Geometry
3.1

The Universe-based Geometry Model in Serpent

The Serpent code uses a universe-based geometry model for describing complicated structures, very similar to MCNP. This means that the geometry is divided into separate levels,
which are all constructed independently and nested one inside the other. This approach allows the complexity of the geometry to be divided into smaller parts, which are much easier
to handle. It also enables the use of regular geometry structures, such as square and hexagonal lattices, commonly encountered in reactor applications.
Perhaps the best example of a universe-based geometry construction is the reactor core. At
the highest level, the geometry consists of fuel pins, in which the fuel pellets are surrounded
by cladding and coolant. Each pin type is described independently in its own universe. The
next level is the fuel assembly, in which the pin universes are arranged in a regular lattice.
The assembly may also comprise flow channel walls, moderator channels or any support
structures. In the next geometry level these assembly universes are arranged in another lattice
to form the core layout, which can be surrounded by radial and axial reflectors and finally
the reactor pressure vessel wall.
The basic building block of the geometry is the cell, which is a region of space determined
using simple boundary surfaces. Each cell is filled with a homogeneous material composition, void or another universe.

Surface Definitions

Serpent provides for various elementary and derived surface types for geometry construction. A “derived” surface type refers here to a surface comprised of two or more elementary
surfaces, such as a cube constructed of six planes. The input format does not make any dif-

3.2 Surface Definitions

ference between elementary and derived surfaces and the description below applies to both.
The syntax of the surface card is:
surf ...
where

is the surface identifier
is the surface type (see Table 3.1)
are the surface parameters

The surface identifier is an arbitrarily chosen number identifying the surface in the cell definitions. Surface types and their use is described in the following subsections.

The present code version contains 20 surface types, listed in Table 3.1. The number of
parameters is fixed and depends on the type. Some surface types have parameters that are
optional.
For the three types of planes, the x0 , y0 and z0 coordinates refer to distances from the origin. For sphere, cube and the cylindrical surfaces these parameters define the coordinates
of the surface center. Sphere, cube and cylinder radii are given by r. The square, hexagonal and cruciform cylinders also include an optional parameter r0 , which defines the radius
of rounded corners. If this parameter is omitted, it is assumed that the corners are sharp.
The optional parameters z1 and z2 are bottom and top planes of truncated cylinder. The
cylindrical surfaces are illustrated in Figure 3.1.
The cuboid is defined by the minimum and maximum coordinates in each direction.
The hexagonal prismatic surfaces are similar to the corresponding cylinders, with the difference that the enclosed space is limited by bottom and top planes at z1 and z2 .
The “pad” is a cylindrical surface type that was included in the code in order to model the
neutron pad in the VENUS-2 reactor dosimetry benchmark [7]. The surface is defined as a
sector between angles θ1 and θ2 cut out from a layer between cylinders of radii r1 and r2 .
The “cone” or “conz” surface type (see Fig. 3.2) is determined by the x0 , y0 and z0 coordinates of the base, the base radius r and the height h. The height of the cone also determines
the orientation: a positive value for a cone pointing in positive direction and a negative value
for a cone pointing in the negative direction of the z-axis. Cones oriented in the x- and y-axes
(“conx” and “cony”, respectively) are defined in a similar manner.
The “dode” and “octa” surface types (see Fig. 3.3) are determined by the x0 and y0 coordinates of the central axis and two distances r1 and r2 from the center. If the second value
is omitted, the surface is a regular octa- or dodecagonal cylinder. The octagonal cylinder
basically consists of two intersecting square and the dodecagonal surface of two intersecting

3.2 Surface Definitions

Table 3.1: Surface types in the Serpent code.
Type
inf
px
py
pz
sph
cylx
cyly
cylz or cyl
sqc
cube
cuboid
hexxc
hexyc
hexxprism
hexyprism
cross
pad
conx
cony
conz or cone
dode
octa
plane
quadratic

Description
all space
plane perpendicular to x-axis
plane perpendicular to y-axis
plane perpendicular to z-axis
sphere
circular cylinder parallel to x-axis
circular cylinder parallel to y-axis
circular cylinder parallel to z-axis
square cylinder parallel to z-axis
cube
cuboid
x-type hexagonal cylinder parallel to z-axis
y-type hexagonal cylinder parallel to z-axis
x-type hexagonal prism parallel to z-axis
y-type hexagonal prism parallel to z-axis
cruciform cylinder parallel to z-axis
(see description below)
cone oriented in the x-axis
cone oriented in the y-axis
cone oriented in the z-axis
dodecagonal cylinder parallel to z-axis
octagonal cylinder parallel to z-axis
general plane
general quadratic surface

Parameters
x0
y0
z0
x0 , y0 , z 0 , r
y0 , z0 , r, x1 , x2
x0 , z0 , r, y1 , y2
x0 , y0 , r, z1 , z2
x0 , y0 , r, r0
x0 , y0 , z 0 , r
x1 , x2 , y1 , y2 , z 1 , z 2
x0 , y0 , r, r0
x0 , y0 , r, r0
x0 , y0 , r, z1 , z2
x0 , y0 , r, z1 , z2
x0 , y0 , r, d, r0
x 0 , y 0 , r1 , r2 , θ 1 , θ 2
x0 , y0 , z0 , r, h
x0 , y0 , z0 , r, h
x0 , y0 , z0 , r, h
x 0 , y 0 , r1 , r2
x 0 , y 0 , r1 , r2
A, B, C, D
A, B, C, D, E, F, G, H, J, K

regular hexagons.
The general plane is defined by equation
Ax + By + Cz = D
This is a simplified case of the general quadratic surface, defined by
Ax2 + By 2 + Cz 2 + Dxy + Eyz + F zx + Gx + Hy + Jz + K = 0

Positive and negative surface sides

The surfaces are used for defining the geometry cells as will be described in the following
section. For this purpose, each surface is associated with a positive side and a negative side.
It is defined that a point is inside a surface if it is located on the negative side of the surface.

3.2 Surface Definitions

Figure 3.1: Basic cylinder types. The surfaces are infinite in the z-direction. The square
cylinder illustrates the definition of rounded corners.

h
x0 y0 z 0
r
cone
Figure 3.2: The cone surface.

For the three types of planes, the positive side is defined in the direction of the positive
coordinate axis. The positive sides of the sphere, cube, cone and the cylindrical surfaces are
defined outside the perimeter of the surface.

Surface examples

A few simple examples of surface definitions are given in the following.

3.2 Surface Definitions

Figure 3.3: The octagonal and dodecagonal cylinder surfaces.

% --- plane perpendicular to x-axis, located at x = 4.0 cm:
surf 1 px
4.000
% --- sphere centered at (1.0, 0.0, 2.0), radius 5.0:
surf 2 sph
1.000 0.000 2.000 5.000
% --- cylinder centered at origin, radius 10.5 cm:
surf 3 cyl
0.000 0.000 10.500
% --- cube at origin with diameter 5.0 cm:
surf 4 cube
0.000 0.000 0.000 2.500
% --- square cylinder centered at origin, radius 10.0 cm,
%
rounded corners with radii 0.2 cm:
surf 5 sqc
0.000 0.000 10.000 0.200
% --- x-type hexagonal cylinder centered at (1.0, 0.0),
%
radius 2.0 cm:
surf 6 hexxc 1.000 0.000 2.000
% --- cruciform cylinder centered at origin, radius 20.0 cm,
%
half-thickness 5.0 cm:
surf 7 cross 0.000 0.000 20.000 5.000
% --- neutron pad used in the VENUS-2 benchmark:
surf 8 pad
0.000 0.000 11.250 54.750 59.073 65.073
% --- cone at origin, base diameter 2.0 cm, height 5.0 cm
surf 9 cone
0.000 0.000 0.000 1.000 5.000

IMPORTANT NOTES ON SURFACES:

3.3 Cell Definitions

1. In code versions earlier than 1.1.8 the cone surface type may only be used with the full
delta-tracking calculation mode (threshold = 1).
2. Reflective and periodic boundary conditions may only be used in geometries where
the outermost boundary is defined by a square or hexagonal cylinder or a cube.
3. The dodecagonal cylinder surface type is available from code version 1.1.4 on.
4. The octagonal cylinder and general plane and quadratic surface are available from code
version 1.1.9 on.
SEE ALSO:
1. Delta-tracking options (Sec. 5.11 on page 66)
2. Boundary conditions (Sec. 5.7 on page 59)

Cell Definitions

The geometry description in the Serpent code consists of two- or three-dimensional regions,
denoted as cells. Each cell is defined using a set of positive and negative surface numbers,
which correspond to the surface identifiers defined in the surface cards. Unlike MCNP and
other Monte Carlo codes, Serpent can only handle intersections of boundary surfaces. This
means that the neutron is inside the cell, if and only if it is on the same side of each boundary
surface as given in the surface list (see the examples below).
The lack of the union operator restricts the generality of the geometry description to some
extent. This limitation is compensated for by providing a large collection of derived surface
types, which in most cases can be used to replace the unions of the elementary surfaces. The
advantage of this approach is that the geometry description remains relatively simple.1

The syntax of the cell card is:
1

It is known that the use of derived surface types may slow down the neutron tracking routine in some cases
when the conventional ray-tracing algorithm is used. Neutron transport in Serpent, however, is primarily based
on the delta-tracking method which is not prone to such limitations. The use of derived surface types reduces
the total number of surfaces, which may actually speed up the delta-tracking routine in complicated geometries.

3.3 Cell Definitions

is the cell name
is the universe number of the cell
is the cell material
are the boundary surfaces

The cell name is a text string that identifies the cell.2 Each cell belongs to a universe, which
is determined by the universe number (lattices and universes are thoroughly described in
Section 3.6 on page 28). Cell material determines the name of the material that fills the cell
(see Chapter 4 for material definitions). There are three exceptions:
1. If the cell is empty, the material name is set to “void”.
2. If the cell describes a region of space that is not part a of the geometry, the material
name is set to “outside”.
3. If the cell is filled by another universe, the material name is replaced by command
“fill” and the number of the filling universe.
The “outside” cells are required for filling the regions of space that are not a part of the actual
geometry. When the neutron streams to such a region, the history is terminated or boundary
conditions are applied.
The cell shape is determined by the list of boundary surfaces. Positive entries refer to positive
(“outside”) surface sides and negative entries to negative (“inside”) surface sides. The cell is
defined as the intersection of all surfaces in the list.

A few simple examples of cell definitions are given in the following.
% --- two half-planes separated by a plane in the z-axis at 5.0 cm:
surf 1 pz
cell 1 1
cell 2 1

5.000
water
air

% lower half-plane filled with "water"
% upper half-plane filled with "air"

% --- solid uranium sphere ("Godiva") of radius 8.7407 cm:
2

When the number of cells in the geometry is large, it is often easier to replace cell names with numerical
constants. This is possible since the code treats cell numbers as any other text strings. This convention is
followed in most example cases in this manual.

3.3 Cell Definitions

surf 1 sph
cell 1 0
cell 2 0

uranium
outside

% uranium inside sphere
% outside world

% --- tungsten-reflected plutonium sphere:
surf 1 sph
surf 2 sph
cell 1 0
cell 2 0
cell 3 0

plutonium
tungsten
outside

0.0
0.0
-1
1 -2
2

5.0419
9.7409
% plutonium inside surface 1
% tungsten between surfaces 1 and 2
% outside world

% --- a segment of LWR fuel rod in water:
surf
surf
surf
surf
surf

cyl
0.0
cyl
0.0
cyl
0.0
pz -50.0
pz
50.0

cell
cell
cell
cell
cell
cell

UO2
void
clad
water
water
water

-1
1 -2
2 -3
3
-4
5

UO2 fuel inside surface 1
gap between fuel and cladding
cladding
water outside cladding
water below the segment
water above the segment

IMPORTANT NOTES ON CELLS:
1. Material names “void”, “outside” and “fill” are reserved for empty cells, cells
not belonging to the geometry and cells filled by another universe, respectively.
2. Only the intersection operator is available for cell definitions. This means that a point
is inside the cell if and only if it is inside (or outside if defined by a negative surface
number) all the boundary surfaces in the list.
SEE ALSO:
1. Material definitions (Chapter 4 on page 47)
2. Boundary conditions (Sec. 5.7 on page 59)

3.4 Fuel pin definitions

Fuel pin definitions

Since Serpent is primarily a lattice physics code, the geometry has a simplified definition for
fuel pins consisting of nested annular material layers. The syntax of the pin card is:
pin

is the pin identifier (universe number)
are the materials
are the outer radii of the material regions

The fuel pin is not an actual geometry object, but rather a macro that is used to define a
pin universe. The material regions and their outer radii are given in ascending and the code
constructs the cells using using cylindrical surfaces. If the radius is negative, it is interpreted
as layer thickness instead of absolute radius. The universe number is set by the pin identifier.
Pin materials can also be other universes, which are defined using the fill command (See
filled cells on page 28).
Pin definitions are most commonly used with lattices to define fuel assemblies. Examples
are given in the following section.
IMPORTANT NOTES ON PIN DEFINITIONS:
1. The pin identifier is a universe number, which must not coincide with another universe.
2. The outermost material regions is given without a radius and it fills the rest of the
universe.
3. Layer thickness are available from version 1.1.13 on.
SEE ALSO:
1. Filled cells (Sec. 3.6 on page 28)
2. Lattice examples (Sec. 3.6.3 on page 32)

Fuel pin and particle (see Sec. 3.8 on page 39) are special cases of the nest geometry type,
defined as:

3.6 Universes and Lattices

is the nest identifier (universe number)
is the surface type
are the materials
are the surface parameters

Nested objects consist of materials or sub-universes separated by similar surfaces. Nests can
be defined using planar (px, py, pz), cylindrical (cyl, sqc, hexxc, hexyc), spherical
(sph) or cubical (cube) surface types. In each case the parameters , , ...
define the main dimension, all other parameters are set to zero.

Universes and Lattices

As mentioned above, a universe-based geometry allows the geometry to be divided into
separate levels. Each universe is defined independently and must cover all space. Regions of
space not belonging to the geometry must be defined using “outside” cells. The universes are
defined by the cell universe numbers and the geometry is layered by replacing the material
name with the fill command:
cell fill ...
where

is the cell name
is the universe number of the cell
is the universe number of the filling universe
are the boundary surfaces

Each universe has its own origin, which can be shifted using the universe transformation
command (see Sec. 3.6.1) The lowest level of the geometry belongs to universe 0, which
must always exist.

Universe transformations and rotations

Each universe is by default centered at the origin, which may sometimes cause difficulties
with filled cells. The origin can be shifted to another location using the universe transformation card:

3.6 Universes and Lattices

is the universe number
is the x coordinate of the new origin
is the y coordinate of the new origin
is the z coordinate of the new origin

Universe transformations are also convenient, for example, for positioning control rods in a
reactor core. Universes filled in a lattice structure are automatically shifted to the appropriate
position and transformations are not needed.
Universe rotations were implemented in Version 1.1.14. The syntax of the transformation
card with rotations has two alternative formats:
trans
trans ...
where

is the universe number
is the x coordinate of the new origin
is the y coordinate of the new origin
is the z coordinate of the new origin
is the rotation angle around x-axis
is the rotation angle around y-axis
is the rotation angle around z-axis
are the coefficients of a rotation matrix

If three values are entered after the coordinates, they are interpreted as rotation angles around
the three coordinate axes. If nine values are entered, they form the rotation matrix, which is
used to multiply the position and direction vectors when the rotation is applied.
The coordinate translation always precedes the rotation.

Lattices are special universes, filled with a regular structure of other universes. The Serpent code has eight lattice types: square lattice, two hexagonal lattices, the circular cluster
array, three infinite 3D lattices filled with a single universe and the vertical stack.

Square and hexagonal lattices
The syntax of the lattice card for the square and hexagonal lattices is:

3.6 Universes and Lattices

is the universe number of the lattice
is the lattice type (= 1, 2 or 3)
is the x coordinate of the lattice origin
is the y coordinate of the lattice origin
is the number of lattice elements in the x-direction
is the number of lattice elements in the y-direction
is the lattice pitch

The lattice card is followed by a list of universe numbers, which determines the layout. The
lattice type numbers are:
1. Square lattice
2. X-type hexagonal lattice (unit cell is the x-type hexagonal cylinder, see Fig. 3.1)
3. Y-type hexagonal lattice (unit cell is the y-type hexagonal cylinder, see Fig. 3.1)
Each lattice defines a universe, which must be embedded inside a cell using the fill command. If the bounding cell is larger than the lattice, neutrons may stream to undefined lattice
positions, which results in a geometry error. This can be avoided by increasing the lattice
size by defining additional positions in the periphery (see examples below).

Circular cluster array
The circular cluster array (lattice type 4) is defined by:
lat
where

is the universe number of the lattice
is the lattice type (= 4)
is the x coordinate of the lattice origin
is the y coordinate of the lattice origin
is the number of rings in the array

The lattice card is followed by a list of rings, which are defined by:
...
where

is the number of sectors in ring
is the central radius of the ring
is the angle of rotation
are the universe numbers filling the sectors

3.6 Universes and Lattices

The circular array is needed for constructing some cluster-type fuel assemblies, used in
CANDU, MAGNOX, AGR and RBMK reactors. The array is also convenient for determining the fuel rod layout in some small research reactors, such as the TRIGA.

Infinite 3D lattices
The infinite 3D lattices are used to construct repeated structures of identical cells that fill
the entire universe. This type of construction can be used, for example, for describing the
microscopic fuel particles inside an HTGR fuel pebble or compact. The syntax is:
lat

is the universe number of the lattice
is the lattice type (= 6, 7 or 8)
is the x coordinate of the lattice origin
is the y coordinate of the lattice origin
is the lattice pitch
is the filler universe

Lattice type 6 is a cubical lattice and types 7 and 8 x- and y-type hexagonal prismatic lattices,
respectively.

Vertical stack
Universes can be vertically stacked, one on top of the other, using lattice type 9:
lat
where

is the universe number of the lattice
is the lattice type (= 9)
is the x coordinate of the lattice origin
is the y coordinate of the lattice origin
is the number of axial layers

The lattice card is followed by a list of axial layers, which are defined by:

where is the axial position (lower boundary of the layer)
is the filler universe
The z-values must be given in ascending order. Space below the lowest z-value is not defined
and the top layer fills the entire space above the highest value.

3.6 Universes and Lattices

Cuboidal 3D lattice
The cuboidal lattice is a 3D structure composed of cuboids with different dimensions in the
x-, y- and z-directions. The syntax is:
lat
where

is the universe number of the lattice
is the lattice type (= 11)
is the x coordinate of the lattice origin
is the y coordinate of the lattice origin
is the z coordinate of the lattice origin
is the number of lattice elements in the x-direction
is the number of lattice elements in the y-direction
is the number of lattice elements in the z-direction
is the lattice pitch in x-direction
is the lattice pitch in y-direction
is the lattice pitch in z-direction

The lattice card is followed by a list of universes. This lattice type is available from version
1.1.17 on.

Universe and lattice examples

The universe and lattice definitions are best described using a few examples. The fist example
is a 17×17 PWR MOX fuel assembly containing three types of fuel pins and empty control
rod guide tubes (see Figure 3.4 on page 45).
% --- MOX pin 1:
pin 1
MOX1
4.36250E-01
void
4.43750E-01
clad
4.75000E-01
water
% --- MOX pin 2:
pin 2
MOX2
4.36250E-01
void
4.43750E-01
clad
4.75000E-01
water
% --- MOX pin 3:
pin 3
MOX3
4.36250E-01

3.6 Universes and Lattices

void
clad
water

4.43750E-01
4.75000E-01

% --- Empry guide tube:
pin 4
water
5.62500E-01
clad
6.12500E-01
water
% --- Pin lattice (pitch = 1.26 cm):
lat 10
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1

1
2
3
3
3
2
3
3
2
3
3
2
3
3
3
2
1

2
3
3
3
3
4
3
3
4
3
3
4
3
3
3
3
2

1 0.0 0.0 17 17 1.26
2
3
3
4
3
3
3
3
3
3
3
3
3
4
3
3
2

2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2

2
2
4
3
3
4
3
3
4
3
3
4
3
3
4
2
2

2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2

2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2

2
2
4
3
3
4
3
3
4
3
3
4
3
3
4
2
2

2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2

2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2

2
2
4
3
3
4
3
3
4
3
3
4
3
3
4
2
2

2
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
2

2
3
3
4
3
3
3
3
3
3
3
3
3
4
3
3
2

2
3
3
3
3
4
3
3
4
3
3
4
3
3
3
3
2

1
2
3
3
3
2
3
3
2
3
3
2
3
3
3
2
1

1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
1
1

The second example is a hexagonal VVER-440 lattice with 126 fuel pins and a central instrumentation tube. Empty lattice positions are filled with water (see Figure 3.5 on page 45).

% --- Fuel pin with central hole:
pin 1
void
fuel
void
clad
water

0.08000
0.37800
0.38800
0.45750

% --- Central instrumentation tube:

3.6 Universes and Lattices

pin 2
water
clad
water

0.44000
0.51500

% --- Empty lattice position filled with water:
pin 3
water
% --- Pin lattice (x-type hexagonal, pitch = 1.23 cm):
lat 10

0.0 0.0 15 15 1.23

3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
3 3 3 3 3 3 3 1 1 1 1 1 1 1 3
3 3 3 3 3 3 1 1 1 1 1 1 1 1 3
3 3 3 3 3 1 1 1 1 1 1 1 1 1 3
3 3 3 3 1 1 1 1 1 1 1 1 1 1 3
3 3 3 1 1 1 1 1 1 1 1 1 1 1 3
3 3 1 1 1 1 1 1 1 1 1 1 1 1 3
3 1 1 1 1 1 1 2 1 1 1 1 1 1 3
3 1 1 1 1 1 1 1 1 1 1 1 1 3 3
3 1 1 1 1 1 1 1 1 1 1 1 3 3 3
3 1 1 1 1 1 1 1 1 1 1 3 3 3 3
3 1 1 1 1 1 1 1 1 1 3 3 3 3 3
3 1 1 1 1 1 1 1 1 3 3 3 3 3 3
3 1 1 1 1 1 1 1 3 3 3 3 3 3 3
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

The third example is a CANDU cluster consisting of 37 pins in 4 rings. The third ring is
rotated by 15 degrees (see Figure 3.6 on page 46).
% --- Fuel pin:
pin 1
fuel 0.6122
clad 0.6540
coolant
% --- Cluster:
lat
1
6
12
18

10 4 0.0 0.0 4
0.0000
0.0 1
1.4885
0.0 1 1 1 1 1 1
2.8755 15.0 1 1 1 1 1 1 1 1 1 1 1 1
4.3305
0.0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

3.6 Universes and Lattices

All three examples are illustrated using the geometry plotter in Section 3.9 on page 42. It
should be noted that the plots contain cell structures not included in the above examples.
The following example demonstrates the use of the vertical stack:
% --- Uranium ball:
surf 1 sph 0.0 0.0 2.5 2.5
cell 1 1 uranium
cell 2 1 void

% --- Stack 5 balls:
lat 2 9 0.0 0.0 5
0.0
5.0
10.0
15.0
20.0

Notice that the origin of universe 1 is positioned at the bottom of each layer.
IMPORTANT NOTES ON UNIVERSES AND LATTICES:
1. Each universe is defined independently and must cover all space. Regions of space not
belonging to the geometry must be defined using “outside” cells.
2. The lowest level of the geometry belongs to universe 0, which must always exist.
3. Each universe has its own origin, which can be shifted using the universe transformation command.
4. Cells in higher geometry levels can only be accessed through filled cells or lattices.
5. Each lattice defines a universe, which must be embedded inside a cell using the fill
command. The lattice must fill the container cell completely to avoid neutron streaming to undefined lattice positions.
6. Hexagonal lattices are defined using a square matrix for the universe layout. To position the lattice cells correctly, a number of empty cells must be defined for each row.
The definition is best described in the example in Sec. 3.6.3 on page 33.
7. Multi-level hexagonal structures (pin-assembly-core) are defined using both x- and
y-type hexagonal lattices with different type at each level.

3.7 Repeated Boundary Conditions

8. If the infinite lattice types are is used in burnup calculation, material volumes must be
set manually (see Sec. 4.1.2 on page 49).
9. The vertical stack lattice type is available from code version 1.1.8 on.
SEE ALSO:
1. Pin definitions (Sec. 3.4 on page 27)
2. Filled cells (Sec. 3.6 on page 28)

Repeated Boundary Conditions

What happens to neutrons that end up in a region defined as being outside the geometry is
dictated by the boundary conditions. There are three options:
1. Black boundary - the neutron is killed
2. Reflective boundary - the neutron is reflected back into the geometry
3. Periodic boundary - the neutron is moved to the opposite side of the geometry
The condition is set by the “bc” parameter, described in Section 5.7 on page 59.
Reflective and periodic boundary conditions can be used to construct infinite and semiinfinite lattice structures. The way these boundary conditions are handled in Serpent is
somewhat different from other Monte Carlo codes. Instead of stopping the neutron at the
boundary surface, reflections and translations are handled by coordinate transformations.
This limits the outermost boundary to a few specific surface types that can be used to define
a square or hexagonal lattice structure. There are basically three options:
Infinite 2D geometry: The geometry has no dependence on the z-coordinate. The outer
boundary is defined by a single square or hexagonal cylinder (“sqc”, “hexxc” or
“hexyc”).
Radially infinite, axially finite 3D geometry: The outer boundary is defined by a square
or hexagonal cylinder (“sqc”, “hexxc” or “hexyc”and two axial planes (“pz”).
The boundary condition takes effect in the radial direction only. The axial boundary
conditions are black.
Infinite 3D geometry: The outer boundary is defined by a single cube, cuboid or hexagonal prism (“cube”, “cuboid”, “hexxprism”or “hexyprism”). The boundary
condition takes effect in all directions.

3.7 Repeated Boundary Conditions

The following examples illustrate the different geometry types. The details of the geometry
are omitted for the sake of simplicity and replaced by a fill command.
An infinite 2D hexagonal geometry can be defined as:
surf

% --- Cells:
cell 1
cell 99

fill 10
outside

Note that the reflective boundary condition is unphysical in a hexagonal geometry. infinite
2D square geometry can be defined as:
surf

-0.233 -0.233 7.68750
fill 10
outside

In both cases the outer boundary is defined by a single surface.
If the geometry is finite in the axial dimension, the system becomes three-dimensional. A
radially infinite square lattice can be defined as:
surf
surf
surf

cell 1
cell 97
cell 98
cell 99

sqc -0.233 -0.233 7.68750
pz -200.0
pz 200.0
0
0
0
0

fill 10
outside
outside
outside

It is also possible to define the outside world as:
cell 97
cell 98
cell 99

outside
outside
outside

3.7 Repeated Boundary Conditions

but the code may run slower because the boundary condition will be handled also for some
neutrons that end up outside the geometry.
As for the infinite 2D geometry, the boundary in an infinite 3D geometry must be defined by
a single surface, such as a cube:
surf

cube 0.0 0.0 0.0 3.0
0
0

fill 10
outside

or a hexagonal prism:
surf

hexxprism 0.0 0.0 1.880 0.0 4.93
0
0

fill 10
outside

In both cases the boundary conditions are enforced in both radial and axial directions.
IMPORTANT NOTES ON REPEATED BOUNDARY CONDITIONS:
1. The outer boundary must be defined by a single surface in infinite 2D and 3D geometries. The allowed surface types for a 2D geometry are square and hexagonal cylinders.
Infinite 3D geometries can be defined using a cube, cuboid or hexagonal prism.
2. Axially infinite, radially finite geometries are defined by a square or hexagonal cylinder
and two axial planes. The way the outside world is defined may affect the running time.
3. The hexagonal cylinder and prismatic surfaces are physically reasonable only with
periodic boundary conditions (reflective boundary conditions work if the geometry
has a 30 degree symmetry). The use of reflective boundary conditions with these
types was enabled in update 1.1.18. In earlier code versions the boundary condition is
automatically changed into periodic.
SEE ALSO:
1. Surface types (Sec. 3.2.1 on page 20)
2. Defining the outside world (Sec. 3.3.1 on page 25)
3. Setting the boundary condition (Sec. 5.7 on page 59)

3.8 HTGR geometry types

HTGR geometry types

The fuels in high-temperature gas-cooled reactors (HTGR) consist of microscopic TRISO
particles dispersed in a graphite matrix. The multi-layer fuel particles can be defined similar
to fuel pins (see Sec. 3.4 on page 27):
particle

is the particle identifier (universe number)
are the materials
are the outer radii of the material regions

The simplest approach is to describe the particle distribution as a regular lattice, using lattice
types 6, 7 or 8 (See the infinite 3D lattices in Sec. 3.6.2). However, the regular arrangement
fails to account for the random distribution of the particles and often leads to a distorted
fuel-to-moderator ratio due to cell cut-off at the outer boundary. For this reason the Serpent
code has two geometry models specifically designed for HTGR fuels.

Implicit particle fuel model

The implicit particle fuel model works by sampling new particles on the neutron flight path
during the tracking process. The input syntax is:
disp ...
where

is the universe number of the dispersed medium
is the universe filling the space between the particles
are the packing fractions of the particle types
are the radii of the particle types
are the universe numbers of the particle types

The number of particle types is not limited, but the sum of the packing fractions must be less
than 1.0 (physical factors set the upper limit much lower, although this is not checked by the
routine).
The implicit particle fuel model was revised in update 1.1.3. It should be noted that the
model is not exact and there are statistically significant differences compared to the explicit
model described below. The implicit model seems to work best for low packing fractions but
no comprehensive validation has been carried out yet.

3.8 HTGR geometry types

Explicit particle / pebble bed fuel model

A better choice for modeling HTGR geometries is the explicit particle fuel model, which
reads the positions of the particles from a separate file. The same model can be used for
setting up reactor-scale pebble-bed geometries. The input syntax is:
pbed "" []
where

is the universe number of the dispersed medium
is the universe filling the space between the particles / pebbles
is the input file containing the particle / pebble coordinates
are the options

The particle / pebble distribution is handled explicitly, so there are no approximations done
in the modeling. Each line in the input file describes the position of a single particle / pebble.
The format is:

is the x coordinate of the particle / pebble
is the y coordinate of the particle / pebble
is the z coordinate of the particle / pebble
is the radius of the particle / pebble
is the universe number of the particle / pebble

The total number of entries is unlimited, although memory or running time may become a
limiting factor if the number exceeds several million.
The options are used to activate the calculation of various particle / pebble-wise parameters.
Currently the only available option is the power distribution, which is requested with option “pow”. The code writes the output in a separate file “.out”, where
“” is the file where the distribution was read. The input data is included for
convenience. The format of the output is:

is the x coordinate of the particle / pebble
is the y coordinate of the particle / pebble
is the z coordinate of the particle / pebble
is the radius of the particle / pebble
is the universe number of the particle / pebble
is the power produced inside the particle / pebble
is the associated relative statistical error

All results depend on source normalization (see Sec. 5.8 on page 61).

3.8 HTGR geometry types

HTGR geometry examples

The following example shows how the particle distribution inside a single PBMR fuel pebble
can be modeled using a regular 3D array and the two particle fuel models in the Serpent code.
The definition of a fuel particle is very similar to the fuel pin:
% --- Definition of a coated fuel particle:
particle 1
fuel
buffer
PyC
SiC
PyC
matrix

0.0250
0.0340
0.0380
0.0415
0.0455

The first option is to describe the particle distribution as a regular cubical lattice:
% --- Option 1: regular 3D array:
lat

10 6 0.0 0.0 0.16341 1

The implicit particle fuel model is defined using a list of packing fractions and particle types:
% --- Filler universe composed of graphite:
surf 1 inf
cell 1 2 matrix

% --- Option 2: implicit particle fuel model:
disp 10 2 0.09043 4.55000E-02 1

The explicit particle fuel model reads particle coordinates from a separate input file (can be
used for pebble distributions at reactor scale as well):
% --- Filler universe composed of graphite:
surf 1 inf
cell 1 2 matrix -1
% --- Option 3: explicit particle fuel model (read coordinates from file):
pbed 10 2 "particles.inp"

3.9 Geometry plotter

Finally the pebble description using one of the three options (all assigned with universe
number 10):
% --- Pebble:
surf 10 sph 0.0 0.0 0.0 2.5
surf 20 sph 0.0 0.0 0.0 3.0
surf 30 cube 0.0 0.0 0.0 3.0
cell
cell
cell
cell

fill 10
matrix
helium
outside

-10
10 -20
20 -30
30

IMPORTANT NOTES ON HTGR GEOMETRY TYPES:
1. The implicit particle fuel model was revised in update 1.1.3. The model is not exact
and should be used with caution. Test calculations show that the model works best for
low packing fractions.
2. If the implicit particle fuel model is used in burnup calculation, material volumes must
be set manually (see Sec. 4.1.2 on page 49).
3. Calculation of particle / pebble-wise power distributions is available from update 1.1.4
on.
SEE ALSO:
1. An earlier version of the implicit particle fuel model in Ref. [8]

Geometry plotter

The geometry plotter uses the GD open source graphics library [1] for producing png format
output files for visualization. In order to use the plotter, the source code must be compiled
with this library included (see the Makefile for detailed instructions). The syntax of the
plotter command is:

3.9 Geometry plotter

is the orientation of the plot plane (1, 2 or 3)
is the width of the plot in pixels
is the height of the plot in pixels
is the position on the axis perpendicular to the plot plane
is the minimum value of the first coordinate
is the maximum value of the first coordinate
is the minimum value of the second coordinate
is the maximum value of the second coordinate

The orientation of the plot plane is defined as:
1. yz-plot (perpendicular to the x-axis)
2. xz-plot (perpendicular to the y-axis)
3. xy-plot (perpendicular to the z-axis)
The plotted area is a rectangle defined by the orientation, the position on the perpendicular
coordinate axis and the coordinates of the two corners. Zero position is assumed if the
position parameter is omitted. If the corner coordinates are not given, the boundary values
of the geometry are used.
Each plotter command produces an output file named “_geom.png”, where
is the name of the input file and is the plot index. The resolution of the figure
is defined by the width and height parameters. Each material is represented by a randomly
selected color (void regions are in black, geometry errors bright green or red). Surfaces are
drawn with black lines, which may overlap cell regions. It should be noted that the plotted
surfaces may not necessarily represent the actual cell boundaries.
Example plots are shown in Figures 3.4–3.7. The lattices in the first three cases are described
in the universe and lattice examples in Sec. 3.6.3. In each case the plotter command was:
plot 3 1000 1000
This generates a 1000 by 1000 pixel plot perpendicular to the z-axis, located at z = 0 and
covering the entire geometry.
IMPORTANT NOTES ON GEOMETRY PLOTTER:
1. The geometry plotter uses the GD open source graphics library [1], which must be
installed in the system.
2. The plotter produces png (portable network graphics) format output files.

3.9 Geometry plotter

3. The colors in the plot represent different materials. The color for each material is
selected randomly (void regions are black, geometry errors bright green or red).
4. Surfaces are drawn with black lines, which may overlap cell regions. Plotted surfaces
may not necessarily represent the actual cell boundaries.
SEE ALSO:
1. Compiling Serpent (Sec. 1.1 on page 8)
2. The GD open source graphics library: http://www.libgd.org

3.9 Geometry plotter

Figure 3.4: A 17×17 PWR MOX fuel assembly with 3 pin types.

Figure 3.5: A hexagonal VVER-440 fuel assembly with 126 fuel pins and a central instrumentation tube in an infinite lattice. The proportions of the assembly are slightly distorted
since the hexagonal assembly is fitted inside a square region.

3.9 Geometry plotter

Figure 3.6: A CANDU cluster with 37 fuel pins in 4 rings. The third ring is rotated by 15
degrees.

Figure 3.7: A 10×10 BWR fuel fuel assembly with 7 pin types and an asymmetrically positioned moderator channel.

Chapter 4
Materials
4.1

Material definitions

The geometry in Monte Carlo codes consists homogeneous material regions, which in Serpent are defined using cells and surfaces (see Chapter 3 for geometry definition).1 Each
material consists of a list of nuclides and each nuclide is associated with a cross section
library, as defined in the directory file (see Sec. 1.4.2 on page 12).
Nuclide temperatures are fixed when the cross section data is generated and cannot be
changed afterwards. It is important to use cross section libraries generated at the right temperature to correctly model the Doppler-broadening of resonance peaks. It is equally (or even
more) important to use the appropriate bound-atom thermal scattering libraries for moderator
nuclides.
Soluble absorbers can be defined by mixing two material compositions. This option is introduced in Sec. 5.14 on page 69. The concentration can be used for critical keff iteration.
Serpent also has the option to use a built-in Doppler broadening routine to adjust nuclide
temperatures before the calculation. This method is described in Sec. 4.3 on page 50.

Nuclide names may be arbitrary aliases defined in the directory file. The usual convention,
also used by MCNP, is:
1

It is, in principle, possible to model continuously varying material compositions when the delta-tracking
method is used for neutron transport. This option is considered for the future versions of the Serpent code.

4.1 Material definitions

.
where
is the element Z

is the isotope mass number (three digits)
is the library id
For example, “92235.09c” refers to 235 U. Natural element cross sections are denoted by
mass number zero (“40000.06c” for natural zirconium). The library id usually refers to
data evaluation or temperature (“60c” for ENDF/B-VI.0 based data, “09c” for data generated at 900K, and so on...).
There is no standard convention on how to name isomeric states. The xsdirconvert-utility
used for producing Serpent directory files assumes a form in which the isotope mass number
is simply increased above 300 (“95342.09c” for 242m Am). In any case it is important to
realize that the nuclide names are used for identification only and they do not contain any
information used by the code in the calculation.

The basic syntax of the material card is:
mat []

is the material name
is the density (mass or atomic)
are the options (depending on case)
are the names of the constituent nuclides
are the corresponding fractions (mass or atomic)

Material name is used to identify the material in cell cards (see Sec. 3.3.1 on page 25). The
nuclide names correspond to the identifier determined in the directory file. These identifiers
define the cross section data used in the calculation. Densities and fractions can be given
as atomic or mass values. Positive entries refer to atomic densities (in units of 1024 /cm3 )
and atomic fractions, respectively, and negative entries to mass densities (in units of g/cm3 )
and mass fractions. Isotopic compositions are normalized before the calculation and mixed
entries are not allowed.
If the material density is set to zero or “sum”, the value is calculated from the isotopic composition. The isotope fractions must then be in absolute density units, not relative fractions.
Material volumes and masses are used for normalizing reaction rates, which is important,
for example, in burnup calculation. The code calculates these automatically for simple pin
structures, but the values must be entered manually for some more complicated geometries.

4.2 Thermal scattering libraries

Material volume is set using option:
mat vol
...
where is the total material volume in cm3
and material mass (alternatively):
mat mass
...
where is the total material mass in g
Colors for the geometry plotter (see Sec. 3.9 on page 42 can be set using:
mat rgb
...
where is the value for red channel (between 0 and 255)
is the value for green channel (between 0 and 255)
is the value for blue channel (between 0 and 255)
if the colors are not set, random values are used in the plots.
Other options are used to set up depletion zones in burnup calculation and to determine thermal scattering libraries for moderator materials and temperatures for Doppler broadening.
Material definitions in burnup calculation is a separate topic in Section 8.2 on page 109.
Thermal scattering and Doppler broadening are discussed below.

Thermal scattering libraries

Thermal scattering cross sections are used to replace the low-energy free-gas elastic scattering reactions for some important bound moderator nuclides, such as hydrogen in water
or carbon in graphite. Thermal systems cannot be modelled using free-atom cross sections
without introducing significant errors in the spectrum and results. Thermal scattering data is
defined using:
therm
where

is the name of the data library
is the library identifier as defined in the directory file

The library identifier is the actual name of the library in the directory file. The library name
is used to associate the data with a material, in which case the material card has an additional
entry:

4.3 Doppler broadening

is the material name
is the density (mass or atomic)
is the name of the thermal scattering data library
is the nuclide ZA of the moderator nuclide
are the names of the constituent nuclides
are the corresponding fractions (mass or atomic)

The nuclide ZA identifies the moderator nuclide (in the form of: 1000*Z + A). The “moder”
entry can be used several times to define thermal scattering libraries for multiple nuclides,
such as hydrogen and deuterium in heavy water (see the example in Sec. 4.4).

Doppler broadening

The Doppler broadening routine is initiated by adding a “tmp” entry in the material card:
mat tmp

is the material name
is the density (mass or atomic)
is the Doppler temperature in Kelvin
are the names of the constituent nuclides
are the corresponding fractions (mass or atomic)

The broadening is performed after the data is read from the ACE format libraries and there
is slight increase in the overall calculation time, depending on the number of nuclides. If the
the same nuclide is broadened to several temperatures in different materials, there is also an
increase in memory usage. The routine works only if the given temperature is above the original one. The cross section libraries provided with the Serpent code are generated between
300K intervals and it is recommended that the closest temperature below the broadened value
is used as the basis.

Material examples

A few simple examples of material definitions are given in the following.

4.4 Material examples

% --- Fuel consisting of enriched UO2 and burnable absorber.
%
Atomic densities given in units of 1/(barn*cm):
mat UO2Gd
92234.09c
92235.09c
92238.09c
64154.09c
64155.09c
64156.09c
64157.09c
64158.09c
64160.09c
8016.09c

sum
4.2940E-06
5.6226E-04
2.0549E-02
4.6173E-05
2.9711E-04
4.1355E-04
3.1518E-04
4.9786E-04
4.3764E-04
4.5243E-02

%
%
%
%
%
%
%
%
%
%
%

Atomic
Atomic
Atomic
Atomic
Atomic
Atomic
Atomic
Atomic
Atomic
Atomic
Atomic

density
density
density
density
density
density
density
density
density
density
density

from composition
of U-234
of U-235
of U-238
of Gd-154
of Gd-155
of Gd-156
of Gd-157
of Gd-158
of Gd-160
of O-16

% --- Zircaloy cladding:
mat clad
40000.06c
24000.50c
26000.55c
28000.42c
50000.42c
8016.06c

-6.55000
-0.98135
-0.00100
-0.00135
-0.00055
-0.01450
-0.00125

Mass
Mass
Mass
Mass
Mass
Mass
Mass

density given in g/cm3
fraction of natural zirconium
fraction of natural chromium
fraction of natural iron
fraction of natural nickel
fraction of natural tin
fraction of O-16

% --- Boronized light water with thremal scattering data:
mat water
1001.06c
8016.06c
5010.06c
5011.06c

-0.7207 moder lwtr 1001
-1.1180E-1
-8.8755E-1
-1.1890E-4
-5.3110E-4

therm lwtr lwtr.04t
% --- Heavy water with thermal scattering data (two libraries):
mat D2O
8016.06c
1002.06c
1001.06c

-0.812120
moder lwtr1 1001 moder hwtr1 1002
-7.99449E-1
-1.99768E-1
-7.83774E-4

therm lwtr1 lwtr.04t
therm hwtr1 hwtr.04t
% --- Doppler broadening from 900K to 1000K:
mat fuel

-10.45700 tmp 1000

4.4 Material examples

92235.09c
92238.09c
8016.09c

-0.03173
-0.84977
-0.11850

IMPORTANT NOTES ON MATERIALS:
1. Nuclide names are used for identification only. All information used in the calculation
is read from the directory file and the cross section data.
2. Positive entries in material density and isotopic composition refer to atomic densities
and atomic fractions, respectively, and negative entries to mass densities and mass
fractions. Typical input errors in material compositions are related to confusing the
two definitions.
3. Isotopic compositions can be given as density values, rather than fractions, since the
compositions are normalized before the calculation.
4. The mass fraction of oxygen in UO2 fuel is ∼0.1185. Natural boron consists of 20%
10
B and 80% 11 B (atomic fractions).
5. Thermal scattering data must be used for moderator materials (water, graphite, etc.)
when modelling thermal systems. The use of free-atom cross sections will introduce
significant errors in the results.
6. Doppler broadening is available from code version 1.1.0 on, and completed in version
1.1.2. The broadened temperature must be above the original nuclide temperature.
SEE ALSO:
1. Directory files and the “xsdirconvert” utility (Sec. 1.4.2 on page 12)
2. Soluble absorber definitios (Sec. 5.14 on page 69)

Chapter 5
Options
5.1

Serpent has various calculation parameters determined using the “set” command:
set ...
where
...

is the parameter name
are the parameter values

The available options are listed in Table 5.1 and described in more detail in the following
sections. Parameters used for burnup calculation are described in Section 8.4 of Chapter 8.
Table 8.2 on page 111 summarizes the options in the burnup calculation mode.

Neutron Population and Criticality Cycles

The default calculation mode in Serpent is the k-eigenvalue criticality source method, in
which the simulation is run in cycles and the source distribution of each cycle is formed by
the fission reaction distribution of the previous cycle. External source simulation is discussed
as a separate topic in Chapter 9. The parameters for criticality source calculation are set
using:
set pop [ ]
where

is the number of source neutrons per cycle
is the number of active cycles run
is the number of inactive cycles run
is the initial guess for keff
is the collection interval

5.2 Neutron Population and Criticality Cycles

The number of source neutrons per cycle is fixed. Since the number of generated source
points usually differs from this value, the source size is increased (keff < 1) or decreased
(keff > 1) to match the given source size.
Inactive cycles are cycles that are run in order to allow the initial fission source distribution to
converge before starting to collect the results. In lattice calculations the convergence is typically reached well within the first 20 cycles. Source convergence in full-core calculations,
however, may take much longer.
The initial source points are randomly selected inside the fissile cells in the geometry and no
source input is needed from the user. The simulation requires an initial guess for keff , which
by default is set to unity. This is usually sufficient, but if the system is far from criticality,
the simulation may die out during the first few cycles. This problem may be overcome by
setting the initial keff guess to a more appropriate value.
If all fissile material is located in a small region compared to the geometry dimensions, initial
source sampling may fail. The default source can be overridden by defining an external
source, as described in Section 9.2 of Chapter 9. If the “nps” parameter is not set, the
user-defined source is used as the initial guess only, and the simulation proceeds in power
iterations.
The statistical accuracy of the results depends on the total number of active neutron histories
run, which is determined by the population size per cycle and the total number of active
cycles. Appropriate values for a typical lattice calculation are 500 active cycles of 5000
source neutrons. If more precision is required or the geometry is larger, it is suggested that
the population size, rather than the number of cycles is increased.
The collection interval defines the number of generations run for each batch of results. By
default this value is set to one, and increasing the number has essentially the same impact as
running more neutrons per generation and fewer generations.1
Serpent uses two buffers to store data for new source points and neutrons produced in multiplying scattering reactions and certain special calculation modes. In some cases these buffers
may be overrun, which terminates the simulation. To overcome such problems, the buffer
size may be increased by setting:
set nbuf
where is the buffer factor (criticality mode) or total size (external source mode)
In criticality source mode the buffer size is population size multiplied by the given factor (set
to 2.5 by default). In external source mode neutron histories are run one at a time and the
value of nbuf sets the absolute size of the buffer (set to 1000 by default).
IMPORTANT NOTES ON NEUTRON POPULATION AND CRITICALITY CYCLES:
1

The difference is that the correlations between adjacent batches could be weaker, which may have some
impact in the statistics in large geometries (the effects have not yet been studied).

5.3 Energy grid reconstruction

1. It is important that a sufficient number of cycles is discarded to allow the initial fission source to converge before starting to collect the results. This number depends on
the size and the complexity of the geometry. Fission source convergence is a complicated research topic, subject to both theoretical and practical considerations [9–14].
It should be noted, however, that problems with source convergence are practically
never encountered in lattice calculations where the neutron migration distance is long
compared to the dimensions of the geometry.
2. The k-eigenvalue criticality source calculation yields physically consistent results only
in the special case that keff = 1. When the system is far from criticality, the importance
of fission neutrons is either over- (keff < 1) or underestimated (keff > 1). The result is
that the neutron population becomes biased in energy and space (and time), which may
affect the final results as well. The problem originates from the basic methodology
and the fact that a physically sub- or super-critical chain reaction is simulated as a
steady-state system. Deterministic lattice transport codes use neutron leakage models
to overcome this problem, but the methodology for Monte Carlo calculation is not well
established.
3. All source neutrons are born in fission. Other neutron-multiplying (n,xn) reactions are
treated as scattering within the criticality cycle.
4. External source definitions are available from version 1.1.11 on.
5. Buffer size and collect interval are options available from version 1.1.13 on.
SEE ALSO:
1. Simulating the neutron chain reaction and the k-eigenvalue criticality source calculation mode in Ref. [15] (Sec. 5.5 on page 112).
2. Discussion on neutron leakage models in Monte Carlo calculation in Ref. [15] (Sec. 9.5
on page 171).
3. External source simulation (Chapter 9).

Energy grid reconstruction

The continuous-energy reaction cross sections in Serpent are reconstructed using a single
unionized energy grid for all nuclides. The reason for this approach is the major speed-up
in calculation, achieved by minimizing the number of grid search iterations.2 The default
2

Each nuclide in the continuous-energy ACE format data is associated with its own energy grid. The
calculation of material total cross sections, for example, is carried out by summing over all the constituent
nuclides. This requires an iterative energy grid search to be performed for each nuclide, which may take a
significant fraction of the overall CPU time, especially since the procedure has to be repeated each time the
neutron enters a new material. The advantage of using the same grid for all nuclides is that the grid search has
to be performed only once, each time the neutron scatters to a new energy.

5.3 Energy grid reconstruction

method for grid reconstruction is that all grid points of all nuclides in the ACE format data
are included in the master grid. The disadvantage of this method is that computer memory is
wasted for storing a large number of redundant data points. The available memory is usually
not a problem in fresh fuel calculations, but the introduction of actinide and fission product
isotopes in burnup calculation may result in an excessively large master grid. Serpent has
a method for avoiding this problem by combining adjacent grid points. The reconstruction
parameters are given by:
set egrid [ ]
where
is the fractional reconstruction tolerance
is the minimum energy in the grid (MeV)
is the maximum energy in the grid (MeV)
The fractional reconstruction tolerance is the minimum relative difference between two grid
points, below which the points are combined. All points below the lower limit and above the
upper limit are discarded.
The default value for the fractional reconstruction tolerance is zero in the transport calculation mode and 5 · 10−5 in the burnup calculation mode. Test calculations have shown that
the results are not significantly affected until the tolerance is raised above 10−3 . There is
no absolute guarantee, however, that the results are valid in all imaginable cases when the
grid size is significantly reduced. It is therefore suggested that the grid reduction is not used
unless necessary because of insufficient computer memory.
The lower limit of the energy grid is by default set to 10−9 MeV and the upper limit to
15 MeV. Very few neutrons are born or scattered to higher or lower energies in fission reactor
applications.
If a reduction in memory size is necessary, an alternative to grid thinning is the double
indexing method, in which the data is stored in the original ACE format and the unionized
grid used only for accessing the nuclide-wise grids. This method is activated by:
set dix
where is 1 if the method is used and 0 if not.
The double indexing method reduces the memory usage, but may lead to an increase in
processing time, which may become significant in burnup calculation. Double indexing is
turned off by default.
IMPORTANT NOTES ON ENERGY GRID:
1. Grid reduction inevitably leads to some loss of data. There is no guarantee that this
reduction does not affect the results.
2. Test calculations have shown that the reduction in grid size does not significantly affect
the overall calculation time.

5.4 Library File Paths

SEE ALSO:
1. Cross section data in the PSG code in Ref. [15] (Sec. 8.2 on page 143). NOTE: Some
of the described methods are outdated.
2. A more recent description of the unionized energy grid formats in Serpent is found in
Ref. [16].

Library File Paths

The Serpent code uses a single directory file for determining the cross sections used in the
transport simulation. The directory file can be generated from an MCNP xsdir file using the
“xsdirconvert” utility (see Sec.1.4 on page 11). The cross section library directory file path
is set using:
set acelib ""
where is the file path for the ACE directory file
A default directory path can be set by defining environment variable SERPENT_DATA. The
code looks for cross section directory files in this path if not found at the absolute location.
IMPORTANT NOTES ON DATA FILES AND FILE PATHS:
1. The cross section library directory file is a text file generated by the “xsdirconvert”
utility.
2. The environment variable feature is available from code version 1.1.8 on.
SEE ALSO:
1. Setting up the directory file (Sec.1.4 on page 11)
2. Setting up file paths for burnup calculation (Sec.8.4 on page 111)

Unresolved resonance data

The use of unresolved resonance probability tables can be switched on and off using:

5.5 Unresolved resonance data

set ures [] [dilu] [ ...]
where

is the option (1 = use data, 0 = omit data)
is the calculation mode
is the infinite dilution cut-off
are the nuclides for which the data is used or omitted

Since the probability table sampling has to be carried out during tracking, the transport cycle
tends to slow down quite significantly.3 There are three options to treat the probability table
data:
1. Sample all cross sections at once, each time the neutron scatters to a new energy, adjust
material totals and majorant.
2. Sample cross sections when the neutron enters a new material. Use a pre-calculated
majorant cross section corresponding to the maximum probability table values.
3. Sample cross sections when the neutron enters a new material. Switch to surface
tracking when neutron is in the unresolved range.
The overall calculation time using the different options depends on the case, in particular the
flux spectrum and the number of nuclides with probability table data. Mode 1 is used by
default. It should also be noted that options 2 and 3 work by sampling the cross sections for
physical materials. If a material is used for detector calculation only, the probability tables
may not be appropriately sampled. This is not a problem for method 1.
Since the overall impact of using unresolved resonance cross sections is a self-shielding
effect, the calculation routine can be optimized by omitting the probability table sampling
for nuclides with low concentration. This limit is given by the infinite dilution cut-off, which
is set to zero by default.
If the options are followed by a list of nuclide names (94239.09c, etc.), the probability table
treatment is used or omitted only for the listed nuclides. If no list is given, the options cover
all nuclides with probability table data.
In order for the methodology to work, the probability table data must be available in the ACE
format cross section libraries. This data is not included, for example, in the default libraries
provided with installation package 1.1.0. The methodology is available from update 1.1.4 on
and is still under development. The mode and infinite dilution cut-off options were added in
update 1.1.5. The treatment is currently switched off by default.
3

Serpent pre-calculates certain material-wise total cross sections to avoid having to sum over the constituent
nuclides during the transport cycle. This pre-calculation cannot be combined with probability table sampling,
which has to be carried out on-the-fly.

5.6 Doppler-Broadening Rejection Correction (DBRC)

Doppler-Broadening Rejection Correction (DBRC)

There is a physical flaw in the ENDF reaction laws of the ACE format data, that ignores
the impact of thermal motion on angular distributions of elastic scattering near resonances.
The phenomenon becomes important in heavy nuclides (A > 200) with scattering resonances
at low energy (< 0.2 keV), and ignoring it may result in a noticeable under-prediction of
resonance absorption and over-prediction of keff . To correct this flaw, Serpent can apply a
Doppler-broadening rejection correction (DBRC) in the thermal free-gas model:
set dbrc [ ...]
where
is the minimum energy for DBRC (MeV)

is the maximum energy for DBRC (MeV)
are the zero-Kelvin cross section data of the nuclides involved
The method uses zero-Kelvin elastic scattering cross sections in the rejection sampling loop
and the provided data tells the code which nuclides should use the correction. If the correction is used with U-238, for example, the entry is the nuclide name for U-238 generated at
0K (“92238.00c”). It is usually sufficient to use DBRC for the primary heavy nuclide only.
The energy range should cover the low resonance peaks. Typical range for U-238 is from 0.4
to 210 eV. The method is not used by default.
IMPORTANT NOTES ON DBRC:
1. The correction increases resonance absorption, which may reduce keff by few hundred
pcm.
2. DBRC is not widely used by other Monte Carlo codes, so switching the correction on
may increase differences to any reference results.
3. The method is available from update 1.1.14 on.
4. Zero-Kelvin cross section data is not available in the cross section libraries distributed
with the current base versions.
SEE ALSO:
1. Theory behind DBRC is discussed in reference [17].

Boundary conditions

Boundary conditions determine the fate of neutrons escaping outside the defined geometry.
The boundary conditions are set using:

5.7 Boundary conditions

set bc
where is the boundary condition
The Serpent code has three available boundary condition options: 1 - black, 2 - reflective
and 3 - periodic. Default is the black boundary, which means that all neutrons streaming into
outside cells are killed. Reflective and periodic boundary conditions can be used for setting
up infinite lattices. When the neutron encounters a reflective boundary, it is diverted back
into the geometry. In the case of a periodic boundary, the neutron is moved to the opposite
surface.
Different boundary conditions can be applied in x-, y- and z- surfaces of square cylinder,
cube and cuboidal boundary. The syntax is then:
set bc
where is the boundary condition in the x-direction
is the boundary condition in the y-direction
is the boundary condition in the z-direction
All three entries must be given, even if the geometry is two-dimensional. This capability is
available in code version 1.1.17 on.
Symmetries in finite geometries can be taken into account using the universe symmetry option:
set usym
where

is the universe number
is symmetry type
is the x-coordinate of symmetry origin
is the y-coordinate of symmetry origin

Present version of Serpent allows only quadrant symmetries ( = 4) in universe 0.
IMPORTANT NOTES ON BOUNDARY CONDITIONS:
1. The reflective and periodic boundary conditions can only be used in geometries where
the outer boundary surface is either a square or a hexagonal cylinder or a cube.
2. Even though the reflective and the periodic boundary conditions produce the same
results in many cases, it should be noted that they are not equivalent when the geometry
is asymmetric. This is the case, for example, for BWR assemblies surrounded by an
asymmetric moderator channel. Infinite BWR lattices must alway be defined using
reflective, instead of periodic boundary conditions.
3. If black boundary conditions are used, the outer geometry boundary must be non reentrant or leakage will produce unphysical results.

5.8 Source rate normalization

4. The universe symmetry option is available from version 1.1.14 on.
SEE ALSO:
1. Outside cells (Sec. 3.3.1 on page 24)

Source rate normalization

The integral reaction rate estimates given by a Monte Carlo simulation are more or less
arbitrarily normalized, unless fixed by a given constant. The Serpent code provides for seven
options for source rate normalization.
Normalization to fission neutron generation rate is set using:
set genrate
where is the number of fission neutrons emitted per second
The neutron generation rate includes only prompt and delayed neutrons emitted in fission.
All (n,xn) reactions are treated as neutron-multiplying scattering within the criticality cycle.
Normalization to source rate is set using:
set srcrate
where is the number of neutrons emitted per second
Normalization to source rate is recommended to be used only in external source calculation
mode, in which case the total source rate refers to the rate at which neutrons are emitted from
the user-defined source. In criticality source mode, the normalization is done for the number
of neutron histories started per generation. Normalization to total fission rate is set using:
set fissrate
where is the number of fission reactions per second
Normalization to total absorption rate is set using:
set absrate
where is the number of neutrons absorbed per second
Absorption includes all reactions in which the incident neutron is lost, i.e. all capture reactions and fission. The default normalization is absorption rate set to unity. Normalization to
total loss rate is set using:

5.8 Source rate normalization

set lossrate
where is the number of fission neutrons lost per second
Loss rate includes absorption rate and leakage. Normalization to total flux is set using:
set flux
where is the total neutron flux
Normalization to total heating power is set using:
set power

where
is the total heating power (W)
The total heating power includes all heat generated in the system. If the geometry is twodimensional, the value is the linear power in W/cm. The source rate normalization can be
changed during burnup calculation by re-defining the value between burnup intervals. The
first value is used during the first interval, the second during the second interval and so on.
It should be noted that the heating power is not the same thing as the total fission power
(recoverable fission energy production rate), mainly because a significant fraction of heat
is produced in (n,γ) reactions. The direct calculation of this heating power is difficult and
Serpent uses an approximation based on the total fission rate and empirical heating values
directly proportional to fission energy. The heating value for U-235 fission is 202.27 MeV
and the values for other nuclides are scaled according to the ratios of fission Q-values. The
U-235 heating value can also be set manually using:
set U235H
where is the heating value for U-235 (MeV)
Heating values for individual actinides can be overridden using:
set fissh
...
where is the actinide ZAI

is the heating value
Power density, instead of power can be used for source normalization by setting:
set powdens
where is the average power density (kW/g)
The value is the total heating power divided by the total initial mass of fissile isotopes. This
mass is calculated automatically by the code. If the calculation is not possible, the value
must be set manually (see Sec. 8.4.2 on page 112).

5.8 Source rate normalization

IMPORTANT NOTES ON SOURCE RATE NORMALIZATION:
1. The source rate normalization affects only integral reaction rates encountered, for example, in detector calculation. Homogenized group constants are unaffected since the
normalization cancels out.
2. The default normalization is unit loss rate. It should be noted that the value generally
differs from source and generation rates due to neutron-producing reactions.
3. If the geometry is two-dimensional, the values are divided by unit length. The total
heating power (W), for example, becomes the linear power (W/cm).
4. Power density is given in units of kW/g, not W/g used in several other codes. The
typical order of magnitude for this parameter in LWR calculations is 20E-3 ... 50E-3.
SEE ALSO:
1. Definition of irradiation history (Sec. 8.3 on page 110)
2. Discussion on source normalization in Ref. [15] (Sec. 9.4 on page 169).
3. Additional options for source rate normalization in burnup calculation (Sec. 8.4.2 on
page 112).

Figure 5.1: Symmetry options.

5.9 Group constant generation

Group constant generation

The universes in which the group constants are calculated can be set by:
set gcu ...
where are the universe numbers
The homogenization is carried out in the given universes and all higher universes accessed
from lattices and filled cells. The results are printed in the output file (see Sec. 6.1 on page 77)
using a different run index for each universe in the list. The default is = 0, i.e. a
single universe that covers the entire geometry. It should be noted that the universe options
affect only some of the output parameters, mainly the homogenized group constants. The
methodology was included in code version 1.1.5 and is still under development.
The statistical error in assembly discontinuity factors can be reduced by taking advantage of
the symmetry of the geometry. The symmetry option is set by:
set sym
where is the symmetry option
The available symmetries are illustrated in Figure 5.1. Options 2, 4 and 8 are used with
square lattice geometries and options 6 and 12 with hexagonal geometries. Default option is
0 (no symmetry).
All group constants are generated using the same few-energy group structure. The default
structure consists of two energy groups: fast group above 0.625 eV and thermal group below
that. This can be overridden by setting the group boundaries manually:
set nfg [ ...]
where
...

is the number of energy groups
are the group boundaries (in MeV)

The boundaries are listed in ascending order without the upper and lower limits and the
number must be consistent with the number of given values ( - 1 values for
groups).
When it comes to multiplying scattering reactions, such as (n,2n), (n,3n) or (n, 2nα), there
is some ambiguity in the way group-to-group scattering matrices and removal cross sections
are defined and used in nodal diffusion codes.
The first option is to reflect only the scattering rate, i.e. to disregard the number of neutrons
produced in each reaction. In this case, the sum of each matrix column equals the group-wise

5.9 Group constant generation

total scattering cross section:
G
X

Σs,g→h = Σs,g = Σela,g + Σinl,g + Σ2n,g + Σ3n,g + · · · = Σtot,g − Σcapt,g − Σfiss,g .

The second option is to include neutron production in the cross sections, so that the product of group flux and the corresponding group-transfer cross section yields the rate at which
neutrons enter group h from scattering reactions in group g, taking into account the multiplication in (n,xn) reactions. The summation to total scattering cross section no longer
holds.
Serpent versions from 1.1.15 on calculate both matrixes (see Sec. 6.1.23). The definition of
the scattering matrix also affects the removal cross section:
Σrem,g = Σtot,g − Σs,g→g ,
and the whether production of secondary neutrons is included or not is selected by:
set remxs
where is the scattering matrix option (0 = include only scattering rate,
1 = include also production)
The option is available from version 1.1.15 on. The methods used in previous versions
correspond to option 0. Option 1 is currently used as the default.
IMPORTANT NOTES ON GROUP CONSTANT GENERATION:
1. The methodology has been thoroughly tested only in cases where group constants are
homogenized over the entire geometry. The calculation may produce incorrect results
for diffusion coefficients and assembly discontinuity factors if the homogenization is
restricted to a higher universe.
2. The list of universes given after the gcu option is exclusive. If a collision point is
located in several universes in the list, only the highest universe is scored.
3. The use of the symmetry option will lead to erroneous results if the geometry is not
truly symmetric.
4. The listed energy values cover only the group boundaries between the minimum and
maximum energy of the cross section data. The absolute boundary values are defined
in the reconstruction of the master energy grid.
5. The energy groups are indexed in increasing lethargy (decreasing energy) (1 = highest
group, = lowest group).
SEE ALSO:

5.10 Full-core power distributions

1. Setting the master energy grid (Sec. 5.3 on page 55)
2. Group constant output (Sec. 6.1 on page 77)

Full-core power distributions

Serpent can calculate assembly or pin-wise power distributions in full-core simulations. This
option is set by:
set cpd [ ]
where

is the number of levels included
is the number axial bins
is the lower axial boundary
is the upper axial boundary

The level argument determines whether the calculation is carried out at assembly only (1) or
both assembly and pin-levels (2). The axial variables determine the number and location of
bins in the z-direction.
The code calculates integral fission power inside nested lattice structures (core and assembly
lattices). The output data is printed in a separate file named “_core.png”,
where is the name of the input file and is the burnup step.
IMPORTANT NOTES ON FULL-CORE POWER DISTRIBUTIONS:
1. This is an experimental feature, available from version 1.1.8 on. The routine has not
been thoroughly tested. The results may not be considered reliable, especially when
used in combination with the the track-length estimator option.
2. When used in full 3D mode with axial binning, the routine produces very large output
files.
SEE ALSO:
1. The track-length estimator option (Sec. 5.18 on page 74)

Delta-tracking options

The Woodcock delta-tracking tracking method used by Serpent loses its efficiency in the
presence of localized heavy absorbers, such as control rods or burnable absorber pins. To
overcome this problem, the code switches to the conventional surface-to-surface ray-tracing

5.11 Delta-tracking options

when the probability of sampling a physical collision falls below a user-defined threshold.
The value is set by:
set dt
where

is the delta-tracking threshold value

This parameter determines the probability limit below which the delta-tracking method is
used (0 = never, 1 = always).4
Finding the optimal value for the threshold parameter can only be accomplished by trial and
error. The default value is 0.9, which seems to work well for most cases.
Serpent also has a built-in optimization routine that tries to find the best value for the cut-off
criterion. From version 1.1.9 on the optimization handles each material separately, which has
shown to improve the efficiency at least in some complicated HTGR full-core geometries.
The optimization is switched on by giving a negative threshold value. This value also serves
as the initial guess, so = -0.9 is the recommended choice for optimization.
The use of delta-tracking can be blocked in given materials by setting:
set blockdt ...
where ...

are the materials where delta-tracking
will not be used

The tracking routine in serpent selects between surface and delta-tracking, based on the cutoff criterion described above. Some geometries may run faster, however, if surface tracking
is always used in very large material regions comprised of simple cells. A good example of
such region is the outer reflector in a full-core geometry. It should be noted, however, that
this option may also impair the efficiency if not properly used.
IMPORTANT NOTES ON DELTA-TRACKING:
1. The cut-off value is set to 0.9 by default in code version 1.1.1 and later. Earlier versions
use the optimization by default. The optimization routine was changed in update 1.1.9
to handle each material separately.
2. The optimization has not been thoroughly tested and the methodology is not guaranteed to result in the optimal threshold value in terms of CPU time.
3. The code should always yield consistent results with and without delta-tracking. If any
discrepancies are observed, please report them by e-mail to Jaakko.Leppanen@vtt.fi
4. The block option is available from version 1.1.8 on.
4

The delta-tracking method is essentially a rejection probability sampling technique, and the threshold
parameter determines the highest rejection probability at which the method is used. If the probability is higher
than the threshold value, the code switches to the conventional ray-tracing method.

5.12 Cross section data plotter

SEE ALSO:
1. Description of the basic Woodcock delta-tracking method in Ref. [15] (Sec. 5.3.3 on
page 100).
2. Description of the extended delta-tracking method used in PSG in Ref. [15] (Sec. 8.3.1
on page 149). NOTE: The described methods are partially outdated.
3. A more recent description of the method is found in Ref. [18].

Cross section data plotter

Serpent has the option to plot all cross sections in a matlab m-file format. The cross section
data plotter is activated using:
set xsplot [ ]
where
is the number of energy points in plot
is the lower limit of the energy grid (MeV)
is the upper limit of the energy grid (MeV)
The energy grid used for the plot is uniform with respect to the lethargy variable. The plotter
produces a file “_xs.png”, where is the name of the input file
and is the burnup step. The file contains the energy grid vector, isotopic reaction cross
sections, material total cross sections and fission nubars.

Fission source entropy

The fission source entropy for convergence studies is calculated by default and the total
entropy is divided in x-, y- and z-components. The entropy mesh is set by:
set entr [ ]
where

is the number of x bins
is the number of y bins
is the number of z bins
is the minimum x-coordinate in mesh
is the maximum x-coordinate in mesh
is the minimum y-coordinate in mesh
is the maximum y-coordinate in mesh
is the minimum z-coordinate in mesh
is the maximum z-coordinate in mesh

5.14 Soluble absorber

The source entropies are written in the history output file as function of criticality cycle.
SEE ALSO:
1. History output (Sec. 6.2 on page 94)
2. Discussion on fission source entropy in Ref. [14].

Soluble absorber

Materials with soluble absorber, most commonly boron in light water, can be defined by
mixing two material compositions. This is considerably simpler than explicitly listing the
associated isotopic fractions. The soluble absorber is defined using:
set abs ...
where

is the soluble absorber material name
is the absorber concentration
are the materials where the absorber is dissolved

The code mixes material “” into materials “”, “” ... in concentration defined by “”. Positive concentrations refer to atomic fractions and negative
concentrations to mass fractions. A simple example is given in the VVER-440 calculation
case in Sec. 11.1.1 on page 134.
The absorber concentration can be changed during burnup calculation by re-defining the
value between burnup intervals. The first value is used during the first interval, the second
during the second interval and so on.
IMPORTANT NOTES ON SOLUBLE ABSORBER:
1. If soluble absorber is used with multiple materials, all must share the same isotopic
composition.
2. Only the total absorption channel of the absorber material is used and fission, scattering
and all the other reaction modes are discarded. This is a good approximation if the
concentration is low and the material is high-absorbing. The maximum concentration
for natural boron in water is around few-thousand ppm per weight. If the concentration
is higher, it is better to determine the isotopic composition explicitly.
3. The methodology is available from code version 1.0.2 on.
SEE ALSO:
1. Definition of irradiation history (Sec. 8.3 on page 110)

keff can be iterated to a desired value by allowing the variation in some geometry, material
or physics variable. Iteration is defined by:
set iter [ ]
where

is the iteration mode
is the target keff
is the leakage spectrum mode (B1 -iteration only)
is number of energy bins in the spectrum (B1 -iteration only)

The iteration modes are:
– Iteration of soluble absorber concentration, = “abs”
– α-eigenvalue calculation, = “alpha”
– Iteration of albedo boundary condition, = “albedo”
– B1 iteration, = “B1”
The soluble absorber iteration works by varying the concentration of soluble absorber (see
Sec. 5.14) to yield the desired keff .
The α-eigenvalue mode is a standard transport technique that allows time-absorption or
-multiplication to balance neutron source and loss rates. The “cross section” for the reaction is equal to the α-eigenvalue divided by neutron speed. The calculation is basically
equivalent with a time-dependent simulation.
The albedo boundary condition iteration is an attempt to simulate the effects of neutron
leakage in an infinite-lattice geometry. The method works by sampling leakage (k > 1) or
multiplication reactions (k < 1) each time a neutron crosses a repeated or periodic boundary. It should be noted that this method is highly experimental, and does not have physical
foundation similar to deterministic leakage models.
The second experimental leakage model is the B1 iteration. The method works similar to
the α-eigenvalue simulation: leakage absorption or multiplication reactions are introduced
to balance neutron source and loss rates. The cross section for the reaction is equal to
the B1 -factor multiplied by the leakage spectrum, given by the energy-dependence of the
volume-integrated diffusion coefficient. Since the diffusion coefficient cannot be defined
as a continuous-energy parameter, the code calculates a fine-group spectrum using an estimate based on the diffusion area and the removal cross section ( = 1) or the P1approximation ( = 2). The energy variable is divided into equal lethargywidth bins (default = 500).
IMPORTANT NOTES ON ITERATION:

5.16 Fundamental mode calculation

1. When iteration is used in burnup calculation mode, the procedure is repeated independently for each burnup step.
2. Soluble absorber must be defined in the absorber iteration mode.
3. The α-eigenvalue calculation, albedo iteration and B1 mode are available from update
1.1.5 on.
4. The albedo- and B1 -iteration modes are experimental, rather than standard Monte
Carlo techniques. The theory is not on a particularly solid foundation and the results
are generally not satisfactory when compared to deterministic calculations.
5. The B1 iteration mode must not be confused with the fundamental mode calculation,
discussed in Section 5.16.
6. The α-eigenvalue simulation is a widely-used method, but the implementation in Serpent has not been validated. The mode does not account for delayed neutron emission.
7. Some of the keff estimates are different from a standard calculation, depending on the
iteration mode used.
SEE ALSO:
1. Definition of soluble absorber (Sec. 5.14 on page 69)
2. Ref. [19] and Sec. 5.5.2 in Ref. [15] for discussion on the α-eigenvalue method.
3. Diffusion coefficients in output (Sec. 6.1.27 and 6.1.29).
4. Discussion on neutron leakage models in Monte Carlo calculation in Sec. 9.5 in Ref. [15]

Fundamental mode calculation

The fundamental mode calculation can be considered as an intermediate solution to the criticality spectrum problem, until the development of a valid Monte Carlo based leakage correction. The calculation consists of two stages. First, the Monte Carlo simulation is run to
produce homogenized micro-group cross sections for B1 equations. The solution of these
equations yields the criticality spectrum, which is used to re-homogenize the cross sections.
The syntax is:
set fum
where is the micro-group structure used for the calculation
The energy grid determines the micro-group structure used to form the B1 equations and it
is set up using the “ene” parameter (see Sec. 7.1.2). The method produces a separate set of

5.17 Equilibrium xenon calculation

output parameters (see Sec. 6.1.30) and does not affect the values of other group constants.
The energy group boundaries in the few-group structure must match the boundaries in the
micro-group structure.
IMPORTANT NOTES ON FUNDAMENTAL MODE CALCULATION:
1. The fundamental mode calculation is available from version 1.1.14 on. The spectrum
correction affects only a set of separately produced few-group constants. Extending
the correction to burnup calculation is under development.
2. The group boundaries in the few-group structure must match the boundaries in the
micro-group structure.
3. Relative statistical errors are not included in the results.
4. The fundamental mode calculation must not be confused with the experimental B1
iteration, discussed in Section 5.15.
SEE ALSO:
1. Definition of energy grids (Sec. 7.1.2 on page 99).
2. Output parameters for fundamental mode calculation (Sec. 6.1.30 on page 92).
3. Definition of the few-group structure (Sec. 5.9 on page 64).

Equilibrium xenon calculation

Serpent can iterate the concentration of fission product poison Xe-135 to an equilibrium
value in transport or burnup calculation. The equilibrium xenon calculation is set by:
set xenon [ ...]
where
...

is the calculation mode (0 = off, 1 = on)
are the materials involved in the calculation

The mode option is followed by a list of materials for which the calculation is turned on
or off. If no list is given, the option affects all fissile materials. Each material is handled
separately.
The calculation is based on the production rates of Xe-135 and its precursors (I-135, Te-135,
Sb-135 and Sn-135), the absorption rate of Xe-135 and the radioactive decay of the isotopes.
The production and absorption rates are normalized to source rate. The decay and fission
yield data are read from external libraries, similar to a burnup calculation.
IMPORTANT NOTES ON EQUILIBRIUM XENON CALCULATION:

5.18 Miscellaneous parameters

1. The equilibrium concentration depends on source rate normalization.
2. When used in the burnup calculation mode, the concentration of Xe-135 is handled
separate from the other nuclides. The equilibrium concentration is copied in the depletion output.
3. The capability was included in code version 1.1.9 and currently it may not work with
unresolved resonance probability table sampling, soluble absorbers or k-eff iteration.
SEE ALSO:
1. Source rate normalization (Sec. 5.8 on page 61)
2. Setting the decay and fission yield library file paths (Sec. 8.4 on page 111)

Miscellaneous parameters

A title string for the calculation can be set using
set title ""
where is the title string
This text string is reproduced in the output files together with date and time and version
information.
The Monte Carlo simulation uses a sequence of random numbers, generated from an initial
seed value. This seed is by default taken from system time. The calculation can be reproduced using the “replay” command line option, which forces the code to use the same seed
as in the previous run. The seed value can also be set manually using:
set seed
where is the seed value (a large integer)
Temperatures used in the free-gas model for elastic scattering are read from the ACE format data. The free-gas temperatures in cells can be overridden by defining a list of cell
temperatures:
set ctmp ...
where

are the cell names
are the temperatures

User-defined variables can be set up for labeling different runs. The syntax is:

5.18 Miscellaneous parameters

set var
where
is the variable name
is the value
The variable name and value are printed in the main output (see Sec. 6.1 on page 77). The
type (numeric or string) is identified from the value.
The use of track-length flux estimate can be forced in place of the collision estimator using:
set tle
where is the option (1 = use tle, 0 = use cfe)
If the track-length estimator is used, delta-tracking is switched off.
By default, Serpent uses various pre-calculated summation cross sections for each material
to speed-up the transport simulation. This increases the overall memory demand per material, which may become a limiting factor in burnup calculation. To reduce the demand, the
calculation can be switched off using:
set sumxs
where is the option (1 = use pre-calculated cross sections, 0 = calculate
cross sections on-the-fly)
It should be noted that switching off the option results in an increase in the overall calculation
time. The option is available from version 1.1.13 on.
The emission of delayed neutrons can be swithed on and off using:
set delnu
where is the option (1 = emission on, 0 = emission off)
This option was added in version 1.1.16. Delayed neutron emission is on by default in
criticality source problems and off in external source problems.
Calculation of fission product poison cross section (production of I-135, Xe-135, Pm-149
and Sm-149 and absorption of Xe-135 and Sm-149) can be switched on and off by setting:
set poi ""
where is the option (0 = off, 1 = on)
Calculation is off by default. Switching the mode on requires setting the file path for fission
yield data (see Sec. 8.4.1). This feature is available from version 1.1.17 on.
SEE ALSO:

5.18 Miscellaneous parameters

1. Running the code in replay mode (Sec. 1.2 on page 9)
2. Main output file (Sec. 6.1 on page 77)

5.18 Miscellaneous parameters

Table 5.1: List of parameters and options.
Option
pop
nbuf
egrid
dix
acelib
ures
dbrc
bc
usym
genrate
srcrate
fissrate
absrate
lossrate
flux
power
powdens
U235H
fissh
gcu
sym
nfg
remxs
cpd
dt
blockdt
xsplot
entr
abs
iter
fum
xenon
title
seed
ctmp
var
tle
sumxs
delnu
poi

(3-4)
(1)
(1-3)
(1)
(1)
(1-N)
(3-N)
(1)
(2-4)
(1)
(1)
(1)
(1)
(1)
(1)
(1)
(1)
(1)
(1-N)
(1)
(1)
(1-N)
(1)
(1)
(1)
(1)
(1-4)
(1-9)
(3-N)
(2)
(2)
(1-N)
(1)
(1)
(1)
(1)
(1)
(1)
(1)
(1)

Description
population size and number of cycles
source buffer
energy grid reconstruction
double indexing of energy grids
file path for xs library directory file
probability table treatment for ures data
DBRC correction for scattering kernel
boundary conditions
universe symmetry
source normalisation to generation rate
source normalisation to source rate
source normalisation to fission rate
source normalisation to absorption rate
source normalisation to loss rate
source normalisation to total flux
source normalisation to total heating power
source normalisation to power density
heating value for U-235 fission
fission heating values for individual actinides
universe for homogenization
symmetry option
few-group structure for homogenization
scattering matrix used with removal cross section
full-core power distributions
delta-tracking threshold
delta-tracking block
cross section data plot file
parameters for source entropy calculation
soluble absorber
keff iterations
fundamental mode calculation
equilibrium Xe-135 calculation
case title
random number seed value
override cell temperatures
user-defined variable
track-length estimate of neutron flux
use pre-calculated summation cross sections
switch delayed neutron emission on and off 5.18
fission product poison cross sections 5.18

Section
5.2
5.2
5.3
5.3
5.4
5.5
5.6
5.7
5.7
5.8
5.8
5.8
5.8
5.8
5.8
5.8
5.8
5.8
5.8
5.9
5.9
5.9
5.9
5.10
5.11
5.11
5.12
5.13
5.14
5.15
5.16
5.17
5.18
5.18
5.18
5.18
5.18
5.18
74
74

Page
53
53
55
56
57
58
59
59
60
61
61
61
61
62
62
62
62
62
62
64
64
64
65
66
67
67
68
68
69
70
71
72
73
73
73
74
74
74

Chapter 6
Output
6.1

Main output file

The main output file contains all results calculated by default during the transport cycle.
User-defined detectors produce another file, described in Section 7.2 on Page 105. Inventory
data in burnup calculation is discussed in Section 8.5 on Page 116.
The file is named “_res.m”, where “” is the name of the input file.
The data is written in matlab m-file format to simplify the simultaneous post-processing of
several files. Each parameter is read to a variable (scalar or vector) and a run index “idx”
is assigned to each file. Each time a new file is read, the index is first increased by, 1 so that
the new data is placed on the next line in the result matrix. The following Octave example
illustrates the procedure:1
octave:1> idx = 0
idx = 0
octave:2> run1_res;
octave:3> FISSXS
FISSXS =
0.0160550 0.0005253 0.0033174 0.0006745 0.0863956 0.0005001
octave:4> run2_res;
octave:5> FISSXS
FISSXS =
0.0160550 0.0005253 0.0033174 0.0006745 0.0863956 0.0005001
0.0158454 0.0005277 0.0032059 0.0006817 0.0833996 0.0005078
1

GNU Octave is a Matlab-compatible open-source language for numerical computations.

6.1 Main output file

octave:6> run3_res;
octave:7> FISSXS
FISSXS =
0.0160550 0.0005253 0.0033174 0.0006745 0.0863956 0.0005001
0.0158454 0.0005277 0.0032059 0.0006817 0.0833996 0.0005078
0.0119486 0.0005737 0.0031909 0.0005741 0.0694696 0.0005300

Three input files: “run1_res.m”, “run2_res.m” and “run3_res.m” are read and the data from
each file is placed on a different row in the variables. Variable “FISSXS” is the homogenized fission cross section, calculated in this case using a two-energy group structure. The
first two columns are the total (one-group) value and the associated relative statistical error,
respectively. The following four columns contain the same data for the two energy groups in
ascending order.
Output data in burnup calculation is written in a single file. The run index is updated for each
burnup step. The variables in the main output file are listed in the following.

Version, title and date

Parameter
VERSION
TITLE
DATE

Description
Code version used in calculation
Case title
Date and time at the beginning

Parameter
POP
CYCLES
SKIP
SRC_NORM_MODE

SEED
MPI_TASKS
MPI_MODE
DEBUG
CPU_TYPE
CPU_MHZ
AVAIL_MEM

Description
Number of source neutrons per cycle
Number of active cycles
Number of inactive cycles
Fission source normalization mode (1 = preserve size, 2 = preserve weight)
Random number seed
Number of MPI taks in parallel calculation
Results collection in MPI mode
Debug mode flag (1 = yes, 0 = no)
CPU type
CPU MHz
Available memory in Mb

6.1 Main output file

1. In parallel calculation mode, the number of source neutrons per cycle is the number
for each parallel task.
2. CPU type and MHz are read from /proc/cpuinfo and available memory from /proc/meminfo.

Parameter
XS_DATA_FILE_PATH
DECAY_DATA_FILE_PATH
NFY_DATA_FILE_PATH

Description
Cross section directory file path
Decay data file path
Fission yield data file path

NOTES:
1. Only the first given xs directory file path is printed

Delta-tracking parameters

Parameter
DT_THRESH
DT_FRAC

DT_EFF
MIN_MACROXS

Description
Probability thresold for using delta-tracking
Fraction of path lengths sampled using deltatracking
Efficiency of DT rejection algorithm
Minimum macroscopic cross section for sampling the collision distance

6.1 Main output file

Parameter
TOT_CPU_TIME
RUNNING_TIME
CPU_USAGE

TRANSPORT_CYCLE_TIME
BURNUP_CYCLE_TIME

CYCLE_IDX
SOURCE_NEUTRONS
MEAN_POP_SIZE
MEMSIZE
SIMULATION_COMPLETED

Description
Total CPU time
Cumulative total running time (wall-clock)
CPU usage (ratio of CPU time to wall-clock
time)
Total initialization time before transport or
burnup cycle
Cumulative transport cycle running time
Cumulative time used for solving the depletion equations
Cumulative time used for data processing between transport cycles
Current cycle index
Number of simulated source neutrons
Mean population size
Size of allocated memory block in megabytes
Flag to idicate that all neutron histories are run
(1 = yes, 0 = no)

NOTES:
1. The total RUNNING_TIME is the sum of INIT_TIME, PROCES_TIME, TRANSPORT_CYCLE_TIME and BURNUP_CYCLE_TIME.

Energy grid parameters

Parameter
ERG_EMIN
ERG_EMAX
ERG_TOL
ERG_NE
ERG_NE_INI
ERG_NE_IMP
ERG_DIX
USE_DBRC

Values
1
1
1
1
1
1
1
1

Description
Minimum energy in unionized grid (MeV)
Maximum energy in unionized grid (MeV)
Fractional grid reconstruction tolerance
Number of grid points
Number of grid points before thinning
Number of important grid points
Double indexed energy grids (1 = yes, 0 = no)
Doppler-broadening rejection correction (1 =
yes, 0 = no)

6.1 Main output file

Unresolved resonance data

Parameter
URES_MODE
URES_DILU_CUT
URES_EMIN

URES_AVAIL
URES_USED

Description
Probability table sampling mode
Infinite dilution cut-off
Minimum energy for unresolved resonance
probability table data (MeV)
Maximum energy for unresolved resonance
probability table data (MeV)
Number of isotopes with ures data available
Number of isotopes with ures data used

Nuclides and reaction channels

Parameter
TOT_ISOTOPES
TOT_TRANSPORT_ISOTOPES

TOT_DECAY_ISOTOPES

TOT_REA_CHANNELS
TOT_TRANSMU_REA

Description
Total number of isotopes
Total number of isotopes with cross section
data
Total number of isotopes without cross section data
Total number of reaction channel
Total number of transmutation reactions

NOTES:
1. TOT_REA_CHANNELS includes neutron reactions only, no decay.

6.1 Main output file

Reaction mode counters

Parameter
COLLISIONS
FISSION_FRACTION
CAPTURE_FRACTION
ELASTIC_FRACTION
INELASTIC_FRACTION
ALPHA_FRACTION

Values
1
1
1
1
1
1

BOUND_SCATTERING_FRACTION
NXN_FRACTION
UNKNOWN_FRACTION
VIRTUAL_FRACTION
FREEGAS_FRACTION

TOTAL_ELASTIC_FRACTION

FISSILE_FISSION_FRACTION
LEAKAGE_REACTIONS
REA_SAMPLING_EFF

Description
Total number of collisions
Fraction of fission reactions
Fraction of capture reactions
Fraction of elastic scattering reactions
Fraction of inelastic scattering reactions
Fraction of time-absorption or -multiplication
reactions in α-eigenvalue calculation mode
Fraction of bound atom scattering reactions
Fraction of (n,xn) reactions
Fraction of unsampled reactions
Fraction of virtual collisions
Fraction of free-gas elastic scattering reactions
Fraction of free and bound atom elastic scattering reactions
Fraction of fission reactions in fissile isotopes
Number of leakage reactions
Reaction mode sampling efficiency

NOTES:
1. Leakage in B1 and albedo iteration modes is counted in LEAKAGE_REACTIONS

Slowing-down and thermalization

Parameter
COL_SLOW

COL_TOT
SLOW_TIME
THERM_TIME
SLOW_DIST
THERM_DIST
THERM_FRAC

Description
Average number of collisions before thermalization
Average number of collisions after thermalization
Average total number of collisions
Average slowing-down time
Average thermal life time
Average slowing-down distance
Average thermal migration distance
Average fraction of neutrons reaching thermalization

6.1 Main output file

Parameters for burnup calculation

Parameter
BURN_MODE
BURN_STEP
BURN_TOT_STEPS
BURNUP
BURN_DAYS
ENERGY_OUTPUT
DEP_TTA_CUTOFF
DEP_STABILITY_CUTOFF
DEP_FP_YIELD_CUTOFF
DEP_XS_FRAC_CUTOFF
DEP_XS_ENERGY_CUTOFF
BURN_MATERIALS
FP_NUCLIDES_INCLUDED

Values
1
1
1
1
1
1
1
1
1
1
1
1
1

FP_NUCLIDES_AVAILABLE

TOT_ACTIVITY
TOT_DECAY_HEAT
TOT_SF_RATE
ACTINIDE_ACTIVITY
ACTINIDE_DECAY_HEAT
FISSION_PRODUCT_ACTIVITY
FISSION_PRODUCT_DECAY_HEAT

DH_N_PREC
DH_PREC_BOUNDS
DH_PREC_LAMBDA
DH_PREC_HEAT

Description
Burnup mode (1 = TTA, 2 = CRAM)
Burnup step index
Total number of burnup steps
Burnup at current step (in MWd/kgU)
Number of burn days at current step
Total cumulative energy output (in J)
TTA cut-off value
Stability cut-off value
Fission product yield cut-off value
Depletion fraction cut-off value
Depletion reaction energy cut-off value
Number of depleted materials
Total number of fission product nuclides included in the calculation
Total number of fission products available before yield cut-off
Total activity at current step
Total decay heat at current step (in W)
Total spontaneous fission rate
Actinide activity at current step
Actinide decay heat at current step (in W)
Fission product activity at current step
Fission product decay heat at current step
(in W)
Number of decay heat precursor groups
Decay heat precursor group boundaries
Decay heat group-wise decay constants
Decay heat group-wise heat production (in W)

NOTES:
1. Precursor-group wise decay heat production is available from version 1.1.17 on. The
option for setting the group boundaries is described in Sec. 8.4.8.

Fission source entropies

Parameter
ENTROPY_X
ENTROPY_Y
ENTROPY_Z
ENTROPY_TOT

Description
X-component of fission source entropy
Y-component of fission source entropy
Z-component of fission source entropy
Total fission source entropy

6.1 Main output file

Fission source center

Parameter
SOURCE_X0
SOURCE_Y0
SOURCE_Z0

Description
X-coordinate of fission source center
Y-coordinate of fission source center
Z-coordinate of fission source center

Description
Atomic fraction of soluble absorber
Mass fraction of soluble absorber

Soluble absorber

Parameter
SOLU_ABS_AFRAC
SOLU_ABS_MFRAC

NOTES:
1. The values are printed only if soluble absorber defined (see Sec. 5.14 on page 69).

Parameter
ITER_MODE
ITER_KEFF
ITER_VAR
B1_MODE

B1_ERG
B1_SPECTR

Description
Iteration mode
Target keff for iteration
Iteration variable
Method used for calculating leakage spectrum
in B1 iteration mode
Number of equal lethargy-width bins in the
leakage spectrum
Energy bin limits for the leakage spectrum
Cycle-averaged leakage spectrum

NOTES:
1. The values are printed only if iteration is in use (see Sec. 5.15 on page 70).

Equilibrium Xe-135 calculation

Parameter
XE135_EQUIL_CONC
I135_EQUIL_CONC

Description
Equilibrium Xe-135 concentration
Equilibrium I-135 concentration

6.1 Main output file

1. The values are printed only if xenon iteration is in use (see Sec. 5.17 on page 72).
2. The concentrations are averaged over all regions involved in the iteration

Criticality eigenvalues

Parameter
ANA_KEFF
IMP_KEFF
COL_KEFF
ABS_KEFF
ABS_KINF
ABS_GC_KEFF

Values
2
2
2
2
2
2

IMPL_ALPHA_EIG
FIXED_ALPHA_EIG

Description
Analog estimate of keff
Implicit estimate of keff
Collision estimate of keff
Absorption estimate of keff
Absorption estimate of k∞
Absorption estimate of keff in group constant
generation universe
Absorption estimate of k∞ in group constant
generation universe
External source multiplication factor in 5 generations
Implicit estimate of α-eigenvalue
Fixed or iterated value in α-eigenvalue calculation
Fixed or iterated value for albedo

NOTES:
1. The absorption estimate of keff is currently used as the implicit estimate.
2. External source multiplication factor is not printed in criticality source mode.

6.1 Main output file

Parameter
TOT_POWER
TOT_GENRATE
TOT_FISSRATE
TOT_ABSRATE
TOT_LEAKRATE
TOT_LOSSRATE
TOT_SRCRATE
TOT_FLUX
TOT_RR
TOT_SOLU_ABSRATE
TOT_XE135_ABSRATE
TOT_FMASS
TOT_POWDENS
BURN_POWER
BURN_GENRATE
BURN_FISSRATE
BURN_ABSRATE
BURN_FLUX
BURN_FMASS
BURN_POWDENS
BURN_VOLUME

Values
2
2
2
2
2
2
2
2
2
2
2
1
2
2
2
2
2
2
1
2
1

Description
Total power
Total neutron generation rate
Total fission rate
Total absorption rate
Total leakage rate
Total loss rate
Total source rate
Total flux
Total reaction rate
Total absorption rate in soluble absorber
Total absorption rate in Xe-135
Total fissile mass
Total power density
Power in burnable materials
Neutron generation rate in burnable materials
Fission rate in burnable materials
Absorption rate in burnable materials
Flux in burnable materials
Fissile mass in burnable materials
Power density in burnable materials
Total combined volume of all burnable materials

NOTES:
1. Normalization is set by the user (see Sec. 5.8 on page 61).
2. By default, the loss rate is normalized to unity.
3. Total power density is printed only if total fissile mass is calculated or given by the
user.
4. Parameters in burnable materials are printed only in burnup calculation mode.
5. Total (external) source rate is not printed in criticality source mode.
6. Xe-135 absorption rate is printed only in equilibrium xenon mode.
7. All flux values are divided by volume

6.1 Main output file

Point-kinetic parameters

Parameter
ANA_PROMPT_LIFETIME
IMPL_PROMPT_LIFETIME
ANA_REPROD_TIME
IMPL_REPROD_TIME
DELAYED_EMTIME

Values
2
2
2
2
2

Description
Analog estimate of prompt neutron lifetime
Implicit estimate of prompt neutron lifetime
Analog estimate of neutron reproduction time
Implicit estimate of neutron reproduction time
Mean delayed neutron emission time

NOTES:
1. The neutron reproduction time is also commonly known as the “neutron generation
time”.
2. The analog estimates and delayed neutron emission time are calculated for the entire
geometry. The implicit estimates are calculated in the universe set by the user.

Six-factor formula

Parameter
SIX_FF_ETA

SIX_FF_F
SIX_FF_P
SIX_FF_EPSILON
SIX_FF_LF
SIX_FF_LT
SIX_FF_KINF
SIX_FF_KEFF

Description
Average number of neutrons emitted per thermal neutron absorbed in fuel
Thermal utilization factor
Resonance escape probability
Fast fission factor
Fast non-leakage probability
Thermal non-leakage probability
Six-factor k∞ (four-factor keff )
Six-factor keff

NOTES:
1. The parameters are calculated using simple analog estimates and inteded mainly for
the demonstration of basic reactor physics phenomena.

Delayed neutron parameters

Parameter
USE_DELNU
PRECURSOR_GROUPS
BETA_EFF
BETA_ZERO
DECAY_CONSTANT

Values
1
1
2Jd + 2
2Jd + 2
2Jd + 2

Description
Delayed neutron emission (0 = off, 1 = on)
Number of delayed neutron precursor groups
Effective delayed neutron fraction
Physical dealyed neutron fraction
Precursor group-wise decay constants

6.1 Main output file

NOTES:
1. The number of precursor groups Jd depends on data. The usual number is 6 or 8. The
first two entries refer to the total value and the associated relative statistical error.
2. Since different precursor group structures cannot be combined, the number of groups
is fixed to the value used in the first actinide in the input. Delayed neutron emission is
entirely omitted for nuclides using a different group structure.

Parameters for group constant generation

Parameter
GC_UNI
GC_SYM
GC_NE
GC_BOUNDS

Values
1
1
1
G+1

Description
Universe for group constant generation
Symmetry option
Number of energy groups
Group boundaries

Few-group cross sections

Parameter
FLUX
LEAK
TOTXS
FISSXS
CAPTXS
ABSXS
RABSXS
ELAXS
INELAXS
SCATTXS
SCATTPRODXS
N2NXS
REMXS
NUBAR
NSF

Values
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2

Description
Integral flux
Leakage rate
Total cross section
Fission cross section
Capture cross section
Absorption cross section
Reduced absorption cross section
Elastic scattering cross section
Inelastic scattering cross section
Total scattering cross section
Total scattering production cross section
(n,2n) cross section
Group-removal cross section
Average number of emitted fission neutrons
Fission neutron production cross section
(νΣfissg )
Inverse mean neutron speed
Average fission heating value (in MeV)

6.1 Main output file

MAJOR FLAW IN CALCULATION METHODS:
Earlier code versions, including base version 1.1.0, contain a serious flaw in group constant
calculation. The collision flux estimator yields zero values in void regions, resulting in
a systematic over-prediction of the homogenized values. The problem was fixed in code
update 1.1.3.
NOTES:
1. The first two entries are the total (one-group) value and the associated relative statistical error. The remaining 2G entries are few-group values.
2. The normalization of group-flux does not work in the pre-release version 1.0.0 of the
Serpent code (corrected in version 1.0.1). The one-group value should be equal to
variable TOT_FLUX (see Sec. 6.1.18).
3. All cross sections are macroscopic.
4. Capture cross section includes all (n,0n) reactions.
5. Absorption cross section includes capture and fission.
6. Elastic scattering includes thermal bound-atom reactions.
7. Group-removal cross section includes absorption and scattering out of the energy
group. Option to include neutron multiplication was added in version 1.1.15 (see Sec.
5.9).
8. Reduced absorption cross section is defined as absorption minus production in (n,xn)
reactions.

Fission product poison cross sections

Parameter
I135PRODXS
XE135PRODXS
PM149PRODXS
SM149PRODXS
XE135ABSXS
SM149ABSXS

Values
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2

Description
Production cross section for I-135
Production cross section for Xe-135
Production cross section for Pm-149
Production cross section for Sm-149
Absorption cross section of Xe-135
Absorption cross section of Sm-149

NOTES:
1. The option to switch on the calculation of fission product poison cross sections is
described in Sec. 5.18

6.1 Main output file

2. All values are microscopic cross sections
3. Available from version 1.1.17 on

Fission spectra

Parameter
CHI
CHIP
CHID

Values
2G
2G
2G

Description
Energy spectrum of all fission neutrons
Energy spectrum of prompt fission neutrons
Energy spectrum of delayed fission neutrons

Group-transfer probabilities and cross sections

Parameter
GTRANSFP
GTRANSFXS
GPRODP
GPRODXS

Values
2G2
2G2
2G2
2G2

Description
Group-transfer probability matrix
Group-transfer cross section matrix
Group-production probability matrix
Group-production cross section matrix

NOTES:
1. The matrices are given in vector format:
P1→1 P2→1 ... PG→1 P1→2 P2→2 ... PG→2 ...
Each probability and cross section is followed by the associated relative statistical
error. Index for reaction j → i is given by:
n = 2(i − 1)G + 2j − 1
2. The production matrixes include neutron multiplication in (n,xn) reactions.

Diffusion parameters

Parameter
DIFFAREA
DIFFCOEF
TRANSPXS
MUBAR
MAT_BUCKLING
LEAK_DIFFCOEF

Values
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2

Description
Diffusion area
Diffusion coefficient
Transport cross section
Average scattering angle
Material buckling
Diffusion coefficient from leakage mode

6.1 Main output file

1. The first two entries are the total (one-group) value and the associated relative statistical error. The remaining 2G entries are few-group values.
2. The values are based on the analog estimate of group-wise diffusion area. The results
usually differ from the P1 -values below.
3. Leakage diffusion coefficient is defined as buckling divided by leakage, which can be
physical or from a leakage model. The theoretical basis is very questionable.

Pn scattering cross sections

Parameter
SCATT0
SCATT1
SCATT2
SCATT3
SCATT4
SCATT5

Values
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2

Description
P0 scattering cross section
P1 scattering cross section
P2 scattering cross section
P3 scattering cross section
P4 scattering cross section
P5 scattering cross section

NOTES:
1. The first two entries are the total (one-group) value and the associated relative statistical error. The remaining 2G entries are few-group values.

P1 diffusion parameters

Parameter
P1_TRANSPXS
P1_DIFFCOEF
P1_MUBAR

Values
2G + 2
2G + 2
2G + 2

Description
Transport cross section
Diffusion coefficient
Average scattering angle

NOTES:
1. The first two entries are the total (one-group) value and the associated relative statistical error. The remaining 2G entries are few-group values.
2. The values are based on the P1 approximation. The results usually differ from values
calculated using the analog estimate of diffusion area (see above).

6.1 Main output file

B1 fundamental mode calculation

Parameter
B1_KINF
B1_BUCKILNG
B1_FLUX
B1_TOTXSXS
B1_NSF
B1_FISSXS
B1_CHI
B1_ABSXS
B1_RABSXS
B1_REMXS
B1_DIFFCOEF
B1_SCATTXS
B1_SCATTPRODXS
B1_I135PRODXS
B1_XE135PRODXS
B1_PM149PRODXS
B1_SM149PRODXS
B1_XE135ABSXS
B1_SM149ABSXS

Values
1
1
2G + 2
2G + 2
2G + 2
2G + 2
2G
2G + 2
2G + 2
2G + 2
2G + 2
4G2
4G2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2
2G + 2

Description
Iterated multiplication factor
Iterated buckling
B1 integral flux
B1 total cross section
B1 fission neutron production cross section
B1 fission cross section
B1 fission spectrum
B1 absorption cross section
B1 reduced absorption cross section
B1 removal cross section
B1 diffusion coefficient
B1 scattering matrix
B1 scattering production matrix
Production cross section for I-135
Production cross section for Xe-135
Production cross section for Pm-149
Production cross section for Sm-149
Absorption cross section of Xe-135
Absorption cross section of Sm-149

NOTES:
1. B1 fundamental mode calculation is performed after the transport cycle using homogenized multi-group cross sections (see Sec. 5.16).
2. The definition of scattering matrix was changed in version 1.1.15 (see Sec. 5.9).
3. Reduced absorption cross section is defined as absorption minus production in (n,xn)
reactions.
4. Scattering production matrix includes neutron multiplication in (n,xn) reactions.
5. The option to switch on the calculation of fission product poison cross sections is
described in Sec. 5.18
6. The capability is available from version 1.1.14 on. Some parameters were added in
versions 1.1.15 and 1.1.17.

6.1 Main output file

Assembly discontinuity factors

Parameter
ADFS
ADFC

Values
2GNV
2GNV

Description
Surface discontinuity factors
Corner discontinuity factors

NOTES:
1. The assembly discontinuity factors are calculated only for square and hexagonal cylinder boundaries. The ADF surface is the outermost surface in the universe where the
group constants are calculated.
2. The number of vertices NV is 4 for square boundary and 6 for hexagonal boundary.
3. The index for vertice (corner) n and group g is given by:
i = 2(n − 1)G + 2g − 1
4. The methodology is tested only group constant generation is extended over the entire
geometry.
5. For square assemblies the numbering of vertices is: 1 - West, 2 - North, 3 - East, 4 South and for the corners: 1 - South-West, 2 - South-East, 3 - North-East, 4 - NorthWest. Also note that geometry plots are inverted in the north-south direction.

Power distributions in lattices

Parameter
LAT
POWDISTR
FG_POWDISTR

Values
3
2NL
2NL (2G + 1)

Description
Lattice type and size
Power distribution in lattice
Power distribution in lattice divided
into energy groups
Peaking factor in lattice

NOTES:
1. Lattice parameters are calculated for each lattice, regardless of the content. Variable
names include the lattice number “”.
2. For square and hexagonal lattices the type and number of rows and columns is given.
For cluster-type lattices the entries are type, number of rings and total number of elements.
3. The values in the power distribution are given as a single vector. The order is determined by the universe map in the lattice definition.

6.2 History output

4. Peaking factor gives the position and the peak value in the lattice.
5. Energy-group wise power distribution is calculated from version 1.1.17 on.

Chapter 7
Detectors
7.1

Serpent uses the collision estimate of neutron flux for calculating user-defined reaction rates
integrated over space and energy:
Z Z Ei
1
f (r, E)φ(r, E)d3 rdE .
(7.1)
R=
V V Ei+1
The response function f (r, E) and the spatial and energy domains of the integration are set
by the detector parameters.1 The syntax is relatively simple:
det ...
where
...

is the detector name
are the detector parameter sets

The parameters are listed in Table 7.1 and they can be combined in different ways as described in the following subsections. Some parameters produce multiple results and some
may be used several times in the definition. In such a case, the results are divided into a
number of separate bins, depending on the combination.
The integral in Eq. (7.1) is divided by detector volume, which is set to unity by default. This
is because in most cases it is the total reaction rate, not the reaction rate density that is
of interest to the user. The volume can be set manually using the “dv” entry. If a negative
number is entered, the code uses a value calculated by the geometry routine (when available).

To be precise, the integration is also carried over time and space-angle, but user-defined limits can be set
for the spatial and energy variables.

7.1 Detector Input

Table 7.1: Detector parameters.
Param.
dr
dv
dc
du
dm
dl
de
dx
dy
dz
dt
ds

Description
Reaction multiplier
Detector volume
Detector cell
Detector universe
Detector material
Detector lattice
Detector energy grid
Detector mesh
Detector mesh
Detector mesh
Detector type
Surface current detector

Comments
Determines the response function
Used for normalization
Defines the cell where the reactions are scored
Defines the universe where the reactions are scored
Defines the material where the reactions are scored
Defines the lattice where the reactions are scored
Defines the energy bins for the response function
Defines the x-mesh where the reactions are scored
Defines the y-mesh where the reactions are scored
Defines the z-mesh where the reactions are scored
Special detector types
Defines surface for current detector

IMPORTANT NOTES ON THE COLLISION FLUX ESTIMATOR:
1. The Serpent code uses the collision estimate of neutron flux, simply because the tracklength estimate is not available when delta-tracking is used for neutron transport. The
two estimates are equally well-suited for typical reactor lattice calculations, in which
the neutron source is distributed over the entire geometry. The efficiency of the collision estimator becomes poor, however, if reaction rates are calculated inside small or
optically thin volumes located in regions of low collision density. This is why the code
is not the best choice for dosimetry calculations (see Ref. [20]). On the other hand,
the use of the collision estimate requires less computational effort, especially for mesh
detectors, which is directly reflected in the overall calculation time.

Setting the Response Function

The detector response function determines the type of the calculation. In the simplest case,
f = 1, and (7.1) is reduced to the neutron flux integrated over space and energy. If a reaction
cross section is used, the result is the corresponding reaction rate. It should be noted that the
absolute value of the integral depends on source normalization (see Sec. 5.8).
The detector response function is defined by the “dr” entry:
det dr
where is the detector name

is the response function number

is the material name (or “void” for void material)

7.1 Detector Input

Table 7.2: Detector response functions. For a complete list of ENDF reaction MT’s, see
Ref. [6].

Material total reactions

ENDF Reaction modes

MT
0
-1
-2
-3
-5
-6
-7
-8
-9
1
2
16
17
18
19
20
51
52
...
90
91
102
103
104
105
106
107

Reaction mode
None
Total
Total capture
Total elastic
Total (n,2n)
Total fission
Total fission neutron production
Total fission energy deposition
Majorant
Total
Elastic scattering
(n,2n)
(n,3n)
Total fission
First-chance fission
Second-chance fission
Inelastic scattering to 1st excited state
Inelastic scattering to 2nd excited state
Inelastic scattering to 40th excited state
Continuum inelastic scattering
(n,γ)
(n,p)
(n,d)
(n,t)
(n,3 He)
(n,α)

If multiple responses are defined for a detector, an equal number of bins are created for the
results. The response functions are listed in Table 7.2. Negative entries define total reaction
rates related to materials. The total cross section (mt = -1), for example, is calculated from:

Z Z Ei X
1
Σtot,j (r, E)φ(r, E) d3 rdE ,
(7.2)
R=
V V Ei+1 j
where the summation is carried over all nuclides in the material. If the material entry is
set to void, the material at each collision point is used in the calculation. This allows the
integration of reaction rates in volumes extending over several material regions.
Positive response numbers are related to isotopic, rather than material total reaction rates,

7.1 Detector Input

and they correspond to the reaction MT’s used in ENDF format data. The list in Table 7.2 is
not complete and a more detailed description is found in Ref. [6]. The detector material for
an isotopic response function must consist of a single nuclide.
Detector values can be multiplied or divided by other values by setting the detector type to 2
or 3, respectively. The type is then followed by the name of the multiplier or divider detector.
The total number of values must be equal for both detectors or the divider / multiplier detector
single-valued.
EXAMPLES:
% Total flux in material "fuel":
det 1 dm fuel
% Detector materials:
mat U235 1.0 92235.09c 1.0
mat U238 1.0 92238.09c 1.0
% Calculate microscopic fission and capture cross sections of
% U-235 and U-238 by dividing the reaction rate by total flux:
det
det
det
det

fuel
fuel
fuel
fuel

dr 18 U235 dt
dr 102 U235 dt
dr 18 U238 dt
dr 102 U238 dt

IMPORTANT NOTES ON DETECTOR RESPONSE FUNCTIONS:
1. If multiple response functions are defined for a detector, an equal number of bins are
created for the results.
2. Dosimetry cross sections (type 2 or ’y’) can be used with detectors and with detectors
only.
3. The ENDF reaction MT numbers are universal and related to isotopic cross sections.
These reactions may not be used with materials consisting of more than one nuclide.
The result is multiplied by the material atomic density and microscopic reaction rates
can be calculated by setting the density to unity.
4. Some high-energy reaction modes, such as (n,3n), are excluded from the transport simulation. These modes are not available in the detector calculation either. All reaction
modes are included for dosimetry cross sections.
5. The negative MT numbers are specific to Serpent and not universally defined. The
reaction rates are calculated by summing over all nuclides in the material. MCNP also

7.1 Detector Input

uses some code-specific negative reaction MT’s, but the interpretations are slightly
different.
6. The fission energy deposition function defined by mt = -8 yields the total energy absorbed in the system (in J). This is not equivalent with the fission Q-value (see source
normalization in Sec. 5.8).
7. The mt’s 0, -9 and -10 are not material-specific and the entry must be set to void.
8. If the “dr” entry is omitted entirely, the result is the total flux integrated over space and
energy.
SEE ALSO:
1. Dosimetry cross sections (Sec. 1.4.1 on page 11)
2. Source rate normalization (Sec. 5.8 on page 61)

Setting the Energy Domain

The energy boundaries [Ei+1 Ei ] of the integration (7.1) are set by a user-defined energy
grid, linked to the detector by the “dt” entry:
det de
where is the detector name

is the grid name
The same energy grid definition is also used with B1 fundamental mode calculation (See
Sec. 5.16).
The number of energy bins is defined by the grid size. There are four types of energy grids
1. arbitrarily defined
2. equal energy-width bins
3. equal lethargy-width bins
4. predefined energy group structure
The grid definition has three entry formats:

7.1 Detector Input

ene 1 ...
ene
ene 4
where

is the grid name
is the grid type
are the bin boundaries in type 1 grid
is the number of bins in type 2 and 3 grids
is the minimum energy in type 2 and 3 grids
is the maximum energy in type 2 and 3 grids
is the name of a predefined structure

The predefined energy grid names and descriptions are listed in Table 7.3. Bin boundaries
are not listed here, but the values are easily readable in Serpent source file “egroups.c”.
The detector energy grid is often used for calculating spectral quantities. There are three
special detector types for spectral calculations, determined by the “dt” detector type entry:
1. Cumulative spectrum (“dt -1”)
2. Division by energy width (“dt -2”)
3. Division by lethargy width (“dt -3”)
In the default mode, the bin values are independent and undivided.
EXAMPLES:
% Flux per lethargy using energy grid 1:
det 1 de 1 dt -3
% Differential capture, fission and production spectra:
det 2 de 1 dt -2 dr -2 void
det 3 de 1 dt -2 dr -6 void
det 4 de 1 dt -2 dr -7 void
% Integral capture, fission and production spectra:
det 5 de 1 dt -1 dr -2 void
det 6 de 1 dt -1 dr -6 void
det 7 de 1 dt -1 dr -7 void

7.1 Detector Input

Table 7.3: Predefined energy grid types.
Grid name
nj2
nj3
nj4
nj5
nj8
nj9
nj11
nj14
nj16
nj17
nj18
nj19
nj20
nj21
nj22
nj23
wms69
wms172
cas70
cas40
cas25
cas23
cas18
cas16
cas14
cas12
cas9
cas8
cas7
cas4
cas3
cas2
mupo43
scale44
scale238

Description
csewg 239 group structure
lanl 30 group structure
anl 27 group structure
rrd 50 group structure
laser-thermos 35 group structure
epri-cpm 69 group structure
lanl 70 group structure
eurlib 100-group structure
vitamin-e 174-group structure
vitamin-j 175-group structure
xmas 172-group structure
ecco 33-group structure
ecco 1968-group structure
tripoli 315-group structure
xmas lwpc 172-group structure
vit-j lwpc 175-group structure
WIMS 69-group structure (equivalent with nj9)
WIMS 172-group structure
CASMO 70-group structure
CASMO 40-group structure
CASMO 25-group structure
CASMO 23-group structure
CASMO 18-group structure
CASMO 16-group structure
CASMO 14-group structure
CASMO 12-group structure
CASMO 9-group structure
CASMO 8-group structure
CASMO 7-group structure
CASMO 4-group structure
CASMO 3-group structure
CASMO 2-group structure
MUPO 43-group structure
SCALE 44-group structure
SCALE 238-group structure

Setting the Spatial Domain

There are five options for setting the spatial domain of the integration:

7.1 Detector Input

1. By defining the cell where the reaction rates are scored using the “dc” parameter.
2. By defining the universe where the reaction rates are scored using the “du” parameter.
3. By defining the material where the reaction rates are scored using the “dm” parameter.
4. By defining the lattice where the reaction rates are scored using the “dl” parameter.
5. By setting up a one-, two- or three-dimensional mesh using the “dx”, “dy” and “dz”
parameters.
All these options can be used without restrictions in various combinations. It should be
noted, however, that some combinations may result in physically impossible configurations
and produce zero results.

Detector cells, materials and universes
Detector cell, material and universe parameters all work on the same principle: the collision
is scored if it occurs inside the cell, material or universe, respectively. A separate bin is
created for each entry and the combination of different types creates a combination of bins.
The syntax is:
det dc dm du
where

is the detector name
is the detector cell
is the detector material
is the detector universe

Detector cells can be either physical or super-imposed on the geometry. Super-imposed cells
are not used for defining material regions. They must contain void material and the universe
number must be set to a negative value. Universes containing super-imposed cells can be
created for defining complicated geometry regions. These universes are not bound by the
restrictions of physical universes discussed in Section 3.6. Leakage rate can be calculated by
scoring collisions in outside cells.
Fuel pin definitions are geometry macros that are converted into ordinary geometry objects
constructed using cells and surfaces. The cells in fuel pins are named using convention:
nstc
where is the pin (universe) number
is the ring index starting from the innermost region (= 1)
Burnable materials in fuel pins are renamed and divided into a user-defined number of annular depletion zones (see Sec. 8.2 on page 109). The naming convention is:

7.1 Detector Input

pr
where is the original material name

is the pin (universe) number

is the ring index starting from the innermost region (= 1)
EXAMPLES:
% Simple cell, material and universe detectors:
det 1 dc 1
det 2 dm fuel
det 3 du 2

% Score collisions in cell 1
% Score collisions in material "fuel"
% Score collisions in universe 2

% Combined detectors:
det 4 dc 1 dc 2
det 5 du 1 dm H2O

% Two bins: collisions in cells 1 and 2
% Collisions in material "H2O" in universe 2

% Super-imposed cells:
cell 10 -1
cell 11 -1
det 6 dc 10
det 7 du -1

void -1
void 1 -2
% Collisions in super-imposed cell 10
% Collisions in super-imposed universe -1

Lattice detectors
The input format for the lattice detector is:
det dl
where is the detector name

is the detector lattice number
A bin is created for each lattice position. The results can be combined with cell, material and
universe bins. For example, the flux distribution in material “clad” in a fuel pin lattice “10”
can be calculated using:
det 1
dm clad
dl 10

Score in material "clad"
Lattice bins in lat 10

7.1 Detector Input

Mesh detectors
The mesh detector creates a super-imposed uniform square mesh over the geometry. The
mesh structure is given separately in x-, y- and z-directions and the input format for the
x-type is:
det dx
where

is the detector name
is the minimum x-coordinate of the mesh
is the maximum x-coordinate of the mesh
is the number of mesh bins in the x-direction

EXAMPLES:
% One-dimensional mesh (axial power distribution in fuel pin):
det
du
dm
dz

1
1
fuel
0.0 120.0 50

% Score in universe (pin) 1
% Score in material "fuel"
% 50 axial bins between z = 0 and z = 120 cm

% Two-dimensional mesh (total fission rate distribution):
det 2
dr -6 void
dx -225.0 225.0 30
dy -225.0 225.0 30

% Multiply by total fission rate
% 30 bins in x-direction
% 30 bins in y-direction

% Three-dimensional mesh (thermal flux distribution):
ene 1 1 1E-11 0.625E-6

% Detector energy grid (single bin)

det 3
de
1
dx -225.0 225.0 30
dy -225.0 225.0 30
dz
0.0 400.0 10

Use energy
30 bins in
30 bins in
10 bins in

grid 1
x-direction
y-direction
z-direction

Surface Current Detectors

Serpent 1.1.17 and later versions have the capability to calculate neutron current through
surfaces. The syntax for the current detector is:

7.2 Detector output

det ds
where is the detector name
is the surface name

is the direction vector (-1 = inward, 0 = net, 1 = outward)
The surface associated with the detector is assumed to be located relative to the origin of
universe zero, and it may or may not be a part of the geometry definition. The direction
vector determines which surface crossings are included in the result. Inward current has
positive and outward current negative value, respectively. Net current is calculated as the
sum of the two.
The surface current detector was added mainly for the calculation of reflector group constants, and the first implementation had some limitations with respect to boundary conditions (see note below). Reflector geometries typically involve the use of partial boundary
conditions (see Sec. 5.7 on page 59), available from code version 1.1.17 on.
IMPORTANT NOTES ON THE SURFACE CURRENT DETECTOR:
1. The surface current detector in version 1.1.17 cannot cope with some of the coordinate transformations performed when repeated boundary conditions are applied, which
limits its use to geometries with black boundary conditions, or reflected or periodic
boundary condition perpendicular to the detector surface. This limitation was lifted in
update 1.1.18, and the most recent implementation should work in all geometry types.

Detector output

The output from all detectors is printed in matlab m-file format in a single file named
“_det.m”, where “” is the name of the input file and “” is
the burnup step.
The results for each detector are written in a 13-column table, one bin value per row. The
variable is named “DET.m”, where “” is the detector name. The values in
each column are:
1. Value index (total number in “DET_VALS”)
2. Energy bin index (total number in “DET_EBINS”)
3. Universe bin index (total number in “DET_UBINS”)
4. Cell bin index (total number in “DET_CBINS”)
5. Material bin index (total number in “DET_MBINS”)

7.2 Detector output

6. Lattice bin index (total number in “DET_LBINS”)
7. Reaction bin index (total number in “DET_RBINS”)
8. Z-mesh bin index (total number in “DET_ZBINS”)
9. Y-mesh bin index (total number in “DET_YBINS”)
10. X-mesh bin index (total number in “DET_XBINS”)
11. Mean value
12. Relative statistical error
13. Total number of scores
Detector volume is given in variable “DET_VOL”. All results have been divided by
this number.
If an energy bin structure is defined, the corresponding bin boundaries are written in variable
“DETE”. The variable has three columns:
1. Lower energy boundary of bin
2. Upper energy boundary of bin
3. Mean energy of bin
The number of rows is equal to the number of energy bins.
If x-, y- or z-bins are defined, the corresponding bin boundaries are written in variables
“DETX”, “DETY”, “DETZ”, respectively. The variables have three
columns:
1. Coordinate of the lower bin boundary
2. Coordinate of the upper bin boundary
3. Coordinate of bin center
The number of rows is equal to the number of x-, y- or z-bins.
IMPORTANT NOTES ON DETECTOR OUTPUT:
1. Some variables are missing and the names are in lower-case in the pre-release version
1.0.0 of the Serpent code (corrected in version 1.0.1).
2. Detector volume is printed in version 1.1.13 on.

7.3 Detectors in Burnup Calculation

Detectors in Burnup Calculation

There are a few things that need to be considered when using detectors in the burnup calculation mode. First, the output is printed in a different file for each burnup step (see previous
section). The file names are separated by the step index, which is set to zero for the initial
composition. Second, when burning materials inside pin and particle structures (see Sec.3.4
and 3.8), the materials are renamed according to pin / particle index and region number if the
material is divided into multiple depletion zones (see Sec. 8.2). The original material names
no longer exist and the new names must be used instead with the “dm” parameter.

Chapter 8
Burnup calculation
8.1

Serpent can be run both as a stand-alone burnup calculation code and as a part of a coupled
system. In the first case, the code uses an internal calculation routine for solving the set of
Bateman equations describing the changes in the material compositions caused by neutroninduced reactions and radioactive decay. In the second case, the code is used as the neutronics
solver in an externally coupled system.
The additional input for burnup calculation consists of identifying the depleted materials
(Sec. 8.2) and setting up the irradiation history (Sec. 8.3). There are also some additional
parameters for determining file paths and options used by the calculation routines (Sec. 8.4).
A few simple examples are given in Sec. 8.7 and complete input listings in Sec. 11.2.
It should be noted that burnup calculations are more sensitive to small changes in the geometry, materials and calculation parameters compared to a steady state simulation. The length
of burnup steps and predictor-corrector calculation (see Sec. 8.3 and Sec. 8.4) may have a
significant impact on the accumulation of certain isotopes, and especially the depletion of
burnable absorbers. In thermal systems, the build-up rate of plutonium is strongly dependent on moderator conditions, such as density and the S(α, β) scattering laws (see Sec. 4.2
on Page 49). As low as a 30K difference in moderator temperature may result in over 1%
discrepancy in Pu-239 concentration at high burnup.1 Differences originating from the evaluated nuclear data should always be taken into account, especially for older libraries, such
as JEF-2.2 and ENDF/B-VI.
1

It should be noted that the thermal scattering data provided with the installation package is generated at
slightly different temperatures for different libraries.

8.2 Depleted materials

Depleted materials

Depleted materials are identified by an additional “burn” entry in the material card:
mat burn

is the material name
is the density (mass or atomic)
is the number of annular regions in depleted
fuel pins
are the names of the constituent nuclides
are the corresponding fractions (mass or atomic)

If the irradiation history is not set up, the “burn” entry activates the coupled calculation mode
and one-group transmutation cross sections, radioactive decay constants and fission yields
are written in a separate output file (see Sec. 8.6) without running the depletion calculation.
The code treats depleted materials in fuel pins different from materials in ordinary cells.
Each pin type is treated separately and further divided into annular depletion zones of
equal volume. The division is important for accounting for the rim-effects caused by spatial
self-shielding. The code automatically renames the depleted pin materials using convention:
pr
where is the original material name

is the pin (universe) number

is the ring index starting from the innermost region (= 1)
Depleted materials in ordinary cells are not renamed or divided into sub-regions.
IMPORTANT NOTES ON DEPLETED MATERIALS:
1. Each fuel pin type containing a depleted material is treated separately and divided into
a user-given number of annular depletion zones.
2. The separation of material regions is based on pin type, not lattice position. If similar
pins in different positions need to be treated as different materials, a new (identical)
pin type must be assigned for each position (See examples in Sec. 8.7.1 on page 117).
3. Fuel pins containing burnable absorber should always be divided into ∼10 rings in
order to account for the rim-effects caused by spatial self-shielding.
4. The current code version can only handle burnup calculation in cylindrical or spherical
material regions, such as fuel pins or HTGR micro particles.

8.3 Irradiation history

SEE ALSO:
1. Material cards (Sec. 4.1.2 on page 48)

Irradiation history

The irradiation history in the independent burnup calculation mode consists of one or several
burnup intervals, defined by the “dep” card:
dep

is the step type
are the burnup steps

The step types are listed in Table 8.1
Table 8.1: Burnup step types.

bustep
butot
daystep
daytot
decstep
dectot

Step values
depletion step, burnup intervals given in MWd/kgU
depletion step, cumulative burnup given in MWd/kgU
depletion step, time intervals given in days
depletion step, cumulative time given days
decay step, time intervals given in days
decay step, cumulative time given in days

Source rate normalization and soluble absorber concentration can be changed between intervals by re-defining the values. The first value is used during the first burnup interval, the
second during the second interval and so on. Examples are given in Sec. 8.7.2 on page 121.
The last two options omit the transport cycle and handle only radioactive decay, which makes
the calculation run significantly faster. This mode is intended to be used for calculating
activities and inventories after the irradiation is completed. Downtime between cycles is
better handled by setting the power to zero.
IMPORTANT NOTES ON IRRADIATION HISTORY:
1. If source rate normalization or soluble absorber concentration are changed between
burnup intervals, it is important that the number of definitions is equal to the number
of intervals.

8.4 Options for Burnup Calculation

2. The structure of the “dep” card is different in the early code versions (before 1.0.2).
3. The soluble absorber definition is available from version 1.0.2 on.
4. The decay mode is available from code version 1.1.10 on.
SEE ALSO:
1. Source rate normalization options (Sec. 5.8 on page 61)
2. Soluble absorber (Sec. 5.14 on page 69)

Options for Burnup Calculation

The calculation parameters in the burnup mode are summarized in Table 8.2.
Table 8.2: List of parameters and options in burnup calculation mode.
Option
declib
(1)
nfylib
(1)
sfylib
(1)
bunorm
(1)
fmass
(1)
bumode
(1)
pcc
(1)
xscalc
(1)
fpcut
(1)
axs
(2)
stabcut
(1)
ttacut
(1)
xsfcut
(1)
xsecut
(1)
inventory (1-N)
printm
(1)
dhprec
(1)

Description
file path for radioactive decay data
file path for fission yield data
file path for spontaneous fission yield data
normalization mode in burnup calculation
total fissile mass
solution method for Bateman equations
flag for predictor-corrector calculation
transmutation cross sections generation
fission product yield cut-off
actinide mass chains included in calculation
stability cut-off
TTA chain cut-off
XS fraction cut-off
XS threshold energy cut-off
nuclide list for burnup calculation output
flag for printing material compositions
precursor-group wise decay heat production

Section
8.4.1
8.4.1
8.4.1
8.4.2
8.4.2
8.4.3
8.4.3
8.4.4
8.4.5
8.4.5
8.4.5
8.4.5
8.4.5
8.4.5
8.4.6
8.4.7
8.4.8

Page
112
112
112
112
112
113
113
113
114
114
114
114
114
114
115
115
115

Library File Paths

In addition to the continuous-energy cross section libraries, burnup calculation requires radioactive decay data and neutron-induced and spontaneous fission product yields. These files
are read in the raw ENDF format. The decay data library file path is set using:

8.4 Options for Burnup Calculation

set declib ""
where is the file path for the ENDF format decay data library
the neutron-induced fission yield library using:
set nfylib ""
where is the file path for the ENDF format fission yield library
and the spontaneous fission yield library:
set sfylib ""
where is the file path for the ENDF format fission yield library
The spontaneous fission yield library is optional. If the file path is not set, the code uses
neutron-induced yields for spontaneous fission. The present code version does not model
spontaneous fission.
A default directory path can be set by defining environment variable SERPENT_DATA. The
code looks for data files in this path if not found at the absolute location.

The normalization of fission source is described in Sec. 5.8 on page 61. In some burnup
calculation problems, the geometry may contain fissile materials that are not depleted, which
may also affect the source normalization. Serpent offers three options, set using:
set bunorm
where is the normalization mode
Mode 1 is the default treatment which normalizes the given reaction rate or power to all
materials. Mode 2 includes only burnable materials and mode 3 only non-burnable materials.
The option is available from update 1.1.5 on and earlier code versions use all materials in the
normalization.
The code automatically calculates the total fissile mass in the system, which is needed for
normalizing the reaction rates. If the calculation fails, the value can be set manually using:
set fmass
where is the total fissile mass in the system (in grams)

8.4 Options for Burnup Calculation

Solution of Depletion Equations

The Serpent code has three options and two methods for solving the Bateman equations
describing the changes in the isotopic compositions caused by neutron-induced reactions
and radioactive decay. The calculation mode is set using:
set bumode
where is the method used for depletion calculation
The first method ( = 1) is Transmutation Trajectory Analysis (TTA), based on the
analytical solution of linearized transmutation chains. The second method ( = 2),
used by default, is an advanced matrix exponential solution based on the Chebyshev Rational Approximation Method (CRAM). The third option ( = 3) is the variation TTA
method, in which cyclic transmutation chains are handled by inducing small variations in the
coefficients instead of solving the extended TTA equations.
Predictor-corrector calculation is activated using:
set pcc
where is the flag for running the corrector step (0 = no, 1 = yes)
The method is used by default and results in a more accurate estimation of isotopic changes
during each burnup step. The drawback is that the transport cycle is repeated, which increases the overall calculation time.

Calculation of Transmutation Cross Sections

There are two options for calculating the isotopic one-group transmutation cross sections:
set xscalc
where is the method used for cross section calculation
In the default method ( = 2), the code calculates these parameters using a highresolution flux spectrum recorded during the transport calculation. This procedure results in
a reduction of calculation time by a factor of 3-4 compared to the direct calculation of the
cross sections during the transport cycle ( = 1). The drawback is that the method is
an approximation and that the information on statistical accuracy is lost.2
2

The flux spectrum is calculated using the main energy-grid structure. The resolution is high and the only
approximation is that the continuous-energy cross sections are assumed constant between two grid points. It
is therefore assumed that the difference to the direct calculation are negligible, although the methodology still
requires some thorough validation. The two methods are automatically compared by setting = 3.

8.4 Options for Burnup Calculation

Burnup calculation uses various cut-offs for reducing the computational effort.
Fission product yield cut-off determines which fission products are included in the calculation. The selection is based on the cumulative yield of each fp mass chain:
set fpcut
where is the limit for fission product yield cut-off
By default, the range of actinide mass chains included in the calculation extends from Amin 1 to Amax + 7, where Amin and Amax are the minimum and maximum actinide mass numbers
in the initial composition. This range can be set manually by:
set axs
where is the lightest actinide mass chain included in the calculation
is the heaviest actinide mass chain included in the calculation
Stability cut-off:
set stabcut
where is the limit for stability cut-off
TTA chain cut-off:
set ttacut
where is the limit for TTA chain cut-off
Cross section fraction cut-off:
set xsfcut
where is the limit for cross section fraction chain cut-off
Threshold energy cut-off:
set xsecut
where is the energy boundary

Nuclide Inventory

The standard output in the independent calculation mode consists of material compositions,
transmutation cross sections, activities and decay heating values. The isotopes, elements,

8.4 Options for Burnup Calculation

etc. included in the output are set by the inventory option:
set inventory ...
where are the identifiers.
The list consists of numerical values that identify the nuclides (1000*Z + 10*A + I) or elements (Z). Isotope and and elemental names and symbols (“Pu-239”, “Gd155”, “PM148M”,
“Cs”, “plutonium”, etc.) are also accepted. Elemental values are calculated by summing
over the isotopes. Table 8.3 lists additional options that can be used in the inventory list to
sum over several elements.
Table 8.3: Special entries in the inventory list. The list entry may consist of name or ID.
ID
201
202
204
208

Name
act
fp
dp
ng

Description
Actinides (Z > 89)
Fission products
Decay products below thorium in the natural actinide decay series
Noble gases (in the fission product range, helium and radon excluded)

Additional Output

The code has an option for writing the compositions of depleted materials in a separate output
file after each step:
set printm
where is the flag for printing material compositions (0 = no, 1 = yes)
The code produces for each step a file named “.bumat”, where
is the name of the input file and is the burnup step. The material compositions can be
used in another Serpent calculation or converted to MCNP format for validation purposes.

Decay heat production in multiple precursor groups

Decay heat production can be divided into multiple precursor groups based on the nuclide
decay constant. The syntax for the option is:
set dhprec [ ...

where are the group boundaries in ascending order
Default values are used if the option is not given. The output is printed in the main output
file (see Sec. 6.1.11). The feature is available from code version 1.1.17 on.

8.5 Output in independent mode

IMPORTANT NOTES ON BURNUP CALCULATION PARAMETERS:
1. Decay and fission yield libraries are raw ENDF data files in ASCII format.
2. Symbolical names can be used in the inventory list from version 1.1.3 on. Elemental
and special identifiers are available from version 1.1.10 on. If the list is empty, only
material total values are printed.
3. The code looks for the daughter nuclide cross section data libraries in the ACE directory file. It is important that the directory file contains as many nuclides as possible.
4. Mode 2 (matrix exponential solution) is available and used by default from version
1.1.0 on.
5. It is important to use the predictor-corrector step in cases involving burnable absorbers.
6. The environment variable feature is available from code version 1.1.8 on.
SEE ALSO:
1. Setting up the cross section library file path (Sec. 5.4 on page 57).
2. Description of the CRAM method in Ref. [21].

Output in independent mode

The burnup calculation output in the independent calculation mode is written in Matlab mfile format in file “_dep.m”, where is the name of the input file. The
variables are summarized in Table 8.4. The number of burnup steps is N and the number of
inventory nuclides I. The material-wise parameters are printed for each depleted material.
IMPORTANT NOTES ON OUTPUT:
1. If the predictor-corrector method is used, the material compositions are given at the
beginning of each step. The transmutation cross sections are not equivalent with the
corrected values used for solving the depletion equations.
2. The variable names are slightly different in the pre-release version 1.0.0 of the Serpent code (corrected in version 1.0.1).
3. The “lost” in the output file refers to data that is lost to undefined nuclides.
SEE ALSO:
1. Setting up burnup inventory list (Sec. 8.4.6 on page 115).

8.6 Output in coupled mode

Table 8.4: Variables in the Matlab m-format burnup calculation output file.
Variable
BU
DAYS
i
iTOT
iLOST
ZAI
NAMES
MAT__VOLUME
MAT__FLUX
MAT__ADENS
MAT__MDENS
MAT__A
MAT__H
MAT__FISSXS
MAT__CAPTXS
MAT__N2NXS
TOT_VOLUME
TOT_ADENS
TOT_MASS
TOT_A
TOT_H

Size
(1, N )
(1, N )
1
1
1
(I + 2, 1)
(I + 2, 8)
(1, N )
(1, N )
(I + 2, N )
(I + 2, N )
(I + 2, N )
(I + 2, N )
(I + 2, N )
(I + 2, N )
(I + 2, N )
1
(I + 2, N )
(I + 2, N )
(I + 2, N )
(I + 2, N )

Contents
Cumulative burnup in MWd/kgU
Cumulative burn time in days
Table index for nuclide “”
Table index for total values
Table index for lost data
Nuclide ZAI’s
Nuclide names (character strings)
Volume of material “”
Volume-integrated flux in material “”
Atomic densities in material “”
Mass densities in material “”
Activities in material “”
Decay heat in material “”
(n,f) cross sections in material “”
(n,γ) cross sections in material “”
(n,2n) cross sections in material “”
Total volume of depleted materials
Total averaged atomic densities
Total mass
Total activities
Total decay heat

Output in coupled mode

Burnup calculation examples

Material and lattice examples

A simple assembly burnup calculation consisting of two pin types:
% --- Fuel pin:
pin 1
UO2
clad
water

8.7 Burnup calculation examples

pin 3
UO2Gd
clad
water

% --- Guide tube:
pin 4
water
tube
water

% --- Pin lattice:
lat 110
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
3
4
1
1
4
1
1
4
3
1
1
1
1

1
1
1
4
1
1
1
1
3
1
1
1
1
4
1
1
1

1
1
1
3
1
1
1
1
1
1
1
1
1
1
1
3
1
1

1
1
4
1
1
4
1
1
4
1
1
4
1
1
4
1
1

1
1
1
1
1
1
3
1
1
1
3
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
4
3
1
4
1
1
4
1
1
4
1
3
4
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
3
1
1
1
3
1
1
1
1
1
1

1
1
4
1
1
4
1
1
4
1
1
4
1
1
4
1
1

1
1
3
1
1
1
1
1
1
1
1
1
1
1
3
1
1

1.265
1
1
1
4
1
1
1
1
3
1
1
1
1
4
1
1
1

1
1
1
1
3
4
1
1
4
1
1
4
3
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

% --- Fuel in normal pins, no division into rings:
mat UO2
92234.09c
92235.09c
92238.09c
8016.09c

6.7402E-02
9.1361E-06
9.3472E-04
2.1523E-02
4.4935E-02

% --- Fuel in Gd pins, division into 10 rings:
mat UO2Gd
92234.09c
92235.09c

6.8366E-02
4.2940E-06
5.6226E-04

8.7 Burnup calculation examples

92238.09c
64154.09c
64155.09c
64156.09c
64157.09c
64158.09c
64160.09c
8016.09c

2.0549E-02
4.6173E-05
2.9711E-04
4.1355E-04
3.1518E-04
4.9786E-04
4.3764E-04
4.5243E-02

Similar case, but each lattice position treated as a separate depletion zone, taking into account
the 1/12 symmetry of the pin layout:
% --- Fuel pins:
pin 10
UO2
clad
water

pin 11
UO2
clad
water

...
(identical definition of pins 12-45 omitted for simpicity)
...
% --- Gd-pins:
pin 50
UO2Gd
clad
water

pin 51
UO2Gd
clad
water

pin 52
UO2Gd
clad
water

8.7 Burnup calculation examples

% --- Guide tube:
pin 90
water
tube
water

% --- Pin lattice:
lat 110

45
44
43
42
41
40
39
38
37
38
39
40
41
42
43
44
45

42
34
27
90
24
23
22
21
51
21
22
23
24
90
27
34
42

41
33
52
24
20
19
18
17
16
17
18
19
20
24
52
33
41

38
30
25
21
17
14
13
11
10
11
13
14
17
21
25
30
38

44
36
35
34
33
32
31
30
29
30
31
32
33
34
35
36
44

43
35
28
27
52
90
26
25
90
25
26
90
52
27
28
35
43

40
32
90
23
19
90
15
14
90
14
15
90
19
23
90
32
40

39
31
26
22
18
15
50
13
12
13
50
15
18
22
26
31
39

38
30
25
21
17
14
13
11
10
11
13
14
17
21
25
30
38

37
29
90
51
16
90
12
10
90
10
12
90
16
51
90
29
37

39
31
26
22
18
15
50
13
12
13
50
15
18
22
26
31
39

40
32
90
23
19
90
15
14
90
14
15
90
19
23
90
32
40

41
33
52
24
20
19
18
17
16
17
18
19
20
24
52
33
41

42
34
27
90
24
23
22
21
51
21
22
23
24
90
27
34
42

43
35
28
27
52
90
26
25
90
25
26
90
52
27
28
35
43

44
36
35
34
33
32
31
30
29
30
31
32
33
34
35
36
44

45
44
43
42
41
40
39
38
37
38
39
40
41
42
43
44
45

% --- Fuel in normal pins, no division into rings:
mat UO2
92234.09c
92235.09c
92238.09c
8016.09c

6.7402E-02
9.1361E-06
9.3472E-04
2.1523E-02
4.4935E-02

% --- Fuel in Gd pins, division into 10 rings:
mat UO2Gd
92234.09c
92235.09c
92238.09c
64154.09c
64155.09c
64156.09c
64157.09c
64158.09c

6.8366E-02
4.2940E-06
5.6226E-04
2.0549E-02
4.6173E-05
2.9711E-04
4.1355E-04
3.1518E-04
4.9786E-04

8.7 Burnup calculation examples

64160.09c
8016.09c

4.3764E-04
4.5243E-02

Irradiation history examples

Irradiation at constant power density, cumulative burnup steps:
set powdens 40.0E-3
dep butot
0.10000
0.50000
1.00000
1.50000
2.00000
2.50000
3.00000
3.50000
4.00000
4.50000
5.00000
5.50000
6.00000
6.50000
7.00000
7.50000
8.00000
8.50000
9.00000
9.50000
10.00000
10.50000
11.00000
11.50000
12.00000
12.50000
13.00000
13.50000
14.00000
14.50000
15.00000
20.00000
25.00000
30.00000
35.00000

8.7 Burnup calculation examples

Similar case with step size given and history divided into 3 irradiation intervals with cooling
period. Nuclide inventory is traced for 1000 years after the fuel is removed from the reactor:
% --- Cycle 1: 650 ppm boron, final burnup 13.5 MWd/kgU
set powdens 40.0E-3
set abs boron -650E-6 water
dep bustep
0.10000
0.40000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
0.50000
% --- Downtime for 80 days:
set powdens 0.0
set abs boron -650E-6 water
dep daystep 80

8.7 Burnup calculation examples

% --- Cycle 2: 300 ppm boron, final burnup 25.0 MWd/kgU
set powdens 40.0E-3
set abs boron -300E-6 water
dep bustep
0.50000
0.50000
0.50000
5.00000
5.00000
% --- Downtime for 80 days:
set powdens 0.0
set abs boron -300E-6 water
dep daystep 80
% --- Cycle 3: no boron, final burnup 40.0 MWd/kgU
set powdens 40.0E-3
set abs boron 0.0 water
dep bustep
5.00000
5.00000
5.00000
% --- Decay after fuel is removed from the reactor
dep decstep
365
365
365
365
365
365
365
365
365
365
3650
3650

%
%
%
%
%
%
%
%
%
%
%
%

1. year
2. year
3. year
4. year
5. year
6. year
7. year
8. year
9. year
10. year
20. year
30. year

8.7 Burnup calculation examples

3650
3650
3650
3650
3650
3650
3650
36500
36500
36500
36500
36500
36500
36500
36500
36500

%
%
%
%
%
%
%
%
%
%
%
%
%
%
%
%

40. year
50. year
60. year
70. year
80. year
90. year
100. year
200. year
300. year
400. year
500. year
600. year
700. year
800. year
900. year
1000. year

Chapter 9
External Source Mode
9.1

External source simulation mode, available from version 1.1.11 on, can be used to replace the
k-eigenvalue criticality source method in sub-critical and non-multiplying systems. Instead
of performing power iterations on the fission source, all source neutrons are started from a
user-defined distribution. The calculation mode is activated by replacing the “pop” input
parameter (see Sec. 5.2 on page 53) with:
set nps [ ]
where

is the total number of source neutrons run
is the number of batches run

By default, the simulation is run by dividing the source size into 200 batches. Apart from the
source definition, described in the following section, the external source simulation works
very similar to the criticality source method. All features, including detectors and burnup
calculation are available.
IMPORTANT NOTES ON EXTERNAL SOURCE SIMULATION:
1. The calculation mode is available from version 1.1.11 on, and still very much under
development.
2. External source simulations can only be run in non-multiplying or sub-critical systems.
Geometries with keff ≥ 1 produce infinite multiplication and the simulation diverges.

9.2 Source definition

Source definition

The external source simulation requires one or several source definitions. A user-defined
source can also be used as the initial guess for criticality source calculations (see Sec. 5.2).
The syntax for the source definition is:
src ...
where
...

is the source name
are the source parameter sets

The parameters are listed in Table 9.1 and they can be combined in different ways as described in the following subsections. If multiple sources are used, the relative importances
are determined by the weights, set to unity by default.
Table 9.1: Detector parameters.
Param.
Description
sw
Source weight
sc
Source cell
sm
Source material
sp
Source point
sx, sy, sz Source boundaries
sd
Source direction
se
Source energy
sb
Source energy bins
sr
Source reaction
ss
Source surface

Comments
Determines the relative importance of the source
Defines the cell where the neutrons are started
Defines the material where the neutrons are started
Defines the coordinates of a point source
Defines the boundaries of the source distribution
Defines the source direction vector
Multiple uses
Defines a bin-wise energy spectrum
Defines the source reaction
Defines a surface source

Setting the Spatial Distribution

If spatial distribution is not defined, neutrons are started uniformly all over the geometry.
The sampling volume can limited by setting the boundaries in x-, y- and z-directions using:

9.2 Source definition

src sx sy sz
where

is the source name
is the minimum boundary in x-direction
is the maximum boundary in x-direction
is the minimum boundary in y-direction
is the maximum boundary in y-direction
is the minimum boundary in z-direction
is the maximum boundary in z-direction

The source can be defined by a single cell using:
src sc
where is the source name
is the cell where the neutrons are started
or to a single material using:
src sm
where is the source name

is the material where the neutrons are started
The cell and material definitions can be used in combination with the boundaries set by “sx”,
“sy” and “sz”.
An alternative to a volume source is the point source, defined as:
src sp
where

is the source name
is x-coordinate of the point source
is y-coordinate of the point source
is z-coordinate of the point source

Surface sources can be defined as:
src ss
where is the source name
is the source surface
The surface is defined using the “surf” card (see Sec. 3.2 on page 19). Positive and negative
entries refer to neutrons being emitted in the direction of positive and negative surface normal, respectively. The feature is available from version 1.1.15 on, and the allowed surface
types include sphere (“sph”) and cylinder (“cyl”).

9.2 Source definition

Setting the Directional Distribution

By default, all source neutrons in point and volume sources are emitted isotropically. To
define a mono-directional source, the direction vector can be set by the “sd” parameter:
src sd
where

is the source name
is direction cosine in the x-direction
is direction cosine in the y-direction
is direction cosine in the z-direction

Directional distributions will be added in future code versions.

Setting the Energy Distribution

A mono-energetic source is defined by setting the “se” parameter:
src se
where is the source name

is neutron energy
By default, the emission energy is set to 1 MeV.
Another option is to take the energy distribution from a nuclear reaction using the “sr” option:
src sr
where is the nuclide identifier

is the reaction mt
The reaction can be any scattering or fission reaction for which the distribution data exists in
the ACE format data (notice that this is not the case for elastic scattering and inelastic level
scattering). If source energy is defined using the “se” option, the value is used as the energy
of the incoming neutron when the emission energy is sampled. If the value is not set, the
minimum value allowed by the distribution is used.
The third option is to define discrete energy bins as:
src sb ...

where is the number of source energy bins
are the energy bin boundaries
are the bin weights
The code samples the energy bin according to the probability calculated from the bin weights,

9.3 Source Examples

and the energy uniformly between the bin boundaries. The energy entries correspond to the
upper boundaries of each bin, and the weight of the first bin must be set to zero. The feature
is available from version 1.1.15 on.

Source distribution can be read from a file using:
src sf
where is the source name
is the source file
is the file type (must be 1)
The source file contains coordinates, direction cosines, energy, weight and time for every
source neutron, one entry per line. This feature was added in version 1.1.17, and the format
of the source file may change in later updates.

Source Examples

Source definition using default parameters – isotropic, mono-energetic 1 MeV source, uniformly distributed over the geometry:
src 1

Setting the spatial and directional distribution:
% Uniform source in a cuboid:
src 2 sx -1.0 1.0 sy -1.0 1.0 sz -1.0 1.0
% Source in cell:
src 3 sc 1
% Source in material, bounded in axial direction:
src 4 sm fuel sz -10.0 10.0
% Point source in origin, directed in the positive x-axis:
src 5 sp 0.0 0.0 0.0 sd 1.0 0.0 0.0

9.3 Source Examples

Setting the energy distribution:
% Three point sources with different energy and importance
src 6 sw 0.5 sp 0.0 0.0 0.0 se 1.0
src 7 sw 0.3 sp 1.0 0.0 0.0 se 2.0
src 8 sw 0.2 sp 0.0 1.0 0.0 se 3.0
% U-235 fission source in material fuel:
src 9 sc fuel sr 92235.03c 18
% U-238 fission source induced by 14 MeV neutrons:
src 10 sr 92238.03c 18 se 14.0
% Histogram energy distribution defined using 5 bins:
src 6
1E-11
1E-6
1E-3
1.0
20.0

sb 5
0.0 %
0.5 %
1.0 %
2.0 %
1.0 %

Energy below 1E-11 MeV (weight must be zero)
Between 1E-11 and 1E-6 MeV, weight 0.5
Between 1E-6 and 1E-3 MeV, weight 1.0
Between 1E-3 and 1.0 MeV, weight 2.0
Between 1.0 and 20.0 MeV, weight 1.0

Chapter 10
Reaction rate mesh plotter
10.1

Serpent has a built-in capability to visualize the neutronics in thermal systems by plotting
the fission power and thermal flux distributions in a single png graphics file. The parameters
for a reaction rate mesh plotter are defined as:
mesh [ ]
where

is the orientation of the plot plane (1, 2 or 3)
is the width of the plot in pixels
is the height of the plot in pixels
is the symmetry option (0, 2, 4 or 8)
is the minimum value of the x-coordinate
is the maximum value of the x-coordinate
is the minimum value of the y-coordinate
is the maximum value of the y-coordinate
is the minimum value of the z-coordinate
is the maximum value of the z-coordinate

The code calculates reaction rates in an by mesh, and projects tha data according to the orientation of the plot plane, defined as:
1. yz-plot (perpendicular to the x-axis)
2. xz-plot (perpendicular to the y-axis)
3. xy-plot (perpendicular to the z-axis)
If the optional coordinate boundaries are not given, the code uses the boundaries of the
defined geometry.

10.2 Mesh output

The symmetry option can be used to attain better statistics. The symmetry types are illustrated in Fig. 5.1 on page 63, and only options 0, 2, 4 and 8 are allowed with mesh plots. The
option is set to zero by default (no symmetry).

Output is written in a png format file “_mesh.png”, where is the
name of the input file and is the plot index. Burnup mode produces new plots for each
depletion step. The files are named “_mesh_bstep.png”, where
is the step index.
The colour scheme consists of “hot” shades of red and yellow, representing relative fission
power, and “cold” shades of blue, representing relative thermal flux (flux below 0.625 eV).
The normalization is fixed after the first burnup step, so changes in flux and power level can
be observed in the color schemes. Examples of reaction rate mesh plots can be found at the
Serpent website: http://montecarlo.vtt.fi/development.htm.
IMPORTANT NOTES ON REACTION RATE MESH PLOTTER:
1. The mesh plots are subject to random noise, and the figures become smoother along
with better statistics.
2. The geometry plotter uses the GD open source graphics library [1], which must be
installed in the system.
3. The plotter produces png (portable network graphics) format output files.
SEE ALSO:
1. Compiling Serpent (Sec. 1.1 on page 8)
2. The GD open source graphics library: http://www.libgd.org
3. Mesh plot gallery at Serpent website:
http://montecarlo.vtt.fi/development.htm.

Chapter 11
Complete Input Examples
11.1

For an experienced Monte Carlo code user the easiest way to get started with Serpent is
to look at the lattice input examples in the following subsections. Installation and running
the code is described in Chapter 1 and a general description of the input syntax is given in
Chapter 2. The input cards used in the example cases include:
– Fuel pin definitions (Sec. 3.4 on page 27)
– Lattice definitions (Sec. 3.6 on page 28)
– Surface definitions (Sec. 3.2 on page 19)
– Cell definitions (Sec. 3.3 on page 24)
– Material definitions (Sec. 4.1.2 on page 48, see also Sec. 4.1.1)
– Thermal scattering libraries (Sec. 4.2 on page 49)
– Soluble absorber (Sec. 5.14 on page 69)
– File paths (Sec. 5.4 on page 57)
– Neutron population and criticality cycles (Sec. 5.2 on page 53)
– Boundary conditions (Sec. 5.7 on page 59)
– Parameters for group constant generation (Sec. 5.9 on page 64)
– Detectors (Chapter. 7 on page 95)

11.1 Quick start

The examples describe the three main lattice types: square and hexagonal lattices and the
circular cluster array. All geometries are two-dimensional and infinite in the axial direction.
The VVER-440 example in Sec. 11.1.1 demonstrates the use of soluble absorber and the
calculation of various spectral quantities using detectors. The BWR case in Sec. 11.1.2
demonstrates the calculation of fast neutron flux (E > 1 MeV) in cladding and flow channel
walls.
A more complicated mixed UOX/MOX lattice example is given in Sec. 11.1.4. The homogenization is carried over the central MOX assembly, but the use of a simple infinite MOX
lattice would result in a distroted flux spectrum near the boundary between the two fuel types.
The input format is free and unrestricted. The only limitation is that command words must
be separated by one or more white space characters. Due to the universe-based approach,
similarities to MCNP input files are easy to see. To differentiate from the other examples,
the mixed lattice case in Sec. 11.1.4 is prepared following a “SCALE-style” formulation.
More example cases are available at the Serpent website: http://montecarlo.vtt.fi.

VVER-440 lattice calculation

% --- VVER-440 Assembly -------------------------------------set title "VVER-440"
% --- Fuel pin with central hole:
pin 1
void
fuel
void
clad
water

0.08000
0.37800
0.38800
0.45750

% --- Central tube:
pin 2
water
clad
water

0.44000
0.51500

% --- Empty lattice position:
pin 3
water

11.1 Quick start

% --- Lattice (type = 2, pin pitch = 1.23 cm):
lat 10 2 0.0 0.0 15 15 1.23
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
3 3 3 3 3 3 3 1 1 1 1 1 1 1 3
3 3 3 3 3 3 1 1 1 1 1 1 1 1 3
3 3 3 3 3 1 1 1 1 1 1 1 1 1 3
3 3 3 3 1 1 1 1 1 1 1 1 1 1 3
3 3 3 1 1 1 1 1 1 1 1 1 1 1 3
3 3 1 1 1 1 1 1 1 1 1 1 1 1 3
3 1 1 1 1 1 1 2 1 1 1 1 1 1 3
3 1 1 1 1 1 1 1 1 1 1 1 1 3 3
3 1 1 1 1 1 1 1 1 1 1 1 3 3 3
3 1 1 1 1 1 1 1 1 1 1 3 3 3 3
3 1 1 1 1 1 1 1 1 1 3 3 3 3 3
3 1 1 1 1 1 1 1 1 3 3 3 3 3 3
3 1 1 1 1 1 1 1 3 3 3 3 3 3 3
3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
% --- Surfaces (assembly pitch = 14.7 cm):
surf 1
surf 2
surf 3

hexyc
hexyc
hexyc

7.100
7.250
7.350

% Shroud tube inner radius
% Shroud tube outer radius
% Outer boundary

% --- Cells:
cell 1
cell 4
cell 5
cell 99

fill 10
tube
water
outside

Pin lattice
Shroud tube
Water in channel
Outside world

% --- UO2 fuel enriched to 3.6 wt-% U-235:
mat fuel
92235.09c
92238.09c
8016.09c

-10.45700
-0.03173
-0.84977
-0.11850

% --- Zr-Nb cladding and shroud tube:
mat clad
40000.06c
41093.06c

-6.55000
-0.99000
-0.01000

mat tube
40000.06c
41093.06c

-6.58000
-0.97500
-0.02500

11.1 Quick start

% --- Water:
mat water
1001.06c
8016.06c

-0.7207
2.0
1.0

moder lwtr 1001

% --- Thermal scattering data for light water:
therm lwtr lwj3.11t
% --- Natural boron (used as soluble absorber):
mat boron
5010.06c
5011.06c

% --- 650 ppm soluble absorber in water:
set abs boron -650E-6 water
% --- Cross section library file path:
set acelib "/xs/sss_jeff31.xsdata"
% --- Periodic boundary condition:
set bc 3
% --- Group constant generation:
% universe = 0 (homogenization over all space)
% symmetry = 12
% 2-group structure (group boundary at 0.625 eV)
set gcu
set sym
set nfg

0
12
2 0.625E-6

% --- Neutron population and criticality cycles:
set pop 2000 500 20
% --- Geometry and mesh plots:
plot 3 500 500
mesh 3 500 500

11.1 Quick start

% --- Detector energy grid (uniform lethargy):
ene 1 3 1000 1E-9 12.0
% --- Flux per lethargy:
det 1 de 1 dt -3
% --- Differential capture, fission and production spectra:
det 2 de 1 dt -2 dr -2 void
det 3 de 1 dt -2 dr -6 void
det 4 de 1 dt -2 dr -7 void
% --- Integral capture, fission and production spectra:
det 5 de 1 dt -1 dr -2 void
det 6 de 1 dt -1 dr -6 void
det 7 de 1 dt -1 dr -7 void
% ------------------------------------------------------------

BWR lattice calculation

% --- Asymmetric BWR assembly with Gd-pins ------------------set title "BWR+Gd"
% --- Fuel Pin definitions:
pin 1
fuel1
void
clad
cool

4.33500E-01
4.42000E-01
5.02500E-01

pin 2
fuel2
void
clad
cool

4.33500E-01
4.42000E-01
5.02500E-01

pin 3
fuel3
void
clad

4.33500E-01
4.42000E-01
5.02500E-01

11.1 Quick start

cool
pin 4
fuel4
void
clad
cool

4.33500E-01
4.42000E-01
5.02500E-01

pin 5
fuel5
void
clad
cool

4.33500E-01
4.42000E-01
5.02500E-01

pin 6
fuel6
void
clad
cool

4.33500E-01
4.42000E-01
5.02500E-01

pin 7
fuel7
void
clad
cool

4.33500E-01
4.42000E-01
5.02500E-01

% --- Empty lattice position:
pin 9
cool
% --- Lattice (type = 1, pin pitch = 1.295):
lat
9 9
9 1
9 2
9 3
9 5
9 5
9 5
9 5
9 5
9 3
9 2
9 9

10
9 9
2 3
3 5
5 7
6 6
6 7
6 6
6 6
7 6
5 6
4 5
9 9

1
9
5
6
6
6
6
6
6
7
6
6
9

0.0 0.0 12
9 9 9 9 9 9
5 5 5 5 3 2
6 6 6 7 5 4
7 6 6 6 6 5
6 6 6 7 6 6
9 9 9 6 7 6
9 9 9 6 6 6
9 9 9 6 6 6
6 6 6 7 6 5
7 6 6 6 6 5
6 6 6 5 5 3
9 9 9 9 9 9

12 1.295
9
9
9
9
9
9
9
9
9
9
9
9

% --- Outer channel (assembly pitch = 15.375):

11.1 Quick start

surf 1
surf 2
surf 3

6.70000
6.93000
7.68750

% --- Channel inside assembly:
surf 4
surf 5

% --- Cell definitions:
cell 1
cell 2
cell 3
cell 4
cell 5
cell 99

moder
box
fill 10
box
moder
outside

-4
4 -5
-1 5
1 -2
2 -3
3

% --- Fuel materials:
mat fuel1
92235.09c
92238.09c
8016.09c

-10.424
-0.015867
-0.86563
-0.1185

mat fuel2
92235.09c
92238.09c
8016.09c

-10.424
-0.018512
-0.86299
-0.1185

mat fuel3
92235.09c
92238.09c
8016.09c

-10.424
-0.022919
-0.85858
-0.1185

mat fuel4
92235.09c
92238.09c
8016.09c

-10.424
-0.026445
-0.85505
-0.1185

mat fuel5
92235.09c
92238.09c
8016.09c

-10.424
-0.029971
-0.85153
-0.1185

mat fuel6
92235.09c

-10.424
-0.032615

Water inside moderator channel
Moderator channel walls
Pin lattice
Channel box wall
Water outside channel box
Outside world

11.1 Quick start

92238.09c
8016.09c

-0.84888
-0.1185

% --- Fuel with Gd:
mat fuel7
92235.09c
92238.09c
64152.09c
64154.09c
64155.09c
64156.09c
64157.09c
64158.09c
64160.09c
8016.09c

-10.291
-3.13109E-02
-8.14929E-01
-6.70544E-05
-7.13344E-04
-5.06012E-03
-7.08860E-03
-5.43718E-03
-8.64341E-03
-7.69426E-03
-1.19056E-01

% --- Cladding and channel box wall:
mat clad
40000.06c
24000.06c
26000.06c
28000.06c
50000.06c
8016.06c

-6.55
-0.98135
-0.00100
-0.00135
-0.00055
-0.01450
-0.00125

mat box
40000.06c
24000.06c
26000.06c
28000.06c
50000.06c
8016.06c

-6.55
-0.98135
-0.00100
-0.00135
-0.00055
-0.01450
-0.00125

% --- Coolant (40% void fraction):
mat cool
1001.06c
8016.06c

-0.443760
0.66667
0.33333

moder lwtr 1001

% --- Moderator:
mat moder -0.739605 moder lwtr 1001
1001.06c
0.666667
8016.06c
0.333333
% --- Thermal scattering data for light water:

11.1 Quick start

therm lwtr lwj3.11t
% --- Cross section data library file path:
set acelib "/xs/sss_jeff31.xsdata"
% --- Reflective boundary condition:
set bc 2
% --- group constant generation:
% universe = 0 (homogenization over all space)
% symmetry = 4
% 4-group structure (3 group boundaries)
set gcu
set sym
set nfg

% --- Neutron population and criticality cycles:
set pop 2000 500 20
% --- Geometry and mesh plots:
plot 3 500 500
mesh 3 500 500
% --- Total power for normalization:
set power 1.96329E+04
% --- Detector energy grid (1 bin, E > 1.0 MeV):
ene 1 1 1.0 20
% --- Average fast flux in cladding:
det 1
de 1
dm clad
dv 16.3361

Use energy grid 1
Score in material "clad"
Volume for normalization

% --- Pin-wise fast flux in cladding:
det 2

11.1 Quick start

1
clad
10
0.17952

Use energy grid 1
Score in material "clad"
Lattice bins in lat 10
Volume for normalization

% --- Fast flux in inner moderator channel wall:
det 3
de 1
dc 2
dv 0.96134

Use energy grid 1
Score in cell 2
Volume for normalization

% --- Fast flux in outer channel wall:
det 4
de 1
dc 4
dv 12.5396

Use energy grid 1
Score in cell 4
Volume for normalization

% ------------------------------------------------------------

CANDU lattice calculation

% --- CANDU cluster -----------------------------------------set title "CANDU"
% --- Fuel pin:
pin 1
fuel 0.6122
clad 0.6540
cool
% --- Lattice (type = 4, 4 rings, 3rd ring rotated 15 deg.):
lat
1
6
12
18

10 4 0.0 0.0
0.0000
0.0 1
1.4885
0.0 1
2.8755 15.0 1
4.3305
0.0 1

4
1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

% --- Surfaces (core pitch = 18.191 cm):
surf 1
surf 2

cyl 0.0 0.0 5.16890
cyl 0.0 0.0 5.60320

% Pressure tube inner wall
% Pressure tube outer wall

11.1 Quick start

surf 3
surf 4
surf 5

cyl 0.0 0.0 6.44780
cyl 0.0 0.0 6.58750
sqc 0.0 0.0 9.09570

% Calandria tube inner wall
% Calandria tube outer wall
% Outer boundary

% --- Cells:
cell
cell
cell
cell
cell
cell

fill 10
tube
void
caltube
moder
outside

Pin lattice
Pressure tube
Void between tubes
Calandria tube
Moderator channel
Outside world

% --- Fuel (UO2, natural uranium, 0.7% U-235):
mat fuel
8016.09c
92235.09c
92238.09c

-10.4375010
-1.18473E+1
-6.27118E-1
-8.75256E+1

% --- Cladding:
mat clad
25055.06c
28000.06c
24000.06c
40000.06c
5010.06c
5011.06c

-6.44
-1.60000E-1
-6.00000E-2
-1.10000E-1
-9.97100E+1
-5.7409e-05
-2.5259E-04

% --- Pressure tube:
mat tube
40000.06c
5010.06c
5011.06c

-6.57
-9.75000E+1
-3.8889E-05
-1.7111E-04

% --- Calandria tube:
mat caltube
25055.06c
28000.06c
24000.06c
40000.06c
5010.06c
5011.06c

-6.44
-1.60000E-1
-6.00000E-2
-1.10000E-1
-9.97100E+1
-5.7409e-05
-2.5259E-04

% --- Coolant water:

11.1 Quick start

mat cool
8016.06c
1002.06c
1001.06c

-0.812120
-7.99449E-1
-1.99768E-1
-7.83774E-4

moder lwtr 1001 moder hwtr 1002

% --- Moderator water:
mat moder
8016.06c
1002.06c
1001.06c

-1.082885
-7.98895E-1
-2.01016E-1
-8.96000E-5

moder lwtr 1001 moder hwtr 1002

% --- Thermal scattering data for light and heavy water:
therm lwtr lwj3.11t
therm hwtr hwj3.11t
% --- Cross section data library file path:
set acelib "/xs/sss_jeff31.xsdata"
% --- Periodic boundary condition:
set bc 3
% --- group constant generation:
% universe = 0 (homogenization over all space)
% symmetry = 2
% 4-group structure (3 group boundaries)
set gcu
set sym
set nfg

% --- Neutron population and criticality cycles:
set pop 2000 500 20
% --- Geometry and mesh plots:
plot 3 500 500
mesh 3 500 500
% ------------------------------------------------------------

11.1 Quick start

Mixed UOX/MOX PWR lattice calculation

% --- PWR MOX/UOX lattice (SCALE-style input formulation) ---% --- Problem title:
set title "MOX assembly in UOX lattice"
% --- Cross section library file path:
set acelib "/xs/sss_jeff31.xsdata"
% -----------------------------------------------------------% --- Material definitions ("comp block"):
% --- UOX fuel, initial enrichment 3.25%, burnup 25 MWd/kgU:
mat UO2
92235.09c
92236.09c
92238.09c
93237.09c
94238.09c
94239.09c
94240.09c
94241.09c
94242.09c
95241.09c
54131.09c
54135.09c
63153.09c
62149.09c
45103.09c
60143.09c
55133.09c
64155.09c
43099.09c
42095.09c
61147.09c
62150.09c
62151.09c
62152.09c
8016.09c

6.585000E-02
3.0000E-04
8.0000E-05
2.0000E-02
7.1000E-06
1.7000E-06
1.2000E-04
3.8000E-05
2.1000E-05
5.3000E-06
4.2000E-07
1.4000E-05
8.0000E-09
2.8000E-06
9.0000E-08
1.8000E-05
2.5000E-05
3.5000E-05
8.4000E-10
3.2000E-05
3.2000E-05
6.4000E-06
7.5000E-06
4.1000E-07
3.2000E-06
4.5100E-02

% --- Low Pu-content (2.9%) MOX fuel:

11.1 Quick start

mat MOX1
92234.09c
92235.09c
92236.09c
92238.09c
94238.09c
94239.09c
94240.09c
94241.09c
94242.09c
95241.09c
8016.09c

6.702700E-02
4.3391E-07
4.9682E-05
8.6782E-07
2.1644E-02
5.4861E-06
4.3144E-04
1.3387E-04
4.8185E-05
1.8859E-05
9.1090E-06
4.4685E-02

% --- Medium Pu-content (4.4%) MOX fuel:
mat MOX2
92234.09c
92235.09c
92236.09c
92238.09c
94238.09c
94239.09c
94240.09c
94241.09c
94242.09c
95241.09c
8016.09c

6.702100E-02
4.2718E-07
4.8271E-05
8.5435E-07
2.1309E-02
8.1476E-06
6.5555E-04
2.0151E-04
7.4065E-05
2.7751E-05
1.4626E-05
4.4681E-02

% --- High Pu-content (5.6%) MOX fuel:
mat MOX3
92234.09c
92235.09c
92236.09c
92238.09c
94238.09c
94239.09c
94240.09c
94241.09c
94242.09c
95241.09c
8016.09c

6.701800E-02
4.2175E-07
4.9766E-05
8.4350E-07
2.1037E-02
1.0815E-05
8.3501E-04
2.5798E-04
9.4430E-05
3.6112E-05
1.7374E-05
4.4678E-02

% --- Zircaloy in cladding and guide tube:
mat can
40000.06c
26000.06c

4.004642E-02
3.9550E-02
1.3830E-04

11.1 Quick start

24000.06c
8016.06c

7.0720E-05
2.8740E-04

% --- Water with 550 ppm boron:
mat water
1001.06c
8016.06c
5010.06c
5011.06c

7.088200E-02
4.7240E-02
2.3620E-02
4.3210E-06
1.7390E-05

moder lwtr 1001

% --- Thermal scattering data for light water:
therm lwtr lwj3.11t
% -----------------------------------------------------------% --- Parameters ("param block"):
% --- Periodic boundary condition:
set bc 3
% --- Group constant generation:
% universe = 200 (homogenization over MOX assembly)
% symmetry = 8
% 2-group structure (group boundary at 0.625 eV)
set gcu
set sym
set nfg

200
8
2 0.625E-6

% --- Neutron population and criticality cycles:
set pop 2000 500 20
% -----------------------------------------------------------% --- Geometry ("geom block"):
% --- UOX Pin ("unit 1"):
pin 1
UO2
0.41260
can
0.47400
water

11.1 Quick start

% --- Guide tube ("unit 2"):
pin 2
water 0.57100
can
0.61300
water
% --- MOX Pins ("units 3-5"):
pin 3
MOX1
can
water

0.41260
0.47400

pin 4
MOX2
can
water

0.41260
0.47400

pin 5
MOX3
can
water

0.41260
0.47400

% --- UOX-assembly ("unit 100"):
surf

cell 100
cell 101

% --- MOX-assembly ("unit 200"):
surf

cell 200
cell 201

% --- Core lattice ("global unit 0"):
surf

cell 300
cell 301

fill 300
outside

% -----------------------------------------------------------% --- Lattices ("array block"):

11.1 Quick start

% --- UOX pin lattice:
lat 110
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
2
1
1
2
1
1
2
1
1
1
1
1

1
1
1
2
1
1
1
1
1
1
1
1
1
2
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
2
1
1
2
1
1
2
1
1
2
1
1
2
1
1

0.0 0.0 17 17
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
2
1
1
2
1
1
2
1
1
2
1
1
2
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
2
1
1
2
1
1
2
1
1
2
1
1
2
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1.262
1
1
1
2
1
1
1
1
1
1
1
1
1
2
1
1
1

1
1
1
1
1
2
1
1
2
1
1
2
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

% --- MOX pin lattice:
lat 210
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3

3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3

3
4
4
4
4
2
4
4
2
4
4
2
4
4
4
4
3

3
4
4
2
4
5
5
5
5
5
5
5
4
2
4
4
3

1
3
4
4
4
5
5
5
5
5
5
5
5
5
4
4
4
3

3
4
2
5
5
2
5
5
2
5
5
2
5
5
2
4
3

0.0 0.0 17 17
3
4
4
5
5
5
5
5
5
5
5
5
5
5
4
4
3

3
4
4
5
5
5
5
5
5
5
5
5
5
5
4
4
3

3
4
2
5
5
2
5
5
2
5
5
2
5
5
2
4
3

3
4
4
5
5
5
5
5
5
5
5
5
5
5
4
4
3

3
4
4
5
5
5
5
5
5
5
5
5
5
5
4
4
3

3
4
2
5
5
2
5
5
2
5
5
2
5
5
2
4
3

3
4
4
4
5
5
5
5
5
5
5
5
5
4
4
4
3

1.262
3
4
4
2
4
5
5
5
5
5
5
5
4
2
4
4
3

3
4
4
4
4
2
4
4
2
4
4
2
4
4
4
4
3

3
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3

% --- Core lattice:
lat 300

3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3

11.2 Burnup calculation examples

100 100 100
100 200 100
100 100 100
% -----------------------------------------------------------% --- Plotters ("plot block"):
% --- Geometry and mesh plots:
plot 3 500 500
mesh 3 500 500
% ------------------------------------------------------------

Burnup calculation examples

Pin-cell burnup calculation

% --- Pin-cell burnup calculation ---------------------------set title "Pin-cell burnup calculation"
% --- Pin definition:
pin 1
fuel
clad
water

% --- Geometry:
surf 1

sqc 0.0 0.0 0.665

% --- Fuel (composition given in atomic densities):
mat fuel -10.045 burn 1
92234.09c
6.15169E+18
92235.09c
6.89220E+20
92236.09c
3.16265E+18
92238.09c
2.17103E+22

11.2 Burnup calculation examples

6012.09c
7014.09c
8016.09c

9.13357E+18
1.04072E+19
4.48178E+22

% --- Zircalloy cladding:
mat clad
40000.06c
50000.06c
26000.06c

-6.560
-0.9791
-0.0159
-0.0050

% --- Water (composition given in atomic densities):
mat water
1001.06c
8016.06c
5010.06c
5011.06c

-0.7569
moder lwtr 1001
5.06153E+22
2.53076E+22
2.75612E+18
1.11890E+19

% --- Thermal scattering data for light water:
therm lwtr lwj3.11t
% --- Cross section library file path:
set acelib "/xs/sss_jeff31.xsdata"
% --- Periodic boundary condition:
set bc 3
% --- Group constant generation:
% universe = 0 (homogenization over all space)
% symmetry = 12
% 2-group structure (group boundary at 0.625 eV)
set gcu
set sym
set nfg

0
12
2 0.625E-6

% --- Neutron population and criticality cycles:
set pop 2000 500 20
% --- Geometry and mesh plots:
plot 3 500 500

11.2 Burnup calculation examples

mesh 3 500 500
% --- Decay and fission yield libraries:
set declib "/xs/JEFF311RDD"
set nfylib "/xs/JEFF31NFY"
% --- Reduce energy grid size:
set egrid 5E-5 1E-9 15.0
% --- Cut-offs:
set
set
set
set

fpcut
stabcut
ttacut
xsfcut

1E-9
1E-12
1E-18
1E-6

% --- Options for burnup calculation:
set
set
set
set

bumode
pcc
xscalc
printm

TTA method
Predictor-corrector calculation on
Cross sections from spectrum
No material compositions

% --- Depletion steps:
% Power density 40 kW/kgU
% Depletion steps given in units of total burnup
set powdens 40.0E-3
dep butot
0.1
0.5
1
5
10
15
20
25
30
35
40
% --- Isotope list for inventory calculation:

11.2 Burnup calculation examples

set inventory
922340
922350
922360
922380
932370
942380
942390
942400
942410
942420
952410
952430
420990
430990
441010
451030
471090
551330
621470
621490
621500
621510
621520
601430
601450
631530
641550
% ------------------------------------------------------------

PWR assembly burnup calculation

set title "PWR Burnup Calculation Based on NEA Benchmark"
% --- Fuel pins:
pin 10
UO2
clad
water
pin 11
UO2
clad

11.2 Burnup calculation examples

water
pin 12
UO2
clad
water

pin 13
UO2
clad
water

pin 14
UO2
clad
water

pin 15
UO2
clad
water

pin 16
UO2
clad
water

pin 17
UO2
clad
water

pin 18
UO2
clad
water

pin 19
UO2
clad
water

pin 20
UO2
clad
water

11.2 Burnup calculation examples

pin 22
UO2
clad
water

pin 23
UO2
clad
water

pin 24
UO2
clad
water

pin 25
UO2
clad
water

pin 26
UO2
clad
water

pin 27
UO2
clad
water

pin 28
UO2
clad
water

pin 29
UO2
clad
water

pin 30
UO2
clad
water

11.2 Burnup calculation examples

pin 31
UO2
clad
water

pin 32
UO2
clad
water

pin 33
UO2
clad
water

pin 34
UO2
clad
water

pin 35
UO2
clad
water

pin 36
UO2
clad
water

pin 37
UO2
clad
water

pin 38
UO2
clad
water

pin 39
UO2
clad
water

11.2 Burnup calculation examples

pin 41
UO2
clad
water

pin 42
UO2
clad
water

pin 43
UO2
clad
water

pin 44
UO2
clad
water

pin 45
UO2
clad
water

% --- Gd-pins:
pin 50
UO2Gd
clad
water

pin 51
UO2Gd
clad
water

pin 52
UO2Gd
clad
water

% --- Guide tube:
pin 90

11.2 Burnup calculation examples

water
tube
water

% --- Pin lattice:
lat 110

45
44
43
42
41
40
39
38
37
38
39
40
41
42
43
44
45

42
34
27
90
24
23
22
21
51
21
22
23
24
90
27
34
42

41
33
52
24
20
19
18
17
16
17
18
19
20
24
52
33
41

38
30
25
21
17
14
13
11
10
11
13
14
17
21
25
30
38

44
36
35
34
33
32
31
30
29
30
31
32
33
34
35
36
44

43
35
28
27
52
90
26
25
90
25
26
90
52
27
28
35
43

40
32
90
23
19
90
15
14
90
14
15
90
19
23
90
32
40

39
31
26
22
18
15
50
13
12
13
50
15
18
22
26
31
39

38
30
25
21
17
14
13
11
10
11
13
14
17
21
25
30
38

37
29
90
51
16
90
12
10
90
10
12
90
16
51
90
29
37

39
31
26
22
18
15
50
13
12
13
50
15
18
22
26
31
39

40
32
90
23
19
90
15
14
90
14
15
90
19
23
90
32
40

% --- assembly data:
surf
surf

cell 110
cell 111
cell 112

0.0 0.0 10.752
0.0 0.0 10.806

fill 110
water
outside

-1000
1000 -1001
1001

% --- Materials:
mat UO2
92234.09c
92235.09c
92238.09c
8016.09c

6.7402E-02
9.1361E-06
9.3472E-04
2.1523E-02
4.4935E-02

mat UO2Gd
92234.09c
92235.09c
92238.09c
64154.09c

6.8366E-02
4.2940E-06
5.6226E-04
2.0549E-02
4.6173E-05

41
33
52
24
20
19
18
17
16
17
18
19
20
24
52
33
41

42
34
27
90
24
23
22
21
51
21
22
23
24
90
27
34
42

43
35
28
27
52
90
26
25
90
25
26
90
52
27
28
35
43

44
36
35
34
33
32
31
30
29
30
31
32
33
34
35
36
44

45
44
43
42
41
40
39
38
37
38
39
40
41
42
43
44
45

11.2 Burnup calculation examples

64155.09c
64156.09c
64157.09c
64158.09c
64160.09c
8016.09c

2.9711E-04
4.1355E-04
3.1518E-04
4.9786E-04
4.3764E-04
4.5243E-02

mat clad
26000.06c
24000.06c
40000.06c

3.8510E-02
1.3225E-04
6.7643E-05
3.8310E-02

mat tube
26000.06c
24000.06c
40000.06c

4.3206E-02
1.4838E-04
7.5891E-05
4.2982E-02

mat water
1001.06c
8016.06c
5010.06c
5011.06c

7.2216E-02
4.8132E-02
2.4066E-02
3.6487E-06
1.4686E-05

moder lwtr 1001

therm lwtr lwj3.11t
% --- Cross section library file path:
set acelib "/xs/sss_jeff31.xsdata"
% --- Periodic boundary condition:
set bc 3
% --- Neutron population and criticality cycles:
set pop 5000 500 20
% --- Geometry and mesh plots:
plot 3 500 500
mesh 3 500 500
% --- Decay and fission yield libraries:
set declib "/xs/JEFF311RDD"
set nfylib "/xs/JEFF31NFY"
% --- Reduce energy grid size:

11.2 Burnup calculation examples

set egrid 5E-5 1E-9 15.0
% --- Cut-offs:
set fpcut
1E-6
set stabcut 1E-12
% --- Options for burnup calculation:
set bumode
set pcc
set xscalc

% CRAM method
% Predictor-corrector calculation on
% Cross sections from spectrum

% --- Irradiation cycle:
set powdens 38.6E-3
dep butot
0.10000
0.50000
1.00000
1.50000
2.00000
2.50000
3.00000
3.50000
4.00000
4.50000
5.00000
5.50000
6.00000
6.50000
7.00000
7.50000
8.00000
8.50000
9.00000
9.50000
10.00000
10.50000
11.00000
11.50000
12.00000
12.50000
13.00000
13.50000

11.2 Burnup calculation examples

14.00000
14.50000
15.00000
17.50000
20.00000
22.50000
25.00000
27.50000
30.00000
32.50000
35.00000
37.50000
40.00000
% --- Nuclide inventory:
set inventory
922340
922350
922360
922370
922380
922390
932360
932370
932380
932390
942360
942380
942390
942400
942410
942420
942430
952410
952420
952430
952440
952421
962420
962430
962440
962450
962460
962470
962480
962490

972490
972500
982490
982500
982510
982520
360830
451030
451050
471090
531350
541310
541350
551330
551340
551350
551370
561400
571400
601430
601450
611470
611480
611490
611481
621470
621490
621500
621510
621520
631530
631540
631550
631560
641520
641540
641550
641560
641570
641600
% ------------------------------------------------------------

Bibliography
[1] GD Graphics Library. URL www.libgd.org.
[2] S. M. Girard, editor. MCNP – A General Monte Carlo N-Particle Transport Code,
Version 5 Volume I: Overview and Theory. LA-UR-03-1987. Los Alamos National
Laboratory, 2003.
[3] The Message Passing Interface (MPI) Standard. URL
www-unix.mcs.anl.gov/mpi/, reviewed March 2006.
[4] OECD/NEA Data Bank Home Page. URL www.nea.fr/html/databank/.
[5] R. E. MacFarlane and D. W. Muir. The NJOY Nuclear Data Processing System.
LA-12740-M. Los Alamos National Laboratory, 1994.
[6] V. McLane, editor. ENDF-102, Data Formats and Procedures for the Evaluated
Nuclear Data File ENDF-6. BNL-NCS-44945-01/04-Rev. Brookhaven National
Laboratory, 2001.
[7] N. Messaoudi and B.-C. Na. VENUS-2 MOX-fuelled Reactor Dosimetry Calculations,
Final Report. NEA/NSC/DOC(2005)22. OECD/NEA, 2004.
[8] J. Leppänen. Randomly Dispersed Particle Fuel Model in the PSG Monte Carlo
Neutron Transport Code. In Proc. Joint International Topical Meeting on
Mathematics & Computation and Supercomputing in Nuclear Applications (M&C +
SNA 2007). Monterey, California, April 15–19 2007.
[9] F. Brown et al. MCNP Calculations for the OECD/NEA Source Convergence
Benchmarks for Criticality Safety Analysis. LA-UR-01-5181. Los Alamos National
Laboratory, 2001.
[10] J. Ueki and F. B. Brown. Stationarity and Source Convergence in Monte Carlo
Criticality Calculation. LA-UR-02-6228. Los Alamos National Laboratory, 2002.
[11] T. Yamamoto, T. Nakamura and Y. Miyoshi. Fission Source Convergence of Monte
Carlo Criticality Calculations in Weakly Coupled Fissile Arrays.
J. Nucl. Sci. Technol., 37 (2000) 41–52.
[12] R. N. Blomquist et al. Source Convergence in Criticality Safety Analyses, Phase I:
Results for Four Test Problems. NEA No. 5431. OECD/NEA, 2006.

[13] The NEA Expert Group on Source Convergence in Criticality-Safety Analysis. URL
www.nea.fr/html/science/wpncs/convergence/, Reviewed March
2007.
[14] F. B. Brown. On the Use of Shannon Entropy of the Fission Distribution for Assessing
Convergence of Monte Carlo Criticality Calculations. In Proc. PHYSOR-2006
American Nuclear Society’s Topical Meeting on Reactor Physics Organized and
hosted by the Canadian Nuclear Society. Vancouver, BC, Canada, Sept. 10–14 2006.
[15] J. Leppänen. Development of a New Monte Carlo Monte Carlo Reactor Physics Code.
D.Sc. thesis, Helsinki University of Technology, 2007. VTT Publications 640.
[16] J. Leppänen. Two practical methods for unionized energy grid construction in
continuous-energy Monte Carlo neutron transport calculation. Ann. Nucl. Energy, 36
(2009) 878–885.
[17] B. Becker, R. Dagan and G. Lochnert. Proof and implementation of the stochastic
formula for ideal gas, energy dependent scattering kernel. Ann. Nucl. Energy, 36
(2009) 470–474.
[18] J. Leppänen. Performance of Woodcock Delta-Tracking in Lattice Physics
Applications Using the Serpent Monte Carlo Reactor Physics Burnup Calculation
Code. Ann. Nucl. Energy, 37 (2010) 715–722.
[19] D. E. Cullen et al. Static and Dynamic Criticality: Are They Different?
UCRL-TR-201506. Lawrence Livermore National Laboratory, 2003.
[20] J. Leppänen. Current Status of the PSG Monte Carlo Neutron Transport Code. In
Proc. PHYSOR-2006 American Nuclear Society’s Topical Meeting on Reactor Physics
Organized and hosted by the Canadian Nuclear Society. Vancouver, BC, Canada,
Sept. 10–14 2006.
[21] M. Pusa and J. Leppänen. Computing the Matrix Exponential in Burnup Calculations.
Nucl. Sci. Eng., 164 (2010) 140–150.

Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.4 Linearized : No Page Count : 164 Page Mode : UseOutlines XMP Toolkit : XMP toolkit 2.9.1-13, framework 1.6 About : 57fa388d-4da8-11f0-0000-55da2356b91 Producer : dvips + GPL Ghostscript 9.04 Keywords : () Modify Date : 2015:06:18 10:29:46+03:00 Create Date : 2015:06:18 10:29:46+03:00 Creator Tool : LaTeX with hyperref package Document ID : uuid:57fa388d-4da8-11f0-0000-55da2356b91 Format : application/pdf Title : () Creator : () Description : () Subject : Author :
EXIF Metadata provided by EXIF.tools

Navigation menu

Upload a User Manual

Versions of this User Manual:

Wiki Guide

HTML

Mobile

Download & Help

Views

User Manual

Discussion / Help

Navigation

© 2024 UserManual.wiki

Contact Us

DMCA