This is the second project for the TALENT course: nuclear theory for astrophysics (2014).
- Analysis: Fei Yuan
- Code: Amber Lauer
- Parameter: Tsunghan Yeh
Modelling X-ray burst nucleosynthesis with the XNet reaction network solver, following the model of Fisker et al (2007): the premade model for a CNO X-ray burst.
Firstly, be sure to run the following script to generate Makefile using
Makefile.in as input:
./configure
The makefile can do several things:
-
download and unpack ReduceReaclib and XNet if necessary
-
compile ReduceReaclib and XNet
-
run XNet and ReduceReaclib with the parameters in:
th: thermodynamic trajectoryab: initial abundancecontrol: control filesunet: nuclide list
Note: XNet is run inside a subdirectory under runs.
Commonly used targets in the makefile are:
-
all: compile bothreducereaclibandxnet -
reducereaclib: compile ReduceReaclib -
xnet: compile XNet -
clean: delete the entiredistdirectory -
run: equivalent torun-xnet -
run-xnet: run XNet -
run-reducereaclib: run ReduceReaclib
When compiling, the code for XNet and ReduceReaclib are automatically
downloaded to the dist directory.
For example, to run XNet in runs/my_run:
make RUN_DIR=my_run run
Here are the input formats inferred from the Fortran code as well as the sample input files. The formats here are shown using an EBNF-like notation.
The ... indicates that Fortran will ignore all trailing garbage on that line
so feel free to put whatever in there.
<control> ::=
"## Problem Description" "\n"
<str:description1> "\n"
<str:description2> "\n"
<str:description3> "\n"
"## Job Controls" "\n"
<int:initial_zone> ... "\n"
<int:num_of_zones> ... "\n"
<int:include_weak_reactions> ... "\n"
<int:include_screening> ... "\n"
<int:process_nuclear_data_at_runtime> ... "\n"
"## Integration Controls" "\n"
<int:choice_of_integration_scheme> ... "\n"
<int:max_num_of_timesteps_before_quit> ... "\n"
<int:max_iterations_per_step> ... "\n"
<int:convergence_condition> ... "\n"
<float:max_abundance_change_per_timestep> ... "\n"
<float:smallest_abundance_used_in_timestep_calculation> ... "\n"
<float:mass_conservation_limit> ... "\n"
<float:convergence_criterion> ... "\n"
<float:lower_abundance_limit> ... "\n"
<float:max_factor_to_change_dt_in_a_time_step> ... "\n"
"## Output Controls" "\n"
<int:diagnostic_output_level> ... "\n"
<int:per_timestep_output_level> ... "\n"
... "\n"
<str:ascii_output_filename_root> "\n"
... "\n"
<str:binary_output_filename_root> "\n"
... "\n"
<str:nuclide>{repeat 14 times} "\n"
"## Input Controls" "\n"
... "\n"
<str:data_dir> "\n"
... "\n"
(<str:ab_filename> "\n"
<str:th_filename> "\n){repeat as many times as num_of_zones}
<ab> ::= <str:description> "\n"
(<item> <item> <item> <item> "\n")*
<item> ::= <str:nuclide> <float:abundance>
<th> ::= <str:description> "\n"
<float:start_time> ... "\n"
<float:stop_time> ... "\n"
<float:dt_initial> ... "\n"
<row>*
<row> ::= <float:time> <float:temperature> <float:density> ... "\n"
<sunet> ::= (<str:nuclide> "\n")*
Note that <str:nuclide> must be right-aligned and occupy exactly 5
characters.
Note: most of the Python codes require Numpy, and some will also require Matplotlib for visualization.
Python module that reads the binary output files ("tso" files) from XNet.
This is based on a script originally written by Kevin Siegl.
Reads a tso file and writes the abundances (as well as time) in an ASCII format. Nuclides that don't participate at all are excluded (i.e. sum of abundances less than a certain threshold).
Dependencies:
tso_reader.pyelements.py
Reads a tso file and plots the abundances against time.
Dependencies:
tso_reader.pyelements.py
Reads a tso file and plots the abundances on the nuclear chart as an animation.
Dependencies:
chart-allchart-stabletso_reader.py
Note: requires ffmpeg to be installed, though you can also change the
encoder used by Matplotlib by modifying the script.
This is based on a script originally written by Kaitlin J. Cook. Nuclear data
was provided by Kaitlin J. Cook (chart-stable) and Alison Dreyfuss
(chart-all).