krome-CSE

The chemistry code KROME is in the process of being made compatible with our AGB circumstellar envelope (CSE) chemistry, usually calculated using the UMIST Rate22-CSE code, see also this link.

Aim: Couple KROME to PHANTOM to run 3D hydro-chem model of AGB outflow.

Why KROME?: KROME is implemented in such a way that it can deal with running in parallel, which is needed for a coupling with PHANTOM.

---

Notes

• KROME was once coupled to PHANTOM by W. Homan, see code here. --> Didn't work as it should, chemically speaking.

• Chemical networks were added to KROME for AGB outflows by J. Bolte, see details here.

---

How to run?

Clone using the recursive option: git clone --recursive, to also clone the KROME package.

1. Run the bash script

build_UMIST.sh chem_netw

with a chemical network as argument.

2. Run the bash script

make_CSEkrome.sh

This will make an executable for file CSE_run_krome.f.

3. Run the bash script

run_CSE_krome.sh

with a inputfile as argument to properly run the executable.

---

Updates

(Latest first)

• 12/09/'24:

  Added benchmark to results from ChemPy, see figure below.

  Discrepancy in CO between UMIST model and ChemPy model is due to CO self-shielding; this is not present in the ChemPy model.

• 06/09/'24:

  Added test input file so that users can do a test run.

  Benchmark: Still something wrong with (probably) the density/photodissociation.

• 06/08/'24:

  1. Found the first bug in the system!! Because of a typo (on my side), the cosmic ray reaction rates were way to high. Fixed this.

  2. In our setup, every timestep the density is lower ($\frac{1}{^2}$-law in density). Hence, we cannot use the -useN flag (using number densities), but have to use the -useX flag (fractions). This fixes a lot!

  --> Results now are getting decent :D

  To do: The photodissociation know seems to set in too late, and too few electrons. See figure, dotted lines are the abundances calculated by the UMIST Rate22-CSE code ("absolute truth"), full lines are the abundances calculated by KROME. The dashed lines are results for the same physical setup by ChemPy.

• 31/07/'24:

  Reading in parent species file + physical input parameters works. Initialising KROME works. Saving output of KROME works.

  KROME chemical evolution does NOT coincide with our UMIST Rate22-CSE code. We (Mats Esseldeurs and me, Silke Maes) think the issue lies within the cosmic ray reaction of H2. In the UMIST Rate22-CSE code it is assumed that H2 is fully self-shielding, so that it stays roughly constant throughout the CSE. At the moment, this assumption is not yet realised in the kromeCSE code.

  Note to self: Abundances (resulting from KROME) change when using -useX (mass fractions) or -useN (number density). Check later why!

• 12/07/'24:

  Correct rates, including radiation parameters $A_V$ and $\xi$ implemented (more info see Maes et al. 2024), via script

  umist2krome_custm.py
  

  originally written by Tommaso Grassi.

• 08/07/'24:

  KROME runs, but the radiation component for the photodissociation reactions does not work yet.