JKCS1_prepare

JKCS1_prepare uses the information from input.txt to create SYS_{system} subfolders in your working directory. Each SYS_{system} subfolder corresponds to a specific cluster composition. JKCS1_prepare also creates parameters.txt inside each of these subfolders. parameters.txt file is similar to input.txt and is used further for additional calculations in each subfolder. JKCS1_prepare has however added all possible combinations of conformers and conjugated acids/bases of monomers, which fulfil the composition and total charge as defined in input.txt, to the end of parameters.txt. User does not have to check those but these monomer combinations will be used in the configurational sampling. JKCS1_prepare does not require any arguments.

Hint

For help use:

JKCS1_prepare -help

The folder/subfolders tree should during configurational sampling look like this:

filesstructure

Hint

Each subsequent script (e.g. JKCS2_explore or JKCS3_run) can either be called from the working directory or from one of the SYS_{system} subfolders. When called from a subfolder, it is only applied to that specific subfolder. When called from the working directory, the script is applied to all subfolders. However, you can also specify to which subfolder you want to apply scripts. See examples:

JKCS3_run -p XTB -m “--opt --gfn 1” SYS_2sa_*am
JKCS4_collect XTB SYS_2sa_1am

if you want to perform a specific command (or sequence of commands) in each subfolder, use:

JKfor head -n 3 resultsXTB.dat
JKfor "head -n 1 resultsDFT_freq.dat > ../all_results.dat"
JKfor 'head -n 50 resultsXTB.dat > resultsXTB_FILTERED.dat; JKCS3_run -p XTB -rf XTB -nf XTB_freq -m "--hess --gfn 1"'

EXAMPLE: Calling JKCS1_prepare for a negatively charged cluster of 1 ammonia and 2 sulfuric acid molecules, SYS_2am_1sa folder would be created. This folder would contain parameters.txt where 5 possible cluster combinations are written:

  • 0_1_0_0_2_0 (NH4+)1(HSO4-)2

  • 0_1_0_1_0_1 (NH4+)1(cis-H2SO4)1(SO4−-)1

  • 1_0_0_1_1_0 (NH3)1(cis-H2SO4)1(HSO4−)1

  • 0_1_1_0_0_1 (NH4+)1(trans-H2SO4)1(SO4-−)1

  • 1_0_1_0_1_0 (NH3)1(trans-H2SO4)1(HSO4−)1

You can easily verify that these clusters all have total charge −1 and that these are all the combinations that can be made for this specific composition and total charge.

Arguments

-s, -sample <integer>

sample from all monomer combinations. In the case of many combinations, you can pick just random combinations (helpful when molecules with a lot of conformers are used). For instance:

JKCS1_prepare -s 100
JKCS1_prepare -s NoC/2

Note

You can use variables “NoC” and “M” to define the parameter as a function of “Number Of (monomer) Combinations” or “(total) number of Molecules”.

However, I recommend lowering the simulation parameters and using all combinations. In the case of 200 conformer combinations, the following two options would lead to a similar computational cost, but the second one would lead to a more thorough configurational search:

# REDUCING CONFORMER COMBINATIONS:
JKCS1_prepare -s 100
JKCS2_explore -pop 100 -gen 200 -lm 500
JKCS3_run

# REDUCING SIMULATION LENGTH (200 NoC)
# -jpt 2 = each submitted job performs 2 simulations/calculations
JKCS1_prepare
JKCS2_explore -pop 100/2 -gen 200/2 -lm 500/2 -jpt 2
JKCS3_run -jpt 2
-o, -overwrite

overwrites parameters.txt in already existing subfolders. Not important for beginners.