Messages

Hi, everyone.
I am trying to get the breakthrough curve for CO2/H2 mixture in a MOF.
I read and followed the instructions present in "S. Sharma, S. Balestra, R. Baur, U. Agarwal, E. Zuidema, M. Rigutto, S. Calero, T.J.H. Vlugt, and D. Dubbeldam, Molecular Simulation Journal, 49(9), 2023". And everything seems right, however, the breakthrough curve time appears to be overestimated. I saw in different papers (theoretical studies GCMC) that CO2 usually start under 50 (Time, t/[min.]) or under 600 (Dimensionless time)(experimental studies), but my results (with isotherms obtained with GCMC/RASPA and fitted with RUPTURA) start around 400 (Time, t/[min.]) or close to 1700 (Dimensionless time). 
I already used several approaches combined: SIAST, IAST, EI, Langmuir, Langmuir-Freundlich, the number of isotherm sites, changes in column entrance velocity, and more. But the results that I get are weird. Curiously, when I tried repeating the examples provided in GitHub, I was able to get the same good results.
I am new in the RUPTURA use, so if someone has some tips, I would appreciate it.
(Attached is one between 21 tries done until now)

Hello everyone!
I am having the same problem when trying to use RASPA 2.0.47. At the time, I am running calculations in a highthroughput screening scheme. However, only a part of the .cif files that constitute the group of structures I am studying are calculated. The other structures fall into a loop that returns the following calculation status in a slurm file:
"
[...]
unknown: 1,000
unknown: 1,000
unknown: 1,000
End reading cif-file
Shift all potentials
'force_field.def' file not found and therefore not used
Realloc MaxNumberOfCoulombicSites: 2048
realloc(): invalid next size
_cell_length_a: 7.119000
_cell_length_b: 21.580000
_cell_length_c: 9.100000
_cell_length_alpha: 90.000000
_cell_length_beta: 90.000000
_cell_length_gamma: 90.000000
_symmetry_space_group_name_H-M: P 1 found space group: 1
_symmetry_Int_Tables_number: 1
space group found from symmetry elements: 1 (nr elements: 1)
unknown: 1,000
unknown: 1,000
[...]
unknown: 1,000
unknown: 1,000
unknown: 1,000
End reading cif-file
Shift all potentials
'force_field.def' file not found and therefore not used
free(): invalid pointer
Realloc MaxNumberOfCoulombicSites: 2048
realloc(): invalid next size
free(): invalid pointer
Realloc MaxNumberOfCoulombicSites: 2048
realloc(): invalid next size
free(): invalid pointer
free(): invalid pointer
Realloc MaxNumberOfCoulombicSites: 2048
realloc(): invalid next size
free(): invalid pointer
"
And then the calculation crashes in an abnormal termination.

I am a beginner with RASPA. I looked in the manual, chapter 5 "the source code" and found no mention of the part of the code that deals with the subject raised (MaxNumberOfCoulombicSites).

Can you tell me where, in which file contained in the /src/ folder I can find the code, and what is the best way to modify it so that I can overcome this problem?

Thank you in advance, thank you very much.

Hi Prof. Dubbeldam and dear RASPA Community.

I am a beginner in using RASPA.
I am trying to simulate adsorption on a MOF with Monte Carlo, and I am following the manual and the RASPA repository on github as guidance. In order to simulate the structure as rigid, I follow the instructions in the manual in the "Framework parameters" section, so I adopt the parameters:

Framework 0
FrameworkName calf-20
UnitCells 1.0 1.0 1.0
FlexibleFramework no
FrameworkDefinitions framework.def

The last parameter was adopted by recommendation of the manual, which states: "The file is read even when 'FlexibleFramework no' is specified (the reason is that framework bond-dipoles are defined using the 'framework.def' file)".

However, the manual does not contain examples or further details about the framework.def file when the intention is to simulate a rigid structure.
In the RASPA github repository, the examples show the specifications of simulation.input and framework.def for adsorption analysis only when the structure is flexible, thus leaving no further information on how to simulate adsorption in rigid structures.

In an attempt to find a solution, my most reasonable guesses were: 1. Consider that the definition of the rigid structure of the framework described in the content of framework.def was exactly the same as the rigid structure of the molecule, in molecule.def. 2. Consider 0 in all #CoreShells bond BondDipoles UreyBradley bend inv tors configurations improper-torsion bond/bond bond/bend bend/bend stretch/torsion bend/torsion, as stated in the framework.def file used as an example for flexible structures.
Nothing worked.

These and countless other attempts proved fruitless, and most of the time they resulted in the same error log: "Error: Unknown group type (should be 'Flexible' or 'Rigid')", except for guess 1, described above, in which I got the error: "ReturnPseudoAtomNumber: Error!!!! :0.22394", referring to the corresponding value of "Acentric factor [-]", as stated in the edited framework.def file, taking as an example the file for the rigid molecule CO2.def, which is in the RASPA github repository.

With no further ideas on how to proceed, I come to ask for your help, a long-time RASPA user, on how to proceed and solve this problem.

How can I correctly define the parameter to establish that the MOF structure used in the adsorption simulation calculated with Monte Carlo is in fact rigid?

Thank you in advance.

Regards,
Antonio Douglas