Surface reaction does not proceed

  • alicankizilkaya
  • Topic Author
  • Offline
  • User
  • User
More
4 months 1 week ago #17 by alicankizilkaya
Surface reaction does not proceed was created by alicankizilkaya
Dear Admin,

I have been trying to simulate CO methanation on a Co(111) surface (CO + 3 H2 -> CH4 + H2O ). In the simulation input file, the ref. states are the clean slab, CO_gas, H2_gas and O2_gas. The energetics input and the mechanism input are prepared according to the DFT results obtained on this surface. When I try to run a KMC simulation, I only get CO and H2 (dissociative) adsorption on the surface. However, there is no HCO formation, which is necessary for the formation of products ( Input and output files are attached in data1.zip). As I thought that this may be related to the pre-exponentials of the reverse reactions, I tried to make the CO & H2 adsorption and HCO formation irreversible, and then I only get a blank line in specnum_output.txt and this time there is not even the adsorption of CO and H2 (input and output files are attached in data2.zip). 

I checked the manual and all tutorials carefully but I am stuck at this issue because no surface reaction occurs in Zacros (and with this energetics and the simulation T and P, it is known to occur for sure). Could you please indicate what is the issue with my simulation (and what does a single blank line of zeros indicate in the specnum_output.txt) ?

Thanks a lot for your support,
Best regards,
Dr. Ali Can Kizilkaya
Attachments:

Please Log in or Create an account to join the conversation.

More
4 months 6 days ago - 3 months 4 weeks ago #18 by hprats
Replied by hprats on topic Surface reaction does not proceed
I can see in your procstat_output.txt file that you do get HCO formation, in fact about 50% of the steps correspond to "HCO_formation_fwd" but the other half are "HCO_formation_rev". It seems that, after an HCO species is formed, it quickly dissociates back to CO+H because the rate constant of this process is very fast. Therefore, your lattice ends up poisoned by CO&H species and most of the simulation time corresponds to forward + reverse HCO formation. Your general_output.txt file shows that the rate constant for HCO_formation_rev is 14 and 11 orders of magnitude higher than that of HCO_formation_fwd and HCO_dissociation_fwd, respectively, so the overall reaction can not proceed. 
As you suggested, the problem seems to be related to the pre-exponential factors. May I ask, why have you chosen e.g. pre_expon = 2.0E-1 and pe_ratio = 1.0 for CO_adsorption step? I can see that you have declared most of the pre-exponential factors to be 1.0E+13, which ignores any thermal contribution to the free energy. I recommend you to calculate them from TST (e.g. see the supporting information of  doi.org/10.1063/1.3596751 ). Most importantly, some pre-exponential factors seem to be many orders of magnitude off from the typical values (e.g. 2.0E-1 for CO_adsorption_rev, when values closer to 1E+13-1E+17 are expected, or 1.0E+18 for HCO_formation_rev, which for consistency should be 1.0E+13).
Regarding your second simulation with irreversible processes, a single line of zeros in the simulation_output.txt in this case indicates that the simulation has reached the maximum allowed wall-clock (real) time of 30 seconds before taking the first snapshot of the lattice, which you specified to take at intervals of 1 second of KMC time. Note that, when your simulation finished, the current KMC time was 3.34E-02 (you can see this information in the general_output.txt). After correcting your pre-exponential factors, you might want to modify the stopping criteria and the sampling frequency. 
 
Last edit: 3 months 4 weeks ago by hprats.

Please Log in or Create an account to join the conversation.

Time to create page: 0.270 seconds

Sorry, this website uses features that your browser doesn’t support. Upgrade to a newer version of Firefox, Chrome, Safari, or Edge and you’ll be all set.