Firstly, thank-you for using MESMER and apologies for my slow response.
I have looked at the input file that you sent me and in particular the thermochemistry. Some of the numbers are quite large which is not a problem in itself but to get a clear picture of the thermochemistry I shifted the baseline so that the HEX1 species is at the energy origin and found the following:
The Hex1 reactant appears to be an ozonide. Give this there seems to be some large differences here in the product values, except for the isomerization, are these expected? Some of the transition states also appear to be below the reactants or products, is this expected?
With regards, Struan
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thank you for this first analysis. You are right, HEX1 is an ozonide. Since this is our first application
of Mesmer, we omitted, for the time being, the first step leading to HEX1 in order to reduce the
complexity of our input. That reaction will be added eventually. One of our interests is the product
ratios. Our data files are a priori correct, verified by IRC.
TS1 and TS2 have higher energies than HEX1, the intermediate I, for example, is lower. Its concentration [I] should start at zero and have disappeared at the end of the reaction, which is
not what we get.
A perhaps minor problem is that the graphical representation of the energy profiles does not show up. And then, how can we set the energy origin?
Thanks for your help.
Best wishes,
Alex
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I attach an updated input file. I have altered the thermochemistry so that that the energy origin is Hex1 on input as indicated in previous email. This still caused problems for the visualization in FireFox. I think the problem was that the colossal energy of the product channels (e.g. 5.0E+05 kJ/mol) was compressing the figure so that one could not see anything. If these channels are really this high in energy, I doubt that they would contribute much, however, I doubt these bonds are that strong, so to get a better view I reduced these channels by a factor of a 1000. I also fixed some of the reaction definitions. You should now be able to view the surface.
Please ignore my last email, I see made a silly mistake in calculating the energies of the product pairs which I shifted individually but should have altered in pairs. Having corrected this thermochemistry now makes sense, my apologies. After a few other changes, I located the actual problem was the precision of the calculation, the low temperature meant that ordinary double precision was insufficient to capture the span of the eigenvalues. I have altered this to double-double. I now get better results, or at least no negative concentrations. I observe that at this low temperature most of the reaction ends up in the low energy isomer. I attach a .zip file with the results.
Dear Mesmer developers,
We are new to Mesmer, and are trying to solve the problem
A --> P1+P2
A <--> I
I --> P2 + P3
I --> P4 + P5
Unfortunately, we get huge positive or even negative values for the intermediate I. Could you help us please.
We attach the xml file.
Thank you very much,
Alex
Last edit: Alexander Alijah 2022-10-07
Dear Prof. Alijah,
Firstly, thank-you for using MESMER and apologies for my slow response.
I have looked at the input file that you sent me and in particular the thermochemistry. Some of the numbers are quite large which is not a problem in itself but to get a clear picture of the thermochemistry I shifted the baseline so that the HEX1 species is at the energy origin and found the following:
HEX1 -1405140.00 0.00E+00
I -1405380.00 -2.40E+02
P1 -898494.40 5.07E+05
P2 -507055.79 8.98E+05
P3 -898492.20 5.07E+05
P4 -701387.14 7.04E+05
P5 -704620.00 7.01E+05
TS1 -1405040.00 1.00E+02
TS2 -1405060.00 8.00E+01
TSa -1405170.00 -3.00E+01
TSk -1405170.00 -3.00E+01
The Hex1 reactant appears to be an ozonide. Give this there seems to be some large differences here in the product values, except for the isomerization, are these expected? Some of the transition states also appear to be below the reactants or products, is this expected?
With regards, Struan
Dear Struan,
Thank you for this first analysis. You are right, HEX1 is an ozonide. Since this is our first application
of Mesmer, we omitted, for the time being, the first step leading to HEX1 in order to reduce the
complexity of our input. That reaction will be added eventually. One of our interests is the product
ratios. Our data files are a priori correct, verified by IRC.
TS1 and TS2 have higher energies than HEX1, the intermediate I, for example, is lower. Its concentration [I] should start at zero and have disappeared at the end of the reaction, which is
not what we get.
A perhaps minor problem is that the graphical representation of the energy profiles does not show up. And then, how can we set the energy origin?
Thanks for your help.
Best wishes,
Alex
Dear Alex,
I attach an updated input file. I have altered the thermochemistry so that that the energy origin is Hex1 on input as indicated in previous email. This still caused problems for the visualization in FireFox. I think the problem was that the colossal energy of the product channels (e.g. 5.0E+05 kJ/mol) was compressing the figure so that one could not see anything. If these channels are really this high in energy, I doubt that they would contribute much, however, I doubt these bonds are that strong, so to get a better view I reduced these channels by a factor of a 1000. I also fixed some of the reaction definitions. You should now be able to view the surface.
I hope this helps a little.
With regards, Struan
Dear Alex,
Please ignore my last email, I see made a silly mistake in calculating the energies of the product pairs which I shifted individually but should have altered in pairs. Having corrected this thermochemistry now makes sense, my apologies. After a few other changes, I located the actual problem was the precision of the calculation, the low temperature meant that ordinary double precision was insufficient to capture the span of the eigenvalues. I have altered this to double-double. I now get better results, or at least no negative concentrations. I observe that at this low temperature most of the reaction ends up in the low energy isomer. I attach a .zip file with the results.
With regards, Struan
Dear Struan,
Thank you very much for your help.
Best wishes,
Alex