Events set up

[Andrew Weller]

Hi All

I am a molecular chemist trying to simulate some kinetics a reaction that also releases H2 in an organometallic catalysis reaction. I have some beautiful holistic fits over 8 data sets, but the issue I have is that H2 does not remain in the solution - it builds up to 1 bar above the solution and then says steady. I would like to set an EVENT that when H2(dissolved) reaches the solubility limit (~0.0056M from Henry's constant) it remains at this concentration (as the pressure of H2 above the solution will be at 1 bar very quickly and remain there as we are using a Eudiometer).

The issue I have is that I can set an EVENT (see below) but this fires just once early on and then the H2 concentration in the model continues to increase. Is there a way of the event being triggered continuously - i.e. looping back to check the trigger expression.

 

Trigger Expression [H2]>0.0056

Target [H2]

Expression 0.0056

Thanks

Andy

[Frank Bergmann]

Hello Andy, 

in theory, the events should already be triggered continuously. I'd probably set the assignment a bit below the trigger threshold (as for the root finder the difference between >= and > are philosophical). Feel free to send me the model and I can have a look (from the technical point of view). 

here an example model, showing that expressions of that form do trigger continuously. 

best

Frank

[Hoops, Stefan (sh9cq)]

Hello Andy,

Events are discrete and thus cannot be continuously triggered. You can
get very close to 'continuous' triggering by the approach Frank
suggests. In your situation I would modify the reaction which creates
the soluable H2. This process should stop when then the solution limit
is reached.

Thanks,
Stefan


On Fri, 2025-01-24 at 06:58 -0800, 'Andrew Weller' via COPASI User

--
Stefan Hoops, Ph.D.
Research Associate Professor
Biocomplexity Institute & Initiative
University of Virginia
995 Research Park Boulevard
Charlottesville, VA 22911

Phone: +1 540 570 1301
Email: sho...@virginia.edu

[Andrew Weller]

Hi Stefan

I will try Franks' approach. I cannot modify the reaction that produces H2 as this will complete change the kinetics. H2 generation (and this inhibition) a real problem in modelling in catalysis and this EVENT mechanism could be a neat way around it.

Andy

[Andrew Weller]

Frank

That seems to be working at keeping the [H2] constant (I have set the trigger to be 0.001 higher than the Expression . Thanks!

But very very slow. Any tips on which is the best (fast) method to use to quickly iterate?

Andy

[Andrew Weller]

Andrew Weller

06:53 (now) 

to COPASI User Forum

Dear Frank and Stefan

With immense thanks to you both. I have now got it working well, and a stable fit. 

To summarise: I wanted to mimic the loss of H2 from a catalytic reaction that produced H2, and in which H2 also acted in an inhibitory manner. As the reaction was run under eudiometric conditions there is a rapid establishment of a constant H2 pressure above the reaction solution (which s THF). This would-be about 1 atm. So using Henry's Constant for H2/THF I can estimate the max. steady state of H2 available in solution. Normally under conditions where H2 is lost to the atmosphere (or removed by a flow of N2) removing H2 in COPSAI is easy (I use H2 -> Z, and set the rate constant appropriately high). However I needed a set of condition where H2 built up then equilibrated at a certain conc. I used EVENTS to do this, and it works well in looping round. 

Events

Trigger = [H2]>0.05

Target = [H2]

Expression = 0.045          <here I found that if I made the expression too close to the target the simulation ran very slowly and also sometimes did not work, I am sure this can be refined>

Method: Genetic Algorithm SR worked well (as does Evolutionary Programming).

The fits (over 8 separate experiments refined holistically) with the added constraint of the [H2] work out well (see attached). Still room for improvement but the model now reflects far more accurately the actual system.

Thanks again for your speedy replies and help.

Andy