Manual ROOH decomposition - how to add it?

[DetlevCM]

Dear Developers,

as I cannot seem to get a handle on the database format for RMG, I would like to ask for some assistance:

ROOH -> RO. + .OH

How can I add this reaction to the library so that I can call it as a user defined library with user defined parameters?

For reference:
Denisov, 2003 - a book
Hydroperoxide decomposition has a comparatively high activation energy in the gas phase, lower in the liquid phase but it also catalysed by many species (sulfur, metals, etc.).

Many thanks
Detlev

[connie]

If you want to put a specific reaction into the database, you can place it into a reaction library by creating a new folder in RMG-Java/databases/RMG_database/kinetics_libraries/. The name of the folder should also be used in your condition file like so:

ReactionLibrary:
Name: NameOfLibrary
Location: NameOfFolder
END

The folder should contain the 3 files: reactions.txt pdepreactions.txt species.txt

Please take a look at any of the existing libraries to find the format for these files (i.e. https://github.com/GreenGroup/RMG-Java/tree/master/databases/RMG_database/kinetics_libraries/wang/)

The species.txt file contains the species dictionaries and names. In the reactions.txt file, write the reaction using the names found in the species.txt file, a reaction's format is:

R1 + R2 = P1 + P2 A n E dA dn dE

If you instead are wanting to create a general rule for ROOH --> RO + OH, then you will have to modify the reaction family R_Recombination, which is actually defined in the reverse direction. Which of these did you need?

[DetlevCM]

Dear Connie,

thank you for getting back to me.

I'd prefer to have the reaction in a general form as a user library - however that might not be possible.
I was trying to find it in the R_Recombination library but the syntax beat me - I understand bits of it but cannot find it...

My reasoning for having it as a user defined library is that Denisov suggests the activation energy is comparatively high in the gas phase and lower, but still high in the liquid phase.
However the decomposition is catalysed - where the parameters would depend on the catalyst employed. So a user library would be more beneficial to testing and tweaking parameters which differ depending on the additive employed.

[connie]

If you want to create a general ROOH --> RO + OH decomposition rate, you would have to alter the R_Recombination rate library. Right now that family is entirely filled with gas phase rates. One thing you could potentially do is substitute the entire library with liquid-phase catalyzed rates. However, the rest of the reaction families will still use gas phase kinetics. Basically, there is no good way to implement what you want to do, unless you made alterations to the code itself.

[DetlevCM]

Hmm, that sounds problematic - I don't have the time to work out all parameters...

Having said that though, my understanding of RMG is that it does a liquid phase correction by itself. So I guess for the remaining parameters it would not cause any issues as long as one does not want any catalytic effects.

Rewriting the code is out of the question (for now - no idea what happens in the future).

I also wonder if there is any better documentation on the format than the manual? Because I cannot make head or toe of it...

[connie]

Database modification documentation can be found here: http://rmg.sourceforge.net/documentation/modification.html

The one for RMG-Py can be found here: http://greengroup.github.io/RMG-Py/users/rmg/database/modification.html

I'm not sure too many people know what the liquid phase kinetics correction is inside RMG-Java.
You can create some rate rules just for your reactions but if they may have side effects on other kinetics due to averaging.

You can try to find out what groups you should design rate rules for by using the search kinetics function here: http://rmg.mit.edu/database/kinetics/search/ Just take a representative real reaction and search for the kinetics generated by rate rules. The estimate should tell you what the most specific nodes are in the tree for that reaction, and that should be the rate rule you need to make. You can also find visual links to the groups here: http://rmg.mit.edu/database/kinetics/families/R_Recombination/groups/