Hydrazine pyrolysis #296

HiroumiTani · 2018-09-27T17:29:06Z

To avoid mistakes, I'm doing the PR by two steps. This PR added new reactions to primaryNitrogenLibrary but did't change the rate constants of other reactions.
The reactions of NH2+N2H3 were commented out to carry out make test-database.

alongd

Looking good.
I only reviewed the format and made a couple of comments, I'll take a look at the rates soon.

alongd · 2018-09-27T23:58:21Z

input/kinetics/libraries/primaryNitrogenLibrary/reactions.py

+""",
+)
+
+#entry(


Let's remove these commented entries from this PR. We also have a reasonable rate for NH2 + N2H3 <=> H2NN(S) + NH3 from D&B, so we might leave these out.

alongd · 2018-09-28T00:00:23Z

input/thermo/libraries/primaryNS.py

@@ -1665,8 +1665,8 @@
    molecule =
 """
 1 N u0 p1 c0 {2,S} {4,S} {5,S}
-2 N u0 p1 c0 {1,S} {3,S} {6,S}
-3 N u0 p2 c0 {2,S}
+2 N u0 p0 c+1 {1,S} {3,D} {6,S}


Could you also change the SMILES in the description below?

Thanks. I corrected it.

HiroumiTani · 2018-09-28T14:32:34Z

I replaced the rate constant of NH2+N2H3=H2NN(S)+NH3 with that proposed by DeanBozz. Also, I removed NH2+N2H3=H2NN(T)+NH3. This reaction with the same rate constant of DeanBozz will be added by RMG as a rate rule.

alongd · 2018-10-22T16:39:21Z

The branch was updated with the following changes:

Added various NxHy rates to the primaryNitrogenLibrary, some of which turned into H_Abs training reactions
Updated forbidden structures in H_Abstraction to address multiple bonds and birads cases
Added two new kinetics families, 1,2_NH3_elimination and 1,3 NH3_elimination with training reactions
Added NxHy thermo to the primaryNS library
Added some NxHy globally forbidden structures
Corrected some of the weird training reactions labelling in H_Abs

ReactionMechanismGenerator/RMG-Py#1496 should be merged before this PR

alongd · 2019-01-28T03:51:45Z

I think this PR is now ready for review.
@mjohnson541, could you take a look and let me know if you have comments?

mjohnson541

I have some general comments, but other than that this looks fine.

mjohnson541 · 2019-02-01T15:57:38Z

input/kinetics/families/1,2_NH3_elimination/groups.py

+entry(
+    index = 0,
+    label = "NNHNH2",
+    group = "OR{non_charged, charged}",


Is it possible to not use an OR?

mjohnson541 · 2019-02-01T16:03:15Z

input/forbiddenStructures.py

+    shortDesc = u"""""",
+    longDesc =
+u"""
+Geometry could not converge at CBS-QB3


I trust you guys have considered the implications of these forbidden groups, but it's worth noting that an inability to converge a proper geometry doesn't mean a structure couldn't be an intermediate in a concerted reaction RMG takes multiple steps to do.

I haven't thought about that. Does it hold even if I define a forbidden Species and not a Group?

Forbidden species would be less risky because they only apply to that one species. The issue isn't that common, but it is worth considering as forbidding a pathway that's real could be a huge problem for someone later on.

alongd · 2019-02-12T18:26:03Z

@mjohnson541, I updated this branch.
What do you mean by a concerted reaction in RMG?

mjohnson541 · 2019-02-12T18:40:55Z

So lets say the elementary steps are A => B => C but B doesn't have a stable geometry. Technically the real reaction becomes A => C, but RMG will always find A => B => C instead. This means that forbidding B will prevent the real reaction A => C from being found in RMG.

I'm not suggesting this is the case for your forbidden species, but it's worth thinking about.

alongd · 2019-02-12T18:47:19Z

Thanks! Definitely something to look out for. I went over the forbidden species/groups list again, I'm relatively comfortable with it.
Let me know if I can squash fixups

alongd · 2019-03-23T00:11:29Z

@mjohnson541, I rebased and squashed.
I also added two small commits: added a forbidden structure and fixed some typos in comments

In NOx2018 and NitrogenDean&Bozzelli this reaction (D&B k31f1) is only the direct route of a proccess that also involves a P-dep route

Data taken from the same source Also corrected units of two training reactions

Added elementary_high_p flags and updated degeneracies

to differentiate it from the less stable form H2NN(T)

Also fixed some labeling

Forbidding cases such as: R + C2H3 = RH + [CH]=[CH] (mul) R + NNH = RH + [N]=[N] (mul) R + [NH][NH] <=> RH + [N][NH] (birad)

alongd · 2019-04-30T13:31:30Z

@mjohnson541, I rebased and it should now be good to go. Let me know if you have further comments

mjohnson541 · 2019-04-30T16:12:16Z

@alongd can you take a look at the RMG-tests failures on this branch?

original A factor in the library was wrong, probably a typo

Currently contains molecules related to hydrazine decomposition calculated using the 1DMin code for N2 bath gas.

also for the reverse T.L. Nguyen, J.F. Staton, IJCK 2019, doi: 10.1002/kin.21255

T.L. Nguyen, J.F. Staton, IJCK 2019, doi: 10.1002/kin.21255

mliu49 · 2019-04-30T21:09:46Z

I deleted the Travis build cache and restarted the push job, which should re-trigger RMG-tests. Let's see what happens.

mliu49 · 2019-04-30T22:57:55Z

input/kinetics/families/recommended.py

@@ -56,6 +56,8 @@
    '6_membered_central_C-C_shift',
    'Intra_R_Add_Exo_scission',
    '1,2_shiftC',
+    '1,2_NH3_elimination',
+    '1,3_NH3_elimination',


Do you think this would be a good opportunity to create a separate "nitrogen" set of families? Or do you think they should be included in the default set?

@mliu49, we don't have that many (I only see intra_NO2_ONO_conversion in addition to the ones added here), I think it's OK to have just one recommended family list to avoid adding another parameter the user has to remember to add in the input file. It could be OK to automatically check for reactive nitrogen in the input first, then let RMG decide which families to use, but it sounds a bit like over-engineering

mliu49 · 2019-04-30T23:21:37Z

I figured out why RMG-tests is failing, and it's an extremely stupefying reason...

Basically, RMG-tests is set up so that the RMG-Py/database branch to be tested is encoded in the name of the RMG-tests branch which is created and then parsed by RMG-tests.

For example, if we're testing the testbranch branch of RMG-database, an RMG-tests branch called rmgdb-testbranch would be created. This is then parsed by splitting the name on hyphens.

You can probably see where this is going...

This branch is called H4N2-Pyrolysis, so the RMG-tests branch is rmgdb-H4N2-Pyrolysis, which RMG-tests parses to obtain H4N2 as the name of the branch. Since there is no branch called H4N2, it fails to clone RMG-database, leading to non-existent thermo libraries as the error messages reported.

The easiest solution might be to create a new PR under a different branch name (without any hyphens). Ideally, we would update RMG-tests to be more robust, but updating the bash script there would take some effort.

alongd · 2019-05-01T01:24:37Z

amazing...
migrated to #326

HiroumiTani requested a review from alongd September 27, 2018 17:29

alongd reviewed Sep 28, 2018

View reviewed changes

alongd force-pushed the H4N2-Pyrolysis branch from da47f9b to 8215285 Compare October 22, 2018 16:27

alongd added Twin RMG-Py PR Topic: Nitrogen labels Oct 22, 2018

alongd mentioned this pull request Oct 22, 2018

Update charge after using the NH3_elimination families ReactionMechanismGenerator/RMG-Py#1496

Merged

alongd force-pushed the H4N2-Pyrolysis branch from 8215285 to ce209cc Compare October 22, 2018 18:43

alongd self-assigned this Oct 22, 2018

alongd force-pushed the H4N2-Pyrolysis branch 4 times, most recently from 2485ddd to 405ab1b Compare October 25, 2018 12:36

alongd force-pushed the H4N2-Pyrolysis branch 3 times, most recently from c3dd0e4 to 71d4b2a Compare January 28, 2019 03:45

alongd requested a review from mjohnson541 January 28, 2019 03:50

mjohnson541 reviewed Feb 1, 2019

View reviewed changes

alongd force-pushed the H4N2-Pyrolysis branch from 60ba3a6 to 92f8441 Compare February 12, 2019 18:18

alongd force-pushed the H4N2-Pyrolysis branch 3 times, most recently from 988dbfd to 7b465ac Compare March 23, 2019 00:11

alongd mentioned this pull request Apr 30, 2019

RMG v2.4.0 (?) Release Planning ReactionMechanismGenerator/RMG-Py#1591

Closed

5 tasks

Added allow_pdep_route flag to N2H3+NH2<=>N2H2+NH3

3b780de

In NOx2018 and NitrogenDean&Bozzelli this reaction (D&B k31f1) is only the direct route of a proccess that also involves a P-dep route

alongd added 10 commits April 30, 2019 09:30

Changed two R_Rec training rxns to the forward direction

8cbd5fe

Data taken from the same source Also corrected units of two training reactions

Added the HydrazinePDep kinetic library

57e59b3

Added NxHy globally forbidden structures

cc45a17

Modified NitrogenDean&Bozzelli

326b8f0

Added elementary_high_p flags and updated degeneracies

Rename H2NN -> H2NN(S) throughout libraries

998bf0f

to differentiate it from the less stable form H2NN(T)

Added NxHy thermo to primaryNS

fb767d9

Added the 1,2 and 1,3 NH3_elimination families

67d1b69

Added nitrogen training reactions to H_abs

80bbc7e

Also fixed some labeling

Added disprop mul_bonds birad forbidden groups to H_abs

1a974c8

Forbidding cases such as: R + C2H3 = RH + [CH]=[CH] (mul) R + NNH = RH + [N]=[N] (mul) R + [NH][NH] <=> RH + [N][NH] (birad)

Added various NxHy rates to the primaryNitrogenLibrary

3e91e0c

alongd force-pushed the H4N2-Pyrolysis branch from 7b465ac to 8559be3 Compare April 30, 2019 13:30

alongd added 7 commits April 30, 2019 13:47

Fixed Disprop training rxn H2N + H3N2 <=> H3N + H2N2-2

0e575fa

original A factor in the library was wrong, probably a typo

Fixed H2NN(S) structure in BurcatNS

14911c7

Added the OneDMinN2 transport library

eea5c80

Currently contains molecules related to hydrazine decomposition calculated using the 1DMin code for N2 bath gas.

Updated H_Abs training for NH3+H with data at the HEAT level

454e07a

also for the reverse T.L. Nguyen, J.F. Staton, IJCK 2019, doi: 10.1002/kin.21255

Added NH2+H2 rate at HEAT level to primaryNitrogenLibrary

d243b53

T.L. Nguyen, J.F. Staton, IJCK 2019, doi: 10.1002/kin.21255

Added N-N(S) as a forbidden species

b1102b3

Minor: Fixed typos in primaryNitrogenLibrary

ffddebf

alongd force-pushed the H4N2-Pyrolysis branch from 8559be3 to ffddebf Compare April 30, 2019 17:47

mliu49 reviewed Apr 30, 2019

View reviewed changes

alongd closed this May 1, 2019

alongd mentioned this pull request May 1, 2019

Hydrazine pyrolysis #326

Merged

alongd deleted the H4N2-Pyrolysis branch July 15, 2020 01:29

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Hydrazine pyrolysis #296

Hydrazine pyrolysis #296

HiroumiTani commented Sep 27, 2018

alongd left a comment

alongd Sep 27, 2018

alongd Sep 28, 2018

HiroumiTani Sep 28, 2018

HiroumiTani commented Sep 28, 2018

alongd commented Oct 22, 2018

alongd commented Jan 28, 2019

mjohnson541 left a comment

mjohnson541 Feb 1, 2019

mjohnson541 Feb 1, 2019

alongd Feb 1, 2019

mjohnson541 Feb 12, 2019

alongd commented Feb 12, 2019

mjohnson541 commented Feb 12, 2019

alongd commented Feb 12, 2019

alongd commented Mar 23, 2019

alongd commented Apr 30, 2019

mjohnson541 commented Apr 30, 2019

mliu49 commented Apr 30, 2019

mliu49 Apr 30, 2019

alongd May 1, 2019

mliu49 commented Apr 30, 2019

alongd commented May 1, 2019 •

edited

Loading

+              """,
+              )
+              #entry(

Hydrazine pyrolysis #296

Hydrazine pyrolysis #296

Conversation

HiroumiTani commented Sep 27, 2018

alongd left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

HiroumiTani commented Sep 28, 2018

alongd commented Oct 22, 2018

alongd commented Jan 28, 2019

mjohnson541 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

alongd commented Feb 12, 2019

mjohnson541 commented Feb 12, 2019

alongd commented Feb 12, 2019

alongd commented Mar 23, 2019

alongd commented Apr 30, 2019

mjohnson541 commented Apr 30, 2019

mliu49 commented Apr 30, 2019

Choose a reason for hiding this comment

Choose a reason for hiding this comment

mliu49 commented Apr 30, 2019

alongd commented May 1, 2019 • edited Loading

alongd commented May 1, 2019 •

edited

Loading