Surface auto tree gen support with metal attrs #2187

davidfarinajr · 2021-07-26T16:48:45Z

Motivation or Problem

Adding auto tree gen support for Surface families and improving kinetics estimates based on metal attrs.

Description of Changes

modified get_bde method to get energies for metal bonds (X-A) using atomic binding energies from MetalDB
modified ArrheniusBM methods to work for surface reactions and created SurfaceArrheniusBM class as subclass of ArrheniusBM
added metal attrs to reaction class
modified tree descend algorithms to include metal attrs
revised tree gen methods to work for surface families
created a split_template family method to break up merged template (created by autogen trees) into difference reactant templates
added facet to rmg input file, and metal attrs to Kinetics Database that are used when descending kinetics trees to match training reactions or nodes in surface kinetics trees
added notebook to estimate A factors for surface reactions

Testing

autogenerated surface families trees here ReactionMechanismGenerator/RMG-database#499 and built CPOX models with different metals and facets for which the appropriate nodes and training reactions are selected based on metal attrs

lgtm-com · 2021-07-26T17:39:16Z

This pull request introduces 4 alerts when merging d96097f into 6932bc8 - view on LGTM.com

new alerts:

2 for Unused import
1 for Unused local variable
1 for Variable defined multiple times

codecov · 2021-07-26T17:51:47Z

Codecov Report

Merging #2187 (47c4e0b) into main (47c4e0b) will not change coverage.
The diff coverage is n/a.

❗ Current head 47c4e0b differs from pull request most recent head 5ed781e. Consider uploading reports for the commit 5ed781e to get more accurate results

@@           Coverage Diff           @@
##             main    #2187   +/-   ##
=======================================
  Coverage   47.21%   47.21%           
=======================================
  Files         104      104           
  Lines       27706    27706           
  Branches     7133     7133           
=======================================
  Hits        13082    13082           
  Misses      13217    13217           
  Partials     1407     1407

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 47c4e0b...5ed781e. Read the comment docs.

lgtm-com · 2021-07-26T18:13:28Z

This pull request introduces 4 alerts when merging 4ce2516 into 6932bc8 - view on LGTM.com

new alerts:

2 for Unused import
1 for Unused local variable
1 for Variable defined multiple times

mazeau · 2021-09-25T00:07:07Z

more as a note to myself, but after the rebase, this passes all py tests locally

mazeau · 2021-09-25T01:03:18Z

probably doesn't matter too much here, but I get errors with the py tests on discovery

======================================================================
ERROR: Test that generate_thermo_data() works correctly on gaussian PM3.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussianTest.py", line 78, in test_generate_thermo_data
    self.qmmol1.generate_thermo_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 363, in generate_thermo_data
    self.qm_data = self.generate_qm_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 265, in generate_qm_data
    success = self.run()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 100, in run
    return self.verify_output_file()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 170, in verify_output_file
    qm_data = self.parse()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 340, in parse
    qm_data = parse_cclib_data(cclib_data, radical_number + 1)  # Should `radical_number+1` be `self.molecule.multiplicity` in the next line of code? It's the electronic ground state degeneracy.
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/qmdata.py", line 101, in parse_cclib_data
    rotational_constants = (cclib_data.rotcons[-1], 'cm^-1')
IndexError: list index out of range

======================================================================
ERROR: Test that generate_thermo_data() can load thermo from the previous gaussian PM3 run.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussianTest.py", line 94, in test_load_thermo_data
    self.qmmol1.generate_thermo_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 363, in generate_thermo_data
    self.qm_data = self.generate_qm_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 254, in generate_qm_data
    if self.verify_output_file():
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 170, in verify_output_file
    qm_data = self.parse()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 340, in parse
    qm_data = parse_cclib_data(cclib_data, radical_number + 1)  # Should `radical_number+1` be `self.molecule.multiplicity` in the next line of code? It's the electronic ground state degeneracy.
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/qmdata.py", line 101, in parse_cclib_data
    rotational_constants = (cclib_data.rotcons[-1], 'cm^-1')
IndexError: list index out of range

======================================================================
ERROR: Test that generate_thermo_data() works correctly for gaussian PM6.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussianTest.py", line 136, in test_generate_thermo_data
    self.qmmol1.generate_thermo_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 363, in generate_thermo_data
    self.qm_data = self.generate_qm_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 265, in generate_qm_data
    success = self.run()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 100, in run
    return self.verify_output_file()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 170, in verify_output_file
    qm_data = self.parse()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 340, in parse
    qm_data = parse_cclib_data(cclib_data, radical_number + 1)  # Should `radical_number+1` be `self.molecule.multiplicity` in the next line of code? It's the electronic ground state degeneracy.
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/qmdata.py", line 101, in parse_cclib_data
    rotational_constants = (cclib_data.rotcons[-1], 'cm^-1')
IndexError: list index out of range

======================================================================
ERROR: Test that generate_thermo_data() can load thermo from the previous gaussian PM6 run.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussianTest.py", line 153, in test_load_thermo_data
    self.qmmol1.generate_thermo_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 363, in generate_thermo_data
    self.qm_data = self.generate_qm_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 254, in generate_qm_data
    if self.verify_output_file():
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 170, in verify_output_file
    qm_data = self.parse()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 340, in parse
    qm_data = parse_cclib_data(cclib_data, radical_number + 1)  # Should `radical_number+1` be `self.molecule.multiplicity` in the next line of code? It's the electronic ground state degeneracy.
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/qmdata.py", line 101, in parse_cclib_data
    rotational_constants = (cclib_data.rotcons[-1], 'cm^-1')
IndexError: list index out of range

======================================================================
ERROR: Test that Gaussian get_thermo_data() works correctly.
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/mainTest.py", line 276, in test_get_thermo_data_gaussian
    thermo1 = self.gauss1.get_thermo_data(mol)
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/main.py", line 224, in get_thermo_data
    thermo0 = qm_molecule_calculator.generate_thermo_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 363, in generate_thermo_data
    self.qm_data = self.generate_qm_data()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 265, in generate_qm_data
    success = self.run()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 100, in run
    return self.verify_output_file()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/gaussian.py", line 170, in verify_output_file
    qm_data = self.parse()
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/molecule.py", line 340, in parse
    qm_data = parse_cclib_data(cclib_data, radical_number + 1)  # Should `radical_number+1` be `self.molecule.multiplicity` in the next line of code? It's the electronic ground state degeneracy.
  File "/home/mazeau.e/Code/RMG-Py/rmgpy/qm/qmdata.py", line 101, in parse_cclib_data
    rotational_constants = (cclib_data.rotcons[-1], 'cm^-1')
IndexError: list index out of range

…or ArrBM this is needed so we can get the correct units for surface reactions

this is needed so we can select reactions that match metals in training and reaction libraries

this is needed for autogen trees with vdw bonds in the template in order to avoid splitting the vdw bonded complex into difference reactants

this method is needed to split merged templates (generated in autotrees) for multi-molecular families when generating reactions

- make deepcopies of adsorbates since thermo is different on different metals - assign metal/facet/site to reverse entries - stick entry index on reaction index (helpful for debugging)

metal attrs are loaded from input and set as attrs on kinetics database. They are taken from kinetics database and set on reactions generated with the kineticsDB

if we fail to split the node any furthur or all the reactions are the same, make child facet nodes

this fixed a bug where surface group parent specified rings, but children did not

… `get_w0` method

…roups

Metal attrs are loaded and saved to entries. They are loaded as Database attr, but not saved. This commit saves DB metal attrs

…b test The surface thermo groups should no longer have metal attrs since the adsorption groups are used for any type of metal/facet and this can mess with descending the trees if the thermo group entry has a metal, and the structure does not. Instead, the metal and facet are saved as thermo group database attrs

…utes` db test

…g_set`

lgtm-com · 2021-10-01T18:51:18Z

This pull request introduces 4 alerts when merging 5ed781e into 47c4e0b - view on LGTM.com

new alerts:

2 for Unused import
1 for Unused local variable
1 for Variable defined multiple times

github-actions · 2023-06-21T22:46:41Z

This pull request is being automatically marked as stale because it has not received any interaction in the last 90 days. Please leave a comment if this is still a relevant pull request, otherwise it will automatically be closed in 30 days.

davidfarinajr added Topic: Kinetics Topic: Catalysis Status: WIP This is currently work-in-progress labels Jul 26, 2021

davidfarinajr requested review from rwest, mjohnson541 and mazeau July 26, 2021 16:49

mazeau force-pushed the surface_arrbm branch from 4ce2516 to cbe3ba4 Compare September 24, 2021 23:08

davidfarinajr changed the base branch from master to main October 1, 2021 17:34

davidfarinajr added 17 commits October 1, 2021 14:20

added get_metal_label entry method

f010e40

use get_rate_coefficient_units_from_reaction_order to get A units f…

f9a43e3

…or ArrBM this is needed so we can get the correct units for surface reactions

get_w0 with metals and vdw bonds

dae5b64

added SurfaceArrheniusBM kinetics class

606e3c0

added add_van_der_waals_bond mol method

58a8995

added is_surface_bond Bond method

eae6857

get_bde for metal bonds using MetalDB

5121668

added test_get_bde unit test

da71141

added metal attrs to Reaction

d30cad6

change logging from warning to info

3fcc61c

added metal attrs to reaction when loading entries

5079f7a

reactions on different metals are not duplicate

f181ca0

added metal and facet to input, RMG and CoreEdgeReactionModel classes

4a66dc4

this is needed so we can select reactions that match metals in training and reaction libraries

select the best kinetics with metal and facet attrs

a3fd3e9

added vdw_bonds family attr

3d2daa8

this is needed for autogen trees with vdw bonds in the template in order to avoid splitting the vdw bonded complex into difference reactants

added split_template family method

c209e60

this method is needed to split merged templates (generated in autotrees) for multi-molecular families when generating reactions

use SurfaceArrheniusBM for surace reactions with autotree gen methods

9fdc4be

davidfarinajr added 19 commits October 1, 2021 14:21

save training reactions with metal attrs

f75f7c8

set family reactant_num to length of template if not specified

dbab14b

using get_metal_label function

cae9258

get_training_set for surface families

ab9d981

- make deepcopies of adsorbates since thermo is different on different metals - assign metal/facet/site to reverse entries - stick entry index on reaction index (helpful for debugging)

get_kinetics_from_depository fixup

468331e

added metal attrs to kinetics database and generate reactions

675c18a

metal attrs are loaded from input and set as attrs on kinetics database. They are taken from kinetics database and set on reactions generated with the kineticsDB

added metal attrs to kinetics Groups and descend tree algorithm

e4822a9

added facets to autotree gen extend node

171b2eb

if we fail to split the node any furthur or all the reactions are the same, make child facet nodes

do not regularize rings if the molecule contains surface site

cdf047c

this fixed a bug where surface group parent specified rings, but children did not

added notebook to estimate A factors for surface reactions

527a9d5

use cathub db branch for tests

dd1eb00

use with for mp pool

c290746

family product num is length of template products if not specified

11c65d1

raise InvalidActionError if nonexistent bond when applying recipe for…

88ed85f

… `get_w0` method

added reactantNum and productNum to intra_H_migration testing g…

f013d37

…roups

saving DB metal attrs

528c24a

Metal attrs are loaded and saved to entries. They are loaded as Database attr, but not saved. This commit saves DB metal attrs

fixup! commented out `check_surface_thermo_groups_have_surface_attrib…

ea49070

…utes` db test

added rxn metal attrs to own reverse reactions in family `get_trainin…

5ed781e

…g_set`

mazeau force-pushed the surface_arrbm branch from cbe3ba4 to 5ed781e Compare October 1, 2021 18:22

github-actions bot added the stale stale issue/PR as determined by actions bot label Jun 21, 2023

github-actions bot added the abandoned abandoned issue/PR as determined by actions bot label Jul 22, 2023

github-actions bot closed this Jul 22, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Surface auto tree gen support with metal attrs #2187

Surface auto tree gen support with metal attrs #2187

davidfarinajr commented Jul 26, 2021

lgtm-com bot commented Jul 26, 2021

codecov bot commented Jul 26, 2021 •

edited

Loading

lgtm-com bot commented Jul 26, 2021

mazeau commented Sep 25, 2021

mazeau commented Sep 25, 2021

lgtm-com bot commented Oct 1, 2021

github-actions bot commented Jun 21, 2023

Surface auto tree gen support with metal attrs #2187

Surface auto tree gen support with metal attrs #2187

Conversation

davidfarinajr commented Jul 26, 2021

Motivation or Problem

Description of Changes

Testing

lgtm-com bot commented Jul 26, 2021

codecov bot commented Jul 26, 2021 • edited Loading

Codecov Report

lgtm-com bot commented Jul 26, 2021

mazeau commented Sep 25, 2021

mazeau commented Sep 25, 2021

lgtm-com bot commented Oct 1, 2021

github-actions bot commented Jun 21, 2023

codecov bot commented Jul 26, 2021 •

edited

Loading