Charge transfer integrals with ADF

To follow along, either

Worked Example

Initial Imports

from scm.plams import Settings, ADFFragmentResults, Molecule, log, ADFFragmentJob, add_to_class

Helper Functions

Add helper results extraction method.

@add_to_class(ADFFragmentResults)
def get_transfer_integrals(self):
    return self.job.full.results.read_rkf_section("TransferIntegrals", file="adf")

Configure Settings

Set up common settings for all 3 jobs, the specific settings for full system job. Note that in the interest of computational speed we use a minimal basis set. For more quantitatively meaningful results, you should use a larger basis.

common = Settings()
common.input.ams.Task = "SinglePoint"
common.input.adf.Basis.Type = "SZ"
common.input.adf.Basis.Core = "None"
common.input.adf.Symmetry = "NoSym"

full = Settings()
full.input.adf.transferintegrals = True

Load Molecule

Load benzene dimer from xyz file and separate it into 2 fragments. Alternatively one could simply load fragments from separate xyz files:

mol1 = Molecule('fragment1.xyz')
mol2 = Molecule('fragment2.xyz')

mol = Molecule("BenzeneDimer.xyz")
mol.guess_bonds()
fragments = mol.separate()
if len(fragments) != 2:
    log("ERROR: Molecule {} was split into {} fragments".format(mol.properties.name, len(fragments)))
    import sys

    sys.exit(1)
else:
    mol1, mol2 = fragments

Run Job and Get Results

job = ADFFragmentJob(name="ADFTI", fragment1=mol1, fragment2=mol2, settings=common, full_settings=full)
results = job.run()

[13.08|16:56:10] JOB ADFTI STARTED
[13.08|16:56:10] JOB ADFTI RUNNING
[13.08|16:56:10] JOB ADFTI/frag1 STARTED
[13.08|16:56:10] JOB ADFTI/frag1 RUNNING
[13.08|16:56:13] JOB ADFTI/frag1 FINISHED
[13.08|16:56:13] JOB ADFTI/frag1 SUCCESSFUL
[13.08|16:56:13] JOB ADFTI/frag2 STARTED
[13.08|16:56:13] JOB ADFTI/frag2 RUNNING
[13.08|16:56:17] JOB ADFTI/frag2 FINISHED
[13.08|16:56:17] JOB ADFTI/frag2 SUCCESSFUL
[13.08|16:56:17] JOB ADFTI/full STARTED
[13.08|16:56:17] JOB ADFTI/full RUNNING
[13.08|16:56:27] JOB ADFTI/full FINISHED
[13.08|16:56:27] JOB ADFTI/full SUCCESSFUL
[13.08|16:56:27] JOB ADFTI FINISHED
[13.08|16:56:27] JOB ADFTI SUCCESSFUL

# TI is a dictionary with the whole TransferIntegrals section from adf.rkf
print("== Results ==")
TI = results.get_transfer_integrals()
for key, value in sorted(TI.items()):
    print("{:<28}: {:>12.6f}".format(key, value))

== Results ==
J(charge recombination 12)  :     0.010744
J(charge recombination 21)  :     0.010744
J(electron)                 :     0.012050
J(hole)                     :    -0.034988
S(charge recombination 12)  :    -0.016229
S(charge recombination 21)  :    -0.016229
S(electron)                 :    -0.018755
S(hole)                     :     0.049852
V(charge recombination 12)  :     0.009867
V(charge recombination 21)  :     0.009867
V(electron)                 :     0.013129
V(hole)                     :    -0.026790
Vtot(charge recombination 12):     0.013193
Vtot(charge recombination 21):     0.013193
Vtot(electron)              :     0.021464
Vtot(hole)                  :     0.034178
e1(electron)                :     0.057279
e1(hole)                    :    -0.165788
e2(electron)                :     0.057280
e2(hole)                    :    -0.165790

Complete Python code

#!/usr/bin/env amspython
# coding: utf-8

# ## Initial Imports

from scm.plams import Settings, ADFFragmentResults, Molecule, log, ADFFragmentJob, add_to_class


# ## Helper Functions
# Add helper results extraction method.


@add_to_class(ADFFragmentResults)
def get_transfer_integrals(self):
    return self.job.full.results.read_rkf_section("TransferIntegrals", file="adf")


# ## Configure Settings
# Set up common settings for all 3 jobs, the specific settings for full system job. Note that in the interest of computational speed we use a minimal basis set. For more quantitatively meaningful results, you should use a larger basis.

common = Settings()
common.input.ams.Task = "SinglePoint"
common.input.adf.Basis.Type = "SZ"
common.input.adf.Basis.Core = "None"
common.input.adf.Symmetry = "NoSym"


full = Settings()
full.input.adf.transferintegrals = True


# ## Load Molecule
# Load benzene dimer from xyz file and separate it into 2 fragments. Alternatively one could simply load fragments from separate xyz files:
# ```
# mol1 = Molecule('fragment1.xyz')
# mol2 = Molecule('fragment2.xyz')
# ```

mol = Molecule("BenzeneDimer.xyz")
mol.guess_bonds()
fragments = mol.separate()
if len(fragments) != 2:
    log("ERROR: Molecule {} was split into {} fragments".format(mol.properties.name, len(fragments)))
    import sys

    sys.exit(1)
else:
    mol1, mol2 = fragments


# ## Run Job and Get Results

job = ADFFragmentJob(name="ADFTI", fragment1=mol1, fragment2=mol2, settings=common, full_settings=full)
results = job.run()


# TI is a dictionary with the whole TransferIntegrals section from adf.rkf
print("== Results ==")
TI = results.get_transfer_integrals()
for key, value in sorted(TI.items()):
    print("{:<28}: {:>12.6f}".format(key, value))