Skip to content

7.4 ONIOM understand the subtraction before trusting the partition

These are unexecuted teaching inputs and starting models. Original diagrams are schematics, not calculated results. Validate version-specific syntax, licensed or authorized data, numerical convergence and the scientific model before using this workflow.

7.4.1 Model, units and provenance

Geometry in Å; electronic energy in hartree; vibrational wavenumbers in cm⁻¹. Check each printed field and keep thermal and standard-state terms distinct.

Shared inputs, conventions and evidence

Original schematic: ONIOM understand the subtraction before trusting the partition. No numerical results are claimed.
Original schematic: ONIOM understand the subtraction before trusting the partition. No numerical results are claimed.

7.4.2 Unexecuted inputs and explicit deltas

Use the accompanying instructions to identify the parent calculation and placement of every delta; a snippet is not automatically a standalone input. Preserve all blank-line and file-provenance requirements.

7.4.2.1 Input block 1

%chk=g28_methanol_oniom.chk
%mem=2GB
%nprocshared=2
#p ONIOM(HF/3-21G:AM1) SP Geom=Connectivity

Methanol two layer QM QM subtraction teaching model

0 1 0 1 0 1
C 0  0.000000  0.000000  0.000000 L H 2
O 0  1.430000  0.000000  0.000000 H
H 0  1.750000  0.900000  0.000000 H
H 0 -0.363000  1.028000  0.000000 L
H 0 -0.363000 -0.514000  0.890000 L
H 0 -0.363000 -0.514000 -0.890000 L

1 2 1.0 4 1.0 5 1.0 6 1.0
2 3 1.0
3
4
5
6

7.4.3 Worked investigation

7.4.3.1 Intuition and prerequisites

Two-layer ONIOM constructs E≈Ehigh(model)+Elow(real)−Elow(model). The low-level description covers the real system, while the high-level calculation corrects a selected model region. The subtraction prevents counting that region twice. Methanol is intentionally too small to need ONIOM, which makes it ideal for checking the bookkeeping against an all-high calculation. This example is QM:QM, using HF/3-21G and AM1; it does not require MM atom types or assert a production-quality embedding model. The boundary cuts the C–O bond, so the high-level O–H region receives a hydrogen link atom in place of carbon. This boundary is educational, not a recommended general partition for alcohol chemistry.

See input block 1 above.

7.4.3.2 Workflow

  1. Draw the real six-atom methanol and the capped high-level model separately. Explain that a link atom is generated for the model calculation and is not an extra real atom in the six-atom coordinate list.
  2. Read the atom-1 suffix L H 2: carbon is low layer, the replacing link species is H, and its high-layer bonded partner is atom 2. The three 0 1 pairs describe real-low, model-high, and model-low charge/spin.
  3. Run a single point first. Inspect all generated subsystems and their link bond placement; do not begin a long optimization before this check.
  4. Record the three component energies and reconstruct the extrapolated ONIOM value with the stated formula, using the units printed for those component records.
  5. Make an all-high HF/3-21G SP input using the same real coordinates but removing freeze/layer/link suffixes and connectivity if not requested. Compare properties at the same geometry.
  6. Move the boundary only in a well-defined follow-up with a chemically saturated model. Compare the target difference or property, not just absolute extrapolated energies from incompatible models.

7.4.3.3 Interpret check and limits

Link-atom distance scaling belongs to the model definition and can affect results. A neat partition picture does not prove the high region contains all relevant charge rearrangement. The high-layer model here is water-like after capping, so its electronic structure differs from intact methanol; the example makes this limitation visible. Extending to QM:MM requires verified atom types, charges, force-field parameters, and a considered choice of mechanical or electronic embedding. The AM1 subcalculation's reported energy conventions should be read from the Gaussian ONIOM component output; do not substitute an unrelated standalone semiempirical heat-of-formation field into the subtraction.

Exercise Explain why a partition chosen for a local O–H vibration might be unsuitable for a C–O cleavage reaction. Specify a larger model and a boundary-sensitivity test.

7.4.5 Sources and further reading