A key component of any schema is a comprehensive dictionary of component names and descriptions.
Some sort of discussion of consistent QM nomenclature. (MP2, Moller-Plesset2, B3LYLP, B3LYP5, etc).
QM Methodologies produce a wide array of variables that need to be named and tagged for downstream use. For example if we think about a simple SCF computation we can produce the following variables:
- Nuclear Repulsion Energy
- SCF Total Energy
- One-Electron Energy
- Two-Electron Energy
- Number of SCF Iterations
- SCF Energy Convergence
- SCF Density Convergence
- Dipole X, Y, Z
- Additional One-Electron Properties
From this list it is likely the SCF Total Energy is the most important quantity; however, other quantities may be more important depending on the circumstances. Some sort of decision of which variables are imporant and how we want to describe them is in order.
If we take the first variable Nuclear Repulsion Energy this has a very basic
form nuc_rep = \frac{1}{2}\frac{q_i * q_j}{r_{ij}}. For most SCF quantities
the closed form is easy to write down; however, this quickly breaks down for
more advanced theories. Im not saying that we need to do this, but bringing up
the point that some decision on how explicit we are needs to be made.
Another possiblity is a word description which tries to explain the spirt rather than the specifics for these keys. A combination of both may be required.
SCF TOTAL ENERGY - result in Hartree of solving the SCF equations for Hartree--Fock or density functional theory. Its definition implies convergence reached but does not specify minimum convergence levels or convergence metric. Includes various self-consistent and additive contributions, such that may be the sum of nuclear repulsion energy, 1e- contributions, 2e- contributions, functional/HF energy, dispersion correction energy, etc.
FAMILY MP2 CORRELATION ENERGY - result in Hartree of the 2nd-order perturbation upon a Hartree--Fock (not DFT) reference such that (mp2 total energy = scf total energy + mp2 correlation energy) is always valid and the derived quantity (mp2 total energy) may be optionally stored (to check consistency). Includes unscaled same-spin doubles, opposite-spin doubles, and (for open-shell) singles contributions. A sufficient number of these will be stored under separate labels so that any SCS-MP2-like energy can be computed.
(T) CORRECTION ENERGY - result in Hartree of the parenthesis-triples correction to the coupled cluster singles and doubles method. (T)-F12, T0, and [T] fall under different labels.
- IUPAC Goldbook (https://goldbook.iupac.org)
- Units/Constants: https://goldbook.iupac.org/lists/list_math.html, https://goldbook.iupac.org/lists/list_goldbook_unit_defs.html
- Electron Density: https://goldbook.iupac.org/html/E/E01986.html IUPAC recommendations for computational chemistry: https://doi.org/10.1351/pac199769051137, https://doi.org/10.1515/pac-2012-1204 IUPAC recommendations are product of IUPAC Projects (https://iupac.org/recommendations/recently-published/) IUPAC InChI related activities beyond organics: https://iupac.org/who-we-are/divisions/division-details/?body_code=802
- CCLibVars
- PsiVars