Skip to contents

Organo-mineral Hamiltonians derived from ab initio molecular models

Source: R/quantum_nature.R
quantum_hamiltonian_organo_mineral_nature.Rd

Curated presets that build an edaphos_quantum_hamiltonian from a realistic molecular geometry through quantum_hamiltonian_from_pyscf(). Three variants are shipped, each chosen to expose a distinct piece of organo-mineral chemistry at a qubit count that is accessible to present-day hardware:

Usage

quantum_hamiltonian_organo_mineral_nature(
  variant = c("formic_acid", "methanediol", "ferric_formate"),
  basis = "sto3g",
  num_active_electrons = NULL,
  num_active_orbitals = NULL,
  mapper = c("parity", "jordan_wigner", "bravyi_kitaev")
)

Arguments

variant

Character — "formic_acid" (default), "methanediol" or "ferric_formate".

basis

Character — PySCF basis set label. Default "sto3g" for speed; upgrade to "631g" for quantitative work.

num_active_electrons, num_active_orbitals

Optional integer overrides for the active-space size. When left NULL the variant-specific default is used ((2, 2) for the closed-shell organics, (4, 4) for ferric formate).

mapper

Character — fermion-to-qubit mapping; see quantum_hamiltonian_from_pyscf().

Value

An edaphos_quantum_hamiltonian_nature (which inherits from edaphos_quantum_hamiltonian) carrying the qubit Hamiltonian plus the nuclear-repulsion, frozen-core and reference-energy attributes needed to reconstruct the total molecular energy.

Details

"formic_acid" (default)

HCOOH in STO-3G with the chemical core frozen and a (2e, 2o) active space. Models the carboxylate functional group that is the dominant coordinating moiety of humic acids. After parity tapering the qubit Hamiltonian is 2 qubits / ~5 Pauli terms — fast enough to run on any hardware and small enough for CI.

"methanediol"

H\(_2\)C(OH)\(_2\) in STO-3G with the core frozen and a (2e, 2o) active space. Models the ortho-diol motif that underpins catechol-type Fe(III) chelation in humic substances. 2 qubits after tapering.

"ferric_formate"

Monodentate Fe(III)–OOCH complex in STO-3G. Open-shell (S = 5/2), requires a (4e, 4o) active space around Fe 3d + carboxylate \(\pi^*\), and produces a 4-qubit Hamiltonian after parity tapering. This variant is the minimum viable representation of an organo-mineral coordination event at the clay–humus interface; because the SCF convergence is sensitive to the initial guess we recommend running it on backend = "aer_shots" or "ibmq" with "SPSA" and at least 100 iterations, and consulting attr(ham, "reference_energy") as a sanity check.

References

Stevenson, F. J. (1994). Humus Chemistry: Genesis, Composition, Reactions. Wiley. (Chapter on carboxylate and catechol-type chelation of Fe(III) at mineral surfaces.)

See also

quantum_hamiltonian_from_pyscf(), quantum_hamiltonian_organo_mineral() (toy 4-qubit variant), quantum_vqe_fit().

Examples

if (FALSE) { # \dontrun{
  ham <- quantum_hamiltonian_organo_mineral_nature("formic_acid")
  attr(ham, "nuclear_repulsion_energy")
  quantum_vqe_fit(ham, seed = 1L)$energy +
    attr(ham, "nuclear_repulsion_energy")
} # }