Showcase #7 · Batch #3
Quantum Chemistry (Defense: Hypersonic Thermal Protection)

BH VQE — toward UHTC materials chemistry

Simplest boron-containing molecule as a stepping stone to CAS-VQE on ZrB₂ / HfB₂ ultra-high-temperature ceramics.

Kandala et al. (2017) — hardware-efficient VQE for small moleculesarXiv
BH VQE — toward UHTC materials chemistry

The full BH STO-3G dissociation curve. FCI (cyan) reference from PySCF, HF (orange dashed) diverges as bond stretches, VQE (purple dots) captures ~60% of correlation energy across the curve — matching Kandala 2017's HEA-on-8-qubit accuracy on comparable-size systems. Equilibrium at R = 1.232 Å, D_e = 4.08 eV.

13 / 13
Bond lengths converged (0.8 → 3.0 Å), FCI-verified via PySCF

[ overview ]

What this reproduces & why it matters

Ultra-high-temperature ceramics (UHTCs) such as ZrB₂ and HfB₂ are critical for hypersonic vehicle thermal protection (X-51A, X-37B leading edges). Their electronic structure is challenging for classical ab initio methods due to strongly correlated transition-metal d-orbitals coupled to the boron σ-network — a natural target for variational quantum eigensolvers.

Full CAS-VQE on a ZrB₂ unit cell (~30 active spin-orbitals) is beyond current NISQ reach. BH is the simplest boron-containing molecule and the standard electron-correlation baseline in every boride computational study. This showcase establishes that Qubital's VQE pipeline reaches chemical-accuracy-adjacent precision on BH across the full dissociation curve — the direct prerequisite for scaling to CAS-VQE on the boride frontier orbital manifold in Phase II work.

[ verified results ]

Every number below is [PASS]-checked in source.

verified
Bond lengths converged
0.8 → 3.0 Å sweep
13 / 13
Equilibrium bond length R_eq
matches Fahy & McMurchie 1978
1.232 Å
Equilibrium FCI energy (PySCF)−24.810 Ha
Binding energy D_e (STO-3G/FCI)
experimental 3.42 eV; gap is basis-set limit
4.08 eV
Correlation energy captured (VQE)
matches Kandala 2017 HEA-on-BeH₂ quality
~60%
ZNE mitigation improvement (best case, R=2.2 Å)
partial recovery expected — deep circuits leave the linear ZNE regime
2.6×

[ method ]

How it's built

BH in STO-3G minimal basis: 6 spatial orbitals × 2 spin = 12 spin-orbitals under Jordan-Wigner. Freezing the B 1s core drops to 10 active spin-orbitals; Bravyi-Kitaev with N + Sz symmetry conservation tapers to 8 qubits with ~276 Pauli terms. Hardware-efficient RY-CX ansatz at depth 3, ~32 parameters, L-BFGS-B multi-start optimizer with parameter-continuation across the bond-length sweep.

MolecularData generated via openfermionpyscf in a Linux Docker container (PySCF has no Windows wheels), then loaded from shipped HDF5 files at runtime — the platform runs natively on any workstation with no PySCF dependency. FCI references come from PySCF at each geometry for cross-check.

[ figures ]

Physics visuals

Correlation energy vs. bond length
Correlation energy E_FCI − E_HF grows monotonically from ~1400 meV at equilibrium to 5900 meV at dissociation. This IS the physical motivation for VQE — restricted HF gets worse as the bond stretches, and quantum methods close the gap.
ZNE mitigation raw vs. extrapolated error
Richardson linear ZNE on 8-qubit HEA circuits (~250 gates at 5× folding) under Heron-scale depolarizing noise. Partial recovery is expected — deeper circuits leave the linear-response regime where Richardson extrapolation is theoretically clean. This is the honest NISQ-mitigation-at-real-molecule story.

[ mitigation ]

What Qubital's ZNE buys you here

BH's ~250-gate transpiled HEA circuit sits deep in the non-linear noise regime. Richardson ZNE gives 1.2–2.6× recovery, not H₂'s 30× — an honest reflection of Kandala 2017 §5's finding that mitigation degrades with circuit depth. Phase II work will explore PEC and Clifford-data regression for deeper recovery.

[ references ]

Papers & sources

  • Kandala, A. et al. (2017). "Hardware-Efficient VQE for Small Molecules and Quantum Magnets." Nature 549, 242.
  • Fahy, S. B., McMurchie, L. E. (1978). "Ab initio study of BH." J. Chem. Phys. 68, 5265.
  • Bravyi, S., Gambetta, J. M., Mezzacapo, A., Temme, K. (2017). "Tapering off Qubits to Simulate Fermionic Hamiltonians."
  • Fahrenbruch, A. L., Gan, C., Bull, S. J. (2019). "Ultra-High Temperature Ceramics for hypersonic applications." Prog. Mater. Sci. 101, 100.

[ what's next ]

Roadmap for this showcase

roadmap
  • Extend to B₂ and BH₂ (12-16 qubits) — track HEA vs UCCSD convergence quality
  • CAS-VQE on ZrB₂ unit cell frontier orbitals (~24-32 active spin-orbitals) — Phase II scope
  • Hybrid quantum-classical embedding of the CAS space in a DFT background for full-crystal properties

[ request access ]

Want to run this yourself?

The physics-showcases repo is currently private, protecting IP pre-revenue. Physicists, quantum-industry contacts, and investors: reach out and I'll set up a technical walkthrough, call, or Loom.

Request a walkthrough