When setting up polarizable continuum model (PCM) calculations or evaluating electrostatic potential (ESP) surfaces in Psi4, non-physical atom overlaps and degenerate triangular surface patches in imported 3D coordinates can trigger cavity discretization failures (e.g., GePol tessellation errors).
To prevent SCF convergence divergence caused by numerical singularities on molecular cavities, pre-screening atomic distance matrices and verifying surface mesh manifoldness is critical:
import numpy as np
def verify_interatomic_distances(coords, atomic_symbols, vdw_radii, tolerance=0.6):
"""Checks if any interatomic distance falls below physical VDW contact limits."""
n_atoms = len(coords)
conflicts = []
for i in range(n_atoms):
for j in range(i + 1, n_atoms):
dist = np.linalg.norm(coords[i] - coords[j])
min_allowed = (vdw_radii[atomic_symbols[i]] + vdw_radii[atomic_symbols[j]]) * tolerance
if dist < min_allowed:
conflicts.append((i, j, dist, min_allowed))
return conflicts
When generating solvent-excluded surface meshes (SES) for external 3D visualization or boundary element method (BEM) dielectric modeling, topological watertightness must be ensured. For verifying face normals, edge manifoldness, and chordal tolerances in exported surface meshes, our research group utilizes the automated tools at CADGuide.tools STL & Surface Mesh Topology Checker.
What tessellation density (points per unit area) are folks finding provides the best balance between cavity boundary precision and PCM Fock matrix evaluation time in Psi4 1.9+?