Research | QTCOVI – Theoretical and Computational Chemistry

Excited States, Catalysis & Biomolecules

Mon, 01 Jan 2024 00:00:00 +0000

Overview

Quantum Chemical Topology and IQA are not limited to ground-state molecules. We actively apply our methods to:

Excited-State Chemistry

Topological analysis of excited-state densities (natural transition orbitals, state-specific densities) reveals how electron rearrangement drives photochemical reactivity. IQA energy decomposition along excited-state reaction paths provides mechanistic insight inaccessible to MO-based approaches.

Homogeneous and Enzymatic Catalysis

IQA along reaction coordinates quantifies how the environment modulates the strengths of bonds being formed and broken. In enzymatic systems, QCT descriptors illuminate how the protein scaffold polarises and stabilises transition states.

Biomolecular Systems

We study hydrogen bond networks, halogen bonds, and π-stacking in DNA, proteins, and drug–receptor complexes using IQA and delocalization indices, complementing classical force-field analyses with a rigorous quantum mechanical perspective.

Collaborations

This line is developed in close collaboration with experimental and computational groups in Spain, the UK, France, Italy, Mexico, and Chile.

Interacting Quantum Atoms (IQA)

Mon, 01 Jan 2024 00:00:00 +0000

Overview

The Interacting Quantum Atoms (IQA) methodology, developed in our group, partitions the total electronic energy of a molecule rigorously into intra-atomic self-energies and pairwise inter-atomic interaction energies. Each inter-atomic interaction is further decomposed into classical electrostatics and quantum exchange–correlation contributions.

Because IQA is grounded in the QTAIM atomic partition, it is invariant to orbital transformations and does not depend on any reference state or arbitrary choices of localisation. This makes IQA particularly valuable for:

Comparing bonding across different molecular environments and bond types.
Tracking energy changes along reaction coordinates and conformational changes.
Establishing connections between bond strength and electron delocalisation.

IQA and Bond Orders

A central result of IQA analysis is the relationship between the inter-atomic exchange–correlation energy V_XC(A,B) and the classical bond order between atoms A and B. The delocalization index δ(A,B)—the number of electron pairs shared between basins—provides an orbital-invariant bond order.

Applications

IQA has been applied in our group to:

Hydrogen and halogen bonds
Metal–ligand bonding in transition metal complexes
π-stacking and non-covalent interactions
Reaction mechanisms and transition state analysis

Machine Learning & Neural Network Potentials

Mon, 01 Jan 2024 00:00:00 +0000

Overview

Machine learning is transforming computational chemistry, and we are harnessing it in two complementary ways:

IQA-Informed Neural Network Potentials

Classical machine-learned interatomic potentials (NNPs) are trained on total energies and forces, but lack chemical interpretability. We develop NNPs informed by IQA energy components (self-energies and interaction energies), resulting in potentials that:

Decompose into physically meaningful atomic and pairwise contributions.
Transfer more reliably to out-of-distribution chemical environments.
Naturally encode the correct physics of bonding interactions.

Topological Descriptors as ML Features

QTAIM atomic properties and IQA energy components serve as physically motivated features for machine learning models targeting molecular properties, reaction barriers, and drug–target binding affinities.

Deep Learning for Electron Density

We explore the use of deep learning models to predict electron densities directly, enabling rapid computation of topological properties for large molecular datasets.

Codes & Tools

Our ML work builds on open-source frameworks (PyTorch, JAX) and is integrated with our in-house topological analysis codes.

Quantum Chemical Topology

Mon, 01 Jan 2024 00:00:00 +0000

Overview

Quantum Chemical Topology (QCT) encompasses a family of methods that use the topology of scalar fields derived from the electron density—such as the gradient of ρ(r), the electron localisation function (ELF), or the non-covalent interaction (NCI) index—to partition molecular space into chemically meaningful regions (atoms, bonds, rings, cages).

The cornerstone of QCT is Bader’s Quantum Theory of Atoms in Molecules (QTAIM), which defines atoms as basin-like regions bounded by zero-flux surfaces of the electron density gradient. Critical points of ρ(r) identify bond, ring, and cage features, while integrated atomic properties (charge, kinetic energy, volume) carry well-defined quantum mechanical meaning.

Our Contributions

Development of PROMOLDEN, a high-performance code for topological analysis of electron densities and pair densities.
Systematic study of basin properties and their relationship to bonding descriptors.
Extension of QCT frameworks to pair and reduced density matrices.

Key References

Selected publications from the group on quantum chemical topology are listed in the Publications section.

Real-Space Electron Correlation Descriptors

Mon, 01 Jan 2024 00:00:00 +0000

Overview

A key insight driving our research is that the statistics of the electron number distribution in real-space atomic basins contain rich information about electron correlation and chemical bonding.

By computing the probability of finding n electrons in basin Ω, we obtain:

The average electron population N(Ω) — the QTAIM atomic charge.
The variance — a measure of electron fluctuation and localisation.
The delocalization index δ(A,B) — the covariance of electron populations between basins A and B, providing an orbital-invariant bond order.
Higher-order cumulants — sensitive to multi-centre bonding and electron correlation beyond mean-field.

Non-Covalent Interactions

We use delocalization indices, variance maps, and IQA inter-atomic energies to characterise and classify non-covalent interactions (hydrogen bonds, halogen bonds, van der Waals, π–π stacking) without reference to molecular orbitals.

Beyond DFT

Our descriptors are applicable with any level of theory (HF, DFT, CISD, CCSD, CASSCF, Quantum Monte Carlo), allowing direct assessment of the effect of electron correlation on bonding descriptors.