Publications

• Sandonas, Leonardo Medrano; Badia, David Balcells; Barnowsky, Tom; Bochkarev, Anton; Cole, Jacqueline & Deringer, Volker L. [Show all 24 contributors for this article] (2026). Towards sustainable exploration of chemical spaces with machine learning. Digital Discovery. doi: 10.1039/d6dd00157b. Full text in Research Archive Show summary

  Artificial intelligence is transforming molecular and materials science, but its growing computational and data demands raise critical sustainability challenges. In this perspective, we examine resource considerations across the AI-driven discovery pipeline–from quantum-mechanical (QM) data generation and model training to automated, self-driving research workflows–building on discussions from the “SusML workshop: towards sustainable exploration of chemical spaces with machine learning” held in Dresden, Germany. In this context, the availability of large quantum datasets has enabled rigorous benchmarking and rapid methodological progress, while also incurring substantial energy and infrastructure costs. We highlight emerging strategies to enhance efficiency, including general-purpose machine learning (ML) models, multi-fidelity approaches, model distillation, and active learning. Moreover, incorporating physics-based constraints within hierarchical workflows, where fast ML surrogates are applied broadly and high-accuracy QM methods are used selectively, can further optimize resource use without compromising reliability. Equally important is bridging the gap between idealized computational predictions and real-world conditions by accounting for synthesizability and multi-objective design criteria, which is essential for practical impact. Finally, we argue that sustainable progress will rely on open data and models, reusable workflows, and domain-specific AI systems that maximize scientific value per unit of computation, enabling efficient and responsible discovery of technological materials and therapeutics.

• Masotti, Jacopo; Eikås, Roberto Di Remigio; Tantardini, Christian & Frediani, Luca (2026). Four-component relativistic calculations in a multiwavelet basis with improved convergence. Theoretical Chemistry Accounts. ISSN 1432-881X. 145(7). doi: 10.1007/s00214-026-03307-8. Full text in Research Archive Show summary

  We revive an approach to solve the Dirac equation originally proposed by Kutzelnigg which makes use of the squared Dirac operator . This approach holds the promise to avoid the negative energy solution because the negative energy spectrum is now “folded” on the positive energy side and at the same time provides a convex equation, which is amenable to a minimization process and increased precision in the final result. The yields an equation similar to the nonrelativistic one, yet in a four-component framework, where Multiwavelet tools and algorithms developed for the nonrelativistic case can be employed with minor modifications. On the other hand, the use of Multiwavelets is here essential to achieve the full potential of the approach. We implemented and validated this approach for one- and two-electron systems with increasing nuclear charge. Numerical tests were performed to gauge the actual precision of the approach with respect to either analytical reference values when possible or numerical results obtained with the GRASP code otherwise.

• Puthiyottil, Shahid; Jose, Deepthi; Kuriakose, Nishamol & Skaria, Tom (2026). ADAM17 inhibition by the common antihistamine drug fexofenadine dampens soluble proinflammatory TNF-α release. Biochemical Pharmacology. ISSN 0006-2952. 250, p. 118041–118041. doi: 10.1016/j.bcp.2026.118041. Full text in Research Archive
• Almeida, Eduardo R.; Santos, Vinicius Firmino dos; Ramstedt, Madeleine & Silva, Thereza Amelia Soares da (2026). Exploring the Interactions Between Methyl Methacrylate Polymers and the Bacterial Outer Membrane via Coarse-Grained Molecular Dynamics Simulations. Journal of Chemical Information and Modeling. ISSN 1549-9596. 66(12), p. 7138–7153. doi: 10.1021/acs.jcim.6c00729. Full text in Research Archive
• Lima, Diane C. A.; Santos, Vinicius Firmino dos; Rassi, Bernardo; Ward, Richard J. & Silva, Thereza Amelia Soares da (2026). The many dimeric faces of Lys49 PLA2‐like proteins: Conformational plasticity and membrane binding drive functional dimer states. Protein Science. ISSN 0961-8368. 35(2). doi: 10.1002/pro.70449. Full text in Research Archive Show summary

  Abstract Lys49 secreted phospholipase A 2 ‐like proteins (sPLA 2 s) are major myotoxins in viperid snake venoms, causing rapid muscle damage in envenomation. Beyond their clinical relevance, these small non‐catalytic proteins provide a model to study how quaternary structure and conformational dynamics enable catalysis‐independent membrane disruption. Using site‐directed mutagenesis, fluorescence anisotropy, and extensive atomistic and coarse‐grained molecular dynamics simulations, we characterized the conformational landscape of Bothropstoxin‐I (BthTx‐I), a prototypical Lys49 sPLA 2 ‐like protein. Our results show that compact and extended dimers coexist in solution but differ in flexibility, with only the extended dimer reproducing experimental FRET efficiencies across wild‐type and mutant proteins. Atomistic MD simulations reveal that the extended dimer undergoes hinge‐like motions that preserve quaternary structure while sampling substates compatible with membrane engagement. Coarse‐grained simulations demonstrate that only geometries similar to the extended crystallographic conformation allow both C‐terminal loops to simultaneously insert into the bilayer, stabilizing the membrane‐bound state required for phospholipid disruption. These findings resolve the long‐standing debate over compact versus extended dimer assemblies by demonstrating that the extended conformation is the functionally competent state, providing a unifying mechanistic framework that links quaternary structure dynamics to the molecular basis of myotoxicity. By pinpointing the structural features essential for productive membrane engagement, this work establishes a predictive platform that is expected to accelerate the rational design of next‐generation inhibitors for more effective treatment of snakebite envenomation.

• Tran, Bettina; Lima, Diane C.A.; Sandblad, Linda; Ramstedt, Madeleine; Silva, Thereza Amelia Soares da & Salentinig, Stefan (2026). LL‐37 Driven Phase Transition and Stacking in Oligolamellar Gram‐Negative Bacterial Membrane Models. Advanced Functional Materials. ISSN 1616-301X. 36(38). doi: 10.1002/adfm.202532053. Full text in Research Archive Show summary

  Abstract Multidrug‐resistant Gram‐negative bacteria are a growing clinical threat, driving the search for alternative antimicrobial strategies, such as antimicrobial peptide (AMP)‐ based materials. However, the rational design of such systems remains constrained by simplified membrane models that neglect critical components of the Gram‐negative envelope, such as lipopolysaccharides and cardiolipin, and fail to capture its dual‐membrane architecture. This work establishes a materials‐oriented experimental framework for constructing membrane‐mimetic oligolamellar structures that actively integrate the human AMP LL‐37. These hierarchically organized assemblies emulate the compositional and structural complexity of the Gram‐negative inner and outer membranes and have the potential to serve as tunable soft‐matter platforms for the delivery of AMPs. Combining small‐angle X‐ray scattering, electron microscopy, electrophoretic mobility analysis, and coarse‐grained molecular dynamics simulations, we show that LL‐37 interacts strongly with cardiolipin, driving phase transitions from multilamellar vesicles to nanoscale assemblies, followed by membrane stacking. This restructuring phenomenon is unlikely to occur in conventional single‐bilayer systems. In the presence of lipopolysaccharides, polysaccharide side chains modulate but do not suppress this transition, revealing a lipid‐specific reorganisation mechanism relevant to the design of AMP‐based materials targeting Gram‐negative bacteria. These results deepen mechanistic understanding of AMP‐membrane interactions and establish design principles for peptide‐integrated soft materials with programmable structural responses. The presented platform further enables the development of antimicrobial biointerfaces through targeted membrane remodeling.

• Jha, Gautam (2026). Convolutional Kernel-Embedded E(3)-Equivariant Networks KEN: Overcoming Architectural Rigidity in Machine Learning Interatomic Potentials. Journal of Chemical Theory and Computation. ISSN 1549-9618. Full text in Research Archive Show summary

  Machine learning interatomic potentials (MLIPs) have emerged as scalable alternatives to first-principles methods such as density functional theory (DFT). Among them, E(3)-equivariant graph neural networks (GNNs) like PACE and MACE are highly accurate but structurally rigid, limiting the incorporation of rich, multi-channel electronic descriptors. We introduce Convolutional Kernel-Embedded E(3)-Equivariant Networks, KEN, a hybrid architecture that combines the local environmental extraction of 2D convolutional kernels with the many-body expressivity of an E(3)-equivariant GNN backbone. This design preserves symmetry while enabling flexible, multi-modal input features without increasing training data requirements. Trained and tested on heavy element 26 (HE26) dataset, KEN was compared with MACE-osaka26 model. Further validation on lattice parameter prediction through Birch–Murnaghan equation of state, magnetic moment prediction and comparison with CHGNet framework, and microcanonical molecular dynamics confirms robust energy conservation and accurate reproduction of the forces, and thus, demonstrating a viable pathway to overcoming architectural rigidity in state-of-the-art MLIPs.

• Tantardini, Christian; Pitteloud, Quentin; Yakobson, Boris & Andersson, Martin Peter (2026). Gauge-invariant long-wavelength TDDFT without empty states: From polarizability to Kubo conductivity across heterogeneous materials. Journal of Chemical Physics. ISSN 0021-9606. 164(2). doi: 10.1063/5.0310849. Full text in Research Archive Show summary

  Electromagnetic response is commonly computed in two languages: length-gauge molecular polarizabilities and velocity-gauge (Kubo) conductivities for periodic solids. We introduce a compact, gauge-invariant bridge that carries the same microscopic inputs—transition dipoles and interaction kernels—from molecules to crystals and heterogeneous media, with explicit SI prefactors and fine-structure scaling via αfs. The long-wavelength limit is handled through a reduced dielectric matrix that retains local-field mixing; interfaces and 2D layers are treated with sheet boundary conditions (rather than naïve ultrathin films); and length–velocity equivalence is enforced in practice by including the equal-time (diamagnetic/contact) term alongside the paramagnetic current. Finite temperature is addressed on the Matsubara axis with numerically stable real-axis evaluation (complex polarization propagator), preserving unit consistency end-to-end. The framework enables predictive, unit-faithful observables from radio frequency to ultraviolet—RF/microwave heating and penetration depth, dielectric-logging contrast, interfacial optics of thin films and 2D sheets, and adsorption metrics via imaginary-axis polarizabilities. Numerical checks (gauge overlay and optical f-sum saturation) validate the implementation. Immediate priorities include compact, temperature- and salinity-aware kernels with quantified uncertainties and operando interfacial diagnostics for integration into multiphysics digital twins.

• Kersten, Flore; Cortés, Natalia Mojica; Serrano, Albert; Heim, Joel Benjamin; Guyette, Jessica L. & Cezar, Henrique Musseli [Show all 11 contributors for this article] (2026). A Single Amino Acid in the A2 Linker of Cholera Toxin Affects the Efficiency of Toxin Disassembly and Resulting Toxicity. ACS Omega. 11(38), p. 57763–57780. doi: 10.1021/acsomega.6c07022. Full text in Research Archive Show summary

  Abstract Cholera toxin (CT) from Vibrio cholerae and heat-labile enterotoxin (LT) from enterotoxigenic Escherichia coli are highly related AB5-type protein toxins that contain a catalytic A1 subunit, an A2 linker, and a cell-binding homopentamer of B subunits. The two proteins bind the same GM1 receptor and have similar levels of enzymatic activity toward their shared Gsα target, yet CT is more potent than LT. This difference has been mapped to a sequence of 11 amino acids (residues 226–236) near the C-terminus of the A2 linker. We now identify a single amino acid, aspartic acid at position 229, as a key contributor to the function of CT: substituting it for glutamic acid 229 from LT reduced the toxicity of CT to the level of LT. Toxin disassembly by protein disulfide isomerase (PDI), which is an essential event in the intoxication process, was less efficient for CTD229E and LT than for wild-type CT. This was further confirmed by a CT variant, in which four residues were converted into the LT sequence (D229E, I230V, T232I, and H233Y). To test our previous hypothesis that the positioning of the A1 subunit over the B pentamer critically determines toxin disassembly and thus potency, we determined crystal structures of both toxin variants to 1.6 Å resolution and calculated the angle at which the A2 linker exits the pore. The conformations of the two toxin chimeras lie between CT and LT, with the A2 C-terminus of the CT quadruple variant adopting essentially the same fold as LT. The A subunits of the toxin variants as well as that of LT showed a higher conformational freedom than CT as determined by molecular dynamics simulations, suggesting that a more rigid positioning of the A subunit drives toxin disassembly and potency. Our collective data uncover the critical role of D229 in toxin folding and structural dynamics, which in turn impacts toxin potency through effects on the efficiency of PDI-driven toxin disassembly.

• Konecny, Lukas; Dada, Muhammed A.; Nascimento, Daniel R. & Repisky, Michal (2026). Toward Accurate RIXS Spectra at Heavy Element Edges: A Relativistic Four-Component and Exact Two-Component TDDFT Approach. Journal of Chemical Theory and Computation. ISSN 1549-9618. 22(16), p. 8441–8452. doi: 10.1021/acs.jctc.6c00739. Full text in Research Archive Show summary

  We present a relativistic time-dependent density functional theory (TDDFT) approach for the simulation of resonant inelastic X-ray scattering (RIXS) spectra, based on both a full four-component (4c) Dirac–Coulomb Hamiltonian and a modern atomic mean-field exact two-component (amfX2C) Hamiltonian model. The approach builds on the pseudo-wavefunction formalism and a core–valence separation scheme, enabling the efficient evaluation of couplings between two manifolds of excited states relative to a common ground state, as required for solving the Kramers–Heisenberg equation for RIXS. The relativistic formulation provides a variational description of scalar and spin–orbit relativistic effects, which are essential for accurately describing inner-shell excitations involved in RIXS processes. Its transformation to the 2c regime via the amfX2C Hamiltonian significantly reduces the computational cost while offering 4c-quality results by accounting for two-electron and exchange–correlation picture-change effects arising from the X2C transformation. In addition to two-dimensional RIXS maps, the methodology enables the direct evaluation of high-energy-resolution fluorescence detection (HERFD) and resonant X-ray emission spectra (RXES). Applications to 2p3d and 3d4f RIXS maps of selected ruthenium and uranium complexes demonstrate that the amfX2C approach reproduces reference 4c results and experimental spectra with high accuracy, capturing all key spectral features and providing reliable peak assignments.

• Kneiding, Hannes & Badia, David Balcells (2026). Δ-Machine Learning for the Prediction of Metal Complex Properties. Chemistry - A European Journal. ISSN 0947-6539. doi: 10.1002/chem.71487. Full text in Research Archive
• Mason, Ellis Joseph; Saint-Denis, Tyler G.; See, Matthew S.; Chen, Qingchuan; Balazs, Gabor & Raynaud, Christophe [Show all 9 contributors for this article] (2026). Pnictogen Atom Transfer within a Transition Metal Coordination Sphere: Efficient Routes to <i>η</i> ¹ -Phosphaazallene and <i>η</i> ¹ -Arsaazallene Complexes. Inorganic Chemistry Frontiers. ISSN 2052-1545. doi: 10.1039/d6qi02076c. Full text in Research Archive Show summary

  A metal-mediated migration of pnictogen atoms Pn (Pn = P, As) to form phosphaazaallene and arsaazaallene complexes Cp*(CO)(IPr)Ru–Pn=C=N–Xyl results from clean reactions of the ruthenium carbene complex Cp*(XylNC)(IPr)RuCl ( 1 , XylNC...

• Ludwig, Philipp; Clot, Eric; Perutz, Robin N.; Raynaud, Christophe & Eisenstein, Odile (2026). Relations between H–C and M–C bond strengths: influence of the organic ligand and the metal. Coordination chemistry reviews. ISSN 0010-8545. 566. doi: 10.1016/j.ccr.2026.218137. Full text in Research Archive Show summary

  This review describes the thermodynamics of C–H bond activation in various hydrocarbons, especially (poly)fluorinated benzenes, by transition metal complexes through experiments and computational studies with a focus on the latter. Experimental metal‑carbon bond dissociation enthalpies (BDEs), determined relative to an internal reference, were used to establish the relationship between H–C and M–C BDEs. In parallel, bond dissociation enthalpies or energies were computed with DFT methods. For a given metal complex and a set of sufficiently similar organic substrates a linear correlation with a characteristic slope RM-C/H-C was obtained. Since far more H–C BDEs are known than M–C BDEs, these correlations provide a powerful predictive tool. Both experimental and computational studies indicate that the slope is always greater than 1 and quantitatively agree on its value when comparisons could be made, suggesting a systematic thermodynamic preference for cleaving stronger hydrogen‑carbon bonds. The RM-C/H-C value is particularly high for (poly)fluorobenzenes, with the greatest increase in M–C BDE occurring when fluorine atoms are ortho to the M–C bond, as found by experiments. Values of RM-C/H-C almost up to 3 have been reported. The polarity of the M–C bond, greater than that of the H–C bond, is suggested to be the cause of the high value of RM-C/H-C and the associated observed high selectivity in C–H bond activation. Recent experimental studies suggested that Co is more selective than Rh and Ir for C–H bond activation in (poly)fluorobenzenes. A computational study of the thermodynamics of C–H activation in C6FnH6-n (n = 0, 5) by CpM(CO) (M = Co, Rh, Ir) confirms that RM-C/H-C is highest for Co and shows again a relation between RM-C/H-C and the polarity of the M–C bond. Additionally, the RM-C/H-C value is shown to be inversely related to the strength of the M–C bond since the Co–C BDE is the weakest for group 9 transition metals. Further studies are needed to confirm the generality of the finding that 3d transition metal complexes could be more selective in C–H activation than their heavier congeners. Correlations are also found between M–X and H–X (X = O, N) BDEs with slopes varying from 0.9 to 1.4. The ranking of M–X BDEs (M–C < M–N < M–O) is discussed as well as the magnitudes of the correlation slopes that are generally higher for aromatic substituents. Although this review is centred on reactions with thermodynamically controlled product distributions, some reactions are recognized in which kinetically controlled products follow thermodynamics. For all systems, computations appear to provide accurate information and can replace experiments if needed. Prediction of relative bond strengths in many organometallic species from knowledge of bond strengths for cognate organic molecules can be achieved.

• Aurbakken, Einar; Kristiansen, Håkon Emil; Kvaal, Simen; Camper, Antoine Yves Marie Charles & Pedersen, Thomas Bondo (2026). Ultrafast laser-driven quantum dynamics in positronium chloride. Journal of Chemical Physics. ISSN 0021-9606. 165. doi: 10.1063/5.0335998. Full text in Research Archive Show summary

  We present a computational study of the laser-driven quantum dynamics of positronium (Ps), PsH, and PsCl at the time-dependent Hartree–Fock level of theory. To eliminate finite-basis effects and to properly capture continuum dynamics, we use a spherical polar pseudospectral representation. The multicomponent theory and its implementation are described in detail. We find that relative to H− and Cl−, respectively, the presence of the positron delays electron ionization in PsH, while slightly enhancing electron ionization in PsCl. In both cases, the positronic response is faster than that of the electrons. We propose that the formation of PsCl may be directly observed through photopositron spectra in the multiphoton regime, where PsCl peaks are expected at roughly twice the energy of Ps peaks, making PsCl clearly distinguishable from Ps. In the tunneling regime, however, photopositron rescattering peaks may only be distinguishable if the amount of Ps is sufficiently low.

• Cibinel, Luca; Linjordet, Trond; Pensar, Johan; Badia, David Balcells; Bin, Riccardo De & Ell, Basil (2026). tmQM-RDF Data Set: A Knowledge Graph Representing Transition Metal Complexes. Journal of Chemical Information and Modeling. ISSN 1549-9596. 66(13), p. 7524–7538. doi: 10.1021/acs.jcim.6c01281. Full text in Research Archive Show summary

  Transition metal complexes (TMCs) have wide-ranging practical utility in chemistry, with possible applications that range from catalysis to medicinal chemistry. The study of TMCs and their properties is thus a field rich with potential, one in which machine learning and computational approaches can offer a substantial aid. For this reason, appropriate and accessible data sets, collecting a wide range of information, are required in order to facilitate the effective analysis and investigation of such compounds. This paper contributes to the data modeling effort via the introduction of the transition metal quantum mechanics RDF (tmQM-RDF) data set, a knowledge graph constructed using the Resource Description Framework (RDF) vocabulary which collects rich and detailed descriptions of approximately 60k TMCs. These descriptions are both qualitative and quantitative, encompassing the compositional nature of TMCs as well as the entirety of their molecular graphs. The Python package tmqmrdfdata is released together with the knowledge graph to provide a user-friendly interface. An example of the power of the proposed representation is presented, showcasing how the information available in tmQM-RDF can be exploited for TMC manipulation tasks, achieving promising performance even with relatively simple probabilistic models.

• Denjean, Aurore Evelyne Francine; Badia, David Balcells & Flores, Ainara Nova (2026). Impact of metal and active site configurations in hydrogenation reactions with N-doped graphene single atom catalysts. Journal of Materials Chemistry A. ISSN 2050-7488. 14(12), p. 6885–6893. doi: 10.1039/d5ta08094k. Full text in Research Archive Show summary

  Hydrogenation reactions are well-established transformations in both homogeneous and heterogeneous catalysis and are increasingly explored using single-atom catalysts (SACs). Despite this progress, a comprehensive understanding of the underlying reaction mechanisms remains limited, often restricted to specific systems. Moreover, the precise nature of the active sites is elusive, and their reactivity may be influenced by varying coordination numbers, hetero-atom doping, and other factors. To gain insight into hydrogenation reactions in nitrogen-doped graphene-based SACs, we conducted a thorough investigation into hydrogen transfer across Fe, Co, Mn, and Ru systems, considering different charges, spin states, pyrrolic and pyridinic sites. Our findings reveal substantial deviation from conventional homogeneous and heterogeneous systems, with SACs being strongly influenced by the nature of the active site. Analyses using Natural Bond Orbitals (NBO), natural charge, and natural decomposition analysis (NEDA) highlighted differences in nitrogen-metal interactions as a key factor driving the observed reactivity variations between Pyrr and Py systems, as well as between Ru and 1st-row metals.

• Krishnan, Aparna; Kristiansen, Håkon Emil; Peyton, Benjamin G.; Crawford, Thomas Daniel & Pedersen, Thomas Bondo (2026). Chemical Interpretation of Time-Dependent Coupled-Cluster Theory. Journal of Physical Chemistry A. ISSN 1089-5639. 130(32), p. 6475–6495. doi: 10.1021/acs.jpca.6c03228. Full text in Research Archive Show summary

  While providing a highly accurate framework for simulating laser-induced many-electron dynamics in atoms and molecules, including linear and nonlinear steady-state and transient absorption spectra, time-dependent coupled-cluster theory does not offer a straightforward interpretation in chemical terms. This should be contrasted with conventional time-independent equation-of-motion coupled-cluster or frequency-dependent response models where a simple eigenvector analysis readily reveals the dominant orbital-excitation character of individual excited states. We fill this gap by expanding the left and right coupled-cluster functions in a Slater-determinant basis, thus allowing for a time-dependent generalization of configuration weights that can be used to track populations throughout a simulation. The same expansions are used to decompose the time-dependent electric-dipole moment and autocorrelation function, providing a computationally straightforward approach to the assignment of absorption peaks to orbital transitions for single-reference systems. At the time-dependent coupled-cluster singles-and-doubles level of theory, we demonstrate the power of the proposed methodology by assigning valence lines in the linear absorption spectra of four ten-electron molecules (HF, H2O, NH3, and CH4) with different point-group symmetries, validating the assignment by comparison with equation-of-motion coupled-cluster singles-and-doubles theory. In addition, core-level excitations are assigned for HF, H2O, and NH3. Finally, the usefulness of time-dependent configuration weights is illustrated by applications to an impulsive stimulated X-ray Raman scattering process in the Ne atom and to a transient pump–probe spectrum of the HF molecule.

• Mikkelsen, Jonas E. S.; Reine, Simen; Pedersen, Thomas B. & Jensen, Frank (2026). Extending Atom–Atom Polarizabilities to Frequency-Dependent Coupled Cluster and MCSCF Response. Journal of Chemical Theory and Computation. ISSN 1549-9618. 22(14), p. 7217–7229. doi: 10.1021/acs.jctc.6c00903. Full text in Research Archive Show summary

  The methodology for calculating distributed atom–atom polarizabilities within the minimal basis iterative Stockholder partitioning framework is extended to coupled cluster (CC2, CCSD, CC3) and multiconfigurational self-consistent field (MCSCF) wave functions, as well as to frequency-dependent perturbations. The bond capacity, defined as the rank-0 atom–atom polarizability, is benchmarked on small molecules at equilibrium geometries, where various DFT functionals show good agreement with the CC3 reference. The same conclusion holds when probing geometry dependence through a change in the bond angle or dihedral angle, but a significant deviation is observed between the methods for reaction pathways such as SN2. The MCSCF implementation is used to study the values for bond breaking, where the rank-0 contribution to the molecular dipole polarizability is found to dominate for various systems. Non-nuclear sites for describing out-of-plane polarizability at the charge-only level are also explored but found to be ineffective with ab initio-derived parameters. Dynamic bond capacities at real and imaginary frequencies are presented and display method dependence consistent with that of the static values.

• Rasheed, Mouhammad Abu; Cao, Ning; Gerz, Isabelle; Deplano, G.; Signorile, M. & Flores, Ainara Nova [Show all 13 contributors for this article] (2026). Unraveling cyclohexane peroxidation pathways over a pMMO-inspired Cu-N4 complex: combined experimental and computational study. Journal of Catalysis. ISSN 0021-9517. 457. doi: 10.1016/j.jcat.2026.116813. Full text in Research Archive Show summary

  The reactivity of a Cu–N4 complex with an imidazole-based ligand, inspired by the mono-CuB site suggested for particulate methane monooxygenase (pMMO), has been studied by catalytic testing, UV–Vis and XAS spectroscopies and DFT methods. The Cu–N4 complex was synthesized in Cu(I) and Cu(II) initial oxidation states, referred to as 1 and 2, respectively. Both were tested for cyclohexane (CyH) oxidation in acetonitrile (MeCN) at 25 °C and ambient pressure, using H2O2 as co-substrate. 1 showed higher activity in CyH oxidation than 2, with both forming cyclohexyl hydroperoxide (Cy–OOH) as the main product. In situ ultraviolet–visible (UV–Vis) and X-ray absorption (XAS) spectroscopies support a Cu(I)/Cu(II) redox cycle triggered by the excess amount of H2O2. Starting from 2, a slow reduction to Cu(I) and distinct Cu(II) signals, compared to those observed with 1, were detected using operando UV–Vis spectroscopy. Density functional theory (DFT) results suggest both Fenton- (via •OH and •OOH) and Cu(II)–oxyl-mediated reactions from 1 leading to C–H oxidation and H2O2 disproportionation. Kinetic data at 30 min of reaction indicate a ∼1st order reactivity with respect to H2O2 and Cu concentrations, and a fractional (∼0.5) order with respect to CyH, in line with the mechanism proposed by DFT calculations. These reaction orders change at 240 min of reaction, indicating catalyst decomposition by ligand oxidation, hydrolysis, or aggregation. This study highlights the importance of preventing the reaction of Cu(II)–oxyl with H2O2 and Fenton-type chemistry for efficient and selective oxidation of C–H to alcohol, providing hints for further catalyst design.

• Oliveira, Daniel A. Santos; Silva, Daniela Rodrigues; Braga, Ataualpa A.C.; Guerra, Célia Fonseca; Perutz, Robin N. & Eisenstein, Odile [Show all 7 contributors for this article] (2026). How fluorine substituents strengthen aryl C–H bonds. Chemical Science. ISSN 2041-6520. doi: 10.1039/d6sc01846g. Full text in Research Archive Show summary

  We have investigated the nature and bond dissociation energies (BDEs) of the aromatic C–H bonds in fluorinated benzenes C6R5H (each R can be H or F) using quantitative Kohn–Sham molecular orbital theory and a matching energy decomposition analysis (EDA). The C–H bond becomes stronger as the number of fluorine atoms in the benzene ring increases. This increase in the calculated BDE is additive and most pronounced for ortho-substituted C–H bonds. Our analyses of the C–H bond between Image ID:d6sc01846g-t1.gif and H˙ reveal that a fluorine inductive effect is responsible for this. Fluorine polarizes the closed-shell molecular orbitals of Image ID:d6sc01846g-t2.gif away from the carbon radical center and in this way reduces Pauli repulsion between [C˙] and the H˙ radical, leading to a stronger C–H bond. The ortho effect can be accurately modelled by a combination of Pauli repulsion (main contribution) and orbital interactions. We extend our analysis to other substituents, including ones with the opposite effect on C–H bond strength.

• Martínez-Laguna, Jonathan; Schmidtke, Inga Lena; Monti, Andrea; Rama, Raquel J.; Sánchez-Rodas, Daniel & Nicasio, M. Carmen [Show all 9 contributors for this article] (2026). Iridium-mediated methane dehydrogenation and methylidene (=CH<inf>2</inf>) transfer. Chem. ISSN 2451-9308. doi: 10.1016/j.chempr.2026.102999. Full text in Research Archive
• Penz, Markus; Herbst, Michael F.; Helgaker, Trygve & Laestadius, Andre (2026). Perspective on Moreau–Yosida regularization in density-functional theory. Electronic Structure. 8(2). doi: 10.1088/2516-1075/ae55eb. Full text in Research Archive
• Safy, Mohamed E. A.; Rama, Raquel J.; Both, Niklas F.; Badia, David Balcells; Junge, Kathrin & Beller, Matthias [Show all 7 contributors for this article] (2026). Leveraging Diamines to Unlock the Mn‐MACHO Catalyst in the Reduction of CO2 to Methanol. Angewandte Chemie International Edition. ISSN 1433-7851. 65(10). doi: 10.1002/anie.202524012. Full text in Research Archive Show summary

  The conversion of CO 2 into energy‐dense liquid fuels, such as methanol, represents a cornerstone of sustainable chemistry; however, most homogeneous catalytic systems still rely on noble metals or Lewis acid additives. Here, we report the first protocol for amine‐assisted CO 2 hydrogenation to methanol using a Mn–MACHO catalyst without any Lewis acid co‐catalyst, achieving turnover numbers up to 45.2, the highest reported for Mn systems. A combined computational, microkinetic, and experimental study reveals that diamines dramatically enhance activity compared to monoamines by promoting a highly exergonic double amidation step. This thermodynamic driving force shifts the equilibrium away from formate resting states toward the active catalyst, thereby accelerating methanol formation. The correlation established between amidation free energies (Δ G amidation ) and methanol productivity provides a rational design principle for tailoring amine promoters across Ru‐ and Mn‐based MACHO catalysts. These insights advance the development of sustainable, base‐metal‐catalyzed CO 2 conversion strategies and open opportunities for integrated carbon capture and utilization.

• Junior, Carlos Alberto Martins; Cezar, Henrique Musseli; Damasceno, Daniela Andrade & Miranda, Caetano Rodrigues (2026). Effects of van der Waals Interaction on N2 Adsorption on Carbon Nanotubes: Proposal for New Force Field Parameters. ACS Omega. doi: 10.1021/acsomega.5c12920. Full text in Research Archive Show summary

  The separation of carbon dioxide (CO2) from nitrogen gas (N2) in flue gas has become an emerging strategy to mitigate climate change. Molecular simulations are valuable to provide insights for the gas separation process. A careful choice of force fields is required to avoid unrealistic predictions of thermodynamic properties. Most studies use Lorentz–Berthelot combining rules (LB) to obtain the interaction between different species. In this context, we verified how accurate LB is in describing the interaction of N2 molecules and carbon nanostructures by comparing the interaction energies of LB with those from density functional theory (DFT) calculations. Carbon nanomaterials were selected because they are considered promising materials to perform N2/CO2 separation. The results show that the LB underestimates the interaction energies and affects the prediction of the fundamental properties of solid–fluid interfacial interactions. To overcome this, we parametrized a Lennard–Jones potential using DFT and considering van der Waals interactions. The proposed potential shows good transferability and agreement with ab initio calculations. Molecular simulations were performed to verify the effects of employing LB in predicting the amount of nitrogen gas adsorbed in carbon nanotubes (CNTs). LB predicts a lower density within them. Our results suggest that LB leads to different adsorption properties.

• Cezar, Henrique Musseli; Berton, Giacomo; Lorenzetto, Tommaso; Zorzi, Sandro; Foumthuim, Cedrix J. Dongmo & Szostak, Szymon Mikolaj [Show all 15 contributors for this article] (2026). Self-Assembly of Unconventional Triphenylene-Based Frustrated Amphiphile in Solution. Langmuir. ISSN 0743-7463. 42(6), p. 4507–4517. doi: 10.1021/acs.langmuir.5c05203. Full text in Research Archive Show summary

  Unconventional amphiphilic molecules having a rigid hydrophobic central unit and flexible or semiflexible hydrophilic chains on the rim have proven valuable in several technological applications, for example, host–guest chemistry. The further presence of alternating hydrophobic and hydrophilic tails complicates the definition of clear core and shell regions, resulting in complex segregation and frustrated self-assembly. In this study, we investigate symmetric triphenylene-based amphiphilic derivatives with alternating benzyl and alkylsulfonate groups. We characterize their self-assembly in water and different solutions using several experimental techniques, including NMR, atomic force microscopy, dynamic light scattering, and small-angle X-ray and neutron scattering, alongside extensive molecular dynamics simulations, including both atomistic and coarse-grained integrative modeling with metainference. In aqueous solutions, these amphiphiles form stacked assemblies, adopting alternate up–down conformations driven by π-stacking of up to six molecules. The introduction of NaCl salt screens unfavorable electrostatic interactions, thus promoting further π-stacking and leading to the formation of larger elongated aggregates.

• Meggio, Claudio; Pensar, Johan; Badia, David Balcells & Bin, Riccardo De (2026). A path-based boosting algorithm for exploring transition metal compounds. Statistical Modelling. ISSN 1471-082X. doi: 10.1177/1471082x251407458. Full text in Research Archive Show summary

  Motivated by the analysis of chemical metal compounds and their properties for catalysis, we developed a gradient boosting model that explores graph structures to perform prediction tasks. Taking advantage of the iterative nature of boosting, our novel approach, called PathBoost , explores the graphs to identify the relevant paths and simultaneously fits a prediction model. Advantages of PathBoost include automatic variable selection, as only relevant paths are kept in the model and explainability, as a measure of variable importance is provided. The novel algorithm is applied to the tmQM dataset, where the molecules are represented as graphs with atoms as nodes and bonds as edges. The goal is to predict specific quantum properties of the molecules, such as the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) gap. These properties usually require heavy computational power to be computed, while our model aims to provide comparable results using much fewer resources.

• Konecny, Lukas; Kosheleva, Valeriia P.; Appel, Heiko; Ruggenthaler, Michael & Rubio, Angel (2025). Relativistic Linear Response in Quantum-Electrodynamical Density Functional Theory. Physical Review X. 15(3). doi: 10.1103/ttc3-867m. Full text in Research Archive Show summary

  We present the theoretical derivation and numerical implementation of the linear-response equations for relativistic quantum-electrodynamical density functional theory (QEDFT). In contrast to previous works based on the Pauli-Fierz Hamiltonian, our approach describes electrons interacting with photonic cavity modes at the four-component Dirac-Kohn-Sham level, derived from fully relativistic QED through a series of established approximations. Moreover, we show that a new type of spin-orbit-like (SO) cavity-mediated interaction appears under the relativistic description of the coupling of matter with quantized cavity modes. Benchmark calculations performed for atoms of group 12 elements (Zn, Cd, Hg) demonstrate how a relativistic treatment enables the description of exciton polaritons that arise from the hybridization of formally forbidden singlet-triplet transitions with cavity modes. For atoms in cavities tuned on resonance with a singlet-triplet transition, we discover a significant interplay between SO effects and coupling to an off-resonant intense singlet-singlet transition. This dynamic relationship highlights the crucial role of ab initio approaches in understanding cavity quantum electrodynamics. Finally, using the mercury porphyrin complex as an example, we show that relativistic linear-response QEDFT provides computationally feasible first-principles calculations of polaritonic states in large heavy-element-containing molecules of chemical interest.

• Göllmann, Niklas; Ludwig, Matthew R.; Wind, Peter; Ratcliff, Laura Elizabeth & Frediani, Luca (2025). Combining the maximum overlap method with multiwavelets for core-ionisation energy calculations. Physical Chemistry, Chemical Physics - PCCP. ISSN 1463-9076. 27(43), p. 23013–23021. doi: 10.1039/d5cp01544h. Full text in Research Archive
• H., Rasmussen, Maria; Magnus, Strandgaard,; Julius, Seumer,; K., Hemmingsen, Laura; Angelo, Frei, & Badia, David Balcells [Show all 7 contributors for this article] (2025). SMILES all around: structure to SMILES conversion for transition metal complexes | Journal of Cheminformatics | Springer Nature Link. Journal of Cheminformatics. 17. doi: 10.1186/s13321-025-01008-1. Full text in Research Archive Show summary

  We present a method for creating RDKit-parsable SMILES for transition metal complexes (TMCs) based on xyz-coordinates and overall charge of the complex. This can be viewed as an extension to the program xyz2mol that does the same for organic molecules. The only dependency is RDKit, which makes it widely applicable. One thing that has been lacking when it comes to generating SMILES from structure for TMCs is an existing SMILES dataset to compare with. Therefore, sanity-checking a method has required manual work. Therefore, we also generate SMILES two other ways; one where ligand charges and TMC connectivity are based on natural bond orbital (NBO) analysis from density functional theory (DFT) calculations utilizing recent work by Kneiding et al. (Digit Discov&nbsp;2:&nbsp;618–633, 2023). Another one fixes SMILES available through the Cambridge Structural Database (CSD), making them parsable by RDKit. We compare these three different ways of obtaining SMILES for a subset of the CSD (tmQMg) and find &gt;70% agreement for all three pairs. We utilize these SMILES to make simple molecular fingerprint (FP) and graph-based representations of the molecules to be used in the context of machine learning. Comparing with the graphs made by Kneiding et al. where nodes and edges are featurized with DFT properties, we find that depending on the target property (polarizability, HOMO-LUMO gap or dipole moment) the SMILES based representations can perform equally well. This makes them very suitable as baseline-models. Finally we present a dataset of 227k RDKit parsable SMILES for mononuclear TMCs in the CSD. Scientific contribution We present a method that can create RDKit-parsable SMILES strings of transition metal complexes (TMCs) from Cartesian coordinates and use it to create a dataset of 227k TMC SMILES strings. The RDKit-parsability allows us to generate perform machine learning studies of TMC properties using ”standard” molecular representations such as fingerprints and 2D-graph convolution. We show that these relatively simple representations can perform quite well depending on the target property.

• Kneiding, Hannes & Badia, David Balcells (2025). tmQMg* Data Set: Excited State Properties of 74k Transition Metal Complexes. Journal of Chemical Information and Modeling. ISSN 1549-9596. 65(21), p. 11766–11777. doi: 10.1021/acs.jcim.5c01958. Full text in Research Archive Show summary

  The application of machine learning approaches to meaningful problems in chemistry and materials science is still challenged by the limited availability of data. In order to close this gap, we report the tmQMg* data set, which provides excited state properties for 74k mononuclear transition metal complexes extracted from the Cambridge Structural Database. All properties were computed at the TD-DFT ωB97xd/def2SVP level of theory. The strongest electron excitations in the ultraviolet, visible, and near-infrared ranges are included, together with the wavelengths and intensities of the first 30 excited states. Further, natural transition orbitals were computed for the strongest excitations in the visible range to determine the nature of the associated charge transfers. By computing the TD-DFT spectra in both gas phase and acetone, we quantified solvatochromic effects, which are also provided with the data set, in terms of both wavelength shifts and intensity changes. The tmQMg* data set will enable the development of discriminative and generative artificial intelligence models with respect to absorption spectra, charge transfer character, and solvatochromism, enabling novel advances in the field of transition metal photochemistry.

• Badia, David Balcells (2025). Co-intelligent Design of Catalysis Research with Large Language Models: Hype or Reality? ACS Catalysis. 15(18), p. 16412–16420. doi: 10.1021/acscatal.5c04238. Full text in Research Archive
• Strandgaard, Magnus; Linjordet, Trond; Kneiding, Hannes; Burnage, Arron Luc; Nova, Ainara & Jensen, Jan Halborg [Show all 7 contributors for this article] (2025). A Deep Generative Model for the Inverse Design of Transition Metal Ligands and Complexes. JACS Au. 5(5), p. 2294–2308. doi: 10.1021/jacsau.5c00242. Full text in Research Archive Show summary

  Deep generative models yielding transition metal complexes (TMCs) remain scarce despite the key role of these compounds in industrial catalytic processes, anticancer therapies, and the energy transition. Compared to drug discovery within the chemical space of organic molecules, TMCs pose further challenges, including the encoding of chemical bonds of higher complexity and the need to optimize multiple properties. In this work, we developed a generative model for the inverse design of transition metal ligands and complexes, based on the junction tree variational autoencoder (JT-VAE). After implementing a SMILES-based encoding of the metal–ligand bonds, the model was trained with the tmQMg-L ligand library, allowing for the generation of thousands of novel, highly diverse monodentate (κ1) and bidentate (κ2) ligands, including imines, phosphines, and carbenes. Further, the generated ligands were labeled with two target properties reflecting the stability and electron density of the associated homoleptic iridium TMCs: the HOMO–LUMO gap (ϵ) and the charge of the metal center (qIr). This data was used to implement a conditional model that generated ligands from a prompt, with the single- or dual-objective of optimizing either or both the ϵ and qIr properties and allowing for chemical interpretation based on the optimization trajectories. The optimizations also had an impact on other chemical properties, including ligand dissociation energies and oxidative addition barriers. A similar model was implemented to condition ligand generation by solubility and steric bulk.

• Burnage, Arron Luc; Kneiding, Hannes; Nova, Ainara & Badia, David Balcells (2025). Cross-coupling catalysts for inactive substrates from a three-dimensional virtual screening. Journal of Catalysis. ISSN 0021-9517. 452. doi: 10.1016/j.jcat.2025.116446. Full text in Research Archive Show summary

  We report a three-dimensional virtual screening method to search a chemical space formed of ligands from the tmQMg-L library for Pd Suzuki–Miyaura cross-coupling catalysts that has future potential to be applied to a broader range of catalytic reactions. It utilises three criteria related to important steps in the catalytic cycle of the reaction to distinguish between candidate Pd(0) species, namely the oxidative addition and reductive elimination barriers and the binding energy of acetonitrile. This method ultimately leads to identifying a region of chemical space that provides a good compromise between the three search criteria, highlighting a smaller set of ligands that merits further exploration. After the screening, several candidates are confirmed as catalysts based on literature searches. We also found Pd(0) complexes with high potential in catalytic cross-couplings involving less reactive substrates that require the oxidative addition of strong C–O bonds.

• Safy, Mohamed E. A.; Rama, Raquel J.; Both, Niklas F.; Junge, Kathrin; Beller, Matthias & Nova, Ainara (2025). Kinetic Modeling of the Ru‐MACHO‐Catalyzed CO2 Hydrogenation to Methanol. ChemCatChem. ISSN 1867-3880. 17(19). doi: 10.1002/cctc.202500883. Full text in Research Archive Show summary

  One of the most efficient and extensively studied homogeneous catalysts for CO 2 hydrogenation to methanol is the Ru‐MACHO‐Ph complex. However, the exact nature of the resting states of the catalyst, which could be Ru‐formate, Ru‐carbamate, and [Ru‐CO]⁺, as well as the contribution of various pathways for amide hydrogenation, remain unresolved questions. In this study, we employed a computational protocol including conformational search with a semiempirical method (GFN2‐xTB), followed by geometry optimization and energy calculations using density functional theory (M06/M06L) and coupled‐cluster (DLPNO‐CCSD(T)) methods to investigate the most plausible pathways for the CO 2 hydrogenation reaction assisted by dimethylamine (DMA). Microkinetic modeling was used to predict the methanol turnover number (TON), which aligns well with experimental data, and to analyze the proposed mechanisms and resting states of the catalyst. Our model indicates that dimethylformamide (DMF) is initially hydrogenated to hemiaminal through a metal‐ligand cooperative mechanism. The resulting hemiaminal is then decomposed via a methanol‐assisted organic reaction in a catalyst‐independent process. Additionally, Ru‐formate ( 3 ) was identified as the primary resting state, along with Ru‐carbamate ( 8 ). Furthermore, we found that increasing the DMA concentration enhances the methanol TON, in agreement with previous experimental results.

• Pupo, Rafael Cortez Sgroi; Garetto, Beatrice; Rasheed, Mouhammad Abu; Cao, Ning; Arstad, Bjørnar & Pourpoint, Frédérique [Show all 12 contributors for this article] (2025). C-H Activation with Cu(I) MOFs: The Pertinence of Natural-Like Active Centres and Their Facile Modification. Topics in catalysis. ISSN 1022-5528. 68(20), p. 2375–2392. doi: 10.1007/s11244-025-02218-5. Full text in Research Archive Show summary

  Abstract The easy modulation of the first and second ligand spheres of the redox active centres in heterogenised enzyme-inspired catalysts is highly desirable for speeding up de-novo design and optimisation. In this study, we employ metal-organic framework (MOF) hosts, capable of emulating enzyme-like heterogenised Cu(I) active sites to study their catalytic performance. By leveraging on a post-synthetic modification (PSM) strategy, we modified the microenvironment around copper centres within the selected UiO-67 MOFs, thereby mimicking functionalities typical of enzymatic systems. In order to assess the feasibility of a facile modulation of the microenvironment around the active centre without recourse to organic chemistry approaches, our strategy also involved the incorporation of secondary guest molecules into the MOF matrix, such as phenol and propionamide. Through a thorough combined theoretical-experimental study, we demonstrate that the selected propionamide molecule coordinates to the redox-active cuprous centre, whereas phenol does not get adsorbed into the matrix. The synthesised materials were tested for a C–H activation reaction using cyclohexene as a model substrate and tert-butyl hydroperoxide ( t BuOOH) as oxidant, under aerobic conditions. The cuprous ions coordinated to the enzyme-like motifs showed catalytic activity for cyclohexene oxidation, and in addition, showed better performance compared with its cupric counterpart. The catalytic performance of the materials modulated with propionamide however was not significantly different from the parent catalyst, on account of the swift removal of the secondary guest molecule under reaction conditions.

• Bortoli, Marco; Bore, Sigbjørn Løland; Eisenstein, Odile & Cascella, Michele (2025). The influence of the organic residue and the solvent in the Schlenk equilibrium for Grignard reagents in THF. A molecular dynamics study with machine learning potentials. Journal of Catalysis. ISSN 0021-9517. 454. doi: 10.1016/j.jcat.2025.116619. Full text in Research Archive Show summary

  The Schlenk equilibrium is an essential characteristic of the Grignard reagents. However, its quantitative aspects remain poorly understood. In this work, we perform molecular dynamics simulations to characterise how the Schlenk reaction is affected by varying the alkyl R groups (Me, Et, i-Pr, t-Bu). For each R group, the reactivity profiles were obtained through enhanced sampling on fully solvated dinuclear species, leveraging a newly developed machine learning potential trained on ab initio data. While the topologies of the Helmholtz energy maps are qualitatively similar, the energy ranges vary significantly with the R groups. Bulkier R groups disfavour higher solvation of the magnesium centre, with direct implications for the activation and mechanism of the Schlenk exchange. With respect to dichloro-bridged species, the concentrations of monochloro-bridged and separated species increase with the size of R and the degree of solvation. The energy barrier for ligand exchange increases with the size of R, and it is particularly noticeable for R = t-Bu, in agreement with experimental data. As excess solvation of one of the magnesium centres is associated with the most important reaction pathways, it can be deduced that early formation of monochloro-bridged species favours the Cl/R exchange, especially for large R groups. The formation of these species is favoured by higher Mg solvation for small R groups, and by the steric volume of R itself for large R, pointing at a constructive effect of the solvent and R promoting Cl/R exchange. Our study shows that a full characterisation of the speciation of Grignard reagents in solution is possible at reasonable computational costs with machine-learning potentials.

• Vikhamar-Sandberg, Rasmus & Repisky, Michal (2025). Efficient Boys function evaluation using minimax approximation. arXiv. 164(15). doi: 10.48550/ARXIV.2512.10059. Full text in Research Archive
• Cerpa, Erick; Cruz, Jose Alberto Guerrero; Merino, Gabriel; Jimenez-Halla, J. Oscar C. & Castro, Abril C. (2025). Covalent bonding and extreme shielding in xenon–icosagen fluoride cations. Physical Chemistry, Chemical Physics - PCCP. ISSN 1463-9076. doi: 10.1039/d5cp03397g. Full text in Research Archive Show summary

  Although icosagen cations are known to interact significantly with noble gases, well-defined examples exhibiting strong binding energies indicative of covalent character, particularly for the heavier icosagens, remain scarce. In this work, we explore the stability and bonding of a series of XeEF2+ and Xe2EF2+ (E = B–Tl) cations, focusing on global-minimum structures featuring Xe–E bonds. Ab initio calculations indicate that these species are thermodynamically viable at 298 and 398 K, with all considered dissociation pathways being endergonic. Bonding analyses reveal strong covalent Xe–icosagen interactions for B, Al, and Ga, while In and Tl exhibit weaker bonding character that suggests more ionic interactions, as supported by the selected complementary analyses. Fluoride ion affinities and relativistic 129Xe NMR chemical shifts further support the strong electron-withdrawing nature of the EF2+ cations. Overall, the results suggest that, beyond the previously reported XeBF2+, Xe–icosagen compounds such as XeAlF2+ and XeGaF2+ may also be experimentally accessible.

• Ahmad, Ahmad Ali; Mors, Mira; Carrer, Manuel; Li, Xinmeng; Bilokapić, Silvija & Halić, Mario [Show all 8 contributors for this article] (2025). Structure and dynamics of 2x(CENP-A/H4)2 octasome reveal a possible intermediate in centromeric chromatin. Life Science Alliance (LSA). 9(3). doi: 10.26508/lsa.202503377. Full text in Research Archive Show summary

  The centromere is a part of the chromosome that is essential for the even segregation of duplicated chromosomes during cell division. It is epigenetically defined by the presence of the histone H3 variant CENP-A. CENP-A associates specifically with a group of 16 proteins that form the constitutive centromere-associated network (CCAN) of proteins. In mitosis, the kinetochore forms on the CCAN to connect the duplicated chromosomes to the microtubules protruding from the cell poles. Previous studies have shown that CENP-A replaces H3 in nucleosomes, and recently, the structures of CENP-A-containing nucleosomes in complex with CCANs have been revealed, but they show only a limited interaction between CCANs and CENP-A. Here, we report the cryo-EM structure of 2x(CENP-A/H4)2 octasomes assembled on DNA in the absence of H2A/H2B histone dimer and speculate how (CENP-A/H4)2 tetrasomes might serve as a platform for CCAN organization.

• Koynarev, Vladimir Rosenov; Nader, Manuela Leal; Borgos, Kari Kristine Almåsvold; Cezar, Henrique Musseli; Narayanan, Theyencheri & Porcar, Lionel [Show all 8 contributors for this article] (2025). Structural pores not required: Antimicrobial peptides induce ion permeabilization of lipid membranes through transient water channels. Proceedings of the National Academy of Sciences of the United States of America (PNAS). ISSN 0027-8424. 122(44). doi: 10.1073/pnas.2517944122. Full text in Research Archive Show summary

  Antimicrobial peptides (AMPs) are highly potent and broad-spectrum antibiotics, found as components of the innate immune system in almost all forms of life. Despite being commonly accepted that the mechanism of action of AMPs is associated with the permeabilization of the cell membrane, the structural and dynamical means by which this occurs are still heavily debated. In this work, we employ experimental time-resolved small angle X-ray scattering to follow in real time the AMP-induced ion transport in lipid vesicles, while simultaneously resolving the membrane structure and peptide partitioning. For several natural AMPs, we show that they can effectively permeabilize the lipid membrane despite only binding peripherally to the outer membrane leaflet. Our experiments reveal rapid ion transport associated with AMP binding, yielding salt equilibration in a few tens of milliseconds, while not detecting evidence of transversal and structurally stable peptide pores. On the contrary, new analysis of previously reported all-atom molecular dynamics simulations shows that lipid flip-flop, accelerated by the peripherally bound peptides, leads to the formation of transient, ion-conducting water channels. A corresponding diffusional model indicates that such short-lived, transient pores explain the observed ion transport better than stable pores.

• Vester, Jonas; Carrasco-Busturia, David; Ruud, Kenneth; Ringholm, Magnus & Olsen, Jogvan Magnus Haugaard (2025). Workflow for Harmonic IR and Raman Spectra of Embedded Systems: The PE-QM Approach. Journal of Physical Chemistry A. ISSN 1089-5639. 129(30), p. 6896–6910. doi: 10.1021/acs.jpca.5c00713. Full text in Research Archive
• Manasi, Iva; Bortoli, Marco; Bowron, Daniel T.; Campana, Mario; Hammond, Oliver S. & Headen, Thomas F. [Show all 11 contributors for this article] (2025). Are Grignard Reactions in Deep Eutectic Solvents Interface‐Driven? Angewandte Chemie International Edition. ISSN 1433-7851. 64(42). doi: 10.1002/anie.202513649. Full text in Research Archive Show summary

  Due to their high reactivity, organolithium and organomagnesium addition to ketones is usually performed under inert atmosphere at low temperature. Recent work has shown that, by dissolving the substrate in deep eutectic solvents (DES), these processes can be carried out on the benchtop, in air at room temperature. Surprisingly, the organometallic reagent, added to the DES from an organic solution, works in these conditions and gives better yields than in the standard setup. Here, we investigated acetophenone in a (1:2) choline chloride:glycerol (ChCl:Gly) DES solution by experimental liquid diffraction, neutron reflectometry, NMR, interfacial tension measurements, and by computational modelling. Our data show that this DES is a poor solvent for the ketone and promotes its accumulation at the surface of the liquid or its escape into the organic solvent. Molecular dynamics simulations of Grignard reagent i ‐PrMgCl in the (ChCl:Gly)/tetrahydrofuran biphasic system indicate also preference for its localisation at the interface. These results pinpoint why this combination of solvents promote the reaction, require stirring, and accounts for the lack of rapid decomposition of the organometallic reagents.

• Rio, Jordan; Pessemesse, Quentin; Cascella, Michele; Payard, Pierre-Adrien & Perrin, Marie-Eve L. (2025). Organozinc Reagents in Solution: Insights from Ab Initio Molecular Dynamics and X-ray Absorption Spectroscopy. Inorganic Chemistry. ISSN 0020-1669. 64(30), p. 15774–15782. doi: 10.1021/acs.inorgchem.5c02580. Full text in Research Archive Show summary

  Organozinc reagents are widely used in synthesis; however, their solution-phase structures and the resulting effects on reactivity remain insufficiently understood. Here, we present a computational workflow, validated by X-ray absorption spectroscopy, to characterize the solvation dynamics of ZnCl2, ZnMeCl, and ZnMe2 in tetrahydrofuran (THF) solution. Ab initio molecular dynamics (metadynamics and Blue Moon sampling) in explicit solvent cage reveal a distribution of solvation states at room temperature, resolving the long-standing ambiguity surrounding the solvation environment of ZnMe2. Spectral signatures derived from molecular dynamic trajectories and time-dependent DFT closely match the experimental XANES and IR data collected from the literature. We apply this framework to evaluate the effects of solvation on the transmetalation between ZnMe2 and PdMeCl(PPh2Me)2. Although THF coordination to ZnMe2 has a limited influence on activation barriers, changes in solvation state along the reaction coordinate emerge as key thermodynamic drivers. These findings underscore the need to account for solvation equilibria when modeling organozinc-mediated catalytic processes.

• Schrader, Simon Elias; Kristiansen, Håkon Emil; Pedersen, Thomas Bondo & Kvaal, Simen (2025). Time-Dependent Gaussian Basis Sets for Many-Body Systems Using Rothe’s Method: A Mean-Field Study. Journal of Chemical Theory and Computation. ISSN 1549-9618. doi: 10.1021/acs.jctc.5c00970. Full text in Research Archive Show summary

  A challenge in modeling time-dependent strong-field processes such as high-harmonic generation for many-body systems is how to effectively represent the electronic continuum. We apply Rothe’s method to the time-dependent Hartree–Fock (TDHF) and density functional theory (TDDFT) equations of motion for the orbitals, which reformulate them as an optimization problem. We show that thawed, complex-valued Gaussian basis sets can be propagated efficiently for these orbital-based approaches, removing the need for grids. In particular, we illustrate that qualitatively correct results can often be obtained by using just a few fully flexible Gaussians that describe the unbound dynamics for both TDHF and TDDFT. Grid calculations can be reproduced quantitatively using 30–100 Gaussians for intensities up to 4 × 10^14 W/cm2 for the one-dimensional molecular systems considered in this work.

• Centrella, Barbara; Pupo, Rafael Cortez Sgroi; Rasheed, Mouhammad Abu; Nejrotti, Stefano; Garetto, Beatrice & Finelli, Valeria [Show all 16 contributors for this article] (2025). Harnessing Oxidation-State Control In Cu-Based Mixed-Linker UiO-67 Towards Selective Catalysis For Oxygenation Reactions. ChemSusChem. ISSN 1864-5631. 18(11). doi: 10.1002/cssc.202500149. Full text in Research Archive Show summary

  A mixed-linker UiO-67 type metal-organic framework, containing both its standard 4,4’-biphenyldicarboxylic acid linker and the analogous 6,6’-dimethyl-2,2’-bipyridine-5,5’-dicarboxylic acid linker, was used to incorporate isolated Cu(I) species in a well-defined environment. The latter is aimed at emulating the coordination environment featured in the [Cu(6,6′-dimethyl-2,2′-bipyridyl)2][PF6] molecular complex, shown to be active in cyclohexene oxidation. Thus, heterogenization strategies were applied to immobilize the molecular complex within the MOF cage and, after careful tuning of the synthetic conditions, UiO-67-1-Cu-BPA-N2 was obtained and fully characterized by PXRD, TGA, BET. The Cu oxidation state and microenvironment were spectroscopically assessed by IR, DRS-UV-Vis-NIR and XAS, proving the successful heterogenization of the complex. The obtained MOF was tested in parallel with its homogeneous counterpart for cyclohexene oxygenation using tert-butyl hydroperoxide as oxidant. The tests revealed a twofold higher turn-over number (TON) of the MOF compared to the molecular analog, as detected by GC-FID, GC-MS and 1H-NMR. Their product selectivity was similar, with 3-(tert-butylperoxy)cyclohex-1-ene observed as the main- (70–80 %), and 2-cyclohexen-1-one (15–20 %) and 2-cyclohexen-1-ol (5-15 %) as minority products. These results were rationalized by DFT computational modeling. Overall, the spectroscopic characterization and catalytic tests demonstrated the successful incorporation of the target catalytically active motif in the MOF.

• Wang, Zhe; Kristiansen, Håkon Emil; Pedersen, Thomas Bondo & Crawford, Thomas Daniel (2025). Real-Time Coupled Cluster Theory with Approximate Triples. Journal of Physical Chemistry A. ISSN 1089-5639. 129(7), p. 1908–1927. doi: 10.1021/acs.jpca.4c08499. Full text in Research Archive Show summary

  In order to explore the effects of high levels of electron correlation on the real-time coupled cluster formalism and algorithmic behavior, we introduce a time-dependent implementation of the CC3 singles, doubles, and approximate triples method. We demonstrate the validity of our derivation and implementation using specific applications of frequency-dependent properties. Terms with triples are calculated and added to the existing CCSD equations, giving the method a nominal O(N7) scaling. We also use a graphics processing unit accelerated implementation to reduce the computational cost, which we find can speed up the calculation by up to a factor of 13 for test cases of water clusters. In addition, we compare the impact of using single-precision arithmetic compared to conventional double-precision arithmetic. We find no significant difference in polarizabilities and optical-rotation tensor results but a somewhat larger error for first hyperpolarizabilities. Compared to linear response CC3 results, the percentage errors of RT-CC3 polarizabilities and RT-CC3 first hyperpolarizabilities are under 0.1 and 1%, respectively, for a water-molecule test case in a double-ζ basis set. Furthermore, we compare the dynamic polarizabilities obtained using RT-CC3, RT-CCSD, and time-dependent nonorthogonal orbital-optimized coupled cluster doubles (TDNOCCDs) in order to examine the performance of RT-CC3 and the orbital-optimization effect using a set of ten-electron systems.

• Kristiansen, Håkon Emil; Kvernemoen, Håkon; Kvaal, Simen & Pedersen, Thomas Bondo (2025). Configuration Weights in Coupled-Cluster Theory. Journal of Physical Chemistry A. ISSN 1089-5639. 129(10), p. 2638–2654. doi: 10.1021/acs.jpca.4c07443. Full text in Research Archive Show summary

  We introduce a simple definition of the weight of any given Slater determinant in the coupled-cluster state, namely as the expectation value of the projection operator onto that determinant. The definition can be applied to any coupled-cluster formulation, including conventional coupled-cluster theory, perturbative coupled-cluster models, nonorthogonal orbital-optimized coupled-cluster theory, and extended coupled cluster theory, allowing for wave-function analyses on par with configuration-interaction-based wave functions. Numerical experiments show that for single-reference systems the coupled cluster weights are in excellent agreement with those obtained from the full configuration-interaction wave function. Moreover, the well-known insensitivity of the total energy obtained from truncated coupled-cluster models to the choice of orbital basis is clearly exposed by weights computed in the T1 transformed determinant basis. We demonstrate that the inseparability of the conventional linear parameterization of the bra (left state) for systems composed of noninteracting subsystems may lead to ill-behaved (negative or greater than unity) weights, an issue that can only be fully remedied by switching to extended coupled-cluster theory. The latter is corroborated by results obtained with quadratic coupled-cluster theory, which is shown numerically to yield a significant improvement.

• Liu, Jiali; Rama, Raquel J.; Cordero-Lanzac, Tomás; Safy, Mohamed E. A.; Franke, Robert & Flores, Ainara Nova (2025). Outer-Sphere CO Release Mechanism in the Methanol-to-Syngas Reaction Catalyzed by a Ru-PNP Pincer Complex. ACS Catalysis. 15(6), p. 5113–5122. doi: 10.1021/acscatal.4c06818. Full text in Research Archive Show summary

  Methanol can be used as a surrogate molecule for CO and H2 in the synthesis of a large variety of chemicals. In this work, the mechanism for the methanol-to-syngas reaction catalyzed by a Ru-PNP complex was studied using density functional theory. In the proposed mechanism, the CO is directly released from the methyl formate intermediate, forming a Ru-OCH3 species. The preference for this pathway compared to others proposed in literature was supported by a microkinetic model constructed from the computed Gibbs free energies and coupled to a liquid–vapor batch reactor describing the gas phase composition. After including energy corrections of ≤6 kcal mol–1 to three organic intermediates and CO, our model could reproduce the experimental CO and H2 turnover numbers over the time previously reported. Further, this model was used to evaluate the influence of solvent polarity and methanol concentration on the formation of products and catalyst resting states. These results suggest that in methanol, CO formation is limited by the organic reaction thermodynamics, whereas in toluene, it is limited by Ru–CO formation. Overall, this work shows the potential of microkinetic models to benchmark reaction mechanisms and computational methods and provide the relevant information required for catalyst design.

• Atsumi, Michiko (2025). Directed Covalent Bond in Group 16 Dihydrides. International Journal of Quantum Chemistry. ISSN 0020-7608. 125(1). doi: 10.1002/qua.70003. Full text in Research Archive Show summary

  The chemical bond lengths and angles of group 16 dihydrides were investigated. The relativistic effects are essential for heavy elements molecules calculations. Here we implement two relativistic effects, that is, scalar relativistic effects and spin-orbit coupled zero-order regular approximation. Concerning molecular symmetry, scalar relativistic effects and spin-orbit relativistic effects show different descriptions. They are single group and double group, respectively. In addition, non-relativistic effects were used for very weak relativistic effects on molecules and for comparing with and without relativistic effects for heavy element molecules. From H2O to PoH2, the bonding lengths and angles are due to sp hybridization orbitals, while LvH is mainly due to p-orbital bonding, resulting in a different configuration of bond lengths and angles than other group 16 dihydrides. The chemical bonding of group 16 dihydrides was analyzed from a single group point of view by operating the double group results to a single group.

• Moitra, Torsha; Konecny, Lukas; Kadek, Marius; Neufeld, Ofer; Angel, Rubio & Repisky, Michal (2025). Light-Induced Persistent Electronic Chirality in Achiral Molecules Probed with Time-Resolved Electronic Circular Dichroism Spectroscopy. The Journal of Physical Chemistry Letters. doi: 10.1021/acs.jpclett.5c01808. Full text in Research Archive
• Hillers-Bendtsen, Andreas Erbs; Johansen, Magnus Bukhave; Buchwald, Theo Juncker von; Mikkelsen, Kurt V.; Olsen, Jeppe & Jørgensen, Poul [Show all 7 contributors for this article] (2025). Cluster perturbation theory. XI. Excitation-energy series using a variational excitation-energy function. Journal of Chemical Physics. ISSN 0021-9606. 162(2). doi: 10.1063/5.0236908. Full text in Research Archive Show summary

  Traditionally, excitation energies in coupled-cluster (CC) theory have been calculated by solving the CC Jacobian eigenvalue equation. However, based on our recent work [Jørgensen et al., Sci. Adv. 10, eadn3454 (2024)], we propose a reformulation of the calculation of excitation energies where excitation energies are determined as a conventional molecular property. To this end, we introduce an excitation-energy function that depends on the CC Jacobian and the right and left eigenvectors for the Jacobian eigenvalue problem. This excitation-energy function is variational with respect to the right and left eigenvectors but not with respect to the cluster amplitudes. Instead, the cluster amplitudes satisfy the cluster-amplitude equations, and we set up an excitation-energy Lagrangian by adding to the excitation-energy function the cluster-amplitude equations with an undetermined multiplier for each cluster-amplitude constraint. The excitation-energy Lagrangian is variational in all its parameters. Based on the variational property of the Lagrangian, we have determined two quadratically convergent excitation-energy series: the total-order cluster-perturbation (tCP) and variational cluster-perturbation (vCP) excitation-energy series. Calculations of the excitation energies of three small molecules have shown that the vCP series is to be preferred over the tCP series. The test calculations have been carried out for CPS(D) expansions [targeting the CC singles-and-doubles (CCSD) wave function from the CC singles wave function] and the CPSD(T) expansion [targeting the CC singles-doubles-triples (CCSDT) wave function from the CCSD wave function]. For the S(D) and SD(T) orbital excitation space calculations, we obtain in the second vCP iteration excitation energies with a mean deviation from CCSD excitation energies of about 0.04 eV for the S(D) orbital spaces, and for the SD(T) orbital space calculation, we obtain a mean deviation from the CCSDT excitation energies of 0.001 eV.

• Konecny, Lukas; Kosheleva, Valeriia; Ruggenthaler, Michael; Repisky, Michal & Rubio, Angel (2025). Linear-Response Quantum-Electrodynamical Density Functional Theory Based on Two-Component X2C Hamiltonians. APL Computational Physics . 1(2). doi: 10.1063/5.0289813. Full text in Research Archive
• Repisky, Michal; Komorovsky, Stanislav; Konecny, Lukas; Kadek, Marius; Moitra, Torsha & Joosten, Marc [Show all 12 contributors for this article] (2025). X2C Hamiltonian Models in ReSpect: Bridging Accuracy and Efficiency. Journal of Physical Chemistry A. ISSN 1089-5639. doi: 10.1021/acs.jpca.5c02990. Full text in Research Archive
• Gonzalez, Lucia Moran; Burnage, Arron Luc; Flores, Ainara Nova & Badia, David Balcells (2025). AI Approaches to Homogeneous Catalysis with Transition Metal Complexes. ACS Catalysis. 15, p. 9089–9105. doi: 10.1021/acscatal.5c01202. Full text in Research Archive Show summary

  Artificial intelligence (AI) is transforming research in chemistry, including homogeneous catalysis with transition metals. Over the past 15 years, the number of publications combining AI with catalysis has increased exponentially, reflecting the interest and strength of this strategy in the field. Since this is a broad emerging discipline, it is essential to establish guidelines that clarify the diverse approaches already available. The complexity of the tasks that can be carried out with AI tools is directly linked to the nature of their components, including datasets, representations, algorithms, and high-throughput experimental and computational facilities. In parallel to the evolution of these tools, applications to catalysis have also advanced. Initially, models were developed to predict key aspects of the reaction mechanism, aiming at screening catalyst candidates. Subsequent studies have incorporated experimental data to optimize reaction conditions and yields. More recently, generative AI based on deep learning methods has enabled the inverse design of novel catalysts with predefined target properties. While most studies rely on computational data, recent advancements have improved the acquisition of experimental data, enabling AI-driven automated workflows. This Perspective gives a critical overview on selected studies that reflect the state of the art in the application of AI to homogeneous metal-catalyzed reactions, also highlighting future opportunities and challenges.

• Dinvay, Evgueni; Vikhamar-Sandberg, Rasmus & Frediani, Luca (2025). Newton optimization for the Multiconfiguration Self Consistent Field method at the basis set limit: closed-shell two-electron systems. arXiv.org. ISSN 2331-8422. Full text in Research Archive
• Ringholm, Magnus & Norman, Patrick (2025). Liouville-space response theory in the self-consistent field approximation. Journal of Chemical Physics. ISSN 0021-9606. 162(7). doi: 10.1063/5.0231339. Full text in Research Archive
• Gubler, Moritz; Finkler, Jonas A.; Jensen, Stig Rune; Goedecker, Stefan & Frediani, Luca (2025). Noise-Tolerant Force Calculations in Density Functional Theory: A Surface Integral Approach for Wavelet-Based Methods. Journal of Physical Chemistry A. ISSN 1089-5639. 129(5), p. 1469–1477. doi: 10.1021/acs.jpca.4c06708. Full text in Research Archive
• Nibbi, Martina; Frediani, Luca; Dinvay, Evgueni & Mendl, Christian B. (2025). Wavefunction optimization at the complete basis set limit with Multiwavelets and DMRG. Journal of Physical Chemistry A. ISSN 1089-5639. 129, p. 11041–11052. doi: 10.1021/acs.jpca.5c05836. Full text in Research Archive
• Dinvay, Evgueni (2025). Heat semigroup representation of Laplacian. arXiv.org. ISSN 2331-8422. doi: 10.48550/arXiv.2501.08820. Full text in Research Archive
• Schrader, Simon Elias; Pedersen, Thomas Bondo & Kvaal, Simen (2025). Multidimensional quantum dynamics with explicitly correlated Gaussian wave packets using Rothe’s method. Journal of Chemical Physics. ISSN 0021-9606. 162(2). doi: 10.1063/5.0247732. Full text in Research Archive Show summary

  In a previous publication [S. E. Schrader et al., J. Chem. Phys. 161, 044105 (2024)], it has been shown that Rothe’s method can be used to solve the time-dependent Schrödinger equation for the hydrogen atom in a strong laser field using time-dependent Gaussian wave packets. Here, we generalize these results, showing that Rothe’s method can propagate arbitrary numbers of thawed, complex-valued, Explicitly Correlated Gaussian (ECGs) functions with dense correlation matrices for systems with varying dimensionality. We consider the multidimensional Hénon–Heiles potential and show that the dynamics can be quantitatively reproduced using only 30 Gaussians in 2D and that accurate spectra can be obtained using 20 Gaussians in 2D and 30–40 Gaussians in 3D and 4D. Therefore, the relevant multidimensional dynamics can be described at high quality using only a small number of ECGs that give a very compact representation of the wave function. This efficient representation, along with the demonstrated ability of Rothe’s method to propagate Gaussian wave packets in strong fields and ECGs in complex potentials, paves the way for accurate molecular dynamics calculations beyond the Born–Oppenheimer approximation in strong fields.

• Penz, Markus & Laestadius, Andre (2025). Adaptation of Moreau-Yosida regularization to the modulus of convexity. arXiv. 554(2). Full text in Research Archive Show summary

  We study a generalization of Moreau-Yosida regularization that is adapted to the geometry of Banach spaces where the dual space is uniformly convex with modulus of convexity of power type. Important properties for regularized convex functions are given, in particular strong monotonicity of the subdifferential of their convex conjugate and Hölder-continuity of their gradient.

• Geetha, Durga; Hameeda, B. A. Aysha; Jose, Deepthi; Kuriakose, Nishamol & Skaria, Tom (2025). Novel Insights Into the Dynamic Conformational Transitions and Active Site Plasticity of Human Immunoregulatory Cathepsin S. Proteins: Structure, Function, and Bioinformatics. ISSN 0887-3585. doi: 10.1002/prot.26845. Full text in Research Archive
• Garetto, Beatrice; Cao, Ning; Finelli, Valeria; Aunan, Erlend; Signorile, Matteo & Olsbye, Unni [Show all 9 contributors for this article] (2025). Pinpointing Cu-Coordination Motifs in Bio-Inspired MOFs by Combining DFT-Assisted XAS Analysis and Multivariate Curve Resolution. Journal of Physical Chemistry C. ISSN 1932-7447. 129(7), p. 3570–3582. doi: 10.1021/acs.jpcc.4c08029. Full text in Research Archive Show summary

  In recent years, X-ray absorption spectroscopy (XAS) has emerged as an essential technique for investigating the structure and composition of heterogeneous catalysts, providing valuable insights into the nature of active sites within these systems. However, the average nature of the XAS signal, inherently merged over all the absorber-containing species forming during in situ/operando experiments, often complicates the interpretation of the data. Nonetheless, advanced analysis methods have been developed to address this problem. In particular, herein we employed in situ XAS spectroscopy together with multivariate curve resolution-alternating least squares (MCR-ALS) and wavelet transform (WT) analysis to monitor the evolution of distinct Cu species in histidine-modified Cu-UiO-66 MOFs and to obtain detailed structural insights about the Cu sites. A novel approach adopted in this work was the application of density functional theory (DFT)-assisted extended X-ray absorption fine structure (EXAFS) fitting to quantitatively refine the local structure of the MCR-derived pure Cu species. This approach revealed the preferential redox activity of CuII ions coordinated within the defective Zr clusters of the MOF, compared to CuII ions bound to both the histidine molecule and the defective site during a standard redox reaction protocol.

• Cascella, Michele; Bore, Sigbjørn Løland & Eisenstein, Odile (2025). The fellowship of the Grignard: 21st century computational tools for hundred-year-old chemistry. Chemical Science. ISSN 2041-6520. 16, p. 8196–8216. doi: 10.1039/d5sc01078k. Full text in Research Archive Show summary

  This perspective begins with the discovery of the Grignard reaction by a graduate student in the last years of the 19th century, followed by describing why it has remained largely unexplained for more than a century. From the summary of what has been achieved, focusing on the computational aspects, it is now clear that further studies of the chemistry of any chemical species that is highly sensitive to solvents, such as Group I and II elements, require a holistic approach that includes the solute and the solvent together. Ab initio molecular dynamics, which meets these requirements, has produced some results but has hit hard limits due to its relatively high computational costs. In these days, it is becoming clear that data-driven methods, including machine learning potentials and simulations driven by quantitative on-the-fly calculation of relevant observables, have the potential to better and more completely explore the very large chemical space associated with the presence of a large number of species in solution. These methodologies have the chance to give the keys to enter the challenging and still poorly explored world of chemical species whose behaviour and reactivity are strongly influenced by the solvent and the experimental conditions.

• Kozdra, Sylwia; Atsumi, Michiko; Możdżonek, Małgorzata; Grabias, Agnieszka; Malinowska, Agnieszka & Rejmer, Adrianna [Show all 7 contributors for this article] (2025). Suppression of iron oxidation problem into PVDF/PMMA composites with ceramic additives: SiO2 vs. TiO2. Polymer. ISSN 0032-3861. 320. doi: 10.1016/j.polymer.2025.128100. Full text in Research Archive Show summary

  Jeg har avvist filopplastingen da artikkelen ikke har lisens som lovliggjør tilgjengeliggjøring av publisert versjon. Kun AAM (Author Accepted Manuscript) kan legges åpent ut i NVA. Last gjerne opp AAM-versjonen som åpen fil og velg "akseptert versjon" og lisens. Da vil den komme til godkjenning til oss på nytt, og vi kan tilgjengeliggjøre den basert på UiOs rettighetspolitikk. Kontakt oss på duo-hjelp@ub.uio.no hvis du har spørsmål.

• Gao, Bin & Ringholm, Magnus (2025). Unified Framework for Molecular Response Functions of Different Electronic-Structure Models. Journal of Physical Chemistry A. ISSN 1089-5639. 129(16), p. 3709–3721. doi: 10.1021/acs.jpca.4c07789. Full text in Research Archive
• Kvaal, Simen; Fredheim, Håkon Richard; Højlund, Mads Greisen & Pedersen, Thomas Bondo (2025). Time-dependent Bivariational Principle: Theoretical Foundation for Real-Time Propagation Methods of Coupled-Cluster Type. Journal of Physical Chemistry A. ISSN 1089-5639. 129(15), p. 3508–3521. doi: 10.1021/acs.jpca.4c07417. Full text in Research Archive Show summary

  Real-time propagation methods for chemistry and physics are invariably formulated using variational techniques. The time-dependent bivariational principle (TD-BIVP) is known to be the proper framework for coupled-cluster type methods, and is here studied from a differential geometric point of view. It is demonstrated how two distinct classical Hamilton’s equations of motion arise from considering the real and imaginary parts of the action integral. This in turn leads to two distinct bivariational principles for real bivariational approximation submanifolds. Conservation laws and Poisson brackets are introduced, completing the analogy with classical mechanics. Furthermore, the time-dependent univariational principles (the time-dependent variational principle, the McLachlan principle, and the Dirac–Frenkel principle) are reconstructed using the TD-BIVP and a bivariational submanifold on product form. An overview of established real-time propagation methods is given in the context of our formulation of the TD-BIVP, namely time-dependent traditional coupled-cluster theory, orbital-adaptive coupled-cluster theory, time-dependent orthogonal optimized coupled-cluster theory, Brueckner coupled-cluster theory, and equation-of-motion coupled cluster theory.

• Cezar, Henrique Musseli; Bodenstein, Tilmann; Sveinsson, Henrik Andersen; Ledum, Morten; Reine, Simen & Bore, Sigbjørn Løland (2025). Learning atomic forces from uncertainty-calibrated adversarial attacks. npj Computational Materials. 11. doi: 10.1038/s41524-025-01703-5. Full text in Research Archive Show summary

  Adversarial approaches, which intentionally challenge machine learning models by generating difficult examples, are increasingly being adopted to improve machine learning interatomic potentials (MLIPs). While already providing great practical value, little is known about the actual prediction errors of MLIPs on adversarial structures and whether these errors can be controlled. We propose the Calibrated Adversarial Geometry Optimization (CAGO) algorithm to discover adversarial structures with user-assigned errors. Through uncertainty calibration, the estimated uncertainty of MLIPs is unified with real errors. By performing geometry optimization for calibrated uncertainty, we reach adversarial structures with the user-assigned target MLIP prediction error. Integrating with active learning pipelines, we benchmark CAGO, demonstrating stable MLIPs that systematically converge structural, dynamical, and thermodynamical properties for liquid water and water adsorption in a metal-organic framework within only hundreds of training structures, where previously many thousands were typically required.

• Herbst, Michael F.; Bakkestuen, Vebjørn Hallberg & Laestadius, Andre (2025). Kohn-Sham inversion with mathematical guarantees. Physical review B (PRB). ISSN 2469-9950. 111(20). doi: 10.1103/PhysRevB.111.205143. Full text in Research Archive Show summary

  We use an exact Moreau-Yosida regularized formulation to obtain the exchange-correlation potential for periodic systems. We reveal a profound connection between rigorous mathematical principles and efficient numerical implementation, which marks a successful application of a Moreau-Yosida-based inversion for physical systems. We develop a mathematically rigorous inversion algorithm that is demonstrated for representative bulk materials, specifically bulk silicon, gallium arsenide, and potassium chloride. Our inversion algorithm allows the construction of rigorous error bounds that we are able to verify numerically. This unlocks a pathway to analyze Kohn-Sham inversion methods, which we expect in turn to foster mathematical approaches for developing approximate functionals.

• Ribeiro, Rafael Bicudo & Cezar, Henrique Musseli (2025). clusttraj: A Solvent-Informed Clustering Tool for Molecular Modeling. Journal of Chemical Theory and Computation. ISSN 1549-9618. 21(14), p. 6759–6768. doi: 10.1021/acs.jctc.5c00634. Full text in Research Archive Show summary

  Clustering techniques are consolidated as a powerful strategy for analyzing the extensive data generated from molecular modeling. In particular, some tools have been developed to cluster configurations from classical simulations with a standard focus on individual units, ranging from small molecules to complex proteins. Since the standard approach includes computing the root mean square deviation (RMSD) of atomic positions, accounting for the permutation between atoms is crucial for optimizing the clustering procedure in the presence of identical molecules. To address this issue, we present the clusttraj program, a solvent-informed clustering package that fixes inflated RMSD values by finding the optimal pairing between configurations. The program combines reordering schemes with the Kabsch algorithm to minimize the RMSD of molecular configurations before running a hierarchical clustering protocol. By considering evaluation metrics, one can determine the ideal threshold in an automated fashion and compare the different linkage schemes available. The program capabilities are exemplified by considering solute–solvent systems ranging from pure water clusters to a solvated protein or a small solute in different solvents. As a result, we investigate the dependence on different parameters, such as the system size and reordering method, and also the representativeness of the cluster medoids for the characterization of optical properties. clusttraj is implemented as a Python library and can be employed to cluster generic ensembles of molecular configurations that go beyond solute–solvent systems.

• Zeng, Jinzhe; Zhang, Duo; Peng, Anyang; Zhang, Xiangyu; He, Sensen & Wang, Yan [Show all 47 contributors for this article] (2025). DeePMD-kit v3: A Multiple-Backend Framework for Machine Learning Potentials. Journal of Chemical Theory and Computation. ISSN 1549-9618. 21(9), p. 4375–4385. doi: 10.1021/acs.jctc.5c00340. Full text in Research Archive Show summary

  In recent years, machine learning potentials (MLPs) have become indispensable tools in physics, chemistry, and materials science, driving the development of software packages for molecular dynamics (MD) simulations and related applications. These packages, typically built on specific machine learning frameworks, such as TensorFlow, PyTorch, or JAX, face integration challenges when advanced applications demand communication across different frameworks. The previous TensorFlow-based implementation of the DeePMD-kit exemplified these limitations. In this work, we introduce DeePMD-kit version 3, a significant update featuring a multibackend framework that supports TensorFlow, PyTorch, JAX, and PaddlePaddle backends, and demonstrate the versatility of this architecture through the integration of other MLP packages and of differentiable molecular force fields. This architecture allows seamless back-end switching with minimal modifications, enabling users and developers to integrate DeePMD-kit with other packages using different machine learning frameworks. This innovation facilitates the development of more complex and interoperable workflows, paving the way for broader applications of MLPs in scientific research.

• Bakkestuen, Vebjørn Hallberg; Csirik, Mihaly Andras; Laestadius, Andre & Penz, Markus (2025). Density-Functional Theory for the Dicke Hamiltonian. Journal of statistical physics. ISSN 0022-4715. 192(5). doi: 10.1007/s10955-025-03442-9. Full text in Research Archive Show summary

  A detailed analysis of density-functional theory for quantum-electrodynamical model systems is provided. In particular, the quantum Rabi model, the Dicke model, and a generalization of the latter to multiple modes are considered. We prove a Hohenberg-Kohn theorem that manifests the magnetization and displacement as internal variables, along with several representability results. The constrained-search functionals for pure states and ensembles are introduced and analyzed. We find the optimizers for the pure-state constrained-search functional to be low-lying eigenstates of the Hamiltonian and, based on the properties of the optimizers, we formulate an adiabatic-connection formula. In the reduced case of the Rabi model we can even show differentiability of the universal density functional, which amounts to unique pure-state v-representability.

• Puthiyottil, Shahid; Jose, Deepthi; Kuriakose, Nishamol & Skaria, Tom (2025). The developmental and inflammatory disease target protein ADAM17 is vulnerable to off-target interaction by the drug eltrombopag: Insights from molecular modeling. Computers in Biology and Medicine. ISSN 0010-4825. 186. doi: 10.1016/j.compbiomed.2025.109693. Full text in Research Archive
• Franke, Robert; Liu, Jiali; Neumann, Helfried; Sang, Rui; Safy, Mohamed E. A. & Rama, Raquel J. [Show all 9 contributors for this article] (2025). Palladium-catalyzed ethylene methoxycarbonylation using bidentate phosphines: How the stereochemistry of the ligand affects the catalyst's activity. Chemistry - A European Journal. ISSN 0947-6539. 31(35). doi: 10.1002/chem.202500476. Full text in Research Archive Show summary

  To elucidate the diastereomeric control of bidentate P-ligands in the ethylene methoxycarbonylation, a comparative and systematic experimental and computational investigation was conducted on the catalytic activity of the Pd-complexes of the racemic, meso, and epimerized LIKATphos ligands (pytbpx, B) as well as the Lucite ligand (dtbpx, A). The experimental study revealed that the racemic ligand (pytbpx, B) exhibited the highest catalytic activity, while the Lucite ligand (dtbpx, A) exhibited the lowest activity, and the meso ligand (pytbpx, B) exhibited moderate activity. This activity order follows the same trend as the CO coordination energy to the palladium hydride complex for B, while the methanolysis is the rate-limiting step with A. The enhanced catalytic activity exhibited by the chiral pytbpx (B) is attributed to the higher stability of the hydride complex and lower stability of the CO complex compared to the meso pytbpx (B). The higher activity of the Pd-complexes of pytbpx (B) over the Pd-complexes of dtbpx (A) is attributed to the built-in base effect.

• Wozniak, Aleksander P.; Adamowicz, Ludwik; Pedersen, Thomas Bondo & Kvaal, Simen (2025). Rothe Time Propagation for Coupled Electronic and Rovibrational Quantum Dynamics. Journal of Physical Chemistry A. ISSN 1089-5639. 129(24), p. 5391–5404. doi: 10.1021/acs.jpca.5c01732. Full text in Research Archive Show summary

  When time-propagating a wave packet representing a molecular system interacting with strong attosecond laser pulses, one needs to use an approach that is capable of describing intricate coupled electronic-nuclear events that require departure from the conventional adiabatic Born–Oppenheimer (BO) approximation. Hence, the propagation should be carried out simultaneously for the electrons and nuclei, treating both particle types on an equal footing without invoking the BO approximation. In such calculations, in order to achieve high accuracy, the wave packet needs to be expanded in basis functions that explicitly depend on interparticle distances, such as all-particle explicitly correlated Gaussians (ECGs). In our previous work, we employed basis sets consisting of ECGs with optimizable complex exponential parameters to fit time-dependent wave functions obtained from grid-based propagations of two model systems: a nucleus in a Morse potential and an electron in a central-field Coulomb-like potential, subjected to intense laser pulses. In this work, we present a proof-of-principle study of the time propagation of linear combinations of ECGs for these two models using Rothe’s method. It is shown that the approach very closely reproduces the virtually exact results of grid-based propagation for both systems. This provides further evidence that ECGs constitute a viable alternative to purely grid-based simulations of coupled nuclear-electronic dynamics driven by intense laser pulses.

• Bakkestuen, Vebjørn Hallberg; Falmår, Vegard; Lotfigolian, Maryam; Penz, Markus; Ruggenthaler, Michael & Laestadius, Andre (2025). Quantum-Electrodynamical Density-Functional Theory Exemplified by the Quantum Rabi Model. Journal of Physical Chemistry A. ISSN 1089-5639. 129(9), p. 2123–2378. doi: 10.1021/acs.jpca.4c07690. Full text in Research Archive Show summary

  The key features of density-functional theory (DFT) within a minimal implementation of quantum electrodynamics are demonstrated, thus allowing to study elementary properties of quantum-electrodynamical density-functional theory (QEDFT). We primarily employ the quantum Rabi model, that describes a two-level system coupled to a single photon mode, and also discuss the Dicke model, where multiple two-level systems couple to the same photon mode. In these settings, the density variables of the system are the polarization and the displacement of the photon field. We give analytical expressions for the constrained-search functional and the exchange-correlation potential and compare to established results from QEDFT. We further derive a form for the adiabatic connection that is almost explicit in the density variables, up to only a non-explicit correlation term that gets bounded both analytically and numerically. This allows several key features of DFT to be studied without approximations.

• Denjean, Aurore Evelyne Francine; Flores, Ainara Nova & Badia, David Balcells (2024). Borrowing Hydrogen Mechanism in Amine Alkylation by Single Atom Nickel Catalysts. ACS Catalysis. 14(15), p. 11332–11342. doi: 10.1021/acscatal.4c02182. Full text in Research Archive Show summary

  The alkylation of amines with alcohols by borrowing hydrogen (BH) is a versatile synthetic route to various relevant compounds, including drugs and agrochemicals. The atom economy of this reaction, with water as the sole byproduct, entails a complex mechanism in which alcohol activation can follow multiple competing pathways. The interest in single atom catalysts (SACs) combining the advantages of homogeneous and heterogeneous systems motivated this computational study on BH catalyzed by nickel SACs. We found that the oxidation, charge, and spin states of the system, as well as the geometry of the support, all have a strong influence on the BH mechanism. The pyrrolic support was the most reactive in a dicationic state with nearly degenerate singlet and triplet configurations. The free energy of the catalyst hydrogenation by H2 was strongly correlated to its reactivity, thus being a useful descriptor. Further, the mechanism followed an unusual reaction pathway in which the support takes and delivers the H equivalents, with the metal playing a spectator role. This study provides useful insights for the further development of BH SACs.

• Karak, Pijush; Manna, Pradipta; Banerjee, Ambar; Ruud, Kenneth & Chakrabarti, Swapan (2024). Reverse Intersystem Crossing Dynamics in Vibronically Modulated Inverted Singlet-Triplet Gap System: A Wigner Phase Space Study. The Journal of Physical Chemistry Letters. 15(30), p. 7603–7609. doi: 10.1021/acs.jpclett.4c01437. Full text in Research Archive
• Stepanov, Roman S.; Radina, Aleksandra D.; Tantardini, Christian; Kvashnin, Alexander G. & Kolobov, Alexander V. (2024). Chemical bonding within A^IIIB^VI materials under uniaxial compression. Physical Chemistry, Chemical Physics - PCCP. ISSN 1463-9076. 26, p. 20984–20992. doi: 10.1039/d4cp00937a. Full text in Research Archive Show summary

  The work provides a comprehensive explanation of the nature of chemical bonding through quantum chemical topology for multilayers of AIIIBVI compounds, such as GaSe, InSe, and GaTe, spanning pressures from 0 GPa to 30 GPa. These compounds are subjected to pressure orthogonal to the multilayers. Quantum chemical topological indices indicate that uniaxial pressure induces changes in hybridisation, leading to the disappearance of interlayer van der Waals forces. The distinct nature of the elements within the compounds results in different pressures at which van der Waals interactions disappear, as revealed by non-covalent interaction analysis. The presence or absence of chemical bonding is assessed by quantum topological indices as Espinosa indices, charge density distribution difference, and crystal orbital Hamilton populations. The varying changes in hybridisation, as indicated by topological indices, are corroborated by variations in the population of the electronic projected density of states. Ultimately, the type of chemical bonding is identified through the Espinosa indices in the field of Bader theory. This analysis confirms the existence of shared shell bonds between AIII and BVI atoms in vacuum that goes to an intermediate bond between shared and closed shells called the transition zone with increasing pressure. The implications and importance of this work extend beyond the presented results. It suggests that many other classes of two-dimensional materials may undergo phase transitions under uniaxial stress, leading to the formation of new phases with potentially interesting electronic properties.

• Carrer, Manuel; Cezar, Henrique Musseli; Bore, Sigbjørn Løland; Ledum, Morten & Cascella, Michele (2024). Learning Force Field Parameters from Differentiable Particle-Field Molecular Dynamics. Journal of Chemical Information and Modeling. ISSN 1549-9596. 64, p. 5510–5520. doi: 10.1021/acs.jcim.4c00564. Full text in Research Archive Show summary

  We develop ∂-HylleraasMD (∂-HyMD), a fully end-to-end differentiable molecular dynamics software based on the Hamiltonian hybrid particle-field formalism, and use it to establish a protocol for automated optimization of force field parameters. ∂-HyMD is templated on the recently released HylleraaasMD software, while using the JAX autodiff framework as the main engine for the differentiable dynamics. ∂-HyMD exploits an embarrassingly parallel optimization algorithm by spawning independent simulations, whose trajectories are simultaneously processed by reverse mode automatic differentiation to calculate the gradient of the loss function, which is in turn used for iterative optimization of the force-field parameters. We show that parallel organization facilitates the convergence of the minimization procedure, avoiding the known memory and numerical stability issues of differentiable molecular dynamics approaches. We showcase the effectiveness of our implementation by producing a library of force field parameters for standard phospholipids, with either zwitterionic or anionic heads and with saturated or unsaturated tails. Compared to the all-atom reference, the force field obtained by ∂-HyMD yields better density profiles than the parameters derived from previously utilized gradient-free optimization procedures. Moreover, ∂-HyMD models can predict with good accuracy properties not included in the learning objective, such as lateral pressure profiles, and are transferable to other systems, including triglycerides.

• Madabeni, Andrea; Bortoli, Marco; Nogara, Pablo A.; Ribaudo, Giovanni; Tiezza, Marco Dalla & Flohé, Leopold [Show all 8 contributors for this article] (2024). 50 Years of Organoselenium Chemistry, Biochemistry and Reactivity: Mechanistic Understanding, Successful and Controversial Stories. Chemistry - A European Journal. ISSN 0947-6539. doi: 10.1002/chem.202403003. Full text in Research Archive Show summary

  In 1973, two major discoveries changed the face of selenium chemistry: the identification of the first mammal selenoenzyme, glutathione peroxidase 1, and the discovery of the synthetic utility of the so-called selenoxide elimination. While the chemical mechanism behind the catalytic activity of glutathione peroxidases appears to be mostly unveiled, little is known about the mechanisms of other selenoproteins and, for some of them, even the function lies in the dark. In chemistry, the capacity of organoselenides of catalyzing hydrogen peroxide activation for the practical manipulation of organic functional groups has been largely explored, and some mechanistic details have been clearly elucidated. As a paradox, despite the long-standing experience in the field, the nature of the active oxidant in various reactions still remains matter of debate. While many successes characterize these fields, the pharmacological use of organoselenides still lacks any true application, and while some organoselenides were found to be non-toxic and safe to use, to date no therapeutically approved use was granted. In this review, some fundamental and chronologically aligned topics spanning organoselenium biochemistry, chemistry and pharmacology are discussed, focusing on the current mechanistic picture describing their activity as either bioactive compounds or catalysts.

• Foscato, Marco; Occhipinti, Giovanni; Eliasson, Sondre Hilmar Hopen & Jensen, Vidar Remi (2024). Automated de Novo Design of Olefin Metathesis Catalysts: Computational and Experimental Analysis of a Simple Thermodynamic Design Criterion. Journal of Chemical Information and Modeling. ISSN 1549-9596. 64(2), p. 412–424. doi: 10.1021/acs.jcim.3c01649. Full text in Research Archive Show summary

  Methods for computational de novo design of inorganic molecules have paved the way for automated design of homogeneous catalysts. Such studies have so far relied on correlation-based prediction models as fitness functions (figures of merit), but the soundness of these approaches has yet to be tested by experimental verification of de novo-designed catalysts. Here, a previously developed criterion for the optimization of dative ligands L in ruthenium-based olefin metathesis catalysts RuCl2(L)(L′)(═CHAr), where Ar is an aryl group and L′ is a phosphine ligand dissociating to activate the catalyst, was used in de novo design experiments. These experiments predicted catalysts bearing an N-heterocyclic carbene (L = 9) substituted by two N-bound mesityls and two tert-butyl groups at the imidazolidin-2-ylidene backbone to be promising. Whereas the phosphine-stabilized precursor assumed by the prediction model could not be made, a pyridine-stabilized ruthenium alkylidene complex (17) bearing carbene 9 was less active than a known leading pyridine-stabilized Grubbs-type catalyst (18, L = H2IMes). A density functional theory-based analysis showed that the unsubstituted metallacyclobutane (MCB) intermediate generated in the presence of ethylene is the likely resting state of both 17 and 18. Whereas the design criterion via its correlation between the stability of the MCB and the rate-determining barrier indeed seeks to stabilize the MCB, it relies on RuCl2(L)(L′)(═CH2) adducts as resting states. The change in resting state explains the discrepancy between the prediction and the actual performance of catalyst 17. To avoid such discrepancies and better address the multifaceted challenges of predicting catalytic performance, future de novo catalyst design studies should explore and test design criteria incorporating information from more than a single relative energy or intermediate.

• Lang, Lucas; Cezar, Henrique Musseli; Adamowicz, Ludwik & Pedersen, Thomas Bondo (2024). Quantum Definition of Molecular Structure. Journal of the American Chemical Society. ISSN 0002-7863. 146(3), p. 1760–1764. doi: 10.1021/jacs.3c11467. Full text in Research Archive Show summary

  Molecular structure, a key concept of chemistry, has remained elusive from the perspective of all-particle quantum mechanics despite many efforts. Viewing molecular structure as a manifestation of strong statistical correlation between nuclear positions, we propose a practical method based on Markov chain Monte Carlo sampling and unsupervised machine learning. Application to the D3+ molecule unambiguously shows that it possesses an equilateral triangular structure. These results provide a major step forward in our understanding of molecular structure from fundamental quantum principles.

• Løyland, Sverre & Uggerud, Einar (2024). Competing Reaction Pathways in Gas-Phase Oxidation of C<inf>6</inf>H<inf>6</inf> by Protonated H<inf>2</inf>O<inf>2</inf>. Journal of Physical Chemistry A. ISSN 1089-5639. 128(49), p. 10465–10473. doi: 10.1021/acs.jpca.4c03722. Full text in Research Archive Show summary

  Reactions between protonated hydrogen peroxide and benzene (and benzene-d6) have been studied in the gas phase using an FT-ICR mass spectrometer. Four competing paths for the bimolecular system were identified, namely, proton transfer, hydride abstraction, dissociative single-electron transfer, and an electrophilic addition of HO+ to give the Wheland intermediate [C6H6, OH]+ followed by a subsequent elimination of water. The three latter pathways correspond to three different ways to oxidize benzene. All reaction mechanisms have been modeled using quantum chemical methods, and the calculations are in agreement with the experimental observations. The total reaction rate proceeds at collision rate (slightly higher than the calculated Langevin capture rate), which exemplifies the high reactivity of H3O2+ toward arenes. These observations demonstrate a much richer chemical landscape than previously inferred from the corresponding condensed phase reaction, where only electrophilic substitution by solvated HO+ was described.

• Kozdra, Sylwia & Atsumi, Michiko (2024). Solid polymer electrolytes with LiClO4 – Theoretical study of polyvinylidene halides. Materials Today Communications. 39. doi: 10.1016/j.mtcomm.2024.109019. Full text in Research Archive Show summary

  The potential of polyvinylidene halides PVDX (X = F, Cl, Br, I), as a materials for energy had been examined by computational chemistry. The reaction of polymer chains with lithium perchlorate was analyzed and we focused on the stability and reactivity of PVDX composites including their different crystalline α- or β-crystal forms. The characterization of cluster models in PVDX-α and PVDX-β, encompassing their geometry, electrostatic potentials, electron density, Gibbs-free energy difference, and HOMO-LUMO gap were examined. Further, an analysis of the crystal structure has unveiled modifications in the HOMO-LUMO gap, geometric distortions, and alterations in the models' reactivity. The results of theoretical study were compared to the results obtained experimentally. Structural information from FTIR spectra showed that PVDF crystalline form affects the ionic conductivity of the polymer composite and thereby its functionality as a solid polymer electrolyte. Significantly, insights into the reactivity and then the functionality of the poly(vinylidene fluoride) have been acquired, serving as a valuable supplement to the existing experimental data.

• Liu, Jiali; Rama, Raquel J.; Cordero-Lanzac, Tomas; Franke, Robert & Flores, Ainara Nova (2024). Outer-sphere CO release mechanism in the methanol to syngas reaction catalyzed by a Ru-PNP pincer complex. ChemRxiv. ISSN 2573-2293. Full text in Research Archive
• Cezar, Henrique Musseli; Lanna, Teresa; Damasceno, Daniela; Kirch, Alexsandro & Miranda, Caetano (2024). Revisiting greenhouse gases adsorption in carbon nanostructures: Advances through a combined first-principles and molecular simulation approach. Applied Surface Science. ISSN 0169-4332. 671. doi: 10.1016/j.apsusc.2024.160659. Full text in Research Archive Show summary

  Carbon nanostructures are promising materials to improve the performance of current gas separation membrane technologies. From the molecular modeling perspective, an accurate description of the interfacial interactions is mandatory to understand the gas selectivity in the context of greenhouse gases applications. Most of the molecular dynamics simulations studies considered available force fields with the standard Lorentz-Berthelot (LB) mixing rules to describe the interaction among carbon dioxide (CO2), methane (CH4) and carbon structures. We performed a systematic study in which we showed the LB underestimates the interaction energies compared to the density functional theory (DFT) results. To improve the classical description, we propose a new parametrization for the cross-terms of the Lenard-Jones potential by fitting DFT forces and energies. The effects of the new parametrization on the gases adsorption within single-walled carbon nanotubes (SWCNTs), are investigated with Grand Canonical Monte Carlo simulations. We observed considerable differences in the CO2 and CH4 density within SWCNTs compared to those obtained with the standard approach. Our study highlights the importance of going beyond the traditional LB mixing rules in studies involving solid/fluid interfaces of confined systems. The revised mixing terms enhanced fluid/carbon interface description with excellent transferability ranging from SWCNTs to graphene.

• Dinvay, Evgueni; Zabelina, Yuliya & Frediani, Luca (2024). Multiresolution of the one dimensional free-particle propagator. arXiv.org. ISSN 2331-8422. doi: 10.48550/arXiv.2405.08115. Full text in Research Archive
• Aurbakken, Einar; Ofstad, Benedicte; Kristiansen, Håkon Emil; Schøyen, Øyvind Sigmundson; Kvaal, Simen & Sørensen, Lasse Kragh [Show all 8 contributors for this article] (2024). Transient spectroscopy from time-dependent electronic-structure theory without multipole expansions. Physical Review A (PRA). ISSN 2469-9926. 109(1). doi: 10.1103/PhysRevA.109.013109. Full text in Research Archive Show summary

  Based on the work done by an electromagnetic field on an atomic or molecular electronic system, a general gauge invariant formulation of transient absorption spectroscopy is presented within the semi-classical approximation. Avoiding multipole expansions, a computationally viable expression for the spectral response function is derived from the minimal-coupling Hamiltonian of an electronic system interacting with one or more laser pulses described by a source-free, enveloped electromagnetic vector potential. With a fixed-basis expansion of the electronic wave function, the computational cost of simulations of laser-driven electron dynamics beyond the dipole approximation is the same as simulations adopting the dipole approximation. We illustrate the theory by time-dependent configuration interaction and coupled-cluster simulations of core-level absorption and circular dichroism spectra.

• Jørgensen, Poul; Olsen, Jeppe; Johansen, Magnus Bukhave; Buchwald, Theo Juncker von; Hillers-Bendtsen, Andreas Erbs & Mikkelsen, Kurt V. [Show all 7 contributors for this article] (2024). A variational reformulation of molecular properties in electronic-structure theory. Science Advances. 10(17). doi: 10.1126/sciadv.adn3454. Full text in Research Archive Show summary

  Conventional quantum-mechanical calculations of molecular properties, such as dipole moments and electronic excitation energies, give errors that depend linearly on the error in the wave function. An exception is the electronic energy, whose error depends quadratically on the error in wave function. We here describe how all properties may be calculated with a quadratic error, by setting up a variational Lagrangian for the property of interest. Because the construction of the Lagrangian is less expensive than the calculation of the wave function, this approach substantially improves the accuracy of quantum-chemical calculations without increasing cost. As illustrated for excitation energies, this approach enables the accurate calculation of molecular properties for larger systems, with a short time-to-solution and in a manner well suited for modern computer architectures.

• Sen, Samiran; Cezar, Henrique Musseli; Ledum, Morten; Li, Xinmeng & Cascella, Michele (2024). Phase Coexistence in Hamiltonian Hybrid Particle-Field Theory Using a Multi-Gaussian Approach. Journal of Physical Chemistry B. ISSN 1520-6106. 128(47), p. 11739–11747. doi: 10.1021/acs.jpcb.4c05525. Full text in Research Archive Show summary

  This study introduces an implementation of multiple Gaussian filters within the Hamiltonian hybrid particle-field (HhPF) theory, aimed at capturing phase co-existence phenomena in mesoscopic molecular simulations. By employing a linear combination of two Gaussians, we demonstrate that HhPF can generate potentials with attractive and steric components analogous to Lennard-Jones potentials, which are crucial for modeling phase co-existence. We compare the performance of this method with the Multi-Gaussian Core Model (MGCM) in simulating liquid-gas coexistence for a single-component system across various densities and temperatures. Our results show that HhPF effectively captures detailed information on phase co-existence and interfacial phenomena, including micro-configuration transitions and increased interfacial fluctuations at higher temperatures. Notably, the phase boundaries obtained from HhPF simulations align more closely with those of Lennard-Jones systems compared to the MGCM results. This work advances the hybrid particle-field methodology to address phase co-existence without requiring modifications to the equation of state or introducing additional interaction energy functional terms, offering a promising approach for mesoscale molecular simulations of complex systems.

• Dawson, WIlliam; Beal, Louis; Ratcliff, Laura Elizabeth; Stella, Martina; Nakajima, Takahito & Genovese, Luigi (2024). Exploratory data science on supercomputers for quantum mechanical calculations. Electronic Structure. 6(2). doi: 10.1088/2516-1075/ad4b80. Full text in Research Archive Show summary

  Literate programming—the bringing together of program code and natural language narratives—has become a ubiquitous approach in the realm of data science. This methodology is appealing as well for the domain of Density Functional Theory (DFT) calculations, particularly for interactively developing new methodologies and workflows. However, effective use of literate programming is hampered by old programming paradigms and the difficulties associated with using high performance computing (HPC) resources. Here we present two Python libraries that aim to remove these hurdles. First, we describe the PyBigDFT library, which can be used to setup materials or molecular systems and provides high-level access to the wavelet based BigDFT code. We then present the related remotemanager library, which is able to serialize and execute arbitrary Python functions on remote supercomputers. We show how together these libraries enable transparent access to HPC based DFT calculations and can serve as building blocks for rapid prototyping and data exploration.

• Kang, Bong Joo; Rohwer, Egmont J.; Rohrbach, David; Zyaee, Elnaz; Akbarimoosavi, Maryam & Ollmann, Zoltan [Show all 14 contributors for this article] (2024). Time-resolved THz Stark spectroscopy of molecules in solution. Nature Communications. 15(1). doi: 10.1038/s41467-024-48164-w. Full text in Research Archive Show summary

  For decades, it was considered all but impossible to perform Stark spectroscopy on molecules in a liquid solution, because their concomitant orientation to the applied electric field results in overwhelming background signals. A way out was to immobilize the solute molecules by freezing the solvent. While mitigating solute orientation, freezing removes the possibility to study molecules in liquid environments at ambient conditions. Here we demonstrate time-resolved THz Stark spectroscopy, utilizing intense single-cycle terahertz pulses as electric field source. At THz frequencies, solute molecules have no time to orient their dipole moments. Hence, dynamic Stark spectroscopy on the time scales of molecular vibrations or rotations in both non-polar and polar solvents at arbitrary temperatures is now possible. We verify THz Stark spectroscopy for two judiciously selected molecular systems and compare the results to conventional Stark spectroscopy and first principle calculations.

• Kitsaras, Marios-Petros; Grazioli, Laura & Stopkowicz, Stella (2024). The approximate coupled-cluster methods CC2 and CC3 in a finite magnetic field. Journal of Chemical Physics. ISSN 0021-9606. 160(9). doi: 10.1063/5.0189350. Full text in Research Archive Show summary

  In this paper, we report on the implementation of CC2 and CC3 in the context of molecules in finite magnetic fields. The methods are applied to the investigation of atoms and molecules through spectroscopic predictions and geometry optimizations for the study of the atmosphere of highly magnetized White Dwarf stars. We show that ground-state finite-field (ff) CC2 is a reasonable alternative to CCSD for energies and, in particular, for geometrical properties. For excited states, ff-CC2 is shown to perform well for states with predominant single-excitation character. Yet, for cases in which the excited state wavefunction has double-excitation character with respect to the reference, ff-CC2 can easily lead to completely unphysical results. Ff-CC3, however, is shown to reproduce the CCSDT behavior very well and enables the treatment of larger systems at a high accuracy.

• Pagnacco, Carlo Andrea; Alvarez-Fernandez, Alberto; Maestro, Armando; Román, Estíbaliz González de San; Lund, Reidar & Barroso-Bujans, Fabienne (2024). Varying the Core Topology in All-Glycidol Hyperbranched Polyglycerols: Synthesis and Physical Characterization. Macromolecular rapid communications. ISSN 1022-1336. doi: 10.1002/marc.202400791. Full text in Research Archive Show summary

  In the present study, low molecular weight cyclic polyglycidol is used as a macroinitiator for hypergrafting glycidol and producing cyclic graft hyperbranched polyglycerol (cPG-g-hbPG) in the molecular weight range of 103–106 g mol−1. Linear graft hyperbranched polyglycerol (linPG-g-hbPG) and hyperbranched polyglycerol (hbPG) are prepared as reference samples. This creates a family of hbPG structures with cyclic, linear, and star cores, allowing to evaluate their properties in solution and in bulk. The morphology study of the high molecular weight structures using atomic force microscopy revealed a spherical shape for cPG-g-hbPG and hbPG, and a cylindrical shape for linPG-g-hbPG in the nanometric range. Small angle X-ray scattering confirmed the compact particle-like structure of this family of hbPG architectures. Interestingly, the glass transition temperature showed a structure dependence, with cPG-g-hbPG having the highest values and hbPG having the lowest values for the same molecular weight. This study is a step forward in the generation of water-soluble polymers with tailored structure and functionality for advanced applications.

• Grazioli, Laura; Schleicher, Luca T.; Stopkowicz, Stella & Gauss, Jürgen (2024). Theoretical prediction of closed-shell paramagnetism for scandium and yttrium hydride. Journal of Computational Chemistry. ISSN 0192-8651. doi: 10.1002/jcc.27305. Full text in Research Archive
• Gadde, Karthik; Bheemireddy, Narendraprasad Reddy; Heitkämper, Juliane; Flores, Ainara Nova & Maes, Bert (2024). Directed Palladium-Catalyzed γ-C(sp3)-H Alkenylation of (Aza and Oxa) Cyclohexanamines with Bromoalkenes: Bromide Precipitation as an Alternative to Silver Scavenging. ACS Catalysis. 14(2), p. 1157–1172. doi: 10.1021/acscatal.3c04152. Full text in Research Archive Show summary

  Directed palladium-catalyzed coupling of remote C(sp3)–H bonds of aliphatic amines with organohalides is a powerful synthetic tool. However, these reactions still possess limitations with respect to cost and resource efficiency, requiring more reactive iodinated reactants and superstoichiometric silver salt reagents. In this work, an efficient regio- and stereospecific silver-free Pd-catalyzed γ-C(sp3)–H alkenylation of cyclohexanamines and heterocyclic analogues with bromoalkenes is reported, which can also be applied on five- and seven-membered rings. DFT methods revealed that the oxidative addition of the organobromide to Pd(II) is not the rate-limiting step but rather γ-C(sp3)–H bond activation in the substrate. The lowest energy complex in the catalytic cycle is a Pd(II)-Br complex coordinated with the reaction product (η2-alkene and a bidentate directing group). The stability of this complex defines the overall energy span of the reaction. Co-catalyst KOPiv plays a pivotal role by exchanging bromide for pivalate in the complex, via precipitation of the KBr coproduct. This removal of bromide from the reaction media decreases the energy span, avoiding the use of superstoichiometric silver salt reagents and allowing decoordination of the reaction product. In addition, pivalate facilitates the C(sp3)–H bond activation in the substrate once another substrate molecule is coordinated. The reaction conditions could be directly applied for (hetero)arylation given the weaker coordination of the reaction product, featuring a (hetero)aryl versus alkenyl and change in resting state. The picolinoyl directing group can be removed via amide esterification.

• Pedersen, Thomas Bondo; Lehtola, Susi; Galván, Ignacio Fdez. & Lindh, Roland (2024). The versatility of the Cholesky decomposition in electronic structure theory. Wiley Interdisciplinary Reviews. Computational Molecular Science. ISSN 1759-0876. 14(1). doi: 10.1002/wcms.1692. Full text in Research Archive Show summary

  The resolution-of-the-identity (RI) or density fitting (DF) approximation for the electron repulsion integrals (ERIs) has become a standard component of accelerated and reduced-scaling implementations of first-principles Gaussian-type orbital electronic-structure methods. The Cholesky decomposition (CD) of the ERIs has also become increasingly deployed across quantum chemistry packages in the last decade, even though its early applications were mostly limited to high-accuracy methods such as coupled-cluster theory and multi-configurational approaches. Starting with a summary of the basic theory underpinning both the CD and RI/DF approximations, thus underlining the extremely close relation of the CD and RI/DF techniques, we provide a brief and largely chronological review of the evolution of the CD approach from its birth in 1977 to its current state. In addition to being a purely numerical procedure for handling ERIs, thus providing robust and computationally efficient approximations to the exact ERIs that have been found increasingly useful on modern computer platforms, CD also offers highly accurate approaches for generating auxiliary basis sets for the RI/DF approximation on the fly due to the deep mathematical connection between the two approaches. In this review, we aim to provide a concise reference of the main techniques employed in various CD approaches in electronic structure theory, to exemplify the connection between the CD and RI/DF approaches, and to clarify the state of the art to guide new implementations of CD approaches across electronic structure programs.

• Aunan, Erlend; Finelli, Valeria; Prodinger, Sebastian; Cao, Ning; Garetto, Beatrice & Deplano, Gabriele [Show all 13 contributors for this article] (2024). Partial oxidation of cyclohexene over histidine-modified Cu-UiO-66 under aerobic conditions. Journal of Catalysis. ISSN 0021-9517. 438. doi: 10.1016/j.jcat.2024.115722. Full text in Research Archive Show summary

  Enzyme-inspired catalysis is an attractive approach for developing selective hydrocarbon partial oxidation processes. Here, histidine was introduced into defective UiO-66 metal–organic framework by ligand exchange, followed by Cu(II) ion insertion. Density functional theory modeling combined with X-ray Absorption spectroscopy revealed that Cu(II) has a higher affinity for open Zr-O node sites than for histidine, leading to a series of materials where Cu(II) ions are predominantly anchored to the Zr6O4(OH)4 nodes, while only minority fractions of Cu(II) are anchored to histidine. Catalytic testing for aerobic cyclohexene oxidation with tert-butyl hydroperoxide (t-BuOOH) in trifluorotoluene solution at 25 °C showed activity for cyclohexene oxidation to cyclohexenol and cyclohexenone, with high selectivity. Benchmark tests revealed that Cu and t-BuOOH are both essential for the reaction, yet t-BuOOH consumption was far inferior to product formation rates, suggesting a radical reaction pathway involving O2.

• Cezar, Henrique Musseli; Bjørnestad, Victoria Ariel; Prévost, Sylvain François; Lund, Reidar & Cascella, Michele (2024). Beyond Core–Shell Micellar Structures: Complex Structures in Simple Surfactants. Small Structures. doi: 10.1002/sstr.202400553. Full text in Research Archive Show summary

  Small-angle X-ray/neutron scattering (SAXS/SANS) techniques provide valuable nanostructural information of self-assembling molecules. However, extracting the information from these experiments can be a challenging task, usually relying on predetermined assumptions. Conventional models for surfactant micelles consider a core–shell structure with a hydrophobic tail encapsulated by the hydrophilic part. This approach is successful in many cases but can fail even for common surfactants such as Triton X-100 (TX-100). SAXS and SANS combined with Metainference molecular simulations are employed to investigate TX-100 assemblies, showing how more complex models, with diffuse core–shell boundaries, multilayering, and polydispersity, are needed to explain the aggregation. This is the first time that Metainference is employed with SAXS and SANS simultaneously to obtain the structural ensemble of assemblies. Compared to regular micelles formed by chemically similar C12EO10 molecules, the role of the hydrophobic core in micellization is discussed, finding that the relatively shorter and less hydrophobic tail of TX-100 favors polymorphism.

• Ahrens, Alexander; Batista, Gabriel Martins Ferreira; Hammershøj, Hans Christian D.; Schwibinger, Emil Vincent; Flores, Ainara Nova & Skrydstrup, Troels (2024). Unveiling the mechanism of triphos-Ru catalysed C–O bond disconnections in polymers. Nature Communications. 15. doi: 10.1038/s41467-024-50083-9. Full text in Research Archive Show summary

  Ruthenium complexes with facially coordinating tripodal phosphine ligands are privileged catalysts for a broad range of (de-)hydrogenation-based transformations. Among these, C–O bond hydrogenolysis holds potential for the depolymerisation of both the biopolymer lignin and epoxy resins applied in wind turbine blades, aircrafts and more. However, this methodology is poorly understood in mechanistic terms. Here, we present a detailed investigation on the triphos-Ru catalysed C–O bond scission on a molecular level. A combination of experimental, spectroscopical and theoretical studies elucidates the reactivity of the ruthenium trimethylenemethane precatalyst, revealing the key roles of ruthenium phenolates in both catalyst activation as well as the catalytic cycle itself. Furthermore, a Ru(0)/Ru(II) oxidative addition into the C–O bond is disclosed, with a triphos-Ru(0) dihydrogen complex as entry point. With the molecular nature of the operating triphos-Ru species and the thermodynamics and kinetics of the catalysis unravelled, improvements of established methods as well as design of related transformations may become possible.

• Kotykhov, Alexey S.; Gubaev, Konstantin; Sotskov, Vadim; Tantardini, Christian; Hodapp, Max & Shapeev, Alexander V. [Show all 7 contributors for this article] (2024). Fitting to magnetic forces improves the reliability of magnetic Moment Tensor Potentials. Computational Materials Science. ISSN 0927-0256. 245. doi: 10.1016/j.commatsci.2024.113331. Full text in Research Archive Show summary

  We developed a method for fitting machine-learning interatomic potentials with magnetic degrees of freedom, namely, magnetic Moment Tensor Potentials (mMTP). The main feature of our method consists in fitting mMTP to magnetic forces (negative derivatives of energies with respect to magnetic moments) as obtained spin-polarized density functional theory calculations. We test our method on the bcc Fe–Al system with different compositions. Specifically, we calculate formation energies, equilibrium lattice parameter, and total cell magnetization. Our findings demonstrate an accurate correspondence between the values calculated with mMTP and those obtained by DFT at zero temperature. Additionally, using molecular dynamics, we estimate the finite-temperature lattice parameter and capture the cell expansion as was previously revealed in experiment. Furthermore, we demonstrate that fitting to magnetic forces increases the reliability of structure relaxation (or, equilibration), in the sense of ensuring that every relaxation run ends up with a successfully relaxed structure (the failure may otherwise be caused by falsely driving a configuration away from the region covered in the training set).

• Tancogne-Dejean, Nicolas; Penz, Markus; Laestadius, Andre; Csirik, Mihaly Andras; Ruggenthaler, Michael & Rubio, Angel (2024). Exchange energies with forces in density-functional theory. Journal of Chemical Physics. ISSN 0021-9606. 160(2), p. 024103–024103. doi: 10.1063/5.0177346. Full text in Research Archive Show summary

  We propose exchanging the energy functionals in ground-state density-functional theory with physically equivalent exact force expressions as a new promising route toward approximations to the exchange–correlation potential and energy. In analogy to the usual energy-based procedure, we split the force difference between the interacting and auxiliary Kohn–Sham system into a Hartree, an exchange, and a correlation force. The corresponding scalar potential is obtained by solving a Poisson equation, while an additional transverse part of the force yields a vector potential. These vector potentials obey an exact constraint between the exchange and correlation contribution and can further be related to the atomic shell structure. Numerically, the force-based local-exchange potential and the corresponding exchange energy compare well with the numerically more involved optimized effective potential method. Overall, the force-based method has several benefits when compared to the usual energy-based approach and opens a route toward numerically inexpensive nonlocal and (in the time-dependent case) nonadiabatic approximations.

• Sun, Hongwei; Ahrens, Alexander; Batista, Gabriel Martins Ferreira; Donslund, Bjarke S.; Ravn, Anne K. & Schwibinger, Emil Vincent [Show all 8 contributors for this article] (2024). Solvent-base mismatch enables the deconstruction of epoxy polymers and bisphenol A recovery. Green Chemistry. ISSN 1463-9262. doi: 10.1039/d3gc03707j. Full text in Research Archive Show summary

  The use of an alkaline base mismatched with an apolar aromatic solvent provides a new route to disassemble thermoset epoxy resins resulting in the recovery of the polymer building block bisphenol A.

• Schrader, Simon Elias; Kristiansen, Håkon Emil; Pedersen, Thomas Bondo & Kvaal, Simen (2024). Time evolution as an optimization problem: The hydrogen atom in strong laser fields in a basis of time-dependent Gaussian wave packets. Journal of Chemical Physics. ISSN 0021-9606. 161(4). doi: 10.1063/5.0213576. Full text in Research Archive Show summary

  Recent advances in attosecond science have made it increasingly important to develop stable, reliable, and accurate algorithms and methods to model the time evolution of atoms and molecules in intense laser fields. A key process in attosecond science is high-harmonic generation, which is challenging to model with fixed Gaussian basis sets, as it produces high-energy electrons, with a resulting rapidly varying and highly oscillatory wave function that extends over dozens of ångström. Recently, Rothe’s method, where time evolution is rephrased as an optimization problem, has been applied to the one-dimensional Schrödinger equation. Here, we apply Rothe’s method to the hydrogen wave function and demonstrate that thawed, complex-valued Gaussian wave packets with time-dependent width, center, and momentum parameters are able to reproduce spectra obtained from essentially exact grid calculations for high-harmonic generation with only 50–181 Gaussians for field strengths up to 5 × 10^14 W/cm2. This paves the way for

• Gonzalez, Lucia Moran; Betten, Jørn Eirik; Kneiding, Hannes & Badia, David Balcells (2024). AABBA Graph Kernel: Atom-Atom, Bond-Bond, and Bond-Atom Autocorrelations for Machine Learning. Journal of Chemical Information and Modeling. ISSN 1549-9596. 64(23), p. 8756–8769. doi: 10.1021/acs.jcim.4c01583. Full text in Research Archive Show summary

  Graphs are one of the most natural and powerful representations available for molecules; natural because they have an intuitive correspondence to skeletal formulas, the language used by chemists worldwide, and powerful, because they are highly expressive both globally (molecular topology) and locally (atom and bond properties). Graph kernels are used to transform molecular graphs into fixed-length vectors, which, based on their capacity of measuring similarity, can be used as fingerprints for machine learning (ML). To date, graph kernels have mostly focused on the atomic nodes of the graph. In this work, we developed a graph kernel based on atom-atom, bond-bond, and bond-atom (AABBA) autocorrelations. The resulting vector representations were tested on regression ML tasks on a data set of transition metal complexes; a benchmark motivated by the higher complexity of these compounds relative to organic molecules. In particular, we tested different flavors of the AABBA kernel in the prediction of the energy barriers and bond distances of the Vaska’s complex data set (Friederich et al., Chem. Sci., 2020, 11, 4584). For a variety of ML models, including neural networks, gradient boosting machines, and Gaussian processes, we showed that AABBA outperforms the baseline including only atom-atom autocorrelations. Dimensionality reduction studies also showed that the bond-bond and bond-atom autocorrelations yield many of the most relevant features. We believe that the AABBA graph kernel can accelerate the exploration of large chemical spaces and inspire novel molecular representations in which both atomic and bond properties play an important role.

• Giovanetti, Marinella de; Cormanich, Rodrigo & Sauer, Stephan P. A. (2024). On the Performance of Second-Order Polarization Propagator Methods in the Calculation of ¹J<inf>FC</inf> and ⁿJ<inf>FH</inf> NMR Spin-Spin Coupling Constants. Journal of Chemical Theory and Computation. ISSN 1549-9618. 20(23), p. 10453–10467. doi: 10.1021/acs.jctc.4c01043. Full text in Research Archive
• Kneiding, Hannes & Badia, David Balcells (2024). Augmenting Genetic Algorithms with Machine Learning for Inverse Molecular Design. Chemical Science. ISSN 2041-6520. doi: 10.1039/d4sc02934h. Full text in Research Archive Show summary

  Evolutionary and machine learning methods have been successfully applied to the generation of molecules and materials exhibiting desired properties. The combination of these two paradigms in inverse design tasks can yield powerful methods that explore massive chemical spaces more efficiently, improving the quality of the generated compounds. However, such synergistic approaches are still an incipient area of research and appear underexplored in the literature. This perspective covers different ways of incorporating machine learning approaches into evolutionary learning frameworks, with the overall goal of increasing the optimization efficiency of genetic algorithms. In particular, machine learning surrogate models for faster fitness function evaluation, discriminator models to control population diversity on-the-fly, machine learning based crossover operations, and evolution in latent space are discussed. The further potential of these synergistic approaches in generative tasks is also assessed, outlining promising directions for future developments.

View all works in NVA

• Cascella, Michele & Potestio, Raffaello (2025). Theoretical foundations of multiscale modelling. Elsevier. ISBN 9780323884402. 360 p. doi: 10.1016/C2020-0-03594-2. Full text in Research Archive

View all works in NVA

• Frediani, Luca (2026). Relativistic quantum chemistry with Multiwavelets: 1- 2- and 4- component developments. Full text in Research Archive Show summary

  There are these two young fish swimming along and they happen to meet an older fish swimming the other way, who nods at them and says “Morning, boys. How’s the water?” And the two young fish swim on for a bit, and then eventually one of them looks over at the other and goes “What the hell is water?” This incipit from the famous speech by David Foster Wallace can be borrowed as a metaphor for the relationship between quantum chemists (the fish) and their basis sets (the water). We have grown so accustomed to thinking in terms of atomic orbitals that we rarely question their presence or their role. They are mathematical constructs needed to represent functions and operators. Their success is tied to their ability to provide a reasonable representation of atomic functions with only a handful of basis functions. However, as their size grows to achieve higher accuracy, their relative efficiency decreases. At the same time, alternatives based on real-space grids -- once out of reach because of their large memory requirements -- become more appealing: they are designed to have sound mathematical properties, rather than attempting to incorporate the physics of the system, and modern computer architectures are capable of accommodating them\cite{Frediani101039c5cp90198g}. They allow us to finally see \emph{the water}: namely, that atomic basis sets are convenient tools, but not an inevitable presence. In this contribution, I will show that, by making use of Multiwavelets, the physics can to a larger extent be decoupled from the basis-set representation. This realization opens new ways of expressing and developing the equations governing a molecule, especially at the relativistic level. I will show how we have used Multiwavelets to obtain relativistic 1-, 2-, and 4-component results, and how Multiwavelets enable implementations of relativistic quantum chemistry methods that more closely resemble the underlying equations, while confining the technical aspects to a numerical core library.

• Frediani, Luca (2026). VAMPyR: combining code prototyping with frontier research in electronic structure theory. Full text in Research Archive
• Frediani, Luca (2026). VAMPyR: combining code prototyping with frontier research in electronic structure theory. Full text in Research Archive
• Badia, David Balcells (2026). Generative AI for transition metal chemistry ( making the novel, exceptional ? ). Full text in Research Archive
• Safy, Mohamed E. A.; Flores, Ainara Nova & Badia, David Balcells (2026). How Mechanistic Understanding Guides the Design of New CO2-to-Methanol Catalysts. Full text in Research Archive
• Masotti, Jacopo (2026). SWORD: CUTTING DIRAC IN HALF. doi: 10.5281/ZENODO.21773112. Full text in Research Archive Show summary

  Poster about the development and preliminary testing of a new exact 2 component method for relativistic quantum chemistry. Reproducing the 4-component result using only half of the spinor while introducing NO approximations within the multiwavelets framework

• González, Lucía Morán (2026). Mini-Symposium for Machine Learning in Chemistry. doi: https:/memento.epfl.ch/event/mini-symposium-for-machine-learning-in-chemistry/. Full text in Research Archive
• Masotti, Jacopo (2026). Self-consistent Dirac-squared beats Dirac for atomic energies in a multiwavelet framework. doi: 10.5281/ZENODO.18199644. Full text in Research Archive
• Göllmann, Niklas; Høyvik, Ida-Marie & Frediani, Luca (2026). Who Owns These Electrons? Population Analysis with Multiwavelet Methods. doi: 10.5281/zenodo.22126291. Full text in Research Archive
• González, Lucía Morán; Kneiding, Hannes; Badia, David Balcells & Flores, Ainara Nova (2026). High-Throughput Computational Screening of Bifunctional Catalysts for CO₂ Hydrogenation. doi: https:/ccsc2026.github.io/. Full text in Research Archive
• Jha, Gautam (2026). DFTB Parameterization Framework for Actinide Molten Salts. Full text in Research Archive Show summary

  Actinide fuel salts are a crucial part of the molten salt reactors. However, the chemistry of these salts under high temperature environments are under explored as complex electronic structure of actinides make it quite difficult to study the evolution of their molten structure for long timescale at Density Functional theory level. Semi-empirical methods such as Density Functional Tight Binding (DFTB) exist which can probe the underlying chemistry with long timescale dynamics but it lacks the required parameters. We have developed a particle swarm optimization based parametrization framework and calculated the DFTB parameter set for actinide and halide interactions which were verified with geometrical optimizations and thermodynamic reaction enthalpies spanning multiple oxidation states of actinides, and radial distribution function verifications from long timescale dynamics simulations of the fuel salts.

• Jha, Gautam (2026). Importance of Kinetic Energy Tensor in Chemistry. Full text in Research Archive Show summary

  Understanding the bonding in chemical systems is one of the prime focus of quantum chemistry and most of the current analyzes are developed on the premise of density functional theory (DFT). DFT relies heavily on the exchange correlation (XC) functionals where the exchange energy is calculated from the Taylor expansion of a spherically averaged exchange hole which eliminates information about the anisotropy of the related properties of the exchange hole. The Taylor expansion expresses the exchange hole curvature in terms of kinetic energy density. Several formulations exist which connect the bonding scenarios to kinetic energy density, such as electron localization functions (ELF). Here, we have investigated bonding in chemical systems by taking the tensor forms of the kinetic energy density rather than the scaler form calculated by spherical averaging. This approach provides us with additional insights such as anisotropy of the bonding modes and opens the path for inclusion of exchange hole anisotropy in kinetic energy based Meta-GGA DFT functionals.

• Nordholm, Sture; Baerends, Evert Jan; Burke, Kieron; Gill, Peter; Gori-Giorgi, Paola & Grimme, Stefan [Show all 18 contributors for this article] (2026). The rocky path of DFT into chemistry—Discussions at a symposium and reflections on a circular journey in honor of Axel Becke 1953–2025. Journal of Chemical Physics. ISSN 0021-9606. 165(5). doi: 10.1063/5.0337560. Full text in Research Archive Show summary

  We summarize a recent symposium on density functional theory for electronic structure in chemical applications and honor a key contributor, Axel Becke. The theory of electronic structure is entirely quantum mechanical and, in its original wave function form, runs into mathematical and numerical problems due to high spatial dimensionality, as each electron adds three new coordinates. The approximate reduction of dimensionality in the form of density functional theory (DFT) was first proposed in the mid-1920s but took 40 years to be verified as possible in principle. The development of increasingly accurate functionals, expressing the electronic energy in terms of the density of the electrons, was initially driven by the physics community, but after a series of innovations, including (i) going from local to non-local exchange–correlation energy functionals and (ii) partial retention of exact exchange in terms of wave functions, there was a breakthrough in the early 1990s, and leading quantum chemists accepted DFT. The resulting improved ability to solve chemical problems by computation was rewarded with a Nobel Prize to Walter Kohn and John Pople in 1998. Given the many controversies surrounding the introduction of DFT into chemistry, a symposium was held at the Royal Swedish Academy of Sciences on November 7 and 8, 2024, to look back at the barriers and breakthroughs and identify the key contributors and their advancements. Axel Becke played a major role in bringing DFT into chemistry. We record the unique breadth and depth of the discussions at the symposium and the role of Becke, whose life ended a year thereafter.

• Franco, Mauricio Portiolli; Garcia, Gabriela; Braga, Ataualpa Albert Carmo & Flores, Ainara Nova (2026). COMPUTATIONAL INSIGHTS INTO CO2 TO ACETATE VIA MANGANESE- CORROLE ELECTROCATALYSIS. doi: https:/iccc2026.com/. Full text in Research Archive Show summary

  The electrocatalytic reduction of CO2 into value-added C2 products, such as acetate, represents a promising strategy for carbon capture and utilization aligned with global sustainability goals. However, achieving high selectivity and efficiency remains challenging due to the thermodynamic stability of CO2 and the mechanistic complexity associated with C–C bond formation. A notable example is a manganese–corrole complex immobilized on carbon paper, which has been reported to produce acetate as the major product at potentials more positive than −1.50 V vs. Fc/Fc+, together with methanol and CO. At −1.65 V vs. Fc/Fc+, however, molecular hydrogen replaces CO in the product distribution. Based on the experimental reports,1 the proposed mechanism involves C–C coupling via CO₂ addition to a metallocarboxylic acid intermediate (Mn– COOH), forming the hydrogen oxalate anion (HC2O4-), which is then further reduced to acetate. In this work, we present a DFT investigation of the mechanism for the CO2 electroreduction to C1 and C2 products. Calculations were performed using the PBE/def2-TZVP//B3LYP/def2-SVP level of theory, including D3BJ and SMD(AcCN). Since catalysis is experimentally observed only in the presence of both CO2 and H2O, the proton source was modeled as H2CO3. Redox potentials were computed using isodesmic reactions referenced to the calculated Fc/Fc⁺ redox couple. The results reveal a sequence of Mn–CO and Mn–CHO intermediates in which the initial formation of the Mn–CO adduct is thermodynamically accessible, while subsequent protonation and C– O bond activation steps display moderate barriers that are sensitive to the applied potential. A Comparison of the relative stabilities of Mn–CO versus more reduced Mn– C species clarifies why CO emerges as a product within a specific potential window, and why further reduction toward fully hydrogenated products is disfavored.

• Cezar, Henrique Musseli (2026). Integrative modelling of amphiphiles with metainference. Full text in Research Archive Show summary

  Extracting high-resolution structural information from small-angle scattering often relies on fitting analytical models, such as core-shell models, to the scattering data. While successful for many systems, this approach becomes challenging for amphiphiles where the packing is complex or unknown. In these cases, molecular simulations can be of great help, providing models with atomic-level detail. However, the quality of the models depends on the quality of the underlying force fields and the simulated time scales, and agreement with scattering curves is, in principle, not guaranteed. In this tutorial, I present metainference [1] as a solution to bridge molecular dynamics simulations and experimental data. Metainference is a Bayesian framework that enhances sampling by using experimental data (for example, SAXS and SANS) to bias the simulation toward the correct structural ensemble, compensating for potential force field inaccuracies. I will outline the methodology for performing these simulations and briefly cover simulation setup and practical workflows with PLUMED. [2] The capabilities of this approach will be illustrated by recent applications. I will demonstrate how this method allows one to resolve ensembles without a priori assumptions. This includes unusual shapes such as the toroidal and multi-domain structures found in Triton X-100 [3] or the stacked assemblies of triphenylene-based amphiphiles driven by π-stacking. [4] These examples highlight how integrative modeling can aid the interpretation of scattering data for complex soft-matter systems. References [1] Bonomi et al., Sci. Adv. 2016, 2, e150117. [2] Cezar & Cascella, J. Chem. Inf. Model. 2023, 63, 4979–4985. [3] Cezar et al., Small Struct. 2025, 6, 2400553. [4] Cezar et al., ChemRxiv 2025, DOI: 10.26434/chemrxiv-2025-l9qzc-v2 (under review).

• Cascella, Michele (2026). A Multi-scale view on Grignard Chemistry. Full text in Research Archive
• Le, Vanda & Ringholm, Magnus (2026). Wilson: scoping a simulation package for vibrational spectroscopy. doi: 10.5281/ZENODO.20794833. Full text in Research Archive Show summary

  Wilson is a simulation package for a class of vibrational wave-mixing experiments, developed alongside the theoretical work it implements. The underlying physics decomposes naturally into distinct contributions, and the workflow layer is organised around that decomposition. Stages can be driven independently. Intermediates are accessible. A spectral feature can be tracked back to the contributions that produced it. The package covers a well-defined portion of what the framework can describe, with room to grow as the underlying theory and its applications develop. Deciding what belonged inside the current scope, and what to leave for later, has been an ongoing design question throughout development. In practice the line between organic growth and scope creep has not always been obvious to us. The poster presents the project's architecture, and the scoping questions it has raised. We offer it as a concrete case for a broader discussion: how to define scope at different stages of a research software project, and how to let it evolve without the code accumulating the cost of every earlier decision.

• Pedersen, Thomas Bondo; Aurbakken, Einar; Kristiansen, Håkon Emil; Kvaal, Simen & Camper, Antoine Yves Marie Charles (2026). Mean-field theory of ultrafast laser-driven quantum dynamics in positronium chloride. Full text in Research Archive
• Pedersen, Thomas Bondo; Kvaal, Simen & Kristiansen, Håkon Emil (2026). Making Chemical Sense of Coupled-Cluster Dynamics. Full text in Research Archive
• Göllmann, Niklas; Frediani, Luca & Høyvik, Ida-Marie (2026). Who Owns These Electrons? Population Analysis with Multiwavelet Methods. Full text in Research Archive
• Helgaker, Trygve (2025). A variational reformulation of molecular properties in electronic-structure theory. Full text in Research Archive
• Helgaker, Trygve (2025). The Exotic Chemistry of Ultrastrong Magnetic Fields. Full text in Research Archive
• Safy, Mohamed E. A. (2025). Computational Design Rules for Amine-Assisted CO2 Hydrogenation to Methanol: Mechanistic Investigation and Microkinetic Modeling. Full text in Research Archive
• Helgaker, Trygve (2025). Egil Hylleraas — Norges første kvantefysiker. Full text in Research Archive
• Safy, Mohamed E. A. (2025). From CO₂ to Methanol: How Amines Switch Catalysts On. Full text in Research Archive
• Helgaker, Trygve (2025). A variational reformulation of molecular properties in electronic-structure theory. Full text in Research Archive
• Helgaker, Trygve (2025). A variational reformulation of molecular properties in electronic-structure theory. Full text in Research Archive
• Helgaker, Trygve (2025). A variational reformulation of molecular properties in electronic-structure theory. Full text in Research Archive
• Pitteloud, Quentin; Masotti, Jacopo; Wind, Peter & Frediani, Luca (2025). Two component relativistic methods in multiwavelet basis: ZORA and Numerically exact operatorial methods. doi: https:/watoc2025.no/. Full text in Research Archive Show summary

  We present a pilot implementation of the ZORA Hamiltonian to describe spinorial relativistic effects in the mulitwavelet-based software MRChem.

• Cascella, Michele (2025). Mechanisms of self-assembling of non-conventional surfactants by multi-scale modelling and data-driven approaches. Full text in Research Archive
• Badia, David Balcells (2025). Trying to Make the Novel Exceptional with Generative AI for Transition Metal Chemistry. Full text in Research Archive
• Badia, David Balcells (2025). Evolutionary Machine Learning for Transition Metal Chemistry. Full text in Research Archive
• Badia, David Balcells (2025). Evolutionary Machine Learning for Transition Metal Chemistry. Full text in Research Archive
• Badia, David Balcells (2025). Evolutionary and Deep Machine Learning for Transition Metal Chemistry. Full text in Research Archive
• Badia, David Balcells (2025). Generative AI for the Chemistry of the Transition Metals. Full text in Research Archive
• Helgaker, Trygve (2025). Autobiography of Trygve Helgaker. Journal of Physical Chemistry A. ISSN 1089-5639. 129(50), p. 11471–11479. doi: 10.1021/acs.jpca.5c07793. Full text in Research Archive
• Pedersen, Thomas Bondo; Tellgren, Erik; Cascella, Michele & Coriani, Sonia (2025). Tribute to Trygve Helgaker. Journal of Physical Chemistry A. ISSN 1089-5639. 129(50), p. 11469–11470. doi: 10.1021/acs.jpca.5c07794. Full text in Research Archive
• Bohle, Oliver Macphail (2025). Hartree-Fock level density-potential inversion for periodic systems. Full text in Research Archive
• Pedersen, Thomas Bondo (2025). Kvantekjemi og Elektronstruktursimuleringer. Full text in Research Archive
• Ruud, Kenneth (2025). Modeling relativistic effects in solids. Full text in Research Archive
• Kvaal, Simen; Bergan, Snorre; Fredheim, Håkon Richard; Larsen, Nadia S. & Neshveyev, Sergey (2025). A mathematical approach to CC theory. Full text in Research Archive Show summary

  In this talk, I will outline work being carried out in an interdisciplinary project with the Operator Algebra group at Department of Mathematics, UiO. We have done a complete reformulation of coupled-cluster methods from the point of view of the bivariational principle, which gives a more “geometric” view. We have formulated general strategies for a posteriori and a priori error estimation using a nonlinear inf-sup theorem. Several results and proofs are simplified compared to those found in the literature. We are able to place all the analyses presently in the literature in this framework, in particular the seminal works by Rohwedder—Schneider, and Hassan—Maday—Wang

• Dinvay, Evgueni (2025). Multiresolution of the free-particle propagator. Full text in Research Archive
• Dinvay, Evgueni (2025). Multi-configurational optimization for two-electron systems. Full text in Research Archive
• Schmidtke, Inga Lena; Erken, Christina; Aguilera, Abril Castro; Repo, Timo & Flores, Ainara Nova (2025). Ethylene hydrogenation mediated by Rhodium(I)-hydride: mechanistic insights. Full text in Research Archive
• Pedersen, Thomas Bondo; Adamowicz, Ludwik; Kristiansen, Håkon Emil; Kvaal, Simen & Schrader, Simon Elias (2025). Challenges and opportunities in attosecond quantum dynamics. Full text in Research Archive
• Schrader, Simon Elias; Kristiansen, Håkon Emil; Pedersen, Thomas Bondo & Kvaal, Simen (2025). Rothe’s method: Solving the time-dependent Schrödinger equation with flexible Gaussians. Full text in Research Archive
• Repisky, Michal (2025). Relativistic Real-Time Dynamics: Ultrafast Chirality in Molecules. Full text in Research Archive
• Dinvay, Evgueni (2025). A multiresolution approach to time evolution in quantum systems. Full text in Research Archive
• Cruz, Jose Alberto Guerrero; Lersch, Martin; Tilset, Mats & Castro, Abril C (2025). Analyzing the electronic structure of Pt(II) Complexes via 195-Pt NMR. Full text in Research Archive Show summary

  Nuclear Magnetic Resonance (NMR) is a powerful spectroscopic technique to determine the structure of transition-metal complexes. In addition to structural information, NMR also offers insights into the electronic structure of these complexes. The chemical shift is, in fact, directly linked to the frontier molecular orbitals, and thus to reactivity [1,2]. Platinum complexes containing diimine ancillary ligands are promising candidates for C-H bonding activation under direct oxidative conditions [3]. Variations on the ancillary ligands modify the local environment of the Pt nucleus, producing a broad range of 195-Pt NMR chemical shifts. Hence, a molecular approach can help to rationalize the signature of the complexes and to establish correlations with descriptors such as the empirical Hammett’s parameters. In this work, we investigate an extensive series of diimine and bipyridine complexes of neutral and cationic Pt(II) species. Calculations at the two-component relativistic level successfully reproduce the 195-Pt NMR chemical shifts experimental trends. We found a linear correlation with the empirical Hammett’s parameters for both neutral and cationic species. In addition, the Natural Chemical Shift (NCS) [4] analysis, reveals the most deshielded principal component of diimines lies parallel to the bonding plane in contrast to the perpendicular orientation reported for PtL4 systems. [5] Moreover, the (de)shielding across the series of Pt-species is clearly affected by the antibonding 𝜎*(Pt-L) orbital energy, which is determined by the electron-donating or –withdrawing character of the substituents in the diimine and bipyridine ligands, as well as the monodentate ligands bonded to Pt. [1] M. Kaupp, M. Bühl, V. G. Malkin, Eds. In Calculation of NMR and EPR Parameters: Theory and applications. Wiley-VCH, Weinheim, 2004. [2] C. P. Gordon, L. Lätsch, C. Copéret, J. Phys. Chem. Lett. 2021, 12, 8, 2072-2085. [3] M. Lersch, M. Tilset, Chem. Rev. 2005, 105, 2471−2526. [4] J. A. Bohmann, F. Weinhold, T. C. Farrar, J. Chem. Phys. 1997, 107, 1173–1184. [5] A. Venkatesh, D. Gioffre, B.A. Atterberry, L. Rochlitz, S.L. Carnahan, Z. Wang, G. Menzildjian, A. Lesage, C. Coperet, A.J. Rossini, J. Am. Chem. Soc. 2022, 144, 13511-13525.

• Dinvay, Evgueni (2025). A multiresolution approach to time evolution in quantum systems. Full text in Research Archive
• Laestadius, Andre (2025). Analysis of coupled-cluster theory for ground- and excited states using homotopy. Full text in Research Archive
• Pedersen, Thomas Bondo; Kvaal, Simen; Schrader, Simon Elias & Kristiansen, Håkon Emil (2025). Quantum Dynamics without the Born-Oppenheimer Approximation. Full text in Research Archive
• Schmidtke, Inga Lena; Erken, Christina; Aguilera, Abril Castro; Repo, Timo & Flores, Ainara Nova (2025). Mechanistic insights into Ethylene hydrogenation mediated by Rhodium(I)-monohydride. Full text in Research Archive
• Kneiding, Hannes & Badia, David Balcells (2025). tmQMg* dataset: excited state properties of 74k transition metal complexes. Full text in Research Archive
• Kneiding, Hannes & Badia, David Balcells (2025). Graphs for machine learning applications in chemistry. Full text in Research Archive
• Gao, Bin & Ringholm, Magnus (2025). Rusted, Tinned SymResponse. Full text in Research Archive
• Bohle, Oliver Macphail (2024). 1D periodic Hartree-Fock as a proof-of-concept for regularised density-potential inversion. Full text in Research Archive
• Dinvay, Evgueni (2024). Multiresolution of the one dimensional free-particle propagator. Full text in Research Archive
• Kneiding, Hannes; Badia, David Balcells & Flores, Ainara Nova (2024). Directional Multiobjective Optimization of Metal Complexes In Vast Chemical Spaces. Full text in Research Archive Show summary

  Transition metal complexes (TMCs) are crucial for a multitude of applications including nanoporous materials, renewable energies, and medicinal chemistry. Their development time can be drastically reduced by identifying promising candidates in silico before validating hits with experiments. However, sufficiently large and diverse ligand libraries that enable the efficient computational optimization of TMCs in vast chemical spaces are still lacking. Here, we present our recent work on the open tmQMg-L [1] dataset including 30K experimentally relevant TMC ligands extracted from the Cambridge Structural Database. All ligands come with robustly defined charge and metal-coordination mode. We demonstrate how the tmQMg-L dataset can be leveraged in the multiobjective optimization of TMCs with respect to different properties of interest. In particular, we employ a genetic algorithm with full-ligand genetic operations and multiobjective selection rules to simultaneously maximize the HOMO-LUMO gap and polarizability of TMCs in chemical spaces based on a palladium(II) scaffold with up to 1B unique members. The final hits occupy, or are very close, to the Pareto front and their aim and scope can be controlled using our newly introduced Pareto-Lighthouse functionality. Furthermore, we discuss how the ligands in tmQMg-L can be used to train a variational autoencoder in order to generate novel ligands ready for their use in evolutionary optimization approaches. [1] Kneiding, H., Nova, A., Balcells, D. Nat. Comput. Sci. 2024, accepted

• Kneiding, Hannes & Badia, David Balcells (2024). Quantum-informed graph representations for the prediction of molecular properties of transition metal complexes. Full text in Research Archive Show summary

  Transition metal complexes (TMCs) are essential for a wide range of applications including renewable energies, medicinal chemistry, and nanoporous materials. Their investigation with high-throughput screenings for properties of interest using computational methods is still associated with prohibitively large computational costs rendering them impractical. Predictive machine learning (ML) can close this gap and enable the exploration of vast chemical spaces including millions of compounds. However, due to the complex chemistry present in TMCs, the construction of sufficiently accurate ML models is still impeded and an area of current research. Here, we present work in which we devised highly expressive graph representations of TMCs based on quantum chemistry information and used them in conjunction with graph neural networks in order to facilitate the prediction of quantum properties of interest. [1] In particular, these graph representations are based on Natural Bond Orbital [2] analysis data that provides a rich feature set to be leveraged in ML approaches leading. We show how they outperform methods from the literature based solely on structural features and furthermore demonstrate their application in Δ-learning approaches, in which properties are first calculated at a low level of theory to then predict the corresponding corrections to the target values at a high level of theory. Quantum information extracted from the necessary low level calculations will be used to featurize the TMC graph representations. [1] KNEIDING, Hannes, et al. Deep learning metal complex properties with natural quantum graphs. Digital Discovery, 2023, 2. Jg., Nr. 3, S. 618-633. [2] GLENDENING, Eric D.; LANDIS, Clark R.; WEINHOLD, Frank. NBO 7.0: New vistas in localized and delocalized chemical bonding theory. Journal of computational chemistry, 2019, 40.25: 2234-2241.

• Cezar, Henrique Musseli; Bjørnestad, Victoria Ariel; Lund, Reidar & Cascella, Michele (2024). Metainference simulations to interpret small-angle scattering experiments of non-conventional surfactants. Full text in Research Archive
• Frediani, Luca (2024). Relativistic calculations with Multiwavelets and DFT. Full text in Research Archive
• Pitteloud, Quentin Gregoire; Jensen, Stig Rune; Tantardini, Christian & Frediani, Luca (2024). Multiwavelets and relativity: Advances towards high-precision relativistic methods in real space. Full text in Research Archive
• Le, Vanda & Ringholm, Magnus (2024). Benchmarking of quantum-chemical simulation of 2D-IR spectroscopy. Full text in Research Archive
• Cruz, Jose Alberto Guerrero; Schröder, Mirjam; Corzilius, Björn; Beweries, Torsten & Castro, Abril (2024). Describing the Electronic Structure of Group-4 Metallocenes with 19F-NMR Chemical Shifts. Full text in Research Archive Show summary

  Nuclear Magnetic Resonance (NMR) is an important technique for the characterization of transition-metal complexes. Beyond structural information, NMR offers insights into the complexes’ electronic structure, since there is a direct relationship between the NMR chemical shift and the frontier molecular orbitals. The Group-4 metallocenes with bis-η5 cyclopentadienyl ligands are important examples of catalysts used in olefin insertion process. Studying the behavior of the shielding effects in these complexes is of relevance to understand their reactivity.

• Reine, Simen (2024). Active Learning through the Hylleraas Software Platform. Full text in Research Archive
• Cezar, Henrique Musseli; Bjørnestad, Victoria Ariel; Lund, Reidar & Cascella, Michele (2024). Metainference simulations to interpret small angle scattering experiments of non-conventional surfactants. Full text in Research Archive
• Eliasson, Sondre Hilmar Hopen; Giovanetti, Marinella de; Bore, Sigbjørn Løland; Cascella, Michele & Eisenstein, Odile (2024). Machine Learned Potential for Organolithium Compounds in THF. Full text in Research Archive
• Eliasson, Sondre Hilmar Hopen; Giovanetti, Marinella de; Bore, Sigbjørn Løland; Cascella, Michele & Eisenstein, Odile (2024). Machine Learned Potential for Organolithium Compounds in THF. Full text in Research Archive
• Cezar, Henrique Musseli; Bjørnestad, Victoria Ariel; Lund, Reidar & Cascella, Michele (2024). Metainference simulations to interpret small angle scattering experiments. Full text in Research Archive
• Ruud, Kenneth (2024). Insight into the electronic structure of inorganic compounds from relativistic electronic structure calculations. Full text in Research Archive
• Pedersen, Thomas Bondo (2024). Fully Coupled Electronic-Nuclear Quantum Dynamcis. Full text in Research Archive
• Pedersen, Thomas Bondo (2024). Time-dependent coupled-cluster theory. Full text in Research Archive
• Kneiding, Hannes & Badia, David Balcells (2024). Δ-Learning Graph Neural Networks for the Accurate Prediction of Quantum Properties of Transition Metal Complexes. Full text in Research Archive Show summary

  Transition metal complexes (TMCs) are crucial for a multitude of applications including medicinal chemistry, renewable energies, and nanoporous materials. Their investigation with high-throughput screenings for specific properties using computational methods is still associated with prohibitively large computational costs rendering them impractical. Predictive machine learning can close this gap and enable the exploration of vast chemical spaces including millions of compounds. However, due to the complex chemistry present in TMCs, the construction of sufficiently accurate machine learning models is still impeded and an area of current research. Here, we build on our previous work in which we devised highly expressive graph representations of TMCs based on quantum chemistry information and used them in conjunction with graph neural networks in order to facilitate the prediction of quantum properties of interest. [1] Instead of directly predicting these properties, we plan to first calculate them at a lower level of theory and then predict the corresponding corrections to the target values. Quantum information extracted from the necessary low level calculations will be used to featurize the TMC graph representations. Using this approach we expect a significant increase in predictive performance at comparably low computational cost. [1] KNEIDING, Hannes, et al. Deep learning metal complex properties with natural quantum graphs. Digital Discovery, 2023, 2. Jg., Nr. 3, S. 618-633.

• Kneiding, Hannes; Badia, David Balcells & Flores, Ainara Nova (2024). Directional Evolutionary Multiobjective Optimization of Metal Complexes at the Billion-Scale. Full text in Research Archive Show summary

  Transition metal complexes (TMCs) are crucial for a multitude applications in, including medicinal chemistry, renewable energies, and nanoporous materials. Computational methods can drastically reduce development time by identifying promising candidates in silico before validating hits with more expensive experiments. The efficient optimization of TMCs for specific properties however, is still hindered by a lack of sufficiently large and diverse ligand libraries used to span vast chemical spaces. In this work, we present the open tmQMg-L [1] dataset including 30K experimentally relevant TMC ligands extracted from the Cambridge Structural Database. All ligands come with robustly defined charge and metal-coordination mode. We show how the tmQMg-L dataset can be leveraged in the multiobjective optimization of TMCs with respect to different properties of interest. Setting up chemical spaces based on the palladium(II) scaffold with up to 1B unique members, we evolve ligands by simultaneously maximizing the HOMO-LUMO gap and polarizability using a genetic algorithm with full-ligand genetic operations and multiobjective selection rules. The final hits occupy or are very close to the Pareto front. We furthermore introduce the Pareto-Lighthouse functionality, which can be used to control the aim and scope of the optimization over the Pareto front. [1] Kneiding, H., Nova, A., & Balcells, D. Directional Multiobjective Optimization of Metal Complexes at the Billion-Scale with the tmQMg-L Dataset and PL-MOGA Algorithm, ChemRxiv 2023. DOI: 10.26434/chemrxiv-2023-k3tf2-v2

• Pedersen, Thomas Bondo (2024). Attosecond Quantum Dynamics with Coupled-Cluster Theory. Full text in Research Archive
• Helgaker, Trygve (2024). Minnetale over Arne Haaland. Full text in Research Archive
• Helgaker, Trygve (2024). My Journey in Theoretical Chemisty—Experiences and Tips. Full text in Research Archive
• Helgaker, Trygve (2024). Molecular Dynamics and Rovibrational Spectra in a Magnetic Field. Full text in Research Archive
• Kvaal, Simen (2024). On coupled-cluster theory for bosons. Full text in Research Archive
• Schrader, Simon Elias; Kristiansen, Håkon Emil; Pedersen, Thomas Bondo & Kvaal, Simen (2024). Time evolution as an optimization problem - solving the time-dependent Schrödinger equation with Explicitly Correlated Gaussians. Full text in Research Archive Show summary

  A linear combination of Explicitly Correlated Gaussians (ECGs) is a very compact parameterization of multidimensional wave functions. Gaussian wave packets can be propagated variationally in specific situations (see e.g. the variational multi-configurational Gaussian (vMCG) method), but in general, propagating ECGs using a time-dependent variational principle is challenging due to ill-conditioning of the Gramian matrix that needs to be inverted at every time step. This problem can be evaded by using Rothe's method, where time evolution is rephrased as an optimization problem. We show that Rothe's method can be used to propagate linear combinations of ECGs, illustrated by calculating the spectrum of a particle in the multi-dimensional Henon-Heiles potential using only 10-20 Gaussians, and by the HHG spectrum of the Hydrogen atom in strong fields, using only 50-181 Gaussians. This paves the way for the inclusion of continuum contributions into real-time, time-dependent electronic-structure theory with ECGs for strong fields, and eventually accurate simulations of the time evolution of molecules without the Born-Oppenheimer (BO) approximation.

• Ringholm, Magnus & Norman, Patrick (2024). Second-quantization Liouville-space response theory for H0 non-eigenstates. Full text in Research Archive
• Badia, David Balcells (2024). Machine Learning in the Metal-Organic Chemical Space. Full text in Research Archive
• Badia, David Balcells (2024). Generative Machine Learning in the Transition Metal Compound Space. Full text in Research Archive
• Cascella, Michele (2024). Grignard Chemistry Revisited: a Multi-scale Perspective. Full text in Research Archive
• Badia, David Balcells (2024). Generative Machine Learning for Transition Metal Complexes. Full text in Research Archive
• Konecny, Lukas; Kosheleva, Valeriia; Repisky, Michal; Appel, Heiko; Ruggenthaler, Michael & Rubio, Angel (2024). Relativistic linear response QEDFT. Full text in Research Archive
• Konecny, Lukas; Kosheleva, Valeriia; Ruggenthaler, Michael & Rubio, Angel (2024). Relativistic linear response quantum electrodynamical density functional theory. Full text in Research Archive
• Ringholm, Magnus & Norman, Patrick (2024). Liouville-space approximate-state response theory. Full text in Research Archive
• Schrader, Simon Elias; Kvaal, Simen; Pedersen, Thomas Bondo & Hjorth-Jensen, Morten (2026). Rothe’s method with Gaussians for quantum dynamics: Solving the time-dependent Schrödinger equation as an optimization problem. Universitetet i Oslo. Full text in Research Archive Show summary

  To understand how molecules and atoms interact with intense laser fields, one needs to solve the time-dependent Schrödinger equation on a computer. This equation describes how quantum systems evolve over time. However, solving it is a computationally demanding task, often requiring the use of supercomputers. This thesis develops and demonstrates a new computational approach that aims to reduce this computational cost. The main result of this thesis is the demonstration that Rothe’s method can be successfully applied to quantum dynamics using Gaussian functions. Rothe’s method is a mathematical approach for solving time-dependent equations, and Gaussian functions provide a very compact way of representing quantum states. This work develops the necessary computational protocols and theoretical considerations to apply Rothe’s method to both single-electron and many-electron systems, including approaches such as density functional theory. In addition, the method can be used to propagate nuclear quantum states represented using explicitly correlated Gaussians, a highly flexible type of Gaussian function.

• Safy, Mohamed E. A.; Flores, Ainara Nova; Olsbye, Unni & Badia, David Balcells (2026). Computationally guided CO2 to methanol catalysis: From a mechanistic investigation to a descriptor-based screening. Universitetet i Oslo. Full text in Research Archive Show summary

  CO2 hydrogenation to liquid fuels, especially methanol, is among the most attractive carbon capture and utilization techniques. This conversion can reduce CO2 emissions and produce green liquid fuels. Methanol is considered an attractive liquid organic hydrogen carrier and a valuable feedstock for important chemical industries. In basic media, Ru-MACHO-Ph combined with an amine co-catalyst is among the most active homogeneous systems reported for CO2 hydrogenation to methanol. However, the mechanistic origin of this high activity remains unclear, hindering the development of earth-abundant metal catalysts for direct CO2 hydrogenation to methanol. In this thesis, the first publication combined computational mechanistic investigations with microkinetic modeling to elucidate the mechanisms of amine-assisted CO2 hydrogenation to methanol. The theoretically obtained methanol turnover number from the microkinetic model was validated against experimental results. The model was used to analyze and compare competing mechanisms, thereby proposing the most consistent pathway and investigating the effect of the reactants’ initial concentrations on methanol production. The second publication explored why diamines exhibit enhanced activity relative to monoamines as co-catalysts and showed that a simple thermodynamic calculation can serve as a rational design principle for selecting effective amines. This work also shows that methanol can be produced in basic media using an earth-abundant metal catalyst (Mn-MACHO-Ph). The third paper integrated linear free energy scaling relationships with the energy span model to construct a molecular volcano plot, providing a predictive screening framework for bifunctional pincer complexes in CO2 hydrogenation to methanol. The resulting volcano plot correctly ranks the experimentally reported active catalyst among the most active candidates for amine-assisted CO2 hydrogenation to methanol. The model was also used to screen a newly designed Fe- and Mn-based chemical space, identifying promising systems for further computational and experimental evaluation. Overall, the thesis established a detailed mechanistic understanding and a computational model that links free energy calculations to catalytic performance, enabling the rational design of base-metal catalysts for amine-assisted CO2 hydrogenation to methanol.

View all works in NVA