Matheus Máximo-Canadas

Accurate reference data for electronically excited states are essential for assessing multireference electronic-structure methods. We benchmark state-averaged completeactive-space self-consistent field (SA-CASSCF), exchange-scaled CASSCF (XS-CASSCF), fully uncontracted multireference configuration interaction with single and double excitations (MR-CISD), and four a posteriori size-consistency corrections against the theoretical best estimates of QUESTDB. The dataset comprises 262 valence excitations of 41 small and medium-sized molecules, including singlet and triplet n*, *, and doubly excited states. All calculations used a consistent COLUMBUS protocol with molecule-specific active spaces and the aug-cc-pVDZ basis set. For the 232 safe transitions, SA-CASSCF and MR-CISD yield mean absolute errors of 0.57 and 0.31 eV, respectively, with predominantly positive biases. The Davidson–Silver/Siegbahn (+DV3) and Pople (+P) corrections reduce the mean absolute errors to 0.10 and 0.11 eV and the root-mean-square errors to 0.13 eV. They also provide a more balanced description across spin multiplicities and excitation characters. Both XSCASSCF improve singlet * states, particularly those with pronounced ionic character, but degrade n* states and do not improve the overall statistics relative to SA-CASSCF. For 36 selected unsafe or doubly excited transitions, MR-CISD+DV3 and MR-CISD+P retain mean absolute errors of 0.16 eV. Their agreement with exFCI-based reference values and discrepancies from several CASPT3- and PC-NEVPT2-based estimates identify states whose current best estimates merit independent re-examination. Overall, size-consistencycorrected MR-CISD provides highly accurate and broadly balanced valence excitation energies.

Understanding aromaticity in excited states remains a fundamental challenge in chemistry. Baird aromaticity describes the stabilization of a molecule's lowest triplet excited state due to cyclic conjugation, similar to how Huckel's rule applies to the ground state. In this work, we employ a new set of descriptors, derived from the Distributed Multipole Analysis (DMA) method, to systematically analyze the electronic delocalization and triplet aromaticity of cyclooctatetraene ( COT ) and its BN/CC isosteres in their lowest triplet states ( T ₁ ). The descriptors are built from the components of the electric quadrupole tensor Q ₂ , the first in the DMA expansion to include out‐of‐plane electron contributions. The Q ₂ ‐based aromaticity descriptors allow for an accurate quantification of the delocalized electron density, thereby providing global and local measures of π‐delocalization. Comparison with magnetic ( NICS ), geometric ( HOMA ), and electronic ( MCI ) descriptors reveals a good agreement, confirming that the new descriptors efficiently capture changes in electronic delocalization in rings containing BN substitutions. The analysis also shows that the connectivity between the B and N atoms, rather than the number of heteroatoms, governs the degree of Baird aromaticity: adjacent BN pairs reduce π‐delocalization, whereas alternating or opposite arrangements enhance it. These results indicate that the Q ₂ ‐based descriptors provide an alternative framework for exploring the multidimensional nature of triplet‐state aromaticity.

Accurately predicting thermophysical properties across various physical states is essential for both industrial and scientific applications. However, experimental data often exhibit variability and noise, requiring robust modeling approaches. In this work, we employ machine learning (ML) techniques to predict methane's thermophysical properties in liquid, vapor, and supercritical phases, including isobaric and isochoric heat capacities, density, specific volume, Joule-Thomson coefficients, enthalpies, sound speed, and shear viscosities, applying an approach recently developed [ACS Eng. Au 5, 226 (2025)10.1021/acsengineeringau.5c00001]. We explore various ML algorithms, including Adaptive Boosting, Bagging, Decision Trees, Extra Trees, Gradient Boosting, Histogram-based Gradient Boosting Regression Tree, K-Nearest Neighbors, Light Gradient Boosting Machine, Nu-Support Vector Regression, Random Forest, Extreme Gradient Boosting, and Artificial Neural Networks. The ML models, which use raw experimental data without removing outliers, produce predictions that are closer to those obtained from the equations of state (EOS) employed by the National Institute of Standards and Technology than to the same raw experimental data used as input for developing the EOS. These results highlight ML's potential to identify and generalize complex patterns, smooth inherent noise, and manage the variability of different thermophysical properties. They indicate that ML models, particularly Extra Trees and Gradient Boosting, can offer a scalable alternative for thermophysical property predictions, offering consistency and efficiency over traditional methods. Although our approach does not eliminate preprocessing, it demonstrates that ML can effectively manage noisy data independently, offering a more efficient and cost-effective alternative to conventional pre- and postprocessing techniques.

The efficient detection of nerve agents is paramount in civilian and war contexts. In this work, we investigated theoretically the mechanisms of fluorescence quenching and charge transfer (CT) in two recently synthesized small molecule fluorescent probes for detecting nerve agents, NMU-1 and 10-hydroxybenzo-[h]-quinoline (HBQ-AE), which are based on the pyridine group as the identifying unit. These fluorescent molecules change their emission pattern upon binding to an organophosphorus compound. We employed density functional theory (DFT), implicit methods for describing the water solvent, and a path integral approach to calculate fluorescence rates and emission spectra from first-principles for analyzing the interaction between the nerve agent simulant diethyl chlorophosphite (DCP) and NMU-1 or HBQ-AE. Upon interacting with DCP, both HBQ-AE and NMU-1 experience strong fluorescence quenching, with intensity reductions of ∼99.65 and ∼99.95%, respectively, due to enhanced CT induced by DCP's electron-accepting nature and formation of a positively charged nitrogen within the probes. Although the measured fluorescence enhancement in HBQ-AE/DCP systems was attributed to the HBQ-DCP product, our results show a different picture: the true fluorescent species is the hydrolysis product HBQ + H, whose calculated emission differs by a blue shift of only 0.23 eV from the experimental data, rather than the HBQ-DCP product as previously thought. For acetylcholinesterase (AChE) detection, fluorescence is confirmed to originate from the HBQ-Keto, with a blue shift deviation of 0.12 eV. The NMU-1 calculations also confirm the experiments, showing a 0.24 eV deviation, thus confirming its quenching-based detection of DCP. Therefore, HBQ-AE is a fluorescence turn-on sensor via its emissive hydrolysis product HBQ+H. In contrast, NMU-1 is a turn-off sensor, as both NMU-DCP and NMU+H are essentially nonemissive. Our theoretical approach is general, accurate, and can be applied to different problems involving fluorescent probes and binding agents.

We investigated the impact of fluorination on the aromaticity of 12 benzene‐based compounds using a recently proposed new set of six descriptors (ACS Omega 2025, 10, 14, 14157–1417) based on the distributed multipole analysis (DMA) partition of the electron density. These descriptors allow the quantification of electron delocalization, an observable property related to aromaticity, and were defined from the components of the DMA quadrupole moment tensor , the first electric multipole moment with out‐of‐plane contributions. We found a decrease in aromaticity that is related to the electron‐withdrawing inductive effects of fluorine substituents, which modify the electron density distribution on the ring, particularly in regions adjacent to the substituents. The analysis by the ‐based aromaticity descriptors revealed a strong correlation between electron distribution and electronic delocalization in fluorinated systems, confirming that fluorine substituents directly affect the symmetry and uniformity of electronic delocalization. A linear relationship between the descriptors and the and ring current strength () descriptors was found, suggesting that these descriptors capture similar aspects of aromaticity. These findings highlight the effectiveness and robustness of the new aromaticity descriptors for rationalizing the aromaticity of fluorinated compounds.

Aromatic compounds are energetically stable and have low reactivity, but an aromaticity concept is difficult to establish because it cannot be measured. For this reason, different descriptors have been developed to rationalize and quantify this phenomenon. Given that cyclic electron delocalization is an essential property of aromatic compounds, in this work, we propose six new aromatic descriptors based on Stone's distributed multipole analysis (DMA) to partition the molecular electron density into electric multipoles localized on different sites of a molecule. The new aromatic descriptors are based on different components of the DMA quadrupole electric moment tensor Q ₂ , the first term of the DMA multipole expansion having contributions from the out-of-plane electron density. The proposed descriptors are straightforward to obtain because the DMA method is implemented on different popular electronic structure packages. For users of Gaussian, the formatted checkpoint file with the calculated molecular electron density is used as input for the GDMA2 program of Stone to compute the necessary Q ₂ components. A Python script is provided to calculate the proposed descriptors. The computer protocol for determining the Q ₂ -based descriptors in either way is presented. To assess the performance of the aromaticity descriptors, we used 12 tests of the Girona benchmark developed by the Solà group involving different distortions of benzene, substitutions, complexation, ring size dependence, atom size dependence, heteroatomic species, Clar systems, and fulvenes. The descriptors normalized by the corresponding benzene values presented the most consistent results. The correct aromaticity trends of the six normalized new indices were predicted entirely in 65% of the cases; for 22%, most trends were predicted and only failed utterly in 13%. The failed cases were related to the molecular symmetry, which led to the partial cancellation of the Q ₂ tensor components, thereby affecting the aromatic descriptors and, in some cases, due to a contamination of sigma electrons. Our proposal joins others to contribute to understanding the important and complex chemical concept of aromaticity.

Context Nitrocellulose, widely used in energetic materials, is prone to thermal and chemical degradation, compromising safety and performance. Stabilizers are molecules used in the composition of nitrocellulose-based propellants to inhibit the autocatalytic degradation process that produces nitrous gases and free nitric acids. Curcumin, (1E,6E)-1,7-bis(4-hydroxy-3-methoxyphenyl)-1,6-heptadiene-3,5-dione, known for its antioxidant properties and a potential green stabilizer, was investigated using Density Functional Theory (DFT) focusing on its interaction with nitrogen dioxide. Two mechanisms were analyzed: aromatic ring nitration and free radical formation. The results indicate that nitration of the aromatic ring of curcumin and the formation of a curcumin-based free radical are viable. The computed Gibbs free activation energy (∆^‡G°) and the activation enthalpy (∆^‡H°) for two different temperatures, 298.15 K (room temperature) and 363.15 K (typical temperature in aging tests), are respectively 43.64 kcal/mol and 44.78 kcal/mol for the first reaction, and 31.54 kcal/mol and 35.31 kcal/mol for the second. The radical-based mechanism favors improved kinetics. These findings demonstrate curcumin’s potential as an effective stabilizer, providing comparable performance to traditional compounds with lower environmental impact. Methods DFT calculations were carried out using Gaussian 09 and Orca 5.0.1 packages. The ωB97M-V, B3LYP, and M062X functionals were employed with the 6–311 + G(d) and 6-311G(d) basis sets. Solvent effects were modeled using the Conductor-like Polarizable Continuum Model (CPCM) and Solvation Model based on Density (SMD) continuum solvent models. Thermochemical data were computed using the same levels of calculation.

Theoretical and computational chemistry (TCC) is a set of theories and models that, over the years, were refined to the point that it is possible to determine measurable quantities with precision, predict experimental results, and provide fundamental insights into chemical phenomena and mechanisms that may be difficult or impossible to observe experimentally. Machine Learning (ML), on the other hand, is a subfield of Artificial Intelligence (AI) that applies different types of statistical methods to a large volume of data or a smaller volume of precise data, combined with high computational power, enabling the discovery of complex patterns and production of explanations inaccessible through human deductive reasoning and intuition alone or using traditional scientific methods. Recently, ML, combined or not with TCC methods, has emerged as a transformative force, bringing significant advances in chemistry and materials science. This review surveys basic ML concepts and their applications in chemistry, focusing on supervised and unsupervised learning approaches, data preprocessing, and model development workflows, exploring the most relevant ML algorithms selected for their specific usefulness in chemical applications. Integrating ML with traditional computational chemistry methods, such as density functional theory, is highlighted as a powerful synergy for accelerating materials discovery and design. Key areas of impact discussed include High-Throughput Virtual Screening (HTVS) of molecules and materials, spectroscopy (including UV-Vis and fluorescence), organic electronics (such as solar cells and organic light-emitting diodes), potential energy surfaces, and molecular dynamics. It also addresses critical aspects of ML in chemistry, including data representation, model interpretability through explainability techniques, and the emerging role of large language models (LLM). Tips for acquiring knowledge of ML for practical applications in chemistry are given.

Intramolecular charge transfer (ICT) effects of para‐nitroaniline (pNA) in eight solvents (cyclohexane, toluene, acetic acid, dichloroethane, acetone, acetonitrile, dimethylsulfoxide, and water) are investigated extensively. The second‐order algebraic diagrammatic construction, ADC(2), ab initio wave function is employed with the COSMO implicit and discrete multiscale solvation methods. We found a decreasing amine group torsion angle with increased solvent polarity and a linear correlation between the polarity and ADC(2) transition energies. The first absorption band involves π → π* transitions with ICT from the amine and the benzene ring to the nitro group, increased by 4%–11% for different solvation models of water compared to the vacuum. A second band of pNA is characterized for the first time. This band is primarily a local excitation on the nitro group, including some ICT from the amine group to the benzene ring that decreases with the solvent polarity. For cyclohexane, the COSMO implicit solvent model shows the best agreement with the experiment, while the explicit model has the best agreement for water.