Articles publicats en revistes (Institut de Química Teòrica i Computacional (IQTCUB))

URI permanent per a aquesta col·leccióhttps://hdl.handle.net/2445/121047

Estadístiques

Examinar

Mostrant 1 - 20 de 229

Density Functional Theory (DFT) and Time-Dependent DFT (TDDFT) Studies of Porphyrin Adsorption on Graphene: Insights on the Effect of Substituents and Central Metal on Adsorption Energies
(Wiley, 2024-12-05) Gara, Rayene; Morales García, Ángel; Arfaoui, Youssef; Illas i Riera, Francesc
Combining metalloporphyrins (MPr) and graphene constitutes key composites in the development of photovoltaic devices. Here, we focus on the analysis of the properties of metalloporphyrins/graphene systems by means of the density functional theory (DFT) and its time-dependent (TDDFT) version, focusing on the ground and singlet excited states. Our benchmark analysis concludes that ωB97XD density functional combined with 6-31G(d)/Def2-TZVP basis set is a better-suited method for simulating accurate MPr adsorption on graphene. It is shown that a reduced atomic model where the external organic shell of the structure is removed provides the same resulting optoelectronic properties of the original model, constituting an important speed-up of the calculations when studying porphyrins-derived molecules. We observe that ZnPr provides the highest light harvesting efficiency (LHE) value. In addition, we find out that the adsorption energy increases monotonically with the size of the graphene flake and the highest stability involves the use of graphene comprising above 500 atoms. Besides, CdPr and HgPr keep their properties as photosensitizers when they are bonded to graphene and show promising values in terms of LHE emerging as suitable solar energy harvesters.
Design of donor–acceptor type benzotrithiophene-based covalent organic frameworks for visible-light-driven overall water splitting
(Elsevier Ltd., 2025-12-23) Wang, Chao; Ontiveros Cruz, Diego; Sousa Romero, Carmen
Benefitting from the abundant accessible catalytic sites and well-defined porous architectures enhancing mass transport, two-dimensional (2D) covalent organic frameworks (COFs) are emerging as promising photocatalysts for overall water splitting (OWS). However, the performance of many known COFs for this application remains unsatisfactory, primarily due to stringent requirements for precise band alignment, the limitation posed by overpotentials in hydrogen evolution reaction (HER) and oxygen evolution reaction (OER), and mutual interference between the two half-reactions. Herein, we propose eight donor–acceptor (D–A) type 2D benzotrithiophene-based COFs (BTT-COFs), constructed from experimentally feasible building blocks via Schiffbase condensation reaction. By incorporating D–A pairs into the frameworks, these BTT-COFs exhibit enhanced intermolecular charge transfer characteristics as anticipated, thereby promoting efficient carrier separation during OWS. Concurrently, the D–A combinations enable precise modulation of the electronic structure, affording band gaps ranging from 2.35 to 2.89 eV with band-edge arrangements appropriately aligned for photocatalytic OWS under neutral conditions (pH = 7). Among them, the BTT-COF1 (incorporating benzotrithiophene and 1,3,5-triaminobenzene), BTT-COF2 (featuring benzotrithiophene and 2,4,6-triamino-1,3-diazine), and BTT-COF3 (consisting of benzotrithiophene and 2,4,6-triamino-1,3,5-triazine) are found to be capable of spontaneously driving OWS under their intrinsic photoinduced bias potentials. The remaining BTTCOFs require external bias to facilitate the reaction. Crucially, the theoretical solar-to-hydrogen (STH) conversion efficiencies of these materials range from 1.8 % to 10.0 %, highlighting their potential as efficient photocatalysts for OWS.
Long-Distance Charge Transport between Cytochrome c and Complex III is Mediated by Protons and Reactive Oxygen Species
(Wiley-VCH, 2025-09-12) Lagunas, Anna; Gomila, Alexandre M. J.; Nin Hill, Alba; Guerra-Castellano, Alejandra; Pérez-Mejías, Gonzalo; Samitier i Martí, Josep; Rovira i Virgili, Carme; Rosa, Miguel A. de la; Díaz Moreno, Irene; Gorostiza Langa, Pablo Ignacio
Electron transfer (ET) between redox proteins is an essential process in the respiratory and photosynthetic transport chains. While intra-protein ET is well characterized, the experimental methods to investigate inter-protein ET are limited by the presence of the solvent and by the transient nature of the protein– protein interaction and ET event, which are averaged in protein ensembles. Wiring precisely oriented redox protein partners to the nanoscale electrodes of an electrochemical scanning tunneling microscope allows recording the time- and distance-dependence of the current flowing between them. These methods have revealed that the current flowing between individual protein pairs extends beyond tunneling distances and that it is electrochemically gated. However, the corresponding mechanism and the identity of the charge carriers in aqueous solution remain to be elucidated. To determine the species involved in long-distance charge transport between the redox partner proteins Cc and Cc1 of the respiratory chain, recordings are performed as a function of pH, in heavy water solutions, and in degassed solutions. It is observed that the spatial span and electrochemical gating of long-distance currents are reduced at high pH, in heavy water, and at low oxygen concentration, showing that the currents are assisted by superoxide anions and by protons.
Rational design of organic diradicals with robust high-spin ground state based on antiaromatic linkers
(Royal Society of Chemistry, 2024-11-21) Santiago, Raul; Carvajal Barba, M. Àngels; Poater i Teixidor, Jordi; Moreira, Ibério de Pinho Ribeiro; Bromley, Stefan Thomas; Deumal i Solé, Mercè; Ribas Ariño, Jordi
Fully-organic molecules with high-spin ground states are promising building blocks for new lightweight flexible magnetic materials for emerging technological applications (e.g. spintronics). In this study, we explore the potential of diradicals made of two diphenylmethyl-based open-shell cores covalently linked via different types of pentalene and diazapentalene-based antiaromatic couplers (including dibenzopentalenes and acene-inserted derivatives). Accurate electronic structure calculations have been employed to target non-bonding and non-disjoint frontier molecular orbitals that favor high-spin configurations, leading to the identification of diradicals displaying robust triplet ground states. These candidates exhibit singlet-triplet energy gaps that are up to ten times the thermal energy at room temperature. These substantial gaps emerge from strong interactions between the p-systems of the open-shell centers and the antiaromatic coupler. These interactions not only result in high spin states but are also found to lead to an enhanced stability of the diradicals by drastically dampening their inherent antiaromatic character as compared to the bare couplers, and promoting a high degree of spin density delocalization. These findings highlight the potential of pentalene-based diradicals as building blocks for developing new advanced fully organic magnetic materials.
Gas-Phase Production of Hydroxylated Silicon Oxide Cluster Cations: Structure, Infrared Spectroscopy, and Astronomical Relevance
(American Chemical Society, 2024-06-20) Donato, Andreu A. de; Ghejan, Bianca-Andreea; Bakker, Joost M; Bernhardt, Thorsten M.; Bromley, Stefan Thomas; Lang, Sandra M.
The interaction of free cationic silicon oxide clusters, SixOy+ (x = 2–5, y ≥ x), with dilute water vapor, was investigated in a flow tube reactor. Product mass distributions indicate cluster size-dependent dissociative water adsorption. To probe the structure and vibrational spectra of the resulting SixOyH2+ (x = 2–4) clusters, we employed infrared multiple photon dissociation spectroscopy and density functional theory calculations. The planar rhombic cluster core of the disilicon oxides (x = 2) appears to be retained upon dissociative adsorption of one H2O unit, whereas a significant structural transformation of the tri- and tetra-silicon oxides (x = 3 and 4) is induced, resulting in an increased coordination of the Si atoms and more 3D cluster structures. In an astronomical context, we discuss the potential relevance of SixOyHz+ clusters as seeds for dust nucleation and catalysts for carbon-based chemistry in diffuse or translucent interstellar clouds, where all the necessary conditions for producing these species are found. In the produced clusters, the frequency of the isolated silanol Si–OH stretching vibrational mode is considerably blue-shifted compared to that in hydroxylated bulk silica and small inorganic compounds. This mode has a characteristic frequency range between 1200 cm–1 (8.3 μm) and 1090 cm–1 (9.2 μm) and is associated with the anomalously small Si–OH bond lengths in these ionised species. In infrared observations such high frequency Si–O stretching modes are usually associated with a pure bulk silica component of silicate cosmic dust. The presence of SixOyH2+ clusters in low silica astrophysical environments could thus potentially be detected via their signature Si−O band using the James Webb space telescope.
Stable 1,3,2-Benzodithiazolyl Radicals: Modification ofReactivity, Crystal Packing, and Solid State MagneticProperties by Fluorination
(Wiley-VCH Verlag, 2026-02-26) Buravlev, Alexander A.; Makarov, Alexander Yu.; Ribas Ariño, Jordi; Carvajal Barba, M. Àngels; Deumal i Solé, Mercè; Balmohammadi, Yaser; Grabowsky, Simon; Shundrina, Inna K.; Zakharov, Boris A.; Irtegova, Irina G.; Uvarov, Mikhail N.; Bogomyakov, Artem S; Bagryanskaya, Irina Yu.; Shundrin, Leonid A.; Zibarev, Andrey V.
Impact of fluorination on crystal and molecular structure, heteroatom reactivity, and solid-state magnetic properties of thermally-stable π-radicals is studied experimentally and computationally with 1,3,2-benzodithiazolyl 1· and its 4,7-difluoro, 4,5,6,7-tetra-fluoro, and 4,7-difluoro-5,6-(hexafluoropropane-1,3-diyl) derivatives 2-4, respectively. Radicals 2 -4 are isolated by vacuum thermolysis of their unusual covalent 2:1 adducts with 7,7,8,8-tetracyanoquinodimethane. The impact of fluorination on reactivity is evidenced by transformation of 2-4 and 2 +-4 + into corresponding 2H-1-oxo-1,3,2-benzodithiazoles under the influence of air’s or solvents’ moisture; back transformation into the cations under the action of protic acids; and formation of a paramagnetic molecular complex between 3 and naphthalene, whereas 1 and octafluoronaphthalene do not exhibit complexation. The crystal structures of 3 and 4 reveal a novel packing motif featuring radical pairs linked by four-center interactions that stack into offset π-columns, forming a unique zip-π-stack synthon that incorporates head-over-tail π-pairs of radicals. Despite the formation of π-pairs, polycrystalline 3 and 4 display a nonzero effective magnetic moment that rises with temperature above 200 K, although the values remain significantly lower than those of the high-temperature polymorphs of magnetically-bistable 1 and 2·. This behavior can be rationalized by different magnetic topologies and values of spin exchange between the radicals.
Engineered SnO 2/BiOI fibers via electrospinning for robust visible-light/peroxymonosulfate -driven multipollutant mineralization
(Elsevier B.V., 2026-03-01) Huidobro, Laura; Allés, Miquel; Abid, Mahmoud; Bechelany, Mikhael; Sousa Romero, Carmen; Gómez, Elvira; Serrà i Ramos, Albert
Engineered photocatalysts capable of operating under visible light and realistic water matrices are needed to address emerging pharmaceutical contaminants. Here, we fabricate SnO2/BiOI fibrous heterostructures by electrospinning SnO2 nanofibers decorated with solvothermally synthesized BiOI followed by calcination. The electrospun fibers provide a mechanically robust, high-surface-area scaffold, while BiOI incorporation enhances visible-light absorption and creates SnO2/BiOI heterointerfaces. Textural, optical, and spectroscopic analyses reveal progressive surface decoration, increased surface area, and defect-rich Bi environments as BiOI loading increases. Using tetracycline (TC) as a model contaminant at neutral pH, SnO2/BiOI composites markedly outperform pristine SnO2 under visible light and/or peroxymonosulfate (PMS), with an optimal BiOI content (SBO2) under single-stimulus conditions and near-complete TC mineralization for the highest loading (SBO3) in the PMS + visible-light system. Radical scavenging indicates that SO4•− and •OH are the dominant reactive species, with O2•−, h+ and e− playing secondary roles. A multipollutant mixture (TC, sulfamethoxazole, levofloxacin, lansoprazole) is mineralized by >80% in both Milli-Q and tap water, and SBO3 retains high activity over nine cycles with Bi and I leaching below 0.05% after 48 h. Density functional theory calculations, combined with XPS, support an S-scheme SnO2/BiOI heterojunction, enabling spatial separation of strongly reducing electrons in BiOI and oxidizing holes in SnO2. Although high PMS loadings can partially mask intrinsic catalyst differences, these results outline a practical design platform for heterogeneous (slurry), visible-responsive, PMS-assisted photocatalysts for pharmaceutical-laden effluents.
Formation of a covalent adduct in retaining β-Kdo glycosyl-transferase WbbB via substrate-mediated proton relay
(Wiley-VCH, 2024-07-22) Sagiroglugil, Mert; Liao, Qinghua; Planas, Antoni; Rovira i Virgili, Carme
The GT99 domain of the membrane-anchored WbbB glycosyltransferase (WbbBGT99) catalyzes the transfer of 3-deoxy-D-manno-oct-2-acid (β-Kdo) to an O-antigen saccharide acceptor with retention of stereochemistry. It has been proposed that the enzyme follows an unprecedented double-displacement mechanism involving the formation of covalent adduct between the Kdo sugar and an active site residue (Asp232) that is properly oriented for nucleophilic attack. Here we use QM/MM metadynamics simulations on recently reported crystal structures to provide theoretical evidence for the formation of such adduct and unveil the atomic details of the chemical reaction. Our results support the interpretation made on the basis of X-ray and mass spectrometry analyses. Moreover, we show that the formation of the β-Kdo-Asp232 adduct is assisted by the sugar Kdo-carboxylate group, which mediates the transfer of a proton from Asp232 towards the phosphate leaving group, alleviating electrostatic repulsion between the two negatively charged carboxylate groups. The computed mechanism also explains whyvHis265, previously proposed to act as a general acid, does not impairvcatalysis. This mechanism can be extended to other related enzymes, expanding the repertoire of GT mechanisms in Nature.
Understanding the Chemical Bond in Semiconductor/MXene Composites: TiO2 Clusters Anchored on the Ti2C MXene Surface
(Wiley-VCH, 2024-02-22) García-Romeral, Néstor; Keyhanian, Masoomeh; Morales García, Ángel; Viñes Solana, Francesc; Illas i Riera, Francesc
First-principles calculations on titania clusters (TiO2)n (n=5 and 10) supported on the pristine Ti2C (0001) surface were carried out to understand the properties of semiconductor/MXene composites with implications in (photo)-catalysis. The reported results reveal a high exothermic interaction accompanied by a substantial charge transfer with a concomitant, notorious, deformation of the titania nanoclusters. The analysis of the density of states analysis of the composite systems evidences a metallic character with titania related states crossing the Fermi level. The picture of the chemical bonds is completed by the analysis of X-Ray Photoelectron Spectra (XPS) features, evidencing clear shifts of the C(1s) and O(1s) related peaks relative to the isolated systems that have a quite complex origin. This detailed analysis provides insights to experimentalists interested in the design and synthesis of these systems with possible applications in catalys
Chemical bonding and electronic properties along Group 13 metal oxides
(Springer Verlag, 2024-05-07) Kapse, Samadhan; Voccia, Maria; Viñes Solana, Francesc; Illas i Riera, Francesc
The present work provides a systematic theoretical analysis of the nature of the chemical bond in Al 2 O 3, Ga 2 O 3 ,and In2O3 group 13 cubic crystal structure metal oxides. The influence of the functional in the resulting band gap is assessed. The topological analysis of the electron density provides unambiguous information about the degree of ionicity along the group which is linearly correlated with the band gap values and with the cost of forming a single oxygen vacancy. Overall, this study offers a comprehensive insight into the electronic structure of metal oxides and their interrelations. This will help researchers to harness information effectively, boosting the development of novel metal oxide catalysts or innovative methodologies for their preparation.
Engineered π⋯π interactions favour supramolecular dimers $\mathrm{X@[FeL_{3}]_{2}(X = Cl, Br, I)}$: solid state and solution structure
(Royal Society of Chemistry, 2024-05-28) Risa, Arnau; Barrios Moreno, Leoní Alejandra; Diego, Rosa; Roubeau, Olivier; Aleshin, Dmitry Y.; Nelyubina, Yulia; Novikov, Valentin; Teat, Simon J.; Ribas Ariño, Jordi; Aromí Bedmar, Guillem
Ditopic bis-pyrazolylpyridine ligands usually react with divalent metal ions ($\mathrm{M^{2+}}$) to produce dinuclear triple- stranded helicates $\mathrm{[M_{2}L_{3}]^{4+}}$ or, via $\mathrm{\pi\cdots\pi}$ interactions, dimers of monoatomic complexes $\mathrm{([ML_{3}]_{2})^{4+}}$. The introduction of an additional benzene ring at each end of ligand L increases the number of aromatic contacts within the supramolecular aggregate by 40%, driving the self-recognition process in an irreversible manner. Consequently, the mixing of new bis-pyrazolylquinoline L2 with $\mathrm{FeX_{2}}$ salts leads to crystallization of the tripartite high-spin assemblies $\mathrm{(X@[Fe(L2)_{3}]_{2})^{3+}(X = Cl, Br or I)}$. The aggregates exhibit exceptional stability, as confirmed by a combination of paramagnetic $\mathrm{^{1}H}$ NMR techniques, demonstrating their persistence in solution. Our investigations further reveal that the guests$\mathrm{Br^{-}}$ and $\mathrm{I^{-}}$ are retained inside the associate in solution but $\mathrm{Cl^{-}}$ is immediately released, resulting in the formation of the empty supramolecular dimer $\mathrm{([Fe(L2)_{3}]_{2})^{4+}}$
Family of Quasi-Isotropic $Mn^{II}$ and ${Mn_2}^{II}$ Complexes Exhibiting Slow Relaxation of the Magnetization
(American Chemical Society, 2024-10-16) Pilichos, Evangelos; Font Bardia, Ma. Mercedes; Aullón López, Gabriel; Mayans Ayats, Júlia; Escuer Fité, Albert
Slow relaxation of magnetization has been studied for a family of mononuclear MnII complexes and one ferromagnetic dinuclear system, all of them presenting very weak anisotropy. Complexes with formula $[\{NiL1Mn-(H_{2}O)_{2}(MeOH)\}\{NiL1\}_{2}](ClO_{4})_{2}$ (1), [Mn{NiL1}2](ClO4)2 (2), $[Mn\{NiL2\}_{2}](ClO_{4})_{2}(RR-L2^{2-}, 3RR, SS-L2^{2-}, 3SS)$, $[Mn\{NiL3\}_{2}](ClO_{4})_{2}(RR-L3^{2-}, 4RR, SS-L3^{2-}, 4SS)$ and $(\mu_{1,1}-N_{3})_{2}[Ni_{2}Mn_{2}(L1)_{2}(N_{3})_{2}]$ (5) are derived from compart- mental Schiff bases, in which the NiII environment is square planar and thus diamagnetic. All of the systems have been structurally and magnetically characterized. Zero field splitting (D) values for the MnII cations have been obtained from EPR spectroscopy and NEVPT2 calculations. The slow relaxation of the magnetization for 1−5 has been studied by means of ac magnetometry and rationalized on the basis of their low, but not zero, anisotropy, providing the first example of a polynuclear MnII complex, with S = 5 ground state, exhibiting slow relaxation.
No Switching Cooperativity between Coordinated Azo Ligands on Complexes Having $M^{II}(phosphane-k^2P)\}^{2+}\ ( M = Pd, Pt)$ Scaffolds
(American Chemical Society, 2024-08-21) Raïch Panisello, Ot; Jover Modrego, Jesús; Puigjaner Vallet, Ma. Cristina; Ferrer García, Montserrat; Martínez López, Manuel, 1957-
A series of square-planar palladium and platinum compounds with cis-blocking phosphanes and terminal azobenzene ligands ${[M(dppp)-(azo)_{2}] (OTf)_{2} (azo = CN(C_{6}H_{4})-N = N-(C_{6}H_{4})CN (iso-cyano), CN(C_{6}H_{4})-N = N-(C_{6}H_{5}) (iso-Ph)) and [{M_{2}(tpbz)}(azo)_{4}](OTf)_{4}(azo = CN(C_{6}H_{4})-N = N-(C_{6}H_{5})}$ have been synthesized and fully characterised. Similarly to the uncoordinated ligands, the new coordination compounds have shown to be photochemically active with respect to their trans-to-cis isomerization process. Their cis-to-trans back spontaneous reaction have been studied as a function of solvent, temperature and pressure and the corresponding activation parameters determined in order to investigate the mechanism of these transformations. The results obtained are indicative of the operation of a rotational mechanism with no cooperativity between the azo ligands attached to the same metal. DFT calculations have been carried out in order to estimate the relative energies of the different photoisomers for the theoretical interpretation of the experimental data.
Unlocking the predictive power of quantum-inspired representations for intermolecular properties in machine learning
(Royal Society of Chemistry (RSC), 2024-01-17) Santiago, Raul; Vela Llausí, Sergi; Deumal i Solé, Mercè; Ribas Ariño, Jordi
The quest for accurate and efficient Machine Learning (ML) models to predict complex molecular properties has driven the development of new quantum-inspired representations (QIR). This study introduces MODA (Molecular Orbital Decomposition and Aggregation), a novel QIR-class descriptor with enhanced predictive capabilities. By incorporating wave-function information, MODA is able to capture electronic structure intricacies, providing deeper chemical insight and improving performance in unsupervised and supervised learning tasks. Specially designed to be separable, the multi-moiety regularization technique unlocks the predictive power of MODA for both intra- and intermolecular properties, making it the first QIR-class descriptor capable of such distinction. We demonstrate that MODA shows the best performance for intermolecular magnetic exchange coupling (JAB) predictions among the descriptors tested herein. By offering a versatile solution to address both intra- and intermolecular properties, MODA showcases the potential of quantum-inspired descriptors to improve the predictive capabilities of ML- based methods in computational chemistry and materials discovery.
Multi-Modal Constrastive Learning for Chemical Structure Elucidation with VibraCLIP
(Royal Society of Chemistry (RSC), 2025-11-11) Rocabert Oriols, Pau; Lopez, Noelia; Heras-Domingo, Javier; Lo Conte, Camilla
Identifying molecular structures from vibrational spectra is central to chemical analysis but remains challenging due to spectral ambiguity and the limitations of single-modality methods. While deep learning has advanced various spectroscopic characterization techniques, leveraging the complementary nature of infrared (IR) and Raman spectroscopies remains largely underexplored. We introduce VibraCLIP, a contrastive learning framework that embeds molecular graphs, IR and Raman spectra into a shared latent space. A lightweight fine-tuning protocol ensures generalization from theoretical to experimental datasets. VibraCLIP enables accurate, scalable, and data-efficient molecular identification, linking vibrational spectroscopy with structural interpretation. This tri-modal design captures rich structure–spectra relationships, achieving Top-1 retrieval accuracy of 81.7% and reaching 98.9% Top-25 accuracy with molecular mass integration. By integrating complementary vibrational spectroscopic signals with molecular representations, VibraCLIP provides a practical framework for automated spectral analysis, with potential applications in fields such as synthesis monitoring, drug development, and astrochemical detection.
Benchmarking Periodic Density Functional Theory Calculations for Spin-State Energies in Spin-Crossover Systems
(American Chemical Society, 2024-07-08) Gómez Coca, Silvia; Ruiz Sabín, Eliseo
Spin energetics is one of the biggest challenges associated with energy calculations for electronic structure methods. The energy differences of the spin states in spin-crossover compounds are very small, making them one of the most difficult systems to calculate. Few methods provide accurate results for calculating these energy differences. In addition, studies have usually focused on calculating energetics of single molecules while spin-crossover properties are usually experimentally studied in the solid phase. In this paper, we have used periodic boundary conditions employing methods based on density functional theory to calculate the high- and low-spin energy differences for a test case of twenty extended systems. Compounds with different metals and ligands have been selected, and the results indicate that a semiquantitative description of the energy differences can be obtained with the combination of geometry optimization using the PBE functional including many-body dispersion approach and the use of meta-GGA functionals, as r2SCAN but especially KTBM24, for the energy calculation. Other hybrid functionals, such as TPSSh, gives generally good results, but the calculation of the exact exchange with periodic boundary conditions involves a huge increase in computer time and computational resources. It makes the proposed non-hybrid functional approach (KTBM24//PBE+MB) a great advantage for the study of periodic systems.
Transfer learning based on atomic feature extraction for the prediction of experimental 13C chemical shifts
(Royal Society of Chemistry (RSC), 2024-09-19) Ivković, Žarko; Jover Modrego, Jesús; Harvey, Jeremy
Forecasting experimental chemical shifts of organic compounds is a long-standing challenge in organic chemistry. Recent advances in machine learning (ML) have led to routines that surpass the accuracy of ab initio Density Functional Theory (DFT) in estimating experimental 13C shifts. The extraction of knowledge from other models, known as transfer learning, has demonstrated remarkable improvements, particularly in scenarios with limited data availability. However, the extent to which transfer learning improves predictive accuracy in low-data regimes for experimental chemical shift predictions remains unexplored. This study indicates that atomic features derived from a message passing neural network (MPNN) forcefield are robust descriptors for atomic properties. A dense network utilizing these descriptors to predict 13C shifts achieves a mean absolute error (MAE) of 1.68 ppm. When these features are used as node labels in a simple graph neural network (GNN), the model attains a better MAE of 1.34 ppm. On the other hand, embeddings from a self-supervised pre-trained 3D aware transformer are not sufficiently descriptive for a feedforward model but show reasonable accuracy within the GNN framework, achieving an MAE of 1.51 ppm. Under low-data conditions, all transfer-learned models show a significant improvement in predictive accuracy compared to existing literature models, regardless of the sampling strategy used to select from the pool of unlabeled examples. We demonstrated that extracting atomic features from models trained on large and diverse datasets is an effective transfer learning strategy for predicting NMR chemical shifts, achieving results on par with existing literature models. This method provides several benefits, such as reduced training times, simpler models with fewer trainable parameters, and strong performance in low-data scenarios, without the need for costly ab initio data of the target property. This technique can be applied to other chemical tasks opening many new potential applications where the amount of data is a limiting factor.
Limitations of free energy diagrams to predict the catalytic activity: the reverse water gas shift reaction catalyzed by Ni/TiC
(Academic Press, Elsevier, 2023) Lozano-Reis, Pablo; Prats Garcia, Hèctor; Sayós Ortega, Ramón; Illas i Riera, Francesc
The temporal evolution at the catalyst surface is a result of an intricate interplay between all involved microscopic events such as adsorption, desorption, diffusion, and bond breaking/formation steps, and the interaction with the surrounding environment. By properly including these effects, kinetic Monte Carlo (kMC) simulations can accurately describe the complexity of real catalysts, unravel the dominant reaction mechanisms and provide fundamental understanding towards the rational design of novel catalysts. In this work, we combine density functional theory (DFT) calculations, statistical thermodynamics and kMC simulations to study the reverse water–gas shift (RWGS) reaction on Ni/TiC, a bifunctional catalyst. The predictions from DFT energy profiles do not coincide with the outcome of the kMC simulations, evidencing the limitations of the former, especially in including the effect of coverage of surface species, which plays a crucial role. The kMC simulations results are in remarkable agreement with the experimental data, proving that the kMC simulations are able to describe the complex chemistry of the RWGS reaction on a bifunctional catalyst.
Interfacial behavior of binary, ternary and quaternary oil/water mixtures described from molecular dynamics simulations
(Elsevier B.V., 2021-02-15) Alonso Benito, Gerard; Gamallo Belmonte, Pablo; Rincón, Cristina; Sayós Ortega, Ramón
The correct description of crude oil/water interfaces is a very complex and an important task, particularly to the oil industry, whosemain difficulty relies on understanding how the interfacial properties (i.e., interfacial tension and interfacial accumulation) of the systemare affected by a very large number of components. To give some additional insight to the oil/water interfacial behavior, eleven oil/water mixtures (i.e., six binary, four ternary and a quaternary mixture) have been modeled through atomistic molecular dynamics simulations at laboratory conditions. All mixtures were built with a model oil based on dodecane, toluene, quinoline and a naphthenic acid, to represent the saturated, aromatic, basic resin and acid resin fractions, respectively. The results from this contribution show that interfacial tensions can be correlated to interfacial accumulation, which can be used as good starting point in predicting interfacial properties of oil mixtures. Additionally, the interfacial properties of mixtures behave similarly to the most polar pure oil/water interface, while all other compounds stay in the oil bulk as spectators. This behavior raises the question of whether using common n-alkane oils is a good enough approximation for modeling the interfacial properties of crude oils.
Time-optimal control of a solid-state spin amidst dynamical quantum wind
(Springer Nature, 2024-11-05) Albareda, Guillermo; Guo, Guang-Can; Sun, Fang-Wen; Dong, Yang; Jiang, Wang; Gao, Xue-Dong; Yu, Cui; Liu, Yong; Zhang, Shao-Chun; Chen, Xiang-Dong; Moreira, Ibério de Pinho Ribeiro; Bofill i Villà, Josep M.; Sentis, Gael; Ramos, Ramon
Time-optimal control holds promise across the full spectrum of quantum technologies, where the rapid generation of unitary gates and state transformations is crucial to mitigate decoherence effects. In practical scenarios, quantum systems are always immersed in an external time-dependent field or potential, either owing to the inevitable influence of the environment or as a sought-after effect for enhanced coherence. The challenge then lies in finding the time-optimal approach to navigate quantum systems amidst dynamical ambient Hamiltonians, a pursuit that has proven elusive thus far. We showcase the implementation of arbitrary quantum state transformations and a universal set of single-qubit gates under a background Landau-Zener Hamiltonian. Leveraging the favorable coherence properties of timedomain Rabi oscillations, we achieve velocities surpassing the Mandelstam-Tamm quantum speed limit and significantly lower energy costs than those incurred by conventional quantum control techniques. These findings highlight a promising pathway to expedite and economize high-fidelity quantum operations.

Examinar

Enviaments recents