Significance Water is vital to our everyday life, but its structure at a molecular level still not fully understood from either experiment or theory. The latter hampered by inability construct purely predictive, first principles model. difficulty in modeling water lies capturing the delicate interplay among many strong and weak forces that govern behavior phase diagram. Herein, simulations with recently proposed nonempirical quantum mechanical approach (the SCAN density functional) yield an...

TiO₂ is a widely used photocatalyst in science and technology its interface with water important fields ranging from geochemistry to biomedicine.

We introduce a scheme based on machine learning and deep neural networks to model the environmental dependence of electronic polarizability in insulating materials. Application liquid water shows that training network with relatively small number molecular configurations is sufficient predict arbitrary close agreement ab initio density functional theory calculations. In combination representation interatomic potential energy surface, allows us calculate Raman spectra along 2-nanosecond...

We explore the role of long-range interactions in atomistic machine-learning models by analyzing effects on fitting accuracy, isolated cluster properties, and bulk thermodynamic properties. Such have become increasingly popular molecular simulations given their ability to learn highly complex multi-dimensional within a local environment; however, many them fundamentally lack description explicit interactions. In order provide well-defined benchmark system with precisely known pairwise...

The photocatalytic activity of TiO2 for water splitting has been known decades, yet the adsorption structure and hydrogen bonding at interface with have remained controversial. We investigate prototypical aqueous anatase (101) using ab initio molecular dynamics (AIMD) strongly constrained appropriately normed (SCAN) density functional, recently shown to provide an excellent description properties bulk liquid water. find that forms a stable bilayer intact molecules ice-like enhanced dipole...

The interaction of water with TiO 2 surfaces is crucial importance in various scientific fields and applications, from photocatalysis for hydrogen production the photooxidation organic pollutants to self-cleaning bio-medical devices. In particular, equilibrium fraction dissociation at –water interface has a critical role surface chemistry , but difficult determine both experimentally computationally. Among surfaces, rutile (110) special interest as most abundant ’s stable phase. While...

The chemical equilibrium between self-ionized and molecular water dictates the acid-base chemistry in aqueous solutions, yet understanding microscopic mechanisms of self-ionization remains experimentally computationally challenging. Herein, Density Functional Theory (DFT)-based deep neural network (DNN) potentials are combined with enhanced sampling techniques a global collective variable to perform extensive atomistic simulations for model systems increasing size. explicit inclusion...

Lipase from Pseudomonas cepacia was covalently attached to magnetite nanoparticles coated with a thin polydopamine film, and employed in the enzymatic conversion of soybean oil into biodiesel, presence methanol. The proposed strategy explored direct immobilization enzyme via Michael addition aldolic condensation reactions at catechol rings, no need using specific coupling agents. In addition, larger amount enzymes could be bound magnetic nanoparticles, allowing their efficient recycling use...

<italic>Ab initio</italic> molecular dynamics of an aqueous electrode interface reveal the electrostatic, structural, and dynamic effects quantifiable voltage biases on water.

The hydrogen-bond network of confined water is expected to deviate from that the bulk liquid, yet probing these deviations remains a significant challenge. In this work, we combine large-scale molecular dynamics simulations with machine learning potential derived first-principles calculations examine hydrogen bonding in carbon nanotubes (CNTs). We computed and compared infrared spectrum (IR) existing experiments elucidate confinement effects. For CNTs diameters >1.2 nm, find imposes...

Improved understanding of proton transfer in nanopores is critical for a wide range emerging applications, yet experimentally probing mechanisms and energetics this process remains significant challenge. To help reveal details process, we developed applied machine learning potential derived from first-principles calculations to examine water reactivity TiO2 slit-pores. We find that confinement within pores smaller than 0.5 nm imposes strong complex effects on transfer. Although the mechanism...

Understanding the molecular-level structure and dynamics of ice surfaces is crucial for deciphering several chemical, physical, atmospheric processes. Vibrational sum-frequency generation (SFG) spectroscopy most prominent tool probing air–ice interface as it a surface-specific technique, but molecular interpretation SFG spectra challenging. This study utilizes machine-learning potential, along with dipole polarizability models trained on ab initio data, to calculate spectrum interface. At...

Combined modeling and experiments uncover the influence of epoxide-functionalization on hydrogen bonding mobility within poly(ethylenimine) CO 2 sorbents, rationalizing antidegradation benefits conferred by functionalization.

The electronic properties and optical response of ice water are intricately shaped by their molecular structure, including the quantum mechanical nature hydrogen atoms. Despite numerous previous studies, a comprehensive understanding nuclear effects (NQEs) on structure at finite temperatures remains elusive. Here, we utilize simulations that harness efficient machine-learning potentials many-body perturbation theory to assess how NQEs impact bands hexagonal ice. By comparing path-integral...

The electrical double layer (EDL) at metal oxide-electrolyte interfaces critically affects fundamental processes in water splitting, batteries, and corrosion. However, limitations the microscopic-level understanding of EDL have been a major bottleneck controlling these interfacial processes. Herein, we use ab initio-based machine learning potential simulations incorporating long-range electrostatics to unravel molecular-scale picture prototypical anatase TiO2-electrolyte interface under...

As a promising layered electrode material, TiS2-based capacitive deionization (CDI) devices for water desalination have attracted significant attention. However, TiS2/H2O interfacial features, potentially important device optimization, remain unidentified. Using Deep Potential Molecular Dynamics (DPMD), we characterized distinct aqueous interfaces introduced by four TiS2 terminations expected to be present as intercalates into TiS2, namely, Armchair, Zigzag, Zigzag-L, and Zigzag-R. First,...

The electrical double layer (EDL) at aqueous solution-metal oxide interfaces critically affects many fundamental processes in electrochemistry, geology and biology, yet understanding its microscopic structure is challenging for both theory experiments. Here we employ ab initio-based machine learning potentials including long-range electrostatics large-scale atomistic simulations of the EDL TiO2-electrolyte interface. Our provide a molecular-scale picture that demonstrates limitations...

Precise determination of atomic structural information in functional materials holds transformative potential and broad implications for emerging technologies. Spectroscopic techniques, such as X-ray absorption near-edge structure (XANES), have been widely used material characterization; however, extracting chemical from experimental probes remains a significant challenge, particularly disordered materials. We present an integrated approach that combines simulations, data-driven measurements...

Amine-functionalized porous solid materials are effective sorbents for direct air capture (DAC) of CO

Amorphous ${\mathrm{TiO}}_{2}$ (a-${\mathrm{TiO}}_{2}$) is widely used in many fields, ranging from photoelectrochemistry to bioengineering, hence detailed knowledge of its atomic structure scientific and technological interest. Here we use an ab initio-based deep neural network potential (DP) simulate large scale models crystalline disordered with molecular dynamics. Our DP reproduces the structural properties all 11 phases, predicts densities factors molten amorphous only a few percent...