The Carbon Dioxide−Water Interface at Conditions of Gas Hydrate Formation

Neural network analysis of neutron and X-ray reflectivity data incorporating prior knowledge

Due to the ambiguity related to the lack of phase information, determining the physical parameters of multilayer thin films from measured neutron and X-ray reflectivity curves is, on a fundamental level, an underdetermined inverse problem. This ambiguity poses limitations on standard neural networks, constraining the range and number of considered parameters in previous machine learning solutions. To overcome this challenge, a novel training procedure has been designed which incorporates dynamic prior boundaries for each physical parameter as additional inputs to the neural network. In this manner, the neural network can be trained simultaneously on all well-posed subintervals of a larger parameter space in which the inverse problem is underdetermined. During inference, users can flexibly input their own prior knowledge about the physical system to constrain the neural network prediction to distinct target subintervals in the parameter space. The effectiveness of the method is demonstrated in various scenarios, including multilayer structures with a box model parameterization and a physics-inspired special parameterization of the scattering length density profile for a multilayer structure. In contrast to previous methods, this approach scales favourably when increasing the complexity of the inverse problem, working properly even for a five-layer multilayer model and a periodic multilayer model with up to 17 open parameters.

Probing the critical nucleus size in tetrahydrofuran clathrate hydrate formation using surface-anchored nanoparticles

Controlling the formation of clathrate hydrates is crucial for advancing hydrate-based technologies. However, the microscopic mechanism underlying clathrate hydrate formation through nucleation remains poorly elucidated. Specifically, the critical nucleus, theorized as a pivotal step in nucleation, lacks empirical validation. Here, we report uniform nanoparticles, e.g., graphene oxide (GO) nanosheets and gold or silver nanocubes with controlled sizes, as nanoprobes to experimentally measure the size of the critical nucleus of tetrahydrofuran (THF) clathrate hydrate formation. The capability of the nanoparticles in facilitating THF clathrate hydrate nucleation displays generally an abrupt change at a nanoparticle-size-determined specific supercooling. It is revealed that the free-energy barrier shows an abrupt change when the nanoparticles have an approximately the same size as that of the critical nucleus. Thus, it is inferred that THF clathrate hydrate nucleation involves the creation of a critical nucleus with its size being inversely proportional to the supercooling. By proving the existence and determining the supercooling-dependent size of the critical nucleus of the THF clathrate hydrates, this work brings insights in the microscopic pictures of the clathrate hydrate nucleation.

Stochastic Cellular Automata Modeling of CO2 Hydrate Growth and Morphology

Carbon dioxide (CO₂) hydrates are important in a diverse range of applications and technologies in the environmental and energy fields. The development of such technologies relies on fundamental understanding, which necessitates not only experimental but also computational studies of the growth behavior of CO₂ hydrates and the factors affecting their crystal morphology. As experimental observations show that the morphology of CO₂ hydrate particles differs depending on growth conditions, a detailed understanding of the relation between the hydrate structure and growth conditions would be helpful. To this end, this work adopts a modeling approach based on hybrid probabilistic cellular automata to investigate variations in CO₂ hydrate crystal morphology during hydrate growth from stagnant liquid water presaturated with CO₂. The model, which uses free energy density profiles as inputs, correlates the variations in growth morphology to the system subcooling ΔT, i.e., the temperature deficiency from the triple CO₂-hydrate-water equilibrium temperature under a given pressure, and properties of the growing hydrate-water interface, such as surface tension and curvature. The model predicts that when ΔT is large, parabolic needle-like or dendrite crystals emerge from planar fronts that deform and lose stability. In agreement with chemical diffusion-limited growth, the position of such planar fronts versus time follows a power law. In contrast, the tips of the emerging parabolic crystals steadily grow in proportion to time. The modeling framework is computationally fast and produces complex growth morphology phenomena under diffusion-controlled growth from simple, easy-to-implement rules, opening the way for employing it in multiscale modeling of gas hydrates.

Molecular Dynamics Study of Clathrate-like Ordering of Water in Supersaturated Methane Solution at Low Pressure

Using molecular dynamics, the evolution of a metastable solution for “methane + water” was studied for concentrations of 3.36, 6.5, 9.45, 12.2, and 14.8 mol% methane at 270 K and 1 bar during 100 ns. We have found the intriguing behavior of the system containing over 10,000 water molecules: the formation of hydrate-like structures is observed at 6.5 and 9.45 mol% concentrations throughout the entire solution volume. This formation of “blobs” and the following amorphous hydrate were studied. The creation of a metastable methane solution through supersaturation is the key to triggering the collective process of hydrate formation under low pressure. Even the first stage (0–1 ns), before the first fluctuating cavities appear, is a collective process of H-bond network reorganization. The formation of fluctuation cavities appears before steady hydrate growth begins and is associated with a preceding uniform increase in the water molecule’s tetrahedrality. Later, the constantly presented hydrate cavities become the foundation for a few independent hydrate nucleation centers, this evolution is consistent with the labile cluster and local structure hypotheses. This new mechanism of hydrogen-bond network reorganization depends on the entropy of the cavity arrangement of the guest molecules in the hydrate lattice and leads to hydrate growth.

Neural network analysis of neutron and X-ray reflectivity data: automated analysis using mlreflect, experimental errors and feature engineering

A Python-based analysis pipeline for the fast analysis of X-ray and neutron reflectivity data using neural networks is presented.

The Python package mlreflect is demonstrated, which implements an optimized pipeline for the automated analysis of reflectometry data using machine learning. The package combines several training and data treatment techniques discussed in previous publications. The predictions made by the neural network are accurate and robust enough to serve as good starting parameters for an optional subsequent least-mean-squares (LMS) fit of the data. For a large data set of 242 reflectivity curves of various thin films on silicon substrates, the pipeline reliably finds an LMS minimum very close to a fit produced by a human researcher with the application of physical knowledge and carefully chosen boundary conditions. The differences between simulated and experimental data and their implications for the training and performance of neural networks are discussed. The experimental test set is used to determine the optimal noise level during training. The extremely fast prediction times of the neural network are leveraged to compensate for systematic errors by sampling slight variations in the data.

Neural network analysis of neutron and x-ray reflectivity data: pathological cases, performance and perspectives

Neutron and x-ray reflectometry (NR and XRR) are powerful techniques to investigate the structural, morphological and even magnetic properties of solid and liquid thin films. While neutrons and x-rays behave similarly in many ways and can be described by the same general theory, they fundamentally differ in certain specific aspects. These aspects can be exploited to investigate different properties of a system, depending on which particular questions need to be answered. Having demonstrated the general applicability of neural networks to analyze XRR and NR data before (Greco et al 2019 J. Appl. Cryst. 52 1342), this study discusses challenges arising from certain pathological cases as well as performance issues and perspectives. These cases include a low signal-to-noise ratio, a high background signal (e.g. from incoherent scattering), as well as a potential lack of a total reflection edge (TRE). By dynamically modifying the training data after every mini batch, a fully-connected neural network was trained to determine thin film parameters from reflectivity curves. We show that noise and background intensity pose no significant problem as long as they do not affect the TRE. However, for curves without strong features the prediction accuracy is diminished. Furthermore, we compare the prediction accuracy for different scattering length density combinations. The results are demonstrated using simulated data of a single-layer system while also discussing challenges for multi-component systems.

Water Wettability Coupled with Film Growth on Realistic Cyclopentane Hydrate Surfaces

Although the wettability of hydrate surfaces and hydrate film growth are key to understanding hydrate agglomeration and pipeline plugging, a quantitative understanding of the coupled behavior between both phenomena is lacking. In situ measurements of wettability coupled with film growth were performed for cyclopentane hydrate surfaces in cyclopentane at atmospheric pressure and temperatures between 1.5-6.8 °C. Results were obtained as a function of annealing (conversion) time and subcooling. Hydrate surface wettability decreased as annealing time increased, while hydrate film growth rate was unaffected by annealing time at any subcooling. The results are interpreted as a manifestation of the hydrate surface porosity, which depends on annealing time and controls water spreading on the hydrate surface. The wettability generally decreased as the subcooling increased because higher subcooling yields rougher hydrate surfaces, making it harder for water to spread. However, this effect is balanced by hydrate growth rates, which increase with subcooling. Also affecting the results, surface heating from heat release (from exothermic crystallization) allows excess surface water to promote spreading. The hydrate film growth rate on water droplets increased with subcooling, as expected from a higher driving force. At any subcooling, the instantaneous hydrate growth rate decreased over time, likely from heat transfer limitations. A new phenomenon was observed, where the angle at the three-phase point increases from the initial contact angle upon hydrate film growth, named the crystallization angle. This is attributed to the water droplet trying to spread while the thin film is weak enough to be redirected. Once the hydrate film grows and forms a "wall" around the droplet, it cannot be moved, and further growth yields a crater on the droplet surface, attributed to water penetrating the hydrate surface pore structures. This fundamental behavior has many flow assurance implications since it affects the interactions between the agglomerating hydrate particles and water droplets.

Understanding the Pathway of Gas Hydrate Formation with Porous Materials for Enhanced Gas Separation

The reason that the stoichiometry of gas to water in artificial gas hydrates formed on porous materials is much higher than that in nature is still ambiguous. Fortunately, based on our experimental thermodynamic and kinetic study on the gas hydrate formation behavior with classic ordered mesoporous carbon CMK-3 and irregular porous activated carbon combined with density functional theory calculations, we discover a microscopic pathway of the gas hydrate formation on porous materials. Two interesting processes including (I) the replacement of water adsorbed on the carbon surface by gas and (II) further replacement of water in the pore by gas accompanied with the gas condensation in the pore and growth of gas hydrate crystals out of the pore were deduced. As a result, a great enhancement of the selectivity and regeneration for gas separation was achieved by controlling the gas hydrate formation behavior accurately.

Generalized skew-symmetric interfacial probability distribution in reflectivity and small-angle scattering analysis

Generalized skew-symmetric probability density functions are proposed to model asymmetric interfacial density distributions for the parameterization of any arbitrary density profiles in the `effective-density model'. The penetration of the densities into adjacent layers can be selectively controlled and parameterized. A continuous density profile is generated and discretized into many independent slices of very thin thickness with constant density values and sharp interfaces. The discretized profile can be used to calculate reflectivitiesviaParratt's recursive formula, or small-angle scatteringviathe concentric onion model that is also developed in this work.

What are the key factors governing the nucleation of CO2 hydrate?

Microsecond molecular dynamics simulations were performed to provide molecular insights into the nucleation of CO₂ hydrate. The adsorption of sufficient CO₂ molecules around CO₂ hydration shells is revealed to be crucial to effectively stabilize the hydrogen bonds formed therein, catalyzing the hydration shells into hydrate cages and inducing the nucleation. Moreover, a high aqueous CO₂ concentration is found to be another key factor governing the nucleation of CO₂ hydrate, and only above a critical concentration can the nucleation of CO₂ hydrate occur. The 4¹5¹⁰6² cages, with size similar to the CO₂ hydration shell and an elliptical space closely matching a linear CO₂ molecule, play a dominant role in initiating the nucleation and remain the most abundant. The incipient CO₂ hydrate is rather amorphous due to the abundance of metastable cages (mostly 4¹5¹⁰6²).

Highly porous CO2 hydrate generation aided by silica nanoparticles for potential secure storage of CO2 and desalination

We report a new way of storing CO₂ in a highly porous hydrate structure, stabilized by silica nanoparticles (NPs).

Direct phase coexistence molecular dynamics study of the phase equilibria of the ternary methane–carbon dioxide–water hydrate system

Grown mixed carbon dioxide–methane hydrate with molecular dynamics.

How Properties of Solid Surfaces Modulate the Nucleation of Gas Hydrate

Molecular dynamics simulations were performed for CO2 dissolved in water near silica surfaces to investigate how the hydrophilicity and crystallinity of solid surfaces modulate the local structure of adjacent molecules and the nucleation of CO2 hydrates. Our simulations reveal that the hydrophilicity of solid surfaces can change the local structure of water molecules and gas distribution near liquid-solid interfaces, and thus alter the mechanism and dynamics of gas hydrate nucleation. Interestingly, we find that hydrate nucleation tends to occur more easily on relatively less hydrophilic surfaces. Different from surface hydrophilicity, surface crystallinity shows a weak effect on the local structure of adjacent water molecules and on gas hydrate nucleation. At the initial stage of gas hydrate growth, however, the structuring of molecules induced by crystalline surfaces are more ordered than that induced by amorphous solid surfaces.

New observations and insights into the morphology and growth kinetics of hydrate films

The kinetics of film growth of hydrates of methane, ethane, and methane-ethane mixtures were studied by exposing a single gas bubble to water. The morphologies, lateral growth rates, and thicknesses of the hydrate films were measured for various gas compositions and degrees of subcooling. A variety of hydrate film textures was revealed. The kinetics of two-dimensional film growth was inferred from the lateral growth rate and initial thickness of the hydrate film. A clear relationship between the morphology and film growth kinetics was observed. The shape of the hydrate crystals was found to favour heat or mass transfer and favour further growth of the hydrate film. The quantitative results on the kinetics of film growth showed that for a given degree of subcooling, the initial film thicknesses of the double hydrates were larger than that of pure methane or ethane hydrate, whereas the thickest hydrate film and the lowest lateral growth rate occurred when the methane mole fraction was approximately 0.6.

X-ray reflectivity measurements of liquid/solid interfaces under high hydrostatic pressure conditions

A high-pressure cell for in situ X-ray reflectivity measurements of liquid/solid interfaces at hydrostatic pressures up to 500 MPa (5 kbar), a pressure regime that is particularly important for the study of protein unfolding, is presented. The original set-up of this hydrostatic high-pressure cell is discussed and its unique properties are demonstrated by the investigation of pressure-induced adsorption of the protein lysozyme onto hydrophobic silicon wafers. The presented results emphasize the enormous potential of X-ray reflectivity studies under high hydrostatic pressure conditions for the in situ investigation of adsorption phenomena in biological systems.

Nucleation of gas hydrates within constant energy systems

The early stage of formation of gas hydrates has recently attracted attention as amorphous intermediate gas hydrate structures have been observed, apparently contrary to a classical model of nucleation and some experimental observations. To date, essentially all reported molecular simulations of the nucleation of gas hydrates have been under constant temperature conditions, which does not consider the possible impacts of heat transfer on the nucleation processes. Here we show, using constant energy molecular simulations, that the nuclei at an early stage of the hydrate formation have relatively more crystalline order in comparison with those observed in previous isothermal (NPT or NVT) work. The current work suggests a more transient role for intermediate amorphous structures during hydrate nucleation, thereby providing a stronger link between molecular simulation and experimental observations. Our NVE results nevertheless support the two-step nucleation mechanism proposed in previous simulation studies under constant temperature conditions which features the initial formation of amorphous hydrate-like structures.

Nucleation of the CO2 hydrate from three-phase contact lines

Using molecular dynamics simulations on the microsecond time scale, we investigate the nucleation and growth mechanisms of CO(2) hydrates in a water/CO(2)/silica three-phase system. Our simulation results indicate that the CO(2) hydrate nucleates near the three-phase contact line rather than at the two-phase interfaces and then grows along the contact line to form an amorphous crystal. In the nucleation stage, the hydroxylated silica surface can be understand as a stabilizer to prolong the lifetime of adsorbed hydrate cages that interact with the silica surface by hydrogen bonding, and the adsorbed cages behave as the nucleation sites for the formation of an amorphous CO(2) hydrate. After nucleation, the nucleus grows along the three-phase contact line and prefers to develop toward the CO(2) phase as a result of the hydrophilic nature of the modified solid surface and the easy availability of CO(2) molecules. During the growth process, the population of sI cages in the formed amorphous crystal is found to increase much faster than that of sII cages, being in agreement with the fact that only the sI hydrate can be formed in nature for CO(2) molecules.

Fundamentals and applications of gas hydrates

Fundamental understanding of gas hydrate formation and decomposition processes is critical in many energy and environmental areas and has special importance in flow assurance for the oil and gas industry. These areas represent the core of gas hydrate applications, which, albeit widely studied, are still developing as growing fields of research. Discovering the molecular pathways and chemical and physical concepts underlying gas hydrate formation potentially can lead us beyond flowline blockage prevention strategies toward advancing new technological solutions for fuel storage and transportation, safely producing a new energy resource from natural deposits of gas hydrates in oceanic and arctic sediments, and potentially facilitating effective desalination of seawater. The state of the art in gas hydrate research is leading us to new understanding of formation and dissociation phenomena that focuses on measurement and modeling of time-dependent properties of gas hydrates on the basis of their well-established thermodynamic properties.

Dynamical and rheological properties of fluorinated surfactant films adsorbed at the pressurized CO2-H2O interface

The dynamics of adsorption, interfacial tension, and rheological properties of two phosphocholine-derived partially fluorinated surfactants FnHmPC, designed to compensate for the weak CO(2)-surfactant tail interactions, were determined at the pressurized CO(2)-H(2)O interface. The two surfactants differ only by the length of the hydrocarbon spacer (5 CH(2) in F8H5PC and 11 CH(2) in F8H11PC) located between the terminal perfluoroalkyl chain and the polar head. The length of this spacer was found to have a critical impact on the adsorption kinetics and elasticity of the interfacial surfactant film. F8H5PC is soluble in both water and CO(2) phases and presents several distinct successive interfacial behaviors when bulk water concentration (C(W)) increases and displays a nonclassical isotherm shape. The isotherms of F8H5PC are similar for the three CO(2) pressures investigated and comprise four regimes. In the first regime, at low C(W), the interfacial tension is controlled by the organization that occurs between H(2)O and CO(2). The second regime corresponds to the adsorption of the surfactant as a monolayer until the CO(2) phase is saturated with F8H5PC, resulting in a first inflection point. In this regime, F8H5PC molecules reach maximal compaction and display the highest apparent interfacial elasticity. In the third regime, a second inflection is observed that corresponds to the critical micelle concentration of the surfactant in water. At the highest concentrations (fourth regime), the interfacial films are purely viscous and highly flexible, suggesting the capacity for this surfactant to produce water-in-CO(2) microemulsion. In this regime, surfactant adsorption is very fast and equilibrium is reached in less than 100 s. The behavior of F8H11PC is drastically different: it forms micelles only in the water phase, resulting in a classical Gibbs interface. This surfactant decreases the interfacial tension down to 1 mN/m and forms a strongly elastic interface. As this surfactant forms a very cohesive interface, it should be suitable for formulating stable water-in-CO(2) emulsions. The finding that the length of the hydrocarbon spacer in partially fluorinated surfactants can drastically influence film properties at the CO(2)-H(2)O interface should help control the formation of microemulsions versus emulsions and help elaborate a rationale for the design of surfactants specifically adapted to pressurized CO(2).

A large-volume gas cell for high-energy X-ray reflectivity investigations of interfaces under pressure

A cell for the investigation of interfaces under pressure is presented. Given the pressure and temperature specifications of the cell, P ≤ 100 bar and 253 K ≤ T ≤ 323 K, respectively, high-energy X-rays are required to penetrate the thick Al(2)O(3) windows. The CH(4)(gas)/H(2)O(liquid) interface has been chosen to test the performance of the new device. The measured dynamic range of the high-energy X-ray reflectivity data exceeds 10(-8), thereby demonstrating the validity of the entire experimental set-up.

Amorphous precursors in the nucleation of clathrate hydrates

The nucleation and growth of clathrate hydrates of a hydrophobic guest comparable to methane or carbon dioxide are studied by molecular dynamics simulations of two-phase systems. The crystallization proceeds in two steps: First, the guest molecules concentrate in "blobs", amorphous clusters involving multiple guest molecules in water-mediated configurations. These blobs are in dynamic equilibrium with the dilute solution and give birth to the clathrate cages that eventually transform it into an amorphous clathrate nucleus. In a second step, the amorphous clathrate transforms into crystalline clathrate. At low temperatures, the system can be arrested in the metastable amorphous clathrate phase for times sufficiently long for it to appear as an intermediate in the crystallization of clathrates. The "blob mechanism" unveiled in this work synthesizes elements of the labile cluster and local structuring hypotheses of clathrate nucleation and bears strong analogies to the two-step mechanisms of crystallization of proteins and colloids.

Stability, adsorption, and diffusion of CH₄, CO₂, and H₂ in clathrate hydrates

We present a study of the adsorption and diffusion of CH₄, CO₂, and H₂ molecules in clathrate hydrates using ab initio van der Waals density functional formalism [M. Dion, Phys. Rev. Lett. 92, 246401 (2004)10.1103/PhysRevLett.92.246401]. We find that the adsorption energy is dominated by van der Waals interactions and that, without them, gas hydrates would not be stable. We calculate the maximum adsorption capacity as well as the maximum hydrocarbon size that can be adsorbed. The relaxation of the host lattice is essential for a good description of the diffusion activation energies, which are estimated to be of the order of 0.2, 0.4, and 1.0 eV for H₂, CO₂, and CH₄, respectively.

Hydrate formation at the methane/water interface on the molecular scale

We report the nucleation process of methane hydrate on the molecular scale. A stationary planar interface separating methane gas and liquid water was studied by using in situ neutron reflectivity. We found that the angstrom-scale surface roughening is triggered as soon as the water phase contacts methane gas under the hydrate forming conditions. In addition, it was found that the microscopic surface structure remains unchanged until a macroscopic hydrate film is developed at the interface. We therefore postulate that the angstrom-scale surface roughening is attributed to the formation of microscopic hydrate "embryos" in a "dynamic equilibrium" manner.

Tetrahydrofuran clathrate hydrate formation

We report on the formation of tetrahydrofuran clathrate hydrate studied by x-ray Raman scattering measurements at the oxygen K edge. A comparison of x-ray Raman spectra measured from water-tetrahydrofuran mixtures and tetrahydrofuran hydrate at different temperatures supports stochastic hydrate formation models rather than models assuming hydrate precursors. This is confirmed by molecular dynamics simulations and density functional theory calculations of x-ray Raman spectra. In addition, changes in the spectra of tetrahydrofuran hydrate with temperatures close to the hydrate's dissociation temperature were observed and may be connected to changes in hydrate's local structure due to the formation of hydrogen bonds between guest and water molecules.