Modeling and scale-bridging using machine learning: nanoconfinement effects in porous media

Colloid Transport in Bicontinuous Nanoporous Media

Colloid transport and retention in porous media are critical processes influencing various Earth science applications, from groundwater remediation to enhanced oil recovery. These phenomena become particularly complex in the confined spaces of nanoporous media, where strong boundary layer effects and nanoconfinement significantly alter colloid behavior. In this work, we use particle dynamics models to simulate colloid transport and retention processes in bicontinuous nanoporous (BNP) media under pressure gradients. By utilizing particle-based models, we track the movement of each colloid and elucidate the underlying colloid retention mechanisms. Under unfavorable attachment conditions, the results reveal two colloid retention mechanisms: physical straining and trapping in low-flow zone. Furthermore, we investigate the effects of critical factors including colloid volume fraction, d, pressure difference, ΔP, interaction between colloids and BNP media, Ec-p, and among colloids, Ec-c, on colloid transport. Analysis of breakthrough curves and colloid displacements demonstrates that higher values of d, lower values of ΔP, and strong Ec-p attractions significantly increase colloid retention, which further lead to colloid clogging and jamming. In contrast, Ec-c has minimal impact on colloid transport due to the limited colloid-colloid interaction in nanoporous channels. This work provides critical insights into the fundamental factors governing colloid transport and retention within stochastic nanoporous materials.

Recent Trends in Polymeric Foams and Porous Structures for Electromagnetic Interference Shielding Applications

Polymer-based (nano)composite foams containing conductive (nano)fillers limit electromagnetic interference (EMI) pollution, and have been shown to act as good shielding materials in electronic devices. However, due to their high (micro)structural complexity, there is still a great deal to learn about the shielding mechanisms in these materials; understanding this is necessary to study the relationship between the properties of the microstructure and the porous structure, especially their EMI shielding efficiency (EMI SE). Targeting and controlling the electrical conductivity through a controlled distribution of conductive nanofillers are two of the main objectives when combining foaming with the addition of nanofillers; to achieve this, both single or combined nanofillers (nanohybrids) are used (as there is a direct relationship between electrical conductivity and EMI SE), as are the main shielding mechanisms working on the foams (which are expected to be absorption-dominated). The present review considers the most significant developments over the last three years concerning polymer-based foams containing conductive nanofillers, especially carbon-based nanofillers, as well as other porous structures created using new technologies such as 3D printing for EMI shielding applications. It starts by detailing the microcellular foaming strategy, which develops polymer foams with enhanced EMI shielding, and it particularly focuses on technologies using supercritical CO2 (sCO2). It also notes the use of polymer foams as templates to prepare carbon foams with high EMI shielding performances for high temperature applications, as well as a recent strategy which combines different functional (nano)fillers to create nanohybrids. This review also explains the control and selective distribution of the nanofillers, which favor an effective conductive network formation, which thus promotes the enhancement of the EMI SE. The recent use of computational approaches to tailor the EMI shielding properties are given, as are new possibilities for creating components with varied porous structures using the abovementioned materials and 3D printing. Finally, future perspectives are discussed.

Predictive scale-bridging simulations through active learning

Throughout computational science, there is a growing need to utilize the continual improvements in raw computational horsepower to achieve greater physical fidelity through scale-bridging over brute-force increases in the number of mesh elements. For instance, quantitative predictions of transport in nanoporous media, critical to hydrocarbon extraction from tight shale formations, are impossible without accounting for molecular-level interactions. Similarly, inertial confinement fusion simulations rely on numerical diffusion to simulate molecular effects such as non-local transport and mixing without truly accounting for molecular interactions. With these two disparate applications in mind, we develop a novel capability which uses an active learning approach to optimize the use of local fine-scale simulations for informing coarse-scale hydrodynamics. Our approach addresses three challenges: forecasting continuum coarse-scale trajectory to speculatively execute new fine-scale molecular dynamics calculations, dynamically updating coarse-scale from fine-scale calculations, and quantifying uncertainty in neural network models.

Neural network predicts ion concentration profiles under nanoconfinement

Modeling the ion concentration profile in nanochannel plays an important role in understanding the electrical double layer and electro-osmotic flow. Due to the non-negligible surface interaction and the effect of discrete solvent molecules, molecular dynamics (MD) simulation is often used as an essential tool to study the behavior of ions under nanoconfinement. Despite the accuracy of MD simulation in modeling nanoconfinement systems, it is computationally expensive. In this work, we propose neural network to predict ion concentration profiles in nanochannels with different configurations, including channel widths, ion molarity, and ion types. By modeling the ion concentration profile as a probability distribution, our neural network can serve as a much faster surrogate model for MD simulation with high accuracy. We further demonstrate the superior prediction accuracy of neural network over XGBoost. Finally, we demonstrated that neural network is flexible in predicting ion concentration profiles with different bin sizes. Overall, our deep learning model is a fast, flexible, and accurate surrogate model to predict ion concentration profiles in nanoconfinement.

A Dataset of 3D Structural and Simulated Transport Properties of Complex Porous Media

Physical processes that occur within porous materials have wide-ranging applications including - but not limited to - carbon sequestration, battery technology, membranes, oil and gas, geothermal energy, nuclear waste disposal, water resource management. The equations that describe these physical processes have been studied extensively; however, approximating them numerically requires immense computational resources due to the complex behavior that arises from the geometrically-intricate solid boundary conditions in porous materials. Here, we introduce a new dataset of unprecedented scale and breadth, DRP-372: a catalog of 3D geometries, simulation results, and structural properties of samples hosted on the Digital Rocks Portal. The dataset includes 1736 flow and electrical simulation results on 217 samples, which required more than 500 core years of computation. This data can be used for many purposes, such as constructing empirical models, validating new simulation codes, and developing machine learning algorithms that closely match the extensive purely-physical simulation. This article offers a detailed description of the contents of the dataset including the data collection, simulation schemes, and data validation.

Machine Learning for Property Prediction and Optimization of Polymeric Nanocomposites: A State-of-the-Art

Recently, the field of polymer nanocomposites has been an area of high scientific and industrial attention due to noteworthy improvements attained in these materials, arising from the synergetic combination of properties of a polymeric matrix and an organic or inorganic nanomaterial. The enhanced performance of those materials typically involves superior mechanical strength, toughness and stiffness, electrical and thermal conductivity, better flame retardancy and a higher barrier to moisture and gases. Nanocomposites can also display unique design possibilities, which provide exceptional advantages in developing multifunctional materials with desired properties for specific applications. On the other hand, machine learning (ML) has been recognized as a powerful predictive tool for data-driven multi-physical modelling, leading to unprecedented insights and an exploration of the system’s properties beyond the capability of traditional computational and experimental analyses. This article aims to provide a brief overview of the most important findings related to the application of ML for the rational design of polymeric nanocomposites. Prediction, optimization, feature identification and uncertainty quantification are presented along with different ML algorithms used in the field of polymeric nanocomposites for property prediction, and selected examples are discussed. Finally, conclusions and future perspectives are highlighted.

Multiscale Modeling of Broadband Perfect Absorbers Based on Gold Metallic Molecules

The modeling of functional photonic devices that rely on molecular materials continues to be a major contemporary challenge. It is a challenge because, in the Maxwell equations, which govern the light-matter interaction, material properties are primarily introduced on phenomenological grounds and not from first principles. To overcome such limitations, we outline a multiscale modeling approach that bridges multiple length scales. We can predict with our approach the optical response of a photonic device that exploits in its design molecular materials whose properties were determined using time-dependent density functional theory. The specifically considered device is a broadband perfect absorber that uses in part a thin film comprising gold molecules made from 144 atoms. Our methodology discloses various chemical and physical effects that define such a device's response. Our methodology is versatile, and a larger number of applications will profit from this development.

A review on slip boundary conditions at the nanoscale: recent development and applications

The slip boundary condition for nanoflows is a key component of nanohydrodynamics theory, and can play a significant role in the design and fabrication of nanofluidic devices. In this review, focused on the slip boundary conditions for nanoconfined liquid flows, we firstly summarize some basic concepts about slip length including its definition and categories. Then, the effects of different interfacial properties on slip length are analyzed. On strong hydrophilic surfaces, a negative slip length exists and varies with the external driving force. In addition, depending on whether there is a true slip length, the amplitude of surface roughness has different influences on the effective slip length. The composition of surface textures, including isotropic and anisotropic textures, can also affect the effective slip length. Finally, potential applications of nanofluidics with a tunable slip length are discussed and future directions related to slip boundary conditions for nanoscale flow systems are addressed.

Nanoscale slip length prediction with machine learning tools

This work incorporates machine learning (ML) techniques, such as multivariate regression, the multi-layer perceptron, and random forest to predict the slip length at the nanoscale. Data points are collected both from our simulation data and data from the literature, and comprise Molecular Dynamics simulations of simple monoatomic, polar, and molecular liquids. Training and test points cover a wide range of input parameters which have been found to affect the slip length value, concerning dynamical and geometrical characteristics of the model, along with simulation parameters that constitute the simulation conditions. The aim of this work is to suggest an accurate and efficient procedure capable of reproducing physical properties, such as the slip length, acting parallel to simulation methods. Non-linear models, based on neural networks and decision trees, have been found to achieve better performance compared to linear regression methods. After the model is trained on representative simulation data, it is capable of accurately predicting the slip length values in regions between or in close proximity to the input data range, at the nanoscale. Results also reveal that, as channel dimensions increase, the slip length turns into a size-independent material property, affected mainly by wall roughness and wettability.

Molecular-Scale Considerations of Enhanced Oil Recovery in Shale

With only less than 10% recovery, the primary production of hydrocarbon from shale reservoirs has redefined the energy equation in the world. Similar to conventional reservoirs, Enhanced Oil Recovery (EOR) techniques could be devised to enhance the current recovery factors. However, shale reservoirs possess unique characteristics that significantly affect the fluid properties. Therefore, we are adopting a molecular simulation approach that is well-suited to account for these effects to evaluate the performance of three different gases, methane, carbon dioxide and nitrogen, to recover the hydrocarbons from rough pore surfaces. Our hydrocarbon systems consists of either a single component (decane) or more than one component (decane and pentane). We simulated cases where concurrent and countercurrent displacement is studied. For concurrent displacement (injected fluids displace hydrocarbons towards the production region), we found that nitrogen and methane yielded similar recovery; however nitrogen exhibited a faster breakthrough. On the other hand, carbon dioxide was more effective in extracting the hydrocarbons when sufficient pressure was maintained. For countercurrent displacement (gases are injected and hydrocarbons are produced from the same direction), methane was found to be more effective, followed by carbon dioxide and nitrogen. In all cases, confinement reduced the recovery factor of all gases. This work provides insights to devise strategies to improve the current recovery factors observed in shale reservoirs.