Multiscale Simulation Approach Research Articles

A Multi-Scale Numerical Simulation Method Considering Anisotropic Relative Permeability

Most of the oil reservoirs in China are fluvial deposits with firm reservoir heterogeneity, where differences in fluid flow capacity in individual directions should not be ignored; however, the available commercial reservoir simulation software cannot consider the anisotropy of the relative permeability. To handle this challenge, this paper takes full advantage of the parallelism of the multi-scale finite volume (MsFV) method and establishes a multi-scale numerical simulation approach that incorporates the effects of reservoir anisotropy. The methodology is initiated by constructing an oil–water black-oil model considering the anisotropic relative permeability. Subsequently, the base model undergoes decoupling through a sequential solution, formulating the pressure and transport equations. Following this, a multi-scale grid system is configured, within which the pressure and transport equations are progressively developed in the fine-scale grid domain. Ultimately, the improved multi-scale finite volume (IMsFV) method is applied to mitigate low-frequency error in the coarse-scale grid, thereby enhancing computational efficiency. This paper introduces two primary innovations. The first is the development of a multi-scale solution method for the pressure equation incorporating anisotropic relative permeability. Validated using the Egg model, a comparative analysis with traditional numerical simulations demonstrates a significant improvement in computational speed without sacrificing accuracy. The second innovation involves applying the multi-scale framework to investigate the impact of anisotropy relative permeability on waterflooding performance, uncovering distinct mechanisms by which absolute and relative permeability anisotropy influence waterflooding outcomes. Therefore, the IMsFV method can be used as an effective tool for high-resolution simulation and precise residual oil prediction in anisotropic reservoirs.

Mechanism of activation and autophosphorylation of a histidine kinase

Histidine kinases (HK) are one of the main prokaryotic signaling systems. Two structurally conserved catalytic domains inside the HK enable autokinase, phosphotransfer, and phosphatase activities. Here, we focus on a detailed mechanistic understanding of the functional cycle of the WalK HK by a multi-scale simulation approach, consisting of classical as well as hybrid QM/MM molecular dynamics simulation. Strikingly, a conformational transition induced solely in DHp leads to the correct activated conformation in CA crucial for autophosphorylation. This finding explains how variable sensor domains induce the transition from inactive to active state. The subsequent autophosphorylation inside DHp proceeds via a penta-coordinated transition state to a protonated phosphohistidine intermediate. This intermediate is consequently deprotonated by a suitable nearby base. The reaction energetics are controlled by the final proton acceptor and presence of a magnesium cation. The slow rates of the process result from the high energy barrier of the conformational transition between inactive and active states. The phosphorylation step exhibits a lower barrier and down-the-hill energetics. Thus, our work suggests a detailed mechanistic model for HK autophosphorylation.

(Invited) Multiscale Modeling and Machine Learning Investigation of Battery Interphase Dynamics

In recent years, the solid electrolyte interphase (SEI) has emerged as a critical component in the development of efficient and durable electrochemical storage systems. Understanding the dynamics and properties of SEI is paramount for the enhancement of battery performance and longevity. This study presents a novel multi-scale simulation approach, incorporating Hybrid Ab Initio Reactive Molecular Dynamics (HAIR), to elucidate the intricate mechanisms governing SEI formation and evolution. Our methodology involves simulation under operando conditions, closely mimicking real-world scenarios, to provide insights into the molecular interactions and structural changes within the SEI layer. By leveraging HAIR, we bridge the gap between quantum mechanical precision and the expansive temporal-spatial scales of molecular dynamics, facilitating a more comprehensive understanding of SEI behavior [1,2]. Furthermore, we integrate machine learning (ML) techniques to refine our predictive models. These ML algorithms are trained on a rich dataset derived from HAIR simulations, enabling the prediction of key electrochemical properties, including chemical composition and electronic structure. The application of machine learning significantly enhances our ability to forecast SEI evolution under various operational conditions. A critical aspect of our research is the prediction of X-ray photoelectron spectroscopy (XPS) and Coulombic Efficiency (CE) outcomes [3]. By simulating these properties, we provide valuable predictions that align closely with experimental data, thereby validating our simulation framework. In summary, our study offers a groundbreaking perspective on SEI analysis, merging the fidelity of HAIR with the predictive power of machine learning. This approach not only deepens our fundamental understanding of SEI properties but also paves the way for the development of more efficient and durable electrochemical storage systems. Our findings hold significant implications for the future of battery technology and underscore the importance of advanced simulation techniques in electrochemical research. Reference 1. Liu Y; Yu PP; Sun QT; Wu Y; Xie M; Yang H; Cheng T*; Goddard WA; ACS Energy Lett. 2021, 6, 2320.2. Liu Y; Sun QT; Yue BT; Zhang YY; Cheng T*; J. Mater. Chem. A 2023, 11, 14640.3. Sun QT; Xiang Y; Liu Y; Xu L; Leng TL; Ye YF; Fortunelli A*; Goddard WA*; Cheng T*; J. Phys. Chem. Lett. 2022, 13, 8047. Figure 1

Thermoelectric properties of graphene through BN-ring doping: A theoretical investigation

Although graphene presents excellent transport properties, its potential for thermoelectric applications is still limited due to its low Seebeck coefficient and high thermal conductivity. Efforts to enhance its thermoelectric properties have involved the usage of carbon-based nanoribbons (Zheng et al., 2012; Hossain et al., 2016) [1,2], strain engineering (Nguyen et al., 2015), and heteroatom co-doping, particularly with nitrogen and/or boron atoms (Wang et al., 2018) [3]. In this work, multiscale simulation approaches combining density functional theory calculations and semi-empirical models are used to investigate the thermoelectric properties of graphene via borazine (B3N3)-ring doping. In particular, the bandgap engineering obtained by this doping technique (Caputo et al., 2022) can separate opposite electron and hole contributions and therefore results in a significant enhancement of the Seebeck effect and, accordingly, of the power factor. The results obtained are not only dependent on the concentration of B3N3 rings but are also notably influenced by their orientation configuration. Furthermore, the effects of distribution and rotational disorders are also considered, clarifying the thermoelectric properties of realistic systems. Lastly, the thermal lattice conductance is estimated revealing a substantial reduction of up to 40% compared to pristine graphene opening up the possibility of enhancing the thermoelectric efficiency of BNC materials. The present theoretical approach highlights how BN-ring doping can refine the thermoelectric properties of graphene, offering a pathway for enhancing its suitability in practical thermoelectric applications.

Interaction between lateral jet and hypersonic rarefied flow

Reaction control systems (RCS) are commonly utilized in a variety of air vehicles, and they give rise to complex flow phenomena. For flows dominated by shock waves and expansion, such as lateral jets, a significant gradient is formed in the local region of the flow field, resulting in a high local Knudsen number, and the continuum hypothesis is no longer applicable, even if the freestream belongs to near-continuum flow. As a result, multi-scale simulation approaches should be used to capture the interactions between these multi-scale structures and appropriately determine aerodynamic forces. Our earlier work, which used the conserved discrete unified gas kinetic scheme (CDUGKS) for monatomic gas, studied the flow field features of lateral jets and aerodynamic forces on blunt bodies under various rarefied freestream conditions. In this research, the Rykov model is used to expand the CDUGKS suit for diatomic gas flow while taking into account the excitation of molecular rotational energy. The impact on the flow field and aerodynamic forces from four parameters, namely freestream Mach number, jet Mach number, jet pressure ratio, and nozzle position, is thoroughly investigated. The results show that even if the incoming Mach number is different, the flow field structure and aerodynamic properties don't change much when the jet momentum ratio stays the same. The solid wall stagnation point value and local peak also show some linear relationships. The jet pressure ratio mainly affects the expansion characteristics of the jet, such as the influence range on the solid wall. In addition, with an increase in the distance between the nozzle position and the torque reference point, higher control efficiency can be obtained. The study of these aspects will help us comprehend lateral jet flow and provide reference for the design of an air vehicle's RCS.

Direct FE2 Based on Single-Integration Point for Modeling Damage Evolution of Materials with Micro-Cracks

The Direct FE2 method is a recently developed concurrent multi-scale simulation approach with a monolithic solution scheme, where the macroscopic and microscopic models are solved in a unit iteration process. The conventional Direct FE2 using quadrilateral element has some limitations in determining the failure state of a macro-element. In this work, we develop a new Direct FE2 method based on the Single-integration Point triangle element (Direct SP-FE2) and apply it to simulate the micro-crack-induced fracture problems for the first time. The cohesive element is inserted into a representative volume element (RVE), so that the failure process of a macro-element can be determined by the micro-crack propagation on RVE. Accordingly, the relationship between macro-deformations and microstructures can be established directly, thus avoiding complex derivation and utilization of a multi-scale damage constitutive model. In Direct SP-FE2, the failure state of a macro-element can be directly predicted by the RVE at the single-integration point inside, avoiding the uncertainty associated with predictions by multiple RVEs. As a result, compared to the conventional Direct FE2 with a quadrilateral element, the Direct SP-FE2 shows advantages in capturing complex geometric boundaries and predicting the damage evolution process and failure state more accurately. Various numerical examples are conducted to comprehensively validate the effectiveness of the present method in describing the influence of micro-cracks on macro-structure deformations. For the macro-damage behaviors, the simulation results by the Direct SP-FE2 show good agreement with the direct numerical simulation (DNS) results of the full FE model at a significantly lower computational time. With the developed Direct SP-FE2, a variety of complex micro-crack fracture problems in composite materials can be successfully modeled by manipulating the interior RVE structure.

Meshfree multiscale method for the infiltration problem in permafrost

This study introduces a multiscale simulation approach, employing the Meshfree Generalized Multiscale Finite Element Method (GMsFEM), to numerically model fluid seepage in permafrost soil conditions. Combining Meshfree and Finite Element Methods, the coarse-grid system is addressed, while the standard Finite Element Method operates on the fine grid. For representing fractures at the fine grid level, the Discrete Fracture Model (DFM) is utilized. The findings demonstrate the method’s efficacy in accurately depicting soil heterogeneity and the impact of fractures on soil temperature.

Multi-scale simulation of microstructural evolution for molten U–Zr–O ternary system

The morphological evolution of the molten corium is critical to the success of the “in-vessel retention” (IVR) strategy, which is of great significance for the safety of the nuclear industry under severe accidents. However, the direct observation of microstructural evolution by experiments is impossible in extreme conditions. Herein, a multi-scale simulation approach was employed to investigate the microstructural evolution and related mechanisms for the molten U–Zr–O ternary system. First, based on the reported calculated phase diagram, a simplified free energy formula describing a phase diagram of U–Zr–O was determined, where only liquid phases are focused. Then, the atomic diffusion coefficients were calculated by ab-initio molecular dynamics simulations. Based on such information, a homemade phase-field model was developed, which was coupled with Cahn–Hillard and Navier–Stokes equations. Accordingly, the microstructural evolution of molten U–Zr–O materials was systematically investigated. The results reveal that the evolution of the molten core was driven by both the decreasing free energy and gravity potential, accessing the final morphology of the system as a two-layer structure, including a metallic layer and an oxidized layer. Moreover, diverse intermediate morphologies could be observed from different initial component distributions, although their average compositions were identical. Finally, some intermediate simulated microstructural patterns agreed well with experimental observation, demonstrating the unstable structures captured by experiments and the validity of the simulations. The present study provides an effective approach to investigate the microstructural evolution of molten core and a guideline for optimizing the IVR strategy based on different morphologies in the corium.

Multiscale Simulations of Self-Assembling Peptides: Surface and Core Hydrophobicity Determine Fibril Stability and Amyloid Aggregation.

Assemblies of peptides and proteins through specific intermolecular interactions set the basis for macroscopic materials found in nature. Peptides provide easily tunable hydrogen-bonding interactions, which can lead to the formation of ordered structures such as highly stable β-sheets that can form amyloid-like supramolecular peptide nanofibrils (PNFs). PNFs are of special interest, as they could be considered as mimics of various fibrillar structures found in nature. In their ability to serve as supramolecular scaffolds, they could mimic certain features of the extracellular matrix to provide stability, interact with pathogens such as virions, and transduce signals between the outside and inside of cells. Many PNFs have been reported that reveal rich bioactivities. PNFs supporting neuronal cell growth or lentiviral gene transduction have been studied systematically, and their material properties were correlated to bioactivities. However, the impact of the structure of PNFs, their dynamics, and stabilities on their unique functions is still elusive. Herein, we provide a microscopic view of the self-assembled PNFs to unravel how the amino acid sequence of self-assembling peptides affects their secondary structure and dynamic properties of the peptides within supramolecular fibrils. Based on sequence truncation, amino acid substitution, and sequence reordering, we demonstrate that peptide-peptide aggregation propensity is critical to form bioactive β-sheet-rich structures. In contrast to previous studies, a very high peptide aggregation propensity reduces bioactivity due to intermolecular misalignment and instabilities that emerge when fibrils are in close proximity to other fibrils in solution. Our multiscale simulation approach correlates changes in biological activity back to single amino acid modifications. Understanding these relationships could lead to future material discoveries where the molecular sequence predictably determines the macroscopic properties and biological activity. In addition, our studies may provide new insights into naturally occurring amyloid fibrils in neurodegenerative diseases.

Accurate prediction of aging effects on microstructure evolution and related mechanical strength of Mg-Zn alloys via multiscale simulations

Age strengthening is a common method to achieve superior mechanical properties employed in magnesium (Mg) alloys. However, the experimental approaches to examine the aging effects on mechanical properties are primarily based on trial and error, posing challenges for the design and development of high-strength Mg alloys. A promising solution to address this challenge is to establish an accurate prediction model for the aging effects on Mg alloys, covering the evolution of microstructure and corresponding mechanical properties. This paper introduces a multiscale simulation approach to accurately predict the aging effects on the Mg-2.3Zn (at.%) alloy. To begin, we establish a phase field model to simulate the microstructure evolution at 150 ℃, demonstrating precipitate growth during aging in alignment with experimental observations. Secondly, based on the microstructures obtained from phase field simulations, the corresponding yield strength is determined through finite element method simulations. The simulation successfully captures the peak-aged behavior of yield strength during aging, resulting in a yield strength of 167 MPa at the peak-aged state, with a deviation of less than 5 % from experimental results. In-depth analysis indicates that the peak value of total yield strength arises from the interplay between solid solution strengthening and precipitate strengthening. Furthermore, precipitate strengthening reaches a peak value at a specific aging time, deviating from the total yield strength. This deviation is primarily attributed to the effect of precipitate size on shear strengthening during aging. This study presents a general method for predicting the microstructure evolution and corresponding mechanical properties of alloys through multiscale simulations, offering valuable insights for designing and predicting the performance of metallic alloys.

Higher-order multi-scale computational approach and its convergence for nonlocal gradient elasticity problems of composite materials

A novel higher-order multi-scale computational approach for efficient nonlocal gradient elasticity simulation of composite materials is proposed in this paper. Based on the use of decoupling technique, raw fourth-order nonlocal gradient equations with periodically oscillatory coefficients are decoupled as the new second-order multi-scale equations for reducing the continuity requirements when implementing multi-scale simulation. Next, a novel macro-micro coupled multi-scale computational model with higher-order corrected terms is rigorously devised for high-resolution multi-scale and nonlocal analysis of composite materials via multi-scale asymptotic analysis. Then, the local error analysis of higher-order multi-scale solutions in the point-wise sense illustrates that establishing higher-order asymptotic corrected terms plays a vital role in investigating the micromechanical mechanisms. Moreover, a rigorous global error estimation with an explicit rate of higher-order multi-scale solutions is first derived in the energy norm sense. In addition, an efficient multi-scale numerical algorithm is presented to effectively simulate nonlocal gradient elasticity problems of composite materials based on finite element method (FEM). And we also derive the convergence estimation of the proposed numerical algorithm. Finally, numerical experiments are conducted to gauge the efficiency and accuracy of the proposed higher-order multi-scale method. This paper offers a unified higher-order two-scale computational framework for enabling nonlocal gradient elasticity behavior simulation of composite materials.

A hybrid Gaussian mixture/DSMC approach to study the Fourier thermal problem

In rarefied gas dynamics scattering kernels deserve special attention since they contain all the essential information about the effects of physical and chemical properties of the gas–solid surface interface on the gas scattering process. However, to study the impact of the gas–surface interactions on the large-scale behavior of fluid flows, these scattering kernels need to be integrated in larger-scale models like Direct Simulation Monte Carlo (DSMC). In this work, the Gaussian mixture (GM) model, an unsupervised machine learning approach, is utilized to establish a scattering kernel for monoatomic (Ar) and diatomic (H2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\hbox {H}_{2}$$\\end{document}) gases directly from Molecular Dynamics (MD) simulations data. The GM scattering kernel is coupled to a pure DSMC solver to study isothermal and non-isothermal rarefied gas flows in a system with two parallel walls. To fully examine the coupling mechanism between the GM scattering kernel and the DSMC approach, a one-to-one correspondence between MD and DSMC particles is considered here. Benchmarked by MD results, the performance of the GM-DSMC is assessed against the Cercignani–Lampis–Lord (CLL) kernel incorporated into DSMC simulation (CLL-DSMC). The comparison of various physical and stochastic parameters shows the better performance of the GM-DSMC approach. Especially for the diatomic system, the GM-DSMC outperforms the CLL-DSMC approach. The fundamental superiority of the GM-DSMC approach confirms its potential as a multi-scale simulation approach for accurately measuring flow field properties in systems with highly nonequilibrium conditions.

The catalytic mechanism of the RNA methyltransferase METTL3.

The complex of methyltransferase-like proteins 3 and 14 (METTL3-14) is the major enzyme that deposits N6-methyladenosine (m6A) modifications on messenger RNA (mRNA) in humans. METTL3-14 plays key roles in various biological processes through its methyltransferase (MTase) activity. However, little is known about its substrate recognition and methyl transfer mechanism from its cofactor and methyl donor S-adenosylmethionine (SAM). Here, we study the MTase mechanism of METTL3-14 by a combined experimental and multiscale simulation approach using bisubstrate analogues (BAs), conjugates of a SAM-like moiety connected to the N6-atom of adenosine. Molecular dynamics simulations based on crystal structures of METTL3-14 with BAs suggest that the Y406 side chain of METTL3 is involved in the recruitment of adenosine and release of m6A. A crystal structure with a BA representing the transition state of methyl transfer shows a direct involvement of the METTL3 side chains E481 and K513 in adenosine binding which is supported by mutational analysis. Quantum mechanics/molecular mechanics (QM/MM) free energy calculations indicate that methyl transfer occurs without prior deprotonation of adenosine-N6. Furthermore, the QM/MM calculations provide further support for the role of electrostatic contributions of E481 and K513 to catalysis. The multidisciplinary approach used here sheds light on the (co)substrate binding mechanism, catalytic step, and (co)product release, and suggests that the latter step is rate-limiting for METTL3. The atomistic information on the substrate binding and methyl transfer reaction of METTL3 can be useful for understanding the mechanisms of other RNA MTases and for the design of transition state analogues as their inhibitors.

The catalytic mechanism of the RNA methyltransferase METTL3

The complex of methyltransferase-like proteins 3 and 14 (METTL3-14) is the major enzyme that deposits N6-methyladenosine (m6A) modifications on messenger RNA (mRNA) in humans. METTL3-14 plays key roles in various biological processes through its methyltransferase (MTase) activity. However, little is known about its substrate recognition and methyl transfer mechanism from its cofactor and methyl donor S-adenosylmethionine (SAM). Here, we study the MTase mechanism of METTL3-14 by a combined experimental and multiscale simulation approach using bisubstrate analogues (BAs), conjugates of a SAM-like moiety connected to the N6-atom of adenosine. Molecular dynamics simulations based on crystal structures of METTL3-14 with BAs suggest that the Y406 side chain of METTL3 is involved in the recruitment of adenosine and release of m6A. A crystal structure with a BA representing the transition state of methyl transfer shows a direct involvement of the METTL3 side chains E481 and K513 in adenosine binding which is supported by mutational analysis. Quantum mechanics/molecular mechanics (QM/MM) free energy calculations indicate that methyl transfer occurs without prior deprotonation of adenosine-N6. Furthermore, the QM/MM calculations provide further support for the role of electrostatic contributions of E481 and K513 to catalysis. The multidisciplinary approach used here sheds light on the (co)substrate binding mechanism, catalytic step, and (co)product release, and suggests that the latter step is rate-limiting for METTL3. The atomistic information on the substrate binding and methyl transfer reaction of METTL3 can be useful for understanding the mechanisms of other RNA MTases and for the design of transition state analogues as their inhibitors.

A multiscale molecular simulation approach for assessing the effects of mutation on protein activity and drug sensitivity of oncogenic kinases

A multiscale molecular simulation approach for assessing the effects of mutation on protein activity and drug sensitivity of oncogenic kinases

Multiscale Molecular Dynamics Simulations of an Active Self-Assembling Material.

Inspired by the adaptability observed in biological materials, self-assembly processes have attracted significant interest for their potential to yield novel materials with unique properties. However, experimental methods have often fallen short in capturing the molecular details of the assembly process. In this study, we employ a multiscale molecular dynamics simulation approach, complemented by NMR quantification, to investigate the mechanism of self-assembly in a redox-fueled bioinspired system. Contrary to conventional assumptions, we have uncovered a significant role played by the monomer precursor in the assembly process, with its presence varying with concentration and the extent of conversion of the monomer to the dimer. Experimental confirmation through NMR quantification underscores the concentration-dependent incorporation of monomers into the fibrous structures. Furthermore, our simulations also shed light on the diverse intermolecular interactions, including T-shaped and parallel π stacking, as well as hydrogen bonds, in stabilizing the aggregates. Overall, the open conformation of the dimer is preferred within these aggregates. However, inside the aggregates, the distribution of conformations shifts slightly to the closed conformation compared to on the surface. These findings contribute to the growing field of bioinspired materials science by providing valuable mechanistic and structural insights to guide the design and development of self-assembling materials with biomimetic functionalities.

Multiphysics-Informed Pharmacokinetic Modeling of Systemic Exposure of Intramuscularly Injected LNPs.

Lipid nanoparticle (LNP) constructs have been widely developed for gene therapy delivery. Understanding local absorption and presystemic clearance kinetics of LNPs, however, remains limited. This subsequently restrains the prediction and assessment of the systemic exposure of locally injected LNPs. As such, a multiscale computational approach was developed by integrating multiphysics simulation of intramuscular absorption kinetics of LNPs with whole-body pharmacokinetics modeling, bridged by a presystemic lymphatic kinetic model. The overall framework was enabled by utilizing physiological parameters obtained from the literature and drug-related parameters derived from experiments. The multiscale modeling and simulation approach predicted the systemic exposure of LNPs administered intramuscularly, with a high degree of agreement between the predicted and the experimental data. Sensitivity analyses revealed that the local absorption rate, pinocytosis presystemic clearance rate, and lymph flow rate of the presystemic lymphatic compartment had the most significant impacts on Cmax. The study yielded refreshing perspectives on estimating systemic exposures of locally injected LNPs and their safety and effectiveness.

Multiscale modeling of functionally graded shell lattice metamaterials for additive manufacturing

In this work, an experimentally validated multiscale modeling framework for additively manufactured shell lattice structures with graded parameters is introduced. It is exemplified in application to the Schwarz primitive triply periodic minimal surface microstructure and 3D printing using masked stereolithography of a photopolymer material. The systematic procedure starts with the characterization of a hyperelastic material model for the 3D printed material. This constitutive model is then employed in the finite element simulation of shell lattices at finite deformations. The computational model is validated with experimental compression tests of printed lattice structures. In this way, the numerical convergence behavior and size dependence of the model are assessed, and the range in which it is reasonable to assume linear elastic behavior is determined. Then, representative volume elements subject to periodic boundary conditions are simulated to homogenize the mechanical behavior of Schwarz primitives with varying aspect ratios and shell thicknesses. Subsequently, the parameterized effective linear elasticity tensor of the metamaterial is represented by a physics-augmented neural network model. With this constitutive model, functionally graded shell lattice structures with varying microstructural parameters are simulated as macroscale continua using finite element and differential quadrature methods. The accuracy, reliability and effectiveness of this multiscale simulation approach are investigated and discussed. Overall, it is shown that this experimentally validated multiscale simulation framework, which is likewise applicable to other shell-like metamaterials, facilitates the design of functionally graded structures through additive manufacturing.Graphical

Design of high-strength martensitic steels by novel mixed-metal nanoprecipitates for high toughness and suppressed hydrogen embrittlement

To obtain a fundamental understanding of mechanisms of hydrogen embrittlement (HE) and its prevention in advanced high-strength steels containing novel nanoscale mixed-metal precipitates, it is necessary to study local microstructure, H trapping, and crack path with new multiscale experimental and simulation approach. Spatially resolved hydrogen mapping via SKPFM is used together with investigation of the crack path using high-resolution EBSD and HMPT, and global trapping behavior of the alloys by TDS. These results are combined with newly introduced method to elucidate real-time distribution of hydrogen in the alloy using high-energy synchrotron X-ray diffraction (HES-XRD). Mixed-metal precipitates improves HE resistance of the alloy, due to nature of the trapping sites, e.g. irreversible H-trapping by carbon vacancies inside novel nanoprecipitates and high total length of PAGBs. This is because of lower possibility of build-up of critical local hydrogen content at PAGBs for intergranular hydrogen-assisted cracking due to hydrogen-enhanced decohesion mechanism. Less weakly trapped hydrogen also reduces frequency of dislocation activation and enhanced dislocation slip in {011} slip plane due to hydrogen-enhanced localized plasticity in regions with affinity for transgranular hydrogen-assisted cracking at lower local hydrogen content. Direct evidence of carbon vacancies in novel nanoprecipitates is observed for the first time via HAADF-STEM.

Machine-learned constitutive relations for multi-scale simulations of well-entangled polymer melts

We investigate the use of a machine-learning-based simulation scheme to predict flows of complex fluids with a non-linear constitutive relation. This scheme, originally proposed by Seryo et al. for general complex fluids [N. Seryo et al., “Learning the constitutive relation of polymeric flows with memory,” Phys. Rev. Res. 2, 033107 (2020)], is used to learn the time derivative of the stress as a function of the stress and velocity gradient. However, previous assessments were limited to fluids with linear constitutive relations. In this study, we employ the dual sliplink model to express the dynamics of well-entangled polymers as a fluid exhibiting a non-linear stress response. We test the simulation scheme by using it to predict the flow of a viscoelastic fluid between two parallel plates with only shear deformations and compare the results with those of a multi-scale simulation using microscopic simulators. Overall, our machine-learning method possesses good predictive capabilities, for both the transient response and the non-linear behavior at steady-state, i.e., the shear-thinning. We are able to accurately track the evolution of the stress, for both the weak and strong elastic cases, although the velocity predictions for the latter show a decreased accuracy in the transient regime. Furthermore, our simulation scheme is more computationally efficient than the conventional multi-scale simulation approach, which uses microscopic simulators, containing a system of coarse-grained polymers, to evaluate the macroscopic stress. We discuss possible extensions and improvements for enhancing the predictive capabilities and generality of the method.