Multiscale Numerical Approach Research Articles

A multi-scale numerical approach to study monoclonal antibodies in solution.

Developing efficient and robust computational models is essential to improve our understanding of protein solution behavior. This becomes particularly important to tackle the high-concentration regime. In this context, the main challenge is to put forward coarse-grained descriptions able to reduce the level of detail, while retaining key features and relevant information. In this work, we develop an efficient strategy that can be used to investigate and gain insight into monoclonal antibody solutions under different conditions. We use a multi-scale numerical approach, which connects information obtained at all-atom and amino-acid levels to bead models. The latter has the advantage of reproducing the properties of interest while being computationally much faster. Indeed, these models allow us to perform many-protein simulations with a large number of molecules. We can, thus, explore conditions not easily accessible with more detailed descriptions, perform effective comparisons with experimental data up to very high protein concentrations, and efficiently investigate protein-protein interactions and their role in phase behavior and protein self-assembly. Here, a particular emphasis is given to the effects of charges at different ionic strengths.

Shakedown-reliability based fatigue strength prediction of parts fabricated by directed energy deposition considering the microstructural inhomogeneities

For mechanical components made by additive manufacturing (AM) techniques such as directed energy deposition (DED), micropores and other defects prevalently exist in the microstructure and they significantly reduce the reliability of these parts. To allow the influence of these microstructural features over the load bearing capacity to be considered in the design stage of AM structures, in the present paper we developed a multiscale numerical approach. The goal of this approach is to predict the failure probability of AM structures subjected to time-varied loadings, and it is realized through combining statistical homogenization, shakedown analyses, and reliability methods such as first-order reliability method and Monte Carlo simulation. Based on this strategy, we employed statistics of material parameters obtained from micromechanical models as inputs, implemented numerical tools and applied them to two exemplary structures, the plate with a hole and an aircraft bracket. Through a series of case studies carried out using different methods and under different assumptions of material randomness, the paper confirmed the robustness of the derived results and explained the mechanism of how micropores influence the structural reliability. The method developed in this paper can be a viable means for the design and optimization of metallic and metal matrix composite structures produced by AM techniques.

Multiscale beam model for simulating fracture in laminated glass structures

This paper presents a novel numerical model for the mechanical behavior of laminated glass structures. The numerical model is designed as the multiscale numerical approach capable of simulating the fracture of laminated glass elements. This approach combines the fine-scale multilayer model and the coarse-scale macro model to achieve accuracy with high computational efficiency. The multilayer model describes the real laminated glass cross-section and further provides the parameters for the macro model that has a monolithic cross-section and behaves according to the constitutive model derived from the fine-scale multilayer model. Thus, using fine-scale material parameters for glass and interlayer, a multiscale approach can increase computational efficiency. The proposed models are discretized using the embedded discontinuity finite element method (ED-FEM) and operator split computations. The discontinuity in the displacement field, typical of cracks as the dissipative mechanism for fracture, is represented with the embedded discontinuity approach. The heterogeneity of glass elements (regarding initial imperfections) is included using five different model setups. The comparison between numerical results and experimental tests (four-point bending tests on laminated glass elements) shows a very good agreement. In addition to these comparisons, simplified engineering approaches for deflection prediction are also tested in the numerical model and compared with experimental results.

Multiscale experimental characterisation of mode-I interfacial fracture behaviour of vitrimer composites

Vitrimer composites with bond- exchange reactions offer significant advantages in terms of repair and recyclability. The mechanical properties of such composites can be designed by modifying the epoxy/anhydride ratio, thus allowing their interface properties to be customised. In this study, the mode-I interfacial fracture properties of vitrimer carbon fibre composites with different epoxy/anhydride ratios are investigated experimentally and numerically. The results show that composites with an epoxy/anhydride group ratio of 1:0.5 exhibit the highest fracture toughness, with a maximum value of 1.6 mJ/mm2. Meanwhile, the analysis of the composites’ microscopic fracture morphology indicates their good bonding performance with carbon fibres. Subsequently, a multiscale numerical approach is developed to simulate the mode-I fracture response of composite laminates and establish a trilinear cohesive model to simulate the macroscopic interfacial fracture behaviour. In this approach, a matrix elastic–plastic constitutive model and the J-integral method are applied to obtain the stress distribution in the fibre-bridging phase via a double cantilever beam fracture test. The macroscopic fracture behaviour obtained via simulation is consistent with the experimental results. Therefore, the proposed multiscale numerical simulation method, which correlates the matrix properties with macroscopic fracture parameters, allows the interlaminar properties of composites with novel matrices to be evaluated effectively.

Modelling the creep behaviour and induced failure of clay rock from microscale viscosity to large-scale time-dependant gallery convergences using a multiscale numerical approach

The large-scale creep behaviour of Callovo-Oxfordian (COx) clay rock is modelled from small-scale viscous mechanisms of the rock using a multiscale numerical approach in the context of radioactive waste repositories. At the mesoscale, the saturated non-homogeneous rock is represented in digital 2D Representative Elementary Areas (REAs) as a cracked composite material consisting of rigid elastic mineral inclusions (quartz, calcite, and pyrite) embedded in a clay matrix. The modelling of these materials is considered within double-scale finite element framework, in which the homogenised responses of the REA to the enforced global kinematics serve as a numerical constitutive relation at the macroscale. To reproduce the rock creeping, two viscous mechanisms have been introduced to consider the creep of the clay matrix: either the viscoplasticity of the clay aggregates or the viscoelasticity of their contacts, or both. Firstly, laboratory biaxial creep tests of clay rock samples are simulated. A three-stage creep stage is reproduced and the creep failure process is discussed. Then, large-scale underground engineering structures are modelled to reproduce its time-dependent behaviour. It is found that the developed multiscale model is able to provide some valuable insights into the large-scale creep behaviour of clay rocks through the morphological and material small-scale characterization of REA.

Multiscale particle simulation of the temporal evolution of heat flux into poloidal gaps of castellated divertor with edge-localized modes

Edge-localized modes (ELMs) produce an intense transient heat flux on castellated divertor mono-blocks (MBs). The parallel transport of electrons and ions from the ELM burst varies due to their different velocities, which significantly influences the power load on the MBs. This study shows that two heat load phases are caused by an ELM burst. In the first phase, the horizontal surface withstands the intense heat flux because the faster electrons generate a high sheath potential drop. The leading edge of the MBs suffers a great power load from energetic ions in the second phase. This study applies a multiscale hybrid numerical approach to investigate the temporal evolution of heat flux on the poloidal gap from the ELM source in the scrape-off layer to the castellated divertor surface. Our results indicate that the power load on both the MB surface and the leading edge should be considered in the future design of castellated divertors.

Woven ceramic matrix composite surrogate model based on physics-informed recurrent neural network

• Recurrent neural network-based surrogate model for nonlinear, non-monotonic CMC constitutive response presented. • Physics-informed constraints employed through regularization to enforce requisite physics. • Surrogate model accurately represents non-monotonic, included reversed loading, behavior of multiple textile CMC material systems. • Efficiency over multiscale numerical approach improved several orders of magnitude. A recurrent neural network (RNN) based surrogate model is developed to emulate the nonlinear constitutive behavior of woven ceramic matrix composites (CMCs) driven by matrix damage at multiple length scales. Physics-informed constraints are introduced into the surrogate model through regularization to ground the prediction in physics and improve its predictive capabilities. Training data is generated using the multiscale generalized method of cells (MSGMC) approach coupled with a matrix damage model. This coupling permits simulating the nonlinear behavior of woven CMCs based on constituent response at the micro-, meso-, and macroscales. The multiscale repeating unit cell is loaded under non-monotonic conditions including multiple load / unload cycles and tension / compression. The fiber volume fraction as well as the intra- and intertow void volume fractions are also varied in the generation of training data. Therefore, the RNN-based surrogate model is tasked with predicting, as a function of variable input strain sequence and fiber and void volume fractions, the resulting stress versus strain response while satisfying physical constraints such as positive semi-definiteness of the tangent stiffness matrix and linear elastic unloading. The trained surrogate model effectively matches the stress versus strain response and successfully predicts the tangent modulus throughout the loading regime. Neural network based surrogate models can offer efficient alternatives to running computationally intensive multiscale material models to simulate the nonlinear response of large structural models. Therefore the presented work provides evidence towards the feasibility of developing, training, and running such models for CMCs with complex architectures, nonlinear multiaxial material response, and under non-monotonic loading conditions.

Finite Element-Based Numerical Simulations to Evaluate the Influence of Wollastonite Microfibers on the Dynamic Compressive Behavior of Cementitious Composites.

This paper investigates the dynamic compressive behavior of wollastonite fiber-reinforced cementitious mortars using multiscale numerical simulations. The rate dependent behavior of the multiphase heterogeneous systems is captured in a multiscale framework that implements continuum damage towards effective property prediction. The influence of wollastonite fiber content (% by mass) as cement replacement on the dynamic compressive strength and energy absorption capacity is thereafter elucidated. An average compressive strength gain of 40% is obtained for mortars with 10% wollastonite fiber content as cement replacement, as compared to the control mortar at a strain rate of 200/s. The rate dependent constitutive responses enable the computation of energy absorption, which serves as a comparative measure for elucidating the material resistance to impact loads. Approximately a 45% increase in the dynamic energy absorption capacity is observed for the mixture containing 10% wollastonite fibers, as compared to the control case. Overall, the study establishes wollastonite fibers as a sustainable and dynamic performance-enhanced alternative for partial cement replacement. Moreover, the multiscale numerical simulation approach for performance prediction can provide an efficient means for the materials designers and engineers to optimize the size and dosage of wollastonite fibers for desired mechanical performance under dynamic loading conditions.

Radiative behavior of low-porosity ceramics: II–Experimental and numerical study of porous alumina up to high temperatures

We compare two methods of characterizing the radiative behavior of 22% porous sintered alumina disks containing μm-sized pores on the 1 µm to 7 µm wavelength range up to very high temperatures. The first consists of direct spectroscopic measurements of the normal-hemispherical reflectance and transmittance at room temperature, as well as the normal emittance up to 1300 ∘C. The second is a two-step multi-scale numerical approach: the volume radiative properties and surface reflectivity are first determined from physical optics computations on tomography-reconstructed microstructures, then applied to the resolution of the radiative transfer equation on an equivalent homogeneous medium to obtain the spectral reflectance, transmittance, and emittance. Uncertainties in the numerical results, mainly due to microstructural variability, are quantified through sensitivity studies. The good agreement between the experimental and numerical results confirm the ability of our proposed numerical approach to give accurate predictions of the high-temperature radiative behavior of porous ceramics.

Dynamic Response of Multitower Suspension Bridge Deck Pavement under Random Vehicle Load

Recently, the multitower suspension bridge has been widely used in long-span bridge construction. However, the dynamic response of the deck and pavement system of the multitower suspension bridge under random vehicle load is still not clear, which is of great significance to steel-bridge deck pavement (SBDP) design and construction. To reveal the mechanical mechanism of the steel-bridge deck pavement of the multitower suspension bridge under traffic load, this paper analyzed the mechanical response of the pavement based on case study through the multiscale numerical approach and experimental program. Firstly, considering the full-bridge effect of the multitower suspension bridge, the finite element model (FEM) of the SBDP composite structure was established to obtain key girder segments. Secondly, the influences of pavement layer, bending moment and torque, random traffic flow, and bridge structure on the stress of the girder segment were analyzed. Thirdly, the mechanical response of the pavement layer to the orthotropic plate under random vehicle load was studied. Finally, a full-scale model of the experimental program was established to verify the numerical results. Results show that (1) the pavement layer reduced the stress of the steel-box girder roof by about 10%. In the case of adverse bending moment and torque, the longitudinal and transverse stresses of the pavement layer were mainly concentrated in the stress concentration area near the suspender. Under the action of the random vehicle flow, the stress response of the pavement layer was increased by 40% compared with that under standard load. (2) Three-tower and two-span bridge structures have a great influence on the vertical deformation of the pavement layer under the action of vehicle load. Thus, the pavement material needs to have great deformation capacity. (3) The full-bridge effect has a significant influence on the longitudinal stress of the local orthotropic plate, but a small influence on the transverse stress. (4) There is a good correlation between the experimental measurement results of the full-size model and that of the numerical model. The research results can provide guidance for SBDP design and construction of the multitower suspension bridge.

Influence of surface roughness on cold formability in bending processes: a multiscale modelling approach with the hybrid damage mechanics model

Experimental and numerical investigations on the description of cold formability of extra abrasion-resistant steel considering surface roughness effects were performed in this study. A novel multiscale numerical approach to quantitatively evaluate the impacts of surface roughness on the cold formability/bendability of heavy plates was proposed and verified. The macroscopic ductile damage behavior of the investigated steel was described by a hybrid damage mechanics model, whose parameters were calibrated by notched round-bar (NRB) tensile tests and single-edge notched bending (SENB) tests. The surface roughness was characterized by confocal microscopy and statistically incorporated into a two-dimensional representative volume element (RVE) model. For the assessment of the bendability of heavy plates in the component level, the critical ratio between the punch radius and the sample thickness r/t in three-point bending tests was predicted and compared with experimental results. After the surface roughness effects were taken into consideration, a significant improvement in the predicted results was achieved. A good match between the simulation and experimental results confirmed the indispensable influences of surface roughness on the bendability of steels and validated the efficiency of the multiscale simulation approach in the quantitative description of surface roughness affected ductile damage evolutions.

Ultra-thin and ultra-strong organic interphase in nanocomposites with supercrystalline particle arrangement: Mechanical behavior identification via multiscale numerical modeling

A key challenge in the development of inorganic-organic nanocomposites is the mechanical behavior identification of the organic phase. For supercrystalline materials, in which the organic phase ranges down to sub-nm areas, the identification of the organic materials' mechanical properties is however experimentally inaccessible. The supercrystalline nanocomposites investigated here are 3D superlattices of self-assembled iron oxide nanoparticles, surface-functionalized with crosslinked oleic acid ligands. They exhibit the highest reported values of Young's modulus, nanohardness and strength for inorganic-organic nanocomposites. A multiscale numerical modeling approach is developed to identify these properties using supercrystalline representative volume elements, in which the nanoparticles are arranged in a face-centered cubic superlattice and the organic phase is modeled as a thin layer interfacing each particle. A Drucker-Prager-type elastoplastic constitutive law with perfectly plastic yielding is identified as being able to describe the supercrystals' response in nanoindentation accurately. As the nanoparticles behave in a purely elastic manner with very high stiffness, the underlying constitutive law of the organic phase is also identified to be Drucker-Prager-type elastoplastic, with a Young's modulus of 13 GPa and a uniaxial tensile yield stress of 900 MPa, remarkably high values for an organic material, and matching well with experimental and DFT-based estimations. Furthermore, a sensitivity study indicates that small configurational changes within the supercrystalline lattice do not significantly alter the overall stiffness behavior. Multiscale numerical modeling is thus proven to be able to identify the nanomechanical properties of supercrystals, and can ultimately be used to tailor these materials' mechanical behavior starting from superlattice considerations.

Structural Screening and Design of Platinum Nanosamples for Oxygen Reduction

Nanocatalyst-by-design promises to empower the next generation of electrodes for energy devices. However, current numerical methods consider individual and often geometrical closed-shell nanoparticles, neglecting how the coexistence of several and structurally diverse isomers in a sample affect the activity of the latter. Here, we present a multiscale numerical approach to calculate, in a fast and high-throughput fashion, the current density and mass activity of individual isomers, as well as predict the activity of morphologically diverse but size-selected samples. We propose specific design rules of platinum nanosamples for the electrochemical reduction of molecular oxygen, identifying the size range up to 5.5 nm as the one where isomerization of individual nanoparticles and the morphological composition of the sample cannot be neglected. We confirm a peak of the activity of defected and concave polyhedra at 2–3 nm while spherical but amorphous isomers become the most active in the range of 3–5 nm, with an astonishing mass activity of 2.7 A/mg. We provide a possible explanation to rationalize the discrepancies in the measured mass activity of size-selected samples, in terms of the different distributions of Pt isomers in each specimen.

Investigation of flow regimes in arc plasma–gas interactions using a two-temperature arc in crossflow model

The perpendicular impingement of a gas stream on an electric arc, a configuration known as the arc in crossflow, is of primary relevance in the study of plasma–gas interactions as well as in industrial applications such as circuit breakers and wire-arc spraying. The flow dynamics in the arc in crossflow are the result of coupled fluid-thermal-electromagnetic phenomena accompanied by large property gradients, which can produce significant deviations from Local Thermodynamic Equilibrium (LTE) among electrons and gas species. These characteristics can lead to the establishment of distinct flow regimes depending on the relative values of the controlling parameters of the system, such as inflow velocity, arc current, and inter-electrode spacing. A two-temperature non-LTE model is used to investigate the arc dynamics and the establishment of flow regimes in the arc in crossflow. The plasma flow model is implemented within a nonlinear Variational Multiscale (VMS) numerical discretization approach that is less dissipative and, hence, better suited to capture unstable behavior than traditional VMS methods commonly used in computational fluid dynamics simulations. The Reynolds and the Enthalpy dimensionless numbers, characterizing the relative flow strength and arc strength, respectively, are chosen as the controlling parameters of the system. Simulation results reveal the onset of dynamic behavior and the establishment of steady, periodic, quasi-periodic, and chaotic or potentially turbulent regimes, as identified by distinct spatiotemporal fluctuations. The computational results reveal the role of increasing the relative arc strength on enhancing flow stability by delaying the growth of fluctuating and unstable flow behavior.

Fracture response of metallic particulate-reinforced cementitious composites: Insights from experiments and multiscale numerical simulations

This paper presents an experimental and numerical investigation into the fracture response of mortars containing up to 30% waste iron powder by volume as OPC-replacement. The iron powder-modified mortars demonstrate significantly improved strength and fracture properties as compared to the control mortars due to presence of elongated iron particulates in the powder. With a view to develop a predictive tool towards materials design of such particulate-reinforced systems, fracture responses of iron powder-modified mortars are simulated using a multiscale numerical approach. The approach implements multi-scale numerical homogenization involving cohesive zone-based damage at the matrix-inclusion interface and isotropic damage in the matrix to obtain composite constitutive response and fracture energy. Consequently, these results serve as input to macro-scale XFEM-based three-point-bend simulations of notched mortar beams. The simulated macroscopic fracture behavior exhibit excellent match with the experimental results. Thus, the numerical approach links the material microstructure to macroscopic fracture parameters facilitating microstructure-guided material design.

Multi-scale fracture analysis of fibre-reinforced composites

Fibre reinforced composite (FRC) have been widely used in engineering applications for the purpose of strengthening or lightening of the structures. Many modern applications such as aircraft structures, automotive components, ships, yacht, and infrastructures etc. are fabricated by FRC materials. Since the FRC is the homogenized mixture of fibre and resin (matrix). The structural performance of the FRC materials depends on the fibre arrangement in the matrix, as well as FRC constituents. Therefore, an accurate prediction of structural response of this material needs prior information of micro to macro-level material properties. This paper presents a multi-scale numerical approach to predict the fracture response of fibre reinforced composites. At the micro-level, analysis of representative volume element (RVE) using DIGIMAT software has been presented to describe the microstructure related material properties. In this analysis, fibre is described as a linear elastic material while the matrix phase is modelled by an exponential linear law. The unknown constitutive relationship at the macro-scale has been obtained by DIGIMAT-FE analysis. Further, the macro-scale constitutive relation is used to predict fracture behaviour of FRC. Virtual crack closure technique (VCCT) of Finite Element Method has been implemented to study fracture parameters of FRC. All presented computational results are obtained by DIGIMAT-FE analysis and ABAQUS software. In order to check the robustness and accuracy of the presented approach, few benchmark problems were solved at the multi-scale level and compared with the reference results. In addition, edge crack and centre crack problems were solved by the multi-scale approach to analyse the fracture response of FRC.

A multiscale numerical approach for the finite strains analysis of materials reinforced with helical fibers

Present study provides a multiscale numerical approach based on representative volume elements (RVE) for the finite strain analyses of materials reinforced with helical fibers. An RVE with wavy-like boundaries bioinspired in the microstructure of tendon fascicles is proposed. Due to the unusual geometry of the RVE, a non-periodic mesh mapping will likely occur, precluding the numerical implementation of the periodic boundary condition in a straightforward manner. Moreover, it is verified that the others classical boundary conditions, namely, the linear boundary displacements model and the minimally constrained model, seem not to be suitable choices for the multiscale analyses of this class of RVEs. Motivated by these facts, two mixed boundary conditions allying characteristics of both, linear and minimal models, are suggested. The kinematic constraints on the RVE are enforced via variational principles and Lagrange multipliers. A displacement-controlled triaxial test performed on a numerical specimen larger than the RVE is proposed as a reference solution for the multiscale responses of the RVE. A set of numerical results concerning microscopic strain fields and macroscopic stress-stretch curves points out that one of the proposed mixed models predicts with great accuracy not only the homogenized quantities but also the kinematic fields developed within the specimen. The computational homogenization strategy addressed in this manuscript can be extended to other fiber-reinforced materials composed of different fiber arrangements also including dissipative effects. Moreover, the boundaries surfaces of the RVE are not restricted to any particular layout and the finite element mesh do not require particular mappings nor additional algorithmic handling.

Emplacement of metamorphic core complexes and associated geothermal systems controlled by slab dynamics

Slab rollback results in the development of low-angle normal faults (detachments) and metamorphic core complexes (MCCs) in back-arc domains. Although the mechanical consequences of slab dynamics on lithospheric and crustal behaviors have already been studied, thermal effects have not been investigated yet. This study shows that slab rollback produces lithospheric-scale thermal perturbations intrinsically associated with emplacement of amagmatic high-enthalpy geothermal systems. Using a multi-scale numerical modeling approach, with lithospheric-scale 3-D thermo-mechanical models of subduction, and 2-D models of fluid flow at the scale of detachments, we demonstrate that subduction-induced extensional tectonics controls the genesis and distribution of crustal-scale thermal domes from the base of the crust, and the location of high-energy geothermal systems. We find that when slab tearing occurs, Moho temperatures can temporarily increase by up to 250 °C due to significant shear heating in the flowing upper mantle. Associated thermal anomalies (with characteristic width and spacing of tens and hundreds of km, for crustal and lithospheric scales, respectively) then migrate systematically toward the retreating trench. These thermal domes weaken the crust, localize deformation and enhance the development of crustal-scale detachments. These thermo-mechanical instabilities mimic genesis of high-temperature MCCs with migmatitic cores in the back-arc domain, such as those of the Menderes (western Anatolia, Turkey) and Larderello (southern Tuscany) provinces in the Mediterranean realm, and those in the Basin and Range (western United States), where detachments control the bulk of the heat transport. At the scale of MCCs, the bulk fluid flow pattern is controlled by topography-driven flow while buoyancy-driven flow dominates within the permeable detachments, focusing reservoir location of high-energy geothermal systems at shallow depth beneath the detachments.

A method to numerically predict the loading ratio dependency of long crack propagation rates under cyclic loading

The investigation of cyclic crack propagation rates (CPR) for different loading ratios is very elaborate. A multiscale numerical approach to predict high cycle fatigue (HCF) strength for a ferritic-pearlitic steel has been modified to be applicable for the numerical calculation of its cyclic CPR. The original approach involves a series of numerical and mechanism-based analytical models. Microstructural features of the material are statistically characterized for the generation of synthetically representative volume element (RVE) models of the microstructure. A combined isotropic and kinematic hardening model for crystal plasticity is used to include the microdeformation behavior under cyclic loading in the model. The simulation of cyclic loading with numerous of these statistically equivalent RVEs results in fatigue indicator parameter (FIP) fields which differ from one RVE to the other. With regard to the weakest link theory, only the highest FIP in each RVE is extracted for further fatigue calculations. Theses FIPs are distributed by an extreme value distribution function which can be uniquely described by two parameters. These parameters are used to determine the crack propagation rate. For the comparison of results from experiment and simulations, a concept is proposed which links the globally applied stress amplitude on the RVEs to the cyclic stress intensity factor. The model needs to be calibrated by cyclic CPR results conducted with arbitrarily chosen loading ratios. The validation of the calibrated model is achieved by the prediction and comparison of cyclic CPR at a different loading ratio. The obtained numerical values are in good agreement with the experimental results.

A multiscale numerical simulation approach for chloride diffusion and rebar corrosion with compensation model

Refined analysis depicting mass transportation and physicochemical reaction and reasonable computing load with acceptable DOFs are the two major challenges of numerical simulation for concrete durability. Mesoscopic numerical simulation for chloride diffusion considering binder, aggregate and interfacial transition zone is unable to be expended to the full structure due to huge number of DOFs. In this paper, a multiscale approach of combining both mesoscopic model including full-graded aggregate and equivalent macroscopic model was introduced. An equivalent conversion of chloride content at the Interfacial Transition Layer (ITL) connecting both models was considered. Feasibility and relative error were discussed by analytical deduction and numerical simulation. Case study clearly showed that larger analysis model in multiscale model expanded the diffusion space of chloride ion and decreased chloride content in front of rebar. Difference for single-scale simulation and multiscale approach was observed. Finally, this paper addressed some worth-noting conclusions about the chloride distribution and rebar corrosion regarding the configuration of rebar placement, rebar diameter, concrete cover and exposure period.