Multiscale Science and Engineering

, Volume 1, Issue 2, pp 108–118 | Cite as

Evaluation of the Moisture Effect on the Material Interface Using Multiscale Modeling

  • Renyuan Qin
  • Denvid LauEmail author


Layered material systems are widely seen in various engineering applications such as thin films circuit boards in electronic engineering, lipid bilayer in biological engineering, and adhesive bonding in aerospace and civil engineering applications. However, the durability of the material interface can be seriously affected due to the prolonged exposure to water. Although the experimental studies have shown the reduction in terms of ultimate bond strength and fracture toughness for material interface, the shift in failure mode found in experiment cannot be explained using conventional fracture theory, which is related to the interaction between the water and material interface. To understand the debonding mechanism from a fundamental and comprehensive aspect and bridge knowledge from the atomistic scale to continuum scale, multiscale modeling approach has been proposed to study the debonding behavior of material interface under moisture effect. A number of studies have been conducted using multiscale modeling approach to investigate the debonding of material interface, and it is necessary to summarize these studies to understand the role of water molecules in weakening and diffusing at the material interface using different atomistic models, force fields and upscaling techniques. This paper provides a comprehensive review on the multiscale modeling of interfacial and delamination behavior of layered material system under moisture attack with the focus on the molecular dynamics simulation and finite element modeling. The FRP bonded concrete system is used as a representative to demonstrate the approach of multiscale modeling. The future research direction is recommended, which involves the consideration of roughness of substrate and structural voids at interface for the better understanding of durability issue for interface in layered material system under different environmental conditions.


Interface Multiscale modeling Molecular dynamics simulation Finite element modeling 


Layered material system is widely seen in different engineering aspects such as the thin films circuit boards in electronic engineering, the lipid bilayer in biological engineering, and the adhesive bonding in aerospace and civil engineering applications. The integrity of layered material systems is governed by the interfacial property of bonded materials. The interfacial delamination is one of the most critical issues in the laminated material systems, such as the delamination of epoxy/copper interface in plastic integrated circuit (IC) packages, the detachment of enamel/polymer interface in dental medicine and debonding of fiber reinforced polymer (FRP) from concrete substrate in repairing or strengthening construction structures. The service condition has significant impact on the integrity of interface in such composite system including environmental temperature or moisture levels. The FRP bonded concrete system is a good example to demonstrate such interfacial degradation, since the FRP bonded concrete system is subject to the various environmental loads, e.g. moisture penetration, temperature change, chloride ion penetration, and alkalinity in its service life. Among these environmental effects, the moisture penetration is reported as one of the most critical factors which significantly reduces the load bearing capacity of the reinforced element from its designed value and causes unexpected failure of structures [1, 2, 3, 4, 5, 6, 7, 8, 9]. Under the moisture condition, deterioration of the mechanical property of composite system could be caused by the strength reduction of the FRP composites, plasticization of the polymer matrix or the degradation of interface between adhesion and substrate [10, 11, 12, 13, 14, 15].

As a representative, the FRP bonded system can be regarded as a layered composite system consisting three components, namely FRP, epoxy and concrete substrate. It has been reported that the concrete/epoxy interface is more critical than FRP/epoxy interface, when the composite system is subjected to moisture penetration [16, 17]. Moreover, it was found that although the crack propagation in the dry specimens could be predicted by the fracture theory according to the material properties, i.e. Young’s modulus and Poisson’s ratio, it failed to explain the strength reduction and the shift in failure mode for wet specimens, in which the interaction at the material level between water, epoxy and concrete could result in the variation of the material and interfacial properties [18]. To further understand the interfacial debonding mechanism under the presence of moisture, a more fundamental approach which can capture the molecular interaction and the change of interfacial property at small scales is needed. Recently, atomistic modeling approach employing molecular dynamics (MD) simulation is studied to address the above issues [19, 20, 21].

Molecular dynamics simulation is a powerful tool to perform simulations at molecular level, which can be used to study the molecular behavior of different substances and the mechanics of interfaces from a fundamental perspective [22, 23]. Moreover, it can provide the information about the mechanical behavior of the interfacial region by monitoring the interaction and deformation of the materials, so that the integrity of the bonded materials can be evaluated comprehensively. However, because of the limitation of computational power, MD simulation is mainly feasible for nanoscale, and the deformation mechanisms at interface can be various from different length scales. This fact requires a robust method that can translate the knowledge obtained from MD simulation to macroscale models such as finite element model, so that a prediction of mechanical behavior of materials and structural behavior of composite system can be achieved using a bottom-up approach [24, 25].

Recently multiscale modeling methods have been proposed to bridge the knowledge from different length scales from MD simulation to finite element simulation [26, 27]. The results from MD simulation, such as surface energy and intrinsic strength, can be converted to continuum interfacial properties by using traction–separation law for cohesive zone elements, which is widely used for modeling the interfacial fracture behavior of layered composite material system. The macroscale simulation can be conducted through finite element method to predict the structural behavior using the interfacial behavior captured by MD simulation. This multiscale modeling method enables the researchers to bridge the continuum and atomistic scale modeling and understand the structural behavior under the environmental attack from a fundamental and comprehensive aspect [28, 29, 30, 31, 32, 33].

The primary objective of this article is to provide a comprehensive review on the multiscale modeling of interfacial and structural behavior of layered material system under moisture attack with the focus on the molecular dynamics simulation and finite element simulation, and the FRP bonded system is used as the representative to demonstrate the multiscale modeling approach. The experimental studies towards the evaluation of epoxy/concrete interface under moisture attack are firstly discussed, in which the conventional fracture theory fails to explain the shift in failure mode in experimental findings, and interaction between water and materials is needed to be considered for the understanding of debonding mechanism in a multiscale manner. The MD simulations in terms of investigating the effect of water molecule on epoxy/concrete interface are reviewed, and upscaling techniques are provided to transfer the knowledge from nanoscale to macroscale scale. Besides, the finite element modeling using the interfacial behavior captured by MD simulation is discussed with a case study on the prediction of structural behavior of FRP bonded concrete under both dry and wet condition. This review work would enrich the understanding of debonding mechanism of layered material system under moisture attack at different length scales, and it could benefit the interested researchers to further use the multiscale modeling method to investigate the durability issues of such material system under different environmental conditions.

Moisture Effect on the Material Interface: FRP Bonded Concrete System

Experimental Studies on the FRP Bonded Concrete Under Moisture Attack

The FRP bonded concrete is a good example to study the delamination of layered material, since it is a typical laminated composite system consisting FRP, epoxy and concrete, and such structures are commonly found under environmental moisture attack. As FRP materials are now widely consumed in civil infrastructure application for strengthening or retrofitting concrete structures, the debonding between FRP and concrete substrate has been a critical issue of interdisciplinary research because of its significant impact on the structural performance of reinforced members. To investigate and quantify the interfacial degradation between FRP and concrete, two approaches are commonly used, i.e. strength-based approach and fracture-based approach. Strength-based approach adopts the calculation of the bond stress distribution in FRP reinforced elements based on the elastic material properties to predict the debonding between FRP and concrete substrate by comparing the calculated maximum stress with the strength of the materials [34, 35, 36, 37, 38, 39, 40]. ACI Committee 440 recommends guidelines on the strain level developed in the FRP to prevent debonding failures [41, 42], which is given as follow:
$$\varepsilon_{\text{fe}} = \varepsilon_{\text{cu}} \left( {\frac{h - c}{c}} \right) - \varepsilon_{\text{bi}} \le {\text{K}}_{\text{m}} \varepsilon_{\text{fu}}$$
where \(\varepsilon_{\text{fe}}\) is the maximum allowed effective strain in FRP, \(\varepsilon_{\text{fu}}\) is the ultimate strain in the FRP, \(\varepsilon_{\text{cu}}\) is the ultimate strain of concrete, h and c are the beam height and neutral axis depth, respectively, \(\varepsilon_{\text{bi}}\) is the concrete strain when the FRP reinforcement installs, and the \({\text{K}}_{\text{m}}\) is the limiting strain coefficient related to the layers, thickness and the elastic modulus of the FRP reinforcement. From the Eq. (1), it is shown that in the current design code developed by strength-based approach, the geometric and elastic material properties are the main concerns, which is feasible to predict the short-term performance of the FRP bonded system. However, the failure process of crack initiation and propagation at local debonding regions are intrinsically neglected in the strength-based approach. The fracture-based approach makes use of the interfacial fracture toughness, Γ, to quantify the debonding between FRP and concrete based on the kink criterion, which is considered as a bond property of the multilayer material system [43, 44]. This approach is considered to be robust in determining the debonding mechanism of FRP bonded system since it is able to predict the crack initiation and propagation at a local region by assessing the bonding property between multilayers on bond lines, in which case the real scale structures can be represented by mesoscale interface fracture specimens.

Interfacial Degradation

Externally bonded FRP system usually involves three materials, i.e. FRP, epoxy and concrete, and two interfaces, i.e. FRP/epoxy and epoxy/concrete. The failure modes of the system can always be summarized into following categories, namely (a) FRP delamination; (b) FRP/epoxy separation; (c) epoxy decohesion; (d) epoxy/concrete separation; and (e) concrete substrate fracture. It is reported that the debonding at epoxy to concrete interface is the vulnerable and the most critical in FRP bonded concrete system under moisture condition, which are shown in Fig. 1.
Fig. 1

Idealization and definition of debonding scenarios, including FRP delamination, FRP/epoxy separation, epoxy decohesion, epoxy/concrete separation and concrete substrate fracture. Among this debonding scenarios, the debonding at epoxy to concrete interface in the most critical under moisture effect

To evaluate the debonding behavior of FRP bonded system under moisture condition, experimental tests have been carried out using peel and shear test to determine the interface fracture toughness [17, 18]. The interface fracture toughness was calculated by specialized fracture energy release rate, G. The interface fracture toughness values can be computed by following equations:
$$\varGamma_{\text{peel}} = \frac{{N_{{{\text{y}},{ \hbox{max} }}}^{2} }}{{B^{2} \bar{E}_{3} h^{3} }}\left( {\frac{{(l + a)^{2} - l^{2} }}{{2I_{1} }} + \frac{{6l^{2} }}{{\mathop \sum \nolimits_{13} }}} \right)$$
$$\varGamma_{\text{shear}} = \frac{{N_{{{\text{x}},{ \hbox{max} }}}^{2} }}{{2B^{2} \bar{E}_{3} }}\left( {\frac{1}{{A_{1} h}} + \frac{{\varPsi^{2} }}{{I_{1} h^{3} }}} \right)$$
where \(\varGamma_{\text{peel}}\) and \(\varGamma_{\text{shear}}\) are the interface fracture toughness for the peel fracture and shear fracture, respectively; \(N_{{{\text{y}},{ \hbox{max} }}}^{ }\) and \(N_{{{\text{x}},{ \hbox{max} }}}^{ }\) are the critical peel and shear load, respectively; B is the width of the bond; h is the thickness of FRP; l is the cantilever length; \(a\) is the pre-crack length; \(\bar{E}_{3}\) is the degraded plain strain elastic modulus of concrete; \(\mathop \sum \limits_{13}\) is the ratio of plain strain modulus of FRP to that of concrete; \(A_{1}\) is the dimensionless composite area; \(I_{1}\) is the dimensionless moment of inertia; \(\varPsi\) is the eccentricity function. For the details of the determination of these parameters, the reader is referred to the literature [17, 18].
The experimental results of peel and shear test for FRP bonded system under moisture condition are summarized in Fig. 2 from literature [17, 45]. It is found that in fracture tests, the different amount of reduction in terms of fracture toughness can be seen after the moisture duration. The fracture toughness was reduced to 37% of that in dry case after 8 weeks conditioning in peel test. For shear fracture scenario, the reduction of fracture toughness is less than that in peel fracture scenario, given that the friction between bilayers plays an important role during the fracture process with a treated rough bond surface. However, still 14% reduction of fracture toughness can be seen after 8 weeks moisture condition comparing with that in dry case for shear fracture scenario. It is noticed that the debonding mode of epoxy/concrete separation was consistently observed after 2 weeks moisture conditioning, while the concrete delamination was observed for all dry specimens. However, when the crack kink criterion theory was adopted, although it could predict the crack propagation observed in the dry cases, it failed to explain the reduction of strength and the shift in debonding mode for wet specimens. Such results indicate that this failure mode is due to an interfacial material toughening or an interface weakening mechanism caused by externally moisture penetration, which involves the complex interactions between the water molecules and materials close to the interface. According to these findings, a more fundamental approach should be used to study the debonding mechanism of the interface under moisture penetration, and molecular dynamics simulation is particularly suitable to be used for such study by providing fundamental information for the interaction between the materials at interface.
Fig. 2

Summarized of peeling and shear test results for FRP bonded concrete conditioning in wet environment for 4 and 8 weeks. It is noticed that the debonding model of epoxy/concrete separation was consistently observed after 2 weeks moisture conditioning, while the concrete delamination was observed for all dry specimens

Modeling of Separation at Interface Under Moisture Attack

Molecular Dynamics Simulations

MD simulations have been applied to study the molecular behavior of different substances as a powerful numerical modeling technique [46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61]. According to the previous experimental studies, the critical region of the FRP debonding from substrate under moisture attack locates at the interface between epoxy and substrate. This fact allows the using of MD simulation to model the composite system as a bilayer structure, which consists of epoxy and substrate, to understand the debonding mechanism [62, 63, 64, 65].

MD simulation is based on the modeling of behavior of atoms and molecules explicitly to capture the nature of the matter at atomistic level, which provides a fundamental description of material property and a dynamic evolution of equilibrium and non-equilibrium processes. The interaction between the fundamental units or atoms, i.e. attraction or repulsion, is governed by interatomic potential or force field (FF), which is determined by a function of distance between the atoms. This approach has been used in solving the small scale structural mechanics problems, such as localized fracture processes in materials [66, 67, 68, 69, 70, 71, 72, 73]. One of the challenges in using MD simulation is to build up the atomistic model that can accurately describe the structure and property of the system, and balance the computational amount in a reasonable range.

To capture the accurate interaction between the epoxy and concrete interface under moisture attack and simplify the system in MD simulation, the diglycidyl ether of bisphenol A (DGEBA) [74, 75, 76, 77, 78, 79, 80, 81] is commonly used to represent the molecular structure of epoxy, which is the key functional group in most of the construction-related epoxy possess. For concrete, it is more complex since it is a heterogeneous composite, which consists of hydrated cement and different size of aggregates. It is reported that silica accounts for more than 40% by weight of the solid ingredients of cement and aggregates in concrete, which can be adopted for construction atomistic model of concrete with well-defined chemical formula [19]. Recent studies have shown that crystalline silica (SiO2) can be used to represent both the cement and aggregate as well [82, 83]. After the construction of the atomistic model, the proper force field should be chosen to well describe the interaction between the atoms in the system. Several force fields have been developed in past decades to simulate the behavior of polymers including epoxy-based material, such as ClayFF, DREIDING, PCFF, CFF91 and CVFF [84, 85, 86, 87, 88, 89, 90]. Among these force fields, it is reported that CVFF is capable to describe the atoms behavior of inorganic materials as well, so it is appropriate to be used in the MD simulation to capture the atomic behavior at epoxy and silica interface.

The steered molecular dynamics (SMD) have been performed on the silica-epoxy system for computational peel and shear tests by fixing the silica substrate and applying the pulling force with the velocity ranging from 2 to 1000 m/s [91]. The pulling force was applied to the end carbon atom in the epoxy by a virtual spring with the normal and parallel direction to the interface to simulate the peeling and shear loading. The moisture attack was simulated by the incorporation of water molecule in the simulation box. The simulation results show that for the dry case, the adhesive energy, Eb, was 9.10 kcal/mol and 7.06 kcal/mol for peeling and shear case, respectively. For the wet case, the corresponding adhesive energy was 7.74 kcal/mol and 5.99 kcal/mol for peeling and shear case, which indicated that the adhesive strength between silica and epoxy reduced at 14.9% and 15.2% for peeling and shear case with the presence of water. Moreover, the phenomena that some water molecules can seep into the gap between silica and epoxy under wet condition was captured by molecular dynamics simulation, which is shown in the Fig. 3.
Fig. 3

The snapshots captured by MD simulation at different time for a better understanding on the interaction among epoxy, silica and water: a time = 0 ps; b time = 200 ps; c time = 450 ps in dry condition and d time = 0 ps; e time = 200 ps; and f time = 450 ps in wet condition. The presence of water molecules hinders the attraction between epoxy and concrete (represented by silica), which results in a weaker bond

The failure mode of silica–epoxy system in molecular dynamics simulations is consistent with the results from experimental studies of FRP bonded concrete specimens under prolonged moisture. Results from MD simulations provides an atomistic-based understanding on the reduction of adhesive strength of the interface, which is due to interaction between epoxy and water. However, the simulation results from this study can only describe the considered system correctly on limited length and time scales, which lead to the needs that can expend the knowledge to a larger length scale [92, 93, 94, 95].

The MD simulation has also been applied to study the transport of moisture at silica-epoxy interface [96, 97]. The condensed-phase optimized molecular potential for atomistic simulation studies (COMPASS) force field was applied to the epoxy and silica system with a ratio of 3:7 in this study, and water molecules were directly arranged manually at interface in the system [98, 99]. After an equilibration step, the MD simulations were performed at several temperatures for 2 ns to study the diffusion coefficients. The results show reasonable agreement between experimental and simulation results, and diffusion of water moisture at interface between silica and epoxy is faster than that in the epoxy bulk. Such results further indicate that the interface between silica and epoxy is the critical region under moisture attack [100].


The multiscale approach used for modeling the interface of bilayer system usually involves a systematic determination of interfacial parameters for upscaling. In the nanoscale modeling, the atomistic model can provide the nano-mechanical properties of bonded system using a free energy approach [101, 102]. The surface energy can be characterized by a free energy surface (FES) from an attached stage referring to the lowest free energy state to a detached stage when the separation between two materials is larger than the cutoff distance of the pair potential. The difference in the free energy between the attached and detached stages can be normalized by the corresponding molecular contacting area between two materials, which results an estimate of the upper bound of the surface between two materials. Such procedure can be achieved using the metadynamics methods [103, 104, 105].

After the calculation of surface energy from MD simulation, the interfacial behavior can be then determined for continuum modeling. One method to convert the results from MD simulation to determine the debonding mechanism is using worm-like-chain (WLC) based fracture model [106]. The WLC fracture model is based on the rubber elasticity concept and can be used to describe the interfacial debonding at molecular level as a molecular peeling process from a single polymer chain. A combination of WLC model and fracture approach enable us to obtain the energy release rate (G), which is equal to the surface energy between two material in the bilayer system at the stage just before the onset of rupture, and the intrinsic strength of interface can be then expressed as follow [91]:
$$F_{\text{break}} = \frac{{k_{B} T}}{{4\xi_{\text{p}} }}[(1 - \alpha_{\text{cr}} )^{ - 2} + 4\alpha_{\text{cr}} - 1]$$
where \(k_{B}\) is equilibrium bond length, \(\xi_{\text{p}}\) is the persistence length, \(\alpha_{\text{cr}}\) is the ratio between chain length and contour length when debonding occurs, and T is the temperature set in the simulation. The surface energy and instrinstic strength of interface can be then used to quantify the interfacial behavior of bilayer system, which provides a bridge between the discrete atomistic model and finite element model.
Another upscaling approach is directly modeling the epoxy into bulk material through cross-linking the single epoxy polymer chains [107, 108]. The single epoxy polymer chain is formed from polymerization among the associated monomers and such polymer chain can then cross-link with each other, which leads to a bulk cross-linked structure. An effective dynamic cross-linking algorithm has been studied for epoxy modeling in MD simulation under three force fields, i.e. CVFF, Dreiding and PCFF [109, 110, 111]. The calculated mechanical properties, including Young’s modulus, bulk modulus, shear modulus and Poisson’s ratio, are in good accordance with experimental observables. Such cross-linking process in MD simulation enables us to model the system as two bulk materials. The atoms in MD simulation are normally modeled as point masses, and the potential energy of the system provides the forces on each atom, which can be adopted to determine the acceleration, velocity and positions of each atom. The stress σ for atom i in the region V can be calculated by the relation [112, 113]:
$$\sigma = \frac{1}{V}\mathop \sum \limits_{i \in V} \frac{{\partial \varPhi_{\text{i}} }}{\partial \varepsilon }$$
where \(\varepsilon\) and V are the strain and volume over the region which the stress is calculated, and \(\varPhi_{\text{i}}\) is the potential energy. To capture the interfacial behavior in MD simulation using such approach, a tensile or shear strain can be applied to the model in each simulation step and the displacement should be maintained before the further displacement is applied. By repeating this procedure in MD simulation until the complete separation of the bilayer is achieved, the tensile or shear stress at interface can be calculated from simulation. With the recorded displacement, the stress–displacement relation can be obtained from MD simulation [113, 114, 115].

Macroscale Model Using Finite Element Method

To predict the structural behaviour of FRP bonded in a macroscale manner, the interfacial parameters generated by MD simulation can be converted into finite element simulation using cohesive zone model (CZM), which is widely used in simulating the fracture process in polymers, metals and composites [116, 117, 118]. CZM was first introduced as a technique to study the fracture in quasi-brittle materials such as ceramic and concrete. The mechanical property of CZM is governed by a given traction–separation relationship, which is defined by initial stiffness, damage initiation threshold, and damage evolution properties. In aforementioned WLC fracture model, these parameters can be determined by maximum debonding stress, σth, which can be predicted as Fbreak/s2, where s is the grid length in MD simulation; the fracture energy, Γs, which is predicted by σ th 2 H/2G, where H and G are the thickness and shear modulus of epoxy; and Young’s Modulus, E, which is predicted by kL/A0, where k is the interfacial stiffness, L is distance between epoxy chain and substrate and A0 is the contact area [60].

In applying tensile or shear strain in bulk material in MD simulation, the stress–displacement curve can be directly calculated from simulation results, and this relationship describes the relationship of interfacial traction force and interfacial bonding opening displacement during delamination of the bilayer system, which is exactly the cohesive constitutive relation of the interface. Apart from quantifying the traction–separation relation for modeling the interface from empirical approaches, the interface property converted from MD simulation provides the scientific support in explaining the interaction between two materials at the interfacial region from the physics and/or chemistry point of view, which is a more efficient and fundamental approach to evaluate the structural behavior under moisture or other environmental attack [15, 21, 119, 120, 121].

Case Study of Delamination of FRP Bonded Concrete

To evaluate the structural performance of FRP bonded concrete under moisture attack, we reproduce the finite element model in literature considering the traction–separation law captured from MD simulation with the presence of water molecular [80, 91]. A two-dimensional finite element model has been built, whose dimensions are shown in Fig. 4.
Fig. 4

The schematic diagram of a the dimensions of FRP bonded concrete system simulated in finite element simulation and b the material properties of FRP, epoxy and concrete are determined from experiment, and the interface between epoxy and concrete is modeled based on the results from MD simulation

The properties of bulk materials including concrete, FRP and epoxy are determined from experimental studies using solid elements. Since the study focus on the interfacial debonding, the concrete is modeled as elastic material with the elastic modulus of 26.2 ± 2.2 GPa. The elastic modulus and tensile strength of CFRP laminates are defined as 148 GPa and 2.7 GPa, respectively, according to the experimental measurement in the literature [17, 18]. The elastic and tensile strength of epoxy adhesive are set as 1.5 ± 0.1 MPa and 23.0 ± 2.2 MPa, respectively. The details of the material properties can be found in the literature.

The traction–separation law for cohesive zone elements is defined according to the results from MD simulation for both dry and wet conditions using aforementioned approach, which is arranged at interface between epoxy and concrete. The predicted material properties at epoxy–silica interface at both dry and wet condition is determined from results of MD simulation available in the literature, which is shown in Table 1. It should be noticed that the FES captured at single molecule level can only reflect the interfacial behavior between epoxy and silica system at certain extent, given that there are various defects with unknown sizes along the interface. For the continuum epoxy–silica bilayer system, it can be visualized as a series of cross-linked polymer chains connected to the silica surface through the adhesion at the chain tail, and the free volume in the cross-linked structure near the interfacial region can be regarded as the local heterogeneity or defects at this region. After carefully reviewing the upscaling approaches and techniques, the results from molecular dynamics simulations using a grid system to illustrate the adhesion behavior between cross-link network of epoxy and silica were used [91]. According to the reported results from MD simulations, when the 2 nm grid size is adopted in the molecular model, a good agreement can be achieved between molecular dynamics simulations and experimental results in terms of material and interfacial property [80, 91]. Moreover, 2 nm is also a reasonable distance between adjacent cross-links for a fully cured epoxy, which can reflect the material morphology.
Table 1

Prediction of the material properties at epoxy–silica interface


E (GPa)

σth (MPa)

Γs (J/m2)









The finite element simulation is performed exactly according to the experimental setup for the peeling test of FRP bonded concrete system and the comparisons between experimental and simulation results are shown in Fig. 5.
Fig. 5

Comparison of experimental and simulation results. a The load–displacement curve of FRP bonded system in dry and wet condition determined in experiment and simulation; and b the comparison of ultimate load for experiment and simulation results, indicates that the traction–separation relation conducted by MD simulation is capable to be used in finite element simulation to predict the structural behavior of FRP bonded concrete system in both dry and wet condition

The simulation results show good agreement with the experimental results in terms of ultimate bond strength as well as the structural behavior. The ultimate bond strength from simulation is determined as 132.9 N and 58.9 N for peeling test in dry and wet condition respectively, whose deviation with the experimental results are 10.75% and 17.80% for samples conditioning in dry and wet environment. Such results indicate that the traction–separation law obtained from MD simulation is capable to predict the structural behavior of FRP bonded concrete system under moisture attack. Moreover, it explains the debonding mechanism from atomistic scale fundamentally with the reason why the interface is degraded when water molecule is presented.

Summary and Future Challenges

This paper has provided a review on the multiscale modeling of layered material interface under moisture attack in terms of MD simulations, upscaling techniques and finite element simulations with a specific focus on the interface separation between epoxy and concrete. The multiscale modeling provides a bottom-up approach to predict and understand the interfacial debonding mechanism. However, there is still extensible space for such multiscale approach in modeling the material interface. In most of the MD simulations, the substrate is modeled as crystalline structures, which assumes a perfectly flat interfaces. In fact, the interface of materials will not be intact and smooth, which accelerates the water diffusion into materials interface. Recently, efforts have been made to construct the coarse-grain models parameterized from molecular models at targeted length scale [122, 123, 124, 125]. It will be advantageous to take such coarse-grained models in the multiscale modeling scheme, which can better understand the interfacial behavior of materials with consideration of interfacial roughness and the heterogeneity of the materials. Moreover, the microscopic structural voids generally existing in the epoxy cross-linked networks, which has a detrimental influence on the epoxy mechanical properties [126, 127]. Such structural voids should be considered at different length scale in the multiscale modeling to better understand the interface of porous materials under moisture attack. By considering a more realistic modeling of material interface, it is envisioned that such multiscale modeling scheme can be further developed and adopted in the design and applications of composite structures in different engineering aspects.



The authors are grateful to the support from the Research Grants Council (RGC) in Hong Kong through the General Research Fund (GRF) with the Grant no. 11255616.


  1. 1.
    V.M. Karbhari, M. Engineer, D.A. Eckel II, On the durability of composite rehabilitation schemes for concrete: use of a peel test. J. Mater. Sci. Springer 32(1), 147–156 (1997)Google Scholar
  2. 2.
    J.R. Cromwell, K.A. Harries, B.M. Shahrooz, Environmental durability of externally bonded FRP materials intended for repair of concrete structures. Constr. Build. Mater. 25(5), 2528–2539 (2011)Google Scholar
  3. 3.
    D. Lau, Q. Qiu, A. Zhou, et al., Long term performance and fire safety aspect of FRP composites used in building structures. Constr. Build. Mater. 126, 573–585 (2016)Google Scholar
  4. 4.
    O. Büyüköztürk, D. Lau, C. Tuakta, Durability and long-term performance modelling of Frp-concrete systems. in The 6th International Conference on FRP Composites in Civil Engineering , pp. 13–15 (2012)Google Scholar
  5. 5.
    R. Qin, D. Lau, D, L.H. Tam, et al., Experimental Investigation on Interfacial Defect Criticality of FRP-Confined Concrete Columns. Sensors, 19(3), 468 (2019)Google Scholar
  6. 6.
    R. Sen, Developments in the durability of FRP-concrete bond. Constr. Build. Mater. 78, 112–125 (2015)Google Scholar
  7. 7.
    S. Amidi, J. Wang, Direct measurement of traction–separation law of concrete-epoxy interfaces subjected to moisture attack under mode-I Loading. J. Compos. Constr. 21(5), 4017028 (2017)Google Scholar
  8. 8.
    S. Amidi, J. Wang, Deterioration of the FRP-to-concrete interface subject to moisture ingress: effects of conditioning methods and silane treatment. Compos. Struct. 153, 380–391 (2016)Google Scholar
  9. 9.
    S. Amidi, J. Wang, Subcritical debonding of FRP-to-concrete bonded interface under synergistic effect of load, moisture, and temperature. Mech. Mater. 92, 80–93 (2016)Google Scholar
  10. 10.
    L. Gautier, B. Mortaigne, V. Bellenger, Interface damage study of hydrothermally aged glass–fibre–reinforced polyester composites. Compos. Sci. Technol. 59(16), 2329–2337 (1999)Google Scholar
  11. 11.
    A. Chateauminois, B. Chabert, J.P. Soulier et al., Hygrothermal ageing effects on the static fatigue of glass/epoxy composites. Composites 24(7), 547–555 (1993)Google Scholar
  12. 12.
    Allred, The effect of temperature and moisture content on the flexural response of Kevlar/epoxy laminates. I. [0/90] filament orientation. J. Compos. Mater. 15, 100–116 (1981)Google Scholar
  13. 13.
    A. Pandian, M. Vairavan, W.J. Jebbas Thangaiah et al., Effect of moisture absorption behavior on mechanical properties of basalt fibre reinforced polymer matrix composites. J. Compos. 2014, 1–8 (2014)Google Scholar
  14. 14.
    L.C. Bank, T.R. Gentry, A. Barkatt, Accelerated test methods to determine the long-term behavior of FRP composite structures: environmental effects. J. Reinf. Plast. Compos. 14(6), 559–587 (1995)Google Scholar
  15. 15.
    A. Zhou, C. L. Chow, D. Lau, Structural behaviour of GFRP reinforced concrete columns under the influence of chloride at casting and service stages. Compos. B Eng. 136, 1–9 (2017)Google Scholar
  16. 16.
    O. Büyüköztürk, O. Gunes, E. Karaca, Characterization and modeling of debonding in Rc beams strengthened with Frp composites. in Proceedings of the 15th ASCE Engineering Mechanics Conference: 1–8 (2002)Google Scholar
  17. 17.
    C. Au, O. Büyüköztürk, Peel and shear fracture characterization of debonding in FRP plated concrete affected by moisture. J. Compos. Constr. 10(1), 35–47 (2006)Google Scholar
  18. 18.
    C. Au, Moisture degradation in FRP bonded concrete systems: an interface fracture approach.PhD Thesis. Massachusetts Institute of Technology (2005)Google Scholar
  19. 19.
    O. Büyüköztürk, M.J. Buehler, D. Lau et al., Structural solution using molecular dynamics: fundamentals and a case study of epoxy–silica interface. Int. J. Solids Struct. 48(14–15), 2131–2140 (2011)Google Scholar
  20. 20.
    D. Lau, R.H.W. Lam, Atomistic prediction of nanomaterials: introduction to molecular dynamics simulation and a case study of graphene wettability. IEEE Nanatechnol. Mag. 6(1), 8–13 (2012)Google Scholar
  21. 21.
    Y.L. Yaphary, Z. Yu, R.H.W. Lam et al., Molecular dynamics simulations on adhesion of epoxy-silica interface in salt environment. Compos. B Eng. 131, 165–172 (2017)Google Scholar
  22. 22.
    Y. Mishin, M. Asta, J. Li, Atomistic modeling of interfaces and their impact on microstructure and properties. Acta Mater. 58(4), 1117–1151 (2010)Google Scholar
  23. 23.
    M.J. Buehler, Atomistic modeling of materials failure (Springer, 2008)Google Scholar
  24. 24.
    D. Lau, Moisture-induced debonding in concrete–epoxy interface. HKIE Trans Hong Kong Inst Eng 19(3), 33–38 (2012)Google Scholar
  25. 25.
    H. Salahshoor, N. Rahbar, Nano-scale fracture toughness and behavior of graphene/epoxy interface. J. Appl. Phys. 112(2) (2012)Google Scholar
  26. 26.
    D. Lau, K. Broderick, M.J. Buehler et al., A robust nanoscale experimental quantification of fracture energy in a bilayer material system. Proc. Natl. Acad. Sci. U.S.A. 111(33), 11990–11995 (2014)Google Scholar
  27. 27.
    O. Gunes, D. Lau, C. Tuakta et al., Ductility of FRP-concrete systems: investigations at different length scales. Constr. Build. Mater. 49, 915–925 (2013)Google Scholar
  28. 28.
    E. Lidorikis, M. Bachlechner, R. Kalia, Coupling of length scales: hybrid molecular dynamics and finite element approach for multiscale nanodevice simulations. MRS Online Proc. Library Arch., 653 (2001)Google Scholar
  29. 29.
    B. Shiari, R.E. Miller, D.D. Klug, Multiscale simulation of material removal processes at the nanoscale. J. Mech. Phys. Solids 55(11), 2384–2405 (2007)zbMATHGoogle Scholar
  30. 30.
    S. Ogata, E. Lidorikis, F. Shimojo et al., Hybrid finite-element/molecular-dynamics/electronic-density-functional approach to materials simulations on parallel computers. Comput. Phys. Commun. 138(2), 143–154 (2001)zbMATHGoogle Scholar
  31. 31.
    A. Nakano, M.E. Bachlechner, R.K. Kalia et al., Multiscale simulation of nanosystems. Comput. Sci. Eng. 3(4), 56–66 (2001)Google Scholar
  32. 32.
    C.R. Dandekar, Y.C. Shin, Modeling of machining of composite materials: a review. Int. J. Mach. Tools Manuf.  57, 102–121 (2012)Google Scholar
  33. 33.
    S.W. Chang, Y.-P. Liao, C.-S. Chen et al., Energy and force transition between atoms and continuum in quasicontinuum method. Interact. Multiscale Mech., 7(1), 543–561 (2014)Google Scholar
  34. 34.
    Z. Wu, H. Yuan, H. Niu, Stress transfer and fracture propagation in different kinds of adhesive joints. J. Eng. Mech. 128(5), 562–573 (2002)Google Scholar
  35. 35.
    M.J. Chajes, W.W. Finch, T.F. Januszka et al., Bond and force transfer of composite material plates bonded to concrete. ACI Struct. J. 93(2), 208–217 (1996)Google Scholar
  36. 36.
    B. Täljsten, Strengthening of beams by plate bonding. J. Mater. Civ. Eng. 9(4), 206–212 (1997)Google Scholar
  37. 37.
    U. Neubauer, F. S. Rostasy, Design aspects of concrete structures strengthened with externally bonded CFRP-plates. in Proceedings of the Seventh International Conference on Structural Faults and Repair, pp. 109–118 (1997)Google Scholar
  38. 38.
    A. Khalifa, W.J. Gold, A. Nanni et al., Contribution of externally bonded FRP to shear capacity of RC flexural members. J. Compos. Constr. 2(4), 195–202 (1998)Google Scholar
  39. 39.
    X.Z. Lu, J.F. Chen, L.P. Ye et al., RC beams shear-strengthened with FRP: stress distributions in the FRP reinforcement. Constr. Build. Mater. 23(4), 1544–1554 (2009)Google Scholar
  40. 40.
    J.F. Chen, J.G. Teng, Anchorage strength models for FRP and steel plates bonded to concrete. J. Struct. Eng. 127(7), 784–791 (2001)Google Scholar
  41. 41.
    H.A. Rasheed, Strengthening design of reinforced concrete with FRP. (CRC Press, 2014)Google Scholar
  42. 42.
    ACI 440.2R-08. Guide for the design and construction of externally bonded FRP systems for strengthening existing structures. ACI committee 440 (2008)Google Scholar
  43. 43.
    M.Y. He, J.W. Hutchinson, Kinking of a crack out of an interface. J. Appl. Mech. Am. Soc. Mech. Eng. 56(2), 270–278 (1989)Google Scholar
  44. 44.
    J. Hutchinson, Z. Suo, Mixed-mode cracking in layered materials. Adv. Appl. Mech. 29, 63–191 (1992)zbMATHGoogle Scholar
  45. 45.
    C. Tuakta, O. Büyüköztürk, Deterioration of FRP/concrete bond system under variable moisture conditions quantified by fracture mechanics. Compos. B Eng. 42(2), 145–154 (2011)Google Scholar
  46. 46.
    P. S and J. K,  Computer simulation of liquids. J. Mol. Liquids 38(3–4), 267 (1988)Google Scholar
  47. 47.
    B.J. Alder, T.E. Wainwright, Phase transition for a hard sphere system. J. Chem. Phys. 27(5), 1208–1209 (1957)Google Scholar
  48. 48.
    E. Chiessi, A. Lonardi, G. Paradossi, Toward modeling thermoresponsive polymer networks: a molecular dynamics simulation study of N-isopropyl acrylamide co-oligomers. J. Phys. Chem. B 114(25), 8301–8312 (2010)Google Scholar
  49. 49.
    V. Varshney, S.S. Patnaik, A.K. Roy et al., A molecular dynamics study of epoxy-based networks: cross-linking procedure and prediction of molecular and material properties. Macromolecules 41(18), 6837–6842 (2008)Google Scholar
  50. 50.
    P.V. Komarov, Y.T. Chiu, S.M. Chen et al., Highly cross-linked epoxy resins: an atomistic molecular dynamics simulation combined with a mapping/reverse mapping procedure. Macromolecules 40(22), 8104–8113 (2007)Google Scholar
  51. 51.
    Q. Wang, N.S. Suraweera, D.J. Keffer et al., Atomistic and coarse-grained molecular dynamics simulation of a cross-linked sulfonated poly(1,3-cyclohexadiene)-based proton exchange membrane. Macromolecules 45(16), 6669–6685 (2012)Google Scholar
  52. 52.
    J.D. Monk, J.B. Haskins, C.W. Bauschlicher et al., Molecular dynamics simulations of phenolic resin: construction of atomistic models. Polymer (United Kingdom) 62, 39–49 (2015)Google Scholar
  53. 53.
    S.H. Min, S.K. Kwak, B.-S. Kim, Atomistic simulation for coil-to-globule transition of poly(2-dimethylaminoethyl methacrylate). Soft Matter 11(12), 2423–2433 (2015)Google Scholar
  54. 54.
    G. Paradossi, I. Finelli, F. Natali et al., Polymer and water dynamics in poly(vinyl alcohol)/poly(methacrylate) networks. A molecular dynamics simulation and incoherent neutron scattering investigation. Polymers 3(4), 1805–1832 (2011)Google Scholar
  55. 55.
    P.H. Lin, R. Khare, Molecular simulation of cross-linked epoxy and epoxy− POSS nanocomposite. Macromolecules 42(12), 4319–4327 (2009)Google Scholar
  56. 56.
    A. N. Rissanou, V. Harmandaris, Structure and dynamics of poly(methyl methacrylate)/graphene systems through atomistic molecular dynamics simulations. J. Nanoparticle Res. 15(5) (2013)Google Scholar
  57. 57.
    D.R. Heine, G.S. Grest, C.D. Lorenz et al., Atomistic simulations of end-linked poly(dimethylsiloxane) networks: structure and relaxation. Macromolecules 37(10), 3857–3864 (2004)Google Scholar
  58. 58.
    D.C. Doherty, B. N. Holmes, P. Leung, et al., Polymerization molecular dynamics simulations. I. Cross-linked atomistic models for poly(methacrylate) networks. Comput. Theor. Polym. Sci. 8(1–2), 169–178 (1998)Google Scholar
  59. 59.
    Z. Yu, D. Lau, Flexibility of backbone fibrils in α-chitin crystals with different degree of acetylation. Carbohyd. Polym. 174, 941–947 (2017)Google Scholar
  60. 60.
    Z. Yu, D. Lau, Development of a coarse-grained α-chitin model on the basis of MARTINI force field. J. Mol. Model. 21(5), 128 (2015)Google Scholar
  61. 61.
    Z. Yu, D. Lau, Molecular dynamics study on stiffness and ductility in chitin–protein composite. J. Mater. Sci. 50(21), 7149–7157 (2015)Google Scholar
  62. 62.
    J.A. Zimmerman, D.J. Bammann, H. Gao, Deformation gradients for continuum mechanical analysis of atomistic simulations. Int. J. Solids Struct. 46(2), 238–253 (2009)zbMATHGoogle Scholar
  63. 63.
    R.E. Miller, E.B. Tadmor, Hybrid continuum mechanics and atomistic methods for simulating materials deformation and failure. MRS Bull. 32(11), 920–926 (2007)Google Scholar
  64. 64.
    V.B. Shenoy, R. Miller, E.B. Tadmor et al., Quasicontinuum models of interfacial structure and deformation. Phys. Rev. Lett. 80(4), 742 (1997)Google Scholar
  65. 65.
    G. Scocchi, P. Posocco, A. Danani, et al., To the nanoscale, and beyond! Multiscale molecular modeling of polymer–clay nanocomposites. Fluid Phase Equilibria 261(1–2), 366–374 (2007)Google Scholar
  66. 66.
    C. Zhang, J. Hankett, Z. Chen, Molecular level understanding of adhesion mechanisms at the epoxy/polymer interfaces. ACS Appl. Mater. Interfaces. 4(7), 3730–3737 (2012)Google Scholar
  67. 67.
    X. Periole, S.J. Marrink, The martini coarse-grained force field. Methods Mol. Biol. 924, 533–565 (2013)Google Scholar
  68. 68.
    C.J. Dickson, B.D. Madej, Å.A. Skjevik et al., Lipid14: the amber lipid force field. J. Chem. Theory Comput. 10(2), 865–879 (2014)Google Scholar
  69. 69.
    S.J. Marrink, H.J. Risselada, S. Yefimov et al., The MARTINI force field: coarse grained model for biomolecular simulations. J. Phys. Chem. B 111(27), 7812–7824 (2007)Google Scholar
  70. 70.
    A.C.T. Van Duin, S. Dasgupta, F. Lorant et al., ReaxFF: a reactive force field for hydrocarbons. J. Phys. Chem. A 105(41), 9396–9409 (2001)Google Scholar
  71. 71.
    K. Vanommeslaeghe, E. Hatcher, C. Acharya et al., CHARMM general force field: a force field for drug-like molecules compatible with the CHARMM all-atom additive biological force fields. J. Comput. Chem. 31(4), 671–690 (2010)Google Scholar
  72. 72.
    T.A. Halgren, Merck molecular force field. J. Comput. Chem. 17(5–6), 490–519 (1996)Google Scholar
  73. 73.
    W. Jian, L.H. Tam, D. Lau, Atomistic study of interfacial creep behavior in epoxy–silica bilayer system. Compos. B Eng. 132, 229–236 (2018)Google Scholar
  74. 74.
    R. Berruet, E. Vinard, A. Calle et al., Mechanical properties and biocompatibility of two polyepoxy matrices: DGEBA-DDM and DGEBA-IPD. Biomaterials 8(3), 162–171 (1987)Google Scholar
  75. 75.
    D.G.D. Galpaya, J.F.S. Fernando, L. Rintoul et al., The effect of graphene oxide and its oxidized debris on the cure chemistry and interphase structure of epoxy nanocomposites. Polymer (United Kingdom) 71, 122–134 (2015)Google Scholar
  76. 76.
    K. Li, Y. Li, Q. Lian et al., Influence of cross-linking density on the structure and properties of the interphase within supported ultrathin epoxy films. J. Mater. Sci. 51(19), 9019–9030 (2016)Google Scholar
  77. 77.
    B. Qi, Q.X. Zhang, M. Bannister et al., Investigation of the mechanical properties of DGEBA-based epoxy resin with nanoclay additives. Compos. Struct. 75(1–4), 514–519 (2006)Google Scholar
  78. 78.
    A. Shokuhfar, B. Arab, The effect of cross linking density on the mechanical properties and structure of the epoxy polymers: molecular dynamics simulation. J. Mol. Model. 19(9), 3719–3731 (2013)Google Scholar
  79. 79.
    M. Tackie, G.C. Martin, The polymerization mechanism and kinetics of DGEBA with BF3-MEA. J. Appl. Polym. Sci. 48(5), 793–808 (1993)Google Scholar
  80. 80.
    L. Tam, D. Lau, A molecular dynamics investigation on the cross-linking and physical properties of epoxy-based materials. RSC Adv. 4(62), 33074–33081 (2014)Google Scholar
  81. 81.
    A. Zhou, L.H. Tam, Z. Yu et al., Effect of moisture on the mechanical properties of CFRP-wood composite: an experimental and atomistic investigation. Compos. B Eng. 71, 63–73 (2015)Google Scholar
  82. 82.
    R.J.M. Pellenq, H. Van Damme, Why does concrete set? The nature of cohesion forces in hardened cement-based materials. MRS Bull. 29(05), 319–323 (2004)Google Scholar
  83. 83.
    E.J. Garboczi, D.P. Bentz, Computer simulation of the diffusivity of cement-based materials. J. Mater. Sci. 27(8), 2083–2092 (1992)Google Scholar
  84. 84.
    J.R. Maple, U. Dinur, A.T. Hagler, Derivation of force fields for molecular mechanics and dynamics from ab initio energy surfaces. Proc. Natl. Acad. Sci. U.S.A. 85(15), 5350–5354 (1988)Google Scholar
  85. 85.
    P. Dauber-Osguthorpe, V.A. Roberts, D. J. Osguthorpe, et al., Structure and energetics of ligand binding to proteins: Escherichia coli dihydrofolate reductase-trimethoprim, a drug–receptor system. Proteins Struct. Funct. Bioinf. 4(1), 31–47 (1988)Google Scholar
  86. 86.
    M.J. Hwang, T.P. Stockfisch, A.T. Hagler, Derivation of class II force fields. 2. Derivation and characterization of a class II force field, CFF93, for the alkyl functional group and alkane molecules. J. Am. Chem. Soc. 116(6), 2515–2525 (1994)Google Scholar
  87. 87.
    A.T. Hagler, C.S. Ewig, On the use of quantum energy surfaces in the derivation of molecular force fields. Comput. Phys. Commun. 84(1–3), 131–155 (1994)Google Scholar
  88. 88.
    H. Heinz, T.J. Lin, R. Kishore Mishra et al., Thermodynamically consistent force fields for the assembly of inorganic, organic, and biological nanostructures: the INTERFACE force field. Langmuir 29(6), 1754–1765 (2013)Google Scholar
  89. 89.
    S.L. Mayo, B.D. Olafson, W.A. Goddard, et al., DREIDING: a generic force field for molecular simulations. J. Phys. Chem. 101, 8897–8909 (1990)Google Scholar
  90. 90.
    R.T. Cygan, J.-J. Liang, A.G. Kalinichev, Molecular Models of hydroxide, oxyhydroxide, and clay phases and the development of a general force field. J. Phys. Chem. B 108(4), 1255–1266 (2004)Google Scholar
  91. 91.
    D. Lau, O. Büyüköztürk, M.J. Buehler, Characterization of the intrinsic strength between epoxy and silica using a multiscale approach. J. Mater. Res. 27(14), 1787–1796 (2012)Google Scholar
  92. 92.
    D. Xin, Q. Han, Adhesion reliability of the epoxy-cu interface by molecular simulations. J. Adhes. 91(5), 409–418 (2014)MathSciNetGoogle Scholar
  93. 93.
    H. Mei, Y. Pang, R. Huang, Influence of interfacial delamination on channel cracking of elastic thin films. Int. J. Fract. 148(4), 331–342 (2007)zbMATHGoogle Scholar
  94. 94.
    W. Xie, S.K. Sitaraman, Investigation of interfacial delamination of a copper–epoxy interface under monotonic and cyclic loading: modelling and evaluation. IEEE Trans. Adv. Packag. 26(4), 441–446 (2003)Google Scholar
  95. 95.
    H. Fan, M. M. F. Yuen, A multiscale approach for interfacial delamination in solid-state lighting. in Solid State Lighting Reliability: Components to Systems, pp. 305–316 (2013)Google Scholar
  96. 96.
    O. Hölck, J. Bauer, T. Braun, et al., Transport of moisture at epoxy–SiO2 interfaces investigated by molecular modeling. in Microelectronics Reliability, pp. 1111–1116 (2013)Google Scholar
  97. 97.
    M. H. Shirangi, B. Michel, Mechanism of moisture diffusion, hygroscopic swelling, and adhesion degradation in epoxy molding compounds. in Moisture Sensitivity of Plastic Packages of IC Devices,  (Springer, USA, 2010), pp. 29–69Google Scholar
  98. 98.
    H. Sun, P. Ren, J.R. Fried, The COMPASS force field: parameterization and validation for phosphazenes. Comput. Theor. Polym. Sci. 8(1–2), 229–246 (1998)Google Scholar
  99. 99.
    H. Sun, COMPASS: an ab initio force-field optimized for condensed-phase applications overview with details on alkane and benzene compounds. J. Phys. Chem. B 102(38), 7338–7364 (1998)Google Scholar
  100. 100.
    O. Hölck, J. Bauer, O. Wittler et al., Comparative characterization of chip to epoxy interfaces by molecular modeling and contact angle determination. Microelectron. Reliab. 52(7), 1285–1290 (2012)Google Scholar
  101. 101.
    A. Laio, F.L. Gervasio, Metadynamics: a method to simulate rare events and reconstruct the free energy in biophysics, chemistry and material science. Rep. Prog. Phys. 71(12), 126601 (2008)Google Scholar
  102. 102.
    A. Laio, M. Parrinello, Escaping free-energy minima. Proc. Natl. Acad. Sci. U.S.A. 99(20), 12562–12566 (2002)Google Scholar
  103. 103.
    P. Tiwary, M. Parrinello, A time-independent free energy estimator for metadynamics. J. Phys. Chem. B 119(3), 736–742 (2015)Google Scholar
  104. 104.
    P. Tiwary, M. Parrinello, From metadynamics to dynamics. Phys. Rev. Lett. 111(23) (2013)Google Scholar
  105. 105.
    A. Barducci, M. Bonomi, M. Parrinello, Metadynamics (Computational molecular science, Wiley Interdisciplinary Reviews, 2011)Google Scholar
  106. 106.
    S. Keten, M.J. Buehler, Asymptotic strength limit of hydrogen-bond assemblies in proteins at vanishing pulling rates. Phys. Rev. Lett. 100(19) (2008)Google Scholar
  107. 107.
    S. Yang, F. Gao, J. Qu, A molecular dynamics study of tensile strength between a highly-crosslinked epoxy molding compound and a copper substrate. Polymer (United Kingdom) 54(18), 5064–5074 (2013)Google Scholar
  108. 108.
    C. K. Y. Wong, S. Y. Y. Leung, R. H. Poelma, et al. Molecular dynamics study of the traction-displacement relations of epoxy-copper interfaces. In: 2011 12th Int. Conf. on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems, EuroSimE (2011)Google Scholar
  109. 109.
    H. Sun, Ab initio calculations and force field development for computer simulation of polysilanes. Macromolecules 28(3), 701–712 (1995)Google Scholar
  110. 110.
    K.-S. Chang, T. Yoshioka, M. Kanezashi et al., A molecular dynamics simulation of a homogeneous organic-inorganic hybrid silica membrane. Chem. Commun. 46(48), 9140–9142 (2010)Google Scholar
  111. 111.
    J.L. Asensio, M. Martin-Pastor, J. Jimenez-Barbero, The use of CVFF and CFF91 force fields in conformational analysis of carbohydrate molecules. Comparison with AMBER molecular mechanics and dynamics calculations for methyl α-lactoside. Int. J. Biol. Macromol. 17(3–4), 137–148 (1995)Google Scholar
  112. 112.
    G.C. Sih, Multiscaling in molecular and continuum mechanics: interaction of time and size from macro to nano (Springer, Dordrecht, 2007)Google Scholar
  113. 113.
    N. Iwamoto, M.M.F. Yuen, H. Fan, Molecular modeling and multiscaling issues for electronic material applications, vol. 57 (Springer, London, 2012)Google Scholar
  114. 114.
    X. W. Zhou, N. R. Moody, R. E. Jones, et al. Molecular dynamics study of interfacial cohesive zone law: elastic constant effects. In: ICF12, pp. 1–7 (2009)Google Scholar
  115. 115.
    X.W. Zhou, N.R. Moody, R.E. Jones et al., Molecular-dynamics-based cohesive zone law for brittle interfacial fracture under mixed loading conditions: effects of elastic constant mismatch. Acta Mater. 57(16), 4671–4686 (2009)Google Scholar
  116. 116.
    Z.H. Jin, C.T. Sun, Cohesive zone modelling of interface fracture in elastic bi-materials. Eng. Fract. Mech. 72, 1805–1817 (2005)Google Scholar
  117. 117.
    A. Zhou, R. Qin, L. Feo et al., Investigation on interfacial defect criticality of FRP-bonded concrete beams. Compos. B Eng. 113, 80–90 (2017)Google Scholar
  118. 118.
    R. Qin, A. Zhou, D. Lau, Effect of reinforcement ratio on the flexural performance of hybrid FRP reinforced concrete beams. Compos. B Eng. 108, 200–209 (2017)Google Scholar
  119. 119.
    E.K.L. Chan, H. Fan, M.M.F. Yuen, Effect of interfacial adhesion of copper/epoxy under different moisture level. in 7th International Conference on Thermal, Mechanical and Multiphysics Simulation and Experiments in Micro-Electronics and Micro-Systems, EuroSimE  (2006)Google Scholar
  120. 120.
    N. Iwamoto, Molecularly derived mesoscale modeling of an epoxy/Cu interface: interface roughness. in Microelectronics Reliability, pp. 1101–1110  (2013)Google Scholar
  121. 121.
    N. Burger, A. Laachachi, M. Ferriol, et al., Review of thermal conductivity in composites: mechanisms, parameters and theory. Prog. Polym. Sci. 61, 1–28 (2016)Google Scholar
  122. 122.
    Z. Yu, D. Lau, Evaluation on mechanical enhancement and fire resistance of carbon nanotube (CNT) reinforced concrete. Coupled Syst. Mech. 6(3), 335–349 (2017)Google Scholar
  123. 123.
    D. Lau, Z. Yu, O. Büyüköztürk, Mesoscale modeling of cement matrix using the concept of building block. in Materials Research Society Symposium Proceedings, pp. 13–19 (2015)Google Scholar
  124. 124.
    Z. Yu, A. Zhou, D. Lau, Mesoscopic packing of disk-like building blocks in calcium silicate hydrate. Sci. Rep. 6(1), 36967 (2016)Google Scholar
  125. 125.
    R. Qin, H. Hao, T. Rousakis et al.,  Effect of shrinkage reducing admixture on new-to-old concrete interface. Compos. B Eng. 167, 346–355 (2019)Google Scholar
  126. 126.
    L. Tam, D. Lau, Effect of structural voids on mesoscale mechanics of epoxy-based materials. Multiscale Multiphys. Mech. 1(2): 127–141 (2016)Google Scholar
  127. 127.
    Z. Yu, D. Lau, Nano- and mesoscale modelling of cement matrix. Nanoscale Res. Lett. 10(1), 173 (2015)Google Scholar

Copyright information

© Korean Multi-Scale Mechanics (KMSM) 2019

Authors and Affiliations

  1. 1.Department of Architecture and Civil EngineeringCity University of Hong KongKowloonHong Kong
  2. 2.Department of Civil and Environmental EngineeringMassachusetts Institute of TechnologyCambridgeUSA

Personalised recommendations