Large-scale Molecular Dynamics Simulations Research Articles

Exploring the constitutive activation mechanism of the class A orphan GPR20.

GPR20, an orphan G protein-coupled receptor (GPCR), shows significant expression in intestinal tissue and represents a potential therapeutic target to treat gastrointestinal stromal tumors. GPR20 performs high constitutive activity when coupling with Gi. Despite the pharmacological importance of GPCR constitutive activation, determining the mechanism has long remained unclear. In this study, we explored the constitutive activation mechanism of GPR20 through large-scale unbiased molecular dynamics simulations. Our results unveil the allosteric nature of constitutively activated GPCR signal transduction involving extracellular and intracellular domains. Moreover, the constitutively active state of the GPR20 requires both the N-terminal cap and Gi protein. The N-terminal cap of GPR20 functions like an agonist and mediates long-range activated conformational shift. Together with the previous study, this study enhances our knowledge of the self-activation mechanism of the orphan receptor, facilitates the drug discovery efforts that target GPR20.

Molecular dynamics simulation for phase transition of CsPbI3 perovskite with the Buckingham potential.

The CsPbI3 perovskite is a promising candidate for photovoltaic applications, for which several critical phase transitions govern both its efficiency and stability. Large-scale molecular dynamics simulations are valuable in understanding the microscopic mechanisms of these transitions, in which the accuracy of the simulation heavily depends on the empirical potential. This study parameterizes two efficient and stable empirical potentials for the CsPbI3 perovskite. In these two empirical potentials, the short-ranged repulsive interaction is described by the Lennard-Jones model or the Buckingham model, while the long-ranged Coulomb interaction is summed by the damped shifted force method. Our molecular dynamics simulations show that these two empirical potentials accurately capture the γ ↔ β ↔ α and δ → α phase transitions for the CsPbI3 perovskite. Furthermore, they are up to two orders of magnitude more efficient than previous empirical models, owing to the high efficiency of the damped shifted force truncation treatment for the Coulomb interaction.

Large-scale molecular dynamics simulation and aggregate behavior research on asphalt

In this comprehensive study, a molecular dynamics (MD) model consisting of approximately 150,000 atoms was developed within the Strategic Highway Research Program (SHRP) framework to investigate the behavior of AAA asphalt, focusing on asphaltene aggregate dynamics behavior over a simulation time span of 0.282 microseconds. The findings reveal that the model achieves a stable state in terms of both energy and microstructure without significant changes or the formation of densely clustered molecules, despite the extensive simulation period. Notably, the patterns of asphaltene aggregation were found to significantly influence the MD system energy, with variations ranging from loosely organized structures due to dispersed asphaltenes to more structured, energy-efficient clusters formed by encapsulated asphaltene pentamers. The size of asphaltene aggregates emerged as a pivotal factor, affecting their encapsulation and leading to distinct formation patterns. Furthermore, the research highlighted that increasing shear rates prompt the depolymerization of asphaltene aggregates, shifting from large-scale structures to smaller ones and adapting dynamically to the flow conditions by aligning parallel to the shear direction. This study underscores the critical impact of asphaltene aggregate behavior and shear rates on the structural integrity and distribution within the asphalt model, offering profound insights into the MD of asphalt materials.

Performance Comparisons of NequIP and DPMD Machine Learning Interatomic Potentials for Tobermorites

Machine Learning Interatomic Potentials (MLIPs) present a significant advancement in fitting molecular potential energy surfaces and predicting molecular crystal structures and mechanical properties by closely approximating first-principles (FP) calculations results while substantially reducing computational time. In this work, utilizing datasets derived from FP calculations, neural equivariant interatomic potentials (NequIP) and Deep Potential Molecular Dynamics (DPMD) was implemented to develop MLIPs for tobermorite 9, 11, 14 Å. The accuracy and efficiency of NequIP and DPMD were assessed by predicting energies and forces as well as performing molecular dynamics (MD) simulations. The findings indicate that NequIP closely fits the results of FP calculations, with both accuracy and efficiency improved by 1∼2 orders of magnitude compared to the DPMD. This demonstrates that MLIPs developed through NequIP for tobermorites have the capacity to enhance the capability of large-scale molecular dynamics simulations.

Molecular dynamics simulations of neutron induced collision cascades in Zr — Statistical modelling of irradiation damage and potential applications

Understanding irradiation damage involves a multi-scale and multi-physics approach, integrating data from experiments, simulations and phenomenological models. This paper focuses on its early stages, specifically neutron-induced collision cascades in zirconium, as nuclear-grade zirconium alloys are widely used in fuel assemblies. We have gathered and analysed a significant sample of results from high-fidelity, large-scale molecular dynamics (MD) simulations, employing existing interatomic potentials and the two-temperature model to account for electron–phonon coupling. Our data can be directly applied to higher-scale methods.Furthermore, we carried out a comprehensive statistical analysis of the features associated with the defect production, including the number of defects, their distribution and size of the affected area. As a result, we developed a generative model of collision cascades that is parametric, hierarchical and stochastic, i.e. it takes into account a statistical nature of the phenomenon, is interpretable and shareable. This model has been developed with three primary objectives: to provide a sufficient descriptor of a cascade, to interpolate data obtained from high-fidelity simulations, and to demonstrate that the statistical model can produce representative distributions of primary irradiation defects. The results can be used to generate synthetic inputs for models at longer length and time scales, to provide fast approximations that take into account the morphology of introduced defects, and in general to serve as a powerful analytical tool.

Fundamental influence of irreversible stress–strain properties in solids on the validity of the ramp loading method

The widely used quasi-isentropic ramp loading technique relies heavily on back-calculation methods that convert the measured free-surface velocity profiles to the stress–density states inside the compressed sample. Existing back-calculation methods are based on one-dimensional isentropic hydrodynamic equations, which assume a well-defined functional relationship P(ρ) between the longitudinal stress and density throughout the entire flow field. However, this kind of idealized stress–density relation does not hold in general, because of the complexities introduced by structural phase transitions and/or elastic–plastic response. How and to what extent these standard back-calculation methods may be affected by such inherent complexities is still an unsettled question. Here, we present a close examination using large-scale molecular dynamics (MD) simulations that include the detailed physics of the irreversibly compressed solid samples. We back-calculate the stress–density relation from the MD-simulated rear surface velocity profiles and compare it directly against the stress–density trajectories measured from the MD simulation itself. Deviations exist in the cases studied here, and these turn out to be related to the irreversibility between compression and release. Rarefaction and compression waves are observed to propagate with different sound velocities in some parts of the flow field, violating the basic assumption of isentropic hydrodynamic models and thus leading to systematic back-calculation errors. In particular, the step-like feature of the P(ρ) curve corresponding to phase transition may be completely missed owing to these errors. This kind of mismatch between inherent properties of matter and the basic assumptions of isentropic hydrodynamics has a fundamental influence on how the ramp loading method can be applied.

Learning dislocation dynamics mobility laws from large-scale MD simulations

By dispensing with all the atoms and only focusing on dislocation lines, the computational method of Discrete Dislocation Dynamics (DDD) gains greatly over Molecular Dynamics (MD) in simulation efficiency of metal plasticity. But whereas in MD dislocations follow natural dynamics of atomic motion, DDD must rely on a dislocation mobility function to prescribe how a dislocation line should respond to the driving force exerted on it. However, reflecting our still incomplete understanding of ways in which dislocations move, mobility functions presently employed in DDD simulations entail simplifications and approximations of limited or, worse still, unknown accuracy and applicability. Here we introduce a data-driven approach in which the dislocation mobility function is modeled as a graph neural network (GNN) trained on large-scale MD simulations of crystal plasticity. We apply our proposed approach to predicting plastic strength of body-centered-cubic (BCC) metal tungsten and show that, once implemented in a DDD model, our GNN dislocation mobility function accurately reproduces the challenging tension/compression asymmetry of plastic flow observed both in ground-truth MD simulations and in experiment. Furthermore, subsequently validated by MD simulations, the same function accurately predicts plastic response of tungsten under conditions not previously seen in training. By demonstrating its ability to learn relevant physics of dislocation motion, our DDD+ML approach opens a promising avenue to bringing fidelity of DDD models closer in line with direct MD simulations at a much reduced computational cost.

Unified understanding of the impact of semiflexibility, concentration, and molecular weight on macromolecular-scale ring diffusion

Conformationally fluctuating, globally compact macromolecules such as polymeric rings, single-chain nanoparticles, microgels, and many-arm stars display complex dynamic behaviors due to their rich topological structure and intermolecular organization. Synthetic rings are hybrid objects with conformations that display both ideal random walk and compact globular features, which can serve as models of genomic DNA. To date, emphasis has been placed on the effect of ring molecular weight on their unusual behaviors. Here, we combine simulations and a microscopic force-level theory to build a unified understanding for how key aspects of ring dynamics depend on different tunable molecular properties including backbone rigidity, monomer concentration, degree of traditional entanglement, and molecular weight. Our large-scale molecular dynamics simulations of ring melts with very different backbone stiffnesses reveal unanticipated behaviors which agree well with our generalized theory. This includes a universal master curve for center-of-mass diffusion constants as a function of molecular weight scaled by a chemistry and thermodynamic state-dependent critical molecular weight that generalizes the concept of an entanglement cross-over for linear chains. The key physics is how backbone rigidity and monomer concentration induced changes of the entanglement length, interring packing, degree of interpenetration, and liquid compressibility slow down space-time dynamic-force correlations on macromolecular scales. A power law decay of the center-of-mass diffusion constant with inverse molecular weight squared is the first consequence, followed by an ultraslow activated hopping transport regime. Our results set the stage to address slow dynamics and kinetic arrest in different families of compact synthetic and biological polymeric systems.

Simultaneous improvement in the elastic and fracture properties of graphene-epoxy nanocomposites–a computational perspective

The incorporation of stiff nano-additives (such as graphene) into a relatively soft polymer material (such as epoxy) usually leads to an improvement in elastic properties (i.e., Young's modulus) at the expense of fracture properties (i.e., tensile strength and toughness). Despite over a decade of research in polymer nanocomposites, we still lack a clear understanding of their structure-property relationships, which limits us from enhancing both elastic and fracture properties concurrently. Here, we performed large-scale reactive molecular dynamics simulations to study the deformation and fracture of model graphene/epoxy systems under uniaxial tension. A computationally efficient reactive force field was developed for graphene-epoxy system, allowing covalent bond formation, and breaking, which is crucial to model cross-linking in model epoxy as well as fracture. It was found the mechanical properties of the nanocomposite are very sensitive to the strength of the graphene-epoxy interface. As expected, elastic modulus increases with the interfacial strength. However, there appears to be an optimal interfacial strength to enhance the tensile strength and toughness. This is due to stress concentration occurring near the graphene edges at high interfacial strength, which leads to premature fracture. We show that by appropriately selecting an intermediate interface strength, one can simultaneously improve the ultimate tensile strength, toughness and the Young's modulus of the nanocomposite epoxy at ultra-low (∼0.2 % by weight) loading fraction of graphene additives. Our findings highlight the critical importance of properly engineered additive-matrix interfacial strength to develop high-performing polymer nanocomposites that are concurrently stiff, strong as well as tough.

Role of dislocations on martensitic transformation temperatures and microstructure: A molecular dynamics study

Microstructure and defects strongly affect martensitic transformations in metallic alloys. Significant progress has been made in understanding the atomic-level processes that control the role of grain boundaries and precipitates in these solid-to-solid phase transformations. Yet, the role of dislocations and their structures on martensitic transformation temperature and the resulting microstructure remains unclear. Therefore, we used large-scale molecular dynamics simulations to study the forward and reverse transformation of a martensitic material modeled after Ni63Al37 under cyclic thermal loading. The simulations reveal that dislocations in the austenite phase act as one-dimensional seeds for the martensite phase, which is present at temperatures significantly above the martensite start value. We find a reduction in the dislocation density during cyclic thermal loading, which results in the increase in martensite and austenite transition temperatures, in agreement with experiments. Importantly, we extracted a critical martensitic nuclei size for developing stable domains and found that relatively low dislocation densities are needed to grow independent martensitic variants resulting in a multi-domain structure.

Heterogeneous Structure and Dynamics of Water in a Hydrated Collagen Microfibril.

Collagen type I is well-known for its outstanding mechanical properties which it inherits from its hierarchical structure. Collagen type I fibrils may be viewed as a heterogeneous material made of protein, macromolecules (such as glycosaminoglycans and proteoglycans) and water. Water content modulates the properties of these fibrils. Yet, the properties of water and the fine interactions of water with the protein constituent of these heterofibrils have only received limited attention. Here, we propose to model collagen type I fibrils as a hydrated structure made of tropocollagen molecules assembled in a microfibril crystal. We perform large-scale all-atom molecular dynamics simulations of the hydration of collagen fibrils beyond the onset of disassembly. We found that the structural and dynamic properties of water vary strongly with the level of hydration of the microfibril. More importantly, we found that the properties vary spatially within the 67 nm D-spacing periodic structure. Alteration of the structural and dynamical properties of the collagen microfibril occur first in the gap region. Overall, we identify that the change in the role of water molecules from glue to lubricant between tropocollagen molecules arises around 100% hydration while the microfibril begins to disassemble beyond 130% water content. Our findings are supported by a decrease in hydrogen bonding, recovery of bulk water properties and amorphization of the tropocollagen molecules packing. Our simulations reveal the structure and dynamics of hydrated collagen fibrils with unprecedented spatial resolution from physiological conditions to disassembly. Beyond the process of self-assembly and the emergence of mechanical properties of collagen type I fibrils, our results may also provide new insights into mineralization of collagen fibrils.

Atomic simulations of crack propagation in Ni-Al binary single crystal superalloy with a central crack

Nickel (Ni)-based single-crystal superalloys are of great importance in the aircraft industry due to their excellent mechanical properties, and cracks as unavoidable defects may affect the mechanical performances of materials dramatically. In this paper, large scale molecular dynamics (MD) simulations are carried out to understand the deformation mechanisms of Ni-based single crystal with a central crack under tension. Here, the effects of matrixes (γ, γ′ and γ/γ′), strain rates (1 × 109 s−1 ∼ 3 × 109 s−1) and temperatures (300 K∼900 K) on the role of crack propagation are considered. It is observed that dislocations and slip systems in the γ′ model are concentrated near the crack, resulting in the rapid expansion of dislocation, which leads to the fastest crack growth speed and early fracture. While the crack propagation rate of γ and γ/γ′ models are relatively slow, due to the combined action of the Lomer-Cottrell lock and stacking fault tetrahedron structure and Stair-rod dislocation, which hinders crack propagation. In addition, deformation at increased strain rates and/or reduced temperatures, lead to superior yield stress and Young′s modulus for models with a central crack at γ/γ′ interface. On the other hand, high temperature and high strain rate will promote crack propagation in the γ phase, and the higher the strain rate and/or temperature, the faster the crack propagation speed will be. These results will enrich our understanding on the crack propagation and evolution mechanisms in Ni-based single crystal superalloy.

Shock response of gradient nanocrystalline CoCrNi medium entropy alloy

A comprehensive investigation of shock-induced deformation and spallation mechanisms in nanocrystalline CoCrNi medium-entropy alloys (MEAs) is conducted by large-scale molecular dynamics simulations, encompassing both gradient and homogeneous nanocrystalline structures. Under the shock velocity ranging from 0.2 km/s to 1 km/s, the effects of grain heterogeneity on shock wave propagation, spallation and defect evolution are examined. An inverse Hall–Petch relationship, where the Hugoniot elastic limit (HEL) increases with larger grain sizes is observed within the grain size range of 3 to 12 nm for homogeneous nanocrystalline samples. Compared to the localized strain distribution either within grain interior (GI) or at grain boundaries (GB) in homogeneous nanocrystalline structures, gradient-grained counterparts exhibit compatible deformation along shock direction, which coordinates the load partitioning between GB and GI. As a result, lower densities of dislocation and stacking fault are triggered within the GI, while shear localization is mitigated at GB, leading to higher spall strength observed in the gradient nanocrystalline MEA samples at the same shock velocity. The results shed light on the insights of deformation and spallation mechanisms in gradient-grained MEAs under dynamic loading, which may open up a new avenue for the design of advanced structural MEAs with high strength and toughness.

Proton Collective Quantum Tunneling Induces Anomalous Thermal Conductivity of Ice under Pressure.

Proton tunneling is believed to be nonlocal in ice, but its range has been shown to be limited to only a few molecules. Here, we measured the thermal conductivity of ice under pressure up to 50GPa and found it increases with pressure until 20GPa but decreases at higher pressures. We attribute this nonmonotonic thermal conductivity to the collective tunneling of protons at high pressures, supported by large-scale quantum molecular dynamics simulations. The collective tunneling loops span several picoseconds in time and are as large as nanometers in space, which match the phonon periods and wavelengths, leading to strong phonon scattering at high pressures. Our results show direct evidence of global quantum motion existing in high-pressure ice and provide a new perspective to understanding the coupling between phonon propagation and atomic tunneling.

Markov State Models Reveal How Folding Kinetics Influence Absorption Spectra of Foldamers.

Self-assembly of platinum(II) complex foldamers is an essential approach to fabricate advanced luminescent materials. However, a comprehensive understanding of folding kinetics and their absorption spectra remains elusive. By constructing Markov state models (MSMs) from large-scale molecular dynamics simulations, we reveal that two largely similar dinuclear alknylplatinum(II) terpyridine foldamers, Pt-PEG and Pt-PE with slightly different bridges, exhibit distinctive folding kinetics. Particularly, Pt-PEG bears bridge-dominant, plane-dominant, and cooperative pathways, while Pt-PE only prefers the plane-dominant pathway. Such preference originates from their difference in intrabridge electrostatic interactions, leading to contrastive distributions of metastable states. We also found that the bridge-dominant pathway for Pt-PEG becomes more favorable when lowering the temperature. Interestingly, based on the comprehensive conformation ensembles from our MSMs, we reveal the conformation-dependent absorption spectra of Pt-PEG and Pt-PE. Our theoretical spectra not only align with experimental results but also reveal the contributions of diverse conformations to the overall absorption bands explicitly, facilitating the rational design of stimuli-responsive smart luminescent molecules.

Precision Atomistic Structures of Actinium-/Radium-/Barium-Doped Lanthanide Nanoconstructs for Radiotherapeutic Applications.

Lanthanide vanadate (LnVO4) nanoconstructs have generated considerable interest in radiotherapeutic applications as a medium for nanoscale-targeted drug delivery. For cancer treatment, LnVO4 nanoconstructs have shown promise in encapsulating and retaining radionuclides that emit alpha-particles. In this work, we examined the structure formation of LnVO4 nanoconstructs doped with actinium (Ac) and radium (Ra), both experimentally and using large-scale atomistic molecular dynamics simulations. LnVO4 nanoconstructs were synthesized via a precipitation method in aqueous media. The reaction conditions and elemental compositions were varied to control the structure, fluorescence properties, and size distribution of the LnVO4 nanoconstructs. LnVO4 nanoconstructs were characterized by X-ray diffraction, Raman spectroscopy, and fluorescence spectroscopy. Molecular dynamics simulations were performed to obtain a fundamental understanding of the structure-property relationship between radionuclides and LnVO4 nanoconstructs at the atomic length scale. Molecular dynamics simulations with well-established force field (FF) parameters show that Ra atoms tend to distribute across the nanoconstructs' surface in a broader coordination shell, while the Ac atoms are arranged inside a smaller coordination shell within the nanocluster. The Ba atoms prefer to self-assemble around the surface. These theoretical/simulation predictions of the atomistic structures and an understanding of the relationship between radionuclides and LnVO4 nanoconstructs at the atomic scale are important because they provide design principles for the future development of nanoconstructs for targeted radionuclide delivery.

Shear-promoted graphite-to-diamond phase transition at the grain boundary of nanocrystalline graphite

High pressure has traditionally been considered essential for the transformation of graphite into diamond. However, reducing the transition pressure required for this graphite-to-diamond (G2D) conversion holds significant appeal in both scientific research and engineering applications. In this study, we conducted large-scale molecular dynamics (MD) simulations using an environment-dependent interaction potential (EDIP) to examine the shear deformation of nanocrystalline graphite (n-graphite) with a grain size of approximately 6.5 nm. We discovered that the G2D transition pressure in n-graphite can be reduced to 2–3 GPa, significantly lower than the ∼90 GPa uniaxial stress required in crystalline graphite. This reduction is primarily due to concentrated local shear stresses at grain boundaries (GBs), which induce substantial rotations of graphite layers. These rotations facilitate the initial formation of diamond bonds at sites of pre-existing imperfections at the GBs, assisted by shear. Once initiated at the GBs, the G2D transition rapidly propagates within grains aligned parallel to the shear components, resulting in the formation of nanocrystalline diamond. Our findings underscore the critical roles of GBs and shear stress in enabling the G2D transition in n-graphite.

Rational Design of Dual-Functionalized Gd@C82 Nanoparticles to Relieve Neuronal Cytotoxicity in Alzheimer's Disease via Inhibition of Aβ Aggregation.

The accumulation of amyloid-β (Aβ) peptides is a major hallmark of Alzheimer's disease (AD) and plays a crucial role in its pathogenesis. Particularly, the structured oligomeric species rich in β-sheet formations were implicated in neuronal organelle damage. Addressing this formidable challenge requires identifying candidates capable of inhibiting peptide aggregation or disaggregating preformed oligomers for effective antiaggregation-based AD therapy. Here, we present a dual-functional nanoinhibitor meticulously designed to target the aggregation driving force and amyloid fibril spatial structure. Leveraging the exceptional structural stability and facile tailoring capability of endohedral metallofullerene Gd@C82, we introduce desired hydrogen-binding sites and charged groups, which are abundant on its surface for specific designs. Impressively, these designs endow the resultant functionalized-Gd@C82 nanoparticles (f-Gd@C82 NPs) with high capability of redirecting peptide self-assembly toward disordered, off-pathway species, obstructing the early growth of protofibrils, and disaggregating the preformed well-ordered protofibrils or even mature Aβ fibrils. This results in considerable alleviation of Aβ peptide-induced neuronal cytotoxicity, rescuing neuronal death and synaptic loss in primary neuron models. Notably, these modifications significantly improved the dispersibility of f-Gd@C82 NPs, thus substantially enhancing its bioavailability. Moreover, f-Gd@C82 NPs demonstrate excellent cytocompatibility with various cell lines and possess the ability to penetrate the blood-brain barrier in mice. Large-scale molecular dynamics simulations illuminate the inhibition and disaggregation mechanisms. Our design successfully overcomes the limitations of other nanocandidates, which often overly rely on hydrophobic interactions or photothermal conversion properties, and offers a viable direction for developing anti-AD agents through the inhibition and even reversal of Aβ aggregation.

The origin of phase separation in binary aluminosilicate glasses

The quest for hard and tough transparent oxide glasses is at the core of glass science and technology. Aluminosilicate glasses exhibiting nanoscale phase separation emerge as promising candidates for such materials. Nevertheless, proper control of the phase separation represents a daunting challenge due to its elusive origins. Here we employ large-scale molecular dynamics simulations and structural analysis to unravel the underlying mechanisms of the phase separation in aluminosilicate. The observed phase separation originates from an arrangement of SiO4 and AlOn polyhedra, which manifests from the second coordination shell and extends to higher shells. This specific arrangement is driven by repulsion between the polyhedra, reaching its maximum at around 50 mol% of Al2O3. This behavior becomes pronounced around and below the glass transition temperature. This work sheds light on the origin of phase separation and provides a route for further exploration across other compositions to develop glasses with adapted mechanical performance.

Understanding anti-corrosion mechanisms of the 2,6-dithiopurine (DTP) inhibitor molecule on iron surfaces

Purine derivatives present a high efficiency for anti-corrosion purposes while the nanoscale mechanisms are not yet fully understood. In this work, the inhibition mechanism of the 2,6-dithiopurine (DTP) molecule was investigated using advanced characterizations and atomic simulations. We revealed that the DTP molecule undergoes a multi-layer adsorption mode and is governed by both chemisorption and physisorption processes. An organometallic layer at a thickness of ∼3.8 nm was captured using TEM, comprised of Fe-N and Fe-S bonds strongly anchored to the iron interfaces, as confirmed by the XPS analysis. Our quantum chemical calculations showed that heteroatoms such as N or S in the DTP molecule offer reactive sites that facilitate its chemisorption onto iron surfaces, to form metal-organic bonds (specifically, Fe-N and Fe-S) at the DTP-iron interfaces. We further employed molecular dynamics simulations to investigate the interaction between DTP molecule and various substrates. Our findings suggest that the hydroxyl groups on the lepidocrocite surface can significantly promote bonding between DTP and iron surfaces. Based on large-scale molecular dynamics simulations, we discovered that DTP can exhibit a strong binding affinity for aggressive ions, leading to a considerable decrease in the self-diffusion coefficient of hydronium ions of up to 70 %. This work provides a new understanding of anti-corrosion mechanisms and offers opportunities to design new corrosion inhibitors possessing multiple reactive sites and strong bonding with the substrate.