Yongsheng Leng

Interpreting single-molecule force spectroscopy (SMFS) data from optical tweezers with steered molecular dynamics (SMD) is hindered by a severe timescale mismatch: SMD trajectories span nanoseconds to microseconds, while SMFS experiments unfold over milliseconds to seconds, forcing SMD to pull six to nine orders of magnitude faster than experiment and systematically distorting rupture forces, molecular pathways, and kinetic intermediates. We introduce force-probe coupled steered molecular dynamics (fp-SMD), which explicitly propagates the force probe (the optically trapped bead) alongside the biomolecule within a single classical Hamiltonian description of the coupled molecule– linker–bead system. A Predict–Correct Trajectory Propagation (PCTP) scheme resolves the resulting computational bottleneck by alternating a bead-propagation step, with the molecular center of mass (COM) held fixed, and a molecular step in which metadynamics and transition-state theory identify the kinetically resolved intermediate reached at each macroscopic time step. On the cholesterol–β-cyclodextrin (β-CD) host–guest complex, fp-SMD reproduces the experimental rupture-force histogram directly at matched conditions, without extrapolation from standard SMD. fp-SMD further resolves how applied force reshapes the molecular escape pathway: we find that cholesterol dissociates via a high-angle route at low force and a direct, small-angle route at high force — providing, to our knowledge, the first direct structural evidence for force-induced pathway switching in a small host–guest system, a mechanism believed to underlie catch-slip bond behavior. Significance Statement: Single-molecule experiments using optical tweezers reveal how mechanical forces unfold proteins, rupture bonds, and drive cell adhesion. Simulations that could explain these experiments at atomic resolution have been hindered by a basic mismatch: they must pull the molecule millions of times faster than a real experiment to be computationally tractable, distorting the forces and pathways they aim to explain. This work introduces a simulation method that explicitly includes the physical probe used in the experiment, not just the molecule, letting simulations match experimental timescales without sacrificing atomic detail. Applied to a cholesterol–cyclodextrin complex relevant to atherosclerosis and drug delivery, the method shows how force redirects a molecule’s escape pathway — a mechanism believed to underlie many force-sensitive switches.

Because atomic-scale junction configurations cannot be directly resolved during break-junction measurements, simulations are needed to connect experimental force and conductance signatures with plausible interfacial structures. Here, we use configurational-bias Monte Carlo, driven molecular dynamics, and DFT+NEGF transport calculations to interpret the electromechanical response of Au–1,8-octanedithiol (ODT)–Au junctions measured by conductive atomic force microscopy break-junction experiments. Under cyclic stretching and compression with a modulation amplitude of 1.2 Å, reproducible force and conductance responses are observed. The simulations generate metastable configurations containing one, two, and three bridging ODT molecules and reproduce characteristic force ranges and conductance levels consistent with experiment. Structural analysis reveals configuration-dependent load distribution, junction deformation, molecular reorientation, and local Au–S coordination. The DFT+NEGF results further show that conductance modulation is governed mainly by changes in the Au–S contacts rather than deformation of the alkane backbone. These findings provide an atomistic interpretation of the measured electromechanical responses and clarify how junction configuration and contact rearrangement influence transport in Au–ODT–Au molecular junctions.

Molecular understanding of the CO₂-CH₄-H₂O mixture in clay interlayers provides crucial insights into many geochemical processes in CO₂-enchanced oil/gas recovery and CO₂ subsurface sequestrations. In this work, we developed new molecular models by optimizing the unlike-pair Lennard-Jones (LJ) parameters for the CO₂-H₂O, CH₄-H₂O and CO₂-CH₄ binary mixtures. The optimization utilizes the coupling parameter method in combination with Gibbs ensemble Monte Carlo (GEMC) simulations. The transferability of binary-derived parameters to ternary CO₂-CH₄-H₂O system is further demonstrated. These optimized parameters are used to study CO₂-CH₄-H₂O mixtures in Na-montmorillonite interlayers at 323 K and 90 bar, through grand-canonical Monte Carlo (GCMC) and molecular dynamics (MD) simulations. Chemical potentials of CO₂, CH₄, and H₂O in clay interlayer are calculated through the equation of state (EOS) with the Lennard-Jones referenced Statistical Associating Fluid Theory (SAFT-LJ), which is suitable for aqueous mixtures. The equilibrium basal spacing distances and species concentrations under different relative humidity (RH) and CO₂/CH₄ mole fractions are determined through extensive GCMC simulations. Importantly, using the new molecular models, we find that the trend of the sorbed CO₂ content versus the sorbed H₂O content shows good agreement with in situ infrared (IR) spectroscopy data by Loring et al. [Environmental Science& Technology, 2021, 55, 11192-11203]. While the predicted CO₂ concentrations in the monolayer hydration state (1W) is higher than the experimental results due to the heterogeneous expansion of the clay mineral in experiments, the CO₂ content in the bilayer hydration state (2W) compares remarkably well with the experimental data. The CH₄ adsorption in clay interlayers is found substantially lower than that of CO₂. In general, GCMC results show that water intercalation under high RH suppresses the sorption of CO₂ and CH₄, and that Na-montmorillonite preferentially adsorbs CO₂ over CH₄.

The effects of termination functional groups of the Ti₃C₂T_x MXene membrane on the structural and dynamics properties of nearby water molecules and foulants are investigated through molecular dynamics simulations. The simulation results show that a much denser water layer can be formed at the vicinity of hydroxyl (OH) termination than that near fluorine (F) or oxygen (O) termination. Particular focus is given to the molecular binding properties of β-d-mannuronic acid (M) and α-l-guluronic acid (G) alginate monomers on the MXene membrane surface with different termination groups. Further steered molecular dynamics (SMD) simulations show that M alginate monomers exhibit significant binding with the MXene membrane surface with O termination, due to the strong electrostatic interaction and the van der Waals attraction. In contrast, the binding between the alginate monomers and the MXene membrane surface with OH termination is negligible, as the stable hydration water network prevents them from direct contact. In addition, SMD simulation results show that calcium (Ca²⁺) ions could significantly enhance the surface fouling between M alginate monomers and the MXene with an O termination through the formation of contact ionic pair (CIP) and solvent-shared ionic pair (SSIP) structures.

We synthesized PtTe₂ nanoflakes via chemical transformation utilizing Te nanoflakes as templates. Aberration-corrected transmission electron microscopy confirmed this assertion and revealed the distinctive in-plane layered structure of the synthesized PtTe₂ nanoflakes. The transformation process and reaction mechanisms were investigated using particle swarm optimization algorithms in combination with first-principles density functional calculations. According to the calculations, it was found that the initial chemical bonds automatically transition to the cross-sectional direction, representing the first experimental confirmation of the spontaneous directional transformation of layers and chemical bonds within multivalent Te layers. It is noteworthy that this transformation exhibits a significant dimension-dependent characteristic. This study has broadened the range of growth processes for two-dimensional materials, contributing to a better understanding of transformation behaviors between chemical stimuli and nanocrystals.

Friction behaviors of an amorphous SiO2 tip sliding on the Au(111) surface in atomic force microscopy (AFM) are investigated through molecular dynamics (MD) simulations. We observed a regime of extremely low, close-to-zero friction at low normal loads with clear stick-slip friction signals. The friction is almost independent of the applied normal load below a threshold value. However, above this loading threshold, friction can remain low or increase sharply. Such an unexpected friction duality is attributed to the high probability of defect formation at the sliding interface that can induce plowing friction in a high-friction state. The energy difference between the low-friction state and the high-friction state is surprisingly low, which is comparable to kT (similar to 25 meV) at room temperature. These findings are consistent with previous AFM friction measurements using silicon AFM tips. Further MD simulations show that one can always use an amorphous SiO2 tip to image the crystalline surface with regular stick-slip friction signals. This is largely due to the fact that there is always a small fraction of contacting Si and O atoms at the sliding interface that are sitting on the relatively stable, close-to-hollow sites of the crystalline Au(111) surface during the stick stage; thus, they are capable of sampling local energy minima. We anticipate that regular stick-slip friction can be achieved even in the intermediate loading range, so long as the low friction state is maintained when friction duality occurs.

Friction measurements by an atomic force micro-scope (AFM) frequently showed regular stick-slip friction signalswith atomic-scale resolutions. Typically, for an AFM metal tipsliding on a metal crystal surface, the microstructure of the tipmade from the thermally evaporated metal coating on a siliconcantilever was polycrystalline. Our detailed molecular dynamics-(MD) simulations of a polycrystalline Pt tip (R= 10 nm in radius)sliding on an Au(111) surface revealed how the geometry of the polycrystalline tip took effect on the friction behavior at the contactinterface. We found that the apex of the Pt tip with multiple grains near the edge of contact could induce severe plastic deformationsof the gold substrate, leading to irregular stick-slip frictions upon sliding. Simulation results showed that in order to achieve a clearstick-slip friction signal with single atomic slips, the apex of the Pt tip must adopt a single crystalline protrusion without anyneighboring grains involved in the metal contact. We showed that such a single crystalline protrusion, which presumably could beachieved during initial run-in or wear-out of high-energy Pt atoms in the neighboring grains, was passivated by a large number ofgold atoms due to metal adhesion in the contact periphery. Using such a crystalline protrusion tip, we demonstrated that the stick-slip friction produced was very"tolerant"to the adhesion of a large number of gold atoms on the tip apex. We further showed thatAFM tip mass used in MD simulations also played an important role in determining the transition between friction regimes, whichcould be well explained by the Prandtl-Tomlinson thermal activation model.

Liquid–vapor molecular dynamics (LVMD) simulations are performed to reinvestigate the phase transition and solvation force oscillation behavior of a simple argon liquid film confined between two solid surfaces. Our simulations present a novel scenario in which the n → n − 1 layering transitions are accompanied by the formation, climb, and annihilation of Frank partial dislocations during the squeeze-out process under compression. This is indicated by the splitting of the repulsive peaks in the solvation force profile. The detailed analysis reveals that the formation–climb–annihilation mechanism of Frank dislocation occurs during approach and disappears during receding, which would result in force hysteresis. In combination with our recent works, this study provides new insights into the physical property of nanoconfined lubricant films in boundary lubrication.