The Journal of Physical Chemistry B最新文献

DNAtetrahedra are versatile programmable nanostructures that canassemble into higher-order architectures through sticky-end hybridization.Here, we use coarse-grained molecular dynamics simulations with theoxDNA model to examine dissociation of hybridized sticky-end linkersin finite linear (1D), looped (2D), and multiply connected (3D) tetrahedronassemblies containing 4–9-bp linkers at 300–330 K. Increasingsticky-end length systematically stabilized linker hybridization,increased hydrogen-bond occupancy, and generally increased the effectiveactivation energy for dissociation. Increasing connectivity suppressedglobal configurational fluctuations but did not monotonically delaythe first linker-opening event. Comparative simulations of isolatedhybridized duplexes further showed that incorporation into tetrahedralframeworks shifts opening times and effective activation energiesin a topology- and linker-length-dependent manner. In particular,the 3D assemblies displayed sequential opening times with the greatesttemporal separation between early and late dissociation events, indicatingthat the kinetic environment of the remaining linkers evolves as connectivityis progressively lost. Overall, the results demonstrate that networktopology regulates linker dissociation through an interplay betweenlocal hybridization energetics, configurational constraints, and topology-dependentstructural reorganization rather than through simple monotonic stabilizationwith increasing connectivity.

Therational construction of discrete supramolecular architectureswith controlled topological complexity remains a significant challengein coordination-driven self-assembly. Herein, we report a modularmulticomponent self-assembly strategy for the construction of a homologousseries of perylene diimide (PDI)-based fluorescent metallacycles withprogressively increasing structural complexity. By simply varyingthe geometry and connectivity of auxiliary carboxylate ligands, discretemonocyclic, bicyclic, and tricyclic metallacycles were successfullyassembled and characterized by nuclear magnetic resonance spectroscopy,electrospray ionization mass spectrometry, and single-crystal X-raydiffraction. Photophysical and frontier-orbital analyses reveal thatthese metallacycles retain strong PDI-centered absorption and highfluorescence quantum yields of 69.5–81.4%, while the tricyclicmetallacycle displays a distinct donor–acceptor-type orbitalseparation. Furthermore, these metallacycles exhibited efficient fluorescencesensing toward picric acid, with the tricyclic metallacycle showingthe highest sensitivity with a detection limit of 2.3 μmol/L.This work demonstrates that topological engineering through multicomponentcoordination-driven self-assembly provides an effective strategy fordeveloping structurally sophisticated fluorescent supramolecular sensors.

Nitrate molten salts are widely appliedin concentratingsolarpower due to their excellent thermodynamic properties. However, conventionalexperimental methods struggle to accurately measure the thermodynamicproperties of molten salts and thoroughly analyze their microscopicstructures. The deep potential molecular dynamics (DPMD) method providesa novel approach for investigating the structures and thermodynamiccharacteristics of molten salts. By integrating machine learning (ML)with the deep potential GENerator (DP-GEN) active learning algorithm,we developed a high-fidelity interatomic potential function of NaNO₃–NaCl–NaF molten salt. The structure and thermodynamicproperties of the molten salt were systematically explored, and theinfluence of temperature on these characteristics was elucidated.The results show that the elevated temperature changes the interionicdistances and the interionic forces. The increase in temperature leadsto a decrease in the coordination number and a reduction in the energybarrier, making it easier for ions to detach from the central ion,thereby enhancing the ion migration ability in molten salts. The simulateddensity and specific heat capacity show excellent consistency withthe experimental results, with relative errors of 2.11% and 0.45%,respectively. When the temperature rises from 573 to 773 K, the simulatedviscosity reduces from 2.803 to 1.526 mPa·s, while thermal conductivitydeclines from 0.576 to 0.504 W/(m·K). Combining deep potentialmolecular dynamics with the DP-GEN active learning algorithm, thisstudy provides an efficient approach to predict the thermodynamicproperties of molten salt thermal energy storage materials.

Many of the time scales of life have speed limits setby quantum-mechanicalconstraints along with nonfundamental quantities, such as the temperatureof the environment, which are however bounded by anthropic considerations.Here, some of such speed limits are examined, including those forthe rates of elementary chemical reactions and biomolecular folding.Limitations of simple back-of-envelope estimates are also discussed.

Recent experimentsshow that protein condensates sitting on oppositesurfaces of a flat lipid membrane move together and prefer to overlap,even though they cannot touch each other. This points to an indirect,membrane-mediated interaction. Two mechanisms could be responsible:a curvature-induced interaction, which is energetic in origin, anda fluctuation-induced interaction, which is entropic. Here we studyboth with coarse-grained molecular dynamics simulations, using Cooke’simplicit-solvent lipid model together with a generic bead–springpolymer model for the condensate. We compute the potential of meanforce between two condensates across the membrane. For condensatesof the same size, full overlap is unfavorable, and the pair insteadsettles into a partially overlapping state that bends the membraneinto an S-like shape. When the two condensates differ strongly insize, full overlap becomes favorable. We explain this with a simplegeometric picture. The condensate wets the membrane as a thin filmand imposes curvature only along its rim, while membrane tension flattensthe membrane under its interior. The resulting ring of curvature cantrap a smaller condensate on the opposite side. We also compare thebending undulations and the effective bending modulus of a bare membrane,a membrane with one condensate, and a membrane with condensates onboth sides. A wetting condensate suppresses the undulation modes andstiffens the membrane, but whether this makes overlap entropicallyfavorable remains inconclusive. Our results indicate that the couplingis driven mainly by curvature, and that it depends on the wettingmechanism and on the membrane tension.

Aromaticamino acids buried inside folded proteins play a criticalrole in stabilizing both the folded conformation and anchoring incomingsubstrate molecules. Moreover, their ring flips serve as sensitiveprobes of protein breathing motions. In this work, we accurately estimatethe ring-flip rates of a phenylalanine residue in cytochrome C fromMD simulations using enhanced sampling methods and advanced rate theories,such as the reactive flux formalism and Grote–Hynes theory(GHT). We demonstrate that the combined steric hindrance from theresidues adjacent to the ring, along with their delayed response inaccommodating free rotation, manifests as memory-dependent friction.Increasing solvent viscosity slows the fluctuations of these cavityresidues, reinforcing the delayed response and producing a fractionalviscosity dependence of the flip rate despite the ring being completelyburied.

Simulating electrochemical systems confined within nanoscaleslabgeometries is a major challenge due to long-range electrostatic interactionsand the breakdown of global charge neutrality when the domain exchangesions with an external bulk electrolyte reservoir. In this work, wepresent a grand-canonical simulation framework designed for slab geometriesthat utilizes a regularized 3D periodic Ewald summation techniqueto systematically converge to the true 2D periodic Green’sfunction without unphysical cell-width (L_z) dependencies. Unlike traditional grand canonicalmethods that rely on neutral-pair insertions to maintain a zero netcharge, our framework allows the net cell charge to fluctuate by implementingdecoupled, single-ion trial moves. This approach bypasses the cooperativesampling bottlenecks that plague neutral-pair insertions inside dense,strongly charged nanopores. Crucially, the method allows us to calculatethe Donnan potential (φ_D) inside the cell, whichcannot be accessed directly by any other available simulation method.The Donnan potential acts as a direct microscopic analog to the electromotiveforce (emf) of a nanoscale electrochemical cell, making the methoduniquely suited for iontronic and nanofluidic applications where nonidealion–ion correlations and strong inhomogeneities modify devicepotentials. The methodology is validated against exact solutions ofthe nonlinear Poisson–Boltzmann equation, demonstrating excellentagreement for both local ionic density profiles and self-consistentpotentials. Finally, we demonstrate the utility of this frameworkby simulating charge regulation and surface titration of weak acidicgroups at the pore wall.

Ions in water near the interface create an electricaldouble layer,in which counterions are attracted, while co-ions are repelled fromthe interface. Adding salt to the subphase water is always expectedto decrease the electric field in the electrical double layer as increasedcounterions screen the surface charge more effectively. We discovereda phenomenon contrary to this common belief in a cationic Langmuirmonolayer upon addition of fluoride salts. A Langmuir monolayer consistingof DPTAP (1,2-dipalmitoyl-3-trimethylammonium-propane chloride) atdifferent concentrations of alkali-metal fluoride salts (LiF, NaF,KF, RbF, and CsF) in the subphase water was investigated by sum-frequencyvibrational spectroscopy (SFVS). The OH signal from interfacial watermolecules increased upon addition of NaF and LiF, reached a maximumsignal (about twice that from pure water) at ∼100 μM,and decreased afterward. This phenomenon was explained by the actionof Li⁺ and Na⁺ co-cations that bind effectivelywith the ubiquitous bicarbonate (HCO₃^–) anions at neutral pH and remove these counterions that initiallyscreen the positively charged headgroups of the DPTAP molecules. Bycontrast, the sum-frequency signal decreased almost monotonicallywith KF addition, indicating that the K⁺ co-cation cannotcapture these bicarbonate counterions effectively, and this behavioris explained in terms of the higher solubility of KHCO₃ as compared to LiHCO₃ and NaHCO₃. This trendtoward even heavier alkali cations did not continue as RbF and CsFalso showed an increase in the sum-frequency signal, in spite of themonotonic increase of the solubility from LiHCO₃ to CsHCO₃. This tendency demonstrates that the polarizability and resultingsurface excess of large alkali cations become important at the interface,making them more accessible to the bicarbonate counterions.

Hydrophobic eutectic solvents (HES) are emerging as sustainablealternatives to conventional organic solvents for the extraction ofmolecular pollutants from water. Yet, their selectivity remains poorlyunderstood, hindering the predictive design of eutectic solvents beyondempirical success. Here, we present a multiscale strategy to rationalizeand predict solute partitioning in HES. Focusing on bisphenol A (BPA)in trioctylphosphine oxide (TOPO):menthol as a prototypical system,we combine monophasic and biphasic molecular dynamics with quantumenergy decomposition of dominant solvation motifs. Our methodologycaptures the experimentally measured spontaneous migration of BPAand its thermodynamic stabilization in the HES phase, and it identifiesthe microscopic origin of selectivity: cooperative hydrogen bondingcouples with strong dispersion and polarization in the hydrophobiceutectic microenvironment. The extension of the workflow to a secondexperimentally relevant HES provides additional evidence supportingthe transferability of the mechanistic interpretation and illustratesits potential as a computational framework for the rational screeningand design of HES formulations.

Antimicrobial peptides (AMPs) hold immense potentialin combatingcommunity-acquired and MDR infections. The therapeutic efficiencyof AMPs depends on their various physicochemical properties. We investigatedhow the hydrophobic–hydrophilic landscape of tryptophan-containingcationic heptapeptides (CAMPs) influenced their ability to bind tomicrobial membranes and their antimicrobial effectiveness againstseveral ESKAPE pathogens. We found that charge delocalization overthe peptide backbone enhanced antimicrobial activity, while chargelocalization at the C- and N-termini reduced efficacy (1.5-2.7 folds)for both the single and double tryptophan peptides. The double-tryptophanpeptide P32* had increased activity over the single tryptophan peptideP32 (∼2.3 fold), and alleviated the effect of charge segregationon activity. Using steered molecular dynamics umbrella sampling simulations,we evaluated the peptides based on their affinity (ΔG) for a membrane-mimetic bilayer model of Pseudomonas aeruginosa. Peptides with higher membrane-bindingaffinity exhibited greater antimicrobial potency. All peptides exhibitednegligible hemolytic activity (HC_10% ≤ 10 μM)and retained membranolytic antibacterial activity. The calculatedbinding free energies of the peptide(s) and the bilayer exhibiteda strong linear correlation with the experimental minimum inhibitoryconcentration (MIC) values, with a negative slope. More favorablebinding free energies (i.e., more negative values) corresponded tolower MICs, as stronger interactions with the membrane enhanced themembranolytic properties of these peptides. We demonstrated that integratingmolecular simulations with experimental validation enables the useof membrane-binding free energy as a quantitative predictor of peptideactivity. This approach provides a rational framework for the futuredesign and optimization of membrane-active CAMPs.