Search

During the process of rapid cooling, two distinct types of supercooled liquids are found. One kind of supercooled liquids destines to crystallize (the crystal-forming liquid (GFL)), and the other forms a glass (the glass-forming liquid (GFL)). Despite having no significant differences in conventional physical quantities such as structure and energy, the distribution of the potential energy after short-time averaging reveals the emergence of distinction. By analyzing the skewness of the potential energy distribution after the short-time average, a new characteristic time ($\Delta_{CTS)}$) emerges. For a given temperature, the skewness reaches an extremum when the duration of the short-time average equals this characteristic time. Interestingly, this characteristic time scale follows a Curie-like law ($\Delta_{CTS}\sim 1/(T-T^{*})^{{\gamma}}$) with an exponent $\gamma$, which effectively distinguishes these two types of supercooled liquids. The value of $\gamma$ undergoes a discontinuous transition at the critical cooling rate. The value of $\gamma_{g}$, corresponding to GFL, is always greater than $\gamma_{c}$ associated with CFL. Remarkably, $T^{*}$ precisely represents the glass transition temperature (Tg ) or the crystallization temperature (Tc ) for GFL and CFL, respectively. Theoretical analysis suggests that, the Curie-like law implies a kinetic phase transition. The essential difference between these two types of supercooled liquids lies in the different in local potential energy landscapes. First, the atomic motion in GFL may have more strong correlation than that in CFL. Second, CFL may possess a lower configurational entropy than GFL. The current study not only delineates the differentiation between the two types of supercooled liquids, but also provides a new perspective for exploring the nature of glasses., Comment: 7 figures 24 pages

Within the scenario of the potential energy landscape (PEL), a thermodynamic model has been developed to uncover the physics behind the Angell plot. In our model, by separating the barrier distribution in PELs into a Gaussian-like and a power-law form, we obtain a general relationship between the relaxation time and the temperature. The wide range of the experimental data in the Angell plot, as well as the molecular-dynamics data, can be excellently fitted by two characteristic parameters, the effective barrier ({\omega}) and the effective width ({\sigma}) of a Gaussian-like distribution. More importantly, the fitted {\omega} and {\sigma}^2 for all glasses are found to have a simple linear relationship within a very narrow band, and fragile and strong glasses are well separated in the {\omega}-{\sigma}^2 plot, which indicates that glassy states only appear in a specific region of the PEL.

The fermion sign problem, the biggest obstacle in quantum Monte Carlo calculations, is completely solved in this paper. Here, we find a strategy, in which the contribution from those negative-weighted paths is thoroughly cancelled or replaced by some positive-weighted paths. The crucial point lies on the Feynman path integral formula proposed in our group, which allows us to deeply analyze the Boltzmann weight of each path. Through mathematical proof, we demonstrate that physical quantities can be exactly calculated within a specific kind of paths, which have positive the Boltzmann weight. With this finding, a new Monte Carlo method is proposed, in which the fermion sign problem is absent. As an example, the current method is applied to the two-dimensional Hubbard model, and the results do manifest the correctness., Comment: 24 pages, 4 figures

By precisely writing down the matrix element of the local Boltzmann operator, we have proposed a new path integral formulation for quantum field theory and developed a corresponding Monte Carlo algorithm. With current formula, the Hubbard-Stratonovich transformation is not necessary, and is not based on the determinant approach, which can improve the computational efficiency. The results show that, the simulation time has the square-law scaling with system sizes, which is comparable with the usual first-principle calculation. The current formula also improves the accuracy of the Suzuki-Trotter decomposition. As an example, we have studied the one-dimensional half-filled Hubbard model at finite temperature. The obtained results are in excellent agreement with the known solutions. The new formula and Monte Carlo algorithm can be used in various studies., Comment: 19 pages, 6 figures

It is generally believed that the intrinsic properties of glasses are intimately related to potential-energy landscapes (PELs). However, little is known about the PELs of glasses below the glass transition temperature (Tg). Taking advantage of lower potential-energy barriers in nanosystems, we have systematically investigated the dynamics behavior of two nanoglasses, Al43 and Al46. Structure transformation is identified in our pure molecular-dynamics simulation far below Tg, which manifests the existence of metabasins in PELs, at least for nanoglasses. Surprisingly, we find that the distribution of potential energies shows a paired Gaussian and long-tailed distribution at temperatures below and approaching Tg; correspondingly, the distribution of the {\alpha}-relaxation time exhibits an exponential-like decay. In contrast to the Gaussian distribution of energy in typical liquids and solids, the unexpected distribution may reflect the intrinsic feature of nanoglasses. Associated with the exponential-like distribution of the {\alpha}-relaxation time, the stretched-exponential structural relaxation is found, and the maximum stretched behavior appears around Tg. Despite our studies focused on nanoglasses, the current finding may shed light on future studies of bulk glasses., Comment: 23 pages, 10 figures

The intrinsic nature of glass states or glass transitions has been a mystery for a long time. Recently, more and more studies tend to show that a glass locates at a specific potential energy landscape (PEL). To explore how the flatness of the PEL related to glass transition, we develop a method to adjust the PEL in a controllable manner. We demonstrate that a relatively flat PEL is not only necessary but also sufficient for the formation of a nanoscale glass. We show that: (1) as long as a nanocluster is located in a region of PEL with local minimum deep enough, it can undergo a first-order solid-liquid phase transition; and (2) if a nanocluster is located in a relatively flat PEL, it can undergo a glass transition. All these transitions are independent of its structure symmetry, order or disorder. Our simulations also uncover the direct transition from one potential energy minimum to another below the glass transition temperature, which is the consequence of flat PELs.

In order to understand the long-standing problem of the nature of glass states, we performed intensive simulations on the thermodynamic properties and potential energy surface of an ideal glass. We found that the atoms of an ideal glass manifest cooperative diffusion, and show clearly different behavior from the liquid state. By determining the potential energy surface, we demonstrated that the glass state has a flat potential landscape, which is the critical intrinsic feature of ideal glasses. When this potential region is accessible through any thermal or kinetic process, the glass state can be formed and a glass transition will occur, regardless of any special structural character. With this picture, the glass transition can be interpreted by the emergence of configurational entropies, as a consequence of flat potential landscapes.

Molecular dynamics simulations have been used to investigate the structures of Lennard-Jones (LJ) nanowires (NWs) encapsulated by carbon nanotubes (CNTs). We found that, when the radius of CNTs is small, the structures of NWs are quite simple, {\it i.e,} only multishell NWs are formed. For CNTs with larger radius, the structure of NWs becomes much richer. In addition to crystal and multishell NWs, three new kinds of NWs are found, {\it i.e,} a hybrid NW with a crystal core coated by a few shells, a crystal NW with droplet-like pits on the side of NW, and a multishell NW with droplet-like pits on the side of NW. The 'phase' diagram, which describes the structure change with the number of LJ atoms and the interaction between NWs and CNTs, is also obtained.

The end section of a carbon nanotube, cut by acid treatment, contains hydrophillic oxygen groups. Water molecules can self-assemble around these groups to seal off a carbon nanotube and form an "aqueous valve". Molecular dynamics simulations on single-wall (12,12) and (15,15) tubes with dangling carboxyl groups show that the formation of aqueous valves can be achieved both in the absence of and in the presence of high pressure hydrogen. Furthermore, significant diffusion barriers through aqueous valves are identified. It indicates that such valves could hold hydrogen inside the tube with GPa pressure. Releasing hydrogen is easily achieved by melting the "aqueous valve". Such a design provides a recyclable and non- destructive way to store hydrogen in GPa pressure. Under the storage conditions dictated by sealing off the container in liquid water, the hydrogen density inside the container is higher than that for solid hydrogen, which promises excellent weight storage efficiency., Comment: 17 pages, 4 figures

Applying external strain is an efficient way to manipulate the site preference of dopants in semiconductors, however, the validity of the previous continuum elastic model for the strain influence on the doping forma- tion energy is still under debate. In this paper, by combining quantum mechanical theoretical analysis and first-principles calculations, we show that if the occupation change of different orbitals caused by the strain is negligible, the continuum elastic model is valid, otherwise it will fail. Our theory is confirmed by first-principles calculation of Mn-doped GaAs system. Moreover, we show that under compressive strain the hole density, thus the Curie temperature TC can increase in Mn-doped spintronic materials., Comment: 4 pages + 2 figures

A molecular sieve for hydrogen is presented based on a carbon nanotube intramolecular junction and a $C_{60}$. The small interspace formed between $C_{60}$ and junction provides a size changeable channel for the permselectivity of hydrogen while blocking $Ne$ and $Ar$. The sieving mechanism is due to the high compressibility of hydrogen.

The capillary-induced structural instability of an elastic circular tube partially filled by a liquid is studied by combining theoretical analysis and molecular dynamics simulations. The analysis shows that, associated with the instability, there is a well-defined length scale (elasto-capillary length), which exhibits a scaling relationship with the characteristic length of the tube, regardless of the interaction details. We validate this scaling relationship for a carbon nanotube partially filled by liquid iron. The capillary-induced structural transformation could have potential applications for nano-devices.

An approximate treatment of exchange in finite-temperature path integral Monte Carlo simulations for fermions has been proposed. In this method, some of the fine details of density matrix due to permutations have been smoothed over or averaged out by using the coarse-grained approximation. The practical usefulness of the method is tested for interacting fermions in a three dimensional harmonic well. The results show that, the present method not only reduces the sign fluctuation of the density matrix, but also avoid the fermion system collapsing into boson system at low temperatures. The method is substantiated to be exact when applied to free particles.

Recently we proposed a new constant-pressure molecular dynamics method for finite systems. In this paper, we discuss the current understanding of this method and its technique details. We also review the recent theoretical advances of nano-system under pressure by using this method.

A simple pair potential, which equilibrium pair separation can be varied under a fixed interaction range, has been proposed. The new potential can make both face-centered-cubic(fcc) and body-centered-cubic(bcc) structure stable by simply changing one parameter. To investigate the general effect of the potential shape on cluster structures, the evolution of cluster structures is calculated for different equilibrium pair separations. The small size clusters($N<25$), which adopt the polytetrahedra, are almost independent on the details of the potential. For the large size clusters($25

In this letter, by writing the volume as a function of coordinates of atoms, we present a new constant-pressure molecular dynamics method with parameters free. This method is specially appropriate for the finite system in which the periodic boundary condition does not exist. Simulations on the carbon nanotube and the Ni nanoparticle clearly demonstrate the validity of the method. By using this method, one can easily obtain the equation of states for the finite system under the external pressure., Comment: RevTex, 5 pages, 3 figures, submitted to Phys. Rev. Lett