Vol. 58, No. 13 (2009)
Prediction on mechanical parameters of inconsistently random composites by statistical second-order two-scale method
2009, 58 (13): 1-S7. doi: 10.7498/aps.58.1
The statistical second-order two-scale method is developed to predicte the mechanical parameters of the composite structure with inconsistently random distribution, including stiffness parameters and elastic limit strength parameters. This kind of structure, in which random inclusion dispersions are not the same everywhere but gradually changing in space, is called the structure with inconsistently random distribution. So, it possesses continuously varying macro-mechanical properties in space. In this paper, the microstructure of the composites with consistently/inconsistently random distributions is represented. The statistical second-order two-scale analysis formulation is established. And the strength criterion of the inconsistent random composite structure is discussed. Finally, the numerical results for different inconsistent composite structures are obtained and compared with the experiment data. Their agreement shows that the statistical second-order two-scale method is feasible to predict mechanical parameters of inconsistently random composites.
2009, 58 (13): 8-S14. doi: 10.7498/aps.58.8
The support vector regression （SVR） approach combined with particle swarm optimization for parameter optimization, is proposed to establish a model for estimating the density of selective laser sintering parts under processing parameters, including layer thickness, hatch spacing, laser power, scanning speed, ambient temperature, interval time and scanning mode. A comparison between the prediction results and the results from the BP neural networks strongly supports that the internal fitting capacity and prediction accuracy of SVR model are superior to those of BP neural networks under the identical training and test samples; the generation ability of SVR model can be efficiently improved by increasing the number of training samples. The minimum error value is provided by leave-one-out cross validation test of SVR. These results suggest that SVR is an effective and powerful tool for estimating the density of selective laser sintering parts.
Study on the influence of process parameters on the synthesis of AlON-TiN composites by using support vector regression
2009, 58 (13): 15-S20. doi: 10.7498/aps.58.15
According to the experimental dataset on the bending strength of AlON-TiN composite synthesized by hot pressing sintering approach under different processing parameters, i.e., mass fraction of TiN, sintering temperature and soaking time, the support vector regression （SVR） approach combined with particle swarm optimization for its parameter optimization, is proposed to simulate the relationship between the bending strength and hot pressing sintering synthesis parameters of AlON-TiN composites. The optimization of process parameters and the multi-factor analysis are also carried out. The prediction result demonstrates that the estimation error of the SVR model is less than that of the artificial neural network（ANN） model under the identical training and test samples and reveales that the generalization ability of SVR model surpasses that achieved by the ANN model. The optimal synthesis parameters are obtained numerically under TiN content 13.5%, sintering temperature 1863.5 ℃ and soaking time 5.8 h. The maximum bending strength is estimated to be 555.452 MPa while the AlON-TiN composite is synthesized at the optimal synthesis parameters. These results suggest that SVR can provide an important theoretical and practical guidance to the research and development of AlON-TiN composite possessing ideal bending strength.
2009, 58 (13): 21-S28. doi: 10.7498/aps.58.21
The physical model and the computational system for powder packing process simulation are presented, and the high performance computing methods of simulating the random packing of mixed particles with different sizes are studied. In the simulation system the effects of gravity, contact forces, damping, friction, van der Waals force etc are take into account, therby forming several kinds of mechanical models and damping models. The system is very suitable for the three\|dimensional simulation of large scale powder packing process. Finally, to demonstrate the usefulness of the simulation system, two typical applications are presented in the paper. One is to simulate the random packing of binary mixture with size ratio 10, and when the number of small particles is 300 times as large as that of big particles, the maximal packing density （volume fraction） is achieved to be 0.824, and the other is to simulate the random mixed packing dynamics of particles with two different densities. The segregation phenomenon and clustering phenomenon take place obviously after the packing process has come to an end. The physical model and the simulation system presented in this paper are not only suitable for the study of powder packing process, but also applicable to the packing process simulation of spherical objects.
2009, 58 (13): 29-S34. doi: 10.7498/aps.58.29
Heat-processed materials undergoing phase transformation are considered to be a mixture of several phases, and each of them possesses thermo-elastoplastic property. A thermo-elastoplastic constitutive equation involving phase transformation is put forward based on the physics of continuous media and the irreversible thermodynamics with internal variables. The material parameters in the constitutive equation and their variations with temperature are determined by a series of short-time-tension tests at different temperatures. The heat conduction equation including phase transformation and the transformation kinetics equation including stresses are also presented. The numerical algorithm and the finite element procedure related to these equations are developed and applied to the investigation of the distribution of the residual stress in the welding slot area of 1Cr12WMoV stainless steel pipe. The obtained results of the residual stress are in good agreement with the measurements by X-ray diffraction.
ATOMIC AND MOLECULAR PHYSICS
A method of fitting shell model interatomic potential for H2 molecule from electronic structure data
2009, 58 (13): 35-S39. doi: 10.7498/aps.58.35
The shell model interionic and interatomic potentials are widely used to perform atomistic simulations of ionic crystals. The parameters of the potentials are obtained by fitting the data from measurements and/or electronic structure calculations of the crystal discussed. A new method of fitting shell model interatomic potential is presented for diatomic molecule from electronic structure data obtained by quantum chemical ab initio calculation in the paper. And the shell model interatomic potential for H is fitted from the electronic structure data of H2 molecule for example. To examine the method, the interatomic potential and force of H+2 molecular ion are calculated from the potential functions. The results demonstrate that the proposed new fitting procedure shows a fairly good feasibility to the shell model potential for covalent bond molecule.
2009, 58 (13): 40-S46. doi: 10.7498/aps.58.40
The structural evolution of a molten Cu55 cluster embedded in face-centred cubic （FCC） bulk is simulated by molecular dynamics method based on the embedded-atom method potential through mean square displacement, pair analysis indices, and the average energy of atoms. The simulated results show that final structure of the embedded Cu55 cluster mainly presents local FCC structure at the quenching temperatures. During solidification, the atoms continuously interchange their positions, and the rearrangement of atom positions is sensitive to the temperature change. As the quenching temperature increases, diffusion abilities of the atoms increase. At 100—500 K, the formed FCC structures are stable, whereas at 700—1100 K, the local structures of the clusters fluctuate with time step increasing.
Molecular dynamics study of structures of a Cu13 cluster supported on a Cu(001) surface at low temperatures
2009, 58 (13): 47-S52. doi: 10.7498/aps.58.47
Structural changes of a Cu13 cluster supported on a Cu（001） surface at low temperatures of 10 and 50 K are investigated through molecular dynamics simulations with using interatomic potentials from embedded atom method presented by Johnson. The calculations of atom density profiles show that the cohesive energy is affected mainly by the number and the configuration of the atoms which are directly contacted with the Cu（001） surface, as well as some different configurations of the atoms at higher positions. At 10 K, the initial position of the supported cluster has a larger effect on the atomic packing and the binding energy of cluster.
Molecular dynamics study on structural change of a Au959 cluster supported on MgO(100) surface at low temperature
2009, 58 (13): 53-S57. doi: 10.7498/aps.58.53
Structural change of a Au959 cluster supported on the MgO（100） surface at a low temperature of 10 K is investigated by molecular dynamics simulation based the on atomic interchange potentials of the metal/MgO interface, where the parameters are from the ab initio energies obtained based on Chen-Mbius inversion method. The analyses on paired atoms and pair distribution functions show that there exists a deformation process in the Au cluster to adjust the interatomic distance of the interface atoms, owing to the mismatching of the interatomic distancs in the cluster to those in the substrate. After the adjustment of atomic positions, the volume of the supported Au cluster becomes larger than its unsupported counterpart.
2009, 58 (13): 58-S66. doi: 10.7498/aps.58.58
Structural changes of three molten CuN（N=57,58,59） clusters during freezing and two Cu55 clusters with perfect icosahedral geometries in the coalescence processing at 300 K are investigated by molecular dynamics simulations based on the embedded atom method. Simulation results show that both freezing and coalescing processes have distinct stages. There exist great differences in atomic movement and microstructure change among the three CuN（N=57,58,59） clusters during freezing, which result in the different patterns of atomic packing in the three clusters. Of the three clusters, the ordered degree of the Cu59 cluster is the lowest. Initially structural changes of the two Cu55 cluster during coalescence result from large position displacements of atoms due to the deformation, then the atomic diffusion plays a mainly role in changing structure. The atoms far from the contact region between the two clusters can remain their origin structures.
2009, 58 (13): 67-S71. doi: 10.7498/aps.58.67
Structural change of a molten Cu55 cluster on Cu（010） substrate during freezing at two different cooling rates is investigated by molecular dynamics simulations based on the embedded atom method. The analyses of energy per atom and pair distribution functions show that cooling processes have great influence on the structural changes. A rapid cooling rate results in the lower energy of the atoms in the cluster. At a slow cooling rate， the cluster atoms in the substrate are more likely to move to the substrate surface at high temperature， and then they form face-centred cubic configuration.
2009, 58 (13): 72-S78. doi: 10.7498/aps.58.72
Geometries of anionic, neutral and cationic small tungsten clusters Wn （n=3—6） are studied by the first principles method based on the density functional theory. For each of clusters studied, a number of low-lying states with specific geometries are found and some possible candidates for the global minimum are determined. It is found that all global minimum candidates have relatively low spins: singlet or triplet for the neutral clusters, and doublet or quartet for the ionic clusters. When n>3, the clusters undergo a transition from two-dimensional structure to three-dimensional structure, which is not the same as sliver clusters. In order to determine the most promising global minimum candidates for anionic clusters, the photoelectron spectra are simulated and the vertical detachment energy is calculated. All results obtained are in good agreement with available experimental data.
CLASSICAL AREA OF PHENOMENOLOGY
2009, 58 (13): 79-S83. doi: 10.7498/aps.58.79
Casting magnesium alloys are heterogeneous materials containing numerous voids. These voids markedly affect the mechanical behaviour of the materials. In this paper, a void-growth equation is obtained based on the analysis of a spherical void-cell model, and a void-nucleation equation is presented which is related to the increment of intrinsic-time measure. The evolution equation of the voids is obtained by combining the growth equation with the nucleation equation. The obtained void-evolution equation is incorporated into a nonclassical elastoplastic constitutive equation through introducing a softening function, thus obtaining a constitutive equation that involves a void evolution. A corresponding finite element procedure is developed and applied to the description of the rule of the void evolution and the mechanical behaviour of casting magnesium alloy ML308. Computed results are shown to be in satisfactory agreement with experimental data.
Three-dimensional numerical simulation of the edger and corner metal’s flow on edging rolling on roughing of hot continuous rolling process
2009, 58 (13): 84-S88. doi: 10.7498/aps.58.84
Based on the seam-defect appearing on the plate edge of rough-rolling medium billet in the Meishan Iron and Steel Companys hot-rolling units, the five-pass reversible vertical-horizontal rolling of different vertical roll shapes process is simulated numerically. The results show that groove vertical rolls rolling can inhibit double bulging deformation better. On the same rolling process conditions, the value of side tumbling on groove vertical rolling is larger than that on flat vertical rolling, and the value of side tumbling increases with the groove fillet radius increasing. During the rolling process, that the metal at low temperature, high strain and high stress cumulates on the edge of the workpiece may bring the seam-defect along the length of the workpiece. The edge interlayers could be avoided by reasonably designing the vertical roll shape and optimizing the vertical rolls screw system, thereby reducing or avoiding the occurrence of seam-defects.
PHYSICS OF GASES, PLASMAS, AND ELECTRIC DISCHARGES
2009, 58 (13): 89-S93. doi: 10.7498/aps.58.89
By using the vacuum ion plating technique, nitriding, titanium nitride and their complex coatings are plated on AISI D2 steel, and the finite element analysis （FEA） and experiments are employed to investigate their wear resistance. Archard adhesive wear law based on classic model is embedded into the commercial FEA software in the manner of Fortran subroutine, thereby calculating the wear depth of each node during the contact friction process. The wear depths of substrate and three kinds of coatings are quantitatively analyzed. It is shown that duplex treatment has an optimal wear resistance, and the simulation results are in good agreement with experimental measurements.
2009, 58 (13): 94-S103. doi: 10.7498/aps.58.94
A mathematical model of mutually coupled welding arc and weld pool is established for moving tungsten inert gas welding process on SUS304 stainless steels to investigate the argon and the helium arc properties and their effects on the weld shapes. The comparisons of the temperature contour, the current density and the heat flux on the anode surface between the argon and the helium arcs show that the helium arc is more constricted than the argon arc and transfers more heat fluxes to the anode. The effects of buoyancy, electromagnetic force, Marangoni force and drag individually on the weld pool are simulated, and simulation results show that the Marangoni force on the pool surface is one of the main forces affecting the weld shape, independent of shielding gases. Under argon arc, an other dominant force is the plasma drag force. However, the effect of electromagnetic force is stronger than that of the plasma drag force in helium arc. The inward convection induced by the electromagnetic force increases the weld depth, thereby leading the weld depth and the D/W ratio in helium arc to be larger than the ones in argon arc under the same oxygen content. The weld D/W ratio increases first and then maintains a constant with the increase of oxygen content for both argon arc and helium arc. The weld D/W ratio obtained by simulation is in good agreement with experimental results.
CONDENSED MATTER: STRUCTURAL, MECHANICAL, AND THERMAL PROPERTIES
2009, 58 (13): 104-S111. doi: 10.7498/aps.58.104
Based on the Eulerian and the volume averaging methods, a mathematical model of macrosegregation formation with considering solid movement during solidification of large steel ingot is developed. The cases with and without solid movement are studied and compared. The result shows that the A segregation band results from a narrow channel that forms around the position where the solidified shell in the horizontal direction meets the one in the vertical direction, which serves as a barrier to the fluid flow, leading to the accumulation of solute. Effects of critical solid fraction and density of solid on evolution of macrosegregation are investigated. The results indicate that both the positive segregation in the A segregation band and the bottom negative segregation become more pronounced with the critical solid fraction increasing. As the density of solid increases, stronger macrosegregations occur at the upper and bottom parts of the ingot. However, the length of A segregation band is not increased with density of solid. Calculated results show good agreement with experimental ones, except a larger bottom negative segregation zone.
2009, 58 (13): 112-S117. doi: 10.7498/aps.58.112
A mathematic model of mold filling and solidification in horizontal centrifugal casting is built up. Based on this model, the temperature field and the flow field in horizontal centrifugal casting are numerically simulated. The hydraulic experimental results are compared with numerical simulations, indicating that the mathematic model is accurate. Experimental results show that liquid fills the mold in the manner of a spiral trajectory in the centrifugal casting process. The filling pattern tends to be more uniform with the increase of rotational speed. The present study has a guiding significance to the production of horizontal centrifugal casting roll.
2009, 58 (13): 118-S123. doi: 10.7498/aps.58.118
Considering the plane elastic problem of one-dimensional hexagonal quasicrystals with point group 6mm, the theory about plane strain parallel to aperiodic direction is proposed and established. As an application of this theory, the analytic solutions of elastic fields of the quasicrystals with an elliptical cavity perpendicular to quasi-periodical direction are given in the form of general complex variable. In the limit case, our solutions degenerate into the ones of crack problems.
2009, 58 (13): 124-S131. doi: 10.7498/aps.58.124
Based on the two-parameter phase field model of polycrystalline, the grain growth in polycrystalline material during isothermal holding is simulated by using the adaptive finite element method. The calculated results demonstrates that the cusp grain boundaries are inclined to planar and some small grains coarsen into larger ones through grain boundary migration due to the curvature effect. When the grain boundary misorientations are small enough and meet certain energy and geometric conditions, two grains can rotate to reduce external energy of grain boundaries and coarsen into a single grain. All the modelling results are in good agreement with some experimental observations, and the phase-field model can be successfully employed to simulate polycrystalline grain growth.
2009, 58 (13): 132-S136. doi: 10.7498/aps.58.132
The relation among grain edge length, grain size and topology is studied by large-scale Potts model-Monte Carlo simulation. The results show that the grain edge length is related to the number of grain faces by a linear relation, which is also observed to hold in Poisson-Voronoi structure and average N-hedra. The distribution of the grain edge lengths maintains self-similarity in the period of quasi-stationary grain growth. The average length of an edge segment in individual grains varies with grain size （or the number of grain faces）,thereby indicating that assuming the lengths of edge segments in all individual grains to be equal averagely as does in some models is subject to some limitations. The data of Monte Carlo simulated grains and pure iron show that the grain size is related to the face number by a curve convex upward.
2009, 58 (13): 137-S140. doi: 10.7498/aps.58.137
MacPherson-Srolovitz's three-dimensional grain growth rate equation proposed recently by MacPherson and Srolovitz indicates that the change rate of grain volume depends on the mean width of grains and the total length of the triple lines. This rate equation is studied by large-scale Potts model-Monte Carlo simulation in this paper. The result shows that the data of averaged grain growth rate fit the MacPherson-Srolovitz’s three-dimensional grain growth rate equation very well, thereby this rate equation is preliminary verified.
2009, 58 (13): 141-S145. doi: 10.7498/aps.58.141
A series of Fe based compounds with NaZn13-type structure is studied by using inversed interatomic potentials. Calculated results show that Si atoms and Co atoms are preferentially substituted for Fe atoms at 96i site. The addition of Si or Co makes the cohesive energy decrease. It is found that the lattice constants of LaFe13-x-yCoySix and NdFe13-x-yCoy Six decrease with the increase of Co content. The investigation on the phonon density of states indicates that the lower frequency modes are mostly excited by the rare-earth atoms, and the higher frequency modes are mostly excited by Si atoms. Moreover, the addition of Co atoms makes the Debye temperature of LaFe11.5-yCoySi1.5 compounds increase.
2009, 58 (13): 146-S150. doi: 10.7498/aps.58.146
The rare earth intermetallic compounds, RFe2Zn20 （R represents rare earth）, have become one of hot spots in research field due to their excellent magnetic properties with lower concentration of rare earth atoms. The addition of quaternary could influence the structural properties. The crystal structures of ternary RFe2Zn20 and site preferences of quaternary RFe2Zn20-xInx compounds are investigated by using interatomic potentials based on the lattice inversion technique. The calculated results show that the lattice constants and cell volumes of RFe2Zn20 decline linearly with atomic weight of rare earth increasing and the linear relation is independent of whether the quaternary element is added into the compounds. The In atoms substituted for Zn atoms preferentially occupy 16c site and then 96g site, without 48f site. The sequence of site preference is coincident with the measurements and the analyses from interatomic potentials.
2009, 58 (13): 151-S155. doi: 10.7498/aps.58.151
A physical metallurgy principle based model to predict flow stress, recovery, recrystallization and precipitation of Nb-containing micro-alloyed steels during hot rolling is developed. Precipitation kinetics is simulated by the model during hot rolling. The driving force and the equilibrium concentrations are calculated by Thermo-Calc software. The volume fraction of precipitation particles and the particle size of Nb-containing micro-alloyed steels during hot rolling are calculated with considering the effect of recovery and recrystallization on precipitation. The model is verified by the experimental results, and the volume fraction of precipitation particles and the particle sizes are predicted in different hot rolling processes.
2009, 58 (13): 156-S160. doi: 10.7498/aps.58.156
The molecular dynamics simulations are used to study the characteristics of misfit dislocation networks in the γ/γ′ phase interface of a Ni-based single-crystal superalloy. From the analyses of the formation of interphase dislocations, the reaction of dislocations, and the evolution of dislocation networks, we find that the shape of the dislocation network finally evolves into a regular hexahedron from an original tetradedron.
2009, 58 (13): 161-S168. doi: 10.7498/aps.58.161
General theory and formula of phase field simulation are discussed to as certain physical meanings of some phenomenological parameters in the basic model. A new concept of boundary range is suggested to explain the physical backgrounds of the phase order parameter gradients at grain boundary and the diffusion grain boundary, separately. It is argued that the boundary range is not the geometrical boundary width of atom disorder and generally believed to be within 3—4 atom sizes. However, the range has an independent boundary feature to represent the grain boundary energy distribution range in which the solute alloy atoms are segregated from interior grain. A model is established to simulate the realistic spatio-temporal microstructure evolution in recrystallization of a magnesium alloy by using the phase field approach. A set of rules has been proposed to determine the real physical value of all parameters in the model. The simulated results are shown to be in good agreement with reported measurements at the temperatures from 300 to 400 ℃ for up to 100 min. The effect of applied strains field on the microstructure produced during α2 phase to O-phase （orthorhombic phase） transformation in Ti-25Al-10Nb alloy is finally studied by phase field simulation. The effects of strain direction on the volume fraction of O-phase and on the microstructure are investigated. It is also found that a full laminar microstructure can be formed when the applied strain is loaded along 〈1120〉 of α2 phase with magnitude greater than a half of the stress-free transformation strain. The significance and the potential application of the new simulation discovery are discussed.
Numerical investigation of deformation-induced dynamic transformation in Fe-C alloy by using a Q-state potts Monte Carlo model
2009, 58 (13): 169-S176. doi: 10.7498/aps.58.169
The deformation induced dynamic transformation （DIDT） of a Fe-C alloy above austenite transformation equilibrium temperature is simulated by using a Q-state potts Monte Carlo （MC） model. The austenite-to-ferrite transformation, dynamic recrystallization （DRX） of austenite and ferrite, and the ferrite-to-austenite reverse transformation can be simulated simultaneously in the same MC model by building suitable MC transition rules. Meanwhile, an affine transformation model based on vector operation is also coupled with the MC model for the first time for tracking the changes in grain shape during dynamic transformation. The formation of ferrite during DIDT and the influence of austenite DRX on DIDT are simulated based on this MC model. The simulation results show that the competition between the DRX of austenite and the austenite-to-ferrite transformation causes the oscillation behaviour of ferrite kinetics.
Based on the density functional theory calculations， the diffusion of oxygen atoms near the Zr（0001）surface is studied by using the nudged elastic band method. Firstly， the active energy of diffusion of oxygen atoms from the surface face-centred cubic （SFCC） sites into surface hexagonal close-packed sites in Zr（0001） surface is calculated. The active energy for the diffusion is 0.77 eV. Secondly， the active energy of diffusion of oxygen atoms from the SFCC into octahedral interstitial sites between the second and third layers Octa（2，3） via octahedral interstitial sites between the first and second layers Octa（1，2） is calculated too. There are two energy barriers for oxygen atom to overcome. The active energies are 2.14 and 2.57 eV， respectively. As a result， the diffusion of oxygen atom above the Zr（0001） surface is easier than the diffusion into the subsurface.
Non-equilibrium grain boundary segregation of phosphorous during high temperature plastic deformation
2009, 58 (13): 183-S188. doi: 10.7498/aps.58.183
The quasi-thermodynamics and the kinetics of deformation-induced non-equilibrium grain boundary segregation are described. This model is used to predict non-equilibrium segregation of P during high temperature plastic deformation in an austenite microstructure. There is a segregation concentration peak of P near 800 ℃ for a strain rate of 1×10-3 s-1 and the deformation magnitude is 20%. When the deformation reaches 20% in magnitude at 1000 ℃ the segregation of P at grain boundaries increases with increasing strain rate. The predictions are generally consistent with some available experimental observations.
2009, 58 (13): 189-S192. doi: 10.7498/aps.58.189
Approximate analytic and numerical studies on depth-sensing indentation response of ordered cellular/foam materials with glass substrate are carried out based on the elasticity theory and the theory of cellular solids. The observation focuses on the pressure distribution and critical value of indentation depth in the contact zone，which are approximately determined. The theoretical results are compared with the experimental measurements. The comparison shows that they are in good agreement with each other. In order to give a reasonable estimation on the integrality of the film system, the deformation and the failure mechanism are analyzed. And further we observe the indentation energy and derive its analytic expression， whose form is very simple and we suggest that it can be used as a parameter characterizing the indentation response.
2009, 58 (13): 193-S198. doi: 10.7498/aps.58.193
Water adsorptions on TiO2-terminated SrTiO3（001） surfaces at four kinds of coverages （0.25 monolayer（ML），0.5ML，0.75ML and 1ML） are investigatedby by using density functional theory calculations. Molecular and dissociative adsorptions of water are comparatively investigated. The nudged elastic band method is employed to calculate the dissociative energy barrier. The obtained results showed that dissociative adsorptions were energetically more favourable at low coverages （0.25ML and 0.5ML）， whereas molecular adsorption was more favourable at high coverage （1ML）， and particularly a special mixed adsorption was more favourable at 0.75ML. Based on these results， the influence of adsorption coverage on water dissociation is analyzed.
2009, 58 (13): 199-S203. doi: 10.7498/aps.58.199
Molecular dynamics is used to simulate the deposition process of MgO molecules on MgO（001） surface，and substrate temperature and molecular incident energy are discussed in terms of their effects on the diffusivity of MgO molecules and the substrate surface coverage ratio. The simulated results show that with the substrate temperature increasing， vacant sites in MgO film decrease, owing to the increase in diffusivity of the deposited molecules on the substrate. At low temperatures， the substrate surface coverage ratio increases with molecular incident energy increasing. At high temperatures， the surface coverage ratio reaches the maximum at an incident energy of 3.0 eV， and then it decreases with the increase of incident energy.
CONDENSED MATTER: ELECTRONIC STRUCTURE, ELECTRICAL, MAGNETIC, AND OPTICAL PROPERTIES
First principles calculations of structure and the electronic properties of fullerene derivative phenyl-C71-butyric acid methyl ester
2009, 58 (13): 204-S209. doi: 10.7498/aps.58.204
Phenyl-C71-butyric acid methyl ester （［70］PCBM） clusters are investigated by using the B3LYP method with 6-31G（d） basis set. The optimized results indicate that the addition of PCBM into the ［6，6］-junction produces a closed methanofullerene which is thermodynamically stable product; and the addition into the ［5，6］-junction results in an enlarged fullerene （an open fulleroid） which is a kinetically controlled product. The first adiabatic electron affinity for ［70］PCBM is similar to that for C70. The energy gaps of ［70］PCBM are reduced compared with those of C70. PCBM derivatives and show increased level of the lowest unoccupied molecular orbital of fullerenes. From the natural charge populations， it is found that adding PCBM unit onto the C70 cages does not change the charge populations remarkably; attaching a PCBM has no effect on the electronic structures of C70. The results of theoretical calculation suggest that PCBM is not involved in the process of photoelectric conversion but it plays a key role in adjusting the level of HOMO-LUMO for increasing photoelectric conversion efficiencies.
2009, 58 (13): 210-S215. doi: 10.7498/aps.58.210
The site preference of Si in TiAl3 is calculated by using a first principles method based on density functional theory. The analyses of the c/a value， the formation energy and density of states of systems with different substitution behaviours show that Si prefers to occupy the Al site. The site preference is primarily determined by the electronic structure， and the calculated result is in good agreement with the experimental result. In order to further study the site preference， the occupied sites are divided into two different distributions， the loose distribution and the compact distribution. The calculations of total energy， the formation energy and density of states of systems indicate that Si has a preference for Al（2） sites with a loose distribution. When the systems with Al（2） sites are substituted， the value of c/a increases with doped Si concentration increasing， which is opposite to the case where Al（1） sites are substituted. The study also indicates that the limited solubility of Si in TiAl3 lies between 12.5at% and 18.75at%.
2009, 58 (13): 216-S223. doi: 10.7498/aps.58.216
First principles calculations have been carried out to investigate the electronic structure and the mechanical properties of α-Mg3Sb2. The optimized structural parameters and formation energy are in good agreement with the experimental values. The electronic structure is also given， indicating that α-Mg3Sb2 is indirect-semiconductor. The obtained energy band gap of α-Mg3Sb2 is 0.303 eV. The elastic constaint Cij of α-Mg3Sb2 has been calculated，then the mechanical parameters such as moduli, Possions ratio, etc are abtained. By analysizing the mechanical parameters it is found that α-Mg3Sb2 has good ductility and relatively poor plasticity. In the end，the volume conservation of α-Mg3Sb2 is discussed in terms of the total density of states before and after deformations. From the results， it can be seen that the shears corresponding to （C11-C12）/4 and C44 are almost volume conserving whereas the shears corresponding to C11+C12，C33/2 and the bulk modulus related to C11+C12+2C13+C33/2 are not volume conserving.
2009, 58 (13): 224-S229. doi: 10.7498/aps.58.224
Nb doping is proved to be the most promising way of enhancing the oxidation resistance of TiAl alloys， yet the intrinsic mechanism is still unclear. To understand the effect of Nb doping, a first principles study of several point defects in γ-TiAl during oxidation is performed. Besides the determination of the stable defect structures it is revealed that （1） the formation energy of Nb doping increases with Nb content increasing， at the cost of degradation of the phase stability， which is a hindrance to the oxidation resistance; （2） the abilities of interstitial oxygen and titanium vacancy formation are lowered by Nb addition， thus oxygen diffusion and vacancy formation are reduced correspondingly， which enhances the oxidation resistance of the alloy; （3） the effect of Nb doping on the formation energy of point defect is rather localized， therefore the influence of Nb doping is correlated with its distribution and content in γ-TiAl.
Structural，elastic and electronic properties of L12 aluminum phases from first principles calculation
2009, 58 (13): 230-S234. doi: 10.7498/aps.58.230
First principles calculations are performed to study structural， elastic and electronic properties of typical L12 precipitates of Al-based alloys （Al3Sc and Al3Zr）. The calculated formation energy and the cohesive energy show that both typical L12 precipitates of aluminum alloys have a strong alloying ability and Al3Zr phase has a higher structural stability than Al3Sc phase. According to the calculated density of states of these phases， it is found that the higher structural stability of Al3Zr is attributed to an increase in the number of bonding electrons below Fermi level. Three independent single-crystal elastic constants （C11，C12 and C44） at zero-pressure as well as polycrystalline mechanical parameters such as bulk modulus B， shear modulus G， Youngs modulus Y， Poissons ratio ν and anisotropy value A for both phases are calculated. The mechanical properties of both phases are further analyzed and discussed in comparison with experimental observations and other theory results.
2009, 58 (13): 235-S240. doi: 10.7498/aps.58.235
The self-consistent discrete variational method and the DMol method， based on the density functional theory， are employed to study the electronic structure and the doping effect （N，O） of the kink in the 1/2［111］（110） edge dislocation of body-centred cubic Fe. Our calculations of energies （impurity segregation energy and structural energy） show that N and O each have a strong segregation tendency to enter a kink region， which is related to the lattice distortion introduced by the kink. Furthermore， we find that there exist some charge accumulations between impurity and its neighbouring Fe atoms， resulting in unhomogeneous charge distribution in the kink. N and O atoms obtain electrons while the neighbouring Fe atoms lose electrons. It is found that the interactions between impurity N atom and neighbouring Fe atoms are strengthened due to the strong hybridizations between N-p and Fe-3d4s4p states. The migrations of kink and dislocation motion are impeded by N， which may be beneficial to an increase in strength of material. While the interaction between O and its neighbouring Fe atoms is weaker. The localized effect of impurity-kink complex distinctly affects the electronic structure and properties of the system.
2009, 58 (13): 241-S245. doi: 10.7498/aps.58.241
A multi-scale quantum chemistry method is used to study field emission from arrays of carbon nanotubes. The image potential is effectively given by the image potential of an ideal metal sphere with atomic size. When the image potential is included， the field emission current density is 6 times stronger than the results without considering the image potential.
2009, 58 (13): 246-S253. doi: 10.7498/aps.58.246
The dynamical shear deformation induced hexagonal close-packed （HCP）face-centred cubic （FCC） processes in TiAl/Ti3Al system are investigated by the molecular dynamics method. The details of the dislocation initiation， the microstructure evolution and the force field effect are presented and discussed. The analyses of the potential variation and the structural snapshots show that the shear deformation is related to the “stick-slip” behaviour. The interface can transit the energy and counterpoise the deformation between the hetero-phases. The HCP-Ti3Al （FCC-TiAl） shows obvious （little） covariant deformation stage before the initiation of the fault transition. For HCP-Ti3Al， the final structure is the continued FCC stacking in a region with the large force field， and the FCC and HCP plates alternatively occur near the interface. For FCC-TiAl， twin and SISF are the main remnants， and the meta-stable HCP phase may occur with the increment of the force field.
First principles study of electron field emission from the system of BN nano tuber capped and doped with carbon atom
2009, 58 (13): 254-S258. doi: 10.7498/aps.58.254
The electron field emission performances of BN nano tube （BNNT） capped and doped with one carbon atom are investigated through the first principles calculations. The results show that the electronic structures of the systems change obviously. The shift of margin of the density of states （DOS） towards low energy position increases as the applied electric field and the highest occupied molecular orbital （HOMO）-lowest unoccupied molecular orbital （LUMO） gap decrease. The analyses of DOS, local DOS，HOMO, LUMO and their gap consistently indicate that CeqBNNT system is more propitious to the electron field emission than other systems.
2009, 58 (13): 259-S265. doi: 10.7498/aps.58.259
First principles calculations are performed to study the spin polarized transport in zigzag-edged graphene nanoribbons with anti-ferromagnetic ordering. It is found that when a single B or an N atom is doped in the central scattering region, their effects on the different spin components of current are completely different, though they play the same role in reducing the magnetic moments of the edge carbon atoms. In the B doping case, the spin-up component in current is much larger than the spin-down component, while the situation is opposite to the N doping case. This originates from the fact that the spin degeneracy is broken and the spin-up energy levels in the valence band and the conduction band are higher than the spin-down energy levels in both cases. B doping introduces a hole so that the fully filled valence band becomes partially filled and the Fermi level shifts down to the spin-up energy levels in the valence band while the spin-down energy levels in it are a little far from the Fermi level. This strengthens the transmission of electrons in the spin-up channel more than it does for the spin-down electrons. N doping introduces an electron which makes the empty conduction band partially filled so that the Fermi level shifts up to the spin-down energy levels in the conduction band. This strengthens the transmission of electrons in the spin-down channel more than it does for the spin-up electrons.
2009, 58 (13): 266-S271. doi: 10.7498/aps.58.266
A formalism of transfer matrix method is presented and used to solve a one-dimensional time-independent Schrdinger equation based on a simple one-band effective mass model and the envelope function approximation. The accuracy of this method is proved by comparing the numerical solution and analytical solution for a GaAs-based type Ⅰ single quantum well system， and its applicability is demonstrated by experimental photoluminescence results of the InAs/GaSb-based type Ⅱ and broken-gap quantum well structures. The formalism is extended to calculating the subband energies and corresponding wavefunctions in the GaAs/GaAlAs-based type Ⅰ coupled multiple quantum well systems, showing that the formalism is universal and practical.
Topological research on the molar magnetic susceptibility of alkali metal compounds with support vector regression
2009, 58 (13): 272-S277. doi: 10.7498/aps.58.272
According to the experimental dataset on the molar magnetic susceptibility χm of 45 alkali metal compounds and the topological descriptor?——magnetic connectivity index mF， which is extracted by the magnetic valence gi of simple ion deduced from classical electrodynamics， support vector regression （SVR） combined with particle swarm optimization for its parameter optimization is proposed to establish a model for predicting the molar magnetic susceptibility of alkali metal compound via 0F and 1F. The performance of SVR model is compared with that of multivariate linear regression （MLR） model. The results show that the mean absolute error， the mean absolute percentage error and the root mean square error for 9-fold cross validation test of SVR models are all smaller than those achieved by MLR models. It is revealed that the generalization ability of SVR model is superior to that of MLR model. This study suggests that magnetic connectivity index is an effective descriptor and the SVR is a powerful approach to the prediction of the molar magnetic susceptibility of alkali metal compounds.
INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY
2009, 58 (13): 278-S284. doi: 10.7498/aps.58.278
A three-dimensional （3D） model for the simulation of solutal dendritic growth in the low Pclet number region is presented. In the model is adopted a solutal equilibrium approach previously proposed by Zhu and Stefanescu to calculate the evolution of the solid/liquid（SL） interface， which allows the accurate simulation of dendritic growth from the initial unstable stage to the steady-state stage with a good computational efficiency. In this approach， the kinetics of dendritic growth is driven by the difference between the local equilibrium composition， calculated from the local temperature and curvature， and the local actual liquid composition，obtained by solving the solutal transport equation. To describe the specific crystallographic orientation of 3D dendritic growth， a weighted mean curvature （WMC） algorithm， which is incorporated with the anisotropy of surface tension， is proposed to calculate the local SL interface curvature. This approach is convenient to be implemented and to make the transformation for WMC calculation from two to three dimensions. The model is verified by the comparison of our numerical resuets with the analytical ones. The simulated steady-state tip velocity and radius varying with the degree of undercooling of an Al-2wt%Cu alloy are found to be close to the ones predicted by the Lipton-Glicksman-Kurz analytical model. The steady-state morphology of the needle dendrite tip is analyzed. It is found that the tip is nonaxisymmetric and deviates from a paraboloid in the manner of the fourfold symmetry. Finally， the simulated 3D multi-equiaxed dendrites with various crystallographic orientations are presented.
2009, 58 (13): 285-S291. doi: 10.7498/aps.58.285
A two-dimensional lattice Boltzmann method （LBM） based model is developed for the modelling of dendritic growth during alloy solidification in the presence of forced and natural convections. Instead of conventional continuum-based Navier-Stokes solvers，the present model adopts a kinetic-based LBM for the numerical computations of transport phenomena during solidification. Three sets of distribution functions are employed to constitute the LBM evolution equations for numerically calculating fluid flow as well as solutal and thermal transports which are controlled by both diffusion and convection. By solving the LBM evolution equations， the local temperature and the composition at the solid/liquid interface can be obtained. The kinetics of dendritic growth is then investigated based on a solutal equilibrium approach proposed by Zhu and Stefanescu. The model is verified by the comparison between the simulations and theoretical predictions. The simulated upstream tip velocities and radii of the dendrite growing in a melt with natural convections are found to be in reasonabl agreement with the predictions from the modified Lipton-Glicksman-Kurz model that takes into account the effects of convection. For the convective dendritic growth in a forced flow， the simulated growth Pclet number of the upstream tip as a function of the flow Pclet number is very close to the Oseen-Ivantsov solution. It is also found that convection transports heat and solute from the upstream region to the downstream region， producing asymmetrical dendrite that grows faster in the upstream direction， whereas slower in the downstream direction.
Effects of oxygen plasma treatment on tensile deformation of nano-SiO2 sol-gel coating ultra-high molecular weight polyethylene filaments
2009, 58 (13): 292-S297. doi: 10.7498/aps.58.292
The effects of oxygen plasma treatment on tensile deformation of nano-SiO2 sol-gel coating ultra-high molecular weight polyethylene（UHMWPE） filaments are investigated and a new concept for the nano-structural interface between fiber surface and nano-coating is provided. The tensile test results show that the activation volumes of UHMWPE filaments untreated and treated with oxygen plasma ranging from 1718.448 to 16603.070 nm3 by using Eyrings equation are an important description of the properties of the nano-structural interface between fiber surface and nano-coating, in addition to the enhanced ductility of UHMWPE filaments after nano-SiO2 sol-gel coating UHMWPE filaments treated by oxygen plasma. From the results of scanning electron microscope and Fourier transform infrared spectroscopy, it is observed that the uniform dispersion of the nano-SiO2 coating on the UHMWPE filaments treated by oxygen-plasma not only fills in the micro-flaws, but also introduces the activated functional groups into the fiber surfaces.
Effects of helium plasma treatment on tensile behaviour of nano-SiO2 sol-gel coating T300 carbon fiber
2009, 58 (13): 298-S305. doi: 10.7498/aps.58.298
The purpose of the present work is to investigate the effects of helium plasma treatment on tensile deformation of nano-SiO2 sol-gel coating T300 carbon fiber and provide a new concept for the nano-structural interphase between fiber surface and nano-coating. The tensile test results show that the activation volumes of T300 carbon fibers untreated and treated with helium plasma ranging from 681.9628 to 32342 nm3 by following Eyrings equation are important descriptors for the properties of the nano-structural interface between fiber surface and nano-coating, and the ductility of the nano-SiO2 sol-gel coating T300 carbon fibers treated by helium plasma is enhanced. From the results of the scanning electron microscope, X-ray photoelectron spectroscopy and Fourier transform infrared spectroscopy, it is found that the uniform dispersion of the nano-SiO2 coating of the T300 carbon fibers treated by helium plasma can not only fill the micro-flaws, but also cause the occurrence of effective activation surface interaction between carbon fibers and nano-SiO2 coating, along with the introduced activated functional groups on the fiber surfaces.
2009, 58 (13): 306-S312. doi: 10.7498/aps.58.306
The hydrogen substituted graphenes are investigated by using the density-functional theory. And the localized vibrational modes（LVMs） are calculated within the framework of density-functional perturbation theory. It is found that the number of LVMs at high-frequency part of the spectrum is equal to the number of hydrogen atoms in one vacancy，and LVMs from samples with different substituents have different frequencies. The results and analyses indicate that the measuring of LVMs is an efficient method to determine the classification of substituents as well as their corresponding concentrations in proton-irradiated graphene/graphite samples. This method can also be applied to other doped systems.
Modelling of solidification microstructure evolution of twin-roll casting magnesium strip using cellular automaton
2009, 58 (13): 313-S318. doi: 10.7498/aps.58.313
Cellular automaton （CA） is applied to simulating the microstructure evolution during the solidification， and finite element method （FEM） was used for solving the conservation equations which govern the macro-transport phenomena during the twin-roll casting process of the magnesium strip in the molten pool. Coupling CA with FEM can predict the microstructure evolution during solidification involving nucleation and growth of the grains. On the basis of this CA-FEM method，the effects of the main parameters involving pouring temperature， casting speed and pool height on the characteristic of solidification microstructure such as grain size and orientation can be quantitatively clarified by numerical simulation， which can provide the theoretical basis for the optimization of the processing parameters and the control of the solidification structure.
Microscopic phase field simulation on the nucleation incubation period of L12 and D022 phases in Ni75Cr25-xAlx alloy
2009, 58 (13): 319-S326. doi: 10.7498/aps.58.319
Based on the microscopic phase field dynamic model， the effect of Al concentration on the nucleation incubation period of Ni75Cr25-xAlx alloy and the relation between the incubation period and the precipitation sequence are investigated by using the atomic pictures， averaged long-range order parameters and averaged composition deviation order parameters. The simulation results show that the nucleation incubation periods of L12 and D022 phases are related not only to the Al concentration but also to the precipitation sequence. When the Al concentration is smaller than 7.5%， for the D022 phase first precipitated from the matrix, its nucleation incubation period becomes longer with the increase of Al concentration, and so is the L12 phase subsequently precipitated. When the Al concentration is larger than 7.5%， for the L12 phase first precipitated from the matrix， its nucleation incubation period become shorter with the increase of Al concentration， so is for the D022 phase subsequently precipitated. When the Al concentration is equal to 7.5%， L12 and D022 phases both precipitate almost simultaneously， and there is no obvious difference berween their nucleation incubation periods.
2009, 58 (13): 327-S337. doi: 10.7498/aps.58.327
The problems of composite structures containing small periodic perforated configurations are often encountered in the development of composite materials. These structures often consist of material with very fine micro-structures and vary sharply within a very small periodic domain. The traditional simulation of these structures involving multi-scale is very difficult because of the requirement for a tremendous amount of computer memory and CPU running time. The two-scale formal asymptotic expansions of the increment of temperature and the displacement for the structure with small periodic perforated configuration of composite material are given. The two-scale finite element algorithm is described， and simple numerical results are evaluated by two-scale finite element computational method. The numerical results show that the basic configuration and the increment of temperature strongly affect local strains and local stresses inside basic cell. A new effective numerical method is presented for thermoelastic problem in a periodic perforated domain.
2009, 58 (13): 338-S342. doi: 10.7498/aps.58.338
The cyclic softening of a reduced-activation ferritic/martensitic steel, JLF-1, is one of main problems for application. The microstructure of JLF-1 steel after cyclic deformation is analysed by transmission electron microscope and the Vickers hardness is tested. Based on the dislocation theory, the numerical relationship between change in hardness and microstructure of JLF-1 （dislocation density, lath width and dislocation cell size） is given.
2009, 58 (13): 343-S348. doi: 10.7498/aps.58.343
Ti-6Al-4V is a typical α+β titanium alloys. It presents the diversiform microstructures and thus various properties under different heat-treatment regimes and thermomechanical conditions. The thorough understanding of the formation mechanisms and the evolution rules of different microstructures is crucial to the optimization of the alloys. In this paper, the formation and the evolution of sideplates in Ti-6Al-4V alloy are simulated by using phase field method, with the thermodynamics and the mobility databases used as the input of the phase field model and the primary α phase assumed to exist at the grain boundary. The simulation results show that under a certain condition the grain boundary α phase can grow into β matrix and form sideplates, and that the morphology of α colony is related to the interfacial energy anisotropy. The grain boundary orientation has a strong effect on the sideplate structure. In addition, the heat-treatment temperature can also change the morphological features of sideplates. Higher temperature can slow down the growth of sideplates and result in wider sideplate spacing as well.