Vol. 50, No. 4 (2001)
2001, 50 (4): 577-585. doi: 10.7498/aps.50.577
A “M?bius” invariant asymptotic expansion approach to solve any nonlinear integrable and nonintegrable models with any dimension is proposed. Many new Painlevé integrable models with the same dimension can be obtained at the same time. Taking the (2+1)-dimensional KdV-Burgers(KdVB) equation, (3+1)-dimensional Kudomtsev-Petviashvili (KP) equation as concrete examples, we obtain some new higher dimensional “M?bius” invariant models with Painlevé property and the approximate solutions of these models. In some special case, some approximate solutions become exact.
2001, 50 (4): 586-592. doi: 10.7498/aps.50.586
We Study the abundant localized coherent structures of the (2+1)-dimensional nonlinear Schr?dinger (NLS) equation which was derived from the fluid dynamics and plasma physics. Using a B?cklund transformation and the variable separation approach, we find there exist much more abundant localized structures for the (2+1)-dimensional NLS equation. The abundance of the localized structures of the model is introduced by the entrance of an arbitrary function of the seed solution. Some special types of the dromion solutions, breathers, instantons and ring type of solitons are discussed by selecting the arbitrary functions appropriately. The dromion solutions can be driven by some sets of straight-line and curved line ghost solitons. The dromion solutions may be located not only at the cross points of the lines, but also at the closed points of the curves. The breathers may breath both in amplitudes and in shapes.
2001, 50 (4): 593-596. doi: 10.7498/aps.50.593
We start with Klein-Gordon equation on the background of the Kim-black hole and calculate the free energy and entropy of a scalar field by the-brick-wall method. It is shown that the entropy is not only related to the area of an outer horizon but also is a function of inner horizon's area. Taking some appr oximation, we can obtain that the entropy only is proportional to the area of an outer horizon. Further more, the entropy expressed by location parameter of the outer and inner horizon approaches zero, when the radiation temperature of a black hole approaches zero. This satisfies the Nernst theorem. It can be taken as Planck absolute entropy of a black hole.
2001, 50 (4): 597-605. doi: 10.7498/aps.50.597
Numerical Simulations for directed polymers in 1+1 and 2+1 dimensions show that the transverse fluctuations and free energy fluctuations tend to the strong coupling limit: Δx∝tν,ΔF∝tω (ν=2/3,ω=1/3 for d=1+1; ν≈0.6, ω≈0.2 for d=2+1) at any finite temperature as their behaviors at zero temperature for the long length t of polymer. At finite temperatures, as a result of that the specific heat in both 1+1 dimensions and 2+1 dimensions scales as t and the entropy fluctuations in 1+1 dimensions scale as t1/2 , the ensemble energy fluctuations in both 1+1 and 2+1 dimensions and the internal energy fluctuations in 1+1 dimensions are more pronounced than the free energy fluctuations and scale as t1/2. There is an indication of a phase transition in 2+1 dimensions. It is plausible that the entropy fluctuations and the internal fluctuations per unit length polymers for t→∞ will become zero value at high temperatures from nonzero values at low temperatures at some temperature T=TC where the entropy fluctuation reaches maximum.
The rounding-off error introduces uncertainty in the numerical solution. A computational uncertainty principle was explained by using climate model and the Rossler and super chaos system, and the maximally effective computation time and optimal stepsize are discussed. Under optimal stepsize in solving nonlinear ordinary differential equations, self-memorization equations of chaos systems constitute a new approach to numerical weather prediction.
2001, 50 (4): 612-618. doi: 10.7498/aps.50.612
Crises of chaotic attractors are typical phenomena in nonlinear dynamical systems. By means of generalized cell mapping digraph (GCMD) method, we show that a chaotic boundary crisis results from a collision between a chaotic attractor and a chaotic saddle in the fractal basin boundary. In such a case the chaotic attractor, together with its basin of attraction, is suddenly destroyed as the parameter passes through a critical value, simultaneously the chaotic saddle also undergoes an abrupt enlargement in its size.
The method of synchronizing chaos based on driving a certain parameter of chaotic systems using the chaotic signal outside is proposed. The Logistic map and the Lorenz systems are taken as two typical examples. The simulation results show that two identical chaotic systems can be synchronized completely when the change scope of the parameter driven is large enough. On the other hand, the different synchronization states depending on the initial conditions and the controlling parameter are found in Lorenz systems.
2001, 50 (4): 624-628. doi: 10.7498/aps.50.624
We have effectively controled spatiotemporal chaos in homogeneous and heterogeneous coupled map lattice systems by the same phase space compression. Numerical simulation results show that there is a definite functional relationship between the control results and control parameters in a certain region of compressed phase space, when we control the spatiotemporal chaos to homogeneous stable state. Using this control equation and selecting different phase-space compression parameters to control spatiotemporal chaos in a coupled map lattice systems, we successfully obtain various desired stable patterns.
2001, 50 (4): 629-632. doi: 10.7498/aps.50.629
This paper presents the Hénon chaotic system tracking control of reference signal based on the theory of stability of the linear system. We discuss the problem of system synchronization and give the new concept of chaotic synchronization of diverse structures. Computer simulations have illustrated the effectiveness of the proposed method.
2001, 50 (4): 633-637. doi: 10.7498/aps.50.633
Surrogate data testing is a popular method to detect nonlinearity and chaos in time series and has been vastly used in many applications with erratic time series. The explicit null hypothesis often used is that the time series is generated from a linear, stochastic, Gaussian stationary process, including a possible invertible nonlinear static observation function. It is pointed out that the rejection of such a hypothesis may not only result from an underlying nonlinear or even chaotic system, but also from, e.g., a linear, stochastic, non-Gaussian and non-minimum phase sequence. We investigate the power of the test against non-minimum phase sequence.
2001, 50 (4): 638-643. doi: 10.7498/aps.50.638
We study the giant resonances of stable and unstable nuclei in the framework of the consistent relativistic random phase approximation built on the nonlinear relativistic mean field (RMF) ground states. The isoscalar and isovector modes of giant resonances for stable nuclei, such as 208Pb,144Sm,116Sn,90Zr,40Ca,16O and Ca isotope chains are investigated. The contribution to the giant resonances from the Dirac sea states and the currents of vector mesons are examined. The results show that the effects of the Dirac sea states on the isoscalar giant modes are pronounced, but become weaker for light nuclei, while the contributions of the isovector modes are negligible. The few sets of parameterizations of nonlinear RMF model, which are commonly used to give a good description of ground-state properties of finite nuclei, could also well describe the nuclear dynamic properties-giant resonances. For nuclei with the extreme value of N/Z, low-lying collective excitations are found in isovector dipole modes, which are mainly due to the ph excitation of the weakly bound states near Fermi surface and the isospin mixture effect.
2001, 50 (4): 644-648. doi: 10.7498/aps.50.644
Measurements of the total reaction cross section for proton drip line nucleus 12N at about 50.4 MeV/u on Si target were performed on RIBLL. Compared with its neighbor 13N, large enhancement of total reaction cross section σr for 12N is observed. This might be caused by nuclear deformation and nucleon pair effect, and the experimental inaccuracy should not be neglected. The σr of 12N calculated by using a microscopic Glauber model is in agreement with the experimental results.
2001, 50 (4): 649-654. doi: 10.7498/aps.50.649
Light charged particles emitted from the reactions of 35MeV/u 40Ar+197Au are measured experimentally. A new observable,R=d/(d+t+3He+4He), which is defined as the reduced yield of deuterons, has a monotone relation with the entropy and it is independent of the breakup density (ρ/ρ0) and N/Z of the system. The entropy values are extracted by using several methods which use the d/p ratios,dlike/plike ratios, reduced yield of isotope deuterons R and the reduced multiplicity of charged particles Mred, respectively. The entropy values from the reduced yield of isotope deuterons R and the reduced multiplicity of charged particles Mred are nearly the same. These methods can be used in low beam energy heavy ion reactions. The charged particles emitted in rear angles are mainly from an equilibrium-state source with temperature about 4.7±1.2MeV, which are extracted from the double isotope yields ratios, and entropy S/A=2.5±0.5.This implies that the Break-up density is slightly less than 0.1(ρ/ρ0).
ATOMIC AND MOLECULAR PHYSICS
2001, 50 (4): 655-659. doi: 10.7498/aps.50.655
The Hatree-Fock relativistic(HFR) method has been used to calculate the cross section of photoionization and the Bethe coefficient of all charge states of Al. The latter, which dominates the cross section of electron-impact ionization at high energy region, is also related to the cross section of photoionization. Variation of the cross section of different charge states is discussed. The data reported here is in good agreement with that of Verner et al. obtained with Hatree-Dirac-Slater method.
2001, 50 (4): 660-666. doi: 10.7498/aps.50.660
The experimental progress on rf evaporative cooling of 133Cs atomic sample in recent year is reviewed, and the difficulty of magnetically trapped 133Cs atomic Bose-Einstein condensation (BEC) is analyzed. In this paper, we propose an all-optically-cooled and-trapped 133Cs BEC scheme, which is composed of a pyramidal-hollow-beam gravito-optical trap (PHB GOT) from a diode laser (λ=0.852 μm) and a conical-hollow-beam (CHB) GOT from an Ar+ laser (λ=0.5013 μm). In the PHB GOT, the cold atoms experience an efficient hollow-beam induced Sisyphus cooling (i.e., intensity gradient cooling) and repumpimg-beam induced geometric cooling, and they will be cooled to a few photon-recoil limits (～2 μK) from MOT's temperature (～60 μK). Whereas in the Ar+ hollow-laser-beam trap, cold atoms will be further cooled by Raman cooling (or velocity-selection coherent population trapping) and compressed by a blue-detuned covering beam. We have performed Monte-Carlo simulations for PHB cooling process, calculated the optical potential for 133Cs atoms in Ar+ hollow laser beam, and estimated total collision loss and atomic density. Our study shows that the realization of an optically-cooled and-trapped BEC of 133Cs atoms may be possible in our all-optical dipole trap.
2001, 50 (4): 667-673. doi: 10.7498/aps.50.667
The linear dynamical response of the valence electrons of Na2 is investigated using time-dependent local density approximation with frozen ionic positions. We focus on the dipole response in two kinds of laser field with polarizing direction parallel or perpendicular to the connective line between two atoms. The results show that there is a strong anisotropy in the response of Na2.
CLASSICAL AREA OF PHENOMENOLOGY
2001, 50 (4): 674-681. doi: 10.7498/aps.50.674
Finding an optimal network size is one of the major concerns when building a neural network. In using the local extended Kalman filter (EKF) algorithm, we propose an efficient approach that combines EKF training and pruning as a whole. In particular, the covariance matrix obtained along with the local EKF training can be utilized to indicate the importance of the network weights. As a result, the network size can be determined adaptively to keep pace with the changes in input characteristics. The effectiveness of this algorithm is demonstrated on generalized XOR logic function and handwritten digit recognition.
2001, 50 (4): 682-685. doi: 10.7498/aps.50.682
Based on the generalized Huygens-Fresnel diffraction integral, the propagation of the Bessel-modulated Gaussian beams with quadratic radial dependence (QBG beams) through a circle-apertured ABCD optical system has been studied.-In our treatment the circle function is expanded into a sum of complex Gaussian functions, thus a closed-form propagation expression of QBG beams has been derived. As an application example, numerical calculations have been performed for the focusing of QBG beams by an apertured lens. The calculatied results have shown the advantage of our method, and are in good agreement with those obtained by the straightforward numerical integration of the generalized diffraction formula.
2001, 50 (4): 686-689. doi: 10.7498/aps.50.686
A scheme is presented for the teleportation of an unknown atomic state through the Raman interaction of Λ-type three-level atom with a coherent state cavity-field of large amplitude.
2001, 50 (4): 690-692. doi: 10.7498/aps.50.690
The higher power squeezing effects of photon field for odd and even q coherent states have been studied. We find that the squeezing effects are influenced by the values of parameter q, and the odd and even q-deformed coherent states can exhibit odd power squeezing effect but no even power squeezing effect when parameter q is taken values departure from 1 greatly, and these squeezing characteristics are different from those of the conventional odd and even coherent states.
2001, 50 (4): 693-696. doi: 10.7498/aps.50.693
A three-dementional(3D) kinetic model method for KrF laser is presented. The major difference between the method and the previous kinetic method for KrF laser is that the space nonuniform for the electronic pump is considered. The pump rate is considered as a function of 3D-space (X,Y,Z) and time (t). Using the 3D model method, 96101702 experiment is simulated . The calculated laser energy output and lens energy distribution is in agreement with the experimental results. The problem that the calculated result of 1D model is obviously-higher than the experimental result is solved.
2001, 50 (4): 697-703. doi: 10.7498/aps.50.697
The general equation for determining the scattered sound wave coefficients by an elastic cylinder encased in solid matrix is derived.A thin interphase layer between the matrix and cylinder is taken into account.The resonance modes and related eigenfrequencies in the backscattering spectra are discerned based on the Flax's theory.By using the spring model for interphase layer,the effect of the shear stiffness constant KT on the resonance scattering modes is evaluated.
PHYSICAL CELLULAR AUTOMATON THEORY FOR SIMULATING THE FAILURE PROCESS OF MICRO-HETEROGENEOUS MATERIAL
2001, 50 (4): 704-710. doi: 10.7498/aps.50.704
Based on the physical and mechanical characters of micro-heterogeneous material such as rock and concrete,a new theory of physical cellular automaton (PCA) for simulating the failure evolving process of micro-heterogeneous materials is set up,according to the basic law of the transfer of energy.PCA breaks through the traditional cellular automaton method which is only a mathematical operation.It is shown by some laboratory tests that the simulating results by PCA can agree with the laboratory results perfectly,thus,PCA is an effective physical and mechanical method.
PHYSICS OF GASES, PLASMAS, AND ELECTRIC DISCHARGES
2001, 50 (4): 711-714. doi: 10.7498/aps.50.711
Experiments were performed to determine the electron temperatures of Mg/Al disk target and gold disk target mixed with tracing Mg/Al materials irradiated by laser beams at a wavelength of 0.35μm from the “Xing Guang Ⅱ”high-power laser facility.X-ray spectra from the targets were measured by a crystal spectrometer.The technique of multi-configuration Dirac-Fock was used to calculate relevant atomic parameters.Variations of relative intensity of isoelectronic lines from the two tracing ions in laser-produced plasmas with the electron temperature were derived according to local-thermodynamic-equilibrium model.Based on the above-mentioned experiment the electron temperatures of the plasmas generated by focusing laser pulse on the targets were determined via isoelectronic line ratios.
2001, 50 (4): 715-720. doi: 10.7498/aps.50.715
The electron thermal conductivity in the HT-7 tokamak is obtained by analyzing the electron power balance. In the HT-7 device, the electron thermal convection loss can be neglected, compared with ohmic input power, and the thermal conduction loss is the main energy loss mechanism. The electron thermal diffusion coefficient increases with the minor radius, and is about 100 times larger than the neo-classical theory calculation. After siliconization, the electron thermal diffusion coefficient decreases obviously at the outer half region, and the energy confinement time is prolonged.
Since electrons can exchange energy via emitting and absorbing plasma waves,the effect of ion sound waves on electron transport is studied.The electron thermal conductivity is calculated by reducing an additional electron-electron collision term from the Balescu-Lenard collision intergal.The calculated result shows that the electron thermal conductivity becomes smaller than that by Braginskii's theory.
2001, 50 (4): 726-729. doi: 10.7498/aps.50.726
Time-integrated spectra of second harmonic (2ω) light emission from plasmas produced with 1.053μm laser are simultaneously measured at several scattering-angles.By analyzing the relationship between scattering energy and scattering angle,also by analyzing the total 2ω light intensity as a function of incident beam intensity,we found that the Langmuir turbulence,which is produced by cascade of parametric decay instability,plays an important role in the laser-plasma interaction.
2001, 50 (4): 730-734. doi: 10.7498/aps.50.730
A new kind of double-disk gold targets were irradiated by 0.35μm intense laser and-the properties of impvoved X-ray spectra were investigated at “Xingguang”laser facilities.-Collision velocities between the-primary and secondary disks' plasmas have been measured.This double-disk target's design has avoided the influence of the spraying plasma from primary disk to a certain extent, and increased the “clearance” for ablated secondary disk. Two soft X-ray spectrometers were used to measure X-ray radiation from the primary and secondary disks, respectively. An X-ray streaked camera was used to obtain time-spactial images emitted from laser-plasma of double-disks targets. Experimental results were-analyzed physically.
2001, 50 (4): 735-740. doi: 10.7498/aps.50.735
Charge separation electric fields and magnetic fields induced by superthermal electrons escaped from plasmas have been evaluated using a very simple model. The magnitude and configuration of magnetic fields have been obtained. Generally, the circular-shaped magnetic fields are weak because of the charge separation electric field and the followed finite number of the escaped superthermal electrons, and they are mainly distributed in laser focus near the surface of the plasma because of the transverse motion of the escaped superthermal electrons. Only when the electron current is very strong (about kA through the cross-section with 1μm radius),can a megagauss magnetic field be obtained.
2001, 50 (4): 741-747. doi: 10.7498/aps.50.741
The-numerical simulation study of an experimental result of laser-driven shock wave propagation in a planar aluminum foil target is performed using a one-dimensional radiation hydrodynamic code MULTI. The effect of spatial mesh size on numerical simulation resulits is found to be significant and optimum mesh size is obtained.-Series of simulations are performed using different sets of laser absorption coefficients and the most appropriate laser absorption coefficients value for two laser intensity values used in the experiment is obtained .The results suggest that shock velocities and peak shock pressures are in good agreement with the experimental results.
2001, 50 (4): 748-751. doi: 10.7498/aps.50.748
The phenomenon of discharging in twice of underwater wire exploding is discovered-by investigating the time characteristics of pulsed discharging current and voltage. Pictures taking by high-speed shadow photography and streak camera have testified the existence of the phenomenon. Parameters related to the phenomenon are presented in this paper.
CONDENSED MATTER: STRUCTURAL, MECHANICAL, AND THERMAL PROPERTIES
2001, 50 (4): 752-757. doi: 10.7498/aps.50.752
The phase diagram of Co-rich Sm-Co-Ti system was obtained by X-ray diffraction analysis and magnetic measurement. It was found that there exist three series of compounds, SmCo7-xTix, Sm2Co17-xTix and SmCo12-x Tix. The formation ranges for these three series are 0.32≤x≤0.54, 0≤x≤1.13 and 1.45≤x≤1.95, respectively. The 3∶29 phase was not found after annealing Sm9.31Co90-xTix (3≤x≤15) prepared with an excess Sm of 15%, 20% and 25% added, at temperatures ranging from 1303 to 1403 K. It was also found that the Curie temperatures TC and saturation magnetization Ms decrease with increasing Ti content for both SmCo12-xTix and Sm2Co17-xTix compounds. Below TC, the uniaxial and planar anisotropies were observed for Sm2Co17-xTix and SmCo12-xTix compounds, respectively. The anisotropy fields Ba measured at 1.5 K for both SmCo12-xTix and Sm2Co17-xTixcompounds as a function of Ti content x exhibits a peak at certain compositions, which is thought to be associated with the preferential substitution of Ti for Co.
LOCAL STRUCTURES OF MECHANICALLY ALLOYED Fe100-xCux SOLID SOLUTIONS STUDIED BY X-RAY ABSORPTION FINE STRUCTURE
2001, 50 (4): 758-764. doi: 10.7498/aps.50.758
The local structures of the immiscible Fe100-xCux alloys (x=0,10,20,40,60,80 and 100)produced by mechanical alloying have been investigated by XAFS.For the Fe100-xCux(x≥40) alloys,the local structures around Fe atoms change from bcc structure to fcc one,and the Cu atoms maintain the original coordination geometry after milling for 160 h. On the contrary, the local structures around Cu atoms in both Fe80 Cu20and Fe90 Cu10 alloys appears a transition from fcc to bcc structure.We found that the disorder factor σ(0.0099nm)of fcc Fe—Cu phase is larger than that (0.0081nm) of bcc Fe—Cu phase,and the σ(0.0099nm) around Fe atoms is larger than that (0.0089nm) of Cu in the Fe100-xCux(x≥40) alloys. This suggests that the mechanically alloyed Fe100-xCux alloys is not a homogeneous supersaturated solid solution,but consists of Fe-rich and Cu-rich regions for various compositions.A possible mechanism for bcc-to-fcc and fcc-to-bcc changes in Fe100-xCux alloys is discussed in relation to the interdiffusion and transition induced by the ball milling.
2001, 50 (4): 765-768. doi: 10.7498/aps.50.765
The silver nanoparticles of (4.18±0.5) nm in diameter were prepared using two-phase liquid-liquid method.-The size distribution and the stability of the silver nanoparticles were studied by the UV-vis spectrum.-The results showed that the silver colloid solution was stable and monodispersed.-The IR spectra indicated that the nanoparticles were capped by 1-nonanethiol.-Two-dimensional ordered structure of the nanoparticles was formed by self-assembly technique.
CONDENSED MATTER: ELECTRONIC STRUCTURE, ELECTRICAL, MAGNETIC, AND OPTICAL PROPERTIES
2001, 50 (4): 769-774. doi: 10.7498/aps.50.769
A series of YBa2Cu3-xFexOy(x=0—0.5)samples were prepared in air by the standard solid-state-reaction method.The crystal structures and physical properties have been studied by means of positron annihilation technology (PAT), scan electronic microscope (SEM) and X-ray diffraction (XRD).Oxygen contents of the samples were determined by a volumetric method.The results of PAT show that the short lifetime component τ1 and long lifetime component τ2 are all changed anomalously between x=0.12 and 0.15 where the compounds undergo the orthorhombic-tetragonal (O-T) phase transition.In addition,the grain size of the sample with x=0.15 is much larger than that of samples with x0.15,the doped Fe atoms and incoming oxygens tend to associate into clusters rather than to randomly distribute over the crystal lattice.The cluster behavior is in agreement with the results of theory and other experimental results.
2001, 50 (4): 775-778. doi: 10.7498/aps.50.775
The infrared photoconductivity spectroscopy has been employed to investigate the deep levels in semi-insulating p-type Zn0.04Cd0.96Te. At the temperature ranging between 4.2 and 165K, photoconductivity peaks at 0.24,0.34,0.38,0.47,0.55 and 0.80eV are observed. In conjunction with the photoluminescence measurement of the sample at 4.2K, the characteristics of the deep levels related to the photoconductivity peaks are discussed.
INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY
2001, 50 (4): 779-783. doi: 10.7498/aps.50.779
Polycrystalline silicon (poly-Si) thin films were prepared by a remote microwave plasma-enhanced chemical vapor deposition technique at low temperatures using SiF4and H2 as reactive gases.It was found that the impact of the charged particles in the plasma on the growing surface significantly affects the surface reactions and the crystallinity of the resultant films.By applying the bias on the substrates to decelerate the particles,high-quality poly-Si films, with hydrogen content less than 0.9 at% and the FWHM of 520 cm-1 Raman peak as narrow as 4.4 cm-1,have been obtained at a low temperature of 360℃.
2001, 50 (4): 784-789. doi: 10.7498/aps.50.784
Amorphous fluorinated carbon (α-C∶F) films with F/C ratios between 0.11 and 0.62 are prepared by microwave electron cyclotron resonance plasma enhanced chemical vapor deposition using trifluoromethane (CHF3) and benzene (C6H6) as source gases.The effect of microwave input powers on the deposition rates,F/C ratios and the bond configurations of films is analyzed by film thickness measured, Fourier transform infrared and X-ray photoelectron spectroscopy analysis.The increase of microwave power results in the increase of deposition rate,the decrease of F/C ratios,and the abundance of CF and CF3 groups of the films.The abundance of CF2 groups almost does not vary as microwave power increases.The films are mainly composed of CF2 groups and CC bonds can be obtained at higher microwave power.The frequency dependence of ε(1×103—1×106Hz) and tan δ(1×102—1×105Hz)all follow a power law.The electron polarization is the main contributor to dielectric polarization of film.
2001, 50 (4): 790-792. doi: 10.7498/aps.50.790
A new method to prepare large amount of high purity single-walled carbon nanotubes is reported. In this method, a fixed-length metallic graphite anode is arranged at a given angle with the cathode, and then discharged for several minutes in high-temperature helium arc. By using this method, more than one-gram product containing about 60 percent single-walled carbon nanotubes can be obtained, which has been confirmed by the repetitive experiments and transmission electron microscopy (TEM) analysis.
2001, 50 (4): 793-796. doi: 10.7498/aps.50.793
A complete thermally induced intermartensitic transformation has already been observed in a single crystalline Ni52Mn24Ga24 sample with martensitic transformation temperature (Ms) near room temperature.As constrast with the single crystalline sample,polycrystalline samples,even those annealed ones,did not show this intermartensitic transformation.The experimental results indicates that the intermartensitic transformation can be suppressed by mechanically grinding the single crystalline sample to smaller sizes.This has been attributed to the internal stress and serious lattice distortion caused by the grinding process.The intermartensitic transformation might be suppressed by the elastic energy stored in distorted lattice.As the evidence,grinding the sample to 50μm,the martensitic transformation was even disappeared in the present measurements.Based on this point,the uncompleted intermartensitic transformation observed in the previous works can be explained.