Vol. 65, No. 15 (2016)
2016, 65 (15): 150501. doi: 10.7498/aps.65.150501
Studies on direct transport of particles not only attribute to understand many processes in the fields of biology, physics, chemistry, etc., but also to provide suitable methods to artificially control particles and micro-devices. In recent decades, direct transport in channels has aroused the interest of an increasing number of researchers. However, the current researches on direct transports in channels mainly focus on static boundary situations. Considering the fact that the time-variable channels exist widely in reality, the corresponding studies in time-variable channels are of distinct value and significance. Therefore, in this paper, direct transport of particles in two-dimensional (2D) asymmetric periodic time-shift corrugated channel is discussed. Firstly, the corresponding Langevin equation describing the motion of particles in a 2D time-shift corrugated channel is established. The channel discussed here is periodic and symmetric in space but follows a periodic and asymmetric time-shift law. Secondly, the transport mechanism and properties of the above model are analyzed by numerical simulation. The average velocity of particles is chosen to evaluate the transport performance. The relationships between the average velocity and typical systematic parameters are discussed in detail. According to the research, the transport mechanism is analyzed as follows. The asymmetric shift of the channel along the longitudinal direction will cause the distribution disparity of particles along the section direction, which can influence the bound effect of the channel on the motion of particles. Specifically, higher concentration of the particles along the section direction implies weaker bound effect of the channel walls, and vice versa. As a result, the particles exhibit different diffusive behaviors along the positive and negative longitudinal directions, thus inducing a direct current. By investigating the relationships between the average velocity and typical systematic parameters, the conclusions are derived as follows. 1) The average current velocity is proportional to the asymmetric degree of channel since increasing asymmetric degree can increase the bound effect disparity, and thus promoting the direct transport behavior. 2) Higher temporal frequency can increase the directional impetus number in a certain period of time, but makes the distribution of particles more concentrated simultaneously. The competition between these two effects leads to generalized resonance transport behavior as the temporal frequency varies. 3) Wider channels allow particles to diffuse freely in larger space. Therefore, as the channel width increases, the bound effect is weakened and the direct transport is hindered, resulting in a decline in average velocity of particles. 4) The average current velocity exhibits generalized resonance behavior as the spatial frequency varies, which is caused by the competition between the diffusion scale of particle and the spatial period of channel. 5) With the growth of the noise intensity, the current velocity will first increase and then decrease, which means that adding proper noise to the system can enhance the direct transport phenomenon.
2016, 65 (15): 150502. doi: 10.7498/aps.65.150502
The formation mechanism of network traffic flow and its evolution law are closely related to daily activities of travelers. The current studies indicate that the law of network traffic flow evolution is day-to-day; therefore, using days as the scale unit is an important way to illustrate the evolution of network traffic flow. In previous studies, travelers in the network were tacitly assumed to be entirely rational. When the rationality of travelers is limited, the dynamics of the evolution law needs to be re-examined. This paper presents the utility maximization hypothesis in a logit model by using the bounded rationality hypothesis and develops a bounded rational binary logit (BRBL) model. We apply the BRBL model to a day-to-day network traffic flow distribution and discuss the evolution law of day-to-day network traffic flow under the assumption of the limited rationality of travelers. Through a numerical experiment, this paper analyzes the evolution characteristics of network traffic flow. The results are as follows. Firstly, the final state of the network traffic flow process is not only correlated to the cost-sensitivity of travelers and dependence on actual cost, but also strongly related to the degree of the nationality of travelers. Secondly, the system will be either bifurcated or chaotic when either cost-sensitivity increases or dependence on actual cost increases. Moreover, within the group of travelers whose rationality level is low, no matter what the cost-sensitivity of travelers and the dependence on actual cost are, the evolution results are asymptotically stable. Finally, in particular, in certain circumstances, it is easy to achieve stability when the rationallty degree of travelers is very high or very low, while it is not easy to achieve stability when the rationality degree of travelers is medium.
2016, 65 (15): 150503. doi: 10.7498/aps.65.150503
Target waves usually emit concentric circular waves, whereas spiral waves rotate around a central core (topological defect) region, the two forms of waves are closely related due to the similarity of their spatial structures. Spiral waves can be generated spontaneously in a homogeneous system, while target waves usually cannot be self-sustained in the same system. Therefore, spiral waves can be found in diverse natural systems, and target waves can be produced from the spirals with special boundary configurations or central pacemakers. The pacemaker of target wave is an oscillatory source or medium inhomogeneity. To model the inhomogeneity in some realistic situations, we introduce local parameter shifts and simulate the transition from spiral waves to target waves. In this research, the evolution of the spiral waves in the complex Ginzburg-Landau equation is investigated by numerical simulations, and the multi-spiral patterns can be transformed into stable target waves with local inhomogeneous parameter shifts in a two-dimensional (2D) spatiotemporal system. The detailed study shows that the initial multi-spiral waves can be influenced by introducing inhomogeneity in the local area of the system space, and the oscillatory frequency of the system plays an important role in changing the pattern. A successful transition from inwardly propagating spirals to target waves can be observed when the oscillatory frequencies of non-controlled and local inhomogeneous region, which have equal values, are both less than the inherent frequency of system. When we inspect the relationship between oscillatory frequencies and the characteristics of the inhomogeneous region, an intriguing V-shaped line is found in parameter-frequency diagram, and the V-shaped area presents three features. Firstly, the left and right sides of the V-shaped area are symmetrical. Secondly, the propagating directions of target waves from the left and right sides are opposite. An inwardly propagating target wave is formed on the left side of the V-shaped area, and an outwardly propagating target wave stably exists on the right side of the line. Thirdly, as local inhomogeneous parameter 2 increases, the V-shaped area moves towards the local inhomogeneous parameter 2 and decreases simultaneously, and the width of the V-shaped area remains approximately the same. To our knowledge, this V-shaped line is a novel observation, hence the changes of the system frequencies are thought to be provoking. This work presents the numerical experiments and theoretical analyses for the stable conditions of target waves, and therefore provides the ideas in the applications of signal propagation and mode competition.
2016, 65 (15): 150601. doi: 10.7498/aps.65.150601
Frequency is one of the most important physical quantities of electromagnetic (EM) waves. With the development of terahertz (THz) technology, high-precision measurement of THz frequency is required in THz laser development, wireless communication and ultra fine spectrum measurement. The traditional Fabry-Perot (F-P) interferometry and heterodyne detection method are both difficult to achieve high-precision measurement of THz frequency. Within the range of light wave band, the femtosecond optical frequency comb has long been applied to the light wave frequency measurement due to its extremely high accuracy and stability. By using frequency comb method, measurement with accuracy in the order of 10-11 can also be achieved in THz band. To generate THz frequency combs with stable and controllable frequency, it is required to conduct precise stabilization control on repetition frequency of the femtosecond laser. As a result, some special designs are needed for the femtosecond laser in addition to repetition frequency control devices, including the reference signal source, servo-control module, HV drive module, temperature control module, etc., resulting in a rather complicated system. In this paper, a new method for THz frequency measurement by using an unstabilized femtosecond laser is introduced. The laser is free running and the repetition frequency continuously reduces approximately 8 kHz in 6 h, which is the result of a lengthened laser cavity due to the thermal expansion caused by temperature rise after the laser has been switched on. The repetition frequency and beat signal frequency are simultaneously and continuously measured by two frequency counters. The THz frequency can be calculated from the data with accuracy in the order of 10-10. Although the measurement precision is reduced by one order compared with that obtained by using stabilized femtosecond laser, the system is greatly simplified. The femtosecond laser and complicated repetition frequency control devices no longer need to be specifically designed. This new method will greatly expand the applicable scope of the frequency comb method in measuring THz frequency.
Experiment research on dynamic response of copper film at high strain rate by chirped pulse spectral interferometry
2016, 65 (15): 150602. doi: 10.7498/aps.65.150602
That the femtosecond laser pulses irradiate metallic materials thereby inducing ultrahigh strain rates, is an important experimental approach to studying the material behavior under extreme conditions. Femtosecond laser-generated shock waves in metal films have rise times of several picoseconds, the corresponding diagnostic technique is required to work with a higher time resolution, which makes the experimental measurements difficult. Chirped pulse spectral interferometry (CPSI) possesses capabilities of ultrafast time resolution and continuous measurement, thus it provides a diagnostic technique for studying the ultrashort shock wave. In this article, we carry out an experiment on femtosecond laser driven shock wave in copper film and the measurement by CPSI. Laser pulse of 25 fs duration at the central wavelength 800 nm is used, the tested samples are copper films of (5025) nm in thickness fabricated by electron beam sputtering deposition onto cover slip substrate of 180 m in thickness, pump beam focuses onto front surface of the copper film through the transparent substrate and this laser intensity is 2.31013 W/cm2. Chirped pulse spectral interferometry is used to detect the movements of the free rear surfaces of the copper films with temporal and spatial resolution. In the spectral interferometry, linearly chirped pulse is required and obtained by stretching the femtosecond laser pulse with a pair of gratings. The relation between frequency and time of the chirped pulse is accurately measured using asymmetric spectral interference method, which is required for explaining the experimental data. Since CPSI is a single shot diagnostic technique, we obtain the displacement and velocity history of the free rear surface with picosecond time resolution in a single measurement. From the results, the average shock velocity is calculated to be (5.60.2) km/s and the shock wave rise time is determined to be 6.9 ps. According to the shock wave relations, impact pressure and strain rate in the copper film are (57.18.8) GPa and 8109 s-1 respectively, the strain rate is so high that it is hard to achieve by long-pulse laser driven or other loading approaches. Additionally, experimental results also show that the free rear surface alternately experiences acceleration and deceleration, which indicates the spallation in the copper target. It is obvious that chirped pulse spectral interferometry is a reliable approach to studying ultrashort shock waves in metal films.
ATOMIC AND MOLECULAR PHYSICS
2016, 65 (15): 153201. doi: 10.7498/aps.65.153201
The question of how long it takes for a particle to tunnel through a barrier, which was first put forward by MacColl (Phys. Rev. 40 621 (1932)), belongs to the fundamental process of quantum physics and has been the subject of intense debate since then. Many efforts have been devoted to addressing this question about how to define, explain and measure this tunneling time, but widespread controversies still exist in theories and experiments. Attosecond physics offers insights into ultrafast electron dynamics in atoms and moleculars on the attosecond (10-18 s) timescales, and therefore, ionization of atoms or moleculars in a strong laser filed allows for tackling this question in an experimentally and conceptually well-defined manner. The tunneling ionization dynamics of electrons plays an extremely important role in the field, since tunneling is the first crucial step in strong field ionizations of atoms and molecules and underlies virtually all present experiments in attosecond science. In the present paper, the tunneling ionization time of a single-active electron tunneling through a He atom subjected to a step static electric field, defined as a nonvanishing positive time delay between the instant of switch-on of the step static electric field and the one of ionization, is obtained from the numerical solution of the time-dependent Schrdinger equation in one dimension. The results show that the time delay between the instant of maximum probability current at the potential barrier exit and the one of switch-on of the step static electric field and the time delay needed by the ground wave function evolving to the continuum, which can be expressed as the transition element of the incident and transmitted parts of the wave function, are both very close to the Keldysh time explained as the time it takes for the bound electron having velocity = iIp/2 to cross the tunneling barrier. Compared with the definition of tunneling time delay in other literature, the one of the ground wave function evolution to the continuous state is much consistent with the actual ionization process. The reason why the electron tunneling time cannot be defined as the time delay between the maximum ionization rate and the instant of the laser peak field is that the wave function could penetrate the tunneling barrier earlier if a few-cycle optical field is adopted in experiment. According to the analysis in this article, an experimental method of measuring the actual electron tunneling ionization time using the optical field synthesis technique is proposed. The results of this paper will be helpful in tackling the problem of tunneling time in strong ionization.
2016, 65 (15): 153202. doi: 10.7498/aps.65.153202
To explore the dynamic properties of Eu 4f76p1/2ns autoionization process, the autoionization branching ratios of ions and the angular distributions of ejected electrons from the Eu 4f76p1/2ns (n=7, 9) autoionizing states are systematically investigated with the combination of the three-step isolated-core excitation (ICE) and the velocity-map imaging techniques The Eu 4f76sns Rydberg states are populated via a two-step laser excitation, from which the Eu 4f76p1/2ns autoionizing states are excited by the wavelength of the third laser around the Eu 6s+6p1/2+ ionic resonance in order to obtain autoionization spectra and the velocity-map images of ejected electrons from the Eu 4f76p1/2ns autoionizing states. Once the velocity-map images have been measured, both the energy distribution and angular distribution of ejected electrons can be acquired. Moreover, the spectra of the branching ratios and the anisotropic parameters within the autoionization resonances are also measured to observe their energy dependence and the relation with the autoionization spectra. Comparisons of the observed spectra of 4f76p1/2ns autoionizing states with n = 7, 8, and 9 manifest that the ICE technique is more suitable for the higher-n members of autoionization series. It is found that the Eu atoms in the 4f76p1/2ns (n = 8, 9) autoionizing states mainly decay into 4f75d+(9D) ionic state, leading to the population inversion between 4f75d+(9D) and 4f76s+ (7S) or 4f76s+ (9S) ionic states, which is significant for developing the autoionization laser. The angular distributions of the ejected electrons from the Eu 4f76p1/2ns autoionizing states show simple patterns at the energy points corresponding to the peaks of autoionization spectra, and have complicated patterns in the energy regions off the peaks of autoionization spectra, especially in the regions corresponding to the sharp increase or decrease in the autoionization spectra. The above phenomena can be explained with the strength of configuration interaction among different autoionization series converging to different ionic states, which is fluctuated within the energy region of autoionization spectra. In addition, within the autoionization resonance both the spectra of branching ratios and anisotropic parameters vary irregularly, and no obvious correlation with the spectra of 4f76p1/2ns autoionizing states can be found.
ELECTROMAGNETISM, OPTICS, ACOUSTICS, HEAT TRANSFER, CLASSICAL MECHANICS, AND FLUID DYNAMICS
Design and experimental verification of single-layer high-efficiency transmissive phase-gradient metasurface
2016, 65 (15): 154101. doi: 10.7498/aps.65.154101
Polarization characteristic is an important feature of electromagnetic (EM) wave. Manipulating polarization state and controlling propagation direction of EM wave by phase-gradient metasurface (PGM) have become a research hotspot in recent years. However, using transmissive PGM for polarization manipulation often suffers a low efficiency. To alleviate this problem, multilayered structure was utilized. However, it often suffered bulky volume and design complexity. Therefore, engineering a thin high-efficiency transmissive PGM with polarization manipulation is a pressing and challenging issue. In this paper, a single-layer high-efficiency transmissive PGM with cross-polarization conversion and anomalous refraction is designed. To illustrate the working mechanism, the PGM is comprehensively investigated through theoretical analysis, EM simulations and experimental measurements. The unit cell evolving from an electric-field-coupled resonator is carefully designed to exhibit a Pancharatnam-Berry phase gradient. Each rotated element irradiated separately by the normally-incident left-handed circularly polarized (LHCP)and right-handed circularly polarized (RHCP) waves is simulated in CST microwave studio. The results show that the cross-polarization transmission magnitude keeps over 0.9 and does not change as the rotation angle varies. Moreover, the phase shift is twice the rotation angles and the direction of refracted beam is opposite under the above two different polarizations. In addition, the cross-polarization conversion ratio is above 0.9 from 14 GHz to 15.8 GHz. On the premise of high transmission magnitude, the phase of the cross-polarized transmission can be freely manipulated via varying axis orientation. By spatially arranging six unit cells in rotation angle steps of 30, a PGM with a phase difference of 60 between adjacent unit cells is designed. As is well known, linearly-polarized (LP) EM waves can be decomposed into LHCP and RHCP waves with equal amplitudes. Therefore, an LP wave through the PGM will be separated into two counterpropagating CP waves. The high-efficiency anomalous refraction of the PGM is verified from simulated near-field electric field distributions and far field normalized power patterns. The simulated refracted angle is 33.5, which is in accordance with the theoretical designed value (33.75). Moreover, the transmissive power intensity spectrum under the normally-incident LP waves is simulated and measured. The simulated and measured results are in good agreement with each other, showing that the transmitted wave is perfectly split into two counterpropagating waves from 14.9 GHz to 15.3 GHz. Compared with the available transmissive PGMs, our proposed PGM features high efficiency and thin structure with only single layer, making the proposed PGM a promising alternative to manipulating propagation and polarization of EM waves.
2016, 65 (15): 154102. doi: 10.7498/aps.65.154102
Manipulating the propagating direction and polarization state of electromagnetic wave is always fascinating and used in a wide field. One of the approaches to achieving this aim is typically based on steering the propagation phase of wave traveling inside an optical medium, such as dielectric lens. Nevertheless, this approach creates new problems, such as high loss, bulky volume and fabrication difficulty. Recently, metasurface was found to be a two-dimensional equivalence of metamaterial, which attracted a great deal of attention because of its unique properties and capability of manipulating and controlling electromagnetic waves on a sub-wavelength scale. So metasurface serves as an alternative approach to dealing with the loss and fabrication issues, and opens a door for bridging the gap between the fundamental research of the artificial structures and their device applications. A reflective phase gradient metasurface (PGM) achieving the linear-to-circular (LTC) polarization conversion and anomalous reflection simultaneously is designed in this paper. Firstly, the conventional cross-shaped structure is modified for enlarging the phase range. Then, six modified cross-shaped structures are designed cautiously to serve as quarter wave-plates, and achieve 60 phase difference between adjacent structures. The reflection phase difference between x-and y-direction components is 90, and their magnitudes are both equal to 0.5. Secondly, a one-dimensional PGM is constructed by distributing six modified cross-shaped quarter wave-plates one by one. Furthermore, an LTC polarization converter with an area of 216 mm216 mm is designed by placing 366 one-dimensional PGMs periodically. The mirror reflectivity and axial ratio are simulated and measured to verify the performances of LTC polarization conversion and anomalous reflection. The measured sample is fabricated by printing circuit board technique through using FR4 substrate, and a free space method is adopted in measurement in the anechoic chamber. In addition, the operating bandwidth can be evaluated from the reflective power density spectra. The measured results of mirror reflectivity, reflective power density spectra and axial ratio characteristic are in good agreement with the corresponding simulations, which shows that the mirror reflectivity is lower than -10 dB; the axial ration is lower than 2 dB within the frequency band of 13.8-14.7 GHz. Meanwhile, the theoretical reflection angles from the generalized Snell law are consistent with the CST microwave studio simulated results and measured results. Compared with the reported LTC polarization converters, the proposed LTC polarization converter not only achieves polarization conversion, but also can manipulate the output wave direction, thereby it has an important promising application value for microwave engineering and communication system.
System of label-free three-dimensional optical coherence tomography angiography with high sensitivity and motion contrast and its applications in brain science
2016, 65 (15): 154201. doi: 10.7498/aps.65.154201
Combining three-dimensional (3D) imaging ability of optical coherence tomography (OCT) with movement recognition ability of dynamic scattering technique, label-free 3D OCT angiography can be realized, which has a wide range of applications in basic science research and clinical diagnosis. At no expense of line scanning speed, the scale of capillaries can be detected by improving the sensitivity through the interframe analysis. However, there exists a certain residual overlap between dynamic flow signals and static tissue beds due to a series of reasons, thus making it difficult to completely distinguish dynamic flow signals from static tissue beds. Thus, when it comes to threshold segmentation for the blood flow signal extraction, classification error rate is inevitable, resulting in the decrease of the motion contrast of angiogram. In order to reduce classification error rate between static tissue beds and dynamic flow signals for high motion-contrast angiography, we propose a method of component compounding in wavelet domain. Three main steps are needed for this method. Firstly, on the basis of two-dimensional (2D) discrete static wavelet transform, a frame image can be decomposed into multiple levels. Each level has four components, i.e., approximation component, horizontal detail component, vertical detail component and diagonal detail component. Different decomposition levels and types of wavelet can be selected according to the demand. Secondly, the algorithm of inverse iteration compounding is used, which contains the arithmetic mean and the geometric mean of the components of adjacent decomposition levels. The adopted order for inverse iteration compounding is from the last level to the first one. The weight of the arithmetic mean to the geometric mean is one to one. In this way, four compounding components can be obtained. Thirdly, a new frame image with higher motion contrast can be obtained by using 2D discrete static wavelet inverse transform of the four compounding components. Both flow phantom and live animal experiments are performed. The results show that classification error rate decreases by 83% and 71% respectively after component compounding in wavelet domain. Besides, the angiogram has an improved motion contrast and a better vessel connectivity, which may contribute to better and wider applications of OCT angiography. Furthermore, based on the developed system, the preliminary imaging studies on the model of local stroke are conducted. In this experiment, we record the 3D data of SD mouse brain before and after the local stroke and on the tenth day. As a consequence, a clear presentation for the whole process of stroke model formation, vessel damage and vessel recovery is achieved, which may be beneficial to studying the mechanism of local stroke model.
2016, 65 (15): 154202. doi: 10.7498/aps.65.154202
As is well known, a typical measure of the quality of an optical beam is the M2 parameter, but characterizing the beam quality only by M2 is insufficient. A low value of M2 is generally considered to be equivalent to the single-mode operation with a stable beam. However, even when a large amount of power is contained in high-order modes, the existence of a low value of M2is still possible. Hence, a low value of M2 does not guarantee the single-mode operation. Therefore, a new measurement technique, which aims at measuring modal content of high power fiber laser, is proposed and demonstrated in this paper. This method is named spatial and spectral Fourier transform, or F2 transform in short, and it is based on measuring Fourier transform of both spatial domain and spectral domain of output laser. The experimental set is simple in structure and high in robustness. Another advantage of the method is that it requires no prior detailed knowledge of the fiber properties. In this paper, the patterns of the high-order modes between and after Fourier transform are simulated. From the graph it is evident that the energy of spot diffuses outward and is convenient to measure. We also simulate and compare the group delay difference curve of F2 with existing S2, which are well matched with each other. Experimentally, the high-order modes are stimulated by extruding the fiber periodically, which ensures that we can measure it. Firstly, by scanning two-dimensional (2D) pattern of beam after spatial domain Fourier transform and recording the experimental data, and then through the Fourier transform of data in spectral domain, the group delay differences between the high-order modes and the fundamental mode can be obtained. Finally, different modes in spatial domain are reconstructed and the relative power of every mode is calculated. Additionally, we set up an automatic measuring device to verify the effectiveness of the method. The reconstructed modal patterns are presented in the final section of this paper. We can clearly identify the fundamental mode and the high-order modes, such as LP01, LP02, LP03, LP21, LP11, LP12, LP13 and LP14. It reconfirms that this method is feasible. Compared with the S2 method, this method reduces the requirement for precision of mobile platform greatly and thus it is suited to measure the modal content of high power fiber laser output beam. This technique can be effectively applied to a wide variety of measurements, such as dispersion compensator of large-mode-area fiber, bend loss measurement of the high-order modes, refractive index profiles measurement of fiber and mode convertor fiber.
Phychography is an important technique in the quantitative phase imaging research domain, which employs the illuminating probes to scan the specimen in an overlapped requirement, and the reconstruction is conducted by using the ptychographic iterative engine. But the contradiction between the imaging efficiency and quality has become a bottleneck for its wide applications. In this paper, we start with the fundamental principle of the iterative algorithms for ptychographical imaging, and propose two parallel schemes based on CPU and GPU, besides the influences of the specimen size, the number of blocks and illuminating beams on the speedup of the two schemes are investigated via simulation experiment. The result shows that the complex amplitude of the specimen can be correctly reconstructed, meanwhile, the speed is significantly improved, which reduces the time consumed by one order of magnitude. This improvement solves the above contradiction, so that we can expect to achieve quasi-real-time imaging. The experimental data also indicate that 1) in optimal partition, parallel speedup is related to the size of the specimen, bigger size is corresponding to more obvious acceleration; 2) the same specimen under different partitions will speed up to different extents, which is closely related to the experimental hardware, however the number of illuminating beams has no significant effect on the speedup.
High repetition rate and high beam quality joule level Nd: YAG nanosecond laser for Thomson scattering diagnosis
2016, 65 (15): 154204. doi: 10.7498/aps.65.154204
A joule-level Nd: YAG nanosecond laser of high repetition frequency and high beam quality is developed for Thomson scattering diagnosis. The laser is designed as a master oscillator power-amplifier system mainly including single longitudinal mode seed, pre-amplifier unit and energy extraction unit. The single-longitudinal-mode Q-switched laser of a high stability is taken as the seed laser of output pulse at J level. The pre-amplifier unit amplifies the J-level pulse laser beam into hundreds of mJ level. In order to obtain the high-quality laser beam output, phase conjugation is adopted to compensate for the laser beam distortion. The ultra-filtered FC-770 is taken as an SBS gain medium of 0.0011 cm-1 absorption coefficient, 197.9 GW/cm2 optical breakdown threshold and 3.5 cm/GW gain coefficient. The double-pass amplification of SBS phase conjugation could realize a real-time repair towards the non-uniformity, deformation and wavefront aberration caused by thermal distortion of the optical components and the laser amplifier to achieve the uniform amplified beam output of high quality close to the diffraction limit. In the energy extraction unit, the amplifier of large-diameter slab is used for energy amplification. The size of the slab is 7 mm 35 mm138.2 mm of 56 cutting angle and 0.6% Nd3+ doping concentration. The slab is plated by a layer of SiO2 against light leak. Horizontal pumping mode is adopted. And the slow axis of the laser diode is almost the same as the length of the slat and the direction of laser transmission. The single-plane array is composed of 8 groups of vertical stacks and each group consists of 12 laser diode bars of power 200 W. At 200 Hz repetition frequency, 250 s pump pulse width and 140 A pump current, the up to 2.3 J stored energy can be achieved The energy extraction unit achieves high gain amplification and finally outputs high-quality laser beam. Under the condition of 200 Hz high repetition frequency and 8.23 J single pulse energy injected by the single longitudinal mode seed, 1.85 J output energy is gained. The energy extract efficiency of the laser system is 52.46%. The output laser possesses a pulse width of 5.36 ns, a far field beam spot 1.72 times the diffraction-limited value, and 1.3% energy stability (RMS).
2016, 65 (15): 154205. doi: 10.7498/aps.65.154205
The 1319 nm lasers have important applications in the fields of optical fiber communication, laser medical treatment and laser color display. The Nd:YAG laser pumped by 808 nm laser diode is an efficient alternative to achieving 1319 nm laser output. In recent years, direct pump technology using 885 nm laser diodes has become more promising due to the dramatically reduced thermal effect and improved optical conversion efficiency. Quasi-continuous sodium beacon laser with microsecond pulse duration generated by the sum-frequency of 1319 nm and 1064 nm lasers can provide a gatable pulse format to eliminate the interference of atmospheric Rayleigh scattering and mitigate the spot elongation of sodium guide star to improve imaging accuracy. However, relaxation oscillation in the microsecond pulse could cause the damage to the nonlinear crystal and reduce the efficiency of sum-frequency generation. It is effective to suppress the relaxation by taking advantage of second harmonic generation, in which a nonlinear crystal is utilized to reduce the pulse peaks with higher intensity. In this paper, we demonstrate a high-power relaxation-oscillation-free quasi-continuous microsecond pulse 1319 nm laser by using the dual-end 885 nmdiode-pumped three-mirror ring-cavity. Intra-cavity etalon and customized mirror coating are employed to prevent the 1064 nmand 1338 nmline of Nd:YAG laser crystal from oscillating. A power tuning device, including a thin-film polarizer and a halfwave plate is implemented as the output mirror of ring cavity, which enables continuous adjustment of the out coupling ratio. The output power of the 1319 nm polarized laser is 22.5 W pumped by 150 W 885 nm laser diode. The repetition rate is 800 Hz and pulse width is 150 s. The corresponding optical conversion efficiency is 15%. The beam quality factor M2 is measured to be Mx2= 1.35 and My2=1.24. By precisely adjusting the temperature of etalon viz. adjusting refractive index as well as thickness of the etalon material, laser wavelength is tuned from 1318.888 nm to 1319.358 nm, corresponding to a tunable range of 470 pm and tuning accuracy of 0.7 pm. A 1319 nm frequency doubling crystal KTiOPO4 (5 mm5 mm15 mm, = 59:8 and ϕ = 0) is inserted into the cavity to suppress the relaxation oscillation. The pulse waveform quickly reaches a smooth regime, followed by a pulse spike at the initial stage and the loss of laser output power is only 1%. It is proved that it can be efficiently suppressed by inserting a frequency doubling crystal with negligible power loss. In conclusion, this paper provides a practical and effective technical means for achieving the high-power relaxation-oscillation-free quasi-continuous 1319 nm laser with microsecond pulse duration.
2016, 65 (15): 154206. doi: 10.7498/aps.65.154206
In recent years, the collisional redistribution of radiation and collision-induced broadening of Rydberg atomic spectral lines by buffer gas perturbation have aroused the renewed interest. Rydberg atoms having a large dipole moment and long lifetime can interact with each other coherently for relatively long time, which makes them a potential candidate for quantum information processing. Besides, collisional redistribution has an important potential application in laser cooling and trapping. Based on previous experimental data, in this paper, two-nondegenerate four-wave mixing (NFWM) for studying atom collision, composed of two-photon resonant NFWM and collisional redistribution NFWM, is reported. The spectrum variation of the two-NFWM affected by the pressure, temperature, detuning and collision-broadening rate coefficient is analyzed. The principle of two-NFWM involving three incident beams is explained as follows. Consider two-NFWM in a |0-|1-|2 cascade three-level system, where states between |0 and |1 and between |1 and |2 ightangle are coupled by resonant frequencies 1 and 2 , respectively. Beam 1 with frequency 1 propagates along the direction opposite to the direction of beam 2, beams 2 and 2' have the same frequency 2, and between their directions there exists a small angle. Assuming that 1 1 and 2 2 so that 1 drives the transition from |0 to |1 while 2 drives the transition from |1 to |2, the simultaneous interactions of atoms with beams 1 and 2 will induce atomic coherence between |0 and |2 through two-photon excitation. This coherence is probed by beam 2', and as a result a two-photon resonant NFWM signal of frequency 1 is generated in the direction almost opposite to the direction of beam 2'. To avoid strong absorption at the resonant frequency of transition from |0 to |1, here the wavelength of beam1 is detuned from the exact resonance. An atom population of level |1 caused by collisional redistribution can be induced when a certain buffer gas pressure is imposed. The collisional redistribution NFWM process also exists in this case. Beam 2 drives the transition from |1 to |2 to induce an atomic coherence which is probed by beam 2' for giving rise to an atomic population grating. A collisional redistribution NFWM signal propagating along the same direction as the two-photon resonant NFWM signal is generated when beam 1 is scattered by the grating. Much information about atomic collisions can be obtained by analyzing the two NFWM signals. In a cascade three-level system composed of ground state, intermediate state and Rydberg state, and the two-NFWM can be used to investigate not only the broadening and shifting of the Rydberg level but also the collisional redistribution of the intermediate state. Unlike other experiments studying the pressure dependence of the longitudinal relaxation rate of atom states, this technique is a purely optical coherent means, and can measure the transverse relaxation rate 20 between Rydberg state and ground state as well as the pressure dependence of the transverse relaxation rate 21 between Rydberg state and intermediate state.
Previously reported chalcogenide glass Raman fiber lasers are made of glass compositions such as As2S3 or As2Se3. However, due to the high toxicity of the element arsenic, there is a potential risk in the glass preparation, fiber drawing, and testing processes. Therefore, we need to explore new environmentally friendly chalcogenide glasses that do not contain As for Raman fiber lasers. Studies have shown that the chalcogenide glasses of Ge-Sb-Se system have excellent infrared transmissions and good environmental friendliness, and thus they are excellent candidates for chalcogenide glass Raman fiber lasers. However, their Raman gains have not been reported. Then Raman gain coefficients can be obtained by experimental measurements and theoretical analyses. The experimental method requires expensive laboratory equipments, a complex optical path, and precision adjustments. Therefore, the design and preparation of new chalcogenide glass fiber with high Raman gain require the theoretical analysis of the Raman gain characteristics in a particular glass component glass. In this work, four chalcogenide glasses, respectively, with compositions of As2S3, As2Se3, Ge20Sb15Se65 and Ge28Sb12Se60 (mol%) are prepared. Refractive indices, infrared transmission and Raman spectra of these glass samples are measured. By using spontaneous Raman scattering theory combined with the measured Raman spectral data, the values of Raman gain coefficient gR of the chalcogenide glasses are calculated and calibrated by a quartz glass sample. Results show that the gR of As2S3 glass is 6010-13 m/W at 230 cm-1 Raman shift and the gR of As2Se3 glass is 22310-13 m/W at 340 cm-1 Raman shift, which are consistent with the experimental results reported in the literature. Compared with the traditional method, the present method used for calculating the fiber Raman gain coefficient provides great convenience for exploring new chalcogenide glasses with high Raman gain. By using this method, we obtain the gR values of Ge20Sb15Se65 and Ge28Sb12Se60glasses at 200 cm-1 Raman shift, which are 21510-13 m/W and 11110-13 m/W respectively. Meanwhile, we analyze the effects of composition and network structure of chalcogenide glass samples on the Raman gain coefficient and gain spectrum. There are two Raman peaks at 165 cm-1 and 200 cm-1 Raman shift, which are attributed to Ge-Ge bond vibration and Ge-Se bond vibration of common apex GeSe4/2 tetrahedral structure respectively. It could be found that the Raman gain coefficient of Ge20Sb15Se65 glass is bigger than that of Ge28Sb12Se60glass at 200 cm-1 Raman shift because of more Ge-Se bonds. By further optimizing the ratio of components of Ge-Sb-Se chalcogenide glass, we could obtain higher Raman gain coefficient at a particular frequency shift. These results show that the Raman gain coefficient of Ge-Sb-Se chalcogenide glass without poisonous element is up to over 200 times that of the ordinary quartz glass, which provides a new possibility for environment-friendly Raman fiber laser material.
2016, 65 (15): 154208. doi: 10.7498/aps.65.154208
Goos-Hnchen shift is a special optical phenomenon. With the development of the nano-optics, Goos-Hnchen shift has become one of the most valuable and hottest issues in optical field. Meanwhile, due to the unique feature of the near-zero-refractive-index material, it has been used in many fields, but the effect of Goos-Hnchen shift has little studied and received less attention. As a result, the purpose of this paper is to analyze the Goos-Hnchen shift based on near-zero-refractive-index material. In the paper, the photonic crystal with specific parameter is used to simulate the near-zero-refractive-index material, and the measurement in the simulation is based on finite difference time domain. We approach the issue by studying whether and how the wavelength and temperature will affect the Goos-Hnchen shift based on near-zero-refractive-index material. After the simulation at different wavelengths and temperatures based on the incidence angle which gives rise to total reflection, the results of the simulation reveal that when wavelength is between 1.648a and 1.848a (not including 1.848a), the Goos-Hnchen shift is positive and increases gradually, and the total reflection angle decreases. When wavelength is between 1.848a and 2.048a, the total reflection angle increases. When the wavelength is in a range between 1.848a and 1.858a, the Goos-Hnchen shift is negative. When the wavelength is above 1.858a, the Goos-Hnchen shift is negative and increases gradually. When the temperature increases from 0 ℃ to 100 ℃, the Goose-Hnchen shift is unsimilar to the situation of different wavelengths, and fluctuates in the interval at wavelengths ranging from 1.648a to 1.848a, and the total reflection angle increases gradually. Goose-Hnchen shift decreases at a wavelength of 2.048, and the total reflection angle decreases gradually, but a little. Based on the simulation result, it is concluded that the variations of the wavelength and temperature will affect the Goos-Hnchen shift based on near-zero-refractive-index material, and the effective value is in a range from about 1a to 4a, which is not a small value to the shift especially in some precision instruments. As a result, the changes of wavelength and temperature should be taken into consideration, when Goos-Hnchen shift based on near-zero-refractive-index materials is measured or used in research. These findings are expected to be instructive for device design and nano-optics.
2016, 65 (15): 154701. doi: 10.7498/aps.65.154701
The lattice Boltzmann method (LBM) was proposed as a novel mesoscopic numerical method, and is widely used to simulate complex nonlinear fluid systems. In this paper, we develop a lattice Boltzmann model with amending function and source term to solve a class of initial value problems of the FitzHugh Nagumo systems, which arises in the periodic oscillations of neuronal action potential under constant current stimulation higher than the threshold value. Firstly, we construct a non-standard lattice Boltzmann model with the proper amending function and source term. For different evolution equations, local equilibrium distribution functions and amending function are selected, and the nonlinear FitzHugh Nagumo systems can be recovered correctly by using the Chapman Enskog multi-scale analysis. Secondly, through the integral technique, we obtain a new method on how to construct the amending function. In order to guarantee the stability of the present model, the L stability of the lattice Boltzmann model is analyzed by using the extremum principle, and we get a sufficient condition for the stability that is the initial value u0(x) must satisfy |u0(x)|1 and the parameters must satisfy i-(1+)(t)/(x), (i=1-4). Thirdly, based on the results of the grid independent analysis and numerical simulation, it can be concluded that the present model is convergent with two order space accuracy. Finally, some initial boundary value problems with analytical solutions are simulated to verify the effectiveness of the present model. The results are compared with the analytical solutions and numerical solutions obtained by the modified finite difference method (MFDM). It is shown that the numerical solutions agree well with the analytical solutions and the global relative errors obtained by the present model are smaller than the MFDM. Furthermore, some test problems without analytical solutions are numerically studied by the present model and the MFDM. The results show that the numerical solutions obtained by the present model are in good agreement with those obtained by the MFDM, which can validate the effectiveness and stability of the LBM. In conclusion, our model not only can enrich the applications of the lattice Boltzmann model in simulating nonlinear partial difference equations, but also help to provide valuable references for solving more complicated nonlinear partial difference systems. Therefore, this research has important theoretical significance and application value.
PHYSICS OF GASES, PLASMAS, AND ELECTRIC DISCHARGES
Experimental and computational study of damage pocess induced by 1064 nm nanosecond laser pulse on the exit surface of fused silica
2016, 65 (15): 155201. doi: 10.7498/aps.65.155201
Material response and the launch of laser plasma during the 1064 nm nanosecond laser pulse induced damage to the exit surface of fused silica are investigated. Employing a polarization-based two-frame shadowgraphy setup with ～ 60 fs probing resolution, the transient material responses from the rising part of nanosecond pumping pulse to several hundred nanosecond timescale are captured. Using a shearing interferometry setup, the evolution of transient phase shift of laser plasma in the expansion process to the ambient air is also investigated. Inhomogeneous distribution of phase shift caused by the electrons and neutrals in the plasma is quantitatively resolved by employing the fast Fourier transform based filtering algorism. To demonstrate the evolutions of important plasma parameters such as pressure, temperature and density, a continuum hydrodynamic model is numerically solved. The initial pressure of plasma is estimated according to the point-explosion model, and the initial plasma temperature is achieved by calculating the difference between simulating shockwave front radius and experimental value at the same delay. The optimal temperature is chosen when the radius difference is minimal. Main conclusions are as follows. 1) Abundant suprathermal electrons are excited in the early energy deposition process. Part of these electrons contribute to the thermal transport process and produce the laser supported solid-state absorption front (LSSAF) which propagates into the bulk silica. Other electrons escape to the air side and contribute to the formation of air plasma through the impact ionization process. Plasma expansion speed is about 20 km/s during this phase. 2) When the pump pulse is terminated, the LSSAF and air plasma lose their energy supplied and experience a rapid decline of the temperature and expansion velocity. As a result, the final damage crater depth exhibits seldomly no increase compared with the transient crater depth during this phase. Hot bulk plasma formed in this phase becomes the damage precursor and induces the ejection of abundant neutrals probably due to the phase explosion mechanism. Inhomogeneous distribution of stress is formed by Rayleigh-Taylor instability at the interface between hot bulk plasma and surrounding bulk material during the expansion of LSSAF. Radial and circumferential cracks are formed due to the release of stress. 3) Evolution of air plasma follows the conventional evolution process of laser-induced plasma, i. e. , internal pressure, temperature and density decrease quickly with time delay. The simulated transient highest pressure is about 600 MPa. Simulation also predicts the formation of the internal shockwave. Our work will be helpful in understanding the laser damage mechanism of the fused silica optical window.
CONDENSED MATTER: STRUCTURAL, MECHANICAL, AND THERMAL PROPERTIES
High magnetic field influence on the molecular orientation and the morphology of iron phthalocyanine thin films
2016, 65 (15): 156101. doi: 10.7498/aps.65.156101
Molecular orientation and stacking mode are commonly considered to have vital influence on the optoelectronic performances of organic semiconductor devices via changing the dynamics of charge carriers transferring among the molecules. Highly ordered and homogeneous stacking would allow a fast band transfer mechanism in the phase domain. Therefore the controls of the molecular orientation and the stacking behavior are of great significance for optimizing the device natures. In this work, the modification and control of iron phthalocyanine (FePc) molecular orientation on Si(111) are accomplished with the aid of high steady magnetic field at room temperature. The FePc films are grown in situ by organic molecular beam deposition on the Si(111) substrates under a high magnetic field strength of 8.5 T. The Si(111) substrates are preserved at room temperature and are kept perpendicular to the magnetic field. The influences of magnetic field on the molecular orientations and the morphologies of FePc thin films are investigated by X-ray diffraction, angle dependent near edge X-ray absorption fine structure (NEXAFS), Raman spectroscopy and atomic force microscopy (AFM). In the presence of the external magnetic field, the deposited FePc films each show a higher crystallinity and slightly closer packing in (002) plane than those without magnetic field. The AFM images verifies more ordered and uniform morphologies of the FePc films grown in the magnetic field. NEXAFS and Raman results both reveale a standing-up configuration of FePc molecules on the Si(111) substrate surface. The average tilting angle of the molecules changes from 63.6 to 67.1 when 8.5 T magnetic field is employed. The results demonstrate that the external high magnetic field distinctly enhances the orientation order of FePc molecules on Si(111) surface due to the magnetic-magnetic interactions between the magnetic field and the molecular magnetic moment. This work also demonstrates that external magnetic field is an efficient means to regulate the orientation and stacking behavior of magnetic molecules, which may open a new way to optimize the performances of the organic semiconductor devices.
2016, 65 (15): 156201. doi: 10.7498/aps.65.156201
Platinum metal Ir-Rh alloy presents a promising candidate as future ultra-high-temperature gas turbine material due to its excellent high-temperature properties. In this paper, the mechanical properties of Ir-xRh (x=0, 10, 20, 30, 40, 50, 60, 70, 80, 90, 100) alloys with different rhodium content are investigated. Self-consistent, periodic, density functional theory calculations, Perdew-Burke-Ernzerhof functional, virtual crystal approximation are employed to calculate the elastic constants C11, C12, C44, Cauchy pressure (C12-C44), Young modulus E, shear modulus G, bulk modulus B and the ratio G/B, anisotropic factor A, and strain energy of dislocation per unit length. These parameters are adopted to characterize and assess the effect of Rh content on the mechanical property of Ir-Rh alloy. The results indicate that it is reasonable to use the virtual crystal approximation to calculate the mechanical properties of Ir-Rh alloys. The Young modulus E, shear modulus G and bulk modulus B increase rapidly with the increase of rhodium content, and the maximum value is reached at rhodium content 10%. Then it fast dereases down to a minimum value at 40% after the slowly rises and then slowly drops down. It is found to be in remarkable agreement with the strain energy of dislocation per unit length. This indirectly explains its changing trend. The Cauchy pressure (C12-C44), G/B value and the Poisson's ratio reflect the change of the brittleness of the alloy. Therefore, we can come to a conclusion: the addition of Rh can cause the brittleness of the Ir-Rh alloys. The value of the brittleness first increases and then decreases with the increase of Rh content, and its maximum value is reached at 50%. The charge densities and the densities of states of pure Ir, Ir-10Rh, Ir-50Rh and pure Rh are calculated and compared. At the same time, we also establish a 2 2 1 solid solution supercell structure of Ir-Rh alloy and calculate its differential charge density. The results show that in the Ir-Rh alloys exists a pseudo covalent bond, which leads to the abnormal mechanical properties. The pseudo covalent bond is not a metal bond nor a covalent bond but a kind of transition bond or a mixed type. Finally, the experimental results show that the calculation method is reasonable and it can play an important role in understanding the microscopic mechanism of the abnormal mechanical properties of Ir-Rh alloys.
2016, 65 (15): 156301. doi: 10.7498/aps.65.156301
The thermoelectric material is a kind of new functional material, which can convert industrial waste heat and automobile exhaust into the available electric energy by the interaction of carriers. It is widely used in energy, environment, national defense and other fields. For the research of thermoelectric materials, it is the most important to improve the conversion efficiency now. Due to their unique structural properties, the ternary chalcopyrite semiconductors I-III-IV2 (I=Ag, Cu; III=Al, Ga, In; IV=S, Se, Te) display the better thermoelectric performances at high temperature. Many studies show that there are many ways to improve their performances. In order to optimize their thermoelectric efficiencies the structural, elastic and thermoelectric properties of CuGaTe2 and CuInTe2 are studied by employing the density function theory and semi-classical Boltzmann transport theory within the constant time approximation. The electronic band structures are calculated using the Tran-Blaha modified Becke-Johnson potential (MBJ-GGA) and the generalized gradient approximation (GGA). The calculated band gaps with MBJ-GGA of CuGaTe2and CuInTe2 are 0.86 and 0.56 eV, which are more accurate than the calculated values with GGA. The shear modulus, and Young's modulus and sound velocities are determined from the obtained elastic constants. The constant-volume heat capacity is estimated based on the quasi-harmonic Debye model. The calculated temperature dependence of heat capacity agrees very well with the experimental result. Below room temperature, the heat capacity increases quickly with the increasing of temperature. Above room temperature, the heat capacity approaches to the Dulong-Petit limit. In paper, we assume that the lattice thermal conductivities of CuGaTe2 and CuInTe2 are mainly from the phonon scattering. And the phonon scattering is dominated by Umklapp scattering. The calculated lattice thermal conductivities can fit the form kl = A/T-Bin the temperature range of 300-800 K. For CuGaTe2, A = 2869.96 and B = 2.86. The fitting result well approaches to the experimental values and other theoretical results. Based on the calculated band structures with mBJ-GGA potential, the transport properties of CuGaTe2 and CuInTe2 each as a function of chemical potential at various temperatures are investigated. The values of Seebeck coefficient S first increase and then decrease for n-type and p-type doping at low carrier concentrations, which are consistent with the previous results. Electrical conductivity divided by scattering time, i.e. / increases monotonically with chemical potential increasing. The power factor divided by scattering time, i.e. S2/ first increases and then decreases with chemical potential increasing. The magnitude of S2/ increases with temperature increasing. Besides, it is found that the value of S2/ for p-type doping is larger than that for n-type doping. These results show that optimizing the carrier concentration can improve their thermoelectric performances. In order to calculate the electrical conductivity, in this paper we estimate the scattering time from the experiments of Ref.. The CuGaTe2 at 700 K possesses a figure of merit 0.63. These calculated results show that CuGaTe2 and CuInTe2 both are good thermoelectric materials with p-type doping.
CONDENSED MATTER: ELECTRONIC STRUCTURE, ELECTRICAL, MAGNETIC, AND OPTICAL PROPERTIES
2016, 65 (15): 157201. doi: 10.7498/aps.65.157201
Nowadays, the studies on absorption spectra and conductivities of Ti doped ZnO systems have presented distinctly different experimental results when the atom fraction of impurity increases in a range from 1.04 at% to 1.39 at% To solve this contradiction, all calculations in this paper are carried out by the CASTEP tool in the Materials Studio software based on the first-principals generalized gradient approximation (GGA) plane wave ultra-soft pseudopotential method of the density functional theory. The supercell geometric structures of ZnO, Zn0.9792Ti0.208O and Zn0.9722Ti0.278O systems are used as the calculation models. For all the geometry optimization models, the band structures, densities of states, electron density differences, population and absorption spectra are calculated by the method of GGA+U. The results show that with the Ti doping amount increasing from 1.04 at% to 1.39 at%, the lattice parameters and also the volume of the doping system increase. The higher the total energy of the doping system, the higher the formation energy of the doping system is, thereby making doping difficult and lower stability of the doping system. The increase of Ti-doping concentration weakens the covalent bond, but strengthens the ionic bond. As the Ti substitutional doping concentration increases, the Mulliken bond populations decrease, but bond lengths of Ti-O increase for the doping system Meanwhile, the higher the Ti doping content, with all the doping systems converted into n-type degenerate semiconductor the wider the band gap of the doping system will be and the more significant the blue shift of absorption spectra of Ti-doped ZnO systems. In this paper the mechanism of band gap widening is reasonably explained. In addition, the higher the Ti doping content, the higher the electronic effective mass of doping systems is The higher the electronic concentration of doping systems, the lower the electronic mobility of doping systems is. The lower the electronic conductivity of doping systems, the worse the doping systems conductivity is. The calculation results of absorption spectrum and conductivity of Ti-doped ZnO system are consistent with the experimental data. And the contradiction between absorption spectrum and conductivity of Ti-doped ZnO system in experiment is explained reasonably by temperature effect. In this paper, the comprehensive optical and electrical properties of Ti-doped ZnO systems are calculated by first-principals GGA+U method. And these results may improve the design and the preparation of photoelectric functional materials for Ti-doped ZnO at quite a low temperature.
Study on tunneling magnetoresistance effects in parabolic well magnetic tunneling junction with double barriers
2016, 65 (15): 157301. doi: 10.7498/aps.65.157301
In this paper, we construct a ferromagnet/semiconductor/ferromagnet parabolic well magnetic tunneling junction with double barriers as research object by inserting another semiconductor as a barrier between ferromagnetic and semiconductor potential wells. On the basis of the quantum coherent transport theory and transfer matrix method, we investigate the spin polarized electron transport and the tunnel magnetic resistance (TMR) in parabolic well magnetic tunneling junction with double barriers. We derive the analytical expressions of transmission probability, tunnel magnetic resistance and spin polarization from the new magnetic tunneling junction mode. The significant quantum size, Rashba spin orbit interaction, the angle effect and the thickness of the double barriers layer are discussed simultaneously. The results indicate that the tunnel magnetic resistance shows periodic variation as the width of the parabolic-well at different angles. The TMR is monotonically decreasing when the angle varying from 0 to up, which reflects the structure of the spin valve effect. Meanwhile, results also show that the spin polarization and the tunnel magnetic resistance oscillate with the same period for different barriers thickness. The phase difference appears after inserting the barriers. With increasing the barriers width, phase difference becomes large. The amplitude and peak to alley ratio of the spin polarization and the tunnel magnetic resistance are increase with the barrier width increases. Furthermore, the spin polarization make quasiperiodic oscillation that the oscillation amplitudes become large, the period and peak to alley ratio are decrease as the Rashba spin-orbit coupling strength increases. It appears the spin flip phenomenon as increasing the thickness of the barriers. The TMR shows the typical properties of resonant tunneling with the increasing of the spin orbit coupling strength. In order to better reveal the role of the symmetry double tunnel barriers in the parabolic well structure, we calculate TMR against the thickness of the double barriers. It is found that the existence of the double tunnel barriers increase the TMR and the spin polarization significantly, which shows that the large TMR value can be obtained with the suitable layer thickness of the double barriers layer and the Rashba spin-orbital coupling coefficients. These characteristics are helpful to promote the development and application of new magnetic tunnel junctions.
Density functional theory study of hydrogen spillover mechanism on Pd doped covalent organic frameworks COF-108
2016, 65 (15): 157302. doi: 10.7498/aps.65.157302
Hydrogen storage remains one of the main challenges in the implementation of a hydrogen-based energy economy. Among various porous materials, hydrogen storage in covalent-organic frameworks (COFs) has attracted the most significant attention since they were first synthesized due to good stability, large surface area, porosity and extremely low density. Although COFs exhibit promising hydrogen storage properties at very low temperatures, their hydrogen storage capacity is not satisfactory at room temperature, which is too low to meet the uptake target set by US-DOE, thereby being unable to have practical applications. Remarkably, hydrogen spillover has been experimentally demonstrated as an effective approach to improving the hydrogen storage capacity on porous materials at ambient temperature. In some of the most promising results the metal-organic frameworks (MOFs) and COFs have been used as substrates. However, the structures of many COFs materials are quite complex and the experimental condition is difficult to control. Furthermore, the sample preparations for these hydrogen spillover experiments are also very difficult. Therefore, only COF-1 is used in experimental study of hydrogen spillover. Although some theoretical work has contributed to understanding the hydrogen spillover mechanism of COFs, many basic problems about hydrogen spillover have not been solved, which hinders its practical application to a large extent. Based on the above reasons, the hydrogen spillover mechanism of Pd doped COF-108 is studied by using density functional theory (DFT) method, which mainly includes the various deposited configurations of Pd4 cluster on COF-108, the adsorption and dissociation of H2 on Pd4 cluster of Pd4@COF-108, the migration of H atom from Pd4 cluster toward the COF-108 and the diffusion of H atom on COF-108 surface. The results show as follows. 1) The larger the number of Pd atoms interacting with HHTP or TBPM cluster, the greater the binding energy of Pd4 deposited on them is. Deposited configuration orientation has little effect on binding energy. The binding energies of all deposition configurations for TBPM cluster are larger than those for HHTP cluster, so Pd4 cluster prefers to deposit on TBPM cluster with face-contact configuration. (2) H2 molecules spontaneously dissociated into Pd4 cluster, i.e., a barrierless H2 dissociation process takes place, which meets the first condition required by hydrogen spillover. 3) Only H atom located at the bridge site may migrate to the substrate surface, and the migration process is an endothermic reaction and less stable, which indicates that H atoms will further diffuse on the substrate surface. Although H atoms located at the top site may not migrate directly to the substrate surface, it will automatically migrate to the bridge site after the H atom on the bridge site has migrated to the substrate surface, so the migration process may proceed continuously. (4) The introduction of transition metal Pd can greatly reduce the diffusion energy barrier of H atoms on substrate surface, which makes it easier for H atoms to diffuse on substrate. These results may help us understand the microscopic mechanism of hydrogen spillover influencing the properties of hydrogen storage on COFs and provide useful guidance for targeted preparing the COFs materials with excellent hydrogen storage properties experimentally.
2016, 65 (15): 157303. doi: 10.7498/aps.65.157303
Bandgap engineering is one of the keys to practical applications of ZnO. Using ternary ZnMeO (Me=Be, Mg, Cd, etc.) alloys to regulate the bandgap of ZnO has been widely studied. Alloying ZnO with CdO to form CdxZn1-xO is an effective way to narrow down the bandgap of ZnO. With its narrower bandgap, CdxZn1-xO is a promising candidate for fabricating optoelectronic devices operable in the UV-visible wavelength region. In this work, we study the thermodynamic properties of CdxZn1-xO alloys of both wurtzite (WZ) and rock salt (RS) structures by first-principles calculations based on density functional theory (DFT) combined with the cluster expansion approach. The effective cluster interactions (ECIs) fitted formation energies agree well with the DFT-calculated formation energies for different compositions and structures correspondingly, validating the cluster expansion approach in calculations of the formation energy for CdxZn1-xO alloys. It is found that, for both WZ-CdxZn1-xO and RS-CdxZn1-xO alloys, the ECIs involve pair, triplet and quadruplet interactions: the pair interactions are dominant and contribute mostly to the formation energy. The first-and second-neighbor pair interaction parameters of WZ-CdxZn1-xO are positive, which indicates a tendency of ordering in WZ-CdxZn1-xO. For RS-CdxZn1-xO alloys, the nearest-neighbor pair interaction is negative, indicating a tendency to phase separation. The dominant positive second-neighbor pair interaction, however, appears to favor the ordering tendency. For both the WZ-CdxZn1-xO and RS-CdxZn1-xO alloys, the calculated formation energy of most structures is positive in the whole composition range, except for WZ-CdxZn1-xO with Cd concentrations of 1/3 and 2/3. Then, the crystal and electronic band structures of the metastable WZ-Cd1/3Zn2/3O and WZ-Cd2/3Zn1/3O are calculated. It turns out that both lattice constants a and c increase while the value of c/a and the bond angle of OZn(Cd)O decrease with increasing Cd concentration in the WZ-CdxZn1-xO alloys. Analyses of the band structures, densities of states (DOSs) and partial densities of states of WZ-CdxZn1-xO alloys reveal that the valence band maximum (VBM) is determined by O-2 p states and the conduction band minimum (CBM) stems from the hybrid Cd-5 s and Zn-4 s orbital. The VBM rises while the CBM declines, leading to the decrease of the bandgap of WZ-CdxZn1-xO with increasing Cd concentration. At finite temperatures, the thermal stability of the solid-state system is determined by Gibbs free energy. The bimodal curve, which indicates the equilibrium solubility limits as a function of temperature, can be calculated by the common tangent approach from the Gibbs free energy. The critical temperatures, above which complete miscibility is possible for some concentrations, are 1000 and 2250 K for WZ and RS phases, respectively. The higher critical temperature implies that it is more difficult to form RS-CdxZn1-xO than to form WZ-CdxZn1-xO. Finally, the phase diagrams of WZ-CdxZn1-xO and RS-CdxZn1-xO are derived based on calculations of the Gibbs free energy. At 1600 K, the solubility of Cd in WZ-ZnO amounts to 0.13, while the solubility of Zn in RS-CdO limits to only 0.01, indicating that it is much easier to incorporate Cd into WZ-ZnO than to incorporate Zn into RS-CdO.
2016, 65 (15): 157501. doi: 10.7498/aps.65.157501
Magnonic crystals with spin waves as information carriers are the magnetic counterparts of photonic and phononic crystals. The studies of spin waves or magnons in magnonic crystals have attracted increasing attention, especially for the characteristics of band gaps. However, most of the previous work has paid attention to the magnonic crystals with simple lattices. In this paper, the model of magnonic crystals with complex lattices which is composed of two different scatterers of ferromagnetic materials periodically embedded in another kind of ferromagnetic matrix material is proposed for the first time. Then, the plane-wave expansion method is developed by using the idea of super cells, in which the Fourior coefficient of exchange constant in the space of reciprocal lattice vector is analytically derived, and this method can be used to numerically investigate the eigen-properties of spin waves in magnonic crystals with complex lattices. Of course, it can be applied to the fields of other artificial crystals with complex lattices after the corresponding process, such as photonic crystals and phononic crystals. Band structures of two-dimensional magnonic crystal with complex lattices consisting of two different sizes of Fe cylinders alternately arranged in Euo matrix, are numerically calculated by using the above plane-wave expansion method. The behaviors of band gaps of spin waves changing with the total filling fraction of volume f and also with the mismatch of the filling fraction of volume of two Fe cylinders in EuO matrix are numerically studied. The results of magnonic crystals with complex lattices are compared with those of magnonic crystal with simple latticeic. Some conclusions are summarized as follows. In the same filling fraction of volume f, the width of band gap B4, 5 in the magnonic crystal with complex lattice is always larger than that with the simple lattice, but the width of band gap B8, 9 in the complex lattice is less than that in the simple lattice. When f = (fA + fB)/2 = 0.5, the width of band gap B4, 5 increases as the mismatch between fA and fB increases, but the behavior of the gap B8, 9 is opposite. Moreover, some new spin-wave gaps can be generated by changing the mismatch between fA and fB. This is because the gaps in our studied systems result from the mechanism of Bragg scattering of spin wave in periodic ferromagnetic materials. When the mismatch between fA and fB increases, the multiple scattering effects become stronger. All of these results show that the width or the frequency of band gap can be optimized or tuned by using the complex lattice. Such an approach through fabricating complex lattices may open a new scope for engineering and designing the band gaps of magnonic crystals.
2016, 65 (15): 157801. doi: 10.7498/aps.65.157801
High time resolution detecting systems for MeV pulsed radiation are essential for inertial confinement fusion diagnostics. Traditional detection of system time resolution is restricted by cable bandwidth. Based on recording excess carrier dynamics in semiconductors, a new detecting mechanism, called RadOptic, was developed by Lawrence Livermore National Laboratory (LLNL). The variation of intensity of pulsed radiation with time was converted into the variation of intensity of infrared laser probe by using this mechanism. The sensing material was InGaAsP quantum wells with severalmicrometer thickness. Picosecond time resolution for several keV pulsed radiation has been demonstrated. The reported system is not suitable for MeV pulses due to its low efficiency to MeV photons. Multiple cascaded structure for MeV photon to electron transformation was proposed by LLNL. Applying bulk material with several-hundredmicrometer thickness is an alternative. Based on transient free carrier absorption, a system recording bulk materials' instantaneous refractive index change is established. The system consists of a probe laser, an interferometer module, a signal transmission module and a signal recording module. The probe is a tunable infrared continuous wave laser whose wavelength is ~1453 nm, guided by single mode fiber to the interferometer. The interferometer consists of a single mode fiber head coupled directly with the polished face of a bulk semiconductor. The interference pattern forms by multiple beams reflected from the front face and the back face of the bulk. Part of interference light is coupled to the single mode fiber and forms the output signal. Pulsed radiation will deposit energy and generate excess carriers in the bulk material. The refractive index of the bulk material changes therewith according to the Drude model. The interference pattern and the light coupled to the single mode fiber also change therewith. The signal is transmitted by a long single mode fiber. The signal recording module consists of photoelectric detectors and a digital oscilloscope. The signal generation process and the time resolution of the system are analyzed. Intrinsic GaAs refractive index change is exploited under electron pulses and X ray pulses. The analysis of signal generation process shows that when the excess carriers recombine much faster/much slower than the pulse width, the output signal/output signal differential can be viewed as a measure of intensity variation with time of the incident pulse. For this prototype system, the time resolution is restricted by the digital oscilloscope to 1 GHz. Bulk intrinsic GaAs demonstrates 30 ns refractive index response time, which is longer than the incident pulse width. The differential signal can be viewed as a measure of incident pulse intensity when GaAs is exposed to 1 ns~0.2 MeV electrons pulses. The differential signal width is shorter than the pulse width when GaAs is exposed to 5 ns~0.2 MeV electrons pulses. Auger recombination process may occur in the pulse duration under this situation. The differential signal width is longer than the pulse width when GaAs is exposed to 1 ns~0.2 MeV X ray pulses. The poor signal to noise ratio affects the signal. The excess carrier generation process may be longer than theoretically estimated one under X ray pulse incident situation. The generation process and recombination process of excess carriers in GaAs show very different characteristics compared with optical excitation. The relationship between the system output signal and the incident pulsed radiation depends on the type of the incident radiation. With carefully considering the effects from incident pulse type and transient carriers density, the system can be used to detect ~MeV pulsed radiation. With an upgraded recording module, the system would demonstrate much higher time resolution.
Adsorption, film growth, and electronic structures of 2,7-dioctylbenzothieno-[3,2-b]benzothiophene (C8-BTBT) on Cu (100)
2016, 65 (15): 157901. doi: 10.7498/aps.65.157901
Using ultraviolet photoemission spectroscopy (UPS), X-ray photoemission spectroscopy (XPS), atomic force microscopy (AFM), and grazing X-ray diffraction measurement(GIXRD), we systematically investigate the correlations of interface energy level structure, film growth and the molecular orientation of 2, 7-dioctylbenzothieno-[3, 2-b]benzothiophene (C8-BTBT) on Cu(100). We find that the adsorption of the first layer of C8-BTBT molecules on Cu(100) is a stable physical one, and there is no chemical shift of the S 2p peaks of XPS and the ratio of the output of C to that of S is the same as the stoichiometric value of the molecular C8-BTBT. The heights of the steps of the upper layers of C8-BTBT in the AFM images are ～ 30 , close to the length of the molecular long c-axis, indicating the standing-up configuration of the upper molecules. AFM image shows that the upper molecules tend to grow into islands while the bottom molecules tend to grow into layer, suggesting an Stranski-Krastanov growth mode of multilayer C8-BTBT on Cu(100). The GIXRD shows an out-of-plane period of 30.21 , which consistently proves the standing-up configuration of the outer molecule layer. There is an electric dipole of 0.41 eV at the very interface pointing from the substrate copper to C8-BTBT, which will reduce the barrier for electron transport and increase the barrier for hole transport from Cu to C8-BTBT. The vacuum level (Evac) starts to bend downward after 16 deposition, and with the increase of the thickness of the film, a total downward shift of 0.42 eV is observed. The downward shift is ascribed to the changing of molecular orientation from lying down before 16 to standing up after 16 , which establishes an outward-pointing layer of C-H bonds and accordingly forms a dipole layer depressing the surface barrier. The shape and leading edge of the hightest occupied molecular orbit (HOMO) also change with the increase of film thickness. These changes are due to the anisotropy of electron ionization from molecular orbit. The total downward shift of the HOMO is about 0.63 eV. The downward bending of 0.42 eV for Evac and 0.63 eV for HOMO with increasing film thickness lead to a slightly decreasing ionization potential (IP) about 0.1 eV before 32 and then an increasing IP about 0.31 eV, which finally results in a total increase of 0.21 eV for IP. The bending electronic structures facilitate electron transport from interface to surface and hole transport from surface to interface. Our Investigation provides valuable information for relevant device design.
INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY
2016, 65 (15): 158101. doi: 10.7498/aps.65.158101
Rapid solidification is a typical non-equilibrium phase transition process, and the crystallization rate of liquid metal is larger than 1 cms-1. If the alloy is solidified in this case, the solute segregation is reduced or even eliminated and the solid solubility can be improved significantly. Rapid solidification technique can be used to refine the microstructures of alloys, which provides an effective method to prepare the novel metastable materials and improve their strengths, plasticities magnetic properties, etc. In this work, the rapid solidification mechanism and magnetic property of ternary equiatomic Fe33.3Cu33.3Sn33.3 alloy are investigated by drop tube and melt spinning techniques. It is known that Fe-Cu-Sn ternary alloy forms a typical immiscible system. However, the experimental results reveal that the liquid phase separation does not take place during the rapid solidification of ternary equiatomic Fe33.3Cu33.3Sn33.3 alloy. The solidification microstructures are all composed of primary Fe dendrites together with Cu3Sn and Cu6Sn5 phases. Under the free fall condition, as the drop tube technique provides microgravity and containerless states, the maximum surface cooling rate and maximum undercooling of alloy droplets are 1.3105 Ks-1and 283 K (0.19 TL), respectively. When the surface cooling rate reaches 1.9103 Ks-1, the primary Fe phase appears as coarse dendrites, and its maximum dendrite length is 41 m. Meanwhile, the Cu3Sn and Cu6Sn5 phases are distributed in the Fe interdendritic spacings. Once the surface cooling rate increases up to 3.3103 Ks-1, the morphology of the primary Fe phase transforms from coarse dendrites into broken dendrites. It is found that the cooling rate and undercooling greatly affect the solidification microstructure of alloy droplets. During the melt spinning experiments, since the large temperature gradient exists between the wheel surface and free surface, the solidification microstructure is subdivided into two crystal zones according to the different microstructure morphologies of Fe phase: fine grain (zone I) and coarse grain (zone II), where zone I is characterized by granular grains while zone II has some dendrites with secondary branch. Under the rapid cooling condition, the microstructures of ternary equiatomic Fe33.3Cu33.3Sn33.3 alloy ribbons are refined significantly and show soft magnetic characteristics. As the surface cooling rate increases from 8.9106 to 2.7107 Ks-1, the lattice constant of Fe solid solution rises rapidly and the coercivity increases from 93.7 to 255.6 Oe. Furthermore, the results indicate that the grain size of Fe phase is the main factor influencing the coercivity of alloy ribbons.
Modeling and simulation of the insulated gate bipolar transistor turn-off voltage slope under inductive load
2016, 65 (15): 158501. doi: 10.7498/aps.65.158501
The insulated gate bipolar transistor (IGBT) has developed rapidly as a key power device for medium power application since it was first introduced. It is well known for its relatively low conduction loss and easy gate control. The IGBT is commonly seen in the inductive load application circuit. Due to the large inductive load, the current of the IGBT will stay high until the voltage rises to the bus voltage during the IGBT turn-off. After that, the current starts to decrease and IGBT goes into the tail-current procedure withstanding high voltage. When evaluating the turn-off loss of IGBT, the fall time and the tail current are commonly taken into consideration because these two features are known as good representations of power loss during tail-current procedure. However, the power loss occurring during the voltage rise, which is usually neglected, can also be a significant contributor to the total turn-off loss. The dv/dt determines the voltage rise time and the power loss during this procedure. Thus, predicting the dv/dt is essential for evaluating the power loss during the IGBT turn-off. In this paper, the turn-off transient is divided into four stages and the physical mechanism which determines the dv/dt during the turn-off transient is carefully investigated. An analytical model to characterize the dv/dt during IGBT inductive turn-off is derived based on the calculated miller capacitance values. The functions of the miller capacitance and the dv/dt against time are presented to predict the collector voltage waveform during the IGBT turn-off. To make the model more accurate, the current dependence is considered when calculating the miller capacitance as well as the voltage dependency. The derived model shows that the dv/dt increases nonlinearly with the time going by and can be influenced by several factors, including the drive circuit conditions, the collector current and the carrier concentration profile in the ON-state. Further investigation indicates that the ON-state carrier concentration is greatly influenced by the IGBT cell structure. Thus, the model presented in this paper is effective in both the estimation of IGBT turn-off loss and the guidance of device structure design. The prediction of the derived model shows good agreement with the two-dimensional numerical simulation by Sentaurus TCAD (with the relative error not exceeding 10%) for the IGBT turn-off over a broad range of the collector current values. The device structure simulated in this paper is based on the 650 V/60 A trench-FS-IGBT. The thickness values of the total structure and the buffer layer are 80 m and 20 m, respectively.
Streak cameras applied to inertial confinement fusion research and flashless imaging lidar require large working areas. However, the larger the working area, the bigger the temporal distortion is. And the temporal distortion has a great influence on the detecting precision of the streak camera, resulting in an image distortion on the screen. Yet previous streak camera design work emphasized shorter time resolution and higher special resolution with paying less attention to the temporal distortion extent. Key factors that may affect the temporal distortion are thoroughly analyzed in this paper. We calculate the electric field of a small-size streak tube with the aid of the Computer Simulation Technology Particle Studio software which is a three-dimensional electromagnetic simulation software based on finite integration technology. Axial electric field distributions at different distances to the axis of the small-size streak tube are displayed. The electron trajectories launched from different points on photocathode of the streak tube are tracked through interpolating pre-calculated electromagnetic field to the particle position. It is known that curved photocathode can reduce the temporal distortion, so we calculate the temporal distortions of streak tubes whose radii of curvature of the photocathode are 30 mm, 35 mm, 40 mm, 45 mm, 50 mm, and 55 mm respectively to ascertain how the curvature influences the temporal distortion. The results show that the temporal distortion is mainly produced in the photocathode-to-deflector region, and it is negligible in the equipotential region. Also, bigger radius of curvature of the photocathode leads to a positive temporal distortion, and smaller one leads to a negative temporal distortion. And the absolute value of the temporal distortion increases with the increase of the slit length. The small-size streak tube whose radius of curvature of the photocathode is 40 mm owns the smallest temporal distortion. We also calculate the temporal distortions of electrons launched from the different positions of the photocathode with different initial energies, and the initial energy has little influence on the temporal distortion. To sum up, the dominating factor that produces the temporal distortion is the curvature of the photocathode. The slit image under a ramp sweeping voltage on screen is curved due to the temporal distortion. And the bigger the temporal distortion, the greater the curvature of the slit image is. Besides, a linear relation between the temporal distortion and deflection of the slit image is displayed. The spatial resolutions of the streak tubes with the radii of curvature of the photocathode 30 mm, 40 mm, 50 mm are calculated respectively. And the small-size streak tube whose radius of curvature of the photocathode is 30 mm has the highest spatial resolution. The radius of curvature of the streak tube photocathode should be carefully selected according to actual requirements for the streak camera. Through the analysis we provide a significant guidance for streak tube design.
An improved algorithm for prediction of protein loop structure based on position specificity of amino acids
2016, 65 (15): 158701. doi: 10.7498/aps.65.158701
Loop region is necessary structural element of protein molecule, and plays significant roles in protein functioning, e.g., in signaling, ligand recognition. Unlike the well-defined secondary structures (i.e., helix, sheet), however, loop regions vary in structure and some of them are even not able to be measured by ordinary experimental methods. For these reasons, computer-aided prediction of loop structure became a hotspot in bioinformatics and biophysics. Sorts of algorithms have been developed for this purpose. So far, however, the prediction of long loop is still a challenge. Among all the common algorithms, LEAP algorithm achieves the highest precision on long loop prediction. Our investigation on a test data set with LEAP algorithm reveals that the ultimate loop structure predicted by LEAP is almost entirely determined by the initial sampling of the conformation of the loop backbone. If all the backbone conformations in the initial sampling are quite distant from the real (native) conformation, the ultimately predicted structure is also distant from the native conformation, and the prediction accuracy cannot be improved obviously only by increasing the computation time. In the original LEAP, the initial sampling is based on the rough distribution of the backbone torsion angle (Ramachandran plot, R-plot) which doesn't consider the sequence information of the loop region. Many conformations which are far from the native conformation are most likely generated in the sampling. So there raises the open question, is it possible to enhance the initial sampling to be more targeted to the native conformation? In this paper, we suggest an approach to introduce the position-specific amino-acid sequence information into the initial sampling of the backbone conformation, which may generate more targeted initial decoys. An algorithm of protein secondary structure prediction, SPINE X, is used to generate rough but reasonable estimates of torsion angles of each amino acid of the loop backbone in sequence-dependent way. We then combine these values with the original R-plot to reconstruct a new R-plot for each amino acid in the loop, and the initial sampling is performed according to the new R-plot. We applied this new algorithm to a test set of loops (generated from single-chain proteins in CASP 10), and found the medians/means of RMSDs can reduce about 0.12 /0.13 , 0.25 /0.27 , 0.47 /0.27 for loop sets of length 10, 11, 12, respectively. Comparing to the original LEAP algorithm, the probability of making more accurate predictions is almost doubled when using the refined algorithm. The logic of our approach is not limited to LEAP, and can be extended to other algorithms which are also significantly dependent on initial sampling.
2016, 65 (15): 158901. doi: 10.7498/aps.65.158901
Online social networks, such as Facebook, Twitter and YouTube, play a vital role in information sharing and diffusion, and recently many dynamics models on social networks have been proposed to model information diffusion. However most models are theoretical, their parameters do not come from realistic data and their validity and reliability have not been evaluated empirically. In the paper we first analyze the users' behaviors of reading and reposting microblog in Sina Weibo, a Twitter-like website in China, and find that users' number of fans, the average reposted number of users' microblog, the intensity of users' interaction and the similarity between microblog topics and users' topic interests can significantly influence reposting behavior. Then we propose an information diffusion model Susceptible-Infected-Recovered based on Users' Behaviors (SIRUB) on microblog networks, compute the users' probability of reading microblog in the model according to the probability of their logging on microblog in a day, and obtain the reposting probability utilizing the logistic regression which considers 16 possible factors influencing users' reposting behavior. The 16 factors can be divided into three categories: the characteristics of microblog publishers, microblog text features and social relationship characteristics. We utilize the beginning 2/3 microblog data to obtain model parameters and logistic regression coefficients, and the remaining 1/3 data to examine the validity of the model. The experiments on Sina Weibo network show that the model can predict users' reposting behavior accurately only when it considers both reading and reposting probabilities. F-score which considers precision and recall is used to assess prediction effect of the model. The highest F-score for the prediction of SIRUB model on users' reposting behavior is 0.228 which is much larger than those of classical Susceptible-Infected-Recovered (SIR, F-score=0.039) and Susceptible-Infected-Contacted-Recovered (SICR, F-score=0.037) models. The prediction on the spreading scope of microblog for SIR and SICR models is related with users' number of fans while for SIRUB model not. For SIRUB model the mean and standard deviation of the errors of prediction on spreading scope are smaller than those of SIR and SICR models. These results indicate that users' behaviors of reading and reposting microblog should be appropriately taken in account when modeling information diffusion on microblog networks, and that, in general, the prediction performance of the data-driven SIRUB model proposed in the paper is better than those of SIR and SICR models regardless of the prediction of users' reposting behavior or diffusion scope of microblog.