高背压超声气体团簇喷流中团簇平均尺寸沿喷流方向演化研究

本文通过对高背压(50 bar, 1 bar = 1.0×10⁵ Pa)氩气经长锥型喷嘴(长度L=30 mm)向真空绝热膨胀所形成的超声气体团簇喷流的数值模拟, 分析比较了由喷嘴喉口起沿喷流方向在喷流中心轴线上团簇平均尺寸的演化情况. 结果表明: 沿喷流方向团簇平均尺寸显示先增长后趋于饱和的变化趋势, 具有较大尺寸团簇的区域出现在距离喷嘴喉口大约20 mm. 据此本文再结合关于喷流中原子密度沿喷流方向变化的模拟结果开展了锥形喷嘴长度的优化研究. 针对由常见构型的锥形喷嘴(喉径～ 0.5 mm, 半张角～ 8.5°)在高背压下形成的团簇喷流, 20 mm左右的长度为锥形喷嘴的适宜长度.     

An experimental investigation on mixing enhancements in non-circular sharp-edged nozzles using the entropy production concept

An experimental study on mixing enhancement in free jets, issuing from sharp-edged nozzles of different geometry, is performed by using particle image velocimetry. The attention is focused on the jet near-field and interaction zones (0 < X/D < 18, where X is the axial coordinate and D the diameter of the equivalent circular jet). The mixing efficiency is evaluated and quantified using the definition of entropy production derived from the velocity field. The effect of Reynolds number is also discussed by performing measurements at Re = 8000 and Re = 35,000. The results are compared to circular nozzle data to evaluate the change in mixing efficiency among axisymmetric and non-symmetric nozzles. While the effect of Reynolds number on mixing is small, at least for the values tested here, the change in geometry is rather crucial. In terms of entropy production, the rectangular and elliptical nozzles show higher mixing for X/D< 7, whereas the other ones attain the best results for larger distances. This behaviour is basically related to the axis-switching phenomenon observed in elongated jets. Different variables are tested and compared as possible velocity and length scales to derive a meaningful non-dimensional entropy production.     

Thermal stability and diffusion processes in Mo x Si y /Si multilayers studied with high-resolution RBS

Mo _x Si _y /Si multilayers with a period thickness of ∼7.5 nm and bilayers Mo _x Si _y /Si have been fabricated by e⁻-beam evaporation in UHV at a deposition temperature of 150°C [1]. The composition of the as-deposited layer systems and changes in the composition after baking the samples have been studied with high-resolution RBS. For a multilayer with a mixing ratioy/x≃2, no interdiffusion is observed up to a baking temperature of 830°C. For samples with a mixing ratioy/x≃1, diffusion is observed up to a baking temperature of 630°C, resulting in a mixing ratio close toy/x≃2. This mixing ratio remains almost stable up to ∼830°C, and considerable interdiffusion is only observed in those systems where regions with a mixing ratio smaller than 2 still exist. Possible reasons for the high thermal stability of the samples are the lack of a concentration gradient for Si in the system and/or the crystallization of MoSi₂.     

Multi-line fluorescence imaging of the rotational temperature field in a shock-tunnel free jet

A quasi-steady, highly underexpanded free jet of argon seeded with nitric oxide (NO) was generated at the exit of a converging, axisymmetric nozzle supplied by a shock-tunnel reservoir at 4200 K and 3.0 atm. During each run of the facility, an isolated transition in theA ² X ² (0, 0) band of NO at 226 nm was pumped with a pulse of frequency-doubled dye laser light formed into a thin sheet and directed perpendicularly through the axis of the jet. The red-shifted components of the resulting fluorescence at 90° with respect to the laser were imaged onto an intensified, charge-coupled device array. A ratio of images obtained by exciting lines originating from two different rotational states could be used to infer the mean rotational temperature field. However, because of the extreme variations in temperature and density present in the free jet, no single pair of lines simultaneously provided adequate signal levels and temperature sensitivity over the flow's entire temperature range (i.e., 100–3100 K). Instead, a combination of images obtained with four different transitions was used. Excellent agreement was observed between multi-line temperature evaluations from single-shot and frame-averaged images and a numerical simulation of the flow performed by the method of characteristics.     

Anderson localization in different one-dimensional systems with off-diagonal disorder and spin-dependence

The localization properties of certain spin-dependent, one-dimensional electronic systems with only off-diagonal disorder are studied. In higher dimensions (d=2,3) the models considered would correspond to different universality classes, whereas ford=1 no qualitative difference is found: ForE=0, all eigenstates are exponentially localized, whereas forE0 the localization length diverges logarithmically, such that exactly atE=0 the geometric average of the transmission coefficient would decay with increasing chain lengthL as exp (-const. ·L ^1/2), instead of the usual, exponential decay.ForE=0, in the interior of the band, the localization lengthr ₀ diverges W ₂ ^–2 in the limit of weak disorder (W ₂0), whereas just at the band edge one has roughlyr ₀W ₂ ^–2/3. A universal recursion relation, depending only on the energy and on certain randomly distributed determinants, determines the localization length and the density of states for all systems considered.     

Hydrodynamics of pulsed supersonic underexpanded jets: Spatiotemporal characteristics

The gasdynamic parameters of nonsteady expansion of He, Ar, N₂, and SiH₄ from a sonic nozzle into a space with reduced background gas pressure were experimentally studied for moderate values of n (10³–10⁶) and the Reynolds number (Re_L∼10⁰–10²). The jet set times necessary for the formation of pulsed jets of a given finite duration are determined. The results are generalized in terms of dimensionless similarity parameters. The laws of motion of the leading and trailing fronts in pulsed jets of various gases are established. The leading front of a pulsed jet propagates at a velocity significantly smaller than the limiting steady value. The jet expansion dynamics is determined by the ratio of the momentum of the expanding gas to that of the background gas displaced from the flow region. The length of the steady flow region in a pulsed jet monotonically decreases downstream from the source and drops with increasing background gas pressure because of the loss of jet particles in the trailing rarefaction wave; this length increases with the initial momentum because the background gas is more intensively displaced from the flow region.     

Existence and renormalization of the Sine-Gordon field theory in all dimensions

The Euclidean field theory of the Sine-Gordon model (SG) is investigated by using methods of statistical physics. The SG model is shown to be the continuum limit of the ferromagnetic Harmonic Rotator model (HR) which itself is equivalent to a classical Coulomb plasma of unit charges on a lattice. Using our recent results for the latter models in all Euclidean dimensionsD we determine the existence of the SG field theory in terms of the HR parameters, temperaturet and magnetic fieldB. The following results are derived: 1. InD=1 the SG exists neart=B=0; the quantum theory in zero space dimensions is just the Mathieu equation as Schrödinger equation. 2. ForD>2 the SG field theory exists for allt nearB=0. The theory is actually constructed and is equivalent to a free massive scalar theory. 3. In the most interesting case ofD=2 the SG field theory exists for allt<t=8 nearB=0, it does not exist fortt. All necessary renormalizations are performed and all necessary subtractions are obtained in closed form which proves that the theory is superrenormalizable when it exists. We also discuss the relations between the structure of the particle spectrum of the SG, the phase transitions of the HR, and the binding properties of the classical Coulomb charges.     

Renormalization group approach for the site-bond percolation in structured stochastic environments

The quenched averaged percolation problem of a lattice with a given structure is analyzed. The structure is described by the static structure factorS(q)q ^–ain the regionq 0. As a result of the renormalization group, it follows that the critical behavior fora < 2 is the same as in the random percolation. In the case ofa=2 second universality class with=0 and=1/2+/8+ ²/32 is predicted.     

Atomic frequency offset locking in a Λ type three-level Doppler broadened Cs system

We here present a comparative study of frequency stabilities of pump and probe lasers coupled at a frequency offset generated by coherent photon-atom interaction. Pump-probe spectroscopy of the Λ configuration in D₂ transition of cesium is carried out to obtain sub-natural (∼2 MHz) electromagnetically induced transparency (EIT) and sub-Doppler (∼10 MHz) Autler-Townes (AT) resonance. The pump laser is locked on the saturated absorption spectrum (SAS, ∼13 MHz) and the probe laser is successively stabilized on EIT and AT signals. Frequency stabilities of pump and probe lasers are calculated in terms of Allan variance σ(2,τ) by using the frequency noise power spectrum. It is found that the frequency stability of the probe stabilized on EIT is superior (σ∼2×10⁻¹³) to that of SAS locked pump laser (σ∼10⁻¹²), whereas the performance of the AT stabilized laser is inferior (σ∼6×10⁻¹²). This contrasting behavior is discussed in terms of the theme of conventional master-slave offset locking scheme and the mechanisms underlying the EIT and sub-Doppler AT resonances in a Doppler broadened atomic medium.     

Scaling behaviour in the truncated Zel'dovich approximation

Summary We analyse the scaling properties of the matter distribution generated by the truncated Zel'dovich approximation (TZA) for a universe dominated by a mixture of cold+hot non-baryonic dark matter. At the present epoch, the dark-matter distribution generated by TZA displays an intermediate scaling behaviour between about 4 and 60h ⁻¹ Mpc with a correlation dimensionD _q∼2 forq≥2. The possible implications of these results are considered.     

Computation of streamwise vorticity in a compressible flow of a winglet nozzle-based COIL device

Chemical oxygen iodine laser (COIL) is a high-power laser with potential applications in both military as well as in the industry. COIL is the only chemical laser based on electronic transition with a wavelength of 1.315 μm, which falls in the near-infrared (IR) range. Thus, COIL beam can also be transported via optical fibers for remote applications such as dismantling of nuclear reactors. The efficiency of a supersonic COIL is essentially a function of mixing specially in systems employing cross-stream injection of the secondary lasing (I₂) flow in supersonic regime into the primary pumping (O₂¹Δ_g) flow. Streamwise vorticity has been proven to be among the most effective manner of enhancing mixing and has been utilized in jet engines for thrust augmentation, noise reduction, supersonic combustion, etc. Therefore, a computational study of the generation of streamwise vorticity in the supersonic flow field of a COIL device employing a winglet nozzle with various delta wing angles of 5°, 10°, and 22.5° has been carried out. The study predicts a typical Mach number of approximately 1.75 for all the winglet geometries. The analysis also confirms that the winglet geometry doubles up both as a nozzle and as a vortex generator. The region of maximum turbulence and fully developed streamwise vortices is observed to occur close to the exit, at x/λ of 0.5, of the winglets making it the most suitable region for secondary flow injection for achieving efficient mixing. The predicted length scale of the scalloped mixer formed by the winglet nozzle is 4λ. Also, the winglet nozzle with 10° lobe angle is most suitable from the point of view of mixing developing cross-stream velocity of 120 m/s with acceptable pressure drop of 0.7 Torr. The winglet geometry with 5° lobe angle is associated with a low cross-stream velocity of 60 m/s, whereas the one with 22.5° lobe angle is associated with a large static and total pressure drop of 1.87 and 9.37 Torr, respectively, making both the geometries unsuitable for COIL systems. The experimental validation shows a close agreement with the computationally predicted values. The studies for the most suitable 10° lobe angle geometry show an observed Mach number of 1.72 with an improved mixing efficiency of 74% due to the occurrence of predicted streamwise vortices in the flow.     

Error performance analysis for an optical heterodyne FSK communication system with a limiter-discriminator-integrator detector

The error probability for a coherent optical heterodyne FSK system with a limiter-discriminator-integrator (LDI) detector is analysed. The analysis includes laser quantum phase noise, the correlated receiver additive white Gaussian noise (AWGN), Gaussian narrow-band IF filtering and intersymbol interference (ISI) caused by it. It is shown that, for 1 dBEb/N ₀penalty at bit error rate (BER) 10^–9, (i) the normalized IF beat spectral linewidth T0.35% for frequency deviation ratioh=0.5(MSK), and T0.5% forh=0.7 (the receiver is insensitive to laser quantum phase noise ath=1.0, if ISI is not included.); (ii) if ISI is incorporated, T0.15% forh=0.5, T0.5% forh=0.7, both with 3dB bandwidth-bit period product (3.0>BT1.5), and T0.5% forh=1.0 with BT1.0 ifT0.35% when ISI exists,h=0.7 is optimum;h=1.0 otherwise.     

Three-dimensional consideration of jet impingement onto the kerf in relation to laser cutting process: Effect of jet velocity on heat transfer rates

In laser cutting process, an assisting gas is used to improve the mass removal rate from the cutting kerf and protect the kerf surfaces from the high temperature exothermic reactions, such as oxidation reactions, during the cutting process. Therefore, heat transfer rates from the kerf wall and the skin friction along the kerf surface are important for quality cutting. In the present study, jet emerging from a conical convergent nozzle and impinging onto the kerf surface is investigated in relation to the laser cutting process. The flow field in the kerf, the heat transfer rates from the kerf wall, and the skin friction along the kerf surface are computed for four average jet velocities at the nozzle exit and two kerf wall wedge angles. The ratio of the stand-off-distance (distance between the nozzle exit and the kerf top surface) to nozzle diameter is selected as H/D=2.2., where H is the stand-off-distance and D is the nozzle exit diameter. The kerf wall temperature is kept at 1500 K to resemble the laser cutting process. It is found that the Nusselt number increases sharply at the kerf inlet and decreases towards the kerf exit for the kerf wall angle of 0°. However, it increases gradually in this region for the kerf wedge angle of 4°. The skin friction decreases along the kerf surface.     

Radiometric observations of the 752.033-GHz rotational absorption line of H2O from a laboratory jet

The intensity and lineshape of the Doppler-broadened 752.033-GHz (2₁₁ 2₀₂) rotational transition of H₂O has been studied passively using a high-resolution two-stage heterodyne radiometer with single-sideband system noise temperature of 45,000 K. The purpose of the experiments was to demonstrate the observability at submillimeter wavelengths of a high-altitude rocket plume simulated by a laboratory H₂O jet in a vacuum chamber. First-stage mixing was accomplished by means of a GaAs Schottky diode with first local-oscillator power supplied by a CO₂-laser pumped formic-acid laser (761.61 GHz), generating and X-band IF signal. Second localoscillator power was provided by a tunable C-band source. One-MHz resolution capability was obtained by means of a 3-GHz waveguide cavity filter with only 9-dB insertion loss. In the H₂O jet experiments, the center frequency of the line was determined to within 1 MHz of the previously reported value. A rotational temperature 75 K, a linewidth 5 MHz, and a Doppler shift 3 MHz (from a 45-degree rotation of the flow direction) were measured with the line-of-sight intersecting the jet axis at a distance downstream of 30 nozzle diameters. These absorption data were ogtained against continuum background radiation sources at temperatures of 1175 and 300 K.     

薄膜自激拍打射流的流动特性实验

在射流喷嘴出口处安装一端固定的柔性薄膜,在流速足够大时,射流和柔性膜相互诱导产生自激拍打作用,针对这一现象,提出了一种新型的自激拍打射流混合技术。在直径D=40 mm的渐缩喷嘴上固定长度L=(0.5～2)D和厚度δ=50μm的氟化乙烯丙烯共聚物(fluorinated ethylene propylene, FEP)薄膜,使用压差计测量了光滑渐缩喷嘴以及薄膜拍打运动所引起的压力损失。通过激光片光源和高速相机进行薄膜运动状态的显示和拍打幅度的测定,探究了拍打幅度受薄膜长度和Re的影响及其变化规律,利用测得的拍打幅度(A)和频率(f)作为Strouhal数(St=fA/U_o,U_o是射流的出口速度)的特征尺寸进行研究。在Re=3×10⁴的条件下,使用热线风速仪测量了在不同薄膜长度下拍打射流轴向速度沿中心线的分布,并对湍流度、概率密度函数等特征进行分析。此外,经过数字迭代滤波后获得射流沿中心线的积分尺度、 Taylor尺度和Kolmogorov尺度等统计量。实验结果显示,拍打射流的湍流度高于自由射流,意味着前者对周围流体具有更强...     

Characterization of an ethylene free jet

The rotational population distribution in a free molecular jet of ethylene, C₂H₄, is found to be thermal over a range of distances below the nozzle (z/d=0.5–13.2, nozzle diameter d=0.50 mm). Results for on-axis rotational temperature, number density and flow velocity are presented. The average number of gas-kinetic collisions experienced by any molecule in travelling some distance along the jet axis is calculated.     

An Experimental Study of In-Situ Phase Fraction in Jet Pump Using Electrical Resistance Tomography Technique

We perform the experiments to investigate in-situ phase fraction in a jet pump using the electrical resistance tomography （ERT） technique. A new jet pump with ERT sensors is designed to measure in-situ phase fraction and flow regime. The study is based on laboratory experiments that are carried out on a 50-mm vertical flow rig for various gas and liquid phase superficial velocities. The different flow patterns of gas liquid in the jet pump and vertical pipe are studied using the ERT technique. The results suggest that the ERT system can be used to successfully produce images of gas-liquid flow patterns with frames rates of 58 fps and the in-situ phase fraction with frame rates of 5 fps can be obtained. The visualizations of a rapid mixing process in the throat of a jet pump obtained in this work provide a reliable basis for theoretical study and optimal design of jet pumps.     

Lineshape analysis of stimulated Raman spectra of the near‐nozzle region of free jet expansions

An anomalous lineshape of stimulated Raman spectra obtained from the region very close to the nozzle of supersonic pulsed expansions of nitrogen is presented. High‐resolution Raman spectra of the Q branch of the fundamental vibration mode of N₂ have been recorded from two different nitrogen expansions at T₀ = 295 K and P₀ = 1.5–3.5 bar, the lasers crossing the jet axis in the range z/D = 0.25–1.25, where D is the effective nozzle diameter. The combination of Doppler shifts and strong gradients of density and temperature in the near‐nozzle region yield an inhomogeneous broadening and a double peak structure of the recorded Raman line profiles. The comparison of the experimental results with the simulation of the Raman spectrum from this region provides valuable information about the near‐nozzle flow field. The lineshape described here is different from another reported previously in the literature, which is based on a depletion of the density of free molecules on the axis due to condensation. Copyright © 2009 John Wiley & Sons, Ltd.     

Self-affine fractal clusters: Conceptual questions and numerical results for directed percolation

In this paper we address the question of the existence of a well defined, non-trivial fractal dimensionD of self-affine clusters. In spite of the obvious relevance of such clusters to a wide range of phenomena, this problem is still open since thedifferent published predictions forD have not been tested yet. An interesting aspect of the problem is that a nontrivial global dimension for clusters is in contrast with the trivial global dimension of self-affine functions. As a much studied example of self-affine structures, we investigate the infinite directed percolation cluster at the threshold. We measuredD ind=2 dimensions by the box counting method. Using a correction to scaling analysis, we obtainedD=1.765(10). This result does not agree with any of the proposed relations, but it favorsD=1+(1- )/ , where and are the correlation length exponents and is a Fisher exponent in the cluster scaling.     

Scaling of Loop-Erased Walks in 2 to 4 Dimensions

We simulate loop-erased random walks on simple (hyper-)cubic lattices of dimensions 2, 3 and 4. These simulations were mainly motivated to test recent two loop renormalization group predictions for logarithmic corrections in d=4, simulations in lower dimensions were done for completeness and in order to test the algorithm. In d=2, we verify with high precision the prediction D=5/4, where the number of steps n after erasure scales with the number N of steps before erasure as n∼N ^D/2. In d=3 we again find a power law, but with an exponent different from the one found in the most precise previous simulations: D=1.6236±0.0004. Finally, we see clear deviations from the naive scaling n∼N in d=4. While they agree only qualitatively with the leading logarithmic corrections predicted by several authors, their agreement with the two-loop prediction is nearly perfect.