31 July 2019 in China Foundry
In the present work, 7075 aluminum alloy (Al-Zn-Mg-Cu) was produced by both conventional casting (CC) and controlled diffusion solidification (CDS) methods. Each sample was subjected to different heat-treatment conditions: as-cast, T4, and T6; and their microstructural and mechanical properties were investigated by optical microscopy (OM), scanning electron microscopy (SEM) and X-ray diffractometry (XRD). It was found that CDS promoted the formation of non-dendritic primary α-Al phase and reduced shrinkage porosity, thus resulting in improved mechanical properties. In addition, the eutectic phase of the CDS samples mainly consisted of T(Al-Zn-Mg-Cu) phase, which manifested a well-developed lamellar eutectic structure. However, in the CC samples, the T(Al-Zn-Mg-Cu) phase was composed of rod-like eutectics. Moreover, the θ(Al2Cu) eutectic contents in the CC samples were greater than those in the CDS samples. Each element in the CC samples had an obvious change in the grain boundary, whereas the change in element content in the CDS samples was gradual. Therefore, the non-dendritic morphology of the primary phase and the presence of rod-like eutectics in the matrix of the CDS samples led to enhanced tensile strength and elongation under different heat treatment conditions.
1 January 2017 in Acta Physica Sinica
With the detailed consideration of electrochemical reactions and collision relations, a direct numerical simulation model of helicon plasma discharge with three-dimensional fluid-dynamic equations is proposed in the present work. It can improve the precision of results and widen the model applicability by discarding the small perturbation theory in previous helicon models which are partially analytical in essence. Under the assumption of weak ionization, the Maxwell equations coupled with the plasma parameters are directly solved in the whole computational domain. Thus the energy deposited from electromagnetic wave to plasma can be then easily calculated. The values of plasma parameters which include electron density, mean electron energy and heavy species density are obtained by solving a set of drift-diffusion equations. Meanwhile, seven kinds of chemical reactions in the plasma and two kinds of surface reactions on the wall are taken into account. All of the partial differential equations are solved by the finite element solver of COMSOL MultiphysicsTM with the full coupled method.#br#The results of numerical cases employing argon as the working medium show that there exists a sharp density jump from a low to high value as the radiofrequency power is raised. The density jump phenomenon is in accordance with the experimental results of Toki (Toki K, Shinohara S, Tanikawa T, Shamrai K P 2006 Thin Solid Films 506-507 597) and Chen (Chen F F 2007 Plasma Sources Sci. Technol. 16 593). The electron temperature decreases with an increase of the gas pressure, which is similar to Toki's (Toki K, Shinohara S, Tanikawa T, Shamrai K P 2006 Thin Solid Films 506-507 597) measurement by a RF compensation probe. In comparison with the classical sheath theory, the simulation result demonstrates that the distribution of parameters such as particle number density, the Deby length, electric potential and electron temperature can be solved exactly. In addition, the phenomenon of low-field density peak in helicon discharge was studied in the work. Previous research by Chen (Chen F F 2003 Phys. Plasmas 10 2586) suggests that this peak is caused by constructive interference from the reflected wave. The effect of length of the discharge chamber on the relation of electron density and background magnetic field is investigated numerically. The results validate the mechanism of wave interference reflected by endplates of the discharge chamber. Furthermore, the time-averaged magnetic energy density has more than one peak on the axial direction. Comparing the distribution of the magnetic energy density to that of the dimensionless amplitude of the helicon wave and the TG wave in the one-dimensional undamped condition, it found that the length of peak to peak of the helicon wave is just as twice as that of the magnetic energy density, which indicates that the substance of wave interference is involved in the standing wave generated by the helicon wave and its reflected wave from endplates.
1 March 2024 in Physics of Fluids
Obtaining a rapid deflagration-to-detonation transition (DDT) within a short smooth tube is a challenging task. Here, an unconventional means of flame acceleration propagating upstream in subsonic and supersonic mixtures within a smooth tube was introduced to acquire a speedy DDT. The Navier–Stokes equations with an adaptive mesh refinement technique and a detailed hydrogen–air chemistry reaction mechanism of 11 species and 27 steps were utilized to resolve the entire DDT characteristics. The effect of the initial Mach number on flame acceleration and DDT mechanism was revealed comprehensively. The results demonstrated that a prompt oblique shock wave (SW) occurs when the flame propagates upstream along the boundary walls due to the boundary layer influence. An intense coupling between the SW and the leading flame front is enhanced by increasing the initial Mach number of the mixture. The speedy generation of the oblique SW is formed at the incipient stage, mainly produced by the boundary layer influence and the coalescences of the compression waves. Consequently, the run-up time to detonation is shortened accordingly through a fierce reflected SW due to the intense leading SW after it reflects from the confined wall. Furthermore, three kinds of DDT evolution are revealed from the obtained results: (1) localized ignition in the upper boundary wall after the reflected and transverse shock waves propagate in the upper wall regions; (2) autoignition is formed in the confined wall corner after the reflected SW; and (3) direct detonation transition occurs at the end wall behind a strongly reflected SW in the supersonic case.
6 September 2019 in IEEE Transactions on Plasma Science
The laser-induced fluorescence (LIF) diagnostic with highly saturated stimulation scheme was conducted to investigate the propulsion performance of a Hall thruster. A Gaussian inverse filter was employed to isolate the Doppler shift and broadening from a measured LIF spectrum by deconvolution. Besides the average velocity and temperature of the singly ionized Xe, parameters about the overall propulsion performance were also evaluated in this work, which indicates a relatively ideal working state with average ion efficiency for acceleration above 60%, comprehensive average velocity of 18 950 m/s, plume divergence angle of 37.26° and ionization degree of 83%. In addition, error analysis in the LIF measurement was conducted quantitatively and detailedly from different aspects, which is not included in previous research. Results show that the average velocity precision is rather high with a high-quality signal, typically with a relative error about 0.64% under the signal SNR 43.96 dB. Especially, the error by laser output fluctuation can be remarkably restrained by a highly saturated stimulation scheme in this article.
1 February 2021 in Journal of Physics: Conference Series
Considering the difficulty of large computation and the characteristic of helicon plasma, a modified collisional-radiative model was proposed for the diagnosis of low-temperature helicon argon plasma. A simplified 47-level is proposed due to the lack of experimental support of transition data at high levels as well as heavy computation to obtain macroscopic parameters of helicon argon plasma, e.g., electron number density ne and electron temperature Te. A creative twice-matching method is proposed in the model because the current double-line intensity ratio method shows significant sensitivity in diagnosing low-temperature electrons. Calculations based on this model shows the spectrum intensity depends on the electron temperature as well as density for low-temperature plasma, especially when it’s below 6eV. The twice matching process based on the priori knowledge chooses 15 spectrums cognizable within the wavelength from 680nm to 860nm, adopting the absolute values of the lines to match with the results calculated by the collisional-radiative model. This method greatly reduces the average error to 13.7%. The result indicates that the precision of the electron temperature and density has been improved a lot and the relative errors are 25% and 40%, respectively. Within the accuracy range above, the research shows when RF power is 500-800W and the pressure is 0.5-1.3Pa, the electron number density rises with the increasing RF power and decreases with the increasing magnetic field strength (450-900G) and gas pressure. Moreover, comparing to the number density of electrons, the electron temperature changes less and rises with the decreasing pressure.
31 October 2018 in IEEE Transactions on Plasma Science