Search for: All records

To achieve high power density and thermodynamic cycle efficiency, the working pressures of liquid-propellant rocket engines, diesel engines, and gas turbines (based on deflagration or detonation) are continuously increasing, which could reach or go beyond the thermodynamic critical pressure of the liquid propellant. For this reason, the studies of trans- and super-critical injection are getting more and more attention. However, the simulation of transcritical phase change is still a challenging topic. The phase boundary, especially near the mixture critical point, needs to be accurately determined to investigate the multicomponent effects on transcritical injection and atomization. This work used our previously developed thermodynamic model based on the vapor-liquid equilibrium (VLE) theory, which can predict the phase separation near the mixture critical point. An \textit{in situ} adaptive tabulation (ISAT) method was developed to accelerate the computationally expensive multicomponent VLE computation such that it can be cheap enough for CFD. The new thermodynamic model was integrated into OpenFOAM to develop a VLE-based CFD solver. In this work, shock-droplet interaction and two-phase mixing simulations are conducted using our new VLE-based CFD solver. The shock-droplet interaction simulation results capture the thermodynamic condition of the surface entering the supercritical state after shock passes through. The atomization of droplets could be triggered by vorticity formed at the droplets' surface. 2D temporal mixing layer simulations show the evolution of the transcritical mixing layer and capture the phase split effect at the mixing layer. 

Experiments were carried out to observe the flow inside counterflow atomizers over a range of operating conditions and fluid properties. Liquids used were water and propylene glycol, while the gas was either air or helium. Liquid flow rates ranged from 10 ml/min to 40 ml/min, with gas liquid ratio (GLR) ranging from 0.1 to 0.6. The primary experiments used the 7-BM line of the Advanced Photon Source in Argonne National Laboratories with a 2.6 mm atomizer produced from (Poly)Ethyl-Ether-Ketone (PEEK). The X-Ray beam was operated in phase contrast mode, leading to interference patterns near the gas-liquid interface and enabling a qualitative understanding of the flow structure. Complementary optical work applied laser shadowgraphy to a 1 mm orifice atomizer constructed with quartz capillary tubing. A diffuse pulsed Nd:YAG laser backlight captured instantaneous gas-liquid interface positions in the internal flow. With both techniques, two distinct flow behaviors are observed corresponding to low and high GLR values. At low GLR, the inertia of the injected gas is insufficient to penetrate the liquid downflow. The gas stream entering the mixing chamber in the upstream direction is immediately deflected by the denser liquid and enters the discharge tube around a central liquid jet, which is sheared and accelerated by the surrounding gas, leading to breakup. A distinct frequency of jet breakup is observed inside the discharge tube, with the liquid jet oscillating and fragmenting against the walls. The situation at high GLR is quite different, however, as the incoming gas stream asymmetrically penetrates upstream into the mixing chamber, taking the form of a high-speed jet confined along one wall, and displaying a flapping instability as it encounters the liquid flowing downstream. This flapping causes violent mixing, resulting in a highly disturbed interface, along with the generation of liquid ligaments and gas bubbles. This two-phase mixture enters the discharge tube with no liquid jet formation evident for this case. The transition between these two regimes is explored by changing the liquid viscosity and gas molar mass, and weak sensitivity to fluid properties is observed. Further, quantitative image analysis techniques applied to the low and high GLR cases allow extraction of the frequencies of the liquid jet in the discharge tube at low GLR, as well as the flapping mode at high GLR. 

To achieve high performance, the working pressure of liquid-fueled rocket engines, diesel engines, and gas turbines (based on deflagration or detonation) is continuously increasing, which could reach the thermodynamic critical pressure of the liquid fuel. For this reason, the studies of trans- and super-critical injection are getting more attention. However, most of the multiphase researches were mainly concentrated on single- or two-component systems, which cannot capture the multicomponent phase change in real high-pressure engines and gas turbines. The phase boundary, especially near the critical points, needs to be accurately determined to investigate the multicomponent effects in transcritical flow. This work used our previously developed thermodynamic model based on the vapor-liquid equilibrium (VLE) theory, which can predict the phase separation near the critical points. An in situ adaptive tabulation (ISAT) method was developed to accelerate the computation of the VLE model such that the expensive multicomponent VLE calculation can be cheap enough for CFD. The new thermodynamic model was integrated into OpenFOAM to build a VLE-based CFD solver. In this work, simulations are conducted using our new VLE-based CFD solver to reveal the phase change effects in transcritical flow. Specifically, shock-droplet interaction are investigated to reveal the shock-driven high pressure phase change. 

Due to the longer auto-ignition time with liquid fuels compared with hydrogen, the understanding of interaction of shock waves with sprays and the subsequent vapor mixing is significant to design ramjets/scramjets with liquid fuel sprays. In this study, an Eulerian-Lagrangian framework is developed based on the OpenFOAM platform. In this solver, detailed multi-component transport models for Eulerian gas-phase species properties are included. In addition, Lagrangian spray break-up, atomization and evaporation models are added to simulate liquid phase. In addition, an equilibrium wall function is added to model the near-wall properties. The newly developed solver is used to conduct large eddy simulations (LES) on non-reactive liquid jets in supersonic crossflow (JISCF) with liquid sprays. The liquid penetration length are compared with the experimental data, showing a very good agreement. Effects of evaporation and fuel properties (e.g., heat capacity and enthalpy of evaporation) on penetration length, temperature, Sauter mean diameter (SMD) and volumetric parcel flux are discussed in this study. It is shown that evaporation effects primarily show up in the temperature field. For n-heptane sprays, such impact could be conducted to other properties of the flow field like spray plume size, particle size distribution and volumetric flux, which is caused by the smaller enthalpy of evaporation and heat capacity comparing to water. Full version of this paper has been published as a journal article: Shufan Zou, Dezhi Zhou, Suo Yang, “Effects of Evaporation and Fuel Properties on Liquid Jets in Supersonic Crossflow: a Computational study using a compressible Eulerian-Lagrangian solver”, Atomization and Sprays 30 (9) (2020) 675-696. https://doi.org/10.1615/AtomizSpr.2020034860 

The studies of transcritical and supercritical injection have attracted much interest in the past 30 years. However, most of them were mainly concentrated on the single-component system, whose critical point is a constant value. To capture the thermophysical properties of multicomponent, a phase equilibrium solver is needed, which is also called a vapor-liquid equilibrium (VLE) solver. But VLE solver increases the computation cost significantly. Tabulation methods can be used to store the solution to avoids a mass of redundant computation. However, the size of a table increases exponentially with respect to the number of components. When the number of species is greater than 3, the size of a table far exceeds the limit of RAM in today's computers. In this research, an online tabulation method based on In Situ Adaptive Tabulation (ISAT) is developed to accelerate the computation of multicomponent fluid. Accuracy and efficiency are analyzed and discussed. The CFD solver used in this research is based on the Pressure-Implicit with Splitting of Operators (PISO) method. Peng-Robinson equation of state is used in phase equilibrium. 

The studies of transcritical and supercritical flow have attracted much interest in the past 30 years. However, most of them mainly focus on the single-component system, whose critical point is constant. We use the vapor-liquid equilibrium (VLE) theory to capture the thermodynamic properties of the mixture and investigate transcritical flows (i.e., supercritical CO2 oxy-combustion systems). In sCO2 oxy-combustion systems, due to the presence of water from the previous cycles, the mixture critical point increases significantly, such that the phase separation could occur in both the compressor and combustor. However, the VLE solver increases the computation cost of fluid simulation significantly, which limited the size of simulations we can conduct. Naturally, tabulation methods can be used to store the VLE solutions to avoids redundant computation. However, the size of the VLE table increases exponentially with respect to the number of components. When the number of species components is greater than three, the size of the VLE table far exceeds the RAM’s limit in today’s standard computers. In this research, an online tabulation method based on In Situ Adaptive Tabulation (ISAT) is developed to accelerate the computation of multicomponent fluids based on VLE theory. Accuracy and efficiency are analyzed and discussed. The CFD solver used in this research is based on the Pressure-Implicit with Splitting of Operators (PISO) method. Peng-Robinson equation of state (EOS) is used in the calculations of phase equilibrium. 

The studies of transcritical and supercritical injection have attracted much interest in the past 30 years. However, most of them were mainly concentrated on the single-component system, whose critical point is a constant value. To capture the thermophysical properties of multicomponent, a phase equilibrium solver is needed, which is also called a vapor-liquid equilibrium (VLE) solver. But VLE solver increases the computation cost significantly. Tabulation methods can be used to store the solution to avoids a mass of redundant computation. However, the size of a table increases exponentially with respect to the number of components. When the number of species is greater than 3, the size of a table far exceeds the limit of RAM in today's computers. In this research, an online tabulation method based on In Situ Adaptive Tabulation (ISAT) is developed to accelerate the computation of multicomponent fluid. Accuracy and efficiency are analyzed and discussed. The CFD solver used in this research is based on the Pressure-Implicit with Splitting of Operators (PISO) method. Peng-Robinson equation of state is used in phase equilibrium. 

We study the enhanced atomization of viscous liquids by employing a novel two-fluid atomizer. The nozzle establishes a countercurrent flow configuration in which the gas and liquid are directed in opposite directions, establishing a two-phase mixing layer. Detailed measurements of droplet size distributions were carried out using laser shadowgraphy, along with high speed flow visualization. The measurements suggest that the liquid emerges as a spray with little further secondary atomization. The performance of this nozzle is compared to the ‘flow-blurring’ nozzle studied by other investigators for four test liquids of viscosity ranging from 1 to 133.5 mPa.s. The counterflow nozzle produces a spray whose characteristics are relatively insensitive to fluid viscosity over the range studied, for gas-liquid mass flow ratios between 0.25 and 1. To gain insight into the mixing process inside the nozzle, simulations are carried out using an Eulerian-Eulerian Volume of Fluid (VoF) approach for representative experimental conditions. The simulation reveals the detailed process of self-sustained flow oscillations and the physical mechanism that generate liquid filaments and final droplets.