Loading...
Please wait, while we are loading the content...
Exploration of Temporal and Time-Averaged Two-Phase Flow Structures Using X-Ray Diagnostics
| Content Provider | Semantic Scholar |
|---|---|
| Author | Inc Beavercreek Kastengren, Alan L. |
| Copyright Year | 2017 |
| Abstract | Time-averaged and temporal structures of the two-phase flows within the nozzle of an aerated-liquid injector were experimentally explored with synchrotron x-ray diagnostics, including pathlength-integrated and confocal x-ray fluorescence and x-ray high-speed imaging. A beryllium nozzle with a constant passage diameter was fabricated to mate with an aerated-liquid injector featuring the outside-in aerating scheme. Water and nitrogen were doped with x-ray fluorescent elements at low concentrations to facilitate the x-ray diagnostics. The present study shows that the twophase flows inside the nozzle section exhibit an annular-like flow pattern, while the aerating gas is distributed in a Gaussian-like pattern. Axial distributions of the average liquid density, gas density, and liquid velocity were also explored by integrating the line-of-sight properties over several cross-sectional areas and axial locations within the nozzle section. For the first time on these sprays, a confocal x-ray fluorescence measurement was carried out with a polycapillary x-ray optic to spatially resolve the cross-sectional mass distributions within the near field of various discharged plumes. Plume separation phenomena were readily observed in both liquid and gas plumes. High-speed imaging shows that large-scale structures are mainly present in the plenum section and can be aerodynamically stretched into fine structures near the nozzle exit. __________________________________________ * Corresponding author, Kuocheng.Lin.ctr@us.af.mil ILASS Americas, 29 Annual Conference on Liquid Atomization and Spray Systems, Atlanta, GA, May 15-18, 2017 INTRODUCTION Liquid aeration, also called ‘effervescence’ or ‘barbotage,’ is a plausible approach to enhance liquid atomization in several applications. The working principle involves the mixing of a relatively small amount of aerating gas (by mass) with liquid for the creation of a two-phase mixture inside the injector prior to discharge. For well-designed injector and injection conditions, a well-dispersed plume can be generated for effective mixing with the ambient air. With the application of optical-based disgnostics, such as shadowgraph and phase Doppler particle analyzer, it has also been shown that the liquid aeration technique can generate a spray that penetrates deeply into high-speed crossflows and produces favorable plume/droplet properties, such as a high-concentration smaller droplets distributed over a larger plume cross section. Recently, efforts to characterize the near-field structures of aerated-liquid jets were carried out with various x-ray diagnostics, including x-ray phase contrast imaging (PCI), x-ray radiography, and x-ray fluorescence. For instance, the x-ray PCI technique is capable of providing qualitative visualization of droplets, ligaments, and even bubbles within the near field of the aerated-liquid jets. Droplet diameter, bubble diameter, and bubble film thickness within the periphery of the spray could also be measured from the x-ray PCI images. The studies with pathlength-integrated x-ray radiography and fluorescence techniques provide quantitative and time-averaged characterization of liquid and mass distributions within the near fields of aerated-liquid jets. Both techniques were also used to quantitatively characterize the two-phase flow structures inside an aerated-liquid injector, which was made out of beryllium and features the outside-in aeration scheme. Beryllium has fairly low x-ray absorption and, therefore, allows for high transmittance of incident, transmitted, and emitted x-ray photons. The objective of the present study is to characterize both time-averaged and temporal two-phase flow structures generated from an aerated-liquid injector featuring the outside-in aeration scheme. For this aeration scheme, the aerating gas flows through an annular passage before entering the central liquid stream for mixing. Time-averaged flow properties within the beryllium nozzle and also the near field of discharged plumes were characterized with pathlength-integrated and confocal x-ray fluorescence techniques. In addition, high-speed imaging with “white beam” x-rays was utilized temporally resolve qualitative flow evolution and even quantitative flow velocity in this study. EXPERIMENTAL METHODS Experimental Setup The experiment was conducted at the 7-BM beamline of the Advanced Photon Source (APS) at Argonne National Laboratory. For the present study, the injector assembly and liquid spray were placed in the path of a small x-ray beam. Water and aerating gas were supplied into the aerated-liquid injector at the desired flow rates to form a two-phase mixture inside the injector before discharge into a quiescent environment. The aerated-liquid injector features the outside-in aeration configuration with the aerating gas flowing through an annular passage before entering the central liquid stream for mixing. Figure 1 shows a schematic of the injection assembly, which has a stainless steel injector body, an aerating tube, and an exchangeable beryllium nozzle. The aerating tube has an internal diameter of 2.2 mm and features 32 small orifices located in 16 rows with two 0.18-mm diameter orifices in each row. Orifices in each row are offset by 90 degrees from the orifices in adjacent rows, resulting in an orifice distribution pattern in four columns on the aerating tube. A beryllium nozzle was fabricated for the present study. Beryllium has high x-ray transmittance compared to other metals and permits interrogation of the twophase flow inside the nozzle passage. Schematics of the nozzle are shown in Fig. 2. The nozzle features a contoured entrance, followed by a cylindrical passage with a constant diameter (do) of 1.0 mm (0.04 in.) and a passage length (L) of 10 mm, which lead to L/do of 10. The nozzle has a short plenum section with a diameter of 2.2 mm to smoothly connect with the aerating tube. The aerated-liquid jet was vertically discharged into a collecting bucket with a small opening on the cap to minimize the probability of stray droplets entering the beam path. The distance between the nozzle exit and the bucket cap was kept around 5 mm for probing inside the beryllium nozzle and 15 mm for probing of the discharged jet, in order to avoid splashing. Both the aeratedliquid injector and the collecting bucket were rigidly mounted on a traversing table, which provided movement normal to the x-ray beam. X-Ray Measurements 7-BM beamline at Argonne National Laboratory is dedicated to ultrafast x-ray radiography, fluorescence, and tomography experiments in fuel sprays and associated phenomena. The x-ray source for the beamline is a synchrotron bending magnet, which produces nearly collimated, broadband x-ray emission. The beamline consists of two radiation enclosures. The first enclosure (7BM-A) houses a pair of slits to limit the x-ray beam size and a double multilayer monochromator (1.2 – 4.3% ∆E/E). The monochromatic beam then passes into the second radiation enclosure (7BM-B), which houses the experimental equipment. More information regarding the beamline performance can be found in the study of Kastengren et al. Based on previous success in probing the two-phase mixtures inside a beryllium nozzle, x-ray fluorescence was used to simultaneously measure gas and liquid concentrations. When an element is illuminated with x-rays, ionization of core shell electrons occurs. To relax from this highly excited state, atoms can emit an x-ray fluorescence photon. The energy of this photon is characteristic of the atomic number of the atom emitting the photon, and the emitted fluorescence flux is proportional to the amount of fluorescent material in the beam. As such, x-ray fluorescence is commonly used for nondestructive elemental analysis. More detail of the x-ray fluorescence process is given in Refs. [16, 18, 19]. X-ray fluorescence from the liquid and gas phases can be used to determine the distributions of liquid and gas independently, by doping the working fluids with bromine (Br) and krypton (Kr) atoms, respectively. Sodium bromide (NaBr) was dissolved into the water at a concentration of 0.13% by mass. This concentration was monitored during the experiment by measuring the electrical conductivity of the water, which was approximately 1500 μS/cm. The gas phase was pre-mixed (by the gas vendor) to a concentration of 3% Kr (and 97% N2) by volume. The fluorescence signals were measured by a silicon drift diode detector, which was oriented 90° to the x-ray beam in order to minimize elastic x-ray scattering. For the current x-ray fluorescence experiments, the mean x-ray beam photon energy was set at 15 keV; a photon energy of at least 14.3 keV is required to excite both Br and Kr. The beam was focused using a pair of 300 mm long Kirkpatrick-Baez focusing mirrors. The beam focus is approximately 5 6 μm FWHM V H, located approximately 500 mm from the center of the horizontal focusing mirror. The effective size of the beam for the current sprays (which are several mm wide) is somewhat greater than this minimum focus size. The experimental setup at the 7-BM beamline is shown in Fig. 3. Two setups for the fluorescence detector were used in the present study. The first setup measured the pathlength-integrated mass distributions within the beryllium nozzles or within the discharged plumes, as illustrated by the schematics in Fig. 4(a). This is similar to the setup used in previous x-ray fluorescence measurements of aerated sprays. Obliquely scattered x-rays were mitigated using a set of Soller slits. In addition, aluminum and selenium filters were used to filter out elastically scattered photons and x-ray fluorescence photons from impurities (e.g., Fe, Cu) in the beryllium nozzle. The detector area measures 100 mm, and was positioned approximately 100 mm from the injector. The second setup used confocal x-ray fluorescence to examine only a small part of the beam path through the spra |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.ilass.org/2/conferencepapers/010_2017.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
