Summary: This grant is aimed at more predictive models for primary breakup of sprays. It combines model development with detailed experiments. Experiments will use transparent nozzles so that interior flows can be studied using high speed long distance microscopic shadowgraphy and micro-PIV. Primary breakup will be studied using ballistic imaging (BI)§. A 3-pulse system will be set up so that we can correlate 2 images to extract the velocity of the liquid-gas interface. Three pulses will give 2 velocity images allowing us to extract the acceleration vectors. In addition, software will provide statistics on surface curvature, void size distributions, surface wave amplitude vs. frequency and so forth. To better understand primary breakup, we will partner with colleagues doing direct numerical simulation (DNS) of breakup (Prof. H. Pitsch at U. Aachen in Germany, Prof. M. Hermann at Arizona State U., and Prof. M. Trujillo at U. Wisconsin). We will purposely operate under conditions appropriate for DNS because it involves some assumptions and the ballistic images will help confirm their appropriateness. We can then learn much more about breakup dynamics from DNS than we can from ballistic imaging. We can then extrapolate that detailed level of understanding to higher Reynolds numbers by use of ballistic imaging alone. Droplet size distributions and velocities will be acquired using phase Doppler interferometry. Overall images will be taken with planar imaging techniques, and vaporizing sprays can be studied using combined elastic scattering /laser induced fluorescence of exciplex fluorescence imaging techniques. The sprays will thus be characterized with a level of detail never before applied. The experiments will begin with steady flows that isolate one primary breakup mechanism at a time (e.g. turbulent breakup, shear, cavitation, etc.). Our computational collaborators help us design the experiments before we begin, to make sure we are at least close to their needs. We will then combine breakup mechanisms and then move to a transient jet.
The modeling program will be based on several recent developments by our collaborators. Because interior flows are critical for the development of a breakup model, we will model those using LES or URANS. For interior cavitation, we plan to adapt the dynamic models by Prof. D. Arcoumanis and Prof. M. Gavaises at City College, London. For primary breakup of the jet after exiting the nozzle, it may be possible to correlate interior flows and to use the correlations to set more reliably the tunable constants in existing breakup models (working with Prof. D. Schmidt at U. Mass. Amherst). Second, we will investigate use of the one-dimensional turbulence (ODT) model for primary breakup of Dr. A. Kerstein (recently retired from Sandia Labs; he is a collaborator and will be partially supported as a consultant). Third, we will evaluate the stochastic breakup methods of Prof. M. Gorokhovski at Ecole Central de Lyon. Various full spray models under development at Chalmers will be coupled to breakup models. In addition, we will collaborate with Prof. Eva Gutheil at the U. Heidelberg on overall spray breakup models.
This grant will support Professors Linne and Oevermann, it will pay for 2 computational PhD students and one experimental post-doc. Dr. Oevermann has an additional grant from the Swedish Energy Board for a related PhD student, and we will support some consulting by Dr. Kerstein. The grant will also pay for:
- A third fs amplifier for 3 pulse ballistic imaging
- A spectroscopic YAG/dye system for species imaging in sprays (important for imaging, especially as we go to transient fuel sprays later in the program)
- A new phase Doppler interferometer
- Several types of scientific camera systems (including a high speed camera)
§ For a description of ballistic imaging, see ““Ballistic Imaging of Liquid Breakup Processes in Dense Sprays”, M. Linne, M. Paciaroni, E. Berrocal and D. Sedarsky, invited review article, Proceedings of the Combustion Institute, Vol. 32, pp. 2147-2161, (2009).