In all combustion devices, the flame stability and thermal efficiency are strongly influenced by the fuel and oxidizer mixing processes. In spray combustion (e.g., in gas turbines), mixing is associated with breakup and atomization of dense sprays. It is necessary to view these processes in detail to obtain knowledge vital for improving them.1 Photography, holography, and x-ray imaging, which have been used for many years in such research,2, 3 have failed to provide clear views because noise from multiple scattering obscures the signal needed to acquire a useful recording.
In recent years, new methods of separating the signal from the noise have been developed, including pseudo-ballistic photon imaging,4–7 in which high-speed optical gates separate ballistic (not scattered) and only-once-scattered, near-ballistic, ‘snake’ photons from multiply scattered photons. Another problem in imaging spray particles is the wide depth of field needed to bring all of them into focus. Current ballistic imaging systems provide only the shadows of particles without resolving the third dimension.4–7
We are combining pseudo-ballistic imaging with digital holography for the first time to resolve the digitally recorded hologram of a dense spray in three dimensions. The hologram is the out-of-focus diffraction pattern of all of the spray particles, which we can reconstruct numerically at different depths throughout the viewed volume. This procedure brings every particle into focus and detects its position.
Figure 1. Picosecond digital holography system design. DM: Dichroic mirror. ω: Fundamental beam. 2ω: Second-harmonic beam. BS: Beam splitter cube. F1: Focal plane. OP: Optical path length compensation. L1: Lens, f=400mm. L2: Lens, f=200mm. WP: Wave plate. P1, P2: Cross polarizer. OKC: Optical Kerr cell.
Figure 1 shows the digital holocamera design. The laser output is split into three different beams: the fundamental 1.06μm beam, which is used to open the Kerr gate, a second-harmonic beam to pass through the particle field as an object wave, and another second-harmonic beam that passes around the particle field to act as a reference wave for the hologram. These two waves must be matched precisely at the recording (hologram) plane, with an error much less than the length of the gated pulse (better than 1mm). The gate pulse is adjusted with a delay line to arrive at the Kerr gate just before the holography pulse. The overlap time of these two pulses will determine the effective gating time. When the Kerr gate is open, the object and reference waves pass through and are recorded on the digital camera.1
Figure 2. Reconstructed images from digital picosecond holograms with varying gate times imaged through a small-particle-emulsion scattering cell. (a) First 10ps of transmitted light, (b) last 10ps of transmitted light, and (c) all transmitted light.
Figure 2 shows reconstructed images of crossed 70μm-diameter hairs separated in space by 100mm. They were produced from holograms recorded in the picosecond holocamera through a turbid cell containing a milk emulsion that scattered about 98% of the light from the system. In Figure 2(a), the delay is set to allow the passage of the front end of the data pulse, which contains ballistic and snake photons, as well as some scattered light that has not had much path length added to it. The reconstructed images of the crossed wires are clear. In Figure 2(b) the delay is set to allow only the end of the beam, which contains ballistic photons and more scattered light, which has been delayed by the scattering action. The image in Figure 2(b) is clearly degraded and low in contrast. In Figure 2(c), the gate remains open for the entire pulse (the polarizers P1 and P2 are parallel), allowing all of the data beam, including all the scattered light, to pass. In this case, the intensity of the signal pulse is stronger, and additional neutral density filters were placed in front of the CMOS camera. This image has higher contrast than image 2(b), in which scattered light is favored, but is clearly inferior to image 2(a), in which noise has been gated out.1
On the basis of the foregoing research and experimental demonstrations, we conclude that a picosecond-gated, digital holocamera can become an extremely powerful diagnostics tool for studying the mechanics of dense sprays.1 The results demonstrated here show that a combination of ballistic imaging and digital holography is a promising method for imaging 3D structures through turbid media. In the next step of our research, we will use shorter laser pulse durations, which will be close to the relaxation time of our Kerr cell (1.8ps), to achieve the optimal noise filtering possible with this technology. We will then investigate the applicability of high-resolution 3D imaging of spray particles.
Derek Dunn-Rankin, Ali Ziaee, John Garman, Wytze van der Veer
University of California
Derek Dunn-Rankin is a professor in the Department of Mechanical and Aerospace Engineering in the Henry Samueli School of Engineering.
Jim Trolinger, Ben Buckner, Ivan Tomov
1. J. Trolinger, B. Buckner, I. Tomov, W. van der Veer, D. Dunn-Rankin, J. Garman, Probing dense sprays with gated, picosecond, digital particle field holography, Int'l J. Spray Combust. Dyn. 3, p. 351-366, 2011.
2. J. D. Trolinger, W. D. Bachalo, Particle field diagnostics systems for high temperature/pressure environments, EPA/ERDA Symp. High Temp./Pressure Particulate Control, 1977.
3. W. Cai, C. F. Powell, Y. Yue, S. Narayanan, J. Wang, M. W. Tate, M. J. Renzi, A. Ercan, E. Fontes, S. M. Gruner, Quantitative analysis of highly transient fuel sprays by time-resolved x-radiography, Appl. Phys. Lett. 83, p. 1671-1673, 2003.
4. M. A. Linne, M. Paciaroni, E. Berrocal, D. Sedarsky, Ballistic imaging of liquid breakup processes in dense sprays, Proc. Combust. Inst. 32, p. 2147-2161, 2009.
5. M. A. Linne, M. Paciaroni, J. R. Gord, T. R. Meyer, Ballistic imaging of the liquid core for a steady jet in crossflow, Appl. Opt. 44, p. 6627-6634, 2005.
6. J. R. Gord, T. R. Meyer, S. Roy, Applications of ultrafast lasers for optical measurements in combusting flows, Annu. Rev. Anal. Chem. 1, p. 663-687, 2008.
7. D. L. Sedarsky, M. E. Paciaroni, M. A. Linne, J. R. Gord, T. R. Meyer, Velocity imaging for the liquid-gas interface in the near field of an atomizing spray: Proof of concept, Opt. Lett. 31, p. 906-908, 2006.