Multiple-scattering effects on infrared scattering measurements used to characterize droplet size and volume fraction distributions in diesel sprays

Abstract

We quantify the maximum error due to multiple-scattering effects for an infrared scattering droplet izing technique. Errors in Sauter mean diameters (SMDs) and liquid volume fractions were estimated lased on experimentally determined polarization properties of the scattered light. Light that is multiply scattered from spherical particles becomes randomly polarized, whereas singly scattered light from a spherical particle contains no cross-polarization scattering component. Therefore measurement of the cross-polarization component (in this case parallel) of the scattering signal is a measure of the multiply scattered light. A ratio of parallel to perpendicular polarized scattered light was experimentally determined and used to calculate an error due to multiple scattering. The infrared scattering measurements and polarization measurements used to quantify the multiple-scattering errors were applied to a typical diesel spray that was injected into three different background conditions: a room ambient condition; a room-temperature, high-pressure condition; and a combusting condition. Droplet SMD, liquid volume fraction, and multiple-scattering errors were determined for a number of locations within the spray; results indicate that the combusting case is negligibly affected by multiple scattering. However, the room ambient case exhibited notable errors due to multiple scattering near the centerline of the spray, and the high-pressure case demonstrated susceptibility to multiple scattering throughout all regions investigated. It is important to note, however, that multiple-scattering errors in many cases translate into relatively small effects on the reported droplet sizes.

Quantitative imaging of a non-combusting diesel spray using structured laser illumination planar imaging

“…the same distance between the laser sheet and the camera and constant camera gain), K a should remain constant at each position m of the laser sheet in the spray. According to equation (11), the camera function K a is calculated from the relationship between the recorded scattered light S sm and the extinction measurements (I i − I f ). Thus, an increased value of S sm due to multiple scattering residuals also results in an increased value of K a .…”

“…In the derivation of the camera function, it is assumed to be constant as a function of x (see equation (11)). Therefore, the value of K a for a specific voxel actually represents the averaged value of K a for all voxels with the same position in Y and Z.…”

“…With the extinction coefficient known, a complementary measurement (using, for instance, laser-induced fluorescence (LIF) [4,5], Raman scattering [8,9], polarization ratio [10], or elastic scattering at a different wavelength [11]) of either size or number density would directly yield the other quantity. Laser sheet illumination is an attractive technique for twodimensional mapping of volumetric two-phase flows thanks to its small depth of field and large field of view.…”

“…One technique reducing the contribution of multiple scattering intensity is to scan the sample with a narrow Gaussian laser beam rather than a laser sheet and to use a small detection acceptance angle [20,[22][23][24]. It may also be partly suppressed by means of polarization filtering [11]. However, for an extensive reduction of the signal generated by multiple scattering and detected in laser sheet imaging, a technique called structured laser illumination planar imaging (SLIPI) has recently been developed and is described in [25][26][27][28].…”

Measurement of Sauter mean diameter in diesel sprays using a scattering–absorption measurement ratio technique

“…Multiple-scattering is a well-recognized source of error in extinction measurement systems utilizing detectors with a finite collection angle. 15–17 This phenomenon is responsible for redirecting the scattered light back into the detection system, leading to erroneous optical thickness measurements and ultimately an overestimation of local SMD. In Phase II of the project, a multiple scattering correction was employed to present an estimation of uncertainty introduced by this phenomenon.…”