Cameron J. Dasch

One-dimensional tomography: a comparison of Abel, onion-peeling, and filtered backprojection methods

It is shown that the Abel inversion, onion-peeling, and filtered backprojection methods can be intercompared without assumptions about the object being deconvolved. If the projection data are taken at equally spaced radial positions, the deconvolved field is given by weighted sums of the projections divided by the data spacing. The weighting factors are independent of the data spacing. All the methods are remarkably similar and have Abelian behavior: the field at a radial location is primarily determined by the weighted differences of a few projections around the radial position. Onion-peeling and an Abel inversion using two-point interpolation are similar. When the Shepp-Logan filtered backprojection method is reduced to one dimension, it is essentially identical to an Abel inversion using three-point interpolation. The weighting factors directly determine the relative noise performance: the three-point Abel inversion is the best, while onion peeling is the worst with approximately twice the noise. Based on ease of calculation, robustness, and noise, the three-point Abel inversion is recommended.

Continuous-wave probe laser investigation of laser vaporization of small soot particles in a flame

Pulsed laser vaporization of small soot particles (<100-nm radius) in a flame is observed with a cw probe laser. Both light scattering and absorption along the pulsed laser beam are measured. Above a threshold of 0.2 J/cm2, both quantities decay exponentially with increased fluence for submicrosecond pulses. These two measurements demonstrate that the number of particles remains constant, but the mean size decreases: that is, small particles vaporize rather than fragment or photophorese. The vaporization process is modeled including transport across the first gas mean free path (Langmuir layer). Numerical integration of the time-dependent conservation equations demonstrates that a simple analytic treatment is adequate. The threshold and fluence dependence are predicted to within experimental uncertainty, assuming soot has the thermal properties of graphite. Laser vaporization of soot has possible application to laser beam profiling, gas velocity measurements, flow visualization, and point measurement of soot absorption.

Physical properties of vapor-grown carbon fibers

Abstract Vapor-grown carbon fibers (VGCF) are produced by depositing a layer of pyrocarbon from the vapor phase on a catalytically grown carbon filament. This morphology determines many properties of the fiber, since the filament is more graphitic than the pyrocarbon. In this paper we compare VGCF produced by a continuous process with those grown on a substrate. Fibers having thicker pyrocarbon layers are less graphitic as measured by X-ray diffractometry, electron diffraction and Raman spectroscopy. The bulk density of the fibers, near 2.03 g/cm3, is relatively high for carbon fibers; it decreases slightly as the pyrocarbon thickness increases. The surface area of the fibers determined by N2 adsorption is not larger than the calculated geometric area, indicating that the surface is relatively smooth and free of pores. Each of these measurements indicates that fibers produced by a continuous process are comparable to those grown on substrates, with respect to graphitization and surface properties.

Atmospheric pressure premixed hydrocarbon-air flames: Theory and experiment

Local measurements of temperature and species concentrations in flames are necessary for the successful development of accurate models of flame propagation and structure. To date, one of the few hydrocarbon fuel flames that has been modeled to include the kinetics of many chemical reactions is the premixed,methaneair flame. These theoretical models include both species difusion and thermal conduction, and they are restricted to laminar propagation only. Most previous comparisons of these theoretical models to experimental data were for low-pressure (i.e., approximately 5 kPa) flame data only. Here, however, temperature and composition profiles of fuel, O2, CO, H2, CO2, H2O, and OH are reported for both atmospheric pressure, premixed, laminar, methane-air, and propane-air flames. The comparison between one of the existing theoretical models and these experiments shows good agreement for fule, O2, H2O, CO, CO2, and OH. Systematic deviations from the theoretical predictions are observed for the H2 concentration profiles.

The decay of soot surface growth reactivity and its importance in total soot formation

Abstract The empirical soot surface growth kinetics of Haynes and Wagner are reconciled to the acetylene kinetics of Harris and Weiner by including the decay of the surface reactivity. The empirical rate constant measured by Haynes and Wagner is the decay rate of the surface growth reactivity, not the surface growth rate itself. This rate has a value of 50–400 s−1 and appears to depend only on temperature, but not pressure and flame composition. This description of surface growth with coagulation but without particle inception predicts the spherule size in soot chains and a near-constant value of the total soot surface area concentration. An analytic expression for the volume fraction history indicates small discrepancies with Haynes and Wagners empirical form. When the initial number density is high (>1011 cm−3), the highly nonlinear dependence of the final volume fraction f v ∗ on the initial “inception” volume fraction fv0 is predicted to be f ∗ 1 3 = (f 0 ) 1 3 + (f sg ) 1 3 , where fsg is dependent on measured kinetic parameters and is of the order 2 × 10−8. Using this kinetic description it should be possible to extrapolate late volume fractions backward to give either initial volume fractions or number densities. Given this sensitivity, it will be important to understand the systematic dependences of inception soot concentrations.

Spontaneous Raman scattering by ground-state oxygen atoms

We report the first known observation of Raman scattering by oxygen atoms. The (3)P(2)?(3)P(1) and (3)P(2)?(3)P(0) transitions in the electronic ground state that produced Raman shifts of 158 and 227 cm(-1) were detected. These transitions were observed in a fuel-lean atmospheric H(2) + O(2) flame. By comparing the O electronic and O(2) pure-rotational Raman-scattering intensities, we measured the polarized cross sections for the two lines to be 6 +/- 1 x 10(-31) and 4 +/- 1 x 10(-31) cm(2)/sr, respectively, with an excitation source at 532.1 nm. These cross sections are two to three times stronger than those predicted by a single-configuration single-excitation Coulomb approximation.

Planar imaging of soot formation in turbulent ethylene diffusion flames: Fluctuations and integral scales

Abstract Planar imaging of soot in a series of atomspheric ethylene diffusion flames ranging from laminar to turbulent is reported. From single shot images of 532 nm light scattered at right angles by soot, the fluctuations, intermittency, and integral scales of soot are determined. The soot in turbulent (Re = 5500) ethylene flames occurs in highly convolved regions, suggestive of roll-up vortices. The regions are thin and elongated, lying somewhat perferrentially in the axial direction. The soot integral scales are smaller than velocity integral scales in corresponding cold jet flows. The soot in the late oxidative stages of the flame occurs in a few, highly intermittent, high concentration regions. The regions of low concentration have been oxidized away. This suggests that the soot emissions of turbulent flame are determined by (1) the regions of highest soot concentration, which may avoid oxidation by radiative cooling, as suggested by Kent, or (2) a few regions of high soot concentration that, by stochastic chance, fail to mix adequately with hot oxidizing gases in the transit time to the cooler gases above the flame. At given axial distances from the flame, there is little radial variation in the integral scale and strength of scattering within soot-containing eddies, but the number density of soot eddies does vary. This suggests that the soot chemistry is fast only within selective eddies, presumably of high temperature, and that the radial profiles are determined by mixing.

New soot diagnostics in flames based on laser vaporization of soot

Laser vaporization of soot has been developed into several new diagnostics for soot in flames. Soot diffusion and nucleation rates are sufficiently slow in most flames that the vaporized soot volumes persist, as optically labeled regions for tens of milliseconds. This provides new means of flow visualization in sooting flames with and without probe lasers; nucleation zone determination; laser doppler velocimetry using the nascient soot rather than seed particles; and a method for “point” soot absorption measurements using crossed vaporization and probe laser beams.

Spatially resolved soot-absorption measurements in flames using laser vaporization of particles

Soot absorption in flames is measured with high spatial resolution using a new technique. In this method, the absorption of a probe laser is modulated by vaporization of the soot using a crossed, pulsed pump laser. A detailed analysis shows that the effective beam overlap or interaction length is insensitive to temperature, particle size, or pulsed-laser fluence (energy/area) above a threshold of 5 J/cm(2). The technique has an absorption-coefficient shotnoise-detection limit of roughly 1 x 10(-5) cm(-1) with 0.1-cm spatial resolution using a single laser pulse. Spatially resolved measurements in premixed and in diffusion laminar flames are presented.

Velocimetry in laminar and turbulent flows using the photothermal deflection effect with a transient grating.

An enhanced-precision, high-signal method for velocity measurements with photothermal deflection spectroscopy is presented. A transient refractive-index grating is formed by the interference of two pulsed, pump-laser beams in an absorbing gas. The motion of the grating is detected by the oscillatory deflection of a probe beam, which has a diameter smaller than the fringe spacing. This two-beam pump improves on single-beam pumps because there are more markers for the velocity determination, and the larger thermal gradients increase the probe deflection. The method is illustrated by velocity maps in a laminar ethylene/nitrogen jet using a CO(2) pump laser. Velocity distributions and noise levels were also measured with grid-induced turbulence above the jet.