An official website of the United States government
Soot formation is quantified in detail (volume fraction, particle size, number concentration, and light emissivity dispersion exponent) in a series of partially premixed counterflow flames of ethylene at equivalence ratios equal to 6.5, 5.0, and 4.0, and with maximum temperature spanning approximately 200 K. The focus is to investigate the effect of peak temperature and equivalence ratio on soot formation while maintaining constant global strain and stoichiometric mixture fraction. Oxygen is progressively displaced from the oxidizer to the fuel stream of a diffusion flame to stabilize partially premixed flames of decreasing, showing a double-flame structure consisting of a rich premixed flame component stabilized on the fuel side of the stagnation plane and a diffusion flame component stabilized on the oxidizer side. Soot is detected in the region sandwiched between the two flame components, is formed in both of them, and is convected away radially at the Particle Stagnation Plane (PSP). At fixed , raising the peak temperature invariably raises the soot volume fraction throughout the probed region. Vice versa, at fixed peak temperature, lowering the equivalence ratio causes the premixed flame component to shift away from the diffusion flame component, with the consequent broadening of the soot forming region and an increase in both soot volume fraction as well as soot particle sizes through an enhancement of surface growth. Detailed probing of the region in the vicinity of the PSP offers evidence of soot oxidation from molecular oxygen. Furthermore, when the maximum temperature is sufficiently low, the net soot production rate turns negative because surface oxidation overwhelms surface growth. Comparing the soot number production rate inferred from experiments to the dimerization rate of benzene, naphthalene, and pyrene reveals that only the smallest aromatics are present in flames at sufficiently large concentrations to account for soot nucleation. This observation applies to both the diffusion flame and the premixed flame components and confirms previous findings in strictly diffusion flames.
more »
« less
The Planar Mixing Layer Flame (PMLF): a canonical configuration for studying non-premixed combustion chemistry and soot inception
The study of fuel chemistry and soot inception in non-premixed combustion can be advanced by characterizing flame configurations in which the advection and diffusion transport can be finely controlled, with the ability to decouple pyrolysis from oxidation. Also, the ideal flames to be investigated should be perturbed minimally by probes and thick enough for sampling techniques to yield spatially resolved measurements of their structure. The Planar Mixing Layer Flame (PMLF) configuration introduced herein is established between a fuel and an oxidizer slot jet adjacent to each other and shielded from the ambient air by annularly co-flowing inert nitrogen. The PMLF flow is kept laminar and steady by an impinging flat plate equipped with a rectangular exhaust slit opening which anchors the position of the hot combustion products via buoyancy. The PMLF is accessible to sampling and its flow stability is preserved when using any tested probe. The experiments are complemented with 2DComputational Fluid Dynamics (CFD) modeling with detailed chemical kinetics. The results demonstrate that the PMLF has a self-similar boundary layer structure whose horizontal cross-sections are equivalent to properly selected and equally thick 1D- Counterflow Flames (CFs). The equivalence allows for excellent predictions of the PMLF thermochemical structure characterized experimentally but at a small fraction of the 2D-CFD computational cost. The 1D-CF equivalence affects even aromatics less than twofold despite their kinetics being known to be very sensitive to the temperature field. Importantly, the PMLF thickness is several millimeters and grows at increasing HABs so that the equivalent 1D-CFs have strain rates as small as 7.0 /s which cannot be studied in CF experiments. As a result, the PMLF emerges as a promising canonical non-premixed flame configuration for studying flame chemistry and soot inception on time scales of tens of milliseconds typical of many combustion applications.
more »
« less
- Award ID(s):
- 2013382
- PAR ID:
- 10553929
- Publisher / Repository:
- https://doi.org/10.1016/j.combustflame.2024.113537
- Date Published:
- Journal Name:
- Combustion and Flame
- Volume:
- 266
- Issue:
- C
- ISSN:
- 0010-2180
- Page Range / eLocation ID:
- 113537
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Soot and its gaseous precursors are quantified in detail (precursors up to 166 amu, volume fraction, particle size, number concentration, and light emissivity dispersion exponent) in a laminar partially premixed counterflow flame of ethylene. The investigated flame has an equivalence ratio Φ = 2.43 and a mixture fraction Zst = 0.4, resulting in a distinct double-flame structure consisting of a rich premixed flame component and a diffusion flame component, both stabilized on the fuel side of the stagnation plane. The value of the equivalence ratio makes the premixed flame the dominant contributor to soot production, with soot being oxidized completely by OH from the diffusion flame component. Particle size is measured to increase quasi-monotonically, but remains within a few nanometers throughout the soot forming region. Aromatic species are primarily formed in the post flame region of the premixed flame. Their mole fractions peak close to the premixed flame and decrease as the diffusion flame is approached. The experimentally measured gaseous species are captured well by kinetic models, with the exception of two critical species in soot chemistry: benzene and naphthalene.more » « less
-
This work describes and tests the calibration process of the chemical-diffusive model (CDM) for the simulation of non-premixed diffusion flames. The CDM is an alternative, simplified approach for incorporating the effects of combustion in a fluid simulation, based on the ideas of regulating the rate of energy release such that the properties of combustion waves (e.g. flames and detonations) are reproduced. Past implementations of the CDM have considered single-stoichiometry fuel-air mixtures or mixtures with variable stoichiom- etry but with premixed modes of combustion. In this work, the CDM is tested and shown to work for non-premixed, low-Mach-number flames (i.e., diffusion flames) by incorporat- ing it into a numerical model which solves the reactive and compressible Navier-Stokes equations with the barely implicit correction (BIC) algorithm, which removes the acoustic limit on the integration time-step size. Simulations of one-dimensional premixed laminar flames reproduce the required premixed laminar flame speed, thickness, and temperature. A two-dimensional, steady-state, laminar coflow diffusion flame is computed, and the result demonstrates the ability of the algorithm to compute a non-premixed flame. Lastly, a two- dimensional simulation of two opposing jets of fuel and air show that the CDM approach can compute the structure of a counter-flow diffusion flame.more » « less
-
Abstract: Lean premixed (LP) combustion systems are currently used for most modern power generation gas turbines. Though this method reduces emissions, specifically nitrogen oxides, and is more efficient than non-premixed systems, LP systems are susceptible to blowoff. The goal of this study is to find out how centerbody geometry plays a role in the lean blowoff process for swirl-stabilized flames. We find that cylindrical centerbodies have higher lean blowoff equivalence ratios than tapered centerbodies. We also find that the dominant flame shape for both centerbodies is M-shape when not anchored and tulip shaped when anchored, though the tapered centerbodies induce V-shape flames as well. The blowoff equivalence ratio and blowoff process are strongly coupled ith the flame shape.more » « less
-
We perform spatially resolved measurements of light scattering of soot in atmospheric pressure counterflow diffusion flames to complement previously reported data on soot pyrometry, temperature and gaseous species up to three-ring polycyclic aromatic hydrocarbons (PAHs). We compare two flames: a baseline ethylene flame and a toluene-seeded flame in which an aliquot of ethylene in the feed stream is replaced with 3500 ppm of pre-vaporized toluene. The goal is twofold: directly adding an aromatic fuel to bypass the formation of the first aromatic ring, widely regarded as the main bottleneck to soot formation from aliphatic fuels, and assessing the impact of a common component of surrogates of transportation fuels on soot formation. The composition of the fuel and oxidizer streams are adjusted to ensure invariance of the temperature-time history, thereby decoupling the chemical effects of the fuel substitution from other factors. The doping approach enables the comparison of very similar flames with respect to combustion products, radicals and critical precursors to aromatic formation (C2–C5 species), in addition to the temperature-time history. Doping with toluene boosts the aromatic content and soot volume fraction relative to the baseline ethylene flame, but, surprisingly, the soot number density and nucleation rate are affected modestly. As a result, the observed difference in volume fraction in the toluene-doped flame is reflective of larger initial particles at the onset of soot nucleation. The nucleation rate when soot first appears near the flame is of the same order as the dimerization rate of single-ring aromatics, in contrast with the expectation that the dimerization of larger PAHs initiates the process. Even though in and of itself nucleation contributes modestly to the overall soot loading, nucleation conditions the overall soot loading by affecting the size of the initial particle, which ultimately affects subsequent growth.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1016/j.combustflame.2024.113537
