Search

The steady burning and stabilization of the boundary layer diffusion flame over a gasifying condensed fuel surface, commonly called the Emmons flame, is an important problem in the study of boundary layer combustion. We investigate herein, theoretically and numerically, the liftoff distance and blowout limit of the Emmons flame, through the corresponding response of the controlling triple flame in the leading edge of the bulk flame. An explicit solution of the flame liftoff distance and the critical blowout limit is derived, with the theoretical results agreeing well with the numerical simulation for an extensive range of the system parameters. In particular, it is shown that the transversal velocity gradient (TVG) ahead of the triple flame renders the flame harder to liftoff and blowout, with this effect increasing for increasing TVG and decreasing triple flame curvature, which is related to the mixture fraction gradient. Furthermore, the Spalding mass transfer number, B v , for the surface segment ahead of the flame front affects the flame stabilization and blowout limit by modifying the similarity structure of the flow and the location of stoichiometry. Thermal expansion of the flow around the triple flame together with the surface viscous drag also significantly promotes flame stabilization. [ABSTRACT FROM AUTHOR]

Abstract While the presence of an impermeable leading segment is normally inevitable for a boundary layer diffusion flame established over a fuel surface, it also offers an opportunity to control and stabilize the flame by design. In this study, the classical Emmons' problem with an impermeable leading section is investigated numerically and theoretically. The local similarity structure of the boundary layer is discussed and an explicit solution of the skin friction, local gasification rate, and flame characteristics is derived. It is shown that the existence of an impermeable leading section of length L reduces the skin friction and the gasification rate, while increases the flame height in the fuel plate section. This history effect of the flow on the boundary layer development is reduced as the streamwise distance, x , increases, and is negligible for x / L > ∼ 10. The theoretical results agree well with the numerical results, both qualitatively and quantitatively. [ABSTRACT FROM AUTHOR]

Abstract The non-monotonic ignition response of counterflowing premixed hydrogen/oxygen mixtures with nitrogen dilution versus heated nitrogen is studied numerically and theoretically. It is shown that the three ignition limits can be theoretically obtained by considering only the linear system involving at most only one radical in each reaction, while the influences of the nonlinear reactions, each involving two radicals, together with thermal feedback, introduce higher-order corrections, particularly for the third ignition limit. It is also demonstrated that the high diffusivity of H 2 promotes ignition at the third limit. On the other hand, the high diffusivity of the H atom suppresses ignition at the first limit, while the assumption of unity Lewis number for H yields remarkably good results for the other two limits. Furthermore, by solving the time evolution of the crucial H and HO 2 radicals, simplified formulations of the three individual limits and the two quadratic double limits are obtained analytically, in analogy with results for the homogeneous explosion problem. [ABSTRACT FROM AUTHOR]

Agglomeration of aluminum nanoparticles (nano-Al) is inevitable due to the strong van der Waals forces. It is supposed to affect the combustion characteristics of nano-Al. However, the agglomeration effect is not fully understood yet due to the worse description of the morphological evolution of agglomerates during reaction. In this paper the morphological changes of agglomerated nano-Al, which mainly results from the oxide shell thickening when the particle temperature is below the melting point of alumina, are confirmed in the thermogravimetric analyzer (TGA) and Hencken experiments. Three factors ( f A , f ht , and f sp ) are defined to describe the morphological changes quantitatively through proposed formulas. Thus the agglomerate could be equivalent to an effective spherical particle to help studying the effect of agglomeration on combustion. The results indicate two combustion modes for different agglomerates. The small Al agglomerate is hard to combust fully due to the strong heat transfer with ambience, corresponding to the high-temperature oxidation mode. However, at the same environment, as the agglomerate becomes larger, the maximum particle temperature ( T p,max ) increases until that the agglomerate burns fully, corresponding to the fully-fledged combustion mode. At that mode, the burning particle temperature is mainly controlled by the viscous sintering of agglomerates. [ABSTRACT FROM AUTHOR]

The response of counterflow diffusion flames to sub-breakdown DC and AC electric fields, as well as their superposition with ns pulse discharge waveforms, is studied in the plane-to-plane electrode geometry. Sub-breakdown DC and low-frequency AC electric fields cause the flame displacement toward the grounded electrode, in the direction of the applied field, indicating that the body force on the positive ions exceeds that on the electrons and negative ions. As the AC frequency increases, the flame response becomes less pronounced, due to the reduction of the electrohydrodynamic (EHD) body force impulse over the AC half-period. The electric field in the electrode gap is determined by ps Electric Field Induced Second Harmonic (E-FISH) generation, with absolute calibration using sub-breakdown ns pulses overlapped with the measured electric field waveform. The results show that the electric field distribution across the flame in the current saturation regime follows the Laplacian field. This indicates that the space charge density in the gap is too low to distort the applied DC or AC field, consistent with the kinetic modeling predictions. Combining a nanosecond pulse discharge with a sub-breakdown DC field generates a diffuse plasma across the entire gap. Time-resolved and spatially resolved measurements of the electric field in the discharge indicate the ionization wave propagation between the electrodes. The present results do not exhibit a detectable flame displacement enhancement by ns discharge pulses combined with a sub-breakdown field, observed previously. Kinetic modeling calculations show that the absence of this effect in the plane-to-plane geometry is due to the rapid plasma self-shielding. This indicates that alternative electrode geometries limiting the self-shielding would be more effective for the plasma / electric field enhanced flameholding and flame stabilization applications. [ABSTRACT FROM AUTHOR]

Repetitive ns pulse, dielectric barrier discharge voltage waveforms, combined with a tail several ms long, are used to induce oscillations of a counterflow atmospheric pressure diffusion flame. A baseline ns pulse discharge operated at 10 Hz results in a relatively modest oscillatory response of the flame, which becomes more pronounced in burst mode operation, at the same burst repetition rate of 10 Hz. This effect is most likely caused by the residual electric field after the discharge pulse, producing the electrohydrodynamic (EHD) force ("ion wind") on the charges generated during the discharge, although plasma chemistry and Joule heating by the discharge may also contribute. Manipulating the external circuit to add a variable duration tail to the discharge pulse, without changing the pulse shape during breakdown or the pulse repetition rate, considerably enhances the impulse of the EHD force and increases the amplitude of the flame oscillations. To quantify this effect, the electric field distribution between the electrodes during and after the discharge pulse is measured by ps Electric Field Induced Second Harmonic (E-FISH) diagnostic. The results show that the electric field is maintained during the voltage tail, although it is lower compared to the Laplacian field due to the charge accumulation on the dielectric sleeves covering the electrodes. The time scale of the flame oscillations at the present conditions, of the order of ∼10 ms, is limited by the relatively slow momentum transfer from the ions to the neutral species. The present results demonstrate feasibility of enhancing the flame control authority, by combining a high peak ionization fraction generated by a ns pulse discharge with the EHD force applied on a long time scale, using a single plasma generator. [ABSTRACT FROM AUTHOR]

Plasma-induced flow perturbation and its subsequent dynamic effects on a flame are decoupled, both experimentally and theoretically, from the plasma assisted combustion system. A coaxial Dielectric-Barrier-Discharge (DBD) plasma generator is designed, and the discharge is characterized using probes and cameras. Particle tracing method is elaborately demonstrated to map the velocity of flow fluctuation induced by the plasma, with uncertainty estimated to be less than 5%. Then the plasma generator is coupled with the fuel nozzle of the non-premixed counterflow burner. By using multiple diagnostics, several turbulent-like features are observed from the upstream laminar flow (Re ≈ 300) and the downstream non-premixed flat flame, including the distorted velocity profile, fluctuation intensity above 50%, wrinkled flame sheet, and near −5/3 slope of frequency spectra for both fluctuation velocity and CH* chemiluminescence intensity. The aerodynamic effect on the flame is resolved by more than 90% over frequency spectra and then, characterized using flame transfer functions (FTFs). The experimental results show a negative linear correlation between the FTFs' gain and perturbation frequency on the logarithmic plot, which is then verified by a theoretical model derived from the Z -equation of non-premixed counterflow flame. Further, the model indicates that for small perturbations, the influence of the global stretch rate on the FTFs is linear, while the effect of the imposed amplitude is negligible, and the dimensionless perturbation frequency (St) scaled by the global stretch rate becomes the only variable. [ABSTRACT FROM AUTHOR]

Coal particles experience a transition from a reducing to oxidizing environment in the near-burner region of pulverized coal (pc) boilers. For the first time, we report a fundamental study of ignition of a coal-particle stream experiencing a flame environment that transitions from a reducing to an oxidizing environment (termed reducing-to-oxidizing environment). High-speed videography is used to observe the particles in situ, and scanning electron microscopy is used to characterize the sampled particles. The effects of particle size on ignition are presented for four size bins (63–74 µm, 75–89 µm, 90–124 µm and 125–149 µm) for PRB subbituminous coal at two nominal gas temperatures (1300 K and 1800 K). An oxidizing environment with 20% molar oxygen composition is used as base-case. In contradistinction to single particle studies where particles are reported to ignite heterogeneously at higher temperatures, this study shows that coal streams ignite homogeneously, irrespective of particle size, in the oxidizing environment. By changing nominal gas temperature from 1300 K to 1800 K, ignition time decreases, on average, by a factor of five for each of the particle size bins. For both gas temperatures, the trend in ignition delays as particle size changes is non-monotonic. However, at 1800 K nominal gas temperature, ignition delays are independent of particle size in the reducing-to-oxidizing environment and ignition delays are doubled on average when compared to those in the oxidizing environment. It is more noticeable at the lower gas temperature of 1300 K that homogeneous ignition of coal streams is oxygen-dependent below 90 µm particle size and temperature-dependent above 90 µm. In general, ignition delay is determined by volatile release rate (controlled by the particle temperature) and the local oxygen concentration. Micrographs of particles also confirm that ignition and char burnout times are longer in the reducing-to-oxidizing environments than those in the oxidizing environments. [ABSTRACT FROM AUTHOR]

Boron is an attractive high-energy fuel additive. But it could not burn efficiently in practical systems due to its high ignition temperature and slow burning velocity. Finding methods to enhance the combustion of boron is desired. This work focused on the combustion characteristics of boron submicroparticles with and without plasma discharges in a hot environment supported by CH 4 /N 2 /O 2 flat flame based on the optical diagnostics. The boron submicroparticles were dispersed by the nebulization method to control the agglomeration. The well-dispersed boron flame exhibited two different burning modes, depending on the ambient temperature. As the ambient temperature was above 1520 K, the boron flame showed definitely two-stage characteristics where the upstream of particle flow was yellow, corresponding to the first-stage flame, while the downstream was green and diffusive, corresponding to the second-stage flame. The first-stage and second-stage burn times were respectively in the range of 0.46–1.08 ms and 0.92–1.87 ms, as the ambient temperature decreased from 1752 K to 1520 K. The chemical kinetics-controlled mechanism was confirmed by the nearly linear size dependence of the burn time (d 1 law). Nevertheless, as the ambient temperature was below 1520 K, the boron submicroparticles were partially burned or oxidized, exhibiting a mildly orange stream. This mild boron flame could be enhanced using a plasma discharge. The ignition delay time was shortened from 3.06 ms to 0.77 ms when the discharge was introduced at the ignition delay stage. The two-stage combustion characteristics occurred when the discharge was introduced at the combustion stage. [ABSTRACT FROM AUTHOR]

Particulate matter (PM) generated from pulverized coal combustion is influenced by a process of vaporization, nucleation and condensation/reaction of semi-volatile minerals. The objective of the work is to identify the controlling mechanism of the enrichment of volatile sodium (Na). Three representative kinds of coal with different rank and mineral content, Zhundong lignite, Hami lignite and high-ash-fusion (HAF) bituminous, were burned in a 25 kW self-sustained pulverized coal combustor. Zhundong lignite, possessing medium rank among the three coal samples, exhibits the highest formation ability of ultrafine PM 0.2 , indicating a prominent effect of mineral content over that of coal rank on PM formation. The size partitioning behavior of sodium is affected by the molar ratio Na 2 O/(SiO 2 +Al 2 O 3 ) instead of the absolute Na content. With this ratio increasing from 0.27 (HAF bituminous) to 20 (Zhundong lignite), the gas-to-particle conversion of Na transits from a surface-reaction controlled process to a nucleation-condensation/coagulation dominated pathway. The Na partitioning behavior of Zhundong lignite is then quantitatively interpreted by a population-balance-based theoretical approach. The model reveals the details on the competing processes of homogeneous nucleation and vapor condensation, resulting in that the simulated final particle mass size distribution exhibits reasonable agreement with the experimental result. The model has also successfully reproduced the d p 0 dependence of Na fraction in the ultrafine size regime from a process of homogeneous nucleation under the decreasing gas temperature from 1420 to 1270 K. [ABSTRACT FROM AUTHOR]

The temporal release of K from single pinewood particles of four different initial masses (20–50 mg) in a Hencken burner was measured in-situ by laser-induced break down spectroscopy (LIBS). Particle temperature during the combustion was measured by two-color pyrometry. A calculation method was developed to determine the K release flux across the plume based on LIBS measurement and computational fluid dynamics (CFD) model. For all the 4 particle sizes, the temporal release rate of K showed a peak during devolatilization, a monotonous increase during char combustion and a monotonous decrease during the ash stage. It is found that the cumulative fraction of released K was below 4% in the devolatilization stage and ∼20% in the char combustion stage. The particle mass did not have significant influence on the K release. The mechanism of the K release during the char combustion was analyzed, based on the initial portioning of K as inorganic K and organically-bound Char-K. The inorganic K releases mainly by thermal decomposition. The Char-K releases mainly by converting first to inorganic K when the char is burnt and then by releasing to gas phase from the inorganic K. A kinetic model for the release of K during pinewood char combustion was developed on this basis. The kinetic parameters for the release of K was found to be influenced by the initial partitioning between the inorganic K and the Char-K. The correlated initial proportion of inorganic K in pinewood char was found to be 0.37, while the correlated pre-exponential factor and the activation energy for K release were 12.5 1/s and 89.9 kJ/mol, respectively. [ABSTRACT FROM AUTHOR]

This work seeks to achieve stable combustion of swirling premixed flows using gliding arc discharges and understand how plasma alters flame dynamics near blow-off. Gliding arc discharges powered by 7 kHz and 25 kHz alternating current (AC) supplies are demonstrated to significantly extend the flammability and stability limits of swirling premixed flames either confined or unconfined within a quartz tube, with time-averaged discharge power close to 1% of the power typically released by a stable flame. Side visualization of OH profiles using the planar laser-induced fluorescence (PLIF) technique illustrates the unsteady evolution of flames approaching lean blow-off, confirming the positive effect of gliding arcs on the ultra-lean residual flames. Contrary to the unconfined flame anchored near the nozzle exit or the bluff body, the confined flame detached from the wall progressively presents different unstable morphologies, e.g., oscillating flames and lifted helicon flames. The activation of gliding arcs provides the additional anchoring mechanism for the lean swirling flame and particularly suppresses the natural flame oscillation with heat-release rate fluctuations of above 30%. The mechanism controlling flame responses to gliding arc includes a nonlocal effect through the externally applied AC electric field displacing flame and a local effect of spark discharge triggering flame kernels by producing heat and active species. The external sub-breakdown electric field can enlarge the spreading angle of OH-PLIF layers in the unconfined flame, but demonstrates few positive effects on the blow-off limit. Regarding the local effects within or near the arc column, top infrared imaging provides visible evidence for flame kernels initiated upstream of the inner recirculation zone, yielding repetitive ignition in the fresh reactants. [ABSTRACT FROM AUTHOR]