Measurements of the oxidation rates of various forms of carbon (soot, graphite, coal char) have often shown an unexplained attenuation with increasing temperatures in the vicinity of 2000 K, even when accounting for diffusional transport limitations and gas-phase chemical effects (e.g. CO2 dissociation). With the development of oxy-fuel combustion approaches for pulverized coal utilization with carbon capture, high particle temperatures are readily achieved in sufficiently oxygen-enriched environments. In this work, a new semi-global intrinsic kinetics model for high temperature carbon oxidation is created by starting with a previously developed 5-step mechanism that was shown to reproduce all major known trends in carbon oxidation, except for its high temperature kinetic falloff, and incorporating a recently discovered surface oxide decomposition step. The predictions of this new model are benchmarked by deploying the kinetic model in a steady-state reacting particle code (SKIPPY) and comparing the simulated results against a carefully measured set of pulverized coal char combustion temperature measurements over a wide range of oxygen concentrations in N2 and CO2 environments. The results show that the inclusion of the spontaneous surface oxide decomposition reaction step significantly improves predictions at high particle temperatures. Furthermore, the simulations reveal that O atoms released from the oxide decomposition step enhance the radical pool in the near-surface region and within the particle interior itself. Incorporation of literature rates for O and OH reactions with the carbon surface results in a reduction in the predicted radical pool concentrations and a very minor enhancement of the overall carbon oxidation rate.
Previous research has provided strong evidence that CO2 and H2O gasification reactions can provide non-negligible contributions to the consumption rates of pulverized coal (pc) char during combustion, particularly in oxy-fuel environments. Fully quantifying the contribution of these gasification reactions has proven to be difficult, due to the dearth of knowledge of gasification rates at the elevated particle temperatures associated with typical pc char combustion processes, as well as the complex interaction of oxidation and gasification reactions. Gasification reactions tend to become more important at higher char particle temperatures (because of their high activation energy) and they tend to reduce pc oxidation due to their endothermicity (i.e. cooling effect). The work reported here attempts to quantify the influence of the gasification reaction of CO2 in a rigorous manner by combining experimental measurements of the particle temperatures and consumption rates of size-classified pc char particles in tailored oxy-fuel environments with simulations from a detailed reacting porous particle model. The results demonstrate that a specific gasification reaction rate relative to the oxidation rate (within an accuracy of approximately +/- 20% of the pre-exponential value), is consistent with the experimentally measured char particle temperatures and burnout rates in oxy-fuel combustion environments. Conversely, the results also show, in agreement with past calculations, that it is extremely difficult to construct a set of kinetics that does not substantially overpredict particle temperature increase in strongly oxygen-enriched N2 environments. This latter result is believed to result from deficiencies in standard oxidation mechanisms that fail to account for falloff in char oxidation rates at high temperatures.
Previous research has provided strong evidence that CO2 and H2O gasification reactions can provide non-negligible contributions to the consumption rates of pulverized coal (pc) char during combustion, particularly in oxy-fuel environments. Fully quantifying the contribution of these gasification reactions has proven to be difficult, due to the dearth of knowledge of gasification rates at the elevated particle temperatures associated with typical pc char combustion processes, as well as the complex interaction of oxidation and gasification reactions. Gasification reactions tend to become more important at higher char particle temperatures (because of their high activation energy) and they tend to reduce pc oxidation due to their endothermicity (i.e. cooling effect). The work reported here attempts to quantify the influence of the gasification reaction of CO2 in a rigorous manner by combining experimental measurements of the particle temperatures and consumption rates of size-classified pc char particles in tailored oxy-fuel environments with simulations from a detailed reacting porous particle model. The results demonstrate that a specific gasification reaction rate relative to the oxidation rate (within an accuracy of approximately +/- 20% of the pre-exponential value), is consistent with the experimentally measured char particle temperatures and burnout rates in oxy-fuel combustion environments. Conversely, the results also show, in agreement with past calculations, that it is extremely difficult to construct a set of kinetics that does not substantially overpredict particle temperature increase in strongly oxygen-enriched N2 environments. This latter result is believed to result from deficiencies in standard oxidation mechanisms that fail to account for falloff in char oxidation rates at high temperatures.
With the anticipated growth in hydrogen generation and use as part of a broad shift in energy use away from fossil fuels, concerns have been raised regarding the impact of increased H2 emissions on global warming. Atmospheric scientists have long recognized that H2 emissions into the atmosphere do have an indirect impact on global warming, largely because a portion of emitted H2 is consumed by the hydroxyl radical (OH), which is the primary reactant that removes the potent greenhouse gas methane from the atmosphere. Therefore, increases in H2 emissions will result in decreases in the average OH concentrations in the atmosphere and an increase in the atmospheric lifetime of methane. Various assessments of the impact of H2 emissions on global warming have been performed over the past couple of decades. These assessments have yielded significant variability and recognized uncertainty in the magnitude of the warming effect of a given quantity of emitted H2, and an even greater uncertainty in the magnitude of H2 leakage and releases that can be anticipated with an expanded H2 infrastructure. Consequently, definitive estimates of the magnitude of the warming effect of additional emitted H2 are lacking. However, given the current understanding of the warming potential of emitted H2 and given reasonable expectations of the emission rate of an expanded H2 infrastructure, it is anticipated that warming effects from emitted H2 will offset no more than 5% of the reduction in warming associated with avoided CO2 emissions from using clean H2. Further, it is highly unlikely that the warming effects from emitted H2 will offset more than 10% of the benefit from avoided CO2 emissions, at least as considered over a typical 100-year accounting period. Because of the short atmospheric lifetimes of H2 and methane, however, the warming effect of emitted H2 is enhanced over the first few years following increases in H2 emission.
Laser-induced incandescence (LII) is a widely used technique for measuring soot concentrations. For flame applications LII is frequently deployed as a planar diagnostic to measure the two-dimensional soot field. However, when the laser sheet is focused, as is typical to reach the requisite laser fluence level and achieve good spatial resolution, the complex laser power dependence of the LII signal generation process can introduce a large variation in LII signal sensitivity across an LII image. In this work, this effect is quantified for the first time as a function of laser pulse fluence, using a typical planar LII excitation scheme with a clipped Gaussian YAG laser beam focused with a 1 m focal length lens. Furthermore, the cross-sectional energy distribution in the laser sheet was measured across the image plane, to relate the details of the laser sheet focal properties with the resultant LII behavior. The results show that a unique laser fluence level (referenced to the focal plane) exists whereby there is essentially no dependence of LII signal on position relative to the focal plane. However, at lower or higher fluences, the radial signals either decrease (low fluence) or increase (high fluence) rapidly with increasing distance away from the focal point. For measurements using an LII 'plateau' laser fluence level, as is usual in environments with significant optical depth (i.e. sufficiently strong soot levels), the LII signals are found to be 2.5X larger 40 mm away from the focal point. An analysis conducted by combining a previously measured LII fluence dependence for a top-hat laser profile with the laser sheet cross-sections measured in this work shows general agreement with the measured results for LII signal variation. Further, the sensitivity of LII signals at high fluences to the laser beam spatial profile, particularly away from the sheet focus, is highlighted.
Design and analysis of practical reactors utilizing solid feedstocks rely on reaction rate parameters that are typically generated in lab-scale reactors. Evaluation of the reaction rate information often relies on assumptions of uniform temperature, velocity, and species distributions in the reactor, in lieu of detailed measurements that provide local information. This assumption might be a source of substantial error, since reactor designs can impose significant inhomogeneities, leading to data misinterpretation. Spatially resolved reactor simulations help understand the key processes within the reactor and support the identification of severe variations of temperature, velocity, and species distributions. In this work, Sandia's pressurized entrained flow reactor is modeled to identify inhomogeneities in the reaction zone. Tracer particles are tracked through the reactor to estimate the residence times and burnout ratio of introduced coal char particles in gasifying environments. The results reveal a complex mixing environment for the cool gas and particles entering the reactor along the centerline and the main high-speed hot gas reactor flow. Furthermore, the computational fluid dynamics (CFD) results show that flow asymmetries are introduced through the use of a horizontal gas pre-heating section that connects to the vertical reactor tube. Computed particle temperatures and residence times in the reactor differ substantially from the idealized plug flow conditions typically evoked in interpreting experimental measurements. Furthermore, experimental measurements and CFD analysis of heat flow through porous refractory insulation suggest that for the investigated conditions (1350 °C, <20 atm), the thermal conductivity of the insulation does not increase substantially with increasing pressure.
Combined picosecond (10-12 s) and femtosecond (10-15 s) laser pulses can give sensitive, low-noise measurements of important quantities in reacting flows, such as species concentrations and temperature. Emmanuel’s work focused on the development of an instrument for tailoring the time profile of picosecond laser pulses for use in nonlinear optical spectroscopic methods created from broad bandwidth femtosecond pulses. In addition to constructing the device, Emmanuel produced a LabView-based automation code, building off skills he developed in a previous CCI internship at Sandia.
Apparent char kinetic rates are commonly used to predict pulverized coal char burning rates. These kinetic rates quantify the char burning rate based on the temperature of the particle and the oxygen concentration at the external particle surface, inherently neglecting the impact of variations in the internal diffusion rate and penetration of oxygen. To investigate the impact of bulk gas diffusivity on these phenomena during Zone II burning conditions, experimental measurements were performed of char particle combustion temperature and burnout for a subbituminous coal burning in an optical entrained flow reactor with helium and nitrogen diluents. The combination of much higher thermal conductivity and mass diffusivity in the helium environments resulted in cooler char combustion temperatures than in equivalent N2 environments. Measured char burnout was similar in the two environments for a given bulk oxygen concentration but was approximately 60% higher in helium environments for a given char combustion temperature. To augment the experimental measurements, detailed particle simulations of the experimental conditions were conducted with the SKIPPY code. These simulations also showed a 60% higher burning rate in the helium environments for a given char particle combustion temperature. To differentiate the effect of enhanced diffusion through the external boundary layer from the effect of enhanced diffusion through the particle, additional SKIPPY simulations were conducted under selected conditions in N2 and He environments for which the temperature and concentrations of reactants (oxygen and steam) were identical on the external char surface. Under these conditions, which yield matching apparent char burning rates, the computed char burning rate for He was 50% larger, demonstrating the potential for significant errors with the apparent kinetics approach. However, for specific application to oxy-fuel combustion in CO2 environments, these results suggest the error to be as low as 3% when applying apparent char burning rates from nitrogen environments.
Knowledge of soot particle sizes is important for understanding soot formation and heat transfer in combustion environments. Soot primary particle sizes can be estimated by measuring the decay of time-resolved laser-induced incandescence (TiRe-LII) signals. Existing methods for making planar TiRe-LII measurements require either multiple cameras or time-gate sweeping with multiple laser pulses, making these techniques difficult to apply in turbulent or unsteady combustion environments. Here, we report a technique for planar soot particle sizing using a single high-sensitivity, ultra-high-speed 10 MHz camera with a 50 ns gate and no intensifier. With this method, we demonstrate measurements of background flame luminosity, prompt LII, and TiRe-LII decay signals for particle sizing in a single laser shot. The particle sizing technique is first validated in a laminar non-premixed ethylene flame. Then, the method is applied to measurements in a turbulent ethylene jet flame.
The increasing interest in gasification and oxy-fuel combustion of biomass has heightened the need for a detailed understanding of char gasification in industrially relevant environments (i.e., high temperature and high-heating rate). Despite innumerable studies previously conducted on gasification of biomass, very few have focused on such conditions. Consequently, in this study the high-temperature gasification behaviors of biomass-derived chars were investigated using non-intrusive techniques. Two biomass chars produced at a heating rate of approximately 104 K/s were subjected to two gasification environments and one oxidation environment in an entrained flow reactor equipped with an optical particle-sizing pyrometer. A coal char produced from a common U.S. low sulfur subbituminous coal was also studied for comparison. Both char and surrounding gas temperatures were precisely measured along the centerline of the furnace. Despite differences in the physical and chemical properties of the biomass chars, they exhibited rather similar reaction temperatures under all investigated conditions. On the other hand, a slightly lower particle temperature was observed in the case of coal char gasification, suggesting a higher gasification reactivity for the coal char. A comprehensive numerical model was applied to aid the understanding of the conversion of the investigated chars under gasification atmospheres. In addition, a sensitivity analysis was performed on the influence of four parameters (gas temperature, char diameter, char density, and steam concentration) on the carbon conversion rate. The results demonstrate that the gas temperature is the most important single variable influencing the gasification rate.
Apparent char kinetic rates are commonly used to predict pulverized coal char burning rates. These kinetic rates quantify the char burning rate based on the temperature of the particle and the oxygen concentration at the particle surface, thereby inherently neglecting the impact of variations in the penetration of oxygen into the char on the predicted burning rate. To investigate the impact of variable extents of penetration during Zone II burning conditions, experimental measurements were performed of char particle combustion temperature and burnout for a common U.S. subbituminous coal burning in an optical laminar entrained flow reactor with either helium or nitrogen diluents. The combination of much higher thermal conductivity and mass diffusivity in the helium environments resulted in substantially cooler char combustion temperatures than in equivalent N2 environments. Measured char burnout was similar in the two environments for a given bulk oxygen concentration but was approximately 60% higher in helium environments for a given char combustion temperature. Detailed particle simulations of the experimental conditions confirmed a 60% higher burning rate in the helium environments as a function of char temperature, whereas catalyst theory predicts that the burning rate in helium could be as high as 90% greater than in nitrogen, in the limit of large Thiele modulus (i.e. near the diffusion limit). For application combustion in CO2 environments (e.g. for oxy-fuel combustion), these results demonstrate that due to differences in oxygen diffusivity the apparent char oxidation rates will be lower, but by no more than 9% relative to burning rates measured in nitrogen environments.
Apparent char kinetic rates are commonly used to predict pulverized coal char burning rates. These kinetic rates quantify the char burning rate based on the temperature of the particle and the oxygen concentration at the particle surface, thereby inherently neglecting the impact of variations in the penetration of oxygen into the char on the predicted burning rate. To investigate the impact of variable extents of penetration during Zone II burning conditions, experimental measurements were performed of char particle combustion temperature and burnout for a common U.S. subbituminous coal burning in an optical laminar entrained flow reactor with either helium or nitrogen diluents. The combination of much higher thermal conductivity and mass diffusivity in the helium environments resulted in substantially cooler char combustion temperatures than in equivalent N2 environments. Measured char burnout was similar in the two environments for a given bulk oxygen concentration but was approximately 60% higher in helium environments for a given char combustion temperature. Detailed particle simulations of the experimental conditions confirmed a 60% higher burning rate in the helium environments as a function of char temperature, whereas catalyst theory predicts that the burning rate in helium could be as high as 90% greater than in nitrogen, in the limit of large Thiele modulus (i.e. near the diffusion limit). For application combustion in CO2 environments (e.g. for oxy-fuel combustion), these results demonstrate that due to differences in oxygen diffusivity the apparent char oxidation rates will be lower, but by no more than 9% relative to burning rates measured in nitrogen environments.
The Reacting Particle and Boundary Layer (RPBL) model computes the transient-state conditions for a spherical, reacting, porous char particle and its reacting boundary layer. RPBL computes the transport of gaseous species with a Maxwell-Stefan multicomponent approach. Mass transfer diffusion coefficients are corrected to account for a non-stagnant bulk flow condition using a factor based on the Sherwood number. The homogeneous gas phase reactions are modeled with a syngas mechanism, and the heterogeneous reactions are calculated with a six-step reaction mechanism. Both homogeneous and heterogeneous reaction mechanisms are implemented in Cantera. Carbon density (burnout) is computed using the Bhatia and Perlmutter model to estimate the evolution of the specific surface area. Energy equations are solved for the gas temperature and the particle temperature. The physical properties of the particle are computed from the fractions of ash, carbon, and voids in the particle. The void fraction is computed assuming a constant diameter particle during the reaction process. RPBL solves a particle momentum equation in order to estimate the position of the particle in a specific reactor. We performed a validation and uncertainty quantification study with RPBL using experimental char oxidation data obtained in an optically accessible, laminar, entrained flow reactor at Sandia National Laboratories. We used a consistency analysis to compare RPBL and experimental data (with its associated uncertainty) for three coal chars over a range of particle sizes. We found consistency for particle temperature and velocity across all experiments.
The proper consideration of the radiation correction for bare-wire thermocouple measurements requires consideration of the convective and radiative heat transfer of the thermocouple with its surroundings, as well as conductive heat transfer between the thermocouple bead and the connecting thermocouple wires. This has rarely been considered in the past, and to do so has involved complex simulation of the complete thermocouple energy balance. This paper reports on a new, easy-to-implement approach for calculating the proper radiant correction for thermocouples, subject to uncertainties associated with the relevant thermocouple and gas properties and limitations to characterizing convective heat transfer to the thermocouple bead and wires via standard correlations. Examples of the radiation correction computed with this new method as a function of temperature and bead and wire size are given, and are compared with traditional approaches considering heat transfer around either the thermocouple bead or the thermocouple wire.
A non-premixed coannular burner in which oxygen concentrations and oxidizer flow rates can be independently varied was developed to investigate the effects of turbulent mixing intensity from oxygen enrichment on soot formation and flame radiation. Local radiation intensities soot concentrations and soot temperatures were measured using a thin-film thermopile planar laser-induced incandescence and two-color imaging pyrometry respectively. The measurements showed that soot formation increased as the oxygen concentration decreases from 100% to 50% helping to moderate a decrease in overall flame radiation. An increase in turbulence intensity had a remarkable effect on flame height soot formation and thermal radiation resulting to decreases in all of these parameters. The soot temperature decreased with a decrease in the oxygen concentration and increased with an increase in turbulent mixing intensity. Thus properly designed oxygen-enriched burners that enhance soot formation for intermediate levels of oxygen purity may be able to achieve thermal radiation intensities as high as 85% of traditional oxy-fuel burners utilizing high-purity oxygen.
Ignitability is important to characterize pulverized coal combustion, as it is directly related to flame stability. The current work describes a practical test rig for rapid laboratory analysis of pulverized coal cloud ignition properties. The system has been designed for conventional coal combustion conditions using air as the oxidant. In the current work, the measurement principle of the device is described and its adaption to and applicability for oxy-fuel combustion tests is demonstrated. Four coals with different rank were measured in air and in oxy-fuel atmospheres containing 20–35 vol% O2 in CO2. The major influencing factors for the investigated samples were found to be the coal rank and the gas-phase oxygen concentration, while a minor influence of particle size was observed.
In this formation of nano-particles during coal char combustion, the vaporization of inorganic components in char and the subsequent homogeneous particle nucleation, heterogeneous condensation, coagulation, and coalescence play decisive roles. Furthermore, conventional measurements cannot provide detailed information on the dynamics of nano-particle formation and evolution, In this study, a sophisticated intrinsic char kinetics model that considers ash effects (including ash film formation, ash dilution, and ash vaporization acting in tandem), both oxidation and gasification by CO2 and H2O, homogeneous particle nucleation, heterogeneous vapor condensation, coagulation, and and coalescence mechanisms is developed and used to compare the temporal evolution of the number and size of nano-particles during coal char particle combustion as a function of char particle size, ash content, and oxygen content in O2/N2 and O2/CO2 atmospheres .
Truly quantifying soot concentrations within turbulent flames is a difficult prospect. Laser extinction measurements are constrained by spatial resolution limitations and by uncertainty in the local soot extinction coefficient. Laser-induced incandescence (LII) measurements rely on calibration against extinction and thereby are plagued by uncertainty in the extinction coefficient. In addition, the LII measurements are subject to signal trapping in flames with significant soot concentrations and/or flame widths. In the study reported here, a turbulent ethylene non-premixed jet flame (jet exit Reynolds number of 20,000) is investigated by a combination of LII and full-flame HeNe laser (633 nm) extinction measurements. The LII measurements have been calibrated against extinction measurements in a laminar ethylene flame. An extinction coefficient previously measured in laminar ethylene flames is used as the basis of the calibration. The time-Averaged LII data in the turbulent flame has been corrected for signal trapping, which is shown to be significant in this flame, and then the line-of-sight extinction for a theoretical 633 nm light source has been calculated acrob the LII-determined soot concentration field. Comparison of the LII-based extinction with that actual measured along the flame centerline is favorable, showing an average deviation of approximately 10%. This lends credence to the measured values of soot concentrations in the flame and also gives a good indication of the level of uncertainty in the measured soot concentrations, subject to the additional uncertainty in the previously measured extinction coefficient, estimated to be ±15%.
Oxy-fuel combustion is a well-known approach to improve the heat transfer associated with stationary energy processes. Its overall penetration into industrial and power markets is constrained by the high cost of existing air separation technologies for generating oxygen. Cryogenic air separation is the most widely used technology for generating oxygen but is complex and expensive. Pressure swing adsorption is a competing technology that uses activated carbon, zeolites and polymer membranes for gas separations. However, it is expensive and limited to moderate purity O2 . MOFs are cutting edge materials for gas separations at ambient pressure and room temperature, potentially revolutionizing the PSA process and providing dramatic process efficiency improvements through oxy-fuel combustion. This LDRD combined (1) MOF synthesis, (2) gas sorption testing, (3) MD simulations and crystallography of gas siting in pores for structure-property relationship, (4) combustion testing and (5) technoeconomic analysis to aid in real-world implementation.