Pershing, DW

2000

Measurement of Soot and Char in Pulverized Coal Fly Ash

Veranth, J.M.; Fletcher, T.H.; Pershing, D.W. and Sarofim, A.F.
Fuel, 79(9), 1067-1075 (2000).
Contact: Veranth

1999

Char Nitrogen Conversion: Implications to Emissions from Coal-Fired Utility Boilers

Molina, A.; Sarofim, A.F.; Eddings, E.G. and Pershing, D.W.
Progress in Energy and Combustion Science, to be submitted, November 1999.

The contribution of the nitrogen present in the char on the production of nitrogen oxides during char combustion was analyzed. A literature review summarized the current understanding of the mechanisms that account for the formation of NO and N2O from the nitrogen present in the char. The review focused on 1) The functionalities in which nitrogen is present in the coal and how they evolve during coal devolatilization; 2) The mechanism of nitrogen release from the char to the homogeneous phase and its further oxidation to NO; and 3) The reduction of NO on the surface of the char. The critical analysis of these three issues allowed to identify uncertainties and well-founded conclusions observed in the literature for this system.

The existent models for the production of nitrogen oxides from char-N were also reviewed. A critic analysis of the assumptions made in these models and how they affect the final predictions is presented. Finally, a simplified version of these models was used to perform a parametric analysis of the incidence of the rate of NO reduction on the char surface, the rate of carbon oxidation, and the instant during the char oxidation when the nitrogen is released; on the total conversion of char-N to NO. The results underscore the importance of the reaction of NO reduction on the char surface on the final conversion of char-N to NO.

Measurement of Soot and Char in Pulverized Coal Fly Ash

Veranth, J.M.; Fletcher, T.H.; Pershing, D.W. and Sarofim, A.F.
Fuel, In Review, 1999.

The unburned carbon in the fly ash produced by low-NOx pulverized coal combustion has been shown by electron microscopy to be a mixture of porous coal char particles and aggregates of submicron particles, which are thought to be soot. The carbon is bimodally distributed with large soot aggregates mixed with the char in the particles larger than 10 microns and dispersed soot found with the submicron particles. A method for determining the mass of soot and char by liquid-suspension gravity separation was used with both laboratory-scale and power plant fly ash samples. For low-NOx, staged, pilot-scale combustion of bituminous coal the soot in the soot in the furnace exit ash was estimated to be 0.2 to 0.6% of the fuel carbon, which was about 35% of the total unburned carbon.

1997

Numerical Modeling of the Temperature Distribution in a Commercial Hazardous Waste Slagging Rotary Kiln

Veranth, J.M.; Silcox, G.D. and Pershing, D.W.
Environmental Science and Technology, 31:2534-539(1997). Funded by ACERC.

The gas, wall and bed temperatures in a hazardous wasted incineration kiln were studied using a commercially available, CFD-based, r4eacting flow code, which included radiation heat transfer. The model was compared to field measurements make on a co-current flow, 35 MW slagging rotary kiln. Cases were run to determine the sensitivity of the predictions to changes in the model assumptions and to simulate the normal variation in combustion inputs. The model predictions of the peak bed temperature, of the axial temperature profile, and of the gas temperature at the exit-plane were consistent with the measurement at a full-scale wasted incinerator during normal operation. The model and the field observations both indicate that the peak bed temperature occurs near the middle of the kiln and that the difference between the peak bed temperature and the exit-plane gas temperature depends on the inlet flow. The geometry of the transition between the kiln and the secondary combustion chamber and the fuel-to-air equivalence ratio have the greatest effect on the calculated temperature distribution. Modeling studies provide useful information such as the relationship between available measurements and the temperature at inaccessible locations inside a full-scale kiln.

Waste Incineration for Resource Recovery in a Regenerative Life Support System

Brouwer, J.; Kemp, G.; Heap, M.P.; Lighty, J.S.; Burton, B.; Sirdeshpande, A.; Inkley, D.; Pershing, D.W.; Fisher, J. and Pisharody, S.
Western States Section of the Combustion Institute, Spring 1997

For the last two years, the University of Utah and Reaction Engineering International, in cooperation with Ames Research Center, have been developing a waste incineration system for regenerative life support systems. The system is designed to burn inedible plant biomass and human waste. The exhaust gas is currently designed to recycle back to the plant growth chamber and will eventually be recycled to the human chamber after passing through a Trace Contaminant Control System. The incineration system, a fluidized bed reactor, has been designed for a 4-person mission. This paper will detail the design of the components of this system. In addition, results will be presented from testing at the University of Utah. Presently, the unit has been shipped to Ames Research Center for more tests prior to delivery to Johnson Space Center for testing in a 90-day, 4-person test.

Waste Incineration for Resource Recovery in a Bioregenerative Life Support System

Lighty, J.S.; Burton, B.; Sirdeshpande, A.; Inkley, D.; Pershing, D.W.; Brouwer, J.; Kemp, G.; Heap, M.P.; Fisher, J. and Pisharody, S.
27th International Conference on Environmental Systems, Lake Tahoe, Nevada, July 14-17, 1997

For the last two years, the University of Utah and Reaction Engineering International, in cooperation with NASA Ames Research Center (ARC), have been developing a waste incineration system for regenerative life support systems. The system is designed to burn inedible plant biomass and human waste. The goal is to obtain an exhaust gas clean enough to recycle to either the plant or human habitats. The incineration system, a fluidized bed reactor, has been designed for a 4-person mission. This paper will detail the design of the units. In addition, results will be presented from testing at the University of Utah. Presently, the unit has been shipped to Ames Research Center for more tests prior to delivery to Johnson Space Center for testing in a 90-day, 4-person test.

1996

The Formation and Control of NOx Emissions from Biomass Combustion

Pershing, D.W.; Lighty, J.S.; Harding, S.N.; Brouwer, J.; Heap, M.P.; Munro, J.M. and Winter, R.M.
Proceedings from the Finnish-Swedish Flame Days, Naantali, Finland, September, 1996. Funded by Environmental Protection Agency, US Department of Energy and National Science Foundation.

Biomass fuels account for a significant fraction of the worldwide energy usage. In 1990 consumption is believed to have exceeded 13 quads according to Tillman (1991). This does not include the combustion of peat that is known to be widely used in some northern European countries and in parts of the former Soviet Union. Biomass energy consumption is also a significant fraction of total energy usage in many developing nations. Hence, emissions from combustion of biofuels are of interest to the environmental community.

Most biomass materials generally produce lower NOx emissions when they are burned than their fossil fuel counterparts, but under certain conditions NOx emissions can be significant. Biomass fuels usually contain relatively limited amounts of organic nitrogen (often 0.1 ± 0.1%) so NOx formation from fuel nitrogen is generally small (except in the case of peat and some plant wastes). Biofuels also tend to burn at cooler combustion temperatures (due to higher moisture contents and the presence of oxygen in the fuel structure), which tends to minimize the formation of NOx by the thermal mechanism (the high temperature fixation of N2 in the combustion air).

The purpose of this paper is review the available information on the formation and the control of NOx emissions during the combustion of biofuels alone and in combination with fossil fuels and/or wastes. Both laboratory and pilot scale studies have been included, as well as full scale field test results where they are generally available.

1994

Emissions of WDF Kilns Evaluated

Pershing, D.W.
Rock Products, Cement Edition, 19-23, April 1994. (Also presented at the Waste Combustion in Boilers and Industrial Furnaces, Kansas City, MO, April 1994.) Funded by Industrial Consortium.

A recent report by objective experts studying the emissions of cement kilns using waste-derived fuel (WDF) shows that under proper operating conditions that use of the fuel in cement kilns can provide a safe and cost effective disposal method for combustible wastes, as well as a net reduction of CO2 emissions.

The study, conducted by a Scientific Advisory Board (SAB) for the Cement Kiln Recycling Coalition (CKRC), was in response to inquiries about the effectiveness of cement kilns for waste disposal. The study also found this reduction in CO2 emissions is relative to incineration since the WDF is replacing a fossil fuel in the cement kilns using raw materials with inherently high hydrocarbon levels, the emission of organics can be controlled by applying well established, good combustion practices. Because of the high alkali content of the raw materials and the scrubbing action of the cement caused by its abrasiveness, metals emissions are generally below 1% of the input, except in the case of volatile metals such as mercury.

The SAB report, entitled "Evaluation of the Origin, Emissions, and Control of Organic and Metal Compounds from Cement Kilns Cofired With Hazardous Wastes," focused on organic and metal emissions from cement kilns using WDF. The following subsections of this article, which is an examination of the findings of the study, briefly discuss the findings of the SAB regarding the fate of metals and organics in these systems.

1993

Thermal Treatment of Hazardous Wastes: A Comparison of Fluidized Bed and Rotary Kiln Incineration

Rink, K.K.; Larsen, F.S.; Kozinski, J.A.; Lighty, J.S.; Silcox, G.D. and Pershing, D.W.
Energy & Fuels, 7 (6):803-814, 1993. Funded by ACERC.

Large volumes of sludge are produced by a wide variety of industrial processes and by municipal wastewater treatment. Interest in incinerating these sludges, either alone, or co-fired with other fuels, is increasing. The issues surrounding sludge incineration in rotary kilns and fluidized beds were identified through a series of pilot-scale tests using two slightly different paper mill sludges. The specific issues examined include hydrocarbon emissions, NOx emissions, and bottom and fly ash properties. A 61-cm i.d. X 61-cm long, 130-kW pilot-scale rotary kiln simulator (RKS) and a 23-cm i.d., 300-kW circulating fluidized bed combustor (CFB) were maintained at a nominal temperature of 1100 K and a stoichiometric ratio of 1.5. The rotary kiln was fed in a batch mode in order to simulate the passage of solids through a kiln. The fluidized bed was fed in both batch and continuous modes. Samples were removed from the kiln (bottom ash) and transition section (fly ash). Samples of the fluidized bed materials were removed from the bed (bottom ash) and after the cyclone (fly ash). The exhaust gases were analyzed continuous for hydrocarbons, CO, O2, NO, and CO2. This paper presents data on these analyses as well as NO conversion and ash properties. The production of NO in the RKS was dependent on the supply of nitrogen (in the sludge) and oxygen (in the gas phase), in the reactor. The availability of oxygen to the sludge was affected by the particle diameter of the sludge, the charge size, and whether a solids bed was present at the time of the incineration. In the CFB, the nitrogen-containing compounds were oxidized primarily downstream of the feedboard region, resulting in elevated levels of NO in the transition and cyclone regions. Carbon monoxide concentrations were high immediately above the bed, which led to the reduction of NO inside the freeboard zone. In both the CFB and RKS tests little unburned hydrocarbons were present in the exhaust gas streams. Formation of fly ash particles was dependent on types of incinerated material (sludge; mixture of sludge and silica sand). Bottom ash material resembled randomly organized skeletons (or cenospheric skeletons), the structure of which was independent of the type of sludge or reactor. Smaller fly ash and bottom ash particles were formed during CFB incineration experiments.

1992

Solid Waste Incineration in Rotary Kilns

Pershing, D.W.; Lighty, J.S.; Silcox, G.D.; Heap, M.P. and Owens, W.D.
Combustion Science and Technology, 1992 (in press). (Previously presented at the First International Conference on Combustion Technologies for a Clean Environment, Vilamoura, Portugal, September 1991.) Funded by the National Science Foundation, Environmental Protection Agency, Gas Research Institute and ACERC.

Rotary kilns are used to dispose of many solid wastes and sludges and to thermally treat contaminated soils. In this communication the fates of hydrocarbon and metal species are examined with a view toward optimization of new kiln designs and maximizing existing unit throughout while minimizing pollutant emissions. Initially, process fundamentals are considered to characterize the controlling phenomena. Pilot- and large-scale data are then examined to define practical system complexities. Finally, techniques for data scale-up and performance prediction are summarized. Temperature is clearly the most important parameter with respect to the fate of both metal and hydrocarbon species; hence, heat transfer is often rate limiting. High temperatures favor hydrocarbon evolution, but can also enhance the formation of toxic metal fumes. Both the solid composition and the moisture content can significantly influence the time at temperature required for hydrocarbon destruction and metal vaporization.

Improving bed mixing helps contaminant release but can also aggravate puffing tendencies with batch charging. Full-scale performance predictions currently require a combination of small-scale data and computer modeling. Future work needs to focus on verification of large-scale predictions for complex mixtures and sludges so that expensive trial burns can be minimized.

The Desorption of Toluene from a Montmorillonite Clay Adsorbent in a Rotary Kiln Environment

Owens, W.D.; Silcox, G.D.; Lighty, J.S.; Deng, X.-X.; Pershing, D.W.; Cundy, V.A.; Leger, C.B. and Jakway, A.J.
Journal of Air Waste Management Assoc., 42:681-690, 1992. Funded by US Environmental Protection Agency, Gas Research Institute and ACERC.

The vaporization of toluene from pre-dried, 6mm montmorillonite clay particles was studied in a 130 kW pilot-scale rotary kiln with inside dimensions of 0.61 by 0.61 meters. Vaporization rates were obtained with a toluene weight fraction of 0.25 percent as a function of kiln fill fractions from 3 to 8 percent, rotation rates from 0.1 to 0.9 rpm, and kiln wall temperatures from 189 to 793ºC. Toluene desorption rates were obtained from gas-phase measurements and interpreted using a desorption model that incorporates the slumping frequency of the solids. Fill fraction of the kiln, binary gas diffusion in the bed, and particle desorption using an Arrhenius-type expression that is a function of bed temperature and average bed concentration. The model included three adjustable desorption parameters which were obtained by fitting the experimental data with a least squares technique. Solid and kiln-wall temperatures were continuously recorded and used by the model to perform toluene desorption predictions. The model was successful at predicting the effects of fill fraction and rotation rate over a range of temperatures. It was shown that an increase in kiln temperature and rotation rate increased toluene desorption rates. A decrease in kiln fill fraction showed an increase n desorption rate. Desorption predictions were performed using both predicted and measured temperature profiles. In addition, the model was used to perform sensitivity tests examine the relative importance of bed diffusion and particle desorption resistances. A methodology for predicting full-scale performance was developed. A full-scale, rotary-kiln heat-transfer model was used to estimate the bed thermal profile. This profile was then utilized by the model tp predict toluene desorption at full-scale.

1991

Fuel Rich Sulfur Capture in a Combustion Environment

Lindgren, E.R.; Pershing, D.W.; Kirchgessner, D.A. and Drehmel, D.C.
Environmental Science and Technology, 1991 (in press). Funded by Environmental Protection Agency.

A refractory-lined, natural gas furnace was used to study the fuel rich sulfur capture reactions of calcium sorbents under typical combustion conditions. The fuel rich sulfur species hydrogen sulfide (H2S) and carbonyl sulfide (COS) were monitored in a nearly continuous fashion using a gas chromatograph equipped with a flame photometric detector and an automatic sampling system which sampled every 30 seconds. Below the fuel rich zone, 25 percent excess air was added, and the ultimate fuel lean capture was simultaneously measured using a continuous sulfur dioxide (SO2) monitor. Under fuel rich conditions high levels of sulfur capture were obtained, and calcium utilization increased with sulfur concentration. The ultimate lean capture was found to be weakly dependent on sulfur concentration and independent of the sulfur capture level obtained in the fuel rich zone.

Rotary Kiln Incineration: Comparison and Scaling of Field-Scale Contaminant Evolution Rates from Sorbent Beds

Lester, T.W.; Cundy, V.A.; Sterling, A.M.; Montestruc, A.N.; Jakway, A.J.; Lu, C.; Leger, C.B.; Pershing D.W.; Lighty, J.S.; Silcox, G.D. and Owens, W.D.
Environmental Science Technology, 25:1142-1152, 1991. Funded by Environmental Protection Agency, Louisiana State University/Hazardous Waste Research Center and ACERC.

A comparison is made, for the first time, between the evolution of hydrocarbons from clay sorbent beds in a field-scale rotary kiln incinerator and in a pilot-scale rotary kiln simulator. To relate the data from the different sized units, due allowance is made for bed dynamical similitude, bed geometrical factors, and bed heat-up. To minimize the effects of disturbances caused by foreign matter in the field scale bed and differences in loading techniques, the rate of evolution is characterized by an "evolution interval" defined as the time required for the middle 80% of the ultimate containment evolution to occur. A comparison of evolution intervals with reciprocal bed temperature reveals that the data are consistent with an analysis that assumes a uniform bed temperature (at any instant of time) and desorption controlled evolution rate. Furthermore, the evolution intervals scale inversely with a modified Froude Number, which characterizes bed dynamics. The success in comparing field and simulator results indicates that pilot scale rotary kilns may be used to simulate certain features of industrial-scale units if dynamical, geometrical and thermal parameters are matched appropriately.

Thermal Analysis of Rotary Kiln Incineration: Comparison of Theory and Experiment

Owens, W.D.; Silcox, G.D.; Lighty, J.S.; Deng, X.-X.; Pershing D.W.; Cundy, V.A.; Leger, C.B. and Jakway, A.J.
Combustion and Flame, 86:101-114, 1991. Funded by Gas Research Institute, ACERC and Louisiana State University/Hazardous Waste Research Center.

A comprehensive heat-transfer model and associated simplified scaling laws are developed and verified using a pilot-scale, directly fired rotary kiln with a slumping bed of dry or wet, 6-mm clay sorbent particles. The kiln operating conditions examined include: rotation rate (0.1 to 0.9 rpm), percent fill fraction (3-8), feed moisture content (0-20 wt.%), and inner-wall temperature (190º to 790ºC). The model is used to determine the relative importance of several heat-transfer mechanisms including radiation, gas-to-solid convection, and wall-to-solid convection. Simple scaling laws are also developed for water vaporization. Generally good agreement is obtained between theory and experiment without adjusting any model parameters. Further, the simplified scaling laws provide a reasonable estimate of the pilot scale performance. The key conclusions of this study for kilns at the conditions examined are (1) water exerts a profound effect on the solids thermal profile, (2) simple geometrical scaling is not sufficient, (3) the assumption of a well mixed (radially isothermal) solids bed for the heat transfer analysis is appropriate, (4) a dimensionless group, which is a function of temperature, can be defined giving the relative importance of radiative and convective modes of heat transfer, and (5) moisture vaporization rates can be roughly approximated by assuming that the water vaporizes at the boiling point at a rate controlled by the rate of heat transfer to the bed. The implications of the scaling laws for scale-up and kiln design are also examined.

Solid Waste Incineration in Rotary Kilns

Pershing, D.W.; Lighty, J.S.; Silcox, G.D.; Heap, M.P. and Owens, W.D.
First International Conference on Combustion Technologies for a Clean Environment, Vilamoura, Portugal, September 1991. Funded by the National Science Foundation, Environmental Protection Agency, Gas Research Institute and ACERC.

Rotary kilns are used to dispose of many solid wastes and sludges and to thermally treat contaminated soils. In this communication the fates of hydrocarbon and metal species are examined with a view toward optimization of new kiln designs and maximizing existing unit throughout while minimizing pollutant emissions. Initially, process fundamentals are considered to characterize the controlling phenomena. Pilot- and large-scale data are then examined to define practical system complexities. Finally, techniques for data scale-up and performance prediction are summarized. Temperature is clearly the most important parameter with respect to the fate of both metal and hydrocarbon species; hence, heat transfer is often rate limiting. High temperatures favor hydrocarbon evolution, but can also enhance the formation of toxic metal fumes. Both the solid composition and the moisture content can significantly influence the time at temperature required for hydrocarbon destruction and metal vaporization.

Improving bed mixing helps contaminant release but can also aggravate puffing tendencies with batch charging. Full-scale performance predictions currently require a combination of small-scale data and computer modeling. Future work needs to focus on verification of large-scale predictions for complex mixtures and sludges so that expensive trial burns can be minimized.

1990

Fundamentals for the Thermal Remediation of Contaminated Soils - Particle and Bed Desorption Models

Lighty, J.S.; Silcox, G.D.; Pershing, D.W.; Cundy, V.A. and Linz, D.G.
Environmental Science & Technology, 24 (5), 750-757, 1990. Funded by the Gas Research Institute, Dave Linz, Project Manager, NSF - Presidential Young Investigator Program, ACERC (National Science Foundation and Associates and Affiliates), the State of Utah, and US Department of Energy.

A major research effort has been initiated to characterize the rate-controlling processes associated with the evolution of hazardous materials from soils. A threefold experimental approach is being used in conjunction with computer modeling to analyze thermal desorption of contaminants. Phenomena occurring both inside particles (intraparticle) and with a bed of particles (interparticle) were studied.

The results obtained suggest that the most important process variable are local thermal environment and gas-phase contaminant concentration because the adsorption equilibrium characteristics of the contaminant/soil pair control the desorption of contaminant from a particle at a given temperature. A mass-transfer/desorption model, which assumes gas/solid equilibrium at all points and time, is proposed and the model was found to predict the measured temperature dependence.

Mathematical and Physical Modeling of Rotary Kilns with Applications to Scaling and Design

Silcox, G.D.; Larsen, F.S. and Pershing, D.W.
Combustion Science and Technology, 1990 (In press). Funded by ACERC and Gas Research Institute.

The heat and mass transfer in an indirectly fired rotary kiln are examined using a combination of physical and mathematical modeling. The physical modeling is used to determine characteristic mixing times in a slumping kiln bed. These times are compared with characteristic times for diffusion of heat and mass through the bed in order to justify a lumped capacitance analysis of heat and mass transfer.

Mathematical models of heat and mass transfer are used to examine the effects of design and operating parameters on bed temperature and desorption rates. Limited comparisons with temperature measurements are presented. The design and operating parameters studied include kiln length, solid residence time, solid feed rate, and feed moisture content. The effects of moisture are particularly important to both heat and mass transfer. Scaling considerations are examined and it is shown that maintaining equivalent wall temperature profiles, fill fractions, moisture levels, and burden residence times does not necessarily result in equivalent bed thermal profiles.

The Effects of Rotary Kiln Operating Conditions and Design on Burden Heating Rates as Determined by a Mathematical Model of Rotary Kiln Heat Transfer

Silcox, G.D. and Pershing, D. W.
J. Air Waste Management Association, 40, 1990. Funded by ACERC.

A mathematical model of heat transfer in a directly-fired rotary kiln is developed and used to examine the effects of operating and design parameters on burden temperature. The model includes a mean beam length radiation model and axial zoning. Conductive and convective heat transfer are also included. Radiation between immediately adjacent zones is permitted. Calculation of heat transfer rates is facilitated by the use of an electric circuit analogue. An iterative solution procedure is adopted to solve the energy balance equations.

At the conditions examined, the model predicts that coflowing gas and solid streams result in higher average burden temperatures than do counter flowing streams. The moisture level of the feed is predicted to be a key operating parameter. The effects of kiln length, burden residence time, firing rate, and flame length are also examined.

High-Temperature, Short Time Sulfation of Calcium-Based Sorbents: I. Theoretical Sulfaction Model

Milne, C.R.; Silcox, G.D. and Pershing, D.W.
I & EC Research, 29, 2192, 1990. Funded by Environmental Protection Agency and ACERC.

A mathematical model for the sulfation of CaO is developed around the overlapping grain concept employed in the calcination and sintering models of Milne et al. (1988). The potential influence of high mass-transfer rates from simultaneous calcination of CaCO3 or Ca(OH)2 is incorporated in the mass-transfer coefficient for SO2 diffusion to the particle. A solution scheme for the nonlinear differential equation governing pore diffusion with changing particle structure is developed. The influence of grain overlap on product-layer diffusion is quantified. The model predictions show good agreement with the differential reactor data of Borgwardt and Bruce (1986) that include the influences of surface area, temperature, and SO2 partial pressure.

High-Temperature, Short Time Sulfation of Calcium-Based Sorbents: II. Experimental Data and Theoretical Model Predictions

Milne, C.R.; Silcox, G.D. and Pershing, D.W.
I & EC Research, 29, 2202, 1990. Funded by Environmental Protection Agency and ACERC.

The fundamental processes for injection of CaCO3 and Ca(OH)2 for the removal of SO2 from combustion gases of coal-fired boilers are analyzed on the basis of experimental data and a comprehensive theoretical model. Sulfation data were obtained in a 30-KW isothermal dispersed-phase reactor at conditions simulating those of upper-furnace injection. The theoretical model accounts for particle structure, calcination, sintering, sulfation, and heat and mass transfer. Pore diffusion, product-layer diffusion, and sintering appear to be the principle processes that govern the rate of SO2 capture for the hydrate particles of interest for commercial dry sorbent injection.

Calcination and Sintering Models for Application to High-Temperature, Short Time Sulfation of Calcium-Based Sorbents

Milne, C.R.; Silcox, G.D. and Pershing, D.W.
I & EC Research, 29, 139, 1990. Funded by Environmental Protection Agency and ACERC.

To simulate the staged availability of transient high surface area CaO observed in high-temperature flow-reactor data, the rate of calcination of CaCO3 of Ca(OH)2 is described by an empirical modification of the shrinking-core model. The physical model depicts particle decomposition by the shrinking-core mechanism. The subsequent time dependent decrease in CaO reactivity (surface area and porosity) due to sintering is simulated by reducing the grain-center spacing for the matrix of overlapping CaO grains. Information from SEM micrographs and from other physical property measurements of the porous particles is incorporated. This submodel simulates the time dependent availability and reactivity of CaO for a comprehensive model used to study sulfation of CaCO3 and Ca(OH)2 particles at upper-furnace injection conditions.

Rate Limiting Processes in the Rotary-Kiln Incineration of Contaminated Solids

Lighty, J.S.; Eddings, E.G.; Lindgren, E.R.; Deng, X.-X.; Pershing, D.W.; Winter, R.M. and McClennen, W.H.
Combustion Science and Technology, 1990 (In press). Funded by ACERC, Gas Research Institute and Remediation Technology.

A study of transport processes during the desorption of organic and metallic contaminants from solids is being conducted using several fundamental experiments. This paper presents results from three experimental systems, a Particle-Characterization Reactor, Bed-Characterization Reactor, and Metals Reactor.

The organic experiments attempt to identify the controlling transports processes within a particle of soil and through a bed of particles, as well as quantify the necessary parameters to model these processes. Gas and solid-phase speciation data for field samples, soils contaminated with a variety of organics (boiling points from 220ºC to 495ºC), are discussed. The data suggest that local temperature and gas/solid contacting are important in the desorption process. As expected, lighter components desorb faster than the heavier hydrocarbons. Moisture content was also important in the desorption of contaminant.

The metals reactor has been used to determine the fate of metals, specifically the partitioning between the gas and solid, for a metal species on an inert solid matrix. Data from a parametric characterization study of partitioning of lead, in the form of lead oxide on an inert matrix, in different gas environments are presented. The results indicate that lead is most volatile in a dilute hydrogen chloride/nitrogen environment.

Rotary Kiln Incineration: Comparison of Field and Pilot Scale Measurements of Contaminant Evolution Rates from Sorbent Beds

Lester, T.W.; Cundy, V.A.; Sterling, A.M.; Montestruc, A.N.; Jakway, A.J.; Leger, C.B.; Pershing D.W.; Lighty, J.S.; Silcox, G.D. and Owens, W.D.
Environmental Science Technology, 1990. Funded by Environmental Protection Agency and Louisiana State University Hazardous Waste Research Center.

A comparison is made, for the first time, between the evolution of hydrocarbons from sorbent beds in an industrial rotary kiln incinerator and in a laboratory scale rotary kiln simulator. To relate the data from the different sized units, due allowance is made for bed dynamical similitude, bed geometrical factors, and bed heat-up. To minimize the effects of disturbances caused by foreign matter in the full scale bed and differences in loading techniques, the rate of evolution is characterized by an "evolution interval," that is defined as the time required for total hydrocarbon evolution at the maximum evolution rate. A comparison of evolution intervals with reciprocal bed temperature reveals that the data are consistent with an analysis that assumes a uniform bed temperature (at any instant of time) and desorption control of the evolution rate. Furthermore, the evolution intervals scale inversely with modified Froude Number, which characterizes bed dynamics. The success in comparing field and simulator results indicates that pilot scale rotary kilns may be used to simulate certain features of industrial scale units if appropriate care is taken in matching dynamical, geometrical and thermal parameters.

Thermal Analysis of Rotary Kiln Incineration: Comparison of Theory and Experiment

Owens, W.D.; Silcox, G.D.; Lighty, J.S.; Deng, X.-X.; Pershing D.W.; Cundy, V.A.; Leger, C.B. and Jakway, A.J.
Combustion and Flame, 1990. Funded by Gas Research Institute, ACERC and Louisiana State University Hazardous Waste Research Center.

A comprehensive heat-transfer model and associated simplified scaling laws are developed and verified using a pilot-scale, directly-fired rotary kiln with a slumping bed of dry or wet, 2 mm clay sorbent particles. The kiln operating conditions examined include: rotation rate (0.1 to 0.9 rpm), percent fill fraction (3 to 8), feed moisture content (0 to 20 wt. percent), and inner-wall temperature (190 to 790º). The model is used to determine the relative importance of several heat-transfer mechanisms including radiation, gas-to-solid convection, and wall-to-solid convection. Simple scaling laws are also developed for water vaporization. Generally good agreement is obtained between theory and experiment without adjusting any model parameters. Further, the simplified scaling laws provide a reasonable estimate of the pilot scale performance.

The key conclusions of this study for kilns at the conditions examined are: (1) water exerts a profound effect on the solids thermal profile, (2) simple geometrical scaling is not sufficient, (3) the assumption of a well mixed (radially isothermal) solids bed for the heat transfer analysis is appropriate, (4) a dimensionless group, which is a function of temperature, can be defined giving the relative importance of radiative and convective modes of heat transfer, and (5) moisture vaporization rates can be roughly approximated by assuming that the water vaporized at the boiling point at a rate controlled by the rate of heat transfer to the bed. The implications of the scaling laws for scale-up and kiln design are also examined.

On-Line GC/MS Sampling of Exhaust Gas from a Rotary Kiln Simulator

Lighty, J.S.; Wagner, D.; Deng, X.-X.; Pershing, D.W.; McClennen, W.H.; Sheya, S.A.N.; Arnold, N.S. and Meuzelaar, H.L.C.
AWMA Specialty Conference on Waste Combustion in Boilers and Industrial Furnaces, Kansas City, MO, 1990. Funded by ACERC.

An on-line, short-column gas chromatography/mass spectrometry (GC/MS) system has been used to monitor the evolution of trace amounts of hydrocarbons evolving from a material which has been combusted in a rotary-kiln simulator. The system uses the isothermally heated, 1-m long transfer line of an Ion Trap Detector (ITD) as the gas chromatograph. The fused silica capillary normally used in the transfer line is replaced by a 1 m, 0 .15-mm inside diameter, 1.2 micron thick methyl silicone stationary phase (DB-1) GC column. Given the short column length and using a direct vapor sampling inlet, the exhaust gas can be sampled quickly, approximately every 10 s in these experiments. Since the column is isothermal, only a limited range of compounds can be analyzed for any given experiment.

1989

Fundamental Experiments on Thermal Desorption of Contaminants from Soils

Lighty, J.S.; Silcox, G.D.; Pershing, D.W.; Cundy, V.A. and Linz, D.G.
Environmental Progress, 8 (1), 1989. Funded by the Gas Research Institute, Dave Linz, Project Manager, National Science Foundation - Presidential Young Investigator Program, ACERC (National Science Foundation and Associates and Affiliates), the State of Utah, and US Department of Energy.

The goals of this research are to develop an understanding of the transport phenomena that occur during the desorption of contaminants from soil and to obtain rate information. This information can then be used to develop a model to predict the performance of full-scale thermal treatment systems for optimization and cost reduction.

The Effects of Rotary Kiln Operating Conditions and Design on Burden Heating Rates as Determined by a Mathematical Model of Rotary Kiln Heat Transfer

Silcox, G.D. and Pershing, D.W.
Accepted for publication in JAPCA, 1989. Funded by ACERC (National Science Foundation and Associates and Affiliates), the State of Utah, and US Department of Energy.

A mathematical model of heat transfer in a directly fired rotary kiln is developed and used to examine the effects of operating and design parameters on burden temperature. The model includes a mean beam length radiation model and axial zoning. Conductive and convective heat transfers are also included. Radiation between immediately adjacent zones is permitted. Calculation of heat transfer rates is facilitated by the use of an electric circuit analogue. An iterative solution procedure is adopted to solve the energy balance equations.

At the conditions examined, the model predicts that coflowing gas and solid streams result in higher average burden temperatures than do counter flowing streams. The moisture level of the feed is predicted to be a key operating parameter. The effects of kiln length, burden residence time, firing rate, and flame length are also examined.

Calcination and Sintering Models for Application to High-Temperature, Short-Time Sulfation of Calcium-Based Sorbents

Milne, C.R.; Silcox, G.D.; Pershing, D.W. and Kirchgessner, D.A.
Accepted for publication in I & EC Res., 1989. Funded by the US Environmental Protection Agency, ACERC (National Science Foundation and Associates and Affiliates), the State of Utah, and US Department of Energy.

To simulate the staged availability of transient high surface area CaO observed in high-temperature flow-reactor data, the rate of calcination of CaCO3 or Ca(OH)2 is described by an empirical modification of the shrinking-core model. The physical model depicts particle decomposition by the shrinking-core mechanism. The subsequent time dependent decrease in CaO reactivity (surface area and porosity) due to sintering is simulated by reducing the grain-center spacing for the matrix of overlapping CaO grains. Information from SEM micrographs and from other physical property measurements of the porous particles is incorporated. This submodel simulates the time dependent availability and reactivity of CaO for a comprehensive model used to study sulfation of CaCO3 or Ca(OH)2 particles at upper-furnace injection conditions.

Rotary Kiln Incineration--Combustion Chamber Dynamics

Cundy, V.A.; Lester, T.W.; Leger, C.B.; Miller, G.; Montestruc, A.N.; Acharya, S.; Sterling, A.M.; Pershing, D.W.; Lighty, J.S.; Silcox, G.D. and Owens, W.D.
Journal of Hazardous Materials, 22, 195-219,1989. Funded by US Environmental Proctection Agency and ACERC (National Science Foundation and Associates and Affiliates).

A multifaceted experimental and theoretical program aimed at understanding rotary kiln performance is underway. The overall program involves university, industry, and government participation and is broken into distinct sub-programs. This paper discusses in some detail the research effort performed to date in two of the sub-programs: Full-scale in situ sampling and kiln-simulator experimentation. Full-scale in situ measurements are obtained from the Louisiana Division rotary kiln facility of Dow Chemical USA, located in Plaquemine, Louisiana. Summary results obtained from controlled experiments that were performed during continuous processing of carbon tetrachloride and preliminary results obtained during batch mode processing of toluene-laden sorbent packs are presented. Kiln-simulator data are obtained by using the facilities of the Chemical Engineering Department at the University of Utah. Recent kiln-simulator work, conducted in support of the full-scale measurements sub-program, has aided in providing an understanding of the results that have been obtained at the full-scale. Modeling efforts, conducted at Louisiana State University and the University of Utah, have concentrated on the development of realistic, fluid-flow and heat-transfer models, near-term chlorinated kinetic models and bed mass-transfer models to be incorporated into a global three-dimensional kiln-simulator model. The paper concludes with an overview of these modeling efforts.

Biomass Combustion: Relationship Between Pollutant Formation and Fuel Composition

Winter, R.M.; Clough, J.; Overmoe, B.J. and Pershing, D.W.
Tappi Journal, 72(4):139-145, 1989. Funded by Weyerhaeuser Corp.

A 65-kW refractory-walled reactor was used to study biomass combustion under conditions typical of the suspension-burning phase in a spreader-stoker-fired boiler. Isothermal combustion data and nitric oxide (NO) emission rates were obtained as a function of temperature, local oxygen concentration, and vertical velocity for sized biomass fuels. Two softwoods, a hardwood, and a North Carolina peat were studied.

The pyrolytic C, H, and N data confirmed the overall high volatility, relative to coal, of these biomass fuels. Particulate emissions were correlated to vertical velocity and particle geometry, but were found to be relatively insensitive to combustion-zone oxygen, temperature, and biomass composition, NO emissions are strongly dependent on combustion-zone oxygen concentration and the nitrogen content of the biomass fuel. NO emissions increased dramatically with increasing excess air and increasing fuel nitrogen; however, these emissions were relatively insensitive to both temperature and moisture content.

Experimental and Theoretical Studies on the Combustion of Waste Zirconium Sponge in a Rotary Kiln Simulator

Lemieux, P.M.; Silcox, G.D. and Pershing, D.W.
Waste Management, 9:125-137, 1989. Funded by Westinghouse Corp.

Previous experimental studies have indicated that rotary kilns may be suitable combustion systems to incinerate small quantities of off grade zirconium sponge produced during the manufacturing of zirconium for the nuclear industry. This paper describes a mathematical model of zirconium sponge combustion in a rotary kiln environment and specifically examines the use of the bed submodel to analyze detailed zirconium combustion data obtained previously in a rotary kiln simulator. The results of this analysis indicated that the experimentally observed burning rates could be predicted within 25% based on available transport correlations and current understanding of zirconium combustion. Without any adjustment of the physical parameters in the model, both the experimental results and the model predictions indicated that the primary combustion parameters are the bulk oxygen concentration within the kin and the local kiln bed temperature. Charge size was found to be less significant. A detailed analysis of the theoretical predictions indicates that the zirconium sponge oxidation rate is controlled by three sequential processes (the convection of oxygen from the bulk gas to the top of the bed, the diffusion of oxygen through the bed to the particle surface, and the diffusion of oxygen through previously formed zirconium dioxide product to the unreacted zirconium metal). Under conditions typical of commercial rotary kiln operation all three of these resistances appear to be significant. Both the experimental data and the model suggest that intrinsic chemical kinetics are fast and not controlling except during the first few minutes after the zirconium is charges. The model assumes that the zirconium oxide product layer increases until it reaches a maximum value after which it remains constant due to continuous formation and abrasion.

A Fundamental Study of the Rate of Thermal Desorption of Contaminants from Soil Beds

Lighty, J.S.; Eddings, E.G.; Deng, X.-X.; VanOs, L.M. and Pershing, D.W.
82nd Annual Air and Waste Management Association Annual Meeting, Anaheim, California, 1989. Funded by the Gas Research Institute, Dave Linz, Project Manager, ACERC (National Science Foundation and Associates and Affiliates), the State of Utah, and US Department of Energy.

Thermal treatment is a proven, permanent destruction technology for most organic wastes and is presently available commercially. Rotary kiln incineration is one of the most common systems for thermal remediation of solids; however, new thermal technologies are being sought. The EPA Superfund Innovative Technology Evaluation (SITE) Program, which focuses on the use of innovative technologies for permanent remediation purposes, has identified seven thermal technologies that have been or will be demonstrated this has identified seven thermal technologies that have been or will be demonstrated this year. These technologies include infrared thermal destruction, circulating fluidized bed combustion, and in situ steam/air stripping. In each of these technologies important transport phenomena, specifically mass and heat transfer, must be understood to enable performance prediction, through modeling, for these technologies as full-scale processes.

The program at the University of Utah attempts to identify the fundamental transport rates in the decontamination of solids using thermal treatment. Once these phenomena are understood, the thermal treatment system can then be optimized. The experimental approach involves three fundamental rate reactors and two reactors designed specifically to address incineration in a rotary kiln environment. The fundamental experiments attempt to address incineration in a rotary kiln environment. The fundamental experiments attempt to identify the controlling processes at a particle level in a Particle-Characterization Reactor (PCR) and the resistances within a bed of solid in two Bed-Characterization Reactors (BCR). The fundamental insights gained from these reactors can be applied to any thermal treatment system, since the systems differ only in the mechanisms of heat transfer and mass treatment system, since the systems differ only in the mechanisms of heat transfer and mass transfer. For example, in infrared thermal destruction the transport limitations within the bed and radiation from the top of the bed are important processes that need to be understood. In a circulating fluidized bed combustor, an understanding of gas/solid contacting and convection needs to be gained. In situ steam/air stripping requires knowledge of the interactions between the water, contaminant, and soil matrix.

Two other experimental systems focus specifically on the unique features of rotary-kiln incineration technology. An ambient temperature mixing reactor and a rotary kiln simulator are used to evaluate mixing and combustion characteristics typical of a large-scale rotary kiln.

The paper focuses on results obtained in the Bed-Characterization Reactors. Mass transfer and heat transfer resistances have been found to be extremely important. The effects of packing density, moisture, and particle size on these resistances will be discussed.

The Effect of Fuel Nitrogen on NOx Emissions From a Rotary-Kiln Incinerator

Lighty, J.S.; Gordon, D.L.; Pershing, D.W.; Owens, W.D.; Cundy, V.A. and Leger, C.B.
Stationary NOx Symposium, San Francisco, California, 1989. Funded by Louisiana State University/Hazardous Waste Research Center, and ACERC (National Science Foundation and Associates and Affiliates).

While fuel NOx formation has been extensively studied for coal combustion, little information is available on NOx formation for nitrogenous waste constituents. These wastes, usually destroyed in hazardous-waste incinerators, are prevalent and exist as solids (plastics, nylongs) or liquids (dyes, process waste).

Results are presented from studies conducted in a scaled, batch rotary-kiln simulator. Constituent parameters, i.e. constituent type and percent fuel nitrogen, were studied at 730ºC. Sorbent was contaminated with a variety of constituents ranging in concentrations from 0.5% to 3.0% nitrogen by weight (for 681 g charge).

NOx exhaust-gas concentrations ranged from 60-80 ppm for a base run (no fuel nitrogen) to 200-1750 ppm for a nitrogenous waste. Results inducated that, at higher concentrations, more NOx was formed, accompanied by an increase in temperature. Higher concentrations also resulted in reduced percent conversions of fuel nitrogen to NOx. Evolution for different constituents, given the same concentration, varied; aniline formed more NOx than pyridine, followed by ethylenediamine.

An In-Depth Study of Incineration - Rotary Kiln Performance Characterization

Cundy, V.A.; Lester, T.W.; Conway, L.R.; Jakway, A.J.; Leger, C.B.; Montestruc, A.N.; Acharya, S.; Sterling, A.M.; Owens, W.D.; Lighty, J.S.; Pershing, D.W. and Silcox, G.D.
Louisiana State University/Hazardous Waste Research Center SAC Review Meeting, Baton Rouge, Louisiana, 1989. Funded by Louisiana State University/Hazardous Waste Research Center (Supported by US Environmental Protection Agency).

A comprehensive study aimed at understanding rotary kiln performance is underway. The program is led by personnel from Louisiana State University. Bench and pilot-scale facilities at the University of Utah are available for use in solids desorption studies. Full-scale in situ measurements are obtained from the Louisiana Division rotary kiln facility of Dow Chemical USA, located in Plaquemine, Louisiana. This paper presents a summary of the project providing some detail of the work that has been accomplished from 1 January through 31 August 1989.

Fundamentals for the Thermal Remediation of Contaminated Soils - Particle and Bed Desorption Models

Lighty, J.S.; Silcox, G.D.; Pershing, D.W.; Cundy, V.A. and Linz, D.G.
Accepted for publication in Environ. Science Tech., 1989. Funded by the Gas Research Institute, Dave Linz, Project Manager, National Science Foundation/Presidential Young Investigators, ACERC (National Science Foundation and Associates and Affiliates), the State of Utah, and US Department of Energy.

A major research effort has been initiated to characterize the rate-controlling processes associated with the evolution of hazardous materials from soils. A threefold experimental approach is being used in conjunction with computer modeling to analyze thermal desorption of contaminants. Phenomena occurring both inside particles (intraparticle) and with a bed of particles (interparticle) were studied.

The results obtained suggest that the most important process variables are local thermal environment and gas-phase contaminant concentration because the adsorption equilibrium characteristics of the contaminant/soil pair control the desorption of contaminant from a particle at a given temperature. A mass-transfer/desorption model, which assumes gas/solid equilibrium at all points and time, is proposed and the model was found to predict the measured temperature dependence.

1988-1987

Rotary Kiln Incineration I: An In-Depth Study - Liquid Injection

Cundy, V.A.; Lester, T.W.; Morse, J.S.; Montestruc, A.N.; Leger, C.B.; Acharya, S.; Sterling, A.M. and Pershing, D.W.
Submitted to Journal of the Air Pollution Control Association, 1988. 39 pgs. Funded by Environmental Protection Agency.

A multifaceted experimental and theoretical program directed toward the understanding of rotary kiln performance is underway. Following a general overview of the program, we describe in more detail the program components including: In-situ measurements from an industrial-scale rotary kiln located at the Louisiana Division of Dow Chemical USA in Plaquemine, Louisiana; laboratory-scale desorption characterization and kiln-simulator studies; and incinerator modeling efforts. Using water-cooled probes, hot-zone samples have been obtained from both the full-scale rotary kiln and the afterburner and have been analyzed subsequently using GC and/or GC/MS techniques. We report on these preliminary measurements in some detail.

Rotary Kiln Incineration II: Laboratory Scale Desorption and Kiln Simulator Studies - Solids

Cundy, V.A.; Lester, T.W.; Morse, J.S.; Montestruc, A.N.; Leger, C.B.; Acharya, S.; Sterling, A.M. and Pershing, D.W.
Submitted to Journal of the Air Pollution Control Association, 1988. 28 pgs. Funded by Gas Research Institute, ACERC (National Science Foundation and Associates and Affiliates), and National Science Foundation/Presidential Young Investigators.

With landfill costs increasing and regulations on landfilling becoming more stringent, alternatives to conventional hazardous waste treatment strategies are becoming more desirable. Incineration is presently a permanent, proven solution for the disposal of most organic contaminants, but also a costly one, especially in the case of solids that require some auxiliary fuel. The goal of this research is to develop an understanding of the phenomena associated with the evolution of contaminants from solids in the primary combustor of an incineration system. A four-fold approach is being used. First, a bench scale Particle Characterization Reactor was developed to study the transport phenomena on a particle basis, where the controlling processes are mainly intraparticle. Second, a Bed Characterization Reactor was built to examine the controlling transport phenomena within a bed of particles, where the processes are primarily interparticle. The results of these studies can be applied to any primary combustor. A pilot-scale rotary kiln was developed to study the evolution of contaminants from solids within a realistic temperature and rotation environment. Finally, in-situ measurements are being obtained from a full-scale rotary-kiln.

This paper describes results obtained in a study using a commercial sorbent contaminated with toluene. The data are from the Particle Characterization Reactor and the Rotary-Kiln Simulator. The results show that the method of contamination and charge size does not have a large effect on desorption, while temperature and contaminant concentration are important parameters in the evolution of contaminants in a rotary kiln.

Rotary Kiln Incineration III: An In-depth Study - Full-Scale Incineration of Carbon Tetrachloride

Cundy, V.A.; Lester, T.W.; Morse, J.S.; Montestruc, A.N.; Leger, C.B.; Acharya, S.; Sterling, A.M. and Pershing, D.W.
Submitted to Journal of the Air Pollution Control Association, 1988. 31 pgs. Funded by Environmental Protection Agency.

Temperature and stable species concentration data are presented from various locations within a full-scale rotary kiln firing natural gas/carbon tetrachloride/air. The data are being collected as part of a cooperative program involving university, industrial and government participation. The overall goal of the program is to develop a rudimentary understanding of and a predictive capability for rotary kiln and afterburner performance as influenced by basic design and operation parameters. The data clearly demonstrates that severe non-uniformities exist at the kiln exit under certain operating conditions. Even so, the data further indicate that high destruction efficiencies were achieved through adequate secondary combustion processing. The data further show that flow perturbations from within the kiln can persist well into the afterburner section and even into the stack.

Transport Fundamentals for the Thermal Remediation of Contaminated Soils

Lighty, J.S.; Silcox, G.D.; Pershing, D.W.; Cundy, V.A. and Linz, D.G.
Submitted for Review to AIChE Journal, 1988. 23 pgs. Funded by Gas Research Institute, ACERC (National Science Foundation and Associates and Affiliates), and National Science Foundation/Presidential Young Investigators.

A major research effort has been initiated to characterize the transport phenomena associated with the evolution of hazardous materials from contaminated soils. A threefold experimental approach is being used in conjunction with computer modeling to analyze thermal desorption of contaminants from soils under a variety of experimental conditions. First, a particle-characterization reactor (PCR) is being employed to characterize intraparticle transport under conditions where the bulk concentration and temperature at the particle surface are known. Secondly, a packed-bed reactor is being used to examine interparticle transport within a bed of particles. In the third portion of the work, a 73 kW pilot-scale rotary kiln is providing time-resolved measurements of the species evolution. The results from the PCR experiments and modeling are reported in this manuscript.

The PCR experimental results obtained suggest that the most important process variable is local thermal environment and that the absorption characteristics of the contaminant/soil pair are important in the desorption process. The adsorption characteristics are a function of local temperature, thus accounting for the steep dependence of the cleanup efficiency of a soil on temperature. In addition, the adsorption characteristics are influenced by soil type and moisture; it has been found that these parameters also effect desorption of contaminants. A mass-transfer model, which assumes gas/solid equilibrium at all points and times, is proposed and the model predictions were found to exhibit temperature dependence similar to the data.

Advanced NOx Reduction Processes Using -NH and -CN Compounds in Conjunction with Staged Air Addition

Chen, S.L.; Cole, J.A.; Heap, M.P.; Kramlich, J.C.; McCarthy, J.M. and Pershing, D.W.
Twenty-Second Symposium (International) on Combustion/The Combustion Institute, 1135-1145, 1988. Funded by Pittsburgh Energy Technology Center.

The effectiveness of combustion modifications for the control of nitrogen oxide emissions from coal fired combustors is most often limited by problems due to carbon burnout or flame impingement. This paper presents new data on the use of selective reducing agents suggesting that a hybrid control scheme is possible which uses combustion modification to provide those conditions that optimize the selective reduction process. Very low emission levels appear possible that can presently only be achieved by catalytic reduction. The experimental studies were conducted in a tunnel furnace that simulated the thermal environment within a pulverized coal boiler. Application of each of the agents (ammonia, urea, cyanuric acid, and ammonium sulfate) to an overall fuel lean environment produced NO reduction behavior very similar to that of thermal deNOx. However, if the agent was added to the fuel rich zone of a rich/lean staged combustor, very high NO reductions were obtained after the leanout point. The result of the staging was to extend the effectiveness of the agent to lower temperatures relative to overall lean injection. Parametric variations indicated that in addition to temperature, the most important variable was the rich zone stoichiometry. Kinetic modeling suggests that the rich zone acts primarily as a source of CO. At the rich/lean transition the CO is oxidized and excess OH is produced by the usual chain branching reactions. For low initial CO concentrations the excess radicals are consumed by:

NH3 + OH = NH2 + H2O

HNCO + H = NH2 + CO

The NH2 is then available for reaction with NO to eventually yield N2. The strong rich zone stoichiometry dependence is exerted mainly through the amount of CO supplied to the lean zone. Insufficient CO will limit the extent of the initial NH3 or HNCO reaction.

Characterization of Thermal Desorption Phenomena for the Clean-Up of Contaminated Soil

Lighty, J.S.; Pershing, D.W.; Cundy, V.A. and Linz, D.G.
Accepted for Nuclear & Chemical Waste Management, 1987. 32 pgs. Funded by ACERC (National Science Foundation and Associates and Affiliates).

The overall goal of this research is to develop an understanding of the fundamental transport phenomena associated with the evolution of hazardous materials from solids, in particular contaminated soils. At the present time, incineration is a relatively costly alternative for the clean up of contaminated soils. An understanding of the mass transfer and heat transfer limitations might lead to a more economical option, where the contaminants are desorbed from the soil at lower temperatures in a primary combustor and then a secondary, high temperature combustor (afterburner) decomposes the hazardous off-gases. This work is aimed at providing fundamental rate information which will be used to model thermal desorption of contaminants from soils under a variety of thermal conditions, soil properties and contaminants.

The experimental approach is threefold. First, a bench scale particle characterization reactor (PCR) has been developed and is being used to characterize intraparticle transport under conditions where the bulk concentration and temperature at the particle surface are known. Following these studies, a packed-bed reactor will be used to examine interparticle transport within a well-characterized bed of particles. In the third portion of the work, a 73 kW pilot-scale rotary kiln will be used to obtain time resolved measurements of trace species evolution. This paper reports recent PCR results that indicate that soil properties, type of contaminant, and temperature are important in desorption of contaminants from soil particles.

The overall goal of this research is to develop an understanding of the fundamental transport phenomena associated with the evolution of hazardous materials from solids, in particular contaminated soils. At the present time, incineration is a relatively costly alternative for the clean up of contaminated soils and it may render the soil inert. A more economical option is to desorb the contaminants from the soil at lower temperatures and then use high temperature incineration to decompose the hazardous off-gases. This work is aimed at providing fundamental rate information which may be used to model the thermal desorption of contaminants from soils under a wide variety of thermal conditions.

The experimental approach is three-fold. First, a bench-scale particle characterization reactor (PCR) has been developed and is being used to characterize intraparticle transport under conditions where the bulk concentration and temperature at the particle surface are known. Following these studies, a packed bed reactor will be used to examine interparticle transport within a well-characterized bed of particles. In the third portion of the work, a 73,000 Watt pilot-scale rotary kiln will be used to reports recent PCR and kiln results; it does not address the packed bed reactor studies since they are just being initiated. The PCR results indicate that soil pore structure is important in desorption of contaminants from soil particles and that the desorption/diffusion step is probably a controlling mechanism.

A Mathematical Model for the Flash Calcination of Dispersed CaCO3 and Ca(OH)2 Particles

Silcox, G.D.; Kramlich, J.C. and Pershing, D.W.
Submitted to I&EC/Research, 1987. 28 pgs. Funded by ACERC (National Science Foundation and Associates and Affiliates).

A mathematical model for the flash calcination of Ca(OH)2 and CaCO3 is presented. The model describes the decomposition of the parent material at the reactant-product interface, the diffusion of CO2 or H2O through the growing CaO layer, and the sintering of the CaO layer. The model is qualitative, but it provides useful estimates of peak CO2 pressures in the CaO layer, relative rates of surface area development and loss for Ca(OH)2 and CaCO3, the effect of particle size, and the effects of time and temperature.

The Design and Construction of a Rotary Kiln Simulator for Use in Studying the Incineration of Hazardous Waste

Lemieux, P.M. and Pershing, D.W.
Submitted to Review of Scientific Instrument, 1987. Funded by Westinghouse and National Science Foundation/Presidential Young Investigators.

Rotary kiln systems are widely used in industrial applications to transfer energy from high temperature flames to irregular solids. Recently these systems have been shown to be suitable for the incineration of hazardous solid waste materials and the thermal treatment of contaminated soils. Destruction and removal efficiencies in excess of 99.99% have been reported for hazardous species, but the rate controlling steps of the incineration process are not well understood.

This paper describes the design, construction and operation of a laboratory scale simulator that was developed to investigate the fundamentals of hazardous incineration in a rotary kiln environment. This 2 ft. x 2 ft. refractory-lined kiln allows time resolved characterization of contaminant evolution and destruction. Continuous thermal and exhaust concentration measurements are used to characterize the fate of the solid charge as a function of residence time within the kiln. Overall destruction efficiency can be measured by subsequent analysis of the solid phase.

The initial performance of this facility has been demonstrated by studying the combustion of waste zirconium metal and by characterizing the thermal clean up of solid sorbent contaminated with toluene. The rotary kiln simulator has been shown suitable for investigation of parameters such as amount of charge, contaminant loading, rotation speed, temperature, excess oxygen, and particle size.

The Combustion of Waste Zirconium Sponge in a Rotary Kiln Simulator

Lemieux, P.M. and Pershing, D.W.
Submitted to Metals, 1987. Funded by Westinghouse and National Science Foundation/Presidential Young Investigators.

The combustion of off-grade zirconium sponge in a rotary kiln environment is described. The purpose of these experiments was to determine the suitability of rotary kiln incineration for the disposal of zirconium wastes, and to determine the design parameters and optimum operating conditions for a zirconium burning rotary kiln. In the rotary kiln simulator, the experiments are done in a batch mode, simulating a control volume of solids moving down the length of a full-scale rotary kiln, and exchanging time for distance as the independent variable. Parametric studies investigating the effect of oxygen concentration, charge size, rotational velocity, particle size and temperature on the burning rate have been completed. The burning rate appears to be controlled by intra- and interparticle diffusion, thus the rate at which the product layer is abraded off and the availability of O2 at the particle surface appear to strongly affect the burning rate. The burning rate is independent of charge size, weakly dependent on oxygen concentration, rotational velocity and particle size, and strongly dependent on temperature.

Time Resolved Sulfation Rate Measurement for Sized Sorbents

Milne, C.R. and Pershing, D.W.
Proc. 4th Annual Pittsburgh Coal Conf., 1987. 14 pgs. Funded by ACERC (National Science Foundation and Associates and Affiliates).

The sulfation of raw and sized calcium-based sorbents under high-temperature, short time conditions typical of injection in pulverized coal fired boilers was studied. A 100,000 Btu/hr refractory walled flow reactor was used to obtain isothermal SO2 capture which occurs in the first 30 ms after injection and results from inherent particle size difference between the two sorbent types. In this instance, the hydrate particle size is one-tenth that of the carbonate, and the hydrate capture at Ca/S = 2 is 30% versus 5% for the carbonate.

Beyond approximately 200 ms the sulfation rate was insensitive to particle size, but the rate of SO2 capture increase with the hydrated sorbent was larger than that measured with the carbonate sorbent. This suggests that the ultimate extent of sulfation is also influenced by structural parameters. Ultimate calcium utilization does not appear to be significantly influenced by the calcination rate differences between hydrates and carbonates for particles smaller than 5 um. For large carbonate particles, the calcination delay slightly retards the initial sulfur capture and reduces the ultimate calcium utilization.

The Clean Up of Contaminated Soil By Thermal Desorption

Lighty, J.S.; Pershing, D.W.; Cundy, V.A.; Groves, F.R. Jr. and Linz, D.G.
Proc. 2nd Int. Conf. on New Frontiers in Hazardous Waste Management, 1987, NUS, Pittsburgh, PA. 10 pgs. Funded by Gas Research Institute, National Science Foundation, and ACERC (National Science Foundation and Associates and Affiliates).

The overall goal of this research is to develop an understanding of the fundamental transport phenomena associated with the evolution of hazardous materials from solids, in particular, contaminated soils. At the present time, incineration is a relatively costly alternative for the clean up of contaminated soils and it may render the soil inert. A more economical option is to desorb the contaminants from the soil at lower temperatures and then use high temperature incineration to decompose the hazardous off-gases. This work is aimed at providing fundamental rate information which may be used to model the thermal desorption of contaminants from soils under a wide variety of thermal conditions.

The experimental approach is threefold First, a bench-scale particle characterization reactor (PCR) has been developed and is being used to characterize intraparticle transport under conditions where the bulk concentration and temperature at the particle surface are known. Following these studies, a packed bed reactor will be used to examine interparticle transport within a well-characterized bed of particles. In the third portion of the work, a 73,000 Watt pilot-scale rotary kiln will be used to obtain time resolved measurements of trace species evolution. This paper reports recent PCR and kiln results; it does not address the packed bed reactor studies since they are just being initiated. The PCR results indicate that soil pore structure is important in desorption of contaminants from soil particles and that the desorption/diffusion step is probably a controlling mechanism.

Fundamentals of Hazardous Solid Waste Incineration in a Rotary Kiln Environment

Lighty, J.S.; Silcox, G.D.; Britt, R.; Owens, W.D.; Pershing, D.W. and Cundy, V.A.
Proc. AFRC Int'l. Symposium on Waste Incineration, 1987, Palm Springs. Funded by Environmental Protection Agency.

With landfill costs increasing and regulations on landfilling becoming more stringent, alternatives to conventional hazardous waste treatment strategies are becoming more desirable. Incineration is presently a permanent, proven solution for the disposal of most organic contaminants, but also a costly one, especially in the case of solids that require some auxiliary fuel. The goal of this research is to develop an understanding of the phenomena associated with the evolution of contaminants from solids in the primary combustor of an incineration system. A threefold approach has been used. First, a bench-scale Particle Characterization Reactor was developed to study the transport phenomena on a particle basis, where the controlling processes are mainly intraparticle. Second, a Bed Characterization Reactor was built to examine the controlling transport phenomena within a bed of particles, where the processes are primarily interparticle. The results of these studies can be applied to any primary combustor. Finally a pilot-scale rotary kiln was developed to study the evolution of contaminants from solids within a realistic temperature and rotation environment.

This paper describes results obtained in a study using a commercial sorbent contaminated with toluene. The data are from the Particle Characterization Reactor and the Rotary-Kiln Simulator. The results show that the method of contamination and charge size does not have a large effect on desorption, while temperature and contaminant concentration are important parameters in the evolution of contaminants in a rotary kiln. Preliminary modeling efforts for the kiln are also discussed.

Fuel Rich Sulfur Capture in a Combustion Environment

Lindgren, E.R. and Pershing, D.W.
Western States Section, Combustion, Combustion Institute, 1987. Funded by Environmental Protection Agency.

A major concern associated with the combustion of coal for heat and electricity is the emission of acid rain precursors, NOx and SO2. Dry calcium based sorbent injection is a potential method for reducing SO2 emissions from existing coal-fired boilers. A great deal of study has been devoted to the fuel lean SO2 reaction:

CaO + SO2 ==> CaSO4

The fuel rich analogies:

CaO + H2S ==> CaS + H2O

CaO + COs ==> CaS + CO2

are kinetically faster than the SO2 reaction. Thermodynamics plays an important role in the rich capture reactions because the gaseous products are also products of combustion. These fuel rich reactions have not been extensively studied, particularly under combustion conditions.

In this study a refractory lined, down fired natural gas furnace was used to study fuel rich sulfur capture as a function of sulfur concentration (750-3000 ppm), molar ratio of calcium to sulfur (Ca/S = 1 to 4), residence time (0.30 to 0.65 sec), quench rate (450 to 900ºF/sec), stoichiometric ratio in the rich zone (SR=0.65, 0.75), and sorbent type (Marblehead hydrate and Fredonia carbonate). The fuel rich sulfur species H2S and COs were monitored in a near continuous fashion using a gas chromatograph equipped with a flame photometric detector (GC-FPD) and an automatic sampling system which sampled every 30 seconds. Below the fuel rich zone, 25% excess air was added and the ultimate fuel lean capture was measured using a continuous SO2 monitor.

Under fuel rich conditions, calcium utilization increases with increasing sulfur concentration and decreasing Ca/S. At low concentrations, fuel rich sulfur capture may be thermodynamically limited. The ultimate lean capture was found to be independent of the sulfur capture level obtained in the fuel rich zone, thus the high capture realized in the rich zone was lost. The results on the lean side are very typical of lean capture data reported by others; i.e. the calcium utilization is weakly dependent on Ca/S and sulfur concentration. These results suggest that the rate-limiting step under fuel rich conditions is different than under fuel lean conditions.

A Theoretical and Experimental Analysis of High Temperature, Fuel Rich Sulfide Formation with Sized, Precalcined, Calcium Based Sorbents

Lindgren, E.R. and Pershing, D.W.
Western States Section, Combustion Institute, 1987, Honolulu, Hawaii. Funded by Environmental Protection Agency.

Sized 3 to 10 µm calcium based sorbent particles (one hydrate and one carbonate) were precalcined to calcium oxide (CaO) and injected into a refractory lined, natural gas fueled furnace operated under fuel rich conditions (SR=0.75). Fuel rich sulfur species, H2S and COs react with CaO to form calcium sulfide (CaS). Gas phase sulfur capture measurements were made at a residence time of 0.44 seconds as a function of sulfur concentration (800 to 3200 PPM) and calcium availability (Ca/S = 1 to 4). For selected runs, solid samples were collected simultaneously at a residence time of 0.65 seconds and N2 porosimetry measurements were made to determine the effect of conversion on the pore size distribution (PSD).

The data are compared to the predictions of a computer model which views a sorbent particle as a sphere made up of a distribution of randomly oriented, highly interconnected, cylindrical pores, (attributed to Christman and Edgar). Both precalcined sorbents exhibit a strong positive dependence on sulfur concentration that the model predicts well for the hydrate but not for the carbonate. These results parallel those found for the same raw sorbents (not precalcined) in an earlier study. The effect of conversion on the PSD was predicted better for the hydrate sorbent than for the carbonate sorbent.

Time Resolved Sulfation Rate Measurements for Sized Sorbents

Milne, C.R. and Pershing, D.W.
Proc. 4th Annual Pittsburgh Coal Conf., 1987. 6 pgs. Funding source ACERC (National Science Foundation and Associates and Affiliates).

The sulfation of raw and sized calcium-based sorbents under high-temperature, short time conditions typical of injection in pulverized coal fired boilers was studied. A 100,000 Btu/hr refractory walled flow reactor was used to obtain isothermal SO2 capture data for a carbonate and a hydrate derived from the carbonate as a function of reactor residence time and particle size.

Experimental results confirmed the importance of sorbent type and particle size. The hydrated sorbent was clearly superior to the carbonate. This difference is primarily attributable to a dramatic difference in prompt SO2 capture which occurs in the first 30 ms after injection and results from inherent particle size difference between the two sorbent types. In this instance, the hydrate particle size is one-tenth that of the carbonate, and the hydrate capture at Ca/S = 2 is 30% versus 5% for the carbonate.

Beyond approximately 200 ms the sulfation rate was insensitive to particle size, but the rate of SO2 capture increase with the hydrated sorbent was larger than that measured with the carbonate sorbent. This suggests that the ultimate extent of sulfation is also influenced by structural parameters. Ultimate calcium utilization does not appear to be significantly influenced by the calcination rate differences between hydrates and carbonates for particles smaller than 5 µm. For large carbonate particles, the calcination delay slightly retards the initial sulfur capture and reduces the ultimate calcium utilization.

A Mathematical Model of Zirconium Combustion in a Rotary Kiln

Lemieux, P.M. and Pershing, D.W.
Submitted to Combustion Science & Technology, 1987. Funded by Westinghouse and National Science Foundation/Presidential Young Investigators.

Previous experimental studies have indicated that rotary kilns may be suitable combustion systems to incinerate small quantities of off grade zirconium sponge produced during the manufacturing of zirconium for the nuclear industry. This paper describes a mathematical model of zirconium sponge combustion in a rotary kiln environment and specifically examines the use of the bed submodel to analyze detailed zirconium combustion data obtained previously in a rotary kiln simulator.

The results of this analysis indicated that the experimentally observed burning rates could be predicted within 25% based on available transport correlations and current understanding of zirconium combustion, without any adjustment of the physical parameters in the model both the experimental results and the model predictions indicated that the primary combustion parameters are the bulk oxygen concentration within the kiln and the local kiln bed temperature. Charge size was found to be less significant.

A detailed analysis of the theoretical predictions indicates that the zirconium sponge oxidation rate is controlled by three sequential processes (the convection of oxygen from the bulk gas to the top of the bed, the diffusion of oxygen through the bed to the particle surface, and the diffusion of oxygen through previously formed zirconium dioxide product to the unreacted zirconium metal). Under conditions typical of commercial rotary kiln operation all three of these resistances appear to be significant. Both the experimental data and the model suggest that the intrinsic chemical kinetics are fast and not controlling except during the first few minutes after the zirconium is charged. The model assumes that the zirconium oxide product layer increases until it reaches a maximum value after which it remains constant due to continuous formation and abrasion.