Computational Fluid Dynamics Evaluation of Good Combustion Performance in Waste Incinerators

Donghoon Shin, Chang Kook Ryu, and Sangmin Choi Mechanical Engineering Department, Korea Advanced Institute of Science and Technology, Taejon, Korea

IMPLICATIONS As to the acceptable operating performance of solid waste incinerators, good combustion practices(GCP or GOP) have been established. These operating (and design) guidelines call for regulations of furnace temperature and residence times of gaseous combustion products. However, measurements and their control in practical systems are not an easy task, if only can be done. The current paper presents a method which quantitatively evaluates the residence times and furnace temperature, based on the computational fluid dynamices(CFD) modeling. This paper further presents usefulness of newly defined parameters(local mixedness parameter and thermal decomposition parameter) in interpreting the massive amount of CFD result data. These numerical computations are intended to help engineers and scientists in the incineration field, who experience frustrated limitations of field measurement data or quantitative guidelines in their engineering decision making.

ABSTRACT Combustion control technique has become a legal requirement to minimize pollution in municipal solid waste incinerators. The conditions for in-furnace destruction of pollutants are stated as: good combustion is achieved when 2-second gas residence time at 850C and 6% O2 is guaranteed. Incinerator performance is analyzed numerically to validate good combustion performance. Computational Fluid Dynamics (CFD) modeling of gas flow inside the furnace chamber provides 3-dimensional temperature, concentration and velocity vectors. General flow

patterns and the presence of recirculation pockets are traditionally observed. Local temperature and oxygen concentration can also be checked. The CFD results are analyzed further in terms of residence time, mixing and thermal decomposition of potential pollutants. The residence time needs to be carefully determined based on the gas inlet position. The statistical variation requires evaluation of the average and the minimum (or shortest) residence time. Mixing is quantified by defining a local mixedness parameter with which the effects of secondary air jets are interpreted. Thermal decomposition parameter integrates the temperature and oxygen availability over the residence time, which may be directly related to good combustion.

INTRODUCTION Pollution minimization has become the major theme in modern applications of municipal solid waste (MSW) incineration. Incinerator designers and operators have to carefully watch the current status of the requirements. Flue gas cleaning devices are added to meet the emission levels, which have become ever more stringent. In addition to this traditional approach to minimize emissions, newly adopted policies emphasize combustion control concepts [1-4]. Infurnace control of pollution requires one to provide a combustion environment that minimize the production of incomplete combustion products while maximize the destruction of already produced pollutants. For example, U.S. EPAs Good Combustion Practice (GCP) is a qualitative description of methods which must be used to control stack gas emissions of trace organics such as polychlorinated dibenzo dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) and to minimize the amounts of these organics in collected fly ash [5]. Achieving GCP involves the integration of design and operating principles to appropriately control the various incinerator performance parameters; the main emphasis is given to generation of conditions which maximize the furnace destruction of organics. Combustion control is also summarized as achieving 3Ts (time, temperature, and turbulence) [1], to provide sufficiently long residence time, at high enough temperature, and the good mixing of unreaced or partially reacted compounds with fresh air.

corresponding author

To meet the above stated concept action plans must include the decisions from the design phase as well as the control of operating mode. For example, the European directives on municipal solid waste incinerators specifically state that the gases resulting from the combustion of waste must be raised, after the last injection of the combustion air, in a controlled and homogeneous fashion and even under the most unfavorable conditions, to a temperature of at least 850 , for at least two seconds, in the presence of at least 6% oxygen [2]. It is not the point of this argument whether it is possible to control toxic emissions only by furnace control or not. However, many questions remain to be answered about this legally binding statement. Issues include; how the residence time of 2 seconds is determined; and, how closely the local temperature and oxygen concentration can be determined. The usual practice to determine residence time is to calculate the simple bulk averaged gas transit time based on the measured furnace exit gas temperature. It would be desirable if the local temperature and oxygen concentration along the path of each individual volume element of the exit gas are known, in addition to the overall residence time. Computational fluid dynamics (CFD) analysis has become powerful, and the detailed thermo-fluid dynamic processes can be modeled with reasonable efforts [6-12]. Though the validity of the model computation still needs to be assured, computation yields detailed information on temperature, velocity, and concentration, etc. A typical data set of CFD results is enormous in quantity, and it is essential to develop a means of interpreting the computational results if one wants to extract useful information. The objective of the current paper is to present methods of interpreting CFD results, so that the incinerator designs and operation modes which will lead to good combustion conditions can be identified.

INCINERATOR AND COMPUTATIONAL MODEL Figure 1 displays the combustion chamber of a typical MSW incinerator of 300 ton/day capacity. It is a relatively simple geometry of the central-flow type, and only the half volume is shown with a plane of symmetry. Primary air is supplied through the grate and secondary air is injected from the wall. Three secondary air nozzle arrays are located in the designated locations, SA1, SA2, and SA3, respectively. The combustion chamber is conceptually divided into two

sub-chambers. In the primary combustion chamber, the bed of solid waste burns on the grate. A mixture of combustion products, combustibles due to incomplete combustion, and some excess air is released from the solid waste bed, and then allowed to pass through the furnace. The secondary combustion chamber is intended to provide an adequate environment so that the incomplete combustion products and the pollutants should be destroyed, while transferring heat to the water-wall of the boiler. For the comparative evaluation of performance, two cases of secondary air nozzle arrangements are selected. In Case 1 and Case 2, numbers of nozzles on SA1 in half depth are 4 and 3, respectively. As can be seen in Fig.1, the nozzles of SA1 are aligned to face the nozzles of SA2 in Case 1. In Case 2, nozzles of SA1 and SA2 are located in a staggered manner. Jets of Case 1 are strong enough to interact each other in the center region of the chamber, while two facing jets of Case 2 can penetrate further into the opposite wall. This difference in jet interaction is expected to significantly affect the overall mixing in the combustion chamber, which will subsequently determine the performance of in-furnace destruction of pollutants. The geometric shape of incinerators is sometimes simplified as 2-dimensional (planar box shape in most MSW incinerators). This simplification is supported by the usual design practices of sizing the incinerator combustion chamber. For example, as an engineering rule of thumb, the throughput of mass burn incinerators is assumed proportional to the surface area of the stoker grate floor. Based on this assumption, 2-dimensional models are sometimes employed experimentally and computationally [6,7,12]. Flow visualizations using the water flow table and 2-dimensional computations provide a qualitative idea of flow pattern as related to the geometric chamber shapes and the air flow conditions. However, there are many reasons that require three dimensional treatments of thermo-fluid dynamic processes. Heat transfer to the wall and the secondary air injection pattern are typical examples. If the two dimensional simplifications need to be introduced, sufficiently reasonable caution should be exercised. In this study, a three dimensional body fitted grid system of 38 54 24 was chosen to match the geometric shape. The layout was selected to avoid the grid dependency on the solution, while minimizing the computational burden. Flow modeling procedures inevitably involve discussions on the turbulence model, gas phase combustion and heat release, convective and radiative heat

transfer. The standard k- model, and DTRM(Discrete Transfer Radiation Model)[13] where a local absorption coefficient is calculated from CO2 and H2O concentration were employed. Mass burning on the bed was modeled by the inlet boundary conditions as shown in Table 1. Bed combustion products were selected by referring to the actual measurement data[14]. The eddy-dissipation model[15] was employed to model turbulent gas phase combustion. Incomplete combustion products generated from the bed were modeled as CnHm, and a one step complete oxidative reaction was assumed. Computations have been performed by utilizing FLUENT.

PARAMETERS FOR PERFORMANCE EVALUATION OF INCINERATORS

Mixedness parameter :
The mixedness parameter is defined [6] as;
2 = X i X o,i / 2 i

1/ 2

(1)

where Xi is the mass fraction of chemical species i and Xo,i is its average mass fraction at the fully mixed condition. A local variable represents the information of local individual species concentration variances from the perfectly mixed concentration. approaches 0 when the mixing is complete.

Residence time :
Residence time of combustion gas is calculated by the Lagrangian turbulent particle trajectory calculation[10]. The flow pattern of the combustion gas is turbulent, and the local instantaneous velocity uj is the sum of the mean velocity Uj and the turbulent fluctuation uj;

u j = U j + u' j where, uj is determined by the turbulent kinetic energy k and the dissipation rate .
u' j = j (2 k / 3) 1/ 2

(2)

(3)

The instantaneous turbulent fluctuation term is assumed to follow the normal distribution,

by assigning j a normally distributed random number. It is further assumed that the turbulent eddy is retained for the time interval t, which is the characteristic life time of an eddy, defined as
C 3/ 4 k t = 1/ 2 2 10000 particles to get a statistically acceptable value. (4)

where, C equals 0.09. Typical residence time calculations require tracking approximately

Thermal Decomposition Parameter :

Toxic organic pollutants are believed to be generated in an oxygen-starved, relatively low temperature environment [1]. These must be destroyed by sufficient thermal decomposition in the oxygen-rich, high temperature environment. The oxidative reaction of the potential pollutants generated in the incinerator is modeled by the one-step Arrhenius type reaction [16];
n dX i / dt = A X i X O 2 exp ( E / RT )

(5)

where, Xi and XO2 are concentrations of fuel i and oxygen, n is the reaction order of oxygen, and, A and E are the preexponential factor and the activation energy. Since the destruction of pollutants is a combined result of time, temperature and oxygen concentration, the thermal decomposition parameter is introduced which is calculated along the turbulent gas trajectory to quantitatively express the non-reacted fraction of a given pollutant[7];

= X i , exit / X i ,inlet = exp

exit

inlet

n AX O 2 exp( E / RT ) dt

(6)

Of immediate concern, however, is the selection of representative kinetic rate, i.e., the preexponential factor, the order of reaction on oxygen concentration, and the activation energy. A reasonable value was chosen based on the data for chlorobenzene [16], which is one of the precursors of dioxin/furan, and has a relatively lower reaction rate than the other potential pollutants generated in incineration.

RESULTS AND INTERPRETATION

Design and Operation Variables

Given the appropriate modeling along with suitable boundary conditions, CFD codes provide numerical results of velocity, temperature and species concentration at each nodal point of the computational domain. Flow patterns and temperature distributions provide important information on the performance of the incinerator gas flow. Local recirculation zones are identified using 2-dimensional streamline plots [6]. Isotemperature contours are also useful to distinguish the locally isolated high and low temperature islands. Since the Good Combustion Strategy specifies the gas temperature and the oxygen concentration, the computed results can be directly used to check if the chosen incinerator would fulfill the requirements. A remaining question is, however, whether local pockets of low temperature and/or low oxygen can be tolerated, and how small these regioins must be. Incinerator designers would need quantifiable parameters for this decision. For 3-dimensional cases, 2-dimensional plots can only be drawn on specifically chosen slices of the computational domain. Figure 2 shows the velocity vector, oxygen concentration, and temperature distribution on the selected planes of the incinerator geometry. One can easily find out that the flow phenomena are typically three dimensional. It is worth noting that 3dimensional flow does not allow one to draw the streamlines, and local recirculation zones as defined from the 2-dimensional plane can not be extended for the 3-dimensional cases. One may call this zone a recirculation pocket. Through extensive evaluation of the computed flow performance of the incinerators, it has been concluded that the flow pattern is strongly affected by two major parameters: geometric shape and operational modes [6-12]. Incinerator designers may deduce optimal incinerator geometric shape and the flow configuration (typically the primary air distribution and the secondary air injection pattern) based on a comparative evaluation using CFD. These traditional interpretative observations of flow behavior are rather qualitative. It may seem strange to find out that the sophisticated computational data can only be used for qualitative evaluation.

Flow passage - Residence Time

Residence time of the gaseous volume element is one of the major concerns in incinerator performance. Typical legislation would impose a minimum residence time in the high temperature chamber of incinerator. The average gas residence time can be calculated from the mass flow rate, the temperature of the gaseous products, and the cross sectional area of the furnace volume. However, the residence time is affected by blockage and channeling, and therefore must be evaluated in terms of distribution [17]. For application to the incinerator, multiple inlet conditions need to be considered, because the gaseous products, originating from the same inlet point, can also have different residence times. Considering the pollutant formation/destruction kinetics point of view, the minimum residence time is as important as the average gas residence time. Meaningful measurements are not easy to find. Injection of trace surrogate gas and measurement at the exit, a noble method of superimposing small fluctuations of methane injection, called the pseudo-random binary sequence signal tracer technique have been proposed [17]. Using the CFD results, residence time of the imaginary volume element of the gaseous product can be calculated. Figure 3 shows the residence time distribution for the two different cases. Furthermore, Figure 4 shows that the local residence time is a function of the position where the gaseous product is released. It is seen that the residence times of the gaseous volume elements generated from the inlet region of the grate are considerably shorter than that of gas generated from the exit region of the grate. This is probably inevitable because of the geometric shape of the incinerator combustion chamber, but should be carefully evaluated along with various design alternatives. It is also noted that the residence time distribution is critically affected by the operating conditions such as the relative amount of air supplied to the various regions of the grate. Although the current model is designed to satisfy a 2.7 seconds average residence time in the secondary combustion chamber, considerable portions do not satisfy even the directives of 2 seconds as shown in Figure 3. For Case 1, 61% of the gas flow escapes the chamber in less than 2 seconds and 83% in less than 2.7 seconds. The difference stems from the flow pattern, such as the size and location of the recirculation pocket, which is affected by the

chamber shape and the air nozzle arrangement. For Case 2, 23% of the gas flow escapes in less than 2 seconds and 59% in less than 2.7 seconds. These results show the capability to enhance incineration performance by optimizing the design and operation parameters.

Flow Mixing

Flow mixing in incinerators is aimed at ensuring availability of oxidant to all organic material in the high temperature combustion environment. Good mixing minimizes the existence of fuel rich pockets of combustion products. Fresh air is supplemented to the flue gas in the form of overfire air (secondary air) jets. The Good Combustion Strategy requires that the overfire air injectors would be designed to penetrate and cover the entire furnace cross section[1]. Although it is intuitively straightforward to visualize the jet injection into the furnace, the penetration is a loosely defined terminology. The boundary of the jet is not clearly identified and the interaction between the adjacent or the opposing jets needs to be considered. Penetration of the jets into the cross-flow flue gas stream and interaction between the adjacent jets has been previously studied [8]. A typical configuration of the cross-jet flow mixing is three dimensional, and numerical simulation results are used to show the effects of the major design variables. The momentum ratio of the jet to the cross flow, and the relative spacing between the adjacent jets are found to be the major parameters which govern the flow and mixing. These investigations are also examples of the qualitative interpretation of the heavily quantitative results of the CFD modeling. Jet mixing was evaluated using the statistical mixing parameter , which is basically a spatial variable representing the standard deviation of individual local species concentration from the perfectly mixed concentration. Figure 5 shows values on a typical cross section (section B-B of Figure 2) in the combustion chamber. As shown by the gray level, an unmixed region is shown with darker shades. The penetration of the four jets of SA1 and SA2 are clearly visible. Since the jet nozzles of SA1 are directed downward at an angle of 45, air reaches further into the centerline of the chamber as it penetrates down. The penetration of SA2 is

shown as a darker gray region due to high oxygen concentration. In this arrangement, the interaction is limited to the jet-side half of the chamber. A designer may choose to increase the jet momentum to increase the penetration.

Thermal Decomposition

Thermal destruction of pollutants is an integral result of oxygen availability, environment temperature, and reaction time. When a volumetric element of pollutants is released from the waste bed on the grate, it would experience turbulent mixing first in the flame zone, followed by the air jet, and then the secondary combustion chamber. The flow mixing is determined by the local and instantaneous turbulent velocity components with statistical randomness, so the particles liberated from an identical location may follow different paths. The pollutant destruction is controlled by the instantaneous and local temperature and oxygen concentration. The thermal decomposition parameter incorporates the 3Ts by integrating the temperature and turbulent mixing with oxygen along the individual and turbulent particle trajectory. Figure 6 (a) and (b) show two typical histories of the temperature, oxygen concentration and destruction rate . The gas volumetric element (simply referred to as the gas particle) for Figure 6 (a) is released from the drying-zone (close to the entrance of the waste feed) of the grate. The particle temperature would increase through the flame zone until it is affected by the secondary air jet. The jet would increase the oxygen concentration sharply (with a large scale turbulence), but decrease the gas temperature. After approximately 1 sec. of residence time, the gas particle would experience gradual temperature decrease in the heat recovery boiler, but the oxygen concentration does not change appreciably, which means that the bulk oxidation reaction is almost complete. The gas particle in Figure 6 (b) is released from the combustion zone of the grate. The gas particle is rapidly heated up to a temperature higher than that of the particle in Figure 6 (a), but it experiences very strong mixing with the secondary air jet. It is clearly seen that the air jet increases the oxygen concentration but decreases the gas particle temperature. This particle experiences its first jet mixing at about 0.7 sec. and another one at about 1.3 sec., and it enters the fully mixed height at about 1.8 sec. The thermal decomposition parameter of a gaseous volume element decreases monotonically after release. Since the

10

parameter is a strong function of temperature, thermal destruction of the gaseous volume element in Figure 6 (b) is much faster, while the one in Figure 6 (a) is gradually approaching completion. If the volumetric element is trapped in a low temperature pocket, the destruction process would be frozen which will ultimately lead to the escape of pollutants before destruction. By statistically averaging values of the gaseous volume elements which travel across the exit plane of the combustion chamber, overall performance of thermal destruction in the entire incinerator can be discussed. Design and operation mode variables affect the incinerator performance and, by definition, low values of represent better incinerator performance. Figure 7 shows that the Case 2 shows values smaller than Case 1, which was expected based on the nozzle jet arrangement setting. Although the gaseous volume elements follow the turbulent flow path in the combustion chamber, they still retain the information related to the source origin, i.e., the location on the grate plane. Figure 8 shows the average values which have been calculated by statistically averaging on each location of the grate plane. values at the location close to the waste inlet are higher, which is the result of the low temperature environment around the drying zone of the grate. The middle region of the grate is the active combustion zone where the released gas temperature is higher, and is closer to the active flame. Potential pollutants emitted from this grate level are completely destroyed. It is important to note that values at locations closer to the side wall are higher than those from the symmetric centerline area. This reflects the fact that the gas temperatures are higher at the centerline. To reduce pollutant emissions, incinerator designers should look into improving the combustion environment, especially in the vivinity of potential cold spots (pockets).

CONCLUSION

Incinerator performance is investigated numerically and CFD analysis results are interpreted in such a way to validate the adequacy of the incinerator performance as related to

11

the in-furnace destruction of organic pollutants. Good combustion performance, usually stated as 2 seconds at 850 C and 6% O2,, can be validated by a comprehensive interpretation of the CFD results. Independent checks of the residence time, temperature and oxygen concentration are not adequate. The statistical distribution of residence time, and the local and instantaneous variation of temperature and oxygen concentration must be included in the consideration. Quantitative evaluation of CFD analysis of the complex gas flow in the incinerator is suggested by utilizing the following parameters: the residence time distribution, the mixedness parameter and the thermal decomposition parameter . Residence time must be carefully determined based on the gas inlet position, and the statistical variation must be carefully interpreted. It is helpful to check the average and the minimum residence time for the designated source locations. Mixing can be quantified by utilizing a local mixedness parameter . The effects of the secondary air jet penetration and the resultant mixing are discussed by using values for the furnace volume of interest. Thermal decomposition parameter integrates the temperature and oxygen availability over the individual trajectory of a gaseous volume element, which can be understood as a direct indication of the pollutant destruction capability. These performance parameters can be best utilized in comparative evaluation of alternatives in the design and operation mode of incinerators.

REFERENCES

1.

Schindler, P. J.; Nelson, L. P. Municipal Waste Combustion Assessment: Technical Basis for Good Combustion Practice; U.S. Environmental Protection Agency, 1989; US EPA-600/8-89-063.

2. 3. 4. 5.

Environmental Protection Act 1990 : Process Guidance Note IPR 5/3, Municipal Waste Incineration, HMSO, London, 1990. Guidelines for Preventing Dioxin Emissions; Japan, 1991. 17. Bundes-Immissionsschutzvorschrift; 17. BImSchV, Germany, 1990. Kilgroe, J. D.; Nelson, L. P.; Schindler, P. J. Combust. Sci. and Tech. 1990, 74, 223244

12

6. 7. 8. 9.

Choi, S.; Lee, J. S.; Kim, S. K.; Shin, D. H. 25th Symposium (International) on Combustion, The Combustion Institute: Irvine, 1994, 317-323 Kim, S. K.; Shin, D. H.; Choi, S. Combustion and Flame, 1996, 106, 241-251 Ryu, C. K.; Choi, S. J. Energy Research, in press Ryu, C. K.; Choi, S. Combust. Sci. and Tech., 1996, 119, 155-170 1991, 11, 249-261.

10. Nasserzadeh, V.; Swithenbank, J. D.; Scott, D. W.; Jones, B. Waste Management, 11. Nasserzadeh, V.; Swithenbank, J.; Jones, B. Combust. Sci. and Tech. 1993, 92, 389422. 12. Ravichandran, M.; Gouldin, F. C. Combust. Sci. and Tech. 1992, 85, 165-185. 13. Cavalho., M. G.; Farias, T.; Fontes, P. Fundamentals of Radiation Heat Transfer, ASME HTD, 1991, 160, 17-26. 14. Santos, A. D. Study of MSW Incinerator: Overall Operation and On-Site Measurements over the Grate, Royal Institute of Technology, Sweden, STEV-FBT91/14, 1991. 15. Magnussen, B. F.; Hjertager, B. H. 16th Symposium (International) on Combustion, The Combustion Institute, Cambridge, 1976, 719-727. 16. Rubey, W.; Torres, J.; Hali, D.; Graham, J. L.; Dellinger, B. Determination of the Thermal Decomposition Properties of 20 Selected Hazardous Organic Compounds, US EPA Cooperative Agreement CR-807815-01-0, 1985. 17. Nasserzadeh, V.; Swithenbank, J.; Lawrence, D.; Garrod, N.; Jones, B. J. of the Institute of Energy, 1995, 68, 106-120.

13

ABOUT THE AUTHORS

Donghoon Shin and Chang Kook Ryu Graduates students in Dept. of Mechanical Engineering, KAIST sdh@amigas.kaist.ac.kr ryu@amigas.kaist.ac.kr

Sangmin Choi Professor, Dept. of Mechanical Engineering, KAIST, 373-1 Kusong-dong, Yusong-gu, Taejon, 305-701, Korea Tel: 82-42)869-3030 Fax: 82-42)862-1284

email: smchoi@hanbit.kaist.ac.kr

14

Table 1. Operating conditions and boundary conditions on the grate in the model

incinerator.

waste feed rate excess air ratio primary air / secondary air waste constitution (W%) inlet on the grate primary air distributions heat release rate on the bed heat release rate above the bed water vaporization ratio temperature

12.5 ton/h 1.85 65 / 35 C: 0.20, H: 0.03, O: 0.14. N: 0.01, Water: 0.47, Ash:0.15 #1 25% 8% 9% 90% 591K #2 35% 42% 14% 10% 1788K #3 30% 17% 5% 0.0% 1405K #4 10% 3% 2% 0.0% 606K total 100% 70% 30% 100% -

15

LIST OF FIGURES Figure 1. Schematic of a typical 300 ton/day incinerator. (Width : 7.5m, Height : 14m, Full

depth : 5.7m)
Figure 2. Typical view of CFD results for Case 1 and Case 2 on the horizontal cross-section

A-A; (a) isoconcentration contour of oxygen, (b) isotemperature contour, (c) velocity vector.
Figure 3. Cumulative probability distribution of the residence time at the chamber exit. Figure 4. Averaged residence time of the gas products released from the bed (Case1). Figure 5. distribution on the horizontal cross-section B-B near the secondary air nozzle of

Case 1.
Figure 6. Time history of typical gas volume elements; (a) released from the drying-zone, (b)

released from the combustion zone.

Figure 7. Probability distribution of values at the exit. Figure 8. Averaged values of the gas products released from the bed (Case 1).

16

Figure 1. Schematic of a typical 300 ton/day incinerator. (Width : 7.5m, Height : 14m, Full

depth : 5.7m)
Front wall Outlet (Flue gas treatment)

Case 2

Secondary Combustion Chamber Secondary Air Jet, SA1 Waste Feeder Primary Combustion Chamber
Gra t e

Rear wall SA2 SA3

Pri ma ry air

Case 1

Ash Hopper

Figure 2. Typical view of CFD results for Case 1 and Case 2 on the horizontal cross section

A-A ; (a) isoconcentration contour of oxygen, (b) isotemperature contour, (c) velocity vector.

A'

B'

(a)

(b)

(c)

17

18

Figure 3. Cumulative probability distribution of the residence time at the chamber exit.

1.0

0.8

cumulative probability

83% 61%

0.6

59%
0.4

0.2

23 %

case 1 (avg.=2.63) case 2 (avg.=3.61)

0.0 0 1 2 2.7 3 4 5

residence time, (sec.)

Figure 4. Averaged residence time of the gas products released from the bed (Case1).

12

10

C L

0 1 2 0 2

er (m ) th e wa st e fe ed Dist an ce fro m

19

Residence time (sec)

D ep th

(m )

Figure 5. distribution on the horizontal cross section B-B near the secondary air nozzle of

Case 1.

A'

B'

Figure 6. Time history of typical gas volume elements; (a) released from the drying-zone, (b)

released from the combustion zone.

1800 0.20 1.00

temperature
1600

(oC)

oxygen

oxygen mass fraction

0.15

0.75

1400

temperature

0.10

0.50

1200

1000

0.05

0.25

800 0.0 0.5 1.0 1.5 2.0 0.00 2.5 0.00

time (sec.)

(a)
1800 0.20 1.00

temperature
1600

oxygen mass fraction

temperarture ( oC)

oxygen

0.15

0.75 0.50

1400 1200 1000 800 0.0 0.5 1.0 1.5 2.0

0.10

0.05

0.25

2.5

3.0

0.00 3.5

0.00

time (sec.)

(b)

20

Figure 7. Probability distribution of values at the exit.

0.08

0.06

probability

0.04

case 1(avg. = 0.000518) case 2(avg. = 0.000252)

0.02

0.00

10-2

10-4

10-6

10-8

10-10

10

Figure 8. Averaged values of the gas products released from the bed (Case 1).

500

400 ( x 10 -6 )

300

200

C L
D

0 100 1 2 0 0

ep th

(m )

er (m ) th e wa st e fe ed Dist an ce fro m

21