Converting moving-grate incineration from combustion to gasification – Numerical simulation of the burning characteristics

Waste Management 27 (2007) 645–655 www.elsevier.com/locate/wasman

Converting moving-grate incineration from combustion to gasification – Numerical simulation of the burning characteristics Yao Bin Yang *, Vida N. Sharifi, Jim Swithenbank Sheffield University Waste Incineration Centre (SUWIC), Department of Chemical and Process Engineering, Sheffield University, Mappin Street, Sheffield S1 3JD, UK Accepted 7 March 2006 Available online 26 May 2006

Abstract Waste incineration is a politically sensitive issue in the UK. The major current technology is based on direct combustion of wastes in a moving-grate furnace. However, general public opinion prefers non-direct burning technologies. Waste gasification is one of those nearest technologies available. By reducing the primary air-flow rate through the grate of a packed-bed system, operation of the existing solid-waste incineration equipment can be easily converted from combustion mode to gasification mode without major modification of the hardware. The potential advantages of this are lower dust carry-over in the flue gases, lower bed temperature (and therefore lower NOx formation in the bed), simplified gas-treatment procedures and lower running cost, among other benefits. The major disadvantages are, however, reduced throughput of the wastes and possibly higher carbon in the ash at exit. In this study, numerical simulation of both combustion and gasification of municipal solid wastes in a full-scale moving grate furnace is carried out employing advanced mathematical models. Burning characteristics, including burning rate, gas composition, temperature and burning efficiency as a function of operating parameters are investigated. Detailed comparisons between the combustion mode and gasification mode are made. The study helps to explore new incineration technology and optimise furnace operating conditions.  2006 Elsevier Ltd. All rights reserved.

1. Introduction Sustainable energy strategy requires diverse sources of energy production. Municipal solid wastes (MSW) constitute a major part of our daily life, and in the UK wastes generated amount 30 million tonnes per year. Although there is an opinion insisting that landfill is the best way for carbon sequestration (Marxsen, 2001), the available energy from these wastes could produce 2000 MW of electricity and 6000 MW of heat, equal to one major power station, i.e., about 10% of UK power requirements, with significantly reduced net CO2 emission to the environment. However, no more than 20% of MSW is being incinerated in the UK, with the rest discharged into landfills (Martin, 2004). So the potential for energy from wastes is very high. On the other hand, waste incineration is a politically sensi*

Corresponding author. Tel.: +44 114 2227523; fax: +44 114 2227501. E-mail address: [email protected] (Y.B. Yang).

0956-053X/$ - see front matter  2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2006.03.014

tive issue in the UK. There is a general misunderstanding in the public about waste incineration, which is perceived to be dirty, smelling and polluting the environment. While this is not true for modern incineration plants, some political and environmental groups demand that wastes not be incinerated but be treated and utilised in a ‘cleaner’ way. General public opinion also shows a preference for nondirect burning technologies. Waste gasification is one of the nearest technologies to incineration available. By reducing the primary air-flow rate through the grate of a packed-bed system, operation of the existing solid-waste incineration equipment can be easily converted from combustion mode to gasification mode without major modification to the hardware. Fig. 1 shows the extracted results that Yang et al. (2004) obtained for a fixed bed reactor burning simulated wastes at 20% moisture. It is seen that as the primary air flux decreases below 0.25 kg m2 s1, the combustion stoichiometry becomes lower than 1.0 and the burning of the bed

646

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

Nomenclature A Ar Av C Cfuel Cp Cmix Cw,g Cw,s Dig dp Eb Er Ev E0 Hevp H hs h0s Iþ x I x ka kd kr kv ks pg Qh Qsh

particle surface area, m2 m3 pre-exponent factor in char burning rate, kg m2 s1 pre-exponent factor in devolatilisation rate, s1 constant; molar fractions of species fuel concentration, kg m3 specific heat capacity of the gas mixture, J kg1 K1 mixing-rate constant, 0.5 moisture mass fraction in the gas phase moisture mass fraction at the solid surface dispersion coefficients of the species Yi, m2 s1 particle diameter, m black body emission, W m2 activation energy in char burning rate, J kmol1 activation energy in devolatilisation rate, J kmol1 effective diffusion coefficient evaporation heat of the solid material, J kg1 enthalpy, J kg1 convective mass transfer coefficient between solid and gas, kg m2 s1 convective heat transfer coefficient between solid and gas, W m2 K1 radiation flux in positive x direction, W m2 radiation flux in negative x direction, W m2 radiation absorption coefficient, m1 rate constants of char burning due to diffusion, kg m2 s1 rate constants of char burning due to chemical kinetics, kg m2 s1 rate constant of devolatilisation, s1 radiation scattering coefficient, m1 gas pressure, Pa heat loss/gain of the gases, W m3 thermal source term for solid phase, W m3

becomes sub-stoichiometric or fuel-rich. In traditional terminology, the conditions of the bed turn to gasification, producing combustible gases which exit from the bed top. The concentration of the major combustible gaseous species (CO + H2) increases as the primary air flux decreases. The potential advantages of this are lower dust carry-over in the flue gases (Beckmann et al., 1997), lower bed temperature (and therefore lower NOx formation in the bed), simplified gas-treatment procedures (fewer stages of dust collectors required) and lower running cost (slower deposition rate on convection heat exchanger tubes and hence prolonged life-cycle of material), among other benefits. As seen in the figure, bed temperature can decrease from a maximum of 1280 C at a primary air flux of 0.30 kg m2 s1 to below 1000 C at a primary air flux of 0.10 kg m2 s1. Lower bed temperature also reduces the

qr R Rmix S Ssg

Syig Syis t T U V VM x y Y es rb v v1 / k k0g

radiative heat flux, W m2 universal gas constant; process rate, kg m3 s1 mixing-rate of gaseous phase in the bed, kg m3 s1 stoichiometric coefficients in reactions conversion rate from solid to gases due to evaporation, devolatilisation and char burning, kg m3 s1 mass sources due to evaporation, devolatilization and combustion, kg m3 s1 source term, kg m3 s1 time instant, s temperature, K x velocity, m s1 y velocity, m s1; volatile matter in fuel, wt% co-ordinate in bed forward-moving direction, m co-ordinate in bed height direction, m mass fractions of individual species system emissivity Stefan–Boltzmann constant, 5.86 · 108 W m2 K4 remaining volatile in solid at time t ultimate yield of volatile void fraction in the bed thermal dispersion coefficient, W m1 K1 effective thermal diffusion coefficient, W m1 K1

Subscripts env environmental g gas phase i identifier for a component in the solid or gaseous phases p particle s solid phase

amount of heavy metal evaporated into the gaseous phase, and thus reduces heavy metal emission into the atmosphere. The major disadvantages are, however, reduced throughput of the wastes and possibly higher carbon in the ash at exit. Fig. 1 shows that as the primary air flux reduces from the stoichiometric point (around 0.25 kg m2 s1) to lower sub-stoichiometric region (<0.1 kg m2 s1), the burning rate falls from the maximum of 0.055 to below 0.03 kg m2 s1. In this study, numerical simulation of both combustion and gasification of municipal solid wastes in a full-scale moving grate furnace is carried out employing advanced mathematical models. Burning characteristics, including burning rate, gas composition, temperature and burning efficiency as a function of operating parameters, are inves-

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

tigated. A detailed comparison between the combustion mode and gasification mode is made. The study helps to explore new incineration technology and optimise furnace operating conditions. Mathematical modelling of a full-scale bed and furnace has been carried out by a number of researchers. Thunman et al. (2001) employed a four-layer model for the fuel bed characterised by the unreacted fuel layer, the reaction front, the char and surface layers. Conservation of energy and species was applied to each layer, and the free-room space above the bed was solved by commercial CFD software. The modelling prediction was compared to data obtained from a large-scale measurement operation. Shin and Choi (2000) carried out numerical simulations of a moving bed and used flux model to calculate the radiation heat transfer in the bed. Goh et al. (1998) also employed the theory of fuel layers to numerically simulate the combustion process of a burning waste bed. Peters (2003) summarised the previous mathematical models on packed bed combustion. Those models can be generally classified into four categories: continuous-medium models (Behrendt, 1992; Kru¨ll et al., 1998; Shin and Choi, 2000) where the solid bed was treated as a continuous medium; neighbouring-layers models (Adams, 1980; Goh et al., 1998) where the packed bed above the grate was divided into four layers representing fuel, drying, pyrolysis and ash; well-stirred reactor models (Beckmann and Scholz, 1995; Stapf et al., 1997) where the bed was simulated by a cascade of wellstirred reactors; and the 1d + 1d model (Wurzenberger, 2001) where a one-dimensional and transient single-particle model in spherical coordinates was implemented in a transient one-dimensional fuel-bed model. Yang et al. (2002a,b, 2003a,b, 2004) have carried out systematic studies on packed bed combustion of solids

employing advanced mathematical models and numerical techniques. Based on the model presented by Peters (1995), Yang et al. (2002a) first implemented the whole set of governing equations and reaction mechanisms to solve the distributions of temperature and species of both gaseous and solid phases inside a travelling bed, where a novel model for the mixing rate of released the volatile gases with the primary air was proposed. A computer code, FLIC, was developed based on this work. The code was then used and improved to solve for a full-scale bed, compared to measured results from a novel in-suite electronic device (Yang et al., 2002b; Ryu et al., 2004). The code also has been used to assess the effects of fuel moisture (Yang et al., 2003a), devolatilisation rate (Yang et al., 2003b) and primary air flux (Yang et al., 2004). The study carried out in this work is based on previous studies and on the experience obtained from extensively using the mathematical models and running the FLIC code. 2. The waste-to-energy plant The simulated bed and furnace geometry, as well as the operating parameters, are based on an existing waste-toenergy plant which has two incineration lines, a capacity of 2 · 23.5 tonnes of MSW per hour and 25 MW of electricity output. The furnace section of the plant is illustrated in Fig. 2. The furnace grate has a total length of 11.34 m and a width of 6.25 m, and the total height of the furnace and radiation shaft above the grate is about 20 m. The grate is a forward reciprocating grate and primary air is supplied to the underside of the grate in five separate zones, each with a different primary air-flow rate. Secondary air is injected through a series of nozzles above the burning waste bed, in the neck of the furnace, each with a diameter

3.5

sub-stoichiometric combustion

1200

3.0 bed temperature, oC

1000

2.5

800

combustion stoichiometry

0.1 x (CO+H2), vol% 600 400

1.5

4

10 x burning rate, kgm -2s -1

1.0

200 0 0.00

2.0

0.5

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

combustion stoichiometry, CO+H2 concentration

Burning rate, bed temperature

1400

647

0.0 0.45

Primary air flux, kgm-2s-1 Fig. 1. Combustion stoichiometry, burning rate, CO + H2 concentration and bed temperature as a function of primary air flux (data are extracted from Yang et al. (2004)).

648

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

of 90 mm. The simulated bed and furnace operating conditions are given in Table 1. Six different cases are simulated. Case #0 is the baseline case of normal combustion conditions where the overall air to fuel stoichiometric ratio is around 1.1 in the bed (combustion mode) and 1.8 for the whole furnace (including the secondary air). Case #1 reduces the primary air supply by 50% and the air to fuel stoichiometric ratio in the bed is reduced to 0.55 (gasification mode) while maintaining the total air supply to the furnace unchanged. Case #2 is similar to Case #1 except that the bed height is kept constant, i.e., the same as the bed height at the feeding point of the bed. Normally, the bed height decreases as the material burns along the grate. However, this situation can be changed by employ-

Hot Flue Gases to Heat Exchanger Waste In

ing slower moving speed in the latter sections of the grate so that the burning fuels accumulate locally, offsetting the effect of mass loss (which causes the decrease in bed height), and a constant bed height can be achieved. By doing this, the residence time of the fuel on the grate can be significantly prolonged, depending on the specific burning conditions. Case #3 further reduces the primary air to 30% of the baseline supply and makes the air to fuel stoichiometric ratio in the bed only 0.33. No simulation in the radiation shaft has to be made and so the use of secondary air is not applicable. Case #4 is similar to Case #3 except that the bed height is maintained constant along the grate length. Case #5 is the case with optimised operating conditions for the gasification mode. The waste feed rate is slightly reduced (90% of the combustion mode) and the air to fuel stoichiometric ratio in the bed is around 0.5. The composition of typical municipal waste feedstock is shown in Table 2. 3. Mathematical modelling

Radiation shaft

The mathematical models used in this work are based on the authors’ previous works (Yang et al., 2002a,b, 2003a,b, 2004; Ryu et al., 2004); all of the equations and reaction rates are summarised in Table 3. Details of the models, parameters used and numerical procedures are presented in the relevant references. Basically, the whole incineration process consists of moisture evaporation, devolatilisation, gas phase combustion and char burnout. Parameters across the bed width are assumed to be uniform so that the problem is simplified to a two-dimensional one. The whole bed is subdivided into 460 · 700 cells and after ignition at the bed top, a single flame front is formed and travels down towards the bed bottom. The whole process is governed

Secondary Air Ram feeder 1

2

3

4

Forward reciprocating grate

5

Ash

Primary Air Fig. 2. Furnace section of a large waste-to-energy plant.

Table 1 Simulated bed and furnace conditions Case ID

#0 #1 #2 #3 #4 #5

Waste rate, tonnes/h

23.5 23.5 23.5 23.5 23.5 21.0

Primary air-flow rate (kg1)/temperature (C)

Secondary air at 20 C, kg/s

Grate zone

Frontwall

Rearwall

7.45 13.0 13.0 n/a n/a 12.2

7.45 13.0 13.0 n/a n/a 12.2

1

2

3

4

5

3.35/110 1.67/110 1.67/110 1.12/110 1.12/110 1.51/110

6.69/110 3.35/110 3.35/110 2.23/110 2.23/110 3.01/110

7.81/20 3.90/20 3.90/20 2.60/20 2.60/20 3.51/20

3.35/20 1.67/20 1.67/20 1.12/20 1.12/20 1.51/20

1.12/20 0.56/20 0.56/20 0.37/20 0.37/20 0.50/20

Constant bed height

No No Yes No Yes Yes

Table 2 Composition of typical waste feedstock Proximate analysis (wt%)

Ultimate analysis (wt%, daf)

LCV (MJ/kg)

Moisture

Volatile matter

Fixed carbon

Ash

C

H

O

N

S

Other

28

40.8

7.2

24

49.6

7.3

38.3

1.5

0.4

2.9

9.05

Table 3 Summary of the transport equations and reaction rates for the mathematical models Reference Process rate equations

Rm = Aphs(Cw,s  Cw,g)

Devolatilisation

when T s ¼ 100  C   Ev Rv ¼ ð1  /Þqs k v ðv1  vÞ k v ¼ Av exp  RT s

Combustion of volatiles

CO þ

Char gasification

Continuity x-Momentum y-Momentum Species

Energy

m 2 O2

1 2 O2

! mCO þ

! CO2

n 2 H2

RCm Hn ¼ 59:8T g P 11

RCO ¼ 1:3  10

17

1 2 O2

0:3

Badzioch et al. (1970) 0:5 expð12200=T g ÞC C C O2 m Hn

0:5 expð62700=T g ÞC CO C 0:5 H2 O C O2

0:85 1:42 expð20500=T g ÞC H C O2 2 

! H2 O RH2 ¼ 3:9  10  h i C D ð1/Þ2=3 V ð1/Þ1=3 fuel R ¼ min½Rkinetic ; Rmix  Rmix ¼ C mix qs 150 g d 2 / þ 1:75 g d p / ; S OO2 min CS fuel 2 p  6420 CO CðSÞ þ aO2 ! 2ð1  aÞCO þ ð2a  1ÞCO2 CO2 ¼ 2500 exp  T     Er k r ¼ Ar exp  RT R4 ¼ Ap C O2 = k1r þ k1d s oðqg /Þ oðqg U g /Þ oðq V /Þ þ goy g ¼ S sg ot þ ox op oðqg U g /Þ oðq U U /Þ oðq U V /Þ þ g oxg g þ g oyg g ¼  oxg þ F ðU g Þ ot op oðqg V g /Þ oðq V U /Þ oðq V V /Þ þ g oxg g þ g oyg g ¼  oyg þ F ðV g Þ ot   oðqg Y i;g /Þ oðq U Y /Þ oðq V Y /Þ oðq Y /Þ o þ g oxg i;g þ g oyg i;g ¼ ox Di;g goxi;g ot Dig = E0 + 0.5dp|Vg| oðqg H g /Þ ot

þ

oðqg U g H g /Þ ox

þ

oðqg V g H g /Þ oy

Continuity Species Energy

Radiation heat transfer

dI  xi dxi

Hautman et al. (1981) Yang et al. (2002a) Arthur (1951) Gray et al. (1974), Field (1969)

  oðq Y /Þ þ oYo Di;g goYi;g þ S Y i;g

    oT oT o ¼ ox kg oxg þ oyo kg oyg þ Qh

Wakao and Kaguei (1982) Peters (1995)

kg = k + 0.5dp|Vg|qgCpg

Wakao and Kaguei (1982)

oðð1/Þqs Þ ot

Yang et al. (2003b)

sU sÞ sV sÞ þ oðð1/Þq þ oðð1/Þq ¼ S s;g ox oy

Us = f(X), pre-defined oðð1/Þqs y i;s Þ oðð1/Þqs U s Y i;s Þ oðð1/Þqs V s Y i;s Þ þ þ ¼ S Y i;s ot ox oy  oT s  o  oT s  oqrx oqry oðð1/Þqs H s Þ oðð1/Þqs U s H s Þ oðð1/Þqs V s H s Þ o þ þ ¼ ox ks ox þ oy ks oy þ ox þ oy þ Qsh ot ox oy

1 1 ¼ ðk a þ k s ÞI  xi þ 2N k a Eb þ 2N k s

ks = 0

Howard et al. (1973)

Peters (1995)

0

Solid phase conservation equations

Siminski et al. (1972) Y.B. Yang et al. / Waste Management 27 (2007) 645–655

Cm Hn þ

H2 þ

Yang et al. (2002a)

Ap bh0s ðT g T s Þþes rb ðT 4env T 4s Þc H evp

Rm ¼

Gas phase conversation equations

when Ts < 100 C

Moisture evaporation

PN

ka ¼

þ i¼1 ðI xi

 d1p

þ I xi Þ;

lnð/Þ

i ¼ 1; N

Gosman and Lockwood (1972) Shin and Choi (2000)

649

650

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

by reaction kinetics, heat transfer (especially radiation heat transfer) and mechanical movement of the grate. Both gas-phase and solid-phase transport equations have to be solved in terms of species, energy and velocity. The kinetic rate of char combustion is taken as (Yang et al., 2005) kr = Acexp(Ec/R), where Ac = 3 kg/ (m2 s kPa) and Ec/R = 10,300 K. The devolatilisation rate parameters for the waste are taken as Av = 3000 s1 and Ev/R = 7950 K. For combustion of gas-phase species such as CmHn, CO and H2 inside the bed, the reaction rate not only depends on kinetics (temperature related) but also the mixing intensity between the reactants (under-grate air and combustible gases). Cmix in the mixing rate equation is taken as 0.5. The whole computation area is divided into bed-combustion-zone, and the rest includes over-bed combustion chamber and radiation shaft. The conservation equations listed in Table 3 are solved numerically for the bed-combustion-zone using the FLIC code that Yang et al. (2002a,b, 2003a,b, 2004, 2005) developed for packedbed waste incineration calculations. For the over-bed combustion chamber and radiation shaft section, full geometry (three-dimensional) calculations were carried out using the FLUENT code. Conservation equations for momentum, heat and mass transfer are in similar forms. Results from the inside-bed calculation are used as the boundary or input conditions for the calculations of the over-bed space. A standard k–e two-equation model for turbulence and a discrete-ordinates (DO) model for radiation are selected. The radiation absorption coefficient is calculated as a function of characteristic cell size and gas concentrations (see FLUENT manual for details). 4. Results and discussion 4.1. The baseline case and the combustion mode Case #0 is the baseline case where the simulated plant operates at typical combustion mode. The simulated results are shown in Fig. 3. It is seen that the bed temperature gets hotter as the waste feed moves along the grate. The hottest part is in the latter part of the bed where char begins to burn intensively after the completion of the devolatilisation process. The maximum bed temperature reaches 1500 K or 1227 C. The ash temperature at the bed-exit ranges from 400 to 900 K (127–627 C). The fixed carbon increases from the initial 7% to a maximum 23% of the solid mass as the waste is deprived of moisture and volatile matter. At the same time, formed char begins to burn, starting from half of the bed length. The fixed carbon mass-percentage then decreases gradually along the bed and at 9.5 m of the grate length, the char burning process is completed. The temperature profile in the over-bed furnace space indicates that the hottest region is located a short distance just above the bed-top in the combustion chamber, the maximum temperature not exceeding the temperature

inside the bed. With the injection of secondary air, temperature in the radiation shaft gradually falls as heat in the gaseous flow is transferred to the water-cooled furnace wall. The gas temperature at the furnace exit is around 1130 K or 857 C. Secondary air jets from the front and rear walls impinge and merge at the furnace neck (Fig. 3d), forming a narrow stream of flow of high velocity at the central line of the furnace. Large recirculation zones are found both near the front wall and the rear wall. 4.2. The gasification mode and the burning characteristics The remaining five cases are devoted to the gasification mode where the primary air supply is only 30–50% of the combustion mode. Fig. 4 shows the burning rate as a function of position along the grate length. Four cases of gasification mode, as well as the baseline case for combustion mode, are illustrated. It is seen that sharp rise of the burning rate occurs from 2.5 to 3 m of the grate length position, indicating the ignition of the fresh waste feed. All of the investigated cases have similar ignition distances except for Case #1 where the ignition point is 0.5 m later than the others. For the main burning stage following the ignition, the burning rate is apparently case sensitive. The baseline case has a relatively high burning rate. As the primary air is reduced to 50% (Case #1), the burning rate is greatly reduced. However, by employing a constant bed height (Case #2), the burning rate is significantly increased, even exceeding the baseline case for the combustion mode. As the primary air flow is further reduced to 30% (Case #3), the burning rate obtained is the lowest, although employing a constant bed height (Case #4) slightly increases the burning rate in the latter part of the bed. Following the primary stage is the final char burning period with a much slower mass loss rate. Ignition is delayed for Case #1 compared to the other three cases in the gasification mode (Case #2, Case #3 and Case #4). This can be explained by the following reasoning. When waste is fed into the grate bed, it first undergoes moisture evaporation before the release of volatile gases (ignition). Moisture evaporation needs heat input to the fuel and normally the heat is provided by over-bed radiation. But the heat input to the wet fuel can be offset by strong local air flow, which carries away part of the heat from the fuel before ignition, so the moisture evaporation process is delayed and hence the start of the ignition. That is what happens to Case #1 where the air-flow rate through zone 1 of the grate is the highest. The CO concentration in the flue gases exiting the grate bed top is shown in Fig. 5 as a function of position along the grate length. For the combustion mode of the baseline case (Case #0), the CO level is around 2.5% for most of the combustion period. For the gasification mode of Case #1, the CO level is around 4–7% for the devolatilisation period and 2% for the char burning period. For

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

651

Fig. 3. Profiles for the baseline case: (a) waste-bed temperature; (b) waste-bed fixed carbon; (c) temperature in the over-bed combustion chamber and radiation shaft; (d) gas velocity.

Case #2, CO is around 5% during the devolatilisation period but increases sharply to about 15% during the char burning stage. Case #3 has CO from 5% to 7% for most

of the burning and for Case #4 a similar profile is obtained except that CO is back to 5% for the char burning stage.

Burning rate

0.30 0.25 -2

burning rate, kgm s

-1

Case 0# Case 1#

0.20

Case 2# Case 3#

0.15

Case 4#

0.10 0.05 0.00 0

2

4

6

8

10

Horizontal distance along grate (x), m Fig. 4. Burning rate as a function of position along the grate length.

12

652

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

CO concentration in the flue gases exiting the bed

20.0 17.5

Case 0#

15.0

Case 1#

CO, vol%

Case 2# 12.5

Case 3# Case 4#

10.0 7.5 5.0 2.5 0.0 0

2

4

6

8

10

12

Horizontal distance along grate (x), m Fig. 5. CO concentration in the flue gases exiting the bed top.

Overall, Case #2 results in the highest CO level in the flue gases exiting the grate bed. CO comes from both the devolatilisation process and the combustion of CmHn (CmHn + O2 ! CO + H2). Case #2 has the highest CO concentration from the grate bed top because much of the CmHn is converted to CO due to the higher local air flow. 4.3. The optimum gasification mode The optimum gasification mode (Case #5) is based on Case #2 but the waste feed rate is reduced to 90% and the primary air flow to 45% of the combustion mode. This is done to get the unburned carbon below 2% at a distance 2 m from the ash exit. For Case #2, the distance is 1 m. Fig. 6 shows the details of the gasification profiles both inside the bed and in the over-bed furnace space. It is seen that the bed is much hotter in the latter part of the bed (Fig. 6a) and there is a large area where the temperature level is around 1500 K (1227 C). This is due to the increased bed height holding up the combustion heat. This is beneficial to the char gasification reactions which require higher temperatures. Fig. 6b shows that the char gasification reactions occur in a band roughly 0.5 m in width, judging from the thickness of the gradient of the fixed carbon concentration. It also suggests that the gasification front travels upwards, in the same direction of the primary air flow. Fig. 6c shows the temperature profile in the over-bed combustion chamber and radiation shaft. There are two hot zones: one in the upper part of the combustion chamber just below the front wall secondary air jets and another near the bed exit. The former indicates strong combustion reactions in the region and the latter is the result of char burning near the bed exit, which turns from gasification mode to combustion mode as the char gasification approaches completion. For the furnace space above the bed top, Fig. 6d shows that the secondary air jets from the rear wall impinge the

secondary jets from the front wall and overcome them, forming a strong upward gas flow next to the front wall. Adjacent to the rear wall, a recirculation zone roughly one-third of the radiation shaft space is observed. The graph also suggests some recirculation of gases in the combustion chamber immediately above the bed. This compares quite differently to Case #0 of the combustion mode (Fig. 3d), where the gases flow centrally in the radiation shaft and there is no obvious recirculation zones in the combustion chamber below the furnace neck. The flow pattern with Case #5 is due to strong secondary air flow (80% of the total combustion air) and weak primary air flow (20% of the total combustion air). The recirculation of gases in the over-bed combustion chamber benefits mixing of the gasification gases (high proportion of combustible species) from the bed top with the strong secondary jets and enhances the burning of these gases. 4.4. Effects of operating parameters The effects of operating parameters are summarised in Table 4. Thermal energy efficiency is defined as the percentage of chemical energy stored in the wastes which is released to the gas-phase either in thermal form or chemical form. For Cases #1–4, computation was not performed in the over-bed combustion chamber and radiation shaft, as the poor conditions in the bed make it meaningless to perform such a computation. So furnace outlet values were not presented for those cases. It is seen that lower primary air flux helps to reduce dust emissions from the bed into the gas stream and benefits the gas cleaning system in the downstream. Cases #3 and #4 have the lowest primary air flux but the unburned carbon in the ash is over 15%. The energy efficiency is around 90%, which is far below the 98% level required normally by the environmental agencies. Cases #2 and #5 have the longest waste residence time in the bed, nearly two times that of the combustion mode (Case #0). This is achieved by reducing the bed speed as the waste feed moves forward along the grate so that a

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

653

Fig. 6. Profiles of the optimum gasification case (Case #5).

Table 4 Summary of the effects of operating conditions on gasification Case#0

Case #1

Case #2

Case #3

Case #4

Case #5

Waste bed Primary air flux (kg m2 s1)

0.32

0.16

0.16

0.11

0.11

0.14

Waste residence time (min)

80

80

150

80

127

150

Unburned carbon in ash (wt%)

3.0

13.3

2.5

19.3

14.4

1.5

Thermal energy efficiency (%)

98.5

93.2

99

88.1

91.4

99.8

Primary gasification stage Gasification rate (kg m2 s1) CO + CmHn + H2 at bed top (vol%)

0.16 7.5

0.12 15

0.16 16.5

0.08 18.7

0.13 18.0

0.15 17

1.8 · 102 15.2

6.9 · 103 1.5

8.7 · 103 5.0

1.5 · 102 16

Char gasification stage Gasification rate (kg m2 s1) CO at bed top (vol%) Gas temperature (K) CO (ppm) CmHn (ppm) H2 (ppm)

1.7 · 102 1.0 · 102 2.3 1.6 Furnace outlet (radiation shaft exit) 1130 933 212 695

1100 283 73 31

654

Y.B. Yang et al. / Waste Management 27 (2007) 645–655

constant bed height is obtained. The longer residence time especially helps the gasification of char which takes a considerable proportion of the burning process (1/3–1/2 of the total grate length). Table 4 shows that the char reaction rate is highest with Cases #0 and #2. Cases #2 and #5 have the highest CO product from the char gasification, around 15–16% by volume, which is 3–10 times higher than the CO product for the other cases. The char reaction rate is, nonetheless, much lower than the devolatilisation rate. The latter constitutes the major part of mass loss during the primary stage. For the optimum gasification mode (Case #5), the primary mass loss rate (0.15 kg m2 s1) is only a bit lower than the combustion mode (Case #0), but the primary air flux employed is significantly lower, i.e., 43% of the combustion mode. As a result, the combustible species in the gaseous products from the bed make up 17% of the gas mass, two times higher than for the combustion mode (Case #0). For the gasification mode, both Cases #2 and #5 produce satisfactorily low carbon in the ash and over 99% of the energy efficiency. It is very interesting to compare the gas temperature and species concentrations in the furnace outlet between the combustion mode and the optimum gasification mode. The gasification mode produces lower exit temperature and lower concentration of residual un-burned combustible gases. This is due to the higher secondary air flux (103 m/s compared to 58 m/s), and the impingement of the secondary air jets from the front and rear walls causes very intensive turbulence mixing in the furnace neck region. Earlier completion of gaseous combustion means that more heat is transferred to the cooled furnace walls. 5. Conclusions Mathematical modelling has been employed to simulate the burning process of a furnace in a waste-to-energy plant. The operating parameters cover typical combustion mode as well as gasification mode where the primary air-flow rate is significantly reduced. The FLIC code, which has been used in many applications, was employed to solve the governing equations for fluid flow, and heat and mass transfer inside a moving packed bed. Different cases were compared. The main conclusions are: 1. By employing the constant bed height technique, the same burning rate of waste can be achieved with only half of the primary air supply. 2. At the optimum gasification conditions, waste residence time in the bed is twice as long as that of the combustion mode, ensuring complete gasification of the waste and 99% of thermal energy efficiency. 3. Higher velocity of the secondary air jets at the gasification mode results in strong mixing of the combustible gases coming from the waste bed with the oxygen from the secondary air, producing much lower CO and hydrocarbon concentrations at the furnace outlet.

Acknowledgement The authors thank the UK Engineering and Physical Sciences Research Council (EPSRC) for the financial support of this project. References Adams, T., 1980. A simple fuel bed model for predicting particulate emissions from a wood waste boiler. Combust. Flame 39, 225–239. Arthur, J.R., 1951. Reactions between carbon and oxygen. Trans. Faraday Soc. 47, 167–178. Badzioch, S., Hawksley, P.G.W., Peter, C.W., 1970. Kinetics of thermal decomposition of pulverized coal particles. Ind. Eng. Chem. Process Des. Develop. 9, 521–530. Beckmann, M., Scholz, R., 1995. Simplified mathematical model of combustion in stoker systems. In: Proceedings of the Third European Conference on Industrial Furnaces and Boilers, Porto, Potrugal, pp. 61–70. Beckmann, M., Scholz, R., Wiese, C., Davidovic, M., 1997. Optimization of gasification of waste materials in grate systems. Available from: . Behrendt, Th., 1992. Thermodynamische Modellierung des Betriebsverhaltens einer Hausmu¨llverbrennungsanlage am Beispiel TAMARA. Fortschrittberichte Reihe 15: Umwelttechnik. VDI Verlag, Du¨sseldorf. Field, M.A., 1969. Rate of combustion of size-graded fractions of char from a low rank coal between 1200 K and 2000 K. Combust. Flame 13, 237–252. Goh, Y.R., Siddall, R.G., Nasserzadeh, V., Zakaria, R., Swithenbank, J., Lawrence, D., Garrod, N., Jones, B., 1998. Mathematical modelling of the waste incinerator burning bed. J. Inst. Energy 71, 110–118. Gosman, A.D., Lockwood, F.C., 1972. Incorporation of a flux model for radiation into a finite-difference procedure for surface calculations. In: Proceedings of the 14th Symposium (International) on Combustion. The Combustion Institute, Pittsburgh, PA, pp. 661–671. Gray, D., Cogoli, J.G., Essenhigh, R.H., 1974. Problems in pulverised coal and char combustion. Adv. Chem. Ser. 131, 72–91. Hautman, A.N., Dryer, F.L., Schlug, K.P., Glassman, I., 1981. A multiple-step overall kinetic mechanism for the oxidation of hydrocarbons. Combust. Sci. Technol. 25, 219. Howard, J.B., William, G.C., Fine, D.F., 1973. Kinetics of carbon monoxide oxidation in postflame gases. In: Proceedings of the 14th Symposium (International) on Combustion. The Combustion Institute, Pittsburgh, PA, pp. 975–986. Kru¨ll, F., Kremer, H., Wirtz, S., 1998. Feuerraumsimulation einer Mu¨llverbrennungsanlage bei gleichzeitiger Simulation der Verbrennung auf dem Rost. VDI Bericht 1390, Modellierung und Simulation von Dampferzeugern und Feuerungen, pp. 199–212. Martin GmbH, 2004. Global use and future prospects of waste-to-energy technologies. In: Waste-to-Energy Research and Technology Council, Fall Meeting Columbia University, October 7–8, 2004. Marxsen, C.S., 2001. Potential world garbage and waste carbon sequestration. Environ. Sci. Policy 4, 293–300. Peters, B., 1995. A detailed model for devolatilisation and combustion of waste material in packed beds. In: Proceedings of the 3rd European Conference on Industrial Furnaces and Boilers (INFUB), Lisbon, Portugal, 18–21 April, 1995. Peters, B., 2003. Thermal Conversion of Solid Fuels. WIT Press, Southampton, Boston. Ryu, C., Yang, Y.B., Sharifi, V.N., Swithenbank, J., 2004. Thermal reaction modelling of a large municipal solid waste incinerator. Combust. Sci. Technol. 176, 1891–1907. Shin, D., Choi, S., 2000. The combustion of simulated waste particles in a bed. Combust. Flame 121, 167–180.

Y.B. Yang et al. / Waste Management 27 (2007) 645–655 Siminski, V.J., Wright, F.J., Edelman, R.B. Economos, C., Fortune, O.F., 1972. Research on methods of improving the combustion characteristics of liquid hydrocarbon fuels. AFAPL TR 72–74, Vols. I and II, Air Force Aeropropulsion Laboratory, Wright Patterson Air Force Base, OH. Stapf, M., Kircherer, A., Kolb, Th., Wo¨lfert, A., Seifert, H., 1997. Verbrennung im Drehrohrofen: Modellierung, Betriebs – und Technikumsversuche, Volume 15 of VDI-Berichte 1313. VDI Verlag, Du¨sseldorf. ˚ mand, L.-E., Ghirelli, F., Leckner, B., 2001. Thunman, H., A Activity 3.1.3 – Modelling and verifying experiments on the whole furnace, in principles and models of solid fuel combustion, Ph.D. Thesis, Chalmers University of Technology, Go¨teborg, Sweden. Wakao, N., Kaguei, S., 1982. Heat and Mass Transfer in Packed Beds. Gorden & Breach Science Publishers, London, New York. Wurzenberger, J.C., 2001. A combined packed bed and single particle model applied to biomass combustion, Ph.D. Thesis, Graz University of Technology.

655

Yang, Y.B., Goh, Y.R., Zakaria, R., Nasserzadeh, V., Swithenbank, J., 2002a. Mathematical modelling of MSW incineration on a travelling bed. Waste Manage. 22, 369–380. Yang, Y.B., Nasserzadeh, V., Goodfellow, J., Goh, Y.R., Swithenbank, J., 2002b. Parameter study on the incineration of municipal solid waste fuels in packed beds. J. Inst. Energy 75, 66–80. Yang, Y.B., Yamauchi, H., Sharifi, V.N., Swithenbank, J., 2003a. Effect of moisture on the combustion of biomass and simulated solid waste in a packed bed. J. Inst. Energy 76, 105–115. Yang, Y.B., Yamauchi, H., Nasserzadeh, V., Swithenbank, J., 2003b. Effects of fuel devolatilisation on the combustion of wood chips and incineration of simulated municipal solid wastes in a packed bed. Fuel 82, 2205–2212. Yang, Y.B., Swithenbank, J., Sharifi, V.N., 2004. Effect of air flow rate and fuel moisture on the burning behaviours of biomass and simulated solid waste in a packed bed. Fuel 83, 1553–1562. Yang, Y.B., Goodfellow, J., Sharifi, V.N., Swithenbank, J., 2005. Study on the transient process of waste fuel incineration in a full-scale moving-bed furnace. Combust. Sci. Technol. 177 (1), 127–150.