Group Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Title: 11.5.1 - Characterization of Inhomogeneous Non-Spherical Alternative Fuel Particles for the Cement Manufacturing Process
ALL VOLUMES CITATION THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00102023/00287
 Material Information
Title: 11.5.1 - Characterization of Inhomogeneous Non-Spherical Alternative Fuel Particles for the Cement Manufacturing Process Particle-Laden Flows
Series Title: 7th International Conference on Multiphase Flow - ICMF 2010 Proceedings
Physical Description: Conference Papers
Creator: Perumalsamy, S.
Schroer, K.
Yilmaz, H.
Oeljeklaus, G.
Görner, K.
Klasen, T.
Bluhm-Drenhaus, T.
Scherer, V.
Zeki, A.
Schäfer, S.
Publisher: International Conference on Multiphase Flow (ICMF)
Publication Date: June 4, 2010
 Subjects
Subject: two-phase flow
non-spherical particles
Lagrangian approach
turbulence
 Notes
Abstract: Solid recovered fuels (SRF) are a highly heterogeneous mixture generated from high calorific fractions of nonhazardous waste materials intended to be fired in existing cement plants and industrial furnaces. They are composed of a variety of materials that include plastics, paper, wood and textile wastes which are offered in particle size up to several centimetres. These materials are commonly irregular (3D and 2D) in shape and size. It is therefore difficult to predict their aerodynamic properties, pyrolysis and burnout behaviour 1 as well as their material properties as a mixture. In view of this, the ability to characterize SRF particles will immensely help in any modelling of particles and their combustion behaviour in a cement plant. Bulk mixture of SRF particles were shorted out to six fraction based on their difference in settling velocity for which pyrolysis at different temperature was measured. To determine the aerodynamic properties of the particles which are essential for air flow and thus promote the distribution in the calcinator, experimental work was carried out to determine the aerodynamic diameters, particle projection area, apparent density and maximum minimum axial length of representative particles 2 which are selected manually from each fraction. Simulations were performed in commercial CFD code, based on these measured properties such as projection area with different degrees of non-sphericity 3, aerodynamic diameters and pyrolysis of each fraction at different temperature. As the first stage, a model for nonspherical particles tracking in a simple nonuniform two-phase flow is derived in this paper, in which forces, including inertia, drag, gravity, lift, a force due to pressure gradient and virtual-mass force are taken into account in particle force balance. The drag coefficient is calculated using the correlation of Haider and Levenspiel (1989) and Ganser (1993) and combustion behavior such as pyrolysis and burnout of non-spherical SRF particles were simulated.
General Note: The International Conference on Multiphase Flow (ICMF) first was held in Tsukuba, Japan in 1991 and the second ICMF took place in Kyoto, Japan in 1995. During this conference, it was decided to establish an International Governing Board which oversees the major aspects of the conference and makes decisions about future conference locations. Due to the great importance of the field, it was furthermore decided to hold the conference every three years successively in Asia including Australia, Europe including Africa, Russia and the Near East and America. Hence, ICMF 1998 was held in Lyon, France, ICMF 2001 in New Orleans, USA, ICMF 2004 in Yokohama, Japan, and ICMF 2007 in Leipzig, Germany. ICMF-2010 is devoted to all aspects of Multiphase Flow. Researchers from all over the world gathered in order to introduce their recent advances in the field and thereby promote the exchange of new ideas, results and techniques. The conference is a key event in Multiphase Flow and supports the advancement of science in this very important field. The major research topics relevant for the conference are as follows: Bio-Fluid Dynamics; Boiling; Bubbly Flows; Cavitation; Colloidal and Suspension Dynamics; Collision, Agglomeration and Breakup; Computational Techniques for Multiphase Flows; Droplet Flows; Environmental and Geophysical Flows; Experimental Methods for Multiphase Flows; Fluidized and Circulating Fluidized Beds; Fluid Structure Interactions; Granular Media; Industrial Applications; Instabilities; Interfacial Flows; Micro and Nano-Scale Multiphase Flows; Microgravity in Two-Phase Flow; Multiphase Flows with Heat and Mass Transfer; Non-Newtonian Multiphase Flows; Particle-Laden Flows; Particle, Bubble and Drop Dynamics; Reactive Multiphase Flows
 Record Information
Bibliographic ID: UF00102023
Volume ID: VID00287
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: 1151-Perumalsamy-ICMF2010.pdf

Full Text



7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


Characterization of Inhomogeneous Non-Spherical Alternative Fuel Particles for
the Cement Manufacturing Process


Solairaj Perumalsamy, Kay Schroer, Hiiseyin Yilmaz, Gerd Oeljeklaus,

Klaus GOmrer*, Thomas Klasen** Tobias Bluhm-Drenhaus, V. Scherer,t

Asian Zeki, S. Schafer
LUAT (Environmental Process Engineering and Plant Design), University of Duisburg-Essen,
45141 Essen, Germany
**InPro-Consult GmbH, 45145 Essen, Germany
SDepartment of Energy Plant Technology, Ruhr-Universitat Bochum, 44780 Bochum, Germany
40474 Dusseldorf, Germany
German Cement Works Association, 40476 Duesseldorf, Germany
solairaj.perumalsamy @stud.uni-due.de and klaus.goerer @uni-due.de
Keywords: Two-phase flow; Non-spherical particles; Lagrangian approach; Turbulence




Abstract

Solid recovered fuels (SRF) are a highly heterogeneous mixture generated from high calorific fractions of nonhaz-
ardous waste materials intended to be fired in existing cement plants and industrial furnaces. They are composed of
a variety of materials that include plastics, paper, wood and textile wastes which are offered in particle size up to
several centimetres. These materials are commonly irregular (3D and 2D) in shape and size. It is therefore difficult
to predict their aerodynamic properties, pyrolysis and burnout behaviour [1] as well as their material properties as a
mixture. In view of this, the ability to characterize SRF particles will immensely help in any modelling of particles
and their combustion behaviour in a cement plant. Bulk mixture of SRF particles were shorted out to six fraction
based on their difference in settling velocity for which pyrolysis at different temperature was measured. To determine
the aerodynamic properties of the particles which are essential for air flow and thus promote the distribution in the
calcinator, experimental work was carried out to determine the aerodynamic diameters, particle projection area,
apparent density and maximum minimum axial length of representative particles [2] which are selected manually
from each fraction. Simulations were performed in commercial CFD code, based on these measured properties
such as projection area with different degrees of non-sphericity [3], aerodynamic diameters and pyrolysis of each
fraction at different temperature. As the first stage, a model for nonspherical particles tracking in a simple nonuniform
two-phase flow is derived in this paper, in which forces, including inertia, drag, gravity, lift, a force due to pressure
gradient and virtual-mass force are taken into account in particle force balance. The drag coefficient is calculated
using the correlation of Haider and Levenspiel (1989) and Ganser (1993) and combustion behavior such as pyrolysis
and burnout of non-spherical SRF particles were simulated.


Introduction ciency and economic fuel supply is very important for
the operators of cement plants. Co-combustion of Solid
Secondary fuels have gained substantial recognition in Recovered Fuels (SRFs), which mainly consist of bio-
Europe and are now used as part of the standard fuel genic components like paper, cardboard and wood (50 %
mix in the cement industry. In the German cement in- 70 %) and of plastics with particle sizes up to several
dustry in 2007 were about 100 million GJ of thermal centimeters in existing cement plants may bring signif-
energy used to produce about 33 million tons of cement. icant economic and environmental benefits. SRFs have
Because of the high energy intensity, questions of effi-







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


been proven to be very advantageous substitute fuel, due
to their low production cost and highly calorific value(14
-18 MJ/kg). Moreover, co-firing in existing plants nor-
mally requires low investment costs, since little extra
infrastructure in the feeding process is usually needed.
SRF utilisation results in valuable fossil fuel savings and
CO2 emissions reduction, as well as in minimization of
waste quantities to be landfilled. The share of alternative
fuels is currently averaging about 52 % of thermal input.
For technical process optimization in the context of a
project funded by the AiF research project (Project No
15407 N), the combustion behavior of SRF was investi-
gated. The focus is on the airworthy fractions prepared
from industrial and commercial waste, as it is available
as a mixture of different materials with relatively con-
stant energy value and are often used. This methodolog-
ical investigations in the laboratory scale and in a ce-
ment factory are being implemented and the developed
methodology is verified by comparing the numerical re-
sults of CFD simulations with the measured data. The
results show that a qualitative and quantitative prediction
of the temporal behavior of combustion of alternative
fuels subject to the uncertainty by the inhomogeneous
composition is possible.


Experimental

To determine the aerodynamic particle properties that
are vital importance for the current flight support, and
hence for the distribution in the precalcinator, in LUAT,
a test plant has been designed in which the settling veloc-
ity of individual fractions can be determined. A scheme
of particle separator is shown in Figure 1. In order to
avoid turbulence in the filter with a combined blower-
filter unit from the stationary air is pumped through the
plant. The air flow is detected at the air inlet via an eddy
current flow meter. The SRF sample can be fed to the
separator via a lateral approach.
The test facility consists essentially of a counter and a
cross-flow. The actual classification of the particles is re-
alized in settling velocity by changing the volume flow
and the flow velocity. The battles with the respective
flow rates in the cross-flow particles are separated and
sent for analysis. The experimental procedure is divided
into two parts: 1. The air classification to separate the
SRF sample into groups and 2. the statistical determi-
nation of additional parameters to the selected particles.
These parameters include both material and geometric
properties, as well as the mass of individual particles of
that groups.
For determination of geometric properties of the parti-
cles from the six groups, each of 30 particles picked out
manually and individual particles mass, length, width,
height were measured. Furthermore, the material of the


Outlet diffuser

Counterflow
separator

SBS


By-pass valve


Stop valve


To
blowers


Gravity classifier


Flow meter


Figure 1: Particle Separator


Fuel

Dosage gas -

Reaction gas .









Cold air --


Exhaust


Figure 2: Field Tube Reactor




particles was recorded. These parameters can be com-
bined with the mean settling velocity of each group to
calculate an aerodynamic diameter as a proxy, which is
used in the CFD simulation. The modeling allows the
aerodynamic properties of particles, and thus the flight
path. The experimental work and simulation were car-
ried out in the field tube reator which is 100 mm diam-
eter and 1.47 m length tube shown in Fig (2), to study
about combustion behaviour of different fractions and
burnout.







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


Table 1: Settling velocity and aerodynamic properties of different fraction


SRF Weight
wt %
Fraction 1 13
Fraction 2 32
Fraction 3 21
Fraction 4 9
Fraction 5 7
Fraction 6 Rest


Settling velocity Aerodynamic diameter
m/s mm
0.81 16
2.13 25
3.13 32
3.80 35
4.46 38
>4.46


Re Cw Apparent density

8.426 2.849 628.45
34.348 0.699 864.55
70.392 0.341 958.79
97.869 0.245 1080.07
126.523 0.19 1338.9


SRF
Plastics
Foils
Paper
wood
Fibers


Weight %
19
26
11.8
4.4
9.7


Table 2: Proximate and Ultimate analysis of SRF
proximate analysis, (wt %) Ultimate analysis,(wt %)
Fixed carbon Volatile C H O N
1.0 99 75.2 10.4 12.4 0.5
5.0 95 78.4 12.9 4.35 0.6
20 80 45.7 6.8 44.5 0.8
16.4 83.6 49.0 6.3 43.8 0.2
17.9 82.1 55.1 6.8 28.8 2.3


1 Forces on the particle

1.1 Particle force balance

The equation governing particle velocity i in a nonuni-
form flow field u was derived by Maxey and Riley
(1983).


inertia
inertia


67al (u V+ V(pp pf)g
drag gravity


Du 1 d



2 t ,dt
pressure gradient virtual mass




Jo 7 _(t V ) V I
67a_/t_ \V Pf 2/ )

Baset history term

In the above equation, we neglected the Faxen correc-
tion term, which becomes ,inilik.IIh only in the event
of large curvature in the velocity profile. pf and p are
the density and viscosity of the surrounding fluid; pp and
V represent the density and volume of the particles; and
i = [u uz] describes the fluid undisturbed veloc-
ity vector at the point occupied by the particles centre
of mass, with respect to the inertial frame. In Eq (1)
the derivative d / dt is used to denote a time derivative


LHV as received
(MJ/Kg)
34.6
39
17.6
18.3
22.6


following the moving particle, so that


dui ( ui ,Oui
dt a +vO)


is the time derivative of the velocity component ui. the
derivative D / Dt is used by contrast to denote the time
derivatives following a fluid element, and

Dui (ui aui\
D Kt-+ UajxIj (3)

is the fluid acceleration as observed at the instantaneous
centre of the particle. The different terms in Eq (3) rep-
resent in order, the force needed to accelerate the par-
ticles, viscous stokes drag, gravity, a pressure gradient
force accounting for the acceleration of the displaced
fluid, virtual mass force, and basset history term. The
order-of-magnitude estimates presented by Lazaro and
Lasheras (1989) indicate that, for small heavy particles
in the dilute regime, drag and inertia effects dominate
over those of the pressure field, the cirtual mass, and
the particles history. According to this, only the terms
describing inertia drag and gravity are retained in the
study of accumulation and dispersion of heavy particles
in the forced mixing layers (Martin Meiburg, 1994; Raju
Meiburg, 1995).
In this study, since our final purpose is to track the
nonspherical SRF particle in SRF fired cement kiln, and
the PVC particles used for experiment is a light particle,
the virtual mass force is retained, as well as the force
arising due to pressure gradient in the fluid. Addition-
ally, since we are tracking nonspherical particles while


Sphericity

0.008
0.2
0.42
0.64
0.8







7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


not sphere, a lift force should also be taken in to account,
as stated above. Therefore the governing equation for
evaluation of the particles velocity then becomes


ldv
PPV d
gT


FD+V(pp-pf)J+FPC+FPVM+PFL, (4)


where PD, FpP, PVM and FL represent drag, force
due to pressure gradient in the fluid, virtual-mass force
and lift force respectively. The forces FpG and FVM
are determined in the same way as shown in Eq (4);
while FD and FL governed by particles orientation (e.g.
Blaser, 2002; Hoerner, 1965), will be introduced below.
In this model, except Faxen correction term and
Basset history term, some other terms, such as ther-
mophoretic force and Brownian force are also neglected,
as well as Saffman lift thanks to the negligible rotation
with respect to the major axis of particles, i.e. z', axis.
It might be acceptable to exclude these forces in the par-
ticle force balance of the gas-solid flow in a SRF fired
cement kiln.

1.2 Drag force
The drag force is commonly expressed over the entire
Reynolds number spectra as

FD CDPf Sef f 1 V(u V), (5)
2
where Seff is the particle area normal to the direc-
tion of the drag force. It changes with the incidence an-
gle (ai) between relative velocity (i v),and particle
major axis direction is determined by

Sefti = ia (cos2 + (43 / i) sin ) (6)

where CD is the drag coefficient. Chhabra, Agarwal,
and Sinha (1999) collected about 1900 individual exper-
imental data of drag coefficient on nonspherical parti-
cles from 19 independent studies, covering wide range
of particle shapes: cyclinders, needles, cones, prisms,
discs, cubes, and so on, and excompassing wide ranges
of physical and kinematics conditions as: sphericity of
0.09-1 and Reynolds number of 10 4- 5.105; examined
critically the performance of five commonly used meth-
ods with those data, including Haider and Levenspiel
(1989) and Ganser (1993); and found that the best meth-
tod was that proposed by Ganser (1993): the resulting
overall mean error is about 16 %. The method of Ganser
(1993) is therefore used here,


24
-24 [1 + 0.1118(RepK1K2)0 6567
RepK1 K
0.4305
1+ 3305/(RepK1K2)'


(7)


where coefficient K1 = (dn /(3d) + 2 /( ;-" '))1; co-
efficient K2 = 10' ' '' ; d, is the equal pro-
jected area circle diameter d, = Re is the
particle Reynolds number, defined on the basis of the
relative velocity between particle and surrounding fluids
and equal-volume sphere diameter; e is particle spheric-
ity, and can be calculated as e = s/S, s is the surface of
a sphere having the same volume as the particle and S is
the actual surface area of the nonspherical particle.

1.3 Lift force
A lift force can be directly related to a drag force; how-
ever, it is still difficult to assign a specific value to the
constant of proportionality. Here, a lift formation is
used, in which the particle major axis (z') is also used
to characterize the force.
Considering the lift force is orthogonal to the relative
velocity (u v) and lies in the plane defined by the par-
ticle major axis direction (z') and the relative velocity
and that the lift must be invariant under a 180 rotation
of the particle major axis z' and vanishes if a, < 0 or i7,
we express the lift forces as

FL 2 CLpf Sfj 1 -8-C-J', (8)

where SeJf2 is the particle area normal to the direc-
tion of the lift force and is calculated by

Seff2 = 7a 2(sin2 i + (4 / Cr)2cos2 ai), (9)

Where CL is the lift coefficient. In this study, it is
determined in the way that the ratio of lift to drag meets
the relationship (Hoerner, 1965)

I = Ssin 2 aos ai (10)
IFD

1.4 Initial heating of biomass particles
During the first stage, a simple heat balance (without any
mass transfer) is used to relate the particle temperature to
the convective heat transfer and the absorption/emission
of radiation at the particle surface
cdT
mpQ-t = hAp(T, -Tp)+epApa(T -TR ), (11)

in which Cp, Tp, h, Ap, To, ~p, a, and OR repre-
sent particle heat capacity, particle temperature, convec-
tive heat transfer coefficient, particle surface area, lo-
cal fluid temperature at particle position, particle emis-
sivity, Stefan-Boltzmann ,and radiation temperature, re-
spectively. The heat transfer coefficient, h, is calculated
by Nu hdp / k o 2 + 0.6Re/ 2Pr /3 where koo











Pr are thermal conductivity and Prandtl number of the
fluid, respectively. The radiation temperature OR is de-
fined by (, )1 /4 where I is the radiation intensity. The
surface area of a cylindrical particle is calculated by,
Ap Trd / e where e is the particle shape factor.

1.5 Modelling species conservation and gas
phase reactions

The mixing and transport of chemical species are mod-
eled by solving the conservation equations describing
convection, diffusion and reaction sources for each com-
ponent species. The standard model usually applied for
the volatiles combustion is the Eddy-Break Up (EBU)
model of Magnussen and Hjertager (1977). It is based
on the assumption that chemical reactions occur rapidly
compared to mixing. The limiting step is therefore the
mixing process, which is governed by the rate of dissi-
pation of the fuel- and oxygen-containing eddies. The
reactions taken into consideration for the combustion of
coal or biomass volatiles are usually global with a maxi-
mum of two reaction steps, as presented in Eq (12). The
volatiles are expressed as a hydrocarbon CHyOzNa,,
where coefficients x, y, z and a derive from the ultimate
analysis of the fuel and the enthalpy of formation of the
volatiles is based on their heating value.


C1HON x y
C H OzNo + 2 4


) 02 -> xCO+
y a
2HO N2


1
CO + 1 02 -- C02 (13)


1.6 Devolatilization of biomass particles

When the temperature of the particle reaches the va-
porization temperature, devolatilization begins to release
volatiles. The homogeneous combustion of the volatiles
takes place once they are released from biomass parti-
cles. A single-rate kinetic devolatilization model is used
to predict the volatiles yield rate, which assumes that the
rate of devolatilization is dependent on the amount of
volatiles remaining in the particle via a first-order reac-
tion
dm k(mp (1 fo)mpo (14)
dt
where k, fvo, and mpo denote the kinetic rate, the mass
fraction of the volatiles that are initially present in the
particle, and the initial particle mass, respectively. The
kinetic rate k is defined by the input of an Arrhenius-
type, pre-exponential factor and an activation energy ex-
pression
k = Al E/RT, (15)


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


where A1 and E are taken to be 10E+06s-
and 74E+07J/Kmol, respectively, for the biomass
(Kaer,2001). The particle diameter changes during de-
volatilization according to the swelling coefficient, C,,


d~,
dp +(C'.
dp, 0


1) mp, mp
fvormp,


where dp,o and C,, are the particle diameter at the start
of devolatilization and the swelling coefficient, respec-
tively. The particle temperature during devolatilization
is updated by including contributions from convection,
radiation and the heat consumed during devolatilization:


CdT
dt


hAp(T, Tp) + dHf,+

_pApa(O4 T4)


where Hfg is the latent heat of the particle.

1.7 Char combustion


The combustion of biomass char is quite complicated
since it is affected not only by the composition of the
biomass fuel but also by the shape and size of the fuel
particles. Biomass char is believed to be more reac-
tive than coal char. Different measures in modeling of
biomass char combustion can be found in literature, for
instance, using diffusion-limited surface reaction rate
model modified by aspect ratio-dependent enhancement
factors (in the range of 1-1.6 for the aspect ratio from
1 to 11) due to the non-sphericity of biomass particles
(e.g., Gera et al. 2002; Yin et al. 2004), and using
Smiths-intrinsic model modified by a constant enhance-
ment factor of 4 in order to represent the high burning
rate of the biomass char particles (e.g., Ma et al. 2007,
2009). In this study, due to the lack of non-spherical
properties of the straw particles and the low char content
in the straw, the diffusion-limited surface reaction rate
model is used in all the simulations, without considering
any enhancement factor.

2 Results and discussion

2.1 Particle track and devolatilization

A Lagrangian particle model (Crowe, 1979) is used to
calculate particle aerodynamics and turbulent dispersion
for the SRF particles in the field tube reactor. The con-
tinuous distribution of discrete particle size, velocity and
material property such as apparent density is represented
by discrete number of particle streams (typically several
thousands). Each size is characterized by an initial posi-
tion, particle size, mass, velocity, temperature and num-
ber flow rate. The flight history of each stream through











the field tube reactor is determined by gravity, the aero-
dynamic forces such as drag, lift and pressure forces and
turbulent dispersion. The model compute the drag force


Table 3: Settling velocity of different fractions m/s
Experimental Simulation
Fraction 1 0.81 0.89
Fraction 2 2.13 2.14
Fraction 3 3.13 3.18


Fraction 4
Fraction 5
Fraction 6


3.8
4.46
>4.46


for the nonspherical particles (Heider and Levenspiel,
1988) such as wood, plastic and fibers which may be
tumble and rotate as they progress in the tube reactor.
A stochastic particle dispersion model (Milojevis, 1990)
accounts for the effects of turbulent fluctuations in the
gas velocity on the instantaneous motion of the particle.
As fuel particles are heated by radiation and convection,
then dry and burn at nearly constant volume and break
up after completion of char combustion. Weight loss and
density variations have a significant effect on the gravita-
tionaly and aerodynamic forces on each particle. Mois-
ture, volatiles, and char products that evolve from the
particles during flight are used in the solution of overall
material and energy balance. The motion of the three


Figure 3: Lagrangian particle track for different fraction
((a) Fr 1, (b) Fr 2, (c) Fr 3)

different fractions are shown in Fig (3). Quite clearly,
it is possible to see that for the fraction 1, dispersion
is a quicker process, by which the particles already in
the quarl section have reached a higher dispersion due


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


to their aerodynamic forces such as drag and lift forces.
Due to the misalignment of the flow streamlines, and the


S0.00953


0.00763


0.00572


0.00381


10.00191


0.00000
(a)


0.00011


0.00009


0.00007


0.00005


0.00002



0.00000
(b)


'I


H 0.00028


0.00023


0.00017


0.00011


0.00006


0.00000
(c)


Figure 4: Mass fractions of volatiles for different frac-
tions ((a) Fr 1, (b) Fr 2, (c) Fr 3)

direction of the aerodynamic forces upon the particles,
they break off from the streamlines, and tumble in a dis-
orderly fashion through the test rig. For fraction 2 and
3 which is having relatively higher mass than fraction
1 carried along the fluid stream and dispersion is little
bit slower due to higher gravitational forces than lift and
drag force. Almost end of the field tube reactor particle
movements for different fractions are similar because of
their weight loss during the devolatilization. Due to their
difference in mass and aerodynamic forces, particle set-
tling velocity vary for each fraction. Settling velocity of
different fraction materials were compared with experi-
mental data which shows good agreement with simula-
tion results shown in table (3) for fraction up to 3. For
fraction 4-6 we can clearly see that the predicted set-
tling velocity deviate from the experimental data due to
the presence of inert metal wastes which are mixed with
SRF particles. Due to their difference in settling veloc-
ity of the particles, residence time within the field tube
reactor reduced. Thus promote different particle heat-
ing and devolatilization behaviour for different fractions
as shown in Fig(4), and also different material such as
plastics, foils, paper and wood has different combustion
behavior. Mass fraction of the volatiles released from
SRF clearly shows the trent that aerodynamic and grav-
itational forces affecting the heating and devolatiliza-
tion within the filed tube reactor. Since the particle vol-
ume is not changed neither during devolatilization due












I400


350.4



300.9



251.3



S201.8


on. /


Figure 5: Track of particles colored by particle density
(kg/m3) ((a) Fr 1, (b) Fr 4)


to C,, = 1 nor in diffusion-limited char combustion.
Therefore burnout fraction will be reflected from history
of particle density as shown in Fig (5). As discussed
above the particle track is affected by two factors: the
rate at which the particle loses its weight (i.e., burning
rate) and the different aerodynamic forces.


3 Conclusion

From the numerical study in a 1.47 m-long field tube re-
actor, the following conclusions could be drawn on SRF
combustion simulation. It is very important to take into
account the particle non-sphericity in order to achieve
a better understanding on SRF combustion. The par-
ticle non-sphericity affects greatly both its motion and
its combustion. These two aspects are interacting with
each other very closely in SRF combustion. If the mo-
tion of a particle is not predicted accurately, it is less
likely to predict the particle combustion correctly. On
the other hand, if the particle burnout is not predicted
correctly, the aerodynamic forces on the particle may
not be sufficient to sustain the weight of the particle and
thus the fate of the particle could be changed totally. To
accurately model the motion of SRF particles, it is es-
sential to include all the important forces in the parti-
cle force balance. In this paper, a drag correlation for
non-spherical particles, an additional lift due to particle
non-sphericity, lift force and virtual-mass force due to
relatively light particles are taken into account as well
as gravity and a pressure-gradient force. Since the drag


Nomenclature

Roman symbols


C,
0D
CL
Csw
d,
T,
To,
FD
FL


Surface area of the particle, (m2)
Heat capacity of particle, (J/(KgK))
Drag coefficient
Coefficients in evaluating lift force
Swelling coefficient
Equal-volume sphere diameter, (m)
Particle temperature, (K)
Local fluid temperature, (K)
Drag force, (N)
Lift force, (N)


FpG Force due to the pressure gradient
in the fluid,(N)
FVM Virtual mass force,(N)
h Convective heat transfer coefficient, (W/(mr2K))
mi Particle mass, (kg)
mpo Initial particle mass, (kg)
Re, Particle Reynolds number, (kg)
Seff 1 Projected area of particle normal
to drag force, (m2)
Seff2 Projected area of particle normal to
lift force, (mn2)
flo Mass fraction of volatiles that are
initially present in the particle (-)
Greek symbols
particle major axis z'
Ep particle emissivity
P viscosity of the surrounding fluid, (kg/ms)
pf density of the surrounding fluid, (kg/m3)
pp particle bulk density, (kg/m3)
a Stefan-Boltzmann constant,
(5.67 10 W/m2K 4)


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


and the lift are both shape factor and particle orientation-
dependent, the coupled particle rotation is also resolved.
To better model the combustion of SRF, it is primary to
take into account the actual particle surface area avail-
able and the average oxygen mass flux at particle sur-
face, both of which are shape factor-dependent. In or-
der to model these SRF particles experimental work were
carried out to characterize SRF particle as a mixture for
better understanding of aerodynamic properties and set-
tling velocity of SRF particles and these properties were
used for the simulation which shows good agreement
with simulation results.


Acknowledgements

The authors would like to thank the AIF for the financial
support of the project (15407 N)


1000


997



994



991.1



988.1



985.1











particle sphericity (= s/S)
Subscripts
max maximum
ref reference

Superscipts
9 Gas
p Particle


References

Chhabra, R.P., Agarwal, L., Sinha, N.K., 1999. Drag on
non-spherical particles: an evaluation of available meth-
ods. Powder Technology 101,288-295

Ganser, G.H., 1993. A rational approach to drag pre-
diction of spherical and nonspherical particles. Powder
Technology 77, 143-152

Gera, D., Mathur, M.P., Freeman, M.C., Robinson, A.,
2002. Effect of large aspect ratio of biomass particles on
carbon burnout in a utility boiler. Energy & Fuels 16,
1523-1532

Michalis Agraniotis; Panagiotis Grammelis; Emmanuel
Kakaras; Co-firing Solid Recovered Fuels (SRFs) with
brown coal in large-scale pulverized fuel power plants
- a simulation approach, International Journal of Global
Warming 2009 Vol. 1, No.1/2/3 pp. 106 128

Dunnu, G.; Hilber, T.; Schnell, U.; Advanced Size
Measurements and Aerodynamic Classification of Solid
Recovered Fuel Particles Energy & Fuels; Heft 20(4),
Seite(n): 1685-1690; 2006

Sommerfeld, M.; Berend van Wachem; Rene Oliemans;
Best Practice Guidelines for Computational Fluid Dy-
namics of Dispersed Multiphase Flows, ERCOFTAC,
Version 1, October 2008

Chungen Yin,; Lasse Rosendahl,; J. Condra, Thomas
J.Condra,; Use of numerical modeling for co-firing
biomass in wall-fired burners, Chemical Engineering
Science 59 (2004), 3281-3292

Sondreal, E.A., Benson, S.A., Hurley, J.P., Mann, M.D.,
Pavlish,J.H. Swanson, M.L., Weber, G.F., Zygarlicke,
C.J., 2001. Review of advances in combustion technol-
ogy and biomass cofiring. Fuel Processing Technology
71, 7-38

Scharler, R., Fleckl, T., Obernberger, I., 2003. Modifi-
cation of a magnussen constant of the Eddy Dissipation
model for biomass grate furnaces by means of hot gas
in-situ FT-IR absorption spectroscopy. Progress in Com-
putational Fluid Dynamics 3, 102-111


7th International Conference on Multiphase Flow,
ICMF 2010, Tampa, FL, May 30 -June 4, 2010


Rosendahl, L., 1998. Extending the modeling frame-
work for gas-particlesystems. Ph.D. Thesis, Aalborg
University.

Xiaofen Guo, Zhiqi Wang, Haibin Li, Haitao Huang,
Chuangzhi Wu, and Yong Chen: A Study on Com-
bustion Characteristics and Kinetic Model of Municipal
Solid Wastes. Energy Fuels, 2001, 15 (6), pp 1441-1446

S. LainI,; M. Sommerfeld, A study of the pneumatic
conveying of non-spherical particles in a turbulent hori-
zontal channel flow, Braz. J. Chem. Eng. vol.24 no.4 Sao
Paulo Oct./Dec. 2007

Andreas HOlzera and Martin Sommerfeld, ; New sim-
ple correlation formula for the drag coefficient of non-
spherical particles, Powder Technology Volume 184, Is-
sue 3, 2 June 2008, Pages 361-365




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - Version 2.9.7 - mvs