Fluidized bed combustion of tyre derived fuel

Fabrizio Scala

a,*

, Riccardo Chirone

a

, Piero Salatino

b

a

Istituto di Ricerche sulla Combustione––CNR, Piazzale Tecchio 80, 80125 Napoli, Italy

b

Dipartimento di Ingegneria Chimica, Universit 

a

a degli Studi di Napoli Federico II, Piazzale Tecchio 80, 80125 Napoli, Italy

Received 6 January 2002; received in revised form 17 June 2002; accepted 9 July 2002

Abstract

Mechanistic aspects of the fluidized bed combustion of tyre derived fuel (TDF) have been analyzed both experimentally and

theoretically. The time-resolved rates of carbon fines elutriation during the fluidized bed combustion of TDF have been measured
during batch experiments in a bench scale reactor at 850

°C, under different operating conditions. Experimental results indicated

that both gas superficial velocity and oxygen partial pressure exert influence upon the overall fixed carbon combustion efficiency.
The efficiency increases slightly with the oxygen concentration and significantly if the gas superficial velocity decreases.

Experimental data are further analyzed in the framework of a fluidized bed combustor model especially suited for high-volatile

solid fuels feedings. The model takes into account phenomena that assume particular importance with this kind of fuels, namely fuel
particle fragmentation in the bed and combustibles segregation and postcombustion above the bed. Experimental and model results
indicate that the efficiency of the fluidized bed combustion of TDF is controlled by the competition between combustion and
entrainment of char fines and volatile matter released in the early stage of fuel conversion.
Ó 2003 Elsevier Science Inc. All rights reserved.

Keywords: Fluidized bed combustion; Tyre derived fuel; Fragmentation

1. Introduction

The use of tyre derived fuel (TDF) from scrap tyres in

combustion processes is very attractive because of its
large heating value and since conventional disposal of
whole tyres (by landfilling) represents a source of con-
siderable health and fire hazards. In this respect fluidized
bed technology seems to be a natural choice due to its
fuel flexibility and the possibility to achieve an efficient
and clean operation [1,2]. A literature survey on oper-
ating experience of pilot and full scale fluidized bed
combustors fueled with tyres highlights the relevance of
appropriate design and operation criteria in the fluidized
bed combustion of TDF [3–5]. The non-negligible sulfur
content calls for careful consideration of emission issues,
which can be conveniently dealt with by in bed desul-
furization with sorbents. The problems related to metal
wire and low melting point fiberglass contained in tyres
have been underlined by Rasmussen and McFee [6].

Arena et al. [7,8] showed that TDF undergoes ex-

tensive fragmentation upon devolatilization, forming a
number of fine fragments easily elutriated from the bed.
Accordingly the bed carbon loading as coarse particles is
negligible and in practice only fines are present in the
bed. As a consequence, combustion efficiency is strongly
dependent on how the combustion time scale compares
with the residence time of char fines in the bed, which in
turn is related to the elutriation rate. A scheme of the
series-parallel phenomena relevant for the size reduction
of TDF particles in fluidized beds is shown in Fig 1[8].
As no coarse particle is present at the steady-state in the
bed, secondary fragmentation as well as attrition are not
relevant for this fuel.

Another important issue during TDF combustion is

the fate of the combustible volatile matter of the fuel
[9,10]. The large contribution to the overall heat release
coming from homogeneous combustion of volatile
matter emphasizes the importance of mixing/segregation
phenomena with respect to the bed. The location of
volatiles combustion significantly affects the heat release
profiles throughout the combustor, influencing the de-
sign of heat exchange surfaces, the course of pollutant

Experimental Thermal and Fluid Science 27 (2003) 465–471

www.elsevier.com/locate/etfs

*

Corresponding author. Tel.: [phone]; fax: [phone]-

6936.

E-mail address:

[e-mail]

(F. Scala).

0894-1777/03/$ - see front matter

Ó 2003 Elsevier Science Inc. All rights reserved.

doi:10.1016/S0894-1777(02)00249-2

generation and the internals material degradation. De-
pending on fluidization conditions and on the combus-
tion parameters, a large fraction of volatiles can bypass
the bed, eventually burning in the freeboard, even in the
case of submerged feeding of TDF particles in the bed.

The purpose of this paper is to outline mechanistic

aspects of the fluidized bed combustion of TDF by
means of an experimental and theoretical analysis. In
order to investigate the performance of fluidized bed
combustors fed with TDF, experiments consisting of
batchwise combustion of TDF charges in a bed fluidized
at different superficial velocities, have been carried out.
The time-resolved rates of carbon fines elutriation dur-
ing the fluidized bed combustion of TDF have been
measured to determine the fixed carbon combustion
efficiency. Experimental results are then analyzed in the
framework of a fluidized bed combustor model em-
bodying features that are relevant to high-volatile solid
fuels combustion [11,12]. Key aspects relevant to the
efficient combustion of TDF are highlighted.

2. Experimental

Experiments have been carried out with a market

available TDF, whose properties are given in Table 1.
The fuel was chopped in the size range 3.35–4.5 mm with
Sauter mean diameter of 4.1mm.

Fig. 2 shows the laboratory scale stainless steel 40

mm ID fluidized bed combustor. It is equipped with a
two-exit head ending with sintered brass filters for the
time-resolved collection of elutriated fines. Details of
apparatus and techniques are given elsewhere [11,13].
The bed consisted of 180 g of silica sand, having a size
range of 0.3–0.4 mm, whose properties are reported in
Table 1.

Two different sets of batch experiments have been

carried out, namely tests under integral and differential
conditions with respect to oxygen. In experiments under
integral conditions TDF particles were injected into the
bed kept at 850

°C in batches corresponding to a con-

stant fixed carbon content of 2.0 g. This large amount of
TDF was charged in the bed at the beginning of the test
in order to get reproducible values of the carbon elu-

triation rate. The bed was initially kept at the minimum
fluidization velocity in inert atmosphere until devolatil-
ization was over. Afterwards, superficial gas velocity
was raised to its final value and the inlet gases were
switched from nitrogen to nitrogen–oxygen mixtures at
preset values of oxygen concentration. Fines escaping
the combustor during the experiments were collected by

Table 1
Properties of the fuel and of bed material

Fuel

TDF

Proximate analysis, % (as received)
Moisture

1.9

Ash

4.3

Volatile Matter

63.4

Fixed Carbon

30.4

Ultimate analysis, % (dry basis)
Carbon

83.8

Hydrogen

6.9

Nitrogen

0.6

Sulfur

2.0

Ash

4.4

Oxygen

2.3

Particle Sauter mean diameter, mm

4.1

Particle density, kg/m

3

1450

Bed material

Silica sand

Particle size range, mm

0.3–0.4

Particle Sauter mean diameter, mm

0.356

Particle density, kg/m

3

2540

Fig. 2. Experimental apparatus: (1) gas preheating section; (2) elec-
trical furnaces; (3) ceramic insulator; (4) gas distributor; (5) thermo-
couple; (6) fluidization column; (7) head with three-way valve; (8)
sintered brass filters; (9) hopper; (10) SO

2

scrubber; (11) stack; (12)

cellulose filter; (13) membrane pump; (14) gas analyzers; (15) personal
computer; (16) manometer; (17) digital mass flowmeters; (18) air
dehumidifier (silica gel).

Fig. 1. Comminution behavior of TDF in fluidized bed [8].

466

F. Scala et al. / Experimental Thermal and Fluid Science 27 (2003) 465–471

means of sintered brass filters and then analyzed to de-
termine their fixed carbon content. This procedure al-
lowed the evaluation of the instantaneous rates of fixed
carbon elutriation. In experiments under differential
conditions TDF batches corresponding to a fixed car-
bon content of 0.2 g were injected into the bed. The
experimental procedure was similar to that of integral
tests. Due to the low amount of fixed carbon fed, the
elutriated fines were collected for each run in a single
filter, so that time-resolved elutriation rate profiles could
not be obtained.

3. Results

Figs. 3–6 show time-resolved carbon elutriation rates

E

c

measured during the batch fluidized bed combustion

of TDF in integral experiments. Fluidizing gas was either
nitrogen or nitrogen–oxygen mixtures having an oxygen
content of 1.5%, 3.0% and 4.5% at fluidizing gas super-
ficial velocities of 0.4 and 0.8 m/s. The elutriation rates
were normalized by the amount of fixed carbon charged
into the bed W

C;0

¼ 2:0 g. No significant difference in the

order of magnitude of elutriation rate is observed when
passing from inert to oxidizing conditions and when in-
creasing oxygen concentration. Integration of curves
obtained in oxidizing conditions over the time interval
from 0 to complete burn-off provides the values of the
overall fixed carbon combustion efficiency:

g

¼ 1 

Z

s

bo

0

E

c

W

C;0

dt

0

ð1Þ

t, min

0

5

10

15

20

25

30

35

40

45

E

c,bed

/W

C

,bed,

0

,m

in

-1

0.00

0.05

0.10

0.15

0.20

0.25

0.8 m/s
0.4 m/s

Superficial velocity

Fig. 3. TDF fixed carbon elutriation rate as a function of time in in-
tegral batchwise experiments at different fluidization velocities.
O

2

¼ 0%.

t, min

0

5

10

15

20

25

30

E

c

,bed

/W

C

,bed,

0

,m

in

-1

0.00

0.05

0.10

0.15

0.20

Superficial velocity

0.8 m/s
0.4 m/s

Fig. 4. TDF fixed carbon elutriation rate as a function of time in in-
tegral batchwise experiments at different fluidization velocities.
O

2

¼ 1:5%.

t, min

0

5

10

15

20

25

30

E

c,

b

e

d

/W

C,be

d,

0

,m

in

-1

0.00

0.05

0.10

0.15

0.20

0.25

0.8 m/s
0.4 m/s

Superficial velocity

Fig. 5. TDF fixed carbon elutriation rate as a function of time in
integral batchwise experiments at different fluidization velocities.
O

2

¼ 3%.

t, min

0

5

10

15

20

E

c,

b

e

d

/W

C,be

d,

0

,m

in

-1

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.8 m/s
0.4 m/s

Superficial velocity

Fig. 6. TDF fixed carbon elutriation rate as a function of time in
integral batchwise experiments at different fluidization velocities.
O

2

¼ 4:5%.

F. Scala et al. / Experimental Thermal and Fluid Science 27 (2003) 465–471

467

where s

bo

is the burn-out time. It can be clearly appre-

ciated how strongly both the time scale for carbon elu-
triation and the area below the curves are affected by the
value of the gas superficial velocity. This finding is
consistent with previous observations of Rasmussen and
McFee [6]. In analyzing the data, it must be considered
that, due to the amount of TDF charged in the bed at
the beginning of the test, selected so as to get repro-
ducible values of the carbon elutriation rate, the
combustor did not behave as a differential reactor with
respect to oxygen. In particular, oxygen concentration
dropped to nearly zero at the beginning of the test,
whatever the oxygen concentration in inlet gases.
Smaller elutriation rates and larger efficiencies could be
expected, for any given inlet oxygen concentration, un-
der differential reactor operating conditions. Fig. 7
summarizes results obtained in all the batch combustion
experiments. Overall fixed carbon combustion efficiency
g

is reported as a function of oxygen concentration in

the feed gas for the two superficial fluidization velocities
(0.4 and 0.8 m/s) investigated. It can be observed that
efficiency increases as oxygen concentration increases.
This trend is more pronounced at the velocity of 0.4 m/s
than at 0.8 m/s. For a given oxygen concentration in the
fluidizing gas, an increase of the gas superficial velocity
results in a significant decrease of the fixed carbon
combustion efficiency, as a consequence of the lower
fines residence time in the combustor.

Experiments were repeated under conditions that

made the reactor differential with respect to oxygen.
This was accomplished by feeding batches of TDF
corresponding to W

C;0

¼ 0:2 g. Fig. 8 summarizes fixed

carbon combustion efficiencies as a function of oxygen
concentration for this series of experiments at the two
superficial fluidization velocities (0.4 and 0.8 m/s) in-

vestigated. Comparison of Fig. 8 with Fig. 7 shows that,
as expected, the combustion efficiency is larger in all
conditions for the differential experiments, but the
overall trends are the same.

The low sensitivity of the fixed carbon combustion

efficiency on the inlet oxygen concentration under both
integral and differential conditions was somewhat
unexpected. One possible explanation relies on the
mechanism of particles segregation during the early
devolatilization stage highlighted by Fiorentino et al.
[9,10]. The authors showed that, whatever the feeding
procedure, TDF particles tend to rapidly rise to the bed
surface upon devolatilization, and remain segregated
thereon as far as gas-emission associated with pyrolysis
is extensive. As devolatilization rate declines, particles
can be re-entrained into the bed. On the other hand
Arena et al. [7,8] demonstrated that devolatilization of
TDF particles is associated with extensive release of
elutriable carbon fines. Altogether, these findings sug-
gest that a large quantity of combustibles (volatile
matter and carbon fines) are released directly into the
freeboard, completely bypassing the bed. Incremental
conversion of the bypassed fuel in the freeboard is ruled
by oxygen macromixing, combustibles being mostly lo-
cated within reducing regions. On the whole, combina-
tion of fuel bypass and afterburning make combustion
efficiency more strongly related to gas superficial ve-
locity, affecting bed hydrodynamics and particle segre-
gation, and less dependent on oxygen concentration.

4. Theoretical analysis

A stationary atmospheric bubbling fluidized bed

combustor model, suitable for high-volatile fuels, has
been developed [11,12]. The fluidized bed combustor is

O

2

, %

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.4 m/s

0.8 m/s

Superficial velocity

Fig. 7. TDF fixed carbon combustion efficiency as a function of inlet
oxygen concentration in integral batchwise experiments at different
fluidization velocities.

O

2

, %

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0.4 m/s

0.8 m/s

Superficial velocity

Fig. 8. TDF fixed carbon combustion efficiency as a function of inlet
oxygen concentration in differential batchwise experiments at different
fluidization velocities.

468

F. Scala et al. / Experimental Thermal and Fluid Science 27 (2003) 465–471

modeled as a series of three reaction zones: the dense
fluidized bed, the splashing region, and the freeboard.
The model takes into account phenomena that assume
particular importance with this kind of fuels, namely
fuel particle fragmentation and attrition in the bed,
volatile matter segregation and turbulent postcombus-
tion above the bed as well as thermal feedback from the
splashing region to the bed.

The model has been slightly modified for application

to the prediction of the fluidized bed combustion of
TDF. In particular bypass of fines into the splashing

region during devolatilization was considered. The as-
sumption was made that the fraction of fines burning in
the bed was equal to the fractional volatile matter
burned in this zone, the remainder bypassing the bed
directly into the splashing region. Input parameters were
selected with reference to geometrical and operating
parameters typical of large-scale bubbling atmospheric
combustors with under-bed feeding. Char combustion
kinetics was expressed following Masi et al. [14]. Values
of base case operating variables and model results are
shown in Table 2.

Table 2
Operating variables and results of model computations

Operating variables

Combustor cross-sectional area, m

2

1.0

Unexpanded bed height, m

1.0

Combustor height (from the distributor), m

5.0

Bed temperature, K

1123

Freeboard temperature, K

973

Pressure, kPa

101

Superficial gas velocity, m/s

1.0

Excess air factor, –

1.2

Fuel feed mean particle size, mm

10.0

Bed solids mean particle size, mm

0.6

Model results

Expanded bed height, m

1.6

Splashing zone height, m

0.3

Fuel feed rate, kg/s

0.022

Fraction of volatiles and fines burned in the bed, %

19.3

Fixed carbon loading, kg

Coarses in:

Bed

–

Fines in:

Bed

0.56e

)2

Splashing zone

1.4e

)2

Freeboard

4.7e

)2

Oxygen mole fraction, –

Bed

0.18

Splashing zone

0.086

Freeboard

0.068

Fixed carbon combustion efficiency in the bed, %

19.1

Total combustion efficiency in the bed, %

19.3

Total combustion efficiency of the combustor, %

81.2

Splashing zone temperature, K

1175

Heat generation rate, kW

Bed:

Coarse char

–

Fine char

49

Volatiles

98

Splashing zone:

Fine char

47

Volatiles

346

Freeboard:

Fine char

15

Volatiles

63

Splashing zone-to-bed heat fluxes (% of heat release)

Convective

81.2

Radiative

1.6

Total

82.8

Splashing zone-to-freeboard heat fluxes (% of heat release)

Convective

5.6

Radiative

11.6

Total

17.2

F. Scala et al. / Experimental Thermal and Fluid Science 27 (2003) 465–471

469

Calculations show that, as a consequence of the fact

that TDF undergoes complete comminution to elutri-
able fines upon devolatilization (primary fragmenta-
tion), the carbon loading as coarse particles in the bed is
equal to zero and only fines are present in the bed. A
large part of the fines is released directly in the splashing
region. It follows immediately that fixed carbon com-
bustion efficiency depends on the competition between
parallel fines combustion and elutriation/entrainment
processes, as indicated by the experimental results. TDF
has an intrinsic reactivity with a value intermediate be-
tween those of a bituminous coal and of a biomass. As a
consequence the fixed carbon loading as fines in the bed
is smaller than that typically found for a coal but larger
than that relative to a biomass. A very low combustion
efficiency in the bed is calculated, as a consequence of
the extensive volatiles and fines bypassing.

Calculated heat generation rates are reported in Ta-

ble 2. The bed and splashing region sections represent
the main location of fixed carbon conversion, due to
fines combustion. Extensive volatile matter segregation
occurs, bringing a significant portion (about 70%) of the
heat release from volatiles combustion into the splashing
region. On the basis of the above considerations, one
can estimate that about 24% of the total heat release
takes place within the bed during combustion of TDF
under the simulated conditions, the remainder being
released mostly (64%) in the splashing region, mainly
because of extensive volatile matter postcombustion.
Only 12% of the heat release takes place in the upper
freeboard region.

It is interesting to assess, at this point, the magnitude

of thermal fluxes that are established at steady state
between the different sections of the combustor. The
main scope is that of determining the extent to which
heat released in the splashing region is fed back to the
bed (where heat recovery is more efficient) and the re-
lated temperature of the splashing region. Results of
computations are summarized in Table 2. The large
importance of volatile matter and fines postcombustion
gives rise to a pronounced overheating of the splashing
region (about 52

°C). However large thermal feedback

to the bed prevents overheating from being even larger.
About 83% of the heat released in the splashing zone is
fed back to the bed. The contribution from solids con-
vection associated to particles ejection/fall-back is by far
dominant. Heat transfer between gas and ejected solids
in the splashing region takes place along parallel path-
ways associated with convection (about 94% of the heat
flux) and radiation (about 6%).

Fluidization velocity, bed material size and the total

excess air factor have been varied in order to assess their
influence on the combustion performance of TDF. Figs.
9–11 show the splashing region temperature and the
total combustion efficiency as a function of the different
operating variables considered. The splashing region

temperature is influenced both by superficial velocity
and bed particle size. Higher superficial velocities lead to

Fluidization velocity, m/s

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Splashing

re

gion

temperature

,

˚C

850

900

950

1000

1050

1100

1150

1200

To

ta

l

c

ombustion

efficiency,

-

0.70

0.75

0.80

0.85

0.90

0.95

1.00

Splashing region temperature
Total combustion efficiency

Fig. 9. Splashing region temperature and total combustion efficiency as
a function of fluidization velocity.

Bed particle size, mm

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Splashing

re

gion

temperature

,

˚C

800

900

1000

1100

1200

1300

1400

To

ta

l

c

ombustion

efficiency,

-

0.70

0.75

0.80

0.85

0.90

0.95

1.00

Splashing region temperature
Total combustion efficiency

Fig. 10. Splashing region temperature and total combustion efficiency
as a function of bed particle size.

Theoretical excess air factor, -

1.0

1.1

1.2

1.3

1.4

1.5

Splashing

re

g

io

n

te

m

pe

rature

,

˚C

850

875

900

925

950

Total

c

om

bustion

efficien

cy

0.70

0.75

0.80

0.85

0.90

0.95

1.00

Splashing region temperature
Total combustion efficiency

Fig. 11. Splashing region temperature and total combustion efficiency
as a function of theoretical excess air factor.

470

F. Scala et al. / Experimental Thermal and Fluid Science 27 (2003) 465–471

higher inert bed particle ejection velocities and longer
residence times in the splashing region, enhancing the
thermal feedback mechanism to the bed. The larger the
particle size, the smaller the particle ejection rate and
residence time in the splashing region, the smaller the
thermal feedback to the bed. At low fluidization veloc-
ities and/or large bed particle sizes the splashing region
temperature can reach very high peak values, leading to
critical conditions above the bed. On the other hand Fig.
11 shows that the splashing zone temperature is not
influenced significantly by the excess air factor.

Superficial velocity has a strong influence on com-

bustion efficiency (Fig. 9), influencing the competition
between fines combustion and entrainment, while the
influence of bed particle size (Fig. 10) is more limited.
Interestingly, Figs. 9 and 10 show a non-monotonic
trend of combustion efficiency. The humps in the curves
are the consequence of non-negligible volatiles post-
combustion in the upper freeboard region for superfi-
cial velocities and bed particle sizes values in the range
0.6–1.3 m/s and 0.5–0.8 mm, respectively. Larger oxy-
gen concentration establish in the splashing region
under these conditions and this is reflected by enhanced
combustion efficiency. As regards the effect of the the-
oretical excess air factor, while at low values the com-
bustion efficiency is barely influenced by this variable,
at large values the combustion efficiency significantly
increases.

5. Conclusions

Mechanistic aspects of the fluidized bed combustion

of TDF have been investigated by means of batchwise
combustion experiments and theoretical calculations
based on a recently proposed fluidized bed combustor
model. Experimental and model results indicate that the
efficiency of TDF fluidized bed combustion is deter-
mined on the one hand by the competition between
combustion and entrainment of char fines generated by
primary fragmentation in the early stage of fuel con-
version, on the other hand by the propensity of volatiles
to bypass the bed and burn in the freeboard region.
High peak temperatures above the bed surface can be
reached as a consequence of extensive fuel bypassing,
possibly leading to troublesome operation. On the basis
of experimental and model results, proposed strategies
for the efficient and environmentally acceptable com-
bustion of TDF in bubbling fluidized beds such as the
staging of combustion air into a primary and secondary
stream [15], or the use of fluidized bed reactors with
internal solids circulation [16] can be explained.

References

[1] S.C. Saxena, C.K. Jotshi, Fluidized-bed incineration of waste

materials, Progress in Energy and Combustion Science 20 (1994)
281–324.

[2] E.J. Anthony, Fluidized bed combustion of alternative solid fuels:

status, success and problems of the technology, Progress in Energy
and Combustion Science 21(1995) 239–268.

[3] C. Dervin, D.J. Waldron, J.C. Haas, Development of a FI

CIRCe fluidized bed boiler to burn waste tyres, in: Proceedings of
the 14th International Conference on Fluidized Bed Combustion,
ASME, New York, 1997, pp. 439–446.

[4] H. Teng, C.S. Chyang, S.H. Shang, J.A. Ho, Characterization of

waste fire incineration in a prototype vortexing fluidized bed
combustor, Journal of the Air and Waste Management Associ-
ation 47 (1997) 49–57.

[5] I. Gulyurtlu, N. Costa, J. Cunha, The use of waste tyres in a

fluidised bed boiler to generate process steam, in: Proceedings of
the 16th International Conference on Fluidized Bed Combustion,
ASME, New York, 2001, pp. 1240–1246.

[6] G.P. Rasmussen, J.N. McFee, Fluidized bed systems for steam

generation from scrap tires, in: Proceedings of the 7th Interna-
tional Conference on Fluidized Bed Combustion, ASME, New
York, 1982, pp. 870–874.

[7] U. Arena, A. Cammarota, R. Chirone, Fragmentation and

attrition during the fluidized bed combustion of two waste-derived
fuels, in: C. Laguerie, J.F. Large (Eds.), Fluidization VIII, Engi-
neering Foundation, New York, 1995, pp. 437–444.

[8] U. Arena, R. Chirone, P. Salatino, The fate of fixed carbon during

the fluidized-bed combustion of a coal and two waste-derived
fuels, Proceedings of the Combustion Institute 26 (1996) 3243–
3251.

[9] M. Fiorentino, A. Marzocchella, P. Salatino, Segregation of fuel

particles and volatile matter during devolatilization in a fluidized
bed reactor-I-, Model development, Chemical Engineering Science
52 (1997) 1893–1908.

[10] M. Fiorentino, A. Marzocchella, P. Salatino, Segregation of fuel

particles and volatile matter during devolatilization in a fluidized
bed reactor––II, Experimental Chemical Engineering Science 52
(1997) 1909–1922.

[11] F. Scala, Fluidized bed combustion of high-volatile solid fuels,

PhD Thesis, Universit

a

a degli Studi di Napoli Federico II, Italy,

1999.

[12] F. Scala, P. Salatino, Modelling fluidized bed combustion of high-

volatile solid fuels, Chemical Engineering Science 57 (2002) 1175–
1196.

[13] R. Chirone, M. DÕAmore, L. Massimilla, A. Mazza, Char

attrition during the batch fluidized bed combustion of a coal,
AIChE Journal 31 (1985) 812–820.

[14] S. Masi, P. Salatino, O. Senneca, Combustion rates of chars from

high-volatile fuels for FBC application, in: Proceedings of the 14th
International Conference on Fluidized Bed Combustion, ASME,
New York, 1997, pp. 135–143.

[15] P. Salatino, F. Scala, R. Chirone, P. Pollesel, Optimization by air

staging of the fluidized bed combustion of tyre derived fuel,
Proceedings of the 4th International Conference on Technologies
and Combustion for a Clean Environment, Lisbon, Portugal,
1997, pp. 23–27.

[16] A. Cammarota, R. Chirone, P. Salatino, F. Scala, O. Senneca,

Fluidized bed combustion of tyre derived fuel, Proceedings of the
International Symposium on Recycling and Reuse of Used Tyres,
Dundee, UK, 2001, pp. 69–80.

F. Scala et al. / Experimental Thermal and Fluid Science 27 (2003) 465–471

471