This article was downloaded by: [North Dakota State University]

On: 24 November 2014, At: 08:18

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Combustion Science and Technology

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gcst20

Simulating the Impact of Oxygen Enrichment in a

Cement Rotary Kiln Using Advanced Computational
Methods
a a a a b
OVIDIU MARIN , OLIVIER CHARON , JACQUES DUGUE , SARAH DUKHAN & WEI ZHOU
a
Air Liquide, Chicago Research Center , 5230 s. East Ave, Countryside, IL, 60525
b
Fluent Inc.
Published online: 27 Apr 2007.

To cite this article: OVIDIU MARIN , OLIVIER CHARON , JACQUES DUGUE , SARAH DUKHAN & WEI ZHOU (2001) Simulating
the Impact of Oxygen Enrichment in a Cement Rotary Kiln Using Advanced Computational Methods, Combustion Science and
Technology, 164:1, 193-207, DOI: 10.1080/00102200108952169

To link to this article: http://dx.doi.org/10.1080/00102200108952169

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the
publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations
or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any
opinions and views expressed in this publication are the opinions and views of the authors, and are not the
views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be
independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses,
actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever
caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 COmblUl. Sci. and T~ch .• 2001. Vol. 164. pp. 193-207 e 2001OPA (Overseas Publishers Association) N.V.
Reprints available directly from the publisher Published by license under the
Photocopyingpermutedby licenseonly Gordon and BreachScience Publishers imprint.
a memberof the Taylor & Francis Group.

Simulating the Impact of Oxygen

Downloaded by [North Dakota State University] at 08:18 24 November 2014

Enrichment in a Cement Rotary Kiln

Using Advanced Computational
Methods
OVIDIU MARINa', OLIVIER CHARON a, JACQUES DUGUE a,
SARAH DUKHAN a and WEI ZHOU b

8Air Liquide, Chicago Research Center 5230 s. East Ave. Countryside, tL 60525,
and bFluent Inc.

(Received June 16, 2000; In final form December 4, 2000)

This work presents the simulation of a rotary kiln used to produce cement clinker. The effort uses an
original approach to kiln operation modeling. Thus, the moving cement clinker is accurately simu-
lated, including exothermal reactions into the clinker and advanced heat transfer correlations. The
simulation includes the normal operation of a cement kiln. using coal in an air-fired configuration.
The results show the flame characteristics. fluid flow. clinker and refractory characteristics. Two
types of coal are employed, one with medium-volatile and nne with low-volatile content, with signif-
icant differences noted in the kiln operation.
A specific goal of this effort is to study the impact of oxygen enrichment on the kiln operation. For
this purpose, oxygen is lanced into the kiln at a location between the load and the main burner, and
the impact of oxygen enrichment on the kiln operation is assessed. Different oxygen injection
schemes are also studied. Thus, varying the angle of the oxygen lance enables to handle various prob-
lems as reducing conditions, overheating in the burning zone or refractory wall. It is concluded that
oxygen has a beneficial role in the fuel combustion characteristics, and its impact on refractory tem-
perature and the clinker is negligible, in conditions of increased productivity and overall efficiency.
The paper presents the impact of dust insufflation into the kiln, such as reduced temperature pro-
file, resulting in a less stable combustion process. The work shows the beneficial influence of oxygen
enrichment on kiln operation in the presence of dust, leading to an increase in the amount of dust
capable of being insufflated into the kiln.
The paper presents the impact of dust insufflation into the kiln, such as reduced temperature pro-
file. resulting in a less stable combustion process. The work shows the beneficial influence of oxygen
enrichment on kiln operation in the presence of dust. leading to an increase in the amount of dust
capable of being insufflated into the kiln.

* Corresponding Author: 5230 S. East Ave., Countryside, IL 60525, e-mail: ovidiu.marin@airliq-

uide.com

193
 194 OVIDIU MARIN et al.

INTRODUCTION

Rotary kilns have a wide application fu the chemical, metallurgical, refractory

and mineral processing industries. The complexity of these processes, the impor-
tant costs related to the construction, operation and maintenance of the machines
require a good understanding of the intrinsic phenomena occurring in rotary
kilns. Due to the size of many industrial rotary kilns, and the fact that measure-
ments are difficult to perform in these hard to access devices, theoretical models
Downloaded by [North Dakota State University] at 08:18 24 November 2014

have been developed to simulate such processes as chemical, thermodynamics,

etc. Thus, mathematical models are capable to predict such effects as heat trans-
fer, temperature profiles, radiative effect, material bed motion etc.
Jenkins and Moles (1981) have developed a mathematical model aimed at pre-
dicting the temperature profiles of the gas and refractory in a kiln. The approach
employs Hottel and Sarofim's (1967) radiative heat transfer zonal method, and
allows the designer and operator to optimize the kiln operation. Lowes and
Evans (1988) present a model for coal flame optimization, and also control strat-
egy based on the level of NOx. Gorog et al. (1983) develop a simplified heat
transfer model in the flame zone, and partially validate it with laboratory experi-
ments. Tscheng and Watkinson (1979) present a method to calculate the convec-
tive heat transfer between the walls and the clinker used during the present effort.
Boateng and Barr (1996) develop a model to predict the particle motion in the
shearing active layer, essential in developing the temperature gradients in the
cement bed. In the same spirit, Henein et al. (l983a) obtain experimental trans-
verse bed motion modes in a rotary kiln, and develop a model (l983b) to predict
the conditions leading to different bed motion modes:~uch as slumping, rolling,
etc. The results presented by Henein et al. (l983b) have been used in the current
effort to justify the transverse motion employed. Silcox and Pershing (1990)
develop a complex heat transfer model for a rotary kiln, including the radiative
heat transfer from the flame, and also the conductive and convective heat transfer
modes.
A significant effort has been allocated to evaluate the impact of oxygen enrich-
ment on kiln operation. Oxygen enrichment has been demonstrated as a viable
method to increase clinker throughput, the clinker quality, to increase the amount
of dust insufflated, etc. One of the first documented uses of combustion air oxy-
gen enrichment in a rotary kiln is reported by Gaydas (1965), including experi-
mental results. High velocity oxygen is introduced between the load and the
main flame, to increase the temperature of the flame near the load. The results
show an increase in production of around 5-6%, with the same amount of fuel
fired into the kiln. Substantial work continues to support the use of oxygen in
cement/lime kilns (Anselm and Koch, 1953, Frankenberger, 1966, McEntee,
 THE IMPAcr OF OXYGEN ENRICHMENT 195

1973, Bruell, 1983, Foalle, 1984, Garrido et al., 1981), particularly for the pur-
pose of increasing production (Fredericks et al., 1971, Grissom et al., 1980, Wat-
kinson and Brimacombe, 1983 and Wrampe and Rolseth, 1976), burning waste
fuels (Coveney and Hicks, 1994 and Hansen et al., 1994), controlling emissions
(Coveney and Hicks, 1994), etc. Limited effort (Salisbury et al., 1993) has
focused on modeling rotary kilns in the presence of oxygen enrichment.
The purpose of this effort is to investigate the operation of a rotary kiln using
advanced modeling tools. The detailed simulation of an actual kiln is performed,
Downloaded by [North Dakota State University] at 08:18 24 November 2014

and novel models are employed, as described below. The base case simulation is
compared to operational data, with good agreement. This calculation is followed
by an investigation of oxygen enriched combustion, oxygen being introduced
into the kiln via a lance situated between the main burner and the clinker surface.
The impact of oxygen injection on the combustion process and on the overall
kiln operation is assessed. The increase in the clinker throughput, impact on the
furnace walls, reducing conditions at the clinker surface, etc. are also investi-
gated. Different oxygen lancing schemes are considered, and the significant dif-
ferences noted are commented. Finally, the impact of dust insufflation on kiln
operation is assessed, and the improvements due to oxygen enrichment are dis-
cussed.

METHODOLOGY

The geometry of the kiln is presented in Fig. I. The diameter of the kiln is around
3.50 meters, and out of the total kiln length, only the first 50 meters from the
burner wall have been modeled, due to computational considerations. The kiln
uses coal as fuel, and the air necessary for combustion is split into two main
streams: the primary air, used to carry the coal into the kiln, with a temperature of
around 350 K, and the secondary air, coming from the clinker cooler, preheated
to around 820 K. The raw material is introduced into the system at the opposite
end from the fuel, and this calculation assumes that the inlet temperature into the
system boundary is I lOOK. This value has been selected from experimental data
presented in [23]. The inlet temperature has also been correlated with the clinker
properties and the clinker outlet temperature of 1723 K. The coal, air and clinker
flow rates are set in accordance with the kiln operation.
The load (clinker) has been modeled as a distinct solid entity in motion. The
average velocity of the clinker can be calculated from the clinker production rate
and geometrical considerations. The surface of the clinker is considered to be
flat, in agreement with the results of Henein et al. (l983a, b), and the clinker
cross-section has been selected from experimental data. Since the geometry con-
 196 OVIDIU MARIN et al.

Fluegas

J.50m

Raw material

50 m
Downloaded by [North Dakota State University] at 08:18 24 November 2014

FIGURE I Schematics of the kiln model (See Color Plate XXIV at the back of this issue)

sidered here does not take into account the enhanced heat transfer between the
clinker and the flue gases/kiln wall, a modified heat conductivity has been
assigned to the solid body representing the moving clinker. The heat transfer
coefficients between the clinker and the walls, and between the clinker and the
flue gases as predicted by Tscheng and Watkinson (1979) have been correlated,
and the result has led to a modified, enhanced value for the heat conductive coef-
ficient of the clinker.
Finally, the exothermal reaction occurring in the clinker has been considered in
a simplified manner. Thus, a user-defined subroutine has been used to create a
heat source into the clinker in a region where the clinker temperature is between
900°C and 1200°C, as suggested by Hewlett (1998). The value of the heat source
is around 100 callg clinker, a value accepted by the industry.
The models used in the kiln simulation include:
• The coal combustion models, taking into account volatilization and char
combustion. Two different coals have been investigated: one with a medium
content of volatiles (27.6%), and a low-volatile coal with 10% volatile con-
tent. The two-competing rate devolatilization model has been employed, as
well as the surface reaction model for the char combustion.
The P I radiation model has been employed, due to the large dimensions of
the kiln (large path lengths). The PI model solves a diffusion equation and it
includes the effect of scattering. The model assumes that all surfaces are dif-
fuse. The weighted sum of gray gas model (WSGGM) is used to calculate the
absorption coefficient.
• The realizable k-s turbulence method is used, in order to appropriately
address the turbulence in the combustion space/flow field.
The boundary conditions have been addressed above, including the tempera-
ture of the primary air/coal mixture, secondary air and the clinker inlet tempera-
ture. The ratio between the primary and secondary air is 40/60%.
 THE IMPACT OF OXYGEN ENRICHMENT 197

The following cases have been investigated: the base case, including air-only
combustion. The results have been validated with the experimental data availa-
ble, and the results are in good agreement in terms of temperature and flow rates
for both clinker and air/fuel data. Differences of less than 3% for the air/fuel flow
rates, and less than 10 K for the temperatures of clinker and the flue gases have
been obtained. This case has been considered to be the reference for all the other
calculations. It is noted that each calculation, for the base case and the oxygen
enrichment cases include two different coals, one with 27.6% volatile content,
Downloaded by [North Dakota State University] at 08:18 24 November 2014

and another with 10% volatiles.

The oxygen enrichment cases investigated include: lancing oxygen horizon-
tally, with a slope of 10 degrees upwards, and 10 degrees downwards, respec-
tively. These calculations attempt to show the impact of oxygen on the kiln
operation, and also to assess whether oxygen lancing can be fine-tuned to address
particular issues, such as reducing conditions at the clinker surface, maximum
flame temperature, impact on refractory temperature profile, etc.
Finally, the issue of dust insufflation is investigated. The impact of dust insuf-
flation on operating conditions for the base case is assessed. The advantages of
using oxygen for balancing the negative influence of dust on the combustion
process are also shown. It is concluded that the oxygen enrichment allows to
increase the amount of dust capable of being insufflated into the kiln, thus lead-
ing to increase in the clinker throughput.

RESULTS AND DISCUSSION

Air Operation

Figure 2 presents the longitudinal temperature profiles in the kiln at the burner
level for the medium volatile coal (Fig. 2a) and for the low volatile coal
(Fig. 2b), respectively. Several important aspects can be identified, such as the
more intense flame in Fig. 2a, localized in a relatively small volume. This is
attributed to the larger amount of volatile content, leading to an enhanced heat
release once the volatilization process is initiated. The flame shape in Fig. 2b (for
the low volatile content) is less structured, and it spreads over a larger volume,
due to the higher inertia in char combustion (more char, harder to ignite and com-
bust).
Significant differences can also be noted in the flow patterns between the two
cases, as presented in Fig. 3. Figure 3a presents the velocity profile in the com-
bustion zone for the 27.6% volatile coal case, while Fig. 3b presents the velocity
 198 OVIDJU MARIN et at.

Temperature [K]

=
1::
u.;,.; 1.8lle+03
Downloaded by [North Dakota State University] at 08:18 24 November 2014

~~~ 148&+03
2a) 27.6% volatile content.
IIlh2!le+03

1.100+03
JP~:_::l
-.~ ,,,,-f'o

1.218+02
-- .. ~:i:-'':J; . ~

5.3!le+02
2b) 10.0% volatile content.

FIGURE 2 Temperature lK] profile in the combustion zone for the air operation, a) 27.6% coal vola-
tile content, and b) 10% coal volatile content (See Color Plate XXV at the back of this issue)

profile for the 10% volatile coal. In the medium-volatile case, there is a signifi-
cant recirculation observed in the upper side of the kiln, above the burner, clearly
marked in Fig. 3a. This recirculation is attributed to the more intense heat release
in the flame region, leading to a high temperature region, and thus to a higher
velocity profile in this area. The higher velocity gradient leads to a relatively
strong recirculation in the upper region of the kiln. For the low volatile coal case,
the temperature increase in the gas is more progressive, and the velocity profile
is more uniform (closer to plug flow characteristics), and the recirculations are
non-existent (Fig. 3b).
The recirculation impact on the medium-volatile combustion as presented in
Fig. 3a explains the apparent paradox noted in Figs. 2a and 2b, namely that the
low volatile case initiates the combustion process faster than the medium volatile
case, at least on the load (inferior) side. This relatively abnormal behavior can be
explained as follows: the recirculation in Fig. 2a makes the incoming secondary
air to flow primarily through the lower side of the kiln, and thus the velocities are
significantly higher on this side by an estimated 30%. This way, the volatilization
and combustion of the coal will be transported farther into the kiln, as seen in
Figs. 2a and 2b.
 THE IMPACT OF OXYGEN ENRICHMENT 199

Velocity [m/s] Velocity [m/s]

Recirculation zone

1= '.818+01

(;:1-4·~01
Downloaded by [North Dakota State University] at 08:18 24 November 2014

f-':~
33) 27.6% volatile content _4.15..01
".st.+01

3b) 10% volatile content

·1.42...01

FlGURE 3 Velocity [m/s] profile in the combustion zone for the air operation. a) 27.6% coal volatile
content. and b) 10% coal volatile content (See Color Plate XXVI at the back of this issue)

The results above show that the fuel quality has an important effect not only on
the combustion characteristics. but also on the flow characteristics of the highly
restricted kiln geometry. This behavior makes necessary to adjust the burner
placement/operation following changes in the fuelJoxidant characteristics. The
results below show the impact of oxygen on the combustion/overall kiln opera-
tion.

Oxygen-Enriched Operation

This section investigates the impact of oxygen enrichment on kiln operation. The
oxygen is introduced into the kiln via a lance located between the main burner
and the load. This lancing procedure is traditional for kiln applications, since it
directs the bulk of the radiative heat transfer towards the load, while the refrac-
tory walls are shielded from possible overheating by the main flame. The oxygen
enrichment impact on the refractory walls and clinker are presented below.
The calculations presented here employ an amount of oxygen leading to an
enrichment of 23% oxygen in oxidant by volume (compared to 21% oxygen in
air). The production rate is increased (the "clinker" solid velocity is increased) to
preserve the same product outlet temperature, based also on the same inlet tem-
perature. The air input is modified such that the overall flue gas flow rate stays
 200 OV1DlU MARIN er al.

constant in the kiln. The fuel input is modified in order to maintain the same
excess oxygen as the base case. Thus, the fuel is increased by 5.75%, while the
production rate is increased by 6.4%, indicating an increase in the kiln efficiency,
in agreement with the results of Gaydas (I965). The results are also in excellent
agreement with the calculations provided by an independent heat and mass bal-
ance software package designed for kiln investigation. Thus, the results pre-
sented above are within 5% of the heat and mass balance calculations.
Figure 4a presents the longitudinal temperature profile in the kiln for the base
Downloaded by [North Dakota State University] at 08:18 24 November 2014

case (air operation, medium-volatile coal), and for the oxygen-enriched case in
Fig. 4b, respectively. It is noted that the maximum temperature of the flame is
around 150 K higher for the oxygen-enriched case, and also that the combustion
process initiates faster when compared to the air operation. The flame is better
defined in the oxygen case, suggesting a more stable operation in non-stationary
conditions, such as change of coal properties, change of load, etc. The same char-
acteristics are noted for the low-volatile content coal when oxy-combustion is
implemented. These results, in line with practical experience, recommend the use
of oxygen enrichment in kilns where large quantities of low-volatile fuels such as
pet-coke are used.

Temperature [K]

2.3&>+03

2.18&+03

I.
" 1.97..03

1.n...03

4a) Air Operation

1.16&+03

9.59e+02

7.sa..ll2

5.530+02

4b) Oxygen Enriched Operation

FIGURE 4 Temperature (K] profile for a) Air operation, and b) Oxygen enriched operation (See
Color Plate XXVII at the back of this issue)

The increase in local temperature may have a negative impact on kiln opera-
tion, particularly with respect to NOx emissions and/or impact on wall and
 THE IMPACf OF OXYGEN ENRICHMENT 201

clinker temperature profile. In order to address these issues, the oxygen lancing
direction has been varied by 10° upwards and 10° downwards, respectively.
Figure S shows the temperature profile a longitudinal plane at the burner level
for the case with 10° downwards lancing (Fig. Sa), and 10° upwards lancing
(Fig. Sb), respectively. The results presented in Figs. S show a clear difference in
the temperature profile between the two cases. The case using lancing upwards -
towards the flame (Fig. 5b) shows that the mixing between the oxygen and the
flame occurs in a relatively small volume, which in tum has a high temperature,
Downloaded by [North Dakota State University] at 08:18 24 November 2014

with possible negative impact on emissions. When oxygen is lanced downwards

(Fig. Sa), the high temperature area is more limited, and thus the impact on kiln
operation is more limited, due to the gradual, staged mixing between the fuel and
the pure oxygen stream. The different profiles for the three oxygen injection
schemes strongly suggest that the lancing strategy is key in obtaining the desired
outcome - such as controlling emissions. The results presented above also sug-
gest that modeling can play an important role in tailoring a suitable oxygen
enrichment scheme.

Temperature [K]

2.39 03

12.19 03

~¥ 1.989+03

1.78<>+03

1.57.... 03

1.379+03 5a) Oxygen lanced 10' downwards

1.17.... 03

9.62 .... 02

7.58&+02

5.54 .... 02
IIj c~ •...
--
, _:'. ",..',:-"'

- !I
- • ~. .?'. • <
:.r~~
;',: .~;

5b) Oxygen lanced to' upwards

FIGURE 5 Temperature [K] profile for the case when oxygen is lanced with an inclination of a) 10°
downwards, and b) 10° upwards (See Color Plate XXVIII al the back of this issue)

The impact of oxygen enrichment on the clinker bed is also investigated.

Figure 6 presents the temperature variation at the clinker surface for the air oper-
 202 OVIDIU MARIN et al.

ation and the oxygen enrichment operation. The results show that the tempera-
ture of the clinker is relatively the same for both cases. This implies that the
oxygen enrichment will not lead to local overheating of the clinker, with negative
impact on-clinker quality. The slight decrease in the clinker temperature profile
for the oxygen enrichment case is attributed to the higher clinker pull rate, pro-
portionally higher than the fuel increase. The process of lancing oxygen between
the main burner and the load leads to a higher oxygen content at the clinker sur-
. face in the final stage of the clinker formation, and thus reduces the risk of reduc-
Downloaded by [North Dakota State University] at 08:18 24 November 2014

ing conditions at this level.

'000
AirOperation
H'OD
Oxygen-Enriched Operation
"00'

teco

Slatlc '000
Temperature
(k) ,,"00

''''''
-ecc
"oo-l---~--~--_--_-......;'---~
o '0
Position (m) '"
Air casa'Oxygen case: Z7 .S'" of ychdiJo content

I 51,,"0 Tom".,.....-.
FLUENT 5.1 (3d.
Aug 05. , ....
Mgr~Iltad.spe5. rk.)
1

FIGURE 6 Temperature profile next to the clinker bed for the air-operation, and for the oxy-
gen-enriched operation, respectively (See Color Plate XXIX at the back of this issue)

Figure 7 presents the impact of oxygen enrichment on the refractory wall. The
longitudinal temperature profiles in the combustion space next to the wall for the
air operation and the oxygen-enriched operation, respectively, are presented.
These profiles correspond to the highest elevation in the kiln. The results indicate
that the oxygen enrichment leads to a slight increase of the temperature next to
the walls, with a maximum variation of around 18 K. This increase is considered
to be relatively small, unlikely to cause refractory problems. It is reminded that
by varying the oxygen lancing angle, the impact on the combustion process can
be controlJed. Thus, by lancing the oxygen towards the flame increases the
refractory temperature by around 20 K in addition to the results in Fig. 7, while
lancing the oxygen downwards reduces the maximum temperature on the refrac-
tory wall by around 5 degrees (when compared to horizontal oxygen lancing).
 THE IMPACT OF OXYGEN ENRICHMENT 203

............. Air Operation

_ _ - Oxygen-Enriched Operation

~
--- -"'-""-""-""'-.":...-,..
.......
' ~~
I --.......
! ---...,
,,
SIalic
Temperature
(k)
Downloaded by [North Dakota State University] at 08:18 24 November 2014

Position (m)
"'r caselOxygen case: 27.6% of volatile content

Sta1ic Temperalur. Aug OS. 1999

FlUENT 5.1 (3d, segTogB1od, ........ rke)

FIGURE 7 Temperature profile next to the refractory wall for the air-operation. and for the oxy-
gen-enriched operation. respectively (See Color Plate XXX at the back of this issue)

It is concluded that oxygen operation can be performed safely and with good
results when the basic concepts of the mixing/combustion processes are well
understood, and the process is optimized.

Operation with Dust Insufflation

Finally. this investigation presents results related to the issue of dust insufflation.
For certain kiln operations, a large amount of dust may leave the kiln entrained in
the flue gases. The dust can be either landfilled, sold at a lower price when com-
pared to the clinker, or re-insufflated into the kiln. The insufflation technique
requires that the dust be entrained in a current capable of transporting it from the
cold end back to the burner end, and lanced into the kiln. The impact of the dust
on the combustion process can be significant, since the inert particles will inhibit
the flame. Occasionally, kiln operators control the kiln by appropriately varying
the dust in-flow - thus, when the outlet clinker temperature increases, the amount
of dust is increased, or inversely when the temperature decreases, the amount of
dust insufflated decreases as well.
The use of oxygen may be highly beneficial for kiln operations including dust
insufflation. The improved combustion characteristics in the presence of oxygen
will ensure a more stable flame, thus the amount of dust capable of being insuf-
 204 OVIDIU MARIN et al.

flated can be increased. Different schemes of combining the dust insufflation and
the oxygen lancing can be envisioned. This effort considers the injection scheme
to be as presented in Fig. 8. Since the clinker surface is considered to be plane,
for the purpose of this modeling effort it has been assumed to be horizontal (in
practice, due to the kiln rotation, it is inclined). The dust insufflation location is
at an angle from the oxygen lance.
Downloaded by [North Dakota State University] at 08:18 24 November 2014

FIGURE 8 The oxygen lancing/dust insufflation scheme (See Color Plate XXXI at the back of this
issue)

Using the scheme presented in Fig. 8, several calculations have been per-
formed, and the results have been compared to the base operation. Thus, dust has
been insufflated in the air operation, and also in the oxygen-enriched operation.
These results are presented in Fig. 9, showing the longitudinal temperature pro-
files at the burner level for the base case and oxygen enriched operation, in the
 TIffi IMPACT OF OXYGENENRICHMENT 205

absence/presence of dust insufflation. The results clearly show the impact of dust
on the temperature profile, with a significant decrease in the temperature magni-
tude noted in the presence of dust, as expected. Thus the maximum temperature
decrease is around 50 K for the oxygen enriched operation, and around 40 K for
the air operation. It is noted that by reducing the temperature magnitude in the
kiln, the dust impact on the oxygen-enriched operation will most likely lower the
NOx emissions. A future effort will attempt to present a parametric study on the
issue of dust insufflation, by analyzing the relationship between the oxygen
Downloaded by [North Dakota State University] at 08:18 24 November 2014

enrichment and dust insufflation, as well as assessing the NOx emissions.

2350 - - Air· No Oust

- •• Air - Oust

-
~
Gl
2300
--Oxygen - No Oust
••• ·Oxygen - Oust

-
~
:::l
III
~
2250
Gl
Q.

...E
Gl
2200

2150
7
FIGURE 9 Dust insufflation and its impact on kiln operation in the air operation and in the presence
of oxygen lancing (See Color Plate XXXII at the back of this issue)

It can be concluded that the oxygen enrichment can be efficiently used to

increase the amount of dust insufflated into the kiln, by ensuring a more stable
combustion process, and also by providing an increased heat transfer to the
load/dust.

CONCLUSIONS

This effort uses advanced computational fluid dynamics concepts and tools to
model the operation of a cement kiln. FLUENT software has been employed in
an innovative way to accurately represent the actual kiln operation. Thus, the
most up-to-date models have been employed to simulate the coal combustion
process, the fluid dynamic and heat transfer phenomena in the gas phase. This
 206 OVIDIU MARIN 'I al.

paper also includes an original treatment of the clinker, by approximating it with

a moving volume, and including accurate properties and heat transfer character-
istics.
The calculations include the air operation, which is validated by experimental
data. Two air-fired operations are envisioned, using a medium-volatile coal and a
low-volatile coal, respectively. The results presented above show that the fuel
quality impacts the combustion process and the flow patterns. Thus, the medium
volatile case presents a recirculation zone in the flame region, while the low-vol-
Downloaded by [North Dakota State University] at 08:18 24 November 2014

atile coal does not. These results may provide important insight on means to
more effectively operate the kiln when different fuels are employed.
The air operation is compared to different alternate operational scenarios. First,
oxygen enrichment impact on the kiln operation has been investigated. Oxygen is
lanced in the kiln between the main burner and the load. The amount of oxygen
lanced is chosen such that the overall flue gas flow rate is constant, in conditions
of increased fuel/heat flux. The oxygen enriched operation leads to an increase in
the clinker output, and also to an increase in overall efficiency, attributed to
enhanced heat transfer rates to the load. The combustion process is improved by
the presence of oxygen, and the resulting flame is considered to be more stable,
thus suitable for fuels with less volatile content, or with variations in the heating
value/moisture and ash content.
The oxygen lanced horizontally in the kiln leads to higher peak temperatures in
the flame, creating the possibility to increase the NOx emissions in the flue
gases. In order to limit this trend, different injection angles have been simulated,
and the results show a powerful impact of the injection direction on the flame
characteristics. This conclusion implies that the oxygen enrichment technique
has to be carefully chosen before implementation, and that advanced tools such
as CFD simulations can positively influence the technology selection process.
The impact of dust insufflation on kiln operation has been also investigated.
The results show that the dust has a strong influence on the flame characteristics,
by lowering the flame temperature profile, and thus by potentially leading to
unstable flames. The augmentation of the combustion process via oxygen
enriched combustion addresses this issue in a satisfactory method. Thus, besides
all the other advantages of the method, the oxygen enrichment can lead to an
increase in the amount of dust which can be recirculated. Additionally, the com-
bined use of oxygen and dust can lower the NOx emissions by reducing the
flame temperature profile, when compared to oxygen enrichment only.

References
Anselm. W. and Koch.A. (1953) Brennteehnikmil Sauerstoff,Zement Kalk Gips. Sonderausgaheor.
3.
 THE IMPACf OF OXYGEN ENRICHMENT 207

Boateng, A. A. and Barr, P. V. (1996) Modelling of Particle Mixing and Segregation in the Transverse
Plane of a Rotary Kiln, Chern. Eng. Science, Vol. 51, pp. 4167-4181.
Bruell, G. M. (1983) Oxygen enrichment aids rotary kiln operator, Pit and Quarry. Coveney, D. F. and
Hicks, J. K. (1994) Oxygen enrichment for improvements in Emissions control while burning
waste fuels, Portland Cement Association National Meeting.
Foalle, K. D. (1984) Oxygen enhanced combustion, World Cement. Frankenberger, R. (1966) Utilisa-
tion de l'oxygene dans l'industrie cimentiere, Silicates Industriels, pp 259-267.
Fredericks, S. L, Stirling, R. D and Blessing, J. C. (1971) Oxy-fuel burners: minimum heat loss for
maximum profits, Rock Products.
Garrido, G. F.. Perkins, A. S. and Ayton, J. R. (1981) Amelioration de la recuperation de chaux grace
a un enrichissment en oxygene, Conference de la CPPA.Montreal.
Downloaded by [North Dakota State University] at 08:18 24 November 2014

Gaydas, R. A. (1965) Oxygen enrichment of combustion air in rotary kilns, Journal of the PCA R &
D Laboratories, pp. 4~6.
Gorog, J. P., Adams, T. N. and Brimacombe, J. K. (1983) Heat Transfer from Flames in a Rotary
Kiln, Metallurgical Transact. B, Vol. 14B, pp, 411-424.
Grissom, L., Service, G. and Vizvary, P. (1980) The use of supplemental oxygen in a lime kiln, CFI
Augusta.
Hansen, R. E.. Leger, C. B. and Ho, M.-D. (1994) Theory and practice with oxygen enrichment in a
cement kiln firing waste deriving fuels, Proceedings of the 13th Incineration Conference.
Henein, H., Brimacombe, J. K. and Watkinson, A. P. (1983a) The Modeling of Transverse Solids
Motion in Rotary Kilns, Metallurgical Transact. B, Vol. 14B, pp, 207-220.
Henein, H., Brimacombe, J. K. and Watkinson, A. P. (l983b) Experimental Study of transverse Bed
Motion in Rotary Kilns, Metallurgical Transact. B, Vol. 148, pp, 191-205.
Hewlett, P. C. (1998) Lea's Chemistry of Cement and Concrete, Fourth Edition, Arnold Publishers.
Hottel, H. C. and Sarofim, A. F. (1967) Radiative Transfer, McGraw-Hill, New York. Jenkins, B. G.
and Moles, F.D. (1981) Modeling of Heat Transfer From a Large Enclosed Flame in a Cylindri-
cal Furnace, Trans. Inst. Chern. Eng., Vol. 59, pp. 17-25.
Lowes, T. M. and Evans, L. P. (1988) Optimization of the design and Operation of Coal Flames in
Cement Kilns, Presented at the British Flame Days Conference, London.
McEntee, F. J. (1973) Oxygen enrichment, CPA Conference, pp 84-96.
Salisbury, J, Ibbotson, A. and Gredley, P. (1993) Mathematical modelling of a rotary lime kiln and the
use of innovative oxygen technology to improve thermal perfnrmance, TOTEM 8 meeting, Nov.
9-11.
Silcox, G. D. and Pershing, D. W. (1990) The Effects of Rotary Kiln Operating Conditions and
Design on Burden Heating Rates as Determined by a Mathematical Model of Rotary Kiln Trans-
fer, J. Air Waste Management Assoc., Vol. 40, pp. 337-344.
Tscheng, S. H. and Watkinson, A. P. (1979) Convective heat trasnfer in Rotary Kilns, Can. J. Chern.
Engng, Vol. 57, pp. 433-443.
Watkinson, A. P. and Brimacombe, J. K. (1983) Oxygen enrichment in rotary lime kilns, Canadian
Journal of Chemical Engineering, vnl. 61.
Wrampe, P. and Rolseth, H. C. (1976) The effect of oxygen upon the rotary kiln's production and fuel
efficiency: theory and practice, IEEE Trans. Ind. App., pp. 568-573.