CHEMICAL ENGINEERING TRANSACTIONS  
 

VOL. 61, 2017 

A publication of 

 

The Italian Association 
of Chemical Engineering 
Online at www.aidic.it/cet 

Guest Editors: Petar S Varbanov, Rongxin Su, Hon Loong Lam, Xia Liu, Jiří J Klemeš 
Copyright © 2017, AIDIC Servizi S.r.l. 

ISBN 978-88-95608-51-8; ISSN 2283-9216 

Using an Advanced Numerical Technique for Improving 

Pulverized Coal Combustion inside an Industrial Furnace 

Hrvoje Mikulčića,*, Eberhard von Bergb, Xuebin Wangc, Milan Vujanovića, 

Houzhang Tanc, Neven Duića  

aFaculty of Mechanical Engineering and Naval Architecture, University of Zagreb, Ivana Lučića 5, 10000 Zagreb, Croatia 
bAVL List GmbH, Hans List Platz 1, Graz, Austria 
bMOE Key Laboratory of Thermo-Fluid Science and Engineering, Xi’an Jiaotong University, Xi’an 710049, China 

 hrvoje.mikulcic@fsb.hr 

When it comes to industrial activities, one of the most greenhouse gas emitting industrial sectors is the cement 

industry. Cement industry as an energy intensive industry is alone responsible for a large amount of CO2 

emissions. Compared to the total global CO2 emissions in 2014, cement production contributed to global 

anthropogenic CO2 emissions with 8 %. Meaning that cement industry needs to adopt more energy efficient 

technologies and reduce its demand for fossil fuels. The aim of this paper is to present an advanced numerical 

technique for improving pulverized coal combustion inside an industrial furnace. With the development of 

increasingly affordable and powerful computers, use of numerical techniques, have become a very useful and 

cost-effective tool that is increasingly used in industry for detailed understanding of the combustion process 

and for the appropriate design and optimisation of different combustion units. In order to study the role of 

pulverized coal combustion inside a cement furnace, appropriate numerical models need to be used. In this 

study a numerical model of the pulverized coal combustion was used for the numerical analysis. A detailed 

three dimensional geometry of a cement furnace was simulated. The results gained by this simulation can be 

used for better understanding of complex thermo-chemical phenomena occurring inside the calculated 

furnace. 

1. Introduction 

Albeit its CO2 emission is higher than that of other fossil fuels, coal as an abundant resource, is the most used 

solid fossil fuel in industry and power generating sector worldwide. In China, world’s largest greenhouse gas 

(GHG) emitter, coal accounts for approximately 70 % of the primary energy consumption, with coal-fired 

power plants producing 97 % of China’s thermal power capacity (Niu et al., 2015). Due to rapid increases in 

the GHG concentrations in the atmosphere, cleaner and more sustainable production is being increasingly 

emphasized within all industrial sectors (Yong et al., 2016). 

Bearing in mind that cement is the most widely used material for housing and modern infrastructure, and that 

coal is the most used solid fossil fuel in the cement industry, the aim of this paper is to numerically analyse the 

coal combustion process inside the rotary kiln. With the development of increasingly affordable and powerful 

computers, use of Computational Fluid Dynamics (CFD) simulations, has become a very useful and cost-

effective tool that is increasingly used in industry for detailed understanding of the combustion process and for 

the appropriate design and optimisation of different combustion units (Juřena et al., 2016).  

Over the past five years there have been several studies that numerically investigated different cement 

production units inside which multi-phase phenomena occur. The cyclone preheating system, used for the 

heat exchange process between the raw material and the flue gases, was studied by Wasilewski (2017). The 

study showed that by optimizing the structure of the cyclone, an increase in separator efficiency can be 

achieved. In the subsequent study by Wasilewski and Brar (2017) the possibilities for improving the process of 

clinker burning was presented. The study showed that better solid particle collection results with reduction in 

the power consumption. In a study by Mikulčić et al. (2014) the reacting two-phase flow inside the cyclone was 

                               
 
 

 

 
   

                                                  
DOI: 10.3303/CET1761037

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Please cite this article as: Mikulčić H., von Berg E., Wang X., Vujanović M., Tan H., Duić N., 2017, Using an advanced numerical technique for 
improving pulverized coal combustion inside an industrial furnace, Chemical Engineering Transactions, 61, 235-240  
DOI:10.3303/CET1761037  

235



investigated. The study showed that CFD simulations are an effective tool for analysing of the heat exchange 

phenomena, and increasing of the separation efficiency. 

The numerical investigation of fuel substitution inside a cement calciner has been the research topic in the 

study by Mikulčić et al. (2015a). The study showed that different operating condition need to be used is a 

standardly used fossil fuel is replaced with an alternative fuel. In the following study by Mikulčić et al. (2015b) 

it was shown that the CFD simulation method can serve plant operators and practical engineers in the 

optimization of cement calciner’s operating conditions. 

When it comes to the numerical investigation of the cement rotary kiln, over the past five years there have 

been only few studies that investigated the combustion process inside of the rotary kiln. As inside of the rotary 

kiln the highest energy demanding process during cement production occurs, the clinkering process, the 

numerical investigations of this furnace are entirely needed (Mikulčić et at., 2016a). Elattar et al. (2014) 

numerically investigated the confined non-premixed jet flames inside rotary kilns for gaseous fuels, to 

understand the flame behaviour and heat transfer. The study resulted with useful design guidelines and 

dimensionless correlations that characterize the flame length. Granados et al. (2014) numerically evaluated 

the flue gas recirculation effect during coal oxy-fuel combustion inside a cement rotary kiln. The study showed 

that the flame length in the oxy-fuel combustion cases were 30 % to 65 % shorter than that in air combustion 

with a higher intensity. Consequently the study reports that this could allow shorter kiln designs and might 

improve high-temperature clinkering reaction process. However the study states that more research is needed 

in this area. Ariyaratne et al. (2015) investigated the effects of meat and bone meal combustion inside a 

cement rotary kiln. The study showed the importance of fine fuel grinding, and that compared to coal particles, 

meat and bone meal particles needed more time to fully combust due to high moisture content and slower 

devolatilisation. 

In order to study the role of pulverized coal combustion inside a rotary kiln, appropriate numerical model 

needs to be used. In this study a numerical model of the pulverized coal combustion was developed and 

implemented into a commercial CFD code AVL FIRE®, which was then used for the numerical analysis. Time-

averaged conservation equations were solved for the gaseous and particulate phase, the former being treated 

by the Eulerian formulation, and the latter by the Lagrangian formulation. A detailed three dimensional 

geometry of a cement rotary kiln was simulated. The results gained by this simulation can be used for the 

improving the understanding of the complex thermo-chemical phenomena occurring inside the calculated 

rotary kiln. Furthermore, it needs to be stressed that this study analyses only the coal flame. For the raw 

material that is undergoing the clinkering process, different physical models need to be developed. The 

moving bed physics need to be carefully modelled in order to properly simulate the chemical reactions taking 

place inside of this bed. This is part of future work. 

2. Numerical models 

In this study the Eulerian-Lagrangian method for solving the multi-phase flow phenomena was used. The 

motion and transport of the particles, through the rotary kiln, are tracked through the flow field using a 

Lagrangian formulation, while the gas phase is described by solving conservation equations using an Eulerian 

formulation. The coupling between the solid and the gaseous phases is taken into account by introducing 

appropriate source terms for interfacial mass, chemical species, momentum and energy exchange. The 

mathematical model used for the coal combustion calculation is treated in the Lagrangian spray module, 

where heterogeneous chemical reactions occur inside a particle as well as homogeneous reactions between 

components released from the particles and gas phase species. 

2.1 Pulverized coal combustion 
When pulverized coal particles travel through gas and interact with the gas in the rotary kiln, they rapidly heat 

up and undergo a series of conversion processes. The pulverized coal combustion model used in this study 

includes four steps: drying, devolatilisation, char burning, and combustion of volatiles. The four-step process 

for the combustion of the biomass and the coal particles has been reported in a recent review study (Williams 

et al., 2012).  

The coal particle first undergoes the drying process (see Eq. 1), after which the devolatilisation starts. After the 

drying process, the particle further heats up, and with an increase in the temperature, the devolatilisation 

process starts (see Eq. 2). During the devolatilisation, an important loss of weight occurs due to the release of 

volatile matter. The quantity and composition of the volatiles depend on the coal composition and the particle 

size and temperature. The composition of coal and the volatile matter released during the devolatilisation are 

estimated from the coal ultimate and proximate analysis. After the devolatilisation, only char and ash are left in 

the solid particle. Parallel to the devolatilisation and depending on the particle size and temperature, char 

oxidizes to CO or CO2 (see Eq. 3), and subsequently, only ash remains. Then, the ash particle is considered 

inert, as only residual ash heating occurs.  

236



The Eulerian-Lagrangian modelling approach for the numerical computation of the gas-solid flow used in the 

presented work was well documented in our previous study (Mikulčić et al., 2016b). In the mentioned study the 

numerical models were described in detail to accurately explain the thermo-chemical processes that govern 

the combustion process. The plausibility checks and quantitative checks and balances of the presented 

combustion model in the form of tests of a single particle in a single mesh cube are also given. Therefore for 

brevity, this paper only provides a brief description of the coal combustion modelling approach. The treated 

chemical reactions are:  

2  
 Coal(s)  Coal(s),

H O evaporation
wet dry  (1) 

 Coal(s) ( ) (s),pyrolysisdry Volatile g Char   (2) 

2 2(s) ( ) ( ),Char O g CO g   (3) 

2 2 2( ) ( ) ( ) ( ).Volatile g O g H O g CO g    (4) 

The heterogeneous reactions (Eq. 1, 2 and 3) cause mass transfer sources and sinks to the gas phase and 

particles. The homogeneous reactions (see Eq. 4) of volatile oxidation are treated within the gaseous phase 

reactions module of the CFD code used. Considered volatile species in this model are CO, CH4, H2, C6H6. For 

the oxidation of volatile species, a detailed chemistry approach is used for each of the homogeneous reaction.  

3. Computational details 

In order to numerically study the role of pulverized coal combustion inside a rotary kiln, a detail three 

dimensional geometry of an industrial cement rotary kiln was used. The rotary kiln is 45 m long and has a 

diameter of 3.4 m. The burner is 1 m inside of the rotary kiln and is centred along its axis. The high mass flow 

of pre-heated secondary air from clinker cooler enters on one side of the rotary kiln and together with the 

primary air and pulverized particles from the burner exits the rotary kiln on the other side of the rotary kiln. 

The mesh quality testing, mesh dependency tests, testing of the time-step size, and validation of modelling 

approach for pulverised coal combustion was analysed in our previous study (Mikulčić et al., 2016b). Based 

on these findings, in the simulation of a rotary kiln 251,393 cells were employed to discretize the 

computational domain. A transient mode of simulation was applied. Due to fast chemical reactions, a time step 

of 2.5x10-4 was used. The differencing scheme used for momentum, continuity and enthalpy balances was 

MINMOD Relaxed and for turbulence and scalar transport equations an Upwind scheme was applied. 

Turbulence was modelled by the standard k-epsilon model. The P-1 radiation model was used to model the 

radiative heat transfer and the effects of particle radiation.  

Figure 1 shows the boundary conditions that were used for simulation of the studied rotary kiln. As can be 

observed on the upper part of Figure 1, at the back of the rotary kiln, the outlet is positioned. Wall boundary 

condition is used for the complete cylindrical area of the rotary kiln. From this figure, the specific burner 

position can also be observed. The burner is positioned one meter inside of the rotary kiln, and after the 

secondary air inlet. The secondary air inlet depicted with red arrows, is the inlet of the hot gas stream coming 

from the clinker cooler. It was assumed that the secondary air enters the rotary kiln at a 45° angle. More detail 

description of the burner inlets can be seen on the lower part of Figure 4. The particular burner has a 

possibility to use three inlets, however in the studied case, only coal as a fuel was analysed, and therefore the 

secondary fuel inlet was considered as a wall boundary condition. It can be seen that the coal particles enter 

together with the carrying air through the inner inlet ring, whereas the swirl air, used for flame stabilization, 

enters through the outer ring.  

237



 

Figure 1: Rotary kiln boundary conditions 

4. Results and discussion 

In Figure 2, the temperature field inside the calculated rotary kiln is shown. It can be seen that due to the 

devolatilisation process, the highest temperatures are located close to the burner. This is due to the 

production of combustible volatiles and constantly available oxygen that is introduced with the secondary air. 

Furthermore, the flame shape can be observed. It can be seen that the flame is leaning towards the bottom of 

the rotary kiln. This is due to the high mass flow of secondary air, that is entering the rotary kiln at a 45° angle. 

The secondary air first goes to the top of the rotary kiln and then partly comes down, pushing the flame to lean 

towards the rotary kiln bottom. The higher temperatures that are observed away from the burner are a result of 

char oxidation. This statement is confirmed when looking at position of the particles in Figure 4.  

 

 

Figure 2: Temperature field inside the calculated rotary kiln 

Figure 3 shows the velocity field inside the calculated rotary kiln. As can be seen the velocity is highest close 

to the burner where the primary air and swirl air inlet are positioned. On the right side of the figure, where the 

secondary air inlet is positioned, it can be observed that this air stream enters the rotary kiln at a 45° angle. 

The air stream from the burner first is pushed down and later at approximately middle of the rotary kiln, it goes 

to the rotary kiln top.  

 

238



 

Figure 3: Velocity field inside the calculated rotary kiln 

Figures 4 shows the ash mass fraction in particles. The distribution of ash particles inside the calculated 

calciner is also shown. In this figure the increase of ash mass fraction towards the outlet can be observed. 

Also, it can be seen that the ash particles are located in the whole rotary kiln. As can be seen, the ash 

particles are blown away by the high velocity air stream from the burner to the kiln outlet. As ash particles 

move through the kiln, due to the velocity profile, they agglomerate towards the kiln outlet. Here should be 

noted that ash deposition and particle interaction were not considered in the current investigation. 

 

 

Figure 4: Ash mass fraction in particles 

Though there are no available field measurements for this rotary kiln, that would allow some comparison with 

numerical predictions, validation of the pulverised coal combustion model, used for this calculation, was 

performed previously. The results gained by this simulation can be used for better understanding of complex 

thermo-chemical phenomena occurring inside the calculated furnace. Furthermore, the simulation results 

showed that, due to the high mass flow of secondary air, that is entering the rotary kiln at a 45° angle, the 

flame is leaning towards the bottom of the rotary kiln. Therefore, inlet condition parameter variation can give 

different operating scenarios that can be explored by numerical simulations, and the corresponding results can 

be used for decision making. 

5. Conclusion 

To satisfy both the need for more cement and the increased governmental and public environmental concerns, 

cement producers are under pressure to reduce their fossil fuel consumption. Precisely because of this reason 

numerical investigations of various industrial processes have become a major area of the current research. 

Particularly the CFD simulations have shown to be a useful tool for the investigation of different operating 

conditions for coal firing. The aim of this paper was to demonstrate that CFD can be used for the analysis and 

investigation of the coal firing process inside the cement rotary kiln. A pulverized coal combustion numerical 

model was developed and implemented into a commercial CFD code AVL FIRE®, taking into account the 

effects of drying, devolatilisation, char oxidation, and volatile combustion. The model is detail enough to 

contain the relevant physical and chemical processes, yet simple enough to run on the big meshes needed for 

detailed CFD simulations of cement furnaces. The paper gives some preliminary results on coal firing 

numerical modelling. The results gained by this study can be used for the investigation and better 

understanding of particle kinetics and distribution and operating conditions of a particular cement rotary kiln. 

By improving the understanding of the coal firing process inside the rotary kiln, the operating process can be 

improved, and consequently a reduction in fuel consumption can be achieved. Thus, the presented numerical 

model can be used for an ex-ante investigation of the complex thermo-chemical phenomena occurring inside 

the rotary kiln, e.g., regarding temperature distribution, flame position and burnout rate of the utilized fuels. 

239



Acknowledgements 

Authors would also wish to thank Dr. P. Priesching and Dr. R. Tatschl, from the CFD Development group at 

AVL-AST, Graz, Austria, for their continuous support and useful discussions during the development of 

numerical models used in this study. 

References 

Ariyaratne W.K.H., Malagalage A., Melaaen M.C., Tokheim L.A., 2015, CFD modelling of meat and bone meal 

combustion in a cement rotary kiln – Investigation of fuel particle size and fuel feeding position impacts. 

Chemical Engineering Science, 123, 596–608. 

Elattar H.F., Stanev R., Specht E., Fouda A., 2014, CFD simulation of confined non-premixed jet flames in 

rotary kilns for gaseous fuels, Computers and Fluids, 102, 62–73. 

Granados D.A., Chejne F., Mejia J.M., Gómez C.A., Berrio A., Jurado W.J., 2014, Effect of flue gas 

recirculation during oxy-fuel combustion in a rotary cement kiln, Energy, 64, 615–625. 

Juřena T., Bělohradsky P., Skryja P., Hudák I., 2016, Numerical modelling of combustion in 1.5 MW Low-NOx 

burner, Chemical Engineering Transactions, 52, 1261-1266. 

Mikulčić H., Vujanović M., Duić N., 2015a, Improving the sustainability of cement production by using 

numerical simulation of limestone thermal degradation and pulverized coal combustion in a cement 

calciner, Journal of Cleaner Production, 88, 262-271. 

Mikulčić H., Wang X., Vujanović M., Tan H., Duić N., 2015b, Mitigation of climate change by reducing carbon 

dioxide emissions in cement industry, Chemical Engineering Transactions, 45, 649-654. 

Mikulčić, Hrvoje, Jiří Jaromír Klemeš, Vujanović M., Urbaniec K., Duić N., 2016a, Reducing greenhouse 

gasses emissions by fostering the deployment of alternative raw materials and energy sources in the 

cleaner cement manufacturing process, Journal of Cleaner Production, 136, 119-132. 

Mikulčić H., von Berg E., Vujanović M., Wang X., Tan H., Duić N., 2016b, Numerical evaluation of different 

pulverized coal and solid recovered fuel co-firing modes inside a large-scale cement calciner, Applied 

Energy, 184, 1292-1305. 

Niu Y., Liu X., Zhu Y., Tan H., Wang X., 2015, Combustion characteristics of a four-wall tangential firing 

pulverized coal furnace, Applied Thermal Engineering, 90, 471-477. 

Wasilewski M., 2017, Analysis of the effect of counter-cone location on cyclone separator efficiency, 

Separation and Purification Technology, DOI:10.1016/j.seppur.2017.02.012 

Wasilewski M., Brar LS., 2017, Optimization of the geometry of cyclone separators used in clinker burning 

process: A Case Study, Powder Technology, DOI: 10.1016/j.powtec.2017.03.025 

Williams A., Jones J.M., Ma L., Pourkashanian M., 2012, Pollutants from the combustion of solid biomass 

fuels, Progress in Energy and Combustion Science, 38, 113–137. 

Yong J.Y., Klemeš J.J., Varbanov P.S., Huisingh D., 2016, Cleaner energy for cleaner production: Modelling, 

simulation, optimisation and waste management, Journal of Cleaner Production, 111, 1-16. 

240