Direct Reduction of Iron Ore Simulation: A Simplified Approach

Revue de Métallurgie 107, 195–204 (2010)
c EDP Sciences
DOI: 10.1051/metal/2010022
www.revue-metallurgie.org
R evue de
Métallurgie
A simplified approach to the simulation
of direct reduction of iron ore
M. Vannucci1, V. Colla1, G. Corbo2 and S. Fera2
1
2
PERCRO-CEIICP, Scuola Superiore S. Anna, 56127 Pisa, Italy
ILVA S.p.A., Genova Works, 16512 Genova, Italy
Abstract – This paper describes a model for the simulation of the process of direct reduction of iron ore for steel production. The model is implemented through stand-alone
software and the simulation results have been compared with real experimental data. The
very good agreement between the actual and simulated data proves that the model, despite its relative simplicity, takes into account all the fundamental phenomena of iron ore
reduction.
Résumé – Une approche simplifiée à la simulation de la réduction directe de minerai de fer. Cet article décrit un modèle pour la simulation du processus de réduction
directe de minerai de fer pour la production d’acier. Le modèle a été mis en application par
l’intermédiaire d’un logiciel indépendant et les résultats de la simulation ont été comparés
à de véritables données expérimentales. L’excellente concordance entre les données réelles
et simulées montre que le modèle, en dépit de sa relative simplicité, prend en considération
toutes les variables fondamentales de la réduction du minerai de fer.
n recent years, the efforts of the steelmaking industry toward a reduction
of CO2 emissions have become more and more intensive. Continuous
technical improvements of existing processes have contributed to lowering such emissions, but drastic reductions can be achieved only through
the development of breakthrough technologies. For this reason, an important project has been developed since 2004 by a large consortium including
most of the European steel producers and many research institutions, within
the 6th Framework Program, which is entitled “Ultra-Low CO2 Steelmaking”
(ULCOS) [1, 2] and aims at investigating technologies capable of cutting the
CO2 emission of the steelmaking industry by an amount in the order of 50%.
Among the technologies that have been studied within ULCOS and are now
in an advanced experimental phase, Direct Reduction (DR) is one of the most
promising ones [3]. The emissions of DR-EAF plants are in fact already close to
the 50% target reduction expected by the ULCOS project and DR is designed
to be suitable for integration with other technologies, such as CCS, aiming for
the reduction of CO2 emissions, which could further improve the appeal of
this technology. DR is going to be a valid alternative to traditional routes from
an economic point of view as well, as in the future the price of NG is expected
to become highly competitive.
DR is an alternative route for the production of steel which was developed
in the late 70s and it is already applied, in countries where there is abundance
of natural gas, in a route including a pre-reduction furnace and an EAF for
melting [4, 5]. The DR process allows the production of Direct Reduced Iron
(DRI) by means of a mixture of reducing gases mainly composed of hydrogen
and carbon monoxide, which play the role of reducing agents.
The main part of the DR plant is the reduction shaft where the reduction
reactions take place. During the production the shaft is charged from the top
with iron ore and the reducing gases are blown from the bottom in order to
allow the reduction, while the produced DRI is collected from the bottom of
the shaft.
Literature results can be found related to the modeling of chemical
and physical transformations involved in the DR processes and, in particular, on the reduction kinetics [6,7]. One of the most efficient DR processes is the
I
Text received 02 March 2010
accepted 24 March 2010
Article published by EDP Sciences
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)

MIDREX
process, which exploits both
lumps and pellets as raw material and
recycles the used gas, thus showing low
consumption and reduced environmental
impact. The shaft furnace reactor of the

MIDREX
process has been simulated in [8].
In this paper a model is proposed, which
has been developed within the ULCOS
project in order to simulate in a fast but reliable way the reduction process which takes
place in the reduction shaft of a DRI production plant.
The driver for developing this model
has been the need for a simple-to-use design tool that could manage the new reducing gas composition envisaged for within
ULCOS and that could also quickly compare different burden materials. In order to
achieve this goal, non-conventional reduction testing procedures were developed and
a software package was implemented encompassing both test result interpretation
and DR reactor modeling.
In particular, given the geometry of the
shaft, and the composition, temperature and
pressure of the inflated gas, the model estimates the hourly DRI production that can
be obtained by using a specified burden material. The reduction kinetics of the burden
material is fundamental information for the
model and is characterized by means of another model named Ilva REduction Simulation (IRES), which is based on a series of laboratory tests carried out on the material.
The DR shaft model has been implemented as a one-dimensional Finite Element
Model (FEM), where the shaft is represented
by a cylinder formed by 50 overlapping elements, where the conditions affecting the
material reduction vary only along the vertical dimension.
The developed model of the DR shaft has
also been implemented in a stand-alone application named SAILORS (Sant’ Anna ILva
Ore Reduction Simulator), which is realized
in Visual Basic and combines a user-friendly
interface and high computation capabilities.
The DR shaft model was tested in order to evaluate the accuracy of the produced
simulations. In order to validate the developed model, some real data from the most
common DRI production plants were compared with the results of the corresponding
simulations performed through the developed model. In particular, the tests refer to a
set of different configurations both for the
dimension of the shaft and for other parameters such as inlet gas composition and
196
temperature, while the burden material used
is for all tests the same kind of commercial
pellets. The variability in the parameters allowed us to test the simulator in a wide range
of conditions. The results of the comparison
show the very good agreement between real
and calculated DRI production and demonstrate the goodness of the developed model.
The paper is organized as follows: Section 2 describes the IRES model, Section 3 is
devoted to the description of the DR shaft
model, Section 4 depicts the software that
has been developed in order to make the
models easy to use in an experimental context and, finally, Section 5 presents some numerical results, by comparing the model predictions with some experimental data from
real industrial plants.
Characterization of the reduction
behavior of the burden material
The DR shaft model has been implemented
as a one-dimensional FEM, which is based
on a subdivision of the shaft in 50 overlapping elements of the same diameter of the
shaft. The basic assumption is made, that in
each element the conditions affecting the material reduction can vary along the axes of the
cylinder and not along its radius. This hypothesis, although quite schematic, allows
considerable simplification of the process
representation and, as a consequence, of the
related computations.
The reduction of burden material which
takes place within each layer is simulated by
using the IRES model. IRES is a model for
simulating the reduction behavior of an arbitrary material depending on temperature
and gas composition, provided that a set of
kinetic characteristics related to the material
have been determined through laboratory
tests.
The general kinetic laws obtained from
tests performed at constant conditions of
temperature and gas composition were subsequently validated by means of other
tests performed at variable temperature and
gas composition conditions. For the characterization of a material various isothermal reducibility tests were run at different temperatures and different reducing gas
compositions. During the tests, the weight
loss was continuously registered and subsequently converted into a reduction percentage (indicated by R in the following). The
variability ranges of conditions affecting reduction as well as some parameters of the lab
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
700°C
900°C
Fig. 1. Sample results of the reduction tests. These tests were carried out at constant temperature and gas composition.
Fig. 1. Résultats d’échantillon des tests de réduction. Ces essais ont été effectués à la composition en température et en gaz constante.
tests are listed in Table 1, while some sample
test results are shown in Figure 1.
As depicted in Table 1, the tests refer to
typical DRI conditions [9]. The pellets used
for the tests were provided by one of the industrial partners of the ULCOS project and
they belong to the common types of pellets
used in DRI plants [10]. On the basis of the
results of the experiments, the kinetic laws
of the reduction reaction were assessed. The
following general kinetic law was considered [6]:
R (t) = 100 1 − e−Kt
(1)
where t is the time and K is the kinetic factor
depending on the material type and size, the
reducing environment and the temperature.
The process of correlating the experimental
results with the main test conditions (such
as temperature and gas composition) underwent various stages. As a conclusion it was
found that:
– the reduction behavior of all materials
tested can be adequately described by
equation (1);
Table 1. Variability in the test conditions for the main parameters.
Tableau 1. Variabilité des conditions d’essai pour les paramètres principaux.
Parameter
Temperature
H2
CO
CO2
Gas speed (STP)
Gas flow (STP)
Reduction section diameter
Initial sample weight
Pellet size
Min
Max
◦
900◦ C
85%
48%
4.5%
700 C
45%
7%
2.5%
0.057 m/s
9.21 L/min
59 mm
307 g
12.5 mm
15 mm
– the gas composition can be represented
by introducing the concept of equivalent
% CO content: 1 mole of H2 = 2 moles of
CO (so that, for instance, 50% CO + 10%
H2 behaves as 70% CO);
– The parameter K can be linearly correlated to the gas composition and temperature, i.e. K = a • X + b, where: a and b are
197
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
Fig. 2. Comparison between a laboratory test and its IRES predicted behavior.
Fig. 2. Comparaison entre un essai en laboratoire et son comportement prévu par IRES.
constants to be determined by fitting the
experimental results and X is a parameter defined as:
T 2.5 [CO]eq
X = 100
(2)
1000
1000
where T is the absolute temperature and
[CO]eq is the percentage of equivalent CO.
Usually the root mean square error between actual test data and those obtained
from the above correlation lies within a few
points in percentage.
Figure 2 depicts the comparison between
a simulation of material reduction and its
measured behavior, showing the good performance of the model for isothermal and
constant gas composition tests. The simulation shown refers to an extreme situation
within given test ranges and highlights the
goodness of the proposed fitting also for borderline conditions.
The kinetic parameters determined
through the above-described tests are stored
in a database which describes the reduction behavior of several materials in different conditions. Such parameters are used to
simulate the reduction of the tested materials as well as mixtures of them for varying
temperature and gas composition in order
to be in line with the conditions of the DRI
production shaft.
Varying conditions are simulated by
considering any temperature-gas profile as
a sequence of steps whose duration is
198
sufficiently short to be assumed as isothermal and with constant gas composition.
Moving from a step to the following one
requires the concept of virtual time, i.e. the
time required to reach under a hypothetical
history with the temperature and gas composition of the next step, the same reduction index already reached during the real
history. The extra reduction gained in the
subsequent step is equivalent to that which
would be obtained in a step of equal duration beginning at virtual time.
The concept is described in Figure 3,
which refers to a 2-step history: the first one
consisting of a 100-minute reduction time
with 20% CO followed by 100 minutes with
50% CO.
The DR shaft model
IRES is used within SAILORS as a reduction module within a more general model designed to simulate the shaft of a DR furnace.
In particular, IRES is used to simulate the reduction of burden material in each element
of the 1-dimensional FEM.
The following main assumptions were
made for the development of the model:
– In order to exploit FEM simulation the
shaft is divided into 50 cylindrical overlapping elements. In each element conditions are assumed to vary only along the
vertical axis of the cylinder and not in the
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
Fig. 3. Illustration of the “virtual time” concept.
Fig. 3. Illustration du concept de « temps virtuel ».
radial direction. Elements are numbered
1 to 50 starting from the bottom.
– Reducing gas and burden material move
in opposite directions: gas enters from
the bottom of the shaft while material is
put in from the top.
– Reducing gas is formed by: H2 , H2 O, CO
and CO2 .
A scheme of the described shaft including
input and output of gases and materials is
shown in Figure 4.
The main inputs needed by the model,
which correspond to the main input variables of the shaft, are the following:
– Dimension of the reduction shaft (height
and diameter).
– Flow rate, composition, temperature and
pressure of inlet gas.
– Inlet burden material temperature.
– Target reduction of outlet material.
On the other hand, the model provides the
following outputs:
– Produced DRI outlet flow.
– Temperature profile for both gas and material in the shaft.
– Reduction profile of material in the shaft.
Within the FEM, representing a single layer
of the shaft, several physical processes and
relations among variables are considered. In
this framework, all reactions are heavily affected by the composition of the reducing
gas. From the thermal point of view, there is
heat exchange between gas and material (gas
cools as it reaches the upper part of the shaft,
while material gets warmer as it descends toward the bottom of the shaft itself). From the
chemical point of view, gas reduces the mineral by subtracting oxygen which combines
with H2 and CO and forms CO2 and H2 O.
Such reactions lead to an energy exchange
which must be taken into consideration (for
instance, the reaction with H2 absorbs energy while the reaction with CO releases energy).
The functioning of the 1-dimensional
FEM managing these thermal and chemicalphysical interactions can be described according to the following steps:
1. A first simulation of reduction is carried
out, by ignoring all thermal exchanges
but assigning an arbitrary initial temperature value to the burden material. The
simulation, through an iterative process
which is based on the exploitation of the
IRES model, calculates an inlet material
flow which is compatible with the target
reduction.
2. Given the inlet material flow calculated
in the previous step, the thermal balance
is calculated in order to obtain the temperature of gas and material in each layer
of the shaft. To this aim, both the thermal
balance within each layer and boundary
conditions must be taken into account.
In particular, two boundary conditions
have to be respected: the first one for the
upper border concerning inlet material
whose temperature is known; the other
one on the lower border concerning inlet
gas temperature, which is known as well.
199
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
Burden
material
Outlet
gases
……
……
Element #50
Element #2
Element #1
Reduced Reducing
material
gas
Fig. 4. The DR shaft FEM representation exploited by the SAILORS model.
Fig. 4. La représentation du modèle d’élément fini de l’axe exploitée par le modèle.
3. Given the thermal profiles calculated in
step 2, a new inlet material flow is calculated as in step 1 but, in this case,
new thermal conditions are exploited.
The computation terminates when the
calculated inlet material flow value is stabilized according to a set of convergence
criteria established by the user.
Figure 5 depicts the flow chart of the abovedescribed calculation.
Within the simulation IRES is used for
the inlet material flow calculation. Material
flow calculation starts from element No.1
and goes on element by element from the
bottom to the top of the shaft. In this situation and at each step of the computation,
the material temperature is known (from the
current thermal profile) as well as the gas
composition in the shaft element, the reduction degree on the bottom of the element
and the residence time of mineral in the element. IRES is used each time to calculate
the reduction degree on the top border of the
considered element, which represents the reduction degree of the mineral in a previous
moment as the mineral goes from the top
200
to the bottom of the shaft. To this purpose,
IRES calculates the reduction profile of the
material until the reduction of the mineral
at the exit of the element is reached under
the conditions of temperature and gas composition of the considered element, then it
looks backward to a time t which represents
the residence time of material in the element.
The obtained result is the reduction degree
of the material once it enters the element. The
reduction degree of the 50th element is used
in order to verify the mass balance for the
current situation: if there is convergence the
calculation can stop, otherwise parameters
are modified until convergence is reached.
The developed software
On the basis of the above-described 1D FEMbased description of the DR shaft, software
for the simulation was implemented. The realization of a stand-alone application was
judged necessary for two main reasons: the
first one is related to the complexity of
mathematics involved in the calculus and
to the heavy computational burden. Such
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
Fig. 5. Flow chart describing the computational structure of SAILORS.
Fig. 5. Organigramme décrivant la structure informatique du modèle SAILORS.
computational effort needs to be afforded by
dedicated software expressly designed and
compiled in order to minimize the computational time. The second reason is related to
the need for having user-friendly software
which makes the selection and modification
of the numerous parameters involved in the
simulation easy and provides a clear visualization of results.
For the implementation of the software
Microsoftc Visual Basic was chosen because
it gives at the same time the possibility of:
1. writing code to be compiled by the
standard VB compiler. The code compilation gives the advantage of creating
portable stand-alone software compatible with MS Windowsc -based systems.
Moreover, the compilation drastically increases the speed of the whole computation.
2. Designing in a very natural way a userfriendly interface similar to most software running under Microsoft operative
systems.
201
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
Fig. 6. The main window of the software interface.
Fig. 6. La fenêtre principale de l’interface du logiciel.
3. Easy interfacing with other programs
used for the loading and storage of data
used by the model.
The software was designed according to
a modular structure, in order to facilitate subsequent modifications or code additions: there are, for instance, procedures for
the mass balance, for the thermal balance
and some for the management of the FEM
engine.
The main interface window of the developed software is shown in Figure 6 and
includes an area for the input of plant and
material parameters in which the main properties of the shaft and the burden material
are chosen and an output area for the monitoring of main information concerning each
of the 50 elements into which the shaft is
divided and the main results of the DRI
simulation. Moreover, a picture showing the
progress of reduction and temperature of the
material and reducing gas inside the shaft is
included.
The complete list of the inputs to the
model inserted through the interface and
202
taken into account by the model is the following:
– material bulk density;
– guess value on burden material flow rate
(needed for the FEM algorithm);
– gas flow rate;
– inlet gas pressure;
– outlet gas pressure;
– burden material inlet temperature;
– reduction chamber height;
– reduction chamber diameter;
– target discharge metallization;
– average burden material size;
– inlet gas composition;
– burden material type.
In addition to these inputs, when a burden
material is chosen for the simulation, the
software exploits some information previously stored in a separate (independent and
editable) Excel file. This file contains information used by the IRES model embedded
in SAILORS and describes the kinetics of the
reduction of the specific material. The list of
burden materials can be extended by simply adding information to the Excel file and
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
in an analogous manner material properties
can be modified.
SAILORS allows the user to save and
subsequently recall an input set in order to
form a library of principal input situations.
The software, after the calculation is complete, returns the following information on
the main window:
– material inlet flow;
– material outlet flow;
– outlet gas composition;
– for each element of the shaft: gas temperature, burden temperature, reduction
rate, residence time (in a table).
When the calculation is completed, the picture area is updated and it is possible to show
both a graphic describing the progress of the
reduction rate in function of time or (by selecting the specific radio-button) the burden
and gas temperature in function of the height
of the shaft. It is also possible to export these
figures in the JPG format. More detailed data
are available by accessing a further window
(see Fig. 7) which provides the following information:
– total production rate of reduced material;
– metallization degree;
– total iron;
– total metallic iron;
– residence time;
– oxides;
– total oxygen transferred.
All the above-described output data can be
exported in a plain ASCII file which is easily readable by most common text editors.
The exported file also includes information
concerning the input parameters of the simulation.
The computation time is in line with the
expectations. A single run of the simulation depends on the whole set of previously
mentioned parameters. Normally, simulations on a 2.4 GHz processor computer with
2GB ram take from 5 to 20 seconds.
Numerical results
In order to validate the developed model,
some real data from the most common DRI
production plants were compared with the
corresponding simulations performed by
SAILORS. Such information is confidential
and comes from the work carried out within
the ULCOS project. In particular, tests refer
to a set of different configurations both for
Fig. 7. The window providing detailed information on
the result of the calculation.
Fig. 7. La fenêtre qui fournit des informations détaillées sur le
résultat du calcul.
Fig. 8. Comparison between calculated and actual hourly
production rate of DRI for the performed tests.
Fig. 8. Comparaison entre la cadence de fabrication horaire calculée et réelle du DRI pour les essais réalisés.
the dimension of the shaft and for other parameters such as inlet gas composition and
temperature, while the burden material used
is for all tests the same kind of commercial pellets. The variability in such parameters, whose ranges are shown in Table 2, is
203
M. Vannucci et al.: Revue de Métallurgie 107, 195–204 (2010)
Table 2. Variability ranges for the main parameters for the reduction tests exploited for the
model validation.
Tableau 2. Gammes de la variabilité pour des paramètres principaux pour les essais de réduction exploités pour la validation modèle.
Parameter
Shaft diameter
Shaft height
Gas temp. (◦ C)
Inlet H2
Inlet CO
Inlet CH4
Min
5m
9m
900◦ C
38%
15%
2%
Max
7m
11.6 m
1078◦ C
70%
36%
10%
important as it has allowed testing the simulator in a wide range of conditions.
The real plant hourly DRI production
rates are compared with the ones calculated
by the SAILOR simulation. The results of this
comparison, which are depicted in Figure 8,
show the very good agreement between real
and calculated DRI production and demonstrate the goodness of the developed model.
Conclusions
A model of a DR shaft has been developed
within the ULCOS project, in order to support the development of the new DR plant,
which is one of the most promising routes for
reducing the CO2 emissions involved in steel
production. The mono-dimensional FEM of
the shaft was exploited coupled with a reduction module in order to simulate the reduction of burden material in each element
of the FEM. Stand-alone software was also
implemented in order to make the model
easy to use and to speed up the computation.
The model was validated by exploiting
real data from the most common DRI production plants, which were compared with
the results of the model. The tests refer to
a set of different configurations which are
representative of a wide range of possible
cases. The good agreement between the results provided by the model and the experimental data encourage the use of the developed software in the design phase of new
concept DR plants.
Acknowledgements
The work described in the present paper was
developed within the project entitled “Ultra-Low
204
CO2 Steelmaking” (ULCOS), which has received
funding from the European Community within
the 6th Framework Program. The sole responsibility for the issues treated in the present paper
lies with the authors; the Commission is not responsible for any use that may be made of the
information contained therein.
The authors also wish to thank Dr. E Knop
for having provided the experimental data that
have been used to validate the model and Dr. E.
Burström for the fruitful discussions which contributed to the development of the described research work.
References
[1]
K. Meijer, M. Denys, J. Lasar, J.P. Birat, G.
Still, B. Overmaat, ULCOS: ultra-low CO2
steelmaking, Ironmaking & Steelmaking 36
(2009) 249-251
[2]
J.P. Birat, F. Hanrot, ULCOS: the European
Steel Industry’s Effort to Find Breakthrough
Technologies to cut CO2 Emissions
Significantly, EU/Asia Workshop on
Clean Production and Nano Technologies,
Seoul, Korea, Oct. 2006
[3]
K. Knop, M. Hallin, E. Burström, ULCORED
SP 12 Concept for minimized CO2 emission,
Rev. Métall. 10 (2009) 419-421
[4]
M. Small, Direct Reduction of Iron Ore, J.
Met. 3 (1981) 67-75
[5]
J. Feinman, D.R. Mac Rae, Direct Reduced
Iron, The AIST Foundation, Pittsburgh, PA,
1999
[6]
K. Mondal et al., Reduction of iron oxide in carbon monoxide atmosphere – reaction controlled kinetics, Fuel Processing
Technology 86 (2004) 33-47
[7]
K. Piotrowsky et al., Effect of gas composition on the kinetics of iron oxide reduction in a hydrogen production process, Int.
J. Hydrogen Energy, 2004
[8]
D.R. Parisi, M.A. Laborde, Modeling of
counter current moving bed gas-solid reactor used in direct reduction of iron ore, Chem.
Eng. J. 104 (2004) 35-43
[9]
F.N. Griscom, G.E. Metius, J.T. Kopfle,
Ironmaking technology for the new millennium, SEAISI Conference 2000, Perth,
Australia, May 2000
[10] N. Eklund, A. Dahlstedt, The choice of pellets in a mixed blast furnace burden and how
it affects process conditions, 14th conference
on Hungarian pig iron and steel making,
Sept. 2002