You are on page 1of 8

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Kinetics and mechanisms of direct reduction of


iron ore-biomass composite pellets with hydrogen
gas

Dabin Guo a,b, Mian Hu a, Chengxi Pu c, Bo Xiao a,*, Zhiquan Hu a,


Shiming Liu a, Xun Wang a, Xiaolei Zhu a,d
a
School of Environmental Science & Engineering, Huazhong University of Science and Technology, Wuhan 430074,
China
b
SafeCleen Technologies Co. Ltd, Wuhan 430074, China
c
Jiangxi Province Architectural Design & Research General Institute, Nanchang 330108, China
d
Hainan Research Academy of Environmental Sciences, Haikou 571126, China

article info abstract

Article history: Direct reduction of iron ore pellets (DRI) with and without biomass was studied using
Received 26 December 2014 hydrogen as the reducing agent. The influences of temperature and time on the reduction rate
Received in revised form of pellets were investigated. X-ray diffraction (XRD), scanning electron microscopy (SEM) and
13 February 2015 BET (Bet specific surface area) tests were adopted to interpret the mechanism of Fe2O3
Accepted 14 February 2015 reduction with H2, as well as explore the role of biomass in pellets during the reduction
Available online 12 March 2015 process. Results show that the biomass not only improve the reduction extent of the pellets,
but also increase the reduction velocity index (RVI) of the pellets by 1.12%/min. The BET test
Keywords: indicates that the existence of biomass can create porosity in pellets by dehydration and
Biomass combustion/pyrolysis, which increases the contact area between iron oxide and reducing gas
Ironmaking during reduction process, and decreases the apparent activation energy of pellets from
Direct reduction 122 kJ mol1 (without biomass) to 111 kJ mol1 (with biomass). For the DRI process, the crucial
Kinetics reduction step FeO / Fe is controlled by the intrinsic interfacial chemical reaction.
Copyright 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights
reserved.

especially concerning the carbon emissions, compared to the


Introduction traditional ironmaking technology [1e3]. To respond to the
growing energy shortage and greenhouse challenges, the
Direct reduction technology of iron ore is considered as a requirement of incorporating renewable energy to the existing
promising and clean production approach for ironmaking. In a metallurgical process has become more and more urgent.
typical direct reduction process, coal and natural gas are used Biomass is an ideal renewable energy with outstanding ad-
as alternative fuels to replace metallurgical coke. However, vantages, such as low content of harmful elements (S, P and
there is no fundamental improvement in the environmental others), natural abundant resources, as well as low level of
pollution control for the direct reduction technology, overall net carbon emissions, which makes it a promising

* Corresponding author. School of Environmental Science & Engineering, Huazhong University of Science and Technology, 1037 Luoyu
Road, Wuhan, Hubei 430074, China. Tel.: 86 13247178429. .
E-mail address: xiaobo_hust@163.com (B. Xiao).
http://dx.doi.org/10.1016/j.ijhydene.2015.02.065
0360-3199/Copyright 2015, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
4734 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0

energy candidate in the direct reduction ironmaking to alle- The reactions of CO and H2 with iron oxide are endothermic,
viate the environmental pollution. which is in favor of the reduction of the pellets due to the
Recently, a number of studies have been conducted to effect of temperature field. Both CO and H2 are ideal reducing
introduce biomass into metallurgical industries [4e10]. Stre- agents for the process. However, compared to H2, the reduc-
zov [4] investigated the fundamental mechanisms of iron ore tion kinetics of CO is relatively poorer. Moreover, the genera-
reduction with biomass wood waste. Reduction tests were tion of a large amount of greenhouse gas CO2 from CO results
conducted using mixtures with different weight ratios of in the oxidation of metallic iron in the porous surface to form
biomass and iron ore. The iron ore was successfully reduced nFeO$SiO2 slag phase, subsequently leading to the decrease of
to predominantly metallic iron phase when 30 wt.% of direct reduction iron (DRI) quality.
biomass was introduced into the mixture. Katyal et al. [5] and In order to identify the optimal addition amount of
Emmerich et al. [6] investigated the possibility of metal biomass in pellets, a roasting process was conducted with
smelting operations with charcoal produced from wood or biomass addition amount varying from 0.5% to 10% in an
other biomass materials. The results showed that the charcoal increment of 0.5%. It was found that the pellets with a 2%e3%
can substitute for coke successfully in the blast furnace. In weight fraction of biomass exhibited the properties that meet
these studies the reduction of metallic oxides was performed the industrial requirements. Therefore, all the samples used
either using biomass in their natural form or char from in the present study have a weight fraction of 2% biomass.
biomass pyrolysis. The reduction process was driven by the In order to explore the kinetics and mechanism of direct
carbonthermic Boudouard reaction: reduction process of ironmaking with bio-gas, the direct reduc-
tion of iron ore pellets with and without biomass to metallic
Fex Oy s yCOs xFes yCO2 g (1) iron was studied with H2 as the reducing agent over a wide
range of temperatures. The effect of temperature and time on
xCs xCO2 g 2xCOg (2) the reduction rate of pellets was investigated. XRD, SEM and BET
tests were used to analyze the reduction mechanism. The effect
Fex Oy s xCs xFes y  xCO2 g 2x  yCOg (3) of biomass in pellets was also discussed in the present study.

However, the intensity of FeeC reduction is weak. Poor


kinetic conditions and low reduction efficiency limit the Experiment and methods
development of direct reduction ironmaking using biomass.
To improve the reduction efficiency, Luo et al. [7] studied the Materials and facilities
possibility of introducing bio-gas produced from catalytic
gasification of biomass into the direct reduction ironmaking Iron ore powder was supplied by the Wuhan Iron and Steel
as reducing agent. Instead of coke, biomass was mixed Corporation (Wuhan City, Hubei Province) and bentonite from
together with iron ore powder, water, limestone and bentonite the Ezhou Iron and Steel Corporation (Ezhou City, Hubei
to produce composite pellets. During the reduction process, Province). The chemical compositions of the bentonite and
the dehydration of the combined water and pyrolysis of the iron ore powder are shown in Table 1. Pine sawdust, the
biomass make the pellets porous, leading to an increase in the biomass used in this study (see Table 1), was collected from
contact area of iron oxide and reducing gas, as well as an the furniture factory of the Huazhong University of Science
improvement in heat and mass transfer. The results showed and Technology (Wuhan City, Hubei Province). The iron ore
that at a reduction temperature of 1373 K and a reaction time powder, bentonite, and sawdust were crushed into fine par-
of 60 min, the TFe content in the product reached the ticles and sieved such that only particles less than 200 mm in
maximum of 94.7 wt.%. However, research about the reduc- diameter were accepted.
tion mechanism and kinetics laws in term of reducing gas and Iron ore powder, pine sawdust and bentonite (96%, 2% and
iron ore-biomass composite pellets have not been referred to 2% respectively) was rolled and pelletized with a 41000 mm
and systematic studies are needed. disc balling machine, and the green balls with particle size of
During the direct reduction process of ironmaking with 10e15 mm were taken as testing samples. The chemical
bio-gas, the effective reducing agents are mainly CO and H2. composition of the pellets is shown in Table 2.

Table 1 e Chemical compositions of iron ore and bentonite and the proximate and elemental analyses of the biomass.
Sample Chemical composition (%)
Fe Si Al Mn Ca Mg Ti S K
Iron ore 67.72 5.39 2.02 0.30 9.39 14.49 0.33 0.27 e
Bentonite 3.25 68.41 18.37 0.04 3.77 3.88 0.69 e 1.55
Biomass Ultimate analysis (wt.%)
C H O N S
48.21 6.54 44.39 0.77 0.09
Proximate analysis
Higher heating value (MJ/kg) Moisture content (wt.%) Volatile matter (wt.%) Fixed carbon (wt.%) Ash (wt.%)
16.83 7.78 73.06 16.33 2.83
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0 4735

Table 2 e Chemical composition of pellets used in present experiments.


Pellet samples Pellet components % Chemical composition, mass percent, %
Iron ore Bentonite Biomass TFe FeO SiO2 Al2O3 MnO CaO MgO TiO2 K2O
With biomass 96 2 2 57.56 0.28 7.78 2.99 0.22 1.66 4.78 0.19 0.14
Without biomass 98 2 0 57.68 0.26 7.63 2.93 0.22 1.63 4.69 0.19 0.14

The green balls were firstly dried in an oven at 378 K for Determination of the extent of reduction
24 h, then pre-sintered to 1173 K for 30 min and roasted at
1523 K for 25 min to afford oxidizing roasting pellets. At last, The reduction extent (f wt.%) is calculated by the following
the roasted pellets were reduced by hydrogen and nitrogen equation [11]:
(both the concentration of H2 and N2 use was 99.999%, pro- f 0:111W1 =0:430W2 m0  mi =m0 0:430W2   100% (4)
vided by Wuhan Xiangyun Industry CO., Ltd) in a specially
where m0 is the mass of the pellet before reduction, mi is the
designed experimental apparatus, which is presented in Fig. 1.
mass of the pellet in reduction process, W1 and W2 are the
mass of FeO and TFe in the pellet before and after the reduc-
tion respectively, 0.11 represents the oxygen demand con-
Experimental procedure and characterization
version coefficient when converting FeO to Fe2O3, 0.43

A weight loss detection technique was used to accurately


weight the oxygen loss during reduction experiments. The
output of the balance was fed to a strip chart recorder to re-
cord the weight changes as a function of time.
The reduction procedures could be described as follows: 1)
The quantitative roasted pellets were filled in the reduction
reactor; 2) The reduction reactor was firstly preheated to the
reaction temperature in the N2 atmosphere till the tempera-
ture became stable; 3) The N2 was then switched to reducing
gas and reacted for 30 min; 4) To avoid the re-oxidation of the
product, the reducing gas was changed back to N2 until the
reactor was cooled to room temperature. For each trial, 20
pellets were loaded at a reducing gas flow rate of 7 cm/s.

Fig. 1 e Schematic diagram of reduction experimental Fig. 2 e Reduction curves of oxidized pellets with
equipment: 1, Electronic balance; 2, reduction chamber; 3, temperature and time. (a). Oxidized pellets with 96 wt.%
composite pellet; 4, electric furnace; 5, Al2O3 particles; 6, N2 iron ore power, 2 wt.% biomass and 2 wt.% bentonite; (b).
reservoir; 7, H2 reservoir; 8, flow controller; 9, control Oxidized pellets with 98 wt.% iron ore power and 2 wt.%
cabinet; 10, computer. bentonite.
4736 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0

represents the oxygen demand conversion coefficient when which happens mainly in the later stage [12,13]. Secondly, the
converting Fe to Fe2O3. reduction process is commonly controlled by gas diffusion
and chemical reaction, and the contact area is the key factor.
In the initial reduction stage, the hydrogen can be easily
Results and discussion diffused to the reaction surface and the reaction area is larger
than that in the later stage, which can be described by the
Reduction behavior of pellets kinetics model e contracting unreduced nucleus model
[14e17].
The roasted pellets were isothermally reduced in the range of It can be seen from Fig. 2, reduction temperature has a
temperature from 1123 K to 1323 K with pure hydrogen and significant effect on the reduction ratedthe higher the tem-
the results are shown in Fig. 2. perature, the faster the reduction rate. Actually, the main
At the initial stage of reduction (0e5 min), the reduction reduction reactions (1e3, 5, 6) during DRI process are endo-
extent of pellets with biomass increased from 44.12% at 1123 K thermic, which are strengthened by increasing the
to 68.45% at 1323 K, while the value changed from 40.23% to temperature.
65.23% for the pellets without biomass. Afterward, the variation It can also be seen from Fig. 2, the addition of biomass has a
of reduction extent with time (10e20 min) became less pro- positive effect on the reaction extent. The time needed for the
nounced, which can be explained with two ways: Firstly, the reduction extent of 99% is shorter when adding biomass into
main reduction reactions in the initial stage were as follows: pellets under the same reduction temperature, reflecting an
accelerated reduction reaction rate.
3Fe2 O3 H2 2Fe3 O4 H2 O (5) In order to discuss the effect of porosity on the reduction
behaviors of the pellets, the pore volume and BET tests were
Fe3 O4 H2 3FeO H2 O (6) conducted. The results showed that the surface area of the
pellets with and without biomass were 6.03 m2/g and
At T > 843 K, The transformation of Fe2O3 to FeO is much
easier to take place compared to the reduction of FeO to Fe,

Fig. 4 e XRD pattern of pellets samples before and after


Fig. 3 e The micropore volume distributions in pellets. a. reduction at 1173 K under different reduction time. a.
Oxidized pellets with biomass; b. Oxidized pellets without Pellets with addition of biomass; b. Pellets without
biomass. addition of biomass.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0 4737

Fig. 5 e SEM images of pellets samples with different components before and after reduction. a. Pellets with biomass before
reduction; b. Pellets with biomass reduced for 5 min; c. Pellets with biomass reduced for 15 min; d. Pellets without biomass
before reduction; e. Pellets without biomass reduced for 5 min; f. Pellets without biomass reduced for 15 min.

16.26 m2/g, respectively. Fig. 3 displayed the nitrogen generally higher than that of pure iron ore. This reflects that
adsorption curve measured by automated gas adsorption adding biomass in pellets can improve the reduction extent of
system, using BJH (BarretteJoynereHalenda) method to illus- the pellets.
trate the pore volume distribution. The ordinate (dV/dD) is the The scanning electron micrographs of the reduced pellets
pore volume distribution function which represents the pore were examined to detect the surface changes of pellets with
volume during this range. biomass at different reduction time, 0 min, 5 min and 15 min
It can be seen from Fig. 3a that the pore distribution was (as shown in Fig. 5aec), and Fig. 5def shows the surface
non-uniform. In the pellets containing biomass, the pore changes of pellets without biomass. The internal average pore
diameter less than 95 nm mainly concentrated in the range size and porosity of pellets with biomass are larger than those
3e5 nm and 13e22 nm with pore volume pore (dV/dD) of about of pellets without biomass. Besides, the size distribution is
0.007 cm3/nm.g and 0.009 cm3/nm.g, respectively. Conversely more uniform for pellets with biomass due to the higher
the pore distribution showed in Fig. 3b was relatively uniform. conductivity for reduction reaction.
The pore diameter less than 95 nm mainly concentrated in the From Fig. 5b and e, it can be found that the small iron
range 2.5e92 nm, mostly in the range of 3e10 nm and the pore particles and large magnetite particles in pellets periphery are
volume (dV/dD) were between 0.0008 cm3/nm g and reduced to the wustite. The core iron ore particles are not
0.0009 cm3/nm g. reduced. But the reduction degree of pellets in Fig. 5b is higher
By comparison, the main difference is that when the pore than that of the reduced the pellets in Fig. 5e. The reasons can
size was 4 nm, the pore volume of pellets containing biomass be interpreted as follows: Firstly, after adding biomass pellets,
was 8 times higher than that of pellets without biomass, and the biomass that contacts with the iron ore was burned off in
when the pore size was 16 nm, this value was 15 times. In the roasting process. So the pellets with biomass inside would
other stages, pore volume of the pellets with biomass were form more pores. Accordingly the specific surface area and the
slightly higher than the pellets without biomass. reaction interface increased. However, for the pellets with no
For the gasesolid reduction reactions, high porosity facili- biomass, the pore between the iron ore didn't increase after
tates the gas molecules diffusion, and further promotes the roast. Secondly, after adding biomass, the porosity of the
reduction reactions. Thus the reduction performances of the pellets was increased and reducing gas is easier to diffuse and
pellets with the addition of biomass were better than the contact with the iron ore. Meanwhile, the gas generated is also
pellets without biomass. easier to flowing out. Therefore, under the condition of 1173 K
XRD analysis of samples before and after reduction at of temperature and 5 min of reducing time, adding biomass
1173 K under different reduction time was carried out and the into pellets reduced the resistance of interfacial chemical re-
results are shown in Fig. 4. It was observed that both kinds of action, resulting in the enhancement of reduction degree.
pellets can be reduced to metallic iron at 1173 K after some- When the reduction time lasted up to 15 min, both kinds of
where within 10e20 min. Under the same reduction condition, the pellets are almost reduced to metallic iron (Fig. 5c and f).
the peak value of products from pellets containing biomass is Compared with the pellets reduced for 5 min, metallic iron
4738 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0

grains have obvious aggregation and growth. However, the where kG is gas phase boundary mass transfer coefficient (cm/
final characteristics of the two reduced pellets are basically s); r0 is characteristic initial radius of pellets (cm); De is
the same. This shows that the addition of biomass has no effective diffusion coefficient (cm2/s) and given by:
influence on the quality of DRI products, but it greatly im- .  
proves the pellets reduction rate and saves energy De r0 r20 6tD C0AH2  CAH2 ; (11)
consumption.
k is reaction rate constant (cm/s) and can be estimated by:
Analysis of the reduction velocity index .  
k Kr0 r0 tC 1 K C0AH2  CAH2 ; (12)

The reduction velocity index (RVI) [11,17] was proposed to K is reaction equilibrium constant; r0 is initial oxygen
evaluate reduction behavior of the pellet. A higher RVI value concentration in the pellet (mol/cm3); C0AH2 and C*AH2 are the
indicates a higher reducibility of the pellets. The RVI equation concentration of H2 at granule surface and in equilibrium
is presented as below: (mol/cm3), respectively; t is reduction time (min).
 According to un-reacted core model, the reduction time t
RVI dRt =dt 40 33:6=t60  t30 (7)
and reduction extent f functions were selected to do curve
where t30 and t60 are the consuming time of reduction extent fitting work. If the reduction process was controlled by inter-
arriving to 30% and 60%, respectively. Under the same facial chemical reaction, the kinetic equation obtained is as
reduction condition, the reducibility index RVI of the pellets follows:
with biomass was 1.12%/min higher than that of pellets " #
MFeO $k$C0AH2
without biomass. 1  1  f
1=3
t; (13)
rFeO $r0
Reduction kinetics analysis

For the iron direct reduction, a series of reactions occurred


1.0
during the process of pellets reduction at T > 843 K:
Fe2O3 / Fe3O4 / FeO / Fe. In the reduction series, the trans-
(a) [1+2(1-f)-3(1-f)2/3]/t
[1-(1-f)1/3] or [1+2(1-f)-3(1-f)2/3]

formations of Fe2O3 / Fe3O4 and Fe3O4 / FeO are more rapid Experiment(1123 K): R2=0.9838
0.8
than that of FeO / Fe, thus the former two reactions can be diffusion control
ignored in the kinetics analysis [19e21]. The step of FeO / Fe chemical control
0.6 Fitting line:
under hydrogen atmosphere can be expressed as follows:
diffusion control
chemical control [1-(1-f)1/3]/t
FeO H2 Fe H2 Og DrGqm 23430  16:16T J=mol (8)
0.4 R2=0.9954

lnK DGq RT 1:943  2818=T (9)
0.2
The reaction (FeO / Fe) equilibrium constant K under
different reduction temperature can be calculated by (9) and
the results are shown in Table 3. 0.0
To study the kinetics of the pellet reduction process, a weight 0 5 10 15 20 25
change detection method was adopted under isothermal con- Time, min
ditions and the experimental apparatus is shown in Fig. 1. In the
reaction process, the gas flow rate was set at 7 cm/s. The effect of (b)
0.8 [1+2(1-f)-3(1-f)2/3]/t
external diffusion can be negligible as the gas flow rate is above Experiment(1123 K):
[1-(1-f)1/3] or [1+2(1-f)-3(1-f)2/3]

R2=0.9781
5 cm/s. Therefore, it is presumed that the reaction rate might be diffusion control
controlled by three mechanisms: 1) the intrinsic interfacial chemical control
0.6
chemical reaction, 2) the diffusion including internal diffusion Fitting line:
diffusion control [1-(1-f)1/3]/t
of reactant and product gas species through solid product layer, R2=0.9995
chemical control
and 3) the mix of 1) and 2). The pellet reduction process is a 0.4
typical gas solid reaction, which can be described by the reaction
rate equation (10) presented by Szekely [18].
h i . h i 0.2
2=3 1=3
f=3kG 11f 21f $r0 6De 11f $K=k1K

C0AH2 C*AH2 t=r0 r0 0.0


(10)
0 5 10 15 20 25
Time, min
Table 3 e Reaction (FeO / Fe) equilibrium constant K Fig. 6 e Comparisons of reduction curve under H2
under different reduction temperature.
atmosphere at 1123 K between chemical-controlled and
T, K 1123 1173 1223 1273 1323 diffusion-controlled models. a. Oxidized pellets with
K, e 0.568 0.632 0.697 0.763 0.830
biomass; b. Oxidized pellets without biomass.
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0 4739

MFeO $k$C0AH2
Suppose x1 ;
rFeO $r0

1=3
1  1  f x1 t; (14)
3
where r(FeO) represents the density of FeO (kg/m ); M(FeO) is the
FeO molecular weight.
The relationships between 1(1f)1/3 and reaction time
t were calculated, fitted and typical results were shown in
Fig. 6a and b. The correlation coefficients of R2 for the reduc-
tion of the pellets with and without addition of biomass were
0.9954 and 0.9995, respectively.
When the reduction process was controlled by diffusion,
the kinetic equation obtained is as follows:

2=3 2MFeO $De $C0H2


1  2f=3  1  f t (15)
rFeO $r20

where C0H2 is the gas film concentration of outer layer of the


2M
pellets (mol/cm3), suppose x2 FeO
$De $C0H
r $r2
2
,
FeO 0

2=3
1  2f=3  1  f x2 t (16)

The relationships between 12f/3(1f)2/3 and reaction


time t were calculated, fitted and typical results were shown
in Fig. 6a and b. The correlation coefficients of R2 for the
reduction of the pellets with and without addition of biomass
were 0.9838 and 0.9781, respectively.
According to the correlation comparison of R2, this
conclusion can be drawn that the rate controlling step should
be the interfacial chemical reaction, whether introducing
biomass into the pellets or not. Experiments verified that the
model was fit to reducing process. Fig. 7 e Reduction curves of mixed controlled model for H2
Supposing the kinetics mechanism is controlled by the at different temperature. a. Oxidized pellets with biomass;
mixed model, the external diffusion can also be ignored. b. Oxidized pellets without biomass.
Considering both internal diffusion and chemical reaction
resistance, then the equation (14) was changed into:
ln k DE=RT lnA (20)
n h i2 h i3 o h i
1=3 1=3 1=3
t tD 3 1  1  f  2 1  1  f tC  1  1  f According to the slope in Fig. 8, the apparent activation
(17) energy of the pellets is calculated. The value of apparent acti-
vation energy is 111 kJ mol1 for the reaction of the pellets with
where tD and tc stands for reaction time under internal diffusion biomass, while the value is 122 kJ mol1 for the pellets without
control and interfacial chemical reaction control, respectively, biomass. Therefore, the presence of biomass in pellets helps to
which can be estimated by the following equations: decrease the apparent activation energy of the pellets.
  Meanwhile, the corresponding reaction rate k can be
tD r0 r20 =6De C0AH2  C*AH2 (18) described as follows:
For the reduction of the pellets with biomass:
.    
tC Kr0 r0 k1 K C0AH2  CAH2 (19) 111  103
k 237043exp T > 843 K (21)
8:3143T
The relationships between t=1  1  f1=3 and
1=3 2=3
For the reduction of the pellets without biomass:
1 1  f  21  f for the pellets with and without
biomass were calculated and shown in Fig. 7. As can be seen in
Fig. 7, the slopes of the five lines are small, which further proves Table 4 e Reaction rate constant k and effective diffusion
that the reduction reaction process under H2 atmosphere was coefficient De of pellets with and without biomass.
mainly controlled by the interfacial chemical reaction. T, K With biomass Without biomass
At the same time, the values of De and k according to (11) and 2
De (cm /s) 2
k (cm /s) De (cm2/s) k (cm2/s)
(12) can be calculated and the results are shown in Table 4.
1123 7.88 1.72 3.35 1.44
Since the rating control step is in interfacial chemical re- 1173 9.65 2.57 3.68 2.16
action control, the Arrhenius equation (20) is applied and the 1223 12.83 3.95 3.92 3.76
Arrhenius plot of chemical reaction rate under H2 atmosphere 1273 13.46 6.08 5.67 5.91
is shown in Fig. 8. 1323 25.88 10.71 15.45 10.68
4740 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 0 ( 2 0 1 5 ) 4 7 3 3 e4 7 4 0

references

[1] Shi JY, Donskoi E, McElwain DLS, Wibberley LJ. Modelling the
reduction of an iron ore-coal composite pellet with
conduction and convection in an axisymmetric temperature
field. Math Comput Model 2005;42:45e60.
[2] Srinivasan NS. Reduction of iron oxides by carbon in a
circulating fluidized bed reactor. Powder Technol
2002;124:28e39.
[3] Liu GS, Strezov V, Lucas JA, Wibberley LJ. Thermal
investigations of direct iron ore reduction with coal.
Thermochim Acta 2004;410:133e40.
[4] Strezov V. Iron ore reduction using sawdust: experimental
analysis and kinetic modelling. Renew Energy
2006;31:1892e905.
[5] Katyal S, Thambimuthu K, Valix M. Carbonisation of bagasse
Fig. 8 e Arrhenius plot of chemical reaction rate under H2 in a fixed bed reactor: influence of process variables on char
atmosphere. yield and characteristics. Renew Energy 2003;28:713e25.
[6] Emmerich FG, Luengo CA. Babassu charcoal: a sulfurless
renewable thermo-reducing feedstock for steelmaking.
  Biomass Bioenerg 1996;10:41e4.
122  103
k 730146exp T > 843 K (22) [7] Luo SY, Yi CJ, Zhou YM. Direct reduction of biomass-Fe2O3
8:3143T
briquettes using biomass-generated syngas. Renew Energy
2011;36:3332e6.
Conclusions [8] Sastri MVC, Viswanath RP, Viswanathan B. Studies on the
reduction of iron oxide with hydrogen. Int J Hydrogen Energy
1982;7:951e5.
In this paper, the reduction process of iron oxide in composite
[9] Carmo de Lima L, Batista Furlan Duarte J, Nejat Veziroglu T. A
pellets under the H2 atmosphere and related reaction kinetics proposal of an alternative route for the reduction of iron ore
were studied, and the following conclusions can be drawn: in the eastern Amazonia. Int J Hydrogen Energy
The reduction temperature had a significant effect on the 2004;29:659e61.
reduction extent. At the same reduction temperature, the [10] Rashid RZA, Mohd H, Salleh Hanafi MA, Yunus NA.
reduction extent of pellets with biomass was relatively higher Reduction of low grade iron ore pellet using palm kernel
shell. Renew Energy 2014;63:617e23.
than those without biomass.
[11] ISO 4695:1995. Iron ores e determination of reducibility. 2nd
Biomass can improve the porosity of the pellet by dehy-
ed. Geneva: International Organization for Standardization;
dration and combustion/pyrolysis, which increases the con- 1995.
tact area between iron oxide and reducing gas during [12] Wang LL. The dynamics research of directly reduced iron ore
reduction process, and leads to a higher interfacial chemical by coke oven gas. Baotou: Inner Mongolia University of
reaction rate. Results show the RVI of pellets with biomass Science and Technology; 2011.
was 1.12%/min higher than those without biomass under the [13] Liu JH, Zhang JY, Zhou TP. Assessment of apparent activation
energies for reduction reactions of iron oxides by hydrogen. J
same reduction conditions.
Iron Steel Res Int 2000;12:6e8.
The reduction process obeys the gasesolid reaction model. [14] MoonI J, Rhee CH, Min DJ. Reduction of hematite compacts by
The rate controlling reaction is FeO / Fe. A kinetics analysis H2-CO gas mixtures. Steel Res Int 1998;69:302e6.
shows that the reduction process follows the chemical inter- [15] Krzysztof P. Effect of gas composition on the kinetics of iron
facial reaction mechanism. The introduction of biomass in- oxide reduction in a hydrogen production process. Int J
creases the porosity of the pellets and decreases the apparent Hydrogen Energy 2005;30:1543e54.
[16] Bonalde A, Henriquez A, Manrique M. Kinetic analysis of the
activation energy from 122 kJ mol1 to 111 kJ mol1.
iron oxide reduction using hydrogenecarbon monoxide
mixture as reducing agent. ISIJ Int 2005;45:1255e60.
[17] Nasr MI, Omar AA, Hessien MM, El-Geassy AA. Carbon
monoxide reduction and accompanying swelling of iron
Acknowledgments oxide compacts. ISIJ Int 1996;36:164e71.
[18] Szekely J, Evens JW, Sohn HY. Gas-solid reaction. New York:
This research was supported by the State 863 projects (No. Academic Press; 1967.
2012AA101809), National Natural Science Foundation of China [19] Hirschfelder JO, Curtiss CF, Bird RB. Molecular theory of
(No. 21276100), Natural Science Foundation of Hainan Prov- gases and liquids. New York: Molecular Theory of Gases and
ince (No. 414194), and The Fundamental Research Funds for Liquids; 1964.
[20] Wang YM, Yuan ZF, Matsuura H. Reduction extraction
the Central Universities (HUST: 2015650011. CX14-011. CX14-
kinetics of titanium and iron from an limonite by H2eAr gas
012. CXY13q030). The authors are grateful to the Analytical
mixtures. ISIJ Int 2009;49:164e70.
and Testing Center of HUST and Jingbo Wang who works in [21] Pang JM, Guo PM, Zhao P. Kinetics of reduction of hematite
Hainan Research Academy of Environmental Sciences and by H2 using nonisothermal thermo gravimetric method. Iron
helps us to improve this paper. Steel 2009;44:11e5.

You might also like