ISIJ International,

ISIJ International, Vol. 57 (2017), No. 3 Vol. 57 (2017), No. 3, pp. 404–412

Effect of Hydrogen Addition on Reduction Kinetics of Iron Oxides

in Gas-injection BF

Yana QIE,1) Qing LYU,1) Jianpeng LI,1) Chenchen LAN2) and Xiaojie LIU1)

1) College of Metallurgy & Energy, North China University of Science and Technology, Tangshan, 063009 China.
2) School of Metallurgy, Northeastern University, Shenyang, 11081 China.
(Received on June 12, 2016; accepted on November 7, 2016; J-STAGE Advance published date:
February 3, 2017)

The gas-injection BF is a new iron-making technology with injecting gas instead of the traditional pulver-
ized coal injection and recycling the BF top gas through the gasifier. Compare with traditional BF, there is
a larger amount of reducing gas such as CO and H2 in gas-injection BF. Effect of H2 addition on reduction
kinetics of iron oxides in gas-injection BF was investigated by Thermogravimetric Analysis. The result
shows that the reduction rate of iron oxides rise with the increase in both temperature and H2 content. By
contrast with the H2 content increasing from 10% to 15% at 700°C and 900°C, the improvement of reduc-
tion rate is more obvious when addition from 15% to 20% at the same temperature. For comparison, it
shows an opposite result in the condition of 1 000°C. In addition, the efficiency of H2 on reduction rate
can be neglected as the content is less than 5%. The reaction mechanism was obtained according to the
unreacted core model. In the condition of 30% CO + 10% H2, both the diffusion rate and interfacial reaction
rate reveal an increasing trend with the increase of temperature, of which the degree of improvement in
diffusion rate is more significant, which leads to the interface reaction gradually being the controlling step.
The same case occurs with the increase of H2 at 900°C. Under the condition of experiment, the activation
energy decreases with H2 addition, which illustrates that the reduction of iron oxides become more easily
to perform with the rich hydrogen in gas-injection BF.

KEY WORDS: reduction kinetics; hydrogen content; unreaction core model; reduction rate.

BF top gas, which implies the additional cost as $56/tCO2.16)

1. Introduction
The blast furnace (BF) is the main process of the iron
and steel industry, which emits over 70% of the CO2 and
consumes the most energy.1) The traditional blast furnace
relies on the coke seriously and the coke rate contributes to
a great extent of the whole production cost. Most of the new
trends in blast furnace technology focused on developing a
new coke alternative such as natural gas injection, pulver-
ized coal, waste plastic, biomass, coke oven gas and other
hydrocarbons injection.2–9) The utilization of these sources,
however, depends strongly on the developed technologies
and wealthily district resources.10) In addition, top gas recy-
cling blast furnace (TGR-BF) have received an increasing
interest in recent decades in order to reach the strict targets
of energy conservation as well as low emission of CO2.10,11)
Top gas recycling and tuyere injection of oxygen are the
primary technologies studied in the ULCOS project.12) Cur-
rently, a number of studies on TGR-BF have been carried
out, such as the analysis of flow and combustion of reducing
gas in tuyere, CO2 emission and energy consumption.11,13–15)
Nonetheless, the clear limitation for the development of
above technology is that how to capture and disposal CO2 of
* Corresponding author: E-mail: 461434073@qq.com
DOI: http://dx.doi.org/10.2355/isijinternational.ISIJINT-2016-356
Currently, pulverized coal injection (PCI) is still the main
means for reducing coke rate.17) The highest amount of pul-
verized coal injection reported is approximately 250 kg/thm
while the amount of coke is similar as previous. However,
the impact of chemical properties of the coal should be not
left out, which may cause problems for complete combus-
tion within raceway, gas permeability in the shaft, dirtying
of deadman zone and, as a consequence, irregular furnace
operation and decreasing productivity.11,18)
The gas-injection BF is a new iron-making technology
with injecting gas into tuyeres and recycles the BF top gas
through the gasifier, which can provide a new gas source
and effectively translates CO2 in BF top gas into CO as a
lower cost. Compared with traditional BF with PCI, the gas
injection technology could simplify the iron making process
and recycle the BF top gas, which can contribute to lower
emissions and high productivity. The technological process
as follows: BF top gas as the gasifying agent is injected
into gasifier→coal gasification occurs in the gasifier→H2-
rich gas is produced→H2-rich gas is heated in gas heating
device→the high temperature gas and hot air are injected
into the furnace through the tuyere. The technological pro-
cess is shown in Fig. 1.
In addition, the inferior coal possessing the properties
high-volatile and high-ashes can be utilized in the gas-

© 2017 ISIJ 404

 ISIJ International, Vol. 57 (2017), No. 3

making process which not only make the blast furnace
smelting get rid of high-quality coke,19,20) but also make the
hydrogen content in gas injected increase to improve the BF
production efficiency.21)
In the gas-injection BF, the reduction behavior of iron
oxides will be dominated by the high concentration of
reducing gas such as CO and H2. This paper focuses on the
effect of hydrogen addition on the reduction kinetics of iron
oxides aiming to clarify the reaction mechanism and deter-
mine the optimum hydrogen content in gas-injection BF.
2. Experimental Method
The pellets produced by industry process were taken as
testing samples. The chemical compositions are shown in
Table 1.
The schematic layout of experimental apparatus which
can measure the iron mass continuously is shown in Fig.
2. The rated power and the highest heating temperature
are 8 kW and 1 200°C, respectively. The output of the bal-
ance was fed to a strip chart recorder to record the weight
changes as a function of time.
In the reduction experiment, 500 g pellets with the size of
10–12.5 mm were put into the reduction reactor. Then the
reduction reactor was heated up at the rate of 10°C/min in
electric furnace. The pellets were heated up to the required
temperature and subsequently kept for 30 min to minimize
the temperature fluctuation. During the whole process of
heating and holding, nitrogen as the shielding gas was filled
into the reduction tube at a rate of 5 L /min. After that,
reducing gas mixtures were used to reduce the pellets with
the flow rate of 15 L/min as shown in Table 2. The reducing
gases were replaced by nitrogen after the reaction and the
reduction tube was taken out from the furnace to cool down
until room temperature.
3. Results and Discussion
3.1. Effect of H2 Content on Reduction Behavior of
Iron Oxides
The reduction of ferric oxides in the blast furnace expe-
riences three steps: Fe2O3→Fe3O4→FeO→Fe.22) The FeO
reduction is the most difficult step and the requisite atmo-
sphere condition is the strictest.23) FeO reduction with CO
is an exothermic reaction while FeO reduction with H2 is
an endothermic reaction, and the reduction behavior of iron
oxides in mixed atmospheres of CO–H2 is complex. On the
basis of the results obtained from thermodynamic calcula-
tion, the highest H2 content is set to 20% to ensure the exo-
thermicity during the reduction of FeO in gas-injection BF.

3.1.1. Effect of H2 Addition on Weight Loss of Pellets

The weight loss of sample under different H2 contents
at 700°C is shown in Fig. 3. It can be clearly seen that the
weight of sample decreases continuously along with the
reduction time, and at the same reaction time, the weight
loss of sample exhibits a positive correlation with H2 con-
tent. In addition, there is a large weight loss of sample at the
Fig. 1. Process of gas-injection BF. initial stages of reduction, followed by a slowing down in
the weight loss until the end of reduction. XRD analysis of
samples before and after reduction at 700°C under different
Table 1. Chemical composition of the pellets (mass%).
atmospheres was carried out and the results are shown in
TFe FeO SiO2 CaO MgO Al2O3 TiO2 Mn P Fig. 4. It is observed that the pellet is mainly composed of
60.15 2.59 6.32 1.65 1.11 1.23 0.24 0.08 0.03
hematite, magnesium ferrites oxide (MgFe2O4) and iron sili-
con oxide phase (Fe2.45Si0.55O4) before reduction. In contrast,
metallic iron is the predominant phase after reduction, fur-
thermore, a small amount of Fe0.9Si0.1, Al0.7Fe3Si0.3, MgFe2O4
and Fe2.95Si0.05O4 are also identified. In addition, the peak
of reduced products becomes remarkable as H2 content
enhances, while that of both Fe3O4 and FeO greatly weaken

Table 2. Chemical composition of the reducing gas (volume%).

CO H2 N2
1# 30 0 70
2# 30 5 65
3# 30 10 60
4# 30 15 55
5# 30 20 50
Fig. 2. Schematic diagram of reduction experimental.

405 © 2017 ISIJ

 ISIJ International, Vol. 57 (2017), No. 3
even disappear, which reflects that H2 addition in atmosphere
can improve the reduction efficiency of the pellets.
3.1.2. Effect of H2 Addition on Reduction Degree and Rate
The relationship between the reduction degree of pellets
and the reduction time under different H2 contents at 700°C,
900°C and 1 000°C are obtained as Figs. 5(a), 5(c) and 5(e)
respectively. And the functional relation of the reduction
degree and reduction time is found to fit the logarithmic
function by trial and error method, which also shows that the
reduction of pellets is in accordance with the first order revers-
ible reaction. The reduction rate was obtained by solving the
first order differential equations of reduction degree, and the
results are shown in Figs. 5(b), 5(d) and 5(f) respectively.
As can be seen from Fig. 5, the reduction rate is faster
and the reduction degree increase rapidly at the initial
stages of reduction, and then became less pronounced in
Fig. 3. Effect of H 2 addition on weight loss of pellets.
Fig. 5. Reduction degree and rate under different atmospheres.
© 2017 ISIJ
the later stage due to the resistance of thick product layer.
The addition of H2 has a positive effect on the reaction
extent. From Figs. 5(a) and 5(b), the reduction degree can
reach to 87.32% after 2.5 h with 30% CO + 20% H2 while
that is only 60.41% with 30% CO + 5% H2 at 700°C, and
the reduction rate of former should be about 2 times than
the latter. The results prove that the hydrogen is a kind of
much more effective reducing gas than CO even though the
temperature is lower than 847°C, which originates from the
molecular size of H2 (collision diameter 2.915 Å) less than
that of CO (collision diameter 3.590 Å) as well as the dif-
fusion coefficient of H2 in solid more than 3 times than CO
diffusion coefficient.
As shown in Figs. 5(b), 5(d) and 5(f), the reduction rate
of iron oxides increase with the increase in both temperature
Fig. 4. XRD patterns of pellets after reduction at 700°C at differ-
ent H 2 content.
406
 ISIJ International, Vol. 57 (2017), No. 3
and H2 content, while in different extents of H2 content, the
increase rate of reduction is different when H2 content goes
up in the same range. When the H2 content less than 5%,
it has a weak effect on the reduction rate. The reduction
rate increase most obviously with the increase of H2 from
15% to 20% at 700°C and 900°C. With regard to 1 000°C,
there is a greater impact on the increase of reduction rate
with the increase of H2 content from 10% to 15% than from
15% to 20% and the reduction rate increase slowly above
15%, which could illustrate that content of H2 is no longer
the primary factor for regulating the increase of reduction
rate when it’s above 15%, and thus the H2 content in gas-
injection BF should be lower than 15%.
3.1.3. Effect of H2 Addition on Microstructure of Pellets
The micro distribution of Fe element with the H2 addition
is shown in Fig. 6. It’s not hard to see Fe elements clus-
ters into classes of spherical or bulk with 30%CO + 0%H2,
and continue to aggregation and growth. According to the
classical nucleation theory,23) only more than the critical
radius (r*) can the nuclei exist stably. Unstable nucleus
would spontaneously move to stable nucleus resulting in
grain aggregation and growth to reduce the surface energy.
Comparing Figs. 6(a) and 6(b), H2 addition enhances the
rate of the nucleation and grain growth. Furthermore, a large
of ferric crystal nucleus could be homogeneous distribution
along with lots of the iron ion saturated appearing among
the original nucleus and growing throughout the old phase
with H2 addition, which results in a dense metallic layer
forming. Comparatively speaking, the reduction process
with 30%CO + 0%H2 is accompanied with a metallic phase
migration and for the reason that the critical radius is larger
and the aggregation and growth of grain will rely on the
stable crystal nucleus.
3.2. Effect of Temperature on Reduction Behavior of
Iron Oxides
Figure 7 groups the isotherms of pellets reduction at dif-
ferent temperature. As can be seen from Fig. 7, the higher
temperature in the range 700–1 000°C corresponds to the
more efficient reaction, which is due to the reduction of
FeO by H2 being the endothermic reaction and the reduction
ability of H2 strengthening with the increase of temperature.
From Fig. 7(b), when the H2 content is 5%, it makes little
contributions on the reduction rate with the increase of
temperature, which could conclude that the H2 content is
Fig. 6. Effect of H2 addition on micro distribution of Fe element at
1 000°C.
407
the reduction restriction factor for the reaction instead of
temperature when H2 content is less than 5%. So the H2
content in the gas-injection BF should be more than 5%.
When the H2 content reaches to 20%, the degree and rate of
reduction at 900°C are close to that at 1 000°C. Obviously
it could conclude that H2 is no longer the limiting factor
for reduction reaction once its content above 20%, and the
H2 content in gas-injection BF should be lower than 20%.
The XRD patterns of the samples at the atmosphere
of CO = 30%, H2 = 10% under different temperatures are
shown in Fig. 8. The main phase of pellets is metallic
iron, Fe0.9Si0.1, Al0.7Fe3Si0.3, MgFe2O4 and it also contains
a small amount of Fe3O4FeO which not been reduced and
Fe2.95Si0.05O4 at 700°C, With the increase of temperature,
however, Fe3O4 and FeO greatly are reduced, and then the
diffraction peaks of MgFe2O4 and Mg2SiO4 remarkably
weaken, which presents that iron oxides are reduced firstly
and followed by the reduction of MgFe2O4 and Mg2SiO4.
3.3. Kinetic Model for Reduction of Iron Oxides
The kinetic modeling for the reduction of iron oxides is
very important for understanding the kinetics of the reduc-
tion process which will help to increase productivity and to
improve the gas utilization in gas-injection BF. While the
kinetic models for the reduction of iron oxides with H2 or
CO have been largely investigated by many researchers,24–26)
the kinetics of the reduction with H2–CO mixtures is rather
limited. What is more the reduction kinetics of iron oxides
with H2–CO mixtures was generally simulated and predicted
by mathematical modeling. K. Piotrowski27,28)investigated
the kinetics of hematite (Fe2O3) to wustite (FeO) based on
Johnson-Mehl-Avrami-Erofe’ev equation. They found that
this initial stage of the reaction process could be interpreted as
a both phase-boundary-controlled reaction. A. Bonalde29) pro-
posed grain model that considers the particle size and poros-
ity to describe the reduction of hematite pellets with Midrex
gas. And the predictions of the model were compared with
the experimental results. It is very important to develop an
optimum model in the investigation of a specific technology.
In this paper, in order to explore the mechanism of iron ore
reduction and to identify the optimal addition amount of H2 in
gas-injection BF, the kinetic model is established based on the
theory of unreacted shrinking core model which is vastly used
for kinetics of iron ore reduction. In addition, the water gas
shift reaction is also taken into account in this kinetics model.
A series of reactions occurring in succession during
the process of ferric oxides reduction in the blast fur-
nace above 570°C: Fe2O3→Fe3O4→FeO→Fe.23,30) In the
reduction series, the transformations of Fe2O3→Fe3O4 and
Fe3O4→FeO are much easier than that of FeO→Fe, thus the
two former reactions can be ignored in the kinetics analy-
sis.31,32) In addition, the water gas shift reaction will also
take place besides for the reduction of iron oxides, which
will restrict the gas utilization in gas-injection BF. The FeO
reduction in the gases mixture and the water gas reaction
can be expressed as follows:
FeO + CO = Fe + CO2 ln KθCO = 1.94 − 2 818.14 / T ..... (1)
FeO + H 2 = Fe + H 2 O ln KθH2 = −2.92 + 2 742.36 / T ..... (2)
© 2017 ISIJ
 ISIJ International, Vol. 57 (2017), No. 3

CO + H 2 O = CO2 + H 2 ln K w = −4.03 + 4 492.75 / T .... (3)

ρo r0  R r0  
2
t=  + 1 − 3(1 − R ) 3 + 2(1 − R )

K′ in the gases mixture of Eqs. (1) and (2) can be pre- 0 ∗
(c − c )  3kg 6 Deff  
sented as follows: ... (7)
1 
Kθ  
K′ =ϕ θ
H2 K H2 +ϕ θ
CO K CO ........................ (4) + × 1 − (1 − R ) 3 
θ  
krea+ (1 + K )   
Where ϕH2 and φCO is mole fraction of H2 and CO respec-
tively. Where R is the reduction extent of pellets, t is the reduc-
The FeO reduction equilibrium constant (Kθ) in the tion time, Deff is effective diffusion coefficient, kg is mass
kinetic model of interface reaction is the minimum of K′ transfer coefficient of gas boundary laryer, Kθ is the equi-
and Kw as described by following:
Kθ = min (K ′, K w ) ........................... (5)
The Kθ for different reduction conditions can be obtained
by Eq. (5) and results are shown in Table 3.
The effect of H2 addition on the reduction kinetics of
iron oxides can be analyzed by the shrinking core model of
unchanging size, the kinetic equation and its integral form
are presented as follows:33)

dR
=
dt
3(c 0 − c∗ ) ... (6)
 1 r0   
1 θ 2
− K −
 + (1− R ) − 1 +
3
θ
(1− R) 3
(r0 ρo )
 kg Deff   krea+ (1+ K )  Fig. 8. XRD patterns of the samples with H 2 =10% at different
temperatures.

Fig. 7. Reduction degree and rate under different temperature.

Table 3. Kθ under different reduction conditions.

T/°C 30%CO + 0%H 2 30%CO + 5%H 2 30%CO +10%H 2 30%CO +15%H 2 30%CO + 20%H 2
700 0.27 0.27 0.31 0.33 0.35
900 0.17 0.20 0.23 0.26 0.29
1 000 0.14 0.18 0.22 0.25 0.28

© 2017 ISIJ 408

 ISIJ International, Vol. 57 (2017), No. 3

librium constant, r0 is characteristic initial radius of pellets, by fitting various kinetic functions (Eqs. (8)–(10)) with the
ρ0 is oxygen density in pellets, (c0–c*) is the concentration experimental data and determining the higher value of R2,
gradient of reducing gas. indicating a good linearity. The best fitting model can be
In the reaction process, the gas flow rate is 7.32 cm/s. further used to derive the kinetic parameters of k and Deff.
The effect of external diffusion can be negligible as the gas Figure 9 shows the functions using the experimental results
flow rate is above 5 cm/s. Therefore, it is presumed that the for interfacial chemical reaction control and product layer
reaction rate might be controlled by three mechanisms:34) 1) diffusions controls respectively at different temperature with
the intrinsic interfacial chemical reaction (Eq. (8)), 2) the 30%CO + 10% H2.
diffusion including internal diffusion of reactant and product It can be seen from Fig. 9, the reduction temperature has a
gas species through solid product layer (Eq. (9)), and 3) the significant influence on reduction mechanism. The value of
combination of 1) and 2) presented by Eq. (10). R2 for two curves in Fig. 9(a) is close (0.987 and 0.990), and
hence, neither the interfacial chemical reaction nor the inter-
 1
 nal diffusion is considered to be the rate-controlling steps
t = k1 1 − (1 − R ) 3  ........................... (8)
  at 700°C. Comparatively, the higher linearity is obtained
when the mix control of reaction and internal diffusions is
Kθ ρ B1r0 assumed at all of reduction at 700°C. In addition, there is
Where k1 = × 0 ∗ .
θ
krea+ (1 + K ) (c − c ) an obvious intersection between two curves at 900°C (Fig.
9(b)). And the chemical-control curve has a higher linearity
 2
 than diffusion control curve within 1.75 hours, afterward,
t = k2 1 − 3(1 − R ) 3 + 2(1 − R) ................... (9)
  the mixed model matches well with the experimental data.
These results indicate that the reduction rate is controlled
ρo r0 2 by the interfacial chemical reaction at the initial stage of
Where k2 = .
6 Deff (c 0 − c∗ ) reduction, and then controlled by the mixed model in the
later stage. However, at 1 000°C, it is easy to observe
t  1 2
 that the reduction of samples would be controlled by the
1
= k1 + k2 1 + (1 − R ) 3 − 2(1 − R ) 3
 .... (10) interfacial chemical reaction. The results indicate that the
  kinetics can be explained by the shrinking core model, and
[1 − (1 − R)]3
the rate-controlling step will transform from mixed model to
chemical reaction with the increase of temperature.
3.3.1. Controlling Step of Reduction Rate In addition, the slopes of these control lines express the
The controlling step of reduction rate can be obtained apparent rate constant of reduction at different tempera-

Fig. 9. Determination of the reaction model based on the shrinking core model at different temperature.

409 © 2017 ISIJ

 ISIJ International, Vol. 57 (2017), No. 3

ture. As shown in Fig. 9, the slops of diffusion-controlled
line are 0.10178, 0.28482 and 0.42849 while the slopes of
chemical-controlled line are 0.12652, 0.265 and 0.34251 at
700°C, 900°C and 1 000°C respectively. The apparent rate
constant of internal diffusions increases more obviously
than the apparent rate constant of chemical reaction with
the increase of temperature. Therefore, rate-controlling step
will transform from mixed model to chemical reaction with
the increase of temperature.
Figure 10 shows the functions using the experimental
results for interfacial chemical reaction control and internal
diffusions controls respectively at 900°C with different H2
addition.
It can be seen from Fig. 10, H2 addition has a significant
effect on reduction mechanism. Similar to the effect of
temperature (Fig. 9), the gap between the two control curves
gradually becomes smaller with the increase of H2, and
the two control curves intersect when H2 addition is above
10%. The rate-controlling step of iron oxides reduction has
been determined by comparing the value of R2. This conclu-
sion can be drawn that the rate-controlling step should be
the mixed model when the H2 addition is below 10%, and
that should be the interfacial chemical reaction when the
H2 addition is above 10%. And there is transformation of
reduction mechanism when H2 addition is 10% (Fig. 10(c)).
The reduction rate is controlled by the interfacial chemical
reaction within 1.75 h (Fig. 10(d)), and then controlled by
the mixed model in the later stage. The results indicate
that, in the gas-injection BF, the rate-controlling step will
transform from mixed model to chemical reaction model
with the increase of H2 addition, which is deduced from the
case that the apparent rate constant of internal diffusions

Fig. 10. Determination of the reaction model based on the shrinking core model with different H 2 addition.

© 2017 ISIJ 410

 ISIJ International, Vol. 57 (2017), No. 3

increases more obviously than the apparent rate constant of 3.3.2. Effect of H2 Addition on Arrhenius Activation
chemical reaction. Energy
The rate-controlling step of iron oxides under different The best fitting model can be further used to derive the
conditions has been shown in Table 4. The reduction rate kinetic parameters of reaction rate constant (k) and effective
is mainly controlled by the mixed model composed of inter- diffusion coefficient (Deff) as shown in Table 5.
facial reaction and internal diffusions without H2 addition. The activation energy can be obtained from the Arrhenius
However, when the H2 addition is 5%–20%, there is a tran- equation (Eqs. (11)–(14)) by using the results of k and Deff
sition from mixed controlled model to chemical-controlled in Table 5.
model with the increase of temperature and H2 addition.
De = D0 e − ED / RT ............................ (11)
At 900°C, the relative resistance of internal diffusion
decrease with the increase of H2 addition for the reason
that the diffusion coefficient of H2 in solid is 3 times than k = Ae − Ea / RT .............................. (12)
that of CO diffusion coefficient. In addition, the phase of
ln De = − E D / RT + ln D0 ..................... (13)
poor reduction, such as MgFe2O4 and FexSiyO4, will faster
generate with the increase of temperature and H2 addition
ln k = − Ea / RT + ln A ....................... (14)
in the gases, which cause the relative resistance of inter-
facial chemical reaction increase during the process of the Where ED and Ea are the activation energy of internal
reduction. diffusion and chemical reaction respectively, A is pre-
exponential factor, R is molar gas constant.
The activation energy can be obtained according to the
plots of lnk and lnDeff vs. 1/T as shown in Table 6 by using
the two element regression method.
Table 5. k and Deff of reduction at different conditions.
From the Table 6, we can see the activation energy values
Atmosphere of reduction T/°C Deff (m 2/h) k/(m/h) decrease with the increase in H2 content, resulting in the
700 0.083 471.360 reaction rate increasing at a given temperature, which indi-
cates that the reduction of iron oxides would easily occur
30%CO + 0%H 2 900 0.736 1 345.935
with the H2 addition in the reducing mixture.
1 000 1.688 1 522.294
700 0.184 571.438
4. Conclusion
30%CO + 5%H2 900 1.068 1 031.464
(1) Both the reduction extent and rate of pellets increase
1 000 2.319 1 555.373
with the increase of temperature and H2 addition in gases, no
700 0.311 636.559 matter is the temperature below or above 810°C. However,
30%CO +10%H2 900 1.501 1 401.189 the effect is different when the H2 content increases in the
1 000 3.129 1 818.794 same range. The reduction rate increases most obviously
with the increase of H2 from 15% to 20% below 900°C.
700 0.414 743.107
However, there is a greater impact on the reduction rate with
30%CO +15%H 2 900 2.437 1 400.378 the increase of H2 content from 10% to 15% than that from
1 000 4.238 2 199.299 15% to 20% at 1 000°C. In addition, when the H2 content
700 0.754 1 098.665
is less than 5%, it has little effect on the reduction rate.
Considering the effect of H2 addition on the reduction rate,
30%CO + 20%H 2 900 4.238 1 799.208
the H2 content should be 5%–15%.
1 000 7.014 2 703.725 (2) The controlling step of reduction rate is obtained

Table 4. Rate-controlling step of reduction at different conditions.

T/°C 30%CO + 0%H 2 30%CO + 5%H 2 30%CO +10%H 2 30%CO +15%H 2 30%CO + 20%H 2
700 Mixed model Mixed model Mixed model Mixed model Mixed model
R≤85.66% chemical
reaction Interfacial chemical Interfacial chemical
900 Mixed model Mixed model
R > 85.66% Mixed reaction reaction
model
Interfacial chemical Interfacial chemical Interfacial chemical Interfacial chemical
1 000 Mixed model
reaction reaction reaction reaction

Table 6. ED and Ea with different H 2 addition/(kJ/mol).

30%CO + 0%H 2 30%CO + 5%H 2 30%CO +10%H 2 30%CO +15%H 2 30%CO + 20%H 2
ED 102.98 86.26 78.31 94.13 77.65
Ea 55.63 50.99 36.32 35.79 30.39

411 © 2017 ISIJ

 ISIJ International, Vol. 57 (2017), No. 3
according to the shrinking core model. The results show
that, when the H2 addition is 5%–20%, there is a transition
from mixed controlled model to chemical-controlled model
with the increase of temperature and H2 addition.
(3) The relative resistance of internal diffusion gradually
increases during the process of reduction with the increase
of product layer. At the same temperature, the relative
resistance of internal diffusion decreases with the increase
of H2 addition. In addition, the phase of poor reduction,
such as MgFe2O4 and FexSiyO4, will generate faster with the
increase of temperature and H2 addition in the gases, which
causes the relative resistance of interfacial chemical reaction
increases during the process of the reduction.
(4) The activation energy value decreases considerably
with H2 addition in reducing mixture, which indicates that
the reduction of iron oxides would easily occur with the H2
addition in the reducing mixture.
Acknowledgement
Thanks are given to the financial supports form the key
Program of National Nature Science Foundation of China
(U1360205), the Graduate Student Innovation Fund of
North China University of Science Technology (2015B01)
and Development Program and Science Technology of
Tangshan (4110101E-5).
REFRENCES
1) K. D. Xu. China: Iron Steel, 45 (2010), 1.
2) E. P. da Rocha, V. S. Guiherme, J. A. de Castro, Y. Sazaki and J.
Yagi: J. Mater. Res. Technol., 2 (2013), 255.
3) K. S. Abdel Halim: J. Iron Steel Res., Int., 20 (2013), 40.
4) J. M. Steer, R. Marsh, M. Greenslade and A. Robinson: Fuel, 151
(2015), 40.
5) S. W. Du, C. P. Yeh, W. H. Chen, C. H. Tsai and J. A. Lucas: Fuel,
143 (2015), 98.
6) V. Trinkel, N. Kieberger, T. Bürgler, H. Rechberger and J. Fellner:
J. Clean. Prod., 94 (2015), 312.
© 2017 ISIJ
7) C. Wanga, M. Larsson, J. Lövgren, L. Nilsson, P. Mellin, W. Yang,
H. Salman and A. Hultgren: Energy Procedia, 61 (2014), 2184.
8) W. H. Chen, C. L. Hsu and S. W. Du: Energy, 86 (2015), 758.
9) W. H. Chen, M. R. Lin, A. B. Yu, S. W. Du and T. S. Leu: Int. J.
Hydrog. Energy, 37 (2012), 11748.
10) H. Ghanbari, F. Pettersson and H. Saxén: Chem. Eng. Sci., 129
(2015), 208.
11) P. Jin, Z. Jiang, B. Cheng, S. Hao and X. Zhang: Resour. Conserv.
Recycl., 3062 (2015), 1.
12) J. van der Stel, M. Hattink, D. Sert and W Küttner: METEC InSteel
Con 2011, Lean CO2 Ironmaking, Session 5, Steel Institute VDEh,
Düsseldolf, (2011), 1.
13) C. P. Yeh, S. W. Du, C. H. Tsai and R. J. Yang: Energy, 42 (2012),
233.
14) E. Tsupari, J. Kärki, A. Arasto and J. Lilja: Int. J. Greenh. Gas Con-
trol, 32 (2015), 189.
15) A. Arasto, E. Tsupari, J. Kärki, J. Liljia and M. Sihvonen: Int. J.
Greenh. Gas Control, 30 (2014), 140.
16) L. Hooeya, A. Tobiesenb, J. Johns and S. Santos: Energy Procedia,
37 (2013), 7139.
17) R. Schott: Iron Steel Technol., 3 (2013), 63.
18) H. Luengen, M. Peters and P. Schmöle: Iron Steel Technol., 9 (2012),
63.
19) Q. Lyu, F. M. Li and X. B. Li: China: Iron Steel, 43 (2008), 17.
20) F. M. Li, Q. Lyu and X. B. Li: China: Iron Steel, 42 (2007), 12.
21) H. Guo, F. M. Li and Q. Lyu: J. Hebei Inst. Technol., 29 (2007), 27.
22) W. K. Jozwiak, E. Kaczmarek, T. P. Maniecki and W. Maniukiewicz:
Appl. Catal. A: Gen., 326 (2007), 17.
23) H. J. Guo: Physical Chemistry in Metallurgy, Press of Metallurgy
Industry, Beijing, (2006), 246.
24) A. Bonalde, A. Henriquez and M. Manrique: ISIJ Int., 45 (2005),
1255.
25) D. B. Guo, M. Hu, C. X. Pu, B. Xiao, Z. Q. Hu, S. M. Liu, X. Wang
and X. L. Zhu: Int. J. Hydrog. Energy, 40 (2015), 4733.
26) F. Wen, H. Wang and Z. Tang: Thermochim. Acta, 520 (2011), 55.
27) K. Piotrowski, K. Mondal, T. Wiltowski, P. Dydo and G. Rizeg:
Chem. Eng. J., 131 (2007), 73.
28) K. Piotrowski, K. Mondal, H. Lorethova, L. Stonawski and T.
Wiltowski: Int. J. Hydrog. Energy, 30 (2005), 1543.
29) K. Mondal, H. Lorethova, E. Hippo, T. Wiltowski and S. B. Lalvani:
Fuel Process. Technol., 86 (2004), 33.
30) J. Zielinski, I. Zglinickab, L. Znaka and Z. Kaszkur: Appl. Catal. A:
Gen., 381 (2010), 191.
31) Y. M. Wang, Z. F. Yuan and H. Matsuura: ISIJ Int., 49 (2009), 164.
32) J. M. Pang, P. M. Guo and P. Zhao: Iron Steel, 44 (2009), 11.
33) J. Szekely, J. W. Evens and H. Y. Sohn: Gas-solid Reaction,
Academic Press, New York, (1967), 321.
34) H. Imanishi, A. Maeda, T. Maegawa, S. Matsuno and T. Aida: Chem.
Eng., 63 (2008), 4947.
412