Preparation of iron Powders by Reduction of Rolling Mill Scale, International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013,1457-1472

Abstract

The objective of this work is the recycling of mill scale formed during the steel hot rolling process with a reducing gas (carbon monoxide) in order to produce iron powder having characteristics required by powder metallurgy. The reduction was carried out at various temperatures (750-1050°C) during different times ranging between 40 and 180 min in an atmosphere of pure CO. The produced iron powder was characterized by chemical analysis, x-rays diffraction, optical microscopy and scanning electron microscopy. These methods of investigation confirm the presence of iron, graphite and iron carbide (Fe3C) as the products of reactions. The maximum iron content (98.40% Fe) in the iron powder was obtained by reduction of mill scale at 1050°C for 180 min. A reduction annealing under hydrogen makes it possible to decrease carbon and oxygen content of the reduced iron powder up to acceptable values.

Full text

Preparation of iron Powders by Reduction of
Rolling Mill Scale
Said Mechachti1, Omar Benchiheub1, Salim Serrai1, Mohamed E.H.Shalabi2
Abstract— The objective of this work is the recycling of mill scale formed during the steel hot rolling process with a reducing gas (carbon monoxide)
in order to produce iron powder having characteristics required by powder metallurgy. The reduction was carried out at variou s temperatures (750-
1050°C) during different times ranging between 40 and 180 min in an atmosphere of pure CO. The produced iron powder was characterized by
chemical analysis, x-rays diffraction, optical microscopy and scanning electron microscopy. These methods of investigation confirm the presence of iron,
graphite and iron carbide (Fe3C) as the products of reactions. The maximum iron content (98.40% Fe) in the iron powder was obtained by reduction of
mill scale at 1050°C for 180 min. A reduction annealing under hydrogen makes it possible to decrease carbon and oxygen content of the reduced iron
powder up to acceptable values.
Index Terms— Carbon monoxide, Iron powder, Mill scale, Recycling, Reduction.
——————————  ——————————
1 INTRODUCTION
HE steelmaking by-products such as dust and mill
scale, very rich in iron (≈ 72% Fe), are currently
produced in large quantities and represent a potential of
almost 5 million tons in the world [1]. Generally, these
by-products are recycled by the metallurgical processes
such as the blast furnace or the direct reduction reactors
that uses coal as reducing agent to produce pre-reduced
pellets intended for the remelt in electric steel plant.
Besides the steelmaking, recycling part of these by-
products is already supported by the powder metallurgy
where the economic recovery is more favorable. During
the last twenty years, powder metallurgy has presented a
continued expansion in all its aspects and in all of its
applications to the industry. Powder metallurgy
comprises a set of processes of forming having for
common denominator a raw material in a powder form.
The reduced iron powder is the most widely used
material in powder metallurgy industry. The direct
reduction process has commonly been used by many
companies (such as Hoeganas in Sweden, Kawasaki in
Japan and Pyron in US) to obtain metallic iron powder by
the reaction of iron oxide (magnetite, hematite ore or mill
scale) and reducing gases (CO/H2) under high
temperatures (>1000°C) [2]. The reduction of iron oxides
and various ores containing iron oxides have been
studied in the past [3-8]. Their researches are conducted
using as reducing agents solid carbon and reducing gas
(CO and/or H2). When the reduction is carried out by
solid carbon or carbon monoxide the final process of
reduction is always written as follows:
FenOm + mCO(g) = nFe + mCO2(g) (1)
The necessary CO is provided directly or by the
Boudouard reaction:
mC(s) + mCO2(g) = 2mCO(g) (2)
The Boudouard reaction, regenerating reducing gas CO,
keeps the PCO/PCO2 ratio always great and there for the
reduction process according to (1) will be maintained at a
constant level. In the event of reduction occurring by
circulation gas over or through a bed of particles of iron
oxides (fixed or mobile) the CO2 formed by the reaction (1)
will be carried away by gas flow, which allows to reach
the quasi-complete reaction of a bed of oxide particles
[9].In addition to the reduction to metallic iron,
disintegration of carbon monoxide by the reverse
Boudouard reaction (2) and carburization are expected to
occur simultaneously. This is confirmed by
thermodynamic studies which indicate that relatively low
temperatures (less than 1000°C) and high concentrations
of CO along with the presence of metallic iron, which acts
as a catalyst, lead to carbon deposition [10]. The increase
in temperature and CO concentration leads to reduction
in iron oxides without carburizing.
T
 1. Foundry Laboratory, Department of metallurgy and materials, Badji-
Mokhtar Univrsity, BP 12-Annaba-23000 Algeria.
Author's Email: saidmechachti@yahoo.fr
 2. Central Metallurgical Research and Development Institute (CMRDI),
Cairo, Egypt.
International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013
ISSN 2229-5518
1467
IJSER

20 30 40 50 60 70 80 90 100 110
0
200
400
600
800
1000
1200
1400
1600
a. u.
12
2
3
2
1
1
2
1
2
2
1
1
1
3
3
3
3
3
a.u.
Other parameters can greatly affect the kinetics of
reduction of iron oxides. The rate of reduction of iron
oxides depends on several factors that can vary from one
process to another, especially the grain size, porosity and
specific surface, mineralogy (Fe2O3 and Fe3O4 are not
reduced in the same way), pressure, gas flow and the
gangue constituents such as silica, alumina and silicates
which may alter the equilibrium [11]. The aim of this
work is to study:
- The optimal conditions for the reduction of mill scale to
obtain iron powder with acceptable purity for the use in
powder metallurgy industry.
- The effect of carbon deposition on the reduction of iron
oxides by pure CO.
2 EXPERIMENTAL
2.1 Materials Used
Mill scale generated in a hot rolling step was heated to
400°C to remove the oil and crushed to the desired
particle size distribution. Three fractions of particles were
used in this study: 2-3.15, 6.3-10 and 10-16 mm. Sieves
were standardized. Chemical analysis of the mill scale is
given in Table 1. Analysis by X-ray diffraction (Fig. 1.I)
and microstructure of the mill scale particles (Fig. 1.II)
shows the presence of hematite (Fe2O3), magnetite (Fe3O4)
and wustite (FeO) phases. Oxidation at 1000°C converts
the iron oxides that are present in the mill scale, FeO and
Fe3O4, to ferric oxide (Fe2O3).
TABLE 1
CHEMICAL COMPOSITIONS OF MILL SCALE
2.2 Reduction System and Procedure
The reduction tests were performed at the center of
applied research in metallurgy and steel (URASM),
Annaba. The reduction device and the method used to
perform these tests are described in several works [12, 13].
In this method, where reduction was done in fixed bed,
the sample was placed in a vertical reduction shaft which
was introduced into an electric furnace maintained at a
fixed temperature of up to 1600°C. As long as the sample
had not reached test temperature, it was kept under
nitrogen to prevent oxidation. When the temperature was
stabilized, purified CO was passed at fixed rate during a
fixed time. Preliminary reduction tests were carried out at
1000°C to determine the optimal gas flow rate which does
not significantly affect the rate of reduction. As a result of
these tests, a 1000 l/h gas flow of CO gas was used in all
the reduction experiments. The reduced samples were
cooled to room temperature in a nitrogen gas atmosphere
for subsequent examination and chemical analysis. The
tests were performed in two steps: (a) To study the effect
of particle size, the CO/N2 ratio and oxidation of mill
scale on its reducibility at 1000°C and 60 min. (b)
Extending the study to investigate the influence of
various temperatures (750-1050°C) and different
reduction times (40-180 min) under an atmosphere of
pure CO.
The evolution of the reduction rate was calculated on
the basis of data obtained by chemical analysis. The
scanning electron microscope (SEM), light microscope
(LM) and X-ray technique were used to characterize the
structure and the different phases formed in each
reduced sample. The total amounts of carbon deposited
in the reduced samples were determined according to
ISO 437 [14].The oxygen content was also measured in
accordance with ISO 4491-2 [15].
3. RESULTS AND DISCUSSION
3.1 Effect of Some Parameters on the Rate of
Reduction
Preliminary tests were carried out in order to study the
effect of CO rate in the CO/N2 mixture, particle size and
oxidation of mill scale on the evolution of reduction at fixed
time (60 min) and 1000°C. The results are presented in Table
2.
The mill scale sample reduced in the conditions of test
N1 has low total iron content and a low rate of reduction,
78.85 and 34.48% respectively. These two indexes evolve
in the same direction (80.30 and 41.80%) when the
fraction 6.3-10 mm was used, keeping the same CO/N2
ratio. For the same fraction (6.3-10 mm), increasing the
CO/N2 ratio from 40 to 100% (corresponding to pure CO),
while maintaining the other parameters unchanged,
increases significantly the rate of reduction (41.80 to
68.06%) and total iron content (80.30 to 88.59%). The
Compounds
Fetotal
FeO
CaO
SiO2
MnO
%
72.13
56.70
0.42
0.14
0.37
Fig.1. Oxide scale formed on the mill scale: (I) XRD spectra
(II) cross-section microstructure.
(1) Hématite (2) Magnétite (3) Wustite
2Theta
b
c
a
II
I
International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013
ISSN 2229-5518
1468
IJSER

amount of metallic iron (Fe0) increased and that of Fe2+
decreased with decreasing particle size and increasing
TABLE 2
EFFECT OF CO/N2 RATIO, PARTICLE SIZE AND MILL SCALE
OXIDATION ON THE RATE OF REDUCTION AND CHEMICAL
COMPOSITION OF REDUCED IRON POWDER AT 1000°C AND 60 min
CO/N2 ratio. Test N°5 shows the effect of prior oxidation
of the mill scale on its reducibility. The mill scale has
undergone an oxidation at 1000°C. The rate of reduction
and the total iron content of reduced iron powder
increases to 86.21 and 92.96% respectively. The maximum
iron content (94.74%) and rate of reduction (90.78%) was
Obtained when we used sample
of particle size 2-3.15 mm (test N°6). The most important
result to be drawn from these tests is the strong influence
of the rate of CO in the CO/N2 mixture, the oxidation of
the mill scale and the particle size on the rate of reduction
and the total iron content of reduced iron powder. The
total amount of iron reaches its maximum value (94.74%)
when oxidized mill scale with particle size 2-3.15 mm is
reduced with pure CO at 1000°C and 60 min.
3.2. Effects of Temperature and Reduction Time
Another series of tests were performed by varying the
temperature (750 to 1050°C) and reduction time (40 to 180
min). Mill scale with particle size 2-3.15 mm undergoes
an oxidation at 1000°C. The reducing gas used was pure
CO.
The results of reduction tests are presented in Fig. 2.
At 750°C it was found that with the increase reduction
time from 40 to 180 min the total iron content of reduced
iron powder decreases from 88.5 to 75.3% respectively. At
the same time interval, the amount of metallic iron
increases (40.6 to 56.9%) and that of Fe2+ decreases (47.8 to
19%). The decrease in total iron content can be explained
as follows: along the reduction of iron oxides by pure
carbon monoxide to wustite and metallic iron a
disintegration of carbon monoxide takes place
Tests
Size
mm
Gas
CO/N2
Rate of
reduction
%
Chemical
composition, %
Fetot Fe2+ Fe0 Fe3+
N1 10-16 40/60 34.48 78.85 54.32 18.04 6.49
N2 6.3-10 40/60 41.80 80.30 51.10 24.63 4.57
N3 6.3-10 60/40 49.38 85.68 48.38 33.60 3.60
N4 6.3-10 100 68.06 88.59 34.93 53.53 0.13
N5 6.3-10 100 86.21 92.96 17.18 75.66 0.12
N6 2-3.15 100 90.78 94.74 12.10 82.54 0.10
Fig. 2. Evolution of Fetotal, Fe0 and Fe2+ content in reduced
iron powders with temperature and reduction time.
20 40 60 80 100 120 140 160 180 200
0
20
40
60
80
100
Reduction time (min)
Fet, Fe0 and Fe2+ (%)
20 40 60 80 100 120 140 160 180 200
0
20
40
60
80
100
Reduction time (min)
Fet, Fe0 and Fe2+ (%)
20 40 60 80 100 120 140 160 180 200
0
20
40
60
80
100
Reduction time (min)
Fet, Fe0 and Fe2+ (%)
20 40 60 80 100 120 140 160 180 200
0
20
40
60
80
100
Reduction time (min)
Fet, Fe0 and Fe2+ (%)
1050°C
950°C
850°C
750°C
International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013
ISSN 2229-5518
1469
IJSER

40
60
80
100
120
140
160
180
200
0
1
2
3
4
5
6
7
8
9
10
Reduction
temperatures
750°C
850°C
950°C
1050°C
Total carbon (%)
Reduction time,
min
Fig. 3. Evolution of total carbon content in reduced iron powders
with temperature and reduction time.
F- Iron
W- Wustite
G- Graphite
C- Fe3C
F
950°C
850°C
F
F
F
F
G
G
1050°C
F
F
F
F
7 50°C
C
F
F
F
W
G
C
W
F
G
2
0
40
0
60 0
80
0
10
0
12
0
a.u.
F
F
F
F
G
2 Theta
Fig. 4. X-Ray spectra of the reduced iron powders at various
temperatures and 180 min.
monoxide takes place according to the reverse
Boudouard reaction [10]:
2(CO)ads → (CO2)g + C
(3)
and accordingly the carbon deposition and its reaction
with metallic iron and wustite leads to the formation of
iron carbide. Carburization of the reduced iron or wustite
grains can take place according to the following reactions:
3Fe + C → Fe3C
(4)
3Fe + 2CO → Fe3C + CO2
(5)
3FeO + 5CO → Fe3C + 4CO2
(6)
Carbon monoxide, adsorbed (reaction 3) onto the surface
of metallic iron grains disintegrates more easily than
when it is free. The metallic iron acts as a catalyst [16].
Thus, as shown in Fig. 2 at 750°C, the decrease in total
iron content is attributed to the increase of amount of
carbon deposition with the increase of reduction time and
amount of metallic iron. This carbon increases the weight
of the sample and consequently the total iron content is
increased. The total amount of carbon in reduced iron
powder at various temperatures and reduction times is
presented in Fig 3.
It shows that the carbon content in the reduced iron
powders decreased with rise in temperature and with
decrease in the reduction time. At 850°C, as shown in Fig.
2, the amount of total iron, metallic iron and Fe2+ evolve
in the same direction as the tests at 750°C. However, the
decrease of amount of total iron content with reduction
time is less pronounced. As a result, the amount of
carbon deposition with reduction time is lower at 850°C.
For reduction tests at 950°C and 1050°C there is already a
significant improvement in the content of total iron and
metallic iron with increasing reduction time. At 180 min,
the content of total iron and metallic iron increases up to
98.30 and 95.50% at 950°C and up to 98.40 and 97.20% at
1050°C respectively. As shown in Fig 3, the carbon
content in the reduced sample decreased up to 1% at
1050°C and 180 min.
The different phases formed in the reduced iron
powders at 750-1050°C and at 180 min, as identified by X-
ray diffraction technique, are shown in Figure 4. The
diffraction pattern of the reduced iron powder at 750°C
shows the presence of metallic iron (FeO), wustite,
graphite and iron carbide (Fe3C). At 850 and 950°C, it was
observed the presence of metallic iron and graphite and
at 1050°C only metallic iron phase was obtained.
Microscopic examination of polished sections of
reduced samples at various temperatures and at 40 min
and 180 min are given in Figures 5 and 6 respectively.
Figures 5a to 5d, shows the photomicrographs of partially
reduced samples at 40 min and 750 to 1050°C. The four
photo-micrographs show frontal reduction of oxides
particles into magnetite (Figs. 5a, b) or wustite (Fig. 5c, d)
in an envelope of metallic iron. At 180 min and at lower
temperatures (750 and 850°C), metallic iron dominated
with a minor phase of wustite were observed as shown in
Figs. 6a and 6b. For reduced samples at 950 and 1050°C,
iron oxides are completely converted to metallic iron as
shown in Figures 6c and 6d.
International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013
ISSN 2229-5518
1470
IJSER

Fig. 5. Photomicrographs of reduced iron powders (x220) at 40
min and at (a) 750°C, (b) 850°C, (c) 950°C and (d) 1050°C.
M: magnetite, W: wustite, F: iron
a
M
F
M
F
b
c
W
c
W
F
d
F
W
Fig. 6. Photomicrographs of reduced iron powders (x220) at 180
min and at (a) 750°C, (b) 850°C, (c) 950°C and (d) 1050°C.
W: wustite F: iron
W
F
b
F
W
a
F
c
d
F
Fig. 7. SEM photomicrographs of completely reduced iron
powder at 180 min and at 1050°C.
a
b
SEM photographs of reduced samples at 1050°C
and 180 min are presented in Figures 7a and 7b.
Irregular shape of the particles surfaces and particle
porosity are visible. This morphology is the typical form
of particles of iron powder obtained by reduction with
carbon monoxide [17, 18].
The results presented in Figures 2 to 7 shows that the
best reduction conditions for reduced iron powder are
1050°C and 180 min. Chemical analysis of these powders
are presented in the Table 3.
To reduce the carbon and oxygen content, the iron
powder is annealed in an atmosphere of hydrogen at
850°C for 1 hr. During annealing the carbon and oxygen
content is considerably reduced and the strain hardening
of particles is removed [16, 17]. After annealing, the
contents of carbon and oxygen in iron powder increases
to 0.23 and 0.28% respectively.
4 CONCLUSION
The results presented and discussed in this study can
be summarized in the following points:
(1)- Different parameters for the reduction of the mill
scale with carbon monoxide were studied to obtain the
best conditions for the production of iron powder with
physico-chemical properties required by powder
metallurgy.
(2)- The best reduction was obtained at 1050°C and 180
min in pure CO atmosphere. This is confirmed by
chemical analysis, X-ray diffraction and light microscope.
The content of total iron and metallic iron in the reduced
iron powder is 98.4 and 97.2% respectively.
(3)- Annealing by hydrogen reduces considerably carbon
and oxygen content in the iron powder up to the desired
values.
(4)- The iron powders obtained by the process of
reduction of mill scale can be used in the powder
metallurgy industry.
REFERENCES
[1] Bienvenu, Y., Rodrigues, S. Manufacture of Metal
Powders from Pulverulent Waste, ENSMP, Centre des
matériaux, CNRS UMR 7633, France (2007).
[2] Bouvard, D. Métallurgie des poudres, Paris, Hermès
Science Publications, (2002).
[3] Pyron Expands Across the Powder Range. MPR, (1995).
[4] Pineau, A., Kanariand, N., Gaballah, I. Kinetics of
Reduction of Iron Oxides by H2, Part I: Low Temperature
Reduction of Hematite, Thermochimica Acta 89, 447 (2006).
[5] Pineau, A., Kanariand, N., Gaballah, I. Kinetics of
Reduction of Iron Oxides by H2, Part I : Low Temperature
Reduction of Magnetite, Thermochimica Acta 75, 456
(2007).
[6] Kohl, H.K., Marincek, B. Kinetics of the Reduction of Iron
Oxides with Graphite, Archiv Fur das Eisen 36, 851 (1965).
TABLE 3
CHEMICAL COMPOSITIONS OF REDUCED IRON POWDER AT 1050°C
AND 180 min
Compounds
Fet
Fe0
C
Si
Mn
P
S
O
%
98.4
97.2
1.08
0.03
0.32
0.04
0.01
0.49
International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013
ISSN 2229-5518
1471
IJSER

[7] El-Geasy, A.A., Nasr, M.I. Influence of Original Structure
on the Kinetics and Mechanisms of Carbon Monoxide
Reduction of Hematite Compacts. ISIJ International, 417
(1990).
[8] Martín, M.I., López, F.A. Obtainment of Sponge Iron by
Reduction of a Steel aking By- Product, 1st Spanish
National Conference on Advances in Materials Recycling
and Eco- energy, Madrid12, (2009).
[9] Philiber, J., Vignes, A. Métallurgie du minerai au
matériau, Dunod, (2002).
[10] Mondal, K. Reduction of Iron Oxide in Carbon Monoxide
Atmosphere, Reaction Controlled Kinetics, Fuel Processing
Technology 33, 86 (2004).
[11] Coudurier, L., Hopkins, D.W., Wilkomirsky,
I.Fundamentals of Metallurgical Processes, Pergamon
press, (1978).
[12] Measurement of the Reducibility of Iron Ore by C.R.M.
Method, Liege, (1980).
[13] Berg, B., Hankart, J., Poos, A. Reducibility of the Sinters.
Universal Review of Mines, 9th Series. T.XVII 3,284 (1962).
[14] Normes ISO 4491-2, Metallic Powders, Determination of
Oxygen Content by Reduction Method, Second Edition
(1997).
[15] Normes ISO 437, Determination of Total Carbon Content,
Combustion gravimetric Method, First Edition, (1982).
[16] Akiminko,V.B. Iron Powder, Technology, Properties and
Economics,Moscow(1982).
[17] Aslanoglu, Z. Direct Reduction of Mechanically Actived
Specular Iron Oxides, Minera Processing and Extractive
Metallurgy (Trans. Inst. Min. Metall. C) 114 240 (2005).
International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May-2013
ISSN 2229-5518
1472
IJSER