International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 353

ISSN 2229-5518

Prodution of Ferrous Sulfate From Steelmaking

Mill Scale
* * * *
Pedro Jorge Walburga Keglevich de Buzin , Eunice Maria Vigânico , Rodrigo de Almeida Silva ,Nestor Cezar Heck
* **
Ivo André Homrich Schneider and Jean Carlo Salomé dos Santos Menezes .

Abstract— Spent pickle liquor from steel manufacturers is usually used to produce ferrous sulfate in the route of acid recovery. This work proposes a
new process to be used industrially by which clean mill scale is used as raw material for the production of ferrous sulfate by acid leaching. Process
simplicity and abundance of raw material may be invoked as factors that encourage its implementation. In this work, after characterization, a sample of
mill scale from a semi-integrated steelmaking plant is leached with sulfuric acid (at concentrations of 5%, 10%, and 15%) inside percolation columns; by
2+
recirculation of the acid solution, a liquor rich in Fe is produced. Next to the leaching step, the liquor is concentrated by evaporation at the temperature
of 80 °C and, in the sequence, is left to cool naturally until room temperature. This procedure allows for the crystallization of ferrous sulfate. To finish, the
crystals are collected, purified with ethanol, and characterized by X-ray diffraction. X-ray analysis indicate Melanterite (FeSO 4 ⋅7H 2 O) as the main
compound present in the solid fraction, followed by Szomolnokite (FeSO 4 ⋅H 2 O) and Rozenite (Fe SO 4 ⋅4H 2 O).

Index Terms— ferrous sulfate; mill scale; leaching; recycling.

——————————  ——————————
1 INTRODUCTION
A Semi-integrated steel mills are responsible for a large 2 THEORICAL CONSIDERATIONS
quantity of residues which have to be recycled or properly 2.1 Reactions
dispose. In this kind of industry main waste materials are: With reference to the iron oxides (wustite, hematite and

IJSER
electric arc furnace dust (EAFD), slags, sludges, and mill scale. magnetite), acid attack with sulfuric acid can be described by
They should be recycled, reused or correctly managed in order the following reactions:
to avoid environmental damages, minimizing risks and en- A FeO(s) + H 2 SO 4 (aq) = H 2 O + Fe2+ (aq) + SO 4 2- (aq)
suring benefits to the society (1). (1)
Mill scale is a residue from steel hot rolling process and is A+B = Fe 3 O 4 + 4 H 2 SO 4 (aq) = 4 H 2 O + Fe2+ (aq) + 2 Fe3+ (aq) +
basically composed of iron oxides and metallic iron with vari- 4 SO 4 2- (aq) (2)
able oil and grease contents. The specific production of this B Fe 2 O 3 + 3 H 2 SO 4 (aq) = 3 H 2 O +2 Fe3+ (aq) + 3SO 4 2- (aq) (3)
residue is about 10-20 kg/t of the steel production (2). Mill scale In presence of oxygen, Fe2+ can oxidize to Fe3+ by the following
can be processed and transformed into environment-friendly reaction:
concrete blocks (paving products for landscaping area and car 2 Fe2+ + 1/2 O 2 (g) + 2 H+ = 2 Fe3+ + H 2 O (4)
parks), eco-bricks, or can be used as iron raw material e.g. in As indicated by equation (4), the chemical equilibrium that
self reducing briquettes for the electric arc furnace (3). Another will establish between the ferrous and ferric ions depends on
possibility – considered in this study – is to employ it as a raw (i) the pH (sulfuric acid concentration) and (ii) the oxygen
material for the production of ferrous sulfate or Iron (II) sul- concentration (redox potential, E h ) in solution. Oxygen (from
fate. the atmosphere) is in plentiful supply (at 0.21 atm); neverthe-
Melanterite (FeSO 4 ⋅7H 2 O) – ferrous sulfate heptahydrate) is less an anoxic media may establish when oxygen is consumed
the major commercial form of ferrous sulfate and is produced by chemical or biochemical reactions and no satisfactory re-
mainly as a by-product of titanium dioxide manufacture (4) and newal takes place (under strongly anoxic condition, its ‘partial
in steel pickling with sulfuric acid. It has been largely used to pressure’ can reach very low levels, up to around 1x10-83 [atm]
control anemia (both in humans and animals), base compound – as indicated by the line pH 2 = 1 [atm] in Pourbaix diagrams).
to produce ferric sulfate as well as a reagent for water and As expected, pH is manipulated via acid addition.
wastewater treatment (5). In Brazil, there is a huge deficit in
sulfur-based chemicals, including the Iron (II) sulfate (6). 2.2 Thermodynamic Simulation
The aim of this study is to develop a hydrometallurgical A thermodynamic simulation based on the software
route for the production of ferrous sulfate heptahydrate from FactSage (version 6.2) was done to anticipate the behavior of
mill scale. The study makes use of a sulfuric acid leaching step the mill scale leaching step in an aqueous solution of sulfuric
to dissolve the mill scale and an evaporation/crystallization acid. In this simulation, for the sake of simplicity, only simple
procedure to produce Melanterite crystals. iron oxides are allowed as solid phases. Following their self-
determining nature, pH an redox potential values varied in-
———————————————— dependently from each other. Values for the logarithm of oxy-
• *PPG3M-Universidade Federal do Rio Grande do Sul, Porto Alegre-RS- gen pressure ranged from -25 to -50 (with pressure given in
Brazil. E-mail: [email protected] atmospheres); the sulfuric acid concentration (molality, m)
• **PPGCB-Universidade do Oeste de Santa Catarina, Videira-SC-Brazil varied from 0.1 up to 5 (simulation tags along to the following
E-mail: [email protected] schema: when O 2 pressure varies, molality is kept constant,
with m = 3, while, for m as variable, log(pO 2 ) = -45). For the
two conditions surveyed, pH and redox potential were
IJSER © 2014
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 354
ISSN 2229-5518
determined; when displaying these values in a diagram, 3.2 Scale Characterization
Figure 1 is obtained. For the sake of completeness, for the After a drying procedure at 105°C, a particle size analysis of
conditions studied, the single stable solid oxide phase found scale was made with a series of Tyler sieves with apertures
in the system is Hematite. ranging from 4 to 270 mesh.
A quartered sample of the mill scale was comminuted in a
Tecnal orbital disk mill, model TE360 to a particle size -0.074
mm and subjected to a mineralogical and elemental analyses.
The majority of the elements present in the mill scale was de-
termined by Inductively Coupled Plasma - Atomic Emission
Spectroscopy (sequential ICP-AES, Liberty RL); sample diges-
tion procedure followed the EPA 3050B (7) method.
Mineralogical analysis was performed by X-ray diffraction
(Siemens D-500). This analysis was complemented by Möss-
bauer spectroscopy for a better quantification of iron oxides
present in the scale. Finally, a wet chemical analysis of the total
iron and zero valent iron was made by the procedure de-
scribed in ABNT Standard NBR 8577 (8)

3.3 Leaching of Mill Scale

The production of a leachate rich in iron from the mill scale
was carried out in the laboratory by means of three Pyrex
Fig. 1 Diagram displaying redox potential, E h, and pH result values
glass leaching columns (30 cm height, 6,8 cm outer diameter
from thermodynamic simultion (black squares: initial fixed H2SO4
and 6,5 cm inner diameter), each filled with 500 g of scale; a
molatity m = 3; white triangles: constant oxygen potential).
glass frit septum at the base of the column, holds the scale in

IJSER
place, Figure 2.
In this diagram, concentration of Fe2+ in aqueous solution Mill scale was leached with sulfuric acid (reagent grade La-
can be seen as a surface which drops from the lower-left to the borsynth) at the following concentrations: 5, 10 and 15 wt.%
upper-right corner; for reference, a line was drawn indicating H 2 SO 4 . All of the experiments were carried out at room tem-
where the concentration equals to 1 mol/kg H 2 O (molality Fe2+ perature (23°C), in duplicate.
= 1). The total volume of leaching solution was kept constant at 1
Consequently, both effects of lowering the oxygen partial liter by the addition of deionized water. The leachate recircula-
pressure or increasing the sulfuric acid concentration lead to a tion was done by a submerged peristaltic pump (Sarlo Better
concentration increase of the ferrous ion in solution; however, 300) connected to the leaching system with flexible rubber
solutions which are rich in Fe2+ are not easily obtained by vary- hoses resistant to acid.The total time of the leaching test was
ing only one of these factors. As a result, the ideal leaching 120 hours each.
procedure must be carried out under oxygen deficit and lower
pH values.
Therefore, for leaching procedure simulation, the chosen
values are: for the logarithm of oxygen pressure -40, -45 and -
50 and sulfuric acid molalities up to m = 10. Simulation results
will be shown later, together with the experimental results.
From a comparison between simulation and experiment, it is
clear that the chosen ranges were well-located.

3 MATERIALS AND METHODS

3.1 Mill Scale

Sampling of the mill scale used in this study was conducted
on-site, at the disposal area of the steel mill. The sample was
collected at varying amounts in different parts of the stack, to
improve its representativity. The sampling followed proce-
dures described in ABNT Standard NBR 10007:2004. For stor-
age, a plastic barrel with lid was used (sample remained
stored in the laboratory for a time period of one week before
the beginning of the experimental work).
As the process for the removal of the mill scale is based on
jets of water, the amount of oil remaining in the residue used
in this study is very low, unable to disturb the leaching Fig. 2 Glass leaching (recirculating) columns.
process. During the leaching process, the concentrations of Fe2+, Fe3+
IJSER © 2014
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 355
ISSN 2229-5518
and Total Fe was monitored by wet titrimetry (9), except for the crystals were characterized by X-ray diffraction.
sample digestion, this procedure also followed ABNT Stand-
ard NBR 8577:1984. After the leaching step, the solution with 4 RESULTS AND DISCUSSION
15 wt.% acid concentration was saturated – as small amount of
crystallized ferrous sulfate was found on the bottom of the The result of particle size analysis of mill scale used in the
beakers. At the end of the experiment, the mass of scale dis- experiments can be seen in Table 1 and Figure 4.
solved by the acid solution was obtained computing the dif-
ference between initial and final sample weight.

3.4 Crystallization of Ferrous Sulfate

TABLE 1
The leachate was heated to approximately 80 °C (using a PARTICLE SIZE CHARACTERIZATION OF MILL SCALE
heating plate without heating rate control) in order to increase Tyler Sieve size Passing
its concentration (with volume reduction up to about 50% of scale (mm) (%)
the original volume). In the sequence, the solution was cooled
to the room temperature (23 °C) by natural cooling to promote 270 0.053 1.39
the formation of crystals of iron sulfate, Figure 3 (a) and (b). 200 0.074 4.55
150 0.105 9.38
100 0.149 11.29
65 0.21 16.23
48 0.297 21.71
35 0.42 29.22
28 0.59 37.74
20 0.84 49.02

IJSER
14 1.41 67.18
4 5 100

(A)

Fig. 4 Granulometric analysis of mill scale.

Mill scale has the form of platelets and sieve analysis

shows that only about 10% of the sample is characterized by
having a particle size below 0.1 mm. This low fines content
makes the material favorable for application of hydrometal-
lurgical techniques (low clogging danger).
Mill scale is constituted mainly of iron oxides; the
chemical characterization of the mill scale by ICP-AES is
meant to give an idea of the chemical elements that may con-
(B)
taminate the ferrous sulfate agent; results can be seen in Table
Fig. 3. Ferrous sulfate as crystallized (a) and after purification with
2 (as expected, Fe and oxygen are the most abundant
ethyl alcohol (b). elements).

After rinsing with ethyl alcohol (for the removal of free Fe3+,
soluble in alcohol (10) without the oxidation of ferrous ions (11))
IJSER © 2014
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 356
ISSN 2229-5518
TABLE 2. TABLE 3
CHEMICAL CHARACTERIZATION OF THE MILL SCALE BY ICP- IRON AND IRON OXIDE PHASES PRESENT IN THE MILL SCALE.
AES.
Phase Amount WT%
Elements WT%
Fe 2 O 3 6.4
Al 1.95
Cr 0.32 Fe 3 O 4 16.5

Fe 70.30 FeO 67.9

Hg ND
Feo 2.2
Mn 0.65
Other 7.0
Ni 0.10
O 25.80
next. Figures 6, 7 and 8 show the concentrations of total iron
Si 0.05 (Fe Total ), divalent iron (Fe2+) and trivalent iron (Fe3+) in the
W 0.83 leachate for the three columns during the process of leaching.
A gradual increase in ion concentrations during the first 72
hours of testing is clearly seen in all trials; from this point on
the values become nearly constant.
The results of the mill scale analyses by XRD are pre- This behavior is an indication that the system reached
sented in Figure 5 and Table 3. XRD analysis detected the fol- equilibrium, i.e., there will be no more dissolution of the iron
lowing Iron crystalline phases: wustite (predominant), mag- mill scale. In this study, this happens for the concentrations of

IJSER
netite and hematite. Based on Mössbauer spectroscopy and on total iron of 36.6, 69.2 and 100 g/L, as shown in Table . These
the chemical caracterization by ICP, Table 1, an adjusted chem- values are applicable for the initial sulfuric acid concentrations
ical iron and iron oxide composition is presented, Table 2. It of 5, 10 and 15%, respectively.
can be verified that iron compounds represent 93% of the
sample. Wustite (FeO) is found in greatest quantity, followed
by magnetite (Fe 3 O 4 ), hematite (Fe 2 O 3 ) and zero valent iron.
The non-ferrous compounds, in amounts of about 7% (identi-
fied in ICP-AES analysis), are in the form of oxides and their
influence was not considered in this work.

total 2+ 3+
Fig. 5 Characterization of Iron crystalline phases in the mill scale by Fig. 6 Fe , Fe e Fe content in the leachate, 5% H2SO4.
XRD.

Results from the leaching procedure will be dealt with

IJSER © 2014
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 357
ISSN 2229-5518
ditions of high acidity and high leaching time,
Figure 8. In this case, a relative slight decrease in
the concentration of Fe2+ over the concentration of
Fe3+ is observable; this decrease is to be related with
the oxidation of Iron (II) sulphate. Table 4 presents
the limiting concentration values for the different
forms of iron and its percentage at the end of the
tests for all the three sulfuric acid concentrations
used in the iron mill scale leaching step.
One can also observe from Table 4 that the
TABLE 4
IRON AND IRON OXIDE PHASES PRESENT IN THE MILL SCALE. LIMITING
CONCENTRATIONS FOR DIFFERENT FORMS OF IRON AND ITS PERCENT-
AGE RELATIVE TO THE END OF LEACHING FOR THREE SULFURIC ACID
CONCENTRATIONS.

Equilibrium Percentage [wt%]

concentration [g/L]
H 2 SO 4 5% 10% 15% 5% 10% 15%

Fig. 7 Fe
total
, Fe2+ e Fe3+ content in the leachate, 10% Fe total 34.6 69.2 100.0 100.0 100.0 100.0

IJSER
H2SO4.

Fe2+ 26.8 42.4 57.5 73.2 61.3 57.5

Fe3+ 7.8 26.8 42.5 26.8 38.7 42.5

limiting amount of leached iron increases almost
linearly as a function of concentration of sulfuric
acid in solution, Figure 9.

total Fig. 9 Limiting amount of iron dissolved in sulfuric acid solution as a

Fig. 8 Fe , Fe2+ e Fe3+ content in the leachate, 15% function of H2SO4 molality for three O2 pressures; black squares =
H2SO4. experimental values.

The amount of theoretically total iron dis-

An exception was observed only under con- solved (from thermodynamic simulation) is also
IJSER © 2014
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 358
ISSN 2229-5518
linearly proportional to the molality of H 2 SO 4 , var- X-ray diffraction analysis of crystals after purifica-
ying slightly depending on the redox potential. tion with ethyl alcohol was done. The results, for
As observed in Table 4, the ferrous ion each of the acid concentrations, are shown in Fig-
despite always being in the majority decreases in ures 11, 12 and 13.
proportion to the ferric ion as a function of the in- For 5% H 2 SO 4 the crystalline compound ob-
creasing concentration of sulfuric acid (given as tained was Melanterite (FeSO 4 ⋅7H 2 O), Figure 11. At
H 2 SO 4 molality), Figure 10. These values fit well 10% H 2 SO 4 , in addition to Melanterite, other two
the values from the thermodynamic simulation re- phases can be identified: Szomolnokite
sults (Fe3+ was taken as the sum of the values of Fe3+ (FeSO 4 ⋅H 2 O) and Rozenite (FeSO 4 ⋅4H 2 O), Figure
and FeSO 4 +); under strongly anoxic coditions (log 12. Under the concentration of 15% H 2 SO 4 the pre-
(p O2 ) = -50) the ratio shows values that greatly fa- dominant crystalline compound is again Melanter-
vors the formation of ferrous ion. ite, Figure 13. According to the literature (12) the
presence of Szomolnokite and Rozenite may be
referred to a certain degree of superheat of the
preagnant solution.

1800

d=4,893
1700

1600

1500

1400 Melanterite
1300

1200

1100

1000

900

IJSER
d=5,468 d=5,282
800

700

600

d=2,471 d=2,447
500

d=3,775 d=3,714
Gypsum

d=1,487
400

d=2,619
d=4,515
d=5,571
300

d=1,798
d=7,587

d=2,748
d=6,763

d=3,249

d=2,931

d=2,258

d=1,954
d=4,020

d=3,117

d=1,824
d=2,077

d=1,542
d=3,379

d=1,884
2+ 3+
Fig. 10 Mass ratio of the iron ions Fe and Fe (the 200

100
3+
latter taken as the sum of the values of Fe and 0
+
FeSO4 ) as a function of H2SO4 molality for three O2 2 10 20 30 40 50 60 70

2-theta
pressures; black squares = experimental values.
Fig. 11 X-ray diffractogram of the sample of ferrous sulfate obtained
In addition to the ferrous sulfate crystals, the by leaching with 5% H2SO4.
aqueous phase can also be seen as co-product of the
Fe (II) sulfate process, and consists basically of a
ferric sulfate solution. Ferric sulfate does not crys-
tallize under the experimental conditions; this was
confirmed using the thermodynamic tool.
The remaining ferric sulfate solution can be
employed, for example, as a coagulant in the treat-
ment of industrial effluents. Upon completion of
the tests, the amounts of dissolved mill scale and
crystallized ferrous sulfate were determined, Table
5.
In order to characterize and identify the de-
gree of hydration of crystallized ferrous sulfate, an
TABLE 5
AMOUNTS OF DISSOLVED SCALE AND CRYSTALLIZED FER-
ROUS SULFATE AS A FUNCTION OF SULFURIC ACID CONCEN-
TRATION.
Fig. 12 X-ray diffractogram of the sample of ferrous sulfate obtained
Amount [g]
by leaching with 10% H2SO4.
H 2 SO 4 5% 10% 15%
Scale dissolved 53.3 72.6 89.0
Crystallized ferrous sulphate 47.7 139.5 197.6

IJSER © 2014
http://www.ijser.org
International Journal of Scientific & Engineering Research, Volume 5, Issue 4, April-2014 359
ISSN 2229-5518
[7] Metcalf & Eddy, 2003. Wastewater Engineering: Treatment and
Reuse. 4th Ed., MGraw-Hill Book Company, New York.
[8] 8 ASSOCIAÇÃO BRASILEIRA DE NORMAS TÉCNICAS.
NBR 8577. Minério de Ferro – Determinação do Teor de Ferro
Total por Dicromatometria, 8 pg, Rio de Janeiro, 1984,
confirmada em 2004.
[9] VOGEL’s, Textbook of Quantitative Chemical Analysis. Fifth
Edition. Longman, Scientific e Technical. New York.
[10] PERRY, R.H.; CHILTON, C.H. Manual de Engenharia Química,
Editora Guanabara Dois, 5ª. Ed., Rio de Janeiro, 1980, pp 3-16.
[11] ROJAHN, C.A. Productos Químicos y Farmacêuticos. Vol I, pp
240, Editorial Atlante, México, 1956.
[12] CALVET E. Química General Aplicada a La Industria com
Practicas de Laboratorio. Tomo II, pp 624-625, 2ª. Ed. Salvat
Editores S.A. Barcelona – Buenos Aires, 1944.

Fig. 13 X-ray diffractogram of the sample of ferrous sulfate obtained

by leaching with 15% H2SO4.

5 CONCLUSIONS
The results from both experimental and theoretical
analyses suggest that Iron (II) sulfate can be produced by the
process of mill scale leaching with solutions of sulfuric acid,

IJSER
without the need for special techniques, complex installa-
tions or great maintenance.

However, for effective industrial application, com-

plementary studies on a pilot scale are necessary, in order to
verify both the technical and economic feasibility of the sug-
gested process.

REFERENCES
[1] DA CUNHA, A.F. et. al. Aspectos Técnicos da Utilização da
Carepa Gerada em Processos Siderúrgicos e Tratada por
Desagregação Ultrassônica. In: Seminário de Fusão, Refino e
Solidificação dos Metais, São Paulo, ABM – Associação
Brasileira de Metalurgia e Materiais, 2005a.
[2] TAKANO, C.; DEODORO, J. T. C.; NASCIMENTO, R. C.;
MOURÃO, M. B.; LENZ, G.; MARTINS, D. S. A reciclagem de
resíduos siderúrgicos sólidos. In: Seminário Nacional sobre
Reuso/reciclagem de Resíduos Sólidos Industriais, 2000, São
Paulo. São Paulo: Departamento de Engenharia Metalúrgica e
de Materiais, Escola Politécnica da USP, 2000.
[3] BUZIN, P. J. W. K. Desenvolvimento de briquetes
autorredutores a partir de carepas de processamento
siderúrgico para a utilização em forno elétrico a arco. Dis-
sertação de mestrado, PPGEM, UFRGS, 136 p., 2009.
[4] MARTINS, G.; ALMEIDA, A. F. Reuso e Reciclo de Águas em
Indústria Química de Processamento Dióxido De Titânio.
Monografia, Salvador (BA), Universidade Federal da Bahia,
Escola politécnica. 69p, 1999.
[5] VIGÂNICO, E.M. Produção de Sulfato Ferroso a Partir de
Rejeitos de Carvão. Dissertação de Mestrado. PPGEM: UFRGS,
2009.
[6] EPA, (EPA-625/2-78-017), Environmental Protection Agency. In-
dustrial Environmental Research Laboratory. Recovery of spent
sulfuric acid from steel pickling operations; capsule report,
United States, Washington, 1978.

IJSER © 2014
http://www.ijser.org