Fuel 280 (2020) 118648

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

The effect of sulfur content of rotary kiln fuel on the composition of nickel T
laterite calcine
⁎
Sahand Sarbishei , Leili Tafaghodi Khajavi
The University of British Columbia, 309-6350 Stores Road, Vancouver, British Columbia V6T 1Z4, Canada

G R A P H I C A L A B S T R A C T

A R T I C LE I N FO A B S T R A C T

Keywords: Rotary kiln-electric furnace (RK-EF) process is the main pyrometallurgical route for ferronickel production from
Coal nickel laterite ores. Coal is used as a common fuel for calcination of nickel laterites in rotary kiln. Due to the
Combustion depletion of high-quality coal with low impurity content, high-sulfur coal has been used as fuel in nickel laterite
Fuel rotary kilns. This investigation examines the behavior of sulfur present in the rotary kiln fuel in the calcination
Calcination
stage. Using coal with higher sulfur content, improving the reducing atmosphere in the furnace, and increasing
Sulfur deportment
the fuel consumption result in an increase in sulfur content of the nickel laterite calcine. Higher sulfur de-
Thermodynamics
portation from the gas phase to the calcine was observed by raising the temperature from 600 to 800 °C;
however, increasing the temperature to 900 °C decreases the sulfur deportation percentage due to the re-
crystallization of the silicate compounds. Iron sulfide was identified as the main sulfur-containing compound in
the calcine.

1. Introduction
Coal is commonly used in high temperature extraction of various
metals and alloys. Nickel is a critical strategic metal that is mainly
consumed in stainless steel, superalloys, non-ferrous alloys, high-tem-
perature alloys, and batteries [1–3]. Nickel resources are divided into
two types of ores; sulfide, and laterite [4]. With the gradual decrease in
the availability of sulfides, and high demand for nickel, laterite deposits
have become the main source of nickel [5]. Nickel production from
laterite ores reached 60% in 2013 [6,7]. Laterite profile includes three
main layers: limonite, smectite, and saprolite. Saprolite ores are mostly
processed to produce ferronickel alloy in order to be used in stainless
steel or other ferroalloys.
The rotary kiln-electric furnace (RK-EF) process is the predominant
route for ferronickel production from saprolitic ores. This process in-
volves drying the ore in a rotary drier, calcination and partial reduction
in fuel-fired rotary kiln, smelting in electric furnace, and refining in
ladle furnace [4]. The final impurity content of ferronickel is affected
by calcination step [8–10]. Calcination of the ore is performed in the
rotary kiln at 800–1000 °C using different types of fuel, such as coal, oil,

⁎
Corresponding author.
E-mail address: sarbishei.s@alumni.ubc.ca (S. Sarbishei).

https://doi.org/10.1016/j.fuel.2020.118648
Received 18 May 2020; Received in revised form 2 July 2020; Accepted 5 July 2020
0016-2361/ © 2020 Elsevier Ltd. All rights reserved.
S. Sarbishei and L. Tafaghodi Khajavi Fuel 280 (2020) 118648

Table 1
Chemical composition of the laterite ore analyzed by XRF.
Component SiO2 Fe2O3 MgO Al2O3 NiO Cr2O3 MnO Na2O TiO2 CaO Co S LOI

Weight % 36.7 34.6 8.74 5.88 2.38 1.62 0.58 0.12 0.1 0.03 0.12 0.02 Balance

Fig. 1. XRD pattern of the raw laterite ore.

Fig. 2. Schematic of the experimental setup used for the calcination tests.

and natural gas [4].
Coal is considered as a primary and inexpensive energy resource in
metals and minerals industries. It is an abundant and accessible fuel
which is extensively used in pyrometallurgical processes, such as iron,
steel, cement, and nickel production [11–13]. Calcination is an energy
intensive step in the pyrometallurgical extraction of nickel where the
required energy is provided by coal combustion [4,14]. Sulfur is a
common impurity in coal with its content in the range of 0.2 to 10%
[15,16]. During coal combustion, sulfur is oxidized to SO2 gas
[15,17–21]. It is hypothesized that SO2 resulted from coal combustion
in the rotary kiln can react with the ore and results in sulfur transfer to
the product of the calcination stage, i.e. the calcine. Since nickel laterite
ores contain negligible sulfur content, it is believed that the sulfur
present in the calcine is mostly introduced as a result of fuel combustion
in the rotary kiln [22]. Sulfur is a harmful impurity that deteriorates
mechanical properties of nickel alloys [23]. Therefore, it is crucial to
investigate the effect of the sulfur content of coal on the calcine com-
position.
Limited investigations have been performed on the behavior of
sulfur during calcination and partial reduction of the oxide ores. The

2
S. Sarbishei and L. Tafaghodi Khajavi Fuel 280 (2020) 118648

Table 2
Experimental conditions for processing the samples.
Sample code CO CO2 SO2 N2 Total Temperature(°C) %S in
(ml/ (ml/ (ml/ (ml/ gas the
min) min) min) min) flow burning
rate coal
(ml/ Fig. 4. The sequence of the sulfur deportment reactions to form iron sulfide.
min)

S1 23.7 5.9 0 27.7 57.3 900 0 2. Experiments

S2 23.7 5.9 0.4 27.3 57.3 900 2
S3 23.7 5.9 0.9 26.8 57.3 900 5 2.1. Materials
S4 23.7 5.9 1.8 25.9 57.3 900 10
S5 23.7 5.9 2.8 24.9 57.3 900 15
C1 9.9 19.7 0.9 26.8 57.3 900 5
The chemical composition of the nickel laterite ore used as the raw
C2 19.7 9.9 0.9 26.8 57.3 900 5 material for the calcination experiments was determined by X-ray
C3 23.7 5.9 0.9 26.8 57.3 900 5 fluorescence (XRF) (Table 1). The ore contains various oxides, including
C4 26.3 3.3 0.9 26.8 57.3 900 5 SiO2, Fe2O3, MgO, Al2O3, and NiO. The compounds present in the raw
C5 27 2.6 0.9 26.8 57.3 900 5
laterite ore were identified by X-ray diffraction (XRD). Based on the
F1 11.8 2.9 0.4 13.4 28.5 900 5
F2 23.7 5.9 0.9 26.8 57.3 900 5 XRD results presented in Fig. 1, the main crystalline phases of the raw
F3 47.4 11.8 1.8 53.6 114.6 900 5 ore are identified as lizardite ((Mg0.947Fe0.053)3Si2O5(OH)4), nontronite
F4 71.1 17.7 2.7 80.4 171.9 900 5 (Na0.3Fe2(Si,Al)4O10(OH)2.n(H2O)), trevorite (NiFe2O4), hematite
F5 94.8 23.6 3.6 107.2 229.2 900 5 (Fe2O3), magnetite (Fe3O4), wustite (FeO), goethite ((Fe,Ni)OOH), and
T1 94.8 23.6 3.6 107.2 229.2 600 5
T2 94.8 23.6 3.6 107.2 229.2 700 5
silicon oxide (SiO2). Due to the presence of both lizardite and non-
T3 94.8 23.6 3.6 107.2 229.2 800 5 tronite, the ore was identified as a mixture of smectite and saprolite.
T4 94.8 23.6 3.6 107.2 229.2 900 5 Prior to the experiments, the ore was dried at 130 °C for 2 h to remove
the free water. The dried ore was mixed with 5 wt% graphite as the
reducing agent.
most relevant one to this study is the removal of iron from bauxite ore
through a sulfurization process [24]. Bolsaitis and Nagata [24] in- 2.2. Apparatus and procedure
vestigated the kinetics of Fe2O3 sulfurization in CO-SO2 gas mixtures
using thermogravimetric analysis (TGA). According to their results, S2 The calcination experiments were carried out in a horizontal tube
and COS are the main sulfur-containing gaseous species in the sulfur- furnace (RSRB 80–750/11, Nabertherm, Lilienthal, Germany). In order
ization reactions. FeS was found to be the main sulfur-containing to mimic the atmosphere resulted from the coal oxidation in the rotary
compound generated as a result of hematite sulfurization [24]. In a kiln, various mixtures of N2, CO, CO2, and SO2 gases were purged into
similar study, iron oxide was removed from the laterite ore by sulfur- the furnace by controlling the flow rate of each gaseous component
ization and chlorination [25]. It should be noted that the sulfurization using mass flow controllers (MC and MCS series, Alicat Scientific,
atmosphere of all the previous studies is considerably different from the Tucson, USA). A gas mixing column was placed after mass flow con-
atmosphere generated by fuel combustion in nickel laterites calcina- trollers to provide a homogenous gas mixture for the experiments.
tion. Particularly, the partial pressure of SO2 in the furnace is sig- Fig. 2 shows the schematic of the experimental setup, including gas
nificantly lower than the aforementioned studies. cylinders, mass flow controllers, mixing column, horizontal tube fur-
Considering the negligible amount of sulfur in lateritic ores, the nace, and extraction arm.
sulfur content in the crude ferronickel produced in the EF is expected to The feed samples included 30 g of the laterite ore and 1.5 g graphite
be originated from the process additives such as rotary kiln fuel [26]. as a reducing agent. Graphite with negligible sulfur content was se-
This study investigates the behavior of sulfur, originated from coal lected as a reducing agent in order to minimize the effect of the sulfur
combustion, in calcination of nickel laterites. The effect of the partial content of the reducing agent on the sulfur deportment experiments.
pressure of SO2, the intensity of the reducing atmosphere, the gas flow The samples were heated to the desired temperature in the range of
rate, and the temperature on the composition of the calcine were in- 600–900 °C with a heating rate of 450 °C in a neutral atmosphere. The
vestigated by thermodynamic analysis and high temperature experi- samples were held at the selected temperature for 1 h under the desired
ments. The sulfurization reactions and the main sulfur-containing mixture of N2, CO, CO2, and SO2 followed by cooling down to room
compound of the calcine were identified by thermodynamic simulation, temperature in an inert atmosphere The effects of the Pso2 in the gas
X-ray diffractometer (XRD), and scanning electron microscope and mixture, the reducing atmosphere, gas flow rate, and temperature on
energy dispersive spectroscopy (SEM-EDS) analyses. the composition of the calcine were investigated. Table 2 shows the

Fig. 3. Equilibrium composition of the main sulfur-containing compounds in the (a) calcine, (b) gases reacted with solids, and (c) input gas mixture without reaction
with solids.

3
S. Sarbishei and L. Tafaghodi Khajavi Fuel 280 (2020) 118648

Table 3
List of the probable reactions associated with sulfur deportment in the system including laterite ore and N2-CO-CO2-SO2 gas mixture at
900 °C.
Reaction Formation of intermediate gas species ΔG°900°C (KJ/mol SO2)

1 3CO (g) + SO2 (g) = COS (g) + 2CO2 (g) −86.9

2 2CO (g) + SO2 (g) = 1/2S2 (g) + 2CO2 (g) −84.7
Reaction Reaction with COS ΔG°900°C (KJ/mol COS)
3 1/2Fe2O3 + 1/2CO (g) + COS (g) = FeS + 3/2CO2 (g) −98.5
4 1/3Fe3O4 + 1/3CO (g) + COS (g) = FeS + 4/3CO2 (g) −79.6
5 FeO + COS (g) = FeS + CO2 (g) −78.8
6 Fe + COS (g) = FeS + CO (g) −85.8
Reaction Reaction with S2 ΔG°900°C (KJ/mol S2)
7 Fe2O3 + 3CO (g) + S2 (g) = 2FeS + 3CO2 (g) −201.3
8 2/3Fe3O4 + 8/3CO (g) + S2 (g) = 2FeS + 8/3CO2 (g) −163.4
9 2FeO + 2CO (g) + S2 (g) = 2FeS + 2CO2 (g) −161.8
10 2Fe + S2 (g) = 2FeS −175.9
Reaction Overall sulfurization reactions ΔG°900°C (KJ/mol SO2)
11 1/2 Fe2O3 + 7/2CO (g) + SO2 (g) = FeS + 7/2CO2 (g) −185.4
12 1/3 Fe3O4 + 10/3CO (g) + SO2 (g) = FeS + 10/3CO2 (g) −166.5
13 FeO + 3CO (g) + SO2 (g) = FeS + 3CO2 (g) −165.6
14 Fe + 2CO (g) + SO2 (g) = FeS + 2CO2 (g) −172.7

experimental conditions for the calcination experiments.
The initial gas flow rate was selected based on the molar ratio of
carbon in coal to the sum of nickel and iron oxides in the ore, i.e. C/
(Fe2O3 + NiO). For example, the total gas flow rates of 57.3 and
229.2 ml/min correspond to C/(Fe2O3 + NiO) of 1 and 4, respectively.
The total gas flow rate corresponds to the amount of coal used as a fuel
in the rotary kiln. The values of the constant parameters in the ex-
perimental series were selected based on the similarity with the op-
erational parameters of industrial rotary kilns.
The effects of the sulfur content of coal and the complete vs. in-
complete combustion are investigated in S and C sample series. Higher
SO2 flow rates imply higher sulfur content of the coal, while higher CO/
lower CO2 flow rates indicate incomplete coal combustion and a
stronger reducing atmosphere. The effects of the process temperature
and total gas flow rate are investigated in T and F sample series, re-
spectively. In order to perform a fundamental study and isolate the
effect of a target parameter on sulfur behavior, all other experimental
parameters were maintained constant.
2.3. Characterization methods
Inductively coupled plasma optical emission spectrometry (ICP-
OES, Varian 725-ES ICP spectrometer, Agilent Technologies, Santa
Clara, USA) was used to measure the sulfur content of the samples with
a typical detection limit in the range of 1–10 ppb [27]. The samples
were fused with KOH to make it soluble in acid. 0.2 g of the ground
calcine was mixed with 3 g of KOH pellets. The mixture was charged in
a zirconium crucible and placed in a muffle furnace at 450 °C for 2 h.
The fused sample was leached in 35 ml dilute nitric acid. Finally, the
sulfur content of the sample was obtained by ICP-OES analysis.
Sulfurization behavior was also investigated by calculating the
percentage of sulfur absorbed from SO2 gas to the calcine. Sulfur de-
portation is defined as:
Mass of sulfur in the calcine
% Sulfur deportation = × 100
Total mass of input sulfur
The phases present in the calcine samples were identified using XRD
(MultiFlex, Rigaku Corporation, Tokyo, Japan) using a Cu Kα radiation
(λ = 1.54056 °A). SEM-EDS (Quanta 650, FEI Company, Oregon, USA)
was used to investigate the elemental composition of the calcine.
3. Thermodynamic analysis
The equilibrium module of HSC Chemistry 5.11 was utilized to
obtain the equilibrium composition of the system [28]. 135 possible
species, including 50 sulfur-containing compounds, were considered in
the thermodynamic investigation. The analysis was performed to pre-
dict the equilibrium composition of the calcine, the main sulfur-con-
taining compounds, and the sulfur content of the calcine in various
experimental conditions.
It should be indicated that the results obtained by HSC simulation
are based on pure thermodynamic analysis and does not account for
kinetics considerations, since the module used, only minimizes Gibbs
free energy of the system. This thermodynamic evaluation also does not
consider the continuous supply of the gas mixture from outside the
furnace and the removal of the gaseous compounds from the system.
3.1. Sulfur deportment
Thermodynamic analysis of sulfur distribution was conducted for
sample S3 as a representative experimental condition. The equilibrium
concentration of the main sulfur-containing compounds of the system in
the range of 600–900 °C is presented in Fig. 3 (a,b). Iron sulfide is the
dominant sulfur-containing compound in the calcine, and COS is vir-
tually the only sulfur-containing gas in the gas mixture.
In order to find the sequence of the sulfurization reactions, equili-
brium composition of the sulfur-containing gases prior to reaction with
solids is shown in Fig. 3 (c). COS and S2 are the main gaseous com-
pounds that transfer sulfur from gas to the solid phase. Considering the
original sulfur-containing compounds presented in Fig. 3 (a,b), and the
equilibrium composition of the sulfur-containing gases of the input gas
mixture (Fig. 3 (c)), it is proposed that SO2 and CO react with each
other to form COS and S2 which subsequently react with iron-con-
taining components in the ore and form iron sulfide. The proposed
mechanism is clarified in Fig. 4.
A list of the main possible reactions in the system is presented in
Table 3. Intermediate gas species are formed by the reactions between
SO2 and CO gases (reactions 1 and 2). Then, COS reacts with iron and
iron oxides in the ore to form iron sulfides (reactions 3–6). S2 is another
(1) gaseous species that can react with iron-containing compounds of the
ore (reactions 7–10). Considering the results presented in Fig. 3 (c),
COS is believed to be the dominant gas for sulfurization reactions in the
range of 600–900 °C. The overall sulfurization reactions are obtained
from the sum of the reactions that correspond to S2 or COS formation
and the reactions between sulfur-containing gases and iron-containing
compounds. The overall sulfurization reactions are provided in Table 3
(reactions 11–14).

4
S. Sarbishei and L. Tafaghodi Khajavi Fuel 280 (2020) 118648

Fig. 5. (a) Predicted %S in the calcine versus PSO2, (b) equilibrium content of sulfur compounds versus PSO2, (c) %S in the calcine versus CO/CO2, and (d) versus gas
flow rate, (e) equilibrium content of sulfur compounds versus gas flow rate and (f) %S in the calcine versus temperature.

3.2. The effect of sulfur content of the burning coal 3.3. The effect of the reducing atmosphere (CO/CO2)

The experimental conditions for the tests on the effect of PSO2 which
is equivalent to the higher sulfur content of the coal are shown in
Table 2 (samples S1-S5). The final % S in the calcine presented in Fig. 5
(a) is obtained by thermodynamic modeling with HSC. As a result of
increasing the sulfur content of the burning coal from 2%
(PSO2 = 6.4 × 10-3 atm) to 15% (PSO2 = 4.8 × 10-2 atm), the % S in
calcine increases from 0.1 to 0.77%. As expected from reactions 11–14
in Table 3, higher amount of SO2 results in an increase in the formation
of iron sulfides. The effect of the sulfur content of coal on the equili-
brium content of FeS and Fe0.87S is presented in Fig. 5 (b).
Higher CO/CO2 is associated with incomplete coal oxidation as a
result of oxygen deficient combustion. To investigate the effect of the
reducing atmosphere on the sulfur content of the calcine, %S in the
burning coal was selected as a constant value of 5% which is considered
a reasonable value for high-sulfur coals [17]. CO plays a critical role in
sulfur deportment by reacting with SO2 and generating COS and S2.
Consequently, improving the reducing conditions, i.e., higher CO/CO2,
is expected to promote the formation of sulfur compounds in the cal-
cine. As can be seen in Fig. 5 (c), CO/CO2 and the % S in the calcine are
directly related to each other. It should be noted that the increase in the
equilibrium sulfur content as a result of higher CO/CO2 is not sig-
nificant.

5
S. Sarbishei and L. Tafaghodi Khajavi Fuel 280 (2020) 118648

Fig. 6. X-ray diffraction patterns of samples (a) S1, and (b) S5.

Fig. 7. SEM image of sample (a) F5 and (b) sulfur-rich zone. EDS map analysis of (c) sulfur, (d) iron, (e) oxygen, (f) silicon, and (g) nickel.

3.4. The effect of the total gas flow rate
The gas flow rate is defined with respect to the molar ratio of carbon
in the burning coal to the sum of Fe2O3 and NiO content of the raw ore.
With a constant amount of Fe2O3 + NiO, higher gas flow rate in the
system corresponds to higher coal consumption per unit mass of the
ore. In order to investigate the effect of gas flow rate on the % S in the
calcine, a gas mixture made up of the amount of each gas for a one-hour
experiment was considered as the gaseous input in the HSC model.
Fig. 5 (d) illustrates the % S in the calcine versus gas flow rate.
Increasing the CO + CO2 to Fe2O3 + NiO ratio from 0.5 to 4 results in
an increase in % S from 0.13 to 1.04. As presented in Fig. 5 (e), in-
creasing the gas flow rate results in an increase in the equilibrium
amounts of iron sulfides. It should be emphasized that the amount of
SO2 as one of the input materials in the system increases with increasing
the total gas flow rate. However, as mentioned earlier, the HSC ther-
modynamic analysis does not consider kinetic factors such as con-
tinuous gas supply to the system and the effect of the gas flow rate on
the gas velocity.

6
S. Sarbishei and L. Tafaghodi Khajavi Fuel 280 (2020) 118648

Fig. 8. Additional SEM image of sample (a) F5 and (b) sulfur-rich zone. EDS map analysis of (c) sulfur, (d) iron, (e) oxygen, (f) silicon, and (g) nickel.

Fig. 9. SEM images and EDS point analysis of various sulfur-rich zones in sample F5.

Table 4
EDS analysis of the selected points in sample F5.
EDS point Fe (wt%) S (wt%) O (wt%) Si Al Mn Ni Mg
(wt (wt (wt (wt (wt
%) %) %) %) %)

1 37.5 35.9 12.7 8.2 1.8 0.2 0 3.7

2 44.4 41.5 9.8 1.3 0 0.1 2.9 0
3 39.3 30.7 21.8 2.3 0 0.8 4.9 0.2
4 31.8 30 23.5 5.2 1.2 0.8 6.9 0.6
5 43.3 44.8 8.8 0.5 0 0.1 1.1 1.4
6 42.8 47.7 7.5 1.6 0 0.2 0 0.2
7 39.7 46.5 6.9 3.9 0.5 0.6 0.5 1.4
8 39.6 42.7 9.6 4.5 0.9 0.1 0.3 2.3

3.5. The effect of temperature

Thermodynamic evaluation of sulfur deportment was carried out at

different temperatures with 5% S in the burning coal, CO to CO2 ratio of Fig. 10. The effect of the Pso2 on the % S in the calcine and sulfur deportation
4, and the total gas flow rate of 229.2 ml/min. The value for gas flow percentage at 900 °C, CO/CO2 of 4, and the gas flow rate of 57.3 ml/min.
rate which is equivalent to C/(Fe2O3 + NiO) of 4 is selected considering
coal consumption in three industrial plants. Coal consumption in nickel
laterite rotary kilns in these plants is reported to be in the range of 80 to

7
S. Sarbishei and L. Tafaghodi Khajavi
Fig. 11. XRD patterns for samples (a) S1, (b) S3, and (c) S5.
Fig. 12. The effect of CO/CO2 on the % S in the calcine and sulfur deportation
percentage at 900 °C, 5% S in burning coal (Pso2 = 0.016 atm), and gas flow
rate of 57.3 ml/min.
115 kg/ton [29]. Considering the carbon content of the coal and the
iron and nickel contents of the ore, the ratio of C to Fe2O3 + NiO was
obtained to be between 3 and 4.28 for these industrial plants.
Considering the negative Gibbs free energy values for the sulfur-
ization reactions, it is expected that the majority of sulfur originated
from SO2 transfers to the calcine. Thermodynamic model shows that the
sulfur deportation is 99.9% in the temperature range of 600–900 °C.
8
Fuel 280 (2020) 118648
The predicted sulfur content of the calcine at various temperatures is
shown in Fig. 5 (f). By increasing the temperature from 600 to 900 °C,
the % S in the calcine is slightly increased, i.e., from 0.95 to 1.03%.
Increasing temperature improves the reduction of iron oxide, which in
turn results in a mass loss in the calcine. The slight increase in the sulfur
content of the calcine by raising temperature is attributed to the higher
mass loss due to reduction reactions while the amount of sulfur trans-
ferred from the gas to the calcine is constant.
4. Results and discussion
4.1. Characterization of the sulfur-containing compound
XRD analysis was used to identify the sulfur-containing compounds
in the calcine. Sulfur-rich regions where further characterized using
SEM/EDS analysis.
4.1.1. XRD analysis
Based on the thermodynamic evaluations, iron sulfide is predicted
as the main sulfur-bearing compound in the calcine. Samples S1 and S5,
with PSO2 equal to 0 and 0.049 atm, respectively, were analyzed by XRD
to identify any distinction in the calcine composition due to the pre-
sence of SO2 in the furnace atmosphere. XRD patterns of the two
samples presented in Fig. 6 show hematite (Fe2O3), magnetite (Fe3O4),
wustite (FeO), iron (Fe), ferronickel (FeNi), graphite (C), trevorite
(NiFe2O4), quartz (SiO2), fayalite (Fe2SiO4), enstenite (MgSiO3), for-
sterite (Mg2SiO4), and kyanite (Al2SiO5) as the main crystalline phases
in the calcine. Since the only difference between the two samples is the
presence of SO2 in the furnace atmosphere, the sulfur-containing
S. Sarbishei and L. Tafaghodi Khajavi
Fig. 13. X-ray diffraction patterns of samples C1, C3, and C4 prepared in various reducing atmospheres.
Fig. 14. The effect of the gas flow rate on the % S in the calcine and sulfur
deportation percentage at 900 °C, 5% S in burning coal (Pso2 = 0.016 atm), and
CO/CO2 of 4.
compound can be identified accurately. The main difference between
XRD patterns of the sample including sulfur, S5, and the one without
sulfur, S1, is a peak that corresponds to iron sulfide (Fig. 6). For further
clarifications, XRD patterns in the range of 27 to 33° are magnified in
the upper right part of Fig. 6. The presence of the iron sulfide peak in
sample S5 is confirmed by comparing the two XRD patterns.
4.1.2. SEM-EDS analysis
SEM analysis was done for sample F5 to examine sulfur-rich areas in
9
Fuel 280 (2020) 118648
the sample. Figs. 7 and 8 show SEM images and map analysis of sulfur,
iron, oxygen, silicon, and nickel. The sulfur-rich zones were identified
from the sulfur maps (Fig. 7 (c) and 8 (c)). EDS analysis was done for
various points on the sulfur-rich compound to find the approximate
elemental composition of the sulfur-rich areas in the calcine. Fig. 9
shows the sulfur-rich phase at various points of the sample. The %S in
the sulfur-rich regions is 30–47.7%, which is orders of magnitude
higher than the typical EDS detection limit, i.e., 0.08–0.5% [30–32].
Considering the EDS results presented in Table 4, it is concluded that
sulfur in the sulfur-rich phase is primarily accompanied by iron. This is
in agreement with XRD results and the results of the thermodynamic
analysis.
4.2. The effect of the sulfur content of the burning coal
As shown in Fig. 10, there is a direct relationship between PSO2 in
the atmosphere and the S content of the calcine. By increasing PSO2
from 6.4 × 10-3 to 4.8 × 10-2 atm, sulfur in the calcine is increased
from 0.04 to 0.14%. It is worth reiterating that higher PSO2 represents a
higher concentration of sulfur in the fuel. Due to kinetics limitations,
the % S in the calcine was considerably lower than the values predicted
by thermodynamic analysis.
As can be seen in Fig. 10 increasing Pso2 results in lower sulfur
deportation percentage. Since the total gas flow rate was maintained
constant by changing Pso2, increasing Pso2 leads to higher SO2 flow
rate. The decline in sulfur deportation percentage can be related to the
higher velocity of SO2 which results in its lower residence time in the
furnace [33]. By increasing the velocity of SO2, the sulfur-containing
gas departs faster from the furnace, hence inhibits the conversion of
sulfur from gas to solid species. With increasing PSO2 from 6.4 × 10-3 to
S. Sarbishei and L. Tafaghodi Khajavi
Fig. 15. X-ray diffraction patterns of samples (a) F1, (b) F2, (c) F3, and (d) F5.
Fig. 16. The effect of temperature on the % S in the calcine and sulfur de-
portation percentage at 5% S in burning coal (Pso2 = 0.016 atm), CO/CO2 of 4,
and the gas flow rate of 229.2 ml/min.
4.8 × 10-2 atm, sulfur deportation percentage was decreased from 34.3
to 17.3%. Considering that the sulfur originated from coal is distributed
between the calcine and the off gas, the above observation indicates
that the percentage of sulfur that reports to the off gas increases with
increasing the sulfur content of coal.
The XRD patterns of the samples processed at various PSO2 are de-
monstrated in Fig. 11. The main crystalline phases are described in
section 4–1-1. The iron sulfide peak was observed for the samples S3
10
Fuel 280 (2020) 118648
and S5 which were processed in an atmosphere containing SO2. The
intensity of ferronickel and fayalite peaks are decreased at higher PSO2
due to their reactions with SO2.
4.3. The effect of the reducing atmosphere (CO/CO2)
The effect of the reducing atmosphere on the % S in the calcine and
sulfur deportation are shown in Fig. 12. The aggressive reducing con-
dition leads to higher S content of the calcine and higher sulfur de-
portation percentage. The sulfur content of the calcine increases from
0.056 to 0.082% by increasing CO/CO2 from 0.5 to 10. Under the same
conditions, sulfur deportment percentage rises from 18.95 to 25.12%. It
is worth mentioning that in industrial plants, CO/CO2 in the rotary kiln
is controlled by adjusting the relative amount of fuel and air.
In order to investigate the effect of the reducing atmosphere on the
composition of the calcine, samples C1, C3, and C4 which were pro-
cessed in atmospheres with different CO/CO2, were examined by XRD
analysis. The results are presented in Fig. 13. As expected, increasing
CO/CO2 provides a more aggressive reducing atmosphere that facil-
itates the reduction of iron oxides and nickel oxide. As shown in Fig. 13,
metallic iron and ferronickel are not present in the sample processed in
CO/CO2 = 0.5. However, by improving the reducing condition, the
intensity of fayalite, metallic iron, and ferronickel peaks are increased
while the intensity of magnetite and hematite peaks are decreased. Iron
sulfide as a main sulfur-containing compound is also observed in the
XRD patterns of samples C3 and C4.
S. Sarbishei and L. Tafaghodi Khajavi
Fig. 17. X-ray diffraction patterns of samples prepared at various temperatures, i.e., T1-T4.
4.4. The effect of the total gas flow rate
For the samples processed at various gas flow rates, CO to CO2 ratio
was fixed at 4. This ratio was selected based on observing the desired
reduction behavior in the previous set of experiments presented in
section 4–3. Changing the gas flow rate was performed by varying the
flow rate of each gaseous species while maintaining their partial pres-
sure. Increasing the gas flow rate in these experiments is analogous with
using higher amount of fuel per unit mass of the laterite ore. As dis-
cussed in section 3–4, the gas flow rate is correlated with the molar
ratio of carbon in the coal to the sum of Fe2O3 and NiO content of the
raw ore. The value of 0.5 and 4 for C/(Fe2O3 + NiO) corresponds to
28.5 and 229.2 ml/min total gas flow rate, respectively. Variation of %
S in the calcine as a function of gas flow rate is illustrated in Fig. 14. The
sulfur content in the calcine is continuously increased by raising the gas
flow rate and reaches 0.25% at the gas flow rate of 229.2 ml/min.
The percentage of sulfur deportation to the calcine presented in
Fig. 14 shows a slight decrease by increasing the gas flow rate. This
observation can be explained by lower residence time in the system.
Increasing the gas flow rate results in higher gas velocity and lower gas
residence time [33]. Therefore, the contact time between gas and solid
is decreased which hinders the kinetics of the sulfurization reactions.
X-ray diffraction patterns of the samples processed in different gas
flow rates are shown in Fig. 15. The intensity of ferronickel and fayalite
peaks is increased by purging a higher flow rate of gases. This indicates
a higher degree of reduction at higher gas flow rates. Iron sulfide peak
is visible in samples F2, F3, and F5.
4.5. The effect of temperature
The variation of the % S in the calcine and sulfur deportation
11
Fuel 280 (2020) 118648
percentage with changing the temperature are presented in Fig. 16.
Increasing the temperature from 600 to 800 °C resulted in higher %S in
the calcine, however increasing the temperature to 900 °C, results in a
decrease in % S in the calcine. The same behavior was observed for
sulfur deportation. Due to the unexpected observation for the %S in the
calcine and sulfur deportation, ICP analysis for each sample was re-
peated twice. According to both thermodynamics and kinetics con-
siderations, increasing temperature is expected to yield higher %S in
the calcine. The decrease in %S at 900 °C is attributed to the phase
transformation. Recrystallization of enstenite (MgSiO3) and forsterite
(Mg2SiO4) which occur at 800–820 °C result in a significant decrease in
the surface area of the solid [10,34,35]. The recrystallization me-
chanism is initiated by prenucleation step in which amorphous grains
start to undergo a structural change. Short-range order is changed to
long-range crystallographic order at the surface. Subsequently, crys-
tallization is occurred at the surface and progressed to the center of the
grain [36]. The smaller surface area of the solid leads to the slower
reaction kinetics between the gas and the solid, which in turn hinders
sulfur transfer from the gas to the calcine.
XRD spectra of the samples calcined at different temperatures are
presented in Fig. 17. Increasing temperature improves the reduction
reactions and hence the formation of ferronickel and fayalite. It is im-
portant to note that when nickel oxide is incorporated in the crystalline
forsterite phase, nickel oxide will be locked up in olivine structure. This
will hinder the reduction of nickel oxide [10,37,38]. Considering the
results presented in Fig. 17 both enstatite and forsterite are formed
during calcination. Formation of crystalline magnesia silicate phases at
high temperature is related to MO to SiO2 ratio in the raw ore where
MO is the divalent metal oxides in the ore [39]. MO/SiO2 of higher than
1.5 implies the formation of forsterite. However, MO/SiO2 of the ore
used in this study is 0.4, which indicates the formation of higher
S. Sarbishei and L. Tafaghodi Khajavi
amounts of enstatite compared with forsterite. Since the reducibility of
nickel oxide in enstatite structure is easier than forsterite [37,40], in-
creasing the temperature from 800 to 900 °C is not expected to sig-
nificantly affect the reducibility of nickel oxide. This is in agreement
with the results of the XRD spectra for samples T3 and T4.
It is worth mentioning that the percentage of sulfur deportation
calculated from the experimental results and the ones from thermo-
dynamic simulations are in the range of 17.26–34.35% and
99.93–99.99%, respectively. This variance is caused by the difference
between non-equilibrium experiments and the equilibrium thermo-
dynamic analysis.
5. Conclusion
Considering the thermodynamic evaluations combined with XRD
and SEM-EDS analysis, iron sulfide was found to be the main sulfur
bearing compound in the calcine. The effects of various parameters
such as the Pso2 resulted from coal combustion, the reducing atmo-
sphere in the furnace, gas flow rate, and temperature on sulfurization
behavior and composition of the calcine were investigated. It is con-
cluded that increasing the sulfur content of the fuel, improving the
reducing atmosphere, and increasing the fuel consumption result in
higher sulfur content in the calcine. By increasing the fuel consumption,
the percentage of sulfur deportation was slightly decreased because of
the shorter residence time of the gas in the furnace. Increasing tem-
perature from 600 to 800 °C leads to higher %sulfur in the calcine and
increases sulfur deportation from the gas phase to the solid; however,
raising the temperature above 800 °C decreases the sulfur deportation
percentage due to the recrystallization of the silicate compounds at
800–820 °C.
CRediT authorship contribution statement
Sahand Sarbishei: Conceptualization, Methodology, Validation,
Formal analysis, Investigation, Writing - original draft, Visualization.
Leili Tafaghodi Khajavi: Supervision, Resources, Conceptualization,
Methodology, Writing - review & editing, Project administration,
Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influ-
ence the work reported in this paper.
References
[1] Pollock TM, Tin S. Nickel-based superalloys for advanced turbine engines: chem-
istry, microstructure and properties. J Propul Power 2006;22(2):361–74.
[2] Hasanbeigi A, Arens M, Price L. Alternative emerging ironmaking technologies for
energy-efficiency and carbon dioxide emissions reduction: a technical review.
Renew Sustain Energy Rev 2014;33:645–58.
[3] R.E. Kirk D.F. Othmer Encyclopedia of Chemical Technology Vol. 2 1953 Inc; New
York The Interscience Encyclopedia.
[4] Crundwell F, Moats M, Ramachandran V, Robinson T, Davenport WG. Extractive
metallurgy of nickel, cobalt and platinum group metals. Elsevier; 2011.
[5] Mudd GM, Jowitt SM. A detailed assessment of global nickel resource trends and
endowments. Econ Geol 2014;109(7):1813–41.
[6] Dalvi, A.D. Bacon, W.G. Osborne, R.C. The past and the future of nickel laterites,
PDAC 2004 International Convention, Trade Show & Investors Exchange, Toronto,
Canada: The prospectors and Developers Association of Canada; 2004, pp. 1-27.
[7] Swinbourne D. Understanding ferronickel smelting from laterites through compu-
tational thermodynamics modelling. Mineral Process Extract Metall
2014;123(3):127–40.
[8] O'connor F, Cheung W, Valix M. Reduction roasting of limonite ores: effect of de-
hydroxylation. Int J Miner Process 2006;80(2–4):88–99.
[9] Valix M, Cheung W. Study of phase transformation of laterite ores at high
12
Fuel 280 (2020) 118648
temperature. Miner Eng 2002;15(8):607–12.
[10] S. Li, Study of nickeliferrous laterite reduction, PhD Thesis; 1999.
[11] Cheng X, Cheng S, Liu K, Zhou D. Measurement study of preheated pulverized coal
injection on combustion in a blast furnace raceway by visual detection. Fuel
2020;271:117626.
[12] Zhong Q, Zhang J, Yang Y, Li Q, Xu B, Jiang T. Thermal behavior of coal used in
rotary kiln and its combustion intensification. Energies 2018;11(5):1055.
[13] Hou LY, Fu WB, Zhang YJ. A theoretical analysis on combustion intensification for
blended coal in rotary cement kiln. Fuel 2001;80(11):1645–50.
[14] Pamungkas BC, Hadi H. Study of rotary kiln off-gas energy recovery for coal mill
hot gas. Surabaya, Indonesia: Publishing; 2019. p. 030010.
[15] Billings P. Emission of hazardous air pollutants from coal-fired power plants.
Environ Health & Eng 2007.
[16] Bolat E, Saǧlam S, Pişkin S. Chemical demineralization of a Turkish high ash bi-
tuminous coal. Fuel Process Technol 1998;57(2):93–9.
[17] Chou C-L. Sulfur in coals: A review of geochemistry and origins. Int J Coal Geol
2012;100:1–13.
[18] Hu Y, Naito S, Kobayashi N, Hasatani M. CO2, NOx and SO2 emissions from the
combustion of coal with high oxygen concentration gases. Fuel
2000;79(15):1925–32.
[19] Zhang S, Jiang X, Lv G, Liu B, Jin Y, Yan J, et al. HCl and PCDD/Fs emissions during
Co-combustion of bituminous coal and pickling sludge in a drop tube furnace. Fuel
2016;186:91–9.
[20] Wu H, Glarborg P, Frandsen FJ, Dam-Johansen K, Jensen PA, Sander B. Co-com-
bustion of pulverized coal and solid recovered fuel in an entrained flow
reactor–General combustion and ash behaviour. Fuel 2011;90(5):1980–91.
[21] Rokni E, Ren X, Panahi A, Levendis YA. Emissions of SO2, NOx, CO2, and HCl from
Co-firing of coals with raw and torrefied biomass fuels. Fuel 2018;211:363–74.
[22] Ogura T, Kuwayama K, Ono A, Yamada Y. Production of Fe- Ni by the rotary kiln-
electric furnace process at Hyuga Smelter. Int J Miner Process
1987;19(1–4):189–98.
[23] Sidorov V, Rigin V, Min P, Folomeikin YI. Removal of a sulfur impurity from
complex nickel melts in vacuum. Russian Metallurgy (Metally)
2015;2015(11):910–5.
[24] Bolsaitis P, Nagata K. Kinetics of sulfidization of iron oxide with SO 2-CO mixtures
of high sulfur potential. Metall Trans B 1980;11(2):185–97.
[25] Nagata K, Bolsaitis P. Selective removal of iron oxide from laterite by sulphurization
and chlorination. Int J Miner Process 1987;19(1–4):157–72.
[26] Ma A, Mostaghel S, Chattopadhyay K. Review of applicable desulphurization
models for crude ferronickel refining. Mineral Process Extractive Metall
2018;127(1):10–28.
[27] Tyler G., Jobin S. Yvon, ICP-OES, ICP-MS and AAS Techniques Compared, ICP
Optical Emission Spectroscopy Technical Note 5; 1995.
[28] Roine A, Mansikka-Aho J, Björklund P, Kentala J, Lamberg P. HSC Chemistry 5.1
and 6.0 Outokumpu Technology, Chemical Reaction and Equilibrium Software with
Extensive Thermochemical Database and Flowsheet Simulation. Outokumpu
Research Oy 2006.
[29] Warner A, Diaz C, Dalvi A, Mackey P, Tarasov A. JOM world nonferrous smelter
survey, part III: Nickel: Laterite. Jom 2006;58(4):11–20.
[30] Makhlouf ASH, Aliofkhazraei M. Handbook of materials failure analysis with case
studies from the aerospace and automotive industries. Butterworth-Heinemann;
2015.
[31] Kuisma-Kursula P. Accuracy, precision and detection limits of SEM–WDS, SEM–EDS
and PIXE in the multi-elemental analysis of medieval glass. X-Ray Spectrom: An Int
J 2000;29(1):111–8.
[32] Newbury DE, Ritchie NW. Performing elemental microanalysis with high accuracy
and high precision by scanning electron microscopy/silicon drift detector energy-
dispersive X-ray spectrometry (SEM/SDD-EDS). J Mater Sci 2015;50(2):493–518.
[33] Tan L, Shariff A, Lau K, Bustam M. Factors affecting CO2 absorption efficiency in
packed column: a review. J Ind Eng Chem 2012;18(6):1874–83.
[34] Yang J, Zhang G, Ostrovski O, Jahanshahi S. Changes in an Australian laterite ore in
the process of heat treatment. Miner Eng 2013;54:110–5.
[35] Sarbishei S, Tafaghodi Khajavi L. Kinetic analysis on nickel laterite ore calcination
using model-free and model-fitting methods. Miner Eng 2019;136:129–39.
[36] Kamitsuji K, Sato T, Suzuki H, Kaito C. Direct observation of crystallization of
amorphous Mg-bearing silicate grains into Mg (forsterite). Astron Astrophys
2005;436(1):165–9.
[37] Hallet, C. A thermodynamic analysis of the solid state reduction of nickel from
laterite minerals, Proceedings of Nickel-Cobalt 97 International Symposium
Pyrometallurgical Fundamentals and Process Development, Sudbury; 1997, pp.
17-20.
[38] Bunjaku A, Kekkonen M, Pietilä K, Taskinen P. Effect of mineralogy and reducing
agent on reduction of saprolitic nickel ores. Mineral Process Extract Metall
2012;121(3):156–65.
[39] Lopez-Rendon, J.E. Geology, mineralogy and geochemistry of the Cerro Matoso
nickeliferous laterite, Cordoba, Colombia, Colorado State University, Master Thesis;
1986.
[40] Bunjaku A, Kekkonen M, Taskinen P. Effect of mineralogy on reducibility of cal-
cined nickel saprolite ore by hydrogen. Mineral Process Extract Metall
2012;121(1):16–22.