JOM

https://doi.org/10.1007/s11837-019-03464-1
Ó 2019 The Author(s)

TECHNICAL ARTICLE

Catalytic Role of Process Dust in SO2-to-SO3 Conversion

in Flash Smelting Heat Recovery Boilers

JUHO LEHMUSTO,1,3 TOR LAURÉN,1 and MARI LINDGREN2

1.—Laboratory of Inorganic Chemistry, Johan Gadolin Process Chemistry Centre, Abo Akademi
University, Piispankatu 8, 20500 Turku, Finland. 2.—Outotec Research Center, Kuparitie 10,
P.O. Box 69, 28101 Pori, Finland. 3.—e-mail: [email protected]

The aim of this work is to examine the catalytic role of copper smelter, copper
converter, and nickel smelter process dusts in SO2-to-SO3 conversion at 750°C.
To clarify the role of specific oxides in greater detail, synthetic dusts con-
taining varying concentrations of either CuO or Fe3O4 were also studied. All
studied dusts catalyzed SO3 conversion. Among the industrial flue dusts, the
highest concentration of SO3 formed in the presence of copper smelter dust.
Comparing individual oxides, CuO had a greater impact on the SO3 formation.
Such catalytic effects of smelter dusts may lead to SO3 concentrations in the
process gas so high that sulfuric acid dew point corrosion may occur.

the dew point temperature of sulfuric acid, the risk

INTRODUCTION
of corrosion becomes more pronounced in the later
The flash smelting process is one of the most sections of the plant, where temperatures may be
widely used methods to extract primary copper1 and very close to the dew point of sulfuric acid. The
nickel2 from sulfide ores using smelting and refining cyclic temperature profile of equipment together
processes. The process is very cost-effective and with possible air leaks can further enhance damage
environmentally sustainable because, once initi- originating from such sulfuric acid dew point corro-
ated, it is continuous and nearly autothermal.3 sion.8 Partly sintered process dust particles may
Smelting of sulfide ores releases large quantities adhere to heat-transfer surfaces in the heat recov-
(20–70 vol.%) of sulfur dioxide (SO2) into the pro- ery area, resulting in deposits that decrease the
cess gas, which is first cleaned then passed to a heat-transfer efficiency and block the gas flow
sulfuric acid plant for use as raw material.4 Under path.9 Such deposits can also act as initiation points
suitable conditions, part of this SO2 can be oxidized for corrosion, if the temperature inside a deposit
to sulfur trioxide (SO3) (Eq. 1), which can react drops below the dew point temperature of H2SO4,
further to sulfuric acid (H2SO4) due to the presence enabling its condensation and resulting in corrosion
of humidity in the process gas, through a reaction and severe material degradation.10,11 In fact, strong
involving either an adduct (Eq. 2)5,6 or a water indications of such sulfuric acid-induced corrosion
dimer (Eq. 3).7 were observed in full-scale measurements carried
out during operation of a copper flash smelter
2SO2 ðgÞ þ O2 ðgÞ ! 2SO3 ðgÞ ð1Þ plant.11
In a previous study, the effect of the process gas
SO3  H2 OðgÞ þ H2 OðgÞ ! H2 SO4 ðgÞ þ H2 OðgÞ ð2Þ temperature on the catalytic SO2-to-SO3 conversion
was addressed in the temperature range of 275–
SO3 þ H2 OðgÞ  H2 OðgÞ ! H2 SO4 ðgÞ þ H2 OðgÞ ð3Þ 900°C,12 revealing that the temperature and pres-
ence of process dust had a remarkable effect on the
Formation of H2SO4 in the heat recovery boiler is
SO3 formation. Among the tested temperatures, the
undesirable for the following reasons: (1) it lowers
concentration of formed SO3 peaked at 750°C, which
the overall recovery of sulfur, (2) once formed, it is
was thus chosen as the temperature in the current
difficult to process into saleable products, and (3) it
work. Thermodynamically, SO2-to-SO3 conversion
may cause corrosion of heat-transfer surfaces. Since
is reported to show a maximum at around 500°C in
the concentration of SO3 in the process gas affects
the absence of catalytic species.13 However, the
 Lehmusto, Laurén, and Lindgren

presence of a suitable catalyst such as fly ash affects The experimental setup consisted of a narrow
the kinetics and thermodynamics of Eq. 1, shifting quartz glass reactor with inner diameter of around
this conversion maximum to about 700°C. 8 mm, positioned inside a horizontal tube furnace.
In addition to temperature, various other vari- The sample and a KCl plug were inserted inside the
ables also affect the concentration of SO3 formed, quartz reactor. All samples were studied at temper-
namely the partial pressures of SO2 and O2, the ature of 750°C, whereas the temperature of the KCl
presence and concentration of humidity, the content plug was kept at around 200°C to avoid condensa-
of alkaline, earth-alkaline, and catalytically active tion of possibly formed H2SO4. The temperatures
compounds in the process dust and deposits, the were monitored using thermocouples positioned in
temperature–residence time profile of the plant, and the furnace but outside the reactor to avoid the
the application of gas cleaning equipment.14–19 Iron possible participation of the thermocouples as cat-
oxides, which are typically present in high concen- alysts in the studied SO2-to-SO3 conversion. Due to
trations (12–30 wt.%) in copper flash smelter the inertness of quartz in SO2-to-SO3 conversion,14
dust,11,20,21 have been reported to catalyze SO2-to- quartz wool was used to keep the deposit and KCl
SO3 conversion.14,16,17 Other oxides that are present plug in place. The measurement principle is based
in large quantities in process dust include copper on the ability of the KCl plug to capture SO3 from
oxides such as CuO and Cu2O, silicon oxide (SiO2), the gas that is fed through it. This originates from
and aluminum oxide (Al2O3).22,23 Regarding SO2-to- the higher stability of K2SO4 compared with KCl,25
SO3 conversion, SiO2 has been reported to be inert, meaning that, if any SO3 interacts with KCl, it will
whereas Al2O3 catalyzes the conversion to some be converted to K2SO4. However, KCl is inert to SO2
extent.14 Although Cu2O has been reported to lower at 200°C, so the original process gas does not
the activation energy of SO3 according to Eq. 1,24 interact with the KCl plug. The measurement
very little information is available on the catalytic technique is described in greater detail in Ref. 12.
properties of copper oxides. The density of the studied dusts varied greatly, so
The aim of this work is to study how the chemical instead of having an equal mass, all samples had
composition affects the catalytic properties of dif- equal length of 6.0 cm. Due to the possible blockage
ferent flash smelter process dusts regarding the of the reactor through dust sintering, a narrow gap
SO2-to-SO3 conversion at 750°C. In addition, the was left above the sample. The mass of the KCl
catalytic effect of two oxides, viz. copper(II) oxide (Merck, suprapur 99.999%) plug was 0.5 g. The gas
(CuO) and magnetite (Fe3O4), often found in process mixture mimicking genuine process gas consisted of
dust, was addressed. The results are expected to 50 vol.% N2, 10 vol.% O2, and 40 vol.% SO2. In
shed more light on the role of specific compounds in addition, one exposure for each industrial flue dust
SO3 formation. Better knowledge of factors affecting was also carried out in an SO2-free atmosphere.
SO3 formation in the heat recovery boiler is desired This was done to examine the possible SO3 release
for process optimization and corrosion prevention. from the deposit itself. In the exposures without
SO2, the gas consisted of 90 vol.% N2 and 10 vol.%
EXPERIMENTAL PROCEDURES O2. The sampling time was 5 min. During the
heating, the reactor was flushed with a constant
The catalytic properties of flue dusts were first
N2 flow at the same rate as during the exposures,
studied using industrial flue dusts collected from a
viz. 1 dm3 min1 (RT). The exposures in the absence
copper and a nickel smelting plant, then with syn-
of SO2 in the gas were carried out once, whereas all
thetic process dusts consisting of either copper(II)
the other exposures were carried out twice.
oxide (CuO) or magnetite (Fe3O4). The industrial flue
A scanning electron microscope (LEO 1530 Gem-
dust samples were collected from the heat recovery
ini) coupled to an x-ray detector (Thermo Scientific
boilers of both copper and nickel plants, in addition to
UltraDry Silicon Drift Detector) and an energy-
one dust sample collected from the converter of the
dispersive x-ray analysis system (Thermo Scientific
copper plant. Prior to exposure of the synthetic dusts,
ThermoNORAN Vantage x-ray) was used to esti-
two oxides, silica (SiO2; Merck, pro analysi grade)
mate the particle size of the industrial flue dusts. In
and alumina (Al2O3; Aldrich, 99%), were exposed
addition, the chemical composition of the dust
under the same conditions as the samples. The goal
samples was determined before and after the expo-
was to identify an inert oxide to mix with the studied
sures using the same instrumentation. The micro-
oxides to obtain different concentrations without
scope was operated under high vacuum (around
changing the sample length. Based on the results
106 mbar) at accelerating voltage of 20 kV in
(described in detail below), SiO2 was chosen. The
backscatter electron mode for imaging and EDX
synthetic dusts contained 2 mol.%, 10 mol.%, or
analysis.
25 mol.% of either CuO or Fe3O4, while the rest of
To identify different sulfur-containing compounds
the sample was inert SiO2. The effect of the type of
in the process dusts, both solid and dissolved
iron oxide on the SO2-to-SO3 conversion was studied
samples were analyzed to quantify the concentra-
preliminary, thus two parallel samples containing
tion of sulfur (Eltra CS 2000 induction furnace
10 mol.% hematite (Fe2O3) were exposed to the same
coupled with IR detectors) and SO42 (ion chro-
conditions as the samples containing magnetite.
matograph, IC). In addition, a magnetite analyzer
Catalytic Role of Process Dust in SO2-to-SO3 Conversion in Flash Smelting Heat Recovery
Boilers

(Satmagan 135) capable of quantifying magnetite observed. All samples contained spherical particles,
concentrations in a nonmagnetic medium was used indicating that the particles were partly molten at
to distinguish hematite from magnetite in the some point of the process line.27 If the particles
industrial flue dusts. reach heat-transfer surfaces while in the partly
X-ray powder diffraction (XRPD) was used to molten state, they might adhere to the surfaces,
identify compounds in the dust samples. The XRPD resulting in deposits that decrease the heat-transfer
measurements were performed using a Malvern efficiency and block the gas flow path.9 Based on
Panalytical Empyrean x-ray diffractometer EDX spot analyses, the spherical particles consisted
equipped with a Cu x-ray tube and PIXcel3D mainly of Cu, S, and O, most likely as anhydrous
detector. Scans (2h) were carried out between 10° copper sulfate (CuSO4).28 In addition to CuSO4, the
and 100°. The diffractograms were matched using round-shaped particles contained small concentra-
Malvern Panalytical HighScore Plus software and tions of Fe, Pb, and Zn. The analyzed dust samples
International Centre for Diffraction Data (ICDD) contained altogether more than 15 different ele-
powder diffraction file (PDF)-4+ 2018.26 ments, the most abundant of which are listed in
The KCl plug was dissolved in a known volume of Table II as averages of three EDX analyses. The
deionized water, and the SO42 concentration was copper smelter process dust consisted mainly of Cu,
quantified using an ion chromatography system (IC, Fe, O, and S, the copper smelter converter dust
Metrohm Compact IC Pro with Metrosep anion contained mostly Cu, O, Pb, and S, while the nickel
Dual 2 column and Metrohm 732 IC detector) with smelter process dust contained Fe, Ni, and O. The
2 mM NaHCO2 and 1.3 mM Na2CO3 solutions as presence of lead most likely originated from con-
eluents. Due to the different retention times of Cl densation lead vaporized earlier in the process line.
and SO42 ions, they can be easily distinguished by Other metals with low melting points that can be
ion chromatography. found in converter dust in larger quantities are zinc
and arsenic.22 Although As, Zn, and Pb have not
RESULTS AND DISCUSSION been reported to catalyze SO2-to-SO3 conversion,
they may play a crucial role in deposit formation
In this study, industrial flue dusts collected from
due to the low melting points of their compounds
a copper and a nickel smelter plant were studied
such as As2O3, ZnCl2, and PbCl2, to name a few. The
first to clarify their chemical composition and role in
sulfur content of the nickel smelter process dust was
catalyzing the SO2-to-SO3 conversion. Based on the
clearly lower than those measured for the copper
results of XRPD analysis, SEM, and chemical
smelter dusts. The sulfur contents of the reference
analysis (Table I), CuO and Fe3O4 were chosen to
samples were significantly lower than in the sam-
further specify the impact of these oxides on SO3
ples exposed in the presence of SO2, indicating that
formation in a simplified chemical environment
especially copper smelter dust released sulfur-con-
provided by the synthetic dusts. Following the
taining species at high temperatures. Based on XRD
timeline of the work, the results for the industrial
analyses, the following species were identified with
flue dusts are presented and discussed first in their
high certainty: Cu2SO4, CuO, and PbSO4 in con-
own subsection, followed by another addressing the
verter dust, Cu2SO4, Fe3O4, Fe2O3, and SiO2 in
synthetic dusts.
copper smelter dust, and Ni0.6Fe2.4O4 in nickel
smelter dust. It should be borne in mind that,
Industrial Flue Dusts
considering the actual complexity of the dusts, their
The morphology and particle size of the dusts did compositions are far more diverse than listed here.
not change significantly during the exposure However, identification of tracer species in process
(Figs. 1 and 2). The dust samples exposed in the dusts lies beyond the scope of this work.
absence of SO2 were also imaged, but no differences Putting the focus on iron and copper, both were
in the morphology of the dust particles were found at concentrations of at least 10 wt.% in copper
smelter dust, whereas the converter dust was rich
in copper and the nickel smelter dust in iron
Table I. Concentrations of total sulfur, sulfates,
(Table II). Based on EDX results, the concentration
and magnetite in the studied process dusts of iron and copper oxides cannot be reliably quan-
tified, but the analyzed SO42 concentrations
Sample Stot (wt.%) SO422 (%) Fe3O4 (%) (Table I) indicated that roughly 30% of the sulfur
in the copper smelter dust was present as sulfates,
Cu dust 9.7 28.8 20.9 whereas 4% and 8% of the sulfur was bound in
Ni dust 1.4 4.1 41.1 sulfates in the nickel smelter and converter dust,
Conv. dust 10.5 7.9 0.9
respectively. Regarding iron oxides, roughly 21%,
The analyses were carried out with an ion chromatograph, a 41%, and 1% of the iron were in the form of
magnetite analyzer, and an induction furnace coupled with IR magnetite in the copper smelter, nickel smelter,
detectors. and converter dust, respectively. Other compounds
containing Fe, Cu, and/or O included, for example,
CuS, FeS, PbSO4, and As2O5.20,22,29 Based on these
 Lehmusto, Laurén, and Lindgren

Fig. 1. Industrial flue dusts before exposure: copper smelter converter dust (a), copper smelter process dust (b), and nickel smelter process dust
(c).

Fig. 2. Industrial flue dusts after exposure at 750°C: copper smelter converter dust (a), copper smelter process dust (b), and nickel smelter
process dust (c).
Catalytic Role of Process Dust in SO2-to-SO3 Conversion in Flash Smelting Heat Recovery
Boilers

Table II. Concentrations of main elements in studied process dusts, analyzed with SEM–EDS; values
presented in weight percentages

Cu dust Conv. dust Conv. Conv. Ni dust

as Cu dust Cu dust as dust dust as Ni dust Ni dust
Element received exposed reference received exposed reference received exposed reference
As 3.7 3.1 4.7 3.1 0.9 1.1 3.4 – –
Cu 23.8 24.5 26.3 27.3 55.8 49.4 3.1 2.5 3.0
Fe 11.4 11.6 24.0 1.1 1.7 1.8 35.2 37.6 41.2
Ni – – – – 1.0 0.6 12.1 12.7 13.6
O 38.5 37.6 26.8 37.1 21.8 24.0 32.4 32.5 29.3
Pb – – – 9.7 3.4 4.3 – – –
S 14.8 13.2 1.4 11.6 4.2 1.1 4.9 3.1 2.3
Si 1.7 3.6 6.7 1.2 6.5 11.3 3.7 6.4 4.7
Zn 3.9 4.1 5.7 4.4 2.2 3.2 0.2 0.1 0.2

concentrations of formed SO3 are somewhat higher

compared with the 1–3% reported previously in the
same temperature range,30 but the concentration of
formed SO3 depends, among other factors, on the
SO2 concentration of the process gas, which can
vary greatly depending on the feed, copper produc-
tion method, and location inside the copper process-
ing line. Despite the fact that the nickel smelter
process dust clearly catalyzed SO2-to-SO3 conver-
sion, according to industry, less corrosion-related
problems are detected in nickel flash smelters
compared with copper flash smelters. The reason
for this remains ambiguous, but differences in
process conditions or feed impurities, or a lower
Fig. 3. Concentrations of SO3 formed at 750°C in the reaction deposit build-up rate, may play a role.
catalyzed by industrial flue dusts. The reference samples were In the case of copper smelter dust, there was also
exposed without SO2 in the atmosphere. notable SO3 formation in the absence of SO2 in the
process gas. Since the deposits exposed in the
absence of SO2 contained very little sulfur after
the exposures (< 1.5 wt.%), the formed SO3 most
analyses and previous results,14,16,17,24 iron and likely stems from temperature-induced decomposi-
copper oxides can be considered responsible for the tion/oxidation of sulfur-containing species such as
catalyzed SO3 formation in the present study. With sulfates and/or sulfides within the deposit. It is
copper content of roughly 24 wt.% and iron content worth mentioning that, from a measurement/tech-
of 11 wt.%, the copper smelter dust catalyzed SO3 nical point of view, the results are very close to
formation the most, compared with copper converter those reported previously,12 indicating good repro-
dust (27 wt.% Cu) or nickel smelter dust (35 wt.% ducibility of the method.
iron). This suggests that the presence of copper as
an oxide has a greater impact on the catalyzed SO3 Synthetic Process Dusts
formation than iron oxide. The role of the two oxides
To select an inert filler oxide for the synthetic
is addressed in detail below when synthetic dusts
dusts, pure SiO2 and Al2O3 were exposed to the gas
are discussed.
mixture at 750°C for 5 min to confirm that the filler
All studied dusts catalyzed SO3 formation, with
oxide would not contribute to SO3 formation. The
23,000 ppm, 13,500 ppm, and 17,000 ppm SO3
concentrations of formed SO3 in the presence of
formed in the presence of copper smelter dust,
SiO2 and Al2O3 were 120 ppm and 80 ppm, respec-
nickel smelter dust, and copper converter dust,
tively. Part of the formed SO3 originates from
respectively (Fig. 3). Since the concentration of SO2
homogeneous (thermal) conversion, induced by the
in the synthetic process gas was 40% (400,000 ppm),
temperature (Eqs. 1 and 4)31–33 and enhanced by
the concentrations of SO2 converted to SO3 accord-
humidity (Eqs. 5 and 6).33 At 750°C, roughly
ing to Eq. 1 were 5.8%, 3.4%, and 4.3% in the
25 ppm SO3 was formed from the gas containing
presence of copper smelter dust, nickel smelter dust,
40 vol.% through homogeneous conversion in
and copper converter dust, respectively. The
 Lehmusto, Laurén, and Lindgren

10 min, indicating a minor effect of homogeneous concentration of CuO in grams is smaller than the
conversion on the overall concentration of formed concentration of Fe3O4 for the same molar fraction.
SO3.12 When the effect of homogeneous conversion This suggests that CuO is a more efficient catalyst
is taken into account, either of the tested inert than Fe3O4 in SO2-to-SO3 conversion. This trend
oxides could be used as a filler oxide in the synthetic can partly be seen in the results for the industrial
dusts, because the concentrations of formed SO3 flue dusts, as the copper smelter dust with high
were expected to be so much higher than the share copper concentration formed more SO3 than the
originating from the inert oxide, and homogeneous nickel smelter dust with high iron concentration.
conversion can be neglected. SiO2 was selected due The higher SO3 concentration formed by the copper
to its better handleability. Al2O3 has been previ- smelter dust than by the copper converter dust can
ously reported to slightly enhance SO3 formation,14 be explained by the presence of both iron and copper
although this could not be observed in the present in the smelter dust.
study. The exact reason for this difference can only The concentration of oxides did not have an
be speculated upon, but the concentrations of oxy- unequivocal effect on the concentration of formed
gen (10 vol.% in this study versus 60 vol.% in SO3; comparing the samples with 2 mol.% and
Ref. 14) and/or SO2 (40 vol.% in this study versus 10 mol.% of either oxide, the SO3 concentration
1000 ppm in Ref. 14) in the gas mixture might increased. However, a decrease was observed when
explain the divergence in the results. comparing the samples with 10 mol.% and 25 mol.%
of either oxide. The concentration of formed SO3 has
SO2 ðgÞ þ HO2 ðgÞ $ SO3 ðgÞ þ OHðgÞ ð4Þ been reported to increase linearly as a function of
Fe2O3 at 700°C up to 25 wt.% of Fe2O3,17 but similar
SO2 ðgÞ þ OHðgÞ $ HOSO2 ðgÞ ð5Þ behavior was not observed in this work. The reason
was not investigated further, but it should be noted
HOSO2 ðgÞ þ O2 ðgÞ $ SO3 ðgÞ þ HO2 ðgÞ ð6Þ that the experimental setups and methods used to
quantify the formed SO3 were completely different.
Based on the XRD results of the industrial flue
Interestingly, the catalyst concentration has also
dusts and previous calculations,34 copper has two
been reported to show a maximum, above which
stable oxides at temperatures relevant to a flash
higher catalyst concentrations result in a lower
smelter heat recovery boiler, viz. cuprous oxide
conversion rate.35 The similar results of the expo-
(Cu2O) and cupric oxide (CuO). Cupric oxide is the
sures with 10 mol.% Fe3O4 or Fe2O3 suggest that
stable form at temperatures below 900°C and was
Fe3O4 oxidized to Fe2O3, which occurs at tempera-
therefore selected for the exposures with synthetic
tures above 500°C36 at oxygen partial pressures
dusts. Fe3O4 was selected on the basis of the
well below that in the present study.37 Formation of
chemical analyses (Table I) and previous
Fe2O3 was verified by XRD analysis after the
findings.20,22
exposure, but the point of the exposure at which
Both studied oxides (CuO and Fe3O4) catalyzed
this transition takes place remains unclear. There-
SO2-to-SO3 conversion, although the impact of CuO
fore, it is better to consider the results in the light of
was remarkably greater (Fig. 4). Due to the lower
iron oxides in general rather than focusing on a
molar mass of CuO compared with Fe3O4, the
specific type of iron oxide. Since both iron and

Fig. 4. Concentrations of SO3 formed at 750°C as a function of Fe3O4 content (left-hand set) or CuO content (right-hand set). In the sample
‘‘Hema10,’’ Fe2O3 was used instead of Fe3O4.
Catalytic Role of Process Dust in SO2-to-SO3 Conversion in Flash Smelting Heat Recovery
Boilers

copper are present in the process dust and in new computational approaches based on an artifi-
deposits of the heat recovery boiler of a smelter, it cial neural network model42 or Vandermonde
is a justifiable assumption that both oxides are matrix43 have been developed to estimate and
responsible for SO3 formation in the heat recovery predict the effect of the SO3 concentration on the
boiler of a smelter plant. sulfuric acid dew point temperature. However, none

1000
Tdewpoint ¼ ð8Þ
f2:276  0:02943  lnðpH2 OÞ  0:0858  lnðpSO3 Þ þ 0:0062  lnðpH2 OÞ  lnðpSO3 Þg

Relationship Between SO3 and Sulfuric Acid of the abovementioned approaches has been devel-
Dew Point oped to predict the sulfuric acid dew point for SO3
concentrations above 1000 ppm. Therefore, before
Based on these results, it is clear that both the
these models can be applied to flash smelting
industrial and synthetic deposits catalyzed and
environments, work should be carried out to vali-
enhanced SO3 formation. In addition to Fe2O3 and
date the existing models for high SO3 concentra-
CuO, other oxides such as PbO are also known to
tions. Another possibility is to develop a new model
catalyze SO2-to-SO3 conversion.30 Although more
which takes high SO3 concentrations into account.
research should be carried out for comprehensive
clarification of which oxides in smelter plant
deposits are relevant for SO2-to-SO3 conversion, Tdewpoint ¼ 365:6905 þ 11:9864lnpH2 O
iron and copper oxide contribute significantly to SO3 þ 4:70336lnpSO3 þ ð0:446 ln pSO3
formation (Eq. 7). In addition to a suitable catalyst,
formation of SO3 in the heat recovery boiler requires þ 5:2572Þ2:19 ð9Þ
oxygen, which is available due to its addition to the
process to ensure effective sulfation of the flue dust. Tdewpoint ¼ 150 þ 11:664lnpSO3 þ 8:13281lnpH2 O
It should also be noted that, in addition to SO3  0:383226lnpSO3  lnpH2 O
formation in the heat recovery boiler, SO3 can also ð10Þ
form in the smelting or converting furnace and
when entering the heat recovery boiler, increasing To provide a perspective on how the measured SO3
the overall concentration of SO3. concentrations might affect the sulfuric acid dew
point, it was calculated using all three equations
1 Fe2 O3 orCuOðcatalystÞ
SO2 ðgÞ þ O2 ðgÞ ! SO3 ðgÞ ð7Þ presented above. The results are presented in
2 Table III. The calculations were carried out for a
As mentioned in the ‘‘Introduction,’’ the formation of humidity level of 5 vol.% (38 mmHg), a typical value
SO3 together with the presence of humidity enables for copper flash smelter heat recovery boilers.
sulfuric acid (H2SO4) formation. The dew point Depending on the equation used, the calculated
temperature of H2SO4 depends on the SO3 and H2O dew point temperatures differ by 30 K at most, with
concentrations.31 Different empirical equations Eq. 8 predicting the highest dew point tempera-
have been proposed to predict the dew point of tures, Eq. 9 the lowest values, and Eq. 10 interme-
sulfuric acid as a function of the concentrations of diate values. As mentioned above, none of these
SO3 and H2O. The Verhoff–Banchero equation equations has been validated for high SO3 concen-
(Eq. 8)38 has been successfully used in applications trations, so it cannot be stated which one would be
addressing lower SO3 concentrations,31,39,40 but is most suitable for application in the flash smelter
not accurate when applied to acid dew point calcu- environment. Nevertheless, the large deviation in
lations with very low SO3 and high H2O concentra- dew point temperatures obtained when using the
tions.41 The partial pressures in Eq. 8 are expressed different equations suggests that a predictive tool
in mmHg, and the sulfuric acid dew point in K. applicable for high SO3 concentrations is required.
Another equation (Eq. 9)9 provides better predic- Among the calculated dew point temperatures, the
tions of the acid dew point at high H2O concentra- highest sulfuric acid dew point of 233°C is only
tions but loses accuracy at low H 2O around 40°C below the lowest heat-transfer surface
concentrations.41 The partial pressures in Eq. 9 temperature of 275°C. Although the calculated
are expressed in mmHg, and the sulfuric acid dew value is most likely an overestimation, it demon-
point in K. To overcome the shortcomings of these strates how close to one another the material
two equations, a more recent equation (Eq. 10), surface temperature and sulfuric acid dew point
based on 188 validated data points, has been temperatures might be. This becomes even more
developed.41 The partial pressures in Eq. 9 are important in situations in which the heat-transfer
expressed in mmHg, and the sulfuric acid dew point surface temperature might decrease below the
in °C. In addition to the equations presented above,
 Lehmusto, Laurén, and Lindgren

Table III. Sulfuric acid dew points (Tdew) calculated using Eqs. 8–10

SO3 p(SO3) Tdew (K) Tdew (K) Tdew (K) Tdew (°C) Tdew (°C) Tdew (°C)
Sample (ppm) (mmHg) Eq. 8 Eq. 9 Eq. 10 Eq. 8 Eq. 9 Eq. 10
Copper 23,040 17.5104 503 468 482 230 195 209
smelter
Copper con- 16,980 12.9048 498 466 479 225 193 206
verter
Nickel smel- 13,481 10.24556 495 464 477 221 191 203
ter
2 mol.% 6878 5.22728 484 459 470 211 185 197
Fe3O4
10 mol.% 11,104 8.43904 492 463 475 218 190 201
Fe3O4
10 mol.% 11,197 8.50972 492 463 475 219 190 202
Fe2O3
25 mol.% 8817 6.70092 488 461 472 215 188 199
Fe3O4
2 mol.% 23,796 18.08496 504 469 482 230 196 209
CuO
10 mol.% 28,369 21.56044 506 470 484 233 197 211
CuO
25 mol.% 24,417 18.55692 504 469 483 231 196 210
CuO

sulfuric acid dew point due to pressure fluctuations  All the industrial flue dusts catalyzed SO2-to-
or maintenance shutdowns, resulting in sulfuric SO3 conversion: copper smelter dust the most
acid condensation and, most likely, material degra- and nickel smelter dust the least.
dation. It should be noted that the calculations  Regarding the synthetic deposits, both studied
presented in Table III were carried out at only one oxides (CuO and Fe3O4) catalyzed formation of
humidity level. Since the sulfuric acid dew point SO3.
temperature depends also on the partial pressure of  Among the two studied oxides, CuO had a
H2O in the process gas, factors such as air leakage greater influence on the SO3 formation than
could increase the concentration of water in the Fe3O4.
process gas and thereby affect the acid dew point  The greater conversion efficiency of CuO ex-
temperature. Despite the uncertainty regarding the plains the higher concentrations of SO3 formed
accuracy of the calculated dew point temperature, it in the presence of industrial flue dusts with high
is obvious that smelter dusts play a role in SO3 copper content.
formation. It has been reported that, at tempera-  The concentration of catalytic oxide in the syn-
tures above 400°C, SO2 is the favored species of thetic dusts did not play a major role in terms of
oxidized sulfur instead of SO3.9 However, the high the conversion efficiency.
concentration of SO2 in the smelter process gas and  The catalytic abilities of the dusts might form
the presence of a suitable catalyst enable formation concentrations of SO3 sufficient to enable sulfu-
of SO3 in quantities high enough to lower the ric acid dew point corrosion at temperatures
sulfuric acid dew point temperature in a flash close to the surface temperatures of heat-trans-
smelter. fer components in flash smelter heat recovery
boilers.
CONCLUSION
The aim of this work is to examine the catalytic ACKNOWLEDGEMENTS
role of both genuine flash smelter flue dusts and Open access funding provided by Abo Akademi
synthetic dusts in SO2-to-SO3 conversion at 750°C University (ABO). This work was carried out within
by exposing copper smelter dust, copper converter the Academy of Finland Project ‘‘Novel Approaches
dust, nickel smelter dust, and synthetic dusts with to Study Corrosion Mechanisms in High-tempera-
different CuO or Fe3O4 content to synthetic process ture Industrial Processes’’ (Decision No. 296435).
gas with 40 vol.% SO2. After exposure, the concen- Financial support from Boliden Harjavalta and
tration of formed SO3 was quantified. The main Outotec is gratefully acknowledged. In addition, the
conclusions can be summarized as follows: authors would like to thank Mr. Janne Hautamäki
Catalytic Role of Process Dust in SO2-to-SO3 Conversion in Flash Smelting Heat Recovery
Boilers

for the dust samples, Mr. Luı́s Bezerra for carrying 20. T. Markova, B. Boyanov, S. Pironkov, and N. Shopov, J.
out the IC analyses, and Mr. Linus Silvander for Min. Metall. Sect. B 36, 195 (2000).
21. R.K. Jana, B. Gorai, and Premchand, Environmental and
operating the SEM apparatus. Waste Management (NML: Jamshedpur, 1998), pp. 167–173.
22. C. Samuelsson and G. Carlsson, CIM Bull. 94, 111 (2001).
OPEN ACCESS 23. D.O. Okanigbe, A.P.I. Popoola, and A.A. Adeleke, Procedia
This article is distributed under the terms of the Manuf. 7, 121 (2017).
24. G.M. Schwab and E. Kaldis, Naturwissenschaften 50, 516
Creative Commons Attribution 4.0 International (1963).
License (http://creativecommons.org/licenses/by/4.0/ 25. A. Pettersson, L.-E. Åmand, and B.-M. Steenari, Fuel 88,
), which permits unrestricted use, distribution, and 1758 (2009).
reproduction in any medium, provided you give 26. ICDD, PDF-4+ 2018 (Database), in ed. by S. Kabekkodu
appropriate credit to the original author(s) and the (International Centre for Diffraction Data, Newtown
Square, 2018).
source, provide a link to the Creative Commons li- 27. U. Kleinhans, C. Wieland, F.J. Frandsen, and H. Spliethoff,
cense, and indicate if changes were made. Prog. Energy Combust. Sci. 68, 65 (2018).
28. A. Morales, M. Cruells, A. Roca and R. Bergo, Characteri-
REFERENCES zation of flue dusts from a copper smelter furnace, copper
1. I.V. Kojo, A. Jokilaakso, and P. Hanniala, JOM 52, 57 recovery and arsenic stabilization, in ed. by P.A. Riveros,
(2000). D.G. Dixon, D.B. Dreisinger, and M.J. Collins The John E.
2. T. Mäkinen and P. Taskinen, Trans. Inst. Min. Metall. Sect. Dutrizac International Symposium on Copper Hydrometal-
C 117, 86 (2008). lurgy, Cu 2007, Volume IV (Canadian Institute of Mining,
3. M.E. Schelsinger, M.J. King, K.C. Sole, and W.G. Daven- Metallurgy and Petroleum, Toronto, 2008), pp. 177–189.
port, Extractive Metallurgy of Copper, 5th ed. (Amsterdam: 29. T. Kurosawa, T. Yagishi, K. Togo, and T. Kato, J-STAGE 89,
Elsevier, 2011), pp. 89–110. 107 (1973).
4. T. Ranki-Kilpinen, E.J. Peuraniemi, M., and Mäkinen, in 30. S. Sarkar, JOM 34, 43 (1982).
Sulfide Smelting 2002 (TMS, Warrendale, 2002), pp. 261– 31. E. Vainio, T. Laurén, N. DeMartini, A. Brink, and M. Hupa,
272. J. For. 4, 14 (2014).
5. E.R. Lovejoy, D.R. Hanson, and L.G. Huey, J. Phys. Chem. 32. R.J.S. Eloi, C.J. Newman, and G. Macfarlane, CIM Bull. 87,
100, 19911 (1996). 77 (1994).
6. K. Morokuma and C. Muguruma, J. Am. Chem. Soc. 116, 33. V. Montenegro, H. Sano, and T. Fujisawa, Miner. Eng. 49,
10316 (1994). 184 (2013).
7. J.T. Jayne, U. Poeschl, Y. Chen, D. Dai, L. Molina, D.R. 34. T. Ranki-Kilpinen, Sulphation of Cuprous and Cupric Oxide
Worsnop, C.E. Kolb, and M.J. Molina, J. Phys. Chem. A 101, Dusts and Heterogeneous Copper Matte Particles in Simu-
10000 (1997). lated Flash Smelting Heat Recovery Boiler Conditions,
8. R. Ebara, F. Tanaka, and M. Kawasaki, Eng. Fail. Anal. 33, Doctoral Thesis (Helsinki University of Technology, 2004).
29 (2013). 35. M.-C. Kim, Appl. Chem. Eng. 23, 588 (2012).
9. A.G. Okkes, Hydrocarb. Process. 66, 53 (1987). 36. K.F. McCarty, M. Monti, S. Nie, D.A. Siegel, E. Starodub, F.
10. J.M. Blanco and F. Pena, Appl. Therm. Eng. 28, 777 (2008). El Gabaly, A.H. McDaniel, A. Shavorskiy, T. Tyliszczak, H.
11. J. Lehmusto, D. Stenlund, M. Lindgren, and P. Yrjas, Oxid. Bluhm, N.C. Bartelt, and J. de la Figuera, J. Phys. Chem. C
Met. 87, 199 (2017). 118, 19768 (2014).
12. J. Lehmusto, E. Vainio, T. Laurén, and M. Lindgren, Metall. 37. H. Sakai, T. Tsui, and K. Naito, J. Nucl. Sci. Technol. 21,
Mater. Trans. B 49, 434 (2018). 844 (1984).
13. L.P. Belo, L.K. Elliott, R.J. Stanger, R. Spörl, K.V. Shah, J. 38. F. Verhoff and J. Banchero, Chem. Eng. Proc. 70, 71 (1974).
Maier, and T.F. Wall, Energy Fuels 28, 7243 (2014). 39. W.M.M. Huijbregts and R.G.I. Leferink, Anti-Corros.
14. T.L. Jørgensen, H. Linbjerg, and P. Glarborg, Chem. Eng. Methods M 51, 173 (2004).
Sci. 62, 4496 (2007). 40. V. Ganapathy, Hydrocarb. Process. 68, 57 (1989).
15. D. Fleig, K. Andersson, and F. Johnsson, Ind. Eng. Chem. 41. B. ZareNezhad, Oil Gas J. 107, 60 (2009).
Res. 51, 9483 (2012). 42. B. ZareNezhad and A. Aminiam, Energ. Convers. Manag.
16. P.M. Foster, Atmos. Environ. 3, 157 (1969). 52, 911 (2011).
17. P. Marier and H.P. Dibbs, Thermochim. Acta 8, 155 (1974). 43. A. Bahadori, Appl. Therm. Eng. 31, 1457 (2011).
18. R. Spörl, J. Walker, L. Belo, K. Shah, R. Stanger, J. Maier,
T. Wall, and G. Scheffknecht, Energy Fuels 28, 5296 (2014). Publisher’s Note Springer Nature remains neutral with re-
19. D. Fleig, M.U. Alzueta, F. Normann, M. Abián, K. Ander- gard to jurisdictional claims in published maps and institutional
sson, and F. Johnsson, Combust. Flame 160, 1142 (2013). affiliations.