International Journal of Mineral Processing 110–111 (2012) 71–73

Contents lists available at SciVerse ScienceDirect

International Journal of Mineral Processing

journal homepage: www.elsevier.com/locate/ijminpro

Thermodynamic analysis and reduction of tin oxide with methane

Senol Cetinkaya, Serafettin Eroglu ⁎
Istanbul University, Engineering Faculty, Department of Metallurgical and Materials Eng., Avcilar, 34320, Istanbul, Turkey

a r t i c l e i n f o a b s t r a c t

Article history: The present study aims to investigate reduction behavior of SnO2 in a flowing methane atmosphere at
Received 26 October 2011 1200 K. Mass measurements and X-ray diffraction were used to characterize the products at various stages
Received in revised form 9 April 2012 of reactions. It was found that the extent of reduction of SnO2 to Sn increased with time. Complete reduction
Accepted 11 April 2012
was achieved within 60 min at 1200 K. Thermodynamic calculations based on the minimization of the Gibbs'
Available online 21 April 2012
free energy revealed that the reduction of SnO2 to Sn proceeds via an intermediate compound of SnO at
Keywords:
1200 K. Possible reactions leading to the formation of Sn were suggested using the results of thermodynamic
Pyrometallurgy calculations. It was demonstrated that CH4 acts as a reducing agent for SnO2 at 1200 K.
Oxide reduction © 2012 Elsevier B.V. All rights reserved.
Tin
Methane
Thermodynamic analysis

1. Introduction function of temperature (1100–1300 K) and input CH4 mole fraction

(~0–1). Input CH4 mole fraction (XCH4) is defined as noCH4/(noSnO2 +
Because tin has low toxicity, a low melting point and high corrosion noCH4) where no represents the initial number of moles of CH4 and
resistance, it has been used extensively for soldering, corrosion preven- SnO2. The principles of the thermodynamic approach are described
tion and food packaging. Tin is conventionally produced by reactions of elsewhere (Eriksson, 1975; Besmann, 1977). In the system of Sn–O–
a tin oxide mineral (SnO2, cassiterite) with solid C (Hampshire, 1990). C–H, 38 species were considered as the constituents of the gas phase
This pyrometallurgical process, however, requires high temperatures including CH4, H2, H2O, CO, CO2, CH3, CH2O, C2H4, H, Sn and SnO. Con-
(>1500 K) and long reaction times because of sluggish reaction rate be- densed phases were assumed to be C(s), SnO2(s), SnO(s), and Sn(l)
tween solid SnO2 and C. This problem may be circumvented by using where the subscripts s and l stand for solid and liquid, respectively.
gaseous reducing agents which provide intimate contact with SnO2. Thermodynamic data (Gibbs' free energy of the formation) of the con-
Thus, enhanced reduction of SnO2 via solid–gas reactions may take stituents considered here were obtained from the published tables
place at low temperatures. Hydrogen is widely employed for the reduc- (Chase et al., 1985; Barin, 1993). Fig. 1 depicts the diagrams constructed
tion of metal oxides. But, it is relatively expensive. Methane has also from the calculated results. An equilibrium stability diagram for con-
been used as a reducing agent for iron oxide (Ghosh et al., 1986), nickel densed phases is shown in Fig. 1a as a function of temperature and
oxide (Alizadeh et al., 2007), cobalt oxide (Khoshandam et al., 2004) XCH4. The diagram displays SnO2 + SnO, SnO and SnO+ Sn phase fields
and zinc oxide (Ebrahim and Jamshidi, 2001). which appear successively as XCH4 increases to 0.37 at temperatures be-
To the best of our knowledge, no report has been published on the tween 1300 K and 1170 K. As also seen from the diagram, Sn co-
synthesis of Sn by reaction of SnO2 with CH4. The present study aims exists with SnO2 at temperatures less than 1170 K. Furthermore,
to investigate pyrometallurgical reduction behavior of SnO2 in a flowing Sn and Sn + C phase fields are predicted at all temperatures studied
pure CH4 atmosphere. Furthermore, thermodynamic analysis in the Sn– for XCH4 higher than 0.37. Fig. 1b shows variations of the concentrations
O–C–H system was carried out in order to get a deeper insight into the of the gaseous products with XCH4 at 1200 K. H2, CO, H2O, CO2 and CH4
thermochemistry of the process. We used undiluted CH4 because it is are predicted to be the major components of the gas phase. Minor
relatively cheap and abundant (a main component of natural gas). gaseous species such as SnO, C2H4, CH2O, CH3, H and Sn are also present
in the system. It is possible to suggest reaction pathways to the forma-
tions of gaseous species and metallic Sn using variations of the contents
2. Thermodynamic analysis in the Sn–O–C–H system of the condensed and gas phases with XCH4. This will be discussed later
in Section 4.
Thermodynamic calculations based on the method of minimization
of the Gibbs' free energy were performed at atmospheric pressure as a 3. Materials and methods

⁎ Corresponding author. Tel.: + 90 212 473 7065; fax: + 90 212 473 7180. The experimental set-up used for the present study essentially
E-mail address: seref@istanbul.edu.tr (S. Eroglu). consists of a hot-wall furnace with SiC heating elements, a quartz tube

0301-7516/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.minpro.2012.04.005
72 S. Cetinkaya, S. Eroglu / International Journal of Mineral Processing 110–111 (2012) 71–73

1300
SnO a
1250
Temperature (K)

SnO2 + SnO SnO + Sn

1200 Sn Sn + C

1150 SnO2 + SnO + Sn

SnO2 + Sn
1100
0 0.1 0. 2 0.3 0. 4 0.5 0. 6 0.7 0. 8 0.9 1
noCH4 / (noCH4 + noSnO2)

100 SnO2 + SnO SnO + Sn Sn Sn + C b

SnO
Gas Composition (mole)

10-2 H2
H2 O CO
10-4
Fig. 2. A scanning electron microscope image of SnO2 powder.
CO2
10-6 CH4
reaction time when the reaction temperature was raised to 1200 K as
10-8 SnO C2H4 seen from Fig. 3. Mass loss increases to 21.56 wt.% within 60 min close
to the theoretical value (21.23 wt.%) for the complete reduction of
H CH3 Sn
10-10
CH2O SnO2 to Sn. A solidified malleable metallic product was obtained after
0 0.1 0. 2 0.3 0. 4 0.5 0. 6 0.7 0.8 0. 9 1 full reduction.
no
/ (n o
+n o
) Fig. 4 displays X-ray diffraction patterns of the starting SnO2 pow-
CH4 CH4 SnO2
der and the products obtained from the reaction of SnO2 with CH4 for
Fig. 1. (a) Equilibrium stability diagram showing condensed phase fields as a function of various reaction times. Published patterns of the matched materials
temperature and input CH4 mole fraction (XCH4) in the SnO2–CH4 system. (b) Variation (JCPDS, 2008) are also shown in the figure as vertical lines. Compara-
of gas composition with XCH4 at 1200 K. tive analysis of the peak intensities in the patterns a–d reveals that
metal content of the product increases with time. As seen from pat-
(20 mm in diameter, 500 mm in length) and gas flow meters. The details tern b, diffraction peaks from Sn phase appear within 15 min. At
of the system were given elsewhere (Cetinkaya and Eroglu, 2010). The 60 min, the XRD pattern consists of only Sn peaks (pattern d). Free
chemicals used for the present study were argon, methane and SnO2 carbon was not detected in the products by XRD. This result contra-
powder. dicts the thermodynamic prediction. The reason for the absence of
A Scanning Electron Microscope (SEM) equipped with Field Emis- free carbon could be open tube flow which allows some CH4 to escape
sion Gun (FEI Quanta FEG 450) was employed to reveal morphology the system without any contact with powder. A kinetically limiting
of the starting oxide powder. Fig. 2 shows that the powder used is step in the pyrolysis of CH4 may also be responsible for the absence
mostly composed of fine, almost round particles with loose agglomer- of free carbon. An intermediate product, SnO, was predicted to form
ation. Mean particle size was determined from the areas of ~ 130 par- at 1200 K by the thermodynamic calculations. No SnO peaks were,
ticles using the SEM images. Equivalent circular diameters of the however, seen in the XRD patterns of the partially reduced products
measured areas were used to calculate the mean particle size (70 ± (patterns b and c). This result may be explained by the equilibrium
42 nm). stability diagram for the condensed phases (Fig. 1). As seen from
An alumina boat was used as a holder for SnO2 powder. SnO2 powder Fig. 1a, SnO is not stable at temperatures below 1170 K. Hence, if
weighing ~150 mg (~0.001 mol) was allowed to react with CH4 for
15–60 min at temperatures of 1100 K and 1200 K. The flow rate of CH4
was held constant at 40 cm3/min. Argon (85 cm3/min) was used during
0
heating and cooling of the samples. A parafocussing X-ray diffractometer
(Rigaku D/Max-2200) equipped with a Cu radiation tube was employed
for the phase analysis. Mass measurements before and after the synthesis -5
Mass Change (%)

experiments were carried out by a calibrated electronic balance (Sartori-

us BP110S) with a sensitivity of ±10− 4 g in order to determine the -10
extent of reactions. Change in sample mass was calculated using the
equation of (m− mo)/mo ×100, where m is the mass at isothermal -15
time t and mo is the starting mass of the oxide powder.
-20 SnO2 Sn
4. Results and discussion

Equilibrium thermodynamic analysis revealed that Sn could be syn- -25

0 15 30 45 60
thesized from SnO2 under CH4 atmosphere at temperatures between
Holding Time (min)
1100 and 1300 K. Experimental studies carried out at 1100 K, however,
showed no change in the sample mass within 60 min. This result indi- Fig. 3. Variation of mass of SnO2 with reaction time in a flowing CH4 atmosphere at
cates that reduction of SnO2 does not take place owing to high stability 1200 K. Horizontal dashed line represents the theoretical mass loss for the complete
of CH4 at 1100 K. The sample mass was observed to decrease with reduction of SnO2 to Sn.
 S. Cetinkaya, S. Eroglu / International Journal of Mineral Processing 110–111 (2012) 71–73 73

: SnO2 (41-1445) ◊ : Sn (04-0673) In the Sn phase field, H2 and CO content of the gas phase increases
while those of H2O and CO2 decrease with rising XCH4. These variations
in the gas phase may occur through the following net reaction:

CO2ðgÞ þ 2H2 OðgÞ þ 3CH4ðgÞ →4COðgÞ þ 8H2ðgÞ ðΔG¨R ¼ −236900JÞ ð5Þ

Overall reactions between SnO2 and CH4 for Sn formation are

d suggested as follows:

SnO2ðsÞ þ 2CH4ðgÞ →SnðlÞ þ 2COðgÞ þ 4H2ðgÞ ðΔG¨R ¼ −186950JÞ ð6Þ

c 2SnO2ðsÞ þ CH4ðgÞ →2SnðlÞ þ CO2ðgÞ þ 2H2 OðgÞ ðΔG¨R ¼ −137000JÞ ð7Þ

Intensity (a.u.)

These reactions may be viewed as total reactions of 1 and 3 or 2, 4

and 5 or 2 and 4.

b 5. Conclusions

Mass measurements revealed that the extent of the reduction of

SnO2 by CH4 increased with reaction time at 1200 K. XRD analysis of
the products indicated that reduction of SnO2 to Sn was completed
within 30 min. The experimental results showed that CH4 acts as a re-
a
ducing agent for SnO2 at 1200 K, but not at 1100 K. Equilibrium thermo-
dynamic analysis by the method of minimization of Gibbs' free energy
Sn (04-0673)
predicted the reaction pathways to Sn and to the product species in
SnO2 (41-1445) the Sn–O–C–H system. It was demonstrated that Sn could be produced
from SnO2 powder in a flowing undiluted CH4 atmosphere.
10 20 30 40 50 60 70 80
2θ (Degree) References
Fig. 4. X-ray diffraction patterns of (a) SnO2 powder and the products obtained by Alizadeh, R., Jamshidi, E., Ale-Ebrahim, H., 2007. Kinetic study of nickel oxide reduction
reaction of the oxide with CH4 at 1200 K for (b) 15, (c) 30 and (d) 60 min. by methane. Chem. Eng. Technol. 30, 1123–1128.
Barin, I., 1993. Thermochemical Data of Pure Substances. VCH Verlagsgesellschaft,
WeinHeim.
SnO is formed at 1200 K, it may transform to Sn and SnO2 during Besmann, T.M., 1977. SOLGASMIX-PV, a computer program to calculate equilibrium re-
lationships in complex chemical systems. Report No: ORNL/TM-5775. Oak Ridge
cooling to the room temperature. National Laboratory.
The variations of the contents of condensed and gas phases with Cetinkaya, S., Eroglu, S., 2010. Thermodynamic analysis and synthesis of porous Mo2C
XCH4 at 1200 K (Fig. 1) were analyzed in order to suggest reaction sponge by vapor-phase condensation and in situ carburization of MoO3. J. Alloy.
Compd. 489, 36–41.
pathways to the formation of condensed and gaseous species. In the
Chase, M.W., Davies, C.A., Downey, J.R., Frurip, D.J., McDonald, R.A., Syverud, A.N., 1985.
SnO2 + SnO phase field, SnO content of the condensed phase and JANAF Thermochemical Tables, third ed. J. Phys. Chem. 14 (Suppl. 1) Ref. Data.
the concentrations of the major products in the gas phase (H2O, Ebrahim, H.A., Jamshidi, E., 2001. Kinetic study of zinc oxide reduction by methane.
Chem. Eng. Res. Des. 79, 62–70.
CO2, H2 and CO) increase with XCH4. Hence, the following reactions
Eriksson, G., 1975. Thermodynamic studies of high temperature equilibrium. Chem.
may occur initially: Scr. 8, 100–103.
Ghosh, D., Roy, A.K., Ghosh, A., 1986. Reduction of ferric-oxide pellets with methane.
SnO2ðsÞ þ CH4ðgÞ →SnOðsÞ þ COðgÞ þ 2H2ðgÞ ðΔG¨R ¼ −93589JÞ ð1Þ T. Iron. Steel. I. Jpn. 26, 186–193.
Hampshire, W.B., 1990. Tin and tin alloys. In: Davis, J.R. (Ed.), ASM Handbook, Volume
02 - Properties and Selection: Nonferrous Alloys and Special-Purpose Materials.
4SnO2ðsÞ þ CH4ðgÞ →4SnOðsÞ þ CO2ðgÞ þ 2H2 OðgÞ ðΔG¨R ¼ −137456JÞ ð2Þ
Materials Park, OH, p. 517.
JCPDS (Joint Committee on Powder Diffraction Standards), 2008. International Center
for Diffraction Data, Swartmore PA.
where ΔGºR is the standard Gibbs' free energy change of the reac- Khoshandam, B., Kumar, R.V., Jamshidi, E., 2004. Reduction of cobalt oxide with methane.
tions, and g stands for gas. Reduction of intermediate phase SnO to Sn Metall. Mater. Trans. B. 35, 825–828.
may take place via reactions 3 and 4.

SnOðsÞ þ CH4ðgÞ →SnðlÞ þ COðgÞ þ 2H2ðgÞ ðΔG¨R ¼ −93361JÞ ð3Þ

4SnOðsÞ þ CH4ðgÞ →4SnðlÞ þ CO2ðgÞ þ 2H2 OðgÞ ðΔG¨R ¼ −136544JÞ ð4Þ