Characterisation of the oxidation products of pyrite by thermogravimetric and evolved gas analysis

Article abstract of DOI:10.1023/A:1025001811801

Museum specimens of pyrite are known to undergo oxidation even during storage. Characterisation of the oxidation products is however not always simple due to amorphous character and varying degrees of hydration of the oxidation products, This paper presen

Full text of DOI:10.1023/A:1025001811801

Journal of Thermal Analysis and Calorimetry, Vol. 72 (2003) 769–776
CHARACTERISATION OF THE OXIDATION
PRODUCTS OF PYRITE BY THERMOGRAVIMETRIC
AND EVOLVED GAS ANALYSIS
P. S. Thomas*, D. Hirschausen, R. E. White, J. P. Guerbois and A. S. Ray
Department of Chemistry, Materials and Forensic Sciences, University of Technology, Sydney,
Australia
Abstract
Museum specimens of pyrite are known to undergo oxidation even during storage. Characterisation
of the oxidation products is however not always simple due to amorphous character and varying de-
grees of hydration of the oxidation products. This paper presents an alternative approach to the char-
acterisation of oxidation products by identifying their presence from their characteristic thermal de-
composition processes using thermogravimetric and evolved gas analysis. Four pyrite specimens
were characterised with varying degrees of oxidation. Iron(II) and iron(III) sulphates were also char-
acterised for comparative purposes. The degradation products were observed to correlate well with
the presence of iron(II) sulphate even though there was found to be some discrepancy in the higher
temperature decomposition reactions.
Keywords: evolved gas analysis, iron sulphate, pyrite, thermogravimetric analysis
Introduction
Bulk samples of the mineral pyrite (iron disulfide, FeS2) oxidise unpredictably, re-
sulting in the destruction of valuable museum specimens and geological records [1].
Although considerable research effort has been expended on the mechanisms of py-
rite oxidation, or pyrite ‘disease’, the factors involved in the initiation of sudden,
rapid oxidation of bulk specimens are poorly understood. Pyrite is known to be oxi-
dised by atmospheric oxygen and ferric ions [2], and that this process is accelerated
by the presence of moisture [3]. The rate and the products of oxidation, and, there-
fore, the mechanisms, are also a strong function of the relative humidity and the par-
tial pressure of oxygen [4]. Since oxidation is initiated at crystal surfaces, fine
grained specimens with a large surface area tend to be more susceptible to degrada-
tion than single crystal specimens. These generalisations, however, do not account
for the sudden and catastrophic degradation of many bulk pyrite samples which may
have been stored carefully.
*
Author for correspondence: E-mail: paul.thomas@uts.edu.au
1388–6150/2003/ $ 20.00
Akadémiai Kiadó, Budapest
© 2003 Akadémiai Kiadó, Budapest
Kluwer Academic Publishers, Dordrecht

770
THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
Characterisation of the oxidation products is also difficult. Due to the interfacial
and heterogeneous nature of the oxidation process, oxidation products tend to be
nanocrystalline or amorphous and are therefore not easily characterised by traditional
techniques such as X-ray diffraction. In this paper the characterisation of the oxida-
tion products by thermogravimetric analysis (TG) coupled with evolved gas analysis
using mass spectroscopy (MS) is presented.
Experimental
Unaffected pyrite specimens and specimens showing signs of oxidation were sup-
plied by the Australian Museum, Sydney, and are listed in Table 1. The samples were
confirmed to be pyrite by X-ray analysis using a Siemens D5000 diffractometer with
CuKα radiation, although no oxidation products were observed in the degraded speci-
mens. Specimen P1 differed from the other specimens in that P1 contained a signifi-
cant proportion of quartz (approximately 25% as determined from the mass loss asso-
ciated with the pyrite to pyrrhotite decomposition at 640°C (Fig. 1)). The ‘as
received’ specimens were crushed by an agate pestle and mortar to produce a powder
which was then characterised by TG. As the oxidation process produces hydrated
sulphates as the products, iron(II) sulphate heptahydrate (FeSO4 7H2O, ACS Reagent
(99+%), Aldrich) and ferric sulphate hydrate (Fe2(SO4)3 xH2O (97+%), Aldrich)
were also characterised in their ‘as received’ state.
Table 1 Classification of the pyrite specimens
Specimen Origin
Description
Kristineberg Mine,
P1
Undegraded pyrite crystals in quartz
Vosterbotten County, Sweden
Sepik River, Papua New Guinea
Huanzala, Peru
P2
P3
Undergraded pyrite crystals
Pyrite Dollar showing signs of pyrite disease
Pyrite nodules extracted from shale,
showing significant signs of pyrite disease
P4
Pannawonika, Western Australia
TG characterisation was carried out in a Setaram Setsys 16/18 TG-DSC fitted
with a capillary in order to sample the volatiles produced during decomposition reac-
tions. The capillary at the base of the TG furnace carried the evolved gases to a
Balzers MS for mass analysis. Temperature calibration on the TG was carried out
with indium, tin, zinc, antimony, gold and silver. The TG experiments were carried
out on 10 to 15 mg specimens in an argon (BOC, high purity, <10 ppm oxygen) atmo-
sphere with a flow rate of 20 cm3 min–1. Baseline calibrations were also carried out
using the same conditions in order to compensate for buoyancy effects. The gas anal-
ysis was carried out on the Channel Tron electron multiplier detector of the MS in or-
der to maximise sensitivity. As this detector is not linear, quantitative analysis using
J. Therm. Anal. Cal., 72, 2003

THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
771
Fig. 1 TG curves for the pyrite and iron sulphate specimens characterised
the MS signal was not possible. The MS data was, therefore, normalised for presenta-
tion purposes.
Results and discussion
The mass loss (TG) and differential TG (DTG) curves for samples of undegraded py-
rite (P1 and P2) and specimens of pyrite exhibiting signs of decay (P3 and P4) are
shown in Figs 1 and 2, respectively. Also included in these figures are the decomposi-
tion curves of the iron(II) and iron(III) sulphates. These latter experiments were car-
ried out for comparative purposes aiding the identification of the oxidation products
in specimens of pyrite. Significant differences are observed in the decomposition pro-
cesses of Specimens P1 to P4 and the iron(II) and iron(III) sulphates. The mass losses
for each step in the TG curves (in % based on the original mass) are listed in Table 2
and the DTG peaks for each step are listed in Table 3.
Table 2 Mass loss (% based on initial mass) measurements for the decomposition steps observed
Sample
Step 1
Eq. (3)
Step 2
Eq. (4)
Step 3
Eq. (5)
Step 4
Eq. (6)
Step 5
Eq. (1)
Reaction mechanism
P1
17.6
23.7
17.8
17.0
P2
P3
1.8
3.0
0.6
1.4
2.8
4.4
1.4
2.5
P4
FeSO4 7H2O
32.9
7.1
19.21
11.3
15.4
48.92
Fe2(SO4)3 xH2O
1Dehydration of iron(III) sulphate hydrate occurs according to: Fe2(SO4)3 xH2O→Fe2(SO4)3+xH2O
2Thermal decomposition of iron(III) sulphate occurs in accordance with Eq. (2)
J. Therm. Anal. Cal., 72, 2003

772
THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
Table 3 DTG peak temperatures (°C) for the decomposition steps listed in Table 2
Sample
Step 1
Step 2
Step 3
Step 4
Step 5
P1
633
640
631
636
P2
P3
105
116
106
233
232
227
192
484
481
580
521
529
659
686
P4
FeSO4 7H2O
Fe2(SO4)3×xH2O
The pyrite specimens that had remained unaffected by pyrite disease in storage
(P1 and P2) exhibited only a single decomposition DTG peak around 637°C. The
mass loss at this decomposition step for the P2 specimen was observed to correspond
to the formation of pyrrhotite of composition Fe10S11:
FeS2→FeS(2–x)+xS
(1)
as has been observed elsewhere [5]. This transformation was confirmed by X-ray diffrac-
tion analysis of the residue after the heating program had been completed. Only MS
peaks at the single decomposition temperature characteristic of the evolution of sulphur
were observed. Specimen P1 also decomposed in a similar manner to P2; however, due
to the presence of quartz in the matrix the observed mass loss is diminished.
Fig. 2 DTG curves for the pyrite and iron sulphate specimens characterised
The samples which were observed to have suffered pyrite disease produced a
number of mass loss steps during the thermal decomposition (Fig. 1). DTG peaks
(Fig. 2) were observed around 110 and 250°C, corresponding to the evolution of wa-
ter, as observed in the MS data; 490 and 530°C corresponding to sulphate decomposi-
tion and 635°C which is ascribed to pyrite decomposition. The last three assignments
are based on the MS data shown in Figs 3 and 4. 32 (16O2+ and 32S+ ions) and 34 (34S+)
atomic mass units (amu) curves are used to differentiate between the presence of sul-
J. Therm. Anal. Cal., 72, 2003

THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
773
phur and oxygen, and 64 amu to identify the presence ofSO+2 ion. A single peak is ob-
served in each curve in Fig. 3 corresponding to sulphur evolution in the decomposi-
tion of pyrite in the manner of Eq. (3). The presence of oxygen in the argon, even
though its concentration is less than 10 ppm, allows oxidation of the sulphur and thus
a peak at 64 amu is observed. Figure 4 shows the same set of curves for the partially
degraded specimen P4. In this case a number of supplementary peaks are observed.
64 and 32 amu curves show peaks in the range 450 to 550°C. These peaks are not ob-
served in the 34 amu curve. The absence of a peak in the 34 amu curve, coupled with
their presence in the 32 amu curve, indicates that the thermal decomposition of the
products of pyrite disease yields both SO2 and O2. It should be noted that, although
not shown in Figs 3 or 4, SO3 was not observed as a product as no peak was observed
in the 80 amu data.
Fig. 3 DTG and MS curves for the thermal decomposition of P2
Fig. 4 DTG and MS curves for the thermal decomposition of P4
The process of decomposition of the iron(III) sulphate with a maximum rate at
686°C corresponds to the formation of iron(III) oxide with the evolution of SO3 as the
evolved product and corresponds to the mechanism reported by Paulik et al. [6].
J. Therm. Anal. Cal., 72, 2003

774
THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
Fe2(SO4)3→Fe2O3+3SO3
(2)
The iron(III) sulphate hydrate also has a dehydration peak around 190°C. This peak
does not correspond well with the decomposition temperatures of the partially oxi-
dised pyrite specimens. It is evident that the iron(III) sulphate is not a major product
of the pyrite decay process even though in aqueous environments it has been shown
to be important in catalysing the oxidation process [2].
The thermal decomposition of iron(II) sulphate heptahydrate is observed to be a
four-step process. The first step ranging from 50 to 110°C corresponds to the removal
of six waters of crystallisation according to:
FeSO4 7H2O→FeSO4 H2O+6H2O
(3)
Three overlapping peaks are observed indicating that at least three dehydration
sub-steps are involved in this process. The mass loss for this step, however, corre-
sponds to the loss of only 5 water molecules (Table 2). As the higher hydrates are
fairly unstable [7] and the sample was observed to dry under the dry flowing argon
purge, this reduced water content is not unexpected.
The mass loss at 227°C corresponds well to the removal of a single water molecule:
FeSO4 H2O→FeSO4+H2O
(4)
Further heating results in the decomposition of the sulphate which is generally ob-
served to be a two-step process irrespective of the gas environment [7, 8]. A number
of mechanisms have been proposed for the two-steps including the formation of hy-
droxides, mixed valence oxides and oxide sulphates [7, 8]. The decomposition mech-
anism is however dependent on the purge gas environment. The MS data, using 80
amu SO+3 ion, for this decomposition process indicated that no SO3 was produced.
The MS curves for the 64, 34 and 32 amu ions are shown in Fig. 5. The 64 and 34 amu
peaks are observed to follow the DTG peaks indicating the evolution of SO2 during
both steps of the decomposition. The 32 amu curve only shows a peak for the higher
temperature step indicating that in addition to the production of SO2, O2 is also a
product. The final product of the decomposition, as determined from X-ray diffrac-
Fig. 5 DTG and MS curves for the thermal decomposition of iron(II) sulphate
J. Therm. Anal. Cal., 72, 2003

THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
775
tion, was iron(III) oxide which is in agreement with the literature [7, 8]. As this de-
composition is carried out in an argon atmosphere, the decomposition process is
likely to be:
2FeSO4→Fe2O2SO4+SO2
(5)
(6)
Fe2O2SO4→Fe2O3+SO2+½O2
giving an overall equation for the decomposition as:
2FeSO4→Fe2O3+2SO2 +1/2O2
(7)
The mass loss data confirms the stoichiometry for each of these steps.
A significant difference is, however, observed in the DTG peak positions for
these decomposition reactions relative to the decomposition of the partially degraded
pyrite specimens. Experimental conditions are known to have a significant effect on
the temperature and rate of reactions which has also been observed in the decomposi-
tion of iron(II) sulphate [7]. It is, therefore, not surprising that a crystalline and chem-
ically pure bulk specimen of iron(II) sulphate decomposes at higher temperatures
than the ‘amorphous’ oxidation products of pyrite decay. The correlation between the
temperatures of the final dehydration of crystalline water (Eq. (4)) is, however, con-
sistent for the pyrite specimens, which showed evidence of pyrite disease, indicating
the presence of iron(II) sulphate as an oxidation product. This correlates with the ob-
servation of sulphate hydrates for significantly oxidised pyrite specimens using X-ray
crystallography [1, 3, 4].
Conclusions
This study set out to identify the oxidation products in pyrite specimens which were
showing signs of pyrite disease, but whose oxidation products could not easily be
identified by other means, such as, crystallography. TG coupled with MS has aided
the identification of the oxidation products and confirmed the presence of hydrated
iron(II) sulphates. The mechanisms of decomposition of ACS grade iron(II) sulphate
in an argon atmosphere has been identified using TG-MS, however, the decomposi-
tion mechanisms, of the species produced by pyrite disease, remain uncertain due to
the heterogeneity of the oxidised material.
* * *
The authors would like to thank the Australian Museum, Sydney, for the provision of samples and
advice in their treatment and characterisation.
References
1 A. Newman, Geo. Curator, 6 (1998) 363.
2 B. J. Reedy, J. K. Beattie and R. T. Lowson, Geochim. Cosmochim. Acta, 35 (1991) 1609.
3 S. L. Borek, ACS Symp. Ser., 550 (1994) 31.
J. Therm. Anal. Cal., 72, 2003

776
THOMAS et al.: OXIDATION PRODUCTS OF PYRITE
4 S. Chander, R. Zhou and A. Briceno, Min. Met. Proc., 11 (1994) 141.
5 I. I. Maes, J. Yperman, H. van den Rul, D. V. Franco, J. Mullens and L. C. van Poucke,
Energy and Fuels, 9 (1995) 950.
6 F. Paulik, J. Paulik and M. Arnold, J. Thermal Anal., 25 (1982) 313.
7 L. Odochian, D. M. Pantea, I. Calugareanu, D. Ionescu and O. Vicol, J. Serb. Chem. Soc.,
64 (1999) 737.
8 J. Straszko, M. Olszak-Humienik and J. Mozejko, Inz. Chem. Proc., 17 (1996) 517.
J. Therm. Anal. Cal., 72, 2003