Thermal decomposition of natural and synthetic plumbojarosites: Importance in 'archeochemistry'

Article abstract of DOI:10.1016/j.tca.2005.04.001

Plumbojarosite and argentoplumbojarosite were sources of lead and silver in ancient and medieval times. The understanding of the chemistry of the thermal decomposition of these minerals is of vital importance in 'archeochemistry'. The thermal decomposition of plumbojarosite was studied using a combination of thermogravimetric analysis coupled to a mass spectrometer. Three mass loss steps are observed at 376, 420 and 502°C. These are attributed to dehydroxylation, loss of sulphate occurs at 599°C, and loss of oxygen and formation of lead occurs at 844 and 953°C. The temperatures of the thermal decomposition of the natural jarosite were found to be less than that for the synthetic jarosite. This is attributed to a depression of freezing point effect induced by impurities in the natural jarosite. Raman spectroscopy was used to study the structure of plumbojarosite. Plumbojarosites are characterised by stretching bands at 1176, 1108, 1019 and 1003 cm^-1 and bending modes at 623 and 582cm^-1. Changes in the molecular structure during thermal decomposition were followed by infrared emission spectroscopy. The technique shows the loss of intensity in the hydroxyl stretching region attributed to dehydroxylation. Loss of sulphate only occurs after dehydroxylation. Lead is formed at higher temperatures through oxygen evolution.

Full text of DOI:10.1016/j.tca.2005.04.001

Thermochimica Acta 432 (2005) 30–35
Thermal decomposition of natural and synthetic plumbojarosites:
Importance in ‘archeochemistry’
Ray L. Frost∗, Rachael-Anne Wills, Matt L. Weier,
Anthony W. Musumeci, Wayde Martens
Inorganic Materials Research Program, School of Physical and Chemical Sciences, Queensland University of Technology,
GPO Box2434, Brisbane, Qld 4001, Australia
Received 31 January 2005; received in revised form 29 March 2005; accepted 8 April 2005
Available online 10 May 2005
Abstract
Plumbojarosite and argentoplumbojarosite were sources of lead and silver in ancient and medieval times. The understanding of the chemistry
of the thermal decomposition of these minerals is of vital importance in ‘archeochemistry’. The thermal decomposition of plumbojarosite was
studied using a combination of thermogravimetric analysis coupled to a mass spectrometer. Three mass loss steps are observed at 376, 420
and 502 ◦C. These are attributed to dehydroxylation, loss of sulphate occurs at 599 ◦C, and loss of oxygen and formation of lead occurs at 844
and 953 ◦C. The temperatures of the thermal decomposition of the natural jarosite were found to be less than that for the synthetic jarosite.
This is attributed to a depression of freezing point effect induced by impurities in the natural jarosite. Raman spectroscopy was used to study
the structure of plumbojarosite. Plumbojarosites are characterised by stretching bands at 1176, 1108, 1019 and 1003 cm−1 and bending modes
at 623 and 582 cm−1. Changes in the molecular structure during thermal decomposition were followed by infrared emission spectroscopy.
The technique shows the loss of intensity in the hydroxyl stretching region attributed to dehydroxylation. Loss of sulphate only occurs after
dehydroxylation. Lead is formed at higher temperatures through oxygen evolution.
© 2005 Elsevier B.V. All rights reserved.
Keywords: Jarosite; Plumbojarosite; Dehydration; Dehydroxylation; Infrared emission spectroscopy; Thermogravimetric analysis; Raman spectroscopy
1. Introduction
archaeological science [7,8] and are also found in mine
drainage sites both ancient and modern [6,8,9]. Such forma-
tion of jarosites has been occurring since the bronze age [10].
The importance of jarosite formation and its decomposition
depends upon its presence in soils, sediments and evaporate
deposits [11]. These types of deposits have formed in acid
soils where the pH is less than 3.0 pH units [12]. Such acid-
ification results from the oxidation of pyrite which may be
from bacterial action or through air-oxidation.
The thermal decomposition of jarosites has been studied
for some time [13–17]. Some studies of Pb-jarosite has been
undertaken [14]. It was shown that decomposition occurred
at 500 and 700 ◦C [14]. A later study showed that Pb-jarosite
dehydrated at 450 ◦C and lost SO2 at temperatures >550 ◦C
with total decomposition at 950 ◦C [6]. A later study reported
the kinetics of decomposition of Pb-jarosite with decompo-
sition occurring in three stages at 490, 750 and 950 ◦C [18].
The jarosite mineral group has been extensively studied
[1]. Jarosite was first discovered in 1852 in ravines in the
mountainous coast of southeastern Spain. Jarosite on Mars
was discovered in 2004. The Mars mission rover known as
Opportunity has been used to discover the presence of jarosite
on Mars, thus providing evidence for the existence or pre-
existence of water on Mars. (http://www.news.cornell.edu/
releases/rover/Mars.jarosite.html)
Lead jarosite also known as plumbojarosite (PbFe6(SO4)4
(OH)12) was identified in relation to jarosite in 1902 [2].
Plumbojarosite is often found in cationic mixed jarosites
[3–6]. Such minerals are of importance in medieval and
∗
Corresponding author. Tel.: +61 7 3864 2407; fax: +61 7 3864 1804.
E-mail address: r.frost@qut.edu.au (R.L. Frost).
0040-6031/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2005.04.001

R.L. Frost et al. / Thermochimica Acta 432 (2005) 30–35
31
It has been stated that the thermal decomposition of jarosite
begins at 400 ◦C with the loss of water [1]. Water loss can
occur at low temperatures over extended periods of time [1].
It is probable that in nature low temperature environments
would result in the decomposition of jarosite. The products
of the decomposition depend upon the particular jarosite be
it K, Na or Pb, etc., but normally goethite and hematite are
formed together with soluble sulphates [19]. Recently ther-
mogravimetric analysis has been applied to some complex
mineral systems and it is considered that TG-MS analyses
may also be applicable to the jarosite minerals [20–25]. In
this work we report the thermal decomposition of natural
and synthetic plumbojarosites.
˚
scan mode, with Cu K␣ radiation (1.54052 A). Patterns were
collected in the range 3–90 ◦ 2s with a step size of 0.02◦ and
a rate of 30 s per step. Samples were prepared as a finely
pressed powder into aluminium sample holders. The Pro-
file Fitting option of the software uses a model that employs
twelve intrinsic parameters to describe the profile, the instru-
mental aberration and wavelength dependent contributions to
the profile.
2.4. Infrared emission spectroscopy
FT-IR emission spectroscopy was carried out on a Nicolet
spectrophotometer equipped with a TGS detector, which was
modified by replacing the IR source with an emission cell. A
description of the cell and principles of the emission experi-
ment have been published elsewhere. Approximately 0.2 mg
of the plumbojarosite mineral was spread as a thin layer (ap-
proximately 0.2 microns) on a 6 mm diameter platinum sur-
face and held in an inert atmosphere within a nitrogen-purged
cell during heating.
In the normal course of events, three sets of spectra are
obtained: first the black body radiation over the tempera-
ture range selected at the various temperatures, secondly
the platinum plate radiation at the same temperatures and
thirdly spectra from the platinum plate coated with the sam-
ple. Normally only one set of black body and platinum ra-
diation data is required. The emittance spectrum (E) at a
particular temperature was calculated by subtraction of the
single beam spectrum of the platinum backplate from that
of the platinum + sample, and the result rationed to the sin-
gle beam spectrum of an approximate blackbody (graphite).
The following equation was used to calculate the emission
spectra.
2. Experimental
2.1. Minerals
To synthesise lead jarosite a 50 mL chloride solution was
prepared. This solution contained 0.5 g PbCl2, 12 mL of
a saturated LiCl solution and 5 mL of 1.23 M FeCl3 so-
lution. (the high chloride concentration was necessary to
prevent the precipitation of lead sulfate). 4.93 g Fe2(SO4)3
was dissolved in the minimal volume of water and slowly
added dropwise to the chloride solution. The final solution
was heated for 21 h at 120 ◦C in an autoclave. A golden
brown precipitate was collected under vacuum and then
dried at 100 ◦C for an hour. 1.25 g of pure lead jarosite was
obtained.
The natural Pb-jarosite originated from Teutonic Bore
Mine, Western Australia. The jarosite minerals were analysed
by X-ray diffraction for phase purity and by electron probe
using energy dispersive techniques for quantitative chemi-
cal composition. The natural plumbojarosite contained 5%
potassium. No other cations were found.
Pt − S
E = −0.5 × log
(1)
Pt − C
2.2. Thermal analysis
This manipulation is carried out after all the data is col-
lected. Emission spectra were collected at intervals of 50 ◦C
over the range 200–750 ◦C. The time between scans (while
the temperature was raised to the next hold point) was ap-
proximately 100 s. It was considered that this was sufficient
time for the heating block and the powdered sample to reach
thermal equilibrium. Spectra were acquired by 1064 scans
over the temperature range 100–300 ◦C and 128 scans over
the range 350–900 ◦C (approximate scan time 45 s), with a
Thermal decomposition of the plumbojarosites was car-
ried out in a TA® Instruments incorporated high-resolution
thermogravimetric analyser (series Q500) in a flowing nitro-
gen atmosphere (80 cm3/min). 34.4 mg of sample underwent
thermal analysis, with a heating rate of 5 ◦C/min, resolution
of 6–1000 ◦C. With the quasi-isothermal, quasi-isobaric heat-
ing program of the instrument the furnace temperature was
regulated precisely to provide a uniform rate of decomposi-
tion in the main decomposition stage. The TGA instrument
was coupled to a Balzers (Pfeiffer) mass spectrometer for gas
analysis. Only water vapour, sulphur dioxide, sulphur triox-
ide, carbon dioxide and oxygen were analysed.
nominal resolution of 4 cm−1
.
Good quality spectra can be obtained providing the sam-
ple thickness is not too large. If too large a sample is
used then spectra become difficult to interpret because of
the presence of combination and overtone bands. Spectro-
scopic manipulation such as baseline adjustment, smooth-
ing and normalisation was performed using the GRAMS®
software package (Galactic Industries Corporation, Salem,
NH, USA).
2.3. X-ray diffraction
X-ray diffraction patterns were collected using a Philips
X’pert wide angle X-ray diffractometer, operating in step

32
R.L. Frost et al. / Thermochimica Acta 432 (2005) 30–35
3. Results and discussion
3.1. X-ray diffraction
The X-ray diffraction pattern for the synthetic plumbo-
jarosite and the reference pattern (39-1353) for the natural
plumbojarosite are shown in Fig. 1. Clearly the synthetic
plumbojarosite corresponds well with the natural sample with
no impurities that can be detected using the XRD technique.
3.2. Thermogravimetric analysis and mass
spectrometric analysis
TheTGandDTGcurvesforasyntheticandnaturaljarosite
are shown in Fig. 2a and b, respectively. The ion current
curves for the evolved gases water vapour and SO3 from the
synthetic plumbojarosite are shown in Fig. 3a and b. In terms
of DTG patterns, there is a striking similarity between that
of the synthetic and natural plumbojarosites. There is an ap-
parent lowering of the temperatures of decomposition for the
natural plumbojarosite. This may be caused by the presence
of other elements in the natural plumbojarosite. EDX analysis
shows the presence of around 5% K in the natural plumbo-
jarosite.
The ion current curves for evolved water vapour show
water is evolved over a broad temperature range from 350
to 450 ◦C. Thus the mass losses that occur at 376, 420 and
502 ◦C for the synthetic plumbojarosite and 391 and 418 ◦C
for the natural sample are attributed to dehydroxylation. The
theoretical mass loss for dehydroxylation is 9.56% based
upon the formula (PbFe6(SO4)4(OH)12) with the loss of 6 hy-
droxyl units as 3 H2O. The experimentally determined mass
losses for the synthetic and natural plumbojarosites are 11.8
and 11.2 %.
Fig. 2. (a) TG and DTG curves for synthetic plumbojarosite; (b) TG and
DTG curves for natural plumbojarosite.
The ion current curve for evolved SO3 (Fig 3b) shows that
the mass loss for the synthetic plumbojarosite at 599 ◦C is
due to the loss of sulphate as SO2. The temperature for the
natural plumbojarosite is 531 ◦C. The difference in tempera-
ture is attributed to a ‘depression of freezing point effect’, i.e.
a lowering of the temperature of the loss of sulphate by the
Fig. 3. (a) Ion current for masses 17 and 18 and (b) ion current for masses
44, 48 and 64.
Fig. 1. XRD pattern for synthetic plumbojarosite and the reference pattern.

R.L. Frost et al. / Thermochimica Acta 432 (2005) 30–35
33
presence of impurities in the natural sample. The experimen-
tal mass loss is 25%. If all four sulphate units are lost simul-
taneously then the theoretical mass loss would be 22.68%.
Higher temperature mass losses are observed at 953 and
844 ◦C for the synthetic plumbojarosite and at 812 and 759 ◦C
for the natural sample. These mass loss steps are attributed to
the loss of oxygen. XRD shows the decomposition products
of plumbojarosite are lead and hematite.
3.3. Mechanism of thermal decomposition
It is proposed that the thermal decomposition of plumbo-
jarosite occurs in three stages as follows: (a) dehydroxylation,
(b) loss of sulphates and (c) loss of oxygen.
Fig. 4. Comparison of the Raman spectra of natural and synthetic plumbo-
jarosite.
• Stage 1
A previous study by Sasaki reported the two Pb-jarosite
symmetric stretching band at 1015 and 1002 cm−1 which
is in excellent agreement with the value reported in this re-
search [26]. The two Raman bands at 1099 and 1062 cm−1 are
assigned to the SO24− antisymmetric stretching modes. The
two equivalent infrared bands are the two bands at 1182 and
1109 cm−1. Sasaki reported the Raman bands for Pb-jarosite
as 1107 and 1161 cm−1 in the Raman spectrum and 1080,
1092 and 1195 cm−1 in the infrared spectrum [26]. The band
• PbFe6(SO4)4(OH)12 → PbFe6(SO4)4O6 + 6H2O
• Theoretical mass loss of 9.56%.
• Experimental mass loss 11.8%.
• Temperature of mass loss 376 and 420 ◦C.
• Stage 2
• Pb(FeO)6(SO4)4 → PbO3.5(FeO)6(SO4)0.5 + 3.5SO3
• Theoretical mass loss of 24.77%.
• Experimental mass loss for the synthesised plumbojarosite
is 25%.
• Temperature of mass loss is 599 ◦C.
• Stage 3
• PbO3.5(FeO)6(SO4)0.5 → Pb + 3Fe2O3 + 0.5O2 + 0.5SO3
• Theoretical mass loss of 4.96%.
• Experimental mass loss for the synthesised plumbojarosite
is 5.6%.
• Temperature of mass loss is 844 and 953 ◦C.
3.4. Raman and infrared spectroscopy
One technique for studying the changes in the molecular
structure during thermal decomposition is infrared emission
spectroscopy. The technique enables the spectrum at the el-
evated temperature to be determined. In order to understand
the spectrum at the elevated temperature it is necessary to un-
derstand the room temperature spectra. A comparison of the
Raman spectra of the natural and synthetic plumbojarosite are
shown in Fig. 4. The Raman and infrared spectra of the lead
jarositeareshowninFig. 5. Sulphatesaswithotheroxyanions
lend themselves to analysis by Raman spectroscopy. In aque-
ous systems, the sulphate anion is of Td symmetry and is char-
acterised by Raman bands at 981 cm−1 (ν1), 451 cm−1 (ν2),
1104 cm−1 (ν3) and 613 cm−1 (ν4). Reduction in symme-
try in the crystal structure of sulphates such as jarosites will
cause the splitting of these vibrational modes. For jarosites
the space group is C53 and six sulphate fundamentals should
be observed. The Raman band at 1036 cm−1 is attributed to
the SO24−symmetric stretching mode. The equivalent infrared
band is forbidden and is not observed.
Fig. 5. (a) Raman and infrared spectra and their component bands of syn-
thetic plumbojarosite in the 520–1220 cm−1 region and (b) Raman and in-
frared spectra and their component bands of synthetic plumbojarosite in the
hydroxyl stretching region.

34
R.L. Frost et al. / Thermochimica Acta 432 (2005) 30–35
Fig. 7. IES of plumbojarosite in the 550–1850 cm−1 region.
Fig. 6. IES of plumbojarosite in the hydroxyl stretching region.
at 619 cm−1 in the infrared spectrum and at 674 cm−1 in the
Raman spectrum is ascribed to the ν4 bending mode of the
SO24−units. The previous study, which reported the Raman
spectrum of Pb-jarosite gave two bands at 623 and 668 cm−1
in the infrared spectrum and at 623 cm−1 in the Raman spec-
trum [26]. In the Raman spectrum bands at 424 and 454 cm−1
are attributed to the ν2 bending modes of the SO42− units.
Four bands are identified at 1194.8, 1106.4, 1073.4 and
1023.0 cm−1. These bands are attributed to the sulphate anti-
symmetric stretching modes. In the 450 ◦C spectrum, a single
broad band is found at 1092 cm−1. No changes are observed
in the spectra until after dehydroxylation has taken place.
4. Conclusions
3.5. Infrared emission spectroscopy
Understanding the chemistry of plumbojarosite and ar-
gentoplumbojarosite is of importance in archeochemistry.
Plumbojarosite was a source of lead in pre- and Roman times.
Argentoplumbojarosite was the major source of silver. It has
been stated that much of the wealth of colonial Spain orig-
inated from the obtaining of silver from silver containing
jarosites. Indeed the wealth of South American civilisations
depended on silver. Such lead and silver could only be formed
by heating the plumbojarosite and argentoplumbojarosite.
For the understanding of the archeochemistry of jarosites it
is important to undertake thermal studies of these minerals.
Thermal studies provide an indication of the temperatures
that ancient civilisations would need to achieve in order to
obtain lead and or silver.
Thermal decomposition of the lead jarosite may also be
followed using infrared emission spectroscopy in which the
infrared spectrum can be obtained in situ at the elevated tem-
perature. The infrared emission spectra of synthetic plumbo-
jarosite are shown in Figs. 6 and 7. The figure shows that the
intensity of the OH stretching region decreases as the tem-
perature is raised. No intensity remains after 500 ◦C. Three
bands are observed in the infrared spectra at 100 ◦C at 3400.6,
3250.2 and 2992.9 cm−1. In the 500 ◦C spectrum only a sin-
gle band at 3503.3 cm−1 is found. The IES data shows that
dehydroxylation is complete by 500 ◦C. Such a result is in
harmony with the results of the DTG. DTG showed three
mass loss steps 376, 420 and 502 ◦C.
The IES data show that the spectra of the sulphate
The thermal decomposition of a natural and synthetic
plumbojarosite has been studied using thermogravimetric
stretching region remains unchanged over the 100–400 ◦C.

R.L. Frost et al. / Thermochimica Acta 432 (2005) 30–35
35
analysis and mass spectrometry. The temperatures of the
thermal decomposition of the natural jarosite were found
to be less than that for the synthetic jarosite. This is at-
tributed to a depression of freezing point effect induced by
impurities in the natural jarosite. The implication is that
lead would be more easily obtained from the plumbojarosite
if the mineral was impure which is more often the case
than not.
[6] T. Taberdar, H. Gulensoy, A.O. Aydin, Marmara Universitesi Fen
Bilimleri Dergisi 2 (1985) 76.
[7] J.L. Amoros, R. Lunar, P. Tavira, Mineralium Deposita 16 (1981)
205.
[8] C. Rewitzer, R. Hochleitner, Rivista Mineralogica Italiana (1989) 83.
[9] D.L. Harris, B.G. Lottermoser, J. Duchesne, Aust. J. Earth Sci. 50
(2003) 797.
[10] K.A. Hudson-Edwards, C. Schell, M.G. Macklin, Appl. Geochem.
14 (1999) 1015.
[11] T. Buckby, S. Black, M.L. Coleman, M.E. Hodson, Mineral. Mag.
67 (2003) 263.
[12] P.A. Williams, Oxide Zone Geochemistry, Ellis Horwood Ltd.,
Chichester, West Sussex, UK, 1990.
Acknowledgements
[13] S. Nagai, N. Yamanouchi, Nippon Kagaku Kaishi 52 (1949) 83,
1921–47.
[14] J.L. Kulp, H.H. Adler, Am. J. Sci. 248 (1950) 475.
[15] G. Cocco, Periodico di Mineralogia 21 (1952) 103.
[16] A.I. Tsvetkov, E.P. Val’yashikhina, Doklady Akademii Nauk SSSR
89 (1953) 1079.
[17] A.I. Tsvetkov, E.P. Val’yashikhina, Doklady Akademii Nauk SSSR
93 (1953) 343.
[18] M. Ozacar, A. Alp, A.O. Aydin, J. Thermal Anal. Calorim. 59 (2000)
869.
The financial and infra-structure support of the Queens-
land University of Technology Inorganic Materials Research
Program of the School of Physical and Chemical Sciences
is gratefully acknowledged. The Australian Research Coun-
cil (ARC) is thanked for funding. The authors wish to thank
and gratefully acknowledge the support of Dr. Allan Pring of
The South Australian Museum for the loan/donation of the
plumbojarosite minerals used in this study.
[19] P.S. Thomas, D. Hirschausen, R.E. White, J.P. Guerbois, A.S. Ray,
J. Thermal Anal. Calorim. 72 (2003) 769.
[20] R.L. Frost, K.L. Erickson, J. Therm. Anal. Calorim. 76 (2004) 217.
[21] R.L. Frost, K. Erickson, M. Weier, J. Therm. Anal. Calorim. 77
(2004) 851.
References
[22] R.L. Frost, M.L. Weier, K.L. Erickson, J. Therm. Anal. Calorim. 76
(2004) 1025.
[23] R.L. Frost, M.L. Weier, J. Therm. Anal. Calorim. 75 (2004) 277.
[24] R.L. Frost, W. Martens, Z. Ding, J.T. Kloprogge, J. Therm. Anal.
Calorim. 71 (2003) 429.
[25] R.L. Frost, Z. Ding, H.D. Ruan, J. Therm. Anal. Calorim. 71 (2003)
783.
[26] K. Sasaki, O. Tanaike, H. Konno, Can. Mineral. 36 (1998) 1225.
[1] J.E. Dutrizac, J.L. Jambor, Jarosites and their Application in Hy-
drometallurgy, The Mineralogical Society of America, 2000. p. 405
(Chapter 8).
[2] W.F. Hilebranad, F.E. Wright, Am. J. Sci. 30 (1910) 191.
[3] F.I. Leach, Mining J. (Phoenix) 20 (1937) 40.
[4] W.G. Mumme, T.R. Scott, Am. Mineral. 51 (1966) 443.
[5] J.E. Dutrizac, O. Dinardo, S. Kaiman, Hydrometallurgy 5 (1980)
305.