Thermal decomposition and X-ray diffraction of sulphate efflorescent minerals from El Jaroso Ravine, Sierra Almagrera, Spain

Article abstract of DOI:10.1016/j.tca.2007.05.011

Two evaporite minerals from the El Jaroso Ravine, Spain have been analysed by thermogravimetry coupled with an evolved gas mass spectrometer. X-ray diffraction results proved the evaporite minerals were a mixture of sulphates including the minerals magnesiocopiapite, coquimbite and possibly alunogen. Thermal decomposition of the unoxidised samples showed steps at 52, 99 and 143 °C confirmed by mass spectrometric results and attributed to adsorbed water, interstitial water and chemically bonded water. This evaporite mineral rock showed two higher temperature decomposition steps at 555 and 599 °C with mass losses of 19.6 and 7.8%. Slightly different temperatures for the thermal decomposition of the oxadada sample were observed at 52, 64.5 and 100 °C. Two higher temperature mass loss steps at 560.5 and 651 °C were observed for the oxidised sample. By comparison of the thermal analysis patterns of halotrichite and jarosite it can be shown that the El Jaroso samples are mineral sulphates and not halotrichite or jarosite.

Full text of DOI:10.1016/j.tca.2007.05.011

Thermochimica Acta 460 (2007) 9–14
Thermal decomposition and X-ray diffraction of
sulphate efflorescent minerals from El Jaroso
Ravine, Sierra Almagrera, Spain
a,∗
a
a
Ray L. Frost , Daria Wain , Wayde N. Martens ,
a
a,b
a,b
Ashley C. Locke , Jesus Martinez-Frias , Fernando Rull
a
Inorganic Materials Research Program, School of Physical and Chemical Sciences, Queensland University
of Technology, GPO Box 2434, Brisbane, Queensland 4001, Australia
b
Cristalografia y Mineralogia, Unidad Asociada al Centro de Astrobiologia INTA-CSIC,
Universidad de Valladolid, 47006 Valladolid, Spain
Received 11 March 2007; received in revised form 4 May 2007; accepted 7 May 2007
Available online 22 May 2007
Abstract
Two evaporite minerals from the El Jaroso Ravine, Spain have been analysed by thermogravimetry coupled with an evolved gas mass spectrometer.
X-ray diffraction results proved the evaporite minerals were a mixture of sulphates including the minerals magnesiocopiapite, coquimbite and
◦
possibly alunogen. Thermal decomposition of the unoxidised samples showed steps at 52, 99 and 143 C confirmed by mass spectrometric results
and attributed to adsorbed water, interstitial water and chemically bonded water. This evaporite mineral rock showed two higher temperature
◦
decomposition steps at 555 and 599 C with mass losses of 19.6 and 7.8%. Slightly different temperatures for the thermal decomposition of the
◦
◦
oxadada sample were observed at 52, 64.5 and 100 C. Two higher temperature mass loss steps at 560.5 and 651 C were observed for the oxidised
sample. By comparison of the thermal analysis patterns of halotrichite and jarosite it can be shown that the El Jaroso samples are mineral sulphates
and not halotrichite or jarosite.
©
2007 Elsevier B.V. All rights reserved.
Keywords: Evaporite; Jarosite; Sulphate; Dehydration; Dehydroxylation; Halotrichite; TG–MS; High-resolution thermogravimetric analysis
1
. Introduction
tunity has been used to discover the presence of jarosite on Mars,
thus providing evidence for the existence or pre-existence of
water on Mars. Interest in evaporite minerals and their thermal
stability rests with the possible identification of these minerals
and related dehydrated paragenetically related mineral on plan-
ets and on Mars. The existence of these minerals on planets
would give a positive indication of the existence or at least pre-
existence of water on Mars. Further such minerals are formed
through crystallization from solutions.
The jarosite mineral group has been extensively studied [1].
Jarosite was first discovered on Earth in 1852 in ravines in the
mountainous coast of south eastern Spain. Since this time, the
jarosite minerals were mined extensively for silver jarosite. This
has resulted in the formation of evaporite minerals as is found in
the El Jaroso Ravine, Sierra Almagrera, Spain. The importance
of jarosite formation and its decomposition depends upon its
presence in mine tailings, soils, sediments and evaporite deposits
The thermal decomposition of evaporite mineral mixtures
such as those found in the El Jaroso Ravine, Sierra Almagrera,
Spain has not been studied [4–8]. There have been many stud-
ies on related minerals such as the Fe(II) and Fe(III) sulphate
minerals [9–14]. It has been stated that the thermal decomposi-
[
2]. These types of deposits are formed in acid soils where the
pH is less than 3.0 pH units [3]. Such acidification results from
the oxidation of pyrite which may be from bacterial action or
through air-oxidation. The Mars mission rover known as oppor-
◦
tion of jarosite begins at 400 C with the loss of water [1]. The
process is apparently kinetically driven. Water loss can occur at
low temperatures over extended periods of time [1]. It is prob-
able that in nature low temperature environments would result
∗
Corresponding author. Tel.: +61 7 3138 2407; fax: +61 7 3138 1804.
E-mail address: r.frost@qut.edu.au (R.L. Frost).
0
040-6031/$ – see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2007.05.011

1
0
R.L. Frost et al. / Thermochimica Acta 460 (2007) 9–14
in the decomposition of jarosite. The products of the decom-
position depend upon the jarosite be it K, Na or Pb, etc. but
normally goethite and hematite are formed together with sol-
uble sulphates [15]. Recently thermogravimetric analysis has
been applied to some complex mineral systems and it is consid-
ered that TG–MS analyses may also be applicable to the jarosite
minerals [16–21]. Raman spectroscopy has proven very useful
for the study of these types of minerals [22–24]. Indeed Raman
spectroscopy has proven most useful for the study of diagenet-
ically related minerals as often occurs with carbonate minerals
electron probe using energy dispersive techniques for quantita-
tive chemical composition. The Na-jarosite was found to contain
some K but at low concentration levels.
2.2. X-ray diffraction
X-ray diffraction (XRD) patterns were recorded using Cu
˚
K␣ radiation (n = 1.5418 A) on a Philips PANalytical X’Pert
◦
PRO diffractometer operating at 40 kV and 40 mA with 0.125
divergence slit, 0.25 anti-scatter slit, between 3 and 15 (2θ)
◦
◦
◦
◦
at a step size of 0.0167 . For low angle XRD, patterns were
[
25–29]. Some previous studies have been undertaken by the
◦
◦
◦
recorded between 1 and 5 (2θ) at a step size of 0.0167 with
authors using Raman spectroscopy to study complex secondary
minerals formed by crystallization from concentrated sulphate
solutions [30]. In this work we report the XRD and thermogravi-
metric analysis of two evaporite mineral mixtures from the El
Jaroso Ravine, Spain.
◦
variable divergence slit and 0.5 anti-scatter slit.
2.3. Thermal analysis
Thermal decomposition of the evaporate rocks were carried
®
2
. Experimental
out in a TA Instruments incorporated high-resolution ther-
mogravimetric analyzer (series Q500) in a flowing nitrogen
3
2
.1. Minerals
atmosphere (80 cm /min). 34.4 mg of sample underwent ther-
◦
◦
mal analysis, with a heating rate of 5 C/min, resolution of 6 C,
to 1000 C. With the quasi-isothermal, quasi-isobaric heating
◦
The evaporite minerals were collected by two of the authors
(
FR and JM). The minerals originated from El Jaroso Ravine,
program of the instrument the furnace temperature was regu-
lated precisely to provide a uniform rate of decomposition in
the main decomposition stage. The TGA instrument was cou-
pled to a Balzers (Pfeiffer) mass spectrometer for gas analysis.
Only water vapour, carbon dioxide and oxygen were analyzed.
Sierra Almagrera, Spain [see http://tierra.rediris.es/jarosite/].
The two mineral mixture containing samples are labelled as
eflorescencia and eflorescencia oxadada. The evaporite miner-
als were analysed by X-ray diffraction for phase purity and by
Fig. 1. X-ray diffraction patterns of the (a) evaporite minerals samples from the El Jaroso Ravine, Spain together with possible mineral components, (b) halotrichites
from California and Spain together with known standards and (c) jarosite from Spain together with known standards.

R.L. Frost et al. / Thermochimica Acta 460 (2007) 9–14
11
3
. Results and discussion
3
.1. X-ray diffraction
The XRD patterns of the evaporite samples eflorescencia,
halotrichite and Na-jarosite are shown in Fig. 1a–c, respec-
tively. XRD patterns are taken from a powdered sample and
so will be representative of a mixture. Another difficulty is
that the sample does contain amorphous and/or non-diffracting
materials. Therefore, powder XRD may not give the total
story of the efflorescent minerals. Hence this is the funda-
mental reason why techniques such as thermal analysis are
used to analyse these types of mineral mixtures. The XRD
pattern of the eflorescencia matches the sum of the XRD pat-
terns of magnesiocopiapite (MgFe4(SO4) ·20H2O), coquimbite
6
(
Fe2(SO4)3·9H2O) and probably alunogen (Al2(SO4)3·17H2O).
The XRD patterns show the eflorescencia is a complex mixture
of sulphates. Fig. 1a shows the XRD patterns of the oxadada
rock. The analysis is similar to that for the eflorescencia sam-
ple. The difference between the two XRD patterns rests with
the change in the ratio of magnesiocopiapite, coquimbite and
possibly alunogen. For comparison, Fig. 1b shows the XRD
patterns of two halotrichites from Spain and California together
with the reference patterns of pickingerite and apjohnite. The
halotrichite from Spain is a mixture of pickingerite, apjohnite
and heinrichite. The Na-jarosite from Spain is predominantly
Na-jarosite together with some impurities. The XRD patterns
of the Na-jarosite do show resemblance to the patterns of the
evaporite minerals.
Fig. 3. Ion current curves for (a) m/z = 18 and 32, and (b) m/z = 38, 44, 48 and
64 for the thermal decomposition of the eflorescencia sample from the El Jaroso
Ravine, Spain.
3
.2. Thermogravimetric and mass spectrometric analysis of
evaporite minerals
The TG and DTG patterns of the eflorescencia sample are
showninFig. 2. Theioncurrentcurvesarereportedin Fig. 3aand
Fig. 4. TG and DTG patterns for the eflorescencia oxadada sample from the El
Jaroso Ravine, Spain.
(m/z = 18 and 32) and b (m/z = 38, 48, 64, 44). The TG and DTG
of the oxidised sample [eflorescencia oxadada] are shown in
Fig. 4. For comparison the TG and DTG curves for a Na-jarosite
are shown in Fig. 6 and the ion current curves for Na-jarosite in
Fig. 5a and b.
Fig. 5. Ion current curves for (a) m/z = 17 and 18, and (b) m/z = 32, 48 and 64
for the thermal decomposition of the eflorescencia sample from the El Jaroso
Ravine, Spain.
Fig. 2. TG and DTG patterns for the eflorescencia sample from the El Jaroso
Ravine, Spain.

1
2
R.L. Frost et al. / Thermochimica Acta 460 (2007) 9–14
◦
Thermal decomposition steps in the low temperature region
are observed for the eflorescencia sample at 52, 99 and 143 C
Two major mass losses at 555 and 599 C are observed with
mass losses of 19.6 and 7.8%. These thermal decomposition
steps are attributed to the decomposition of sulphate. The ion
current curves for m/z = 48 (SO) and 64 (SO2) show a complex
◦
(
Fig. 2). The mass losses are 6.3, 2.2 and 9.6%. Ion current
◦
curves for m/z = 18 shows a maxima at 119 C and for m/z = 32
◦
◦
at 52, 96 and 142 C. These peaks are attributed to the loss of
pattern with maxima at 547, 602, 637, 659 and 741 C. These
water. The decomposition steps for the efflorescent sample may
be given as follows:
ion current maxima are attributed to the loss of sulphate from
magnesiocopiapite, coquimbite and possible alunogen or some
other sulphate of aluminium.
◦
Step (1) Temperature 53 C:
The TG and DTG patterns of the oxadada sample are shown
in Fig. 4. The thermal decomposition steps and chemical reac-
tions are similar to those above. Three mass loss steps at 52, 64.5
MgFe4(SO4) ·20H2
6
◦
O → MgFe4(SO4) ·xH2O + (20 − x)H2O
and 100 C are attributed to the loss of adsorbed water and water
6
of hydration. Minor mass losses are observed at 176, 260 and
◦
3
58 C. The ion current curves for m/z = 17 and 18 (Fig. 5a) show
Fe2(SO4)3·9H2
◦
maxima at 51.8, 64 and 100 C indicating the loss of water from
the oxadada sample. Another mass loss at 358 C is observed
◦
O → Fe2(SO4)3·yH2O + (9 − y)H2O
which the ion current curves may be ascribed to the loss of water.
Such water molecules must be strongly bonded in the minerals.
This step represents the loss of water or partial loss of
water from the hydrated minerals. The chemical reac-
tions are exemplary only and do not show alunogen
which may not be present or where some amorphous
mineral of a similar formula is present.
◦
Two thermal decomposition steps at 560.5 and 651 C with mass
losses of 28.4 and 7.6% are observed with a higher temperature
◦
mass loss of 2.9% at 736 C. The ion current curves indicate
◦
show some maxima for m/z at 569, 650 and 730 C. These ther-
◦
Step (2) Temperature 99 and 143 C:
mal decomposition steps are attributed to the decomposition of
sulphate.
MgFe4(SO4) ·xH2O → MgFe4(SO4) + xH2O
6
6
3.3. Comparison with the thermal analysis of jarosite and
halotrichite
Fe2(SO4)3·yH2O → Fe2(SO4)3 + yH2O
A comparison may be made with the thermal decomposi-
This step represents the complete loss of water.
Step (3) Temperature 555 and 599 C:
tion of sodium jarosite from a similar origin as the eflorescencia
samples. The reason for this is that the mineral mixture may not
be sufficiently crystalline to give an XRD pattern. Na-jarosite
◦
3+
MgFe4(SO4) → MgO + 2Fe2O3 + 4SO3
6
2 4 4 12
is of a theoretical formula Na (Fe ) (SO ) (OH) . Electron
6
probe analysis showed that the Na-jarosite contained some 17%
K. Thus the formula of the Na-jarosite would be better written
3+
Fe2(SO4)3 → Fe2O3 + 4SO3
6
0.83 2 4 4 12
as (K0.17Na ) (Fe ) (SO ) (OH) . The thermogravimetric
analysis of natural Na-jarosite is shown in Fig. 6. Thermal
◦ ◦
decomposition steps are observed at (a) 50 and 91 C, (b) 214 C,
These thermal decompositions represent the loss of
sulphate as SO3.
It is likely that a second step occurs as follows:
◦
◦
(c) 375 C and (d) 541 C. The ion current curves (Fig. 5a and
b) for Na-jarosite shows that water vapour is the evolved gas at
the first four temperatures and SO2 is lost at the higher temper-
◦
SO3 → SO2 + (1/2)O2
atures. There is a 2.7% mass loss due to water loss at 214 C.
◦
There is a mass loss of 19.6% at 555 C and 7.8% at
◦
5
99 C.
◦
Step (4) Temperature 743 C:
Fe2O3 → 2FeO + (1/2)O2
At this temperature the hematite loses oxygen.
The next decomposition step is observed at around
◦
◦
2
70 C. A corresponding ion current peak at 267 C for
m/z = 32 is observed. The XRD patterns show the pres-
ence of magnesiocopiapite (MgFe4(SO4) ·20H2O), coquimbite
6
(
Fe2(SO4)3·9H2O) and possibly alunogen (Al2(SO4)3·17H2O)
andthusthelossofsignificantamountsofwateristobeexpected.
Fig. 6. TG and DTG patterns of Na-jarosite from the El Jaroso Ravine, Spain.

R.L. Frost et al. / Thermochimica Acta 460 (2007) 9–14
13
4. Conclusions
Thermogravimetric analysis coupled to a mass spectrome-
ter has been used to study selected mineral samples from the
El Jaroso Ravine, Spain. Two samples were analysed firstly an
unoxidised sample labelledas eflorescencia andsecondly an oxi-
dised sample labelled as eflorescencia oxadata. A comparison
is made with the thermal decomposition of a sodium jarosite.
X-ray diffraction showed the sample was a complex mixture
including the sulphates of magnesiocopiapite, coquimbite and
possibly alunogen or some sulphate of aluminium. A difficulty
of analysing such mineral materials is that the sample contains
amorphous and/or non-diffracting materials. Therefore powder
XRD may not give the complete story of the efflorescent min-
erals. Hence, this is the reason why other techniques such as
thermal analysis may be used to analyse these types of mineral
mixtures.
Fig. 7. TG and DTG patterns of a natural halotrichite.
This water may be described as “excess water” [31,32]. It is
possible that this water results from the deprotonation of the
Na-jarosite. Jarosites have a hydroxyl surface which is capable
of adsorbing protons creating a hydrogen-jarosite. Thus the Na-
jarosite is a sodium protonated jarosites and the formula may be
Thermal analysis coupled to evolved gas mass spectrome-
try showed that the TG of the unoxidised samples had thermal
◦
decomposition steps at 52, 99 and 143 C confirmed by the mass
spectrometric results which are attributed to adsorbed water,
interstitial water and chemically bonded water. The eflores-
cencia sample showed two higher temperature decomposition
3+
written as: (HxK0.17 − (x/2)Na0.83 − (x/2))2(Fe ) (SO4)4(OH)12.
6
The theoretical mass loss of Na-jarosite based upon the
3+
◦
formula Na2(Fe ) (SO4)4(OH)12 is 11.13%. If the ini-
steps at 555 and 599 C with mass losses of 19.6 and 7.8%.
6
tial 4% mass loss is neglected the mass loss at 316
and 352 C is 10%. This value is close to the theoret-
Slightly different temperatures of the thermal decomposition
◦
◦
of the oxadada sample are observed at 52, 64.5 and 100 C.
◦
ical value. The reaction may be summarised as follows:
Two higher temperature mass loss steps at 560.5 and 651 C
3+
Na2(Fe ) (SO4)4(OH)12 → 2NaFe(SO4)2 + 2Fe2O3 + 6H2O.
are observed for the oxidised sample. The latter value is signif-
icantly higher than for the unoxidised sample. A comparison of
the thermal decomposition of jarosite and halotrichite confirms
that the evaporite minerals are not jarosite or halotrichite but are
a mixture of sulphates.
6
The thermogravimetric and differential thermogravimetric
analysis of halotrichite are shown in Fig. 7. Seven major ther-
mal decomposition steps are observed. Steps (1)–(4) occur at
◦
temperatures of 53, 72 and 330 C. A mass loss of 14.28%
◦
is observed at quite low temperatures around or up to 42 C.
A further mass loss of 9.20% occurs at 53 C and 18.91%
up to 300 C. A small mass loss of 3.18% is observed at
30 C. Each of these thermal decomposition steps is attributed
◦
Acknowledgments
◦
◦
3
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 Council
to dehydration of the halotrichite. The ion current curves for
◦
halotrichiteofH2OandOHunitsshowmaximaat72and330 C.
Using the formula (Fe0.752 ,Mg0.25)SO4·Al2(SO4)3·22H2O
the total theoretical mass loss for water is 44.90%. The
total experimental mass loss is 45.57% which is in good
agreement.
+
(ARC) is thanked for funding the thermal analysis facility.
References
Steps (4)–(8) occur at temperatures of 546, 625, 697 and
◦
7
38 C. Four thermal decomposition steps are observed for
[
[
[
1] J.E. Dutrizac, J.L. Jambor, Jarosites and their application in hydrometal-
lurgy, 2000, p. 405 (Chapter 8).
2] T. Buckby, S. Black, M.L. Coleman, M.E. Hodson, Mineral. Mag. 67(2003)
263.
3] P.A. Williams, Oxide Zone Geochemistry, Ellis Horwood Ltd., Chichester,
West Sussex, England, 1990.
◦
halotrichite of the above formula at 546, 625, 697 and 738 C.
Each of these steps is assigned to the decomposition of sulphate
anions, the formation of the metal oxides and the evolution of
SO2 and O2. The mass losses at these temperatures are 1.69,
1
9.03, 7.72 and 7.21% making a total of mass loss of 35.65%.
[4] S. Nagai, N. Yamanouchi, Nippon Kagaku Kaishi (1921-47) 52 (1949) 83.
[5] J.L. Kulp, H.H. Adler, Am. J. Sci. 248 (1950) 475.
[6] G. Cocco, Period. Mineral. 21 (1952) 103.
The theoretical mass loss calculated using the above formula
is 36.29%. The measured mass loss is in agreement with the
theoretical mass loss. This agreement confirms the formula of
halotrichite. If the formula was different then the numbers would
not be in such close agreement. These decomposition steps are
confirmed by the ion current curves of the evolved gases SO2
[7] A.I. Tsvetkov, E.P. Val’yashikhina, Doklady Akademii Nauk SSSR 89
1953) 1079.
8] A.I. Tsvetkov, E.P. Val’yashikhina, Doklady Akademii Nauk SSSR 93
1953) 343.
9] M.S.R. Swamy, T.P. Prasad, B.R. Sant, J. Therm. Anal. 16 (1979) 471.
(
[
[
(
◦
and O2 where ion current maxima at 546, 625, 697 and 738 C
[
10] M.S.R. Swamy, T.P. Prasad, B.R. Sant, J. Therm. Anal. 15 (1979) 307.
are observed.
[11] S. Bhattacharyya, S.N. Bhattacharyya, J. Chem. Eng. Data 24 (1979) 93.

1
4
R.L. Frost et al. / Thermochimica Acta 460 (2007) 9–14
[
[
[
12] M.S.R. Swami, T.P. Prasad, J. Therm. Anal. 19 (1980) 297.
13] M.S.R. Swamy, T.P. Prasad, J. Therm. Anal. 20 (1981) 107.
14] A.C. Banerjee, S. Sood, Therm. Anal. Proc. Int. Conf. 7th, vol. 1, 1982, p.
[23] R.L. Frost, D.A. Henry, M.L. Weier, W. Martens, J. Raman Spectrosc. 37
(2006) 722.
[24] R.L. Frost, A.W. Musumeci, J.T. Kloprogge, M.O. Adebajo, W.N. Martens,
J. Raman Spectrosc. 37 (2006) 733.
7
69.
[
15] P.S. Thomas, D. Hirschausen, R.E. White, J.P. Guerbois, A.S. Ray, J.
Therm. Anal. Calorim. 72 (2003) 769.
[25] R.L. Frost, J. Cejka, M. Weier, W.N. Martens, J. Raman Spectrosc. 37
(2006) 879.
[
[
16] R.L. Frost, K.L. Erickson, J. Therm. Anal. Calorim. 76 (2004) 217.
17] R.L. Frost, K. Erickson, M. Weier, J. Therm. Anal. Calorim. 77 (2004)
[26] R.L. Frost, M.L. Weier, J. Cejka, J.T. Kloprogge, J. Raman Spectrosc. 37
(2006) 585.
8
51.
18] R.L. Frost, M.L. Weier, K.L. Erickson, J. Therm. Anal. Calorim. 76 (2004)
025.
[27] R.L. Frost, J. Cejka, M.L. Weier, W. Martens, J. Raman Spectrosc. 37
(2006) 538.
[28] R.L. Frost, M.L. Weier, B.J. Reddy, J. Cejka, J. Raman Spectrosc. 37 (2006)
816.
[
1
[
[
19] R.L. Frost, M.L. Weier, J. Therm. Anal. Calorim. 75 (2004) 277.
20] R.L. Frost, W. Martens, Z. Ding, J.T. Kloprogge, J. Therm. Anal. Calorim.
[29] R.L. Frost, M.L. Weier, W.N. Martens, J.T. Kloprogge, J. Kristof, J. Raman
Spectrosc. 36 (2005) 797.
[30] R.L. Frost, R.-A. Wills, M.L. Weier, W. Martens, J. Raman Spectrosc. 36
(2005) 435.
[31] C. Drouet, A. Navrotsky, Geochim. Cosmochim. Acta 67 (2003) 2063.
[32] J. Kubisz, Mineral. Polon. 2 (1971) 51.
7
1 (2003) 429.
21] R.L. Frost, Z. Ding, H.D. Ruan, J. Therm. Anal. Calorim. 71 (2003)
83.
22] R.L. Frost, S.J. Palmer, J.M. Bouzaid, B.J. Reddy, J. Raman Spectrosc. 38
2007) 68.
[
[
7
(