Quick
Search: 		  
advanced search
 GSW Home   GeoRef Home   My GSW Alerts   Contact GSW   About GSW   Journals List   Help 
American Mineralogist Signup for GSW Email News
JOURNAL HOME HELP CONTACT PUBLISHER SUBSCRIBE ARCHIVE SEARCH TABLE OF CONTENTS
American Mineralogist; November-December 2009; v. 94; no. 11-12; p. 1629-1637; DOI: 10.2138/am.2009.3182
© 2009 Mineralogical Society of America
This Article
Right arrow Abstract
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Xu, W.
Right arrow Articles by Parise, J. B.
GeoRef
Right arrow GeoRef Citation

Humidity-induced phase transitions of ferric sulfate minerals studied by in situ and ex situ X-ray diffraction

Wenqian Xu1, Nicholas J. Tosca2, Scott M. McLennan1 and John B. Parise1,3,*

1 Department of Geosciences, Stony Brook University, Stony Brook, New York 11794-2100, U.S.A.
2 Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, Massachusetts 02138, U.S.A.
3 Department of Chemistry, Stony Brook University, Stony Brook, New York 11794-3400, U.S.A.

Correspondence: * E-mail: john.parise{at}stonybrook.edu


   ABSTRACT
Top
 Abstract
Introduction
 Experimental methods
 Results
 Discussion
 Acknowledgments
 References cited
 
Phases encountered in the hydration of monoclinic and trigonal anhydrous Fe2(SO4)3 and evaporation of Fe2(SO4)3 solutions at room temperature were determined using in situ and ex situ X-ray diffraction (XRD) under dynamic relative humidity (RH) control at room temperature (22–25 °C). Both monoclinic and trigonal forms of Fe2(SO4)3 remain anhydrous at 11% RH or below, and undergo the following phase evolution sequence: anhydrous Fe2(SO4)3 -> (ferricopiapite, rhomboclase) -> kornelite -> paracoquimbite at RH between 33 and 53% as a function of time. Evaporation of aqueous Fe2(SO4)3 solutions at 40% < RH < 60% results in precipitation of ferricopiapite and rhomboclase during evaporation, followed by a transition to kornelite and then paracoquimbite. Evaporation at RH < 33% produced an amorphous ferric-sulfate phase. The presence of some iron sulfate hydrates and their stability under varying RH are not only determined by the final humidity level, but also the intermediate stages and hydration history (i.e., either ferricopiapite or paracoquimbite can be a stable phase at 62% RH depending on the hydration history). The sensitivity to humidity change and path-dependent transitions of ferric sulfates make them potentially valuable indicators of paleo-environmental conditions and past water activity on Mars. The phase relationships reported herein can help in understanding the diagenesis of ferric sulfate minerals, and are applicable to geochemical modeling of mineral solubility in multi-component systems, an endeavor hindered by the need for fundamental laboratory studies of iron sulfate hydrates.

Key Words: Ferric sulfate • humidity • ferricopiapite • rhomboclase • kornelite • paracoquimbite


   INTRODUCTION
Top
 Abstract
 Introduction
Experimental methods
 Results
 Discussion
 Acknowledgments
 References cited
 
An exciting result of recent Mars exploration is the observation of hydrated minerals (including hydrous sulfate minerals and phyllosilicates) that constitute an important fraction of sedimentary minerals on Mars (Bibring et al. 2005; Gendrin et al. 2005; McLennan and Grotzinger 2008; Poulet et al. 2005; Squyres et al. 2004). The presence of these hydrated minerals implies water once played a significant role in martian surficial evolution. Detailed understanding of the formation, stability, and post-depositional evolution of these minerals is crucial in reconstructing ancient martian environments.

Among these hydrated minerals, iron sulfates are capable of yielding uniquely detailed information on the environmental conditions of the past and present martian surface, because their stability is governed by important variables such as pH, redox conditions, temperature (T), and relative humidity (RH). To date, several iron sulfate minerals have been confirmed or strongly suspected at various places at the Spirit and Opportunity sites from Mössbauer or infrared spectroscopy data, including jarosite [(Na,K,H3O)+Fe3(SO4)2(OH)6], rhomboclase [(H3O+)Fe(SO4)2·3H2O], ferricopiapite [Fe4.67(SO4)6(OH)2·20H2O], fibroferrite [Fe(OH)(SO4)·5H2O], and schwertmannite [Fe16O16(SO4)2(OH)12·10H2O] (Farrand et al. 2007; Johnson et al. 2007; Klingelhofer et al. 2004; Lane et al. 2008; Morris et al. 2006; Poulet et al. 2008; Yen et al. 2008). The presence of sulfate minerals in evaporative settings implies that a warmer, wetter, and acidic environment might have existed in the early evolution of Mars (Bibring et al. 2006; Hurowitz and McLennan 2007; King and McSween 2005). Some workers have also suggested that the formation and distribution of ferric sulfate minerals on the martian surface may be influenced by post-depositional diagenetic aqueous processes (Bibring et al. 2007; Tosca et al. 2008). In addition, ferric sulfates found within soils in the Columbia Hills and Inner Basin of Gusev Crater are likely associated with hydrothermal and fumarolic processes (Johnson et al. 2007; Morris et al. 2008; Wang et al. 2008).

Sulfate minerals are known to display path-dependent phase transitions as a function of RH. For example, kieserite (MgSO4·H2O) transforms to hexahydrite (MgSO4·6H2O) or epsomite (MgSO4·7H2O) when exposed to humid air, while subsequent desiccation produces an amorphous phase rather than kieserite (Vaniman et al. 2004). These results suggest that kieserite might not occur in places influenced by surface cycles of hydration and dehydration (Vaniman et al. 2004). Similarly, Fe-sulfate minerals crystallize in multiple hydration states that are also sensitive to the changes in RH. Furthermore, as we show in this study, Fe3+ hydrolysis can significantly complicate hydration and dehydration reactions, but characterizing these complex reaction pathways allows iron sulfates to be used as important tracers of paleo-environmental conditions.

Two parallel experimental protocols were followed to study the phase stabilities and transformations in the ferric sulfate system: (1) ex situ RH equilibration followed by characterization with powder XRD (X-ray diffraction) methods, and (2) in situ monitoring of the diffraction signature as RH is varied. Although the ex situ method is well suited for the study of long-term stability, in situ XRD with dynamic RH control provides a more efficient way to examine the details of phase changes and elucidate transient phases that may form over the course of the reaction. The RH-controlled in situ method was previously used to follow the phase transitions or structural change of inorganic clay minerals (Chipera 1997) and sulfates, including Mg-sulfate (Vaniman et al. 2004), Fe2+-sulfate (Peterson and Grant 2005), and Na-sulfate (Linnow et al. 2006). Previous studies conducted by Chou and Seal (Chou and Seal 2003a, 2003b, 2005a, 2005b, 2007; Chou et al. 2002) determined the equilibrium T and RH boundaries between neighboring hydration states of several bivalent metal sulfates by measuring the weight gain or loss caused by hydration or dehydration. To date, little work has been done to characterize the Fe3+-sulfate interaction with water vapor, a common interaction in acid mine drainage (AMD) environments on Earth as well as on the martian surface. In this study, we explored the stability and phase evolution among Fe3+-sulfate hydrates as a function of RH at room temperature. Results will also be used to establish a database that will enable a more thorough understanding of the X-ray scattering behavior of iron sulfate minerals to robustly interpret data from the CheMin instrument on the Mars Science Laboratory (MSL) (Bish et al. 2007), which is scheduled to launch in 2011.


   EXPERIMENTAL METHODS
Top
 Abstract
 Introduction
 Experimental methods
Results
 Discussion
 Acknowledgments
 References cited
 
Sample preparation
Ferric sulfate [Fe2(SO4)3] was selected as the starting material for hydration experiments. It has two polymorphs: a monoclinic and a trigonal phase (Christidis and Rentzeperis 1975, 1976). Only the trigonal phase has been found in nature as the mineral mikasaite (Miura et al. 1994). The anhydrous ferric sulfate with the simple Fe2(SO4)3 stoichiometry provides an opportunity to explore a phase evolution map (Levi 1998), which is not necessarily related to the phase (equilibrium) diagram. Phase evolution maps also characterize the intermediate phases encountered as a function of time and other system variables (RH, T, P, etc.). The iron sulfate hydration (or dehydration) evolution map will point out the different evolved iron sulfate phases depending on the path taken during the evolution of RH.

The monoclinic ferric sulfate [M-Fe2(SO4)3] was purchased from Alfa-Aesar (Puratronic, 99.998%, metals basis) and was used without further purification. The bulk sample was verified to be M-Fe2(SO4)3 by powder XRD. The trigonal ferric sulfate [T-Fe2(SO4)3] was prepared by heating the ferric sulfate hydrate from J.T. Baker [Baker Analyzed, Assay Fe2(SO4)3 > 73.0%] at 350 °C for 2 h. The purity and the anhydrous state of the prepared samples were further confirmed by XRD and thermogravimetric analysis (TGA). The TGA results show a weight loss of <2% before commencement of desulfation to form hematite at 640 °C, with a final weight loss of 39.5%, consistent with the theoretical value of 40% calculated from the ratio of molecular weights—MWFe2O3/MWFe2 (SO4)3.

In situ XRD with dynamic RH control
Phase characterization was performed with two laboratory X-ray diffractometers, each of which was equipped with sample chambers capable of humidity control. A Bruker General Area Detector Diffraction System (GADDS) equipped with a HI-STAR area detector was configured as shown in Figure 1Go. The diffractometer has a xyz translation stage on which is mounted a polyimide (Kapton) capillary enclosed in a custom-built sample environment chamber. A V-Gen II RH Generator (InstruQuest, not shown in the Fig. 1Go) is used to dynamically control the humidity in the sample chamber. The instrument uses the so-called two-temperature principle, which passes water-saturated air through a condenser and then to the experiment cell, so that the desired RH is achieved by adjusting the condenser temperature. Humid air generated from the V-Gen II first passes through a heat-sealed transfer tube, through a manifold and finally into a polyimide capillary (1 mm in diameter) sealed in an 8 cm long Kapton film chamber (Fig. 1Go). Sample powders, adsorbed on the amorphous borosilicate glass wool substrates, are loaded in the top end of the capillary. The use of the substrate facilitates better contact with humid air by maximizing surface area, and avoids blocking the air flow. During the experiment, the airflow rate was controlled at 100 cc/min. The hygrometer is placed just above the capillary to measure both T and RH. To identify temperature gradients, an additional thermocouple is mounted in the manifold to measure the temperature of the gas before it enters the capillary. CoK{alpha} radiation ({lambda} = 1.7903 Å) was used as the X-ray source.


Figure 1
View larger version (51K):
[in this window]
[in a new window]

  		FIGURE 1. The humidity chamber on the GADDS diffractometer with dynamic RH control.

 
Humidity-buffer method
In addition to in situ studies, samples were also stored at a range of RH levels from 11 to 75% at room temperature (22–25 °C) to investigate reactions over prolonged time scales. The humidity-buffer method, where a saturated salt solution produces a stable RH, was used to control the RH inside the containers where powder samples were disseminated at the base of a trial and stored (Fig. 2Go). The equilibrium RH is specific to the salt, and varies with temperature. Greenspan (1977) evaluated previous experimental data on various saturated salt solutions, and estimated the equilibrium RH as a function of temperature. The buffers were also used to test the calibration of the RH probe (Rotronic, Hygroclip SC05; accuracy: ±1.5 % RH, ±0.3 °C at 23 ± 5 °C), using a setup similar to Figure 2Go except for provision of an opening in the cover for inserting the probe. Table 1Go lists humidity buffer salts used in this study, along with the documented and measured RH for each.


Figure 2
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 2. Schematic illustration of the humidity container used to preserve samples under constant RH.

 

View this table:
[in this window]
[in a new window]

  		TABLE 1. Humidity buffers used in this study
 
XRD patterns of the powder samples preserved in RH buffer cells were periodically collected and analyzed with the Scintag PAD-X X-ray diffractometer. The sample chamber on the Scintag (Fig. 3Go) differs from the one on the GADDS instrument in that it operates in Bragg-Brentano rather than Debye-Scherrer geometry. The chamber body is made from lucite with X-ray-transparent Kapton film windows. The hygrometer is placed 1 cm above the sample plate through the opening in the front panel to monitor RH and T. Two other openings on the panel allow mounting of additional RH and T probes, or are used as gas inlet and outlet when dynamic RH control is needed. Figure 3Go shows the static RH control mode, achieved by simply placing a humidity buffer inside the chamber. Heating tape wrapped around the sample produces a low thermal-gradient field (±2 °C) up to 90 °C. To collect XRD data, powder samples were quickly transferred into the chamber, in which the RH was pre-adjusted to match the condition inside the RH buffer cells, to prevent hydration or dehydration of the sample during data collection. CuK{alpha} radiation ({lambda} = 1.5406 Å) was used throughout experiments performed on the Scintag diffractometer. Data were collected using a step mode with a step size of 0.02° and a count rate of 3.0 s per step.


Figure 3
View larger version (50K):
[in this window]
[in a new window]

  		FIGURE 3. Setup for the Scintag Diffractometer using the static RH control mode.

 

   RESULTS
Top
 Abstract
 Introduction
 Experimental methods
 Results
Discussion
 Acknowledgments
 References cited
 
Long-term ex situ study using humidity-buffer method
Hydration of M-Fe2(SO4)3.
Powder samples of M-Fe2(SO4)3 stored in different buffer cells underwent different hydration pathways over time. Figure 4Go shows the evolved iron sulfate phases in different RH conditions vs. storage time. For crystalline phases, estimated weight proportions were given. These numbers were obtained from Rietveld refinement (Rietveld 1969) using the powder XRD data, where scale factors, cell parameters for each phase were refined along with specimen shift. A maximum error of ±10% was estimated. The details of the phase evolution are described below.


Figure 4
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 4. Phase evolution map of M-Fe2(SO4)3 under static RH. Figures in the pie chart indicate the weight percentages of the phases present. These values are not given for liquid phases or the crystalline phase that coexists with a liquid.

 
The anhydrous state of M-Fe2(SO4)3 appeared to remain stable over all reaction times at 11% RH. At 33% RH, a slow transition to a mixture of rhomboclase and ferricopiapite was observed. The mixture appeared to be a yellow aggregate showing the color expected for ferricopiapite. Examination of the product under a petrographic microscope suggests the crystallites are no larger than 2 µm. Rhomboclase was finely intergrown with ferricopiapite so that its white color could not be observed under microscope; its presence was confirmed by XRD. The transition rate from M-Fe2(SO4)3 to ferricopiapite and rhomboclase increased with increasing RH (i.e., at 43 and 53% RH; Fig. 4Go). The mixture of rhomboclase and ferricopiapite subsequently transformed to kornelite [Fe2(SO4)3·7.25H2O] then paracoquimbite [Fe2(SO4)3·9H2O] after 130 days. Figure 5Go shows the change of the sample XRD profile along this transition at 53% RH. At 62% RH, ferricopiapite formed as a wet yellow paste less than two days after the hydration experiment started. However, rhomboclase, which co-formed with ferricopiapite at lower RH, did not appear, probably because rhomboclase, a highly soluble mineral (Tosca et al. 2007), was dissolved under 62% RH condition. This explanation was confirmed by an in situ experiment on the GADDS showing the deliquescence point of rhomboclase was between 58 and 60% RH at 25 °C. Also, the ferricopiapite paste with its coating liquid appeared to be stabilized at 62% RH, did not transform to kornelite. At 75% RH, the sample was completely dissolved and a clear amber solution was formed.


Figure 5
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 5. XRD data showing the hydration process of M-Fe2(SO4)3 at 53% RH. Several peaks were marked with the initials of corresponding mineral names, either followed by the index numbers in parentheses or not. F = Ferricopiapite; R = Rhomboclase; K = Kornelite; P = Paracoquimbite. The change of peak intensities clearly shows a phase transformation from ferricopiapite and rhomboclase to kornelite, then to paracoquimbite.

 
Hydration of T-Fe2(SO4)3.
The phase evolution for T-Fe2(SO4)3 at different RH levels involved the same hydrated phases as identified in the M-Fe2(SO4)3 experiments—ferricopiapite, rhomboclase, kornelite, and paracoquimbite (Fig. 6Go). But this series was distinct from M-Fe2(SO4)3 (Fig. 4Go) in two ways. First, the T-Fe2(SO4)3 phase absorbed water much more rapidly than M-Fe2(SO4)3. At 33, 43, and 53% RH, T-Fe2(SO4)3 turned to a solid-liquid suspension within hours. Solid phases formed after 3 to 5 days. The second difference is that an unidentified intermediate phase(s) formed along with ferricopiapite and rhomboclase at 43 and 53% RH. Figure 7Go shows the XRD profiles for the transformations at 53% RH containing the unknown phase. The XRD signature of this phase is further addressed in the discussion below.


Figure 6
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 6. Phase evolution map of T-Fe2(SO4)3 at different RH levels. Figures in the pie chart indicate the proportions of the phases present in weight percentages. These values are not given for amorphous, liquid, or unknown phases, or the crystalline phases that coexist with them.

 

Figure 7
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 7. XRD data showing the hydration process of T-Fe2(SO4)3 at 53% RH. F = Ferricopiapite; R = Rhomboclase; K = Kornelite; P = Paracoquimbite; U = Unknown phase.

 
Dehydration from Fe2(SO4)3 solution.
As a test of the effects of fluid evaporation and the reversibility of the observed phase changes in the hydration experiments, the clear solution formed from T-Fe2(SO4)3 in the 75% RH buffer cell was transferred to lower RH conditions. The storage was the same as shown in Figure 2Go except that the samples were liquid drops. The starting solution contains 41 wt% Fe2(SO4)3. The concentration is calculated by measuring the weight of starting T-Fe2(SO4)3 and the resultant solution equilibrated in the 75% RH cell. Figure 8Go shows the evolution of phases precipitated at different RH along with post-precipitation phase evolution (phase changes observed after initial precipitates were formed from fluid evaporation).


Figure 8
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 8. Precipitation and post-precipitation phase transitions at different RH levels over time. Figures in the pie chart indicate the proportions of the phases present in weight percentages. These values are not given for amorphous or liquid phases, or the crystalline phases that coexist with them.

 
At 62% RH, ferricopiapite precipitated as a yellow paste reminiscent of the sample observed at the same RH during the hydration process of both anhydrous forms of Fe2(SO4)3. In the 53 and 43% RH cells, ferricopiapite and rhomboclase crystallized between 36 and 48 h and no visible liquid phase was left. A slow transformation to kornelite and paracoquimbite occurred thereafter, but ferricopiapite and rhomboclase still dominated the overall phase assemblage after 60 days. At 33% RH, the liquid became extremely viscous and gradually (over 25–30 days) solidified to form a hard transparent amber-like amorphous material. Rhomboclase appeared after 50 days of subsequent reaction. At RH < 11%, the amber-colored amorphous solid formed within 24 h retaining the morphology of the starting liquid drops. No crystalline phases formed over two months of observation.

Dehydration and rehydration between 62 and 53% RH.
The ferricopiapite paste stored in the 62% RH cell was loaded onto the Scintag Diffractometer with the RH inside the chamber stabilized at 53%. XRD data were taken over 72 h without removing the sample (Fig. 9Go). At first, rhomboclase and kornelite formed and grew, as the ferricopiapite decreased. After 12 h, rhomboclase XRD peak intensities decreased, while kornelite peak intensities increased. After 60 h, both rhomboclase and ferricopiapite diminished, and kornelite was the only phase in the product. The kornelite product was then placed in the 62% RH buffer cell. The sample rehydrated to paracoquimbite, instead of reverting to ferricopiapite paste with co-existing liquid (Fig. 10Go).


Figure 9
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 9. In situ XRD data showing the dehydration of the sample from 62% RH bottle at 53% RH. Four lines indicate 0, 12, 36, and 60 h stay at 53% RH from bottom to top. Percentages are the weight proportions of each phase estimated from Rietveld refinement. F = Ferricopiapite; R = Rhomboclase; K = Kornelite. It can be seen that rhomboclase crystallized at the beginning, but converted to kornelite after 60 h.

 

Figure 10
View larger version (14K):
[in this window]
[in a new window]

  		FIGURE 10. XRD data showing kornelite transformation to paracoquimbite at 62% RH. Five lines indicate 0, 7, 23, 43, and 79 days preservation of kornelite powders at 62% RH from bottom to top. Percentages are the weight proportions of each phase estimated from Rietveld refinement. K = Kornelite; P = Paracoquimbite.

 
In situ time-resolved study of the hydration of iron sulfate
Hydration of M-Fe2(SO4)3.
Figures 11a, 11b, and 11cGo show time resolved XRD data depicting the hydration of M-Fe2(SO4)3 from 40 to 68% over 92 h, the same sequence of phase evolution as was observed as in the ex situ studies: M-Fe2(SO4)3 -> (rhomboclase, ferricopiapite) -> kornelite at RH < 60%. At 68% RH, kornelite dissolved as ferricopiapite increased. XRD data shown in Figure 11dGo indicates the subsequent dehydration produced kornelite from wet ferricopiapite.


Figure 11
View larger version (33K):
[in this window]
[in a new window]

  		FIGURE 11. The hydration process of M-Fe2(SO4)3 as recorded by time-resolved XRD data from the GADDS diffractometer. 2{theta} ranging from 4 to 36.5° is plotted on the horizontal axis. The vertical time axis is marked with the RH steps and their duration. Some peaks are marked with the initials of corresponding mineral names and diffraction index: F = Ferricopiapite; R = Rhomboclase; K = Kornelite; P = Paracoquimbite. Note that the broad peak at 7° is produced by the scattering from Kapton capillary and the chamber wall. (a) M-Fe2(SO4)3 transformed to ferricopiapite and rhomboclase at 50% RH. (b) Kornelite crystallized at 55% RH. Rhomboclase quickly disappeared when RH was raised to 60%, indicated by a sudden drop of the peak intensity at 11.3° along with an apparent shift to higher 2{theta}, caused by the loss of R(020) contribution. Ferricopiapite diffraction peaks also decreased slowly in intensity at 60% RH as kornelite peaks intensified. (c) Kornelite dissolution and ferricopiapite crystallization with a residual solvent phase as RH increased to 68%. (d) The subsequent dehydration process reversed the transition in c as ferricopiapite transformed to kornelite.

 
Hydration of T-Fe2(SO4)3.
Figure 12Go shows time-resolved XRD data collected during the hydration process of T-Fe2(SO4)3 at 45% RH. The T-Fe2(SO4)3 diffraction peaks disappeared rapidly as the starting sample hydrated to form a suspension. After 10 h, both rhomboclase and ferricopiapite began crystallizing. The peak at 2{theta} = 12° could not be indexed to either rhomboclase or ferricopiapite. This reflection has the same d-spacing (8.4 Å) as the strongest reflection from the unknown phase as observed in the ex situ experiment (Fig. 7Go). Note that the differences in 2{theta} values are due to different X-ray wavelengths used on the Scintag and GADDS instruments. These observations indicate they are very likely the same unknown phase.


Figure 12
View larger version (27K):
[in this window]
[in a new window]

  		FIGURE 12. Time-resolved XRD data depicting the hydration of T-Fe2(SO4)3 at 45% RH at room T. The starting T-Fe2(SO4)3 dissolved quickly at 45% RH (lower patterns). At 10 h, the XRD pattern has few diffraction peaks, indicating a lack of long-range order in the sample, which corresponds to the intermediate suspension phase. After 10 h, rhomboclase, ferricopiapite, and a small amount of the unidentified phase(s) crystallized out (peak marked with U).

 

   DISCUSSION
Top
 Abstract
 Introduction
 Experimental methods
 Results
 Discussion
Acknowledgments
 References cited
 
All experiments performed in this study are chemically constrained to the simplest Fe-SO4-H2O system: molar ratio Fe/SO4 is 2/3 (determined by the purity of the starting material), the amount of H2O as varied with RH.

Hydration series
At least five iron sulfate hydrate phases were involved in the hydration of Fe2(SO4)3 and subsequent dehydration experiments conducted in this study: ferricopiapite, rhomboclase, kornelite, and paracoquimbite, plus an unknown phase particularly found in the hydration of T-Fe2(SO)4. One major difference from other metal sulfate hydrate systems is that the hydrolysis strongly affects phase evolution. This can be seen from the formation of hydronium-bearing rhomboclase and hydroxyl-bearing ferricopiapite from anhydrous Fe2(SO4)3 exposed to water vapor:


Formula 1(1)

For the reaction with T-Fe2(SO4)3, the product side also includes the unknown phase. The reverse reaction of Equation 1 was not observed in this study. The mixture of the ferricopiapite and rhomboclase tends to neutralize each other to eventually form kornelite:


Formula 2(2)

The forward reaction happens at modest RH from around 33 to 53%. Since both are solid phases, this process is slow and equilibrium is not attained under dry conditions. The reverse reaction happens when the RH is higher than the deliquescence point of kornelite but still lower than that of ferricopiapite (refer to "hydration of M-Fe2(SO4)3" section of results—68% RH step). While kornelite is not the most stable form at intermediate RH conditions, it does finally transform to paracoquimbite:


Formula 3(3)

The RH-T boundary between kornelite and paracoquimbite was not a primary focus of this study, although it can be concluded that the equilibrium RH should be below 43% at room temperature (22–25 °C), since the hydration transition was observed in this range [refer to "hydration of M-Fe2(SO4)3" section of results].

Path-dependant reactions
The iron sulfate phases encountered are determined by the RH, and also the path taken to the final RH. Reactions 1 and 2 show that one-step hydration of the anhydrous ferric sulfate at 62% RH produced ferricopiapite, which was stable if there was no further change in RH. On the other hand, kornelite formed by dehydrating the ferricopiapite paste transformed to paracoquimbite rather than reverting to ferricopiapite upon rehydration at 62% RH (reaction 3). We speculate that differences in the pH of the coating solution might be responsible for ferricopiapite transforming to kornelite. The paucity of coating solution and its close adsorption on ferricopiapite prohibited an effective measurement of pH. Previous studies predicted the lower pH boundary for ferricopiapite at about 0 (Marion et al. 2008; Tosca et al. 2007). One-step hydration at 62% RH produced ferricopiapite as the only solid phase. The remnant liquid had to be acidic to balance the hydroxyls in ferricopiapite, but the pH might still be within the stability range for ferricopiapite. Reducing RH to 53% evaporated more water, so that the coating solution finally became sufficiently acidic to convert ferricopiapite to kornelite.

Dehydration series
Evaporation of a concentrated Fe2(SO4)3 solution is also affected by environmental RH. At RH > 33%, ferricopiapite is the first to crystallize. Rhomboclase may or may not precipitate, depending on whether the RH is lower than the deliquescence point (58–60% at 25 °C). This precipitation process can also be described by Equation 1. Kornelite was not observed to crystallize directly from solution in this study, but forms by subsequent alteration of initially crystallized ferricopiapite. At RH conditions from 43 to 62%, kornelite gradually transforms to paracoquimbite over extended storage time (Figs. 4Go, 6Go, and 8Go). Paracoquimbite appears to be the most thermodynamically stable form of ferric sulfate hydrate at this RH range. This initial precipitation of ferricopiapite rather than kornelite or paracoquimbite may provide insight into the polymerization process in concentrated iron sulfate solution. Evaporation at 33% RH or lower results in the formation of amorphous iron sulfate. This may result from a high evaporation rate under low RH conditions, which reduces the water activity of the solution so quickly that it inhibits the formation of any crystalline iron sulfate hydrates. As the diurnal fluctuations of RH on the martian surface ranges broadly from 10 to 90% (Peterson and Wang 2006), iron sulfate, if by any chance precipitated from a transient fluid phase formed under high RH conditions, would most likely stay as the initial precipitation phase like ferricopiapite or the amorphous state.

Unresolved questions
One of the transitional phases formed during the T-Fe2(SO4)3 hydration process is not identified. This phase appears as a white powder, which is finely intergrown with ferricopiapite and rhomboclase, and is therefore difficult to obtain a pure phase powder. In the XRD data (Fig. 7Go), this unknown phase has two distinct low-angle peaks at d = 16.7 and 8.4 Å. The d-spacings suggest the higher-angle peak probably has an index that doubles the lower one. For comparison, ferricopiapite has peak 010 (d = 18.3 Å) and 020 (d = 9.2 Å) to the left of those of the unknown phase, also shown in Figure 7Go. The unknown phase does not occur in the hydration of M-Fe2(SO4)3, which implies a different hydration mechanism, probably due to the structural difference of the two anhydrous iron sulfates. However, no conclusions can be drawn until the structure of the unidentified phase is solved. Efforts are underway to grow single crystals for a structure solution.

Some known ferric sulfate minerals were not found in this study, such as lausenite [Fe2(SO4)3·5H2O] (Majzlan et al. 2005), coquimbite [Fe2(SO4)3·9H2O] (Fang and Robinson 1970), and quenstedtite [Fe2(SO4)3·11H2O] (Thomas et al. 1974). Their formation may require a temperature other than room T. These phases will be targets for further studies as we turn our attention to phase evolution at elevated temperatures. Furthermore, the stability relationship between M-Fe2(SO4)3 and T-Fe2(SO4)3 is not yet clear, and no direct conversion between the two phases has yet been found.

In the future, we will explore the phase evolution in ferric sulfate hydrate, including those phases with Fe3+/SO42– ratios other than 2:3 and under higher and lower temperatures. A complete understanding of the stability of iron sulfate minerals and the transitions induced by RH and other environmental parameters will provide a reliable base for interpretation of field data from Mars.


   ACKNOWLEDGMENTS
Top
 Abstract
 Introduction
 Experimental methods
 Results
 Discussion
 Acknowledgments
References cited
 
The authors acknowledge the support from NASA grant MFRP07-0022.


   Footnotes
 
MANUSCRIPT HANDLED BY RICHARD WILKIN

MANUSCRIPT RECEIVED December 23, 2008; MANUSCRIPT ACCEPTED July 4, 2009


   REFERENCES CITED
Top
 Abstract
 Introduction
 Experimental methods
 Results
 Discussion
 Acknowledgments
 References cited
 

Adams, J.R. and Merz, A.R. (1929) Hygroscopicity of fertilizer materials and mixtures. Industrial and Engineering Chemistry, 21, 305–307.

Bibring, J.P., Langevin, Y., Gendrin, A., Gondet, B., Poulet, F., Berthe, M., Soufflot, A., Arvidson, R., Mangold, N., Mustard, J., Drossart, P., and OMEGA Team (2005) Mars surface diversity as revealed by the OMEGA/Mars Express observations. Science, 307, 1576–1581.[Abstract/Free Full Text][CrossRef][Web of Science][Medline][GeoRef]

Bibring, J.P., Langevin, Y., Mustard, J.F., Poulet, F., Arvidson, R., Gendrin, A., Gondet, B., Mangold, N., Pinet, P., Forget, F., and OMEGA Team (2006) Global mineralogical and aqueous mars history derived from OMEGA/Mars express data. Science, 312, 400–404.[Abstract/Free Full Text][CrossRef][Web of Science][Medline][GeoRef]

Bibring, J.P., Arvidson, R.E., Gendrin, A., Gondet, B., Langevin, Y., Le Mouelic, S., Mangold, N., Morris, R.V., Mustard, J.F., Poulet, F., Quantin, C., and Sotin, C. (2007) Coupled ferric oxides and sulfates on the Martian surface. Science, 317, 1206–1210.[Abstract/Free Full Text][CrossRef][Web of Science][Medline]

Bish, D.L., Blake, D., Sarrazin, P., Treiman, A.H., Hoehler, T., Hausrath, E.M., Midtkandal, I., and Steele, A. (2007) Field XRD/XRF mineral analysis by the MSAL CheMin instrument. Lunar and Planetary Science XXXVIII, Abstract no. 1163.

Chipera, S.J., Carey, J.W., and Bish, D.L. (1997) Controlled-humidity XRD analyses: Application to the study of smectite expansion/contraction. In J.V. Gilfrich, I.C. Noyan, R. Jenkins, T.C. Huang, R.L. Snyder, D.K. Smith, M.A. Zaitz, and P.K. Predecki, Eds., Advances in X-ray Analysis, 39, p. 713–722. Plenum Press, New York.

Chou, I-M. and Seal II, R.R. (2003a) Acquisition and evaluation of thermodynamic data for morenosite-retgersite equilibria at 0.1 MPa. American Mineralogist, 88, 1943–1948.[Abstract/Free Full Text][Web of Science][GeoRef]

——— (2003b) Determination of epsomite-hexahydrite equilibria by the humidity-buffer technique at 0.1 MPa with implications for phase equilibria in the system MgSO4-H2O. Astrobiology, 3, 619–630.[CrossRef][Web of Science][Medline]

——— (2005a) Acquisition and evaluation of thermodynamic data for bieberite-moorhouseite equilibria at 0.1 MPa. American Mineralogist, 90, 912–917.[Abstract/Free Full Text][CrossRef][Web of Science][GeoRef]

——— (2005b) Determination of goslarite-bianchite equilibria by the humidity-buffer technique at 0.1 MPa. Chemical Geology, 215, 517–523.[CrossRef][Web of Science][GeoRef]

——— (2007) Magnesium and calcium sulfate stabilities and the water budget of Mars. Journal of Geophysical Research: Planets, 112, E11004, DOI: 10.1029/2007JE002898.[CrossRef]

Chou, I-M., Seal II, R.R., and Hemingway, B.S. (2002) Determination of melanterite-rozenite and chalcanthite-bonattite equilibria by humidity measurements at 0.1 MPa. American Mineralogist, 87, 108–114.[Abstract/Free Full Text][Web of Science][GeoRef]

Christidis, P.C. and Rentzeperis, P.J. (1975) Crystal-structure of monoclinic Fe2(SO4)3. Zeitschrift für Kristallographie, 141, 233–245.[Web of Science][GeoRef]

——— (1976) Crystal-structure of rhombohedral Fe2(SO4)3. Zeitschrift für Kristallographie, 144, 341–352.[Web of Science][GeoRef]

Fang, J.H. and Robinson, P.D. (1970) Crystal structures and mineral chemistry of hydrated ferric sulfates. Crystal structure of coquimbite. American Mineralogist, 55, 1534–1540.[Web of Science][GeoRef]

Farrand, W.H., Bell III, J.F., Johnson, J.R., Jolliff, B.L., Knoll, A.H., McLennan, S.M., Squyres, S.W., Calvin, W.M., Grotzinger, J.P., Morris, R.V., Soderblom, J., Thompson, S.D., Watters, W.A., and Yen, A.S. (2007) Visible and near-infrared multispectral analysis of rocks at Meridiani Planum, Mars, by the Mars Exploration Rover Opportunity. Journal of Geophysical Research: Planets, 112, E06S02, DOI: 10.1029/2006JE002773.[CrossRef]

Gendrin, A., Mangold, N., Bibring, J.-P., Langevin, Y., Gondet, B., Poulet, F., Bonello, G., Quantin, C., Mustard, J., Arvidson, R., and LeMouélic, S. (2005) Sulfates in martian layered terrains: The OMEGA/Mars Express view. Science, 307, 1587–1591.[Abstract/Free Full Text][CrossRef][Web of Science][Medline][GeoRef]

Greenspan, L. (1977) Humidity fixed-points of binary saturated aqueous-solutions. Journal of Research of the National Bureau of Standards Section a: Physics and Chemistry, 81, 89–96.[Web of Science]

Hurowitz, J.A. and McLennan, S.M. (2007) A similar to 3.5 Ga record of water-limited, acidic weathering conditions on Mars. Earth and Planetary Science Letters, 260, 432–443.[CrossRef][Web of Science][GeoRef]

Johnson, J.R., Bell III, J.F., Cloutis, E., Staid, M., Farrand, W.H., McCoy, T., Rice, M., Wang, A., and Yen, A. (2007) Mineralogic constraints on sulfur-rich soils from Pancam spectra at Gusev crater, Mars. Geophysical Research Letters, 34, L13202, DOI: 10.1029/2007GL029894.[CrossRef]

King, P.L. and McSween Jr., H.Y. (2005) Effects of H2O, pH, and oxidation state on the stability of Fe minerals on Mars. Journal of Geophysical Research: Planets, 110, E12S10, DOI: 10.1029/2005JE002482.[CrossRef]

Klingelhofer, G., Morris, R.V., Bernhardt, B., Schroder, C., Rodionov, D.S., de Souza, P.A., Yen, A., Gellert, R., Evlanov, E.N., Zubkov, B., Foh, J., Bonnes, U., Kankeleit, E., Gutlich, P., Ming, D.W., Renz, F., Wdowiak, T., Squyres, S.W., and Arvidson, R.E. (2004) Jarosite and hematite at Meridiani Planum from Opportunity’s Mössbauer spectrometer. Science, 306, 1740–1745.[Abstract/Free Full Text][CrossRef][Web of Science][Medline][GeoRef]

Lane, M.D., Bishop, J.L., Dyar, M.D., King, P.L., Parente, M., and Hyde, B.C. (2008) Mineralogy of the Paso Robles soils on Mars. American Mineralogist, 93, 728–739.[Abstract/Free Full Text][CrossRef][Web of Science][GeoRef]

Levi, C.G. (1998) Metastability and microstructure evolution in the synthesis of inorganics from precursors. Acta Materialia, 46, 787–800.[CrossRef][Web of Science]

Linnow, K., Zeunert, A., and Steiger, M. (2006) Investigation of sodium sulfate phase transitions in a porous material using humidity- and temperature-controlled X-ray diffraction. Analytical Chemistry, 78, 4683–4689.[Medline]

Majzlan, J., Botez, C., and Stephens, P.W. (2005) The crystal structures of synthetics Fe2(SO4)3(H2O)5 and the type specimen of lausenite. American Mineralogist, 90, 411–416.[Abstract/Free Full Text][CrossRef][Web of Science][GeoRef]

Marion, G.M., Kargel, J.S., and Catling, D.C. (2008) Modeling ferrous-ferric iron chemistry with application to Martian surface geochemistry. Geochimica et Cosmochimica Acta, 72, 242–266.[Web of Science][GeoRef]

McLennan, S.M. and Grotzinger, J.P. (2008) The Sedimentary Rock Cycle of Mars, in the Martian Surface: Compostion, mineralogy, and physical properties, p. 541–577. Cambridge University Press, U.K.

Miura, H., Niida, K., and Hirama, T. (1994) Mikasaite, (Fe3+,Al)2(SO4)3, a New Ferric Sulfate Mineral from Mikasa City, Hokkaido, Japan. Mineralogical Magazine, 58, 649–653.[Abstract][CrossRef][Web of Science][GeoRef]

Morris, R.V., Klingelhofer, G., Schroder, C., Rodionov, D.S., Yen, A., Ming, D.W., de Souza Jr., P.A., Fleischer, I., Wdowiak, T., Gellert, R., Bernhardt, B., Evlanov, E.N., Zubkov, B., Foh, J., Bonnes, U., Kankeleit, E., Gutlich, P., Renz, F., Squyres, S.W., and Arvidson, R.E. (2006) Mössbauer mineralogy of rock, soil, and dust at Gusev crater, Mars: Spirit’s journey through weakly altered olivine basalt on the plains and pervasively altered basalt in the Columbia Hills. Journal of Geophysical Research: Planets, 111, E02S13, DOI: 10.1029/2005JE002584.[CrossRef]

Morris, R.V., Klingelhofer, G., Schroder, C., Fleischer, I., Ming, D.W., Yen, A., Gellert, R., Arvidson, R.E., Rodionov, D.S., Cumpler, L., Clark, B.C., Cohen, B.A., McCoy, T.J., Mittlefehldt, D.W., Schmidt, M.E., de Souza Jr., P.A., and Squyres, S.W. (2008) Iron mineralogy and aqueous alteration from Husband Hill through Home Plate at Gusev crater, Mars: Results from the Mössbauer instrument on the Spirit Mars Exploration Rover. Journal of Geophysical Research: Planets, 113, E12S42, DOI: 10.1029/2008JE003201.[CrossRef]

Peterson, R.C. and Grant, A.H. (2005) Dehydration and crystallization reactions of secondary sulfate minerals found in mine waste: In situ powder-diffraction experiments. Canadian Mineralogist, 43, 1171–1181.[CrossRef][Web of Science]

Peterson, R.C. and Wang, R.Y. (2006) Crystal molds on Mars: Melting of a possible new mineral species to create Martian chaotic terrain. Geology, 34, 957–960.[Abstract/Free Full Text][CrossRef][Web of Science][GeoRef]

Poulet, F., Bibring, J.P., Mustard, J.F., Gendrin, A., Mangold, N., Langevin, Y., Arvidson, R.E., Gondet, B., Gomez, C., and OMEGA Team (2005) Phyllosilicates on Mars and implications for early martian climate. Nature, 438, 623–627.[CrossRef][Medline][GeoRef]

Poulet, F., Arvidson, R.E., Gomez, C., Morris, R.V., Bibring, J.P., Langevin, Y., Gondet, B., and Griffes, J. (2008) Mineralogy of Terra Meridiani and western Arabia Terra from OMEGA/MEx and implications for their formation. Icarus, 195, 106–130.[CrossRef][Web of Science][GeoRef]

Rietveld, H.M. (1969) A profile refinement method for nuclear and magnetic structures. Journal of Applied Crystallography, 2, 65–71.[CrossRef][Web of Science]

Squyres, S.W., Grotzinger, J.P., Arvidson, R.E., Bell III, J.F., Calvin, W., Christensen, P.R., Clark, B.C., Crisp, J.A., Farrand, W.H., Herkenhoff, K.E., Johnson, J.R., Klingelhofer, G., Knoll, A.H., McLennan, S.M., McSween Jr., H.Y., Morris, R.V., Rice Jr., J.W., Rieder, R., and Soderblom, L.A. (2004) In situ evidence for an ancient aqueous environment at Meridiani Planum, Mars. Science, 306, 1709–1714.[Abstract/Free Full Text][CrossRef][Web of Science][Medline][GeoRef]

Thomas, J.N., Robinson, P.D., and Fang, J.H. (1974) Crystal structures and mineral chemistry of hydrated ferric sulfates. IV. The crystal structure of quenstedtite. American Mineralogist, 59, 582–586.[Web of Science][GeoRef]

Tosca, N.J., Smirnov, A., and McLennan, S.M. (2007) Application of the Pitzer ion interaction model to isopiestic data for the Fe2(SO4)3-H2SO4-H2O system at 298.15 and 323.15 K. Geochimica et Cosmochimica Acta, 71, 2680–2698.[CrossRef][Web of Science][GeoRef]

Tosca, N.J., McLennan, S.M., Dyar, M.D., Sklute, E.C., and Michel, F.M. (2008) Fe oxidation processes at Meridiani Planum and implications for secondary Fe mineralogy on Mars. Journal of Geophysical Research: Planets, 113, E05005, DOI: 10.1029/2007JE003019.[CrossRef]

Vaniman, D.T., Bish, D.L., Chipera, S.J., Fialips, C.I., Carey, J.W., and Feldman, W.C. (2004) Magnesium sulfate salts and the history of water on Mars. Nature, 431, 663–665.[CrossRef][Medline]

Wang, A., Bell III, J.F., Li, R., Honson, J.R., Farrand, W.H., Cloutis, E.A., Arvidson, R.E., Crumpler, L., Squyres, S.W., McLennan, S.M., Herkenhoff, K.E., Ruff, S.W., Knudson, A.T., Chen, W., and Greenberger, R. (2008) Light-toned salty soils and coexisting Si-rich species discovered by the Mars Exploration Rover Spirit in Columbia Hills. Journal of Geophysical Research: Planets, 113, E12S40, DOI: 10.1029/2008JE003126.[CrossRef]

Yen, A.S., Morris, R.V., Clark, B.C., Gellert, R., Knudson, A.T., Squyres, S., Mittlefehldt, D.W., Ming, D.W., Arvidson, R., McCoy, T., Schmidt, M., Hurowitz, J., Li, R., and Johnson, J.R. (2008) Hydrothermal processes at Gusev Crater: An evaluation of Paso Robles class soils. Journal of Geophysical Research: Planets, 113, E06S10, DOI: 10.1029/2007JE002978.[CrossRef]



This Article
Right arrow Abstract
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Xu, W.
Right arrow Articles by Parise, J. B.
GeoRef
Right arrow GeoRef Citation

JOURNAL HOME HELP CONTACT PUBLISHER SUBSCRIBE ARCHIVE SEARCH TABLE OF CONTENTS
Copyright © 2010 by Mineralogical Society of America