39Ar geochronology of jarosite: The effectiveness of HF in removing silicate contaminants

Chemical Geology 314–317 (2012) 23–32

Contents lists available at SciVerse ScienceDirect

Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo

Research paper 40

Ar/ 39Ar geochronology of jarosite: The effectiveness of HF in removing silicate contaminants Kimberly E. Samuels-Crow a,⁎, Virgil W. Lueth b, Lisa Peters b, William C. McIntosh b a b

Department of Earth and Environmental Sciences, New Mexico Institute of Mining and Technology, 801 Leroy Place, Socorro, NM 87801, United States New Mexico Bureau of Geology and Mineral Resources, 801 Leroy Place, Socorro, NM 87801, United States

a r t i c l e

i n f o

Article history: Received 23 November 2011 Received in revised form 24 April 2012 Accepted 27 April 2012 Available online 5 May 2012 Editor: J.D. Blum Keywords: Supergene jarosite Sample preparation 40 Ar/39Ar geochronology

a b s t r a c t Supergene and hypogene jarosite have been dated successfully, but accurate dating of weathering-derived jarosite and its application to landscape-evolution studies has been limited because of difficulties in obtaining high-purity mineral separates. Hydrofluoric acid (HF) can remove potassium-bearing silicates from supergene samples, but its effects on jarosite crystal chemistry and age dating are unquantified. Three experiments were conducted to determine whether HF treatment removes silicates without altering the potassium or argon composition of the jarosite prior to dating using the 40Ar/39Ar method. In the first two experiments, pure hypogene jarosite from Peña Blanca Mexico (PB; 9.42 ± 0.22 Ma) and a mixture of 85% PB:15% Fish Canyon sanidine (FC-2; 28.02 Ma) were crushed and treated with 25% HF for 0, 30, 240, and 480 min. Jarosite partially dissolves during HF treatment with grains becoming increasingly pitted and rounded with time in acid, but PB's potassium concentration and apparent age remained constant regardless of treatment time. FC-2 was absent from all treated samples, suggesting that 30 min in HF is sufficient to remove mechanically mixed sanidine crystals from jarosite. Methods developed during experiments conducted on PB jarosite were applied to fine-grained supergene jarosite from the Red River Valley (RRV) in Northern New Mexico. RRV jarosite is mixed with Oligocene (~ 24.86 Ma) potassium-bearing silicates that make up 50% or more of each sample. Four RRV jarosite samples that yielded age spectra with clear signs of contamination by older phases were treated with HF for 30 min and re-dated. Despite treatment, back-scattered electron images show that silicates, including sanidine and illite, continued to comprise approximately 30% of HF-treated RRV samples, and the age spectra produced during laser step heating continued to show signs of contamination in higher-wattage steps. However, the integrated ages of these samples were consistently less than 1 Ma, at least 7 Ma younger than the expected integrated age if the 30:70 mixture of RRV silicates and jarosite degassed completely. The young ages suggest that young jarosite controls the apparent age of these samples and that these ages can be interpreted as the maximum timing of supergene jarosite formation. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Jarosite (KFe3(SO4)2(OH)6), an alunite group mineral that can have potassium concentrations greater than 9 weight percent (wt.%), forms in both hypogene and supergene settings under highly oxidizing, acidic conditions. The high potassium concentration makes jarosite readily datable by the 40Ar/39Ar method. Hydrothermal jarosite has been dated with a high degree of precision using the 40Ar/39Ar method (Lueth et al., 2004), but supergene jarosite does not consistently produce accurate ages due to difficulties in removing other potassiumbearing phases, including silicates (Vasconcelos and Conroy, 2003;

⁎ Corresponding author at: Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, NM 87131, United States. Tel.: + 1 505 277 4204; fax: + 1 505 277 8843. E-mail address: [email protected] (K.E. Samuels-Crow). 0009-2541/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.chemgeo.2012.04.032

Lueth et al., 2006). New sample preparation techniques may improve the accuracy of supergene jarosite ages, providing new insights to landscape evolution studies that focus on acidic, oxidizing settings on both Earth and Mars (e.g. Klingelhöfer et al., 2004; Kula and Baldwin, 2011). Hydrofluoric acid (HF) can dissolve silicates, but its effects on jarosite composition and apparent age are unquantified. HF has been used to remove silicates from both jarosite and alunite prior to geochronology and stable isotope analysis (e.g. Wasserman et al., 1992; Arehart and O'Neil, 1993; Polyak et al., 1998; Lueth et al., 2006; Polyak et al., 2006), but some studies warn against using HF to chemically separate fine-grained, supergene jarosite from clay and other potassium-bearing, silicate contaminants, citing concerns that HF treatment leads to potassium and argon loss (Vasconcelos, 1999). Similarly, other studies have shown that synthetic jarosite dissolves incongruently at pH 2 and 8, preferentially losing K + and SO42 − (Smith et al., 2006). Recent studies by Elwood-Madden et al. (2009)

24

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

Table 1 Neutron-induced interfering reactions correction factors. Interfering reaction (39Ar/37Ar)Ca (36Ar/37Ar)Ca (38Ar/39Ar)K (40Ar/39Ar)K

0.0007 ± 5 × 10− 5 0.00028 ± 1 × 10− 5 0.013 0.01 ± 0.002

also demonstrate jarosite's relatively rapid dissolution rates in many environments. Accordingly, potassium and/or argon loss would make it impossible to accurately date HF-treated jarosite samples. We conducted three complementary experiments to quantify HF's effects on jarosite. The first experiment evaluated the effects of HF on pure, hydrothermal jarosite of known, reproducible age in order to test the effects of HF on jarosite crystal chemistry and apparent age. In the second experiment, jarosite of known age was mixed with a fixed percentage of Fish Canyon sanidine (FC-2; 28.02 Ma) in order to test the impact of silicate contaminants on jarosite's apparent age, whether HF preferentially dissolves the silicates, and to determine the amount of time necessary to dissolve sanidine. Methods refined by these experiments were used to try to separate supergene jarosite from hydrothermal illite and volcanic sanidine that formed ~ 25 Ma in order to determine whether HF can effectively remove silicates without making supergene jarosite undatable. 2. Methods 2.1. Sample preparation We used jarosite from Peña Blanca Mexico (PB) to evaluate the effects of HF on coarsely crystalline, hypogene jarosite. PB is predominantly unzoned jarosite with a low natrojarosite component (Papike et al., 2007) and microscopic fluid inclusions. Four aliquots of 400 mg each of PB (9.42 ± 0.22 Ma; Lueth et al., 2004, 2005) and four aliquots of 85% PB mixed with 15% FC-2 were crushed and sieved to maintain grain size less than ~ 63 μm. Three aliquots of each experimental sample were treated with 40 mL of 25% HF for 30, 240, and 480 min with the fourth aliquot left as an untreated control. Another sample of pure PB and FC-2, crushed to maintain grain size between 125 μm and 75 μm, was mixed following irradiation to ensure that the 85:15 mixture was maintained during analysis. This sample was not treated with HF. Additionally, 73 mg of crushed FC-2 (≤63 μm) was placed in a beaker with 40 mL of 25% HF to determine the amount of time needed for complete dissolution. Fine-grained supergene jarosite was collected from drill cores and ferricretes in the Red River Valley in Northern New Mexico (RRV) and sieved to b63 μm to help separate it from larger silicates and iron oxides. The amount of jarosite separated from RRV samples was approximately four times smaller than the amount of PB jarosite used in our experiments, so 72.3 to 101.5 mg of RRV supergene jarosite were treated with 10 mL of 25% HF for 30 min. HF was decanted after treatment of both PB and RRV samples, and samples were rinsed with deionized water three times before being dried overnight in a 70 °C oven, which is below the closure temperature for argon in fine-grained jarosite (Kula and Baldwin, 2011). 2.2. Sample characterization All samples were characterized by electron microprobe analysis (EMPA) at the New Mexico Bureau of Geology and Mineral Resources (NMBGMR) to assess the impact of HF treatment on both jarosite morphology and crystal chemistry. Quantitative chemical analysis was conducted on polished grain mounts. Samples were mounted on

carbon tape for secondary electron imaging to monitor morphological changes. EMPA was conducted on a Cameca SX-100 microprobe. The microprobe is equipped with three wavelength dispersive spectrometers in addition to secondary electron and high-speed back-scattered electron (BSE) detectors. The microprobe was operated with a 10 nA beam current and 10 μm spot size at 15 keV to minimize Na volatilization (Papike et al., 2006). Counting times were 20 s for major elements, 40 s for minor elements (F, Cl), and 60 s for trace elements (As, Mo). F was analyzed to determine whether or not F-bearing minerals formed after HF treatment; no evidence of new mineral precipitation was found. 2.3.

40

Ar/ 39Ar geochronology

Following HF treatment, samples were wrapped in copper foil and placed in 6-hole, machined aluminum disks with FC-2 as the fluence monitor and were then irradiated at the USGS TRIGA reactor in Denver. PB and PB + FC-2 samples were irradiated for 1 h while RRV samples were irradiated for only 30 min. Irradiation times were based on the known ages of PB and FC-2, 9.5 Ma and 28.02 Ma respectively, and the much younger inferred ages for RRV jarosite. Analyses were corrected for neutron-induced interfering reactions using correction factors listed in Table 1. After irradiation, samples were step-heated with a CO2 laser equipped with a 6 × 6 mm homogenizing lens at the NMBGMR New Mexico Geochronology Research Lab (NMGRL). In order to maximize the amount of sample exposed to the laser, 6–8 mg of irradiated material was distributed evenly across the bottom of 4 × 4 mm squares in a 9-hole laser tray. The heating schedule for each sample is listed in Table 2 and Appendix A. Sample gas extracted from experimental PB samples A-0, A-480, and B-30 was cleaned using a manual cryotrap, consisting of a Dewar flask of liquid nitrogen placed on a stainless steel “finger” in the extraction line between the laser and second stage. This cold trap effectively removed condensable gases. Gas was further cleaned by 2 SAES GP-50 getters for 900 s. Once sample gas was isolated in the mass spectrometer, the extraction line was pumped out, and cold trap gases were qualitatively analyzed using a quadrupole mass spectrometer attached to the roughing line. The quadrupole showed that the cold trap had removed measurable concentrations of both sulfur dioxide and water vapor, but trapped argon was below quadrupole detection limits. When RRV samples and PB sample C-0 were analyzed, volatile gases were trapped in a mechanized cryotrap operated at −140 °C before the gas was cleaned by two SAES GP-50 getters for 900 s. Following volatile gas cleanup, sample gas was expanded into NMGRL's MAP-215-50 mass spectrometer where it was analyzed in static mode. Blank values were monitored throughout the analysis and interpolated to the time of sample analysis. Plateaus in stepheating age spectra are defined as three or more contiguous degassing steps that overlap at 95% confidence level (2-sigma) and contain at least 50% of the 39Ar released.

Table 2 Heating schedule for dated jarosite samples. Sample

Laser power (W)

A-0 A-480 C-0 B-30 PIT-VWL-0007 PIT-VWL-0005 ESS-VWL-0001 CAS-VWL-0002

1 1 – – – – – –

2 2 2.5 2 2.5 2.5 2.5 2.5

3 3 3 – 3 3 3 3

4 4 4 4 4 4 4 4

5 5 – – – – – –

6 6 6 6 6 6 6 6

7 7 7 – 7 7 7 7

8 8 – 8 – – – –

9 9 9 – 9 9 9 9

10 10 – 12 – – – –

– – – 14 – – – –

– – – 15 – – – –

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

25

Table 3 Experimental samples. Sample

Description

Time in HF (min)

Initial mass (mg)

Final mass (mg)

Mass loss (mg)

% mass loss

A-0a A-30 A-240 A-480a B-0a B-30a B-240 B-480 PIT VWL 0007a PIT VWL 0005a ESS VWL 0001a CAS VWL 0002a

PB jarosite PB jarosite PB jarosite PB jarosite PB + FC-2 PB + FC-2 PB + FC-2 PB + FC-2 RRV jar + sil RRV jar + sil RRV jar + sil RRV jar + sil

0 30 240 480 0 30 240 480 30 30 30 30

398.4 399.3 400.0 399.6 469.4 469.9 470.0 472.0 72.3 99.2 101.5 82.0

398.4 215.6 174.7 165.6 469.4 272.1 203.9 184.3 29.8 27.6 41.0 21.0

0 183.7 225.3 224.9 0 197.8 266.1 287.7 42.5 71.6 60.5 61

0 46 56 56 0 42 57 61 59 72 60 26

PB jar = pure jarosite; FC-2 = Fish Canyon sanidine; RRV jar = supergene jarosite; sil = silicates (clay, kspar, quartz). a Dated samples.

3. Results 3.1. Mass loss All samples experienced significant mass loss during HF treatment. Mass loss may be due to sample dissolution, mechanical removal during decanting, or to some combination of both. PB samples that underwent longer treatment times experienced greater mass loss, so some jarosite dissolution is likely (Table 3). In the case of PB + FC-2 and RRV samples, mass loss most likely resulted from a combination of both jarosite and silicate dissolution in addition to minor mechanical losses during decanting. 3.2. Potassium concentrations in treated and untreated samples EMPA analysis shows that neither pure jarosite nor RRV supergene jarosite experienced significant potassium loss with time in HF

(Table 4). PB jarosite treated for 480 min has K2O concentrations identical within analytical uncertainty to untreated PB jarosite (Fig. 1A). K2O concentrations for points analyzed on two treated RRV samples, PIT-VWL-0007 and CAS-VWL-0002, were slightly lower than the range of K2O concentrations found in untreated aliquots of the same samples. These points had relatively high SiO2 and Al2O3 concentrations, so it is likely that these points were set on mixtures of jarosite and clay rather than jarosite alone (Fig. 1B). 3.3. Morphology of treated and untreated samples of PB jarosite Although HF treatment did not significantly alter PB jarosite crystal chemistry, treatment did cause morphological changes. Grains from all treated PB samples are pitted and rounded relative to untreated samples, and the degree of pitting increases with treatment time. Pits are triangular and aligned along discrete planes, similar to the orientation of fluid inclusions evident under the petrographic microscope (Fig. 2).

Table 4 Representative chemical analyses for pure jarosite (wt.% oxides and cations). Sample name

A-0-09

A-480-9

PIT VWL 0007

PIT VWL 0007

ESS VWL 0001

ESS VWL 0001

CAS VWL 0002

CAS VWL 0002

Time in HF (min)

0

480

0

30

0

30

0

30

K2O Na2O FeO (measured) Fe2O3 (calculated)a Al2O3 MoO3 As2O3 SO2 (measured) SO3 (calculated)b P2O5 F Cl SiO2 H2O Total Cations K Na A (Na + K) Fe Al Mo B (Fe + Al + Mo) S P As X (S + P + As)

9.43 0.01 42.1 46.74 1.34 0.29 0.1 24.81 31 0.07 0.12 0.01

9.43 0.01 43.63 48.43 0.69 0.18 0.29 24.78 30.97 0.05 0.09 0.01

10.88 88.98

9.84 90.06

7.15 0.11 39.76 44.13 0.27 0.00 0.04 23.34 29.16 0.43 0.01 0.09 0.78 10.88 82.83

6.82 0.13 33.40 37.07 0.75 0.00 0.04 19.60 24.49 0.16 0.00 0.41 1.22 9.42 71.94

6.79 0.07 37.69 41.83 0.33 0.00 0.01 22.23 27.78 1.00 0.07 0.11 0.82 10.53 79.65

7.17 0.21 39.57 43.92 0.25 0.00 0.00 22.93 28.65 1.22 0.09 0.06 0.48 10.91 82.88

7.03 0.12 33.51 37.20 0.34 0.00 0.03 20.69 25.86 0.18 0.00 0.29 0.14 9.35 71.68

5.57 0.43 34.99 38.84 0.37 0.00 0.05 20.98 26.22 0.11 1.31 0.49 0.65 9.62 74.55

0.93 0 0.93 2.72 0.12 0.01 2.85 1.8 0 0 1.8

0.93 0 0.93 2.81 0.06 0.01 2.88 1.79 0 0.01 1.81

a b

Fe2O3 = FeO × 1.11. SO3 = SO2 × 1.25.

0.75 0.02 0.77 2.75 0.03 0.002 2.78 1.81 0.03

0.83 0.02 0.85 2.67 0.08 0.003 2.75 1.76 0.01

0.74 0.01 0.75 2.69 0.03 0.000 2.73 1.78 0.07

0.75 0.03 0.79 2.73 0.03 0.000 2.75 1.77 0.09

0.86 0.02 0.89 2.70 0.04 0.002 2.74 1.87 0.02

0.66 0.08 0.74 2.74 0.04 0.003 2.78 1.84 0.01

1.84

1.77

1.85

1.86

1.88

1.85

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

8

100

6 150

4

200

2

mass loss (mg)

50

K2O (wt %)

A

10 K2O (wt % oxide)

0

PB

0 0

100

200

300

400

250 500

Time in HF (min) Explanation K O concentration 2 mass loss

12 10 8 6 4 2 10 8 6 4 2 10 8 6 4 2 0

0 10 20 30 40

B PIT VWL 0007

0 15 30 45 60 ESS VWL 0001 0 15 30 45 60 CAS VWL 0002 75 0 5 10 15 20 25 30 35

mass loss (mg) mass loss (mg) mass loss (mg)

12

K2O (wt %) K2O (wt %)

26

Time in HF (min)

Fig. 1. Potassium concentration and mass loss in untreated and treated (A) PB and (B) RRV samples.

3.4. Removal of silicates with HF BSE images show changes in the modal abundance of silicates present in the artificial mixture of PB and FC-2 (B-samples) and in RRV samples. Although FC-2 is clearly visible in sample B-0 (Fig. 3), only one grain of FC-2 was evident in BSE images of polished B-30. No FC-2 was found in B-samples treated for more than 30 min in HF. Additionally, FC-2 grains placed alone in a beaker of 25% HF dissolved completely within 25 min. We chose to treat RRV samples for 30 min based on these dissolution times for sanidine mixed with pure jarosite. Both treated and untreated RRV samples consist of a mixture of jarosite, quartz, potassium feldspar, clay (primarily illite), and goethite with intergrown jarosite and clay, in some cases, comprising more than 50% of the sample (Fig. 4). HF-treated samples have fewer contaminants than untreated samples, suggesting that HF treatment dissolves some of the silicate minerals. Although HF treatment removed some of the contaminants, BSE images indicate that

potassium-bearing silicates, including illite and sanidine, continue to make up approximately 20 to 30% of the treated samples. 3.5.

40

Ar/ 39Ar results for PB jarosite

40

Ar/ 39Ar ages for PB jarosite, like major oxide compositions, do not change significantly with time in HF. Untreated PB jarosite (A-0) produced a plateau age of 9.58 ± 0.06 Ma (all uncertainties are 2sigma) while PB jarosite treated with HF for 8 h (A-480) had a plateau age of 9.64 ± 0.04 Ma. These ages overlap with each other and with previously published ages for PB (9.42 ± 0.22 Ma; Lueth et al., 2005) within 2-sigma uncertainty. The untreated experimental mixture of PB + FC-2 (C-0) approached, but did not meet, plateau criteria, with apparent age overlapping with pure PB at low-power steps (2.5 to 4 W; 0.63 to 1 W/mm 2) and climbing when laser power exceeded 6 W (1.5 W/mm 2). In contrast, the experimental mixture with no minimum grain size control treated with HF for 30 min (B-30) had a plateau age of 9.58 ± 0.05 Ma across ~80% of the 39Ar released. Age

Fig. 2. Secondary electron images of unpolished samples show the effects of HF on PB jarosite after (A) 0, (B) 30, (C) 240, and (D) 480 min.

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

27

Table 5 Summary of age results for PB jarosite and PB + FC-2 mixture. Sample

Description

Time in HF (min)

Percent of spectrum in plateau

Apparent age ± 2σ error (Ma)

A-0 A-480 C-0 B-30

PB PB PB + FC-2 PB + FC-2

0 480 0 30

100 79.3 0 79.1

9.59 ± 0.05a 9.68 ± 0.08a 10.43 ± 0.13b 9.58 ± 0.05a

PB = Peña Blanca jarosite. FC-2 = Fish Canyon sanidine. a Plateau age. b Integrated age.

0.14 Ma (33.5% of 39Ar released) for PIT VWL 0005 were calculated from steps that approached, but did not meet, plateau criteria. Despite discordance in the age spectra, treated RRV samples show no sign of argon loss in early steps. Fig. 3. BSE images of unpolished B-0 show the relative amounts of PB (high-Z, bright phase) and FC-2 (lower-Z, darker phase).

spectra of treated and untreated PB jarosite show no evidence of argon loss. Age results for pure jarosite and experimental PB + FC-2 mixtures are summarized in Table 5 and age spectra are shown in Fig. 5. 3.6.

40

Ar/ 39Ar results for RRV jarosite

Apparent age in both HF-treated and untreated RRV samples generally increases in higher wattage steps (Fig. 6). However, treated RRV samples yielded imprecise apparent ages (Table 6) that are significantly younger than untreated RRV samples analyzed in Lueth et al. (2006). Samples ESS VWL 0001, and CAS VWL 0002 produced weighted mean plateau ages of 0.36 ± 0.17 Ma, and 0.15 ± 0.18 Ma over 65.9%, and 79.3% of the cumulative 39Ar released respectively. Age spectra for samples PIT VWL 0007 and PIT VWL 0005 (Fig. 6A and C) were slightly discordant. Weighted mean ages of 0.24 ± 0.12 Ma (49.7% of 39Ar released) for PIT VWL 0007 and 0.09 ±

A

4. Discussion This study aimed to determine 1) the effectiveness of removing silicate contaminants from jarosite using HF, 2) the effects of HF on jarosite crystal chemistry, and 3) the effects of HF treatment on supergene jarosite's apparent age. Based on EMPA, HF removed sanidine from the experimental mixture of sanidine and jarosite within thirty minutes without altering jarosite chemistry or affecting the apparent age of PB jarosite. However, thirty minutes of treatment was insufficient to completely remove sanidine and other silicates from RRV jarosite. This difference in the effectiveness of HF in removing silicates is likely due to the following factors: 1) RRV samples are significantly more contaminated than experimental samples, with non-jarosite minerals comprising more than 50% of the untreated mixture; 2) RRV jarosite grows on clay and sanidine surfaces, armoring these minerals and protecting them from complete dissolution in HF. In addition to silicate dissolution, jarosite partially dissolved in HF. Despite textural evidence of dissolution, however, neither PB nor RRV

B

jar

jar

clay ksp

chl

qtz 100 µm BSE 15 kV

100 µm BSE 15 kV

D

C ksp

clay

gth

jar

gth

500 µm BSE 15 kV

50 µm BSE 15 kV

Fig. 4. BSE images of untreated RRV samples show silicates and goethite mixed with jarosite in (A) PIT-VWL-0007, (B) ESS-VWL-0001, and (C) CAS-VWL-0002. (D) HF-treated sample PIT VWL 0007 has fewer silicate contaminated than the untreated sample.

28

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

%40Ar*

80

B

A

60 40 20

Apparent Age

A-480

A-0

0 12

1

9.58 ± 0.06 Ma (MSWD = 1.52, p = 0.14)

10 1

8

3

2

4

5

6

8

7

9.64 ± 0.04 Ma (MSWD = 0.29, p = 0.96)

9 10 2

3

6 80 40 0

5

4

6

8 9 10

7

Integrated Age = 9.78 ± 0.06 Ma

Integrated Age = 9.58 ± 0.09 Ma

C

D

14

B-30

9

C-0

12

100 80 60 40 20 0 18 16 14 12 10 8 6 100 80 60 40 20 0 12

7

10

9.58 ± 0.05 Ma (MSWD = 0.69, p = 0.56)

6

8

2

4

3

4

6

10

8 12

2.5

Integrated Age = 9.48 ± 0.08 Ma

Integrated Age = 9.98 ± 0.09 Ma

6 0

10

20

30

40

50

60

70

80

90 100 0

10

Cumulative %39Ar released

20

30

40

50

60

70

14

15

8

6 90 100

80

Cumulative %39Ar released

Apparent Age (Ma)

%40Ar*

Fig. 5. Age spectra for PB and PB + FC-2 experimental samples (A) A-0, (B) A-480, (C) C-0 and (D) B-30. Each heating step is labeled with laser power in watts. Plateau ages produced by the HF-treated samples (B and D) overlap plateau ages produced by untreated PB (A) within analytical uncertainty.

100 60 20 50

A

%40Ar*

100 60 20 20

ESS-VWL-1

PIT-VWL-7

16

40

12

30 6.4

5.6

0.8

20

0.8

8

7.2

4

7

6

8.5

4

4

0.24 ± 0.12 Ma (MSWD = 3.89) 3

2.5

3

2 6

4

9

7

Integrated Age (treated) = 0.67 ± 0.11 Ma Integrated Age (untreated) = 16.9 ± 0.5 Ma

-10 100

8

0.36 ± 0.17 Ma (MSWD = 2.49)

2.4

10 0

Apparent Age (Ma)

B

2.5 1.5

9

0

7

6

4

3

Integrated Age (treated) = 0.9 ± 0.2 Ma Integrated Age (untreated) = 3.1 ± 0.3 Ma

-4

C

60 20 20

D

PIT-VWL-5

100 60 20 4

CAS-VWL-7

16

8.5

1.1 ± 0.4 Ma (MSWD = 4.83)

2

12

2

4 3

0.8

8

2.5

6 0.09 ± 0.14 Ma (MSWD = 2.53) 2 3

4 0

2.5 1.5

-4 0

3

4

7

4

7

0 6

9

0.15 ± 0.18 Ma (MSWD = 1.78) 7

4

40

Cumulative

60

%39Ar

-2

9

7

6

Integrated Age (treated) = 0.64 ± 0.13 Ma Integrated Age (untreated) = 2.6 ± 0.3 Ma

20

3

8.5

6

80

Integrated Age (treated) = 0.0 ± 0.2 Ma Integrated Age (untreated) = 1.5 ± 0.5 Ma

100 0

20

Released Explanation

40

60

80

Cumulative

%39Ar

Released

-4 100

untreated sample (Lueth et al., 2006) treated sample (this study) Fig. 6. Age spectra for HF-treated RRV samples (A) PIT-VWL-0007, (B) ESS-VWL-0001, (C) PIT-VWL-0005, and (D) CAS-VWL-0002. Each heating step is labeled with laser power in watts. HF-treated RRV jarosite produced plateau ages. Apparent age increased in higher wattage steps in samples PIT VWL 0007, ESS VWL 0001, and PIT VWL 0005 (A, B, C). This increase in apparent age is accompanied by an increase in radiogenic yield (% 40Ar*), suggesting that a different, older phase is degassing in these steps.

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32 Table 6 Untreated and treated RRV sample ages. Sample PIT VWL 0007 PIT VWL 0005 ESS VWL 0001 CAS VWL 0002 a b

Table 7 Furnace degassing temperatures of different minerals.

Untreated apparent age ± 2σ (Ma)a

Treated apparent age ± 2σ (Ma)b

8.45 ± 0.38 0.80 ± 0.13 1.24 ± 0.17 No age assigned

0.24 ± 0.12 0.09 ± 0.14 0.36 ± 0.17 0.15 ± 0.18

Assuming that the previously published age reflects the actual age of the PB jarosite, the expected integrated age of sample C-0 is

%39Ar released in each step

50 45 40 35 30 25 20 15 10 5 0 6

8

10

12

14

16

laser power, W PB, less than 63 um

450 to 700a >850b 260 to 1100c

c

%FC-2 percent Fish Canyon sanidine 28.02 Ma age of Fish Canyon sanidine %PB percent Peña Blanca jarosite 9.42 Ma age of Peña Blanca jarosite (Lueth et al., 2004) [K2O]FC-2 12.2 wt.% (Bachmann et al., 2002) [K2O]PB 8.89 wt.% (average) [K2O]mixture weighted mean potassium concentration of the mixture (9.28 wt.%)

4

Furnace temperature, °C

Jarosite Sanidine Illite b

jarosite underwent systematic decreases in the concentration of any major oxides, including K2O, with treatment time. Sample A-480, pure PB treated for 8 h in HF, produced a plateau age identical within analytical uncertainty to untreated PB, indicating that HF does not lead to significant argon loss from highly crystalline jarosite. Although the bulk K2O concentration in PB remained unchanged, the 10-μm beam used to analyze jarosite in this study made it impossible to determine whether HF-treatment created fine-scale gradients in potassium concentrations within grains. Elevated apparent age in the first three steps of the age spectrum for sample A-480 may be due to such finescale gradients in K2O concentrations along grain edges. The degassing behavior of silicate contaminants controls their ability to affect jarosite's apparent age. Sample C-0 allowed us to monitor the influence of sanidine on jarosite's apparent age. If the mixture of sanidine and jarosite degassed completely at the same laser power, the sample's apparent age would be controlled by a mixing equation that takes into account K2O concentration and known age for each phase in addition to the relative amount of each mineral in the mixture: (%FC-2×28.02 Ma)([K2O]FC-2 /[K2O]mixture)+(%PB×9.42 Ma)([K2O]PB /[K2O]mixture)=expected integrated agemixture Where:

2

Mineral

a

From Lueth et al. (2006). This study.

0

PB, 73-125 um

29

FC-2, 73-125 um

FC-2, less than 73 um

Fig. 7. Graph showing gas extracted during laser step-heating of jarosite and sanidine of different size fractions. PB jarosite, regardless of control on minimum grain size, releases most of its argon between 2 and 6 W (0.5 to 1.5 W/mm2) when heated by a defocused laser. When minimum grain size is not controlled, sanidine also releases most of its 39Ar at power lower than 6 W (1.5 W/mm2). Sanidine with grain size between 73 and 125 μm releases most of its argon between 6 and 9 W (1.5 to 2.25 W/m2) when heated by a defocused laser.

Alpers et al. (1992). McIntosh et al. (1990). Wilson and Parry (1990); Jaboyedoff and Cosca (1999).

13.20 Ma. Instead, sample C-0 produced apparent ages that increased over the course of the entire age spectrum with an integrated age of 9.98 ± 0.09 Ma. The 3 W and 4 W steps produced apparent ages within analytical uncertainty of pure PB. Apparent age continued to rise in the higher power steps but did not reach the age of FC-2. This pattern suggests that jarosite dominates the signal in lowpower steps while sanidine exerts a stronger influence over apparent age when laser power is at or above 6 W. The low integrated age may be an artifact of reactive material expelled from the jarosite accumulating on the cover slip. As this material accumulated, the cover slip became opaque to the laser, so the laser cannot interact with the sample at higher wattages. Fig. 7 shows that sanidine with grain size controlled between 73 and 125 μm releases 88% of its 39Ar above 6 W while jarosite in the same grain size range releases ~84% below 6 W. When minimum grain size is not controlled, the degassing pattern for jarosite is similar, but sanidine degasses at lower wattages. The initial step of degassing in C-0 (Fig. 5) may reflect initial degassing by jarosite alone. Subsequent and rising steps can be attributed to increasing sanidine contribution at higher wattages. Jarosite/sanidine mixtures with other silicates, particularly clays (e.g. illite), degas at lower temperatures than sanidine when stepheated in a furnace (Table 7) also influencing the apparent age of the jarosite mixture. RRV jarosite is mixed with Oligocene-age sanidine and hydrothermal clay minerals (particularly illite) (24.86 ± 0.15 Ma; Lueth et al., 2006). When heated isothermally in a furnace, the lowest temperatures at which illite degasses overlap with temperatures at which jarosite degasses (Table 7). Although illite can degas at lower temperatures than sanidine, HFtreated RRV samples, which have fewer silicate contaminants, yield significantly younger apparent ages than untreated RRV samples. Young apparent ages in early heating steps suggest that Oligocene phases did not degas significantly in the low wattage steps. This may be due to the differences in grain size and diffusion rates between jarosite and the silicate contaminants (Fig. 4). It is likely that jarosite degassed at lower power, dominating the young plateau steps in samples where apparent age does not climb at low laser power, and that the plateau ages of the low-power steps can be interpreted as the maximum age of jarosite formation. Additionally, total gas ages of the treated RRV samples are younger than the >7.5 Ma apparent age expected for a 70:30 mixture of RRV jarosite and Oligocene silicates. This estimate of the approximate minimum total gas age of the mixture is based on the equation: % silicates × 24.86 Ma ≅ minimum expected integrated age of the mixture. Potassium concentrations were not taken into consideration in this calculation because contaminant concentration and type differ in each sample. If Oligocene silicates, including clays, dominated the 40Ar/ 39Ar analysis, the apparent age would be approximately 7.5 Ma, more than 7 Ma older than the Pleistocene ages determined for RRV samples. This result suggests that the mixture of jarosite and silicate contaminants did not degas completely, and that material younger than the Oligocene illite dominates the apparent age. Although young jarosite most likely dominates the apparent age of the RRV samples, the continued presence of silicates and the degassing behavior of illite require that these results be treated as maximum ages in landscape evolution studies.

30

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

5. Conclusions This series of experiments and observations shows that jarosite can be treated with HF without compromising the ability to date it accurately. Neither coarsely crystalline, pure jarosite nor finer grains of RRV jarosite underwent significant potassium loss during HF treatment. Although HF-treated samples showed no major potassium loss and yielded reproducible ages, long treatment times may leach potassium along grain edges, leading to older apparent ages in early heating steps. This partial dissolution along grain edges had an impact on the size of the plateau but no effect on the plateau age of PB jarosite. Therefore, removing silicates from jarosite using HF does not compromise the integrity of dating the sulfate. Although jarosite can be dated following HF treatment, treatment times and the effectiveness may vary by contaminant type and textural relationship between jarosite and silicates. Sanidine can dissolve in 25% HF within 30 min, implying that long treatment times are unnecessary. However, 30 min of HF treatment did not remove 100% of the clay and feldspar contaminants from RRV

Appendix A.

ID

40

Power (W)

samples. RRV jarosite does not show evidence of potassium-loss as a function of HF treatment, so longer treatment times in more HF are feasible when preparing fine-grained supergene jarosite intergrown with silicates for 40Ar/ 39Ar dating. Combining HF treatment with gravitational settling (Polyak et al., 1998, 2006) may improve the purity of jarosite mineral separates and lead to increased precision and accuracy in dating weathering jarosite. Acknowledgments Chevron Mining Corporation paid for dating RRV samples. The NM Geochronology Research Laboratory provided funding for student work during the Practical Aspects of Mass Spectrometry class in 2007. The Geological Society of America and New Mexico Geological Society's graduate student research grant programs provided additional funding. Gabriel Graf, Ryan Crow, Jana Stankova, and Kira Lueth provided laboratory and field assistance. Thanks to two anonymous reviewers whose comments improved the manuscript.

Ar/ 39Ar analytical data.

40

Ar/39Ar

37

Ar/39Ar

36 Ar/39Ar (×10− 3)

A-0, jarosite powder, 8.97 mg, J = 0.0002386 ± 0.06%, D = 1.002 ± 0.001, NM-204C, A 1 71.17 − 0.0083 166.7 B 2 50.90 − 0.0004 97.36 C 3 41.29 − 0.0004 63.14 D 4 36.70 − 0.0004 48.21 E 5 35.87 0.0008 45.44 F 6 34.97 0.0009 43.09 G 7 34.48 0.0002 42.15 H 8 35.63 0.0024 45.43 I 9 37.76 0.0026 52.72 Xi J 10 40.65 0.0044 63.86 Integrated age ± 2σ n = 10 Plateau ±2σ Steps A-I n=9 MSWD = 1.52 Isochron ± 2σ Steps A-I n=9 MSWD = 1.65

39 ArK (× 10− 15 mol)

Lab# = 56892-03 0.501 2.39 2.73 3.50 3.36 3.44 1.75 1.497 1.140 0.998 21.3 20.3

A-480, jarosite powder, 9.58 mg, J = 0.0002372 ± 0.07%, D = 1.002 ± 0.001, NM-204C, Lab# = 56895-01 Xi A 1 41.43 − 0.0097 23.97 0.164 XiB 2 28.45 − 0.0006 16.28 1.334 C 3 26.30 − 0.0007 12.40 2.63 D 4 25.84 − 0.0005 11.11 2.90 E 5 26.01 0.0000 11.62 4.52 F 6 29.25 0.0010 22.41 3.69 G 7 36.78 0.0024 48.31 1.64 H 8 49.89 0.0059 91.62 0.794 I 9 44.37 0.0039 72.64 0.825 J 10 43.12 0.0018 69.80 0.898 Integrated age ± 2σ n = 10 19.4 Plateau ±2σ Steps C-J n=8 MSWD = 0.29 17.9 Isochron ± 2σ Steps C-J n=8 MSWD = 0.28 C-0, jarosite, J = 0.0024979 ± 0.08%, D = 1.002 ± 0.001, NM-249J, Lab# = 61180-02 XA 1 72.54 0.0595 224.8 0.015 XB 3 5.418 0.0014 11.59 16.5 XC 3 2.661 0.0003 2.019 14.2 XD 4 2.767 0.0004 2.303 13.4 XE 6 3.119 0.0006 2.855 9.72 XF 7 3.823 0.0017 4.263 8.21 XG 9 5.184 0.0025 6.342 4.56 Integrated age ± 2σ n=7 66.6 Isochron ± 2σ Steps A-G n=7 n = 7 MSWD = 404.23 C-0, jarosite, J = 0.0024979 ± 0.08%, D = 1.002 ± 0.001, NM-249J, Lab# = 61180-02 B-30, jarosite powder, 8.8 mg, J = 0.000238 ± 0.07%, D = 1.002 ± 0.001, NM-204C, Lab# = 56897-01 A 2 34.47 0.0010 40.27 1.12 B 4 31.68 − 0.0011 31.30 1.57 C 6 28.86 − 0.0004 22.06 2.37 D 8 28.41 − 0.0007 20.65 1.45

K/Ca

40 Ar* (%)

39 Ar (%)

Age (Ma)

± 1σ (Ma)

– – – – 628.9 579.7 2280.8 214.7 192.7 115.5 1134.6 771.6 ± 1286.6 40 Ar/36Ar =

30.8 43.5 54.8 61.2 62.6 63.6 63.9 62.3 58.7 53.6 K2O = 3.82%

2.4 13.6 26.4 42.8 58.6 74.7 82.9 90.0 95.3 100.0

9.40 9.500 9.712 9.639 9.631 9.543 9.451 9.532 9.520 9.35 9.566 9.584 9.60

0.22 0.094 0.064 0.052 0.052 0.054 0.072 0.073 0.099 0.12 0.087 0.057 0.15

– – – – – 493.5 213.2 85.8 132.1 289.1 875.0 332.2 ± 253.5 40 Ar/36Ar =

82.9 83.1 86.1 87.3 86.8 77.4 61.2 45.7 51.6 52.1 K2O = 3.28%

14.63 10.083 9.655 9.623 9.629 9.654 9.600 9.73 9.77 9.59 9.717 9.639 9.628

0.51 0.071 0.038 0.037 0.032 0.043 0.088 0.14 0.16 0.14 0.060 0.038 0.057

8.6 361.6 1808.6 1377.4 819.6 304.0 203.3 527.6 40 Ar/36Ar =

8.4 36.7 77.5 75.3 72.9 67.0 63.8

0.0 24.8 46.1 66.2 80.8 93.2 100.0

375.6 ± 6.1

27.33 8.92 9.23 9.33 10.18 11.47 14.81 9.98 8.80

18.25 0.10 0.04 0.04 0.05 0.07 0.11 0.09 0.09

512.6 – – –

65.5 70.8 77.4 78.5

13.7 32.8 61.8 79.5

9.658 9.600 9.561 9.548

0.070 0.053 0.037 0.050

95.3 294.8 ± 6.2

0.8 7.7 21.3 36.2 59.5 78.5 87.0 91.1 95.4 100.0 92.3

297.0 ± 5.5

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

31

Appendix A. (continued) K/Ca

40 Ar* (%)

39 Ar (%)

Age (Ma)

± 1σ (Ma)

876.2 – 453.8 − 1435.902 88.0 ± 0.0 40 Ar/36Ar =

57.2 48.3 32.1 K2O = 1.50%

90.2 96.5 100.0

9.141 9.07 8.83 9.478 9.578 9.44

0.098 0.14 0.25 0.074 0.050 0.21

PIT VWL 0007, jarosite, 6.27 mg, J = 0.0001143 ± 0.48%, D = 1.0006 ± 0.001, NM-211A, Lab# = 57355-01 A 3 21.71 0.0146 71.16 0.295 B 3 18.83 0.0144 58.52 0.202 C 4 27.31 0.0204 87.20 0.222 XD 6 54.11 0.0194 166.9 0.528 XE 7 161.0 0.0575 526.6 0.108 XF 9 238.5 0.0313 778.4 0.091 Integrated age ± 2σ n=6 1.45 Plateau ±2σ Steps A-C n=3 MSWD = 3.89 0.719 Isochron ± 2σ Steps A-F n=6 MSWD = 15.44

35.0 35.4 25.1 26.3 8.9 16.3 23.8 32.039 ± 11.705 40 Ar/36Ar =

3.1 8.1 5.6 8.8 3.4 3.5 K2O = 0.77%

0.138 0.314 0.317 0.986 1.12 1.74 0.67 0.24 0.133

0.049 0.058 0.057 0.062 0.22 0.24 0.11 0.123 0.031

PIT VWL 0005, jarosite, 7.92 mg, J = 0.0001146 ± 0.34%, D = 1.0006 ± 0.001, NM-211A, Lab# = 57351-01 A 2 284.5 0.1694 951.4 0.024 B 3 43.30 0.0077 146.6 0.111 C 3 26.77 0.0138 91.87 0.157 D 4 27.42 0.0064 89.92 0.285 XE 6 58.39 0.0200 192.5 0.370 XF 7 86.67 0.0145 274.0 0.460 XG 9 122.6 0.0218 392.3 0.264 Integrated age ± 2σ n=7 1.67 Plateau ±2σ Steps A-D n=4 MSWD = 2.53 0.576

3.0 66.6 36.9 79.6 25.5 35.1 23.4 29.7 62.322 ± 68.142

1.2 − 0.1 − 1.4 3.1 2.6 6.6 5.5 K2O = 0.71%

0.69 0.00 − 0.079 0.174 0.312 1.180 1.38 0.64 0.09

0.54 0.14 0.087 0.058 0.077 0.098 0.14 0.13 0.144

ESS VWL 0001, jarosite, 6.25 mg, J = 0.000115 ± 0.25%, A 2 1290.8 0.5157 B 3 191.4 0.0068 C 3 76.39 0.0042 D 4 68.74 0.0116 E 6 116.2 0.0111 Xi F 7 139.5 0.0187 Xi G 9 282.9 0.0314 Integrated age ± 2σ n=7 Plateau ±2σ Steps A-E n=5

0.99 75.3 121.3 44.0 45.9 27.3 16.2 26.7 62.8 ± 88.9

1.1 0.1 1.6 2.6 2.4 4.6 5.8 K2O = 0.91%

2.8 0.05 0.251 0.364 0.58 1.33 3.43 0.92 0.36

1.4 0.22 0.094 0.080 0.12 0.13 0.36 0.24 0.17

0.34 0.06 0.35 0.021 − 1.18 − 0.06 0.00 0.15 − 0.044

0.18 0.17 0.14 0.097 0.21 0.22 0.22 0.18 0.041

ID

Power (W)

40

Ar/39Ar

37

Ar/39Ar

36 Ar/39Ar (×10− 3)

39 ArK (× 10− 15 mol)

Xi F 12 37.29 0.0006 53.94 0.877 XiG 14 43.88 − 0.0023 76.81 0.520 Xi H 15 64.12 0.0011 147.2 0.282 Integrated age ± 2σ n=7 8.18 Plateau ±2σ Steps A-D n=4 MSWD = 0.69 6.50 Isochron ± 2σ Steps A-D n=4 MSWD = 0.01 B-30, jarosite powder, 8.8 mg, J = 0.000238 ± 0.07%, D = 1.002 ± 0.001, NM-204C, Lab# = 56897-01

D = 1.0006 ± 0.001, NM-211B, Lab# = 57358-01 4322.5 0.018 647.0 0.138 254.4 0.216 226.7 0.310 383.6 0.441 450.3 0.428 901.4 0.154 1.71 MSWD = 2.49 1.12

CAS VWL 0007, jarosite, 8.03 mg, J = 0.0001141 ± 0.36%, D = 1.0006 ± 0.001, NM-211A, Lab# = 57353-01 A 3 186.1 − 0.0031 624.0 0.437 – B 3 107.9 − 0.0130 364.0 0.313 – C 4 84.60 − 0.0102 280.5 0.393 – D 6 87.94 0.0198 297.2 0.850 25.7 Xi E 7 142.4 0.0807 501.2 0.274 6.3 Xi F 9 183.7 0.0794 622.8 0.246 6.4 Integrated age ± 2σ n=6 2.51 26.1 Plateau ±2σ Steps A-D n=4 MSWD = 1.78 1.99 25.7 ± 0.0 40 Isochron ± 2σ Steps A-D n=4 MSWD = 2.21 Ar/36Ar = Notes: Isotopic ratios corrected for blank, radioactive decay, and mass discrimination, not corrected for interfering reactions. Errors quoted for individual analyses include analytical error only, without interfering reaction or J uncertainties. Integrated age calculated by summing isotopic measurements of all steps. Integrated age error calculated by quadratically combining errors of isotopic measurements of all steps. Plateau age is inverse-variance-weighted mean of selected steps. Plateau age error is inverse-variance-weighted mean error (Taylor, 1982) times root MSWD where MSWD > 1. Plateau error is weighted error of Taylor (1982). Decay constants and isotopic abundances after Steiger and Jäger (1977). ‘X’ preceding sample ID denotes analyses excluded from plateau age calculations. Weight percent K2O calculated from 39Ar signal, sample weight, and instrument sensitivity. Ages calculated relative to FC-2 Fish Canyon Tuff sanidine interlaboratory standard at 28.02 Ma. Decay constant (LambdaK (total)) = 5.543e−10/a. Correction factors: (39Ar/37Ar)Ca = 0.0007 ± 5e−05 (36Ar/37Ar)Ca = 0.00028 ± 2e−05 (38Ar/39Ar)K = 0.013 (40Ar/39Ar)K = 0.01 ± 0.0002

79.5 307.8 ± 18.0

20.4 34.4 49.7 86.2 93.7 100.0 49.7

308.8 ± 2.7

1.4 8.1 17.4 34.5 56.7 84.2 100.0 34.5

1.1 9.2 21.8 40.0 65.9 91.0 100.0 65.9

0.9 0.3 2.0 0.1 − 4.0 − 0.2 K2O = 1.05%

17.4 29.8 45.5 79.3 90.2 100.0 79.3

298.2 ± 5.7

32

K.E. Samuels-Crow et al. / Chemical Geology 314–317 (2012) 23–32

References Alpers, C.N., Rye, R.O., Nordstrom, D.K., White, L.D., King, B.S., 1992. Chemical, crystallographic, and isotopic properties of alunite and jarosite from acid hypersaline Australian lakes. Chemical Geology 96, 203–226. Arehart, G.B., O'Neil, J.R., 1993. D/H ratios of supergene alunite as an indicator of paleoclimate in continental settings. In: Swart, P.K., Lohmann, K.C., McKenzie, J., Savin, S. (Eds.), Climate change in continental isotopic records: AGU Geophysical Monograph, 78, pp. 277–284. Bachmann, O., Dungan, M.A., Lipman, P.W., 2002. The Fish Canyon magma body, San Juan volcanic field, Colorado: rejuvenation and eruption of an upper-crustal batholiths. Journal of Petrology 43 (8), 1469–1503. Elwood-Madden, M.E., Madden, A.S., Rimstidt, J.D., 2009. How long was Meridiani Planum wet? Applying a jarosite stopwatch to determine the duration of aqueous diagenesis. Geology 37 (7), 635–638. Jaboyedoff, M., Cosca, M.A., 1999. Dating incipient metamorphism using 40Ar/39Ar geochronology and XRD modeling: a case study from the Swiss Alps. Contributions to Mineralogy and Petrology 135, 93–113. Klingelhöfer, G., Morris, R.V., Bernhardt, B., Schröder, C., Rodinov, D.S., de Souza, P.A., Yen, A., Gellert, R., Evlanov, E.N., Zobkov, B., Foh, J., Bonnes, U., Kankeleit, E., Gütlich, P., Ming, D.W., Renz, F., Wdowiak, T., Squyers, S.W., Arvidson, R.E., 2004. Jarosite and hematite at Meridiani Planum from Opportunity's Mössbauer spectrometer. Science 306, 1740–1745. Kula, J., Baldwin, S.L., 2011. Jarosite, argon diffusion, and dating aqueous mineralization on Earth and Mars. Earth and Planetary Science Letters 310 (3–4), 314–318. http:// dx.doi.org/10.1016/j.epsl.2011.08.006. Lueth, V.W., Rye, R.O., Peters, L., 2004. 40Ar/39Ar data repository for “Sour gas” hydrothermal jarosite: ancient to modern acid-sulfate mineralization in the southern Rio Grande rift. NMBGMR Open-File Report OF-AR-22. 9 pp. Lueth, V.W., Rye, R.O., Peters, L., 2005. “Sour gas” hydrothermal jarosite: ancient to modern acid-sulfate mineralization in the southern Rio Grande rift. Chemical Geology 215, 339–360. Lueth, V.W., Campbell, A.R., Peters, L.M., 2006. Final report (Phase I) on the geochronological (40Ar/39Ar) dating of jarosite and alunite samples from the

Red River area alteration scars and the Questa Mine: the timing of alteration scar formation and weathering: Questa Rock Pile Weathering and Stability Project, Unpublished report, 18 pp. McIntosh, W.C., Sutter, J.F., Chapin, C.E., Kedzie, L.L., 1990. High-precision 40Ar/39Ar geochronology of ignimbrites in the Mogollon–Datil volcanic field, southwestern New Mexico. Bulletin of Volcanology 52, 584–601. Papike, J.J., Karner, J.M., Shearer, C.K., 2006. Comparative planetary mineralogy: implications of martian and terrestrial jarosite. A crystal chemical perspective. Geochimica et Cosmochimica Acta 70, 1309–1321. Papike, J.J., Burger, P.V., Karner, J.M., Shearer, C.K., Lueth, V.W., 2007. Terrestrial analogs of martian jarosites: major, minor element systematics and Na–K zoning in selected samples. American Mineralogist 92, 444–447. Polyak, V.J., McIntosh, W.C., Güven, N., Provencio, P., 1998. Age and origin of Carlsbad Cavern and related caves from 40Ar/39Ar of alunite. Science 279, 1919–1922. Polyak, V.J., McIntosh, W.C., Provencio, P.P., Güven, N., 2006. Alunite and natroalunite tell the story — the age and origin of Carlsbad Cavern, Lechuguilla Cave, and other sulfuric-acid type caves in the Guadalupe Mountains. New Mexico Geological Society 57th Field Conference Guidebook, pp. 203–209. Smith, A.M.L., Hudson-Edwards, K.A., Dubbin, W.E., Wright, K., 2006. Dissolution of jarosite [KFe3(SO4)2(OH)6] at pH 2 and 8: insights from batch experiments and computational modeling. Geochimica et Cosmochimica Acta 70, 608–621. Vasconcelos, P.M., 1999. K–Ar and 40Ar/39Ar geochronology of weathering processes. Annual Review of Earth and Planetary Sciences 27, 183–229. Vasconcelos, P.M., Conroy, M., 2003. Geochronology of weathering and landscape evolution, Dugald River valley, NW Queensland, Australia. Geochimica et Cosmochimica Acta 67, 2913–2930. Wasserman, M.D., Rye, R.O., Bethke, P.M., Arribas, A., 1992. Methods for separation and total stable isotope analysis of alunite. USGS Open File Report 92-9. 20 pp. Wilson, P.N., Parry, W.T., 1990. Mesozoic hydrothermal alteration associated with gold mineralization in the Mercur district, Utah. Geology 18, 866–869.