low&temperature and low atmospheric pressure infrared ... · journal of geophysical research,...

JOURNAL OF GEOPHYSICAL RESEARCH, VOL. 100, NO. E3, PAGES 5369-5379, MARCH 25, 1995

Low-temperature and low atmospheric pressure infrared reflectance spectroscopy of Mars soil analog materials

Janice L. Bishop • Departments of Chemistry and Geological Sciences, Brown University, Providence, Rhode Island

Carl6 M. Pieters

Department of Geological Sciences, Brown University, Providence, Rhode Island

Abstract. Infrared reflectance spectra of carefully selected Mars soil analog materials have been measured under low atmospheric pressures and temperatures. Chemically altered montmorillonites containing ferrihydrite and hydrated ferric sulfate complexes are examined, as well as synthetic ferrihydrite and a palagonitic soil from Haleakala, Maui. Reflectance spectra of these analog materials exhibit subtle visible to near-infrared features, which are indicative of nanophase ferric oxides or oxyhydroxides and are similar to features observed in the spectra of the bright regions of Mars. Infrared reflectance spectra of these analogs include hydration features due to structural OH, bound H20, and adsorbed H20. The spectral character of these hydration features is highly dependent on the sample environment and on the nature of the H20/OH in the analogs. The behavior of the hydration features near 1.9 gm, 2.2 gm, 2.7 gm, 3 gm, and 6 gm are reported here in spectra measured under a Marslike atmospheric environment. In spectra of these analogs measured under dry Earth atmospheric conditions the 1.9-gm ..band depth is 8-17%; this band is much stronger under moist conditions. Under Marslike atmospheric conditions the 1.9-gm feature is broad and barely discernible (1-3 % band depth) in spectra of the ferrihydrite and palagonitic soil samples. In comparable spectra of the ferric sulfate-bearing montmorillonite the 1.9-gm feature is also broad, but stronger (6% band depth). In the low atmospheric pressure and temperature spectra of the ferrihydrite-bearing montmorillonite this feature is sharper than the other analogs and relatively stronger (6% band depth). Although the intensity of the 3-gm band is weaker in spectra of each of the analogs when measured under Marslike conditions, the 3-gm band remains a dominant feature and is especially broad in spectra of the ferrihydrite and palagonitic soil. The structural OH features observed in these materials at 2.2-2.3 gm and 2.75 gm remain largely unaffected by the environmental conditions. A shift in the Christiansen feature towards shorter wavelengths has also been observed with decreasing atmospheric pressure and temperature in the midinfrared spectra of these samples.

1. Introduction

Spectral experiments on particulate, ferric-containing analogs to Martian soil have been performed to facilitate interpretation of remotely acquired spectra. These analog materials include chemically altered montmorillonites that contain nanophase ferric oxyhydroxides or ferric sulfate complexes, a palagonitic soil and ferrihydrite. Since reflectance spectra of such minerals and • soils are environmentally sensitive, accurate spectral measurements of these samples as Mars analogs requires measurement under Marslike atmospheric conditions. The results presented here encompass spectral experiments of laboratory analog materials to Martian bright region soils measured in ar

1Now at Deutsche Forschungsanstalt ftir Luft- und Raumfahrt, Berlin, Germany.

Copyright 1995 by the American Geophysical Union.

Paper number 94JE03331. 0148-0227/95/94JE-03331 $05.00

environmental chamber under controlled atmospheric and temperature conditions. As the atmospheric pressure and temperature are reduced in these experiments the partial pressure of H:O in the sample chamber is decreased and the samples are dehydrated.

The presence of smectites in bright region surface material on Mars has been suggested based on analyses of chemical measurements from Viking [Toulrnin et al., 1977; Baird et al., 1977] and is further supported by identification of smectites in some of the SNC meteorites [Gooding et al., 1991; Treirnan et al., 1993]. Spectroscopic analyses in the visible to near infrared have shown that chemically altered, ferric-enriched smectites prepared in the laboratory (e.g., ferrihydrite-bearing montmorillonites) exhibit important similarities to the soils on Mars [Bishop et al., 1993a; Bishop, 1994]. Visible absorption bands centered near 0.91 I.tm are found in ferrihydrite and ferrihydrite-bearing montmorillonites (J. L. Bishop et al., Reflectance spectroscopy of ferric sulfate- bearing montmorillonites as Mars soil analog materials, submitted to Icarus, 1994; hereinafter referred to as submitted manuscript) and also in the bright soils of Arabia on Mars [McCord et al., 1982; Murchie et al., 1993a]. Montmorillonites containing hydrated ferric sulfate

5369

5370 BISHOP AND PIETERS: REFLECTANCE SPECTRA OF MARS SOIL ANALOGS

complexes exhibit a ferric absorption feature near 0.88 I. tm (J. L. Bishop et al., submitted manuscript, 1994), which is also observed in the surface material at Lunae Planum [McCord et al., 1977; Murchie et al., 1993b]. Geochemical and spectral analyses of palagonitic soils indicate that they may have formed on Mars as well [Allen et al., 1981; Singer, 1982]. Spectral measurements of palagonitic soils indicate that these materials have a ferric band centered near 0.86 gm and a spectral brightness of about 30-40% reflectance, near that of soils on Mars in the bright regions [Singer, 1982; Morris et al., 1990].

Smectites, ferrihydrite, and palagonitic soils all contain structural OH bonded to metal cations, adsorbed H20, and bound H20. The infrared absorption features due to OH and H20 have been measured under variable environmental conditions

in smectites [Prost, 1975; Bruckenthal, 1987; Coyne et al., 1990a,b; Johnston et al., 1992; Bishop et al., 1994] and in palagonitic soils [Bruckenthal, 1987]. Preliminary experiments involving a variety of Mars analogs under differing environmental conditions examined the influence of exposure history on water content and absorption features due to H20 [Bishop et al., 1993b]. These experiments showed that reflectance spectra measured under dry conditions of ferrihydrite, ferric sulfate-bearing montmorillonite, and palagonitic soils contain sufficient bound H20 to, retain a strong 3-gm band similar to that observed in spectra of Mars. Spectra measured under similar conditions of naturally occurring smectites (on Earth) and also many ferric-bearing smectites exhibit a weaker 3-I. tm band.

Low-temperature reflectance spectra of hydrated minerals have exhibited changes in the band shape of the hydration features [Clark, 1981a]. Reflectance spectra of Mars soil analog materials dehydrated under reduced atmospheric pressures have also shown variations in the spectral character of the hydration features [Bishop et al., 1993b]. In the low atmospheric pressure experiments, as well as in some of the low-temperature experiments, these changes are due to dehydration of the sample. Additional experiments involving low-pressure and low-temperature reflectance spectra found changes in the hydration features and in the Christiansen feature of Mars soil analog materials [Bishop and Pieters, 1994]. The effects of the environment on water in clays, oxides, and salts are manifested in the spectral hydration bands measured via reflectance spectroscopy. Environmental influences on the spectral hydration features are especially important in interpreting observations of Mars where subtle spatial variations in the strengths of cation-OH and H20 absorptions have been observed in telescopic spectra [Bell and Crisp, 1993] and in spectra measured by the French near- infrared imaging spectrometer ISM on the Phobos 2 spacecraft [Bibring et al., 1990; Arnold, 1992; Murchie et al., 1992, 1993a]. For these reasons, reflectance spectra are measured in this study of Mars analog materials under the pressure and temperature conditions of temperate regions on Mars.

2. Background

2.1. Spectral Properties of Mars

Soderblom [1992] and Roush et al. [1993] have recently produced reviews of the spectroscopic properties of the surface of Mars. The spectral properties of the Martian atmosphere [e.g., Rosenqvist et al., 1992], H20 ice in the north polar cap

[e.g., Clark and McCord, 1982], and CO2 ice in the south polar cap [e.g., Calvin, 1990] have also been studied. Reflectance spectra of the bright areas on Mars are characterized by ferric crystal field transitions in the extended visible region [Adams and McCord, 1969; Singer et al., 1979; Sherman et al., 1982; Bell et al., 1990]. These spectra include absorptions near 0.5 I. tm and 0.9 I. tm, an inflection near 0.6 I. tm, and a reflectance maximum at 0.7-0.8 I. tm. The absorption features in the spectra of bright regions on Mars have been assigned to octahedrally coordinated Fe 3+ transitions [Sherman et al., 1982; Burns, 1993] and exhibit band centers ranging from 0.85 to 0.92 gm [McCord et al., 1977; Singer, 1982; Bell et al., 1990; Murchie et al., 1993a,b]. This implies heterogeneous ferric-containing phases or possibly heterogeneous ferric mineralogy in the Martian bright regions.

The shape of the 3-gm absorption band indicates a prevalence of bound molecular H20 in Martian surface materials [Houck et al., 1973; Singer et al., 1990; Blaney and McCord, 1989; Erard et al., 1991]. Since CO 2 in the Martian atmosphere obscures the 2.7-2.8 I. tm region, features due to structural OH species cannot be detected in currently available spectra of the surface materials. A weak 2.2-gm feature, characteristic of structural OH in aluminous clays, has been observed in telescopic spectra of Mars [Bell and Crisp, 1993] and in ISM spectra of some areas on Mars [Arnold, 1992; Murchie et al., 1993a,b]. This suggests that aluminous clay silicates commonly found on Earth occur in some of the bright regions of Mars as well. Variation in the strengths of the 2.2- I. tm and 3-gm features in spectra of bright and dark regions on Mars [Murchie et al., 1993a] indicates heterogeneity in the type, crystallinity or quantity of silicates present in the surface material, as well as variations in the water content.

2.2. Atmospheric Properties of Mars

Surface temperatures on Mars span the range 130 to 290 K [Kieffer et al., 1977]. Spring temperatures vary from 170 to 230 K in north temperate latitudes, and from 200 to 250 K in south temperate latitudes [Davies, 1979]. Atmospheric pressures on the surface of Mars were measured by the Viking landers and vary from 680 Pa (6.7 mbar) to 1000 Pa (9.9 mbar) [Hess et al., 1979]. From mass spectrometry on the Viking landers, Owen et al. [1977] report a CO2(g) abundance in the Martian atmosphere near 95%. Climatic models indicate that atmospheric CO2 pressures vary from 0.1 to 15 mbar [Fanale et al., 1982]. Atmospheric water vapor on Mars varies diurnally and regionally across the planet from about 1 to 100 perceptible microns [Farmer et al., 1977]. This water vapor abundance indicates the depth of water precipitation that would result if all of the water in the atmosphere precipitated on the surface [Carr, 1981 ].

2.3. Spectral Experiments Under Low Temperatures

Clark [1981a] presented reflectance spectra of montmorillonites mixed with ice at low-temperatures. A decrease in the intensity of the 1.4-gm and 1.9-gm features and a decrease in the albedo were observed as the temperature was lowered [Clark, 198 la]. This is explained by low-temperature adsorption isotherms for montmorillonite, which showed decreases in the amount of adsorbed water under reduced

temperatures [Anderson et al., 1967]. Reflectance experiments

BISHOP AND PIETERS: REFLECTANCE SPECTRA OF MARS SOIL ANALOGS 5371

involving H20 or CO2 frosts have observed temperature dependent and grain size dependent spectral behavior [Kieffer, 1970; Clark, 198 lb; Lucey and Clark, 1985]. The bandwidths and intensities of the spectral features varied with grain size and temperature, but the band positions remained constant in these studies.

2.4. Spectral Properties of Mars Analogs

Reflectance spectra are shown in Figure 1 for a group of Mars soil analogs from 0.3 to 3.5 gm. These spectra of ferrihydrite, ferrihydrite-bearing montmorillonite, ferric sulfate-bearing montmorillonite and a palagonitic soil from Haleakala, Maui, were measured using a biconical configuration in the Nicolet Fourier transform interferometer (FTIR) under H20 and CO2-purged conditions from 1.2 to 3.5 gm, then scaled to bidirectional spectra measured from 0.3 to 1.2 gm using the reflectance experiment laboratory (RELAB) spectrometer at Brown University. Composite observational spectra of Olympus Mons, Mars [Mustard and Bell, 1994] are shown in Figure 1 for comparison. The telescopic spectrum was scaled to the ISM spectrum at 0.77 gm, and a model was used to remove the strong atmospheric bands. Absorption features due to OH and H20 are observed in these spectra near 1.4, 1.9, 2.2 or 2.3, 2.75, and 2.8-3.1 gm. The visible and near-infrared spectral features of the altered montmorillonites and ferrihydrite are described in detail by Bishop et al. [1993a, submitted manuscript, 1994] and of palagonitic soils by Singer [1982], Bruckenthal [1987], Morris et al. [1990] and Golden et al. [ 1993].

2.4.1. Ferrihydrite. Ferrihydrite is a nanophase, reddish brown ferric oxyhydroxide. The structure contains Fe 3+ in tetrahedral and octahedral sites bound to O, OH, or H20, and the formula Fe4.3(O,OH,H20)•2 has been assigned based on X ray powder diffraction [Eggleton and Fitzpatrick, 1988].

Reflectance spectra of ferrihydrite exhibit subtle ferric absorption features in the extended visible region including a reflectance maximum at 0.80 gm and a reflectance minimum at 0.92-0.93 gm [Bishop et al., 1993a]. Infrared spectra of ferrihydrite include weak features due to structural OH at 1.41 gm, 2.3-2.4 gm, and 2.75 gm, while stronger and broader features due to water are present near 1.95 gm and 3 gm [Bishop et al., 1993a].

2.4.2. Ferric-bearing montmorillonites. Montmorillonite belongs to the smectite mineral group, which is characterized spectrally by features due to adsorbed H20, bound H20, and structural OH. Chemisorbed water in smectites is bound to the interlayer surfaces or cations and is a necessary part of the smectite structure, while physisorbed water is adsorbed on the interlayer or grain surfaces with longer (looser) bonds at lower energies (longer wavelengths). This chemisorbed and physisorbed water will be informally referred to as "bound" and "adsorbed" water. The term

"structural OH" indicates the OH molecules octahedrally coordinated to A1 and other cations in the smectite layers.

Spectral features in smectites due to bound water are observed at 1.41 gm, 1.91 gm, ~2.9 gm, ~3.1 gm, and near 6 gm [Prost, 1975; CarJarl et al., 1983; Bishop et al., 1994], due to adsorbed water at 1.46 gm, 1.97 gm, ~3.0 gm, and near 6 gm [Prost, 1975; CarJarl et al., 1983; Bishop et al., 1994], and due to structural OH at 1.41 gm, 2.2-2.35 gm, 2.75 gm, and 10-12 gm [Farmer, 1974; Bishop et al., 1994]. Montmorillonites doped with ferric salts (e.g., ferrihydrite- bearing and ferric sulfate-bearing montmorillonite) have shown ferric absorptions near 0.5 gm and near 0.9 gm, that are dependent on the chemical conditions of the doping procedure [Banin et al., 1993; Bishop et al., 1993; J. L. Bishop et al., submitted manuscript, 1994]. The spectrum of ferrihydrite- bearing montmorillonite (Figure 1) exhibits a ferric

1.00

.

0.80

0.60

0.40

0.20 z

0.00 0 500 1000 1500 2000 2500 3000 3500

Wavelength (nm)

Figure 1. Reflectance spectra are shown from 0.3 to 3.5 gm for Mars analogs and Mars. The spectrum of Olympus Mons, Mars, is a composite spectrum created from ISM and telescopic data [Mustard and Bell, 1994]. The reflectance spectra of the ferrihydrite, ferrihydrite-bearing montmorillonite, ferric sulfate- bearing montmorillonite and a palagonitic soil from Haleakala, Maui, were all measured under H20- and CO2- purged conditions at 23øC.


absorption band centered near 0.92 gm, a reflectance maximum near 0.75 gm and a subtle shoulder near 0.6 •m. The ferric sulfate-bearing montmorillonite spectrum (Figure 1) exhibits a ferric absorption band at 0.88 •m, a reflectance maximum at 0.70 •m, and a more pronounced shoulder near 0.6 •m.

2.4.3. Palagonitic soil. Palagonite is defined as weathered basaltic volcanic glass [Bougka, 1993] and is formed via heating, hydration, and devitrification of this glass [Bates and Jackson, 1984]. During palagonitization the ferric to ferrous iron ratio increases and the formation of smectites

and zeolites is common [Bou•ka, 1993]. Based on the thermodynamics of weathering processes, Gooding and Keil [1978] suggested that basaltic glass may have weathered to form clay-rich products on Mars. Palagonitic soils have been proposed as analogs to bright soils on Mars based on basaltic elemental abundances and spectral properties [e.g., Allen et al., 1981; Singer, 1982]. Mineralogical studies of variably palagonitized basaltic tephra from Mauna Kea, Hawaii, have found smectites and iron oxides in the most altered fine-

grained particles of several samples [Golden et al., 1993]. Reflectance spectra of palagonitic soil exhibit subtle ferric

absorption features in the extended visible region that are highly dependent on the nature of the sample. Features such as a broad shallow absorption band near 0.86 gm, a reflectance maximum near 0.75 •m, and an inflection near 0.6 •m are observed in the fine particle size fractions of many samples [Singer, 1982; Morris et al., 1990; Golden et al., 1993]. Infrared spectra of palagonitic soils vary depending on the composition and crystallinity of the sample, but generally contain broad features near 1.9 •m and 3 •m, as well as one or more weak features in the range 2.2-2.35 gm [Singer, 1982; Morris et al., 1990; Bishop et al., 1993b]. Infrared absorption spectra of <0.2 •m size fractions of palagonitic soils exhibit features near 3620 cm -• (2.76 •m) and 880 cm -• (11.4 •m), indicative of smectite clays [Golden et al., 1993]. The spectrum of the palagonitic soil from Haleakala (Figure l) exhibits a broad, weak band extending from 0.8 to 0.95 •m, a reflectance maximum near 0.78 gm, and a rounded shoulder near 0.6-0.65 •m.

2.5. Christiansen Feature

The Christiansen frequency is defined as the frequency at which the real part of the refractive index of individual particles equals that of the surrounding medium [Henry, 1948; Conel, 1969]. For powdered silicates the Christiansen frequency occurs at the frequency of maximum emissivity and maximum transmission [Salisbury, 1993]. This feature is characterized by a sharp decrease in albedo in reflectance spectra and occurs where the real index of refraction, n, is near unity and the extinction coefficient, k, is decreasing strongly.

Spectral experiments [Logan et al., 1973] have shown shifts in the Christiansen feature of quartz and other minerals as a function of particle size, sample texture, background temperature and atmospheric pressure. In this Logan et al. study, emission spectra exhibited shifts in the Christiansen feature toward longer wavelengths with decreasing particle size and toward shorter wavelengths with decreasing atmospheric temperature and pressure. Variations in particle size (0-5 •m versus 0-74 •m) produced the largest effect on the Christiansen feature in this suite of emission spectra [Logan et al., 1973]. In a more recent study involving reflectance spectra of fine- grained olivine, the Christiansen feature was observed to

broaden with decreasing particle size and perhaps shift toward shorter wavelengths as well [Hayes and Mustard, 1994].

3. Experimental Procedures

3.1. Sample Preparation

3.1.1. Ferrihydrite. The ferrihydrite was synthesized by gradually adding potassium hydroxide to a ferric nitrate solution until a p H of 7-8 was reached, according to the procedure for 2-line ferrihydrite described by Schwertmann and Cornell [1991]. The resulting moist sample was rinsed with H20 over a porous frit, lyophilized, and dry sieved to <45 gm. Additional information about this ferric oxyhydroxide (Fe5HO8o4H20) sample can be found in the work by Bishop et al. [1993a].

3.1.2. Ferric-bearing montmorillonites. SWy-1 montmorillonite (The Clay Minerals Society, Source Clays Repository) was chemically altered in the laboratory to produce a large collection of montmorillonites containing a variety of interlayer ferric oxyhydroxide and ferric sulfate complexes [Bishop, 1994; J. L. Bishop et al., submitted manuscript, 1994]. A suspension was formed from the SWy montmorillonite and H20, and repeated washing with dilute HC1 produced an H-exchanged form of the montmorillonite. The ferrihydrite-bearing and ferric sulfate-bearing montmorillonites included in this study were prepared by adding 0.5 N ferric chloride or ferric sulfate solutions at pH-2 to the H-montmorillonite. Dilute NaOH was added dropwise until pH 5 and pH 3.5, respectively. The suspensions were then centrifuged and lyophilized to form a fine powder, which was dry sieved to <45 gm. Additional information about the ferrihydrite-bearing montmorillonite (# 35) and ferric sulfate- bearing montmorillonite (# 117) used in these experiments can be found in the work by Bishop [ 1994] and J. L. Bishop et al. (submitted manuscript, 1994).

3.1.3. Palagonitic soil. This sample (#72) was collected from the Haleakala crater, Maui, as clumps several millimeters in diameter coated by a fine-grained, reddish brown soil. This sample was selected because of the smectitelike absorption features observed near 2.2 gm in reflectance spectra. Only that part of the sample able to pass through a <45 gm dry sieve was used in this study.

3.2. Reflectance Spectra

3.2.1. Nicolet. Reflectance spectra were measured relative to a rough gold surface (Infragold) using a Nicolet 740 FTIR in a H20- and CO2-purged environment. A PbSe detector was used from 0.9 gm to 3.2 gm and a deuterated triglycine sulfate (DTGS) detector from 1.8 gm to 25 gm. The sample chamber, including a remotely controlled sample train with multiple sample positions, was purchased from SpectraTech. Bulk powdered samples are measured horizontally in this system. Spectra are averaged from two locations on each of two replicates for each kind of sample.

3.2.2. RELAB. Additional spectra were measured under ambient conditions relative to Halon in the visible and near-

infrared using a photomultiplier tube with the bidirectional RELAB spectrometer at Brown University (see J. L. Bishop et al., submitted manuscript, 1994 for analyses of these spectra). A detailed description of this instrument is provided by Mustard and Pieters [1989]. The bidirectional spectrometer allows for variable illumination (i) and emergence (e) angles.


Spectra included here were measured at i - 30 ø and e - 0 ø. The sample dish is rotated during the measurement to eliminate orientation effects.

3.2.3. Environmental chamber. Reflectance spectra were also measured under controlled atmospheric pressure and temperature conditions using an environmental chamber attachment (for the Nicolet 740 FTIR) built by Connecticut Instruments. Reflectance spectra were measured using a liquid nitrogen-cooled mercury-cadmium-telluride (MCT) detector in the range 1.5-20 gm for each sample under ambient pressure and then again at several increments under reduced atmospheric pressures. Care was taken to deflect the IR beam during pressure adjustment to avoid additional heating of the sample. Dry conditions (i.e., the unbound water in natural smectites was lost) were achieved under atmospheric pressures in the range 0.1 to 10 mbar. The humidity in the lab was controlled to approximately 40% relative humidity giving a PH2o of-10 torrs (-13 mbar) at 1-bar atmospheric pressure. The partial pressure of H20 is then lower under reduced atmospheric pressures and temperatures.

The measurement procedure entailed first measuring the gold standard, then the sample at ambient atmospheric pressure and temperature, in each case after purging the air in the chamber for at least 30 min. A mechanical pump was used to evacuate the air from the environmental chamber, and the atmospheric pressure was measured using three pressure gauges, each having different active ranges and units. For low atmospheric pressure experiments spectra were measured of the sample at several pressure settings: P (atmospheric pressure) = Parnb (ambient atmospheric pressure) -10 inches (in) Hg, P = Pamb -20 in Hg, P = Pamb -30 in Hg; P = 10 mm Hg, P = 5 mm Hg, P = 1 mmHg, P=0.1 mmHg;P= 100gmHg, 50gmHg, 10gm Hg (-10 -6 bar).

The spectra shown in Figure 2 were measured under reduced atmospheric pressures and temperatures according to the following procedure. The temperature of the sample chamber was cooled by adding N2(/) to a dewar attached to the environmental chamber. Atmospheric pressures below about 1 mbar are required for the low-temperature experiments to prevent H20 condensation on the internal surfaces. Spectra were measured in these experiments at a temperature, T = 23øC and P = 10 mm Hg, 1 mm Hg, 0.1 mm Hg; then at T = 20øC, 15øC, and continuing at 5øC intervals as low as possible in an air environment (-10 to -20øC). At this point the atmospheric pressure was pumped down to ~10 -6 bar, then CO:(g) was added up to pressures near 1 mbar. The CO: molecules assist in heat transfer but do not cause the condensation problems associated with H:O molecules. Therefore cooler temperatures (near -40øC) were achieved in a CO: environment. The temperature range in this system is limited by a thermally conductive disk between the environmental chamber and the N2(/) dewar, which can be exchanged to achieve the desired temperature range.

Water vapor fine structures are seen near 2.7 pm and 6 pm in the low-pressure and temperature spectra due to differences in the atmospheric water vapor concentration between ambient and reduced atmospheric pressures. Spectra of this atmospheric water vapor were derived by measuring spectra of the gold standard under variable atmospheric pressure and temperature conditions, and dividing these spectra by the initial gold spectrum. In order to remove the water vapor fine structure the spectra acquired under low atmospheric pressure and temperature were divided by an atmospheric water vapor spectrum (ratio of gold spectra) and smoothed using a variable

width boxcar smoothing program. Absolute reflectance values were obtained for the calibrated environmental chamber

spectra by scaling these to RELAB values.

4. Results

Reflectance spectra are shown of ferrihydrite, ferrihydrite- bearing montmorillonite, ferric sulfate-bearing montmorillonite and a palagonitic soil under variable atmospheric and environmental conditions in Figure 2. Spectra of each sample were measured under ambient atmospheric air pressure at 23øC, under -1 mbar air pressure at 23øC, and under 1 mbar CO2 pressure at near -40øC. The spectra shown in Figure 2, panels 2a, 3a, 4a, and 5a, extend from 5500 to 2500 cm -• (1.8 to 4.0 gm); the spectra in Figure 2, panels 2b, 3b, 4b, and 5b, extend from 1800 to 1250 cm -• (5.5 to 8.0 gm); and the spectra in Figure 2, panels 2c, 3c, 4c, and 5c, extend from 1400 to 800 cm -• (7 to 12 gm).

In Figure 2, panel 2a, the broad feature at 1.95 gm and the shoulder at 2.5 gm observed in the ambient ferrihydrite spectrum are reduced in intensity in the spectra measured under low-pressures and temperatures. The spectra measured under low-pressure and temperature conditions also show a weak 2.75-gm feature that is not resolvable in ambient spectra. In Figure 2, panel 3a, the spectra of ferrihydrite-bearing montmorillonite measured under low-pressure and temperature conditions show little change in the features at 2.2 and 2.75 gm due to structural OH, but suppression of the features at 1.9, 2.5 and ~3 gm due to adsorbed and bound H20 is observed. Similarly, the spectra of ferric sulfate-bearing montmorillonite in Figure 2, panel 4a, show little change at 2.2 and 2.75 gm; however, significant loss in intensity at 1.9, 2.5 and 3 gm is observed. In spectra of the palagonitic soil shown in Figure 2, panel 5a, the 1.9-gm feature is essentially removed, the 2.2-gm band is enhanced, and the broad 2.5- and 3-gm features are weakened in spectra measured under low- pressure and temperature conditions.

The water-bending vibration near 6.1 gm is seen in spectra of each of these samples in Figure 2, panels 2b, 3b, 4b, and 5b. In the ferrihydrite spectra (Figure 2, panel 2b) this feature nearly disappears under reduced atmospheric pressures and temperatures. In Figure 2, panels 3b, 4b, and 5b, this water feature in the spectra of the ferrihydrite-bearing montmorillonite, the ferric sulfate-bearing montmorillonite, and the palagonitic soil is substantially reduced upon decreasing the atmospheric pressure, and reduced again slightly upon lowering the temperature.

For spectra of the montmorillonites and the palagonitic soil shown in Figure 2, panels 3c, 4c, and 5c, a shift in the Christiansen feature was observed towards shorter wavelengths as the atmospheric pressure and temperature were lowered. In the ferrihydrite spectra (Figure 2, panel 2c) there is an absorption feature near 1350 cm -• (7.5 gm) that disappears under dry conditions, thus complicating detection of shifts in the Christiansen feature for ferrihydrite. The CO2(g ) was required to achieve the lower temperatures in these experiments, but did not appear to influence the position of the Christiansen feature (e.g., no change in the Christiansen feature was observed as a function of the CO 2 pressure).

Band depths have been calculated using the method of Clark and Roush [1984] for the hydration features in the spectra shown in Figure 2. For the ferrihydrite spectra band minima of 1.95 gm, 2.32 gm, and 2.95 gm were used. For the spectra of


0.7o I 0.so

Wavelength (urn) Wavelength (urn) &00 4,00 &00 7.00

ferrihydrite I

Wavelength (urn) 8.oo 0.o0 • 0.oo • tom P..oo

o.$o

• 0.4o

• 0.•o ambient air. n- 1 mbar air, 23• ,

0•.0 1 mbar C02, -44 •---

0.10

0.00,

• • ferrihydrite-bearing

0.80! __ •,•,,.•.._•_.• .• ••,v• • montmorillonlte 0.70 - -

i ambient air :. l mbar air, 23 •- I •"•'

OJO 3a 0.00

0.•0 •erd½ •ff•-beadn• montmorillonite

0.70

ß

ambient air, / •..

:. 0.3 mbar air, 23 • /// : :. 1 mbar C02, -42•--- / .

:4.

palagonific aoil, Haleakala, Maul

1 mbar air, 23 • ---- '

0.10 1 mbar C02, - 38 1

• 0.60 c: 0.50

m 0.40'

i•. oJo,

0.60

0.50

o.1 o o.oo•

0.30

ß • oJ.o

4•o &.o ,so •7o t;o •o t4o •o •o •o u o too 0..o 0.8o Wavenumber (cm-1) X 10E-3 Wavenumber (cm-1) X 10E-3 Wavenumber (cm-1) X 10E-3

Figure 2. Reflectance spectra of ferrihydrite (panel 2), ferrihydrite-bearing montmorillonite (panel 3), ferric sulfate-bearing montmorillonite (panel 4), and a palagonitic soil (panel 5) measured under variable atmospheric conditions are shown from: (a) 5500 cm -1 to 2500 cm '• (1.8 I. tm to 4.0 I. tm), (b) 1800 cm -• to 1250 cm -• (5.5 I. tm to 8.0 I. tm), and (c) 1400 cm -• to 800 cm -• (7 I. tm to 12 I. tm). The spectra shown here were measured under ambient pressure and temperature (solid curve), under reduced atmospheric pressure (patterned curve), and under reduced atmospheric pressures and temperatures (dashed curve).

ferrihydrite-bearing montmorillonite and ferric sulfate-bearing montmorillonite, band minima of 1.91 I. tm, 2.21 I. tm, 2.95 I. tm and -1630 cm -• (6.14 I. tm) were used. For the palagonitic soil spectra band minima of 1.93 I. tm, 2.21 I. tm, 2.95 I. tm, and 1645 cm -• (6.08 I. tm) were used. The band depths for these spectral features are summarized in Table 1.

5. Discussion

5.1. Infrared Features due to Bound and Adsorbed

HzO

Reducing the atmospheric pressure and temperature in these experiments decreases the water vapor in the sample


Table 1. Band Depths for Hydration Features

Feature, •tm

1.91 1.93 1.95 2.21 2.32 2.95 6.08 6.14 6.15

Ferrihydrite Ambient 14 8 9 4

23 ~C 6 9 90 -44 ~C 3 11 80

Ferrihydrite-bearing montmorillonite Ambient 12 12 84

23 ~C 8 14 72 -47 ~C 6 13 73

55

48

51

Ferric sulfate-bearing montmorillonite Ambient 17 1 1 89

23 ~C 8 12 78 -42 ~C 6 112 7 4

63

40

32

Palagonitic Soil Ambient 8 1 90

23 ~C 3 6 83 -38 ~C 1 6 79

64

42

34

All values are in percent.

environment. This induces the H20 molecules not chemically bound to cations or grain surfaces to desorb from the sample. Lowering the atmospheric pressure was found to dehydrate montmorillonites exchanged with several different interlayer cations [Bishop et al., 1994] and montmorillonites containing ferric oxyhydroxides and hydrated ferric sulfate complexes (J. L. Bishop et al., submitted manuscript, 1994). Changes in the spectral hydration features in these experiments depended on the nature of the bonds binding H20 molecules to the mineral surfaces or interlayer cations. For example, Mg-exchanged montmorillonites exhibit greater decreases in the intensities of the !.9-•tm and 3-•tm features than Fe-exchanged montmorillonites, because the interlayer Mg cations are bound to more H:O molecules [Bishop et al., 1994]. Additional spectral experiments using a Wyoming bentonite showed decreases in the intensities of the features

near 1.4 •tm and 1.9 •tm as the temperature was lowered [Clark, 1981a].

Variation in the behavior of the hydration features in spectra of the ferrihydrite, ferrihydrite-bearing and ferric sulfate-bearing montmorillonites and palagonitic soil measured under low atmospheric pressure and temperature conditions (Figure 2) indicates differences in how water is present in these samples. Detailed descriptions of the spectral character of these Mars analogs as a function of atmospheric conditions follow. First, however, there is one trend shared by spectra of all of these samples: reduction in the 2.5 I.tm shoulder under reduced pressures and temperatures. This feature at 2.5-•tm in the ambient spectra of ferrihydrite, montmorillonite, and palagonite is thus apparently due to H20 adsorbed in a similar manner in each of these samples.

The spectral character of the hydration features in chemically altered montmorillonites is dependent on the ferric

interlayer components present. The spectra of the ferrihydrite- bearing montmorillonite in Figure 2, panels 3a and 3b, show a decrease in the intensity of the 3-•tm and 6-•tm features when the atmospheric pressure was reduced to 1 mbar, but no additional intensity loss was observed upon reducing the pressure further or lowering the temperature. In contrast the spectra of the ferric sulfate-bearing montmorillonite in Figure 2, panels 4a and 4b, show continual loss of intensity of the 3 •tm and 6-•tm features as the pressure and temperature are lowered. This indicates that the adsorbed H:O molecules are removed from the ferrihydrite-bearing montmorillonite under relatively mild dehydrating conditions and that additional H20 in this sample is tightly bound (chemisorbed). For the ferric sulfate-bearing montmorillonites there must be additional H:O molecules in the system that are more tightly bound than the adsorbed H:O and less tightly bound than the chemisorbed H:O. The spectra of both the ferrihydrite-bearing montmorillonite and the ferric sulfate-bearing montmorillonite in Figure 2, panels 3a and 4a exhibit subtle features near 3400 cm '• (2.95 •tm) and 3200 cm '• (3.12 •tm) when measured under dehydrated conditions. However, spectra of these samples measured in a moist environment show a broad 3-•tm band lacking distinguishable features (J. L. Bishop et al., submitted manuscript, 1994).

The H:O stretching and bending combination band near 1.9 •tm lost intensity primarily as a result of dehydration under reduced atmospheric pressure, with only secondary loss in intensity from dehydration under reduced temperature, in spectra of the ferrihydrite-bearing and ferric sulfate-bearing montmorillonites. The band depths decreased from 12% to 6% for the ferrihydrite-bearing montmorillonite spectra and from 17% to 6% for the ferric sulfate-bearing montmorillonite spectra. Although the band minimum is near 1.91 •tm in


spectra of both samples, this feature is much broader toward longer wavelengths in spectra of the ferric sulfate-bearing montmorillonite, especially noticeable under dehydrated conditions. This long-wavelength shoulder must be linked to the long-wavelength broadening of the 3-gm band due to the interlayer sulfate complexes in this sample.

These experiments indicate that bound H20 in the ferrihydrite- bearing and ferric sulfate- bearing montmorillonites exhibit spectral features near 1.91, 2.9, 3.1, 5.9, and 6.1 gm. Adsorbed H20 and H20 coordinated to the sulfate groups exhibit broad spectral features near 1.97, 3, and 6 gm that hinder identification of the features due to bound H20. Water coordinated to the sulfate groups in the ferric sulfate-bearing montmorillonites is more firmly bound than the adsorbed H20, and therefore, broad features near 1.97 and 3 gm can be observed in spectra measured under lower pressures and temperatures. Analyses of these ferric sulfate-bearing montmorillonites, including M6ssbauer spectroscopy, differential thermal analysis and X ray diffraction, suggest that small particles of the mineral Schwertmannite and hydrated ferric sulfate complexes surround the montmorillonite grains (J. L. Bishop et al., submitted manuscript, 1994).

The broad combination band centered near 1.95 gm in spectra of ferrihydrite (Figure 2, panel 2a) loses intensity (14% to 3% band depth) and shifts towards 1.9 gm as the pressure and temperature are lowered. The broad 3-gm band due to H20 stretching vibrations in these spectra decreases in intensity as well (94% to 80% band depth), and a band minimum near 2.95 gm ca n be detected under dehydrated conditions. The 6.0-gm H20 bending feature (Figure 2, panel 2b) disappears as the pressure and temperature are reduced, and another feature near 1340 cm '• (7.5 gm) becomes weaker under lower pressures and finally appears as two features at 1400 cm '• (7.1 gm) and 1300 cm -• (7.7 gm) in the low-temperature spectra. It is unclear whether these features at 7.1 and 7.6 gm were present, but masked, in the ambient spectra, or whether a change in the ferrihydrite structure is responsible for these features. Another feature near 1490 cm -• (6.7 gm) broadens and shifts slightly toward shorter wavelengths as the pressure and temperature are decreased. Reflectance spectra of hematite [Salisbury et al., 1991] exhibit absorption features near 6 and 6.5 gm that may be related to those observed for ferrihydrite. Absorption features in reflectance spectra of goethite are observed near 5.5 and 6 gm [Salisbury et al., 1991].

The 1.9-gm H20 combination band weakens such that it is barely discernible in spectra of the palagonitic soil measured under reduced pressures and temperatures. The broad 3-gm band exhibits continued H20 loss as the pressure and temperature are decreased (90% to 79% band depth), and a band center is observed near 2.95 gm under low-pressures and temperatures. The H20 bending feature near 6 gm decreased in intensity, as well, with decreasing pressure and temperature.

In summary, the ferrihydrite and palagonitic soil contain a large amount of adsorbed H20 and bound H20 that is not constrained to claylike interlayer sites. Both forms of water are manifested in the spectra of these materials as a broad, strong (90-94% band depth) feature near 3 gm under ambient atmospheric conditions. Upon dehydration under Marslike conditions ferrihydrite and palagonitic soil exhibit a very weak or absent feature near 1.9 gm and a weakened, less broad 3-gm band (although intensity is lost this feature remains very strong). In contrast, spectra of the ferrihydrite-bearing and ferric sulfate-bearing montmorillonites exhibit primarily a

sharpening of the 1.9-gm and 3-gm bands when measured under Marslike conditions. These materials contain H20 bound in specific interlayer sites resulting in features near 1.9 gm, 2.95 gm, and 3.1 gm under dehydrated conditions. Additional H20 molecules coordinated to sulfate complexes, or adsorbed in the interlayer region or along grain surfaces give broadening near 1.97 gm and 3 gm.

5.2. Infrared Features due to Structural OH

The fundamental OH stretching feature at 2.75 gm and the combination band at 2.2 gm in spectra of the montmorillonites in Figure 2, panels 3a and 4a, remain unchanged as a result of the decreases in atmospheric pressure and temperature. The band depth of the 2.21-gm structural OH combination band is -12% in spectra of the ferrihydrite- bearing montmorillonite and ferric sulfate-bearing montmorillonite. The fundamental OH bending features at 920 cm -• (10.9 gm), 880 cm -• (11.4 gm), and 850 cm -• (11.8 gm) in spectra of the montmorillonites in Figure 2, panels 3c and 4c, were also unaffected by the environmental changes. Octahedrally coordinated OH groups in smectites are responsible for these features [Farmer, 1974]. In montmorillonites, primarily A1 cations occupy these octahedral sites, giving rise to the features observed at 2.2, 2.75, and 10.9 gm. Small amounts of Fe and Mg substituting for A1 in these sites are responsible for the features at 11.4 and 11.8 gm [Farmer, 1974] and the long-wavelength shoulder of the 2.2-gm band [Bishop et al., 1993a].

Spectra of ferrihydrite measured under low-pressures and temperatures (Figure 2, panel 2a) exhibit a fundamental OH stretching feature at 2.75 gm that is masked by the strong 3- gm water band in spectra measured under ambient conditions. The strength of the combination band at 2.3 gm is slightly enhanced in spectra of ferrihydrite under dehydrated conditions. The OH groups are octahedrally coordinated to Fe instead of A1 and vibrate at a lower energy (longer wavelength) because the O is more tightly bonded to the Fe cation. This feature observed at 2.3 gm is similar to that observed in nontronite, a smectite that has primarily Fe in the octahedral sites [Farmer, 1974]. However, nontronite has strong crystal field bands in visible spectra [Singer, 1982; Burns, 1993], which are not observed in the spectra of ferrihydrite. Additional mid-IR OH bending features are seen in nontronite near 820 cm -• (12.2 gm) [Farmer, 1974; Stubican and Roy, 1961], but not seen in the spectra of ferrihydrite in Figure 2, panel 2c, or in Bishop et al. [1993a]. Ferrihydrite has only nanophase crystalline order [Eggleton and Fitzpatrick, 1988], which may explain why the features observed in spectra of nontronite near 0.6, 0.9, 2.3, 2.75, and 12 gm [Singer, 1982; Burns, 1993; Farmer, 1974; Stubican and Roy, 1961], due to octahedrally coordinated Fe and OH species, appear only weakly if at all in the spectra of ferrihydrite.

The spectra of the palagonitic soil in Figure 2, panel 5a, exhibit an OH combination band at 2.2 gm despite the fact that the fundamental OH stretching and bending features are not distinguishable in Figure 2, panel 5c. The 2.2-gm band strength increases from 1% to 6% in spectra of the palagonitic soil as the atmospheric pressure and temperature were reduced. Asymmetry in the broad 3-gm band is probably due to a fundamental OH stretching feature at 2.75 gm that is unresolvable. The presence of a feature near 2.2 gm and possible feature near 2.75 gm indicate a phyllosilicate


component in this soil having A1 in octahedral sites coordinated to OH groups. Since smectites have been found in palagonitic soils from Hawaii [Golden et al., 1993], it is likely that the palagonitic soil studied here has partially weathered to montmorillonite.

5.3. Christiansen Feature

Under low atmospheric pressure and low-temperature conditions the Christiansen feature is observed to broaden and

shift towards shorter wavelengths in these experiments. The Christiansen feature occurs at 1245 cm -• (8.0 •tm)in both of the montmorillonite samples under 1-bar air and 23øC (Figure 2, panels 3c and 4c). This feature shifts gradually to 1270 cm -• (7.9 !.tm) under lower pressures and temperatures. A similar shift in the Christiansen feature is observed in the palagonitic soil (Figure 2, panel 5c) from 1150 cm -• (8.7 !.tm) to 1200 cm -• (8.3 !.tm). The shifts in the Christiansen feature observed here parallel previous observations by Logan et al. [1973] of shifts in the Christiansen feature in emission spectra of silicates toward shorter wavelengths with decreasing atmospheric pressure and temperature. The shifts towards shorter wavelength in the Christiansen feature observed here, however, were not accompanied by increased spectral contrast, as was observed by Logan et al. [1973].

Possible explanations for the shifts in wavelength of the Christiansen feature in the low-temperature reflectance spectra shown here include apparent changes in the optical properties of the bulk materials as a function of the temperature or a temperature gradient, as a function of dehydration, or as a function of effective sample particle size or texture. There is no precedent for dehydration of a material effecting optical properties. Decreasing the particle size in fine-grained olivine, however, has been observed to broaden the Christiansen feature in reflectance spectra with an apparent shift towards shorter wavelengths [Hayes and Mustard, 1994]. In the case of fine-grained soils, the sample texture is a complex mixture of tiny grains clustered together as aggregates. If water molecules adsorbed on the grain surfaces play a role in the ability of these small particles to cluster together, then dehydrating the soils may influence the. sample texture and result in an effectively smaller particle size. These and other possible environmental effects on fine-grained soils need to be further evaluated.

6. Conclusions

The Mars soil analogs examined in these experiments are compositionally consistent with the chemical analyses of Viking and exhibit spectral features in the visible part of the spectrum that are similar to features observed in the bright regions on Mars. Reflectance spectra of these analog materials were measured under reduced atmospheric pressure and temperature environments comparable to the temperate zones on Mars. These experiments showed that the structural OH features near 2.2-2.3 •m and 2.75 •m in spectra of the palagonitic soil and ferrihydrite became enhanced when measured under Marslike atmospheric conditions. The spectral features due to structural OH in ferrihydrite-bearing and ferric sulfate-bearing montmorillonite remained unchanged as a result of decreasing the atmospheric pressure and temperature.

The spectral features due to adsorbed H20 in these samples were lost or weakened under low-pressure and temperature

conditions. Therefore broad features centered at 2.9 !.tm in the ferrihydrite and palagonitic soil are due to bound H20, and features at 1.9 !.tm, 2.9 !.tm, and 3.1 !.tm in the low-pressure and temperature spectra of the ferrihydrite-bearing and ferric sulfate-bearing montmorillonites are due to bound H:O. The ferrihydrite-bearing montmorillonite contains adsorbed H20 that is lost under relatively mild dehydrating conditions and bound H:O that is a necessary part of the structure and is resistant to dehydration under low atmospheric pressures and temperatures. In addition to adsorbed and bound H20, the ferric sulfate-bearing montmorillonites contain H:O molecules coordinated to sulfate groups, which are loosely bound to the montmorillonite system.

Spectra of the palagonitic soil when examined under atmospheric pressures and temperatures of temperate regions on Mars, exhibit enhanced near-infrared features, suggesting the presence of montmorillonite. This is further evidence that palagonitization is linked to smectite formation and that palagonitization of basaltic volcanic glass on Mars may be a mechanism for smectite formation in the surface material

there.

7. Applications to Mars

Based on chemical composition, visible reflectance spectra, and infrared reflectance spectra acquired under low atmospheric pressures and temperatures, these samples are promising analog materials for the bright regions on Mars. The ferrihydrite-bearing montmorillonite or a mixture containing ferrihydrite and a silicate phase may be present in bright regions such as Arabia where the ferric absorption occurs at 0.91-0.92 !.tm. The ferric sulfate-bearing montmorillonite may be present in bright regions such as Lunae Planum where the ferric absorption occurs near 0.88 !.tm. The nature of the ferric and silicate components vary greatly in palagonitic soils; however, many exhibit hematiticlike absorptions and may be present in bright regions having a ferric absorption near 0.85 •m.

The experiments presented here show the sensitivity of the spectral hydration features in soils to the atmospheric environment of the sample, and the importance of measuring analog soils under the appropriate environmental conditions. Under Marslike atmospheric conditions, the palagonitic soil exhibited a structural OH feature with a 6% band strength near 2.2 !.tm. A feature near 2.2 !.tm has also been observed by multiple researchers in some of the Martian bright regions [Arnold, 1992; Murchie et al., 1992, 1993a; Bell and Crisp, 1993]. Spectra of ferrihydrite-bearing montmorillonite and ferric sulfate-bearing montmorillonite samples are characterized by a stronger 2.2-!.tm feature (11-12% band strength). Spectra of ferrihydrite include a broad feature near 2.3 !.tm, which may be related to a feature in this region observed in some bright regions on Mars. Under Marslike atmospheric conditions (1) the 1.9-!.tm feature is substantially weaker in spectra of each of the samples, especially for the palagonitic soil and ferrihydrite, where the 1.9-!.tm feature is barely discernible, (2) the feature near 2.2-2.3 !.tm is stronger than the 1.9-!.tm feature in spectra of each of these analog materials, and (3) a strong 3-!.tm band is present in the spectra of ferrihydrite, palagonitic soil, and ferric sulfate-bearing montmorillonite. The results presented here are an important step in constraining interpretations of the hydration features in Mars bright region spectra.


Acknowledgments. The authors would like to express gratitude to S. Pratt for technical assistance and to J. Mustard and S. Murchie for

discussions of the results and helpful editorial suggestions. Support through the ZONTA Foundation, the NASA Graduate Student Researchers Program at NASA-ARC, Moffett Field, and the Alexander von Humboldt Foundation, Germany, is greatly appreciated by J. Bishop. NASA support for this research under grant NAGW-28 is gratefully acknowledged by C. Pieters. RELAB is a multi-user facility supported by NASA under grant NAGW-748. The Nicolet system was acquired under a grant from the Keck Foundation.

References

Adams, J., and T. B. McCord, Mars: Interpretation of spectral reflectivity of light and dark regions, J. Geophys. Res., 74, 4851- 4856, 1969.

Allen, C. C., J. L. Gooding, M. Jercinovic, and K. Keil, Altered basaltic glass: A terrestrial analog to the soil of mars, Icarus, 45, 347-369, 1981.

Anderson, D. W., E. S. Gaffney and P. F. Low, Frost phenomena on Mars, Science, 155, 319-322, 1967.

Arnold, G., Infrarotsondierung planetarer Oberfl•ichen, Dtsch. Luff Raumfahrt-Nachrichten, 69, 25, 1992.

Baird, A. K., A. J. Castro, B.C. Clark, P. Toulmin III, H. Rose Jr., K. Keil, and J. Gooding, The Viking X ray fluorescence experiment: Sampling strategies and laboratory simulations, J. Geophys. Res., 82, 4595-4624, 1977.

Banin, A., T. Ben-Shlomo, L. Margulies, D. F. Blake, R. L. Mancinelli, and A. U. Gehring, The nanophase iron mineral(s) in Mars soil, J. Geophys. Res., 98, 20,831-20,853, 1993.

Bates, R. L., and J. A. Jackson (Ed.), Dictionary of Geological Terms, Anchor, New York, 1984.

Bell, J. F., III, and D. Crisp, Groundbased imaging spectroscopy of Mars in the near-infrared: Preliminary results, Icarus, 104, 2-19, 1993.

Bell, J. F., III,, T. B. McCord, and P. D. Owensby, Observational evidence of crystalline iron oxides on Mars, J. Geophys. Res., 95, 14,447-14,461, 1990.

Bibring, J.-P., et al., ISM observations of Mars and Phobos: First results, Proc. Lunar Planet, Sci. Conf., 20th, 461-471, 1990.

Bishop, J. L., Spectroscopic analyses of chemically altered rnontmorillonites and applications to the soils on Mars, doctoral thesis, Brown Univ., Providence, R.I., 1994.

Bishop, J. L., and C. M. Pieters, Reflectance spectra of Mars soil analogs measured under reduced atmospheric pressures and temperatures, (abstract),Lunar Planet. Sci. Conf. XXV, 117-118, 1994.

Bishop, J. L., C. M. Pieters, and R. G. Burns, Reflectance and M/3ssbauer spectroscopy of ferrihydrite-montmorillonite assemblages as Mars soil analog materials, Geochim. Cosmochim. Acta, 57, 4583-4595, 1993a.

Bishop, J. L., S. L. Murchie, S. Pratt, J. F. Mustard, and C. M. Pieters, The importance of environmental conditions in reflectance spectroscopy of laboratory analogs for Mars surface materials, LPI Tech. Rep., 93-06, 4-5, 1993b.

Bishop, J. L., C. M. Pieters, and J. O. Edwards, Infrared spectroscopic analyses on the nature of water in montmorillonite, Clays Clay Miner., 42, 701-715, 1994.

Blaney, D. L., and T. B. McCord, An observational search for carbonates on Mars, J. Geophys. Res., 94, 10,159-10,166, 1989.

Bou•ka, V., Natural Glasses, 354 pp., Ellis Horwood, Chichester, England, 1993.

Bruckenthal, E. A., The dehydration of phyllosilicates and palagonites: Reflectance spectroscopy and differential scanning calorimetry, M.S. thesis, Univ. of Hawaii, Honolulu, 1987.

Burns, R. G., Mineralogical Applications of Crystal Field Theory, 2nd ed., Cambridge University Press, New York, 1993.

Calvin, W. M., Additions and corrections to the absorption coefficients of CO 2 ice: Applications to the Martian south polar cap, J. Geophys. Res., 95, 14,743-14,750, 1990.

Cariati, F., L. Erre, G. Micera, P. Piu, and C. Gessa, Polarization of water molecules in phyllosilicates in relation to exchange cations as studied by near-infrared spectroscopy, Clays Clay Miner., 31, 155- 157, 1983.

Cart, M. H., The Surface of Mars, 232 pp., Yale University Press, New Haven, Conn., 1981.

Clark, R. N., The spectral reflectance of water-mineral mixtures at Low-temperatures, J. Geophys. Res., 86, 3074-3087, 198 la.

Clark, R. N., Water frost and ice: The near-infrared spectral reflectance 0.65-2.5 •tm, J. Geophys. Res., 86, 3087-3096, 1981b.

Clark, R. N., and T. B. McCord, Mars residual north polar cap: Earth- based spectroscopic confirmation of water-ice as a major constituent and evidence for hydrated minerals, J. Geophys. Res., 87, 367-370, 1982.

Clark, R. N., and T. L. Roush, Reflectance spectroscopy: Quantitative analysis techniques for remote sensing applications, J. Geophys. Res., 89, 6329-6340, 1984.

Conel, J. E., Infrared emissivities of silicates: Experimental results and a cloudy atmosphere model of spectral emission from condensed particular mediums, J. Geophys. Res., 74, 1614-1634, 1969.

Coyne, L. M., J. L. Bishop, T. Scattergood, A. Banin, G. Carle, and J. Orenberg, Near infrared correlation spectroscopy: Quantifying iron and surface water in a series of variably cation-exchanged montmorillonite clays, in Spectroscopy of Minerals and Their Surfaces, edited by L. Coyne, S. McKeever, and D. Blake, pp. 407- 429, American Chemical Society, Washington, D.C., 1990a.

Coyne, L. M., J. L. Bishop, L. Howard, and T. Scattergood, Recent spectroscopic findings concerning clay-water interactions at low humidity: Possible applications to models of Martian surface reactivity, in 4th Symposium on Chemical Evolution and the Origin of Life, NASA-Ames Research Center, Moffett Field, Calif., 1990b.

Davies, D. W., The relative humidity of Mars' atmosphere, J. Geophys. Res., 84, 8335-8342, 1979.

Eggleton, R. A., and R. W. Fitzpatrick, New data and a revised structure model for ferrihydrite, Clays Clay Miner., 36, 111-124, 1988..

Erard, S., J.-P. Bibring, J. Mustard, O. Forni, J. W. Head, S. Hurtrez, Y. Langevin, C. M. Pieters, J. Rosenqvist, and C. Sotin, Spatial variations in composition of the Valles Marineris and Isidis Planitia regions of Mars derived from ISM data, Proc. Lunar Planet. Sci. Conf., 21st, 437-455, 1991.

Fanale, F. P., J. R. Salvail, W. B. Banerdt, and R. S. Saunders, Mars: The regolith-atmosphere-cap system and climate change, Icarus, 50, 381- 407, 1982.

Farmer, C. B., D. W. Davies, A. L. Holland, D. D. LaPorte and P. E.

Doms, Mars: Water vapor observations from the Viking orbiters, J. Geophys. Res., 82, 4225-4248, 1977.

Farmer, V. C., The layer silicates, in The Infrared Spectra of Minerals, edited by V. C. Farmer, 539 pp., Mineralogical Society, London, 1974.

Golden, D.C., R. V. Morris, D. W. Ming, H. V. Lauer Jr., and S. R. Yang, Mineralogy of three slightly palagonitized basaltic tephra samples from the summit of Mauna Kea, Hawaii, J. Geophys. Res., 98, 3401-3411, 1993.

Gooding, J. L., and K. Keil, Alteration of glass as a possible source of clay minerals on Mars, Geophys. Res. Lett., 5, 727-730, 1978.

Gooding, J. L., S. J. Wentworth, and M. E. Zolensky, Aqueous alteration of the Nakhla meteorite, Meteoritics, 26, 135-143, 1991.

Hayes, J. E. and J. F. Mustard, Effects of fine particles (<25 •tm) on reflectance spectra from 0.3 to 25 •tm, (abstract), Lunar Planet. Sci. Conf., XXV, 519-520, 1994.

Henry, R. L., The transmission of powder films in the infrared, J. Opt. Soc. Am., 38, 775-789, 1948.

Hess, S. L., R. M. Henry, and J. E. Tillman, The seasonal variation of atmospheric pressure on Mars as affected by the south polar cap, J. Geophys. Res., 84, 2923-2927, 1979.

Houck, J. R., J. B. Pollack, C. Sagan, D. Schaack, and J. A. Decker, Jr., High-altitude infrared spectroscopic evidence for bound water on Mars, Icarus, 18, 470-480, 1973.

Johnston, C. T., G. Sposito, and C. Erickson, Vibrational probe studies of


water interactions with montmorillonite, Clays Clay Miner., 40, 722- 730, 1992.

Kieffer, H. H., Spectral reflectance of CO2-H20 frosts, J. Geophys. Res., 75, 501-509, 1970.

Kieffer, H. H., T. Z. Martin, R. Peterfreund, and B. M. Jakosky, Thermal and albedo mapping of Mars during the Viking primary mission, J. Geophys. Res., 82, 4249-4291, 1977.

Logan, L. M., G. R. Hunt, J. W. Salisbury, and S. R. Balsamo, Compositional implications of Christiansen frequency maximums for infrared remote sensing applications, J. Geophys. Res., 78, 4983- 5003, 1973.

Lucey, P. G., and R. N. Clark, Spectral properties of water ice and contaminants, in Ices in the Solar System, edited by J. Klinger et al., pp. 155-168, D. Reidel, Norwell, Mass., 1985.

McCord, T. B., R. L. Huguenin, D. Mink, and C. Pieters, Spectral reflectance of Martian areas during the 1973 opposition: Photoelectric filter photometry, Icarus, 31, 25-39, 1977.

McCord, T. B., R. N. Clark, and R. B. Singer, Mars: Near-infrared spectral reflectance of surface regions and compositional implications, J. Geophys Res., 8Z 3021-3032, 1982.

Morris, R. V., J. L. Gooding, H. V. Lauer Jr., and R. B. Singer, Origins of Marslike spectral and magnetic properties of a Hawaiian palagonitic soil, J. Geophys. Res., 95, 14,427-14,434, 1990.

Murchie, S., S. Erard, J. Bishop, J. Mustard, J.-P. Bibring, Y. Langevin, J. Head, and C. Pieters, The forms and evolution of water in Martian

soil: Evidence from ISM imaging spectroscopy, (abstract), Lunar Planet. Sci. Conf, XXIII, 941-942, 1992.

Murchie, S., J. Mustard, J. Bishop, J. Head, C. Pieters, and S. Erard, Spatial variations in the spectral properties of bright regions on Mars, Icarus, 105, 454-468, 1993a.

Murchie, S., J. Mustard, S. Erard, and R. Singer, Variations in the Fe mineralogy of bright Martian soil,(abstract), Lunar Planet. Sci. Conf, XXIV, 1029-1030, 1993b.

Mustard, J. F., and J. F. Bell III, New composite reflectance spectra of Mars from 0.4 to 3.14 !xm, Geophys. Res. Lett., 21, 353-356, 1994.

Mustard, J. F., and C. M. Pieters, Photometric phase functions of common geologic minerals and applications to quantitative analysis of mineral mixture reflectance spectra, J. Geophys. Res., 94, 13,619- 13,634, 1989.

Owen, T., K. Bieman, D. R. Rushneck, J. E. Biller, D. W. Howarth, and A. L. Lafleur, The composition of the atmosphere at the surface of Mars, J. Geophys. Res., 82, 4635-4639, 1977.

Prost, R., l•tude de l'hydration des argiles' Interactions eau-min6ral et m6canisme de la retention de l'eau., II., l•tude d'une smectite, Ann. Agron., 26, 463-535, 1975.

Rosenqvist, J., P. Drossart, M. Combes, T. Encrenaz, E. Lellouch, J.-P. Bibring, S. Erard, Y. Langevin, and E. Chassefi•re, Minor constituents in the Martian atmosphere from the ISM/Phobos experiment, Icarus, 98, 254-270, 1992.

Roush, T. L., D. L. Blaney, and R. B. Singer, The surface composition of Mars as inferred from spectroscopic observations, in Remote Geochemical Analyses: Elemental and Mineralogical Composition, edited by C. M. Pieters and P. Englert, pp. 367-393, Cambridge University Press, New York, 1993.

Salisbury, J. W., Mid-infrared spectroscopy: Laboratory data, in Remote Geochemical Analysis: Elemental and Mineralogical Composition, edited by C. M. Pieters and P. A. J. Englert, pp. 79-98, Cambridge University Press, New York, 1993.

Salisbury, J. W., L. S. Walter, N. Vergo, and D. M. D'Aria, Infrared (2.1-25 lure) Spectra of Minerals, Johns Hopkins Univ. Press, 1991.

Schwertmann, R. M., and R. M. Cornell, Iron Oxides in the Laboratory, VCH Publications, Weinheim, Germany, 1991.

Sherman, D. M., R. G. Burns, and V. M. Burns, Spectral characterization of the iron oxides with application to the Martian bright region mineralogy, J. Geophys Res., 87, 10,169-10,180, 1982.

Singer, R. B., Spectral evidence for the mineralogy of high-albedo soils and dust on Mars, J. Geophys. Res., 87, 10,159-10,168, 1982.

Singer, R. B., T. B. McCord, R. N. Clark, J. B. Adams, and R. L. Huguenin, Mars surface composition from reflectance spectroscopy: A summary, J. Geophys Res., 84, 8415-8426, 1979.

Singer, R. B., J. S. Miller, K. W. Wells, and E. S. Bus, Visible and near- IR spectral imaging of Mars during the 1988 opposition, (abstract), Lunar Planet. Sci. Conf., XXI, 1164-1165, 1990.

Soderblom, L. A., The composition and mineralogy of the Martian surface from spectroscopic observations: 0.3 •tm to 50 •tm, in Mars, edited by H. Kieffer et al., pp. 557-593, University of Arizona Press, Tucson, 1992.

Stubican, V., and R. Roy, A new approach to the assignment of infrared absorption bands in layer silicates, Z. Kristallogr., 115, 200-214, 1961.

Toulmin, P., III, A. K. Baird, B. C. Clark, K. Keil, H. J. Rose Jr., R. P. Christian, P. H. Evans, and W. C. Kelliher, Geochemical and mineralogical interpretation of the Viking inorganic chemical results, J. Geophys Res., 82, 4625-4634, 1977.

Treiman, A. H., R. A. Barret, and J. L. Gooding, Preterrestrial aqueous alteration of the Lafayette (SNC) meteorite, Meteoritics, 28, 86-87, 1993.

J. L. Bishop, DLR, Institute for Planetary Exploration, Rudower Chaussee 5, D-12489 Berlin, Germany. (email: [email protected]. dlr.de)

C. M. Pieters, Brown University, Box 1846, Department of Geological Sciences, Providence, RI 02912. (email: [email protected]. brown.edu)

(Received March 3, 1994; revised November 17, 1994; accepted December 16, 1994.)

low&temperature and low atmospheric pressure infrared ... · journal of geophysical research,...

Documents