Transcript Parente2008Phyllo2_JB

Lab experiments on phyllosilicates
and comparison
with CRISM data of Mars
Mario Parente, Janice L. Bishop and Javier Cuadros
1
Outline
• Observation 1: diversity in clay spectroscopy in
the NIR not mirrored by equal diversity in visible
range
• Hypothesis: dust coating and mixtures dust /
clays
• Observation 2: the ferrous slope of Mawrth
Vallis
• Hypothesis: hydrothermal alteration of glass
into smectite as a possible spectroscopic
evidence for slope
• Comparison with CRISM
• Conclusions
2
Phyllosilicate diversity in VNIR
• In CRISM images, we
detect different clays
using Metal-OH bands
near 2.2 µm (Alsmectite, blue) and
2.3 µm (Fe/Mgsmectite, red)
• In Mawth Vallis,
phyllosilicates often
accompanied by
hydrated silica (opal
and amorphous
Al/SiOH) ~ green
• also detected using
2.2 µm band depth
Ferrous phase +
Fe/Mg-smectite ~
yellow
3
Little spectral variation in the
visible range
Lab spectra of montmorillonite and nontronite
show differences in metal-OH band positions
(2.2 vs 2.3 µm) + variation in the extended
visible region due to Fe electronic transitions
in the nontronite.
CRISM spectra of Mawrth Vallis (from McKeown
et al 2008) show variation in metal-OH bands
but almost identical spectra from 0.4 to 1.2 µm
4
Little spectral variation in the
visible range
• Further investigation
showed subtle differences
in the position of the peak
near 0.7-0.8 µm
5
Dust coating experiments
• We simulated dust deposition by
sprinkling
progressively
increasing amounts of dust on
top
of
nontronite
and
montmorillonite rocks using a
spatula.
• We measured the spectra from
0.3 to 2.5 mm of the rock
surfaces and the rocks with
different
amounts
of
both
nanophase ferrihydrate dust and
fine-grained Haleakala ash on a
black Teflon dish using an ASD
spectrometer
under
ambient
conditions.
• In order to qualitatively assess
the amount of dust present on
the rocks, we captured images of
the samples with a microscope at
multiple magnifications after
dust deposition
6
Nontronite with Haleakala
coating
•
•
•
•
Number of stars represents level of
dust.
Effect on NIR region is minimal.
Significant and gradually increasing
attenuation of absorptions near 600
and 900 nm.
Position of nontronite peak near 780
nm shifts only slightly.
7
Nontronite with Ferrihydrate
coating
• Effect on NIR region is
minimal.
• Significant attenuation of
absorption near 600 nm.
• Position of nontronite peak
near 780 nm does not shift
8
Montmorillonite with
Haleakala coating
•
•
•
Primary effect of Haleakala ash
coating is darkening of the spectra.
Haleakala coating produces change
in slope between 500 and 700 nm
and a slight shift of the
montmorillonite peak around 800
nm
Note- montmorillonite has strong
bands due to adsorbed water.
9
Montmorillonite with
Ferrihydrate coating
•
•
•
Dominance of ferrihydrite
coating in visible region.
Slight shift of
montmorillonite peak
towards longer WL
Note- montmorillonite
has strong bands due to
adsorbed water.
10
Comparing coatings with
mixtures
• Fine-grained
mixture of
nontronite and
ferrihydrate
(prepared by
Alicia Muirhead).
• In both coating
and mixture we
observe the
dominating effect
of the iron-oxide
before 0.7 µm
11
The ferrous component in
spectra of Mawrth Vallis
• Ferrous slope observed at transition from nontronite to
Al-clay/hydrated silica layers.
• Likely to be mica (could also be carbonate, sulfate,
olivine).
• Extremely unusual on Earth without microbes.
• On Mars implies active aqueous chemistry:
– Rapid input of Fe2+ into system that converted to mineral
quickly before Fe could be oxidized.
– Closed system where Fe3+ in nontronite was reduced by
unknown source at upper boundary.
12
The ferrous component in the
spectra of Mawrth Vallis
•
•
•
•
•
•
Red is Fe/Mg smectite
(Fe3+)
Green is ferrous phase
(Fe2+)
Blue is Alphyllosilicate and/or
hydrated silica
Fe/Mg-smectite found in phyllosilicate deposits at
lower elevations.
OH bands indicate FeMg-OH in octahedral layers.
Fe2+ slope typically observed at boundaries of
Fe/Mg-smectite deposit.
43EC
CRISM image 43EC draped over MOLA with 20X vertical enhancement
13
Alteration of volcanic glass:
spectral features
•
•
•
•
De la Fuente, Cuadros et al.,
explored hydrothermal alteration
as possible scenario for
formation of smectites and
preservation of ferrous features.
The study used volcanic tuff
samples primarily composed of
rhyolitic glass and were exposed
to hydrothermal alteration in the
lab at temperatures ranging from
60 to 160 C for up to a year
Most prominent effect of
abundant smectite present in the
matrix is a shift in Fe band
position from 1.1 µm towards
shorter WL.
Ferrous slope in spectra of these
sample is less than that observed
at Mawrth Vallis; however, other
ferrous-bearing glasses may have
a stronger ferrous absorption.
14
Alteration of volcanic glass:
spectral features
•The alteration does not affect
the spectrum before 0.7 µm
•The band center near 1.1µm
shifts towards lower WL for
more smectite formed
15
Direct comparison with CRISM
• CRISM spectra have the band
around 1.2 µm removed
• We compare with selected
spectra from the studies
described in the previous
slides
16
Direct comparison with CRISM
(close-up)
• Subtle changes observed in
extended visible region
(max near 0.8 and min
near 1 µm) for
montmorillonite and
nontronite rocks implying
changes in Fe mineralogy
with these clay minerals.
• The concave feature near
0.5 µm is better explainred
by the presence of iron
oxides
• The band minimum around
1.0 µm could be due to a
mixture
17
Conclusions
• Coatings: only a few grains of ferrihydrate or altered
volcanic ash coating a rock surface are sufficient to
alter the spectral properties of the rock in the visible
region. At the same time these coatings have only a
minor effect on the glass bands from 1.9 to 2.5.
• Mixtures: fine grains of nontronite mixed with
ferrihydrite show only a hint of 0.6 µm nontronite
shoulder with 10% ferrihydrite (work by REU intern Alicia
Muirhead).
• Chemical alteration: hydrothermally altered ferrous
glass mixed with phyllosilicates in rocks could
account for the presence of the ~ 1 μm band in the
clay units at Mawrth Vallis
• There are possibly multiple causes of the spectral
features observed
18