Abstract - GEORaman 2014

S. Shkolyar1, J. D. Farmer1, and J. Blacksberg2, 1School of Earth and Space Exploration, Arizona State University,
Tempe, AZ. [email protected], 2Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA.
Introduction: The NRC’s 2013-2022 Planetary
decadal survey for Solar System Exploration assigned
a high priority to Mars sample return (MSR). A Mars
sample selection and caching mission, to be launched
in 2020, has been identified as the first step in this endeavor. This mission will identify aqueously formed
sedimentary rocks at a landing site on Mars, and based
upon their enhanced potential for biosignature capture
and preservation, will prioritize and cache them for
eventual return to Earth. High priority will be given to
caching samples that show evidence for past habitable
environments and conditions favorable for preservation
of fossil biosignatures. One of the most compelling
criteria for sample selection will be evidence for fossilized carbonaceous matter (kerogen) of possible biological origin. Such samples will be cached and returned
to Earth to look for evidence of fossil biosignatures
using more detailed analyses in terrestrial laboratories.
The returned sample volume and mass will likely
be restricted to ~30 samples and a total mass of <500
grams [1]. For an effective sample return campaign for
Astrobiology, it will be essential that samples selected
for the MSR cache will have a high potential for capturing and preserving fossil biosignatures.
Raman spectroscopy is a nondestructive surface
technique for analyzing materials in situ, with minimal
sample preparation. It provides information about both
the mineral composition and organic matter present in
rock samples.
Study: To better understand the complexities of
identifying kerogen in aqueous sedimentary materials
using Raman spectroscopy, we surveyed a broad range
of sample materials generally regarded to represent
high priority habitable environments for the astrobiological exploration of Mars [3]. Lithotypes studied
included cherts, carbonates, sulfates, and clay-rich
shales. Samples studied covered a range of geological
ages and kerogen maturities.
Samples used for the study were also characterized
by X-ray Diffraction (XRD) and thin section petrography to provide independent assessments of mineralogy
and micropaleontology, and to map the distribution of
kerogenous fossil materials present in samples.
Results: The study showed that reliable kerogen
identification in most of the lithologies studied is often
complicated by high background fluorescence originating from both long (>10 ns to ms) and short (<1 ns)
lifetime effects arising from both organic matter as
well as the host mineral matrixes.
Some lithotypes, such chert dikes within the Archean Strelley Pool chert, Pilbara region of western
Australia, displayed consistently low background fluorescence, enhancing both mineral identifications and
kerogen detection (Fig. 1). One possible reason for the
lower fluorescence in this sample may be the fluorescence-quenching effects of increased concentrations of
poly-aromatic hydrocarbons present in some kerogen
types, such as thermally mature kerogen [2] seen in the
Strelley Pool chert. Other samples allowed mineral and
kerogen identification through high but non-prohibitive
fluorescence (Fig. 1).
Fig. 1. Non-prohibitive fluorescence in Raman spectra in minerals and kerogen. Raman spectra are
shown for kerogen in a carbonate-rich lamination of
the Castille Formation sulfate which exhibited high
fluorescence but non-prohibitive detection of both
1350 cm-1 (D band) and 1600 cm-1 (G band) kerogen
peaks, as well as dolomite. The chert from Strelley
Pool allowed both kerogen bands as well as quartz to
be detected. The chert exhibited low fluorescence.
In contrast, carbonate lithotypes (e.g., organic-rich
calcareous shale from Green River Formantion, WY)
consistently showed high background fluorescence due
to contributions from both immature kerogen, and the
surrounding carbonate matrix (Fig. 2). In other carbonate examples, such as the stromatolitic limestone
sample from Walker Lake, NV, high flurescence prohibited kerogen detection, except at high beam energies, where both matrix minerals and kerogen may
have been thermally degraded.
Abstract for 11th GeoRaman International Conference, June 15-19, 2014, St. Louis, Missouri, USA
Fig. 2. Prohibitive fluorescence in Raman spectra
in minerals and kerogen. In the spectrum of a muscovite grain cemented by carbonate in the Alum
sandstone, high fluorescence impeded its detection.
Kerogen in an organic-rich lamination between calcite-rich laminations in the Green River Fmn. (GRF)
calcareous shale exhibited high fluorescence, allowing only a weak kerogen peak at 1600 cm -1 (G band)
to be visible, but prohibiting the peak at 1350 cm-1 (D
band) to be visible due to fluorescence. (Two peaks
from pyrite were captured in the sampling volume.)
It is important to note that both carbonate and chert
lithologies are among the most common host rocks for
the Precambrian microfossil record on Earth and arealso high-priority targets for in situ analysis and Mars
sample return [3].
Work by Blacksberg et al [4, 5] has shown the potential of time-resolved Raman spectroscopy for reducing the effects of high background fluorescence and
improving capabilities to detect both kerogen and minerals. In this study, the same time-resolved Raman
spectroscopy instrument showed promise in reducing
fluorescence where it was prohibitive in some of our
Mars analog samples. For example, in the Alum sandstone, glauconite, a phyllosilicate indicative of a marine depositional (habitable) environment, was suspected in microscopy analyses based its color in plane
light, birefringence, grain morphology, and friability. It
was confirmed as glauconite only with time-resolved
Raman (Fig. 3). CW (standard, continuous wave) Raman was unable to confirm its identity due to high fluorescence and XRD did not detect it due to its low
abundance. If glauconite were present and detectable
by Raman on a Mars lander, along with the 1350 and
1600 cm-1 peaks, this represents an ideal example of a
sample for caching and return to Earth. It would represent a hydrated phyllosilicate phase indicative of a
habitable environment, and one that showed an identification for a carbon biosignature.
Conclusions and Recommendations: Because the
Fig. 3. Potential of time-resolved Raman in identifying minerals through high fluorescence. Glauconite
(photomicrograph shown in inset) was detected in the
Alum sandstone sample with Time-Resolved Raman
spectroscopy. XRD and CW Raman were not successful in definitively identifying it.
sample mass for samples returned to Earth will be limited, effective in situ sample selection will be critically
important for caching those samples with the highest
potential for biosignature preservation at a site. While
some lithotypes do not show high background fluorescence, our studies have shown that most of the high
priority lithotypes do. If Raman is used to screen for
the best samples, it will be important to minimize fluorescence effects in samples to ensure success.
This study recommends: (1) continued research to
further evaluate time-resolved Raman techniques to
reduce the effects of fluorescence in high-priority target lithologies for Mars and (2) following a sciencedriven approach to identify and understand the impacts
of sample-dependent issues when defining future exploration strategies.
Acknowledgement: Part of the research described
here was carried out at the Jet Propulsion Laboratory,
California Institute of Technology, under a contract
with the National Aeronautics and Space Administration (NASA). Erik Alerstam and Yuki Maruyama of
JPL are acknowledged for assistance with data collection on the time-resolved Raman.
References: [1] Mars 2020 SDT (2013), Committee members: Mustard, J. F. (chair), et al.: Report of
the Mars 2020 Science Definition Team, 154 pp.,
T_Report_Final.pdf. [2] Bertrand, P. et al. (1985) Advances in Organic Geochemistry, 10, 641-7. [3]
Farmer, J. D. & Des Marais, D. J. (1999) JGR 104,
26977. [4] Blacksberg, J. et al. (2014) 45th LPSC, Abstract #1544. [5] Blacksberg, J. et al. (2013) 2013
AGU, Abstract # P51G-1803.
Abstract for 11th GeoRaman International Conference, June 15-19, 2014, St. Louis, Missouri, USA