Research Article

ASTROBIOLOGY Volume 15, Number 8, 2015 ª Mary Ann Liebert, Inc. DOI: 10.1089/ast.2015.1329

Raman Spectroscopic Techniques for Planetary Exploration: Detecting Microorganisms through Minerals Mattheus F.C. Verkaaik,1 Jan-Hein Hooijschuur,1,2 Gareth R. Davies,2 and Freek Ariese1

Abstract

Raman spectroscopy can provide highly specific chemical fingerprints of inorganic and organic materials and is therefore expected to play a significant role in interplanetary missions, especially for the search for life elsewhere in our solar system. A major challenge will be the unambiguous detection of low levels of biomarkers on a mineral background. In addition, these biomarkers may not be present at the surface but rather inside or underneath minerals. Strong scattering may prevent focusing deeper into the sample. In this paper, we report the detection of carotenoid-containing microorganisms behind mineral layers using time-resolved Raman spectroscopy (TRRS). Two extremophiles, the bacterium Deinococcus radiodurans and the cyanobacterium Chroococcidiopsis, were detected through translucent and transparent minerals using 440 nm excitation under resonance conditions to selectively enhance the detection of carotenoids. Using 3 ps laser pulses and a 250 ps gated intensified CCD camera provided depth selectivity for the subsurface microorganisms over the mineral surface layer and in addition lowered the contribution of the fluorescent background. Raman spectra of both organisms could be detected through 5 mm of translucent calcite or 20 mm of transparent halite. Multilayered mineral samples were used to further test the applied method. A separate tunable laser setup for resonance Raman and a commercial confocal Raman microscope, both with continuous (non-gated) detection, were used for comparison. This study demonstrates the capabilities of TRRS for the depth-selective analysis through scattering samples, which could be used in future planetary exploration to detect microorganisms or biomarkers within or behind minerals. Key Words: Time-resolved Raman spectroscopy—Resonance Raman—Deinococcus radiodurans—Chroococcidiopsis—Extremophiles—Mineral inclusions. Astrobiology 15, xxx–xxx.

a Raman instrument (Wang et al., 1995), various Raman spectroscopic techniques like remote RS (Angel et al., 2012), resonance Raman spectroscopy (RRS) (Marshall et al., 2007), and time-resolved Raman spectroscopy (TRRS) (Blacksberg et al., 2010; Hooijschuur et al., 2013) have been applied to optimize detection of both minerals and organic compounds. Ongoing research of life in extreme terrestrial environments is of great help for the development of analytical techniques for the detection of life signatures in planetary exploration. Examples of extreme environmental parameters are regions with extreme temperatures: deserts (Edwards et al., 2010) and arctic regions ( Jorge Villar et al., 2005), extreme pH levels that typify acidic (Edwards et al., 2007) or alkaline rivers and lakes, extremely dry environments (Vı´tek et al., 2014), and brine lakes. The extremophiles in these environments are (micro)organisms that can survive and in some cases only live in (a combination of) the previously

1. Introduction

R

aman spectroscopy (RS) for future application in planetary sciences has been the subject of a large number of exploratory studies. RS is a nondestructive technique for the identification of organic and inorganic compounds based on molecular vibrations and can be used for detection of life in planetary exploration ( Jorge Villar and Edwards, 2006). Both the European Space Agency and the National Aeronautics and Space Administration have assigned a Raman instrument to one of their scheduled Mars missions, ExoMars in 2018 with 532 nm excitation (Edwards et al., 2012) and Mars 2020 (Beegle et al., 2014), respectively. NASA will equip their rover with two Raman instruments: one for the SHERLOC system with 248.6 nm excitation and one for the SuperCam system with 532 nm excitation and gated detection that can be used for stand-off measurements (Wiens et al., 2014). Since the first proposal to equip a planetary rover with

1 2

LaserLaB, Faculty of Sciences, VU University Amsterdam, Amsterdam, the Netherlands. Deep Earth and Planetary Science, Faculty of Earth and Life Sciences, VU University Amsterdam, Amsterdam, the Netherlands.

1

2

mentioned extreme environmental conditions (Harrison et al., 2013). These environments are good analogues for environmental conditions on other bodies in our solar system. Mars, Enceladus, Europa, and Titan are potential candidates for viable environmental conditions (McKay, 2011). On Earth, various strategies have evolved over time to allow for survival under extreme conditions, for instance, physical protection via mobilization or transformation of minerals (Edwards et al., 1997; Jorge Villar et al., 2005) or hiding in and on different minerals (epiliths), inside porous rocks (endoliths), or inside cracked rocks (chasmoliths) (Edwards, 2010). Chemical protection against ultraviolet radiation is offered by compounds such as scytonemin or carotenoids (Edwards et al., 2012). To prevent desiccation, organisms may have high concentrations of organic osmotic solutes like glycerol, simple sugars, amino acids in their cytoplasm or use extracellular oxalate precipitates (Bo¨ttger et al., 2013). The search for life focuses on so-called biomarkers. Biomarkers are molecules that are unambiguously produced by living organisms and cannot be synthesized via an abiotic route (Edwards et al., 2014). The potential and limitations of RS for carotenoid biomarker detection were discussed by Jehlicka et al. (2014), and competing processes like photodegradation and self-absorption during RRS of carotenoids were described by Hooijschuur et al. (2015a). The ideal method for the detection of biomarkers would be one that is sensitive, highly specific, non-invasive, and capable of detecting molecules hidden inside (nontransparent) minerals. Fluid inclusions are microscopic bubbles that are trapped within minerals and consist of fluids or gasses. These inclusions may contain microorganisms that have been trapped for hundreds of millions of years (Lowenstein et al., 2011). Raman analysis of microorganisms in halite (Conner and Benison, 2013), gypsum (Benison and Karmanocky, 2014), and ancient extremely acidic lake evaporates (Benison et al., 2008) have been reported, all involving 0.5–2 mm thick sections. Fluorescence and photodegradation processes were observed in many cases. These processes are believed to have compromised the analytical data, resulting in poorer detection limits. Laboratory-grown gypsum crystals containing fluid inclusions with biomarkers, for example, b-carotene for Raman microscopic analysis, were synthesized in 3.3, 5.2, and 8.5 mm thick pieces by Osterrothova and Jehlicka (2011). Field samples and laboratory samples of the previously mentioned research consisted of transparent crystals that were measured with microscope objectives with a long working distance. In this study, we show that clear, transparent samples are not a prerequisite for Raman detection of microorganisms in and behind minerals. In addition, we show that the sample thickness can be larger than the focal distance of a microscope objective. Raman spectroscopy is based on the inelastic scattering of photons from a molecule. Energy from the photon is transferred to the molecule in roughly 1 out of every 107 to 108 incident photons as a result of its inherently low Raman cross section. The amount of transferred energy corresponds with one of its molecular vibrations. The combination of molecular vibrations as shown in a Raman spectrum is characteristic for each compound (molecular fingerprint), for organic as well as inorganic materials. RS is a nondestructive and non-invasive technique that usually needs no

VERKAAIK ET AL.

sample preparation. Raman can also be used for detection of microorganisms, since water is not a source of interference. Detecting and identifying biomarkers (if present) on or below a planetary surface will be a tremendous challenge. Raman spectra of samples that consist of microorganisms inside minerals will be heavily dominated by the Raman signal of the surrounding matrix (the mineral) and show only little signal from the organism itself. Techniques like RRS can be used to selectively enhance the Raman signals of compounds with chromophoric groups. Sensitivity can be enhanced by a few orders of magnitude if the excitation wavelength of the laser overlaps with an electronic transition of the molecule (Efremov et al., 2008; Robert, 2009). Other advanced analytical techniques have been developed to obtain depth information of multilayered samples up to a few centimeters thick. Spatially offset Raman spectroscopy (SORS) can provide depth information based on changing peak ratios when applying a lateral offset between the excitation and collection point (Buckley and Matousek, 2011). Alternatively, TRRS utilizes fast gated detection modes like Kerr-gating (Matousek et al., 2005) or fast intensified charge-coupled device (ICCD) detectors (Ariese et al., 2009) to obtain depth information of layered samples and decrease the fluorescent background. Both effects are based on temporal differences between two signals: in a back-scattering geometry the surface-layer Raman signals arrive earlier at the detector than subsurface Raman signals, and Raman is a fast process compared to fluorescence. TRRS has been successfully used in a variety of research areas like biomedical (Iping Petterson et al., 2015) and forensic applications (Iping Petterson et al., 2011). Gated detection for planetary sciences can be used for different reasons, for example, fluorescence suppression (Soleimaninejad et al., 2012), day-time measurements (Misra et al., 2005), and stand-off measurements (Babin et al., 2011; Skulinova et al., 2014). The latter approach will also be used with the NASA SuperCam system, which is designed to yield a 100 ns time resolution (Wiens et al., 2014). Gaft and Nagli (2009) described the necessity of using gated detection in fundamental and applied mineralogy for suppressing the luminescence and fluorescence contribution to a Raman spectrum. The luminescence lifetimes of minerals can be in the order of 10-3 to 10-6 s for minerals, and fluorescence lifetimes can be as short as 10-9 s for organic compounds. This study utilizes an ICCD detector with a fast, 250 ps gate in combination with 3 ps pulses from a frequency-doubled Ti:sapphire laser (k = 440 nm) at a repetition rate of 76 MHz. The ICCD detector is activated after a specific delay following the arrival of a trigger pulse as shown in pink in the schematic overview of events during a TRRS measurement (Fig. 1a). Although the CCD detector is activated during the entire exposure time, the signals are only collected during the times that the intensifier is on (gray). Amplification of the signal is achieved by an image intensifier consisting of a photocathode to convert incident photons from the sample to electrons, a microchannel plate (MCP) to multiply these electrons, followed by a phosphorescent screen to convert the electrons back to photons. These amplified photons are detected by a cooled CCD chip. The trigger pulse (after a defined delay) starts the buildup of potential over the image intensifier, switching it on. At relatively short delays, only photons from the first layer are detected. Raman photons from deeper layers (yellow) are detected at relatively longer delays.

RAMAN OF MICROORGANISMS THROUGH MINERALS

3

FIG. 1. (a) Schematic representation of the time events in TRRS of a two-layered sample in backscattering geometry. After the 3 ps laser pulse (blue), both Raman (red for the first layer and yellow for the second layer) and fluorescence (green) signals are produced. The 250 ps gate is synchronized to the laser pulse, initiated by the trigger (pink), and opens after an adjustable delay. This prevents a large contribution of fluorescence signals to the Raman spectrum and provides depth information about multilayered samples. Relative intensities and lifetimes of the different processes may vary (up to orders of magnitude) for different samples. (b) Illustration of fluorescence suppression of the cyanobacterium Deinococcus radiodurans using TRRS with a 250 ps detector gate. With continuous detection (top spectrum), an additional high fluorescent background is observed. For the TRRS measurement (lower spectrum), the 250 ps gate reduces the additional fluorescent signal and shows a 4.5· improvement of the signal-to-baseline ratio for the C = C stretch vibration at 1513 cm-1. Note that the intensity scale of the two detectors cannot be directly compared. Both spectra were recorded using the same instrument, at a 440 nm excitation wavelength with three accumulations of 20 s. (Color graphics available at www .liebertonline.com/ast) Fluorescence (green) often competes with Raman scattering. This fluorescence can originate from the analyte itself or from compounds in the surrounding matrix. A temporal distinction can be made between fluorescence and Raman, since the fluorescence process is orders of magnitude slower than Raman. The detector opens and closes before the majority of the fluorescence photons arrive. This makes TRRS an appropriate technique for reducing the fluorescent background in Raman spectra, as demonstrated in Fig. 1b. This research elaborates on our previous work on depth analysis of multilayered mineral samples utilizing TRRS (Hooijschuur et al., 2013) and detection of carotenoids and microorganisms in mineral matrices with RRS (Hooijschuur et al., 2015a). Here, we demonstrate for the first time how two species of microorganisms, D. radiodurans and the cyanobacterium Chroococcidiopsis, can be detected through different layers of translucent and transparent minerals using RRS and TRRS. For comparison, Raman spectra were also recorded by using a continuous-wave (CW) Raman setup with resonance excitation and a confocal Raman microscope, both with continuous (non-gated) detection. 2. Materials and Methods 2.1. First layers (minerals)

Calcite and gypsum slices of precise thicknesses were obtained from our own mineral collection at the VU University Amsterdam. Halite was obtained from the Wereldwinkel, Amsterdam. The calcite slice was 2.50 mm thick, the gypsum slices 5.0 and 6.0 mm thick, and the halite 20.0 mm thick. Two gypsum slices could be combined to make an 11 mm thick

layer. The calcite was used as a proxy for translucent (i.e., highly scattering) minerals, the gypsum and halite as proxies for transparent minerals. The centrosymmetric halite crystal did not show any Raman or fluorescence signals. The mineral slices were not sanded or polished. Labradorite (Ca,Na)(Al,Si)4O8 and spectrolite (Ca,Na)(Al,Si)4O8, both of which are plagioclase group minerals from Madagascar, along with spessartine Mn3Al2(SiO4)3, a garnet mineral from China, were acquired at Het Kwartskristal, Gouda, the Netherlands, and were sawn into plates of varying thicknesses. Labradorite was used as a proxy for transparent minerals, spessartine as a proxy for scattering minerals, and spectrolite as a proxy for almost opaque minerals. 2.2. Second layers (microorganisms)

A wild strain of D. radiodurans was provided by Wilfred Ro¨ling, VU University Amsterdam. Bacteria were grown on a microbiologically tested nutrient agar plate for 3 days in a dark environment at 30C. One isolated colony was transferred from the plate to 1.50 L of aseptic 13 g$L-1 Nutrient Broth #3 (Sigma-Aldrich, St. Louis, USA) solution and was set to proliferate in a dark environment for 4 days at 30C. After the proliferation phase, the suspension was centrifuged (3500g for 3 min), washed with normal saline, and centrifuged again (3500g for 3 min). Wet bacterial paste was put in a cuvette with 1.20 mm thick glass walls (OS 0,501, Hellma Analytics, Mu¨llheim, Germany). A wild Chroococcidiopsis strain from the Culture Collection Yerseke was provided by Gerard Muyzer, University of Amsterdam. The culture was proliferated in self-made liquid BG-11 medium, under constant aeration and in

4

luminescent light. Harvesting and storage were performed as for D. radiodurans. For resonance Raman measurements, a deposit of D. radiodurans was prepared on a glass microscope slide. Concentrated D. radiodurans paste (0.30 g) was suspended in 1.50 mL of normal saline and diluted to 0.10 g$mL -1. A total of 50 lL of this dilution was pipetted onto a defined 12.7 · 12.7 mm area on a glass microscope slide. The deposit was dried in ambient air and covered with a second glass plate. To quantify the density of the bacterial layer, a surface area of 254 · 187 lm was automatically mapped using the confocal Raman microscope with the 532 nm laser, and the number of bacteria was manually counted, yielding a total of *9.5$107 bacteria/cm2. Raman mapping was necessary since the optical view did not yield sufficient contrast between bacteria and background. Preparation of a similar Chroococcidiopsis sample plate with a specific density of bacteria was not possible since this bacterium did not form a smooth, homogeneous dry deposit. Reference spectra for the minerals and bacteria are shown in Fig. 2. The carotenoid Raman peaks at 1151 and 1513 cm-1 for D. radiodurans and at 1151 and 1523 cm-1 for Chroococcidiopsis are characteristic for these bacteria. The appearance of both peaks and their relative intensities can serve as proof of detection when combined with minerals in test samples. A third carotenoid peak that is often observed near

FIG. 2. Raman reference spectra of D. radiodurans (black), Chroococcidiopsis (green), gypsum (blue), and calcite (red). Major Raman peaks are labeled. All spectra were recorded in TRRS mode at a 440 nm excitation wavelength and 70 mW laser power with three accumulations of 20 s each at a relative time delay of 150 ps. The upper two spectra have been multiplied by 3, and all spectra are shown with a y axis offset for legibility. As expected, no Raman signals were obtained for halite. (Color graphics available at www.liebertonline .com/ast)

VERKAAIK ET AL.

1007 cm-1 is weaker and easily obscured by significantly stronger Raman bands from calcite and gypsum. 2.3. Instrumentation

Resonance Raman measurements with continuous detection were made with a 532 nm, frequency-doubled Nd:YVO4 laser (Coherent Paladin Advanced 532-20000, Santa Clara, CA, USA) operated at 20 W and pumping an OPO laser (APE Levante Emerald, Angewandte Physik & Elektronik GmbH, Berlin, Germany) with a tunable wavelength range for the Signal from 690 to 990 nm and an Idler range from 1150 to 2300 nm. Both Signal and Idler can be frequency doubled/tripled (APE HarmoniXX). For the RRS measurements of this work, the OPO signal was tuned to 880 nm and frequency doubled to 440 nm. The signal was collected in backscatter mode through a 20 · , 0.65 NA microscope objective (Partec GmbH, Mu¨nster, Germany) and a dielectric stack long-pass filter, Semrock 450 AELP (Semrock Inc., Lake Forest, IL, USA), focused at a spectrograph (Andor Shamrock, SR-303i-A, Belfast, UK) and detected with an Andor Newton 920 CCD detector (Andor DU920P-BR-DD) cooled to 213 K. Raman microscope spectra were recorded using a Renishaw InVia Reflex confocal Raman microscope (Wottonunder-Edge, UK) with a Peltier-cooled CCD detector (203 K). The maximum output power of the 532 nm laser is 80 mW; the powers actually applied at the sample are given in the figure captions. The instrument included a Leica light microscope with 5 · , 20 · , and 50· air objectives. The TRRS measurements were conducted with a 532 nm, frequency-doubled Nd:YVO4 laser (Coherent Verdi-V18, Santa Clara, CA, USA) operated at 10 W, which pumps a tunable Ti-sapphire laser that produces 3 ps pulses at a repetition rate of 76 MHz (Coherent Mira 900P). The laser was tuned to 880 nm and frequency doubled to 440 nm (TP2000B, U-Oplaz). The first lens (diameter 25.4 mm and f = 50 mm) collects the backscattered light and sends a parallel bundle through the dielectric stack long-pass filter, Semrock 450 AELP (Semrock Inc., Lake Forest, IL, USA), used to block the reflected laser line and Rayleigh scattering. A second lens focuses the backscattered Raman emission onto the 70 lm entrance slit of a 50 cm single-stage spectrograph (SpectraPro, Acton, MA, USA). Time-gated detection of the Raman spectra was performed using an ICCD camera (LaVision Picostar HR, Go¨ttingen, Germany) with a MCP operated at a potential difference of 900 V and a CCD chip cooled to 262 K. The ICCD is triggered by a photodiode that receives a fraction of the laser output via a glass plate. The applied potential over the photocathode, MCP, and phosphorescent screen of the intensifier in combination with other settings determines the gate width of the detector, that is, the time window in which the ICCD opens and closes during a single collection event (Fig. 1), apart from the delay relative to the laser pulse. Theoretically, the imaginary shape of the gate would be a block pulse, that is, with instantaneous opening and closing of the detector response. However, since the buildup and decay of the potential take significant time on a picosecond timescale, the time gate will have a bell-shaped profile. The hardware parameters were optimized to achieve maximum signal collection in the shortest possible time window. This

RAMAN OF MICROORGANISMS THROUGH MINERALS

reduces the convolution of signals from varying locations in the sample, thereby increasing the depth resolution of the Raman signals. This optimization was performed by measuring a delay series of a 1.0 mm cuvette with cyclohexane (Acros Organics, Fisher Scientific, Geel, Belgium) and iteration of the parameters, resulting in a gate width of 250 ps full width at half maximum. For comparative measurements, continuous wave detection (but with the same pulsed laser system for excitation) was carried out by a non-gated CCD camera (model DV420-OE, Andor Technology, Belfast, UK) side-mounted to the same spectrograph and operated at a temperature of 223 K. The intensity units of the gated and non-gated detectors are not directly comparable. 2.4. Measurements

The tunable Ti-sapphire laser was set to an output wavelength of 880 nm. This wavelength was frequency doubled to 440 nm, matching the absorption maximum of the carotenoids in both D. radiodurans (Hooijschuur et al., 2015b) and Chroococcidiopsis (Erokhina et al., 2002). The laser power was *70 mW at the sample. The spectral readout was manually calibrated in wavenumbers by using acetaminophen (98%, Sigma-Aldrich, Saint Louis, MO, USA). In our TRRS measurements, the first Raman signals would appear at a delay time of 2650 ps after the trigger pulse, originating from photons scattered directly from the mineral surface. This delay time will be used as t0, and throughout this paper the delays will be expressed relative to that point. All TRRS measurements were performed in backscatter mode, using 25 ps delay steps with three acquisitions of 20 s per step. Time-gated measurements were performed at increasing delays between -150 and 750 ps, incremented in 25 ps steps unless indicated otherwise. For some experiments, a spatial offset between the excitation point and the collection point was created in order to have less contribution from signals of the first layer (Buckley and Matousek, 2011). To this end, the small prism that steers the excitation beam was moved a few millimeters parallel to the sample surface. For more details and a sketch of the setup, see the

5

work of Hooijschuur et al. (2013). The reference spectra of Fig. 2 from both bacterial pastes and minerals were recorded at 150 ps delay for identification purposes. 3. Results

For RRS measurements with continuous detection, the glass plate with dried D. radiodurans or Chroococcidiopsis deposits was placed behind mineral plates of varying thicknesses and excited at 440 nm with the CW Raman setup. The minerals used in this subseries of experiments were spectrolite, labradorite, and spessartine. Ten accumulations of 15 s each were recorded in non-gated mode, with the focus of the backscatter collection lens set to the bacterial layer. The resulting RRS spectra are shown in Fig. 3 for both microorganisms. Maximum thicknesses of the mineral layers allowing the bacterial layers to be detected were determined as 0.5 mm for spectrolite (i), 4 mm for labradorite (ii), and 1 mm for the spessartine (iii). Using a confocal Raman microscope with continuous detection, we recorded spectra of a cuvette containing a wet paste of D. radiodurans or Chroococcidiopsis placed under 2.50 mm of translucent calcite, using the 532 nm laser operating at 31 mW in combination with the 5· air objective (Fig. 4). A wet, nonquantified bacterial sample was chosen because Raman signals of the dry bacterial samples were below the detection limit. Spectra were recorded both focused on the upper mineral surface and focused 14 mm into the sample. Only in the latter case did we detect the carotenoid signals from both bacteria; when focusing at the surface, the spectra were overwhelmed by the 50-fold enhanced calcite signal of the first layer. Defocusing the objective in order to improve detection through scattering layers was shown earlier by Eliasson et al. (2007). With this approach, we could observe bacteria through 2.5 mm of translucent calcite, 20 mm of transparent halite (results not shown), or 11 mm of transparent gypsum (results not shown). Figure 4 also shows the strong fluorescence background in both spectra, as commonly observed in conventional CW Raman measurements.

FIG. 3. Resonance Raman spectra of (a) D. radiodurans and (b) Chroococcidiopsis behind different mineral layers. The microorganisms could be detected through 0.5 mm of spectrolite (i), 4 mm of labradorite (ii), or 1 mm of spessartine (iii). All spectra were recorded at a 440 nm excitation wavelength and 21 mW laser power with 10 accumulations of 15 s (non-gated detection). (Color graphics available at www.liebertonline.com/ast)

6

FIG. 4. Out-of-focus confocal Raman microscope spectra of D. radiodurans (i) and Chroococcidiopsis (ii) behind 2.5 mm of translucent calcite. Raman peaks could not be detected from either microorganism while focusing at the surface of the calcite layer but could be observed when the objective was positioned close to the calcite surface. In addition, the Raman signals of the calcite layer decreased almost 50-fold during defocusing. Both spectra were recorded at an excitation wavelength of 532 nm with 31 mW laser power at the sample, 1.0 s exposure time, and 60 accumulations. For the TRRS measurements, first a 20.0 mm thick block of halite was used as a model for transparent minerals. Since the centrosymmetric NaCl crystal itself does not produce any Raman signals, all three major Raman signals of D. radiodurans at 1007, 1151, and 1513 cm-1 are clearly detectable through the mineral layer (Fig. 5a). The first Raman

VERKAAIK ET AL.

signals of D. radiodurans appear at a delay of 2900 ps, corresponding to a relative delay of 250 ps. Taking into account the refractive index of halite (n = 1.5443), nonscattered photons will take 103 ps to travel through 20 mm of perfectly transparent halite. With the discrete 25 ps interval steps of the used delay times, the fact that photons will have to cross the mineral layer back and forth, and the presence of inevitable imperfections in the mineral lattice, a measured relative delay of 250 ps can be expected with this sample composition. Measurements of Chroococcidiopsis behind 20.0 mm of halite yielded similar results; the specific Raman bands from the carotenoids at 1151 and 1523 cm-1 are clearly visible in the spectrum (Fig. 5b). Also these signals first appear at 250 ps relative delay, matching those of the Raman signals of D. radiodurans carotenoids. Background fluorescence resulting from the bacteria is also clearly visible in both spectra, starting at the same delay times but stretching over a longer timescale. Time-resolved Raman spectra of both bacterial samples were also recorded behind 11 mm of transparent gypsum (data not shown). The gypsum signals at 1008 and 1141 cm-1 completely overwhelmed the carotenoid signals of both bacteria at the same wavenumber range. However, the carotenoid signals at 1513 cm-1 (D. radiodurans) and 1523 cm-1 (Chroococcidiopsis) were clearly visible. Calcite was used as a proxy for scattering minerals. The maximum thickness through which it was possible to detect the carotenoid Raman signals from both bacterial species was determined to be 5.0 mm. This mineral shows a strong Raman signal at 1086 cm-1, potentially overlapping with the carotenoid Raman bands at 1151 cm-1. This indeed happened in both combinations with D. radiodurans (Fig. 6a) and Chroococcidiopsis (Fig. 6b). However, the bands at 1513 and 1523 cm-1 respectively do not overlap with the calcite band at 1433 cm-1, and these were positively identified as carotenoid signals.

FIG. 5. Time-resolved Raman spectroscopy depth profile measurements of samples with transparent 20 mm halite as first layer and (a) D. radiodurans or (b) Chroococcidiopsis as second layer. As expected, no Raman spectrum of halite was observed, and the carotenoid peaks of the D. radiodurans (1513 cm-1) and Chroococcidiopsis (1523 cm-1) appeared after a relative delay of 250 ps in both cases. All spectra were recorded at a 440 nm excitation wavelength and 70 mW laser power with three accumulations of 20 s. Reference spectra of D. radiodurans (bottom left) and Chroococcidiopsis (bottom right) are also shown. (Color images available at www.liebertonline.com/ast)

RAMAN OF MICROORGANISMS THROUGH MINERALS

7

FIG. 6. Time-resolved Raman spectroscopy depth profile measurements of samples with 5 mm translucent calcite as first layer and (a) D. radiodurans or (b) Chroococcidiopsis as second layer. The Raman spectrum of calcite (1086 cm-1) was observed starting at t0, and the carotenoid peaks of the D. radiodurans (1513 cm-1) and Chroococcidiopsis (1523 cm-1) both appeared after a relative delay of 150 ps (indicated with a dotted ellipse). All spectra were recorded at a 440 nm excitation wavelength and 70 mW laser power with three accumulations of 20 s. Reference spectra of D. radiodurans (in black), Chroococcidiopsis (in green), and calcite (in red) are also shown. (Color graphics available at www.liebertonline.com/ast) When two different mineral layers are stacked, the laser light has to span four material interfaces with different refractive indices and different scattering properties, resulting in a larger laser spot size and a decrease in collected signal. To simulate this, we placed the bacterial samples behind a slice of 5 mm transparent gypsum that was in its turn placed behind a slice of 2.5 mm translucent calcite. Since the sample included calcite, the carotenoid signal at 1151 cm-1 could not be identified because of overlap with the calcite band. However, the D. radiodurans carotenoid signal at 1513 cm-1 was still detectable (Fig. 7a). When Chroococcidiopsis was placed behind the two mineral layers, the weaker RRS signal from

that bacterium (see also Fig. 2) could not be detected (Fig. 7b). This is due to the lower signal-to-baseline ratio of the carotenoid peaks for this bacterium, as is also shown in the reference spectra (see Fig. 2). 4. Discussion

This research showed that detection of microorganisms through transparent and, more important, translucent minerals is possible. We showed that microorganisms could be detected through mineral layers thicker than the focal distance of a confocal Raman microscope. Detecting microorganisms

FIG. 7. Time-resolved Raman spectroscopy depth profile measurements of samples with 2.5 mm translucent calcite as first layer, 5 mm gypsum as second layer and (a) D. radiodurans or (b) Chroococcidiopsis as third layer. The Raman spectrum of calcite (1086 cm-1) started at t0, the Raman spectrum of gypsum (1008 cm-1) after 75 ps, and the carotenoid peaks of D. radiodurans (1513 cm-1) appeared after a relative delay of 175 ps. Chroococcidiopsis (carotenoid Raman peak expected at 1523 cm-1) was not detected behind these mineral layers. All spectra were recorded at a 440 nm excitation wavelength at 70 mW laser power with three accumulations of 20 s. Reference spectra of D. radiodurans (black in 7a), Chroococcidiopsis (green in 7b), calcite (red), and gypsum (blue) are also shown. (Color graphics available at www.liebertonline.com/ast)

8

through minerals inherently means that these microorganisms could also be detected when they are trapped within that mineral. Resonance Raman spectroscopy measurements with the CW Raman setup at an excitation wavelength of 440 nm showed that detection of microorganisms through minerals is possible, but no information about the sample distribution could be obtained from this setup. Due to instrumental configurations, no discrimination could be obtained from the Raman spectra whether the microorganisms were located at the surface or within the sample. Depth discrimination could be made with focusing and defocusing of a confocal Raman microscope. Changing Raman peak ratios upon changing the focus indicates a heterogeneous sample, whereas no change indicates homogeneity. In the case of a multilayered sample, larger changes in Raman peak intensity upon (de)focusing indicate Raman signals originating from surface material; smaller changes upon (de)focusing indicate subsurface signals. Both the CW Raman setup and the Raman microscope use continuous detection. Therefore, a fluorescent background could interfere with detection of low amounts of microorganisms. TRRS provides better depth discrimination than the (de)focusing of the microscope and, in addition, rejects the majority of the fluorescent contribution when using a 250 ps gated detector. Changing the focus of the RRS setup did not yield any improvement in depth discrimination of the sample. From the contour plots shown in Figs. 5–7, additional information can be obtained from the various cross sections. Horizontal cross sections show the TRRS spectra at a specific delay, which corresponds to a specific depth in the sample. This is illustrated in Fig. 8a for the measurement of D. radiodurans behind 5 mm of calcite (cross sections through Fig. 6a). At a delay of 150 ps, a Raman spectrum is observed that predominantly reflects the chemical compo-

VERKAAIK ET AL.

sition of the first layer (calcite). At 500 ps, the Raman spectrum still contains calcite bands, but the relative contribution of the carotenoid signals from the subsurface is larger. Vertical cross sections (see Fig. 8b) show the temporal response for the TRRS signals at 1433 cm-1 (calcite) and at 1513 cm-1 (carotenoid). The rising flank and position of the maximum of the carotenoid signal relative to that of the calcite signal illustrate that depth selectivity can be obtained with TRRS. The TRRS contour plots show a slight increase in delay times for peaks at either side of each Raman spectrum (i.e., at high and low wavenumbers) compared to Raman peaks in the middle of the spectrum. In principle, signals from a specific layer should have comparable delay times. This difference is visible in, for example, Fig. 6a as a slope of the background in the vertical (time delay) direction. Measurements of a 1 mm cyclohexane sample with different spectrograph positions confirmed that signals at both edges of the spectral window appeared some tens of picoseconds later than Raman bands that were more in the middle of the spectrum (data not shown). A similar iris effect had also been observed for the ICCD without a spectrograph (factory test sheet); the edges respond about 40 ps faster than the main area, which is why the arrival of the Raman photons at the sides of the spectral observation range appears delayed. This means that the observed differences in delay times at the sides of the spectrum are not due to the sample itself but are an artifact of the intensifier. In this work, the Raman signals from Chroococcidiopsis were generally lower than those of D. radiodurans, although the bacterium density and the resonance enhancement efficiencies at 440 nm were roughly identical. Possible explanations for this disparity are differences in chemical environment and carotenoid content. Whereas the carotenoids of D. radiodurans are located in the cell membrane to

FIG. 8. (a) Horizontal, spectral profiles and (b) vertical, temporal profiles of TRRS measurements of D. radiodurans behind a layer of 5 mm translucent calcite, derived from the contour plot shown in Fig. 6a. The spectrum observed at 150 ps is more selective for the calcite layer (1086 and 1433 cm-1), whereas the spectrum recorded at 500 ps is more selective for D. radiodurans (1513 cm-1). Frame (b) shows the temporal profiles of the calcite (triangles) and carotenoid (circles) peaks, measured at 1433 and 1513 cm-1, respectively. All spectra were recorded at a 440 nm excitation wavelength and 70 mW laser power with three accumulations of 20 s. A y offset is applied in (a) for legibility.

RAMAN OF MICROORGANISMS THROUGH MINERALS

prevent photodegradation, the carotenoids in the phototrophic Chroococcidiopsis are also embedded in the thylakoid membranes and play a role in chlorophyll quenching (Vermaas et al., 2008). In addition, the difference in Raman signal intensities between the two bacteria might be due to different carotenoid content. Sample heterogeneity can be a limiting factor in Raman measurements through minerals: minerals naturally have inclusions, impurities, crystallographic defects, and fractures that influence the optical path of the light. We have observed large differences between spectra of the same sample, depending on the position of the mineral relative to the laser beam. Shifting the mineral layer 1 mm sideways can result in total loss of the second-layer signals and the appearance of peaks from minor surface impurities (results not shown). TRRS helps determine whether signals arise from superficial layers and hence might be due to fouling of the mineral. Since minerals often show optical anisotropy, detection results may also depend on the orientation of the crystal lattice relative to the laser beam. For example, each optical axis of gypsum has its own refractive index. This effect cancels out when measured through a mixture of randomly orientated grains. In addition, morphology also plays a role; the gypsum used in these experiments shows the typical fishbone structure. When measuring between the ridges, the laser beam is not refracted. When measuring through the ridges, the laser beam becomes increasingly deformed, resulting in a larger cross section of the beam and hence a lower laser power incident at deeper layers. Furthermore, different lattice directions can show different Raman signals or a change in relative Raman intensities. Since this research has an exploratory character, the influence of the crystal lattice orientation was not studied in detail. Real-life situations may differ strongly from the laboratory conditions as applied in this study. Natural rocks are usually covered by dust or soil; are infinitely thick from a spectroscopic point of view; are often composed of more than one specific mineral that, on average, will not have a preferred crystallographic orientation; can have a rough surface; and are exposed to ambient light. In addition, minerals occur in a granular, sandlike size, further decreasing the overall transparency of the rock. All these factors have a negative influence on the possibilities of successfully recording a Raman spectrum from superficial or deeper layers. We have shown that TRRS has a significant value in discriminating between distinctive mineral layers and is able to detect important carotenoid biomarkers with high sensitivity. The possibilities of TRRS mainly seem to depend on the absolute transparency of the mineral, that is, the total thickness, scattering, and absorption properties of the sample. Future technological developments are expected to yield compact Raman spectrometers with a wide range of available laser wavelengths, which enables selective resonance enhancement of any Raman-active analyte on planetary missions. In addition, high-powered lasers can clean dust or soil from rock surfaces by using laser ablation, eventually making the instrument also suitable for laser-induced breakdown spectroscopy (LIBS) (Sharma et al., 2003; Wiens et al., 2014). This might expand the possibilities of a planetary rover on one hand while reducing the size, mass,

9

and power consumption on the other. To prevent photodegradation due to the use of high laser powers, the irradiation spot size can be enlarged. This will also increase the sampling speed. 5. Conclusions

By using RRS with continuous detection, it is possible to detect the carotenoid signals from D. radiodurans and Chroococcidiopsis through minerals. However, 4 mm of transparent mineral was the maximum thickness of the first layer in our setup. This maximum thickness is reduced to 0.5 mm in the case of a highly scattering mineral like spectrolite. The signal-to-baseline of Chroococcidiopsis is generally weaker due to fluorescence background. Detection of this cyanobacterium required slightly thinner first layers. Confocal Raman microscopy proved to be a tool for selective depth detection. Focusing deeper into the sample yielded an improved signal-to-noise ratio for the subsurface Raman signals from the carotenoids in the bacterial layers due to a much smaller contribution of the Raman signals of the surface-layer material. However, the use of defocusing for depth discrimination was limited by the thicknesses of the samples due to the short focal length of the microscope objective and the small lens diameter of the objective. A long-distance objective could help in the case of highly transparent samples but would be of limited use in the case of strong scattering. Additionally, the available 532 nm laser excitation wavelength in the Raman microscope setup shows less resonance enhancement for both bacteria compared to the 440 nm excitation used in the other setups. Time-resolved Raman spectroscopy with 250 ps gated detection provided improved discrimination between two or even three different layers of minerals and bacteria compared to both RS techniques with continuous detection. Raman signals from both D. radiodurans and Chroococcidiopsis are clearly detectable through 20 mm of transparent halite or 5 mm of translucent calcite. Additionally, D. radiodurans was detected through a sample that consisted of 2.5 mm translucent calcite and 5 mm gypsum. The Raman signal from each of the three different layers arriving at a different relative delay time showed depth resolution at the millimeter scale. In addition, TRRS reduces the fluorescent background that originates from the sample. From an astrobiological perspective, TRRS shows great potential for the detection of microorganisms. The physical size and power consumption of the described system is currently too large for installation on a planetary rover. Further technological developments, however, are expected to overcome those problems, and a time-gated Raman spectrometer (albeit with poorer time resolution) is already scheduled to be launched in 2020. NASA’s SuperCam system can be gated at 100 ns timescales (Wiens et al., 2014). This system will be helpful to distinguish Raman signals from long-lived mineral luminescence and ambient light but would probably be too slow for depth resolution. TRRS at picosecond timescales offers the possibility for non-invasive detection of microorganisms through or inside minerals with millimeter resolution and could be combined with other spectroscopic techniques, like LIBS. In addition, it offers the possibility to reduce the fluorescence contribution to Raman spectra.

10 Acknowledgments

We would like to express our appreciation to our colleagues of the mineral workshop of the VU Amsterdam for cutting the minerals for our experiments, to Wilfred Ro¨ling of the Department of Molecular Cell Physiology of the VU Amsterdam for providing us with a strain of wild-type Deinococcus radiodurans, to Gerard Muyzer of Microbial Systems Ecology at the University of Amsterdam for providing the Chroococcidiopsis from the Culture Collection Yerseke, and to Joost B. Buijs for technical support and developing the TRRS data collection software. We also thank NWO—Netherlands Space Office for funding this project ALW-GO-PL/14, NWO for a contribution to the Raman microscope (BAZIS grant # 10.017.) and for the OPO system funded by NWO Chemical Sciences, Mid-size Instrumentation grant # 700.59.103. Author Disclosure Statement

The authors declare that no competing financial interests exist. References

Angel, S.M., Gomer, N.R., Sharma, S.K., and McKay, C. (2012) Remote Raman spectroscopy for planetary exploration: a review. Appl Spectrosc 66:137–150. Ariese, F., Meuzelaar, H., Kerssens, M.M., Buijs, J.B., and Gooijer, C. (2009) Picosecond Raman spectroscopy with a fast intensified CCD camera for depth analysis of diffusely scattering media. Analyst 134:1192–1197. Babin, F., Hoˆ, N., Paradis, P.F., Deblois, S., and Chaˆteauneuf, F. (2011) Stand-off UV laser induced fluorescence and UV enhanced Raman spectroscopy for mineral analysis [abstract 1800]. In 42nd Lunar and Planetary Science Conference, Lunar and Planetary Institute, Houston. Beegle, L.W., Bhartia, R., DeFlores, L., Darrach, M., Kidd, R.D., Abbey, W., Asher, S., Burton, A., Clegg, S., Conrad, P.G., Edgett, K., Ehlmann, B., Langenhorst, F., Fries, M., Hug, W., Nealson, K., Popp, J., Sobron, P., Steele, A., Wiens, R., and Williford, K. (2014) SHERLOC: Scanning Habitable Environments with Raman and Luminescence for Organics and Chemicals, an investigation for 2020 [abstract 5101]. In 11th International GeoRaman Conference, Lunar and Planetary Institute, Houston. Benison, K.C. and Karmanocky, F.J. (2014) Could microorganisms be preserved in Mars gypsum? Insights from terrestrial examples. Geology 42:615–618. Benison, K.C., Jagniecki, E.A., Edwards, T.B., Mormile, M.R., and Storrie-Lombardi, M.C. (2008) ‘‘Hairy blobs’’: microbial suspects preserved in modern and ancient extremely acid lake evaporites. Astrobiology 8:807–821. Blacksberg, J., Rossman, G.R., and Gleckler, A. (2010) Timeresolved Raman spectroscopy for in situ planetary mineralogy. Appl Opt 49:4951–4962. Bo¨ttger, U., Meessen, J., Martinez-Frias, J., Hu¨bers, H.W., Rull, F., Sa´nchez, F.J., de la Torre, R., and de Vera, J.P. (2013) Raman spectroscopic analysis of the calcium oxalate producing extremotolerant lichen Circinaria gyrosa. International Journal of Astrobiology 13:19–27. Buckley, K. and Matousek, P. (2011) Non-invasive analysis of turbid samples using deep Raman spectroscopy. Analyst 136: 3039–3050.

VERKAAIK ET AL.

Conner, A.J. and Benison, K.C. (2013) Acidophilic halophilic microorganisms in fluid inclusions in halite from Lake Magic, Western Australia. Astrobiology 13:850–860. Edwards, H.G.M. (2010) Raman spectroscopic approach to analytical astrobiology: the detection of key geological and biomolecular markers in the search for life. Philos Trans A Math Phys Eng Sci 368:3059–3065. Edwards, H.G.M., Russell, N.C., and Wynn-William, D.D. (1997) Fourier transform Raman spectroscopic and scanning electron microscopic study of cryptoendolithic lichens from Antarctica. J Raman Spectrosc 28:685–690. Edwards, H.G.M., Vandenabeele, P., Jorge Villar, S.E., Carter, E.A., Perez, F.R., and Hargreaves, M.D. (2007) The Rı´o Tinto Mars analogue site: an extremophilic Raman spectroscopic study. Spectrochim Acta A Mol Biomol Spectrosc 68: 1133–1137. Edwards, H.G.M., Sadooni, F., Vı´tek, P., and Jehlicka, J. (2010) Raman spectroscopy of the Dukhan sabkha: identification of geological and biogeological molecules in an extreme environment. Philos Trans A Math Phys Eng Sci 368:3099–3107. Edwards, H.G.M., Hutchinson, I.B., and Ingley, R. (2012) Raman spectroscopy and the search for life signatures in the ExoMars Mission. International Journal of Astrobiology 11: 269–278. Edwards, H.G.M., Hutchinson, I.B., Ingley, R., and Jehlicka, J. (2014) Biomarkers and their Raman spectroscopic signatures: a spectral challenge for analytical astrobiology. Philos Trans A Math Phys Eng Sci 372, doi:10.1098/rsta.2014.0193. Efremov, E.V., Ariese, F., and Gooijer, C. (2008) Achievements in resonance Raman spectroscopy; review of a technique with a distinct analytical chemistry potential. Anal Chim Acta 606: 119–134. Eliasson, C., Claybourn, M., and Matousek, P. (2007) Deep subsurface Raman spectroscopy of turbid media by a defocused collection system. Appl Spectrosc 61:1123–1127. Erokhina, L.G., Shatilovich, A.V., Kaminskaia, O.P., and Gilichinskii, D.A. (2002) The absorption and fluorescence spectra of the cyanobacterial phycobilins of cryptoendolithic lichens in the high-polar regions of Antarctica. [in Russian] Mikrobiologiia 71:697–704. Gaft, M. and Nagli, L. (2009) Gated Raman spectroscopy: potential for fundamental and applied mineralogy. European Journal of Mineralogy 21:33–42. Harrison, J.P., Gheeraert, N., Tsigelnitskiy, D., and Cockell, C.S. (2013) The limits for life under multiple extremes. Trends Microbiol 21:204–212. Hooijschuur, J.H., Iping Petterson, I.E., Davies, G.R., Gooijer, C., and Ariese, F. (2013) Time resolved Raman spectroscopy for depth analysis of multi-layered mineral samples. J Raman Spectrosc 44:1540–1547. Hooijschuur, J.H., Verkaaik, M.F.C., Davies, G.R., and Ariese, F. (2015a) Raman spectroscopy for future planetary exploration: photodegradation, self-absorption and quantification of carotenoids in microorganisms and mineral matrices. J Raman Spectrosc, online ahead of print, doi:10.1002/jrs.4647. Hooijschuur, J.H., Verkaaik, M.F.C., Davies, G.R., and Ariese, F. (2015b) Will Raman meet bacteria on Mars? An overview of the optimal Raman spectroscopic techniques for carotenoid biomarkers detection on mineral backgrounds. Netherlands Journal of Geosciences, online ahead of print, doi:10.1017/ njg.2015.3. Iping Petterson, I.E., Lo´pez-Lo´pez, M., Garcı´a-Ruiz, C., Gooijer, C., Buijs, J.B., and Ariese, F. (2011) Noninvasive detection of concealed explosives: depth profiling through

RAMAN OF MICROORGANISMS THROUGH MINERALS

opaque plastics by time-resolved Raman spectroscopy. Anal Chem 83:8517–8523. Iping Petterson, I.E., Esmonde-White, F.W., de Wilde, W., Morris, M.D., and Ariese, F. (2015) Tissue phantoms to compare spatial and temporal offset modes of deep Raman spectroscopy. Analyst 140:2504–2512. Jehlicka, J., Edwards, H.G., Osterrothova, K., Novotna, J., Nedbalova, L., Kopecky, J., Nemec, I., and Oren, A. (2014) Potential and limits of Raman spectroscopy for carotenoid detection in microorganisms: implications for astrobiology. Philos Trans R Soc Lond B Biol Sci 372:1–17. Jorge Villar, S.E. and Edwards, H.G.M. (2006) Raman spectroscopy in astrobiology. Anal Bioanal Chem 384:100–113. Jorge Villar, S.E., Edwards, H.G., and Cockell, C.S. (2005) Raman spectroscopy of endoliths from Antarctic cold desert environments. Analyst 130:156–162. Lowenstein, T.K., Schubert, B.A., and Timofeeff, M.N. (2011) Microbial communities in fluid inclusions and long-term survival in halite. GSA Today 21:4–9. Marshall, C.P., Leuko, S., Coyle, C.M., Walter, M.R., Burns, B.P., and Neilan, B.A. (2007) Carotenoid analysis of halophilic archaea by resonance Raman spectroscopy. Astrobiology 7:631–643. Matousek, P., Everall, N., Towrie, M., and Parker, A.W. (2005) Depth profiling in diffusely scattering media using Raman spectroscopy and picosecond Kerr gating. Appl Spectrosc 59:200–205. McKay, C.P. (2011) The search for life in our Solar System and the implications for science and society. Philos Trans A Math Phys Eng Sci 369:594–606. Misra, A.K., Sharma, S.K., Chio, C.H., Lucey, P.G., and Lienert, B. (2005) Pulsed remote Raman system for daytime measurements of mineral spectra. Spectrochim Acta A Mol Biomol Spectrosc 61:2281–2287. Osterrothova, K. and Jehlicka, J. (2011) Feasibility of Raman microspectroscopic identification of biomarkers through gypsum crystals. Spectrochim Acta A Mol Biomol Spectrosc 80:96–101. Robert, B. (2009) Resonance Raman spectroscopy. Photosynth Res 101:147–155. Sharma, S.K., Lucey, P.G., Ghosh, M., Hubble, H.W., and Horton, K.A. (2003) Stand-off Raman spectroscopic detection of minerals on planetary surfaces. Spectrochim Acta A Mol Biomol Spectrosc 59:2391–2407. Skulinova, M., Lefebvre, C., Sobron, P., Eshelman, E., Daly, M., Gravel, J.F., Cormier, J.F., Chaˆteauneuf, F., Slater, G., Zheng, W., Koujelev, A., and Le´veille´, R. (2014) Timeresolved stand-off UV-Raman spectroscopy for planetary exploration. Planet Space Sci 92:88–100. Soleimaninejad, H., Matroodi, F., and Tavassoli, S. (2012) Raman spectroscopy of Iranian region calcite using pulsed laser: an approach of fluorescence suppression by time-gating method. Journal of Spectroscopy 2013, doi:10.1155/2013/ 254964.

11

Vermaas, W.F.J., Timlin, J.A., Jones, H.D.T., Sinclair, M.B., Nieman, L.T., Hamad, S.W., Melgaard, D.K., and Haaland, D.M. (2008) In vivo hyperspectral confocal fluorescence imaging to determine pigment localization and distribution in cyanobacterial cells. Proc Natl Acad Sci USA 105:4050–4055. Vı´tek, P., Jehlicka, J., Ascaso, C., Masek, V., Gomez-Silva, B., Olivares, H., and Wierzchos, J. (2014) Distribution of scytonemin in endolithic microbial communities from halite crusts in the hyperarid zone of the Atacama Desert, Chile. FEMS Microbiol Ecol 90:351–366. Wang, A., Jolliff, B.L., and Haskin, L.A. (1995) Raman spectroscopy as a method for mineral identification on lunar robotic exploration missions. J Geophys Res 100:21189–21199. Wiens, R.C., Maurice, S., Johnson, J.R., Clegg, S.M., Sharma, S.K., Rull, F., Montmessin, F., Anderson, R.B., Beyssac, O., Bonal, L., Deflores, L., Dromart, G., Fischer, W., Forni, O., Gasnault, O., Grotzinger, J., Mangold, N., Martinez-Frias, J., McLennan, S., McCabe, K., Cais, P., Nelson, T.E., Angel, S.M., Beck, P., Benzerara, K., Bernard, S., Bousquet, B., Bridges, N., Cloutis, E., Fabre, C., Fouchet, T., Grasset, O., Lanza, N., Lasue, J., Le Mouelic, S., Leveille, R., Lewin, E., McConnochie, T., Melikechi, N., Meslin, P.-Y., Misra, A., Montagnac, G., Newsom, H., Ollila, A., Pinet, P., Poulet, F., Sautter, V., and Sobron, P. (2014) The SuperCam remote sensing suite for Mars 2020: co-aligned LIBS, Raman, and near-IR spectroscopies, and color micro-imaging [abstract 1086]. In International Workshop on Instrumentation for Planetary Missions (IPM-2014), NASA Goddard Space Flight Center, Greenbelt, MD.

Address correspondence to: Freek Ariese LaserLaB Faculty of Sciences VU University Amsterdam De Boelelaan 1083 1081 HV Amsterdam The Netherlands E-mail: [email protected] Submitted 30 March 2015 Accepted 15 June 2015 Abbreviations Used CW ¼ continuous-wave ICCD ¼ intensified charge-coupled device LIBS ¼ laser-induced breakdown spectroscopy MCP ¼ microchannel plate RRS ¼ resonance Raman spectroscopy RS ¼ Raman spectroscopy TRRS ¼ time-resolved Raman spectroscopy

Raman Spectroscopic Techniques for Planetary Exploration: Detecting Microorganisms through Minerals.

Raman spectroscopy can provide highly specific chemical fingerprints of inorganic and organic materials and is therefore expected to play a significan...
1MB Sizes 4 Downloads 21 Views