A Historical Perspective of the Development of the CheMin Mineralogical Instrument for the Mars Science Laboratory Mission

by David Blake, NASA Ames Research Center

Introduction by Paul Mahaffy (NASA Goddard Space Flight Center)
Volumes of multispectral infrared imaging data presently flowing in from Mars orbiting spacecraft are giving us a new view of the planet (Ehlman, Geochemical News 142) and pointing toward candidate landing sites for surface rovers. Highly ruggedized and miniaturized instruments on future rovers will carry out an even more detailed exploration of the chemistry and mineralogy at the most interesting sites to elucidate geological and geochemical processes that may point toward habitable environments for past or present life. One such instrument planned for use on the Curiosity rover that is planned to land on Mars in 2012 is the x-ray fluorescence/x-ray diffraction instrument CheMin described in this contribution from the Principle Investigator for this investigation, David Blake. Some of the robust field-testing of this instrument on remote Mars analog sites in the Arctic Svalbard archipelago is also described.

In our solar system, Mercury, Venus, Earth and Mars are rocky planets; Jupiter, Saturn, Uranus and Neptune are gas giants with no solid surfaces, and Pluto (now, sadly, demoted to ‘planetesimal' status) is icy and cryogenically cold. Early in Solar System history, surface conditions on Mars and Venus diverged from what could have been a ‘habitable' status. Venus is not presently habitable due to its high surface temperature and inhospitable atmospheric composition. Furthermore, present day surface temperatures and atmospheric conditions on Venus would have likely destroyed any evidence of earlier habitable conditions. Mars cooled off and lost most of its atmosphere, but still preserves geological evidence of its early history - a time when its climate was much more clement than the present day. With the exception of ‘special' zones such as the proposed subsurface ocean of Jupiter's moon Europa or the organic-rich surface of Saturn's moon Titan, ‘habitable zones' are more or less confined to the surface and near-surface environments of Earth and Mars. For the foreseeable future then, Mars seems to be the most promising place in the solar system to look for evidence of past or present extraterrestrial life.

An important difference between Mars and Earth is the apparent absence of extensive plate tectonics on Mars. On Earth, as a result of plate tectonic movements, most surface crustal components older than ~3 billion years have been subsumed by convergent plate margins or metamorphosed beyond recognition of origin or both. Truly ancient crust that existed in the habitable zone when early terrestrial life developed and radiated can only be found in highly deformed slivers of rocks in a few places in the world. By contrast, much of the Mars surface from that early epoch has been preserved relatively unaltered. Thus, there is the possibility on Mars to study relatively unaltered crust in which habitable zones were preserved at the time when terrestrial life originated and radiated.

Our recent knowledge of early geology and potentially habitable environments on Mars is derived from decades of study of Mars meteorites such as ALH 84001, from the orbital observations of Mars Observer, Mars Reconnaissance Orbiter and the European Space Agency's Mars Express Orbiter, from Mars Pathfinder and from the Mars Exploration Rovers. We hope to continue the rich heritage of Mars science during the upcoming Mars Science Laboratory Mission.

The Mars Science Laboratory Mission

Mars Science Laboratory (MSL '11) is NASA's follow-on landed mission to the Mars Exploration Rovers Spirit and Opportunity, which, more than 6 years after their intended 90-day lifetimes, are still making fundamental scientific discoveries on the Mars surface. MSL is intended to ‘...Explore and quantitatively assess a local region on the Mars surface as a potential habitat for life, past or present.' The duration of the primary mission of MSL will be 670 sols, or one Mars year (about two Earth years). During this time, the plan is for the MSL rover (named ‘Curiosity') to traverse to at least 3 geologically distinct sites within its 20 km diameter landing ellipse and determine the ‘habitability' of these sites (habitability is defined in this context as the ‘capacity of the environment, past or present, to sustain life').

The CheMin instrument will be principally engaged in the following MSL science objectives:

  • Characterizing the geology and geochemistry of the landed region;
  • Investigating the chemical and mineralogical composition of Martian surface and near-surface geological materials;
  • Interpreting the processes that have formed and modified rocks and regolith.

MSL carries four types of instruments: ‘Environmental instruments' such as the Radiation Assessment Detector (‘RAD') and the Rover Environmental Monitoring Station (‘REMS'), ‘remote observation' instruments such as the Mast Camera (‘Mastcam'), the Mars Descent Imager (‘MARDI') the Chemistry-and-Camera (‘ChemCam') and the Dynamic Albedo of Neutrons (‘DAN'), and ‘arm instruments' including the Alpha Particle X-ray Spectrometer (‘APXS') and the Mars Hand Lens Imager (‘MAHLI'). Inside the body of the rover is the ‘Analytical Laboratory' which is comprised of the Chemistry and Mineralogy (‘CheMin') instrument and the Sample Analysis at Mars (‘SAM') instrument suite. Each of these instruments, as well as the mission itself, is described in detail on the MSL mission website.1

Together, the environmental, remote and arm instruments of Curiosity operate much like a field geologist would on Earth - identifying interesting rocks or formations from a distance (even obtaining an elemental analysis from up to 10 meters using ChemCam), then approaching them to make in situ hand lens observations and elemental analyses. These operations and observations are similar to those performed by Spirit and Opportunity for the past 6 years. However, once these operations are performed, Curiosity goes further. Curiosity can collect, prepare and analyze samples in much the same way that this is done in a terrestrial laboratory. When a sample is found to be interesting by scientists on Earth, Curiosity will deploy an arm-mounted percussion drill called ‘PADS' (Powder Acquisition Drill System) in concert with a sample processing system called ‘CHIMRA' (Collection and Handling for Interior Martian Rock Analysis) to collect material for further analysis. Drilled or scooped material is then sieved and a portion of this sieved material is transferred to the Analytical Laboratory. Curiosity's Analytical Laboratory can conduct quantitative mineralogical analysis (CheMin) and organic and isotopic analysis (SAM) of material delivered through funnels into the body of the rover. Once a complete analysis is obtained by the Analytical Laboratory, results are sent back to Earth for interpretation.

How does CheMin work?

CheMin determines the mineralogy of crushed or powdered samples through X-ray diffraction (XRD) and elemental composition through X-ray Fluorescence (XRF). XRD is an extraordinarily powerful technique that identifies the structures of crystalline materials from first principles, and is the preferred method for mineralogical analysis in terrestrial laboratories.

During a mineralogical analysis, a ~50 µm collimated X-ray beam from an X-ray tube source is directed through powdered or crushed sample material. An X-ray sensitive CCD imager on the opposite side of the sample from the source directly detects X-rays diffracted or fluoresced by the sample (Figure 1a) < Details of the CCD operation are shown in Apendix A>.

During an X-ray Diffraction analysis, CheMin's CCD detector is exposed to the X-ray flux, read out and erased many times (100-1000 exposures). The detected X-rays are used to produce both diffraction and fluorescence data. Diffracted primary beam X-rays strike the detector and are identified by their energy. A two-dimensional image of these X-rays constitutes the diffraction pattern (Fig. 1a). At incremental radii this pattern is summed circumferentially about the central undiffracted beam to yield a 1- dimensional 2θ plot comparable to conventional diffractometer data (Fig. 1b). All of the X-rays detected by the CCD are summed into a histogram of number of photons vs. photon energy that constitutes an XRF analysis of the sample (Fig. 1c).

The requirement for CheMin to perform X-ray Fluorescence (XRF) elemental analysis was removed early in the spacecraft instrument design phase. The energy-dispersive X-ray histograms obtained during an analysis will only be used to construct energy-selected diffraction patterns, such as CoKα. When this is done, sample fluorescence, multiple photon detections in a single pixel, tracks from cosmic rays, etc. can be removed from the patterns. While the requirement to perform XRF elemental analyses was descoped by the MSL project, these data are still obtained by the instrument and will be analyzed on a ‘best effort' basis.

Quantitative mineralogical results are obtained from XRD data by Rietveld refinement and other full-pattern fitting techniques. 2,3 Both crystalline and amorphous materials can be analyzed in this way. The duration of a single XRD experiment sufficient to quantitatively analyze a single mineral such as quartz or olivine, is less than one hour and consumes a few tens of Watts. Complex assemblages such as basalts having 4 or more minerals plus glass may require the summation of 4-8 of these experiments over 1 or more Mars sols.

Prototypes and Terrestrial Analogs of the CheMin Instrument

During the development of the CheMin instrument concept, several prototypes were built which had increasing fidelity to what would become the flight instrument. The CheMin I instrument consisted of a commercial CCD camera in an evacuated housing with rudimentary sample handling mechanisms, interfaced to a commercial X-ray tube tower (Fig. 2a). An evacuated housing was necessary because the CCD was cooled to -60 C during operation. The electronics and power supplies for this machine - and the X-ray source - occupied a small room. CheMin II was an improvement relative to CheMin I in that a self-contained X-ray tube and power supply was interfaced to a sealed and evacuated commercial CCD camera (Fig. 2b). A thin beryllium window allowed X-rays to enter the camera and expose the CCD. Because the samples were analyzed in room air, we could experiment more freely with sample types and sample delivery / sample movement mechanisms.

A breakthrough in sample handling occurred between CheMin II and III. For a sample to be optimally prepared for powder XRD, it must be ground to somewhat less than 10 µm particle size, a difficult task even in terrestrial laboratories (this is basically the same grain size as flour or face powder). The reason for this is that in order to get a pattern with all diffraction peaks present at their correct intensities, random orientations of myriads of tiny crystallites are required. We had experimented with finely ground pressed powders translated with an x,y stage, but nothing seemed to work particularly well. CheMin Co-I Philippe Sarrazin proposed a sample holder in which loose powder is held between two X-ray transparent plastic windows spaced ~200 µms apart. By vibrating the sample holder at sonic frequencies with a piezoelectric device, (we first used modified buzzers from Radio Shack) the powder was made to flow like a liquid inside the cell. This solved our two most difficult sample handling problems - exposing a representative amount of the sample to the 50 µm diameter X-ray beam during the analysis, and rotating the grains to all orientations as they passed through the beam. We found that powders ground and sieved to less than 150 µm diameter produced excellent patterns. Rock powder produced by a variety of techniques (drilling, grinding, crushing) has a significant fraction in this size range and below.

CheMin III was our first field demonstration model, and consisted of the basic components of CheMin (X-ray tube, Sample handling system and CCD camera) held in a portable frame along with control electronics and power supplies (Fig. 2c). The instrument was powered through a cable connected to several motorcycle batteries, and operated through a laptop computer. CheMin III was demonstrated in Death Valley, CA in 2003. However, the entire system still occupied the trunk of a car and had limited durability at remote localities. CheMin IV was the first truly portable CheMin system. While it appears much like the earlier model, it was operated by an integrated microcomputer and powered by on-board Li-ion batteries and a power management system. Data were transferred to a laptop computer using a USB thumb drive, and processed off-line. CheMin IV was used to obtain the first remote quantitative mineralogical analysis by David Bish of Indiana University during a field expedition to Svalbard, Norway, in 2006 (Fig. 2d).

The CheMin Flight Instrument

Building a successful flight instrument is a task that is breathtakingly hard, and best left to professionals. MSL instruments are required to operate for a full Martian year on the Mars surface, in temperatures that vary from -70 C to +50 C (instruments outside the body of the rover are required to survive temperatures as low as -130 C). They must survive intense vibration during launch, transit to Mars in vacuum and ultimately operate under varying thermal and atmospheric conditions, as well as survive the intense neutron flux from the DAN instrument and from Curiosity's Radioisotope Thermoelectric Generator (RTG). Every fastener and component is tested under the full range of conditions and spares of critical components are ‘life-tested' for 1.5 times the nominal mission duration. The CheMin flight instrument (also called the ‘Flight Model,' or ‘FM') was built by scientists, engineers and technicians at NASA's Jet Propulsion Laboratory during 2005-2009, and delivered to the MSL project in June, 2010. A CheMin ‘Demonstration Model', or ‘DM' is also being constructed that is an exact copy of the FM, and will be used for testing Mars analog samples, as well as for studying potential instrument anomalies during the MSL mission. '

An exploded view of the spacecraft instrument is shown in Fig 3. The CheMin sample handling system consists of a funnel, a sample wheel (which carries 27 reusable sample cells and 5 permanent reference standards), and a sample sump where material is dumped after analysis (Fig. 4). CheMin receives sieved drill powders or scoop samples from PADS/CHIMRA (Details of sample delivery and processing are described in Appendix B).

A full analysis of an individual sample is called a ‘major frame' and will require as many as 10 hours of analysis time, accumulated over multiple sols. Once a major frame of data is sent to ground and accepted, the analyzed material is emptied from the cell and that cell is ready to be reused. CheMin does not have the capability to store previously analyzed samples for later re-analysis (Details of CheMin sample handling and analysis are described in Appendix C).

X­-ray Diffraction Mode

The CCD is placed in the forward-scattered direction relative to the X-ray beam so that mineral phases with large interplanar spacings (and hence narrow diffraction cones at low 2θ), such as clays, can be detected. In addition, low-index lines (which are commonly the most intense and most definitive for phase identification) occur in the forward-scattered direction. Table 1 shows the expected 2θ range (for Co Kα radiation) and 2θ Full Width at Half-Maximum (FWHM ) for X-ray diffraction.

A special case of X-ray detection by the CCD is the detection of Co Kα characteristic photons from the primary source. When Co Kα photons are detected, the X,Y pixel location on the CCD is identified and the corresponding X,Y location in a 600x582 counting number array is incremented by one. This process results in a Co Kα diffraction image. Various strategies are used in on-board data processing to optimize the quality or quantity of diffraction data returned (e.g., ‘single pixel' detection, and ‘split pixel' detection, etc.

An additional 600x582 array stores an image of all of the photons detected by the CCD regardless of energy. This array acts very much like a piece of photographic film, recording the Co Kα XRD pattern as well as background, X-ray Fluorescence from the sample, and Bremsstrahlung radiation from the X-ray source.

XRD/XRF Calibration and Characterization

Five permanent cells are loaded with calibration standards (Fig. 4). Three of these cells are loaded with single minerals or a synthetic ceramic and two are loaded with differing quartz/beryl mixtures. Basic calibration, completed prior to delivery of the instrument to MSL Assembly, Test, and Launch Operations (ATLO), was performed using only the five permanent standards loaded into the sample cells of the FM.

Calibration of these standards entails measurement of 2θ range and 2θ FWHM for XRD, and of the required XRF energy range and FWHM for elemental peaks, in particular Fe Kα, Co Kα and Co Kβ. Quantitative accuracy, precision and detection limits are evaluated using the quartz-beryl standards (CheMin's requirements for detection limit, accuracy and precision of analyses are shown in Table 2). Figures 6-7 show data from Cryo-Vac tests of the flight instrument. Fig. 6a-c show single frame, minor frame and major frame data from the amphibole standard. Fig. 6d shows an energy dispersive X-ray histogram for a single frame, and Fig. 6e shows the energy dispersive X-ray histogram for all of the single frames summed into the major frame. Figure 6f shows a 1-D diffractogram obtained for a major frame (full analysis) of the amphibole. Figure 7 illustrates the ‘best case' X-ray Fluorescence capability of the instrument. These data were obtained by analyzing the synthetic ceramic standard, prepared to evaluate the XRF capability of the instrument.

Quantitative XRD calibration of the Development Model (DM) and other CheMin Testbeds

Quantitative XRD (QXRD) calibration will be performed using the DM and various other testbeds. For QXRD calibration, synthetic mixtures that mimic real samples likely to be encountered on Mars have been prepared from minerals mixed in known weight fractions, or natural materials of known composition.

For characterization of CheMin operation across a broad spectrum of samples, synthetic and natural, the DM will be supported by testbeds and facilities that replicate various parts of the DM/FM function with varying levels of fidelity:

The Development Model: The DM will be set up in a testbed configuration at JPL in CheMin Co-I Albert Yen's laboratory. Prior to launch, the DM will be used to test algorithms, establish calibrations, develop operation scenarios, and characterize Mars analog samples. During landed operations the DM unit will be used to test new command sequences, develop operational scenarios, characterize Mars analog samples and reproduce any instrument anomalies that might occur during the MSL mission.

Analytical facility for Mars analog rocks: The Planetary Mineralogy and Spectroscopy Laboratory at NASA Ames Research Center (ARC) will house several CheMin analog instruments. The principal instruments in this laboratory are a CheMin IV instrument4 and a Terra instrument5 (a field-deployable instrument based on the CheMin design that was developed by InXitu, Inc). These instruments will be used to analyze Mars analog rocks in a geometry similar to the CheMin FM and the DM instruments. A commercial InelTM X-ray diffractometer at Ames Research Center is configured to analyze Mars analog rocks in a geometry nearly identical to the CheMin flight instrument. This instrument is equipped with a Co X-ray tube and a 120-degree parallel detection system capable of collecting XRD patterns with a higher 2θ resolution than the spacecraft instrument (but which can be degraded to MSL CheMin resolution for comparison and pattern matching). A Mars atmospheric pressure chamber is installed with a carousel and MSL funnel, and a CheMin transmission sample cell capable of being filled, piezo-electrically shaken during analysis and dumped. A large number of patterns of Mars analog rocks and soil are being collected for analysis prior to, during and after the prime MSL mission. The CheMin IV and Terra instruments have the resolution and diffraction geometry of the MSL Flight instrument and will be used in supporting tests.

CheMin Instrument Modes

During a nominal 10-hour analysis, CheMin collects and stores X-ray data as individual 600x582 pixel CCD images of 5-30 seconds exposure each. A ‘minor frame' consists of 30 minutes of these images, nominally 360-60 frames depending on integration time. A complete 10-hour analysis of a sample comprises 20 such minor frames and is called a ‘major frame.'

There is insufficient bandwidth to deliver all of CheMin's raw data to Earth. When commanded, CheMin will deliver raw data to the Rover Compute Element (RCE) that in turn partially processes the raw data for each minor frame, in order to reduce the data volume. Each minor frame of data transmitted to Earth contains one or more raw frames in order to assess the health of the detector, a variety of engineering and health information about the instrument, and one or more of three possible processed data products. The three types of data products are described below:

  • ‘Fully processed mode': Each image is reduced to a pixel map containing ones and zeros, where ‘1' represents the detection of a photon within a specific energy window (e.g., Co Kα), and ‘0' represents everything else. Each pixel map is summed into a 600x582 counting number array of pixel positions; the result is a 2-D energy-filtered diffraction pattern. In addition to the energy-filtered diffraction pattern, ‘fully processed mode' also provides a histogram of all of the photons detected vs. energy, which amounts to an X-ray energy-dispersive spectrum of the sample material.
  • ‘Film mode': Each image is summed into a 600x582 array as raw data. A single real number array holds the summed image for each minor frame.
  • ‘Modified raw mode': Pixels below a selected threshold are set to zero, and pixels that are above that threshold are run-length encoded with x, y, and intensity information preserved.

Analysis of diffraction data in the tactical time frame: Immediately after a downlink of CheMin data, the downlink lead will process the minor frames to create 1-D 2θ plots. These 1-D plots or ‘diffractograms' will be analyzed and compared with the ICDD (International Centre for Diffraction Data) PDF-2 powder diffraction file and the AMCSD (American Mineralogist Crystal Structure Database) to determine major mineral components. CheMin Science Team members will offer preliminary identifications of any major or clearly discernable mineral components during this tactical cycle. Periodically these data will support a ‘drive away' or ‘stay' decision for rover operations.

Analysis and refinement of diffraction data in the strategic time frame. Rietveld computational refinement methods and full pattern fitting, among others, will be utilized to perform a quantitative analysis of each pattern. These patterns will be compared with library patterns to identify mineral components and to derive quantitative mineral abundances. Analyses will be updated on a continuous basis as insights are made as to the identity of major and minor phases, and X-ray amorphous materials.

Mars on Earth - Utilization of CheMin Prototype Instruments during the AMASE (Arctic Mars Analog Svalbard Expeditions) to Svalblard, Norway

The first deployment of CheMin IV to Svalbard was thought provoking for us, because while we had approached it principally as a technology demonstration, we discovered quickly that in situ mineralogical analyses were actually useful for the geologists conducting fieldwork. The change from the usual procedure of first conducting field work and later performing laboratory mineralogical analysis suffers for the fact that hypotheses generated in the field can typically only be evaluated after the geologists have left the field area. Geologists commonly collect a comprehensive set of samples from the field and make the assumption that the mineralogical information necessary to support or modify an already formed hypothesis is in hand. With in situ mineralogical analyses, field geologists are able to test hypotheses while in the field, and alter their collection/analysis strategy (or even modify their working hypotheses) based on the field data. However, in order for CheMin IV to be truly useful, user-friendliness, portability and pattern acquisition speed had to be improved. The next generation system was called mini-CheMin, and was made smaller and more portable, and contained all the hardware and software necessary to provide diffractograms as data were collected (Fig. 8). Diffraction data are transmitted wirelessly to a laptop system and analyzed using commercial programs like MDITM Jade or XpowderTM. MiniCheMin proved to be highly useful, and a commercial version called ‘Terra' is now offered by a company called inXitu, Inc. With increased tube power and optimized geometry, XRD patterns of complex mineralogies can be acquired in 5-15 minutes, and single minerals such as quartz can be identified in as little as 20 seconds. With the increase in data collection speed, field-deployed XRD is now possible on the same time scale as other field-portable techniques such as XRF Spectrometry, laser-Raman and IR imaging. The advantage of XRD over these other techniques is that it is a true phase identification technique, and it is possible to quantify mineralogical results.

When a rover is deployed on Mars, time is money. If one divides the full cost of a Mars rover by the number of sols (days) in its active mission, the result is typically millions of dollars per sol. Decisions to move the rover, collect a sample or deploy one instrument vs. another are made on a time-critical, tactical basis. Because data are relayed to an orbiting Mars satellite only when the rover is in sight of the orbiter, then downlinked to the Deep Space Network (DSN) on Earth only when the orbiter is in sight of Earth, just a few hours are available for the Science Operations Working Group (SOWG) to analyze the data before the next command sequence has to be uplinked to the rover. During this time, the SOWG must converge on an interpretation of the downlinked data, come to a consensus as to the next day's operations, then code and relay those instructions through the DSN to the Mars orbiting satellite and down to the rover. Each day's activities are proscribed by the total energy and time made available to the instrument suite, the energy and time required for data acquisition for each individual instrument, the priority of desired measurements, and the total data volume that can be transmitted back to Earth. All of this means that instrument deployment and data interpretation, as well as the sequence in which particular measurements are made, require a thorough knowledge of instrumental capabilities, limitations and instrument synergisms one to another.

Many instrument providers have not had prior experience in instrument operations on Mars (the author included). So that potential instrument providers can become proficient in these activities, NASA initiated a program called the Astrobiology Science and Technology for Exploring Planets (ASTEP). Typically, ASTEP deployments involve instrument operations in remote areas of Earth, on Mars-like (or other extraterrestrial) terrains. Two types of activities are conducted: In one type of activity, a scientific theme is pursued that has relevance to future missions: Habitability, Life Detection or Sample Return among them. A variety of instruments is utilized in the field to collect data useful in solving the multidisciplinary science or engineering problem that is posed. In a second type of activity, called a ‘Fast-Motion' field test (‘FMFT'), a suite of Mars analog instruments is deployed in the field by one group of scientists (‘the rover team') while a second group of scientists (the Science Operations Working Group, or ‘SOWG') is sequestered away from the field area. The SOWG learns about the field area by deploying ‘rover instruments' and analyzing data returned by the rover team under the same constraints that are faced by a SOWG during a real Mars mission (except that several Mars sols of activity can be completed in a single day).

Svalbard, Norway is an ideal location for ASTEP instrument deployments. At nearly 80° North latitude, the archipelago is almost devoid of plants larger than mosses, and mountain glaciers have created deeply scoured valleys, effectively the world's largest ‘road cuts' (Fig. 9). Rock exposures are breathtaking, and the lack of human habitation (with the exception of a few isolated towns) has left much of the landscape pristine. Svalbard has official status as an International Science Preserve, managed by the Norwegian government.

Initial interest in Svalbard for NASA research activities was engendered by the petrologic field work of Hans Amundsen (then at the University of Oslo).6 In 1997-1999, in the midst of the flurry of research activities associated with characterizing the ALH84001 meteorite (which was purported to have evidence of life7), Amundsen collaborated with Allan Treiman of the Lunar and Planetary Institute and the author (Ames Research Center) to characterize carbonate globules similar to those described from ALH84001, found in ultramafic xenoliths from Svalbard.8 Now in its 7th year, AMASE conducts yearly expeditions to Svalbard, providing a base of operations for scientists and technologists from many countries. Both NASA and ESA (European Space Agency) use AMASE as a testing ground for flight instrument prototypes and for conducting interdisciplinary science similar to that which occurs during landed planetary missions.

There is a variety of sedimentary, metamorphic and volcanic terrains in Svalbard.9 CheMin prototype instruments such as Terra have principally been deployed in volcanic areas, analyzing basalts, volcanic soils,10 ultramafic xenoliths and their weathering products11 and secondary Fe-Mg carbonates associated with late-stage hydrothermal events.12 Hydrothermal activity associated with volcanism was probably common on early Mars, which featured abundant basaltic rocks, water as ice or liquid, and heat from volcanoes and asteroid impacts. The most primitive forms of life on Earth still prefer hydrothermal environments, and such environments - and their habitability - can be studied either as presently active systems, or as fossil systems in the geologically near-recent of Svalbard. Early organisms were probably chemolithotrophic (they derived their energy from inorganic reactions in rocks), and a second possible habitable zone for early chemolithotrophic organisms in Svalbard is created by the weathering of ultramafic minerals such as olivine to form serpentine with the release of hydrogen.13

The style of volcanism in Svalbard is unique, in that the eruptive activity apparently occurred under ice, and the source magmas were volatile rich. Ultramafic xenoliths comprise as much as 20% of the volume of eruptive material, suggesting a deep source for the magma. Several Quaternary volcanic centers are prominent in the field areas studied by AMASE. Sverrefjell is a stratovolcano or large cinder cone ~500 m tall on the shore of Bokfjord (Fig. 9). Horizons of pillow lavas, indicative of eruption in water or under ice, exist nearly to the top of the volcano, and glacial erratics can be found at Sverrefjell's summit, indicating that the volcano was covered with thick ice either during or soon after its eruption. The style of eruption of Siggurdfjell is less obvious but probably was a volcanic neck or fissure, exposed up to 1000 m above present sea level. Fe-Mg carbonates are found in a variety of petrologic rock types and regimes on both Sverrefjell and Siggurdfjell. Ultramafic xenoliths, abundant at both localities, contain carbonate globules nearly identical to those found in ALH84001.8 Several styles of volcanic vents, 1-10 m in width (Fig. 10) contain films, rinds or thick crusts of carbonate material, clearly precipitated from water in localized hydrothermal systems during the waning stages of eruptive activity. The Terra instrument was used for in situ analysis of these carbonates both at the sites of deposition, and aboard the Research Vessel Lance (our field station) directly after collection. These on-site mineralogical analyses allowed the formulation of hypotheses and collection/analysis strategies while scientists were still in the field, and also were used to direct the collection of samples used for other purposes, such as life marker chip, IR, laser Raman, Laser-Induced Breakdown Spectroscopy (LIBS) and Gas Chromatography-Mass Spectrometry (GC-MS).14

The value of Svalbard as a Mars analog site was recently proved out in a paper identifying carbonate-rich outcrops on Mars from data obtained during operations of the MER rover Spirit.15 In the paper, the authors reported finding carbonates with nearly identical composition to the ALH84001 meteorite and to those reported from Svalbard11 and concluded that the Martian carbonates were formed in basaltic hydrothermal systems, analogous to those described from Svalbard.

Less obvious but equally valuable are the personal bonds formed by the various interactions, arguments, compromises and group decisions made by geologists, biologists, chemists, physicists and engineers during many days of joint field activity and Fast Motion Field Tests. Interdisciplinary cooperation in the field between different scientific specialties, as well as between scientists of different nationalities, is also a hallmark of joint expeditions such as AMASE. As a result of the common interests of NASA and ESA, and the high cost of ‘going it alone' when it comes to planetary exploration, a decision was recently reached by these agencies to work together on future Mars exploration. The collaboration, cooperation and respect engendered by joint research endeavors on Earth such as AMASE will pay dividends when the real missions come to fruition.

Future Versions of CheMin for Mars and Other Solar System Destinations

A major drawback of CheMin XRD/XRF instruments developed to date is that samples must be prepared and delivered to the instrument as fine-grained powder. Two next-generation CheMin-like instruments funded for development by NASA are intended to minimize or overcome these sample handling problems. Luna (Fig. 11) is an XRD/XRF instrument intended for robotic missions to the Moon or any other airless body. Luna is a reflection mode instrument that will analyze soils without sample preparation. While there is still a requirement for fine-grained samples (as there is for all powder XRD instruments), the reflection geometry allows for less sophisticated sample delivery and sample handling systems. Due to the geometrical considerations of a compact reflection instrument, the low 2θ limit is 8° 2θ, too high for the detection of some clays. However, for almost all other minerals of geological interest, the lowest 2θ peaks of interest are not lower than 15° 2θ. A second Hybrid single crystal / powder XRD instrument is being designed and built as an arm instrument for small rovers (Fig. 12).16 No sample preparation will be necessary for this instrument, except perhaps for the creation of a flat surface much as the Rotary Abrasion Tool (RAT) creates on the Spirit and Opportunity rovers. As-received powders such as soils can be analyzed via powder XRD, much like Luna, but with multiple detectors (and multiple wavelengths of radiation). Rocks with mineral grain sizes larger than fine-grained powder (most crystalline rocks would fall into this category) would be analyzed using the single crystal Laue technique. This capability is made possible through the use of multiple energy-discriminating area detectors. Single crystal maxima can be identified by their energy and 3-dimensional position in space, and related back to the structure (and therefore identity) of the diffracting mineral phase.

Spinoff CheMin Instruments for Terrestrial Use

Several instruments based on the CheMin design are now in commercial production. Terra is used as a CheMin analog instrument in NASA-sponsored field campaigns as well as a number of ESA and other field expeditions throughout the world. Terra is also used for mud-logging applications at oil drilling sites, in museums and at remote sites for analyzing geological materials for industrial processes.17

Pharmaceutical products are typically crystalline, and can be identified and quantified by XRD just as minerals are. The author is working with scientists from the Centers for Disease Control (CDC) and elsewhere to develop applications of Terra for the identification of counterfeit drugs in developing nations. First and foremost of these are the Malaria drugs. Nearly 1.5 million people, mostly children in developing nations, die of Malaria each year.18 Tens of thousands of these deaths are due to the scourge of counterfeit drugs, which in some places in SE Asia and equatorial Africa account for more than 50% of the drugs available.19 Simple and effective tests have been devised to detect the presence or absence of the active ingredient in malaria pills, artesunate. However, an increasing number of these counterfeit malaria drugs are being ‘salted' with a small amount of artesunate to escape detection by these tests. This unfortunate practice is creating resistance in the malaria parasite for what is now the only drug that is effective in fighting the worst strains of the disease.20 Terra is quite good at quantifying artesunate in malaria pills and can do so in remote locations, at the source of the problem.21

CheMin-type instruments are also being used in the world of Art and Antiquities. A reflection XRD instrument called ‘Duetto' was commissioned by Giacomo Chiari of the Getty Conservation Institute for characterizing art objects and intiquities.17,22 This instrument is now being used world-wide to characterize sensitive archeological artifacts, to evaluate the causes of deteriorization of ancient murals and other artifacts in places such as Tutankhamen's tomb and the Roman ruins of Herculaneum, Italy (Fig. 13).


1. http://msl-scicorner.jpl.nasa.gov/Instruments/.
2. Bish, D.L. and J.E. Post (1993). ‘Quantitative mineralogical analysis using the Rietveld full-pattern fitting method.' American Mineralogist 78, 932-942.
3. Chipera, S.J. and D.L. Bish (2002). ‘FULLPAT: a full-pattern quantitative analysis program for X-ray powder diffraction using measured and calculated patterns.' J. Applied Crystallography 35, 744-749.
4. Bish, D.L., Blake, D., Sarrazin, P., Treiman, A.H., Hoehler, T., Hausrath, E.M., Midtkandal, I., and Steele, A. (2007). ‘Field XRD/XRF mineral analysis by the MSL CheMin instrument.' LPSC XXXVIII, Abstract #1163.
5. Sarrazin, P., W. Brunner, D. Blake, M. Gailhanou, D.L. Bish, D. Vaniman, S. Chipera, D.W. Ming, A. Steele, I. Midtkandal, H.E.F. Amundsen and R. Peterson (2008). ‘Field studies of Mars analog materials using a portable XRD/XRF instrument.' LPSC XXXIX, Abstract #2421.
6. Amundsen, H. E. F. (1987). ‘Evidence for liquid immiscibility in the upper mantle.' Nature, 327, 692-695.
7. McKay, D.S., E.K Gibson Jr., K.L. Thomas-Keptra, H. Vali, C.S. Romanek, S.J. Clemett, X. D.F. Chillier, C.R. Maechling and R.N. Zare (1996). ‘Search for Past Life on Mars: Possible Relic Biogenic Activity in Martian Meteorite ALH84001.' Science 273, 924-930.
8. Treiman, A.H., H.E.F. Amundsen, David F. Blake and Ted Bunch (2002). ‘Hydrothermal origin for carbonate globules in Martian meteorite ALH84001: a terrestrial analogue from Spitsbergen (Norway). EPSL 204 (2002), 323-332.
9. Harland, W.B. (1997). ‘The Geology of Svalbard.' Memoir 17, Geological Society of London (521 p.).
10. Hausrath, E.M., Treiman, A., Bish, D.L., Blake, D., Sarrazin, P., Vincenzi, E., Midtkandl, I., Steele, A., and Brantley, S.L. (2009). ‘Short and long-term olivine weathering in Svalbard, and implications for Mars.' Astrobiology 8(6), 1061-1069.
11. Treiman, A.H., K.L. Robinson, D.F. Blake and D. Bish (2010). ‘Mineralogy Determinations by CheMin XRD, Tested on Ultramafic Rocks (Mantle Xenoliths).' AbSciCon 2010, Abstract #5351.
12. Blake, D.F., H.E.F. Amundsen, L. Benning, D. Bish, P. Conrad, M. Fogel, I. Midtkandal, D. Ming, A. Steele, A.H. Treiman and the AMASE team (2010). ‘Carbonate cements from the Sverrefjell and Sigurdfjell volcanoes, Svalbard Norway; Terrestrial analogs for Martian carbonates?' AbSciCon 2010, Abstract #5119.
13. Schulte, M., D. Blake, T. Hoehler and T. McCollom (2006). ‘Serpentinization and its implications for life on the early Earth and Mars.' J. Astrobiology 6(2), 364-376.
14. Steele, A., H.E.F. Amundsen, P.G. Conrad, L. Benning and the AMASE ‘09 Team (2010). ‘Arctic Mars Analogue Svalbard Expedition (AMASE) 2009.' AbSciCon 2010, Abstract #5674.
15. Morris, R.V. et al (2010). ‘Identification of Carbonate-Rich Outcrops on Mars by the Spirit Rover.' Sciencexpress, / www.sciencexpress.org / 3 June 2010 / Page 1-8 / 10.1126 / science / 1189667.
16. Sarrazin, P., P. Dera, R.T. Downs, D. Blake, D. Bish and M. Gailahou (2009). ‘Hybrid X-ray Diffraction for Planetary Mineralogical Analysis of Unprepared Samples.' LPSC XXXX, Abstract #1496.
17. Wilkinson, M. (2010). ‘Beyond Terra Firma.' Chemistry World, March 2010 pp. 50-53 (www.chemistryworld.org).
18. Marshall, A. (2009). ‘The Fatal Consequences of Counterfeit Drugs.' Smithsonian Magazine, Oct. 2009; Finkle, M., (2007). ‘Malaria: Stopping a Global Killer.' National Geographic, July, 2007.
19. Newton, P.N. et al (2008). ‘A Collaborative Epidemiological Investigation into the Criminal Fake Artesunate Trade in South East Asia.' PLoS Medicine | www.plosmedicine.org, February 2008, Vol. 5, Issue 2, pp. 209-219.
20. Newton, P. N. et al (2006). ‘Manslaughter by Fake Aretesunate in Asia - Will Africa Be Next? Plos Medeine 3(6):e197. DOI:10.1371 / Journal.pmed.0030197.
21. Blake, D.F., Philippe Sarrazin, Bradley W. Boyer and Tyler C. Jennison (2010). ‘The use of field-portable p-XRD for the rapid identification of counterfeit pharmaceutical products and subsequent excipient identification and quantification.' PPXRD IX, Abstracts w/program, Hilton Head Island, SC.
22. http://www.inxitu.com/new/html/duetto.html, http://www.inxitu.com/images/Duettowhitepaper.pdf
23. Chiari, G. (2009) ‘Saving Art in situ.' Nature Vol. 453, 159.
24. The author is grateful for nearly 20 years of support from NASA-sponsored research and SBIR activities that made this instrument possible. Thanks also to the many dedicated engineers, technicians, scientists and managers at the Jet Propulsion Laboratory, Pasadena, CA who made the CheMin FM a reality.


A. The CCD imager is operated in single photon counting mode (the CCD is exposed and read often enough so that in the vast majority of cases, pixels contain either background noise or the charge from a single photon). When a single X-ray photon is absorbed in a single pixel of the CCD, the energy of the photon is dissipated in the volume of the pixel as electron-hole pairs. Each electron-hole pair has about 3.65 eV of charge associated with it. The energies of X-ray photons from elements of geological interest range from a few hundred to many thousands of electron volts. A calcium Kα x-ray fluoresced from the sample, for example, has an energy of 3.68 KeV, and would create slightly more than 1,000 electron hole pairs in the CCD. A cobalt Kα photon from the X-ray source that has an energy of 6.99 KeV, would produce 1,917 electron hole pairs in the sample.

B. A maximum of 65 mm3 of sample material is delivered to the piezoelectrically vibrated funnel system that penetrates through the rover deck (during the time period when CheMin is not receiving samples, the CheMin inlet is protected by a cover). The funnel contains a 1 mm mesh screen to keep larger than expected grains from entering the CheMin sample handling system. Grains that cannot pass through the screen will remain there for the duration of the mission (samples will have been prescreened first to150 µm that pass through the screen will pass into the upper reservoir portion of the sample cell, where they will remain until the cell is inverted and they are dumped into the sump. However, under nominal conditions, the funnel will only receive material that has passed through the CHIMRA's 150 µm sieve. For the lifetime of the mission, CheMin is required to accept and analyze material delivered from CHIMRA with no more than 5% internal contamination between samples. Self-generated contamination originates from material that has remained in the funnel from previously delivered samples (and co-mingled with subsequent samples), or from material that has remained in previously used analysis cells (each cell will be used two to three times to accommodate 74 analyses during the nominal mission). CheMin empties used sample cells by inverting and vibrating the cell into a sump inside the instrument. Contamination is reduced by sample dilution; aliquots of sample material can either be dumped into the funnel and delivered directly to the sump through a shunt in the wheel without entering a sample cell (to remove funnel contamination), or a previously used sample cell can be filled with an aliquot of sample material, shaken and emptied to the sump prior to receiving a second aliquot of sample for analysis (to remove sample cell contamination). This processes requires coordination with CHIMRA to deliver more than one aliquot of a given sample.

C. The collimated ~50 µm diameter X-ray beam illuminates the center of an 8 mm diameter, 175-µm thick sample cell bounded by 6 µm thick Mylar or Kapton windows. The sample introduced into the funnel consists of

During the moderate shaking which results in grain convection, it is possible that phase segregation will occur as a result of size or density differences between individual mineral grains. To reduce this problem CheMin can episodically use larger shaking amplitudes (we call this ‘chaos mode') to homogenize particle size or density segregations in the sample chamber.

The CheMin sample cells are constructed in dual-cell ‘tuning-fork' assemblies with a single horizontally driven piezoelectric actuator in each assembly (Fig. 5). Sixteen of the dual-cell assemblies are mounted around the circumference of the sample wheel. Five of the cells will be devoted to carrying standards; the other 27 cells are available for sample analysis and may be reused by dumping samples into the sump after analysis (Fig. 4).

Both Mylar and Kapton window cells are mounted on the wheel. Each window type has both benefits and drawbacks, and the choice of window type will be made based on the nature of the sample that is being analyzed. Mylar windows have a very flat diffraction background but Mylar is less durable than Kapton under severe vibration and is susceptible to destruction if highly acidic samples (e.g., the iron sulfate hydrate mineral copiapite) are analyzed. Kapton windows are extremely durable under severe vibration and are not susceptible to acid attack, but have a small diffraction contribution at ~6-7° 2θ (CoKα radiation) which could be mistaken for the (001) diffraction peak of some clay minerals. Windows of both Kapton (in 13 cells) and Mylar (in 14 cells) are used in the Flight Model (FM) and Development Model (DM), as shown in Fig. 4.

CheMin uses a 600x600 pixel E2V CCD-224 frame transfer imager operated with a 600x582 data collection area. The pixels in the array are 40x40 µm2, with an active region of deep depleted silicon 50 µms thick. The front surface passivation layer is thinned over a substantial fraction of the active pixel area. The E2V CCD-224 imager is a modern version of the E2V CCD-22 that was specially built for an X-ray astronomy application. The large size of the individual pixels causes a greater percentage of X-ray photons to dissipate their charge inside a single pixel rather than splitting the charge between pixels. The enhanced deep depletion zone results in improved charge collection efficiency for high energy X-rays. The thin passivation layer makes the CCD sensitive to relatively low-energy X-rays (AlKα at 1.49 KeV, SiKα at 1.75 KeV).

In order to keep the CCD from being exposed to photons in the visible energy range (from X-ray induced optical fluorescence from the sample) during analysis, a 150 nm Al film supported on a ~200 nm polyimide film is suspended in front of the detector. The detector itself is cooled to ~-60 °C by a cryocooler. The actual temperature of the CCD depends on the temperature of the RAMP (Rover Avionics Mounting Platform) into which the cryocooler's thermal load is dissipated. By cooling the CCD, dark current is reduced, as well as the effects of damage to the silicon lattice caused by neutrons from the RTG and the DAN instrument.

Figure and Table Captions

Figure 1. Geometry of the CheMin XRD/XRF instrument. a) (left) overall geometry of CheMin; b) (above right) XRD 2θ plot obtained by summing diffracted photons from the characteristic line of the X-ray source (colored magenta in Figure 1a); c) (below right) X-ray fluorescence spectrum obtained by summing all of the X-ray photons detected by the CCD (XRF photons from the sample shown schematically in green and red in Figure 1a).

Figure 2 (a). CheMin I instrument. A Princeton InstrumentsTM camera is interfaced to an evacuated sample chamber. A modified collimator from a Debye-Scherrer powder camera provided a 50 µm diameter collimated beam. X-rays are provided by a Norelco-Philips Tube tower. The detector is a Tektronix TK-512 backthinned, back illuminated 512 X 512 pixel CCD imager. Initially, analyses required several thousand CCD frames, with a total energy of thousands of watts hours. Since the sample chamber had to be taken apart to load samples, it typically took several days to align the X-ray beam after sample exchange. Power supplies, X-ray generator, camera controller and chilled water supplies for the X-ray tube tower and camera are not shown.

(b). CheMin II instrument. The instrument consisted of an AndorTM CCD camera interfaced to a sample chamber (kept at atmospheric pressure), bolted to an Oxford X-ray Technology Group micro-focus X-ray tube. The detector is an E2V 5530 1250 X 1150 pixel CCD imager. Samples could be exchanged in air, without disturbing the alignment of the instrument. Power supplies for the X-ray tube and camera are not shown. The instrument is air cooled.

(c). CheMin III. The instrument is shown in operation in Badwater Basin, Death Valley in 2003. The X-ray tube and camera are the same as used in CheMin II, however, miniature power supplies and camera control electronics are housed in a portable frame with the instrument. Samples are moved by piezo-shaking in a mylar film-bounded transmission cell. Not shown are storage batteries and laptop computer that provide power and instrument control. Data are collected as individual frames and analyzed off-line by the laptop computer.

(d). CheMin IV. (left) As deployed in Svalbard, Norway in 2006. (right) CheMin Co-I Dave Bish of Indiana University performs the first complete XRD/XRF data collection and quantitative analysis in the field. The instrument is fully self-contained, operated with an integrated computer and on-board lithium batteries and power supplies. Once an analysis is completed, the raw data are downloaded onto a USB storage drive and transferred to a laptop computer for processing.

Figure 3. Exploded view of CheMin spacecraft instrument. Major components of the instrument are labeled: (1), sample delivery funnel; (2), sample wheel containing 32 sample cells; (3), X-ray tube and collimator housed in pressurized SF6 dielectric (tube, collimator and 28 KV high voltage power supply weigh less than 1 Kg); (4), CCD imager with support structure; (5), CCD housing with cryocooler. Other components include electronics, sample wheel drive motor, etc. The entire instrument, which is essentially a 10' cube, weighs ~10 Kg and operates on ~40 watts.

Figure 4. Sample wheel. Sample cells (27 total) are loaded and analyzed at the top, then rotated 180 degrees and dumped in the sump after analysis. Permanent reference standards (5 total) are shown colored red, green cells are available for analysis of unknowns. The ‘bypass cell,' shown in pink, is used to pass multiple sample aliquots through the funnel for contamination reduction by dilution. Similarly, individual sample cells can be loaded, shaken and dumped without analysis, then refilled to dilute any contamination that may be present in the cell from previous analyses. If necessary, cells can be used 3 or more times during the mission.

Figure 5. CheMin sample cells: (left), exploded view of dual-sample cell. Each dual-sample cell is a tuning fork driven at resonant frequency by a piezoelectric stack. The amplitude of shaking can be changed as a function of input voltage to the piezo. Each cell is an 8mm diameter X 170 µm thick volume bounded by X-ray transparent plastic films. During analysis, the piezo stack vibrates the cell at resonance, creating a turbulent flow of grains in the cell. Over time a myriad of grains from the sample flow past the 50 µm X-ray beam in random orientations. A reservoir above each sample cell holds excess material. The dual-cell assemblies are machined from solid titanium, and have a reproducible positioning accuracy (sample-to-detector path length) of less than 10 microns, 1/10th the diameter of a human hair).

Table 1. Diffraction pattern range and peak resolution (for CoKα radiation) of the CheMin FM. Range and resolution were chosen to allow for the successful identification and quantification of virtually all minerals.

Table 2. CheMin FM mineralogical analysis requirements. Analytical capabilities are defined by a ‘flow down' of requirements derived from the science goals and objectives of the mission.

Figure 6 (a). Stack of raw single frames (10-second exposures) of 600X582 pixel CCD data. Each frame is background subtracted, and shows individual X-ray photon detections as white or gray spots. In any single exposure, only a few hundred pixels out of 350,000 record X-ray photon detections.

(b). Stack of energy-selected CoKα minor frames of CCD data. Each minor frame is a 600X582 array in which the x,y array elements store the integer sum of CoKα photons detected at that x,y pixel location during 100-200 individual frames (exposures).

(c). Major frame of CCD data, representing the sum of the minor frames. Major frames are used to construct a 2-D diffraction pattern that is translated into a more conventional 1-D diffractogram used for mineral identification and quantitative analysis. Minor frames are used to estimate the precision and accuracy of an analysis and to detect instrumental drift and transient anomalies in the data can be identified and removed.

(d). 1-D diffractogram constructed from a major frame of CCD data. Intensities from the 2-D pattern are summed circumferentially about the central beam to create an intensity vs. 2θ diffractogram, similar to that obtained in conventional powder X-ray diffractometry. The diffractogram shown is from the mineral amphibole, one of the reference standards in the CheMin FM.

(e). Raw Energy-Dispersive X-ray (EDX) histogram of a single frame of CCD data. The x-axis represents X-ray photon energy and the y-axis represents the number of counts per unit energy (in this case, represented as ‘DN' or digital numbers from the CCD). Observed peaks include the primary CoKα flux at ~1960 DN, sample-generated FeKα at ~1890 DN, and sample-generated CaKα and CaKβ at ~1525 and ~1640 DN respectively.

(f). Summed EDX histogram for major frame of CCD data. The X-axis is now converted from DN to energy in thousands of electron volts (KeV). Elemental peaks present include CoKα (6.92KeV), CoKβ (7.70 KeV), FeKα (6.4 KeV), TiKα (4.51 KeV), TiKβ (4.93 KeV), CaKα (3.68 KeV), CaKβ (4.01 KeV), ArKα (2.96 KeV), SiKα (1.75 KeV) and AlKα (1.49 KeV). Fe, Ti and Ca are elements detected from the sample. Argon originates from the 10 mbar of Ar present in the FM during the analysis, Si originates from the detector itself and Al is generated from the light shield in front of the detector. The CheMin instrument is sensitive to elements above ~ Z=15 (phosphorus). While the CCD itself is supremely sensitive to low-energy X-rays such as AlKα, the transmission geometry of the instrument and the presence of a light shield precludes detection of these elements.
Figure 7. A synthetic ceramic was developed as a permanent standard for the instrument. This sample is used to measure the energy response of the CCD detector to sample-generated X-rays. X-ray sensitivity appears to be quite good from Sr to S (Al and Si are instrument generated). The FWHM of individual X-ray peaks, and ultimately X-ray detection and discrimination will be degraded by neutron damage from the RTG during the mission.

Figure 8. Use of Terra during NASA and other field campaigns. a). The author (NASA/ARC) during the Scarab/RESOLVE field test on Mauna Kea, HI (2008). Terra was used to analyze as-received volcanic soils and partial run products from In Situ Resource Utilization (ISRU) experiments. Lunar oxygen-generating equipment in the background b). Doug Ming (NASA/JSC) on an expedition to the dry valleys of Antarctica (2008). Terra was used to analyze soils and other materials to understand pedogenesis (soil generation) in cold, dry climates. c). Ron Peterson (Canadian Geological Survey), in the high Arctic of Canada (2009). Searching for low-temperature stability minerals such as the hydrates of iron and magnesium sulfate. d). The author (NASA/ARC) during the AMASE expedition in Svalbard, Norway (2009). Analyzing carbonate sediments and cements in volcanic vents for clues as to their origin.

Figure 9. Sverrefjell Volcano, Svalbard, Norway. Sverrefjell was apparently erupted under a thick ice sheet and later bisected by a km-thick glacier. Pillow lavas indicative of eruption at sea level or below are evident at several levels on the cindercone, and glacial eratics can be seen at all levels, including at the summit.

Figure 10. Volcanic vent. One wall of a breccia-filled vent showing carbonate cementation (orange) of the basaltic breccia. In all likelihood, original volcanic vents such as this one acted as a plumbing system for late stage hydrothermal systems under the ice. Large Norwegian (Hans Amundsen) shown for scale.

Figure 11. Cartoon of LUNA instrument. The LUNA instrument is a powder XRD with a reflection geometry. Angular range is 8-55° 2θ, sufficient to characterize all minerals thought to exist on the lunar surface.

Figure 12. Concept drawing of the hybrid single crystal Laue / powder XRD contact instrument. (left), Placement of detectors and X-ray source in the instrument. Fine-grained powders are characterized by powder XRD, similar to the LUNA instrument. Larger crystals are characterized by energy-resolved Laue single crystal patterns. (right), Placement of instrument on the arm of a mid-rover instrument similar to that proposed for the MAX-C 2018 mission to Mars.

Figure 13. Duetto Instrument with Mummy. Duetto, commissioned by the Getty Conservation Institute, is a commercial spinoff of CheMin, is used for in situ art characterization and art conservation.