Chemistry and Mineralogy (CheMin)

Scientific Objectives

CheMin is designed to perform mineralogical analyses of rocks and soils to pursue the identification and characterization of past habitable environments on Mars. Understanding the mineralogy of samples allows an assessment of the involvement of water in the formation, deposition and alteration of surface materials. CheMin can also help to identify mineral biosignatures and to determine if mineral indicators of energy sources for life have been present.

Instrument Overview

The Chemistry and Mineralogy (CheMin) instrument located within the body of the Mars Science Laboratory establishes the mineralogy of powdered samples using x-ray diffraction spectroscopy. CheMin consists of a funnel for receiving powdered samples, 27 reusable sample cells, an x-ray source, a cooled charge-coupled device (CCD) detector, and analog and digital electronics. Software for onboard processing of data runs in the rover computer. CheMin contains 5 internal calibration standards for instrument characterization and establishing performance trends.

Principal Investigator

The CheMin principal investigator is David Blake, Ames Research Center.

Samples

The MSL rover robotic arm can acquire powdered samples by drilling rocks or scooping soils. These samples are sieved through a screen with 150 micrometer holes prior to delivery to the CheMin funnel. After opening the inlet cover, the funnel is vibrated with piezoelectric actuators during the delivery of samples. Approximately 75 cubic millimeters of material is delivered in one portion, but the individual sample cells need only 15 cubic millimeters for an analysis. Multiple portions may be delivered to help clean previous samples from either the funnel or a reused cell. Two types of sample cells are available on the sample wheel: 14 with mylar windows, and 13 with kapton windows. Kapton is more durable (mechanically and chemically) than mylar but has a signature which may interfere with certain types of samples. Sample cell selection is an operational choice. Piezoelectric actuators are used to agitate the samples in the cells and present random crystal faces to the x-ray beam. After a sample is analyzed, it will be dumped into the instrument sump, allowing the cell to be reused.

X-ray source (XRS)

The x-ray tube has a cobalt anode producing characteristic x-rays at 6.9 keV (K-alpha) and 7.6 keV (K-beta). The sample in the cell is exposed to this collimated x-ray beam, and diffracted and fluoresced x-rays are measured by the CCD detector. The x-ray tube is nominally operated at 28 kV and requires approximately 15 minutes to ramp up before data collection and a similar amount of time to ramp down after use.

Detectors

X-rays are captured on a CCD cooled with a mechanical cryocooler. Frames recorded during an analysis are 600 pixels by 582 pixels, which is not quite the entire array, allowing frames to conveniently fit within available memory pages. The frequency of image collection is a commandable parameter, and will typically be one frame every ~15 seconds. Each image records the location and energy of the incident x-ray, allowing the extraction of both diffraction and fluorescence data from the sample. The optimum operational temperature is below -60 C to eliminate dark current and charge transfer issues due to accumulated radiation damage. The ability to achieve the optimum detector temperature depends on the temperature of the instrument mounting interface.

Electronics

The analog voltages corresponding to each pixel of the CCD are fed through commandable gain stages and converted to digital format for storage and processing. Up to 2730 individual frames can be stored within CheMin, corresponding to over 11 hours of data collection, assuming a 15 second integration time. Housekeeping data (temperatures, voltages, parameters, etc.) are periodically captured and stored for subsequent downlink whenever the instrument is powered.

Software

The individual CCD images collected during an analysis fill a data volume which is too large to downlink in its entirety. For example, a 5 hour integration consisting of 1200 frames is well over 800 MBytes. This large data volume is processed into smaller products which will fit into the downlink. These products are described in the Experiment Data Record (EDR) Software Interface Specification (SIS) and are produced for each 'minor frame.' The size of a minor frame is commandable and may consist of approximately 30 minutes of data. In the 5 hour example, the data might be grouped into 10 minor frames, each consisting of 120 images (or 30 minutes of data). The data processing software is located within the MSL rover compute element, not within CheMin itself: The data interface between the instrument and the rover is the transfer of raw image frames.

Calibration

Five separate calibration standards are carried within CheMin. Two beryl-quartz mixtures (beryl:quartz=88:12 and 97:3) are used to establish the 2-theta scale for the data and will be analyzed periodically to check the internal alignment. A synthetic ceramic designed to have a variety of fluorescence lines can be used to establish and track the energy scaling of the data. A natural amphibole (pargasite) with minor additional mineral phases has a number of closely spaced diffraction peaks and is useful for evaluating 2-theta resolution. An arcanite sample was also added to characterize instrument performance in the presence of a significant sulfur fluorescence background. Integral to the calibration approach is the CheMin Development Model (DM) which is similar to the flight instrument, but will reside in a terrestrial laboratory and provide extensive analyses of a variety of relevant samples.

Operational Considerations

The primary considerations for commanding CheMin during surface operations are ambient temperature and available energy. The detector produces better data when it is colder, and its temperature is related to that of the instrument mounting interface. Thus, the ideal time to run a CheMin analysis from a thermal perspective is generally immediately prior to sunrise on any given sol. Receiving samples from the robotic arm do not require a cold detector and will most likely be performed in the daytime while the arm actuators are warmer. Depending upon the complexity of the sample and the signal to noise levels necessary to identify and quantify minor mineral phases, the total integration time will likely be 8 to 10 hours. This analysis duration does not have to be continuous, as data from different sols can be combined on the ground to produce an analysis representing the total integration time. CheMin consumes approximately 60 Watts during an analysis, so the integration time on a given sol may be limited by the available energy. The MSL rover can go to 'sleep' after it initiates an analysis to minimize the energy usage. The rover needs to be awake to turn off CheMin after the analysis and to transfer and process the data for downlink.

Ground Software

After downlink of the data products in the format described in the accompanying EDR SIS, ground software tools are used to monitor the health of the instrument, determine the mineralogy, and identify chemical contributors to the sample. The two-dimensional diffraction images are converted into one-dimensional plots of counts versus 2-theta and read by commercial fitting programs to determine the mineralogy of the sample. The energy dispersive histograms produced in onboard processing are used to characterize the elemental chemistry.

References: [BLAKEETAL2010], [VANIMANETAL1998]