The FAMARS instrument,
an AFM for planetary exploration, part II
With Daniel Parrat, Institute of Microtechnology University of Neuchâtel

Sensor chip
The microfabricated sensor chip consists of eight cantilevers with integrated, piezoresistive deflection sensors. These cantilevers are aligned in a row and mounted on the scanner with two tilt angles relative to the sample. Thus, they can be engaged one after the other to provide redundancy in case of tip or cantilever failure. Silicon tips at the end of the cantilevers are used for probing the sample. Images can be recorded in both, static and dynamic operation mode. In the latter case, excitation of the resonance frequencies of the active cantilever is achieved by vibrating the whole chip by means of a piezoelectric disk.


Figure 5. FAMARS chip. The cantilever on the right is the first to be used. A silicon tip is located at the end of each cantilever. Click to enlarge.

The electronic board measures 300mm x 110mm x 10mm and weights 190g. It is a digital controller which generates the feedback and scanning signals. The onboard processor executes AFM specific, low-level commands and communicates via a serial interface with the Lander computer. The latter sequentially requests the execution of the low-level commands, based on predefined, special algorithms for autonomous AFM measurements. Measured images are stored on the Lander computer from where they will be downloaded to Earth via one of the satellites currently orbiting Mars.
The measured data for the whole mission is down linked only once every day.


Figure 6. Image of the controller board (300mm x 110mm). Click to enlarge.

Preliminary measurements
In order to interpret the data that hopefully will be sent back from Mars, we currently perform calibration and characterization experiments. For example, we evaluate the capabilities of FAMARS by imaging small spheres (typically less than 2 microns), put also other samples, such has small crystals with characteristic shapes will be looked at. Typical images are composed of 256 x 256 data points. An example is shown in figure 7, where silica beads (about 100 nm in diameter) are imaged with FAMARS. The beads appear to have an oval distortion, which we attribute to the imperfect shape of the probing tip.


Figure 7. 4µm x 4µm AFM image of silica nanobeads taken with FAMARS. Click to enlarge.

Measurements procedure on Mars
Before we can look at samples, they have to be delivered to the microscope station. This will be done by a robotic arm to which a scoop is attached. The contents of this scoop will be poured onto a section of the sample wheel, where a set of six substrates will receive the particles.

The substrates are designed to distinguish between different adhesion mechanisms and include magnets, sticky polymers, and “buckets" for bulk sampling. The wheel will then be rotated, presenting the samples to the optical microscope and the AFM. Upon power up of the AFM, the Lander computer downloads the operation software into the RAM of the AFM controller. It then sequentially requests the execution of low-level commands, based on predefined, special algorithms. The latter consist of block-level commands which allow autonomous measurements. First, the health of all cantilevers is verified by checking the integrated deflection sensors. If the active cantilever is alive, it is initialized in dynamic mode. The vibration amplitude is verified to ensure that the dynamic mode is usable. If the amplitude is too weak, the AFM either initializes the lever for the static mode or aborts, depending on what is asked by the operator.

After the mode selection, all the parameters of the feedback loop (setpoint, proportional and integral gains, etc) are set for the approach procedure. The sample, which is placed in front of the AFM by a rotation of the sample wheel, is then slowly stepped towards the AFM tip. When the sample enters in contact with the tip, i.e. when the setpoint of the feedback loop is reached, the AFM controller sends a signal to the MECA controller to stop the stepper-motor. Then some of the parameters are adjusted in order to take an AFM image after the first image, other images can be taken by doing zooms in areas of interest. When the measurements are finished, the wheel is moved away from the AFM.

In order to have a right interpretation of the acquired data, AFM measurements of a Martian sample will be preceded and followed by measurements of calibration targets. These two substrates are also located on the sample wheel but will not be exposed to soil particles. They comprise a silicon grid with a period of 10 microns and an array of sharp silicon tips. The silicon grid allows calibrating the size and the orthogonality of the scan and the array of tips allows having an estimation of the AFM tip’s shape. Figure 8 shows AFM images of both calibration targets.


Figure 8. AFM image (tapping mode) of a) a calibration grid in silicon (period = 10 microns) and b) an array of silicon tips (period = 2.1 microns) with few contaminations. The height of these images is coded in a grey scale. Click to enlarge.

State of the project
The first measurements showed that the FAMARS instrument, in combination with the delivery system of MECA, should allow characterizing Martian particles with high resolution. FAMARS passed successfully all environmental tests. The flight model was delivered to the Jet Propulsion Laboratory in 2005 and is now integrated on the Phoenix Lander, waiting for the launch. Future work will mainly consist in characterizing the AFM in end-to-end tests and cataloging different samples and Mars-analogues using test-bed set-ups.

History
In April 2001, NASA's Mars Surveyor Lander, intended to realize several in-situ experiments on Mars, was ready to be launched, but was cancelled due to the back to back loss of two NASA missions in the end of 1999, Mars climate Orbiter (MCO) and Mars Polar Lander (MPL). On the Surveyor Lander, a payload called MECA (Mars Environment Compatibility Assessment) was designed to gauge possible physical dangers that the Martian surface may present to human explorers. MECA included a microscopy station, in particular the FAMARS instrument.

After cancellation of the mission, the MECA consortium sought for a new flight opportunity by joining the PHOENIX project in a competition for the first Scout mission to Mars, announced by NASA in 2002. The Phoenix mission was selected, and MECA will see its second chance for a journey to Mars onboard of the 2007 Lander. As Phoenix is designed to study the history of water and habitability potential in the Martian arctic's ice-rich soil, the former meaning of “MECA” is no more appropriate and “MECA” now stands for “Microscopy, Electrochemistry, and Conductivity Analyzer”.

Acknowledgements
This work is financially supported by the Space Center at EPFL, the Wolfermann-Nägeli Foundation and the Canton and Republic of Neuchâtel”.

The first part of this article can be read here.