In a bid to improve on the soil sampling methods used in planetary exploration, researchers at Los Alamos National Laboratory are developing a laser-based system suitable for use on Mars that would work rapidly and remotely. The instrument will scan martian topogra-phy from a surface rover using an optical imaging system and a red aiming beam to identify potential samples. After the ground crew on Earth selects a target, a 6-ns laser pulse fires at the sample, creating a small plasma from the martian soil. The optical imaging system then collects the spark light from the plasma. A microspectrograph measures it and a charge-coupled device camera made by Santa Barbara Instrument Group, Santa Barbara, Calif., records the spectrum, which is downloaded to Earth for analysis. Laser plasma is formed on a soil sample in a surrounding atmosphere of 5 t CO2 to simulate the conditions on Mars. Courtesy of Dave Cremers. David Cremers, a technical staff member with the project, said the instrument can take samples up to 19 m away, using 35-mJ pulses. Lenses will expand each laser pulse 10 or 20 times to focus it on the sample. "We would fire the laser every 5 or 10 seconds and average 20 to 30 sparks on the same sample," he said. The project is working to characterize the technique and its limits. Such a remote technique is a boon to planetary exploration because the rover can collect samples from difficult terrain, like cliff faces or inside craters, without having to get close. And it's efficient. Scientists can make a measurement in 3 to 5 minutes. A measurement of elemental composition on the Pathfinder rover took several hours. What's more, the ground crew on Earth can take another sample by repositioning the aiming system instead of moving the rover. Yet making the sampling loop work reliably involves ironing out some issues on the drawing board. One such concern is the choice of laser. Cremers said the instrument will probably incorporate an Nd:YAG laser operating at 1064 nm because it is small and compact. Although they have not decided on a specific one, the researchers are trying a Q-switched model made by Kigre Inc. in Hilton Head, S.C. Getting reliable results will depend on being able to calibrate the instrument. So the researchers are testing the system on samples that simulate the martian mantle, and in conditions that mimic the CO2-rich martian atmosphere. This will provide the test data to enable remote calibration. Cremers said the researchers are also working to integrate a Raman system that compares scattered light with known molecular responses. The resulting mineralogical data would be useful to geologists for understanding planet-building processes as they have unfolded on Mars.