© NASA / JPL-Caltech / University of Arizona / Texas A&M UniversityThis image was taken by NASA's Phoenix Mars Lander's Surface Stereo Imager on June 5, 2008, the eleventh day after landing. It shows the robotic arm scoop, with a soil sample, poised over the partially open door of the lander's oven
Have Mars landers been destroying signs of life? Instead of identifying chemicals that could point to life, NASA's robot explorers may have been toasting them by mistake.

In 1976, many people's hopes of finding life on Mars collapsed when the twin Viking landers failed to detect even minute quantities of organic compounds - the complex, carbon-containing molecules that are central to life as we know it. "It contributed, in my opinion, to the fact that there were no additional [US lander] missions to Mars for 20 years," says Jeff Moore of NASA's Ames Research Center in Moffett Field, California.

The result also created a puzzle. Even if Mars has never had life, comets and asteroids that have struck the planet should have scattered at least some organic molecules - though not produced by life - over its surface.

Some have suggested that organics were cleansed from the surface by naturally occurring, highly reactive chemicals such as hydrogen peroxide. Then last year, NASA's Phoenix lander, which also failed to detect organics on Mars, stumbled on something in the Martian soil that may have, in effect, been hiding the organics: a class of chemicals called perchlorates.

At low temperatures, perchlorates are relatively harmless. But when heated to hundreds of degrees Celsius they release a lot of oxygen, which tends to cause any nearby combustible material to burn. For that very reason, perchlorates are used in rocket propulsion.

The Phoenix and Viking landers looked for organic molecules by heating soil samples to similarly high temperatures to evaporate them and analyse them in gas form. When Douglas Ming of NASA's Johnson Space Center in Houston, Texas, and colleagues tried heating organics and perchlorates like this on Earth, the resulting combustion left no trace of organics behind. Ming's team presented their results at the recent Lunar and Planetary Science Conference in Houston.

Iron oxides have also been suspected of interfering with the detection of organics, but perchlorates are probably far more effective, says Chris McKay of Ames. Even if organics make up a few parts per thousand of the soil, Viking or Phoenix could have missed them, he adds, so it is too soon to conclude that these materials are not there. "We haven't looked the right way," he says.

Jeffrey Bada of the University of California, San Diego, agrees that a new approach is needed. He is leading work on a new instrument called Urey for the European Space Agency's ExoMars rover, due to launch in 2016, which will be able to detect organic material at concentrations as low as a few parts per trillion. The good news is that, although Urey heats its samples, it does so in water, so the organics cannot burn up.

Mystery of the missing salt

Organic chemicals are not the only substance that we may have missed on the Red Planet. We should have seen carbonate salts littering the surface.

Weathering breaks down basalt, the dominant rock in the planet's crust, into a clay plus positive ions. These ions should react with carbon dioxide in the Martian atmosphere to form carbonate salts, explains Ralph Milliken at NASA's Jet Propulsion Laboratory in Pasadena, California.

Orbiters have spotted clay on Mars but few carbonates or other salts. We shouldn't assume that they aren't there, however, Milliken says.

Milliken and his colleagues have calculated that weathered Mars basalt should produce equal amounts of clay and salt. Thus in the planet's southern highlands, where thousands of clay deposits have been identified, there should be at least as much salt (Geophysical Research Letters, DOI: link). "Chemistry has shown that you can't draw conclusions from observations alone, because you are still missing pieces of the puzzle," says Milliken.

Some argue that the lack of known carbonate salt deposits points to a different atmospheric composition in the past, but Milliken says we should study the rocks directly before making any conclusions.