Correction to “Reanalysis of the Viking results suggests perchlorate and organics at midlatitudes on Mars”

Abstract

[1] In the paper “Reanalysis of the Viking results suggests perchlorate and organics at midlatitudes on Mars” by R. Navarro-Gonzalez et al. (Journal of Geophysical Research, 115, E12010, doi:10.1029/2010JE003599), the concentrations of carbon dioxide (CO2), oxygen (O2), nitric oxide (NO), nitrous oxide (N2O), benzene (C6H6), toluene (C6H5-CH3), chloromethane (CH3Cl) and dichloromethane (CH2Cl2), produced by thermal volatilization (TV) of Atacama soils with and without added perchlorate, were estimated from a calibration curve of carbon dioxide assuming that all chemical species would have a similar response to CO2 by gas chromatography coupled to mass spectrometry (GC-MS). This was an oversimplification and the concentrations of all species, except for carbon dioxide, should be taken as a first order of approximation, particularly since no calibration curves were done for each individual compound detected. A known amount of CO2 was introduced into the pyrolyzer probe cavity by varying the gas pressure of a calibration mixture (396 ppm carbon dioxide in helium) inside a sampling gas loop of fixed volume at a constant room temperature. The volume of the sampling loop could be changed to increase/decrease the amount of gas injected into the pyrolyzer cavity in the experiments. This volume was fixed for all experiments reported; however, it was incorrectly taken to be 40 cm3 when it was actually 10 cm3. This resulted in an over estimation of the concentration of all chemical species measured experimentally by a factor of 4 and reported in Figure 4. We have revised Figure 4 to show the actual estimated concentrations of CO2, O2, NO, N2O, C6H6, C6H5-CH3, CH3Cl and CH2Cl2. In addition, the conversion factor from organic carbon to benzene was incorrectly given as 1 ppm organic carbon = 6.5 ppm of benzene, when in fact, it is 1.1 ppm of benzene. This error was only made in the legend and does not alter any data cited in the main text of the paper. Figure 4 and its legend are completely revised. [2] The revised values of chemical species released by TV and measured by GC-MS agree stoichiometrically with the estimated concentrations of soil organic carbon determined independently by its oxidation to carbon dioxide by permanganate and measured by GC-MS. The Atacama soil from Yungay is characterized by 32 ± 6 ppm of organic carbon that upon TV at 750°C releases mainly 36.5 ± 14.8 ppm benzene and 6.2 ± 2.1 ppm toluene (1 ppm organic carbon = 1.1 ppm of toluene), equivalent to about 39.4 ± 15.6 ppm of organic carbon. The agreement is significant considering the different analytical methods employed in the determination of the soil carbon and the products released by TV. [3] Similar to Figure 4, there was an error in the conversion factor of soil carbon to methane reported in Figure 6. 1 ppb organic carbon is equivalent to 1.3 ppb CH4 and not 3.7 ppb as reported previously. This error was only made in the legend and does not alter any data cited in the main text of the paper. Furthermore, the oxidant considered in the computations was 1% wt magnesium perchlorate and not magnesium peroxide as stated. Therefore, the legend of Figure 6 is completely revised. [4] The conclusions and implications of our paper are not affected by these corrections. Our results do show that when Mars-like soils from the Atacama Desert with 32 ± 6 ppm of organic carbon are mixed with 1 wt% magnesium perchlorate and heated nearly all the organics present are decomposed to water and carbon dioxide, but a small amount are chlorinated forming 400 ppb of chloromethane and 5 ppb of dichloromethane at 500°C. The detection of chloromethane and dichloromethane by the Viking Lander 1 and Viking Lander 2, respectively, was due to the combined presence of perchlorate and organics in the soil sample during heating. The corrections given here do not change or alter the conclusions given originally that reinterpretation of the Viking results suggests that chloromethane or dichloromethane were not due to the presence of cleaning fluids but if fact were produced by the interaction of perchlorates and organics were the Martian soil was heated to examine the presence of organics. [5] Figure 6. Predicted production of methane (CH4), formaldehyde (CH2O), carbon monoxide (CO), carbon dioxide (CO2), chloromethane (CH3Cl) and dichloromethane (CH2Cl2) as a function of time during the TV step at 500°C from a Martian soil containing 40 ppb organic carbon, 0.5% wt water, 0.5% of hydroxyl groups, and 1% wt magnesium perchlorate; 1 ppb organic carbon is equivalent to 1.3 ppb CH4, 2.5 ppb CH2O, 2.3 ppb CO, 3.7 ppb CO2, 4.2 ppb CH3Cl, and 7.1 ppb CH2Cl2. [6] We would like thank Klaus Biemann from the Massachusetts Institute of Technology and Jeffrey L. Bada from University of California, San Diego, for pointing out these errors to us.