By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
 Recent laboratory tests, analog studies and numerical simulations all suggest that Martian dust devils and larger dusty convective storms generate and maintain large-scale electric fields. Such expected E-fields will have the capability to create significant electron drift motion in the collisional gas and to form an extended high energy (u ≫ kT) electron tail in the distribution. We demonstrate herein that these energetic electrons are capable of dissociating any trace CH4 in the ambient atmosphere thereby acting as an atmospheric sink of this important gas. We demonstrate that the methane destruction rate increases by a factor of 1012 as the dust storm E-fields, E, increase from 5 to 25 kV/m, resulting in an apparent decrease in methane stability from ∼ 1010 sec to a value of ∼1000 seconds. While destruction in dust storms is severe, the overall methane lifetime is expected to decrease only moderately due to recycling of products, heterogeneous effects from localized sinks, etc. We show further evidence that the electrical activity anticipated in Martian dust storms creates a new harsh electro-chemical environment.
 Dust storms on Mars are expected to generate and maintain large-scale electric dipole fields, with interior E-field values approaching the atmospheric breakdown levels of ∼25 kV/m [Melnik and Parrot, 1998; Farrell et al., 2003, 2006; Zhai et al., 2006]. The E-fields form via grain-grain contact electrification that tends to leave smaller dust particles with a net negative charge and larger sandy grains with a net positive charge [Ette, 1971]. Vertical winds in the storm then mass-stratify (and charge-stratify) the grain distribution as a function of height, creating a downward-directed electric dipole moment and large-scale dipole electric field structure [Crozier, 1964; Farrell et al., 2004].
 Evidence for this charge creation/separation process is in the form of analytical models [Melnik and Parrot, 1998; Farrell et al., 2003, 2006; Zhai et al., 2006] and terrestrial analog studies [Freier, 1960; Crozier, 1964; Schmidt et al., 1998; Delory et al., 2002; Farrell et al., 2004; Jackson and Farrell, 2006]. The former modeling studies all suggest that E-fields at many tens of kV/m can be generated in a convective Martian storm with charging exponential growth time scales on the order of 10's of seconds under ideal conditions. The latter analog studies consistently find relatively large electric fields in small terrestrial dust devils and saltating dust fronts, with dust devil fields > 100 kV/m and dust front fields > 150 kV/m. The E-fields in terrestrial dust devils/fronts are not large enough to create atmospheric breakdown on Earth, but well above the atmospheric breakdown fields of ∼ 25 kV/m for the low-pressure Martian atmosphere. Given that very similar analogous aeolian processes occur on Mars, large fields approaching breakdown are anticipated in that planet's dust devils/storms and dust frontal systems.
 A complementary set of papers by Delory et al.  and Atreya et al. [2006b] recently examined the behavior of ambient electrons in the Martian atmosphere under the influence of large E-fields (like that in dust devils/storms/fronts). They found that the electrons evolve in three ways: First, the electron density increases geometrically with increasing E, this due to increasing CO2 ionizations from energetic E-field driven electrons. Second, the electron drift velocity steadily increases with E reaching 5 × 105 m/s near 20 kV/m. Finally, a distinct high energy electron tail develops in the electron distribution, with its power-law rolloff in the 10's of eVs decreasing directly with increasing E [Nighan, 1970; Delory et al., 2006]. Basically, a collisional plasma develops in a mature dust devil/storm in the low pressure CO2 atmosphere. This plasma may even glow [Eden and Vonnegut, 1973; Farrell et al., 2005].
 As described by Delory et al.  and Atreya et al. [2006b], the energetic electrons also affect the ambient molecules in the low pressure gas. Specifically, electron dissociation of CO2 into CO and O− occurs near 4.4 eV. Without the E-field, the electrons do not possess the intrinsic energy to dissociate the molecule. However, with an electric field, an energetic portion of the electron population now exists (i.e., the tail of the distribution) with the required energy to dissociate the CO2 and this population (between 4 and 5 eV) is found to increase geometrically with increasing E. Water also undergoes a similar electron dissociation attachment at a number of resonant energies between 6 and 12 eV (the narrow resonance at 6.6 eV is dominant), and Delory et al.  found that the production of dissociated water products OH and H− also geometrically increases with increasing E-field. Atreya et al. [2006b] examined the subsequent recombination pathways for the dust storm E-field created species CO, O−, OH, and H− in the Martian near-surface atmosphere and found that anomalously large amounts of the reactive species H2O2 can be created in the atmosphere at concentrations exceeding 200 times that of regular photochemical production. The reactive product H2O2 was found to be at condensation levels and hence they conclude that the peroxide coats aeolian dust and the surface, and may possibly account for the presumed highly-reactive soil found by the Viking landers [Oyama et al., 1977].
 These tandem models suggest that the atmospheric conditions in Martian dust devils/storms/fronts are relatively harsh, with large E-fields, the creation of a collisional plasma (energetic electrons, CO2+, and negative ions), and the formation of reactive species. Given this new meteorology-driven electrochemical environment on Mars that is so very different from the fair-weather photochemical environment, we ask an obvious question: What is the stability of methane in Martian dust storms?
2. Model and Results
 The behavior of electrons in a low pressure CO2 gas requires the solution of the Boltzmann equation with inclusion of inelastic and elastic collisions. The resulting electron distribution has been derived by Nighan  and Delory et al.  with similar results. Figure 1 shows the distribution from Delory et al.  for the case of electrons in the Mars near-surface atmosphere. Note the development of an extended high energy electron tail (energy, u > 4 eV) that increases in population with increasing E-field.
 The electron impact ionization of CO2 occurs for those electrons with energies, u > 14 eV. Each impact will create an added electron thereby allowing exponential growth in electrons density defined by:
where α is Townsend's first ionizing coefficient (in units of inverse length) and x is an anode/cathode distance (or separation of the charging centers). The inverse of α represents the distance required for an electron to impact-ionize CO2 thereby freeing a second electron. As quantified by Delory et al. [2006, equations 7 and 8 and Figure 4], this distance ranges from 1010 cm for fields at 5 kV/m to 3 cm at 25 kV/m. We note that expression (1) applies for E-fields in pre-breakdown conditions. To consider discharges/full breakdown situations, a denominator term of the form (1 − F(ω)) would then be included in (1), where F(ω) is a function dependent upon the number of secondary electrons produced per unit length, ω. Note that when enough secondaries are produced to have 1 − F(ω) = 0, the electron density becomes infinite, signifying complete breakdown of the gas. Since any modeled secondary processes is not unique and would create an unnecessary contrivance, we simply exclude breakdown/discharge situations in our model (set F(ω) = 0), and consider our formalism in (1) as applicable to the pre-breakdown electrostatic environment in the dust devil.
 The energetic electrons are also capable of dissociating CO2 at energies near 4.4 eV [Delory et al., 2005]. At this resonance, an electron will dissociate CO2 into CO and O−. The production of CO/O− is describable via
where ke/CO2 is the chemical rate for the reaction and n is the density of the designated species. The chemical rate is [Ferreira, 1983]
where σd is the (e, CO2) dissociation cross section of ∼ 0.15 x 10−22 m2 [Itakawa, 2002], at uo of 4.4 eV and δu is the relatively-narrow cross section width of ∼ 1.0 eV. We make use of the narrow line width to derive the approximate form of k in (3). Delory et al.  found that the E-field production of CO/O− ranged from 1010m−3 s−1 at 5 kV/m to ∼1018m−3 s−1 at 25 kV/m. A similar calculation was made therein for (e, H2O), and subsequent products OH/H− were found to be at 107 m−3 s−1 at 5 kV/m to ∼1015m−3 s−1 at 25 kV/m (for a water loss Δn/n of ∼10−5 every second in large fields). As described in the complementary paper of Atreya et al. [2006b], the enhanced production of O and H in large E-fields was used as inputs to an atmospheric chemistry model which found that for large electric fields the atmospheric abundance of the recombination product H2O2 may exceed 200 times that of maximum photochemical model, with column density values of H2O2 at 7 × 1017 cm−2 and intermediate product OH at 8 × 1013 cm−2 [Atreya et al., 2006b]. This enhancement of the peroxide has implications in the biogeochemistry of Mars soil/surface (see discussion therein). Clearly, the electro-chemistry model indicates that the collisional plasma that forms in dust storms is a harsh electrochemical environment creating very reactive species via subsequent recombination of electron-dissociated products.
 Energetic electrons are also capable of dissociating methane at energies > 10 eV. Figure 2 shows the methane dissociation cross section (adapted from Ohmori et al. ). Unlike CO2 and H2O, the methane dissociation cross section is relatively broad, starting near 10 ev, peaks at 20 eV, and slowly decreasing beyond 100 eV. The methane loss via electron dissociation can be described via
where σd(u) is the methane dissociation cross section shown in Figure 2 and f(u) is shown in Figure 1. The electron density is defined by equation (1) and we assume that the ambient density of CH4 is nCH4 ∼ 10 ppb ∼ 2 × 1015 m−3.
Figure 3 shows the destruction of methane (Figure 3a) and methane destruction time ((ke/CH4ne)−1) (Figure 3b) as a function of dust devil/storm/front electric field. The line in Figure 3b indicates the photochemical lifetime of 1010 seconds (∼300 years). Note that as the E-field increases, the CH4 destruction time geometrically decreases and is close to ∼1000 seconds for large E-fields (Figure 4).
 The meteorological-driven electrochemical environment in dust devils/storms/fronts is a clear sink of atmospheric methane. The molecular destruction time of methane exponentially decreases with driving electric field as t(E) ∼ to exp(−β(E − Eo)) where to is 1011 seconds, Eo is 10 kV/m and β is 1.2 × 10−3 m/kV. The value of 1/β represents an e-folding of the value of t every ∼1 kV/m. We note that the global methane lifetime, even with dust storm losses, may only display a moderate decrease since we have not included sources from recycling of dissociated products back into CH4, including but not limited to such unexplored processes as reactions between energetic (hot) hydrogen atoms and methyl radicals, heterogeneous chemistry, atmospheric dynamics, etc. Also, the volumetric regions of largest E-fields still remained undefined for Mars. Once these elements of the problem become known, a new large-scale lifetime can be derived.
 It has been proposed (via the Mars Scout program) to fly a high resolution atmospheric composition spectrometer from orbit specifically targeting the sources and sinks of methane. Given the results herein, such an orbiting spectrometer should see temporal and spatial variations in CH4 in association with dust activity. At the largest time and spatial scales, we suggest that the global content of methane may have a seasonal dependence, with abundance being smaller during and immediately following the major dust storm season (just after southern summer), depending upon how fast methane recovers. On intermediate temporal/spatial scales (weeks/100s km), it is anticipated that methane concentrations will decrease in spatial regions behind dust fronts and in the near-vicinity of the dusty “cores” that form in the southern highlands in southern summer, just prior to the onset of major dust storms. Mesoscale missions that examine atmospheric composition in a region (i.e., airplane, balloon) should observe a correlation between passing dust storms and fronts, E-fields, and methane production. On the small temporal/spatial scales (10's minutes/∼1km), methane concentrations are anticipated to decrease at locations of local dust devil activity. This smallest scale might not be resolvable via orbiting spectrometers. However, landed assets such as the mass spectrometer and the tunable laser spectrometer that are part of the Sample Analysis at Mars (SAM) suite on the 2009 Mars Science Laboratory (MSL) are expected to note some decrease in methane with the passing of a moderate sized dust devil (lasting a few minutes) or in conjunction with a set of dust devils like that recently observed passing in succession by MER (http://marsrovers.jpl.nasa.gov/gallery/press/spirit/20050819a/dd_enhanced_568b-B558R1.gif). Laboratory simulation studies and further modeling are required to fully characterize the effect of electric fields on the nature of methane on Mars.
 The authors are grateful to their sponsored funding sources for this research: W.M.F. and G.T.D. are supported by NASA's Mars Fundamental Research Program and S.K.A. is supported by NASA's Mars Express and Mars Science Laboratory Mission programs.