Hot oxygen and carbon escape from the martian atmosphere

Hannes Gröller; Herbert Lichtenegger; Helmut Lammer; Valery I. Shematovich

doi:10.1016/j.pss.2014.01.007

Hot oxygen and carbon escape from the martian atmosphere

Gröller, Hannes;Lichtenegger, Herbert;Lammer, Helmut;Shematovich, Valery I. 2019-11-04 00:00:00 The escape of hot O and C atoms from the present martian atmosphere during low and high solar activity conditions has been studied with a Monte-Carlo model. The model includes the initial energy distribution of hot atoms, elastic, inelastic, and quenching collisions between the suprathermal atoms and the ambient cooler neutral atmosphere, and applies energy dependent total and diﬀerential cross sections for the determination of the collision probability and the scattering angles. The results yield a total loss rate 25 −1 of hot oxygen of 2.3 − 2.9 × 10 s during low and high solar activity conditions and + + is mainly due to dissociative recombination of O and CO . The total loss rates of 2 2 24 −1 carbon are found to be 0.8 and 3.2×10 s for low and high solar activity, respectively, with photodissociation of CO being the main source. Depending on solar activity, the obtained carbon loss rates are up to ∼ 40 times higher than the CO ion loss rate inferred from Mars Express ASPERA-3 observations. Finally, collisional eﬀects above the exobase reduce the escape rates by about 20-30% with respect to a collionless exophere. Key words: Mars, hot atom escape, hot corona density, heavy atoms, oxygen, carbon 1. Introduction The dynamics of the upper atmosphere and its interaction with the solar radiation and plasma environment are important clues for the stability and long-term evolution of terrestrial atmospheres. In a recent study Lammer et al. (2013) showed that the evo- lution and escape of the martian CO atmosphere and the planet’s H O inventory can 2 2 be separated from the planet’s origin into an early and a later evolutionary epoch. The ﬁrst epoch started after Mars’ formation and lasted ≤ 500 Myr (Lammer et al., 2013). Because of the high soft X-ray and EUV ﬂux of the young Sun (e.g. Ribas et al., 2005; Claire et al., 2012) and Mars’ low gravity the nebula-based and outgassed hydrogen- rich protoatmosphere was lost by hydrodynamic blow-oﬀ of H atoms and strong ther- mal escape rates of dragged heavier species such as O and C atoms within several Principal corresponding author Email address: herbert.lichtenegger@oeaw.ac.at (H. Lichtenegger) Preprint submitted to Planetary and Space Sciences December 25, 2013 arXiv:1911.01107v1 [astro-ph.EP] 4 Nov 2019 4 10 years (Erkaev et al., 2013, this issue). These recent results are in support of the hypotheses of Tian et al. (2009) that early Mars could not build up a dense CO atmo- sphere during the early Noachian. However, after the planet lost its protoatmosphere the atmospheric escape rates were most likely balanced with the volatiles (H O, CO , etc.) which have been out- 2 2 gassed by volcanic activity and delivered by impacts until the activity of the young Sun decreased, so that the atmospheric sources could dominate over the losses ∼ 4 − 4.2 Gyr ago (Lammer et al., 2013). If a secondary outgassed CO atmosphere of several 100 mbar was indeed present at the end of the Noachian, its evolution during the second epoch which lasted from the end of the Noachian until today should have been deter- mined by a complex interplay of various nonthermal atmospheric escape processes, impacts, carbonate precipitation, and serpentinization during the Hesperian and Ama- zonian epochs which led to the present day surface pressure (Chasseﬁe`re and Leblanc, 2011a,b). Atmospheres may get eroded through the loss of light particles like H or He, when their thermal velocities exceed the escape velocity at the exobase. Heavier species like O or C, are stronger bound to the planet and can therefore either directly escape through nonthermal processes like dissociative recombination of molecular ions or may form extended coronae which in turn can be eroded by the solar wind. From Mars Express ASPERA-3 data Barabash et al. (2007) estimated the present escape of CO 22 −1 molecular ions and yield low escape rates of ∼ 8 × 10 s . Based on this rate these authors estimated that the fraction of CO molecular ions lost to space since the end of the Noachian when the martian dynamo stopped to work would represent an equivalent surface pressure of only about ∼ 0.2 − 4 mbar. Lammer et al. (2013) compared these ion escape rates with atmospheric sputtering and exothermic escape processes of heavy neutrals and suggested that the loss of pho- tochemically produced suprathermal atoms such as O, C may be responsible for the main loss of the martian CO atmosphere at present and most likely also during the planet’s history until thermal escape took over. The simulation of this nonthermal photochemically related escape of heavy atmo- spheric atoms at terrestrial planets is not only challenged by the physical and chem- ical complexity of the upper atmosphere but is additionally hindered by our sparse knowledge of many of the processes which govern the direct escape of atmospheric constituents. Although Mars has been visited by many spacecraft for more than 40 years, observational data of the hot particle corona at Mars are still rather limited (e.g. Leblanc et al., 2006; Lichtenegger et al., 2006); the only recent observation of the ex- tended oxygen corona of Mars was made with the Alice far-ultraviolet imaging spec- trograph during the Rosetta swing-by of Mars (Feldman et al., 2011). Therefore, nu- merical modeling of exosphere densities and of escape rates are still little constrained and suﬀer from the various assumptions made by diﬀerent authors. The aim of this work is to obtain a better understanding on the escape of hot car- bon and oxygen atoms from present Mars. For this purpose we consider a number of possible sources of hot O and C atoms in the thermosphere, calculate the production rate proﬁles and the energy distribution functions at the exobase for low and high so- lar activity conditions and discuss the associated escape ﬂuxes and exosphere density proﬁles. Because there are little observational hints of the rotational and vibrational 2 excitation distributions of the particles and because these eﬀects have negligible inﬂu- ence on the escape rates (Gro¨ller et al., 2010), we assume all particles in the rotational and vibrational ground states. The simulations are based – if available – on the most recent data for both the atmospheric input and the collision cross sections. The model is limited by various circumstances: (a) the background atmosphere is represented by 1D ion and neutral density proﬁles (i.e. all variables depend only on radial distance) assumed to be valid for 60 solar zenith angle, (b) the hot particles are treated as test particles, i.e. their inﬂuence on the background gas is neglected, (c) due to the lack of data, some types of collisions are simulated by using approximate values of the parameters (see Tables 9 and 10). 2. Hot Atom Modeling 2.1. Monte Carlo model The simulation of the hot particle corona is initiated by calculating for a speciﬁc reaction the corresponding velocity distribution of the products at discrete altitudes be- tween 130 km and 450 km above the surface. These products are represented by a number of particles (typically ∼ 10 per 2.5 km altitude step) and their 3-dimensional motion is followed through the thermosphere up to 450 km in the gravitational ﬁeld of Mars. On their way the hot particles can interact with the background neutral at- mosphere via elastic, inelastic and quenching collisions and will loose on average part of their initial energy. Background particles which acquire an energy exceeding the thermal velocity of the atmospheric species upon collisions are also treated as new hot particles. The collision probability and energy transfer is calculated by means of total and diﬀerential cross sections. At the exobase (for comparison with other models) and at 380 km altitude, the energy distribution function of the suprathermal particles is de- termined which in turn serves as input for the exosphere density calculations. A more detailed description of the Monte-Carlo model can be found in Gro¨ller et al. (2010) and Gro¨ller et al. (2012). Although some 3D simulation results of the martian thermosphere have been pub- lished in the recent past (Valeille et al., 2010; Yagi et al., 2012), our background ther- mosphere model is assumed to be spherically symmetric for various reasons: (a) we do not focus on local variations such as solar zenith angle eﬀects; (b) the primary aim of our paper is the study of a number of chemical reactions producing hot particles in view of their relevance for escape as well as the treatment of the various types of collisions on the basis of total and diﬀerential cross sections rather than on hard sphere or universal potential models, and this analysis can be done most easily in 1D; (c) to facilitate the comparison with older models which are to a large extend 1D models; (d) due to the lack of suﬃcient observations, realistic 3D thermosphere models of Mars are yet diﬃcult to construct. 2.2. Input Proﬁles The neutral and ion density as well as the temperatures for both high and low solar activity are taken from Fox and Hac´ (2009) for a solar zenith angle of 60 . In their 3 model, the ambipolar diﬀusion for ions and the molecular and eddy diﬀusion for neu- trals are included as well as more than 220 reactions. The altitude dependent density proﬁles of the background atmosphere are based on measurements of the Viking 1 neutral mass spectrometer together with simulations of the Mars Thermospheric Gen- eral Circulation Model (MTGCM). According to Fox and Hac´ (2009), the O density proﬁles measured by Viking can only be reproduced by imposing an upward velocity boundary condition (”eroded” ionospheres); therefore, only these eroded models are used in the present paper as input for the ion densities. The solar ﬂux is taken from the Solar Irradiance Platform (SIP) version 2.36 (Tobiska, 2004), where we have chosen DOY 32 of 2002 to represent high solar ac- tivity (HSA) and DOY 110 of 1997 to represent low solar activity (LSA). For these dates, HSA and LSA correspond to a solar ﬂux of F = 245.6 and F = 69.7, 10.7 10.7 respectively. Finally, for each considered neutral element in the atmosphere, the cor- responding photo cross sections are taken from the photo ionization/dissociation rates database of Huebner et al. (1992). 2.3. Sources of Hot O Atoms Although dissociative recombination (DR) of O ions is supposed to be the main source of hot oxygen atoms (e.g. McElroy, 1972; Nagy and Cravens, 1988; Kim et al., 1998; Fox and Hac´, 2009; Gro¨ller et al., 2010, 2012), we have also considered some additional possible sources of hot O provided that the corresponding density proﬁles are available. These additional reactions are listed in Tables 2 to 6. 2.3.1. Dissociative and Radiative Recombination Dissociative recombination of ground-state O proceeds via the ﬁve exothermic 1 1 channels shown in Table 1. The excited states O( D) and O( S) are 1.97 eV and Table 1: Excess energies (e.g. Fox and Hac´, 2009) and branching ratios (Peverall et al., 2001) for dissociative + 2 recombination of O (X Π ). The branching ratios correspond to approximately zero collision energy. 2 g Reaction Channels Excess Energy [eV] Branching Ratios [%] + 3 3 O + e O( P) + O( P) 6.99 20.4 3 1 O( P) + O( D) 5.02 44.0 1 1 O( D) + O( D) 3.05 31.5 3 1 O( P) + O( S) 2.80 0.0 1 1 O( D) + O( S) 0.83 4.1 4.19 eV above the ground state O( P), respectively. For the ground vibrational state of the oxygen molecule O (v = 0), the branching ratios are taken from Peverall et al. (2001) for an electron-ion collision energy of nearly 0 eV. There are more recent val- ues published by Petrignani et al. (2005) but they state that “The branching fractions and quantum yields do not entirely agree with earlier measurements from our group that were taken at energies between 0 and 40 meV in a cold ion beam, nor with mea- surements performed at ASTRID by Kella et al. (1997).” Therefore and because the listed branching ratios are comparable to those of Kella et al. (1997), the values from Peverall et al. (2001) are chosen. 4 + The dissociative recombination of CO and the corresponding excess energies and branching ratios are listed in Table 2. The branching ratios are taken from Viggiano et al. (2005), where the uncertainties are given in parentheses. Of all channels, only the third and fourth produce atomic oxygen, although the latter is of no interest since it is an endothermic process and its probability is close to zero. The channels for dissociative recombination of NO are given in Table 3. The branching ratios taken from Vejby-Christensen et al. (1998) are valid for the vibrational ground state and for an electron-ion collision energy close to 0 eV. The excited states 1 2 O( D) and N( D) have energies of 1.97 eV and 2.38 eV above the corresponding ground states, respectively. The excess energies and branching ratios for the dissociative recombination of CO are listed in Table 4. The branching ratios for collision energies of 0.0, 0.4, 1.0, and 1.5 eV are taken from Rose´n et al. (1998). The altitude dependent production rate P(r) of the dissociative recombination of an ion can be calculated by P(r) = n (r) n (r) k(T (r)), (1) i e e where n (r) and n (r) are the altitude dependent ion and electron densities, respectively, i e and k(T (r)) is the rate coeﬃcient which depends on the electron temperature T (r). e e The value of k is determined according to the Arrhenius equation as T (r) 3 −1 k(T (r)) = α cm s . (2) The coeﬃcients α and β used for the dissociative recombination processes considered in this paper are listed in Table 5. Another possible source of hot O is radiative recombination of O + 4 3 O ( S) + e −−−→ O( P) + hν, where the parameters α and β for the rate coeﬃcient (2) are taken from the UDFA06 database (Woodall et al., 2007) based on Nahar (1999) and given in Table 5. However, radiative recombination appears to be rather ineﬃcient, yielding maximum production −4 −5 −3 −1 rates of about 1×10 and 2×10 cm s for high and low solar activity, respectively. The production rates as a function of altitude for the processes listed in Table 5 are illustrated in Figure 1 for low (a) and high (b) solar activity. Table 2: Excess energies (e.g. Fox, 2004) and branching ratios (Viggiano et al., 2005) for dissociative re- combination of CO ; the uncertainties are given in parentheses. Reaction Channels Excess Energy [eV] Branching Ratios [%] + 1 CO + e CO ( Σ ) 13.78 0.0 (+ 2.0) 2 2 g 3 3 C( P) + O ( Σ ) 2.29 0.0 (+ 4.0) 2 g 1 3 CO( Σ) + O( P) 8.27 100.0 (- 6.0) 3 3 C( P) + 2O( P) -2.87 0.0 5 Table 3: Excess energies and branching ratios (Vejby-Christensen et al., 1998) for dissociative recombination + 1 + of NO (X Σ ); the latter are valid for zero electron-ion collision energy. Reaction Channels Excess Energy [eV] Branching Ratios [%] + 4 3 NO + e N( S) + O( P) 2.78 15.0 4 1 N( S) + O( D) 0.81 0.0 2 3 N( D) + O( P) 0.39 85.0 2.3.2. Photodissociation A further possible source of hot oxygen is photodissociation of O O + hν −−−→ O + O, Table 4: Excess energies (e.g. Fox and Bakalian, 2001) and branching ratios (Rose´n et al., 1998) for disso- + 2 + ciative recombination of CO (X Σ ) for collision energies of 0.0, 0.4, 1.0, and 1.5 eV. Reaction Channels Excess Energy [eV] Branching Ratios [%] 0.0 eV 0.4 eV 1.0 eV 1.5 eV + 3 3 CO + e C( P) + O( P) 2.90 76.1 53.0 39.0 38.0 1 3 C( D) + O( P) 1.64 14.5 34.0 35.0 35.0 3 1 C( P) + O( D) 0.93 9.4 8.0 15.0 11.0 1 3 C( S) + O( P) 0.22 0.0 0.0 5.0 5.0 1 1 C( D) + O( D) -0.32 5.0 6.0 11.0 3 1 C( P) + O( S) -1.28 0.0 Table 5: Possible sources of hot O atoms due to dissociative and radiative recombination and references for the rate coeﬃcients. 3 −1 Source Rate coeﬃcient [cm s ] Dissociative recombination O + e −−−→ O + O Sheehan and St.-Maurice (2004) −7 T ≤ 1200 K : α = 2.40 × 10 , β = −0.70 −7 T > 1200 K : α = 1.93 × 10 , β = −0.61 CO + e −−−→ CO + O Viggiano et al. (2005) −7 α = 4.20 × 10 , β = −0.75 NO + e −−−→ N + O Sheehan and St.-Maurice (2004) −7 T ≤ 1200 K : α = 3.50 × 10 , β = −0.69 −7 T > 1200 K : α = 3.02 × 10 , β = −0.56 CO + e −−−→ C + O Rose´n et al. (1998) −7 α = 2.75 × 10 , β = −0.55 Radiative recombination O + e −−−→ O + hν Nahar (1999) −12 α = 3.24 × 10 , β = −0.66 6 300 + + O + e CO + e (a) 2 2 C + O CO + O + + NO + e CO + e N + NO N + O + + O + N NO + O + + O + O O + O 3 3 O + h ν O( P) + O( P) 1 1 O + h ν O( S) + O( S) CO + h ν C + O 0.01 0.10 1.00 10.00 100.00 1000.00 10000.00 -3 -1 Production rate [cm s ] + + O + e CO + e 2 2 (b) C + O CO + O + + NO + e CO + e N + NO N + O + + O + N NO + O O + O O + O 3 3 O + h ν O( P) + O( P) 1 1 O + h ν O( S) + O( S) CO + h ν C + O 0.1 1.0 10.0 100.0 1000.0 10000.0 -3 -1 Production rate [cm s ] Figure 1: Production rates of hot O atoms for low (a) and high (b) solar activity due to various sources of oxygen. for which an energy of at least 5.16 eV is needed (see Bond Dissociation Energies in Simple Molecules - B. deB. Darwent. NSRDS-NBS 31, (1970)) . The altitude dependent production rate P(r) for photodissociation of a molecular neutral n can be obtained by the relation pd P(r) = n (r) σ (λ)F(r, λ) dλ (3) n n where n (r) is the altitude dependent neutral density, F(r, λ) the solar ﬂux at the altitude pd r and σ (λ) the photodissociation cross section for the neutral at wavelength λ. The photodissociation and absorption cross sections are taken from Huebner et al. (1992) http://www.nist.gov/data/nsrds/NSRDS-NBS31.pdf Altitude [km] Altitude [km] 2 who provide them in the Photo Rate Coeﬃcient Database . For low and hight solar activity, the production rate of photodissociation of O and CO valid for a solar zenith angle of 60 is illustrated in Figure 1. As can be seen, photodissociation of CO – though signiﬁcant for the production of hot C (see Figure 2) – is of little importance for the production of hot O. 2.3.3. Chemical Reaction and Charge Transfer The rate coeﬃcient k for a two-body chemical reaction as well as for charge transfer is determined by means of the Arrhenius equation T −γ 3 −1 k = α exp cm s , (4) 300 T where T is the gas temperature. The coeﬃcients α, β, and γ are taken from the UDFA06 database described in Woodall et al. (2007) and are listed in Table 6. The production rate for the charge transfer between an ionized and a neutral oxygen atom is calculated −11 0.5 3 −1 using a rate coeﬃcient equal to 1.6×10 (T + T ) cm s taken from Terada et al. n i (2002) with T and T being the neutral and ion temperature, respectively. n i Table 6: Possible sources of hot O atoms with their corresponding coeﬃcients α, β, and γ to calculate the 3 −1 rate coeﬃcient k in cm s . The excess energies for chemical reactions are given for the ground state of the 3 1 + products, e.g. O( P), CO(X Σ ). 3 −1 3 −1 Source Rate coeﬃcient [cm s ] Production rate [cm s ] α β γ HSA LSA Chemical reaction −11 0 0 C + NO −−−→ CN + O + 1.22 eV 6.00 × 10 −0.16 0.0 < 3 × 10 < 2 × 10 1 a −11 C + O −−−→ CO + O( D) + 4.04 eV 4.70 × 10 −0.34 0.0 Figure 1b Figure 1a b −12 2.48 × 10 1.54 613.0 Figure 1b Figure 1a + + −11 −1 −1 C + O −−−→ CO + O + 4.04 eV 5.20 × 10 0.00 0.0 < 5 × 10 < 2 × 10 + + −10 0 −1 C + O −−−→ CO + O + 3.23 eV 3.80 × 10 0.00 0.0 < 2 × 10 < 10 + + −10 −1 0 H + CO −−−→ HCO + O + 0.83 eV 2.90 × 10 0.00 0.0 < 3 × 10 < 2 × 10 + + −9 −2 −2 He + CO −−−→ C + O + He + 2.19 eV 1.60 × 10 0.00 0.0 < 10 < 10 + + −10 −2 −2 He + CO −−−→ CO + O + He + 1.04 eV 8.70 × 10 0.00 0.0 < 10 < 10 + + −9 −2 −2 He + NO −−−→ N + O + He + 3.51 eV 1.40 × 10 0.00 0.0 < 10 < 10 + + −9 −2 −2 He + O −−−→ O + O + He + 5.81 eV 1.10 × 10 0.00 0.0 < 10 < 10 −11 N + NO −−−→ N + O + 3.26 eV 3.75 × 10 0.00 26.0 Figure 1b Figure 1a + + −10 N + O −−−→ NO + O + 4.17 eV 1.80 × 10 0.00 0.0 Figure 1b Figure 1a + + −11 −4 −5 N + NO −−−→ N + O + 2.23 eV 7.90 × 10 0.00 0.0 < 10 < 3 × 10 + + −10 −2 −3 N + O −−−→ NO + O + 6.65 eV 2.63 × 10 0.00 0.0 < 10 < 3 × 10 Charge transfer + + −10 −2 −1 O + H −−−→ H + O + 0.02 eV 5.66 × 10 0.36 −8.6 < 3 × 10 < 2 × 10 + + −11 0 −2 O + O −−−→ O + O + 1.55 eV 1.90 × 10 0.00 0.0 < 10 < 3 × 10 2 2 + + 0.5 −11 O + O −−−→ O + O + 0.00 eV 1.6 × 10 (T + T ) Figure 1b Figure 1a n i for T ≤ 300K for T > 300K http://phidrates.space.swri.edu (retrieved Nov 2012) 8 Table 6 includes all oxygen producing reactions – besides photodissociation and the reactions listed in Tables 1-5 – which are considered in this study. We note that 3 1 the second reaction in Table 6, i.e. C( P) + O mainly produces excited O( D) as mea- sured by Ogryzlo et al. (1973) and Costes and Naulin (1998), although some portion of ground state O( P) cannot be ruled out. For the present purpose we assume all oxygen products of this reaction beeing excited. The change of enthalpy ΔH which is equal to the excess energy and listed in Ta- ble 6 is obtained via Hess’s law. According to this law, the change of enthalpy for a chemical reaction can be calculated by means of the diﬀerence ΔH between the heat of formation of the products and the heat of formation of the reactants, i.e. X X ◦ ◦ ΔH = ΔH − ΔH . (5) f f products reactants If ΔH is negative, the reaction is exothermic, otherwise it is endothermic. The values for the heat of formation ΔH of the products and reactants are taken from the CCCBDB (Computational Chemistry Comparison and Benchmark Database) website . Figure 1 also includes the production rates according to the reactions listed in Table −3 −1 6 for which the maximum value is at least 1 cm s . 2.4. Main Sources of Hot C Atoms 2.4.1. Dissociative and Radiative Recombination Dissociative recombination of CO proceeds through the channels listed in Table 2. Atomic carbon is only produced via the second and fourth channel, however, since the fourth channel represents an endothermic process only the second channel is taken into account. The probability for this channel is low, however, because, as noted by Viggiano et al. (2005), it requires an unlikely rearrangement of the two oxygen atoms. + + − − The CO ion is a linear molecule O C O where the carbon is located in between − − the two oxygen atoms. In order to form the O molecule, the two oxygen atoms have to bond in a very short time. It should be noted, however, that according to Seiersen et al. (2003) the branching ratio of the second channel is 9%, in contrast to almost zero percent as reported by Viggiano et al. (2005). Dissociative recombination of CO takes place via the channels shown in Table 4. 3 1 C atoms can be produced in the ground state C( P) and in the excited states C( D) and C( S). The latter states have energies of 1.26 eV and 2.68 eV, respectively, above the ground state. Radiative recombination of C proceeds via the reaction + 3 C + e −−−→ C( P) + hν, where the parameters of equation (2) are taken from the UDFA06 database (Woodall et al., 2007) based on Nahar and Pradhan (1997). As for O , the production rates for radia- + −5 −3 −1 tive recombination of C are low, having a maximum of 10 cm s at 150 km for −6 −3 −1 high solar activity and of 3 × 10 cm s at 140 km for low solar conditions. NIST Computational Chemistry Comparison and Benchmark Database. NIST Standard Reference Database Number 101 Release 15b, August 2011; Editor: Russell D. Johnson III; http://cccbdb.nist.gov/ 9 The coeﬃcients α and β for dissociative and radiative recombination are given in Table 7. The correponding atomic carbon production rates are illustrated in Figure 2 for low and high solar activity. 2.4.2. Photodissociation In addition to dissociative recombination, photodissociation of CO is another im- portant source of hot C atoms 3 3 CO + hν −−−→ C( P) + O( P) with a threshold energy of 11.14 eV (see Bond Dissociation Energies in Simple Molecules - B. deB. Darwent. NSRDS-NBS 31, (1970)) . The production rate proﬁles of atomic carbon due to photodissociation of CO are included in Figure 2. 2.4.3. Chemical Reaction and Charge Transfer The coeﬃcients α, β, and γ for some additional reaction that produce hot atomic carbon are listed in Table 8. The corresponding production rates for high and low solar activity are also given, showing that hot carbon is created by these reactions with a low rate only; therefore they are not displayed in Figure 2. 2.5. Total and Diﬀerential Cross Sections The motion of the hot particles through the thermosphere is dominated by colli- sions with the background atmosphere. Thus the appropriate treatment of collisions is a crucial task in the simulation of hot coronae. Depending on the collision model, e.g. a hard sphere approximation or energy dependent cross sections, a wide range of outcome can be obtained (e.g. Lichtenegger et al., 2009; Gro¨ller et al., 2010). In the present paper, collisions are modeled by using total and diﬀerential cross sections. Collisions between energetic particles and neutral background constituents can be either elastic, inelastic, or quenching. In elastic collisions the kinetic energy is con- served, only the scattering angle can change. During inelastic collisions kinetic energy http://www.nist.gov/data/nsrds/NSRDS-NBS31.pdf Table 7: Possible sources of hot C atoms due to dissociative and radiative recombination and references of the rate coeﬃcients. 3 −1 Source Rate coeﬃcient [cm s ] Dissociative recombination CO + e −−−→ C + O Viggiano et al. (2005) 2 2 −7 α = 4.20 × 10 , β = −0.75 CO + e −−−→ C + O Rose´n et al. (1998) −7 α = 2.75 × 10 , β = −0.55 Radiative recombination C + e −−−→ C + hν Nahar and Pradhan (1997) −12 T ≤ 7950 K : α = 4.67 × 10 , β = −0.60 10 300 (a) CO + e CO + h ν C + O CO + e 0.01 0.10 1.00 10.00 100.00 -3 -1 Production rate [cm s ] CO + h ν C + O (b) CO + e CO + e 0.01 0.10 1.00 10.00 100.00 -3 -1 Production rate [cm s ] Figure 2: Production rates of hot C atoms for low (a) and high (b) solar activity due to diﬀerent sources of carbon. can be converted into vibrational energy and into electronic excitation, the latter being usually less likely. Quenching collisions describe the de-excitation of the reactant by collision, i.e. the conversion of the internal energy of the reactant into kinetic energy. The references for the used total and diﬀerential collision cross sections (σ and dσ) as well as the fraction of energy ( f ) which goes into internal energy can be found in Tables 9 and 10. Inelastic collision of two oxygen atoms are not considered because the excitation energy in this case is higher than the energy released during the collision. 3 1 1 2.5.1. Collisions between O( P, D, S) and O Collisions between energetic (hot) oxygen and a neutral thermal atom are either elastic 3 1 1 3 1 1 O ( P, D, S) + A → O ( P, D, S) + A , (6) th h h h Altitude [km] Altitude [km] where A denotes the thermal background particles and O the hot oxygen, or quench- th h ing 1 1 3 O ( D, S) + A → O ( P) + A + ΔE , (7) th h q h h 1 1 with the released energy ΔE being 1.97 eV and 4.19 eV for O( D) and O( S), respec- tively. Since no data are available for the diﬀerential cross section of quenching colli- 1 1 sions of excited O( D) and O( S) atoms, we assume them to be similar to the elastic collisions of O( D) atoms. The cross sections simulated by Yee and Dalgarno (1985), Yee and Dalgarno (1987), Yee et al. (1990), and Tully and Johnson (2001) for various types of collisions are illustrated in Figure 3a. Since hot particles may gain energies up to about 10 eV, the values for the cross sections are extrapolated to this interval. 3 1 1 2.5.2. Collisions between O( P, D, S) and N Collisions between a suprathermal O atom and a neutral atmospheric background molecule can be Table 8: Possible sources of hot C atoms and the corresponding coeﬃcients α, β, and γ to calculate the rate 3 −1 coeﬃcient k in cm s . The excess energies for chemical reactions are given for the ground state of the 3 1 + products, e.g. C( P), CO(X Σ ). 3 −1 3 −1 Source Rate coeﬃcient [cm s ] Production rate [cm s ] α β γ HSA LSA Chemical reaction + + −10 −3 −3 C + NO −−−→ NO + C + 1.99 eV 5.20 × 10 0.0 0.0 < 6 × 10 < 4 × 10 + + −11 −4 −4 He + CO −−−→ O + C + He + 1.04 eV 1.10 × 10 0.0 0.0 < 10 < 2 × 10 2 2 + + −10 −2 −2 N + CO −−−→ NO + C + 0.67 eV 1.45 × 10 0.0 0.0 < 3 × 10 < 10 Table 9: Total and diﬀerential collision cross sections (σ, dσ) for elastic and quenching collisions between O atoms together with the available energy ranges. Hot Particle Collision Type References O( P) elastic Tully and Johnson (2001) σ: 1 − 10 eV (Figure 4) dσ: 1, 5, and 10 eV (Figure 2) O( D) elastic Yee and Dalgarno (1987) σ: 0 − 5 eV (Figure 4) dσ: 0.1 and 1.0 eV (Figure 2b and 3b) quenching Yee et al. (1990) σ: 0.001 − 5 eV (Figure 4) dσ: like O( D) elastic O( S) elastic Yee and Dalgarno (1985) σ: 0 − 5 eV (Figure 4) dσ: 1 and 5 eV (Figure 2b and 3b) quenching σ: like O( D) elastic dσ: like O( D) elastic 12 Table 10: Total and diﬀerential collision cross sections (σ, dσ) used in this study for elastic, inelastic, and quenching collisions between O and N , as well as the fraction ( f ) of the kinetic energy which changes into internal energy of the colliding molecule together with the available energy ranges. Hot Particle Collision Type References O( P) elastic Balakrishnan et al. (1998b) σ: 0 - 3 eV (Figure 1) Balakrishnan et al. (1998a) dσ: 1, 2, and 3 eV (Figure 3 to 5) inelastic Balakrishnan et al. (1998b) σ: 0 - 3 eV dσ: 3 eV (Figure 2) f : 0 - 3 eV (Figure 8) O( D) elastic Balakrishnan et al. (1999) σ: 0 - 2 eV (Figure 1) dσ: like O( P) inelastic σ: like O( P) dσ: like O( P) Tachikawa et al. (1997) f : 0 - ∼ 0.8 eV (Table 1) quenching Matsumi and Sarwaruddin Chowdhury (1996) σ: 0 - ∼ 0.9 eV (Figure 9) dσ: like O( P) Zahr et al. (1975) f : 0 - 1.2 eV (Table II) O( S) elastic, inelastic, and quenching like O( D) (a) elastic 3 1 1 3 1 1 O ( P, D, S) + M → O ( P, D, S) + M , (8) th h h h where M denotes the thermal background molecule and O the hot oxygen atom, th h (b) quenching 1 1 3 O ( D, S) + M → O ( P) + M + ΔE , (9) th h q h h with the released energy ΔE being a fraction of the excitation energies, or (c) inelastic 3 1 1 3 1 1 O ( P, D, S) + M → O ( P, D, S) + M − ΔE , (10) th h i h h where ΔE is the kinetic energy which is converted into internal energy of the products. The total and diﬀerential collision cross sections for the collisions between O( P, 1 1 D, S) atoms and N molecules are taken from the references listed in Table 10. In addition, energy ranges for which the cross sections are available are given together with the fraction f of the kinetic energy which goes into internal energy of the col- liding molecule. The total collision cross section for inelastic O( P)−N collisions is 13 10000.00 Tully and Johnson (2001) Yee and Dalgarno (1987) Yee and Dalgarno (1985) 1000.00 Yee et al. (1990) 100.00 -15 2 3 x 10 cm 10.00 1.00 0.10 0.01 0.001 0.010 0.100 1.000 10.000 Energy [eV] 1000.0 Balakrishnan et al. (1998b) elastic: Balakrishnan et al. (1999) 100.0 -15 2 3 x 10 cm 10.0 1.0 inelastic: Balakrishnan et al. (1998b) quenching: Matsumi and Sarwaruddin Chowdhury (1996) 0.1 0 2 4 6 8 10 Energy [eV] Figure 3: Total collision cross sections for (a) elastic and quenching collisions between O atoms, (b) elastic, inelastic, and quenching collisions between O atoms and N molecules. The indicated constant value of −15 2 3 × 10 cm was used by some others in the past. calculated according to the fraction of the frequencies for elastic and inelastic colli- sions given in Figure 6 of Balakrishnan et al. (1998b). The fraction f of the energy which is deposited into N vibrational and rotational excitation in percent of quenching O( D)−N collisions is obtained from Figure 3 and Equation 4.2 of Zahr et al. (1975) for the three given collision energies of 0.03, 0.6, and 1.2 eV. For elastic collisions between O( P) and N no diﬀerential cross sections are available, therefore the dif- ferential cross sections for N−N collisions published in Balakrishnan et al. (1998a) (Figures 3 to 5) are used. The cross sections of Zahr et al. (1975), Tachikawa et al. (1997), Balakrishnan et al. (1998b), and Balakrishnan et al. (1999) are based on sim- ulations whereas those of Matsumi and Sarwaruddin Chowdhury (1996) are obtained from measurements. The total collision cross sections are shown in Figure 3b together with the constant cross section used by e.g. Fox and Hac´ (2009). Since the simulated particles can achieve energies up to about 10 eV, the cross sections are extrapolated to this interval. -16 2 -16 2 Cross section [10 cm ] Cross section [10 cm ] 3 1 1 2.5.3. Collisions between O( P, D, S) and other atoms and molecules Due to the lack of data for collisions between hot oxygen and CO or CO, the same total and diﬀerential cross sections and energy loss rates as for O−N collisions are employed. Collisions of hot oxygen with H and He are treated by means of the elastic collision cross section for O−H collisions given in Zhang et al. (2009). For the collisions of hot oxygen atoms with other background atoms, the data for O−O collisions are used. 3 1 1 2.5.4. Collisions between C( P, D, S) and atoms or molecules Since no data for collisions between hot atomic carbon and atoms or molecules of the background atmosphere are available, the same total and diﬀerential cross sections and energy loss rates as for collisions of hot oxygen with atoms and molecules are assumed, see Tables 9 and 10. 3. Results and Discussion According to the input parameters used in our simulations, the calculated exobase altitude for low and high solar activity at Mars is located at about 220 and 260 km, respectively, above the surface. Once the energy distribution functions of hot O and C atoms at the exobase, at 380 km and at 450 km altitude are determined, escape ﬂuxes as well as density proﬁles of the suprathermal atoms can be estimated. Assuming that the SZA=60 eroded models represent an average over the martian dayside, global escape rates are obtained by integrating the escape ﬂuxes based on the according EDF and by multiplying the resulting total escape ﬂux by the hemispheric area of Mars. Table 11: Escape ﬂuxes and hemispheric loss rates of hot oxygen atoms according to the present study. The numbers are based on the EDFs at the exobase altitude, 380 km and 450 km altitude, respectively. −2 −1 −1 Source Escape ﬂux [cm s ] Escape rate [ s ] Low solar activity 220 km 380 km 450 km 220 km 380 km 450 km + 7 7 7 25 25 25 O + e −−−→ O + O 1.8 × 10 1.4 × 10 1.3 × 10 1.5 × 10 1.2 × 10 1.2 × 10 + 7 7 7 25 25 25 CO + e −−−→ CO + O 1.5 × 10 1.2 × 10 1.1 × 10 1.3 × 10 1.0 × 10 1.0 × 10 1 5 5 5 23 23 23 O + C −−−→ CO + O( D) 8.4 × 10 6.1 × 10 5.8 × 10 6.9 × 10 5.5 × 10 5.3 × 10 + 5 5 5 23 23 23 O + N −−−→ NO + O 1.0 × 10 0.7 × 10 0.7 × 10 0.8 × 10 0.6 × 10 0.6 × 10 7 7 7 25 25 25 total 3.4 × 10 2.7 × 10 2.5 × 10 2.9 × 10 2.3 × 10 2.3 × 10 High solar activity 260 km 380 km 450 km 260 km 380 km 450 km 7 7 7 25 25 25 O + e −−−→ O + O 2.5 × 10 2.1 × 10 2.0 × 10 2.1 × 10 1.9 × 10 1.9 × 10 + 7 7 7 25 25 25 CO + e −−−→ CO + O 1.3 × 10 1.0 × 10 1.0 × 10 1.1 × 10 0.9 × 10 0.9 × 10 + 5 5 5 23 23 23 O + N −−−→ NO + O 11.4 × 10 9.2 × 10 8.5 × 10 9.6 × 10 8.2 × 10 7.8 × 10 1 5 5 5 23 23 23 O + C −−−→ CO + O( D) 8.1 × 10 5.9 × 10 5.5 × 10 6.8 × 10 5.2 × 10 5.1 × 10 7 7 7 25 25 25 total 4.0 × 10 3.3 × 10 3.1 × 10 3.4 × 10 2.9 × 10 2.9 × 10 15 + Table 12: Comparison of hot oxygen escape rates of various studies due to DR of O for low (LSA) and high solar activity (HSA). 25 −1 References Escape rate × 10 s LSA HSA present study (at the exobase level) 1.5 2.1 present study (at 450 km) 1.2 1.9 Kim et al. (1998) 3.4 8.5 Hodges (2000) 2.8 - Hodges Jr. (2002) 4.4 18.0 Krestyanikova and Shematovich (2006) 4.5 - Cipriani et al. (2007) 3.4 8.5 Cipriani et al. (2007) 0.55 2.6 Chaufray et al. (2007) 1.0 4.0 Fox and Hac´ (2009) 0.66 0.68 Fox and Hac´ (2009) 14.4 21.0 Valeille et al. (2010) 3.8 13 Yagi et al. (2012) 2.9 - 5.3 - corrected by a factor of 6.5 by Nagy et al. (2001); two-stream model and constant collision cross section of −15 2 1.2 × 10 cm b −15 2 constant collision cross section of 2 × 10 cm for O−O and O−CO c −15 2 1 constant collision cross section of 2 × 10 cm for O−O and O−CO and including quenching of O( D) and O( S) Model B: including elastic, inelastic and quenching collisions; collision cross sections are taken from Balakrishnan et al. (1998b) and Kharchenko et al. (2000) atmosphere inputs from Kim et al. (1998) (corrected values) and the collision cross sections taken from Kharchenko et al. (2000) and Tully and Johnson (2001) atmosphere inputs from Krasnopolsky (2002) and the collision cross sections taken from Kharchenko et al. (2000) and Tully and Johnson (2001) single interaction potential and only atomic background atmosphere eroded ion proﬁles, exobase approximation, isotropic scattering and constant collision cross section of −15 2 3 × 10 cm eroded ion proﬁles, forward scattering using diﬀerential cross sections from Kharchenko et al. (2000) and −15 2 constant collision cross section of 3 × 10 cm j −15 2 3D thermospheric inputs from Bougher et al. (2006); constant collision cross section of 2 × 10 cm for O−O and O−CO collisions universal potential model for collisions; variations due to diﬀerent martian seasons 16 3.1. Hot O Atoms The escape ﬂuxes together with the escape rates for various sources of hot oxygen are listed in Table 11 for LSA and HSA. Since the escape rates are calculated from the EDF, they depend on the altitude at which the EDF is taken. At suﬃciently high altitudes, however, where collisions become insigniﬁcant, the escape rates converge towards a constant value. To illustrate this eﬀect, the escape rates based on the EDFs at three diﬀerent altitudes are shown in Table 11. Although the production rates are generally larger during HSA, there is little dif- ference in the escape of hot O between HSA and LSA. This is mainly due to the fact that at HSA also the neutral background density is higher, leading to an increase in the number of collisions between the suprathermal particles and the cold atmosphere and thus to a more eﬃcient thermalization of the hot particles compared to LSA conditions. It should be noted that this eﬀect is only found for the eroded ionosphere models. In case of the non-eroded ionosphere models of Fox and Hac´ (2009) the higher production rates of hot O at HSA is not totally compensated by the more eﬃcient thermalization in the denser atmosphere; indeed in that case the escape rate of hot oxygen produced from O is about 65% higher at HSA than at LSA. Moreover, inspection of Table 11 also reveals that the escape ﬂux due to DR of CO is smaller for HSA than for LSA. This is due to the fact that the production rates in the region close to the exobase are comparable for LSA and HSA (∼0.1 and ∼0.3 −3 −1 cm s ), while the initial energies of the newly produced hot O is diﬀerent because of the diﬀerent ion temperatures at LSA and HSA. The loss rates for both solar conditions are mainly sustained by particles with initial energies between 4.7 . E . 5.8 eV whose production rates are higher at LSA than at HSA. Thus, although the number of particles with E > 5.8 eV is increased at HSA with respect to LSA, the number of particles with E < 5.8 eV is decreased, leading to a net escape ﬂux at HSA smaller than at LSA. The same eﬀect can also be seen for the O + C reaction. In case of O + N, 2 2 however, the escape ﬂux for HSA is higher than for LSA. In this case, the eﬀect of the initial energy distribution plays a minor role with respect to the higher production rate at higher altitudes. Table 11 also illustrates the eﬀect of ignoring the production of hot particles and their collisions with the background atmosphere above the exobase. The loss rates are found to be higher when based on the EDF at the exobase than when based on the EDF at 450 km altitude. This shows that the hot particles produced above the exobase only marginally contribute to the total escape despite the fact that most of these particles will be lost to space if their energy exceeds the escape energy. Rather, the decrease of escaping oxygen due to energy loss in collisions above the exobase somewhat exceeds the gain due to production, so that the escape rate is in fact reduced at 450 km by ∼ 20− 30% with respect to the values at the exobase altitude. In general our simulation results suggest that collision eﬀects are still noticeable up to ∼ 5 − 10 scale heights above the exobase. In Table 12 our escape rates are compared to previous simulations. The diﬀer- ences in the results mainly reﬂect the diﬀerences in the input neutral and ion density proﬁles and the diﬀerent treatment of collisions. Although we use the neutral density 17 350 escape - LSA (eroded) source - LSA (eroded) escape - HSA (eroded) source - HSA (eroded) escape - LSA (non-eroded) source - LSA (non-eroded) 250 escape - HSA (non-eroded) source - HSA (non-eroded) (a) O + e 0.01 0.10 1.00 10.00 100.00 1000.00 10000.00 -3 -1 Production rates [cm s ] escape - LSA (eroded) source - LSA (eroded) escape - HSA (eroded) source - HSA (eroded) escape - LSA (non-eroded) source - LSA (non-eroded) 250 escape - HSA (non-eroded) source - HSA (non-eroded) (b) CO + e 0.001 0.010 0.100 1.000 10.000 100.000 1000.000 -3 -1 Production rates [cm s ] + + Figure 4: Production rates of escaping particles generated by DR of O (a) and CO (b) as a function of 2 2 altitude for both solar activities and for the eroded and non-eroded ionosphere models. The dashed-dotted lines represent the production rates of all (escaping and non-escaping) oxygen atoms created via DR of O (a) and CO (b). and ionosphere model of Fox and Hac´ (2009), the escape rates diﬀer by a factor of 0.5 − 10. The main reason may be attributed to the diﬀerent collision models: While −15 2 Fox and Hac´ (2009) uses a constant collision cross section of 3×10 cm for all types of collisions, we invoke diﬀerent energy dependent collision cross sections (see Tables 9 and 10). In particular, the cross section for collisions with molecules – which are the most important types of collisions up to ∼ 200 km – is about one order of magnitude higher (see Figure 3b) than the above constant value. This higher value signiﬁcantly increases the average number of collisions, thus leading to more eﬃcient thermaliza- tion and to lower escape rates. Finally, while most previous studies reveal a distinct dependence of the hot oxygen loss rates on solar activity, this is neither the case in our results nor in those of Fox and Hac´ (2009) and may be attributed to the use of an eroded ionosphere model. Figure 4 illustrates the production rates of escaping O re- Start Altitude [km] Start Altitude [km] 6000 total O + e CO + e CO + h ν 4000 C + O CO + O + + O + N NO + O CO + e NO + e (a) -1 0 1 2 3 4 5 10 10 10 10 10 10 10 -3 Density [cm ] total O + e CO + e CO + h ν 4000 C + O CO + O + + O + N NO + O CO + e NO + e (b) -1 0 1 2 3 4 5 10 10 10 10 10 10 10 -3 Density [cm ] Figure 5: Exosphere densities of hot oxygen atoms for various sources for low (a) and high (b) solar activity. + + sulting from the DR of O and CO as a function of altitude for both solar activities. 2 2 For comparison, the results based on the eroded and non-eroded ionosphere models of Fox and Hac´ (2009) are shown. The solid lines represent the production rates at each height of those particles (consisting of hot particles generated by photochemical reactions and of background particles which became ”hot” upon collisions with these particles) which arrive at 380 km altitude with an energy exceeding the escape energy. As can be seen, the production rates for the non-eroded models become larger than those for the eroded ones approximately at altitudes, where the O density proﬁles become sensitive to the upward velocity. At higher altitudes, the production rate of escaping hot O created by collisions can even exceed the local production rate of O due to DR of CO . The exosphere density proﬁles based on the energy distribution functions of hot oxygen at the exobase for low and high solar activity are shown in Figure 5. For both low and hight solar activity conditions the total exosphere density is mainly governed Altitude [km] Altitude [km] 1500 (a) this study Kim et al. (1998) Hodges (2000) - noon Krestyanikova and Shematovich (2006) Cipriani et al. (2007) Chaufray et al. (2007) Valeille et al. (2010) - SZA = 60° Feldman et al. (2011) Yagi et al. (2012) 2 3 4 5 10 10 10 10 -3 Density [cm ] (b) this study Kim et al. (1998) Hodges (2000) - noon 750 Cipriani et al. (2007) Chaufray et al. (2007) Valeille et al. (2010) - SZA = 30° Valeille et al. (2010) - SZA = 90° 2 3 4 5 10 10 10 10 -3 Density [cm ] Figure 6: Comparison of exosphere densities of hot oxygen atoms for a number of studies. The curves of Hodges (2000), Chaufray et al. (2007), and Yagi et al. (2012) include also the cold component. + + by DR of O and CO . During HSA, however, at altitudes below ∼ 4000 km, the 2 2 oxygen density due to photodissociation of CO becomes comparable with the density of O produced via DR of CO . It should be noted that, although the production rate of hot O due to NO + e is between one and two orders of magnitudes higher than the one due to CO + e (see Figure 1), the obtained exosphere densities are comparable. The main reason is that the initial energies of the O atoms originating from NO +e are on average much lower than those originating from CO + e. Moreover, although the oxygen production rates due to the reactions C + O and NO + e are comparable above ∼ 130 km (Figure 1), the corresponding exosphere density arising from C + O for HSA and LSA is nearly one order of magnitude higher than that arising from NO + e (Figure 5). This is due to the initial velocities of the oxygen atoms gained via DR of NO , which are much lower than those acquired via the reaction between C and O . Fur- ther, although the production rate of atomic oxygen originating from charge transfer between an ionized and a neutral O becomes dominant at higher altitudes (Figure 1), the contribution of this reaction to the exospheric density is negligible because the ini- Altitude [km] Altitude [km] tial energies involved are similar to the average ion energies of ∼ 0.3 eV. For HSA as well as LSA the production rate of oxygen atoms with an excess energy of 4.17 eV due to reaction O + N is about three orders of magnitude smaller than the production rate due to DR of of O and thus leading also to much lower exosphere densities. Finally, at low solar activity, the oxygen exosphere density resulting from the reaction O + N is about 5-10 times smaller than the O density produced via C+O (Figure 5a) because above ∼ 130 km the production rate is also smaller by a similar amount (Figure 1a). At high solar activity, however, the production rates of these two reactions above ∼ 180 km become comparable (Figure 1b), making also the densities comparable (Figure 5b). In Figure 6 our exosphere densities are compared with previous studies. This Figure reﬂects the variety of results obtained based on diﬀerent assumptions. The use of en- ergy dependent total and diﬀerential cross sections and the inclusion of elastic, inelastic and quenching collisions as done in this study (see Tables 9 and 10) appear to favour lower exosphere densities than other assumptions (Table 12). In addition, in our sim- ulations the collision of ground-state oxygen atoms with the atmospheric background is modeled by means of the total and diﬀerential cross sections of Tully and Johnson (2001) (see Table 9). As shown by Gro¨ller et al. (2010) the use of the corresponding cross sections published by Kharchenko et al. (2000) would lead to somewhat higher densities. 3.2. Hot C Atoms The escape ﬂuxes as well as the escape rates for the most eﬃcient sources of hot C + + are listed in Table 13. As opposed to oxygen, where DR of O and CO are the most 2 2 important suppliers for hot O escape, the C loss is mainly due to photodissociation of CO. As in the case of oxygen, the higher production rates at HSA are at least partially compensated by the simultaneously increased number of collisions with the denser background atmosphere, leading to escape ﬂuxes at HSA of about four times the ﬂux at LSA. Regarding DR of CO , the corresponding carbon escape ﬂux at HSA is even lower than at LSA. It should also be noted that the carbon production and escape rate due to photodissociation of CO, i.e. of a neutral molecule, is not aﬀected by the erosion of the ionosphere, in contrast to the production and escape of oxygen due to DR of O - and CO -ions. Finally, the total loss rates calculated by means of the EDFs at 450 km altitude are about ∼ 25− 30% smaller than the loss rates obtained by using the EDFs at the exobase level. Thus collisions above the exobase are again still suﬃcient to reduce the ﬂux of escaping carbon by a noticeable amount. In Table 14 the escape rates of hot atomic carbon due to various source terms obtained in previous simulations are shown. As discussed in Fox (2004), a detailed analysis of the diﬀerences among the escape ﬂuxes obtained by Nagy et al. (2001), Fox and Bakalian (2001), and Fox (2004) is diﬃcult, but may be related to the use of diﬀerent solar ﬂuxes, dissociative recombination coeﬃcients and exobase altitudes. The results of Cipriani et al. (2007) are signiﬁcantly lower than the aforementioned values; unfortunately, these authors do not discuss any possible reasons for this dis- crepancy. Upon comparing our computed carbon escape ﬂuxes (Table 13) with those pub- lished in Fox (2004) (Table 14), it is seen that the ﬂuxes at the exobase due to pho- todissociation of CO are higher by a factor of ∼ 1.5 and ∼ 1.3 for LSA and HSA, 21 Table 13: Escape ﬂuxes and hemispheric loss rates of hot carbon atoms according to the present study. The numbers are based on the EDFs at the exobase altitude, 380 km and 450 km altitude, respectively. −2 −1 −1 Source Escape ﬂux [cm s ] Escape rate [ s ] Low solar activity 220 km 380 km 450 km 220 km 380 km 450 km 5 5 5 23 23 23 CO + hν −−−→ C + O 10.6 × 10 7.8 × 10 7.1 × 10 8.7 × 10 6.9 × 10 6.5 × 10 + a 5 5 5 23 23 23 CO + e −−−→ O + C 1.4 × 10 1.0 × 10 0.9 × 10 1.1 × 10 0.9 × 10 0.8 × 10 2 2 5 5 5 23 23 23 CO + e −−−→ C + O 0.8 × 10 0.7 × 10 0.6 × 10 0.7 × 10 0.6 × 10 0.6 × 10 5 5 5 23 23 23 total 12.8 × 10 9.5 × 10 8.6 × 10 10.5 × 10 8.4 × 10 7.9 × 10 High solar activity 260 km 380 km 450 km 260 km 380 km 450 km 5 5 5 24 24 24 CO + hν −−−→ C + O 45.1 × 10 34.4 × 10 30.4 × 10 3.8 × 10 3.1 × 10 2.8 × 10 + 5 5 5 24 24 24 CO + e −−−→ C + O 3.4 × 10 3.3 × 10 3.3 × 10 0.3 × 10 0.3 × 10 0.3 × 10 a 5 5 5 24 24 24 CO + e −−−→ O + C 1.2 × 10 0.8 × 10 0.8 × 10 0.1 × 10 0.1 × 10 0.1 × 10 2 2 5 5 5 24 24 24 total 49.7 × 10 38.5 × 10 34.5 × 10 4.2 × 10 3.5 × 10 3.2 × 10 the branching ratio of this process is assumed to be 4% respectively. Our escape ﬂuxes at the exobase related to DR of CO are about a factor of ∼ 4.5 and ∼ 2.4 higher for low and high solar activity, respectively, whereas our escape ﬂuxes due to DR of CO are about a factor of ∼ 1.6 and ∼ 2.3 lower than the one given in Fox (2004) who assumes the exobase locations at 184 and 220 km altitude for low and high solar acitivity. Besides the diﬀerences in the exobase heights, these latter lower values can be explained mainly by our lower production rates (Figure 1) which are based on the CO density proﬁles of Fox and Hac´ (2009) being smaller than those given in Fox (2004). The higher escape rates due to DR of CO compared to Fox (2004) may be some- what surprising at ﬁrst glance, since we assume a branching ratio of only 4% for the C + O channel of DR of CO , which is lower than the 9% assumed by Fox (2004). 2 2 Moreover, our production rates of carbon are smaller than those calculated in Fox (2004) because we use the CO proﬁles of Fox and Hac´ (2009) which correspond to lower densities than those in Fox (2004). That our escape rates of carbon produced by DR of CO are nevertheless higher is at least partially due to the fact that we consider a higher initial velocity for carbon. While Fox (2004) assumes that the vibrational- rotational states of the resulting O molecule and the translational states of the C and O fragments are populated statistically, we assume O to be in the electronic and 2 2 vibrational-rotational ground state, which can be considered as an upper limit for the resulting velocities of the C atoms. In the latter case the velocity distribution peaks at ∼ 1.7 eV, in the former case at ∼ 0.6 eV. Another reason for our higher carbon escape rates is due to the diﬀerent determination of these rates. Fox (2004) applies the exobase approximation, where the escape ﬂux is estimated by integrating the production rates above the exobase. Our estimation, however, is based on the integration of the energy distribution function, yielding higher escape rates than the former method – especially in case of ﬂat production rate proﬁles like the one for CO + e. 22 Table 14: Escape ﬂuxes of hot atomic carbon obtained in previous studies. 5 −2 −1 Reaction Escape ﬂux × 10 cm s low solar activity high solar activity Nagy et al. (2001) hemispheric escape ﬂuxes CO + hν −−−→ C + O - 60 CO + e −−−→ C + O - 17 total 5.33 77 Fox and Bakalian (2001) global average escape ﬂuxes CO + hν −−−→ C + O 1.65 18 CO + e −−−→ C + O 0.29 6.2 total 1.94 24.2 total (including additional sources) 2.1 26 Fox (2004) hemispheric escape ﬂuxes CO + hν −−−→ C + O 7.3 35 CO + e −−−→ C + O 1.28 7.7 CO + e −−−→ C + O 0.31 0.49 2 2 total 8.89 43.19 total (including additional sources) 11.1 64 Cipriani et al. (2007) hemispheric escape ﬂuxes CO + e −−−→ C + O 0.005 0.45 a −15 2 cross section of 1.2 × 10 cm ; for LSA only the sum of the rates for the two reactions is given by Nagy et al. (2001) b −15 2 cross section of 3 × 10 cm ; exobase at ∼ 188 km for LSA and ∼ 220 km for HSA hemispheric escape ﬂuxes are twice as large atmosphere inputs from Kim et al. (1998) and the collision cross sections used from Kharchenko et al. (2000) and Tully and Johnson (2001) e −1 calculated from the escape rates [s ] given in Table 1 of Cipriani et al. (2007) by dividing through the martian hemispheric surface at 400 km altitude 23 6000 total CO + h ν CO + e CO + e (a) 0.1 1.0 10.0 100.0 1000.0 -3 Density [cm ] total CO + h ν CO + e CO + e (b) 0.1 1.0 10.0 100.0 1000.0 -3 Density [cm ] Figure 7: Exosphere densities of hot carbon atoms for various sources for low (a) and high (b) solar activity. The exosphere density proﬁles for diﬀerent source terms of the hot C atoms are shown in Figure 7 for low and high solar activity. In Figure 8 the carbon exosphere den- sity obtained by a number of studies is illustrated. It should be noted that Nagy et al. (2001) include photodissociation of CO, dissociative recombination of CO and colli- sions between energetic oxygen and thermal carbon atoms as sources of hot C, while Cipriani et al. (2007) consider only dissociative recombination of CO . Our results suggest – in agreement with previous studies – that the main loss of atomic carbon at present Mars is due to photodissociation of CO. According to the 24 −1 present simulations, the total carbon loss is about 0.8 and 3.2 × 10 s for low and high solar activity (Table 13) , respectively, which is ∼ 5−15 times higher compared to the estimated average atmospheric sputtering of C atoms (Chasseﬁe`re et al., 2007) and up to ∼ 40 times higher compared to the estimated CO molecular ion escape from ASPERA-3 (Barabash et al., 2007). The present results are in accord with the sugges- tion of Lammer et al. (2013) that the escape of photochemically produced suprathermal Altitude [km] Altitude [km] 6000 this study - LSA this study - HSA Nagy et al. (2001) - LSA Nagy et al. (2001) - HSA Cipriani et al. (2007) - HSA 1 10 100 1000 -3 Density [cm ] Figure 8: Comparison of the obtained exosphere densities of hot carbon atoms with previous studies. + + C atoms, which originate from the dissociation of CO , CO and CO are the most eﬃ- cient processes for the loss of the martian CO atmosphere at present. Therefore, these exothermic reactions may play a major role for the escape of the martian atmosphere since the end of the Noachian. It is important to note that the modeled hot C and O and the related CO losses depend on a complex interplay of many physical and chemical processes – e.g. the change of the solar XUV ﬂux and the corresponding response of the upper neutral and ionized atmosphere – and that the present escape rates can not be easily extrapolated backwards during the planet’s history. Since CO +e dissociates mainly into CO and O and only with 4% probability into C and O (see Table 2), it follows O/C=24/1. On the other hand, the corresponding ratio 25 23 of the escape rates (for e.g. LSA) is 1.0 × 10 /(0.8 × 10 ) = 125, i.e. oxygen can escape ﬁve times more eﬃcient than carbon. This imbalance in the loss rates of O and C can be at least partially attributed to the diﬀerence in the energies released by the two channels producing atomic oxygen and carbon (Table 2), providing oxygen with a 3.3 times higher energy than carbon. 4. Summary and Conclusion A number of possible sources for hot C and O atoms and the resulting escape rates during low and high solar activity conditions from present Mars have been studied by applying a Monte-Carlo model to simulate the stochastic motion of these particles through the upper atmosphere. The main sources of escaping oxygen are dissociative + + 25 −1 recombination of O and CO leading to a total loss of ∼ 2 − 3 × 10 s over 2 2 the solar cycle when an eroded ionosphere model is assumed. The vast majority of escaping carbon is delivered through photodissociation of CO with a total loss of the 24 −1 order of 0.5 to 3 × 10 s over the solar cycle. This value is distinctly higher than the Altitude [km] + estimated present loss due to atmospheric sputtering and the observed CO escape at Mars. The calculated loss rates based on the energy distribution function at 450 km alti- tude are reduced by ∼ 20 − 30% with respect to the values calculated by means of the EDFs taken at the exobase level. This suggests that collisions still play a role above the exobase in a region with an extension of ∼ 5 − 10 times the local scale height. It should also be emphasized that the modeled oxygen loss is due to the reactions + + listed in Tables 1-6 – above all due to dissociative recombination of O and CO – 2 2 and is therefore not directly related to the loss of water. On the other hand, the O escape cannot be balanced by C escape since the former appears to be 10–30 times higher than the latter. This suggests – assuming the martian atmosphere to be in steady state over long time scales – that all oxygen lost by escape may be ﬁnally replenished by means of H O via a hydrogen-carbon chemistry rather than converting CO into CO, thereby 2 2 stabilizing the CO atmosphere. Acknowledgement The authors wish to thank two anonymous referees, whose constructive comments helped to improve the paper. H. Gro¨ller, H. Lammer and H. Lichtenegger acknowledge support from the Helmholtz Alliance project ”Planetary Evolution and Life”, H. Gro¨ller and H. Lichtenegger also acknowledge support by the Austrian Science Fund (FWF) under the project P24247-N16. H. Lammer acknowledges also the support by the FWF NFN project S116 “Pathways to Habitability: From Disks to Active Stars, Planets and Life”, and the related FWF NFN subproject, S116607-N16 ”Particle/Radiative Interac- tions with Upper Atmospheres of Planetary Bodies Under Extreme Stellar Conditions”. V. Shematovich acknowledges support from the Basic Research Program of the Presid- ium of the Russian Academy of Sciences (Program 22), Russian Foundation for Basic Research (Project 11-02-00479a), Federal Targeted Program ”Science and Science Ed- ucation for Innovation in Russia 2009-2013”. References Balakrishnan, N., Kharchenko, V., Dalgarno, A., 1998a. Quantum mechanical and semiclassical studies of N+N collisions and their application to thermalization of fast N atoms. The Journal of Chemical Physics 108 (3), 943–949. Balakrishnan, N., Kharchenko, V., Dalgarno, A., 1998b. Slowing of energetic O( P) atoms in collisions with N . Journal of Geophysical Research 103 (A10), 23393– Balakrishnan, N., Kharchenko, V., Dalgarno, A., 1999. Translational energy relaxation of hot O( D) atoms. The Journal of Physical Chemistry A 103, 3999–4001. Barabash, S., Fedorov, A., Lundin, R., Sauvaud, J.-A., 2007. Martian atmospheric erosion rates. Science 315 (5811), 501–503. 26 Bauer, S. J., Lammer, H., 2004. Planetary Aeronomy - Atmosphere Environments in Planetary Systems. Springer. Bougher, S. W., Bell, J. M., Murphy, J. R., Lopez-Valverde, M. A., Withers, P. G., 2004. Polar warming in the Mars thermosphere: Seasonal variations ow- ing to changing insolation and dust distributions. Geophysical Research Letters 32, doi:10.1029/2005GL024059. L02203. Chasseﬁe`re, E., Leblanc, F., Langlais, B., 2007. The combined eﬀects of escape and magnetic ﬁeld histories at Mars. Planetary and Space Science 55 (3), 343–357. Chasseﬁe`re, E., Leblanc, F., 2011. Constraining methane release due to serpentiniza- tion by the observed D/H ratio on Mars. Earth and Planetary Science Letters 310, 262–271. Chasseﬁe`re, E., Leblanc, F., 2011. Explaining the redox imbalance between the H and O escape ﬂuxes at Mars by the oxidation of methane. Planetary and Space Science 59, 218–226. Chaufray, J. Y., Modolo, R., Leblanc, F., Chanteur, G., Johnson, R. E., Luhmann, J. G., 2007. Mars solar wind interaction: Formation of the Martian corona and atmospheric loss to space. Journal of Geophysical Research 112 (E9), E09009. Cipriani, F., Leblanc, F., Berthelier, J. J., 2007. Martian corona: Nonthermal sources of hot heavy species. Journal of Geophysical Research 112 (E7), E07001. Claire, M. W., Sheets, J., Cohen, M., Ribas, I., Meadows, V. S., Catling, D. C., 2012. The evolution of solar ﬂux from 0.1 nm to 160 μm: Quantitative estimates for plan- etary studies. The Astrophysical Journal 757 (1), 95. Costes, M., Naulin, C., 1998. State-to-state cross sections for the C( P ) + 3 – 1 + 1 O (X Σ ) −−−→ CO(X Σ ) + O( D ) reaction at kinetic energies between 4.4 and 2 g 2 90 meV. Comptes Rendus de l’Acadmie des Sciences - Series IIC - Chemistry 1 (12), 771–775. Erkaev, N. V., Lammer, H., Elkins-Tanton, L., Odert, P., Kislyakova, K. G., Kulikov, Y. N., Leitzinger, M., Gu¨del, M., 2013. Escape of the martian protoatmosphere and initial water inventory. Planetary and Space Science, submitted. Feldman, P. D. et al., 2011. Rosetta-Alice observations of exospheric hydrogen and oxygen. Icarus 214, 394–399. Fox, J. L., 2004. CO dissociative recombination: A source of thermal and nonthermal C on Mars. Journal of Geophysical Research 109 (A8), A08306. Fox, J. L., Bakalian, F. M., 2001. Photochemical escape of atomic carbon from Mars. Journal of Geophysical Research 106 (A12), 28785–28795. Fox, J. L., Hac´, A. B., 2009. Photochemical escape of oxygen from Mars: A compari- son of the exobase approximation to a Monte Carlo method. Icarus 204 (2), 527–544. 27 Gro¨ller, H., Lammer, H., Lichtenegger, H. I. M., Pﬂeger, M., Dutuit, O., Shematovich, V. I., Kulikov, Y. N., Biernat, H. K., 2012. Hot oxygen atoms in the Venus nightside exosphere. Geophysical Research Letters 39 (3), L03202. Gro¨ller, H., Shematovich, V. I., Lichtenegger, H. I. M., Lammer, H., Pﬂeger, M., Ku- likov, Y. N., Macher, W., Amerstorfer, U. V., Biernat, H. K., 2010. Venus’ atomic hot oxygen environment. Journal of Geophysical Research 115, E12017. Hodges, R. R., 2000. Distributions of hot oxygen for Venus and Mars. Journal of Geo- physical Research 105 (E3), 6971–6981. Hodges Jr., R. R., 2002. The rate of loss of water from Mars. Geophysical Research Letters 29 (3), 1038–1040. Huebner, W. F., Keady, J. J., Lyon, S. P., 1992. Solar photo rates for planetary atmo- spheres and atmospheric pollutants. Astrophysics and Space Science 195 (1), 1–294. Kella, D. P., Johnson, P. J., Pedersen, H. B., Vejby-Christensen, L., Andersen, L. H., 1997. The source of green light emission determined from a heavy-ion storage ring experiment. Science 276, 1530–1533. Kharchenko, V., Dalgarno, A., Zygelman, B., Yee, J., 2000. Energy transfer in colli- sions of oxygen atoms in the terrestrial atmosphere. Journal of Geophysical Research 105 (A11), 24899–24906. Kim, J., Nagy, A. F., Fox, J. L., Cravens, T. E., 1998. Solar cycle variability of hot oxygen atoms at Mars. Journal of Geophysical Research 103 (A12), 29339–29342. Krasnopolsky, V. A., 2002. Mars’ upper atmosphere and ionosphere at low, medium, and high solar activities: Implications for evolution of water. Journal of Geophysical Research 107 (E12), 5128. Krestyanikova, M. A., Shematovich, V. I., 2006. Stochastic models of hot planetary and satellite coronas: A hot oxygen corona of Mars. Solar System Research 40 (5), 384–392. Lammer, H., Chasseﬁe`re, E., Karatekin, O., Morschhauser, A., Niles, P. B., Mousis, O., Odert, P., Mo¨stl, U. V., Breuer, D., Dehant, V., Grott, M., Gro¨ller, H., Hauber, E., Pham, L. B. S., 2013. Outgassing history and escape of the martian atmosphere and water inventory. Space Science Reviews 174 (1-4), 113–154. Leblanc, F., Chaufray, J. Y., Lilensten, J., Witasse, O., Bertaux, J.-L., 2006. Martian dayglow as seen by the SPICAM UV spectrograph on Mars Expresss. Journal of Geophysical Research 111 (E9), E09S11. Lichtenegger, H., Lammer, H., Kulikov, Y., Kazeminejad, S., Molina-Cuberos, G., Ro- drigo, R., Kazeminejad, B., Kirchengast, G., 2006. Eﬀects of low energetic neutral atoms on martian and venusian dayside exospheric temperature estimations. Space Science Reviews 126, 469–501. 28 Lichtenegger, H. I. M., Gro¨ller, H., Lammer, H., Kulikov, Y. N., Shematovich, V. I., 2009. On the elusive hot oxygen corona of Venus. Geophysical Research Letters 36 (10), L10204. Matsumi, Y., Sarwaruddin Chowdhury, A. M., 1996. Translational relaxation and elec- tronic quenching of hot O( D) by collisions with N . The Journal of Chemical Physics 104 (18), 7036–7044. McElroy, M. B., 1972. Mars: An evolving atmosphere. Science 175 (4020), 443–445. Nagy, A. F., Cravens, T. E., 1988. Hot oxygen atoms in the upper atmospheres of Venus and Mars. Geophysical Research Letters 15 (5), 433–435. Nagy, A. F., Liemohn, M. W., Fox, J. L., Kim, J., 2001. Hot carbon densities in the exosphere of Mars. Journal of Geophysical Research 106 (A10), 21565–21568. Nahar, S. N., 1999. Electron-Ion recombination rate coeﬃcients, photoionization cross sections, and ionization fractions for astrophysically abundant elements. II. Oxygen ions. The Astrophysical Journal Supplement Series 120 (1), 131–145. Nahar, S. N., Pradhan, A. K., 1997. Electron-Ion recombination rate coeﬃcients, pho- toionization cross sections, and ionization fractions for astrophysically abundant ele- ments. I. Carbon and nitrogen. The Astrophysical Journal Supplement Series 111 (1), 339–355. Ogryzlo, E., Reilly, J., Thrush, B., 1973. Vibrational excitation of CO from the reaction C + O . Chemical Physics Letters 23 (1), 37–39. Petrignani, A., Hellberg, F., Thomas, R. D., Larsson, M., Cosby, P. C., van der Zande, W. J., 2005. Electron energy-dependent product state distributions in the dissocia- tive recombination of O . The Journal of Chemical Physics 122 (23), 234311–1– 234311–8. Peverall, R., Rose´n, S., Peterson, J., Larsson, M., Al-Khalili, A., Vikor, L., Semaniak, J., Bobbenkamp, R., Le Padellec, A., Maurellis, A., van der Zande, W., 2001. Dis- sociative recombination and excitation of O : Cross sections, product yields and implications for studies of ionospheric airglows. The Journal of Chemical Physics 114 (15), 6679–6689. Ribas, I., Guinan, E. F., Gudel, M., Audard, M., 2005. Evolution of the solar activ- ity over time and eﬀects on planetary atmospheres. I. High-energy irradiances (1- 1700 Å). The Astrophysical Journal 622 (1), 680–694. Rose´n, S., Peverall, R., Larsson, M., Le Padellec, A., Semaniak, J., Larson, ., Strmholm, C., van der Zande, W. J., Danared, H., Dunn, G. H., 1998. Absolute cross sections and ﬁnal-state distributions for dissociative recombination and excitation of CO (v=0) using an ion storage ring. Physical Review A 57 (6), 4462–4471. Seiersen, K., Al-Khalili, A., Heber, O., Jensen, M. J., Nielsen, I. B., Pedersen, H. B., Safvan, C. P., Andersen, L. H., 2003. Dissociative recombination of the cation and dication of CO . Physical Review A 68 (2), 022708. 29 + + + Sheehan, C. H., St.-Maurice, J., 2004. Dissociative recombination of N , O , and NO : 2 2 Rate coeﬃcients for ground state and vibrationally excited ions. Journal of Geophys- ical Research 109 (A3), A03302. Tachikawa, H., Ohnishi, K., Hamabayashi, T., Yoshida, H., 1997. Translational re- laxation of hot O( D) by inelastic collision with N molecule: Ab initio MO and classical trajectory studies. The Journal of Chemical Physics A 101, 2229–2232. Terada, N., Machida, S., Shinagawa, H., 2002. Global hybrid simulation of the Kelvin−Helmholtz instability at the Venus ionopause. Journal of Geophysical Re- search 107 (A12), 1471. Tian, F., Kasting, J. F., Solomon, S. C., 2009. Thermal escape of carbon from the early martian atmosphere. Geophysical Research Letters 36 (2), L02205. Tobiska, W. K., 2004. SOLAR2000 irradiances for climate change research, aeronomy and space system engineering. Advances in Space Research 34 (8), 1736–1746. Tully, C., Johnson, R. E., 2001. Low energy collisions between ground-state oxygen atoms. Planetary and Space Science 49 (6), 533–537. Valeille, A., Combi, M. R., Tenishev, V., Bougher, S. W., Nagy, A. F., 2010. A study of suprathermal oxygen atoms in Mars upper thermosphere and exosphere over the range of limiting conditions. Icarus 206 (1), 18–27. Vejby-Christensen, L., Kella, D., Pedersen, H. B., Andersen, L. H., 1998. Dissociative recombination of NO . Physical Review A 57 (5), 3627–3634. Viggiano, A. A., Ehlerding, A., Hellberg, F., Thomas, R. D., Zhaunerchyk, V., Geppert, W. D., Montaigne, H., Larsson, M., Kaminska, M., Osterdahl, F., 2005. Rate con- stants and branching ratios for the dissociative recombination of CO . The Journal of Chemical Physics 122 (22), 226101. Woodall, J., Agu´ndez, M., Markwick-Kemper, A. J., Millar, T. J., 2007. The UMIST database for astrochemistry 2006. Astronomy and Astrophysics 466 (3), 1197–1204. Yagi, M., Leblanc, F., Chaufray, J., Gonzalez-Galindo, F., Hess, S., Modolo, R., 2012. Mars exospheric thermal and non-thermal components: Seasonal and local varia- tions. Icarus 221 (2), 682–693. 1 3 Yee, J., Dalgarno, A., 1985. Energy transfer of O( S) atoms in collision with O( P) atoms. Planetary and Space Science 33 (7), 825–830. Yee, J., Guberman, S. L., Dalgarno, A., 1990. Collisional quenching of O( D) by O( P). Planetary and Space Science 38 (5), 647–652. 1 3 Yee, J. H., Dalgarno, A., 1987. Energy transfer of O( D) atoms in collision with O( P) atoms. Planetary and Space Science 35 (4), 399–404. Zahr, G. E., Preston, R. K., Miller, W. H., 1975. Theoretical treatment of quenching in O( D) + N collisions. The Journal of Chemical Physics 62 (3), 1127–1135. 30 Zhang, P., Kharchenko, V., Jamieson, M. J., Dalgarno, A., 2009. Energy relaxation in collisions of hydrogen and deuterium with oxygen atoms. Journal of Geophysical Research 114 (A7), A07101. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Astrophysics arXiv (Cornell University) http://www.deepdyve.com/lp/arxiv-cornell-university/hot-oxygen-and-carbon-escape-from-the-martian-atmosphere-V06Ax3kd0q

Loading next page...

References (67)

J. Yee, A. Dalgarno (1985)
ENERGY TRANSFER OF O('D) ATOMS IN COLLISION WITH 0(3P) ATOMS
S. Rosén, R. Peverall, M. Larsson, A. Padellec, J. Semaniak, Å. Larson, C. Strömholm, W. Zande, H. Danared, G. Dunn (1998)
Absolute cross sections and final-state distributions for dissociative recombination and excitation of CO + (v=0) using an ion storage ring
Physical Review A, 57
E. Ogryzlo, J. Reilly, B. Thrush (1973)
Vibrational excitation of CO from the reaction C + O2
Chemical Physics Letters, 23
N. Terada, S. Machida, H. Shinagawa (2002)
Global hybrid simulation of the Kelvin–Helmholtz instability at the Venus ionopause
Journal of Geophysical Research, 107
(2004)
SUBMITTED TO APJ Preprint typeset using L ATEX style emulateapj EVOLUTION OF THE SOLAR ACTIVITY OVER TIME AND EFFECTS ON PLANETARY ATMOSPHERES: I. HIGH-ENERGY IRRADIANCES (1–1700 A)
M. Costes, C. Naulin (1998)
State-to-state cross sections for the C(3PJ) + O2(X3∑g−) → CO(X1∑+) + O(1D2) reaction at kinetic energies between 4.4 and 90 meV
Comptes Rendus De L Academie Des Sciences Serie Ii Fascicule C-chimie, 1
(1998)
Slowing of energetic O(3P
J. Fox (2004)
CO2+ dissociative recombination: A source of thermal and nonthermal C on Mars
Journal of Geophysical Research, 109
J. Fox, F. Bakalian (2001)
Photochemical escape of atomic carbon from Mars
Journal of Geophysical Research, 106
E. Chassefière, F. Leblanc (2011)
Constraining methane release due to serpentinization by the observed D/H ratio on Mars
Earth and Planetary Science Letters, 310
H. Tachikawa, Koichi Ohnishi, Takayuki Hamabayashi, H. Yoshida (1997)
Translational Relaxation of Hot O(1D) by Inelastic Collision with N2 Molecule: Ab Initio MO and Classical Trajectory Studies
Journal of Physical Chemistry A, 101
single interaction potential and only atomic background atmosphere
D. McElroy, C. Walsh, A. Markwick, M. Cordiner, M. Cordiner, K. Smith, T. Millar (2012)
The UMIST database for astrochemistry 2012
Astronomy and Astrophysics, 550
R. Peverall, S. Rosén, J. Peterson, M. Larsson, A. Al-Khalili, L. Víkor, J. Semaniak, Rolf Bobbenkamp, A. Padellec, A. Maurellis, W. Zande (2001)
Dissociative recombination and excitation of O2+: Cross sections, product yields and implications for studies of ionospheric airglows
Journal of Chemical Physics, 114
H. Lichtenegger, H. Lammer, Y. Kulikov, S. Kazeminejad, Gregorio Molina-Cuberos, R. Rodrigo, B. Kazeminejad, G. Kirchengast (2006)
Effects of Low Energetic Neutral Atoms on Martian and Venusian Dayside Exospheric Temperature Estimations
Space Science Reviews, 126
S. Barabash, A. Fedorov, R. Lundin, J. Sauvaud (2007)
Martian Atmospheric Erosion Rates
Science, 315
P. Feldman, A. Steffl, J. Parker, M. A’Hearn, J. Bertaux, S. Stern, H. Weaver, D. Slater, M. Versteeg, H. Throop, N. Cunningham, L. Jhu, SwRI, Umd, Latmos, Jhuapl, Nwu (2011)
Rosetta-Alice observations of exospheric hydrogen and oxygen on Mars
Icarus, 214
W. Tobiska (2004)
SOLAR2000 irradiances for climate change research, aeronomy and space system engineering
Advances in Space Research, 34
V. Krasnopolsky (2002)
Mars' upper atmosphere and ionosphere at low, medium, and high solar activities: Implications for evolution of water
Journal of Geophysical Research, 107
J. Yee, S. Guberman, A. Dalgarno (1990)
Collisional quenching of O(1D) by O(3P)
Planetary and Space Science, 38
A. Nagy, T. Cravens (1988)
Hot oxygen atoms in the upper atmospheres of Venus and Mars
Geophysical Research Letters, 15
S. Guberman (1983)
FOR DISSOCIATIVE RECOMBINATION
M. Claire, J. Sheets, Martin Cohen, I. Ribas, V. Meadows, D. Catling (2012)
THE EVOLUTION OF SOLAR FLUX FROM 0.1 nm TO 160 μm: QUANTITATIVE ESTIMATES FOR PLANETARY STUDIES
The Astrophysical Journal, 757
N. Erkaev, N. Erkaev, H. Lammer, L. Elkins‐Tanton, A. Stökl, P. Odert, P. Odert, E. Marcq, E. Dorfi, K. Kislyakova, Y. Kulikov, M. Leitzinger, M. Güdel (2013)
Escape of the martian protoatmosphere and initial water inventory
Planetary and Space Science, 98
Bougher (2004)
Polar warming in the Mars thermosphere
Geophysical Research Letters, 32
V. Shematovich (2004)
Stochastic Models of Hot Planetary and Satellite Coronas
Solar System Research, 38
N. Balakrishnan, V. Kharchenko, A. Dalgarno (1999)
Translational Energy Relaxation of Hot O(1D) Atoms
Journal of Physical Chemistry A, 103
Y. Matsumi, A. Chowdhury (1996)
Translational relaxation and electronic quenching of hot O(1D) by collisions with N2
Journal of Chemical Physics, 104
H. Lichtenegger, H. Gröller, H. Lammer, Y. Kulikov, V. Shematovich (2009)
On the elusive hot oxygen corona of Venus
Geophysical Research Letters, 36
Y. Teuff (2000)
The UMIST database for astrochemistry
, 146
Jhoon Kim, A. Nagy, J. Fox, T. Cravens (1998)
Solar cycle variability of hot oxygen atoms at Mars
Journal of Geophysical Research, 103
M. Jensen, R. Bilodeau, O. Heber, H. Pedersen, C. Safvan, X. Urbain, D. Zajfman, L. Andersen (1999)
Dissociative recombination and excitation of H2O+ and HDO+
Physical Review A, 60
F. Tian, J. Kasting, S. Solomon (2009)
Thermal escape of carbon from the early Martian atmosphere
Geophysical Research Letters, 36
H. Gröller, V. Shematovich, H. Lichtenegger, H. Lammer, M. Pfleger, Y. Kulikov, W. Macher, U. Amerstorfer, H. Biernat (2010)
Venus' atomic hot oxygen environment
Journal of Geophysical Research, 115
J. Chaufray, R. Modolo, F. Leblanc, G. Chanteur, R. Johnson, J. Luhmann (2007)
Mars solar wind interaction: Formation of the Martian corona and atmospheric loss to space
Journal of Geophysical Research, 112
F. Leblanc, J. Chaufray, J. Lilensten, O. Witasse, J. Bertaux (2006)
Martian dayglow as seen by the SPICAM UV spectrograph on Mars Express
Journal of Geophysical Research, 111
G. Zahr, R. Preston, W. Miller (1975)
Theoretical treatment of quenching in O(1D) + N2 collisions
Journal of Chemical Physics, 62
H. Lammer, E. Chassefière, Ö. Karatekin, A. Morschhauser, P. Niles, O. Mousis, P. Odert, U. Möstl, D. Breuer, V. Dehant, M. Grott, H. Gröller, E. Hauber, L. Pham (2015)
Outgassing History and Escape of the Martian Atmosphere and Water Inventory
Space Science Reviews, 174
F. Cipriani, F. Leblanc, J. Berthelier (2007)
Martian corona: Nonthermal sources of hot heavy species
Journal of Geophysical Research, 112
Walter Huebner, J. Keady, S. Lyon, P. Dickinson (1984)
Solar photo rates for planetary atmospheres and atmospheric pollutants
Astrophysics and Space Science, 195
E. Chassefière, F. Leblanc, B. Langlais (2007)
The combined effects of escape and magnetic field histories at Mars
Planetary and Space Science, 55
A. Nagy, M. Liemohn, J. Fox, Jhoon Kim (2001)
Hot carbon densities in the exosphere of Venus
Journal of Geophysical Research, 106
D. Kella, L. Vejby-Christensen, P. Johnson, H. Pedersen, L. Andersen (1997)
The Source of Green Light Emission Determined from a Heavy-Ion Storage Ring Experiment
Science, 276
E. Chassefière, F. Leblanc (2010)
Explaining the redox imbalance between the H and O escapes fluxes at Mars by the oxidation of methane
S. Bougher, J. Bell, J. Murphy, M. López‐Valverde, P. Withers (2006)
Polar warming in the Mars thermosphere: Seasonal variations owing to changing insolation and dust distributions
Geophysical Research Letters, 33
R. Hodges (2000)
Distributions of hot oxygen for Venus and Mars
Journal of Geophysical Research, 105
J. Schleipen, J. Meulen, A. Offer (1993)
State-to-state cross sections for rotational excitation of ortho- and para-NH3 by ortho- and para-H2. Experiment versus theory
Chemical Physics, 171
C. Tully, Robert Johnson (2001)
Low energy collisions between ground-state oxygen atoms
Planetary and Space Science, 49
M. Yagi, F. Leblanc, J. Chaufray, F. González-Galindo, S. Hess, R. Modolo (2012)
Mars exospheric thermal and non-thermal components: Seasonal and local variations
Icarus, 221
N. Balakrishnan, V. Kharchenko, A. Dalgarno (1998)
Quantum mechanical and semiclassical studies of N+N2 collisions and their application to thermalization of fast N atoms
Journal of Chemical Physics, 108
R. Hodges (2002)
The rate of loss of water from mars
Geophysical Research Letters, 29
K. Seiersen, A. Al-Khalili, O. Heber, M. Jensen, I. Nielsen, H. Pedersen, C. Safvan, L. Andersen (2003)
Dissociative recombination of the cation and dication of CO 2
Physical Review A, 68
A. Petrignani, F. Hellberg, R. Thomas, M. Larsson, P. Cosby, W. Zande (2005)
Electron energy-dependent product state distributions in the dissociative recombination of O2+.
The Journal of chemical physics, 122 23
A. Valeille, M. Combi, V. Tenishev, S. Bougher, A. Nagy (2010)
A study of suprathermal oxygen atoms in Mars upper thermosphere and exosphere over the range of limiting conditions
Icarus, 206
Y. Yung (2005)
Planetary Aeronomy: Atmosphere Environments in Planetary Systems
Astrobiology, 5
C. Sheehan, J. St.‐Maurice (2004)
Dissociative recombination of N2+, O2+, and NO+: Rate coefficients for ground state and vibrationally excited ions
Journal of Geophysical Research, 109
S. Nahar (1999)
Electron-Ion Recombination Rate Coefficients, Photoionization Cross Sections, and Ionization Fractions for Astrophysically Abundant Elements. II. Oxygen Ions
The Astrophysical Journal Supplement Series, 120
V. Kharchenko, A. Dalgarno, B. Zygelman, J. Yee (2000)
Energy transfer in collisions of oxygen atoms in the terrestrial atmosphere
Journal of Geophysical Research: Space Physics, 105
J. Fox, A. Hać (2009)
Photochemical escape of oxygen from Mars: A comparison of the exobase approximation to a Monte Carlo method
Icarus, 204
A. Nagy, M. Liemohn, J. Fox, Jhoon Kim (2001)
Hot Carbon Densities in the Exosphere of Mars Hot Carbon Densities in the Exosphere of Mars
L. Vejby-Christensen, D. Kella, H. Pedersen, L. Andersen (1998)
Dissociative recombination of NO
Physical Review A, 57
H. Gröller, Helmut Lammer, H. Lichtenegger, M. Pfleger, Odile Dutuit, V. Shematovich, Y. Kulikov, H. Biernat (2012)
Hot oxygen atoms in the Venus nightside exosphere
Geophysical Research Letters, 39
M. McElroy (1972)
Mars: An Evolving Atmosphere
Science, 175
Peng Zhang, V. Kharchenko, M. Jamieson, A. Dalgarno (2009)
Energy relaxation in collisions of hydrogen and deuterium with oxygen atoms
Journal of Geophysical Research, 114
M. Krest’yanikova, V. Shematovich (2006)
Stochastic models of hot planetary and satellite coronas: A hot oxygen corona of Mars
Solar System Research, 40
A. Viggiano, A. Ehlerding, F. Hellberg, R. Thomas, V. Zhaunerchyk, W. Geppert, H. Montaigne, M. Larsson, M. Kamińska, F. Osterdahl (2005)
Rate constants and branching ratios for the dissociative recombination of CO2(+).
The Journal of chemical physics, 122 22
N. Balakrishnan, V. Kharchenko, A. Dalgarno (1998)
Slowing of energetic O(3 P) atoms in collisions with N2
Journal of Geophysical Research, 103

ISSN: 0032-0633
eISSN: ARCH-3330
DOI: 10.1016/j.pss.2014.01.007
Publisher site: See Article on Publisher Site

Abstract

The escape of hot O and C atoms from the present martian atmosphere during low and high solar activity conditions has been studied with a Monte-Carlo model. The model includes the initial energy distribution of hot atoms, elastic, inelastic, and quenching collisions between the suprathermal atoms and the ambient cooler neutral atmosphere, and applies energy dependent total and diﬀerential cross sections for the determination of the collision probability and the scattering angles. The results yield a total loss rate 25 −1 of hot oxygen of 2.3 − 2.9 × 10 s during low and high solar activity conditions and + + is mainly due to dissociative recombination of O and CO . The total loss rates of 2 2 24 −1 carbon are found to be 0.8 and 3.2×10 s for low and high solar activity, respectively, with photodissociation of CO being the main source. Depending on solar activity, the obtained carbon loss rates are up to ∼ 40 times higher than the CO ion loss rate inferred from Mars Express ASPERA-3 observations. Finally, collisional eﬀects above the exobase reduce the escape rates by about 20-30% with respect to a collionless exophere. Key words: Mars, hot atom escape, hot corona density, heavy atoms, oxygen, carbon 1. Introduction The dynamics of the upper atmosphere and its interaction with the solar radiation and plasma environment are important clues for the stability and long-term evolution of terrestrial atmospheres. In a recent study Lammer et al. (2013) showed that the evo- lution and escape of the martian CO atmosphere and the planet’s H O inventory can 2 2 be separated from the planet’s origin into an early and a later evolutionary epoch. The ﬁrst epoch started after Mars’ formation and lasted ≤ 500 Myr (Lammer et al., 2013). Because of the high soft X-ray and EUV ﬂux of the young Sun (e.g. Ribas et al., 2005; Claire et al., 2012) and Mars’ low gravity the nebula-based and outgassed hydrogen- rich protoatmosphere was lost by hydrodynamic blow-oﬀ of H atoms and strong ther- mal escape rates of dragged heavier species such as O and C atoms within several Principal corresponding author Email address: herbert.lichtenegger@oeaw.ac.at (H. Lichtenegger) Preprint submitted to Planetary and Space Sciences December 25, 2013 arXiv:1911.01107v1 [astro-ph.EP] 4 Nov 2019 4 10 years (Erkaev et al., 2013, this issue). These recent results are in support of the hypotheses of Tian et al. (2009) that early Mars could not build up a dense CO atmo- sphere during the early Noachian. However, after the planet lost its protoatmosphere the atmospheric escape rates were most likely balanced with the volatiles (H O, CO , etc.) which have been out- 2 2 gassed by volcanic activity and delivered by impacts until the activity of the young Sun decreased, so that the atmospheric sources could dominate over the losses ∼ 4 − 4.2 Gyr ago (Lammer et al., 2013). If a secondary outgassed CO atmosphere of several 100 mbar was indeed present at the end of the Noachian, its evolution during the second epoch which lasted from the end of the Noachian until today should have been deter- mined by a complex interplay of various nonthermal atmospheric escape processes, impacts, carbonate precipitation, and serpentinization during the Hesperian and Ama- zonian epochs which led to the present day surface pressure (Chasseﬁe`re and Leblanc, 2011a,b). Atmospheres may get eroded through the loss of light particles like H or He, when their thermal velocities exceed the escape velocity at the exobase. Heavier species like O or C, are stronger bound to the planet and can therefore either directly escape through nonthermal processes like dissociative recombination of molecular ions or may form extended coronae which in turn can be eroded by the solar wind. From Mars Express ASPERA-3 data Barabash et al. (2007) estimated the present escape of CO 22 −1 molecular ions and yield low escape rates of ∼ 8 × 10 s . Based on this rate these authors estimated that the fraction of CO molecular ions lost to space since the end of the Noachian when the martian dynamo stopped to work would represent an equivalent surface pressure of only about ∼ 0.2 − 4 mbar. Lammer et al. (2013) compared these ion escape rates with atmospheric sputtering and exothermic escape processes of heavy neutrals and suggested that the loss of pho- tochemically produced suprathermal atoms such as O, C may be responsible for the main loss of the martian CO atmosphere at present and most likely also during the planet’s history until thermal escape took over. The simulation of this nonthermal photochemically related escape of heavy atmo- spheric atoms at terrestrial planets is not only challenged by the physical and chem- ical complexity of the upper atmosphere but is additionally hindered by our sparse knowledge of many of the processes which govern the direct escape of atmospheric constituents. Although Mars has been visited by many spacecraft for more than 40 years, observational data of the hot particle corona at Mars are still rather limited (e.g. Leblanc et al., 2006; Lichtenegger et al., 2006); the only recent observation of the ex- tended oxygen corona of Mars was made with the Alice far-ultraviolet imaging spec- trograph during the Rosetta swing-by of Mars (Feldman et al., 2011). Therefore, nu- merical modeling of exosphere densities and of escape rates are still little constrained and suﬀer from the various assumptions made by diﬀerent authors. The aim of this work is to obtain a better understanding on the escape of hot car- bon and oxygen atoms from present Mars. For this purpose we consider a number of possible sources of hot O and C atoms in the thermosphere, calculate the production rate proﬁles and the energy distribution functions at the exobase for low and high so- lar activity conditions and discuss the associated escape ﬂuxes and exosphere density proﬁles. Because there are little observational hints of the rotational and vibrational 2 excitation distributions of the particles and because these eﬀects have negligible inﬂu- ence on the escape rates (Gro¨ller et al., 2010), we assume all particles in the rotational and vibrational ground states. The simulations are based – if available – on the most recent data for both the atmospheric input and the collision cross sections. The model is limited by various circumstances: (a) the background atmosphere is represented by 1D ion and neutral density proﬁles (i.e. all variables depend only on radial distance) assumed to be valid for 60 solar zenith angle, (b) the hot particles are treated as test particles, i.e. their inﬂuence on the background gas is neglected, (c) due to the lack of data, some types of collisions are simulated by using approximate values of the parameters (see Tables 9 and 10). 2. Hot Atom Modeling 2.1. Monte Carlo model The simulation of the hot particle corona is initiated by calculating for a speciﬁc reaction the corresponding velocity distribution of the products at discrete altitudes be- tween 130 km and 450 km above the surface. These products are represented by a number of particles (typically ∼ 10 per 2.5 km altitude step) and their 3-dimensional motion is followed through the thermosphere up to 450 km in the gravitational ﬁeld of Mars. On their way the hot particles can interact with the background neutral at- mosphere via elastic, inelastic and quenching collisions and will loose on average part of their initial energy. Background particles which acquire an energy exceeding the thermal velocity of the atmospheric species upon collisions are also treated as new hot particles. The collision probability and energy transfer is calculated by means of total and diﬀerential cross sections. At the exobase (for comparison with other models) and at 380 km altitude, the energy distribution function of the suprathermal particles is de- termined which in turn serves as input for the exosphere density calculations. A more detailed description of the Monte-Carlo model can be found in Gro¨ller et al. (2010) and Gro¨ller et al. (2012). Although some 3D simulation results of the martian thermosphere have been pub- lished in the recent past (Valeille et al., 2010; Yagi et al., 2012), our background ther- mosphere model is assumed to be spherically symmetric for various reasons: (a) we do not focus on local variations such as solar zenith angle eﬀects; (b) the primary aim of our paper is the study of a number of chemical reactions producing hot particles in view of their relevance for escape as well as the treatment of the various types of collisions on the basis of total and diﬀerential cross sections rather than on hard sphere or universal potential models, and this analysis can be done most easily in 1D; (c) to facilitate the comparison with older models which are to a large extend 1D models; (d) due to the lack of suﬃcient observations, realistic 3D thermosphere models of Mars are yet diﬃcult to construct. 2.2. Input Proﬁles The neutral and ion density as well as the temperatures for both high and low solar activity are taken from Fox and Hac´ (2009) for a solar zenith angle of 60 . In their 3 model, the ambipolar diﬀusion for ions and the molecular and eddy diﬀusion for neu- trals are included as well as more than 220 reactions. The altitude dependent density proﬁles of the background atmosphere are based on measurements of the Viking 1 neutral mass spectrometer together with simulations of the Mars Thermospheric Gen- eral Circulation Model (MTGCM). According to Fox and Hac´ (2009), the O density proﬁles measured by Viking can only be reproduced by imposing an upward velocity boundary condition (”eroded” ionospheres); therefore, only these eroded models are used in the present paper as input for the ion densities. The solar ﬂux is taken from the Solar Irradiance Platform (SIP) version 2.36 (Tobiska, 2004), where we have chosen DOY 32 of 2002 to represent high solar ac- tivity (HSA) and DOY 110 of 1997 to represent low solar activity (LSA). For these dates, HSA and LSA correspond to a solar ﬂux of F = 245.6 and F = 69.7, 10.7 10.7 respectively. Finally, for each considered neutral element in the atmosphere, the cor- responding photo cross sections are taken from the photo ionization/dissociation rates database of Huebner et al. (1992). 2.3. Sources of Hot O Atoms Although dissociative recombination (DR) of O ions is supposed to be the main source of hot oxygen atoms (e.g. McElroy, 1972; Nagy and Cravens, 1988; Kim et al., 1998; Fox and Hac´, 2009; Gro¨ller et al., 2010, 2012), we have also considered some additional possible sources of hot O provided that the corresponding density proﬁles are available. These additional reactions are listed in Tables 2 to 6. 2.3.1. Dissociative and Radiative Recombination Dissociative recombination of ground-state O proceeds via the ﬁve exothermic 1 1 channels shown in Table 1. The excited states O( D) and O( S) are 1.97 eV and Table 1: Excess energies (e.g. Fox and Hac´, 2009) and branching ratios (Peverall et al., 2001) for dissociative + 2 recombination of O (X Π ). The branching ratios correspond to approximately zero collision energy. 2 g Reaction Channels Excess Energy [eV] Branching Ratios [%] + 3 3 O + e O( P) + O( P) 6.99 20.4 3 1 O( P) + O( D) 5.02 44.0 1 1 O( D) + O( D) 3.05 31.5 3 1 O( P) + O( S) 2.80 0.0 1 1 O( D) + O( S) 0.83 4.1 4.19 eV above the ground state O( P), respectively. For the ground vibrational state of the oxygen molecule O (v = 0), the branching ratios are taken from Peverall et al. (2001) for an electron-ion collision energy of nearly 0 eV. There are more recent val- ues published by Petrignani et al. (2005) but they state that “The branching fractions and quantum yields do not entirely agree with earlier measurements from our group that were taken at energies between 0 and 40 meV in a cold ion beam, nor with mea- surements performed at ASTRID by Kella et al. (1997).” Therefore and because the listed branching ratios are comparable to those of Kella et al. (1997), the values from Peverall et al. (2001) are chosen. 4 + The dissociative recombination of CO and the corresponding excess energies and branching ratios are listed in Table 2. The branching ratios are taken from Viggiano et al. (2005), where the uncertainties are given in parentheses. Of all channels, only the third and fourth produce atomic oxygen, although the latter is of no interest since it is an endothermic process and its probability is close to zero. The channels for dissociative recombination of NO are given in Table 3. The branching ratios taken from Vejby-Christensen et al. (1998) are valid for the vibrational ground state and for an electron-ion collision energy close to 0 eV. The excited states 1 2 O( D) and N( D) have energies of 1.97 eV and 2.38 eV above the corresponding ground states, respectively. The excess energies and branching ratios for the dissociative recombination of CO are listed in Table 4. The branching ratios for collision energies of 0.0, 0.4, 1.0, and 1.5 eV are taken from Rose´n et al. (1998). The altitude dependent production rate P(r) of the dissociative recombination of an ion can be calculated by P(r) = n (r) n (r) k(T (r)), (1) i e e where n (r) and n (r) are the altitude dependent ion and electron densities, respectively, i e and k(T (r)) is the rate coeﬃcient which depends on the electron temperature T (r). e e The value of k is determined according to the Arrhenius equation as T (r) 3 −1 k(T (r)) = α cm s . (2) The coeﬃcients α and β used for the dissociative recombination processes considered in this paper are listed in Table 5. Another possible source of hot O is radiative recombination of O + 4 3 O ( S) + e −−−→ O( P) + hν, where the parameters α and β for the rate coeﬃcient (2) are taken from the UDFA06 database (Woodall et al., 2007) based on Nahar (1999) and given in Table 5. However, radiative recombination appears to be rather ineﬃcient, yielding maximum production −4 −5 −3 −1 rates of about 1×10 and 2×10 cm s for high and low solar activity, respectively. The production rates as a function of altitude for the processes listed in Table 5 are illustrated in Figure 1 for low (a) and high (b) solar activity. Table 2: Excess energies (e.g. Fox, 2004) and branching ratios (Viggiano et al., 2005) for dissociative re- combination of CO ; the uncertainties are given in parentheses. Reaction Channels Excess Energy [eV] Branching Ratios [%] + 1 CO + e CO ( Σ ) 13.78 0.0 (+ 2.0) 2 2 g 3 3 C( P) + O ( Σ ) 2.29 0.0 (+ 4.0) 2 g 1 3 CO( Σ) + O( P) 8.27 100.0 (- 6.0) 3 3 C( P) + 2O( P) -2.87 0.0 5 Table 3: Excess energies and branching ratios (Vejby-Christensen et al., 1998) for dissociative recombination + 1 + of NO (X Σ ); the latter are valid for zero electron-ion collision energy. Reaction Channels Excess Energy [eV] Branching Ratios [%] + 4 3 NO + e N( S) + O( P) 2.78 15.0 4 1 N( S) + O( D) 0.81 0.0 2 3 N( D) + O( P) 0.39 85.0 2.3.2. Photodissociation A further possible source of hot oxygen is photodissociation of O O + hν −−−→ O + O, Table 4: Excess energies (e.g. Fox and Bakalian, 2001) and branching ratios (Rose´n et al., 1998) for disso- + 2 + ciative recombination of CO (X Σ ) for collision energies of 0.0, 0.4, 1.0, and 1.5 eV. Reaction Channels Excess Energy [eV] Branching Ratios [%] 0.0 eV 0.4 eV 1.0 eV 1.5 eV + 3 3 CO + e C( P) + O( P) 2.90 76.1 53.0 39.0 38.0 1 3 C( D) + O( P) 1.64 14.5 34.0 35.0 35.0 3 1 C( P) + O( D) 0.93 9.4 8.0 15.0 11.0 1 3 C( S) + O( P) 0.22 0.0 0.0 5.0 5.0 1 1 C( D) + O( D) -0.32 5.0 6.0 11.0 3 1 C( P) + O( S) -1.28 0.0 Table 5: Possible sources of hot O atoms due to dissociative and radiative recombination and references for the rate coeﬃcients. 3 −1 Source Rate coeﬃcient [cm s ] Dissociative recombination O + e −−−→ O + O Sheehan and St.-Maurice (2004) −7 T ≤ 1200 K : α = 2.40 × 10 , β = −0.70 −7 T > 1200 K : α = 1.93 × 10 , β = −0.61 CO + e −−−→ CO + O Viggiano et al. (2005) −7 α = 4.20 × 10 , β = −0.75 NO + e −−−→ N + O Sheehan and St.-Maurice (2004) −7 T ≤ 1200 K : α = 3.50 × 10 , β = −0.69 −7 T > 1200 K : α = 3.02 × 10 , β = −0.56 CO + e −−−→ C + O Rose´n et al. (1998) −7 α = 2.75 × 10 , β = −0.55 Radiative recombination O + e −−−→ O + hν Nahar (1999) −12 α = 3.24 × 10 , β = −0.66 6 300 + + O + e CO + e (a) 2 2 C + O CO + O + + NO + e CO + e N + NO N + O + + O + N NO + O + + O + O O + O 3 3 O + h ν O( P) + O( P) 1 1 O + h ν O( S) + O( S) CO + h ν C + O 0.01 0.10 1.00 10.00 100.00 1000.00 10000.00 -3 -1 Production rate [cm s ] + + O + e CO + e 2 2 (b) C + O CO + O + + NO + e CO + e N + NO N + O + + O + N NO + O O + O O + O 3 3 O + h ν O( P) + O( P) 1 1 O + h ν O( S) + O( S) CO + h ν C + O 0.1 1.0 10.0 100.0 1000.0 10000.0 -3 -1 Production rate [cm s ] Figure 1: Production rates of hot O atoms for low (a) and high (b) solar activity due to various sources of oxygen. for which an energy of at least 5.16 eV is needed (see Bond Dissociation Energies in Simple Molecules - B. deB. Darwent. NSRDS-NBS 31, (1970)) . The altitude dependent production rate P(r) for photodissociation of a molecular neutral n can be obtained by the relation pd P(r) = n (r) σ (λ)F(r, λ) dλ (3) n n where n (r) is the altitude dependent neutral density, F(r, λ) the solar ﬂux at the altitude pd r and σ (λ) the photodissociation cross section for the neutral at wavelength λ. The photodissociation and absorption cross sections are taken from Huebner et al. (1992) http://www.nist.gov/data/nsrds/NSRDS-NBS31.pdf Altitude [km] Altitude [km] 2 who provide them in the Photo Rate Coeﬃcient Database . For low and hight solar activity, the production rate of photodissociation of O and CO valid for a solar zenith angle of 60 is illustrated in Figure 1. As can be seen, photodissociation of CO – though signiﬁcant for the production of hot C (see Figure 2) – is of little importance for the production of hot O. 2.3.3. Chemical Reaction and Charge Transfer The rate coeﬃcient k for a two-body chemical reaction as well as for charge transfer is determined by means of the Arrhenius equation T −γ 3 −1 k = α exp cm s , (4) 300 T where T is the gas temperature. The coeﬃcients α, β, and γ are taken from the UDFA06 database described in Woodall et al. (2007) and are listed in Table 6. The production rate for the charge transfer between an ionized and a neutral oxygen atom is calculated −11 0.5 3 −1 using a rate coeﬃcient equal to 1.6×10 (T + T ) cm s taken from Terada et al. n i (2002) with T and T being the neutral and ion temperature, respectively. n i Table 6: Possible sources of hot O atoms with their corresponding coeﬃcients α, β, and γ to calculate the 3 −1 rate coeﬃcient k in cm s . The excess energies for chemical reactions are given for the ground state of the 3 1 + products, e.g. O( P), CO(X Σ ). 3 −1 3 −1 Source Rate coeﬃcient [cm s ] Production rate [cm s ] α β γ HSA LSA Chemical reaction −11 0 0 C + NO −−−→ CN + O + 1.22 eV 6.00 × 10 −0.16 0.0 < 3 × 10 < 2 × 10 1 a −11 C + O −−−→ CO + O( D) + 4.04 eV 4.70 × 10 −0.34 0.0 Figure 1b Figure 1a b −12 2.48 × 10 1.54 613.0 Figure 1b Figure 1a + + −11 −1 −1 C + O −−−→ CO + O + 4.04 eV 5.20 × 10 0.00 0.0 < 5 × 10 < 2 × 10 + + −10 0 −1 C + O −−−→ CO + O + 3.23 eV 3.80 × 10 0.00 0.0 < 2 × 10 < 10 + + −10 −1 0 H + CO −−−→ HCO + O + 0.83 eV 2.90 × 10 0.00 0.0 < 3 × 10 < 2 × 10 + + −9 −2 −2 He + CO −−−→ C + O + He + 2.19 eV 1.60 × 10 0.00 0.0 < 10 < 10 + + −10 −2 −2 He + CO −−−→ CO + O + He + 1.04 eV 8.70 × 10 0.00 0.0 < 10 < 10 + + −9 −2 −2 He + NO −−−→ N + O + He + 3.51 eV 1.40 × 10 0.00 0.0 < 10 < 10 + + −9 −2 −2 He + O −−−→ O + O + He + 5.81 eV 1.10 × 10 0.00 0.0 < 10 < 10 −11 N + NO −−−→ N + O + 3.26 eV 3.75 × 10 0.00 26.0 Figure 1b Figure 1a + + −10 N + O −−−→ NO + O + 4.17 eV 1.80 × 10 0.00 0.0 Figure 1b Figure 1a + + −11 −4 −5 N + NO −−−→ N + O + 2.23 eV 7.90 × 10 0.00 0.0 < 10 < 3 × 10 + + −10 −2 −3 N + O −−−→ NO + O + 6.65 eV 2.63 × 10 0.00 0.0 < 10 < 3 × 10 Charge transfer + + −10 −2 −1 O + H −−−→ H + O + 0.02 eV 5.66 × 10 0.36 −8.6 < 3 × 10 < 2 × 10 + + −11 0 −2 O + O −−−→ O + O + 1.55 eV 1.90 × 10 0.00 0.0 < 10 < 3 × 10 2 2 + + 0.5 −11 O + O −−−→ O + O + 0.00 eV 1.6 × 10 (T + T ) Figure 1b Figure 1a n i for T ≤ 300K for T > 300K http://phidrates.space.swri.edu (retrieved Nov 2012) 8 Table 6 includes all oxygen producing reactions – besides photodissociation and the reactions listed in Tables 1-5 – which are considered in this study. We note that 3 1 the second reaction in Table 6, i.e. C( P) + O mainly produces excited O( D) as mea- sured by Ogryzlo et al. (1973) and Costes and Naulin (1998), although some portion of ground state O( P) cannot be ruled out. For the present purpose we assume all oxygen products of this reaction beeing excited. The change of enthalpy ΔH which is equal to the excess energy and listed in Ta- ble 6 is obtained via Hess’s law. According to this law, the change of enthalpy for a chemical reaction can be calculated by means of the diﬀerence ΔH between the heat of formation of the products and the heat of formation of the reactants, i.e. X X ◦ ◦ ΔH = ΔH − ΔH . (5) f f products reactants If ΔH is negative, the reaction is exothermic, otherwise it is endothermic. The values for the heat of formation ΔH of the products and reactants are taken from the CCCBDB (Computational Chemistry Comparison and Benchmark Database) website . Figure 1 also includes the production rates according to the reactions listed in Table −3 −1 6 for which the maximum value is at least 1 cm s . 2.4. Main Sources of Hot C Atoms 2.4.1. Dissociative and Radiative Recombination Dissociative recombination of CO proceeds through the channels listed in Table 2. Atomic carbon is only produced via the second and fourth channel, however, since the fourth channel represents an endothermic process only the second channel is taken into account. The probability for this channel is low, however, because, as noted by Viggiano et al. (2005), it requires an unlikely rearrangement of the two oxygen atoms. + + − − The CO ion is a linear molecule O C O where the carbon is located in between − − the two oxygen atoms. In order to form the O molecule, the two oxygen atoms have to bond in a very short time. It should be noted, however, that according to Seiersen et al. (2003) the branching ratio of the second channel is 9%, in contrast to almost zero percent as reported by Viggiano et al. (2005). Dissociative recombination of CO takes place via the channels shown in Table 4. 3 1 C atoms can be produced in the ground state C( P) and in the excited states C( D) and C( S). The latter states have energies of 1.26 eV and 2.68 eV, respectively, above the ground state. Radiative recombination of C proceeds via the reaction + 3 C + e −−−→ C( P) + hν, where the parameters of equation (2) are taken from the UDFA06 database (Woodall et al., 2007) based on Nahar and Pradhan (1997). As for O , the production rates for radia- + −5 −3 −1 tive recombination of C are low, having a maximum of 10 cm s at 150 km for −6 −3 −1 high solar activity and of 3 × 10 cm s at 140 km for low solar conditions. NIST Computational Chemistry Comparison and Benchmark Database. NIST Standard Reference Database Number 101 Release 15b, August 2011; Editor: Russell D. Johnson III; http://cccbdb.nist.gov/ 9 The coeﬃcients α and β for dissociative and radiative recombination are given in Table 7. The correponding atomic carbon production rates are illustrated in Figure 2 for low and high solar activity. 2.4.2. Photodissociation In addition to dissociative recombination, photodissociation of CO is another im- portant source of hot C atoms 3 3 CO + hν −−−→ C( P) + O( P) with a threshold energy of 11.14 eV (see Bond Dissociation Energies in Simple Molecules - B. deB. Darwent. NSRDS-NBS 31, (1970)) . The production rate proﬁles of atomic carbon due to photodissociation of CO are included in Figure 2. 2.4.3. Chemical Reaction and Charge Transfer The coeﬃcients α, β, and γ for some additional reaction that produce hot atomic carbon are listed in Table 8. The corresponding production rates for high and low solar activity are also given, showing that hot carbon is created by these reactions with a low rate only; therefore they are not displayed in Figure 2. 2.5. Total and Diﬀerential Cross Sections The motion of the hot particles through the thermosphere is dominated by colli- sions with the background atmosphere. Thus the appropriate treatment of collisions is a crucial task in the simulation of hot coronae. Depending on the collision model, e.g. a hard sphere approximation or energy dependent cross sections, a wide range of outcome can be obtained (e.g. Lichtenegger et al., 2009; Gro¨ller et al., 2010). In the present paper, collisions are modeled by using total and diﬀerential cross sections. Collisions between energetic particles and neutral background constituents can be either elastic, inelastic, or quenching. In elastic collisions the kinetic energy is con- served, only the scattering angle can change. During inelastic collisions kinetic energy http://www.nist.gov/data/nsrds/NSRDS-NBS31.pdf Table 7: Possible sources of hot C atoms due to dissociative and radiative recombination and references of the rate coeﬃcients. 3 −1 Source Rate coeﬃcient [cm s ] Dissociative recombination CO + e −−−→ C + O Viggiano et al. (2005) 2 2 −7 α = 4.20 × 10 , β = −0.75 CO + e −−−→ C + O Rose´n et al. (1998) −7 α = 2.75 × 10 , β = −0.55 Radiative recombination C + e −−−→ C + hν Nahar and Pradhan (1997) −12 T ≤ 7950 K : α = 4.67 × 10 , β = −0.60 10 300 (a) CO + e CO + h ν C + O CO + e 0.01 0.10 1.00 10.00 100.00 -3 -1 Production rate [cm s ] CO + h ν C + O (b) CO + e CO + e 0.01 0.10 1.00 10.00 100.00 -3 -1 Production rate [cm s ] Figure 2: Production rates of hot C atoms for low (a) and high (b) solar activity due to diﬀerent sources of carbon. can be converted into vibrational energy and into electronic excitation, the latter being usually less likely. Quenching collisions describe the de-excitation of the reactant by collision, i.e. the conversion of the internal energy of the reactant into kinetic energy. The references for the used total and diﬀerential collision cross sections (σ and dσ) as well as the fraction of energy ( f ) which goes into internal energy can be found in Tables 9 and 10. Inelastic collision of two oxygen atoms are not considered because the excitation energy in this case is higher than the energy released during the collision. 3 1 1 2.5.1. Collisions between O( P, D, S) and O Collisions between energetic (hot) oxygen and a neutral thermal atom are either elastic 3 1 1 3 1 1 O ( P, D, S) + A → O ( P, D, S) + A , (6) th h h h Altitude [km] Altitude [km] where A denotes the thermal background particles and O the hot oxygen, or quench- th h ing 1 1 3 O ( D, S) + A → O ( P) + A + ΔE , (7) th h q h h 1 1 with the released energy ΔE being 1.97 eV and 4.19 eV for O( D) and O( S), respec- tively. Since no data are available for the diﬀerential cross section of quenching colli- 1 1 sions of excited O( D) and O( S) atoms, we assume them to be similar to the elastic collisions of O( D) atoms. The cross sections simulated by Yee and Dalgarno (1985), Yee and Dalgarno (1987), Yee et al. (1990), and Tully and Johnson (2001) for various types of collisions are illustrated in Figure 3a. Since hot particles may gain energies up to about 10 eV, the values for the cross sections are extrapolated to this interval. 3 1 1 2.5.2. Collisions between O( P, D, S) and N Collisions between a suprathermal O atom and a neutral atmospheric background molecule can be Table 8: Possible sources of hot C atoms and the corresponding coeﬃcients α, β, and γ to calculate the rate 3 −1 coeﬃcient k in cm s . The excess energies for chemical reactions are given for the ground state of the 3 1 + products, e.g. C( P), CO(X Σ ). 3 −1 3 −1 Source Rate coeﬃcient [cm s ] Production rate [cm s ] α β γ HSA LSA Chemical reaction + + −10 −3 −3 C + NO −−−→ NO + C + 1.99 eV 5.20 × 10 0.0 0.0 < 6 × 10 < 4 × 10 + + −11 −4 −4 He + CO −−−→ O + C + He + 1.04 eV 1.10 × 10 0.0 0.0 < 10 < 2 × 10 2 2 + + −10 −2 −2 N + CO −−−→ NO + C + 0.67 eV 1.45 × 10 0.0 0.0 < 3 × 10 < 10 Table 9: Total and diﬀerential collision cross sections (σ, dσ) for elastic and quenching collisions between O atoms together with the available energy ranges. Hot Particle Collision Type References O( P) elastic Tully and Johnson (2001) σ: 1 − 10 eV (Figure 4) dσ: 1, 5, and 10 eV (Figure 2) O( D) elastic Yee and Dalgarno (1987) σ: 0 − 5 eV (Figure 4) dσ: 0.1 and 1.0 eV (Figure 2b and 3b) quenching Yee et al. (1990) σ: 0.001 − 5 eV (Figure 4) dσ: like O( D) elastic O( S) elastic Yee and Dalgarno (1985) σ: 0 − 5 eV (Figure 4) dσ: 1 and 5 eV (Figure 2b and 3b) quenching σ: like O( D) elastic dσ: like O( D) elastic 12 Table 10: Total and diﬀerential collision cross sections (σ, dσ) used in this study for elastic, inelastic, and quenching collisions between O and N , as well as the fraction ( f ) of the kinetic energy which changes into internal energy of the colliding molecule together with the available energy ranges. Hot Particle Collision Type References O( P) elastic Balakrishnan et al. (1998b) σ: 0 - 3 eV (Figure 1) Balakrishnan et al. (1998a) dσ: 1, 2, and 3 eV (Figure 3 to 5) inelastic Balakrishnan et al. (1998b) σ: 0 - 3 eV dσ: 3 eV (Figure 2) f : 0 - 3 eV (Figure 8) O( D) elastic Balakrishnan et al. (1999) σ: 0 - 2 eV (Figure 1) dσ: like O( P) inelastic σ: like O( P) dσ: like O( P) Tachikawa et al. (1997) f : 0 - ∼ 0.8 eV (Table 1) quenching Matsumi and Sarwaruddin Chowdhury (1996) σ: 0 - ∼ 0.9 eV (Figure 9) dσ: like O( P) Zahr et al. (1975) f : 0 - 1.2 eV (Table II) O( S) elastic, inelastic, and quenching like O( D) (a) elastic 3 1 1 3 1 1 O ( P, D, S) + M → O ( P, D, S) + M , (8) th h h h where M denotes the thermal background molecule and O the hot oxygen atom, th h (b) quenching 1 1 3 O ( D, S) + M → O ( P) + M + ΔE , (9) th h q h h with the released energy ΔE being a fraction of the excitation energies, or (c) inelastic 3 1 1 3 1 1 O ( P, D, S) + M → O ( P, D, S) + M − ΔE , (10) th h i h h where ΔE is the kinetic energy which is converted into internal energy of the products. The total and diﬀerential collision cross sections for the collisions between O( P, 1 1 D, S) atoms and N molecules are taken from the references listed in Table 10. In addition, energy ranges for which the cross sections are available are given together with the fraction f of the kinetic energy which goes into internal energy of the col- liding molecule. The total collision cross section for inelastic O( P)−N collisions is 13 10000.00 Tully and Johnson (2001) Yee and Dalgarno (1987) Yee and Dalgarno (1985) 1000.00 Yee et al. (1990) 100.00 -15 2 3 x 10 cm 10.00 1.00 0.10 0.01 0.001 0.010 0.100 1.000 10.000 Energy [eV] 1000.0 Balakrishnan et al. (1998b) elastic: Balakrishnan et al. (1999) 100.0 -15 2 3 x 10 cm 10.0 1.0 inelastic: Balakrishnan et al. (1998b) quenching: Matsumi and Sarwaruddin Chowdhury (1996) 0.1 0 2 4 6 8 10 Energy [eV] Figure 3: Total collision cross sections for (a) elastic and quenching collisions between O atoms, (b) elastic, inelastic, and quenching collisions between O atoms and N molecules. The indicated constant value of −15 2 3 × 10 cm was used by some others in the past. calculated according to the fraction of the frequencies for elastic and inelastic colli- sions given in Figure 6 of Balakrishnan et al. (1998b). The fraction f of the energy which is deposited into N vibrational and rotational excitation in percent of quenching O( D)−N collisions is obtained from Figure 3 and Equation 4.2 of Zahr et al. (1975) for the three given collision energies of 0.03, 0.6, and 1.2 eV. For elastic collisions between O( P) and N no diﬀerential cross sections are available, therefore the dif- ferential cross sections for N−N collisions published in Balakrishnan et al. (1998a) (Figures 3 to 5) are used. The cross sections of Zahr et al. (1975), Tachikawa et al. (1997), Balakrishnan et al. (1998b), and Balakrishnan et al. (1999) are based on sim- ulations whereas those of Matsumi and Sarwaruddin Chowdhury (1996) are obtained from measurements. The total collision cross sections are shown in Figure 3b together with the constant cross section used by e.g. Fox and Hac´ (2009). Since the simulated particles can achieve energies up to about 10 eV, the cross sections are extrapolated to this interval. -16 2 -16 2 Cross section [10 cm ] Cross section [10 cm ] 3 1 1 2.5.3. Collisions between O( P, D, S) and other atoms and molecules Due to the lack of data for collisions between hot oxygen and CO or CO, the same total and diﬀerential cross sections and energy loss rates as for O−N collisions are employed. Collisions of hot oxygen with H and He are treated by means of the elastic collision cross section for O−H collisions given in Zhang et al. (2009). For the collisions of hot oxygen atoms with other background atoms, the data for O−O collisions are used. 3 1 1 2.5.4. Collisions between C( P, D, S) and atoms or molecules Since no data for collisions between hot atomic carbon and atoms or molecules of the background atmosphere are available, the same total and diﬀerential cross sections and energy loss rates as for collisions of hot oxygen with atoms and molecules are assumed, see Tables 9 and 10. 3. Results and Discussion According to the input parameters used in our simulations, the calculated exobase altitude for low and high solar activity at Mars is located at about 220 and 260 km, respectively, above the surface. Once the energy distribution functions of hot O and C atoms at the exobase, at 380 km and at 450 km altitude are determined, escape ﬂuxes as well as density proﬁles of the suprathermal atoms can be estimated. Assuming that the SZA=60 eroded models represent an average over the martian dayside, global escape rates are obtained by integrating the escape ﬂuxes based on the according EDF and by multiplying the resulting total escape ﬂux by the hemispheric area of Mars. Table 11: Escape ﬂuxes and hemispheric loss rates of hot oxygen atoms according to the present study. The numbers are based on the EDFs at the exobase altitude, 380 km and 450 km altitude, respectively. −2 −1 −1 Source Escape ﬂux [cm s ] Escape rate [ s ] Low solar activity 220 km 380 km 450 km 220 km 380 km 450 km + 7 7 7 25 25 25 O + e −−−→ O + O 1.8 × 10 1.4 × 10 1.3 × 10 1.5 × 10 1.2 × 10 1.2 × 10 + 7 7 7 25 25 25 CO + e −−−→ CO + O 1.5 × 10 1.2 × 10 1.1 × 10 1.3 × 10 1.0 × 10 1.0 × 10 1 5 5 5 23 23 23 O + C −−−→ CO + O( D) 8.4 × 10 6.1 × 10 5.8 × 10 6.9 × 10 5.5 × 10 5.3 × 10 + 5 5 5 23 23 23 O + N −−−→ NO + O 1.0 × 10 0.7 × 10 0.7 × 10 0.8 × 10 0.6 × 10 0.6 × 10 7 7 7 25 25 25 total 3.4 × 10 2.7 × 10 2.5 × 10 2.9 × 10 2.3 × 10 2.3 × 10 High solar activity 260 km 380 km 450 km 260 km 380 km 450 km 7 7 7 25 25 25 O + e −−−→ O + O 2.5 × 10 2.1 × 10 2.0 × 10 2.1 × 10 1.9 × 10 1.9 × 10 + 7 7 7 25 25 25 CO + e −−−→ CO + O 1.3 × 10 1.0 × 10 1.0 × 10 1.1 × 10 0.9 × 10 0.9 × 10 + 5 5 5 23 23 23 O + N −−−→ NO + O 11.4 × 10 9.2 × 10 8.5 × 10 9.6 × 10 8.2 × 10 7.8 × 10 1 5 5 5 23 23 23 O + C −−−→ CO + O( D) 8.1 × 10 5.9 × 10 5.5 × 10 6.8 × 10 5.2 × 10 5.1 × 10 7 7 7 25 25 25 total 4.0 × 10 3.3 × 10 3.1 × 10 3.4 × 10 2.9 × 10 2.9 × 10 15 + Table 12: Comparison of hot oxygen escape rates of various studies due to DR of O for low (LSA) and high solar activity (HSA). 25 −1 References Escape rate × 10 s LSA HSA present study (at the exobase level) 1.5 2.1 present study (at 450 km) 1.2 1.9 Kim et al. (1998) 3.4 8.5 Hodges (2000) 2.8 - Hodges Jr. (2002) 4.4 18.0 Krestyanikova and Shematovich (2006) 4.5 - Cipriani et al. (2007) 3.4 8.5 Cipriani et al. (2007) 0.55 2.6 Chaufray et al. (2007) 1.0 4.0 Fox and Hac´ (2009) 0.66 0.68 Fox and Hac´ (2009) 14.4 21.0 Valeille et al. (2010) 3.8 13 Yagi et al. (2012) 2.9 - 5.3 - corrected by a factor of 6.5 by Nagy et al. (2001); two-stream model and constant collision cross section of −15 2 1.2 × 10 cm b −15 2 constant collision cross section of 2 × 10 cm for O−O and O−CO c −15 2 1 constant collision cross section of 2 × 10 cm for O−O and O−CO and including quenching of O( D) and O( S) Model B: including elastic, inelastic and quenching collisions; collision cross sections are taken from Balakrishnan et al. (1998b) and Kharchenko et al. (2000) atmosphere inputs from Kim et al. (1998) (corrected values) and the collision cross sections taken from Kharchenko et al. (2000) and Tully and Johnson (2001) atmosphere inputs from Krasnopolsky (2002) and the collision cross sections taken from Kharchenko et al. (2000) and Tully and Johnson (2001) single interaction potential and only atomic background atmosphere eroded ion proﬁles, exobase approximation, isotropic scattering and constant collision cross section of −15 2 3 × 10 cm eroded ion proﬁles, forward scattering using diﬀerential cross sections from Kharchenko et al. (2000) and −15 2 constant collision cross section of 3 × 10 cm j −15 2 3D thermospheric inputs from Bougher et al. (2006); constant collision cross section of 2 × 10 cm for O−O and O−CO collisions universal potential model for collisions; variations due to diﬀerent martian seasons 16 3.1. Hot O Atoms The escape ﬂuxes together with the escape rates for various sources of hot oxygen are listed in Table 11 for LSA and HSA. Since the escape rates are calculated from the EDF, they depend on the altitude at which the EDF is taken. At suﬃciently high altitudes, however, where collisions become insigniﬁcant, the escape rates converge towards a constant value. To illustrate this eﬀect, the escape rates based on the EDFs at three diﬀerent altitudes are shown in Table 11. Although the production rates are generally larger during HSA, there is little dif- ference in the escape of hot O between HSA and LSA. This is mainly due to the fact that at HSA also the neutral background density is higher, leading to an increase in the number of collisions between the suprathermal particles and the cold atmosphere and thus to a more eﬃcient thermalization of the hot particles compared to LSA conditions. It should be noted that this eﬀect is only found for the eroded ionosphere models. In case of the non-eroded ionosphere models of Fox and Hac´ (2009) the higher production rates of hot O at HSA is not totally compensated by the more eﬃcient thermalization in the denser atmosphere; indeed in that case the escape rate of hot oxygen produced from O is about 65% higher at HSA than at LSA. Moreover, inspection of Table 11 also reveals that the escape ﬂux due to DR of CO is smaller for HSA than for LSA. This is due to the fact that the production rates in the region close to the exobase are comparable for LSA and HSA (∼0.1 and ∼0.3 −3 −1 cm s ), while the initial energies of the newly produced hot O is diﬀerent because of the diﬀerent ion temperatures at LSA and HSA. The loss rates for both solar conditions are mainly sustained by particles with initial energies between 4.7 . E . 5.8 eV whose production rates are higher at LSA than at HSA. Thus, although the number of particles with E > 5.8 eV is increased at HSA with respect to LSA, the number of particles with E < 5.8 eV is decreased, leading to a net escape ﬂux at HSA smaller than at LSA. The same eﬀect can also be seen for the O + C reaction. In case of O + N, 2 2 however, the escape ﬂux for HSA is higher than for LSA. In this case, the eﬀect of the initial energy distribution plays a minor role with respect to the higher production rate at higher altitudes. Table 11 also illustrates the eﬀect of ignoring the production of hot particles and their collisions with the background atmosphere above the exobase. The loss rates are found to be higher when based on the EDF at the exobase than when based on the EDF at 450 km altitude. This shows that the hot particles produced above the exobase only marginally contribute to the total escape despite the fact that most of these particles will be lost to space if their energy exceeds the escape energy. Rather, the decrease of escaping oxygen due to energy loss in collisions above the exobase somewhat exceeds the gain due to production, so that the escape rate is in fact reduced at 450 km by ∼ 20− 30% with respect to the values at the exobase altitude. In general our simulation results suggest that collision eﬀects are still noticeable up to ∼ 5 − 10 scale heights above the exobase. In Table 12 our escape rates are compared to previous simulations. The diﬀer- ences in the results mainly reﬂect the diﬀerences in the input neutral and ion density proﬁles and the diﬀerent treatment of collisions. Although we use the neutral density 17 350 escape - LSA (eroded) source - LSA (eroded) escape - HSA (eroded) source - HSA (eroded) escape - LSA (non-eroded) source - LSA (non-eroded) 250 escape - HSA (non-eroded) source - HSA (non-eroded) (a) O + e 0.01 0.10 1.00 10.00 100.00 1000.00 10000.00 -3 -1 Production rates [cm s ] escape - LSA (eroded) source - LSA (eroded) escape - HSA (eroded) source - HSA (eroded) escape - LSA (non-eroded) source - LSA (non-eroded) 250 escape - HSA (non-eroded) source - HSA (non-eroded) (b) CO + e 0.001 0.010 0.100 1.000 10.000 100.000 1000.000 -3 -1 Production rates [cm s ] + + Figure 4: Production rates of escaping particles generated by DR of O (a) and CO (b) as a function of 2 2 altitude for both solar activities and for the eroded and non-eroded ionosphere models. The dashed-dotted lines represent the production rates of all (escaping and non-escaping) oxygen atoms created via DR of O (a) and CO (b). and ionosphere model of Fox and Hac´ (2009), the escape rates diﬀer by a factor of 0.5 − 10. The main reason may be attributed to the diﬀerent collision models: While −15 2 Fox and Hac´ (2009) uses a constant collision cross section of 3×10 cm for all types of collisions, we invoke diﬀerent energy dependent collision cross sections (see Tables 9 and 10). In particular, the cross section for collisions with molecules – which are the most important types of collisions up to ∼ 200 km – is about one order of magnitude higher (see Figure 3b) than the above constant value. This higher value signiﬁcantly increases the average number of collisions, thus leading to more eﬃcient thermaliza- tion and to lower escape rates. Finally, while most previous studies reveal a distinct dependence of the hot oxygen loss rates on solar activity, this is neither the case in our results nor in those of Fox and Hac´ (2009) and may be attributed to the use of an eroded ionosphere model. Figure 4 illustrates the production rates of escaping O re- Start Altitude [km] Start Altitude [km] 6000 total O + e CO + e CO + h ν 4000 C + O CO + O + + O + N NO + O CO + e NO + e (a) -1 0 1 2 3 4 5 10 10 10 10 10 10 10 -3 Density [cm ] total O + e CO + e CO + h ν 4000 C + O CO + O + + O + N NO + O CO + e NO + e (b) -1 0 1 2 3 4 5 10 10 10 10 10 10 10 -3 Density [cm ] Figure 5: Exosphere densities of hot oxygen atoms for various sources for low (a) and high (b) solar activity. + + sulting from the DR of O and CO as a function of altitude for both solar activities. 2 2 For comparison, the results based on the eroded and non-eroded ionosphere models of Fox and Hac´ (2009) are shown. The solid lines represent the production rates at each height of those particles (consisting of hot particles generated by photochemical reactions and of background particles which became ”hot” upon collisions with these particles) which arrive at 380 km altitude with an energy exceeding the escape energy. As can be seen, the production rates for the non-eroded models become larger than those for the eroded ones approximately at altitudes, where the O density proﬁles become sensitive to the upward velocity. At higher altitudes, the production rate of escaping hot O created by collisions can even exceed the local production rate of O due to DR of CO . The exosphere density proﬁles based on the energy distribution functions of hot oxygen at the exobase for low and high solar activity are shown in Figure 5. For both low and hight solar activity conditions the total exosphere density is mainly governed Altitude [km] Altitude [km] 1500 (a) this study Kim et al. (1998) Hodges (2000) - noon Krestyanikova and Shematovich (2006) Cipriani et al. (2007) Chaufray et al. (2007) Valeille et al. (2010) - SZA = 60° Feldman et al. (2011) Yagi et al. (2012) 2 3 4 5 10 10 10 10 -3 Density [cm ] (b) this study Kim et al. (1998) Hodges (2000) - noon 750 Cipriani et al. (2007) Chaufray et al. (2007) Valeille et al. (2010) - SZA = 30° Valeille et al. (2010) - SZA = 90° 2 3 4 5 10 10 10 10 -3 Density [cm ] Figure 6: Comparison of exosphere densities of hot oxygen atoms for a number of studies. The curves of Hodges (2000), Chaufray et al. (2007), and Yagi et al. (2012) include also the cold component. + + by DR of O and CO . During HSA, however, at altitudes below ∼ 4000 km, the 2 2 oxygen density due to photodissociation of CO becomes comparable with the density of O produced via DR of CO . It should be noted that, although the production rate of hot O due to NO + e is between one and two orders of magnitudes higher than the one due to CO + e (see Figure 1), the obtained exosphere densities are comparable. The main reason is that the initial energies of the O atoms originating from NO +e are on average much lower than those originating from CO + e. Moreover, although the oxygen production rates due to the reactions C + O and NO + e are comparable above ∼ 130 km (Figure 1), the corresponding exosphere density arising from C + O for HSA and LSA is nearly one order of magnitude higher than that arising from NO + e (Figure 5). This is due to the initial velocities of the oxygen atoms gained via DR of NO , which are much lower than those acquired via the reaction between C and O . Fur- ther, although the production rate of atomic oxygen originating from charge transfer between an ionized and a neutral O becomes dominant at higher altitudes (Figure 1), the contribution of this reaction to the exospheric density is negligible because the ini- Altitude [km] Altitude [km] tial energies involved are similar to the average ion energies of ∼ 0.3 eV. For HSA as well as LSA the production rate of oxygen atoms with an excess energy of 4.17 eV due to reaction O + N is about three orders of magnitude smaller than the production rate due to DR of of O and thus leading also to much lower exosphere densities. Finally, at low solar activity, the oxygen exosphere density resulting from the reaction O + N is about 5-10 times smaller than the O density produced via C+O (Figure 5a) because above ∼ 130 km the production rate is also smaller by a similar amount (Figure 1a). At high solar activity, however, the production rates of these two reactions above ∼ 180 km become comparable (Figure 1b), making also the densities comparable (Figure 5b). In Figure 6 our exosphere densities are compared with previous studies. This Figure reﬂects the variety of results obtained based on diﬀerent assumptions. The use of en- ergy dependent total and diﬀerential cross sections and the inclusion of elastic, inelastic and quenching collisions as done in this study (see Tables 9 and 10) appear to favour lower exosphere densities than other assumptions (Table 12). In addition, in our sim- ulations the collision of ground-state oxygen atoms with the atmospheric background is modeled by means of the total and diﬀerential cross sections of Tully and Johnson (2001) (see Table 9). As shown by Gro¨ller et al. (2010) the use of the corresponding cross sections published by Kharchenko et al. (2000) would lead to somewhat higher densities. 3.2. Hot C Atoms The escape ﬂuxes as well as the escape rates for the most eﬃcient sources of hot C + + are listed in Table 13. As opposed to oxygen, where DR of O and CO are the most 2 2 important suppliers for hot O escape, the C loss is mainly due to photodissociation of CO. As in the case of oxygen, the higher production rates at HSA are at least partially compensated by the simultaneously increased number of collisions with the denser background atmosphere, leading to escape ﬂuxes at HSA of about four times the ﬂux at LSA. Regarding DR of CO , the corresponding carbon escape ﬂux at HSA is even lower than at LSA. It should also be noted that the carbon production and escape rate due to photodissociation of CO, i.e. of a neutral molecule, is not aﬀected by the erosion of the ionosphere, in contrast to the production and escape of oxygen due to DR of O - and CO -ions. Finally, the total loss rates calculated by means of the EDFs at 450 km altitude are about ∼ 25− 30% smaller than the loss rates obtained by using the EDFs at the exobase level. Thus collisions above the exobase are again still suﬃcient to reduce the ﬂux of escaping carbon by a noticeable amount. In Table 14 the escape rates of hot atomic carbon due to various source terms obtained in previous simulations are shown. As discussed in Fox (2004), a detailed analysis of the diﬀerences among the escape ﬂuxes obtained by Nagy et al. (2001), Fox and Bakalian (2001), and Fox (2004) is diﬃcult, but may be related to the use of diﬀerent solar ﬂuxes, dissociative recombination coeﬃcients and exobase altitudes. The results of Cipriani et al. (2007) are signiﬁcantly lower than the aforementioned values; unfortunately, these authors do not discuss any possible reasons for this dis- crepancy. Upon comparing our computed carbon escape ﬂuxes (Table 13) with those pub- lished in Fox (2004) (Table 14), it is seen that the ﬂuxes at the exobase due to pho- todissociation of CO are higher by a factor of ∼ 1.5 and ∼ 1.3 for LSA and HSA, 21 Table 13: Escape ﬂuxes and hemispheric loss rates of hot carbon atoms according to the present study. The numbers are based on the EDFs at the exobase altitude, 380 km and 450 km altitude, respectively. −2 −1 −1 Source Escape ﬂux [cm s ] Escape rate [ s ] Low solar activity 220 km 380 km 450 km 220 km 380 km 450 km 5 5 5 23 23 23 CO + hν −−−→ C + O 10.6 × 10 7.8 × 10 7.1 × 10 8.7 × 10 6.9 × 10 6.5 × 10 + a 5 5 5 23 23 23 CO + e −−−→ O + C 1.4 × 10 1.0 × 10 0.9 × 10 1.1 × 10 0.9 × 10 0.8 × 10 2 2 5 5 5 23 23 23 CO + e −−−→ C + O 0.8 × 10 0.7 × 10 0.6 × 10 0.7 × 10 0.6 × 10 0.6 × 10 5 5 5 23 23 23 total 12.8 × 10 9.5 × 10 8.6 × 10 10.5 × 10 8.4 × 10 7.9 × 10 High solar activity 260 km 380 km 450 km 260 km 380 km 450 km 5 5 5 24 24 24 CO + hν −−−→ C + O 45.1 × 10 34.4 × 10 30.4 × 10 3.8 × 10 3.1 × 10 2.8 × 10 + 5 5 5 24 24 24 CO + e −−−→ C + O 3.4 × 10 3.3 × 10 3.3 × 10 0.3 × 10 0.3 × 10 0.3 × 10 a 5 5 5 24 24 24 CO + e −−−→ O + C 1.2 × 10 0.8 × 10 0.8 × 10 0.1 × 10 0.1 × 10 0.1 × 10 2 2 5 5 5 24 24 24 total 49.7 × 10 38.5 × 10 34.5 × 10 4.2 × 10 3.5 × 10 3.2 × 10 the branching ratio of this process is assumed to be 4% respectively. Our escape ﬂuxes at the exobase related to DR of CO are about a factor of ∼ 4.5 and ∼ 2.4 higher for low and high solar activity, respectively, whereas our escape ﬂuxes due to DR of CO are about a factor of ∼ 1.6 and ∼ 2.3 lower than the one given in Fox (2004) who assumes the exobase locations at 184 and 220 km altitude for low and high solar acitivity. Besides the diﬀerences in the exobase heights, these latter lower values can be explained mainly by our lower production rates (Figure 1) which are based on the CO density proﬁles of Fox and Hac´ (2009) being smaller than those given in Fox (2004). The higher escape rates due to DR of CO compared to Fox (2004) may be some- what surprising at ﬁrst glance, since we assume a branching ratio of only 4% for the C + O channel of DR of CO , which is lower than the 9% assumed by Fox (2004). 2 2 Moreover, our production rates of carbon are smaller than those calculated in Fox (2004) because we use the CO proﬁles of Fox and Hac´ (2009) which correspond to lower densities than those in Fox (2004). That our escape rates of carbon produced by DR of CO are nevertheless higher is at least partially due to the fact that we consider a higher initial velocity for carbon. While Fox (2004) assumes that the vibrational- rotational states of the resulting O molecule and the translational states of the C and O fragments are populated statistically, we assume O to be in the electronic and 2 2 vibrational-rotational ground state, which can be considered as an upper limit for the resulting velocities of the C atoms. In the latter case the velocity distribution peaks at ∼ 1.7 eV, in the former case at ∼ 0.6 eV. Another reason for our higher carbon escape rates is due to the diﬀerent determination of these rates. Fox (2004) applies the exobase approximation, where the escape ﬂux is estimated by integrating the production rates above the exobase. Our estimation, however, is based on the integration of the energy distribution function, yielding higher escape rates than the former method – especially in case of ﬂat production rate proﬁles like the one for CO + e. 22 Table 14: Escape ﬂuxes of hot atomic carbon obtained in previous studies. 5 −2 −1 Reaction Escape ﬂux × 10 cm s low solar activity high solar activity Nagy et al. (2001) hemispheric escape ﬂuxes CO + hν −−−→ C + O - 60 CO + e −−−→ C + O - 17 total 5.33 77 Fox and Bakalian (2001) global average escape ﬂuxes CO + hν −−−→ C + O 1.65 18 CO + e −−−→ C + O 0.29 6.2 total 1.94 24.2 total (including additional sources) 2.1 26 Fox (2004) hemispheric escape ﬂuxes CO + hν −−−→ C + O 7.3 35 CO + e −−−→ C + O 1.28 7.7 CO + e −−−→ C + O 0.31 0.49 2 2 total 8.89 43.19 total (including additional sources) 11.1 64 Cipriani et al. (2007) hemispheric escape ﬂuxes CO + e −−−→ C + O 0.005 0.45 a −15 2 cross section of 1.2 × 10 cm ; for LSA only the sum of the rates for the two reactions is given by Nagy et al. (2001) b −15 2 cross section of 3 × 10 cm ; exobase at ∼ 188 km for LSA and ∼ 220 km for HSA hemispheric escape ﬂuxes are twice as large atmosphere inputs from Kim et al. (1998) and the collision cross sections used from Kharchenko et al. (2000) and Tully and Johnson (2001) e −1 calculated from the escape rates [s ] given in Table 1 of Cipriani et al. (2007) by dividing through the martian hemispheric surface at 400 km altitude 23 6000 total CO + h ν CO + e CO + e (a) 0.1 1.0 10.0 100.0 1000.0 -3 Density [cm ] total CO + h ν CO + e CO + e (b) 0.1 1.0 10.0 100.0 1000.0 -3 Density [cm ] Figure 7: Exosphere densities of hot carbon atoms for various sources for low (a) and high (b) solar activity. The exosphere density proﬁles for diﬀerent source terms of the hot C atoms are shown in Figure 7 for low and high solar activity. In Figure 8 the carbon exosphere den- sity obtained by a number of studies is illustrated. It should be noted that Nagy et al. (2001) include photodissociation of CO, dissociative recombination of CO and colli- sions between energetic oxygen and thermal carbon atoms as sources of hot C, while Cipriani et al. (2007) consider only dissociative recombination of CO . Our results suggest – in agreement with previous studies – that the main loss of atomic carbon at present Mars is due to photodissociation of CO. According to the 24 −1 present simulations, the total carbon loss is about 0.8 and 3.2 × 10 s for low and high solar activity (Table 13) , respectively, which is ∼ 5−15 times higher compared to the estimated average atmospheric sputtering of C atoms (Chasseﬁe`re et al., 2007) and up to ∼ 40 times higher compared to the estimated CO molecular ion escape from ASPERA-3 (Barabash et al., 2007). The present results are in accord with the sugges- tion of Lammer et al. (2013) that the escape of photochemically produced suprathermal Altitude [km] Altitude [km] 6000 this study - LSA this study - HSA Nagy et al. (2001) - LSA Nagy et al. (2001) - HSA Cipriani et al. (2007) - HSA 1 10 100 1000 -3 Density [cm ] Figure 8: Comparison of the obtained exosphere densities of hot carbon atoms with previous studies. + + C atoms, which originate from the dissociation of CO , CO and CO are the most eﬃ- cient processes for the loss of the martian CO atmosphere at present. Therefore, these exothermic reactions may play a major role for the escape of the martian atmosphere since the end of the Noachian. It is important to note that the modeled hot C and O and the related CO losses depend on a complex interplay of many physical and chemical processes – e.g. the change of the solar XUV ﬂux and the corresponding response of the upper neutral and ionized atmosphere – and that the present escape rates can not be easily extrapolated backwards during the planet’s history. Since CO +e dissociates mainly into CO and O and only with 4% probability into C and O (see Table 2), it follows O/C=24/1. On the other hand, the corresponding ratio 25 23 of the escape rates (for e.g. LSA) is 1.0 × 10 /(0.8 × 10 ) = 125, i.e. oxygen can escape ﬁve times more eﬃcient than carbon. This imbalance in the loss rates of O and C can be at least partially attributed to the diﬀerence in the energies released by the two channels producing atomic oxygen and carbon (Table 2), providing oxygen with a 3.3 times higher energy than carbon. 4. Summary and Conclusion A number of possible sources for hot C and O atoms and the resulting escape rates during low and high solar activity conditions from present Mars have been studied by applying a Monte-Carlo model to simulate the stochastic motion of these particles through the upper atmosphere. The main sources of escaping oxygen are dissociative + + 25 −1 recombination of O and CO leading to a total loss of ∼ 2 − 3 × 10 s over 2 2 the solar cycle when an eroded ionosphere model is assumed. The vast majority of escaping carbon is delivered through photodissociation of CO with a total loss of the 24 −1 order of 0.5 to 3 × 10 s over the solar cycle. This value is distinctly higher than the Altitude [km] + estimated present loss due to atmospheric sputtering and the observed CO escape at Mars. The calculated loss rates based on the energy distribution function at 450 km alti- tude are reduced by ∼ 20 − 30% with respect to the values calculated by means of the EDFs taken at the exobase level. This suggests that collisions still play a role above the exobase in a region with an extension of ∼ 5 − 10 times the local scale height. It should also be emphasized that the modeled oxygen loss is due to the reactions + + listed in Tables 1-6 – above all due to dissociative recombination of O and CO – 2 2 and is therefore not directly related to the loss of water. On the other hand, the O escape cannot be balanced by C escape since the former appears to be 10–30 times higher than the latter. This suggests – assuming the martian atmosphere to be in steady state over long time scales – that all oxygen lost by escape may be ﬁnally replenished by means of H O via a hydrogen-carbon chemistry rather than converting CO into CO, thereby 2 2 stabilizing the CO atmosphere. Acknowledgement The authors wish to thank two anonymous referees, whose constructive comments helped to improve the paper. H. Gro¨ller, H. Lammer and H. Lichtenegger acknowledge support from the Helmholtz Alliance project ”Planetary Evolution and Life”, H. Gro¨ller and H. Lichtenegger also acknowledge support by the Austrian Science Fund (FWF) under the project P24247-N16. H. Lammer acknowledges also the support by the FWF NFN project S116 “Pathways to Habitability: From Disks to Active Stars, Planets and Life”, and the related FWF NFN subproject, S116607-N16 ”Particle/Radiative Interac- tions with Upper Atmospheres of Planetary Bodies Under Extreme Stellar Conditions”. V. Shematovich acknowledges support from the Basic Research Program of the Presid- ium of the Russian Academy of Sciences (Program 22), Russian Foundation for Basic Research (Project 11-02-00479a), Federal Targeted Program ”Science and Science Ed- ucation for Innovation in Russia 2009-2013”. References Balakrishnan, N., Kharchenko, V., Dalgarno, A., 1998a. Quantum mechanical and semiclassical studies of N+N collisions and their application to thermalization of fast N atoms. The Journal of Chemical Physics 108 (3), 943–949. Balakrishnan, N., Kharchenko, V., Dalgarno, A., 1998b. Slowing of energetic O( P) atoms in collisions with N . Journal of Geophysical Research 103 (A10), 23393– Balakrishnan, N., Kharchenko, V., Dalgarno, A., 1999. Translational energy relaxation of hot O( D) atoms. The Journal of Physical Chemistry A 103, 3999–4001. Barabash, S., Fedorov, A., Lundin, R., Sauvaud, J.-A., 2007. Martian atmospheric erosion rates. Science 315 (5811), 501–503. 26 Bauer, S. J., Lammer, H., 2004. Planetary Aeronomy - Atmosphere Environments in Planetary Systems. Springer. Bougher, S. W., Bell, J. M., Murphy, J. R., Lopez-Valverde, M. A., Withers, P. G., 2004. Polar warming in the Mars thermosphere: Seasonal variations ow- ing to changing insolation and dust distributions. Geophysical Research Letters 32, doi:10.1029/2005GL024059. L02203. Chasseﬁe`re, E., Leblanc, F., Langlais, B., 2007. The combined eﬀects of escape and magnetic ﬁeld histories at Mars. Planetary and Space Science 55 (3), 343–357. Chasseﬁe`re, E., Leblanc, F., 2011. Constraining methane release due to serpentiniza- tion by the observed D/H ratio on Mars. Earth and Planetary Science Letters 310, 262–271. Chasseﬁe`re, E., Leblanc, F., 2011. Explaining the redox imbalance between the H and O escape ﬂuxes at Mars by the oxidation of methane. Planetary and Space Science 59, 218–226. Chaufray, J. Y., Modolo, R., Leblanc, F., Chanteur, G., Johnson, R. E., Luhmann, J. G., 2007. Mars solar wind interaction: Formation of the Martian corona and atmospheric loss to space. Journal of Geophysical Research 112 (E9), E09009. Cipriani, F., Leblanc, F., Berthelier, J. J., 2007. Martian corona: Nonthermal sources of hot heavy species. Journal of Geophysical Research 112 (E7), E07001. Claire, M. W., Sheets, J., Cohen, M., Ribas, I., Meadows, V. S., Catling, D. C., 2012. The evolution of solar ﬂux from 0.1 nm to 160 μm: Quantitative estimates for plan- etary studies. The Astrophysical Journal 757 (1), 95. Costes, M., Naulin, C., 1998. State-to-state cross sections for the C( P ) + 3 – 1 + 1 O (X Σ ) −−−→ CO(X Σ ) + O( D ) reaction at kinetic energies between 4.4 and 2 g 2 90 meV. Comptes Rendus de l’Acadmie des Sciences - Series IIC - Chemistry 1 (12), 771–775. Erkaev, N. V., Lammer, H., Elkins-Tanton, L., Odert, P., Kislyakova, K. G., Kulikov, Y. N., Leitzinger, M., Gu¨del, M., 2013. Escape of the martian protoatmosphere and initial water inventory. Planetary and Space Science, submitted. Feldman, P. D. et al., 2011. Rosetta-Alice observations of exospheric hydrogen and oxygen. Icarus 214, 394–399. Fox, J. L., 2004. CO dissociative recombination: A source of thermal and nonthermal C on Mars. Journal of Geophysical Research 109 (A8), A08306. Fox, J. L., Bakalian, F. M., 2001. Photochemical escape of atomic carbon from Mars. Journal of Geophysical Research 106 (A12), 28785–28795. Fox, J. L., Hac´, A. B., 2009. Photochemical escape of oxygen from Mars: A compari- son of the exobase approximation to a Monte Carlo method. Icarus 204 (2), 527–544. 27 Gro¨ller, H., Lammer, H., Lichtenegger, H. I. M., Pﬂeger, M., Dutuit, O., Shematovich, V. I., Kulikov, Y. N., Biernat, H. K., 2012. Hot oxygen atoms in the Venus nightside exosphere. Geophysical Research Letters 39 (3), L03202. Gro¨ller, H., Shematovich, V. I., Lichtenegger, H. I. M., Lammer, H., Pﬂeger, M., Ku- likov, Y. N., Macher, W., Amerstorfer, U. V., Biernat, H. K., 2010. Venus’ atomic hot oxygen environment. Journal of Geophysical Research 115, E12017. Hodges, R. R., 2000. Distributions of hot oxygen for Venus and Mars. Journal of Geo- physical Research 105 (E3), 6971–6981. Hodges Jr., R. R., 2002. The rate of loss of water from Mars. Geophysical Research Letters 29 (3), 1038–1040. Huebner, W. F., Keady, J. J., Lyon, S. P., 1992. Solar photo rates for planetary atmo- spheres and atmospheric pollutants. Astrophysics and Space Science 195 (1), 1–294. Kella, D. P., Johnson, P. J., Pedersen, H. B., Vejby-Christensen, L., Andersen, L. H., 1997. The source of green light emission determined from a heavy-ion storage ring experiment. Science 276, 1530–1533. Kharchenko, V., Dalgarno, A., Zygelman, B., Yee, J., 2000. Energy transfer in colli- sions of oxygen atoms in the terrestrial atmosphere. Journal of Geophysical Research 105 (A11), 24899–24906. Kim, J., Nagy, A. F., Fox, J. L., Cravens, T. E., 1998. Solar cycle variability of hot oxygen atoms at Mars. Journal of Geophysical Research 103 (A12), 29339–29342. Krasnopolsky, V. A., 2002. Mars’ upper atmosphere and ionosphere at low, medium, and high solar activities: Implications for evolution of water. Journal of Geophysical Research 107 (E12), 5128. Krestyanikova, M. A., Shematovich, V. I., 2006. Stochastic models of hot planetary and satellite coronas: A hot oxygen corona of Mars. Solar System Research 40 (5), 384–392. Lammer, H., Chasseﬁe`re, E., Karatekin, O., Morschhauser, A., Niles, P. B., Mousis, O., Odert, P., Mo¨stl, U. V., Breuer, D., Dehant, V., Grott, M., Gro¨ller, H., Hauber, E., Pham, L. B. S., 2013. Outgassing history and escape of the martian atmosphere and water inventory. Space Science Reviews 174 (1-4), 113–154. Leblanc, F., Chaufray, J. Y., Lilensten, J., Witasse, O., Bertaux, J.-L., 2006. Martian dayglow as seen by the SPICAM UV spectrograph on Mars Expresss. Journal of Geophysical Research 111 (E9), E09S11. Lichtenegger, H., Lammer, H., Kulikov, Y., Kazeminejad, S., Molina-Cuberos, G., Ro- drigo, R., Kazeminejad, B., Kirchengast, G., 2006. Eﬀects of low energetic neutral atoms on martian and venusian dayside exospheric temperature estimations. Space Science Reviews 126, 469–501. 28 Lichtenegger, H. I. M., Gro¨ller, H., Lammer, H., Kulikov, Y. N., Shematovich, V. I., 2009. On the elusive hot oxygen corona of Venus. Geophysical Research Letters 36 (10), L10204. Matsumi, Y., Sarwaruddin Chowdhury, A. M., 1996. Translational relaxation and elec- tronic quenching of hot O( D) by collisions with N . The Journal of Chemical Physics 104 (18), 7036–7044. McElroy, M. B., 1972. Mars: An evolving atmosphere. Science 175 (4020), 443–445. Nagy, A. F., Cravens, T. E., 1988. Hot oxygen atoms in the upper atmospheres of Venus and Mars. Geophysical Research Letters 15 (5), 433–435. Nagy, A. F., Liemohn, M. W., Fox, J. L., Kim, J., 2001. Hot carbon densities in the exosphere of Mars. Journal of Geophysical Research 106 (A10), 21565–21568. Nahar, S. N., 1999. Electron-Ion recombination rate coeﬃcients, photoionization cross sections, and ionization fractions for astrophysically abundant elements. II. Oxygen ions. The Astrophysical Journal Supplement Series 120 (1), 131–145. Nahar, S. N., Pradhan, A. K., 1997. Electron-Ion recombination rate coeﬃcients, pho- toionization cross sections, and ionization fractions for astrophysically abundant ele- ments. I. Carbon and nitrogen. The Astrophysical Journal Supplement Series 111 (1), 339–355. Ogryzlo, E., Reilly, J., Thrush, B., 1973. Vibrational excitation of CO from the reaction C + O . Chemical Physics Letters 23 (1), 37–39. Petrignani, A., Hellberg, F., Thomas, R. D., Larsson, M., Cosby, P. C., van der Zande, W. J., 2005. Electron energy-dependent product state distributions in the dissocia- tive recombination of O . The Journal of Chemical Physics 122 (23), 234311–1– 234311–8. Peverall, R., Rose´n, S., Peterson, J., Larsson, M., Al-Khalili, A., Vikor, L., Semaniak, J., Bobbenkamp, R., Le Padellec, A., Maurellis, A., van der Zande, W., 2001. Dis- sociative recombination and excitation of O : Cross sections, product yields and implications for studies of ionospheric airglows. The Journal of Chemical Physics 114 (15), 6679–6689. Ribas, I., Guinan, E. F., Gudel, M., Audard, M., 2005. Evolution of the solar activ- ity over time and eﬀects on planetary atmospheres. I. High-energy irradiances (1- 1700 Å). The Astrophysical Journal 622 (1), 680–694. Rose´n, S., Peverall, R., Larsson, M., Le Padellec, A., Semaniak, J., Larson, ., Strmholm, C., van der Zande, W. J., Danared, H., Dunn, G. H., 1998. Absolute cross sections and ﬁnal-state distributions for dissociative recombination and excitation of CO (v=0) using an ion storage ring. Physical Review A 57 (6), 4462–4471. Seiersen, K., Al-Khalili, A., Heber, O., Jensen, M. J., Nielsen, I. B., Pedersen, H. B., Safvan, C. P., Andersen, L. H., 2003. Dissociative recombination of the cation and dication of CO . Physical Review A 68 (2), 022708. 29 + + + Sheehan, C. H., St.-Maurice, J., 2004. Dissociative recombination of N , O , and NO : 2 2 Rate coeﬃcients for ground state and vibrationally excited ions. Journal of Geophys- ical Research 109 (A3), A03302. Tachikawa, H., Ohnishi, K., Hamabayashi, T., Yoshida, H., 1997. Translational re- laxation of hot O( D) by inelastic collision with N molecule: Ab initio MO and classical trajectory studies. The Journal of Chemical Physics A 101, 2229–2232. Terada, N., Machida, S., Shinagawa, H., 2002. Global hybrid simulation of the Kelvin−Helmholtz instability at the Venus ionopause. Journal of Geophysical Re- search 107 (A12), 1471. Tian, F., Kasting, J. F., Solomon, S. C., 2009. Thermal escape of carbon from the early martian atmosphere. Geophysical Research Letters 36 (2), L02205. Tobiska, W. K., 2004. SOLAR2000 irradiances for climate change research, aeronomy and space system engineering. Advances in Space Research 34 (8), 1736–1746. Tully, C., Johnson, R. E., 2001. Low energy collisions between ground-state oxygen atoms. Planetary and Space Science 49 (6), 533–537. Valeille, A., Combi, M. R., Tenishev, V., Bougher, S. W., Nagy, A. F., 2010. A study of suprathermal oxygen atoms in Mars upper thermosphere and exosphere over the range of limiting conditions. Icarus 206 (1), 18–27. Vejby-Christensen, L., Kella, D., Pedersen, H. B., Andersen, L. H., 1998. Dissociative recombination of NO . Physical Review A 57 (5), 3627–3634. Viggiano, A. A., Ehlerding, A., Hellberg, F., Thomas, R. D., Zhaunerchyk, V., Geppert, W. D., Montaigne, H., Larsson, M., Kaminska, M., Osterdahl, F., 2005. Rate con- stants and branching ratios for the dissociative recombination of CO . The Journal of Chemical Physics 122 (22), 226101. Woodall, J., Agu´ndez, M., Markwick-Kemper, A. J., Millar, T. J., 2007. The UMIST database for astrochemistry 2006. Astronomy and Astrophysics 466 (3), 1197–1204. Yagi, M., Leblanc, F., Chaufray, J., Gonzalez-Galindo, F., Hess, S., Modolo, R., 2012. Mars exospheric thermal and non-thermal components: Seasonal and local varia- tions. Icarus 221 (2), 682–693. 1 3 Yee, J., Dalgarno, A., 1985. Energy transfer of O( S) atoms in collision with O( P) atoms. Planetary and Space Science 33 (7), 825–830. Yee, J., Guberman, S. L., Dalgarno, A., 1990. Collisional quenching of O( D) by O( P). Planetary and Space Science 38 (5), 647–652. 1 3 Yee, J. H., Dalgarno, A., 1987. Energy transfer of O( D) atoms in collision with O( P) atoms. Planetary and Space Science 35 (4), 399–404. Zahr, G. E., Preston, R. K., Miller, W. H., 1975. Theoretical treatment of quenching in O( D) + N collisions. The Journal of Chemical Physics 62 (3), 1127–1135. 30 Zhang, P., Kharchenko, V., Jamieson, M. J., Dalgarno, A., 2009. Energy relaxation in collisions of hydrogen and deuterium with oxygen atoms. Journal of Geophysical Research 114 (A7), A07101.

Journal

Astrophysics – arXiv (Cornell University)

Published: Nov 4, 2019

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Hot oxygen and carbon escape from the martian atmosphere

Hot oxygen and carbon escape from the martian atmosphere

Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Hot oxygen and carbon escape from the martian atmosphere

Hot oxygen and carbon escape from the martian atmosphere

References (67)

Abstract

Journal

Recommended Articles

There are no references for this article.

Our policy towards the use of cookies