1. PARTICLE CIRCULATION MODEL IN THE MARTIAN/TITAN ENVIRONMENTS: ATMOSPHERIC SPUTTERING G. Rinaldi 1, A. Mura 1, V. Mangano 1, A. Milillo 1, S. Orsini 1 1 IFSI-INAF via del Fosso del Cavaliere 100 00133 Roma Italy Abstract: Atmospheric sputtering is a well-known process acting on planetary atmospheres in a similar way in which ion-sputtering acts on surfaces of airless bodies: solar energetic ions impact on the upper regions of planetary atmospheres and may cause significant escape of atmospheric particles. Mars and Titan do not possess an intrinsic magnetic field (Garnier et al. 2007); for this reason, atmospheric sputtering is expected to act more effectively on their atmospheres. To study this process we developed a Montecarlo single-particle model analogous to the one already done for Mercury (Mura et al. 2007) that simulates the cascade process. First results and comparisons are shown here. Numerical Model: In this work, to study the atmospheric sputtering, we apply a Montecarlo single- particle model. This model uses a four-dimensional grid (radius r; latitude φ; longitude λ and energy E) and an associated four-dimensional matrix Q. For a given source process, we define the surface S where the process occurs, then we launch a total number N tp of test-particles. For each test particle, we randomly choose a starting point P o, the starting velocity v 0, according to the velocity distribution function of the source, which is reproduced by using a von Neumann algorithm (1951). A weight w is associated to the test-particle, which takes into account the number of real particles that it represents. The trajectory of the test particle is computed using the classical equation of motion. The test-particle trajectory interacts with the exosphere- atmosphere system producing a significant escape of atmospheric particles. To describe the collisions between the plasma energetic ions and the exosphere-atmosphere system we use different collision-processes: 1) elastic collision; 2) ionization; 3) e- loss; 4) charge exchange. Cross sections : For a more realistic approach we describe the collision processes, cited above, with the esperimental cross section (listed in the Table 1). In this table we show the kind of processes that better explain the escape of atmospheric particles. For each non- thermal mecchanism we shown the cross section values for 3 energy levels (100 eV, 1KeV, 10 KeV) and the related references. In the Figure 2 it is shown the trend versus energy of the total cross section, respectively for neutral target (left) and ionized target (right) ( Noel and Prolss 1993 ). All collisions involving the same target has been grouped in the same plot. These values have been put in Table 1. FIGURE 1. Here we plotted the three different scenarios that may occur (from left to right) : 1) an energetic plasma ion crosses the exosphere and impinge directly on an atmospheric atom or molecule, and it is scattered (still ionized) back to the outer space, while a cascade process is generated in the atmosphere; 2) an energetic plasma ion impinge on a exospheric atom and it is neutralized via charge exchange, then it impacts with the atmospheric atom/molecule causing a cascade process; 3) an energetic plasma ion impinge directly on atmospheric atom/molecule which is scattered back and neutralized via charge exchange in the exosphere; the cascade process occurs in any case. FIGURE 2: from Noel and Prolss 1993 TABLE 1: Experimental cross sections for the case of: elastic collisions, ionization, e-loss and charge-exchange. Atmospheric sputtering: Sputtering process  energetic particles that impinge on a surface and let the atoms - located close to the surface- escape by overcoming their binding energy Atmospheric sputtering  same process but acting on atmospheric particles. In this case atoms -located in the upper regions of the atmosphere- must overcome the gravitational attraction of the planet Sputtering on a surface  Complete analogy  Sputtering on atmosphere First monolayer of surface  Same approach  Atmospheric exobase Hence, also in the upper atmosphere, particles may escape directly or after a series of bouncing, or they may loose velocity and form an atmospheric corona. In particular, a collision cascade below the exobase is expected, and the yield of the process may be very high, allowing a consistent flux outward from the atmosphere. Proje ctile Target Cross-section ProcessRefer. 100 eV (cm-2) 1 keV (cm-2) 10 keV (cm-2) HH5*10 -18 1.5*10 - 16 1*10 -17 e - loss Fit from Newman et al. 1986 HO2O2 1*10 -17 3*10 -16 4*10 -17 e - loss Fit from Newman et al. 1986 HH2H2 5*10 -18 1.5*10 - 16 1*10 -17 e - loss Fit from Newman et al. 1986 OO2O2 5*10 -18 5*10 -17 2*10 -16 e - loss Schafer (come N2) HO4*10 -17 1*10 -16 3*10 -16 e - loss Noel and Prolss 1993 HN2N2 1*10 -17 1.5*10 - 16 4*10 -16 e - loss Noel and Prolss 1993 OO3*10 -17 1*10 -16 3*10 -16 e - loss Noel and Prolss 1993 ON2N2 5*10 -18 5*10 -17 2*10 -16 e - loss Noel and Prolss 1993 N2N2 O2*10 -18 3*10 -17 4*10 -16 e - loss Noel and Prolss 1993 N2N2 N2N2 1*10 -18 2*10 -17 3*10 -16 e - loss Noel and Prolss 1993 HH5*10 -19 4*10 -17 1.3*10 - 17 ionization Fit from Newman et al. 1986 HO2O2 1*10 -18 8*10 -17 4.6*10 - 17 ionization Fit from Newman et al. 1986 HH2H2 5*10 -19 4*10 -17 1.3*10 - 17 ionization Fit from Newman et al. 1986 OO2O2 3*10 -18 5*10 -17 5*10 -16 ionization Schafer (come N2) HO1*10 -19 1*10 -17 1*10 -16 ionization Noel and Prolss 1993 HN2N2 1*10 -18 4*10 -17 5*10 -16 ionization Noel and Prolss 1993 OH+H+ 3*10 -18 5*10 -17 5*10 -16 ionization Noel and Prolss 1993 N2N2 O2*10 -17 9*10 -17 3*10 -16 ionization Noel and Prolss 1993 H+H+ H3*10 -15 2*10 -15 1.5*10 - 15 Charge- exchange Noel and Prolss 1993 H+H+ O1*10 -15 Charge- exchange Noel and Prolss 1993 H+H+ H2H2 3*10 -17 4*10 -16 1*10 -15 Charge- exchange Stebbings ? Barnett? H+H+ N2N2 2*10 -17 7*10 -16 1*10 -15 Charge- exchange Noel and Prolss 1993 O+O+ H1*10 -15 Charge- exchange Noel and Prolss 1993 O+O+ O1.5*10 -15 1*10 -15 Charge- exchange Noel and Prolss 1993 O+O+ N2N2 8*10 -16 1*10 -15 2*10 -15 Charge- exchange Noel and Prolss 1993 N2+N2+ H3*10 -15 2*10 -15 Charge- exchange Noel and Prolss 1993 N2+N2+ O3*10 -16 5*10 -16 8*10 -16 Charge- exchange Noel and Prolss 1993 N2+N2+ N2N2 3*10 -15 2*10 -15 Charge- exchange Noel and Prolss 1993 Proje ctile Target Cross-section ProcessRefer. 100 eV (cm-2) 1 keV (cm-2) 10 keV (cm-2) HO2O2 1*10 -15 4*10 -16 4*10 -18 elastic Fit from Newman et al. 1986 OO2O2 2*10 -15 9*10 -16 4*10 -16 elastic Schafer et al. 1987 HH2H2 5*10 -16 2*10 -16 1*10 -18 elastic Fit from Newman et al. 1986 HH5*10 -16 2*10 -16 1*10 -18 elastic Fit from Newman et al. 1986 N+N+ N2N2 1*10 -14 2*10 -15 5*10 -16 elastic Lammer et Bauer 1993 O O1*10 -15 5*10 -16 3*10 -16 elastic Noel and Prolss 1993 ON2N2 2*10 -15 9*10 -16 4*10 -16 elastic Noel and Prolss 1993 HO1*10 -15 2*10 -16 4*10 -17 elastic Noel and Prolss 1993 HN2N2 1*10 -15 2*10 -16 4*10 -17 elastic Noel and Prolss 1993 O+O+ H9*10 -16 1.5*10 -16 4*10 -17 elastic Noel and Prolss 1993 O+O+ O1*10 -15 5*10 -16 2.5*10 - 15 elastic Noel and Prolss 1993 O+O+ N2N2 2*10 -15 8*10 -16 3.5*10 - 16 elastic Noel and Prolss 1993 H+H+ H4*10 -16 6*10 -17 1*10 -17 elastic Noel and Prolss 1993 H+H+ O1*10 -15 2*10 -16 4*10 -17 elastic Noel and Prolss 1993 H+H+ N2N2 1*10 -15 2*10 -16 5*10 -17 elastic Noel and Prolss 1993 N2N2 O2*10 -15 8*10 -16 3.5*10 - 16 elastic Noel and Prolss 1993 N2N2 N2N2 2*10 -15 1*10 -15 6*10 -16 elastic Noel and Prolss 1993 N2+N2+ H5*10 -16 1.5*10 -16 3*10 -17 elastic Noel and Prolss 1993 N2+N2+ O2*10 -15 8*10 -16 3*10 -16 elastic Noel and Prolss 1993 N2+N2+ N2N2 2*10 -15 1.5*10 -15 7*10 -16 elastic Noel and Prolss 1993 HO1*10 -19 1*10 -17 1*10 -16 elastic Noel and Prolss 1993 HN2N2 1*10 -18 4*10 -17 5*10 -16 elastic Noel and Prolss 1993 ON2N2 3*10 -18 5*10 -17 5*10 -16 elastic Noel and Prolss 1993 N2N2 O2*10 -17 9*10 -17 3*10 -16 elastic Noel and Prolss 1993 H+H+ H3*10 -15 2*10 -15 1.5*10 - 15 Charge- exchange Noel and Prolss 1993 H+H+ O1*10 -15 Charge- exchange Noel and Prolss 1993 H+H+ H2H2 3*10 -17 4*10 -16 1*10 -15 Charge- exchange Stebbings ? Barnett? H+H+ N2N2 2*10 -17 7*10 -16 1*10 -15 Charge- exchange Noel and Prolss 1993 O+O+ H1*10 -15 Charge- exchange Noel and Prolss 1993 O+O+ O1.5*10 -15 1*10 -15 Charge- exchange Noel and Prolss 1993 O+O+ N2N2 8*10 -16 1*10 -15 2*10 -15 Charge- exchange Noel and Prolss 1993 N2+N2+ H3*10 -15 2*10 -15 Charge- exchange Noel and Prolss 1993 N2+N2+ O3*10 -16 5*10 -16 8*10 -16 Charge- exchange Noel and Prolss 1993 N2+N2+ N2N2 3*10 -15 2*10 -15 Charge- exchange Noel and Prolss 1993 Project ile Targe t Cross-section ProcessRefer 100 eV (cm-2) 1 keV (cm-2) 10 keV (cm-2)

2. Results: Proton Nitrogen FIGURE 4: On the left H+, on the right N+. The upper two panels show the latitude and longitude map of the escaping N 2 flux at a surface five Titan Radii (R T ) far from the planet. The middle two panels are a bi-dimensional distribution of the N2 density (log10 scale) in the first two R T as they are produced by the two different projectiles. The N+ effects appear to be much more remarkable (by two orders of magnitude). Finally, in the lower two panels the two energy distributions of ion-sputtered proton and nitrogen atoms for a impact energy Ei of 1keV are plotted. The red curve is the energy distribution for the particles inside a exospheric volume with a size of 5 R T and the blue curve is the energy distribution for the particles escaping from the same region. Titan: Titan is the largest Saturn’s satellite and it has a dense atmosfere, which interact either with corotating plasma of Saturn (when Titan is inside the magnetosphere), or with the solar wind (when it is outside the magnetosphere). Model Plasma parameters: we used two different profjectiles representative of these two main situations during Titan’s orbit (outside and inside Saturn’s magnetosphere, respectively). 1.Protons: the protons of solar wind parameters has been taken from Ledvina et al. 2003 1.Nitrogen: the nitrogen ions N + of the Saturn’s magnetosphere has been taken from Lammer et Bauer 1993 Atmospfere/exosphere system: The atmosphere model has been extracted from the thermal profile measured by Huygens spacecraft, and the exospheric one is derived from Garnier et al 2007 (Figure 3). For this model we consider only the main atmospheric constituent N 2. Collisional process: in this simulation only the elastic processes and related cross sections (table 1) have been used. Atmospheric Sputtering Computed Yields : The yield of the atmospheric sputtering for the different species involved in the Titan and Mars simulations were derived as the ratio between the simulated escaping N2 flux and incoming projectiles (H+ or N+) for Titan, and by using the theoretical formulas in Lammer et Bauer (1993) for the case of martian CO2. In particulare the computed values for Titan agree with the values given by Lammer et Bauer (1993). proton nitrogen Results FIGURE 5: The upper figure shows the latitude and longitude map of the escaping CO 2 + H flux at a surface five R M (Mars Radii) far from the planet. The middle figure is the bi-dimensional distribution of the CO 2 density (log10 scale) in the first two R M out of the planet, as it is produced by elastic process. The lower figure is the bi-dimensional distribution of the H+ density (log10 scale) in the first two R M out of the planet, as it is produced by charge exchange. Atmosperic speciesYield Titan N 2 10 Titan N 2 50 martian CO 2 0.15 Mars Mars is the fourth planet in the Solar Sistem, it is a rocky planet with a tenuous atmosphere which interacts with the solar wind plasma. Plasma Model: The empirical model of the proton flow is based on ASPERA three-dimensional proton velocity measurements, and is ascribed to Kallio [1996]. Atmospfere/exosphere system: Our simplified model considers Mars’ atmosphere and exosphere as composed only of CO 2. The neutral profile for CO 2 population is approximated by an exponential function [Krasnopolsky and Gladstone,1996]. Collisional process : in this simulation only the elastic and charge exchange processes have been used. We use a geometrical cross section with a CO 2 radius of 0.33 nm (Araki et al. 2007) REFERENCES: Sadao Arakia, Norito Mohrib, Yuichi Yoshimitsub and Yoshikazu Miyakeb; Synthesis, characterization and gas permeation properties of a silica membrane prepared by high-pressure chemical vapor deposition Garnier, P, Dandouras,I, D. Toublanc, et al; The exosphere of Titan and its interaction with the Kronian magnetosphere: MIMI observations and modeling;Planetary and Space Science 55 (2007) 165-173 H. Lammer and S. J. Bauer; Atmospheric mass loss from Titan by sputtering Planet. Space Sci., Vol.41, No. 9, pp. 6577663, 1993 S.A. Ledvina, J.G. Luhmann, S.H. Brecht, T.E. Cravens Titan’s induced magnetosphere Advances In Space Research 33 (2004) 2092–2102 Mura,A, Milillo, A., Orsini, S., Massetti.S, Numerical and analytical model of Mercury’s exosphere: Dependence on surface and external conditions, Planetary and Space Science 55 (2007) 1569–1583 Newman.J.H., Y.,S. Cher, K. A. Smith, and R. F. Stebbing, Differential cross-section for scattering of 0.5, 1.5, and 5.0 keV hydrogen atoms by He, H 2, N 2 and O 2, J. Geophys. Res,91,8947,1986 Noel,S. and G.W. Prolss, Heating and radiation production by neutralised ring current particles, J. Geophys. Res., 98,17,317,1993 Kallio, E., An empirical model of the solar wind flow around Mars, J. Geophys. Res., 101, 11,133, 1996. Krasnopolsky, V. A., and G. R. Gladstone, Helium on Mars: EUVE and PHOBOS data and implications for Mars’ evolution, J. Geophys. Res.,101, 11,765, 1996. Stebbings, R. F., C. H. Smith, and H. Ehrahardt, Charge transfer between oxygen atoms and O+ and H+ ions, J. Geophys. Res., 69, 2349,1964. CONCLUSIONS : This preliminary work was intended to study the behaviour of atmospheric particles via a process called ATMOSPHERIC SPUTTERING. We applied our model to simulate this process in two very different environments of the Solar System. The calculated yields are in agreement with some previous calculations for Titan (Lammer and Bauer, 1993). The results show a remarkable contribuiton by atmospheric sputtering in the planetary atmospheric loss, hence underlying the importance of this process in the study of atmospheric evolution.