Evaluation of isotopic fractionation of oxygen ions escaping from terrestrial thermosphere

Abstract

Oxygen isotopic ratios in contemporary lunar soils indicate a highly mass-independent fractionation (MIF) of $\mathrm{\Delta }{}^{17}\mathrm{O}\approx 25$ per mil. This fraction is comparable to the ratio for stratospheric ozone and is thus supposed to have originated from the terrestrial upper atmosphere through transport of oxygen ions by the intense Earth wind. In this paper, we develop a one-dimensional chemical model that can solve for the ionic and neutral compositions of oxygen isotopes by including the isotopic effects in dynamical, photolytic, and chemical processes. We first confirm that the isotopic ratios of ${\mathrm{O}}^{+}$, ${\mathrm{O}}_2^{+}$, O, and ${\mathrm{O}}_2$ are nearly mass dependent, $\mathrm{\Delta }{}^{17}\mathrm{O}\approx 0$ per mil, when the isotopic effects of multiple (eddy, molecular, and ambipolar) diffusion processes are taken into account. We examine which photolytic and chemical processes can produce a strong MIF for oxygen species and select four candidates: photo-dissociative ionization of ${\mathrm{O}}_2$, photo-dissociation of ${\mathrm{O}}_2$ by Lyman-$\alpha$ solar emissions, charge exchange between ${\mathrm{O}}^{+}$ and ${\mathrm{O}}_2^{+}$, and atomic exchanges between O and ${\mathrm{O}}_2$. We estimated the oxygen isotopic ratio to be $\mathrm{\Delta }{}^{17}\mathrm{O}\approx 1''–''4$ per mil at the height of 100–400 km on the basis of a statistical treatment of atom mechanics. These values are smaller than those in the lunar case, and we conclude that a more rigorous quantum mechanical treatment of photo-dissociation processes will be necessary before we can make an evaluation of the role of Earth-escaping oxygen isotopes.

Introduction

Oxygen is the third most abundant element in the solar system and it is one of the most crucial aspects to understanding evolution of the system. However, its mean elemental and isotopic compositions are still not well understood (cf. Clayton, 2002a, Clayton, 2002b, Yurimoto and Kuramoto, 2004, Lyons and Young, 2005, Hashizume and Chaussidon, 2005, Ozima et al., 2007). Recent measurements of contemporary lunar soils have indicated that the oxygen isotopic ratio in the metallic particles has a highly mass-independent fractionation (MIF) of $\mathrm{\Delta }{}^{17}\mathrm{O}\approx 25$ per mil (Ireland et al., 2006). There is also a negative anomaly in the same metal grains as well, thus the value remains controversial. Here, $\mathrm{\Delta }{}^{17}\mathrm{O}=\mathrm{\delta }{}^{17}\mathrm{O}-0.52\mathrm{\delta }{}^{18}\mathrm{O}$ stands for the strength of MIF where $\mathrm{\delta }{}^{\text{n}}\mathrm{O}=\{([{}^{\text{n}}\mathrm{O}]/[{}^{16}\mathrm{O}])/([{}^{\text{n}}\mathrm{O}]/[{}^{16}\mathrm{O}]{)}_{\mathrm{SMOW}}-1\}\times {10}^3$ in per mil with $\text{n}=17$ or 18; the density ratio denoted by “SMOW” means the Standard Mean Ocean Water content. Ireland et al. (2006) emphasized that this value must be indigenous to the solar wind-implanted component and accordingly it represents the mean isotopic ratio of the solar system. Two predictions have thus far been made for the mean value of $\mathrm{\Delta }{}^{17}\mathrm{O}$; one predicts a negative MIF of −20 per mil (Clayton, 2002a; Hashizume and Chaussidon, 2005; Lyons and Young, 2005), and another predicts $\mathrm{\Delta }{}^{17}\mathrm{O}=0$ per mil (Ozima et al., 2007). The measured lunar oxygen isotopic ratio can be taken as a third prediction. This exotic oxygen has roughly the same isotopic ratio as is commonly found in the terrestrial atmosphere in stratospheric nitric oxides and ozone (Thiemens, 2006; and references therein). These facts motivated us to study if the exotic oxygen could be attributable to the Earth-escaping wind, as suggested by our previous studies (Ozima et al., 2007, Ozima et al., 2008; Hiraki et al., 2008b). This would be the first attempt to assess the direct interaction between the terrestrial atmosphere and the lunar surface, which in turn would yield a new means to trace the evolution of oxygen in the terrestrial atmosphere.

Such speculation is supported by recent observations of the Earth-escaping ion flow by the GEOTAIL mission (Seki et al., 2001). Since the ambient magnetic field line opens in the high latitude (cusp) region, a measurable amount of energetic oxygen ions ${\mathrm{O}}^{+}$ can escape Earth’s gravitation and reach far beyond lunar orbit. Seki et al. (2001) estimated the escaping ${\mathrm{O}}^{+}$ flux at the lunar orbit, and the flux of O that hits the surface is $\approx 3.7\times {10}^3$ ${\mathrm{cm}}^{-2}{\mathrm{s}}^{-1}$. Their numerical simulations indicate that the escaping ${\mathrm{O}}^{+}$ flux is more or less consistent with the amount of the exotic oxygen implanted in metallic particles. If the MIF of exotic oxygen originates from the present-day Earth, this would give robust support for the Earth-wind hypothesis. Another problem is whether the implantation process on the Moon can change the oxygen isotopic ratio, but this is beyond our present scope.

The lunar surface is exposed to the solar wind as well as the Earth wind. There are a few estimates of the solar-wind oxygen flux in interplanetary space (e.g. von Steiger et al., 2000; Hashizume and Chaussidon, 2005). Taking the average abundance ratio of He/O and assuming the shielding factor of the geomagnetic tail, the O flux at the lunar surface was estimated to be $1.5\times {10}^4$ ${\mathrm{cm}}^{-2}{\mathrm{s}}^{-1}$, which is a factor of four larger than the estimated Earth-wind O flux. If the solar-wind O has a different value of $\mathrm{\Delta }{}^{17}\mathrm{O}$ from the Earth-wind O, the oxygen isotopic signature of the Earth wind would be diluted by a factor of five; note that the dilution rate depends on the Earth wind conditions and the solar activity. In order for our hypothesis to be tenable, the $\mathrm{\Delta }{}^{17}\mathrm{O}$ values in the Earth atmosphere should be at least comparable to those observed for exotic oxygen.

Numerical studies were made to estimate the isotopic fractionation rates of oxygen for minor species in the middle atmosphere (e.g. Lyons, 2001; Zahn et al., 2006; Liang et al., 2007). These estimates showed maximum values of $\mathrm{\Delta }{}^{17}\mathrm{O}=$ 30–80 per mil for N${\mathrm{O}}_x$, H${\mathrm{O}}_2$, H₂O, OH, and ${\mathrm{O}}_3$ in the stratosphere (20–30 km). It was also found that the MIF ratio of oxygen inversely correlates with concentrations of nitrous oxide and methane since these molecules are sinks for the reactions related to the above minor species. Direct observations of ${\mathrm{CO}}_2$ have been made using balloons, rockets, and aircraft (e.g. Boering et al., 2004; Thiemens, 2006; and references therein). The estimated $\mathrm{\Delta }{}^{17}\mathrm{O}$ is $\approx 10$ per mil for ${\mathrm{CO}}_2$ and 40–60 per mil for ${\mathrm{O}}_3$ in the stratosphere. Laboratory analyses showed that the oxygen isotopic ratio of the sampled ${\mathrm{CO}}_2$ linearly increases with altitude up to $\mathrm{\Delta }{}^{17}\mathrm{O}\approx 12$ per mil at 60 km, which suggests even higher values at higher altitudes (Thiemens et al., 1995). However, because of the low abundance of ${\mathrm{CO}}_2$ in the upper atmosphere, this ratio may not be directly relevant to the isotopic signature of the Earth-wind oxygen. Thiemens’ group intends to gather information on oxygen in the middle atmosphere (90–110 km) with their latest rocket measurements.

In spite of the vigorous measurements on light oxygen $(\mathrm{\Delta }{}^{17}\mathrm{O}>0)$, the current information on the isotopic composition that is available from measurements and theoretical calculations is limited to less than 100 km and is not useful for assessing the isotopic characteristics of the Earth-wind O. The purpose of our study is to determine if the origin of the large MIF of lunar oxygen is the terrestrial upper atmosphere. We focused on the high-latitude thermosphere (100–300 km) where oxygen ions start to escape. We developed an ion-neutral chemical model that includes oxygen isotopes for atmospheric minor species such as ${\mathrm{O}}^{+}$. We calculated the amount and the sign of the isotopic ratios $\mathrm{\Delta }{}^{17}\mathrm{O}$ and clarified the source mechanisms (see Section 2.1). The MIF for neutral major species of ${\mathrm{O}}_2$ can largely affect ionic species; as a result, composition changes in major neutral species are self-consistently treated in this model. In Section 2.2, we present a test calculation including the isotopic effect of multiple (eddy, molecular, and ambipolar) diffusion coefficients. In Section 3.1, we assess the key photolytic and chemical processes in the altitude range of 100–300 km and determine the candidates that can produce a large MIF. In Section 3.2, we perform several test simulations including the isotopic effects of the photolytic and chemical reaction coefficients. We present estimates for isotopes in the upper atmosphere based on a statistical treatment of atomic reaction mechanics.

Before showing the details of our chemical model, we briefly mention recent studies on the generation of large MIFs due to chemical reactions or photolyses. Gao and Marcus (2001) proposed a semi-empirical theory based on measured isotope-specific rate coefficients for the formation of ${\mathrm{O}}_3$ that explains the observed large MIF ratios for stratospheric ozone. They pointed out that the reaction speed difference due to the asymmetry of ${\mathrm{O}}_3$ isotopes leads to a large MIF. Lyons and Young (2005) suggested a self-shielding effect in photo-dissociation of CO molecules in the solar nebula. Navon and Wasserburg (1985) examined the self-shielding effect of photo-dissociation of ${\mathrm{O}}_2$ in oxygen-rich low pressure gas and found that it cannot occur in the nebula because of the rapidity of isotope exchange. Similar processes would be expected in the upper atmosphere, owing to the chemical reactions associated with ⩾3 oxygens or photolysis in the ultraviolet wavelength range.