H fractionation of kerogens during experimental electron irradiation: Comparison with chondritic organic matter

Icarus 226 (2013) 101–110

Contents lists available at SciVerse ScienceDirect

Icarus journal homepage: www.elsevier.com/locate/icarus

Amorphization and D/H fractionation of kerogens during experimental electron irradiation: Comparison with chondritic organic matter Corentin Le Guillou a,⇑, Laurent Remusat b, Sylvain Bernard b, Adrian J. Brearley a, Hugues Leroux c a

Department of Earth and Planetary Sciences, University of New Mexico, Albuquerque, NM 87131, USA Laboratoire de Minéralogie et Cosmochimie du Muséum, UMR CNRS 7202, MNHN, CP 52, 57 rue Cuvier, 75231 Paris Cedex 05, France c Unité Matériaux et Transformations, CNRS UMR 8207 – Université Lille 1, 59655 Villeneuve d’Ascq, France b

a r t i c l e

i n f o

Article history: Received 25 October 2012 Revised 24 April 2013 Accepted 3 May 2013 Available online 16 May 2013 Keywords: Organic chemistry Cosmic rays Solar radiation Cosmochemistry Experimental techniques

a b s t r a c t Irradiation is common in the interstellar medium and the protosolar nebula. We have investigated the effects of electron irradiation on kerogens of type I and III in a 200 kV transmission electron microscope (TEM), at 293 K and 92 K, using various fluences. Using synchrotron-based scanning transmission X-ray microscopy (STXM) and NanoSIMS, we have demonstrated a progressive amorphization coupled with hydrogen loss and a significant deuterium to hydrogen ratio (D/H) fractionation, with dD increasing by up to 1000‰. Hydrogen loss is non-linearly related to the fluence. Irradiation under cryogenic conditions (92 K) hinders amorphization and D/H elevation. We suggest that these effects are controlled by radiolysis (carbonAhydrogen bonds are broken and hydrogen is lost), coupled with recombination. The amorphization and hydrogen loss are rate-limited by defect diffusion which controls the recombination probability. The D/H increase appears to follow a Rayleigh distillation law with an apparent fractionation factor similar to the equilibrium fractionation factor of the isotopic exchange reaction CH4 + HD M CH3D + H2. This study represents a first step to estimate the kinetics and timescales of D/H fractionation under ionizing radiation. Extrapolatation of this fractionation behavior to natural environments remains difficult at this point because simultaneous irradiation by protons and other cosmic rays at various energies also occurs. However, the present results show that isotopic fractionation by electron irradiation at 200 kV alone might have been kinetically inhibited at the low temperatures of the interstellar medium and the outer region of the protosolar nebula. In addition, we show that STXM or NanoSIMS experiments should not be performed on organic samples that have already been investigated using TEM, even under low flux electron irradiation conditions. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction Carbonaceous chondrites contain complex refractory organic material (OM; Hayatsu and Anders, 1981). The reaction pathways that led to the formation of this extraterrestrial material are still debated. In chondrites, OM is finely distributed among other materials such as phyllosilicates, oxides and sulfides (Garvie and Buseck, 2007; Zega et al., 2010; Le Guillou et al., 2011). A component of this OM exhibits high D/H and 15N/14N ratios compared to the Sun or terrestrial materials (Robert and Epstein, 1982; Meibom et al., 2007; Charnley and Rodgers, 2008; Floss and Stadermann, 2009; Alexander et al., 2007; Aléon, 2010; Marty, 2012). Two environments have been proposed to explain where ⇑ Corresponding author. Present address: Institüt für Geologie, Mineralogie und Geophysik, Ruhr-Universität Bochum, Universitätsstr. 150, 44780 Bochum, Germany. E-mail address: [email protected] (C. Le Guillou). 0019-1035/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.icarus.2013.05.003

such deuterium enrichments could have been produced: the protosolar nebula (PSN) (Remusat et al., 2006, 2009, 2010; Willacy and Woods, 2009) and the interstellar medium (ISM), possibly during the dense molecular cloud phase (Robert and Epstein, 1982; Floss and Stadermann, 2009). In the ISM, ion–molecule or gas-grain reactions, in irradiated environments at very low temperature (<70 K), could have favored deuterium enrichment (Geiss and Reeves, 1981; Aikawa and Herbst, 1999; Sandford et al., 2001; Watanabe and Kouchi, 2008). In the outer region of the PSN, small molecules such as those present in comets could have also undergone cold chemistry in a similar way as in the ISM (Aikawa and Herbst, 1999). This OM is then accreted into asteroidal parent bodies, together with water ice and other more refractory materials, where it encounters aqueous and thermal processes (Huang et al., 2007; Garvie and Buseck, 2007; Remusat et al., 2008; Alexander et al., 2010; Cody et al., 2011; Le Guillou et al., 2012).

102

C. Le Guillou et al. / Icarus 226 (2013) 101–110

Irradiation studies of both experimental and natural materials are crucial to understand the molecular and isotopic evolution of the OM and shed new light on its original state. Irradiation of OM in extraterrestrial environments by proton or other ions has been previously experimentally investigated (Strazzulla and Baratta, 1992; Cooper et al., 2001; Moroz et al., 2004; Gomis and Strazzulla, 2005; Muñoz Caro et al., 2006; Brunetto et al., 2009; Godard et al., 2011). These studies demonstrated that ion irradiation induces the formation of refractory carbons from smaller molecules (by gradual dehydrogenation and polymerization) or the progressive amorphization of soots. However, none of these studies have discussed the evolution of isotopic signatures during irradiation. UV irradiation is also widely studied since UV photolysis of OM is assumed to be an efficient source of energy for cryogenic organic chemistry (e.g., Bernstein et al., 1995, 1999; Sandford et al., 2000; Dartois et al., 2005; Muñoz Caro et al., 2006; Mennella, 2008). In addition to proton and UV, electron fluxes are known to occur in both the ISM and the PSN (Meyer et al., 1998; Padovani et al., 2009; Godard et al., 2011). Electron irradiation has been mostly studied with applications to material science (Hobbs, 1987; Banhart, 1999; Egerton and Rauf, 1999; Egerton et al., 2004; Ammar et al., 2010). These studies have provided a conceptual basis for electron-organics physico-chemical interactions like knock-on damage, radiolysis, amorphization, and mass loss. Recently, De Gregorio et al. (2010) showed that electron irradiation of carbon-based polymers (epoxy and cyanoacrylate) in a TEM could produce significant dD increases (up to 1000‰). They thus speculated that some of the deuterium enrichment of chondritic OM particles in the ISM or the PSN could have been driven by similar irradiation processes. However, they did not thoroughly investigate the reaction mechanisms responsible for the fractionation and its kinetics. Understanding those mechanisms would be a first step to model isotopic fractionation induced by electrons in space. An improved understanding of electron irradiation mechanisms requires better constraints on the influences of the main experimental parameters on given organic precursors. Such constraints may eventually allow the fractionations caused by electron irradiation to be extrapolated with greater confidence to extraterrestrial environments. In this work, we irradiated Focused Ion Beam-prepared (FIB) sections of terrestrial type I and III kerogens in a TEM under various fluence rates at both room temperature and at 92 K, in order to constrain more thoroughly the irradiation-induced structural modifications and isotopic fractionation mechanisms. Experiments at 92 K are relevant to investigate the impact of low temperature encountered in the dense molecular cloud and the outer region of the PSN (where temperatures could drop below 50 K – Aikawa and Herbst (1999) and Sandford et al. (2001)). The molecular structure, the chemical and the isotopic evolution of the irradiated samples have been investigated using synchrotron-based STXM at the carbon K-edge and NanoSIMS.

2. Sample and experimental conditions 2.1. Kerogens: description and relevance Two reference kerogens have been selected for this study: a type I kerogen from the Green River shale formation (H/C = 1.6; O/C = 0.07; N/C = 0.025, dD = 150‰) and a type III kerogen from the Appalachians Eastern Coal Province (H/C = 0.7; O/C = 0.06; N/ C = 0.016, dD = 92‰). These reference materials have been provided by IFP Energies Nouvelles. The investigated type I kerogen mainly consists of aliphatic chains with a significant contribution of olefinic carbons (C@C), while the investigated type III kerogen, classified as a mature one, is more aromatic (Vandenbroucke,

2003; Vandenbroucke and Largeau, 2007). Although its molecular composition is different from that of chondritic insoluble organic matter from type 1 and 2 carbonaceous chondrites (Cody et al., 2008a), this type III kerogen has an elemental composition within the same range (Alexander et al., 2007). In contrast to chondritic organic matter (Busemann et al., 2006; Remusat et al., 2009; Okumura and Mimura, 2011), the investigated type III kerogen exhibits a homogeneous isotopic composition at the submicrometer scale. 2.2. Focused Ion Beam sample preparation Samples for TEM irradiation were prepared using the Focused Ion Beam (FIB) technique, with a FEI Quanta 3D FEGSEM/FIB instrument installed at the Department of Earth and Planetary Sciences, University of New Mexico. The FIB Ga ion milling was carried out at an ion beam voltage of 30 kV and beam currents from 3 nA down to 10 pA for the final thinning step. Ultrathin sections were lifted out in situ using an Omniprobe 200 micromanipulator and transferred to a Cu TEM half grid for final ion milling. The final FIB sections thickness was slightly larger than 100 nm. Detailed descriptions of the FIB extraction procedures can be found in Heaney et al. (2001) and Wirth (2004, 2009). FIB preparation may induce some modifications of the materials. A few nanometers of amorphized materials can form on each side of the section. Bernard et al. (2009) have shown that the FIB milling does not induce significant changes in the speciation of carbon in model polymers as measured by STXM-based C-XANES spectroscopy. The use of a dual beam FIB apparatus may also lead to the degradation of OM by SEM imaging, especially during the last steps of the preparation, but although significant for polymers, this effect remains very limited for kerogen-like materials (Bassim et al., 2012). To minimize this damage as much as possible, imaging of the sample was carried out at low voltage (5 kV) and low electron beam current (50 pA). 2.3. Irradiation conditions in the TEM The TEM irradiation was carried out using a JEOL 2010F FEGTEM/STEM in the Electron Microbeam Analysis Facility at the Department of Earth and Planetary Sciences and Institute of Meteoritics, University of New Mexico. The JEOL 2010F was operated at 200 kV. Three different FIB sections were prepared. Two FIB sections of the type I kerogen were irradiated, respectively, at 293 K and at cryogenic temperature (181 °C, 92 K) using a GATAN cryogenic sample holder cooled by liquid nitrogen. Temperature was monitored by a thermocouple located at the tip of the holder. A type III kerogen FIB section was also irradiated at room temperature. For each section, we used the low magnification mode and the smallest condenser aperture to locate the sample. This low electron exposure at a fluence rate of about 1011 e cm2 s1 resulted in a total fluence of 6  1012 e cm2. After location of the sample, 1 micron diameter spots were irradiated in bright field mode at beam currents from 1 to 600 pA (measured by the probe recording the illumination of the screen) for 30 min duration (Table 1), resulting in fluence rates from 6  1014 to 4  1017 e cm2 s1 and total fluences from 1.1  1018 to 6.7  1020 e cm2. The irradiated volume is a wide and short cylinder (1 lm in diameter  100 nm thick; Fig. 1). 2.4. Analytical methods 2.4.1. Scanning transmission electron microscopy (STXM) 2.4.1.1. Data collection. Following Bernard et al. (2012a, 2012b), STXM experiments were done on beamline 5.3.2.2 (STXM Polymer beamline – Kilcoyne et al., 2003) at the Advanced Light Source

103

C. Le Guillou et al. / Icarus 226 (2013) 101–110 Table 1 Summary of the different irradiation conditions and the resulting chemical and isotopic (D/H) compositions as obtained by NanoSIMS. Beam current (pA)

Fluence rate (e cm2)

H/C (at.)

H remaining fraction (f)

dD (%)

Kerogen I (293 K)

1 (bulk) 600

6E+12 6.7E+20

0.84 0.39

1 0.56 (±0.09)

38 (±33) 881 (±132)

Kerogen I (92 K)

1 (bulk) 1 11 20 100 600

6E+12 1.1E+18 1.2E+19 2.2E+19 1.1E+20 6.7E+20

0.75 0.81 0.79 0.74 0.74 0.65

1 1.08 1.05 0.98 0.99 0.86

(±0.1) (±0.1) (±0.09) (±0.09) (±0.09)

250 132 385 102 675 341

1 (bulk) 1 11 100 600

6E+12 1.1E+18 1.2E+19 1.1E+20 6.7E+20

0.56 0.44 0.45 0.31 0.31

1 0.79 0.81 0.57 0.57

(±0.06) (±0.06) (±0.05) (±0.05)

250 (±23) 809 (±229) 901 (±189) 1305 (±290) 1099 (±293)

Kerogen III (293 K)

Fig. 1. (a) Schematic representation of a 100 nm thick kerogen FIB section irradiated by a 1 lm diameter e beam in the TEM operating in bright field imaging mode; (b) SEM (secondary electron) image of the FIB section extracted from type III kerogen.

(ALS, Berkley, CA, USA). We followed the procedures for X-ray microscopy studies of radiation sensitive samples recommended by Wang et al. (2009), although radiation damage per unit of analytical information has been shown to be typically 100–1000 times lower in STXM-based XANES spectroscopy than in TEM-based EELS (Rightor et al., 1997; Braun et al., 2005, 2009; Hitchcock et al., 2008). As detailed by Bernard et al. (2012a, 2012b), the ALS 5.3.2.2 STXM beamline uses soft X-rays (250–700 eV) generated via a bending magnet while the electron current in the storage ring is held constant in top-off mode at 500 mA at a storage ring energy of 1.9 GeV. After pumping the chamber to vacuum, the microscope chamber was filled with He. Energy calibration was performed using the 3p Rydberg peak at 294.96 eV of gaseous CO2 for the C K-edge. X-ray absorption near edge structure (XANES) spectra were obtained by collecting image stacks with energy increments of 0.1 eV over the energy range of interest (270–340 eV at the C K-edge). Counting times are of the order of a few milliseconds per pixel. Alignment of stacks and extraction of XANES spectra were done using the aXis2000 software (ver2.1n – available on http://unicorn.mcmaster.ca/aXis2000.html). 2.4.1.2. Spectral deconvolution. Following the procedure described in detail by Bernard et al. (2010, 2012a), the collected C-XANES spectra have been first normalized to the total amount of carbon (absorption at 320 eV) to facilitate the comparison between the

(±43) (±161) (±176) (±158) (±186) (±167)

molecular signatures of irradiated areas. Spectral intensities were deconvolved (Fig. 2) using the Athena software package (Ravel and Newville, 2005) using an arctangent function to model the absorption edge itself and 27 Gaussian functions to model the absorption signal (Bernard et al., 2012a). The center position, the amplitude and the width of the arctangent function are fixed to 291.5, 1.0 and 0.4 eV, respectively. Although the natural lineshape of a XANES peak is a Lorentzian function, Gaussian functions have been used to take into account the contribution to the linewidth produced by instrumental broadening effects (Braun et al., 2006). Each Gaussian function has a fixed energy position and a constant half width at half maximum (0.4 eV below 295 eV and 2 eV above). Assuming that the oscillator strength of a given functional group is the same in organic materials of similar chemistry, variations in the intensity of the corresponding Gaussian function can be interpreted as relative variations in functional group concentration. However, it should be noted that some Gaussian functions used for the deconvolution of XANES signals do not account for the absorption of given functional groups. Instead, they may account for broad spectral features corresponding to highly delocalized excited states, sometimes referred to as 1s–r⁄ virtual state transitions, or for the overlapping contribution of Feshbach resonances (Stohr, 1991; Cody et al., 2008b). 2.4.2. NanoSIMS Imaging of D/H and H/C ratios was performed on the Cameca NanoSIMS 50 installed at the MNHN (Muséum Nationale d’Histoire Naturelle). This instrument combines high sensitivity, high mass resolution and high spatial resolution allowing quantitative isoto-

Fig. 2. Example of the deconvolution of a spectrum (kerogen type 3, ‘‘bulk spectra’’). The spectrum is first normalized to the absorption at 320 eV to account for the total carbon abundance, then an arctangent function (green) is used to model the absorption edge. The various Gaussian functions (fixed position and constant widths – 0.4 eV below 295 and 2 eV above) used to model the absorption at different energies are also displayed. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

104

C. Le Guillou et al. / Icarus 226 (2013) 101–110

pic images to be obtained at the submicron scale. TEM grids were stuck on Al disks using Cu tape, and then gold-coated (10 nm thick). We acquired 256  256 pixels images covering areas of around 400 lm2. The sample surface was rastered by a primary Cs+ beam set to a beam current of 7 pA. Secondary ion images of 1  2  H , D and 12C were simultaneously recorded using the combined analysis mode i.e. with magnetic peak switching between two different B-fields; this was done to facilitate measurements of masses 1 and 12, to investigate both the D/H and the C/H ratios during the same analysis at one location. Counting time was set to 10 ms pixel1 for 1H and 2D and 1 ms pixel1 for 12C for each frame. Between 20 and 40 frames were collected to obtain one image, in order to improve the counting statistics. The mass spectrometer was tuned to reach a mass resolving power of 2300 on mass 1 and 6000 on mass 12; this is enough to resolve any mass interference around the targeted ions. Prior to each analysis, the sample was presputtered with a 50 pA Cs+ current for about 5 min to remove surface contamination, the gold coating and to implant Cs+ into the surface of the sample. The instrumental stability for elemental and isotopic ratios was checked using the type III kerogen pressed on indium that was analyzed several times a day (n P 2). Images and data were processed using the LIMAGEÒ software (developed by L. Nittler, Carnegie Institution, Washington DC, USA) and corrected for detector dead time (44 ns, set electronically) and secondary ion images drift in X and Y positions (a few pixels drift may eventually occur between each frame). Uncertainties reported here correspond to 68% interval of confidence (1 sigma) and are based on the quadratic sum of counting statistics and external reproducibility achieved on the type III kerogen standard mounted in indium.

3. Results 3.1. Scanning transmission X-ray microscopy 3.1.1. Irradiation at room temperature The ‘‘bulk’’ absorption of each FIB section is defined as the averaged absorption of the areas which have not been irradiated by the 1 micron electron beam spots. The C-XANES spectrum (Fig. 3a) of the bulk type I kerogen (room temperature) shows a main absorption centered at 285 eV revealing the presence of aromatics and/or olefinic groups and a wide absorption around 287.7 eV and 288.6 eV corresponding to the absorption of aliphatic and carboxylic groups, respectively (Bernard et al., 2009, 2010). The C-XANES spectra of the irradiated areas of the type I kerogen also exhibit an absorption feature centered around 285 eV, but shifted by about 0.3–0.4 eV towards lower energies compared to the non-irradiated sample (Fig. 3a). With increasing fluence, this absorption feature becomes broader and covers a larger energy range. In parallel, a slight decrease of the aliphatic and carboxylic absorptions is observed. The evolution of the spectra of the type III kerogen is mostly similar to those of type I, although less significant at the lowest fluence (Fig. 3b).

3.1.2. Irradiation of type I kerogen at 92 K Cryogenic temperature seems to largely reduce the effects of electron irradiation: irradiation of type I kerogen at 92 K results in much less modification compared to irradiation at room temperature (Fig. 3c). Only a slight shift of the maximum of the absorption centered around 285 eV to lower energies (< 0.3 eV) and a slight decrease of the aliphatic and carboxylic absorption are observed. Remarkably, at 92 K, the increase of fluence does not significantly modify the molecular composition of type I kerogen as indicated by C-XANES spectroscopy.

3.1.3. Amorphization induced by electron irradiation A strictly pure amorphous carbon solid is a mixture of sp2 and sp3 hybridized bonds and does not have discrete molecular units such as aromatic rings, for instance. Accordingly, a typical amorphous carbon spectrum shows a featureless area over the range 284–287 eV rather than discrete peaks (e.g., Jiménez et al., 2003). In the present study, progressive amorphization of the kerogens is revealed by the shift towards lower values of the center of the aromatic absorption originally located at 285 eV and by its broadening, leading to a large absorption feature covering the 284– 287 eV range. To follow this progressive amorphization, we introduce here a simple parameter: the standard deviation of the mean of the areas of the 5 five Gaussian functions used to deconvolve the signal between 284 and 287 eV located, respectively, at 284.4, 285.1, 285.5, 286.1 and 286.7 eV (Fig. 3d). If formation and growth of aromatic units were ongoing, this standard deviation should increase, as it does during graphitization where a sharp absorption develops at 285.1 eV (Jiménez et al., 2003; Bernard et al., 2010). In contrast, the amorphization is revealed by the decrease (from 0.35 to 0.18) of this parameter with increasing fluence for the type I and III kerogen irradiated at room temperature. C-XANES spectroscopy demonstrates that type I and type III kerogens are evolving toward pure amorphous carbon solids displaying a broad range of variation of CAC bond lengths and distortions. When irradiation is performed at 92 K, this parameter remains roughly constant, reflecting the low impact of irradiation at cryogenic temperatures.

3.2. NanoSIMS We mapped H/C and D/H ratios of the FIB sections by NanoSIMS after STXM measurements (Table 1; Figs. 4 and 5). The general evolution is that the H/C ratio decreases with increasing fluence while the D/H ratio increases (Fig. 4). As the 12C images do not show depletion in the irradiated region, variations of H/C ratios are interpreted as resulting simply from hydrogen loss. The remaining fraction of hydrogen (‘‘final H content’’ over ‘‘initial H content’’) is used to better compare the type I and III kerogens which have different initial H/C values. To perform the TEM irradiations, we have to first locate the sample at low magnification before we can focus the beam to a 1 micron spot. This results in an irradiation at low fluence rate of the whole section for a short time. Consequently, we defined the reference H/C and dD as the value obtained from the whole FIB section which is called the ‘‘bulk value’’ below (irradiated spots are subtracted from the ROI). This reference bulk H/C value is lower than the powder values while the dD value is higher by 100‰ or more. This result indicates that the FIB sample preparation and TEM localization procedure have modified the material to some extent. However, the ‘‘bulk value’’ serves here as a reference point to evaluate further modifications that have occurred during irradiation under controlled conditions. At room temperature, the remaining fraction of hydrogen decreases and the D/H ratio increases with increasing fluence for both type I and III kerogens (Figs. 4 and 5). The Type I kerogen FIB section broke during analysis and only one irradiation spot could be measured but this data point shows the same evolution as the type III kerogen. The most intensively irradiated areas show a dD increase of about 1000‰, in association with about 50% of hydrogen loss. At 92 K, irradiation induced almost no chemical evolution of the type I kerogen (hydrogen loss lower than 10%). The corresponding increase of dD values is also much smaller than at room temperature (<400‰). These results are consistent with the STXM data describing limited amorphization at cryogenic conditions (see Fig. 6).

C. Le Guillou et al. / Icarus 226 (2013) 101–110

105

Fig. 3. STXM data showing the evolution of the carbon bonding environment of the kerogens with increasing fluence rate: (a) XANES spectra at the CAK edge of the type I kerogen before and after electron irradiation at 293 K; (b) XANES spectra at the CAK edge of the type III kerogen after electron irradiation at 293 K; (c) XANES spectra at the CAK edge of the type I kerogen after electron irradiation at 92 K; (d) standard deviation of the five deconvoluted band surfaces in the range 284–287 eV vs. fluence for the three experiments. The decrease by half of the standard deviation for irradiated areas reveals the evolution of the XANES spectra which tends to display a featureless area over the range 284–287 eV rather than discrete peaks, as expected during amorphization.

4. Discussion STXM and NanoSIMS results show that electron irradiation induces amorphization as illustrated by the molecular, chemical and isotopic evolution of both type I and III kerogens. Several processes can be invoked, including electron-nuclei (elastic) and electron–electron (inelastic) interactions (Hobbs, 1987; Banhart, 1999; Egerton et al., 2004). The respective impact of each of these processes depends on the irradiation conditions (energy of the electrons, total fluence, fluence rate, temperature, etc.) and on the specific response of a given material. In the following, we discuss the possible mechanisms responsible for the irradiation-induced modifications reported here before trying to extrapolate the respective influence of temperature, fluence and material properties to extraterrestrial conditions.

4.1. Hydrogen loss and isotopic fractionation 4.1.1. Elastic scattering ‘‘Knock-on’’ damage consists of elastic electron-matter interactions and occurs when the energy of the incoming electron (Eo, 200 keV here) is sufficient to transfer an amount of energy to the

nuclei of the targeted atom higher than its displacement energy (Ed). The latter depends on both the atom and the material (carbon in diamond: Ed = 80 eV; carbon in graphite: Ed = 30 eV, see Egerton et al. (2004) for a comprehensive review of the process). In a crystalline phase, knock-on damage results in atom displacements to interstitial sites and in the simultaneous creation of vacancies. The recombination of an atom within its original bonding environment can occur, but if the atom is displaced to a distance larger than the typical recombination radius, the Frenkel pair (interstitial/vacancy) is stabilized. Although not crystalline, kerogens should not be called amorphous since they contain distinct aromatic units that contrast in structure to amorphous carbon (Vandenbroucke, 2003). Knock-on damage is probably responsible for some of the observed amorphization of the kerogen by the destruction of the existing aromatic units through random redistribution of carbon atoms. Atomic displacements can be estimated using displacement cross sections (McKinley and Feshbach, 1948). The values of displacement energies (Ed) remain poorly constrained for organic material. They are usually within the range of 10–30 eV (e.g., Lee, 1999). In order to calculate a reasonable order of magnitude of atomic displacement cross sections, we used default Ed displacement energy values provided by the SRIM software for organic

106

C. Le Guillou et al. / Icarus 226 (2013) 101–110

for the observed hydrogen loss. It could, however, marginally help desorption of surface molecules. 4.1.2. Inelastic scattering 4.1.2.1. Sample heating. Electron–electron interactions can result in plasmon excitation and heat release. As formulated by Egerton et al. (2004), the temperature rise can be estimated by the following equation:

IEl ðt=kÞ ¼ 4pKtðT  T o Þ=ð0:58 þ 2 lnð2Ro =dÞÞ where d is the beam diameter; I the current; K the thermal conductivity and k the electron mean free path. Heat is deposited in a sample of thickness t and the associated energy can be written IEl(t/k). El is the energy loss per collision, and heat dissipates over a dimension Ro (that of the FIB section). The emissivity is negligible. Assuming a thermal conductivity similar to that of coal, e.g. 0.41 W m K1 (Herrin and Deming, 1996), the calculated temperature rise is limited to a few degrees at the highest beam current. The effective thermal conductivity could be overestimated because of the particular geometry of the FIB section, i.e. only two corners of the section are welded to the copper grid. Nevertheless, lowering the thermal conductivity by one order of magnitude does not change the result significantly and the temperature rise remains lower than 40 K at the highest beam current whereas hydrogen loss and fractionation are also observed at the lowest currents. Therefore, hydrogen loss resulting from thermal degradation of the kerogens can also be excluded as a significant mechanism.

Fig. 4. Remaining hydrogen fraction and dD plotted against fluence rate for the different irradiation experiments. Whereas H loss and dD increase are clearly observed at room temperature, the electron irradiation experiment carried out at liquid nitrogen temperature shows much less evolution of the kerogen with increasing fluence.

materials (polymers), i.e., 28 eV for C and 20 eV for H. These displacement cross sections are found to be around 1.3  1023 cm2 for H and 7.2  1024 cm2 for C. From these values, a displacement per atom (dpa) can be calculated by multiplying the atomic displacement cross section by the fluence. For the highest electron fluence, the dpa is around 102 for H and 0.5  102 for C. These values are too low to account for the observed hydrogen loss. If an atom located at the surface of a sample is knocked out, it results in sputtering and, ultimately, in mass loss. Using the same formulation as in Egerton et al. (2004), we calculated the sputtering rate for the present irradiation conditions. The sputtering rate S (monolayer per second) is given approximately by:

S ¼ ðJ=eÞðZ 2 =AE0 Þð1=Es  1=Emax Þð3:54  1017 Þ in which J/e represents the current density in electrons cm2 s1; Z the atomic number; A the mass number; Es the sputtering threshold energy and Emax the maximum energy transmitted to nuclei by electrons of energy E0. Given the conditions used in this study, the highest sputtering rate for carbon is of the order of 5  107 nm s1 (considering Ed = 28 eV, and converting monolayer per second to thickness per second assuming a density of about 1.6 for the kerogens) and of 107 for hydrogen (considering Ed = 20 eV). Such very low values prove that sputtering cannot be responsible

4.1.2.2. Radiolysis coupled with recombination and H2 diffusion. Radiolysis is often invoked to explain structural changes and hydrogen loss during electron irradiation in TEMs (Egerton et al., 2004). It corresponds to the interaction of the incoming electrons with the core and valence electrons of the atoms and results in local bonding instabilities and finally in bond breakage. For organic material, radiolysis rates are several orders of magnitude higher (up to five) than displacement rates from knock-on damage (Hobbs, 1987). Radiolysis induces defects (such as vacancies, interstitial atoms or radicals) after the bonds have been broken. The hydrogen loss budget results from a competition between secondary processes such as recombination (H conservation), H-bearing species and/ or defect diffusion and desorption (H loss). Every radical resulting from radiolysis is likely to react with its immediate environment if defects are within recombination distance. If recombination occurs within refractory molecules (aromatic units for instance), then hydrogen remains within the sample. In contrast, if hydrogen exists as a radical or a proton, or recombines as volatile molecules (H2 for instance), it can be lost by diffusion and desorption from the surface, a process which is thermally activated. Defects within carbon-bearing molecules also recombine, but they mostly remain within refractory molecules (carbon loss in the irradiated samples is not observed by NanoSIMS analysis). Radiolysis coupled with recombination of carbon atoms is likely responsible for amorphization, through random recombination between carbon atoms into hybridized sp3 and sp2 bonds. This process is further favored by the loss of hydrogen. Clark et al. (1980) for instance, observed progressive fading of the electron diffraction pattern of initially crystalline organic solids due to progressive amorphization during an electron irradiation experiment. Investigations of ion irradiation of hydrogenated aromatic carbons and soots have also led to the conclusion that the radiolysis/recombination process was relevant to model the rate of hydrogen loss, with the assumption that electronic interactions dominate over nuclear ones (Godard et al., 2011). Hobbs (1987) suggested that mass loss in organic material could be controlled by recombination coupled with defect diffusion. Indeed, non-thermal, uncorrelated recombination (recombi-

C. Le Guillou et al. / Icarus 226 (2013) 101–110

107

Fig. 5. NanoSIMS images of the H/C ratio (a) and D/H ratio (b) of the type III kerogen after irradiation at room temperature. Irradiated spots show a lower H/C ratio and a higher D/H ratio; (c) plot of the remaining H fraction vs. dD ratios of the kerogens. Theoretical curves for Rayleigh fractionation controlled either by diffusion or by equilibrium fractionation of H2 M CH4 at different temperatures (Bottinga, 1969) are presented for comparison with the obtained data.

Fig. 6. Processes controlling hydrogen loss and fractionation at different temperatures.

nation with a vacant site other than the original site) is likely to occur. At a given temperature, the radiolysis rate is expected to be linearly related to the fluence rate, while the recombination rate depends on the defect concentration, which increases faster with increasing fluence. Details of the formulation of this kinetic behavior can be found in Hobbs (1987). We infer that because of this effect, defect diffusion is the rate limiting mechanism and controls the recombination rate, which ultimately limits the hydrogen loss. The notion of critical dose (total fluence) is often used to describe the sensitivity of different materials to electron irradiation (Hobbs, 1987; Egerton, 1999). It is defined as the minimum total dose necessary to alter a crystalline lattice, independently of the fluence rate. The critical dose is expected to be higher for aromatic-rich materials than for aliphatic-rich materials, because of the higher bonding energy of aromatic C@C bonds than of aliphatic CAC bonds, which is related to a lower destruction rate of the former than for the latter. This kinetic effect could explain the early hydro-

gen loss by preferential destruction of aliphatic groups, which have a higher H/C ratio than aromatic groups. 4.1.3. Fractionation by Rayleigh distillation The multiple radiolysis/recombination reactions lead to the formation of volatile hydrogen-bearing species and a residual hydrogen-depleted carbon solid. The TEM column being continuously pumped, the system is open regarding volatile species: no re-equilibration between volatile species and the hydrogen-depleted carbon residue can occur. This configuration is typical of systems in which Rayleigh isotopic fractionation models are applied. Fig. 5c shows that the evolution of the deuterium enrichment as a function of H loss for both type I and type III kerogens follows a Rayleigh fractionation law which can be written: dD = 1000  ((dD0/1000 + 1)  f(1/a1)  1), where f is the remaining fraction of hydrogen and a the fractionation factor, defined as D=H1 =D=H2 .

108

C. Le Guillou et al. / Icarus 226 (2013) 101–110

Hydrogen seems to leave the kerogen faster than deuterium. As a result, kerogens progressively become isotopically heavier. Possible fractionation mechanisms, which determine the a value, are: (1) kinetic fractionation by diffusion-controlled loss of gaseous molecules (H diffuses faster than D); and (2) fractionation at equilibrium through reactions favoring the formation of H-bearing, rather than D-bearing, volatile molecules. Diffusion-controlled fractionation of gas depends on the square root of the ratio of the mass of the respective isotopologues (H2 vs. D2), and is theoretically independent of temperature. Such a mechanism does not fit our data (Fig. 5c, plain blue1 curve) and can thus be excluded. Fractionation at equilibrium appears more likely. To test this hypothesis, we plotted the theoretical Rayleigh fractionation distillation curve at two different temperatures (Fig. 5c), using the equilibrium fractionation factor of the simple CH4 + HD M CH3D + H2 isotopic exchange reaction: 1000  ln a = 25  106/T2 + 346  103/ T  223 (Bottinga, 1969). This isotopic exchange reaction is only one of the multiple possible pathways for isotopic fractionation, but it is the best available data available for isotopic exchange in this system. Kerogens being chemically complex, many other reactions with unknown equilibrium fractionation factors are ongoing simultaneously. The theoretical H2 M CH4 equilibrium fractionation curves reasonably fit the data reported here, which indicates that equilibrium may be reached through radiolysis/recombination reactions. Given the high fluence rates of our experiments, the reaction rates of bond breaking and recombination are very high and thus favor the achievement of isotopic equilibrium. This result could also indicate that hydrogen might recombine and diffuse out as H2. 4.1.4. Temperature dependence In the experiments performed at 92 K, the amorphization process, the hydrogen loss and the D/H increase are reduced. Defect formation by radiolysis does not directly depend on temperature but the behavior observed here strongly does. This observation provides evidence for the important influence of defect mobility and recombination on these processes. Since diffusion of H species is not a limiting mechanism (see isotopic fractionation section), we infer that: (i) at room temperature, defect diffusion is efficient and defects are desorbed at the surface of the TEM foil and lost into vacuum, thus lowering the probability of recombination of hydrogen within refractory molecules and favoring hydrogen loss; (ii) at 92 K, the mobility of defects generated by irradiation is much slower, thus enhancing the recombination probability since the mobile species are not desorbed into vacuum, and reducing hydrogen loss (Fig. 5). 4.2. Comparison to the ISM and PSN conditions 4.2.1. Comparison of experimentally irradiated samples to chondrite data In our experiment, D/H fractionation proceeds by Rayleigh distillation of the residue leading to hydrogen loss and amorphization. If organics from meteorites have encountered similar processes, the same effects should be observable. In carbonaceous chondrites, however, the trend is opposite. CR chondrites have the highest D/H, but also the highest H/C ratio of all groups, while CM and CI have lower H/C but lower D/H ratio as well (Alexander et al., 2007). At the sub-micrometer scale, combined in situ NanoSIMS/STXM/TEM observations have shown that D-rich particles of chondritic organic matter are not strictly amorphous, nor H depleted (Busemann 1 For interpretation of color in Fig. 5, the reader is referred to the web version of this article.

et al., 2006; Remusat et al., 2009; Le Guillou et al., 2011). In contrast, their XANES spectra resemble more pristine type III kerogen and display discrete peaks of C@C, C@O and ACOOH functional groups (Flynn et al., 2003; De Gregorio et al., 2010; Cody et al., 2011; Le Guillou et al., 2011). Organics from interplanetary dust particles (IDPs), collected in the Earth’s stratosphere have been compared to carbonaceous analogs experimentally irradiated with UV and protons (Muñoz Caro et al., 2006). Similarities between the Raman and infra-red signatures of IDPs and irradiated carbonaceous analogs have led to the suggestion that irradiation could have affected IDPs in the PSN, at least. However, XANES spectra of IDPs indicate that these particles are not amorphous (Flynn et al., 2003; Keller et al., 2004). Individual submicrometric amorphous organic particles (no O, S or N) have only been reported in one chondrite (ALH77307; Brearley, 2008) but their isotopic composition was not determined. It has been shown that benzylic radicals and, to some extent, the aliphatic CAH, were the carrier of the deuterium in chondritic organics (Remusat et al., 2006, 2009; Gourier et al., 2008). Those functional groups are likely the most sensitive to irradiation (compared to aromatic CAH; Egerton et al., 2004; Bassim et al., 2012), and would thus be the first ones to be lost during amorphization. If chondritic organics had experienced electron irradiation to a significant degree, small H-depleted ‘‘strictly amorphous’’ grains should be observed. Since these features are not widespread in chondrites, in contrast to D-rich and aliphatic-rich organics, Rayleigh isotopic fractionation driven by electron irradiation appears unlikely to reproduce the global D-enrichments of OM in carbonaceous chondrites. 4.2.2. Extrapolation of irradiation to the ISM and PSN conditions The study of electron irradiation with various fluences and at variable temperatures has allowed the identification of possible mechanisms for amorphization, H loss and D/H fractionation. It constitutes a first step to fully constrain the effects of irradiation in the ISM and the PSN that could be integrated into a more complex model including D/H modifications by other particles and energies when the relevant experimental data become available. The present study is indeed restricted to irradiation (i) of kerogens, (ii) by electrons, and (iii) at one single energy (200 kV). In contrast, in ISM and PSN environments, various other cosmic rays (among which electrons are not the most abundant) are present and heterogeneous carbonaceous compounds might be affected. The spectral range of electrons and other cosmic rays is broad, and their respective fluxes can be variable. For instance, electron irradiation at lower energy would reduce the relative rate of radiolysis damage formation over knock-on damage, and would impact the D/H evolution in a different manner. Ionizing particle fluences in the PSN and the ISM are poorly constrained, and can vary strongly depending on the environments and events that may occur locally (nebulae shock waves in the ISM or flaring events during the T-Tauri stage in the PSN, for instance). The flux of highly energetic particles can be estimated and the energy spectrum can only be extrapolated from higher energies (Yoshida, 2008; Padovani et al., 2009). Fluence rates of 101–105 e cm2 s1, i.e. 6–18 orders of magnitude lower than in our experiments are expected in the ISM. In the PSN, fluxes are even more difficult to constrain. Present measurements of solar winds indicate total electron fluxes of the order of 105 e cm2 s1 at 1 AU (Lin et al., 2008). Based on such estimates, De Gregorio et al. (2010) suggested that it would take from 107 to 1015 years in the ISM and 107–109 years in the PSN to attain the same total fluence as in TEM irradiation experiments. However, in the PSN, the total fluence might have been larger due to the higher activity of the young star (Feigelson et al., 2002), but only on short timescales.

C. Le Guillou et al. / Icarus 226 (2013) 101–110

The aforementioned timescale estimations are based on the extrapolation of the total electron fluence. However, several parameters could increase the required timescales, such as the temperature and the fluence rate. We have shown that at low temperature, D/H fractionation is smaller than at room temperature, mainly because of reduced hydrogen loss. As a consequence, at 90 K, a similar fractionation might only be attained over longer timescales. At even lower temperature, the extreme isotopic fractionation induced by the zero point energy (Geiss and Reeves, 1981) could, however, overcome the limitation due to low H loss. Although not specifically investigated here, we postulate that a lower fluence rate (for the same total fluence) may reduce the hydrogen loss. If, as suggested here, several simultaneous displacements are required to induce damages (the so-called cooperative damages reported in aromatic-bearing OM; Hobbs, 1987), a lower fluence rate (and the same total fluence) would produce a lower level of damage, e. g. less hydrogen loss, and thus a lower D/H fractionation as well. In addition, with a fluence rate tending to 0 for instance, i.e. with electrons colliding one at a time with the organic particle, hydrogen or deuterium would be lost without fractionation. Fractionation mechanisms deduced here involve equilibrium fractionation which requires ‘‘radiolysis/recombination’’ reactions reaching equilibrium, which is only possible if enough events are happening simultaneously. If fluence rate tends to 0, equilibrium cannot be achieved and D/H ratios would not be modified, or at least not in a similar manner. Altogether, it appears that timescales have to be longer than stated above to account for significant fractionation. Under the standard fluence conditions of the ISM (and also maybe for the PSN), the irradiation by 200 kV electrons alone is not likely responsible for the high deuterium content of organic matter assuming that they formed at low temperature. More experiments under various other conditions are required to be able to fully model irradiation-induced fractionation in the PSN. 5. Conclusions The combination of state-of-the-art analytical techniques is increasingly used to study organic materials in chondritic meteorites (Busemann et al., 2006; Zega et al., 2010; Le Guillou et al., 2011). Such nanoscale in situ isotopic and chemical (NanoSIMS), molecular (STXM) and mineralogical (TEM) characterizations is necessary to constrain the origin and evolution of organic grains. However, modifications to OM caused during TEM analysis are recognized issues in this characterization sequence. The present study demonstrates that even very minimal electron irradiation at low magnification (very low fluence and fluence rate) has an impact on OM molecular and isotopic signatures (dD increase of at least 100‰ between the powder value and the least irradiated FIB sections). When higher beam currents, long exposure times, or smaller beam sizes are used (high resolution EDX measurements, etc.), much more serious damage is expected. Therefore, STXM or NanoSIMS experiments on natural organic-rich samples should be performed prior to any TEM characterization. It is, however, worthwhile mentioning that imaging under cryogenic conditions significantly reduces the extent of the degradation. Here, the combined analyses by STXM and NanoSIMS of electronirradiated FIB foils demonstrate the amorphization of natural organic matter during irradiation in the TEM, coupled with a drastic decrease of H/C ratios of natural organic matter (as much as several tens of percents) and a huge increase of dD signatures (as high as several hundreds of permil). We suggest that D/H ratios increase by progressive hydrogen loss (rate-limited by defect diffusion) combined with a Rayleigh isotopic distillation of the lost H (likely at equilibrium). The D/H fractionation is strongly reduced at low temperature, suggesting that irradiation by 200 kV electrons alone in cold envi-

109

ronments (ISM and outer PSN), is probably not a very efficient process to explain the observed deuterium enrichments in chondritic OM. Acknowledgments Research funded by NASA Cosmochemistry Grant NNG06GG37G to AJB (PI). Electron beam analysis was carried out in the Electron Microbeam Analysis Facility, Department of Earth and Planetary Sciences, University of New Mexico. The facility is supported by funds from the State of New Mexico, the National Science Foundation and NASA. The National NanoSIMS facility at the Muséum National d’Histoire Naturelle was established by funds from the CNRS, Région Île de France, Ministère délégué à l’Enseignement supérieur et à la Recherche, and the Muséum itself. We are gratefull to the the PNP (programme nationale de planetologie, INSU CNRS, France). STXM data were acquired at beamline 5.3.2.2 at the ALS, which is supported by the Director of the Office of Science, Department of Energy, under Contract No. DE-AC02-05CH11231. References Aikawa, Y., Herbst, E., 1999. Molecular evolution in protoplanetary disks ultranationalistes distributions and column densities of gaseous molecules. Astron. Astrophys. 351, 233–246. Aléon, J., 2010. Multiple origins of nitrogen isotopic anomalies in meteorites and comets. Astrophys. J. 722, 1342–1351. Alexander, C.M.O’D., Fogel, M., Yabuta, H., Cody, G.D., 2007. The origin and evolution of chondrites recorded in the elemental and isotopic compositions of their macromolecular organic matter. Geochim. Cosmochim. Acta 71, 4380–4403. Alexander, C.M.O’D., Newsome, S.D., Fogel, M.L., Nittler, L.R., Busemann, H., Cody, G.D., 2010. Deuterium enrichments in chondritic macromolecular material— Implications for the origin and evolution of organics, water and asteroids. Geochim. Cosmochim. Acta 74, 4417–4437. Ammar, M.R., Rouzaud, J.N., Vaudey, C.E., Toulhoat, N., Moncoffre, N., 2010. Characterization of graphite implanted with chlorine ions using combined Raman microspectrometry and transmission electron microscopy on thin sections prepared by Focused Ion Beam. Carbon 48, 1244–1251. Banhart, F., 1999. Irradiation effects in carbon nanostructures. Rep. Prog. Phys. 62, 1181–1221. Bassim, N.D. et al., 2012. Minimizing damage during FIB sample preparation of soft materials. J. Microsc. 245, 288–301. Bernard, S., Benzerara, K., Beyssac, O., Brown, G.E., Stamm, L.G., Duringer, P., 2009. Ultrastructural and chemical study of modern and fossil sporoderms by scanning transmission X-ray microscopy (STXM). Rev. Palaeobot. Palynol. 156 (1–2), 248–261. Bernard, S., Beyssac, O., Benzerara, K., Findling, N., Brown Jr., G.E., 2010. XANES, Raman and XRD signatures of anthracene-based cokes and saccharose-based chars submitted to high temperature pyrolysis. Carbon 48, 2506–2516. Bernard, S., Horsfield, B., Schulz, H.-M., Wirth, R., Schreiber, A., Sherwood, N., 2012a. Geochemical evolution of organic-rich shales with increasing maturity: A STXM and TEM study of the Posidonia Shale (Lower Toarcian, northern Germany). Mar. Petrol. Geol. 31 (1), 70–89. Bernard, S., Wirth, R., Schreiber, A., Schulz, H.-M., Horsfield, B., 2012b. Formation of nanoporous pyrobitumen residues during maturation of the Barnett Shale (Fort Worth Basin). Int. J. Coal Geol. 103, 3–11. Bernstein, M.P., Sandford, S.A., Allamandola, L.J., Chang, S., Sharberg, M., 1995. Organic compounds produced by photolysis of realistic interstellar and cometary ice analogs containing methanol. Astrophys. J. 454, 327–344. Bernstein, M.P., Sandford, S.A., Allamandola, L.J., Gillette, J.S., Clemett, S.J., Zare, R.N., 1999. Ultraviolet irradiation of polycyclic aromatic hydrocarbons in ices: Production of alcohols, quinones, and ethers. Science 283, 1135–1138. Bottinga, Y., 1969. Calculated fractionation factors for carbon and hydrogen isotope exchange in the system calcite–CO2–graphite–methane–hydrogen and water vapor. Geochim. Cosmochim. Acta 33, 49–64. Braun, A. et al., 2005. Advantages of soft X-ray absorption over TEM-EELS for solid carbon studies: A comparative study on diesel soot with EELS and NEXAFS. Carbon 43 (1), 117–124. Braun, A., Huggins, F.E., Kelly, K.E., Mun, B.S., Ehrlich, S.N., Huffman, G.P., 2006. Impact of ferrocene on the structure of diesel exhaust soot as probed with wideangle X-ray scattering and C(1s) NEXAFS spectroscopy. Carbon 44 (14), 2904– 2911. Braun, A., Kubatova, A., Wirick, S., Mun, S., 2009. Radiation damage from EELS and NEXAFS in diesel soot and diesel soot extracts. J. Electron Spectrosc. Relat. Phenom. 170, 42–48. Brearley, A.J., 2008. Amorphous carbon rich grains in the matrices of the primitive carbonaceous chondrites, ALH77307 and Acfer 094. Lunar Planet. Sci. XXXIX. Abstract #1494.

110

C. Le Guillou et al. / Icarus 226 (2013) 101–110

Brunetto, R. et al., 2009. Comparison of the Raman spectra of ion irradiated soot and collected extraterrestrial carbon. Icarus 200, 323–337. Busemann, H., Young, A.F., Alexander, C.M.O’D., Hoppe, P., Mukhopadhyay, S., Nittler, L.R., 2006. Interstellar chemistry recorded in organic matter from primitive Meteorites. Science 312, 727–730. Charnley, S.B., Rodgers, S.D., 2008. Interstellar reservoirs of cometary matter. In: Origin and Early Evolution of Comet Nuclei Space Sciences Series of ISSI, vol. 28, pp. 59–73. Clark, W.R.K., Chapman, J.N., Macleod, A.M., Ferrier, R.P., 1980. Radiation damage mechanisms in copper phthalocyanine and its chlorinated derivatives. Ultramicroscopy 5, 195–208. Cody, G.D. et al., 2008a. Organic thermometry for chondritic parent bodies. Earth Planet. Sci. Lett. 272, 446–455. Cody, G.D. et al., 2008b. Quantitative organic and light-element analysis of Comet 81P/Wild 2 particles using C-, N-, and O-l-XANES. Meteorit. Planet. Sci. 43, 353–365. Cody, G.D. et al., 2011. Establishing a molecular relationship between chondritic and cometary organic solids. Proc. Natl. Acad. Sci. 108, 19171–19176. Cooper, J.F., Johnson, R.E., Mauk, B.H., Garrett, H.B., Gehrels, N., 2001. Energetic ion and electron irradiation of the icy Galilean satellites. Icarus 149, 133– 159. Dartois, E., Munoz-Caro, G.M., Deboffle, D., Montagnac, G., d’Hendecourt, L., 2005. Ultraviolet photoproduction of ISM dust. Laboratory characterisation and astrophysical relevance. Astron. Astrophys. 432, 895–908. De Gregorio, B.T., Stroud, R.M., Nittler, L.R., Alexander, C.M.O’D., Kilcoyne, A.L.D., Zega, T.J., 2010. Isotopic anomalies in organic nanoglobules from Comet 81P/ Wild 2: Comparison to Murchison nanoglobules and isotopic anomalies induced in terrestrial organics by electron irradiation. Geochim. Cosmochim. Acta 74, 4454–4470. Egerton, R.F., Rauf, I., 1999. Dose-rate dependence of electron-induced mass loss from organic specimens. Ultramicroscopy 80, 247–254. Egerton, R.F., Li, P., Malac, M., 2004. Radiation damage in the TEM and SEM. Micron 35, 399–409. Feigelson, E.D., Garmire, G.P., Pravdo, S.H., 2002. Magnetic flaring in the pre-main sequence Sun and implications for the early Solar System. Astrophys. J. 572, 335–349. Floss, C., Stadermann, F.J., 2009. High abundance of circumstellar and interstellar Canomalous phases in the primitive CR3 chondrites QUE 99177 and MET 00426. Astrophys. J. 697, 1242–1255. Flynn, G.J., Keller, L.P., Feser, M., Wirick, S., Jacobsen, C., 2003. The origin of organic matter in the Solar System: Evidence from the interplanetary dust particles. Geochim. Cosmochim. Acta 67 (24), 4791–4806. Garvie, L.A.J., Buseck, P.R., 2007. Prebiotic carbon in clays from Orgueil and Ivuna (CI), and Tagish Lake (C2 ungrouped) meteorites. Meteorit. Planet. Sci. 42, 2111–2117. Geiss, J., Reeves, H., 1981. Deuterium in the Solar System. Astron. Astrophys. 93, 189–199. Godard, M. et al., 2011. Ion irradiation of carbonaceous interstellar analogues. Effects of cosmic rays on the 3.4 lm interstellar absorption band. Astron. Astrophys 529, A146. Gomis, O., Strazzulla, G., 2005. CO2 production by ion irradiation of H2O ice on top of carbonaceous materials and its relevance to the Galilean satellites. Icarus 177, 570–576. Gourier, D. et al., 2008. Extreme deuterium enrichment of organic radicals in the Orgueil meteorite: Revisiting the interstellar interpretation? Geochim. Cosmochim. Acta 72 (7), 1914–1923. Huang, Y., Alexandre, M.R., Wang, Y., 2007. Structure and isotopic ratios of aliphatic side 1026 chains in the insoluble organic matter of the Murchison carbonaceous chondrites. Earth Planet. Sci. Lett. 259, 517–525. Hayatsu, R., Anders, E., 1981. Organic compounds in meteorite and their origin. Top. Curr. Chem. 99, 1–37. Heaney, P.J., Vicenzi, E.P., Giannuzzi, L.A., Livi, K.J.T., 2001. Focused Ion Beam milling: A method of site-specific sample extraction for microanalysis of Earth and planetary materials. Am. Miner. 86 (9), 1094–1099. Herrin, J.M., Deming, D., 1996. Thermal conductivity of US coals. J. Geophys. Res. – Solid Earth. 101, 25381–25386. Hitchcock, A.P., Dynes, J.J., Johansson, G., Wang, J., Botton, G., 2008. Comparison of NEXAFS microscopy and TEM-EELS for studies of soft matter. Micron 39 (3), 311–319. Hobbs, L.W., 1987. Radiation effects in analysis by TEM. In: Hren, J.J., Goldstein, J.I., Joy, D.C. (Eds.), Introduction to Analytical Electron Microscopy. Plenum Press, New York, pp. 399–445. Jiménez, I., Gago, R., Albella, J.M., 2003. Fine structure at the X-ray absorption p⁄ and r⁄ bands of amorphous carbon. Diam. Relat. Mater. 12, 110–115. Keller, L.P., Messenger, S., Flynn, G.J., Clemmet, S., Wirick, S., Jacobsen, C., 2004. The nature of molecular cloud material in interplanetary dust. Geochim. Cosmochim. Acta 68, 2577–2589. Kilcoyne, A.L.D. et al., 2003. Interferometer-controlled scanning transmission X-ray microscopes at the Advanced Light Source. J. Synchrotron Radiat. 10, 125–136. Le Guillou, C., Remusat, L., Bernard, S., Brearley, A.J., 2011. Redistribution and evolution of organics during aqueous alteration: NanoSIMS-STXM-TEM analyses of FIB sections from Renazzo, Murchison and Orgueil. Lunar Planet. Sci. XXXXII. Abstract #1996.

Le Guillou, C., Rouzaud, J.-N., Bonal, L., Quirico, E., Derenne, S., Remusat, L., 2012. High resolution TEM of chondritic carbonaceous matter: Metamorphic evolution and heterogeneity. Meteorit. Planet. Sci. 47, 345–362. Lee, E.H., 1999. Ion-beam modification of polymeric materials – Fundamental principles and applications. Nucl. Instrum. Methods Phys. Res. Sect. B: Beam Interact. Mater. Atoms 151, 29–41. Lin, R.P. et al., 2008. The STEREO IMPACT suprathermal electron (STE) instrument. Space Sci. Rev. 136, 241–255. Marty, B., 2012. The origins and concentrations of water, carbon, nitrogen and noble gases on Earth. Earth Planet. Sci. Lett. 313–314, 56–66. McKinley, W.A., Feshbach, H., 1948. The coulomb scattering of relativistic electrons by nuclei. Phys. Rev. 74, 1759–1763. Meibom, A. et al., 2007. Nitrogen and carbon isotopic composition of the Sun inferred from a high-temperature solar nebular condensate. Astrophys. J. 656 (1), L33–L36. Mennella, V., 2008. Activation of the 3.47 lm band by H atom irradiation of carbon grains covered with a water ice layer at 12 K. Astrophys. J. 682, L101. Meyer, J., Drury, L.O., Ellison, D.C., 1998. A cosmic-ray composition controlled by volatility and A/Q ratio. SNR shock acceleration of gas and dust. Space Sci. Rev. 86, 179–201. Moroz, L., Baratta, G., Strazzulla, G., Starukhina, L., Dotto, E., Barucci, M.-A., Arnold, G., Di Stefano, E., 2004. Optical alteration of complex organics induced by ion irradiation: 1. Laboratory experiments suggest unusual space weathering trend. Icarus 170, 214–228. Muñoz Caro, G.M. et al., 2006. Nature and evolution of the dominant carbonaceous matter in interplanetary dust particles: Effects of irradiation and identification with a type of amorphous carbon. Astron. Astrophys. 459, 147–159. Okumura, F., Mimura, K., 2011. Gradual and stepwise pyrolyses of insoluble organic matter from the Murchison meteorite revealing chemical structure and isotopic distribution. Geochim. Cosmochim. Acta 75, 7063–7080. Padovani, M., Galli, D., Glassgold, A.E., 2009. Cosmic-ray ionization of molecular clouds. Astron. Astrophys. 501, 619–631. Ravel, B., Newville, M., 2005. Athena, Artemis, Hephaestus: Data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat. 12 (4), 537–541. Remusat, L., Palhol, F., Robert, F., Derenne, S., France-Lanord, C., 2006. Enrichment of deuterium in insoluble organic matter from primitive meteorites: A Solar System origin? Earth Planet. Sci. Lett. 243, 15–25. Remusat, L., Le Guillou, C., Rouzaud, J.-N., Binet, L., Derenne, S., Robert, F., 2008. Molecular study of insoluble organic matter in Kainsaz CO3 carbonaceous chondrite: Comparison with CI and CM IOM. Meteorit. Planet. Sci. 43, 1099– 1111. Remusat, L. et al., 2009. Proto-planetary disk chemistry recorded by D-rich organic radicals in carbonaceous chondrites. Astrophys. J. 698, 2087–2092. Remusat, L., Guan, Y., Wang, Y., Eiler, J.M., 2010. Accretion and preservation of Drich organic particles in carbonaceous chondrites: Evidence for important transport in the early Solar System nebula. Astrophys. J. 713, 1048–1058. Rightor, E.G. et al., 1997. Spectromicroscopy of poly(ethylene terephthalate): comparison of spectra and radiation damage rates in X-ray absorption and electron energy loss. J. Phys. Chem. B 101 (11), 1950–1960. Robert, F., Epstein, S., 1982. The concentration and isotopic composition of hydrogen, carbon and nitrogen in carbonaceous meteorites. Geochim. Cosmochim. Acta 46, 81–95. Sandford, S.A., Bernstein, M.P., Allamandola, L.J., Gillette, J.S., Zare, R.N., 2000. Deuterium enrichment of polycyclic aromatic hydrocarbons by photochemically induced exchange with deuterium-rich cosmic ices. Astrophys. J. 538, 691–697. Sandford, S., Bernstein, M.P., Dworkin, J.P., 2001. Assessment of the interstellar processes leading to deuterium enrichment in meteoritic organics. Meteorit. Planet. Sci. 36, 1117–1133. Stohr, J., 1991. NEXAFS Spectroscopy. Springer, New York, 291pp.. Strazzulla, G., Baratta, G.A., 1992. Carbonaceous material by ion irradiation in space. Astron. Astrophys. 266, 434–438. Vandenbroucke, M., 2003. Kerogen: from types to models of chemical structure. Oil Gas Sci. Technol. – Rev. IFP 58 (2), 243–269. Vandenbroucke, M., Largeau, C., 2007. Kerogen origin, evolution and structure. Org. Geochem. 38 (5), 719–833. Wang, J., Morin, C., Li, L., Hitchcock, A., Scholl, A., Doran, A., 2009. Radiation damage in soft X-ray microscopy. J. Electron Spectrosc. Relat. Phenom. 170, 25–36. Watanabe, N., Kouchi, A., 2008. Ice surface reactions: A key to chemical evolution in space. Prog. Surf. Sci. 83, 439–489. Willacy, K., Woods, P.M., 2009. Deuterium chemistry in protoplanetary disks. II. The inner 30 a.u. Astrophys. J. 703, 470–499. Wirth, R., 2004. Focused Ion Beam (FIB): A novel technology for advanced application of micro- and nanoanalysis in geosciences and applied mineralogy. Eur. J. Mineral. 16 (6), 863–876. Wirth, R., 2009. Focused Ion Beam (FIB) combined with SEM and TEM: Advanced analytical tools for studies of chemical composition, microstructure and crystal structure in geomaterials on a nanometre scale. Chem. Geol. 261 (3–4), 217– 229. Yoshida, K., 2008. High-energy cosmic-ray electrons in the Galaxy. Adv. Space Res. 42, 477–485. Zega, T.J. et al., 2010. Mineral associations and character of isotopically anomalous organic material in the Tagish Lake carbonaceous chondrites. Geochim. Cosmochim. Acta 74, 5966–5983.