Free Access
Issue
A&A
Volume 561, January 2014
Article Number A73
Number of page(s) 9
Section Interstellar and circumstellar matter
DOI https://doi.org/10.1051/0004-6361/201322010
Published online 03 January 2014

© ESO, 2014

1. Introduction

The key to interstellar chemistry lies in the formation of molecules on dust grain surfaces. These dust grains are thought to be mainly silicate or carbonaceous in nature. In dense regions, such as in the interior of dense molecular clouds, dust grains provide shielding from the external galactic UV radiation field, allowing the formation of icy mantles surrounding the dust grains. The interior of molecular clouds is penetrated by cosmic rays, which are ubiquitous throughout the interstellar medium. Cosmic rays may be protons, alpha particles, nuclei of heavy atoms, electrons, and positrons (Swordy 2001). The interaction of interstellar material and cosmic rays in turn produce a number of secondary particles, such as low-energy cosmic rays, electrons, and UV photons. Thus, interstellar ices are processed simultaneously by mostly internal but also by attenuated external UV radiation as well as by fast-moving ions from the galactic cosmic radiation field. Both processes induce chemical and structural changes within the ice mantles of interstellar dust grains (e.g., Grim et al. 1989; Jenniskens et al. 1993; Palumbo et al. 1998; Hudson et al. 2008; Palumbo et al. 2010).

The distinct processes of ion bombardment and UV irradiation on solids of astrophysical interest are well characterized (e.g., Hagen et al. 1979; Brown et al. 1982; Moore et al. 1983; Strazzulla et al. 1983a; Allamandola et al. 1988; Baratta et al. 1991; Westley et al. 1995a; Gerakines et al. 1996; Strazzulla et al. 2001; Moore et al. 2001; Leto & Baratta 2003; Öberg et al. 2009; Modica & Palumbo 2010; Garozzo et al. 2011). Some investigations on the comparison between the two processes have also been carried out. Experimental results have shown that from a qualitative point of view, ion bombardment and UV photolysis generate similar changes in interstellar ice analogs. However, quantitative differences between the two processes have been observed. Cottin et al. (2001) examined organic residues produced from H2O:CO:CH3OH:NH3 ices after UV photolysis or proton bombardment to find that hexamethylenetetramine was present in the residue after both processes. Gerakines et al. (2000) investigated the formation of carbonic acid (H2CO3) from H2O:CO2 ice mixtures and found that H2CO3 is readily formed by proton bombardment. However, in the case of UV photolysis, the penetration of UV photons into the ice limits the production of H2CO3. Gerakines & Moore (2001) investigated the formation of carbon suboxide (C3O2) in ices analogous to cometary nuclei and interstellar regions to give formation and destruction rates for C3O2, CO and CO2 for both proton bombardment and UV irradiation. They found that Lyman-alpha photons are 3.5 times more efficient than 800 keV protons in synthesizing CO2 on pristine CO per unit incident energy. Differently to Gerakines & Moore (2001), Loeffler et al. (2005) analyzed their data as a function of absorbed energy and found that the production rates (i.e., molecules formed per unit of absorbed energy) of CO2, formed after processing of CO ice, by both 200 keV protons and Lyman-alpha photons are similar, but at saturation (i.e., when the exponential trend approaches the asymptotic value) the concentration of CO2 is three times higher for ion bombardment.

In studies of N2 ices, Hudson & Moore (2002) discovered that while the N3 (azide) radical was readily formed after bombardment with 0.8 MeV protons, photolysis by far-UV did not result in the detection of N3. However, studies by Wu et al. (2012) have shown that UV photolysis of thick (about 160 μm) solid N2 samples does indeed produce N3, thus it is most likely that some N3 was produced in the experiments of Hudson & Moore, albeit in very small quantities below their detection limit. Furthermore, Kobayashi et al. (2008) irradiated gas mixtures of carbon monoxide, nitrogen, and water using UV light and particle beams of hydrogen (protons), helium, carbon, neon, and argon ions to investigate the production of amino acids. They found that when N2 molecules were used as the nitrogen source, amino acids were not detected after photolysis with UV light, but were detected after processing with the particle beams. Kobayashi et al. ascribed this effect to the inability of UV light to dissociate either CO or N2 molecules. From the results reported by Wu et al. (2012), assuming the density of N2 to be about 0.83 g cm-3 (Roux et al. 1980), it is possible to estimate that after UV photolysis, at saturation (i.e., when the exponential trend approaches the asymptotic value) the ratio between the column densities of N3 and initial N2 is about 10-6. Sicilia et al. (2012) reported the results of 200 keV proton bombardment of solid N2. From their results, assuming that the band strength value for the 1657 cm-1 feature of N3 is 4 × 10-17 cm/molecule (Jamieson & Kaiser 2007), it is possible to estimate that after ion bombardment with 200 keV protons at saturation the ratio between the column densities of N3 and initial N2 is about 10-4. These experimental results show that given the same initial column density of N2, the column density of N3 after ion bombardment is two orders of magnitude higher than after UV photolysis.

Therefore, although ion bombardment and UV photolysis both induce similar chemical changes to interstellar ice, there are marked differences in the yields of the two processes, and as yet, it is unknown whether the two processes can affect each other.

To the best of our knowledge, only one combined experiment has been conducted with both ion bombardment and UV photolysis of interstellar-relevant ices. Baratta et al. (2002) investigated the destruction of CH4 and CH3OH by 30 keV He+ ions and 10.2 eV Lyman-alpha photons. The effects of the two processes were found to be comparable at low doses (≤20 eV/16u), such that the same number of CH4 and CH3OH molecules are destroyed at the same doses (see their Figs. 4 and 7). However, at high doses (>20 eV/16u), UV photolysis is less efficient in the destruction of CH4 and CH3OH, whereas ion bombardment continues to induce chemical changes. This is because a residue is formed after continuous exposure to ion bombardment and UV irradiation (Moore et al. 1983; Strazzulla et al. 1983a; Strazzulla & Johnson 1991; Jenniskens et al. 1993; Ferini et al. 2004). The formation of the residue changes the optical properties of the pristine material that becomes progressively optically thick to UV radiation as processing continues. This causes the absorption coefficient to increase, therefore most UV photons are absorbed by the uppermost layers of the sample and do not penetrate the bulk of the ice, whereas the release of energy from ions into the ice remains unaffected. When a sample of CH4 ice was first irradiated with UV up to a dose of 35 eV/16u and was then bombarded with 30 keV He+ ions (from 37 eV/16u to 89 eV/16u) in a combined experiment, Baratta et al. (2002) found that the rate of CH4 destruction by ions between doses of 40 eV/16u to 60 eV/16u increased in comparison with an experiment where the ice was bombarded by ions alone (see their Fig. 5). CH4 molecules were found to be more efficiently destroyed by ion bombardment in an ice sample that had previously undergone UV photolysis. Therefore, it is possible that UV photolysis can enhance the process of ion bombardment.

The concept of synergy between different energetic processes in space has been considered in previous work. Westley et al. (1995a and 1995b) irradiated water-ice with Lyman-alpha photons to investigate photodesorption. They speculated that sputtering of ices by ions may be enhanced due to precursor radicals produced by UV photolysis, which may explain the difference between observations of OH near Tethys, a satellite of Saturn (Shemansky et al. 1993), and sputtering calculations (Johnson et al. 1993). In addition, they considered the possibility that an impinging cosmic ray could cause a transient temperature increase to enhance the sublimation of photolyzed ice to augment photodesorption rates.

As yet, it is unknown to what extent the combined effect of ion bombardment and UV photolysis influences the chemical pathways in icy mixtures. In this paper, we aim to investigate the effect of the occurrence of these two processes in tandem and separately on interstellar ices. To simplify this study we have considered the case of a CH3OH:N2 mixture. This mixture was chosen because the effects of ion bombardment and UV photolysis on methanol have been extensively studied in previous investigations (e.g., Gerakines et al. 1996; Baratta et al. 2002) and because Lyman-alpha photons have a very low efficiency in breaking the dinitrogen bond (Hudson & Moore 2002; Moore & Hudson 2003; Wu et al. 2012) because N2 is almost transparent at Lyman-alpha wavelengths. Based on the experimental results reported by Moore & Hudson (2003) and Wu et al. (2012), we expect that the number of N-bearing species formed by UV photolysis is on the order of 10-4 with respect to pristine species. This amount is too low to be detected in the experimental conditions adopted in this work. On the other hand, fast ions can efficiently break N2 molecules, giving rise to a significant number of N-bearing species. At this point, N-bearing radicals and molecules could be excited and/or broken by Lyman-alpha photons, giving rise to detectable additional (i.e., synergistic) effects. For clarity, throughout the manuscript we use the term ion bombardment to indicate ion processing, and UV photolysis or UV irradiation to indicate UV processing. The term processing is used to mean either process.

2. Experimental method

The experimental apparatus has been described in detail elsewhere (Strazzulla et al. 2001; Baratta et al. 2002; Palumbo et al. 2004). Experiments were carried out in a stainless steel vacuum chamber shown in Fig. 1 at a base pressure of about 10-7 mbar. Transmission spectra were taken using a Bruker Equinox 55 FTIR spectrometer at a resolution of 1 cm-1 in the range 4400−400 cm-1 (2.27−25 μm). A silicon substrate of thickness 500 μm was cooled to 16 K by a cold-finger in thermal contact with a closed-cycle helium cryostat. Gases were admitted into the vacuum chamber through a needle valve, resulting in a background deposition of ice onto the silicon crystal. This method has the advantage that the film deposited has an uniform thickness but has the disadvantage that gas phase molecules can freeze out onto all cold surfaces inside the chamber. To avoid any deposition on the backside of the substrate, this is protected by a 2 cm long copper tube aligned with the IR beam and with the central hole with a diameter of about 4.5 mm. The thickness of the deposited ice was determined from the interference fringe curve of a reflected He-Ne laser beam following Baratta & Palumbo (1998), Fulvio et al. (2009), and Modica & Palumbo (2010).

thumbnail Fig. 1

Schematic view of the experimental apparatus used for in situ IR spectroscopy of frozen gases after ion bombardment and UV photolysis.

Open with DEXTER

Bombardment by 200 keV protons was performed using an electrostatic 200 kV ion implanter (Danfysik) with magnetic mass separation. The ion beams were scanned electrostatically to ensure a uniform fluence over the sample. The ion beam and the infrared beam were mutually perpendicular and formed an angle of 45 degrees with the substrate plane, thus before, during, and after bombardment, spectra could be obtained in situ without tilting the sample. The ion beam produced a spot on the target larger than the area probed by the IR beam. The current density of the ion beam must be sufficiently low to avoid macroscopic heating of the sample and typically ranged from 100 nA cm-2 to a few μA cm-2. The ions passed through the sample without stopping, as the penetration depth of impinging ions is known to be much deeper than the sample thickness from SRIM calculations (Ziegler et al. 2008). Typically, the thickness of our samples was about 250 nm and the penetration depth of 200 keV H+ in a CH3OH:N2 mixture was about 2 μm.

Table 1

Details of experiments.

Photolysis of the sample by UV photons was carried out by means of a microwave-discharge hydrogen lamp interfaced with the vacuum chamber through a MgF2 window. The discharge was produced by a continuous Opthos microwave generator operating at 2.45 GHz. The lamp was operated with a gas mixture of H2 (30%) and Ar (70%) and a gas pressure P = 0.5 mbar. The lamp used in this work as well as the operating conditions were the same as used by Baratta et al. (2002) and Loeffler et al. (2005). An aluminium light collector was placed at the end of the lamp to increase the number of UV photons that reached the sample. We considered that the discharge lamps are not monochromatic and that the spectrum of the emitted radiation depends on experimental conditions, such as pressure in the discharge tube and gas mixture ratio. For example, the operating conditions adopted by Westley et al. (1995b, see their Fig. 1) produced UV photons mainly at Lyman-alpha wavelength. The operating conditions of the UV lamp adopted in this work were very similar to those used by Westley et al. (1995a,b), and we assumed that most of the photons were emitted at 121.6 nm (10.2 eV, Lyman-alpha). We developed a monitoring system of the UV flux to obtain a reliable measure of the total number of UV photons that reach the sample during photolysis (see Baratta et al. 2002). This system is described in detail in the online appendix. The flux of the lamp can be adjusted in the range 1012−5 ×  1014 cm-2 s-1. The typical fluence of UV photons used is on the order of 1018 cm-2.

In this study CH3OH and N2 were coadsorbed on a silicon substrate held at 16 K in a sandwich between layers of Ar to protect the sample from contamination by water and other background gases in the chamber. On the basis of different calibration experiments we have shown that contamination species are H2O and CO2, which deposit on top of the sample during processing (Loeffler et al. 2005). After further processing, CO can be formed from CO2. Because of the Ar layer, these additional deposits do not affect the chemical changes induced by processing, which take place in the bulk of the sample. As shown by Loeffler et al. (2005), the deposition of background gases is significantly larger during UV photolysis than ion bombardment. In particular, after a photon fluence equal to 2.5 × 1018 photons cm-2, the CO2 column density can be as high as 2 × 1016 molecules cm-2 (Loeffler et al. 2005). As a consequence of this phenomenon, in this study we did not consider the trend of CO and CO2 column density during processing.

The IR spectra may be taken by selecting the parallel (P polarization) or perpendicular (S polarization) component of the electric vector with respect to the plane of incidence using a rotatable polarizer placed in front of the detector. For each experiment, background spectra in P and S polarization were acquired before ice deposition and subtracted from all spectra with the same polarization (Baratta & Palumbo 1998; Baratta et al. 2000). P polarization spectra are presented here because the signal-to-noise ratio is higher than for S polarization.

3. Results

We deposited about 250 nm thick CH3OH:N2 ice sandwiched between two Ar layers on a silicon substrate held at 16 K. The samples were processed by 200 keV H+ and 10.2 eV Lyman-alpha photons separately and in tandem. For the simultaneous processing, two different experiments were performed with a different ratio between UV and ion fluence. A list of the experiments is given in Table 1.

Figure 2 shows the spectra, in optical depth scale, relative to the CH3OH:N2 mixture as deposited (panel a), after ion bombardment (panel b), after UV photolysis (panel c), and after simultaneous processing (panel d). Changes in the intensity of the methanol bands at 1039 cm-1 (C-O stretch), 1129 cm-1 (-CH3 rock), 1460 cm-1 (C-H deformation and O-H bend), 2830 cm-1 (symmetric C-H stretch), 2961 cm-1 (asymmetric C-H stretch), and 3260 cm-1 (O-H stretch) indicate that destruction of methanol occurs after processing.

thumbnail Fig. 2

IR transmittance spectra in optical-depth scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K as deposited; after bombardment with 3.91 × 1014 200 keV H+ cm-2 (dose = 10 eV/16u); after photolysis with 2.42 × 1018 photons cm-2 (dose = 41 eV/16u); after simultaneous processing with 1.29 × 1014 200 keV H+ cm-2 and 1.76 × 1018 photons cm-2 (total dose = 37 eV/16u).

Open with DEXTER

To compare the effects that are induced by fast ions and UV photons, the irradiation dose was derived (energy released onto the target molecules, i.e., eV molecule-1). As suggested by Strazzulla & Johnson (1991), the dose given in units of eV per small molecule (16u) is a convenient way to characterize chemical changes and to compare the results obtained after processing of different samples. The energy released into the sample by impinging ions or photons (dose) is then given in eV/16u, where u is the unified atomic mass unit defined as 1/12 of the mass of an isolated atom of carbon-12. For the ion bombardment, the dose (eV/16u) was obtained from the ion fluence values (ions cm-2) measured during the experiment and the stopping power S (eV cm2 molecule-1) calculated using the SRIM code (Ziegler et al. 2008). For the UV photolysis, the dose (eV/16u) was obtained from the UV fluence (photons cm-2) measured during the experiment and the absorption cross-section of Lyman-alpha photons in CH3OH (σa = 6.37 × 10-18 cm2) given by Baratta et al. (2002) assuming that the absorption coefficient of Lyman-alpha photons in N2 is negligible. Two experiments with simultaneous processing were performed. In the first experiment, the dose given by UV photons was on average 15 times higher than the dose given by 200 keV protons. In the second experiment the dose ratio was about 1.5.

thumbnail Fig. 3

Normalized methanol column density as a function of dose for the UV photolysis, ion bombardment, and simultaneous processing.

Open with DEXTER

After processing the pristine CH3OH:N2 ice, the IR spectra showed a reduction in methanol abundance. Figure 3 shows the column density of methanol normalized to the initial methanol column density as a function of dose for the UV photolysis, ion bombardment, and simultaneous processing. The column density of methanol was derived from the band at about 1039 cm-1 after the transmittance spectrum was converted to optical depth scale and using the band strength value A = 1.3 × 10-17 cm molecule-1 (Palumbo et al. 1999). As discussed by Baratta et al. (2002), at low doses the trend of methanol column density is very similar in all cases, while at high doses the destruction of methanol is slightly slower in the case of UV photolysis. This effect can be ascribed to the change of the optical properties of the sample during processing. The optical constants of the sample change in all cases due to energetic processing. It is well known that C-bearing species evolve toward a polymer-like material and eventually to a refractory residue (Moore et al. 1983; Strazzulla et al. 1983a; Strazzulla & Johnson 1991; Jenniskens et al. 1993; Ferini et al. 2004). The energy released by ions is independent of the optical constants of the sample, and at high doses impinging ions continue to release energy through the whole thickness of the sample and thus an additional modification of the sample occurs; on the other hand, the absorption coefficient of the refractory residue is higher than that of the pristine ice, and UV photons cause minor additional modifications to the deeper layers of the sample as photolysis proceeds.

thumbnail Fig. 4

IR transmittance spectra in optical-depth scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K before processing (dashed line), after bombardment with 1.36 × 1015 200 keV H+ cm-2 (dose = 36 eV/16u; dotted line), and after photolysis with 2.42 × 1018 photons cm-2 (dose = 41 eV/16u; solid line). HNCO, N2O, OCN, HCN, and CH2N2 are not detected after UV irradiation.

Open with DEXTER

The experimental data were fitted by the exponential curve (1)where y is the normalized column density of methanol, σ is the cross section (in 16u/eV), and D is the dose in (eV/16u). We obtained an average destruction cross-section for methanol equal to 0.19 ± 0.03 16u/eV.

After ion bombardment, newly formed species such as CO2 (2343 cm-1), CO (2138 cm-1), H2CO (1728 cm-1), CH4 (1305 cm-1), N2O (2235 cm-1), HNCO (2263 cm-1), and OCN (2168 cm-1) were detected in the ice mixture (Figs. 2 and 4). There is also a broad feature at about 2080 cm-1 that may be attributed to some combination of HCN and CH2N2 (Grim & Greenberg 1987; Sandford & Allamandola 1990; Elsila et al. 1997; Demyk et al. 1998; Novozamsky et al. 2001; Hudson et al. 2001; Moore & Hudson 2003). The formation of N-bearing species indicates that the N ≡ N bond has been broken by ions. A weak feature is present at about 2327 cm-1, which is assigned to N2. In fact, N2 has no infrared active bands because it is a homonuclear molecule; but in the solid phase the symmetry is broken and a weak band appears because of the interaction with neighboring molecules (Sandford et al. 2001).

Although methanol is destroyed for the UV photolysis, only species such as CO2, CO, H2CO, and CH4 are detected while the amount of any N-bearing species is below the detection limit (Figs. 2 and 4). As expected from the work of Hudson & Moore (2002), N3 was not detected. Wu et al. (2012) have shown that for very thick layers of N2 ice (160 μm), N3 can be detected after UV photolysis. Given that the ratio of N3 to N2 formed by UV photolysis is approximately 10-6 (Wu et al. 2012), for our sample thickness of 250 nm, where N2 constitutes about half of the ice, we predict that any N3 band at 1657 cm-1, assuming that the width is 3 cm-1, will have a peak optical depth of about 5 × 10-6, which is well below the level of noise in our spectra (τnoise ~ 10-4).

In comparing panels (b) and (c) in Fig. 2, it is apparent that relatively more CO and CO2 is produced by UV photolysis than by ion bombardment. As discussed in Sect. 2, the amount of CO and CO2 detected by IR spectroscopy is affected by the problem that UV photons cause desorption of H2O and CO2 from the walls of the vacuum chamber, followed by deposition of these species on top of the Ar layer of the sample. CO is then formed by UV photolysis of CO2 molecules. This effect is negligible for the ion bombardment experiments because the ion beam is well focused on the sample and does not hit the walls of the vacuum chamber.

An IR spectrum taken after the simultaneous processing of CH3OH:N2 ice by both H+ ions and UV photons is shown in Figs. 2 and 5. The simultaneous processes destroy methanol and cause the formation of species such as CO2, CO, H2CO, CH4, N2O, HNCO, OCN, and CH2N2. As expected, ion bombardment efficiently breaks up the N2 molecule to produce N-bearing species. Also, comparatively large abundances of CO and CO2 are produced due to UV photolysis.

thumbnail Fig. 5

IR transmittance spectra in optical depth-scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K before (dashed line), after bombardment with 1.58 × 1014 200 keV H+ cm-2 alone (dose = 4 eV/16u; dotted line) and after simultaneous processing with 1.29 × 1014 200 keV H+ cm-2 and 1.76 × 1018 photons cm-2 (total dose = 37 eV/16u; solid line). N-bearing species such as HNCO, N2O, OCN, HCN, and CH2N2 are formed.

Open with DEXTER

As discussed above, N-bearing species are not expected to be detected after UV photolysis in our experimental conditions. Therefore a comparison was made between the spectra of samples processed by ions and UV, and spectra relative to a sample bombarded only by ions at the same ion fluence to study the formation of N-bearing species (namely HNCO and OCN) after simultaneous processing. This spectral region also contains strong 13CO2 (2278 cm-1) and CO (2138 cm-1) bands that overlap with the HNCO (2263 cm-1) and OCN (2168 cm-1) bands, respectively. To better compare the effects of the different processing, the spectrum relative to the UV photolyzed sample was subtracted from the spectrum of the ion- and UV processed sample. This corrected spectrum for simultaneous ion bombardment and UV irradiation was compared with the spectrum for the ions-only experiment at the same ion fluence, as shown in Fig. 6. Clearly, the spectra in Fig. 6 exhibit very similar band profiles and intensities, indicating that the abundances of HNCO and OCN are not affected by UV photolysis when simultaneous energetic processing takes place.

thumbnail Fig. 6

IR transmittance spectra in optical-depth scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K after bombardment with 1.58 × 1014 200 keV H+ cm-2 (dose = 4 eV/16u; dashed line) and after processing by a combination of 1.29 × 1014 200 keV H+ cm-2 and 1.76 × 1018 photons cm-2 (total dose = 37 eV/16u) with the band profile contribution from CO and 13CO2 produced by UV photolysis removed (solid line). The band profiles and intensities of HNCO and OCN are similar for both spectra.

Open with DEXTER

Figure 7 shows the column density of CH4 as a function of dose. The column density of CH4 was derived from the spectra in optical-depth scale from the band at about 1310 cm-1 using the band strength value A = 6.4 × 10-18 cm molecule-1 (Mulas et al. 1998). For each experiment the CH4 column density was normalized to the initial CH3OH column density. At the beginning of processing, the CH4 column density increased rapidly; it reached the highest value when the dose increased and after further processing it started to decrease. The figure shows that at the beginning (i.e., at low dose) the formation rate of CH4 is the same regardless of the processing. On the other hand, the highest CH4 column density value reached after UV photolysis was about three times higher than the maximum amount reached after ion bombardment. As discussed above, for CH3OH destruction, this is caused by the change of the optical constants of the sample after processing. Furthermore, similarly to the discussion by Loeffler et al. (2005), in the case of 13CO2 formation after processing of pure 13CO, this could be caused by the different ratio between the formation and destruction cross-section of CH4 by 200 keV protons and Lyman-alpha photons. For the simultaneous processing, the highest CH4 column density depends on the ratio between the UV and ion dose during the experiments. When the UV dose prevails (average dose ratio = 15), the highest CH4 column density is similar to the value reached after UV photolysis whereas when UV and ion doses are similar (dose ratio = 1.5), the highest CH4 column density is an intermediate value between the value obtained after UV photolysis and the value obtained after ion bombardment.

thumbnail Fig. 7

Column density of CH4 formed after processing of CH3OH:N2 mixtures at 16 K. The CH4 column density is normalized to the initial CH3OH column density for each sample. The solid line is the fit of the experimental data relative to UV photolysis by an exponential curve. The dashed line connecting data points relative to ion bombardment has been drawn to guide the eye.

Open with DEXTER

The experimental data (from dose = 0 to the dose value at which the highest CH4 column density is reached) were fitted with an exponential curve (2)where y is the asymptotic value, σ is the cross section (in 16u/eV) and D is the dose (in eV/16u). The formation rate (R in 16u/eV) was obtained from the equation (3)The parameters resulting from the fit and R values are listed in Table 2.

Table 2

Best-fit parameters for the formation of CH4 after energetic processing of a CH3OH:N2 mixture at 16 K.

4. Discussion

Interstellar ices are known to be simultaneously processed by both cosmic-ray bombardment and UV photolysis. It has been estimated that in dense molecular clouds the flux of cosmic ions is 1 cm-2 s-1 in the approximation of effective monoenergetic 1-MeV protons (Mennella et al. 2003). This flux must be regarded as an effective quantity. It represents the flux of 1 MeV protons, which gives rise to the ionization rate produced by the cosmic-ray spectrum if hydrogen were the only source for ionization. As discussed by Prasad & Tarafdar (1983), an internal UV flux, F(UV), is present in dense molecular clouds due to the cosmic-ray-induced fluorescence of molecular hydrogen. Depending on the assumed cosmic-ray spectrum at low energy, Shen et al. (2004) obtained F(UV) = 2 × 103−3 × 104 photons cm-2 s-1.

In our laboratory investigations we used 200 keV protons; based on experimental results, the best assumption we can make is that the chemical effects depend on the total dose (eV molecule-1) deposited onto the sample, therefore these effects scale with the stopping power of impinging ions (e.g., Palumbo 2006; Sicilia et al. 2012).

Recently, Shen et al. (2004) estimated the energy deposition onto water-ice grain mantles by cosmic rays and by UV photons in dense molecular clouds. They found that depending on the assumed cosmic-ray spectrum at low energy, after 107 years the dose deposited by UV photons varies in the range 100−10 eV/molecule and the dose deposited by cosmic rays varies in the range 10−1 eV/molecule. Based on these results, the energy input by UV photons is about an order of magnitude higher than the energy input by cosmic-ray particles. This means that our experimental results are fully representative of the dose range deposited by UV photons and cosmic rays on icy grain mantles in dense molecular clouds.

Moore (1999) showed that in diffuse regions UV photons deposit more energy in the icy mantles than 1 MeV protons. In particular, after 107 years the dose deposited by UV photons is on the order of 106 eV/molecule and the dose deposited by low-energy cosmic rays is on the order of 30 eV/molecule. This result is obtained assuming that the flux of 1 MeV protons is about 10 cm-2 s-1 (Strazzulla & Johnson 1991; Indriolo & McCall 2012) and the flux of UV photons is about 9.6 × 107 photons cm-2 s-1 (Jenniskens et al. 1993). As suggested by Westley et al. (1995a), in diffuse regions ice mantles are not observed because the photon induced desorption rate is higher than the gas adsorption rate, implying rapid removal of a pre-existing ice mantle.

In a recent work, Pedersen & Gómez De Castro (2011) have modeled the disk of a young stellar object assuming that the UV radiation field of a T Tauri star at 500 AU from the central object is 2.9 × 1010 photons cm-2 s-1. Strazzulla et al. (1983b) have estimated that the flux of 1 MeV protons in a T Tauri star is 2 × 1010 cm-2 s-1 at 1 AU. Assuming that the flux decreases with distance as d-2, at 500 AU the value is 8 × 104 cm-2 s-1. Using the same hypothesis as Moore (1999), UV photons deposit more energy in the icy mantles than 1 MeV protons also in circumstellar regions. However, when planetesimals form, UV photons are absorbed in the external layers (on the order of 102 nm) while energetic ions can penetrate the bodies and release their energy in deeper layers.

As another example, it has been shown that cosmic rays can alter the outer few meters of a cometary nucleus. On the other hand, the penetration depth of UV photons is lower than that of ions and their contribution can be neglected (e.g., Strazzulla & Johnson 1991; Modica et al. 2012). It has been estimated that near the surface the dose deposited by cosmic rays after 109 years is as high as hundreds of eV/16u; therefore, volatile species eventually formed by cosmic-ray bombardment would be quickly destroyed by the same mechanism. At deeper layers from the surface, the absorbed dose decreases and doses as low as a few eV/16u, still able to produce noticeable chemical effects, are accumulated at depths on the order of tens of meters.

We have considered a CH3OH:N2 mixture. It might be argued that observations indicate that water ice is the most abundant solid-phase species in cometary and interstellar ices. Accordingly, the study of water-rich ice mixtures should be considered. However, in our experiments the initial thickness of the ice samples was chosen as a compromise between a sample thick enough to obtain infrared spectra with a good signal-to-noise ratio and thin enough to be uniformly processed by ions and UV photons. If we add water to the mixture (without increasing the thickness of the sample), we reduce the amount of CH3OH and N2, reducing in turn the amount of N-bearing species formed after processing. This reduction would make the detection of these species even more difficult. On the other hand, recent laboratory experiments (e.g., Fuchs et al. 2009) have shown that CH3OH molecules efficiently form after hydrogenation of CO molecules in CO-rich and water-poor ices. Therefore it is reasonable to assume that CH3OH-rich and water-poor ice mantles may exist along the line of sight of dense molecular clouds (e.g., Skinner et al. 1992; Palumbo & Strazzulla 1992; Cuppen et al. 2011).

Most of our knowledge on the effects of UV photons and low-energy ions on astrophysically relevant ices is based on laboratory simulations. In the past 35 years, several experiments have been performed in different laboratories to study these effects separately but, to the best of our knowledge, never simultaneously. Here we have presented an experimental study of the effects of simultaneous processing on a CH3OH:N2 mixture at 16 K. Our results confirm that when UV photolysis and ion bombardment act separately, from a qualitative point of view their effects are very similar but significant quantitative differences may exist. We did not detect any synergistic effect after simultaneous processing, but from a quantitative point of view the behavior of newly formed molecules (such as CH4, see Fig. 7) can in some instances significantly depend on the UV/ions dose ratio. Hence the effects of simultaneous processing on other astrophysically relevant mixtures should be experimentally investigated to better understand the chemical evolution of solid phase molecules in space.

Acknowledgments

This research has been financially supported by the Italian Ministero dell’Istruzione, dell’Università e della Ricerca (PRIN-MIUR) and by the LASSIE Initial Training Network, which is supported by the European Community’s Seventh Framework Programme under Grant Agreement Number 238258.

References

Appendix A: Monitoring system of the UV flux during irradiation

We developed a monitoring system of the UV flux to obtain a reliable measure of the total number of UV photons that reach the sample during photolysis (see Baratta et al. 2002). This was obtained by measuring the current produced by the photoelectric effect on a platinum wire placed at the end of the aluminium light collector. The integrated current C (μC) measured by the wire detector is proportional to the photon fluence Φph (cm-2), which is the total UV photon number per square centimeter that reaches the sample. In particular, (A.1)where Mcal is the calibration constant. The calibration constant of the wire was determined by measuring the O2 to O3 conversion rate in a UV photolyzed O2 ice film (see Gerakines et al. 1996, 2000; Baratta et al. 2002). Following Gerakines et al. (2000), we assumed a conversion efficiency of O2 to O3 equal to the quantum yield Y measured for the gas phase by Groth (1937), corresponding to 1.92 ozone molecules produced for each absorbed UV photon in the O2 film. The ozone production was monitored through the 1040 cm-1 band by using a band strength value of 1.4 × 10-17 cm molecule-1 (Smith et al. 1985). If we consider an O2 film of thickness t (cm), the column density of O3 molecules Nph, t) (cm-2) formed in the film after a photon fluence Φph (cm-2) can be computed through the equation (A.2)where n is the O3 number density at saturation (cm-3), that is, when the number of O3 molecules formed (O2 to O3) is equal to the number of O3 molecules destroyed (O3 to O2); α is the absorption coefficient (cm-1) at Lyman-alpha of the irradiated O2 film and σ is the cross section of the process (cm2). Equation (A.2) takes into account the exponential decay of the UV flux with the depth x (cm) within the film. By substituting Eq. (A.1) in Eq. (A.2), the column density can be written versus the integrated current, (A.3)where s = σ·Mcal. Equation (A.3) can be used to fit the experimental data acquired during the calibration experiment. In Fig. A.1, the O3 column density at different photon fluences versus the integrated current is reported together with the theoretical fit obtained by a nonlinear least-squares routine that provides n, α, and s as best-fit parameters. The film thickness t in Eq. (A.3) was computed by using the laser interference method.

thumbnail Fig. A.1

O3 column density at different photon fluences as a function of integrated current detected. A theoretical fit obtained by a nonlinear least-squares routine is used to provide n, α, and s as best-fit parameters.

Open with DEXTER

The quantum yield Y can be related to the initial O3 formation rate when t approaches ∞ in Eq. (A.3). Indeed, in this case, all the UV photons are absorbed by an oxygen molecule since the amount of O3 molecules is negligible at very low fluence.

It follows that (A.4)From Eq. (A.3): (A.5)From Eq. (A.1): (A.6)and finally, (A.7)where the β factor (=0.93) takes into account a 10% UV photon loss by reflection and diffusion and the increased optical path due to the IR beam oblique incidence. The calibration constant depends on the geometry of the experimental set-up, that is, the relative positions of the microwave cavity of the lamp, the aluminium light collector, and the target. The geometry can be slightly different each time the UV lamp is mounted on the vacuum chamber; for this reason we performed the calibration every time the UV lamp was mounted. We found that the calibration constant was roughly the same before and after each UV photolysis section, and it varied in the range between 4 × 1015 and 6 × 1015 cm-2μC-1 among different sections. The flux of the lamp can be adjusted in the range 1012–5 × 1014 cm-2 s-1.

All Tables

Table 1

Details of experiments.

Table 2

Best-fit parameters for the formation of CH4 after energetic processing of a CH3OH:N2 mixture at 16 K.

All Figures

thumbnail Fig. 1

Schematic view of the experimental apparatus used for in situ IR spectroscopy of frozen gases after ion bombardment and UV photolysis.

Open with DEXTER
In the text
thumbnail Fig. 2

IR transmittance spectra in optical-depth scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K as deposited; after bombardment with 3.91 × 1014 200 keV H+ cm-2 (dose = 10 eV/16u); after photolysis with 2.42 × 1018 photons cm-2 (dose = 41 eV/16u); after simultaneous processing with 1.29 × 1014 200 keV H+ cm-2 and 1.76 × 1018 photons cm-2 (total dose = 37 eV/16u).

Open with DEXTER
In the text
thumbnail Fig. 3

Normalized methanol column density as a function of dose for the UV photolysis, ion bombardment, and simultaneous processing.

Open with DEXTER
In the text
thumbnail Fig. 4

IR transmittance spectra in optical-depth scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K before processing (dashed line), after bombardment with 1.36 × 1015 200 keV H+ cm-2 (dose = 36 eV/16u; dotted line), and after photolysis with 2.42 × 1018 photons cm-2 (dose = 41 eV/16u; solid line). HNCO, N2O, OCN, HCN, and CH2N2 are not detected after UV irradiation.

Open with DEXTER
In the text
thumbnail Fig. 5

IR transmittance spectra in optical depth-scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K before (dashed line), after bombardment with 1.58 × 1014 200 keV H+ cm-2 alone (dose = 4 eV/16u; dotted line) and after simultaneous processing with 1.29 × 1014 200 keV H+ cm-2 and 1.76 × 1018 photons cm-2 (total dose = 37 eV/16u; solid line). N-bearing species such as HNCO, N2O, OCN, HCN, and CH2N2 are formed.

Open with DEXTER
In the text
thumbnail Fig. 6

IR transmittance spectra in optical-depth scale (P polarization) of CH3OH:N2 ice (about 250 nm thick) deposited at 16 K after bombardment with 1.58 × 1014 200 keV H+ cm-2 (dose = 4 eV/16u; dashed line) and after processing by a combination of 1.29 × 1014 200 keV H+ cm-2 and 1.76 × 1018 photons cm-2 (total dose = 37 eV/16u) with the band profile contribution from CO and 13CO2 produced by UV photolysis removed (solid line). The band profiles and intensities of HNCO and OCN are similar for both spectra.

Open with DEXTER
In the text
thumbnail Fig. 7

Column density of CH4 formed after processing of CH3OH:N2 mixtures at 16 K. The CH4 column density is normalized to the initial CH3OH column density for each sample. The solid line is the fit of the experimental data relative to UV photolysis by an exponential curve. The dashed line connecting data points relative to ion bombardment has been drawn to guide the eye.

Open with DEXTER
In the text
thumbnail Fig. A.1

O3 column density at different photon fluences as a function of integrated current detected. A theoretical fit obtained by a nonlinear least-squares routine is used to provide n, α, and s as best-fit parameters.

Open with DEXTER
In the text

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.