Skip to main content

Molecular synthesis in ices triggered by dissociative electron attachment to carbon monoxide

A Publisher Correction to this article was published on 11 January 2022

This article has been updated


Chemical reactions in mixed molecular ices as relevant in the context of astrochemistry can be initiated by electron-molecule interactions. Dissociative electron attachment (DEA) as initiating step is identified from the enhancement of product yields upon irradiation at particular electron energies. Herein, we show that DEA to CO leads to the formation of HCN in mixed CO/\(\hbox {NH}_{{3}}\) ice at electron energies around 11 eV and 16 eV. We propose that this reaction proceeds via insertion of the neutral C fragment into a N–H bond. In the case of CO/\(\hbox {H}_{{2}}\)O and CO/\(\hbox {CH}_{{3}}\)OH ices, a resonant enhancement of the yields of HCOOH and \(\hbox {CH}_{{3}}\)OCHO, respectively, is observed around 10 eV. In both ices, both molecular constituents exhibit DEA processes in this energy range so that the energy-dependent product yield alone does not uniquely identify the relevant DEA channel. However, we demonstrate by comparing with earlier results on mixed ices where CO is replaced by \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) that DEA to CO is again responsible for the enhanced product formation. In this case, \(\hbox {O}^{\cdot -}\) activates \(\hbox {H}_{{2}}\)O or \(\hbox {CH}_{{3}}\)OH which leads to the formation of larger products. We thus show that DEA to CO plays an important role in electron-induced syntheses in molecular ices.

Graphical abstract


Electron-molecule interactions are typically associated with their dissociative nature which has been studied for many cases and reviewed repeatedly [1,2,3,4,5,6,7,8]. However, when these interactions occur in dense environments, reactive species resulting from such electron-induced dissociation can undergo further chemical reactions. This leads to the formation of new products [9,10,11] or can be exploited to fabricate novel materials with interesting properties [12, 13]. Electron-induced chemical processes also receive attention with respect to research on the origin of organic molecules in space. It is nowadays recognized that low-energy electrons released under the effect of ionizing radiation can drive much of the interstellar chemistry within molecular ice layers that form on the surface of cold dust grains [11, 14, 15].

In fact, electron-molecule interactions can trigger reactions that lead to the synthesis of larger molecules from smaller building blocks. In ideal cases, the products can even incorporate all atoms of the initial reactants [16]. As prototypical examples, the syntheses of ethylamine (\(\hbox {C}_{{2}}\hbox {H}_{{5}}\hbox {NH}_{2})\) from ethylene (\(\hbox {C}_{{2}}\hbox {H}_{4})\) and ammonia \((\hbox {NH}_{3})\) [17], ethanol (\(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH) from ethylene (\(\hbox {C}_{{2}}\hbox {H}_{4})\) and water (\(\hbox {H}_{{2}}\)O) [18], as well as formamide (\(\hbox {H}_{{2}}\)NCHO) from carbon monoxide (CO) and ammonia \((\hbox {NH}_{3})\) [19] have been observed under low-energy electron irradiation in cryogenic condensed ices containing the starting compounds. It was proposed in all three cases that electron ionization (EI) of one of the reactants creates an attractive force between adjacent molecules. This initiates a reaction sequence in which bonds form between the reactants, one of the hydrogen atoms migrates and the resulting product radical cation is neutralized by thermalized electrons thus leading to the neutral stable product.

While this ionization-driven reaction mechanism is temptingly simple, a deeper insight into the reactions occurring in the ice mixtures is of utmost importance to arrive at a comprehensive understanding of the electron-induced chemistry in ices. The potential complexity of the reaction networks is obvious considering the wealth of products that can be formed even in a pure ice as exemplified by the case of pure methanol (\(\hbox {CH}_{{3}}\)OH) ice [15, 20]. As another example, ethanol (\(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH) formation was also observed as consequence of electron attachment (EA) to \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) in the presence of \(\hbox {H}_{{2}}\)O [18] indicating that different reaction mechanisms can contribute to the formation of a particular product.

In an effort to unravel electron-induced chemistry in ices, we have recently conducted comprehensive studies by post-irradiation thermal desorption spectrometry (TDS) on thin layers of different pure and mixed ices [20,21,22,23]. In each case, the dependence of product formation on electron energy was monitored for several products. This approach not only gives insight into the electron-molecule interactions that initiate the formation of the products by releasing reactive intermediate species. It also allows us to draw conclusions on the reaction mechanisms that occur after this initial event. This is achieved by comparing the energy dependences of different products that are potentially formed via the same intermediates. Most interestingly, resonance structures around 10 eV in the energy dependences have suggested that dissociative electron attachment (DEA) to CO leads to formation of formic acid (HCOOH) in mixtures of CO and \(\hbox {H}_{{2}}\)O [22] and of methyl formate (\(\hbox {CH}_{{3}}\)OCHO) in mixed ices of CO and \(\hbox {CH}_{{3}}\)OH [23]. Here, we substantiate the claim that DEA to CO is an efficient pathway to product formation in ices. To this end, we present new results on mixed CO and \(\hbox {NH}_{{3}}\) ices that reveal the role of DEA to CO in electron-induced chemistry by monitoring the formation of hydrogen cyanide HCN. Furthermore, we compare the results on CO/\(\hbox {H}_{{2}}\)O and CO/\(\hbox {CH}_{{3}}\)OH ices [22, 23] with the data on mixed ices where CO is replaced by \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) [18, 21] to demonstrate that a resonance around 10 eV is only visible when CO is present. We thus show that DEA to CO plays an important role in electron-induced syntheses in molecular ices and propose mechanisms for the reactions that are initiated by the fragments resulting from DEA.

Fig. 1
figure 1

Overview of gas phase dissociative electron attachment to CH\(_3\)OH (from Ref. [24] reporting DEA ion yields in arbitrary units), H\(_2\)O (from Ref. [25] reporting absolute DEA cross sections), NH\(_3\) (from Ref. [26] reporting absolute DEA cross sections), CO (10 eV resonance from Ref. [27] reporting DEA ion yields in arbitrary units), and C\(_2\)H\(_4\) (10 eV resonance from Refs. [28, 29] reporting DEA ion yields in arbitrary units). In the case of \(\hbox {C}_2\hbox {H}_4\), only the most intense signals are displayed. See Ref. [30] for data on minor fragments. Also included is non-dissociative electron attachment at 2 eV to CO (from Ref. [31] reporting charge trapping in arbitrary units) and \(\hbox {C}_2\hbox {H}_4\) (from Ref. [32] derived from energy-dependent cross section for vibrational excitation). Each curve has been normalized to its maximum. The figure does not represent the relative yields of individual fragments from same molecule but merely serves to illustrate to resonance positions. Tick marks on the y-axis indicate the vertical offset 

Fig. 2
figure 2

a Thermal desorption spectra of CO/\(\hbox {NH}_{{3}}\) mixed ice (thickness 10–14 monolayers) without electron exposure (denoted 0 \(\upmu \)C/\(\hbox {cm}^{2})\) and after electron irradiation with 500 \(\upmu \)C/\(\hbox {cm}^{2}\) at an electron energy \(({E}_{0})\) of 20 eV. The peaks at \(\sim \)133 K in the m/z 26 and 27 curves were assigned to HCN. The signals at \(\sim \)50 K and above 220 K in the m/z 27 curve result from desorption of CO from the ice and from parts of the sample holder that warm up more slowly than the Ta foil itself. CO is detected at m/z 27 due to its high abundance which leads to spillover of intensity to mass channels adjacent to the main signal at m/z 28. Tick marks on the vertical axis indicate the baseline level. b Reference mass spectrum of HCN [35]


The electron energy-dependent formation of products upon electron irradiation of condensed molecular ices was monitored by post-irradiation thermal desorption spectrometry (TDS). The experimental procedures were described in detail previously [16,17,18,19,20,21,22,23]. Briefly, all experiments were performed in an ultrahigh vacuum chamber with a base pressure of 10\(^{-10}\) mbar. Molecular ices were prepared by leaking vapours of the reactants onto a Ta substrate (this work and Refs. [19,20,21,22,23]) or an Au substrate (Refs. [16,17,18,19]) held at \(\sim \) 30–35 K either as a gas mixture or sequentially, the latter relying on rapid diffusion into pores of the first component. The amounts of gases leaked into the chamber were monitored as pressure drop in the gas handling manifold measured with an MKS Baratron type 622B capacitance manometer. Film thickness was estimated from TDS for the individual components by observing at which pressure drop the transition from monolayer to multilayer desorption signal occurs. TDS data were acquired by resistively heating the substrate with heating rates of \(\sim \) 1 K/s while monitoring characteristic m/z ratios of the investigated neutral desorbing species using a quadrupole mass spectrometer (QMS) residual gas analyser (Stanford Research System RGA 200). The QMS has an electron-impact ion source operating at an electron energy of 70 eV.

The targeted mixing ratio of the two components in the ice can be established by different procedures. In the simplest case, a 1:1 gas mixture was leaked with the aim of producing an ice layer of the same composition. This procedure was applied previously for \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {NH}_{{3}}\) [17], \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {H}_{{2}}\)O [18], as well as \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {CH}_{{3}}\)OH [21] ices. The actual stochiometric ratio of the two reactants in the ice can be deduced by integrating the area under the TDS curves and correcting for the partial electron ionization cross sections of the characteristic fragments that are monitored in the QMS [21]. The thus determined ice composition may deviate from the mixing ratio of the gases if one of the compounds adsorbs more strongly to the walls of the gas injection tube and to parts of the vacuum chamber than the other. In the case of \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {CH}_{{3}}\)OH ice, the gas mixture was consequently adjusted to produce a 1:1 composition of the ice [21]. Alternatively, the two components can be leaked sequentially. This is advantageous when one component tends to form a porous ice such as \(\hbox {H}_{{2}}\)O. The pores can then accommodate the molecules of the subsequently dosed gas and this can enhance its sticking probability. This procedure was applied previously to prepare CO/\(\hbox {H}_{{2}}\)O ices [22] and also to study the formation of HCN in CO/\(\hbox {NH}_{{3}}\) ices reported herein. In detail, the ice was prepared by leaking first an amount of \(\hbox {NH}_{{3}}\) corresponding to a pressure drop of 28 mTorr in the manifold followed by CO (15 mTorr). According to TDS, this produced an ice with a comparable composition as used previously to investigate the formation of formamide [19].

Fig. 3
figure 3

a Peak areas for the m/z 27 desorption signal of HCN at 133 K obtained by TDS without electron exposure (plotted at 0 \(\upmu \)C/\(\hbox {cm}^{2})\) and after increasing electron irradiation of CO/\(\hbox {NH}_{{3}}\) mixed ice at an electron energy \((\hbox {E}_{0})\) of 20 eV. Error bars denote the estimated error for the peak area. b Energy dependence of the relative yield of HCN formed after electron irradiation of CO/\(\hbox {NH}_{{3}}\) ice with an electron dose of 500 \(\upmu \)C/\(\hbox {cm}^2\) as derived from the integrated m/z 27 and m/z 26 desorption signals. The thickness of the ice layer was 10–14 monolayers. Tick marks on the vertical axes indicate a peak area of zero

Electron irradiation was performed using a commercial STAIB NEK-150-1 electron source with an energy resolution of 0.5 eV. After irradiation, the ice was desorbed in a TDS experiment, again with a heating rate of \(\sim \) 1 K/s, while monitoring characteristic m/z ratios of the investigated neutral reaction products. After each experiment, the substrate temperature was held at 450 K for 2 min to desorb remaining species on the sample holder, which warms up more slowly than the substrate. A relative product yield was obtained by integrating the area under the characteristic desorption signals in the TDS curves. Typically, the same experiment was repeated three to four times to establish an experimental error. This overall procedure was then repeated for a range of electron energies to obtain energy-dependent product yields. Here, attention was paid to select an electron exposure that falls within the linear regime where product yields increase linearly with exposure. This ensures that the product yields obtained at different electron energies can be directly compared.

Results and discussion

DEA processes of the investigated molecular ice constituents

As a reference for the discussion of resonant product formation in the molecular ices studied herein, Fig. 1 provides an overview of the gas phase DEA processes of the individual ice components. It is well-established that the energetic position and the cross sections of the individual DEA channels can change in a condensed environment, the latter even by orders of magnitude [2, 33]. Nonetheless, the overview given in Fig. 1 clearly illustrates the wealth of DEA processes that exist in the vicinity of 10 eV (grey band in Fig. 1). In fact, in both CO/\(\hbox {H}_{{2}}\)O and CO/\(\hbox {CH}_{{3}}\)OH mixed ices, both components exhibit DEA resonances around this energy. The position of the maximum product yield alone is therefore not a sufficient criterion to identify which DEA channel enhances the formation of a particular product in a mixed ice. Moreover, we note that electron-stimulated desorption of \(\hbox {O}^{\cdot -}\) from condensed multilayer films of CO is resonantly enhanced around 12 eV [34, 35] and thus about 2 eV above the maximum product yield in the mixed ices described above. Also, and in contrast to the gas phase, ESD from condensed CO evinces a second resonance around 16 eV [34, 35]. We demonstrate in the following that these DEA channels indeed play a role in the investigated mixed ices. The neutral C fragment leads to the production of HCN in CO/NH\(_3\) ice (Sect. 3.2) while the \(\hbox {O}^{\cdot -}\) radical anion initiates both the formation of formic acid (HCOOH) in CO/H\(_2\)O ice and methyl formate (CH\(_3\)OCHO) in CO/CH\(_3\)OH ice (Sect. 3.3).

Molecular synthesis triggered by the neutral C fragment

The electron-induced chemistry of CO/\(\hbox {NH}_{{3}}\) mixed ice has been investigated previously with particular focus on the formation of \(\hbox {H}_{{2}}\)NCHO [19]. A resonant enhancement of this product was observed in a broad range around 10 eV. Aiming at a more comprehensive insight, we have now investigated the production of HCN in CO/\(\hbox {NH}_{{3}}\) ice and its dependence on the electron energy. Figure 2a shows representative TDS data obtained before and after electron irradiation at 20 eV. A distinct desorption signal with maximum at 133 K is seen in the TDS curves obtained for both m/z 26 and 27. The relative intensity of these signals is 1:5 and thus very similar to the intensity ratio of these two signals in a reference mass spectrum of HCN [36] (Fig. 2b) indicating that this product is indeed formed under electron irradiation of the CO/\(\hbox {NH}_{{3}}\) ice.

To establish the relative product yields as function of electron energy, the area under the TDS signal at 133 K in the m/z 27 data was evaluated. First, it was verified that the exposure of 500 \(\upmu \)C/\(\hbox {cm}^{2}\) applied in Fig. 2a falls within the linear regime so that electron-induced depletion of the reactants CO and \(\hbox {NH}_{{3}}\) as well as decomposition of the product HCN can still be neglected. Therefore, the integral intensity of the desorption signal was first plotted for increasing electron exposure at 20 eV (Fig. 3a). The product yield in fact increases linearly with electron exposure. As reaction rates are typically lower at lower electron energies, an electron exposure of 500 \(\upmu \)C/\(\hbox {cm}^{2}\) was thus also used in all subsequent experiments at lower electron energies. The resulting energy-dependent integral desorption signals of HCN are shown in Fig. 3b. The constant intensity ratio between the signals at m/z 27 and m/z 26 shows that the data are representative of HCN at all electron energies. The curves evince resonant enhancement of the product yield around 11 eV and 15 eV. DEA processes in \(\hbox {NH}_{{3}}\) that could contribute to this enhancement have been reported around 10 eV (see Fig. 1) but not at higher energy. In contrast, apart from a shift to lower energy, the resonances observed here compare well with the maxima in the ESD yields of \(\hbox {O}^{\cdot -}\) from condensed multilayer films of CO around 12 eV and 16 eV [34, 35]. This shift is rationalized by considering that desorption of the anion requires a certain excess energy to overcome the polarization forces in the condensed phase. In contrast, HCN is produced within the layer and the energy required for desorption is supplied thermally in the subsequent TDS experiment. We thus conclude that DEA to CO is the initiating electron-molecule interaction that leads to the formation of HCN in the CO/\(\hbox {NH}_{{3}}\) mixed ice. We propose that the neutral C fragment released together with \(\hbox {O}^{\cdot -}\) upon DEA to CO reacts with \(\hbox {NH}_{{3}}\) by insertion into one of the N-H bonds and subsequent elimination of \(\hbox {H}_{{2}}\) to produce HCN (Fig. 4) as also reported earlier [37].

Fig. 4
figure 4

Proposed reaction mechanism for the electron-induced formation of HCN in CO/\(\hbox {NH}_{{3}}\) mixed ice

Molecular synthesis triggered by the \(\hbox {O}^{\mathbf {\cdot -}}\) radical anion fragment

Figure 5a shows the energy-dependent yield of HCOOH in mixed CO/\(\hbox {H}_{{2}}\)O ice [22] and Fig. 5b the yield of \(\hbox {CH}_{{3}}\)OCHO in CO/\(\hbox {CH}_{{3}}\)OH ice [23], both obtained from the areas under the characteristic TDS signals. For comparison, the data for ices in which CO is replaced by \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) [18, 21] is also included. For details of the experiments and data analysis we refer to the previous studies [18, 21,22,23]. The resonant enhancement of the yields of HCOOH and \(\hbox {CH}_{{3}}\)OCHO around 10 eV clearly indicates that an electron attachment process contributes to the product formation. We show now that the sequence of reactions leading to these products must be initiated by the \(\hbox {O}^{\cdot -}\) radical anion resulting from DEA to CO. This is, however, not a priori obvious. In fact, both \(\hbox {H}_{{2}}\)O and \(\hbox {CH}_{{3}}\)OH also exhibit DEA channels in this energy range (Fig. 1). Furthermore, as discussed in Sect. 3.1, both the energies and the cross sections of DEA processes can change under the effect of the condensed medium.

Fig. 5
figure 5

Electron-energy dependent yields of product formation upon electron irradiation of different mixed ices: a HCOOH in CO/\(\hbox {H}_{{2}}\)O ice (bottom, from Ref. [22]) and \(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH in mixed \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {H}_{{2}}\)O ice (top, from Ref. [18]), b \(\hbox {CH}_{{3}}\)OCHO in CO/\(\hbox {CH}_{{3}}\)OH ice (top, from Ref. [23]) and \(\hbox {C}_{{2}}\hbox {H}_{{5}}\hbox {OCH}_{{3}}\) in \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {CH}_{{3}}\)OH ice (bottom, from Ref. [21]). The thickness of the ice layers in monolayers (ML) is stated in the figures. In each set of experiments, the electron exposures were sufficiently small to ensure that the product formation increased linearly with exposure. Product yields at different energies thus reflect the relative values of the cross section for formation of each particular product. For details of the experiments see [18, 21,22,23]

The formation of HCOOH and \(\hbox {CH}_{{3}}\)OCHO through DEA must involve reactions of radical species with CO. This should be more favourable than reactions of negative ion fragments with the electron-rich CO molecule. Regarding HCOOH formation in CO/\(\hbox {H}_{{2}}\)O mixed ice, gas phase data suggest that DEA to \(\hbox {H}_{{2}}\)O around 10 eV produces either \(\hbox {O}^{\cdot -}\) or \(\hbox {OH}^{-}\), the latter accompanied by release of \(\hbox {H}^{\cdot }\) (Fig. 1). In the mixed ice, \(\hbox {H}^{\cdot }\) can react with CO to yield the \(\hbox {HCO}^{\cdot }\) radical. This radical is a key intermediate to the formation of formaldehyde (\(\hbox {H}_{{2}}\)CO) as has also been deduced from the finding that \(\hbox {H}_{{2}}\)CO is resonantly formed around 4 eV [22] (see also Fig. 6). It was proposed that at this energy, the \(\hbox {HCO}^{\cdot }\) radical results from electron attachment (EA) to CO yielding the \(\hbox {CO}^{\cdot -}\) radical anion which in turn acts as a strong base and is thus converted to the \(\hbox {HCO}^{\cdot }\) radical by transfer of a proton from \(\hbox {H}_{{2}}\)O. However, a corresponding resonant structure at 4 eV was not observed in the yield of HCOOH (Fig. 6) thus ruling out its formation by reaction of \(\hbox {HCO}^{\cdot }\) with \(\hbox {H}_{{2}}\)O. Therefore, the DEA channel near 10 eV yielding \(\hbox {OH}^{-}\) and \(\hbox {H}^{\cdot }\) cannot account for the 10 eV resonant signal in the yield of HCOOH (Fig. 5a, bottom and Fig. 6) and is also not dominant with respect to the formation of \(\hbox {H}_{{2}}\)CO (Fig. 6). In consequence, DEA yielding \(\hbox {O}^{\cdot -}\) is the most conceivable reaction channel leading to HCOOH. \(\hbox {O}^{\cdot -}\) can accept a proton from \(\hbox {H}_{{2}}\)O to yield \(\hbox {HO}^{\cdot }\) which in turn can add to CO yielding a \(\hbox {HOCO}^{\cdot }\) radical. The latter can react with further \(\hbox {H}_{{2}}\)O to yield the product HCOOH.

Having identified \(\hbox {O}^{\cdot -}\) as the most likely species to initiate formation of HCOOH in CO/\(\hbox {H}_{{2}}\)O ice alone, however, does not provide a unique assignment of the DEA process that initiates the formation of HCOOH. In the gas phase, \(\hbox {O}^{\cdot -}\) is produced both by DEA to CO around 10 eV and to \(\hbox {H}_{{2}}\)O at slightly higher energy (Fig. 1). Considering again that gas phase resonance energies can be shifted and cross sections modified in an ice layer [2, 33], it was thus previously unclear which of these DEA processes drives the formation of HCOOH [22]. A comparison with the energy dependent yield of \(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH in mixed ices of \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) and \(\hbox {H}_{{2}}\)O [18] can resolve this question (Fig. 5a). The decisive step in the formation of \(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH is the addition of \(\hbox {HO}^{\cdot }\) radicals to the double bond \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) [18]. As noted above, \(\hbox {HO}^{\cdot }\) can be formed by reaction of \(\hbox {O}^{\cdot -}\) with \(\hbox {H}_{{2}}\)O. However, a resonant enhancement around 10 eV is absent from the energy-dependent yield of \(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH (Fig. 5a, top). Instead, the yield of \(\hbox {C}_{{2}}\hbox {H}_{{5}}\)OH increases steadily above 8 eV (Fig. 5a) pointing to a process initiated by EI or possibly ND. Comparing the cases of CO/\(\hbox {H}_{{2}}\)O mixed ice with the \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {H}_{{2}}\)O ice thus gives evidence that \(\hbox {O}^{\cdot -}\) must be formed with much higher yield by DEA to CO than to \(\hbox {H}_{{2}}\)O in the mixed ice. In conclusion, DEA to CO is the initiating electron-molecule interaction that leads to formation of HCOOH in the CO/\(\hbox {H}_{{2}}\)O mixed ice. We note that \(\hbox {C}^{-}\) has also been observed in ESD from condensed multilayer films of CO but with yield peaking at 14 eV and thus about 2 eV higher than that of \(\hbox {O}^{\cdot -}\) [34]. In contrast, the energetic shift between the maximum yield of HCOOH at 10 eV and the resonance position in ESD of \(\hbox {O}^{\cdot -}\) is in line with the earlier conclusion that the ion needs about 2 eV excess energy to be able to desorb [34]. Therefore, we rule out a significant contribution of the resonance producing \(\hbox {C}^{-}\) to the formation of HCOOH around 10 eV seen in Figs. 5a and 6.

In the case of the CO/\(\hbox {CH}_{{3}}\)OH mixed ice [23], DEA channels of \(\hbox {CH}_{{3}}\)OH near 10 eV lead to \(\hbox {H}^{-}\) accompanied by \(\hbox {CH}_{{3}}\hbox {O}^{\cdot }\) or fragments thereof, to \(\hbox {CH}_{{3}}\hbox {O}^{-}\) and \(\hbox {H}^{\cdot }\), or to \(\hbox {O}^{\cdot -}\) and corresponding neutral species (Fig. 1). Again, it is therefore not immediately obvious if DEA to CO or to \(\hbox {CH}_{{3}}\)OH triggers the resonant production of \(\hbox {CH}_{{3}}\)OCHO around 10 eV (Fig. 5b, top). A dominant contribution of the \(\hbox {CH}_{{3}}\hbox {O}^{-}\) and \(\hbox {H}^{\cdot }\) DEA channel has been excluded. This was again deduced from the lack of resonant formation of the concurrent product \(\hbox {H}_{{2}}\)CO [23], anticipated to be produced by reaction of \(\hbox {H}^{\cdot }\) with CO and subsequent abstraction of a hydrogen atom from \(\hbox {CH}_{{3}}\)OH by the resulting \(\hbox {HCO}^{\cdot }\) radical. It is, however, more difficult to identify which of the remaining DEA channels is responsible for formation of \(\hbox {CH}_{{3}}\)OCHO. This product results when a \(\hbox {CH}_{{3}}\hbox {O}^{\cdot }\) radical adds to CO to produce \(\hbox {CH}_{{3}}\hbox {OCO}^{\cdot }\) which, in turn, abstracts a hydrogen atom from an adjacent \(\hbox {CH}_{{3}}\)OH molecule to yield the final product. \(\hbox {CH}_{{3}}\hbox {O}^{\cdot }\) can result directly from DEA to \(\hbox {CH}_{{3}}\)OH or via reaction of \(\hbox {O}^{\cdot -}\) with \(\hbox {CH}_{{3}}\)OH. Note again that \(\hbox {O}^{\cdot -}\) is potentially formed by DEA to both \(\hbox {CH}_{{3}}\)OH and CO. These different reaction pathways can again be distinguished by comparing with the \(\hbox {C}_{{2}}\hbox {H}_{{4}}\)/\(\hbox {CH}_{{3}}\)OH mixed ice (Fig. 5b, bottom). Here, addition of \(\hbox {CH}_{{3}}\hbox {O}^{\cdot }\) to \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) and abstraction of hydrogen from an adjacent molecule yields \(\hbox {CH}_{{3}}\hbox {OC}_{{2}}\hbox {H}_{{5}}\) [21]. However, as deduced from the threshold-type behavior of the energy dependent product yield, this reaction is not dominated by DEA (see also [21]). In conclusion, this supports again that DEA to CO rather than DEA to \(\hbox {CH}_{{3}}\)OH is responsible for the 10 eV resonant enhancement of the yield of \(\hbox {CH}_{{3}}\)OCHO in CO/\(\hbox {CH}_{{3}}\)OH mixed ice (Fig. 5b, top). Overall, the comparison of different ice systems provides clear evidence that the \(\hbox {O}^{\cdot -}\) DEA fragment from CO activates molecules such as \(\hbox {H}_{{2}}\)O and \(\hbox {CH}_{{3}}\)OH to initiate further chemical reactions. DEA to CO thus plays a dominant role in the electron-induced synthesis of larger products.

Fig. 6
figure 6

Energy dependence of the relative yields of \(\hbox {CO}_{{2}}\), HCOOH, and \(\hbox {H}_{{2}}\)CO formed after electron irradiation of mixed CO/\(\hbox {H}_{{2}}\)O ice with an electron dose of 250 \(\upmu \)C/\(\hbox {cm}^2\). Adapted with permission from Ref. [22]. Copyright 2019 American Chemical Society

We finally comment on the different energetic position of the resonance maximum in the case of reactions driven by \(\hbox {O}^{\cdot -}\) (10 eV) and the first maximum of the HCN production driven by the neutral C fragment (11 eV). It is possible that, in the latter case, the resonant process overlaps with contributions of ND and/or DI that distort the DEA signal. As another explanation, the reactions initiated by \(\hbox {O}^{\cdot -}\) and C may encounter differently high activation barriers. However, a comprehensive evaluation of this effect is beyond the scope of this work.


We have provided evidence that DEA to CO in mixed molecular ices is an important initiating electron-molecule interaction leading to different reaction products. In the case of mixed CO/\(\hbox {NH}_{{3}}\) ice, the neutral carbon fragment resulting from DEA to CO reacts with \(\hbox {NH}_{{3}}\) to yield HCN. This is supported by the observed resonant enhancement of the HCN yield around 11 eV and 15 eV which is in good agreement with earlier ESD studies on condensed CO [34, 35]. In CO/\(\hbox {H}_{{2}}\)O and CO/\(\hbox {CH}_{{3}}\)OH ices [22, 23], the corresponding anionic DEA fragment \(\hbox {O}^{\cdot -}\) initiates reactions leading to the formation of HCOOH and \(\hbox {CH}_{{3}}\)OCHO, respectively. This is evident from the comparison with mixed ices where CO is replaced by \(\hbox {C}_{{2}}\hbox {H}_{{4}}\) [18, 21]. In these latter systems, the 10 eV resonance seen in product formation in the ices containing CO is absent. We thus show that DEA to CO plays an important role in electron-induced syntheses in molecular ices. The understanding gained herein contributes to the general goal to unravel the reactions sequences in astrochemical molecular ices.

Data availability statement

This manuscript has no associated data or the data will not be deposited. [Authors’ comment: The data sets generated during and/or analysed during the current study are available from the corresponding author upon request].

Change history


  1. 1.

    L.G. Christophorou, J.K. Olthoff, Fundamental Electron Interactions with Plasma Processing Gases, 1st edn. (Springer, Boston, 2004)

    Book  Google Scholar 

  2. 2.

    I. Bald, J. Langer, P. Tegeder, O. Ingólfsson, Int. J. Mass Spectrom. 277, 4 (2008)

    Article  Google Scholar 

  3. 3.

    J.W. McConkey, C.P. Malone, P.V. Johnson, C. Winstead, V. McKoy, I. Kanik, Phys. Rep. 466, 1 (2008)

    Article  ADS  Google Scholar 

  4. 4.

    C. Winstead, V. McKoy, Radiat. Phys. Chem. 77, 1258 (2008)

    Article  ADS  Google Scholar 

  5. 5.

    J.H. Moore, P. Swiderek, S. Matejcik, M. Allan, in Nanofabrication Using Focused Ion and Electron Beams – Principles and Applications, edited by P. Russell, I. Utke, S. Moshkalev, 1st edn. (Oxford University Press, New York, 2012)

  6. 6.

    N.J. Mason, J. Phys.: Conf. Ser. 565, 012001 (2014)

    Google Scholar 

  7. 7.

    R.M. Thorman, T.P. Ragesh Kumar, D.H. Fairbrother, O. Ingólfsson, Beilstein J. Nanotechnol. 6, 1904 (2015)

  8. 8.

    O. Ingólfsson, in Low-Energy Electrons: Fundamentals and Applications, edited by O. Ingólfsson, 1st edn. (Pan Stanford Publishing, Singapore, 2019)

  9. 9.

    A.D. Bass, L. Sanche, Low Temp. Phys. 29, 202 (2003)

    Article  ADS  Google Scholar 

  10. 10.

    A. Lafosse, M. Bertin, R. Azria, Prog. Surf. Sci. 84, 177 (2009)

    Article  ADS  Google Scholar 

  11. 11.

    C.R. Arumainayagam, H.-L. Lee, R.B. Nelson, D.R. Haines, R.P. Gunawardane, Surf. Sci. Rep. 65, 1 (2010)

    Article  ADS  Google Scholar 

  12. 12.

    A. Turchanin, A. Gölzhäuser, Prog. Surf. Sci. 87, 108 (2012)

    Article  ADS  Google Scholar 

  13. 13.

    A. Turchanin, Chimia 73, 473 (2019)

    Article  Google Scholar 

  14. 14.

    M.C. Boyer, N. Rivas, A.A. Tran, C.A. Verish, C.R. Arumainayagam, Surf. Sci. 652, 26 (2016)

    Article  ADS  Google Scholar 

  15. 15.

    C.R. Arumainayagam, R.T. Garrod, M.C. Boyer, A.K. Hay, S.T. Bao, J.S. Campbell, J. Wang, C.M. Nowak, M.R. Arumainayagam, P.J. Hodge, Chem. Soc. Rev. 48, 2293 (2019)

    Article  ADS  Google Scholar 

  16. 16.

    E. Böhler, J. Warneke, P. Swiderek, Chem. Soc. Rev. 42, 9219 (2013)

    Article  Google Scholar 

  17. 17.

    T. Hamann, E. Böhler, P. Swiderek, Angew. Chem. Int. Ed. 48, 4643 (2009)

    Article  Google Scholar 

  18. 18.

    J. Warneke, Z. Wang, P. Swiderek, J.H. Bredehöft, Angew. Chem. Int. Ed. 54, 4397 (2015)

    Article  Google Scholar 

  19. 19.

    J.H.Bredehöft, E.Böhler, F. Schmidt T. Borrmann, P. Swiderek, ACS Earth Space Chem. 1, 50 (2017)

  20. 20.

    F. Schmidt, P. Swiderek, J.H. Bredehöft, ASC Earth Space Chem. 5, 391 (2021)

    Article  ADS  Google Scholar 

  21. 21.

    F. Schmidt, P. Swiderek, J.H. Bredehöft, J. Phys. Chem. A 123, 37 (2019)

    Article  Google Scholar 

  22. 22.

    F. Schmidt, P. Swiderek, J.H. Bredehöft, ASC Earth Space Chem. 3, 1974 (2019)

    Article  ADS  Google Scholar 

  23. 23.

    F. Schmidt, P. Swiderek, T. Scheele, J.H. Bredehöft, Phys. Chem. Chem. Phys. 23, 11649 (2021)

    Article  Google Scholar 

  24. 24.

    B.C. Ibanescu, O. May, A. Monney, M. Allan, Phys. Chem. Chem. Phys. 9, 3163 (2007)

    Article  Google Scholar 

  25. 25.

    Y. Itikawa, N. Mason, J. Phys. Chem. Ref. Data 34, 1 (2005)

    Article  ADS  Google Scholar 

  26. 26.

    P. Rawat, V.S. Prabhudesai, M.A. Rahman, N.B. Ram, E. Krishnakumar, Int. J. Mass Spectrom. 277, 96 (2008)

    Article  Google Scholar 

  27. 27.

    K. Gope, V. Tadsare, V.S. Prabhudesai, N.J. Mason, E. Krishnakumar, Eur. Phys. J. D 70, 134 (2016)

    Article  ADS  Google Scholar 

  28. 28.

    J.C.J. Thynne, K.A.G. MacNeil, J. Phys. Chem. 75, 2584 (1971)

    Article  Google Scholar 

  29. 29.

    L. von Trepka, H. Neuert, Z. Naturforsch. A: Phys. Sci. 18, 1295 (1963)

  30. 30.

    E. Szymańska, N.J. Mason, E. Krishnakumar, C. Matias, A. Mauracher, P. Scheier, S. Denifl, Int. J. Mass Spectrom. 365–366, 356 (2014)

    Article  Google Scholar 

  31. 31.

    R.D. Rempt, Phys. Rev. Lett. 22, 1034 (1969)

    Article  ADS  Google Scholar 

  32. 32.

    I.C. Walker, A. Stamatovic, S.F. Wong, J. Chem. Phys. 69, 5532 (1978)

  33. 33.

    A.D. Bass, L. Sanche, Radiat. Environ. Biophys. 37, 243 (1998)

    Article  Google Scholar 

  34. 34.

    L. Sanche, Phys. Rev. Lett. 53, 1638 (1984)

    Article  ADS  Google Scholar 

  35. 35.

    R. Azria, L. Parenteau, L. Sanche, J. Chem. Phys. 88, 5166 (1988)

    Article  ADS  Google Scholar 

  36. 36.

    NIST Mass Spectrometry Data Center, W. E. Wallace, director, in NIST Chem. WebBook, NIST Stand. Ref. Database Number 69, Eds. P.J. Linstrom W.G. Mallard, Gaithersburg MD, 20899, (retrieved March 13, 2020)

  37. 37.

    P.B. Shevlin, D.W. McPherson, P. Melius, J. Am. Chem. Soc. 105, 488 (1983)

    Article  Google Scholar 

Download references


The work presented here was funded by the Deutsche Forschungsgemeinschaft (DFG) under Project Number SW26/15-2.

Open Access

This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.


Open Access funding enabled and organized by Projekt DEAL.

Author information




FS performed the reevaluation of data for Sect. 3.3 and prepared Figs. 1, 2, 3 and 5, 6. MPM performed and analysed the experiments described in Sect. 3.2 and prepared Fig. 4. JHB supervised all experiments. PS wrote the manuscript. All authors have proof-read and discussed the final version of the manuscript.

Corresponding author

Correspondence to Petra Swiderek.

Ethics declarations

Conflict of interest

The authors declare no competing financial interest.

Additional information

The original online version of this article was revised due to one wrong formula and incorrect in-text citations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Schmidt, F., Mues, M.P., Bredehöft, J.H. et al. Molecular synthesis in ices triggered by dissociative electron attachment to carbon monoxide. Eur. Phys. J. D 75, 302 (2021).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: