Radiation-induced chemistry in the C₂H₂–H₂O system at cryogenic temperatures: a matrix isolation study

ABSTRACT

Investigations of the low-temperature radiation-induced transformations in the C₂H₂–H₂O system are relevant to the chemistry occurring in interstellar and cometary ices. In this work, we applied a matrix isolation technique to study the radiation-driven evolution of this system at molecular level in order to get new mechanistic insight. The 1:1 C₂H₂⋅ ⋅ ⋅H₂O complexes were prepared in various solid noble-gas matrices (Ar, Kr, Xe) and these icy matrices were subjected to X-ray irradiation at 5 K. Decomposition of initial complex and formation of products were monitored by Fourier-transform infrared (FTIR) spectroscopy. It was found that complexation with H₂O resulted in significant enhancement of the radiolytic decay of C₂H₂ molecules and provided new channels for its radiation-induced transformations. Ketene (both isolated H₂CCO and in the form of H₂CCO–H₂ pair), ketenyl radical (HCCO), carbon monoxide (CO), and methane (CH₄) were observed as main products of the C₂H₂⋅ ⋅ ⋅H₂O radiolysis. In addition, vinyl alcohol (CH₂CHOH) was detected in an Ar matrix. The mechanistic interpretation (reaction routes leading to formation of these products) is discussed on the basis of consideration of kinetic dependences and matrix effects. Conversion of C₂H₂⋅ ⋅ ⋅H₂O to CH₄ is a prominent example of water-mediated cleavage of the C≡C triple bond, which may occur under prolonged irradiation in rigid media. Possible astrochemical implications of the obtained results are highlighted.

astrochemistry, molecular processes, methods: laboratory: solid state, techniques: spectroscopic, ISM: molecules

1 INTRODUCTION

The simplest alkyne, acetylene (C₂H₂) widely occurs in the Universe and presumably plays an important role in extraterrestrial organic chemistry. It was detected in the gas phase by absorption and emission spectroscopy in various interstellar media including massive young stellar objects (Lahuis & van Dishoeck 2000; Carr & Najita 2008), molecular clouds (Lacy et al. 1989), and cometary comae (Mumma et al. 2003). It should be noted that concentration of acetylene in the gas phase significantly exceeds the value predicted by the cold-gas steady-state chemical models and sublimation of C₂H₂ from the surfaces of interstellar ices could be the main source of this discrepancy (Lahuis & van Dishoeck 2000). This explanation implies that acetylene should occur in low-temperature interstellar ices. Pure acetylene ices were found on Titan, the largest moon of Saturn (Singh et al. 2016). Meanwhile, commonly, acetylene should exist as a component of ices based on highly abundant interstellar molecules (Knez et al. 2012). Water (H₂O) is known to be one of the most widespread molecules in the interstellar medium and water dominant ices are rather abundant in dark clouds (Gibb et al. 2004; Knez et al. 2005; Boogert et al. 2011) and protostellar envelopes (Öberg et al. 2011). C₂H₂ content (relative to H₂O) was determined to be 0.1–1.0 per cent in cometary comae (Mumma et al. 2003) and ca. 2 per cent in interstellar ices (Sonnentrucker, González-Alfonso & Neufeld 2007). Thus, acetylene is a minor but not negligible component of the water-based extraterrestrial ices (Cuylle et al. 2014).

Interstellar ices are continuously exposed to various types of radiation, including electrons, positive ions, and high-energy photons. This impact results in complex radiation-induced chemistry, which particularly leads to formation of new interstellar molecules (Herbst 2017). C₂H₂ is considered to be a starting point and a chain linker in the formation of various cyanopolyynes (Didriche & Herman 2010) and a building block for PAH formation in the interstellar medium (Frenklach & Feigelson 1989; Cherchneff et al. 1992). Astrochemistry of acetylene attracts considerable interest for several decades, which stimulated extensive laboratory studies on the radiation-induced processes in acetylene-containing ices. These systems include pure acetylene ices (Floyd, Prince & Duley 1973; Kaiser & Roessler 1998; Strazzulla et al. 2002; Compagnini et al. 2009; Zhou et al. 2010; Cuylle et al. 2014; Ryazantsev, Zasimov & Feldman 2018), acetylene–carbon monoxide mixed ices (Zhou et al. 2008), acetylene in solid nitrogen media (Wu et al. 2014), and acetylene in water dominant ices (Hudson & Moore 1997; Moore & Hudson 1998; Wu et al. 2002; Hudson & Moore 2003; Hudson & Loeffler 2013; Cuylle et al. 2014).

The principal products detected after radiolysis of the C₂H₂–H₂O ices with a 0.8 MeV proton beam at T < 20 K are C₂H₆, C₂H₄, CH₄, CH₃OH, C₂H₅OH, H₂CO, CH₃COH, CO, and CO₂ (Hudson & Moore 1997; Moore & Hudson 1998), whereas photolysis of this system with the VUV radiation at T < 15 K resulted in formation of CO, CO₂, H₂CO, CH₃COH, and CH₄ (Wu et al. 2002; Cuylle et al. 2014). Further experiments on photolysis and radiolysis of mixed acetylene–water ices (Hudson & Moore 2003) revealed formation of vinyl alcohol (CH₂CHOH) and ketene (H₂CCO). The latter astrochemically important species can be formed either from dehydrogenation of vinyl alcohol (Hawkins & Andrews 1983) or through a free-radical mechanism implying reaction of oxygen atom with acetylene (Haller & Pimentel 1962; Hudson & Loeffler 2013; Bergner, Öberg & Rajappan 2019). It is worth noting that ketene may be a precursor of complex organic molecules (such as CH₃COOH). In addition, the process of ketene formation is supposed to be an intermediate step in the solid-phase oxidation of C₂H₂ to CO in mixed acetylene–water ices (Hudson & Loeffler 2013).

Generally speaking, the laboratory studies on astrophysically relevant ices containing acetylene (and, particularly, acetylene–water mixed ices) have demonstrated diverse and efficient radiation-induced chemistry of these systems, which may lead to formation of a wide variety of new organic molecules in the interstellar medium. However, despite the recent progress in this field, detailed reaction mechanisms lying behind the observed transformations are not well understood. From mechanistic point of view, one can address two main issues in common laboratory simulation of the radiation chemistry of interstellar ices monitored by Fourier-transform infrared (FTIR) spectroscopy. First, there is a lack of direct spectroscopic identification of primary intermediates (due to poor spectral resolution, strong absorptions of ice medium, and typically high irradiation doses). Second, the role of ice medium in the energy transfer and secondary reactions remains unclear. Both these issues can be resolved using matrix isolation as a complementary method to direct simulations of ice chemistry (Feldman et al. 2016). This approach was recently applied in our group for studying the radiation-induced transformations of several small astrochemically important molecules such as H₂O, CO₂, HCOOH, HCN, CH₃OH, CH₃CN, C₂H₂, С₂H₄, C₂H₆ (Ryazantsev & Feldman 2015a, b; Feldman et al. 2016; Kameneva, Tyurin & Feldman 2016a; Saenko & Feldman 2016; Kameneva, Volosatova & Feldman 2017a; Ryazantsev et al. 2018) under the action of X-rays. Regarding the application of X-ray irradiation for simulation of the processes occurring in astrophysical conditions, we have to comment that the energy spectrum of photons and even the radiation type are not crucial as we are focusing on the qualitative description of basic chemical mechanisms. From chemical point of view, we believe that the obtained results may be relevant to the processes induced by high-energy photons, electrons, and protons over a wide energy range since the chemical effects are mainly induced by low-energy secondary electrons (>10³ secondary electrons are produced per primary photon, electron, or ion). Indeed, no qualitative difference was found between the radiation-induced transformation of matrix isolated molecules induced by fast electrons and X-rays (Feldman 1999; Feldman et al. 2016).

Although the noble gas matrices used in this approach are not directly relevant to interstellar ices, they may provide important mechanistic insights in different aspects. High structural informativity and sensitivity are well-known features of the matrix isolation spectroscopy, so this approach was used for identification of astrochemically relevant intermediates for several decades (Allamandola 1987; Zack & Maier 2014). Meanwhile, somewhat less common and even more important for the ice simulations, the matrix isolation approach may be very helpful for elucidation of the details of energy transfer and primary chemical steps of the radiation-induced reactions in solids. Indeed, the ionizing radiation (in contrast to the UV light) is primarily absorbed non-selectively by an icy medium, while the dopant molecules (like acetylene) are activated mainly due to positive hole and excitation transfer. It was found (Feldman 1999) that the energy transfer was extremely efficient in solid noble gas matrices, which was not the case for water-based and other molecular ices. Thus, in contrast to molecular ice simulations, in matrix isolation one can follow the step-by-step transformations of the studied molecules over a wide range of conversion degree (up to >90 per cent) using moderate irradiation doses in the laboratory time-scale (see e.g. Saenko & Feldman 2016; Kameneva et al. 2017a; Ryazantsev et al. 2018). Furthermore, application of noble gas matrices with different ionization energy (IE) and polarizability makes it possible to discriminate between neutral and ionic channels and to probe the environment effect on the primary ionic transformations (Feldman 1999; Feldman et al. 2000a, b; Feldman et al. 2016).

Considering the application of matrix isolation to mechanistic simulations of the radiation-induced transformations in mixed interstellar ices one can come to an idea of using isolated 1:1 intermolecular complexes as a starting point. Indeed, matrix isolation is known to be a powerful method for stabilization and characterization of various weak intermolecular complexes (Young 2013; Khriachtchev 2015). Investigation of the radiation chemistry of such complexes provides new opportunities for elucidation of the elementary steps of transformations in complex ices. In this way, we have shown that the H₂O⋅ ⋅ ⋅CO₂ complex could be the precursor of HOCO radical, one of the fundamentally important species in astrochemical prebiotic evolution (Ryazantsev & Feldman 2015a). Further application of this approach to the radiation chemistry of HCN⋅ ⋅ ⋅CO₂ and HCN⋅ ⋅ ⋅CO complexes (Kameneva et al. 2016b; Kameneva, Tyurin & Feldman 2017b) revealed significant effect of relatively inert carbon oxide molecules on the kinetics of HCN transformations, but did not show formation of the species including the fragments of both components. Thus, the role of complexation in the formation of new molecules in rigid ices remains an open question.

In this work, we present the results of FTIR matrix isolation studies of the transformations of the C₂H₂⋅ ⋅ ⋅H₂O complex under X-ray irradiation at 5 K and discuss their possible astrochemical implications.

2 EXPERIMENTAL DETAILS

Acetylene (¹²C₂H₂, 99.6 per cent SIAD; ¹³C₂H₂, 99.6 per cent, 99 per cent at ¹³C, Aldrich), water (H₂O, double distilled), nitrous oxide (N₂O, 99.8 per cent, Fluka), argon (Ar, 99.9995 per cent, Voessen), krypton (Kr, 99.9998 per cent, Akela-N), and xenon (Xe, 99.9994 per cent, Medxenon) were used without further purification. Gaseous mixtures were prepared by standard manometric technique. ¹²C₂H₂/H₂O/Ng (typical ratio – 1:1:1000, Ng = Ar, Kr, Xe) mixtures were used in the main set of experiments. Complimentary experiments were carried out with ¹³C₂H₂/H₂O/Ar and ¹²C₂H₂/N₂O/Ar 1:1:1000 mixtures.

The experiments were performed using an original closed cycle helium cryostat based on an SHI RDK-101E cryocooler (detailed description of the experimental set-up can be found elsewhere, Feldman 2014). The deposition line was saturated with water vapour through several filling-keeping-evacuation cycles prior to the mixture deposition. This procedure prevents H₂O loss due to absorption on to glassy walls. Prepared mixtures were slowly deposited on to a cooled KBr substrate (flow density ca. 2 mmol h⁻¹ per 1 cm² of the substrate). The sample temperature was controlled using a Lakeshore 325 temperature controller connected to the Cernox-type sensor and resistive heater. The deposition temperature was adjusted specifically for each matrix to obtain less scattering icy films containing sufficient amount of the 1:1 C₂H₂⋅ ⋅ ⋅H₂O complexes; these values were typically about 18, 25, and 29 K for Ar, Kr, and Xe matrices, respectively. Thickness and composition of the depositing matrices were monitored by recording the FTIR spectra during the deposition up to achieving the sample thickness of 80–120 µm (as judged by interference pattern in a spectrum).

The obtained matrix samples were slowly cooled down to 5 K (minimal attainable temperature) and then irradiated with X-rays through a 45-µm thick aluminium foil window mounted in the cryostat. Irradiation was performed using a 5BKhV-6(W) tube with a tungsten anode (45 kVp, anode current 80 mA, effective X-ray energy ca. 20 keV). The irradiation time varied from 3 to 180 min, so the samples were X-irradiated up to different absorbed doses (from 1 to 100 kGy, typically 5 points in each experiment). The absorbed dose rate in matrix samples was estimated in our earlier work (Ryazantsev et al. 2018) by using data from calibrated Fricke dosimeter. It is worth noting that the absorbed dose rate is determined by the mass absorption coefficients of the matrix material (µ_en/ρ), which vary considerably in the photon energy range used in this study (e.g. µ_en/ρ = 8.1, 35.1, and 24.7 cm² g⁻¹ at the photon energy of 20 keV for Ar, Kr, and Xe, respectively). One should also bear in mind that the precise determination of absolute dose absorbed in matrix samples is problematic because X-ray spectrum is, in fact, continuous and dose-depth distribution could be significantly non-uniform (particularly in Kr and Xe). In order to avoid the related uncertainty, we used invariant coordinates (normalized concentration of product versus reagent conversion degree) for the comparative kinetic analysis of the production of various new species (Ryazantsev & Feldman 2015b; Saenko & Feldman 2016).

The FTIR spectra of the X-irradiated samples were recorded at 5 K in the 4000–500 cm⁻¹ range using a Bruker Tensor II spectrometer equipped with a cooled MCT detector (resolution of 1 cm⁻¹, averaging by 144 scans). The normalized concentrations of isolated species in a given matrix sample were obtained by scaling the absorption intensity of a selected spectral band to the corresponding maximum value achieved in the experiment with this sample. Relative uncertainty of the integrated infrared absorption values was estimated as 3 per cent, which gives relative error of 4.3 per cent for the normalized concentration values, as determined using the variance formula (Ku 1966).

3 RESULTS

The FTIR spectra of deposited ¹²C₂H₂/H₂O/Ng samples (Ng = Ar, Kr, Xe) demonstrate several absorption bands of isolated C₂H₂ and H₂O molecules as well as weak bands corresponding to their associates – (C₂H₂)_n and (H₂O)_n (Ryazantsev & Feldman 2015a; Ryazantsev et al. 2018). In addition to these absorptions, new features were found (see Fig. 1), which should be attributed to the C₂H₂⋅ ⋅ ⋅H₂O complex. The C₂H₂⋅ ⋅ ⋅H₂O complex was studied previously in argon and neon matrices (Engdahl & Nelander 1983; Andersen et al. 2017) as well as in the gas phase (Peterson & Klemperer 1984; Block et al. 1992; Rezaei, Moazzen-Ahmadi & McKellar 2012; Didriche & Földes 2013) and characterized by ab initio theory (Frisch, Pople & Del Bene 1983; Block et al. 1992; Tzeli, Mavridis & Xantheas 2000). The absorption bands of the C₂H₂⋅ ⋅ ⋅H₂O in Ar matrix were identified on the basis of available literature data (Engdahl & Nelander 1983). These features characterized by specific complexation-induced shifts of the band positions (see Table 1) were found in the spectral regions of the ν₃(C₂H₂) and ν₅(C₂H₂) vibrations. In Kr and Xe matrices, we also identified the presence of the C₂H₂⋅ ⋅ ⋅H₂O complexes manifested by new absorptions with similar complexation-induced shifts (see Table 1 and Fig. S1 of Supplementary Material). This spectral interpretation was also supported by the complimentary experiments with the ¹³C₂H₂/H₂O/Ar mixtures, where the ¹³C₂H₂⋅ ⋅ ⋅H₂O complex was detected in addition to ¹³C₂H₂ and H₂O isolated molecules (as well as their associates). In experiments with C₂H₂/H₂O/Ar matrices, the absorptions of the C₂H₂⋅ ⋅ ⋅H₂O complexes in the spectral regions of water stretching and bending vibrations were detected at 3640.4 and 1593.0 cm⁻¹, respectively. This observation is in full agreement with literature data (Engdahl & Nelander 1983). Additional absorption band with maximum at 3721.5 cm⁻¹ was tentatively assigned to the ν₃(H₂O) vibration of the C₂H₂⋅ ⋅ ⋅H₂O complex. In Kr and Xe matrices, we tentatively assigned some weak features (Kr: 3706.7 and 1591.4 cm⁻¹, Xe: 3699.2 and 1590.8 cm⁻¹) to the ν₃(H₂O) and ν₂(H₂O) vibrations of the C₂H₂⋅ ⋅ ⋅H₂O complexes, respectively. Meanwhile, we failed to distinguish the corresponding absorptions in the ν₁(H₂O) spectral region (see Fig. S2 of Supplementary Material). We should emphasize that, according to our observations in Ar matrix, the ν₁(H₂O) absorption of the C₂H₂⋅ ⋅ ⋅H₂O complex is much lower in intensity than the ν₂(H₂O) and ν₃(C₂H₂) bands. Moreover, possible overlapping with multiple rotational-structure bands of the matrix-isolated H₂O molecules may complicate the unequivocal identification of some additional absorptions in the spectral region of water vibrations. Despite this lack, we believe that overall data provide solid spectral justification of the formation of C₂H₂⋅ ⋅ ⋅H₂O complexes in the prepared C₂H₂/H₂O/Ng matrices. A strong structureless vibration band corresponding to the ν₃(C₂H₂) mode in the C₂H₂⋅ ⋅ ⋅H₂O complex was used in our experiments to monitor the decomposition kinetics. According to rough estimate, the total amount of the C₂H₂⋅ ⋅ ⋅H₂O complex in our matrix samples was ca. 5–20 per cent of the amount of C₂H₂ monomer. This estimate is based on integrated intensities of the ν₃(C₂H₂) bands for complex and monomer taking into account the corresponding absorption coefficients (values calculated at MP2/aug-cc-pVTZ level are 254 km mol⁻¹ and 96 km mol⁻¹, respectively, as taken from Tzeli et al. 2000). It is worth noting that complexation may lead to changes in the intensity of specific vibrational bands. The available computational data (Tzeli et al. 2000) show that in the C₂H₂⋅ ⋅ ⋅H₂O complex, the ν₃(C₂H₂) absorption intensity increases by a factor of ∼3 (as compared to monomer), whereas the ν₅(C₂H₂) absorption intensity slightly decreases. Indeed, such effect was observed experimentally and it can be clearly seen from Fig. 1. In the case of C₂H₂ monomer, the ν₅(C₂H₂) absorption is more intense than the ν₃(C₂H₂) band, whereas the opposite is true for the C₂H₂⋅ ⋅ ⋅H₂O complex.

Figure 1.

Fragments of the FTIR spectra of the deposited ¹²C₂H₂/H₂O/Ar 1:1:1000 (upper trace) and ¹²C₂H₂/Ar 1:1000 (lower trace) matrices. Spectra were recorded at 5 K. Absorptions due to acetylene associates ((¹²C₂H₂)_n) are marked with asterisks. In the case of the ¹²C₂H₂/Ar sample, a weak absorption of the ¹²C₂H₂⋅ ⋅ ⋅H₂O complex at 3240 cm⁻¹ appears due to the presence of impurity water.

Table 1.

Absorption maxima of the C₂H₂⋅ ⋅ ⋅H₂O complex and C₂H₂ isolated molecules and the corresponding complexation-induced shifts (cm⁻¹) in Ar, Kr, and Xe matrices. Additional features due to site splitting are provided in italic. Unperturbed positions of ν₃(C₂H₂) band (ν₃⁰) calculated as described elsewhere (Rutkowski et al. 2002) and the corresponding complexation-induced shifts in studied matrices are presented in parentheses.

12C2H2									13C2H2
Ar			Kr			Xe			Ar
Monomer	Complex	Shift	Monomer	Complex	Shift	Monomer	Complex	Shift	Monomer	Complex	Shift
3302.8a			3293.2a			3280.4a
3288.9a			3280.0a			3266.8a
(3294.6)	3240.1	(−54.5)	(3287.5)	3234.0	(−53.5)	(3274.5)	3229.3	(−45.2)	3285.0	3228.4	−56.6
	792.7			790.1			783.2			790.2
736.9	785.9	+49.0	732.6	782.4	+49.8	727.6	777.1	+49.5	735.0	783.1	+48.1
734.6sh	783.4			779.2			773.7		732.9sh	780.7
731.1	780.2			774.8			770.7		729.1	777.7

Note. sh—shoulder.

a

FR(ν₃, ν₂₊₄₊₅).

The radiolysis of deposited C₂H₂/H₂O/Ng samples results in decomposition of both C₂H₂ and H₂O isolated molecules as well as the C₂H₂⋅ ⋅ ⋅H₂O complex and formation of new species. Comparison of the decay kinetics for C₂H₂ monomer and its complex with water clearly demonstrates that complex decomposes faster than isolated acetylene molecules (the case of Ar matrix is presented in Fig. 2). Such effect was observed in all the studied matrices (see Fig. S3 of Supplementary Material). This pronounced sensibilization of acetylene radiolysis in complex could be generally explained by involvement of new channels of the radiation-induced transformations due to complexation. It should be noted that similar-type effect was observed previously for HCN decomposition in the HCN⋅ ⋅ ⋅CO complex (Kameneva et al. 2017b).

Figure 2.

Kinetics of the radiation-induced decay of isolated C₂H₂ and H₂O molecules and C₂H₂⋅ ⋅ ⋅H₂O complexes in an Ar matrix.

Radiolysis of matrix-isolated acetylene and water molecules was investigated in our previous works (Ryazantsev & Feldman 2015a; Ryazantsev et al. 2018). It was found that both these compounds underwent the radiation-induced dehydrogenation resulting in formation of C₂H radicals and C₂ species from acetylene, and OH radicals and O atoms from water. We have detected C₂H and OH radicals as well as C₂/Xe species in the FTIR spectra of X-irradiated C₂H₂/H₂O/Ng matrices and these products apparently result from the decomposition of C₂H₂ and H₂O isolated molecules. The traces of C₂H₃ and C₄H radicals (Forney, Jacox & Thompson 1995) were also observed in all studied matrices due to the radiolysis of (C₂H₂)₂ associates. Moreover, a weak absorption due to the C₂H^– species (Andrews et al. 1999) was additionally detected in the irradiated C₂H₂/H₂O/Ar samples. In addition to the absorptions of the above-mentioned species, the FTIR spectra of X-irradiated C₂H₂/H₂O/Ng matrices reveal the appearance of several new features, which originate from the radiolysis products of the C₂H₂⋅ ⋅ ⋅H₂O complex (as illustrated by Fig. 3 for the C₂H₂/H₂O/Ar matrix). These absorptions were attributed to ketene (H₂CCO), ketenyl radical (HCCO), carbon monoxide (CO), and methane (CH₄). The band positions are summarized in Table 2 and vibrational assignment relied mostly on literature data (references are given in Table 2). Vinyl alcohol (CH₂CHOH) was also detected in Ar matrices, but it was not found in Kr and Xe. We should also emphasize that, in addition to a characteristic infrared absorption of matrix-isolated ketene (Haller & Pimentel 1962; Moore & Pimentel 1963), we observed some blueshifted satellite appearing in all the studied matrices. We suggest that these satellite absorptions could be attributed to a perturbed ketene molecule, presumably due to interaction with molecular hydrogen trapped in the same matrix cage (this species is denoted below as H₂CCO–H₂). The arguments in favour of this assumption will be presented in the Discussion section. It is worth noting that we did not observe any evidence for the formation of acetaldehyde (or any related carbonyl species) in the irradiated C₂H₂/H₂O/Ng (Ng = Ar, Kr, Xe) samples (see Fig. S4 of Supplementary Material). Thus, acetaldehyde is either not formed or not stabilized under these experimental conditions.

Figure 3.

Fragments of the difference FTIR spectra showing the result of X-irradiation of the ¹²C₂H₂/H₂O/Ar 1:1:1000 matrices. Note that the spectra in the right-hand panel are multiplied by the factor 5.

Table 2.

Absorption maxima (cm⁻¹) of the radiolysis products of matrix-isolated C₂H₂⋅ ⋅ ⋅H₂O complex. Numbers in italic indicate additional features due to site splitting.

		12C2H2⋅ ⋅ ⋅H2O			13C2H2⋅ ⋅ ⋅H2O
Species	Assignment	Ar	Kr	Xe	Ar
H2CCOa,b	ν1(νs CH2)	3063.1	–	–	3057.2
	ν2(ν C = O)	2142.4	2139.7	2136.9	2080.2
	ν6(ρ CH2 ip)	975.6	–	–	–
HCCOc	ν2(νas CCO)	2023.5	2020.5	2016.1	1965.3
		2019.6	2017.1	2013.7	1961.7
COd	ν1(ν C≡O)	2137.8	2135.6	2133.5	2090.5
CH4d	ν4(δd CH4)	1305.3	1302.4	1300.4	1296.8
H2CCO–H2	ν1(νs CH2)	3072.2	–	–	3067.8
	ν2(ν C = O)	2149.1	2145.6	2139.9	2087.1
		–	2143.6	–	–
	ν6(ρ CH2 ip)	971.4	–	–	–
CH2CHOHe	ν9(mix)	1078.9	–	–	1070.6
	ν13(H2C = C oop)	813.6	–	–	805.4

Notes. ^aHaller & Pimentel (1962).

b

Moore & Pimentel (1963).

c

Forney et al. (1995).

d

Saenko & Feldman (2016).

e

Hawkins & Andrews (1983).

Experiments with ¹³C₂H₂/H₂O/Ar 1:1:1000 samples confirmed the identification of observed radiolysis products for the C₂H₂⋅ ⋅ ⋅H₂O complex. It was found that radiolysis of the ¹³C₂H₂⋅ ⋅ ⋅H₂O complex resulted in formation H₂¹³C¹³CO, H¹³C¹³CO, ¹³CO, ¹³CH₄, H₂¹³C¹³CO–H₂ pair, and ¹³CH₂¹³CHOH (see Table 2). The fragments of the FTIR spectra illustrating formation of methane from the two isotopologues are presented in Fig. 4.

Figure 4.

Fragments of the difference FTIR spectra showing the result of X-irradiation of the ¹²C₂H₂/H₂O/Ar 1:1:1000 (lower trace) and ¹³C₂H₂/H₂O/Ar 1:1:1000 (upper trace) matrices. Note that spectra in the right-hand panel are multiplied by the factor 10.

In order to understand the mechanisms of radiation-induced transformations of the C₂H₂⋅ ⋅ ⋅H₂O complex, we have investigated the build-up profiles of different products. These profiles are qualitatively similar in Ar, Kr, and Xe matrices; the case of Ar matrix is presented in Fig. 5 (the profiles for the Kr and Xe matrices are presented in Figs S5 and S6 of Supplementary Material). Prolonged radiolysis of the C₂H₂/H₂O/Ar samples results in almost complete decomposition of the C₂H₂⋅ ⋅ ⋅H₂O complex (maximum conversion ∼97 per cent). As can be seen from Fig. 5, growth of the bands assigned to H₂CCO–H₂ (perturbed ketene) and CH₂CHOH is followed by their subsequent radiation-induced decay. It should be noted that the concentration of H₂CCO–H₂ and CH₂CHOH reach their maxima at ca. 80 and 60 per cent of the parent complex conversion, respectively. The accumulation curves of H₂CCO isolated molecules and HCCO radicals flatten out after decomposition of ca. 80 per cent of the C₂H₂⋅ ⋅ ⋅H₂O complex. On the other hand, the concentration of CH₄ and CO molecules increases even after almost complete decomposition of the parent complex. Moreover, an induction period is observed for accumulation of CH₄ and CO. Such behaviour implies high contribution of the secondary reaction channels in formation of these species. We have estimated the ratio of CO/CH₄ yields using molar absorption coefficients ε(δ_d CH₄) = 31 km mol⁻¹ and ε(ν CO) = 72 km mol⁻¹ (these values were taken from the available computational data obtained at the CCSD/cc-pVTZ level). At the highest conversion of the parent complex, the CO/CH₄ ratio was found to be ca. 4:1, 6:1, and 8:1 in Ar, Kr, and Xe, respectively. These rough estimates definitely show that radiolysis of C₂H₂⋅ ⋅ ⋅H₂O yields larger amount of CO than CH₄ in all the studied matrices. On the other hand, production of CH₄ was found to be more efficient than production of HCCO [ε(ν_as CCO) in HCCO is 740 km mol⁻¹ according to the CCSD/cc-pVTZ calculations]. The CH₄/HCCO ratio was determined to be ca. 10:1, 10:1, and 5:1 for Ar, Kr, and Xe matrices, respectively.

Figure 5.

Build-up profiles of CH₂CHOH, CH₄, CO, H₂CCO–H₂, H₂CCO, and HCCO generated under radiolysis of the C₂H₂/H₂O/Ar 1:1:1000 matrix.

4 DISCUSSION

The radiolysis of C₂H₂/H₂O/Ng samples results in formation of various C₂ species (H₂CCO, HCCO, and CH₂CHOH) and C₁ species (CH₄ and CO). As follows from the build-up curves of the radiolysis products, the ‘synthetic’ products (ketene and its perturbed analogue, vinyl alcohol) predominate at low absorbed doses, whereas the absorptions of degradation products (methane and carbon monoxide) become most prominent at high conversion degree. This result corresponds to a general scheme implying formation of new bonds followed by the CC bond cleavage.

Ketene, a primary ‘synthetic’ product, is clearly observed even at very low absorbed doses and it definitely plays a very important role in the chemical evolution of the studied system. Remarkably, as shown above (Fig. 3), the major part of ketene is formed in the perturbed state. It is known (Haller & Pimentel 1962; Hudson & Loeffler 2013; Bergner et al. 2019) that ketene could be formed by reaction of oxygen atoms with C₂H₂ molecules. The oxygen atoms apparently result from the radiolysis of water molecules in a single-step or two-step radiation-induced dehydrogenation (1) (Ryazantsev & Feldman 2015a).

$$\begin{eqnarray} {{\rm{H}}_2}{\rm{O}}\, \rightsquigarrow \,{\rm{O}} + 2{\rm{H}}/{{\rm{H}}_2};{\rm{O}} + {{\rm{C}}_2}{{\rm{H}}_2} \to {{\rm{H}}_2}{\rm{CCO}} \end{eqnarray}$$

(1)

In the case of C₂H₂⋅ ⋅ ⋅H₂O complex, a two-step dissociation of water with intermediate formation of OH radical would lead to formation of isolated ketene, while the hydrogen atoms would escape the matrix cage. Meanwhile, for a single-step process one should expect the formation of ketene with hydrogen molecule within the same matrix cage (H₂CCO–H₂ pair) through intermolecular oxygen transfer in the C₂H₂⋅ ⋅ ⋅H₂O complex (2). We may note that this process is somewhat similar to an intermolecular oxygen transfer observed in photoinduced conversion of the HCCO⋅ ⋅ ⋅CO complex to the HCC⋅ ⋅ ⋅CO₂ complex in low-temperature matrices (Ryazantsev et al. 2017).

$$\begin{eqnarray} {{\rm{C}}_2}{{\rm{H}}_2} {\cdots} {{\rm{H}}_2}{\rm{O}}\, \rightsquigarrow \,{{\rm{H}}_2}{\rm{CCO}}\!-\!{{\rm{H}}_2} \end{eqnarray}$$

(2)

Thus, from the phenomenological point of view, it is logical to assign the shifted absorptions in the ketene vibration regions (perturbed ketene) to the H₂CCO–H₂ species. It is worth noting that the maxima of these new absorptions are in good agreement with the positions of absorptions observed after photolysis of ethylene oxide (Schriver et al. 2004). In addition, it should be noted that similar blueshifted component of ketene absorption was found after photolysis of the C₂H₄/O₃/Ar system, where vinyl alcohol or the ^•CH₂CH₂O^• biradical were suggested as a ketene precursor (Hawkins & Andrews 1983). In both cases, the authors of the cited works attributed the observed absorption bands to isolated ketene molecules. However, they are clearly different from the isolated ketene bands observed in other studies (Haller & Pimentel 1962; Moore & Pimentel 1963) and the formation of ketene paired with molecular hydrogen (H₂CCO–H₂) is expected under photolysis of isolated ethylene oxide and vinyl alcohol molecules just from atomic balance reasons. Moreover, it is well known that photolysis of H₃C₂OX species results in efficient formation of H₂CCO⋅ ⋅ ⋅HX complexes (Macoas et al. 2004; Guennoun et al. 2005), so the case of X = H would correspond to the H₂CCO⋅ ⋅ ⋅H₂ complex.

The spectroscopic and structural justification of this assignment is less straightforward. According to our preliminary ab initio calculations, an intermolecular H₂CCO⋅ ⋅ ⋅H₂ complex should be almost unbound. The interaction energy found for two structures revealed in calculations (see Table S1 of Supplementary Material) is 0.5–0.6 kcal mol⁻¹ without zero-point vibrational energy correction and it may become close to zero (or even negative) with zero-point vibrational energy correction. The estimated complexation-induced shift of the C = O stretching vibrations in harmonic approximation should be very small (less than 1 cm⁻¹). Thus, the existence of the complexes as true minima at the potential energy surface is questionable and analysis of their possible spectroscopic manifestations requires more sophisticated computations (out of the scope of this work). However, most probably, the experimentally observed differences between absorption maxima of the H₂CCO–H₂ and isolated ketene molecules are not due to complexation itself. We may suggest that the shifts result mainly from distortion of the matrix cage due to trapping of a hydrogen molecule close to ketene, that is, the two observed bands in this region result from perturbed and unperturbed H₂CCO molecules, which are formed from different precursors (C₂H₂⋅ ⋅ ⋅H₂O complex and isolated C₂H₂ molecules, respectively). The different shifts between main absorption bands of the H₂CCO–H₂ and H₂CCO in Ar, Kr, and Xe matrices (6.7, 5.9, and 3.0 cm⁻¹, respectively) are in qualitative agreement with this explanation, because a looser xenon matrix cage should exhibit smaller distortion. To further verify the assignment, we obtained ketene by acetylene oxidation with oxygen atoms produced from radiolysis of N₂O (complimentary experiments with C₂H₂/N₂O/Ar samples) and found that irradiation of this system resulted in formation of both isolated ketene (3063.2, 2145.2, and 974.8 cm⁻¹) and additional features (3073.9, 2151.0, and 973.0 cm⁻¹), which exhibit even larger blue shifts than those found in irradiated C₂H₂/H₂O/Ar samples (see Fig. 6). This result looks reasonable: in the case of C₂H₂/N₂O/Ar samples, the radiolysis may result in trapping of the nitrogen molecule in the vicinity of ketene (H₂CCO–N₂) and the matrix distortion is expected to be larger than that for H₂CCO–H₂. It should be mentioned that prolonged radiolysis of C₂H₂/N₂O/Ar samples results in formation of CH₂N₂ molecules (2095.4 cm⁻¹, Moore & Pimentel 1964), which could be formed due to dissociation of ketene to CH₂ and CO followed by in-cage reaction of methylene with molecular nitrogen. Thus, we suggest that in both systems (C₂H₂/H₂O/Ar and C₂H₂/N₂O/Ar) unperturbed ketene is formed by reaction of oxygen atoms with acetylene, whereas the perturbed ketene bands with different shifts results from the H₂CCO–H₂ and H₂CCO–N₂ species, respectively.

Figure 6.

Fragments of the difference FTIR spectra showing the result of X-radiolysis of the C₂H₂/H₂O/Ar 1:1:1000 (upper trace) and C₂H₂/N₂O/Ar 1:3:1000 (lower trace) samples. Note that the spectra in the right-hand panel are multiplied by the factor of 5.

The ketenyl (HCCO) radical is related to ketene and, most probably, it is formed by hydrogen elimination from H₂CCO (3) (Hudson & Loeffler 2013)

$$\begin{eqnarray} {{\rm{H}}_2}{\rm{CCO}}\, \rightsquigarrow \,{\rm{HCCO}} + {\rm{H}}. \end{eqnarray}$$

(3)

We suggest that radiolysis of H₂CCO–H₂ could result in formation of HCCO–H₂ species and a blueshifted component of the HCCO doublet could belong to HCCO–H₂ (4)

$$\begin{eqnarray} {{\rm{H}}_2}{\rm{CCO}}\!-\!{{\rm{H}}_2}\, \rightsquigarrow \,{\rm{HCCO}}\!-\!{{\rm{H}}_2} + {\rm{H}}. \end{eqnarray}$$

(4)

The absence of a clear induction period for HCCO accumulation could be tentatively explained by the formation of H₂CCO in excited state and its subsequent dehydrogenation without stabilization (see e.g. radiolysis of matrix-isolated CH₃OH, Saenko & Feldman 2016). The limitation of the HCCO accumulation at high absorbed doses probably occurs to its radiation-induced degradation.

The other primary ‘synthetic’ radiolysis product of the C₂H₂⋅ ⋅ ⋅H₂O complex is vinyl alcohol (CH₂CHOH). This species was detected only in an Ar matrix. Although the mechanism of its formation is not fully clear, such a matrix effect generally may imply involvement of ionic channels. Indeed, Ar has the highest IE among the noble gas matrices used in this work (15.8 eV, NIST Chemistry Webbook,¹, Lias et al. 2018) and it is known that the positive hole transfer to guest species in this matrix is highly efficient and substantially exothermic (Feldman 1999). Thus, we may suggest that positive hole transfer results in formation of the ionized complex (radical cation) with an excess energy, which undergoes rapid rearrangement to the vinyl alcohol radical cation followed by its neutralization with electrons:

$$\begin{eqnarray} {\left[ {{{\rm{C}}_2}{{\rm{H}}_2} {\cdots} {{\rm{H}}_2}{\rm{O}}} \right]^{ + \bullet }} \to {\rm{C}}{{\rm{H}}_2}{\rm{CHO}}{{\rm{H}}^{ + \bullet }} \to {\rm{C}}{{\rm{H}}_2}{\rm{CHOH}}. \end{eqnarray}$$

(5)

An interesting result of our experiments is that acetaldehyde (which is a thermodynamically more stable tautomer of vinyl alcohol) is not formed (or not stabilized) upon radiolysis of the matrix-isolated C₂H₂⋅ ⋅ ⋅H₂O complex. This is opposite to the case of the irradiated C₂H₂–H₂O ices, where both CH₃COH and CH₂CHOH are among the principal radiation-induced products (Moore & Hudson 1998; Wu et al. 2002; Hudson & Moore 2003). Non-observation of CH₃COH in our matrix-isolation experiments could be rationalized in two different ways. On the one hand, if we assume that acetaldehyde is not formed, this result would mean that ionized complex [C₂H₂⋅ ⋅ ⋅H₂O]^+• may relax to the CH₂CHOH^+• radical cation, but not to the CH₃COH^+• radical cation. Such hypothetical mechanism is an evidence of selectivity in the radiation-induced chemistry of the system under consideration. On the other hand, we cannot exclude a possibility that CH₃COH is formed in excited state and undergoes rapid decomposition instead of relaxation. Decomposition of CH₃COH* may yield H₂CCO and CH₄ (Schriver et al. 2004), the species, which were detected in our experiments. Meanwhile, we have to note that from the available experimental data, it is difficult to judge on possible involvement of acetaldehyde in the actual reaction pathways.

The induction period in the accumulation of degradation products (CH₄ and CO) indicates high contribution of secondary reactions of the radiolytic species to formation of these products. The probable reaction channel leading to CH₄ production is radiation-induced dissociation of perturbed ketene (H₂CCO–H₂) to CH₂ and CO species and subsequent reaction of methylene with hydrogen molecule (6). It is worth noting that CH₄ is one of the main products of gas-phase ketene photolysis in the presence of hydrogen (Gesser & Steacie 1956). Similar way leads to formation of CH₂N₂ in the above-discussed experiments with C₂H₂/N₂O/Ar samples.

$$\begin{eqnarray} {{\rm{H}}_2}{\rm{CCO}}\!-\!{{\rm{H}}_2}\, \rightsquigarrow \,\left[ {{\rm{CO}} + {\rm{C}}{{\rm{H}}_2}\!-\!{{\rm{H}}_2}} \right] \to {\rm{C}}{{\rm{H}}_4} + {\rm{CO}} \end{eqnarray}$$

(6)

We have to note that the radiation-induced conversion of H₂CCO–H₂ to CH₄ + CO seems to be rather efficient process. As mentioned in the Results section, at high doses the yield of CH₄ exceeds the yield of HCCO several times in all the studied matrices, i.e. the reaction (6) strongly dominates over reaction (4). This observation agrees well with the data on gas-phase photolysis of H₂CCO at different wavelength, where the branching ratio of the H₂CCO → CH₂ + CO and H₂CCO → H + HCCO channels were determined to be from 10:1 (Glass, Kumaran & Michael 2000) to 3:1 (Fockenberg 2005) at 193 nm and ca. 50:1 at 157.6 nm (Lu et al. 2006).

Another possible secondary reaction channel resulting in formation CH₄ and CO is the radiation-induced degradation of vinyl alcohol (7). It should be noted that UV-photolysis of isolated acetaldehyde (the tautomer of vinyl alcohol) results in quite efficient formation of CH₄ and CO (Védova & Sala 1991).

$$\begin{eqnarray} {\rm{C}}{{\rm{H}}_2}{\rm{CHOH}}\, \rightsquigarrow \,{\rm{C}}{{\rm{H}}_4} + {\rm{CO}} \end{eqnarray}$$

(7)

To recall, the excited acetaldehyde molecules, if they are intermediately formed (see above), could provide an important source of CH₄ + CO. Moreover, both CH₂CHOH and CH₃COH could decompose to H₂CCO and H₂ (Hawkins & Andrews 1983, Schriver et al. 2004) and replenish the amount of the H₂CCO–H₂ consumed in the course of radiolysis:

$$\begin{eqnarray} {\rm{C}}{{\rm{H}}_2}{\rm{CHOH}}\, \rightsquigarrow \,{{\rm{H}}_2}{\rm{CCO}}\!-\!{{\rm{H}}_2} \end{eqnarray}$$

(8)

Reactions (4), (6) and (7), (8) should be responsible for the observed decrease in concentration of H₂CCO–H₂ and vinyl alcohol at high absorbed doses (Fig. 5). The observed lack in balance between CH₄ and CO (see above) can be due to additional formation of CO from isolated ketene molecules due to reaction (9A). Another reason for higher amount of CO (in comparison with CH₄) at high absorbed doses may be concerned with methane decomposition under radiolysis, whereas CO is known to be extremely stable under X-ray irradiation in matrices (Kameneva et al. 2017b). Indeed, we observed the formation of methyl radicals in the trace amounts at high absorbed doses.

It should be noted that the methylene–carbon monoxide pair produced from isolated H₂CCO molecules under radiolysis has a high probability to stay within the same matrix cage, which may lead to ketene recovery due to the matrix cage effect (reaction 9B). This effect could explain the apparent ‘radiation resistance’ of isolated H₂CCO molecules (as well as H₂CCO–H₂). However, if methylene or CO leaves the matrix cage, the recombination with CO molecule would become impossible. The isolated CH₂ species are hardly detectable by FTIR spectroscopy in the noble gas matrices (Sander, Bucher & Wierlacher 1993).

$$\begin{eqnarray} {{\rm{H}}_2}{\rm{CCO}}\, \rightsquigarrow \,\left[ {{\rm{C}}{{\rm{H}}_2} + {\rm{CO}}} \right] \to {\rm{C}}{{\rm{H}}_2} + {\rm{CO}} \end{eqnarray}$$

(9A)

$$\begin{eqnarray} {{\rm{H}}_2}{\rm{CCO}}\, \rightsquigarrow \,\left[ {{\rm{C}}{{\rm{H}}_2} + {\rm{CO}}} \right] \to {{\rm{H}}_2}{\rm{CCO}} \end{eqnarray}$$

(9B)

To sum up, one may consider formation of CH₄ and CO as ultimate fate of an isolated C₂H₂⋅ ⋅ ⋅H₂O complex irradiated to high absorbed dose in a very rigid environment. The whole process may be described as ‘oxidative-reductive disproportionation’ of acetylene formally leading to the cleavage of a very strong triple C≡C bond. However, this description actually cannot be applied rigorously to any real system because of possible migration of intermediates (atoms and small molecular species) escaping the matrix cage. Migration of hydrogen atoms leads to low-temperature hydrogenation reactions, which could result in different products reacting with both acetylene and products of its radiolysis (e.g. C₂H₃, C₂H₄, C₂H₆, HCO, H₂CO, CH₃OH, etc.). Migration of oxygen atoms may yield different oxidation products. Finally, migration of methylene (not confirmed yet) may provide rich and diverse synthetic chemistry.

5 ASTROCHEMICAL IMPLICATIONS

This study has directly demonstrated the formation of a number of key astrochemically important species as a result of radiation-induced evolution of isolated C₂H₂⋅ ⋅ ⋅H₂O complex in rigid inert environment at very low temperatures. As mentioned in the Introduction, a large number of products in different amounts were found after radiolysis of mixed acetylene–water ice at high absorbed doses and most of them probably originate from complex sequence of chemical reactions. Meanwhile, our experiments make it possible to determine principal primary products appearing at the initial stages under irradiation at 5 K. In particular, ketene formation from the reaction of oxygen atom with acetylene molecule occurring in the parent complex could be one of the major sources of interstellar ketene and recently detected ketenyl radical (Agundez, Cernicharo & Guélin 2015) in regions with high amount of water dominated ices containing acetylene. Ketene is considered as an important intermediate in astrochemistry that could be readily converted into more complex organic molecules in space, such as CH₃COH, CH₃COOH, CH₃CONH₂ (Hudson & Loeffler 2013). In addition, formation of vinyl alcohol detected in an Ar matrix indicates that acetylene–water systems may contribute to production of this species and its tautomer (acetaldehyde) in the interstellar medium.

The main secondary products predominating at high absorbed doses are the C₁ species (CO and CH₄), which looks somewhat surprising. As shown in our study, this is an interesting example of carbene (methylene) chemistry, which could occur in the interstellar ices. We have found that ketene readily dissociated to CO and CH₂ under radiolysis. Considering possible fate of methylene in the acetylene–water mixed ices, we may suggest that it could react with surrounding simple molecules to produce different species, e.g. methanol from water and methylacetylene from acetylene. The former reaction could be an additional source of methanol in mixed acetylene–water ices. In a more general context, the reactions of methylene produced from ketene may provide an important source of new chemical compounds in mixed complex ices containing water and acetylene as well as other components. Thus, there is a challenge for experimental and theoretical simulation of the methylene mediated astrochemistry.

6 CONCLUSIONS AND OUTLOOK

In this work, we have demonstrated that the study of radiation chemistry of weak intermolecular complexes using a matrix isolation approach can provide a powerful tool for modelling the radiation-induced processes occurring in mixed interstellar ices and elucidation of their detailed mechanisms. The investigation of the radiation-induced transformations of the C₂H₂⋅ ⋅ ⋅H₂O complex isolated in noble gas matrices has revealed that complexation with water results in sensibilization of the acetylene decomposition under irradiation with X-rays and appearance of new channels of the radiation-induced transformations. The main primary products of radiolysis of this complex in matrices at 5 K determined by FTIR spectroscopy are H₂CCO, H₂CCO–H₂, and vinyl alcohol (the last species was found only in an Ar matrix). The increasing contribution of other species (HCCO, CO, and CH₄) observed at higher absorbed doses should be attributed to secondary reactions. Generally speaking, it implies that the first key step is synthesis and decomposition of the products becomes predominating at high conversion degree, when the parent complex is essentially consumed.

Judging from our results, we may suggest that C₂H₂/H₂O ices could be an important source of ketene and ketenyl radical in the interstellar medium, which readily gives methylene (CH₂) under irradiation. The reactions of methylene with components of interstellar ices could play a significant role in the extraterrestrial evolution of matter leading to formation of complex molecules. Also, analysis of the accumulation profile of C₁ species (CO and CH₄) and ratio of ketene dissociation channels under radiolysis allows us to suggest that a two-step C≡C bond cleavage in acetylene could be significant at high doses typical for the astrochemical time-scale. An important finding is that the formation of methane is probably related to the existence of a H₂CCO–H₂ pair, which is stabilized in a rigid matrix due to cage effect.

Regarding more general implications, we may note that actually the radiation-chemical evolution starting from a single isolated complex may account for the formation of many products observed in mixed systems. This approach can be applied to unravel the mechanistic issues for other complex ices of astrochemical interest.

SUPPORTING INFORMATION

Figure S1. Fragments of FTIR spectra of the deposited C₂H₂/H₂O/Ng and C₂H₂/Ng matrices (Ng = Kr, Xe) showing the spectral regions of the acetylene vibrational modes.

Figure S2. Fragments of FTIR spectra of the deposited C₂H₂/H₂O/Ng matrices (Ng = Ar, Kr, Xe) showing the spectral regions of the water vibrational modes.

Figure S3. Kinetics of radiation-induced decay of C₂H₂ and H₂O isolated molecules and C₂H₂⋅ ⋅ ⋅H₂O complexes in Kr and Xe matrices.

Figure S4. Fragments of the difference FTIR spectra (2025–1575 cm⁻¹ region) showing the result of X-irradiation of the C₂H₂/H₂O/Ng matrices (Ng = Ar, Kr, Xe).

Figure S5. Build-up profiles of CH₄, CO, H₂CCO, HCCO, and H₂CCO–H₂ generated under radiolysis of the C₂H₂/H₂O/Kr matrices.

Figure S6. Build-up profiles of CH₄, CO, H₂CCO, HCCO, and H₂CCO–H₂ generated under radiolysis of the C₂H₂/H₂O/Xe matrices.

Table S1. Computational data on the H₂CCO⋅ ⋅ ⋅H₂ intermolecular complex.

Please note: Oxford University Press is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

ACKNOWLEDGEMENTS

We are thankful to I.V. Tyulpina for her contribution to the experimental procedure. The work was supported by the Russian Foundation for Basic Research (project no. 19-03-00579a).

Footnotes

1

NIST Computational Chemistry Comparison and Benchmark Database, NIST Standard Reference Database Number 101, Release 20, August 2019, Editor: Russell D. Johnson III. Available at: http://cccbdb.nist.gov.