http://pubs.acs.org/journal/aesccq Article Non-energetic Formation of Ethanol via CCH Reaction with Interstellar H2O Ices. A Computational Chemistry Study Jessica Perrero, Joan Enrique-Romero,* Berta Martínez-Bachs, Cecilia Ceccarelli, Nadia Balucani, Piero Ugliengo, and Albert Rimola* Cite This: ACS Earth Space Chem. 2022, 6, 496−511 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. Downloaded via UNIV AUTONOMA DE BARCELONA on May 9, 2022 at 14:51:19 (UTC). ABSTRACT: Ethanol (CH3CH2OH) is a relatively common molecule, often found in star-forming regions. Recent studies suggest that it could be a parent molecule of several so-called interstellar complex organic molecules (iCOMs). However, the formation route of this species remains under debate. In the present work, we study the formation of ethanol through the reaction of CCH with one H2O molecule belonging to the ice as a test case to investigate the viability of chemical reactions based on a “radical + ice component” scheme as an alternative mechanism for the synthesis of iCOMs, beyond the usual radical− radical coupling. This has been done by means of DFT calculations adopting two clusters of 18 and 33 water molecules as ice models. Results indicate that CH3CH2OH can potentially be formed by this proposed reaction mechanism. The reaction of CCH with H2O on the water ice clusters can be barrierless (because of the help of boundary icy water molecules acting as proton-transfer assistants), leading to the formation of vinyl alcohol precursors (H2CCOH and CHCHOH). Subsequent hydrogenation of vinyl alcohol yielding ethanol is the only step presenting a low activation energy barrier. We finally discuss the astrophysical implications of these findings. KEYWORDS: interstellar medium, astrochemistry, DFT, iCOMs, grains ■ INTRODUCTION Interstellar complex organic molecules (iCOMs) are com- occurring on grain surfaces,25,26 although other paradigms have been postulated, like those based on the condensation of pounds between 6 and 12/13 atoms, in which at least one is atomic C,27,28 excited O atom insertion,29,30 or the formation carbon, conferring the organic character.1−3 These molecules of HCO on surfaces as a parent precursor of other iCOMs.31,32 are important because (i) they can be considered as the Both prevailing reaction models have the same initial step: simplest organic compounds that are synthesized in space formation of hydrogenated species (e.g., CH3OH, NH3, H2CO, (hence representing the dawn of organic chemistry) and (ii) and CH4) by H-addition on icy grain surfaces in the cold they are the precursors of more complex organic molecules, prestellar phase. In the gas-phase paradigm, the process follows which can be of biological relevance, such as amino acids, with the desorption of these species, either thermally when the nucleobases, and sugars. Indeed, there is robust evidence that grain surface reaches ca. 100 K like in hot cores/corinos, the iCOMs formed in the interstellar medium (ISM) were nonthermally by photodesorption because of UV incidence on inherited by small objects of the solar system4−8 (e.g., the grains,33−35 by chemical desorption once they are carbonaceous chondrites), which upon alteration mechanisms formed,36−41 or induced by cosmic rays (CRs).42−45 In the (e.g., hydrothermal) can be converted into more evolved gas phase they react with other gaseous species forming organic species,9−12 thereby providing a potential contribution iCOMs. In the on-grain paradigm, the hydrogenated species to the emergence of life on Earth. act as parent species of molecular radicals (e.g., CH3, HCO, The first detection of iCOMs took place in 1971 in massive and NH2) formed by the irradiation of UV photons and/or star formation regions,13 but we had to wait for the beginning energetic ions, partial hydrogenation, and H-abstraction of this century for detections in regions that could eventually evolve in solar-like planetary systems.6,14 Currently, complex Special Issue: Chemical Complexity in Planetary organic molecules have been detected in different astrophysical Systems environments such as prestellar cores, protostellar outflows, protoplanetary disks, and even external galaxies.3,8,15−18 Received: November 1, 2021 Two different prevailing paradigms have been postulated for Revised: February 21, 2022 the formation of iCOMs: one based on gas-phase reac- Accepted: February 27, 2022 tions19−24 and the other based on radical−radical couplings Published: March 7, 2022 © 2022 The Authors. Published by American Chemical Society https://doi.org/10.1021/acsearthspacechem.1c00369 496 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article reactions2,25,26,46−50 during the cold preprotostellar stage. not present any other possible reaction). However, they Later on, these radicals, because of the warm up of the present energy barriers because the reaction is between a protostar surroundings (ca. 30 K), can diffuse on the icy radical and a neutral closed-shell species. surfaces and encounter each other to couple and form the The present work aims to investigate this alternative on- iCOMs. grain mechanism by simulating with quantum chemical These two reaction paradigms have been largely used in computations the reactivity of the CCH radical with a water combined observational/astrochemical modeling studies to molecule of the ice. The goal is to study the possibility of rationalize the presence of iCOMs in the given sources (e.g., forming ethanol (EtOH), with the formation of vinyl alcohol see refs 3 and 51−53). Additionally, they have also been (VA) as intermediate species, through the following reactions: assumed to study the formation of iCOMs by means of CCH + H 2O → H 2CCOH/HCCHOH (1) quantum chemical simulations.54−57 By compiling all the available works, it seems that both paradigms are necessary to H 2CCOH/HCCHOH + H → H 2CCHOH (VA) (2) explain the presence, distribution, and abundance of the wide and rich organic chemistry in the ISM. Thus, knowing whether H 2CCHOH + 2H → CH3CH 2OH (EtOH) (3) the formation of an iCOM is dominated by surface or gas- phase chemistry is a case-by-case problem. It depends on the We consider this path toward ethanol because the electronic nature of the iCOMs, and each one has to be addressed as a structure of CCH is isoelectronic with CN, hence exhibiting a particular case. similar reactivity with water as shown in Rimola et al.58 With the different quantum chemical studies addressing on- CCH was one of the first interstellar polyatomic molecules grain radical−radical couplings, some drawbacks of this to be detected.59 It is a fundamental and common carbon paradigm have been identified. One is that the chance for chain (in fact, the simplest) species in the ISM. It is found in the coupling is a delicate trade-off between the diffusion of the regions near UV sources, the so-called photon dominated radicals and their desorption. That is, the temperature increase regions (PDRs) (e.g., see refs 60 and 61); in diffuse and that enables the radical diffusion must be lower than their translucent clouds (e.g., see refs 62 and 63); in protostellar desorption temperatures. This can give rise to a small objects;64,65 and in the cavities of protostellar outflows.66,67 In temperature window in which the coupling can take place, addition, CCH has been detected in protoplanetary because this coupling temperature window is defined by the disks30,68−71 and external galaxies (e.g., see ref 72) probably lowest diffusion temperature and the lower desorption in their PDR regions/skins. CCH is also abundant toward the temperature between the two radicals. For instance, in lukewarm (≤40 K) envelopes of the so-called warm carbon acetaldehyde formation by the coupling of CH3 and HCO, chain chemistry (WCCC) objects, which are young class 0/I the coupling temperature window was found to be between the sources characterized by higher abundances of carbon chains temperatures at which CH3 becomes mobile on the surface and lower abundances of iCOMs than those observed in hot (between 9 and 15 K, depending on the adopted diffusion corinos (see for example ref. 65). CCH appears at scales of a barrier) and the temperature at which the methyl radical would few thousand AU around the protostellar center at densities of desorb (30 K).55 Another drawback is that radical−radical some 106 cm−3 (e.g., see ref 73). Particularly relevant to this couplings are often assumed to be barrierless because they are work, CCH is also relatively abundant in cold molecular clouds driven by the coupling of the opposite electronic spins of the (e.g., see refs 74 and 75) and prestellar cores.76 Typical CCH radicals. However, the reactions can exhibit energy barriers column densities range from 1012−1015 cm−2,61,77 with the because the radicals, to proceed with the coupling, need to greatest values in PDRs.61 In molecular clouds, the CCH break the interactions with the icy surfaces. Obviously, as a rule column density is around 1014 cm−2, equivalent to abundances of thumb, the stronger the interaction, the higher the energy of ∼10−9−10−8.74,75 Similar abundances are found in prestellar barrier. A third limitation of this paradigm is that radical− cores, as CCH probably resides in the least depleted zone, radical couplings can have competitive channels inhibiting the similar to the molecular clouds.76,78,79 In summary, CCH is efficiency of the iCOM formation. The competitive reactions abundant (∼10−9−10−8) in cold (≤20 K) regions where the are based on H-abstractions from one radical to the other. For interstellar dust grains are enveloped by icy mantles. Thus, it is instance, the CH3 + HCO coupling can give CH3CHO worth investigating the possibility that it interacts with the (acetaldehyde as iCOM) but also CH4 + CO (the H- water molecules of the ice to form ethanol, following the path abstraction product). described above (reactions 1−3). In a previous work by some of us,58 an alternative on-grain The reaction of CCH + H2O has been studied as a gas-phase mechanism different from the radical−radical coupling was process at high temperatures, both experimentally and proposed. It is based on the reaction of a radical (coming from theoretically. Seminal experiments by Van Look and Peeters80 the gas phase or generated by UV irradiation) with neutral, suggested that the outcome of the reaction was not the result entire components of the ice, i.e., H2O and CO as the most of a direct H-abstraction forming HCCH + OH, and hence, abundant components. In that work, this mechanism was the authors proposed a mechanism based on, first, an tested to form formamide (NH2CHO) through the reaction of association between the two reactive partners, forming the radical CN with a water molecule belonging to the ice, i.e., H2CCHOH or HCCHOH, followed by an elimination CN + H2O(ice) → NH2CO, in which the resulting species can giving rise to H2CCO + H and/or HCCH + OH. Subsequent be easily hydrogenated to form formamide. This alternative theoretical calculations,81 however, indicated that among the mechanism overcomes the problems of (i) the coupling different reactive channels, the direct H-abstraction was the temperature window (the diffusion of the radical is not needed most kinetically favorable one, in detriment to the association− because it reacts with an abundant ice component, thus elimination mechanism. Investigations regarding this reaction increasing the chance of the encountering among the two ended with a combined experimental/theoretical study carried reactants) and (ii) the competitive channel (they a priori do out by some of the first paper’s authors, concluding that the H- 497 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article Figure 1. Structures of the 18 and 33 water molecule clusters, optimized at the ωB97X-D3/6-311++G(d,p) level of theory. abstraction producing HCCH + OH is indeed the most facile significance, this work focuses on the formation of this chemical reaction.82 ancestor molecule following reactions 1−3 on water ice In contrast, to the best of our knowledge, no experimental surfaces by means of quantum chemical simulations to works exist on the reactivity of CCH with water ices. However, determine if they are energetically feasible. ■ the addition of OH radicals to C2H2 ices (isoelectronic with CCH + H2O) at temperatures below 20 K, followed by H- METHODOLOGY additions, has been studied by several authors, resulting in the formation of several products, among them vinyl alcohol and All the calculations were based on density functional theory ethanol. This reactivity usually involves an energetic (DFT) and run with the Gaussian16 software package.92 Geometry optimizations and frequency calculations were all preprocessing of the ice analogues (i.e., irradiation with ions, performed by combining the DFT methods with the Pople- electrons, and photons) to generate the OH radicals. Among based 6-311++G(d,p) basis set.93,94 These energies were these works, we can find (1) ion radiolysis experiments (MeV subsequently refined at the 6-311++G(2df,2pd)95 level by protons) that generate mainly CO, CO2, methanol and performing single-point energy calculations on the optimized ethanol;83 (2) MeV protons and far UV photon irradiation geometries. In order to identify the DFT method that better that yield vinyl alcohol formation;84 (3) UV irradiation and describes our systems (and hence to use it for the reaction proton radiolysis of the ices that form vinyl alcohol, simulations on the water ice surface models), we carried out a acetaldehyde, ketene, and ethanol;85 (4) ice irradiation with preliminary benchmarking study using the CCH + H2O and extreme UV photons that leads to the formation of some CH2CHOH + H gas-phase reactions as models. Five different iCOMs like ethane, methanol, and ethanol, together with some hybrid DFT methods were used, which were corrected with simpler species (e.g., CO, CO2, and methane);86 and (5) Grimme’s D3 term or, if possible, with the D3(BJ) radiolysis of the C2H2:H2O ices with less energetic protons version96−98 to account for dispersion interactions. The tested (200 keV) at 17 K that produce several iCOMs, like vinyl DFT-D methods were BHLYP-D3(BJ),99,100 M062X-D3,101 alcohol, acetaldehyde, ketene and ethanol, and some other MPWB1K-D3(BJ),102 PW6B95-D3(BJ),103 and ωB97X-D3.104 species, such as C2H4, C2H6, C4H2, and C4H4.87 In this later By comparing the results with those obtained with single work, it was proposed that once vinyl alcohol is formed by the energy points at the CCSD(T)/aug-cc-pVTZ105 level of attack of an OH radical to C2H2, two possible situations may theory, known as the “gold-standard” in quantum chemistry, take place: either an intramolecular isomerization step forming we found that the ωB97X-D3 method showed the best acetaldehyde or successive H-additions on vinyl alcohol to performance when modeling the water addition to CCH, while form ethanol. More recently, nonenergetic processes have also MPWB1K-D3(BJ) described better the hydrogenation of been explored, in which C2H2:O2 ices exposed to H atoms at CH2CHOH (see Results). Accordingly, the CCH + H2O 10 K produce most of the products found in Chuang et al.,87 and the CH2CHOH + H reactions on the water ice cluster including acetaldehyde, vinyl alcohol, ketene, and ethanol. models were computed at the ωB97X-D3/6-311++G- Other experiments indicate that vinyl alcohol and acetaldehyde (2df,2pd)//ωB97X-D3/6-311++G(d,p) and MPWB1K- can also be formed through other chemical reactions. That is, D3(BJ)/6-311++G(2df,2pd)//MPWB1K-D3(BJ)/6-311++G- the O addition to C 2 H 2 mainly results in ketene (d,p) theory levels, respectively. formation,30,85,88,89 while the O addition to more saturated All the stationary points of the potential energy surfaces hydrocarbons (acetylene, ethane, and ethylene) leads to the (PESs) were characterized by their analytical calculation of the formation of different iCOMs (ketene, ethanol and acetalde- harmonic frequencies as minima (reactants, products, and hyde, and acetaldehyde and ethylene oxide, respectively).30,90 intermediates) and saddle points (transition states). When Ethanol formation has recently received much attention needed, intrinsic reaction coordinate (IRC) calculations at the because it has been postulated to be a parent molecule through level of theory adopted in the geometry optimizations were which other iCOMs can be formed by different gas-phase carried out to ensure that a given transition state actually reactions, such as acetaldehyde, glycolaldehyde, formic acid, connects with the corresponding minima. Thermochemical acetic acid, and formaldehyde (the so-called genealogical tree corrections to the potential energy values were carried out of ethanol; see refs 23, 24, and 91). Because of this using the standard rigid rotor/harmonic oscillator formulas to 498 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article Figure 2. Stationary points identified in the benchmarking study. Reaction pathways R′1, R′2, and R′3 refer to gas-phase CCH + H2O reaction and are optimized at ωB97X-D3. H′1 and H′2 correspond to the two hydrogenation pathways of vinyl alcohol, optimized at MPWB1K-D3(BJ). Distances are in Å. compute the zero-point energy (ZPE) corrections.106 Because (d,p) level, followed by single-point energy calculations at the the systems are open-shell in nature, calculations were improved ωB97X-D3/6-311++G(2df,2pd) level. Basis set performed within the unrestricted formalism. superposition error (BSSE) was corrected following the Boys We additionally calculated the tunnelling crossover temper- and Bernardi counterpoise method. The final, corrected, atures, i.e., the temperature below which quantum tunnelling adsorption energy (ΔECP ads) was calculated as becomes the main mechanism for trespassing the potential CP energy barriers. To this end, we used eq 4,107 where ΔH‡ is the ΔEads (CCH/surf) ZPE-corrected barrier height, ω‡ is the frequency associated = ΔEads + BSSE(CCH) + BSSE(surf) + ΔZPE (5) with the TS, and kB and ℏ are the Boltzmann’s and reduced Planck’s constants. This temperature indicates what reactions where ΔEads = E(CCH/surf) − E(CCH) − E(surf) refers to may actually have an important role at low temperatures the BSSE-noncorrected adsorption energy. Note that we used despite having an activation barrier. the same sign convention as in ref 54; namely, the adsorption energy is the negative of the binding energy: ΔECP ads = −BE. Tc = ℏω‡ΔH ‡/kB 2π ΔH ‡ − ℏω‡ ln(2) (4) ■ RESULTS Benchmarking Study. As mentioned above, we carried The surfaces of amorphous solid water (ASW) ice coating out a preliminary benchmarking analysis for the reactivity using interstellar grains were simulated by two different cluster two gas-phase model reactions, CCH + H2O and CH2CHOH models: one consisting of 18 water molecules and the other + H, to find the DFT method that describes better the reaction consisiting of 33 water molecules (hereafter termed W18 and properties. For CCH + H2O, we found three possible reaction W33; Figure 1). While the former represents a compact, flat pathways, labeled as R′1, R′2, and R′3, the stationary points of water ice surface, the latter presents a cavity structure of about which are shown in Figure 2. 6 Å. For more details, we refer the reader to refs 54, 58, and R′1 follows a stepwise mechanism: the first step involves the 108. formation of acetylene (HCCH) and the hydroxyl radical For the calculation of the binding energies (BEs) of CCH (OH) as intermediate species, and the second step consists of interacting with the H2O ice surface models (i.e., CCH/surf the condensation of these intermediates to form the complexes), we adopted the same electronic structure HCCHOH radical. In contrast, both R′2 and R′3 adopt a methodology as for reactivity, namely, for each CCH/surf concerted mechanism. These two reactions involve a C−O complex and corresponding isolated components, geometry bond formation followed by a H-transfer to the other C atom optimizations and frequency calculations (and hence ZPE to form the H2CCOH radical (isomer of HCCHOH, the corrections) were computed at the ωB97X-D3/6-311++G- product of R′1). The difference between R′2 and R′3 is that, 499 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article Table 1. Relative Energies (in kJ mol−1) of the Stationary Points Involved in the Reaction Pathways Found for the Gas-Phase CCH + H2O Reaction Model (R′1, R′2, and R′3) and for the Gas-Phase Hydrogenation (H′1 and H′2) for All the DFT-D Methods and the CCSD(T) Method Used for the Benchmarking Studya reaction step BHLYP-D3(BJ) M062X-D3 MPWB1K-D3(BJ) PW6B95-D3(BJ) ωB97X-D3 CCSD(T) R′1 H2O + CCH 0.0 0.0 0.0 0.0 0.0 0.0 TS1 28.5 8.9 25.6 19.6 29.0 29.1 OH + HCCH −91.6 −57.0 −72.5 −63.0 −62.3 −69.0 TS2 −68.5 −62.8 −60.2 −61.1 −53.5 −51.5 HCCHOH −217.0 −222.0 −217.7 −203.6 −205.9 −193.8 TS1 % R′1 2.1 69.5 11.8 32.6 0.4 TS2 % R′1 33.2 22.1 17.0 18.8 4.0 Avg % R′1 20.0 30.9 11.5 16.3 5.1 R′2 H2O + CCH 0.0 0.0 0.0 0.0 0.0 0.0 TS1 117.4 107.5 96.4 90.8 108.0 99.5 H2CCOH −249.3 −246.8 −266.3 −253.2 −247.8 −241.1 TS % R′2 18.0 8.1 3.1 8.7 8.5 Avg % R′2 10.7 5.2 6.8 6.9 5.6 R′3 H2O + CCH 0.0 0.0 0.0 0.0 0.0 0.0 TS 120.3 87.7 98.6 123.5 113.0 105.4 H2CCOH −240.3 −244.8 −241.9 −228.8 −229.7 −214.1 TS % R′3 16.7 16.8 6.5 17.1 7.2 Avg % R′3 14.5 15.6 9.7 12.0 7.2 Global Avg % 15.1 17.2 9.3 11.7 6.0 H′1 CH2CHOH + H 0.0 0.0 0.0 0.0 0.0 0.0 TS 0.3 4.1 2.3 11.4 6.5 CH3CHOH −192.6 −183.3 −177.1 −188.1 −176.0 TS % H′1 116.4 43.6 78.5 93.0 Avg % H′1 62.9 23.9 39.6 49.9 H′2 CH2CHOH + H 0.0 0.0 0.0 0.0 0.0 0.0 TS 8.0 14.2 12.5 21.6 15.9 CH2CH2OH −157.9 −143.0 −177.1 −149.0 −142.0 TS % H′2 54.2 11.9 23.7 38.5 Avg % H′2 32.7 6.3 24.1 21.7 Global Avg % 47.8 15.1 31.8 35.8 a See Methodology for more details. Rows indicated as “TS % RX” show the unsigned error (in percent) relative to the predicted energy of the TS. Rows indicated as “Avg % RX” show the averaged unsigned errors (in percent and accounting for all the stationary points) of each DFT-D method with respect to the CCSD(T)//ωB97X-D3 for R′1, R′2, and R′3 and CCDS(T)//MPWB1K-D3(BJ) for H′1 and H′2 reference values. The last row indicates the global averaged unsigned error (in percent). in the former, the H-transfer comes first, followed then by a different products (CH3CHOH and CH2CH2OH, respec- spontaneous C−O bond formation, while in the latter, the C− tively), depending on which C atom the H addition takes O bond formation and the proton transfer evolve in a place. H′1 and H′2 share a similar mechanism, in which the H synchronized way. atom in the reactant structures is located at ca. 3.5 Å from the The computed energetics of these reaction pathways using C atom with which it will react. Results show that path H′1 is the different quantum chemical methods are shown in Table 1. more favorable than H′2, both for the stability of the product The optimized geometries of the stationary points are available and for presenting a lower activation energy barrier (see Table in the Supporting Information. As can be seen, the overall best 1). This is because the H′1 product (CH3CHOH) exhibits a functional for these reactions of water addition to CCH is better delocalization of the unpaired electron with respect to ωB97X-D3, with an average unsigned error of ∼5% in the the H′2 product (CH2CH2OH). Among the used DFT energy barriers and ∼6% in the whole energetics. The methods, the functional with the smallest average unsigned performance of the other functionals is, from better to error compared to CCSD(T) single-point energy calculations worse, MPWB1K-D3(BJ), PW6B95-D3(BJ), BHLYP-D3(BJ), is MPWB1K-D3(BJ). The performance of the other func- and M062X-D3. tionals, from better to worse, is PW6B95-D3(BJ), ωB97X-D3, For the hydrogenation steps (reactions 2 and 3), we and BHLYP-D3(BJ). M062X-D3 was discarded for con- considered only the hydrogenation of vinyl alcohol, namely, vergence problems of the reactant structures. Therefore, CH2CHOH + H, because it is the unique step that can exhibit according to this benchmarking study, the ωB97X-D3 DFT an energy barrier due to involving a closed-shell molecule method has been chosen to simulate the addition of water to (vinyl alcohol) with a radical (H atom). In contrast, the other CCH on the W18 and W33 cluster models, and MPWB1K- steps consist of the hydrogenation of radical species, which are D3(BJ) has been adopted for the hydrogenation step of vinyl barrierless processes because of the spontaneous spin−spin alcohol. coupling. For this H-addition reaction, we found two possible Adsorption of CCH on Water Ice and Binding pathways (labeled as H′1 and H′2 in Figure 2), leading to two Energies. The complexes formed when CCH adsorbs on 500 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article W18 and W33 are shown in the R structures of Figures 3 and and maps, which clearly indicate a delocalization of the 4, respectively. Table 2 reports the computed binding energies unpaired electron between the two centers (see data in the and the different contributions, as detailed in Methodology. Supporting Information). The distances of these hemibonds In most of the cases, a nonclassical hemibond between the vary between 2.1 and 2.3 Å (see the R structures of the R2, R3, CCH species and an H2O of the ice is established because of R2-1, and R2-2 sequences of Figures 3 and 4). The only the formation of a two-center, three-electron system between exception not presenting a hemibonded complex is the the unpaired electron of CCH and a lone pair of H2O. This structure R of R1 (Figure 3), in which classical, weak hydrogen interaction is highlighted by the computed spin density values bond (H-bond) interactions are established between CCH and the W18 ice model. Interestingly, any attempt to find a pure H- bonded complex (i.e., without any hemibonded interaction) on W33 failed, with the geometry optimizations collapsing into the hemibonded structures. This reinforces the idea that the more surface water molecules, the higher the possibility to form hemibonded complexes, although most of the hemi- bonded complexes also present H-bond interactions. This is because the outermost water molecules of the cluster are unsaturated in terms of H-bonds, presenting H- and O- dangling atoms ready to establish H-bond and hemibonded interactions, respectively. Remarkably, hemibonded systems also form in the reactant structures of the gas-phase reactions (see above). However, these gas-phase systems present a lower hemibonded character than the complexes on W18 and W33 because the spin is less delocalized between the two centers (see spin density values and maps in the Supporting Information). As expected, hemibonded complexes present BE values larger than those of H-bonded ones (see Table 2). The difference in the BEs between the complexes shown in R2 and R3 (i.e., on W18) arises from the orbital occupied by the CCH unpaired electron. In the former, the unpaired electron belongs to the π system of the CCH (the spin density is shared between the two C atoms, and the linearity of CCH becomes broken upon hemibond formation), while in the latter it belongs to a σ orbital of the C-end of CCH, in both cases interacting with a lone pair of H2O to form the hemibond. As the orbital overlap is more efficient in σ orbitals than in π orbitals, the latter complex presents a larger BE than the former (49.9 and 37.9 kJ mol−1, respectively). On W33, both hemibonded complexes have similar BE values (89.7 and 86.0 kJ mol−1) because in both systems the unpaired electron occupies a π orbital. Remarkably, the values on W33 are notably higher than those on W18, and we investigated what might be the cause. We checked for a correlation between the BE and the number of water molecules forming the cluster, carving several structures from our W33 model, that is, by removing water molecules from the edges of the model in order to build a set of water clusters with decreasing dimensions (and blocking some O atoms of the cluster to prevent the cavity from collapsing). By proceeding inthis way, we found evidence that the increasing number of interactions between CCH and the ice, together with the cooperativity of H-bonds, is responsible for the increment of the BE. Data of this analysis are provided in the Supporting Information. Finally, it is worth mentioning that we performed a preliminary benchmarking study on the binding energies of the dimeric CCH/H2O system. Results (available in the Supporting Information) indicate that while the H-bonded dimer is very well described by most of the DFT methods, this Figure 3. Computed potential energy surfaces (PESs) of the reactions R1-3 between CCH and the W18 ASW ice cluster model. Bare energy is not the case for the hemibonded one. However, for the values correspond to relative ZPE-corrected values, while values in particular case of ωB97X-D3, the computed binding energies parentheses correspond to those missing this correction. The are somewhat overestimated, belonging to the group of the miniature panels sketch the ZPE-corrected PESs. Energies are in kJ functionals that better compare with CCSD(T). Accordingly, mol−1, and distances are in Å. the computed binding energies of the hemibonded complexes 501 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article Figure 4. Computed potential energy surfaces (PESs) of the reactions R2-1 and R2-2 between CCH and W33 ASW ice cluster model. Bare energy values correspond to relative ZPE-corrected values, while values in parentheses correspond to those missing this correction. The miniature panels sketch the ZPE-corrected PESs. Energies are in kJ mol−1, and distances are in Å. Table 2. Binding Energy (BE) Values (in kJ mol−1) of CCH arising from CCH in interaction with W18 and W33 should be on the W18 and W33 Cluster Models According to the considered overestimated by some amount. Despite this Computed Complexes Shown in Figures 3 and 4 (the R drawback, we would like to stress that the main scope of the Structures of the Reactions (Rx) R1, R2, R3, R2-1, and R2- work is on the reactivity between CCH and H2O followed by 2)a H-additions and that the used DFT methods are actually surface Rx ΔEads ΔEdisp ΔZPE ΔBSSE BE accurate enough for this purpose, as shown above. In this case, W18 R1 −21.2 −3.3 −0.6 1.4 23.7 the error in the binding energies self-canceled when deriving R2 −32.8 −10.1 3.0 2.1 37.9 the energy features (energy barriers and reaction energies) of R3 −32.2 −22.1 2.2 2.2 49.9 the reactions. W33 R2-1 −97.1 −7.2 11.2 3.3 89.7 Reactivity between CCH and H2O on the Ice Surface R2-2 −81.6 −19.0 11.2 3.5 86.0 Models. W18 ASW Ice Cluster Model. Following the three a The contributions from the pure potential energy values (ΔEads), reaction types found in the benchmark study, we tried to dispersion corrections (ΔEdisp), zero-point energy corrections reproduce them on top of the W18 ice cluster model. (ΔZPE), and BSSE corrections (ΔBSSE) are also shown. However, significant differences in the mechanisms have been found, precisely because they occur on the ice model. Indeed, 502 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article the larger number of water molecules indicates that (i) there that R2 is an overall effectively barrierless reaction on W18. In are no concerted mechanisms at work and (ii) water-assisted contrast, R1 has a low-energy intermediate (associated with proton-transfer reactions are operative. Water-assisted proton- the formation of HCCH), and consequently, the second step transfer mechanisms were elucidated theoretically in the latter (TS2) encounters a relatively high energy barrier of 34.6 kJ years of the last century109 and can induce important decrease mol−1 (with respect to the intermediate). For R3, the first step in the energy barriers with respect to the nonwater-assisted already presents a non-negligible energy barrier, even when it is analogues. This is because the assisting water molecules bridge ZPE-corrected (14.1 kJ mol−1). For the second steps, TS2-1 the accepting/donating proton processes with their neighbor- has an energy lower than that of TS2-2 because the geometry ing molecules and, at the same time, reduce the strain of the associated with the latter saddle point is more strained, because rings in the transition-state structures. Hence, this mechanism the cycle created by the water assistant molecules is smaller. All is also called proton-transport catalysis. Such catalytic effects the reaction paths present negative and very large reaction have also been observed in processes of interstellar interest in energies, indicating that the formed vinyl alcohol radicals are icy surfaces.58,110,111 It is worth mentioning that the water- more stable than the reactant states. We observed that, in some assisted proton transfers, to be catalytically effective, need a cases, the ZPE corrections added to the potential energies are proper H-bond connectivity among the bridging water very large (see, for instance, TS1 in R1, TS2 in R2, and TS2-1 molecules, from the first to the last proton transfer. This and TS2-2 in R3). Interestingly, these transition states involve means that, if the H-bonding network is truncated because of the motion of H atoms of the water molecules, either as a the presence of interstitial unpurities, the mechanism is not direct proton transfer (case of TS1 in R1) or as a water- operative. This is an aspect not to overlook in interstellar ices assisted proton-transfer mechanism (cases of TS2 in R2 and because they contain minor volatile species in the ices that can TS2-1 and TS2-2 in R3). In contrast, when a C−O bond is obstruct the chained proton relays. Remarkably, it is worth formed (namely, without involving any proton motion) it stressing that the catalytic transfers involve H atoms with a results in a slight ZPE correction. In order to understand this proton character and not atomic radicals. This is because the H behavior, we simulated the IR spectra of R, TS1, I, and TS2-1 species exchanged during the transfers are proton-like atoms of R3 (see the Supporting Information). The spectra of R and chemically bonded to a more electronegative O atom. TS1, where only the C−O bond forms, show mostly common Because the identified reaction channels on W18 differ features, with the exception of a couple of bands around 3000 significantly from the gas-phase ones, we redefine the R′1-3 cm−1, because of O−H stretching of the water molecules model channels into R1-3, which adopt the following surrounding the CCH. On the other hand, comparing the simplified mechanistic steps: spectrum of TS2-1 with R highlights the presence of a number R1 H2O + CCH → OH + HCCH → HCCHOH of features at a shorter wavenumber, arising from the water- assisted proton-transfer. Given that the ZPE arises from the R2 H2O + CCH → CC(H)−OH2 → HCCHOH sum of the energy of all the vibrational modes, the shorter R3 H2O + CCH → H2O−CCH → H2CCOH wavelength features of TS2-1 explain why its ZPE is smaller The stationary points and the energy profiles of these reaction than R and hence why the resulting ΔZPE is negative. pathways are shown in Figure 3. As a final comment, we stress the influence of the saturated The R1 path remains the same as the R′1 one, namely, state (in terms of H-bonding) of the reacting water molecule in formation of HCCH and OH as intermediate species by H- the energy barriers. Indeed, according to the potential energy transfer from the reactive H2O molecule to CCH, and then values, the TS1 barrier in R2 is actually lower than that in R3, coupling of the OH to HCCH to form the HCCHOH radical 0.7 and 16.0 kJ mol−1, respectively (see Figure 3). This is as vinyl alcohol precursor. Both R2 and R3 start with the because, in the former, TS1 is more reactant-like than in the formation of the hemibonded systems (see the Supporting latter because, in the R structure of R2, the reacting water Information for their spin densities). However, because the molecule is not fully saturated by H-bonds (i.e., 2 H-bonds as involved C atoms are different, the pathways proceed in a donor +1 H-bond as acceptor) while this is the case in R3 (i.e., different way as well. The hemibonded structure of R2 evolves 2 H-bonds as donor +2 H-bonds as acceptor). The lack of the toward the formation of the CC(H)−OH2 intermediate, which H-bond in R of the R2 reaction allows the water “unsaturated” is followed by a proton transfer, from the OH2 moiety and lone pair to form the hemibonded system with CCH, in which adopting a water-assisted proton-transfer mechanism, to form the C−O bond is half-formed. Thus, to reach the TS1 the final HCCHOH radical. In contrast, the hemibonded structure, no significant energy requirements and geometrical structure of R3 transforms into H 2 O−CCH as the changes are needed, and hence the low energy barrier. In intermediate species. From this intermediate, two different contrast, in R3, to form TS1 from the R structure, the reacting paths are possible, namely, from the OH2 moiety, a proton water molecule has to break one of the H-bonds to form the transfer to (i) the central C atom (TS2-1) or (ii) the terminal new C−O bond, implying a more energetic cost and relevant C atom (TS2-2), forming HCCHOH or H 2 CCOH, geometrical changes. Nevertheless, the associated quantum respectively. Interestingly, both paths proceed through a tunnelling crossover temperature is rather high (244 K), water-assisted proton-transfer mechanism. The fact that R3 indicating that this channel could be relevant at interstellar exhibits these two paths is because there are well-oriented H- temperatures. bonds in the water cluster allowing for the water-assisted W33 ASW Ice Cluster Model. After modeling the reaction mechanism in two directions, while this is not possible in R2 on the flat surface of W18, we set out to investigate the because of the geometry of the intermediate structure. possible effects of a cavity structure like that of W33. We tried In relation to the energetics (see also Figure 3), the first to reproduce the same three reaction paths as those taking steps (TS1) of paths R1 and R2 become submerged below the place on W18, but we found only two mechanisms, and both energy of reactants once they are ZPE-corrected. The same are similar to R2; hence, they are termed R2-1 and R2-2. The happens for the second step (TS2) of path R2. This means stationary points and the energy profiles of these reaction 503 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article pathways are shown in Figure 4. It is worth mentioning that, to model the reaction path R2-2, it was necessary to fix the position of some of the oxygen atoms placed at the edge of the model because full geometry relaxation led to the collapse of the cavity. In both cases, CCH is located in the cavity, with the central C atom interacting with the oxygen atom of the reactant water molecule of the surface, forming the aforementioned hemi- bonded systems (see the Supporting Information for spin densities). Along the reactions, these structures evolve to form the CC(H)−OH2 intermediate. For both cases, this step has a very low ZPE-corrected energy barrier (2.1 and 0.8 kJ mol−1 for TS1 and TS2, respectively). The second step involves, for both paths, the proton transfer from the OH2 moiety to the terminal C atom forming the HCCHOH radical, also by means of a water-assisted proton-transfer mechanism. The two paths lead to the same radical because the H-bond network enables the water-assisted proton-transfer to connect with the terminal C atom and not the central one. These second steps are energetically below the corresponding intermediates (by including ZPE corrections), and accordingly, the water-assisted proton transfers proceed in a barrierless fashion. Because the reaction energies are very large and negative, the computed reactions on W33 are energetically very favorable, similarly to the R2 reaction occurring on W18. Toward the Formation of Ethanol: Hydrogenation of Vinyl Alcohol. As shown above, reaction of CCH with an icy water molecule leads to the formation of CHCHOH and H2CCOH. From these species, to reach ethanol, a set of hydrogenation steps are necessary, as sketched in reactions 6−8. HCCHOH/H 2CCOH + H → CH 2CHOH (6) CH 2CHOH + H → CH3CHOH/CH 2CH 2OH (7) CH3CHOH/CH 2CH 2OH + H → CH3CH 2OH (8) The first hydrogenation forms vinyl alcohol (VA). As this H- addition is a radical−radical coupling, we consider it as Figure 5. Computed potential energy surfaces (PESs) of the barrierless. The second hydrogenation step is the H-addition hydrogenation reactions H1 and H2 on W18 ASW ice cluster to the VA. It involves the reactivity between a closed-shell model. Bare energy values correspond to relative ZPE-corrected species (namely, VA) with a radical (H), and accordingly, it is values, while values in parentheses correspond to those missing this correction. The miniature panels sketch the ZPE-corrected PESs. expected to have an activation barrier. Interestingly, depending Energies are in kJ mol−1, and distances are in Å. on the C in which the H-addition takes place, the species formed is either CH3CHOH or CH2CH2OH. The third and final hydrogenation step leads to the formation of ethanol, and reaction energies. Remarkably, the computed energy irrespective of the initial radical species. This H-addition is, barriers on W18 are higher (3−4 kJ mol−1) than those again, a radical−radical coupling, and accordingly, it is obtained in the gas-phase. This is probably because the expected to be barrierless in a similar fashion as the first adsorption of VA with the surface induces an enhanced hydrogenation. According to this reactive scheme, to stabilization of the former due to the intermolecular forces investigate on ethanol formation, we simulated only reaction between VA and the icy surface. Thus, chemically strictly 7 on the W18 ice model. speaking, the surface does not act as a chemical catalyst but We identified two reaction paths, termed H1 and H2 (see slightly inhibits the process. However, it is worth highlighting Figure 5), the difference of which being the C atom that that the H1 reaction presents a very low potential energy undergoes the H-addition, in analogy to the gas-phase H′1 and barrier and, because of the involvement of a H atom and the H′2 reactions (see Figure 2). Both reactions start from a very low temperatures of the ISM, tunneling effects can prereactant structure in which the H atom is ca. 3.3 Å from the operate as well. Indeed, its associated quantum tunnelling VA because of the weak interactions between the two partners. crossover temperature is 118 K. For reaction H2, the crossover The computed potential energy surfaces indicate that the two temperature is slightly higher, 174 K; therefore, quantum hydrogenation reactions present a non-negligible energy tunneling could also play a role. barrier. In agreement with gas-phase results, the path leading Isomerization between Vinyl Alcohol and Acetalde- to the formation of CH3CHOH is more favorable than that hyde. Vinyl alcohol and acetaldehyde are tautomers (i.e., resulting with CH2CH2OH, in terms of both energy barriers structural isomers), and therefore, we have here considered the 504 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article conversion from one to the other according to reaction 9. This isomerization reaction has been computed at the MPWB1K- D3(BJ)//6-311++G(2df,2pd)//MPWB1K-D3(BJ)//6-311+ +G(d,p) level for consistency with the methodology applied to the hydrogenation of vinyl alcohol. CH 2CHOH ⇔ CH3CHO (9) Among them, acetaldehyde is the most stable isomer (by ∼41 kJ mol−1). In the gas phase, the barrier connecting vinyl alcohol with acetaldehyde is very high (237.9 kJ mol−1, see the Supporting Information). This is because the mechanism involves an intramolecular H-transfer from the OH of vinyl alcohol to the terminal C atom, with a highly strained transition state. If we now consider this reaction to take place through the water molecules of W18, as shown in Figure 6, we can see that a water-assisted proton-transfer mechanism can take place, lowering the activation energy barrier down to 73.5 kJ mol−1 when the reaction starts from vinyl alcohol (I1) and to 57.7 kJ mol−1 when it starts from its less hydrogenated precursor (I2). The latter reaction produces CH2CHO, which can be successively hydrogenated to form acetaldehyde. The differ- ence in the two activation barriers is probably due to the structure of the ring through which the proton is transferred; that is, in I1 it is more strained. This is a great example of how interstellar ices act as chemical catalysts. However, computed energy barriers are very high to be surmountable under interstellar conditions, and accordingly, these channels seem unlikely. Nevertheless, we note that the barrier of I2 is narrower and lower than that of I1 (with barriers widths of about 1.6 and 2.2 Å, respectively, assuming the asymmetric Eckart barrier model; see ref 112). Therefore, I2 would be the most efficient mechanism among the two, if assuming that quantum tunneling does play a role. ■ DISCUSSION AND ASTROPHYSICAL IMPLICATIONS CCH is a highly reactive species that easily reacts with water. Figure 6. Computed potential energy surfaces (PESs) of the Because it is a C-centered radical and contains both donor and isomerization between vinyl alcohol and acetaldehyde (I1) and the acceptor H-bond atoms (the H and the C-end atoms, analogue reaction involving their precursor, i.e., the product P2 of R3 respectively), it tends to form H-bonded and hemibonded (I2) on W18 ice cluster model at MPWB1K-D3(BJ)/6-311+ complexes with water ice molecules. The most relevant +G(2df,2pd)//MPWB1K-D3(BJ)/6-311++G(d,p). Bare energy val- computed energetics of CCH reactivity on our ASW ice ues correspond to relative ZPE-corrected values, while values in models are summarized in Table 3. On W18, R3 presents an parentheses correspond to those missing this correction. The miniature panel sketches the ZPE-corrected PESs. Energies are in activation barrier of 14.4 kJ mol−1, and accordingly, it is not a kJ mol−1, and distances are in Å. priori an efficient path to form vinyl alcohol, although it could still be relevant if quantum tunnelling effects work. The first Table 3. Summary of the Energetics of the Simulated step of R1 is barrierless to form HCCH + OH, but the reaction Reactions and Their Products stops at this stage. This is because the second step has an intrinsic energy barrier of 34.6 kJ mol−1 (the energy difference reaction barrier product between TS2 and I; see Figure 3). Moreover, the energy W18 R1 NO HCCH + OH released by the first reaction step will no longer be available for R2 NO HCCHOH the second step, as water ices tend to efficiently dissipate R3 14.4 kJ mol−1 H2CCOH/HCCHOH chemical energy fast.113,114 Therefore, R1 will lead to the W33 R2-1 2.1 kJ mol−1 HCCHOH formation of a highly stabilized acetylene, which will hardly R2-2 0.8 kJ mol−1 HCCHOH react with OH remaining stuck on the ice. However, this W18 H1 7.4 kJ mol−1 CH3CH2OH reaction can be an effective channel toward the formation of H2 17.3 kJ mol−1 CH3CH2OH OH radicals on the ice surfaces without the need of a direct energy processing. This can be of importance because the generated OH could participate in further surface reactions in modeled on W33 to study the effects of the cavity on the the form of OH additions. R2, at variance with the other two reactivity. On W33, we found that the reactions are no more reactions, is an effective barrierless reaction path toward vinyl barrierless, although the two identified paths present low alcohol formation. Because of that, this reaction has also been activation barriers (2.1 and 0.8 kJ mol−1 for R2-1 and R2-2, 505 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article respectively). Because these two pathways are equivalent to R2 at variance with most of the experimentally proposed on W18, the effect of the cavity is almost insignificant for the mechanisms (see Introduction), the formation of the energetics of the path. Therefore, we can consider that the aforementioned products does not require the energetic mechanistic steps represented by the R2, R2-1, and R2-2 processing of interstellar ices. Regarding the formation of pathways constitute the most likely channel to form vinyl iCOMs, our proposal complements the nonenergetic reaction alcohol. scheme of Chuang et al.,123 in which OH radicals attack C2H2 The hydrogenation of vinyl alcohol on grains leads to the to form iCOMs. We note that CCH (and the other formation of ethanol, as occurring in methanol formation from intermediate radicals) could also be destroyed by other CO115−119 and ethane formation from C2H2 and C2H4.120−122 competitive surface reactions, e.g., by H-abstraction reactions The H-addition to vinyl alcohol is the only hydrogenation step with H2, or H-additions. This will be considered in the future. not involving a barrierless radical−radical coupling and In general, the astrochemical processes in cold regions such presents an energy barrier of 7.4 and 17.3 kJ mol−1 for the as the ones described in this work are important to improve H1 and H2 channel (see Table 3). Thus, H1 is the most our understanding of the presence of complex species detected energetically favored pathway for the formation of the ethanol in the cold (<20 K) outskirts of prestellar cores during the past radical precursor. Despite the computed relatively high energy decade (e.g.16,124−128). In this vein, ethanol has recently gained barriers considering very low temperatures, these reactions can some attention, as it has been advocated to be the precursor of proceed through tunneling and are accordingly operative in the several iCOMs formed by cold gas-phase reactions (the reaction chain starting from the adsorption of CCH on water genealogical tree of ethanol23,91). Indeed, the correlation of ice. glycoladehyde and acetaldehyde abundances observed toward These quantum chemical results can partly be related to the a number of interstellar sources has been shown to follow very experimental findings mentioned in the Introduction. Indeed, well the theoretical predictions when their synthesis takes place part of the experimental synthetic routes have been simulated through this gas-phase scheme.24 In the present work, we here, providing an atomistic interpretation (including the showed that an efficient path exists for the formation of energetics) for the formation of vinyl alcohol followed by it ethanol on the surfaces of interstellar ices. However, the hydrogenation to form ethanol. The main difference between nonthermal desorption of ethanol (and of the other products) the experiments and our computations is that, in the remains as a crucial missing step in the sequential events experiments, the C2H2/H2O ices need to be processed to linking the chemistry of iCOMs occurring on the surface of trigger reactivity (probably due to generating the radical grains and in the gas phase in cold regions; this issue is reactive species like CCH and OH), while in our simulations undoubtedly a central matter of further investigation. the assumption is that the reaction does not require energetic processing of ices, because the CCH is readily available in the gas phase in a wide variety of environments. ■ CONCLUSIONS Interstellar complex organic molecules (iCOMs) have been We also considered the isomerization of vinyl alcohol into detected in different astrophysical environments. However, the acetaldehyde (and the same for their less hydrogenated chemistry leading to their formation is not unambiguously precursors), because in the experiments these reactions were known. Two prevailing paradigms have been largely used to suggested to explain the presence of both vinyl alcohol and rationalize their presence in the interstellar medium: one acetaldehyde. According to our calculations, however, these advocates reaction in the gas-phase, whereas the other presents reactions, although being catalyzed by the surfaces thanks to a reaction on the surface of icy grains. In this work, we focused water-assisted proton-transfer mechanism, present high energy on the latter by computing the reaction of CCH with a H2O barriers, rendering them poorly competitive to the final H molecule forming part of the ice structure, leading to vinyl additions. However, experimental authors pointed out that the alcohol (CH2CHOH), which upon hydrogenation is converted isomerization could take place thanks to the exothermicity of into ethanol (CH3CH2OH). This reaction is proposed as an the previous steps, in which the energy released along the alternative synthetic route for iCOMs beyond the commonly reaction steps can be used to overcome the isomerization assumed radical−radical couplings. Investigations have been energy barrier. Our computed energetic data is consistent with performed by means of DFT quantum chemical simulations this view. The very favorable reaction energies shown by the and adopting cluster models of 18 and 33 water molecules reactions indicate that the energy barriers of the isomerization (W18 and W33) to mimic the icy surfaces. For the reaction of processes lay below the prereactive asymptotic states, and CCH with H2O, three different reaction pathways have been therefore, they can be overcome by making use of the nascent elucidated, leading to the formation of HCCH + OH, reaction energies. However, one should bear in mind that CHCHOH, and H2CCOH. Some cases present small water ice surfaces are extraordinary third bodies113,114 and activation barriers, but in others the reactions are barrierless accordingly, the direct transfer of the previous reaction when zero-point energy corrections are accounted for. energies to surmount the isomerization energy barriers is Hydrogenation of vinyl alcohol on the W18 cluster has been doubtful. To shed some light on this topic, dedicated ab initio found to present activation barriers of a certain significance molecular dynamics simulations are needed, which is outside but, for these reactions, quantum tunneling is likely to be at scope of the present work. work, speeding them up. Isomerization between vinyl alcohol Finally, results presented in this work are very relevant in the and acetaldehyde have also been simulated on W18, with the framework of cold astrochemistry. The presence of abundant results indicating that, despite the strong catalytic role played (∼10−9−10−8) gaseous CCH radicals in cold (≤20 K) regions, by the water ice, they have a barrier of significant height. where water ices envelope the refractory cores of the Additionally, the direct H-abstraction from the water molecule interstellar dust grains (see Introduction), can lead to the to CCH leading to the formation of HCCH and the OH formation of vinyl alcohol and ethanol in addition to HCCH radical on the surface has been found to be an energetically and OH through a competitive reaction channel. Remarkably, competitive channel. It is worth mentioning that a chemical 506 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article kinetics treatment of these results is underway in order to Cecilia Ceccarelli − Univ. Grenoble Alpes, CNRS, Institut de compute the rate constants (including tunneling effects Planétologie et d’Astrophysique de Grenoble (IPAG), 38000 explicitly) for each of the proposed reactive channels and Grenoble, France; orcid.org/0000-0001-9664-6292 hence elucidate branching ratios and formation efficiencies of Nadia Balucani − Dipartimento di Chimica, Biologia e the simulated paths. Biotecnologie, Università di Perugia, 06123 Perugia, Italy; In summary, results from our calculations indicate that the Univ. Grenoble Alpes, CNRS, Institut de Planétologie et reaction of CCH with water ice can lead to the formation of d’Astrophysique de Grenoble (IPAG), 38000 Grenoble, vinyl alcohol and, lately, to the production of ethanol, and France; Osservatorio Astrosico di Arcetri, 50125 Firenze, likely acetaldehyde, without the need of ice energy processing. Italy; orcid.org/0000-0001-5121-5683 This conclusion is of relevance in the context of iCOM Piero Ugliengo − Dipartimento di Chimica and formation because, according to the genealogical tree of Nanostructured Interfaces and Surfaces (NIS) Centre, ethanol, this species is the parent molecule through which Università degli Studi di Torino, 10125 Torino, Italy; different iCOMs (e.g., formic acid, formaldehyde, glycolalde- orcid.org/0000-0001-8886-9832 hyde, and others) can form by means of gas-phase Complete contact information is available at: processes.23,24 Thus, in this work, we provide quantum https://pubs.acs.org/10.1021/acsearthspacechem.1c00369 chemical evidence on the feasibility of our mechanistic proposal, in which ethanol can be formed on interstellar icy Notes grain surfaces, hence linking the two paradigms in the synthesis The authors declare no competing financial interest. ■ of iCOMs, at least in this particular case. The missing link between on-grain and gas-phase chemistry stands in the ACKNOWLEDGMENTS nonthermal desorption of ethanol and its precursors, which This project has received funding within the European Union’s should be a subject for further investigation. Horizon 2020 research and innovation programme from the ■ * ASSOCIATED CONTENT sı Supporting Information European Research Council (ERC) for the projects “The Dawn of Organic Chemistry” (DOC), Grant Agreement No. 741002 and “Quantum Chemistry on Interstellar Grains” (QUANTUMGRAIN), Grant Agreement No. 865657. The The Supporting Information is available free of charge at authors acknowledge funding from the European Union’s https://pubs.acs.org/doi/10.1021/acsearthspace- Horizon 2020 research and innovation program Marie chem.1c00369. Sklodowska-Curie for the project “Astro-Chemical Origins” Structures and errors of the benchmarking study; (ACO), Grant Agreement No. 811312. A.R. is indebted to energetics relative to the CCH + H2O reactions on “Ramón y Cajal” program. MINECO (project CTQ2017- W18 and W33; contributions to the binding energies of 89132-P) and DIUE (project 2017SGR1323) are acknowl- CCH on W18 and W33; Mulliken spin densities and edged. Finally, we thank Prof. Gretobape for fruitful and spin density maps of the complexes CCH + H2O in the stimulating discussions. Most of the quantum chemical gas phase and on W18 and W33; energetics relative to calculations presented in this paper were performed using the hydrogenation of vinyl alcohol; energetics relative to the GRICAD infrastructure (https://gricad.univ-grenoble- the isomerization of vinyl alcohol into acetaldehyde alpes.fr), which is partly supported by the Equip@Meso (PDF) project (reference ANR-10-EQPX-29-01) of the programme Investissements d’Avenir supervised by the Agence Nationale ■ pour la Recherche. Additionally, this work was granted access AUTHOR INFORMATION to the HPC resources of IDRIS under allocation 2019- A0060810797 attributed by GENCI (Grand Equipement Corresponding Authors National de Calcul Intensif). CSUC supercomputing center Joan Enrique-Romero − Univ. Grenoble Alpes, CNRS, is acknowledged for allowance of computer resources. ■ Institut de Planétologie et d’Astrophysique de Grenoble (IPAG), 38000 Grenoble, France; Departament de Química, REFERENCES Universitat Autònoma de Barcelona, Bellaterra 08193 (1) Herbst, E.; van Dishoeck, E. F. Complex Organic Interstellar Catalonia, Spain; orcid.org/0000-0002-2147-7735; Molecules. Annu. Rev. Astron. Astrophys. 2009, 47, 427−480. Email: juan.enrique-romero@univ-grenoble-alpes.fr (2) Herbst, E. The synthesis of large interstellar molecules. Int. Rev. Albert Rimola − Departament de Química, Universitat Phys. Chem. 2017, 36, 287−331. Autònoma de Barcelona, Bellaterra 08193 Catalonia, Spain; (3) Ceccarelli, C.; et al. Seeds Of Life In Space (SOLIS): The orcid.org/0000-0002-9637-4554; Email: albert.rimola@ Organic Composition Diversity at 300−1000 au Scale in Solar-type uab.cat Star-forming Regions. Astrophys. J. 2017, 850, 176. (4) Caselli, P.; Ceccarelli, C. Our astrochemical heritage. Astron. Authors Astrophys. Rev. 2012, 20, 56. Jessica Perrero − Departament de Química, Universitat (5) Ceccarelli, C.; Caselli, P.; Bockelée-Morvan, D.; Mousis, O.; Autònoma de Barcelona, Bellaterra 08193 Catalonia, Spain; Pizzarello, S.; Robert, F.; Semenov, D. Deuterium Fractionation: The Ariadne’s Thread from the Precollapse Phase to Meteorites and Dipartimento di Chimica and Nanostructured Interfaces and Comets Today. In Protostars and Planets VI; 2014. DOI: 10.2458/ Surfaces (NIS) Centre, Università degli Studi di Torino, azu_uapress_9780816531240-ch037 10125 Torino, Italy (6) Cazaux, S.; Tielens, A. G. G. M.; Ceccarelli, C.; Castets, A.; Berta Martínez-Bachs − Departament de Química, Wakelam, V.; Caux, E.; Parise, B.; Teyssier, D. The Hot Core around Universitat Autònoma de Barcelona, Bellaterra 08193 the Low-Mass Protostar IRAS 16293−2422: Scoundrels Rule! Catalonia, Spain Astrophys. J. 2003, 593, L51−L55. 507 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article (7) Ligterink, N. F. W.; Calcutt, H.; Coutens, A.; Kristensen, L. E.; (25) Garrod, R. T.; Herbst, E. Formation of methyl formate and Bourke, T. L.; Drozdovskaya, M. N.; Müller, H. S. P.; Wampfler, S. F.; other organic species in the warm-up phase of hot molecular cores. van der Wiel, M. H. D.; van Dishoeck, E. F.; Jørgensen, J. K. The Astron. Astrophys. 2006, 457, 927−936. ALMA-PILS survey: Stringent limits on small amines and nitrogen- (26) Garrod, R. T.; Weaver, S. L. W.; Herbst, E. Complex Chemistry oxides towards IRAS 16293−2422B. Astron. Astrophys. 2018, 619, in Star-forming Regions: An Expanded Gas-Grain Warm-up Chemical A28. Model. Astrophys. J. 2008, 682, 283−302. (8) Bianchi, E.; Codella, C.; Ceccarelli, C.; Vazart, F.; Bachiller, R.; (27) Ruaud, M.; Loison, J. C.; Hickson, K. M.; Gratier, P.; Hersant, Balucani, N.; Bouvier, M.; De Simone, M.; Enrique-Romero, J.; F.; Wakelam, V. Modelling complex organic molecules in dense Kahane, C.; Lefloch, B.; López-Sepulcre, A.; Ospina-Zamudio, J.; regions: Eley-Rideal and complex induced reaction. Mon. Not. R. Podio, L.; Taquet, V. The census of interstellar complex organic Astron. Soc. 2015, 447, 4004−4017. molecules in the Class I hot corino of SVS13-A. Mon. Not. R. Astron. (28) Krasnokutski, S. A.; Jäger, C.; Henning, T. Condensation of Soc. 2019, 483, 1850−1861. Atomic Carbon: Possible Routes toward Glycine. Astrophys. J. 2020, (9) Alexander, C. M. O.; Cody, G. D.; Kebukawa, Y.; Bowden, R.; 889, 67. Fogel, M. L.; Kilcoyne, A. L. D.; Nittler, L. R.; Herd, C. D. K. (29) Bergner, J. B.; Ö berg, K. I.; Rajappan, M. Methanol Formation Elemental, isotopic, and structural changes in Tagish Lake insoluble via Oxygen Insertion Chemistry in Ices. Astrophysical Journal 2017, organic matter produced by parent body processes. Meteorit. Planet. 845, 29. Sci. 2014, 49, 503−525. (30) Bergner, J. B.; Ö berg, K. I.; Rajappan, M. Oxygen Atom (10) Rotelli, L.; Trigo-Rodríguez, J. M.; Moyano-Cambero, C. E.; Reactions with C2H6, C2H4, and C2H2 in Ices. Astrophys. J. 2019, 874, Carota, E.; Botta, L.; di Mauro, E.; Saladino, R. The key role of 115. meteorites in the formation of relevant prebiotic molecules in a (31) Fedoseev, G.; Cuppen, H. M.; Ioppolo, S.; Lamberts, T.; formamide/water environment. Sci. Rep. 2016, 6, 38888. Linnartz, H. Experimental evidence for glycolaldehyde and ethylene (11) Yabuta, H.; Williams, L. B.; Cody, G. D.; Alexander, C. M. O. glycol formation by surface hydrogenation of CO molecules under D.; Pizzarello, S. The insoluble carbonaceous material of CM dense molecular cloud conditions. Mon. Not. R. Astron. Soc. 2015, 448, chondrites: A possible source of discrete organic compounds under 1288−1297. hydrothermal conditions. Meteorit. Planet. Sci. 2007, 42, 37−48. (32) Simons, M. A. J.; Lamberts, T.; Cuppen, H. M. Formation of (12) Callahan, M. P.; Smith, K. E.; Cleaves, H. J.; Ruzicka, J.; Stern, COMs through CO hydrogenation on interstellar grains. Astron. J. C.; Glavin, D. P.; House, C. H.; Dworkin, J. P. Carbonaceous Astrophys. 2020, 634, A52. meteorites contain a wide range of extraterrestrial nucleobases. Proc. (33) Bertin, M.; Fayolle, E. C.; Romanzin, C.; Poderoso, H. A. M.; Natl. Acad. Sci. U. S. A. 2011, 108, 13995−13998. Michaut, X.; Philippe, L.; Jeseck, P.; Ö berg, K. I.; Linnartz, H.; Fillion, (13) Rubin, R. H.; Swenson, J. G. W.; Benson, R. C.; Tigelaar, H. L.; J.-H. Indirect Ultraviolet Photodesorption from CO:N2 Binary Ices − Flygare, W. H. Microwave Detection of Interstellar Formamide. an Efficient Grain-gas Process. Astrophys. J. 2013, 779, 120. Astrophys. J. 1971, 169, L39. (34) Bertin, M.; Romanzin, C.; Doronin, M.; Philippe, L.; Jeseck, P.; (14) Ceccarelli, C.; Loinard, L.; Castets, A.; Faure, A.; Lefloch, B. Ligterink, N.; Linnartz, H.; Michaut, X.; Fillion, J.-H. UV Photo- Search for glycine in the solar type protostar IRAS 16293−2422. desorption of Methanol in Pure and CO-rich Ices: Desorption Rates Astron. Astrophys. 2000, 362, 1122−1126. of the Intact Molecule and of the Photofragments. Astrophys. J. Lett. (15) McGuire, B. A. 2018 Census of Interstellar, Circumstellar, 2016, 817, L12. Extragalactic, Protoplanetary Disk, and Exoplanetary Molecules. (35) Fayolle, E.; Bertin, M.; Romanzin, C.; Michaut, X.; Oberg, K.; Astrophys. J.s 2018, 239, 17. Linnartz, H.; Fillion, J. H. Photodesorption of interstellar ice (16) Bacmann, A.; Taquet, V.; Faure, A.; Kahane, C.; Ceccarelli, C. analogues: a wavelength-dependent study. In The Molecular Universe; Detection of complex organic molecules in a prestellar core: a new 2011; p 157. challenge for astrochemical models. Astron. Astrophys. 2012, 541, L12. (36) Duley, W. W.; Williams, D. A. The formation of H2 on (17) Arce, H. G.; Santiago-García, J.; Jørgensen, J. K.; Tafalla, M.; interstellar dust. Mon. Not. R. Astron. Soc. 1993, 260, 37−42. Bachiller, R. Complex Molecules in the L1157 Molecular Outflow. (37) Garrod, R. T.; Wakelam, V.; Herbst, E. Non-thermal desorption Astrophys. J. 2008, 681, L21. from interstellar dust grains via exothermic surface reactions. Astron. (18) Muller, S.; Beelen, A.; Black, J. H.; Curran, S. J.; Horellou, C.; Astrophys. 2007, 467, 1103−1115. Aalto, S.; Combes, F.; Guélin, M.; Henkel, C. A precise and accurate (38) Minissale, M.; Dulieu, F. Influence of surface coverage on the determination of the cosmic microwave background temperature at z chemical desorption process. J. Chem. Phys. 2014, 141, No. 014304. = 0.89. Astron. Astrophys. 2013, 551, A109. (39) Minissale, M.; Dulieu, F.; Cazaux, S.; Hocuk, S. Dust as (19) Charnley, S. B.; Tielens, A. G. G. M.; Millar, T. J. On the interstellar catalyst I. Quantifying the chemical desorption process. Molecular Complexity of the Hot Cores in Orion A: Grain Surface Astron. Astrophys. 2016, 585, A24. Chemistry as “The Last Refuge of the Scoundrel. Astrophys. J. 1992, (40) Takahashi, J.; Williams, D. A. Chemical Desorption of CO in 399, L71. the Vicinity of H2 Forming Sites on Dust. In The Physics and (20) Charnley, S. B.; Tielens, A. G. G. M.; Rodgers, S. D. Deuterated Chemistry of the Interstellar Medium; 1999; p 375. Methanol in the Orion Compact Ridge. Astrophys. J. 1997, 482, (41) Vasyunin, A. I.; Herbst, E. Reactive Desorption and Radiative L203−L206. Association as Possible Drivers of Complex Molecule Formation in (21) Taquet, V.; Wirström, E. S.; Charnley, S. B. Formation and the Cold Interstellar Medium. Astrophys. J. 2013, 769, 34. Recondensation of Complex Organic Molecules during Protostellar (42) Leger, A.; Jura, M.; Omont, A. Desorption from interstellar Luminosity Outbursts. Astrophys. J. 2016, 821, 46. grains. Astron. Astrophys. 1985, 144, 147−160. (22) Balucani, N.; Ceccarelli, C.; Taquet, V. Formation of complex (43) Hasegawa, T. I.; Herbst, E. Three-Phase Chemical Models of organic molecules in cold objects: the role of gas-phase reactions. Dense Interstellar Clouds - Gas Dust Particle Mantles and Dust Mon. Not. R. Astron. Soc. 2015, 449, L16−L20. Particle Surfaces. Mon. Not. R. Astron. Soc. 1993, 263, 589. (23) Skouteris, D.; Balucani, N.; Ceccarelli, C.; Vazart, F.; Puzzarini, ̅ (44) Kalvans, J. Temperature Spectra of Interstellar Dust Grains C.; Barone, V.; Codella, C.; Lefloch, B. The Genealogical Tree of Heated by Cosmic Rays. I. Translucent Clouds. Astrophys. J., Suppl. Ethanol: Gas-phase Formation of Glycolaldehyde, Acetic Acid, and Ser. 2016, 224, 42. Formic Acid. Astrophys. J. 2018, 854, 135. (45) Dartois, E.; Chabot, M.; Id Barkach, T.; Rothard, H.; Augé, B.; (24) Vazart, F.; Ceccarelli, C.; Balucani, N.; Bianchi, E.; Skouteris, Agnihotri, A. N.; Domaracka, A.; Boduch, P. Cosmic ray sputtering D. Gas-phase formation of acetaldehyde: review and new theoretical yield of interstellar H2O ice mantles. Ice mantle thickness depend- computations. Mon. Not. R. Astron. Soc. 2020, 499, 5547−5561. ence. Astron. Astrophys. 2018, 618, A173. 508 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article (46) Shingledecker, C. N.; Tennis, J.; Gal, R. L.; Herbst, E. On (64) Sakai, N.; Sakai, T.; Hirota, T.; Yamamoto, S. Abundant Cosmic-Ray-driven Grain Chemistry in Cold Core Models. ApJ. Carbon-Chain Molecules toward the Low-Mass Protostar IRAS 2018, 861, 20. 04368 + 2557 in L1527. Astrophys. J. 2008, 672, 371−381. (47) Shingledecker, C. N.; Herbst, E. A general method for the (65) Sakai, N.; Yamamoto, S. Warm Carbon-Chain Chemistry. inclusion of radiation chemistry in astrochemical models. Phys. Chem. Chem. Rev. 2013, 113, 8981−9015. Chem. Phys. 2018, 20, 5359−5367. (66) Oya, Y.; Sakai, N.; Sakai, T.; Watanabe, Y.; Hirota, T.; (48) Taquet, V.; Ceccarelli, C.; Kahane, C. Multilayer modeling of Lindberg, J. E.; Bisschop, S. E.; Jørgensen, J. K.; van Dishoeck, E. F.; porous grain surface chemistry: I. The GRAINOBLE model. A&A Yamamoto, S. A Substellar-Mass Protostar and Its Outflow of IRAS 2012, 538, A42. 15398−3359 Revealed by Subarcsecond-Resolution Observations of (49) Minissale, M.; Moudens, A.; Baouche, S.; Chaabouni, H.; H2 CO and CCH. ApJ. 2014, 795, 152. Dulieu, F. Hydrogenation of CO-bearing species on grains: (67) Zhang, Y.; Higuchi, A. E.; Sakai, N.; Oya, Y.; López-Sepulcre, unexpected chemical desorption of CO. Mon. Not. R. Astron. Soc. A.; Imai, M.; Sakai, T.; Watanabe, Y.; Ceccarelli, C.; Lefloch, B.; 2016, 458, 2953−2961. Yamamoto, S. Rotation in the NGC 1333 IRAS 4C Outflow. ApJ. (50) Jin, M.; Garrod, R. T. Formation of Complex Organic 2018, 864, 76. Molecules in Cold Interstellar Environments through Nondiffusive (68) Kastner, J. H.; Hily-Blant, P.; Rodriguez, D. R.; Punzi, K.; Grain-surface and Ice-mantle Chemistry. Astrophys. J., Suppl. Ser. Forveille, T. Unbiased Millimeter-Wave Line Surveys of TW Hya and V4046 Sgr: The Enhanced C2H and CN Abundances of Evolved 2020, 249, 26. Protoplanetary Disks. ApJ. 2014, 793, 55. (51) Codella, C.; et al. Seeds of Life in Space (SOLIS). II. (69) Kastner, J. H.; Qi, C.; Gorti, U.; Hily-Blant, P.; Oberg, K.; Formamide in protostellar shocks: Evidence for gas-phase formation. Forveille, T.; Andrews, S.; Wilner, D. A Ring of C2H in the Molecular Astron. Astrophys. 2017, 605, L3. Disk Orbiting TW Hya. ApJ. 2015, 806, 75. (52) Ligterink, N. F. W.; Calcutt, H.; Coutens, A.; Kristensen, L. E.; (70) Miotello, A.; Facchini, S.; van Dishoeck, E. F.; Cazzoletti, P.; Bourke, T. L.; Drozdovskaya, M. N.; Müller, H. S. P.; Wampfler, S. F.; Testi, L.; Williams, J. P.; Ansdell, M.; van Terwisga, S.; van der Marel, van der Wiel, M. H. D.; van Dishoeck, E. F.; Jørgensen, J. K. The N. Bright C 2 H emission in protoplanetary discs in Lupus: high ALMA-PILS survey: Stringent limits on small amines and nitrogen- volatile C/O > 1 ratios. A&A 2019, 631, A69. oxides towards IRAS 16293−2422B. Astron. Astrophys. 2018, 619, (71) Guzmán, V. V.; et al. Molecules with ALMA at Planet-forming A28. Scales (MAPS). VI. Distribution of the Small Organics HCN, C 2 H, (53) McGuire, B. A.; Shingledecker, C. N.; Willis, E. R.; Burkhardt, and H 2 CO. Astrophys. J., Suppl. Ser. 2021, 257, 6. A. M.; El-Abd, S.; Motiyenko, R. A.; Brogan, C. L.; Hunter, T. R.; (72) Meier, D. S.; Turner, J. L. Spatially Resolved Chemistry in Margulès, L.; Guillemin, J.-C.; Garrod, R. T.; Herbst, E.; Remijan, A. Nearby Galaxies. I. The Center of IC 342. ApJ. 2005, 618, 259−280. J. ALMA Detection o f Interstellar Methoxymethanol (73) Sakai, N.; et al. A Chemical View of Protostellar-Disk (CH3OCH2OH). Astrophys. J. 2017, 851, L46. Formation in L1527. ApJ. 2014, 791, L38. (54) Enrique-Romero, J.; Rimola, A.; Ceccarelli, C.; Ugliengo, P.; (74) Friberg, P.; Madden, S. C.; Hjalmarson, A.; Irvine, W. M. Balucani, N.; Skouteris, D. Reactivity of HCO with CH3 and NH2 on Methanol in dark clouds. Astron. Astrophys. 1988, 195, 281−289. Water Ice Surfaces. A Comprehensive Accurate Quantum Chemistry (75) Dickens, J. E.; Irvine, W. M.; Snell, R. L.; Bergin, E. A.; Study. ACS Earth Space Chem. 2019, 3, 2158−2170. Schloerb, F. P.; Pratap, P.; Miralles, M. P. A Study of the Physics and (55) Enrique-Romero, J.; Ceccarelli, C.; Rimola, A.; Skouteris, D.; Chemistry of L134N. Astrophys. J. 2000, 542, 870−889. Balucani, N.; Ugliengo, P. Theoretical computations on the efficiency (76) Padovani, M.; Walmsley, C. M.; Tafalla, M.; Galli, D.; Müller, of acetaldehyde formation on interstellar icy grains. Astron. Astrophys. H. S. P. C{2}H in prestellar cores. Astron. Astrophys. 2009, 505, 2021, 655, A9. 1199−1211. (56) Enrique-Romero, J.; Á lvarez-Barcia, S.; Kolb, F. J.; Rimola, A.; (77) Higuchi, A. E.; Sakai, N.; Watanabe, Y.; López-Sepulcre, A.; Ceccarelli, C.; Balucani, N.; Meisner, J.; Ugliengo, P.; Lamberts, T.; Yoshida, K.; Oya, Y.; Imai, M.; Zhang, Y.; Ceccarelli, C.; Lefloch, B.; Kästner, J. Revisiting the Reactivity between HCO and CH3 on Codella, C.; Bachiller, R.; Hirota, T.; Sakai, T.; Yamamoto, S. Interstellar Grain Surfaces. Mon. Not. R. Astron. Soc. 2020, 493, 2523− Chemical Survey toward Young Stellar Objects in the Perseus 2527. Molecular Cloud Complex. Astrophys. J., Suppl. Ser. 2018, 236, 52. (57) Lamberts, T.; Markmeyer, M. N.; Kolb, F. J.; Kästner, J. (78) Liu, X. C.; Wu, Y.; Zhang, C.; Liu, T.; Yuan, J.; Qin, S. L.; Ju, B. Formation of Acetaldehyde on CO-Rich Ices. ACS Earth Space Chem. G.; Li, L. X. C2H N = 1 - 0 and N2H+ J = 1 - 0 observations of Planck 2019, 3, 958−963. Galactic cold clumps. Astron. Astrophys. 2019, 622, A32. (58) Rimola, A.; Skouteris, D.; Balucani, N.; Ceccarelli, C.; Enrique- (79) Taniguchi, K.; Plunkett, A.; Herbst, E.; Dobashi, K.; Shimoikura, T.; Nakamura, F.; Saito, M. Investigation of chemical Romero, J.; Taquet, V.; Ugliengo, P. Can Formamide Be Formed on differentiation among the NGC 2264 cluster-forming clumps. Mon. Interstellar Ice? An Atomistic Perspective. ACS Earth Space Chem. Not. R. Astron. Soc. 2020, 493, 2395−2409. 2018, 2, 720−734. (80) Van Look, H.; Peeters, J. Rate Coefficients of the Reactions of (59) Tucker, K. D.; Kutner, M. L.; Thaddeus, P. The Ethynyl C2H with O2, C2H2, and H2O between 295 and 450 K. J. Chem. Phys. Radical C2H-A New Interstellar Molecule. Astrophys. J. Lett. 1974, A 1995, 99, 16284−16289. 193, L115. (81) Ding, Y.-h.; Zhang, X.; Li, Z.-s.; Huang, X.-r.; Sun, C.-c. Is the (60) Teyssier, D.; Fossé, D.; Gerin, M.; Pety, J.; Abergel, A.; Roueff, C2H+H2O Reaction Anomalous? J. Chem. Phys. A 2001, 105, 8206− E. Carbon budget and carbon chemistry in Photon Dominated 8215. Regions. Astron. Astrophys. 2004, 417, 135−149. (82) Carl, S. A.; Minh Thi Nguyen, H.; Elsamra, R. M. I.; Tho (61) Bouvier, M.; López-Sepulcre, A.; Ceccarelli, C.; Kahane, C.; Nguyen, M.; Peeters, J. Pulsed Laser Photolysis and Quantum Imai, M.; Sakai, N.; Yamamoto, S.; Dagdigian, P. J. Hunting for hot Chemical-Statistical Rate Study of the Reaction of the Ethynyl Radical corinos and WCCC sources in the OMC-2/3 filament. A&A 2020, with Water Vapor. J. Chem. Phys. 2005, 122, 114307. 636, A19. (83) Moore, M. H.; Hudson, R. L. Infrared Study of Ion-Irradiated (62) Lucas, R.; Liszt, H. S. Comparative chemistry of diffuse clouds. Water-Ice Mixtures with Hydrocarbons Relevant to Comets. Icarus I. C_2H and C_3H_2. Astron. Astrophys. 2000, 358, 1069−1076. 1998, 135, 518−527. (63) Gerin, M.; Kaźmierczak, M.; Jastrzebska, M.; Falgarone, E.; (84) Hudson, R. L.; Moore, M. H. Solid-Phase Formation of Hily-Blant, P.; Godard, B.; de Luca, M. The tight correlation of CCH, Interstellar Vinyl Alcohol. Astrophys. J. 2003, 586, L107−L110. and c-C3H2 in diffuse and translucent clouds. Astron. Astrophys. 2011, (85) Hudson, R. L.; Loeffler, M. J. Ketene Formation in Interstellar 525, A116. Ices: A Laboratory Study. Astrophys. J. 2013, 773, 109. 509 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article (86) Wu, C. R.; Judge, D.; Cheng, B.-M.; Shih, W.-H.; Yih, T.-S.; Ip, (105) Raghavachari, K.; Trucks, G. W.; Pople, J. A.; Head-Gordon, W. Extreme ultraviolet photon-induced chemical reactions in the M. A fifth-order perturbation comparison of electron correlation C2H2−H2O mixed ices at 10 K. Icarus 2002, 156, 456−473. theories. Chem. Phys. Lett. 1989, 157, 479−483. (87) Chuang, K. J.; Fedoseev, G.; Scirè, C.; Baratta, G. A.; Jäger, C.; (106) McQuarrie, D. A. Statistical mechanics; Harper & Row New Henning, T.; Linnartz, H.; Palumbo, M. E. Formation of complex York, 1975. organic molecules in molecular clouds: acetaldehyde, vinyl alcohol, (107) Fermann, J. T.; Auerbach, S. Modeling proton mobility in ketene, and ethanol via the “energetic” processing of C2H2 ice. Astron. acidic zeolite clusters: II. Room temperature tunneling effects from Astrophys. 2021, 650, A85. semiclassical rate theory. J. Chem. Phys. 2000, 112, 6787−6794. (88) Haller, I.; Pimentel, G. C. Reaction of Oxygben Atoms with (108) Rimola, A.; Taquet, V.; Ugliengo, P.; Balucani, N.; Ceccarelli, Acetylene to Form Ketene. J. Am. Chem. Soc. 1962, 84, 2855−2857. C. Combined Quantum Chemical and Modeling Study of CO (89) Zasimov, P. V.; Ryazantsev, S. V.; Tyurin, D. A.; Feldman, V. I. Hydrogenation on Water Ice. Astron. Astrophys. 2014, 572, A70. Radiation-induced chemistry in the C2H2-H2O system at cryogenic (109) Chalk, A. J.; Radom, L. Proton-Transport Catalysis:A temperatures: a matrix isolation study. Mon. Not. R. Astron. Soc. 2019, Systematic Study of the Rearrangement of the Isoformyl Cation to 491, 5140−5150. the Formyl Cation. J. Am. Chem. Soc. 1997, 119, 7573−7578. (90) Bennett, C. J.; Osamura, Y.; Lebar, M. D.; Kaiser, R. I. (110) Molpeceres, G.; Kästner, J.; Fedoseev, G.; Qasim, D.; Laboratory Studies on the Formation of Three C2H4O Isomers- Schömig, R.; Linnartz, H.; Lamberts, T. Carbon Atom Reactivity Acetaldehyde (CH3CHO), Ethylene Oxide (c-C2H4O), and Vinyl with Amorphous Solid Water: H2O-Catalyzed Formation of H2CO. Alcohol (CH2CHOH)-in Interstellar and Cometary Ices. Astrophys. J. J. Phys. Chem. Lett. 2021, 12, 10854−10860. PMID: 34727500 2005, 634, 698−711. (111) Rimola, A.; Sodupe, M.; Ugliengo, P. Deep-space glycine (91) Vazart, F. Can Interstellar CH3OCH3 + H+ Lead to Protonated formation via Strecker-type reactions activated by ice water dust Acetaldehyde CH3CHOH +? 19th International Conference on mantles. A computational approach. Phys. Chem. Chem. Phys. 2010, Computational Science and Its Applications, ICCSA 2019, Saint 12, 5285−5294. Petersburg, Russia, July 1−4, 2019, Part VII. 2019; pp 138−139. (112) Peters, P. S.; Duflot, D.; Faure, A.; Kahane, C.; Ceccarelli, C.; (92) Frisch, M. J. et al. Gaussian 16, Rev C.01; Gaussian Inc.: Wiesenfeld, L.; Toubin, C. Theoretical Investigation of the Isomer- Wallingford CT, 2016. ization of trans-HCOH to H2CO: An Example of a Water-Catalyzed (93) Hehre, W. J.; Ditchfield, R.; Pople, J. A. SelfConsistent Reaction. J. Phys. Chem. A 2011, 115, 8983−8989. Molecular Orbital Methods. XII. Further Extensions of Gaussian (113) Pantaleone, S.; Enrique-Romero, J.; Ceccarelli, C.; Ugliengo, P.; Balucani, N.; Rimola, A. Chemical Desorption versus Energy Type Basis Sets for Use in Molecular Orbital Studies of Organic Dissipation: Insights from Ab Initio Molecular Dynamics of HCO· Molecules. J. Chem. Phys. 1972, 56, 2257−2261. (94) Hariharan, P. C.; Pople, J. A. The influence of polarization Formation. Astrophys. J. 2020, 897, 56. (114) Pantaleone, S.; Enrique-Romero, J.; Ceccarelli, C.; Ferrero, S.; functions on molecular orbital hydrogenation energies. Theor. Chim. Balucani, N.; Rimola, A.; Ugliengo, P. H2 Formation on Interstellar Acta 1973, 28, 213−222. Grains and the Fate of Reaction Energy. Astrophys. J. 2021, 917, 49. (95) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. Self- (115) Hidaka, H.; Watanabe, M.; Kouchi, A.; Watanabe, N. Reaction consistent molecular orbital methods. XX. A basis set for correlated Routes in the CO-H2CO-dn-CH3OH-dmSYSTEM Clarified from wave functions. J. Chem. Phys. 1980, 72, 650−654. H(D) Exposure of Solid Formaldehyde at Low Temperatures. (96) Grimme, S. Semiempirical GGA-type density functional Astrophys. J. 2009, 702, 291−300. constructed with a long-range dispersion correction. J. Comput. (116) Watanabe, N.; Kouchi, A. Efficient Formation of Form- Chem. 2006, 27, 1787−1799. aldehyde and Methanol by the Addition of Hydrogen Atoms to CO in (97) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. A consistent and H2O-CO Ice at 10 K. Astrophys. J. 2002, 571, L173−L176. accurate ab initio parametrization of density functional dispersion (117) Osamura, Y.; Roberts, H.; Herbst, E. On the possible correction (DFT-D) for the 94 elements H-Pu. J. Chem. Phys. 2010, interconversion between pairs of deuterated isotopomers of methanol, 132, 154104. its ion, and its protonated ion in star-forming regions. Astron. (98) Grimme, S.; Ehrlich, S.; Goerigk, L. Effect of the damping Astrophys. 2004, 421, 1101−1111. function in dispersion corrected density functional theory. J. Comput. (118) Nagaoka, A.; Watanabe, N.; Kouchi, A. H-D Substitution in Chem. 2011, 32, 1456−1465. Interstellar Solid Methanol: A Key Route for D Enrichment. (99) Becke, A. D. A new mixing of Hartree−Fock and local density- Astrophys. J. 2005, 624, L29−L32. functional theories. J. Chem. Phys. 1993, 98, 1372−1377. (119) Fuchs, G. W.; Cuppen, H. M.; Ioppolo, S.; Romanzin, C.; (100) Lee, C.; Yang, W.; Parr, R. G. Development of the Colle- Bisschop, S. E.; Andersson, S.; van Dishoeck, E. F.; Linnartz, H. Salvetti correlation-energy formula into a functional of the electron Hydrogenation reactions in interstellar CO ice analogues - A density. Phys. Rev. B 1988, 37, 785−789. combined experimental/theoretical approach. Astron. Astrophys. (101) Zhao, Y.; Truhlar, D. The M06 suite of density functionals for 2009, 505, 629−639. main group thermochemistry, thermochemical kinetics, noncovalent (120) Bennett, J. E.; Mile, B. Studies of radical−molecule reactions interactions, excited states, and transition elements: Two new using a rotating cryostat. Reactions of hydrogen atoms with organic functionals and systematic testing of four M06 functionals and 12 substrates at 77 K. J. Chem. Soc., Faraday Trans. 1 1973, 69, 1398− other functionals. Theor. Chem. Acc. 2008, 119, 525. 1414. (102) Zhao, Y.; Truhlar, D. G. Hybrid Meta Density Functional (121) Hiraoka, K.; Takayama, T.; Euchi, A.; Handa, H.; Sato, T. Theory Methods for Thermochemistry, Thermochemical Kinetics, Study of the Reactions of H and D Atoms with Solid C2H2, C2H4, and Noncovalent Interactions: The MPW1B95 and MPWB1K and C2H6at Cryogenic Temperatures. Astrophysical Journal 2000, Models and Comparative Assessments for Hydrogen Bonding and 532, 1029−1037. van der Waals Interactions. J. Phys. Chem. A 2004, 108, 6908−6918. (122) Kobayashi, H.; Hidaka, H.; Lamberts, T.; Hama, T.; Kawakita, (103) Zhao, Y.; Truhlar, D. G. Design of Density Functionals That H.; KÃ ďstner, J.; Watanabe, N. Hydrogenation and Deuteration of Are Broadly Accurate for Thermochemistry, Thermochemical C2H2and C2H4on Cold Grains: A Clue to the Formation Kinetics, and Nonbonded Interactions. J. Phys. Chem. A 2005, 109, Mechanism of C2H6with Astronomical Interest. Astrophysical Journal 5656−5667. PMID: 16833898 2017, 837, 155. (104) Chai, J.-D.; Head-Gordon, M. Systematic optimization of (123) Chuang, K. J.; Fedoseev, G.; Qasim, D.; Ioppolo, S.; Jäger, C.; long-range corrected hybrid density functionals. J. Chem. Phys. 2008, Henning, T.; Palumbo, M. E.; van Dishoeck, E. F.; Linnartz, H. 128, No. 084106. Formation of complex molecules in translucent clouds: acetaldehyde, 510 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511 ACS Earth and Space Chemistry http://pubs.acs.org/journal/aesccq Article vinyl alcohol, ketene, and ethanol via “nonenergetic” processing of C2H2 ice. Astron. Astrophys. 2020, 635, A199. (124) Cernicharo, J.; Marcelino, N.; Roueff, E.; Gerin, M.; Jiménez- Escobar, A.; Muñoz Caro, G. M. Discovery of the Methoxy Radical, CH3O, toward B1: Dust Grain and Gas-phase Chemistry in Cold Dark Clouds. Astrophys. J. 2012, 759, L43. (125) Vastel, C.; Ceccarelli, C.; Lefloch, B.; Bachiller, R. The Origin of Complex Organic Molecules in Prestellar Cores. Astrophys. J. 2014, 795, L2. (126) Jaber, A. A.; Ceccarelli, C.; Kahane, C.; Caux, E. The Census of Complex Organic Molecules in the Solar-type Protostar IRAS16293−2422. Astrophys. J. 2014, 791, 29. (127) Jiménez-Serra, I.; Vasyunin, A. I.; Caselli, P.; Marcelino, N.; Billot, N.; Viti, S.; Testi, L.; Vastel, C.; Lefloch, B.; Bachiller, R. The Spatial Distribution of Complex Organic Molecules in the L1544 Pre- stellar Core. Astrophys. J. 2016, 830, L6. (128) Scibelli, S.; Shirley, Y. Prevalence of Complex Organic Molecules in Starless and Prestellar Cores within the Taurus Molecular Cloud. Astrophys. J. 2020, 891, 73. 511 https://doi.org/10.1021/acsearthspacechem.1c00369 ACS Earth Space Chem. 2022, 6, 496−511