Skip to main content
NCBI home page
Frontiers in Chemistry logoLink to Frontiers in Chemistry
. 2022 Feb 8;9:827085. doi: 10.3389/fchem.2021.827085

Energy Redistribution Following CO2 Formation on Cold Amorphous Solid Water

Meenu Upadhyay 1, Markus Meuwly 1,*
PMCID: PMC8861491  PMID: 35211461

Abstract

The formation of molecules in and on amorphous solid water (ASW) as it occurs in interstellar space releases appreciable amounts of energy that need to be dissipated to the environment. Here, energy transfer between CO2 formed within and on the surface of amorphous solid water (ASW) and the surrounding water is studied. Following CO(1Σ+) + O(1D) recombination the average translational and internal energy of the water molecules increases on the ∼10 ps time scale by 15–25% depending on whether the reaction takes place on the surface or in an internal cavity of ASW. Due to tight coupling between CO2 and the surrounding water molecules the internal energy exhibits a peak at early times which is present for recombination on the surface but absent for the process inside ASW. Energy transfer to the water molecules is characterized by a rapid ∼10 ps and a considerably slower ∼1 ns component. Within 50 ps a mostly uniform temperature increase of the ASW across the entire surface is found. The results suggest that energy transfer between a molecule formed on and within ASW is efficient and helps to stabilize the reaction products generated.

Keywords: reactive molecular dynamics, amorphous solid water, interstellar chemistry, energy redistribution, CO2 formation

1 Introduction

The motion of adsorbates in and on amorphous solid water (ASW) is essential for chemistry at astrophysical conditions. Typically, bulk water is present in the form of ASW which is the main component of interstellar ices. (Hagen et al., 1981). The structure of ASW is usually probed by spectroscopic measurements (Hagen et al., 1981; Jenniskens and Blake, 1994) although interference-based methods have also been employed. (Bossa et al., 2012). ASWs are porous structures characterized by surface roughness and internal cavities of different sizes which can retain molecular or atomic guests. (Bar-Nun et al., 1987). Under laboratory conditions the water ices have been reported to be porous (He et al., 2016; Kouchi et al., 2020) or non-porous (Oba et al., 2009; He et al., 2016; Kouchi et al., 2020) ASW whereas the morphology of ices in the interstellar medium is more debated. (Keane et al., 2001; Kouchi et al., 2021).

The high porosity of ASW (Bossa et al., 2014; Bossa et al., 2015; Cazaux et al., 2015) makes it a good catalyst for gas-surface reactions involving oxygen (Ioppolo et al., 2011a; Romanzin et al., 2011; Chaabouni et al., 2012; Minissale et al., 2013a; Dulieu et al., 2017; Pezzella et al., 2018; Pezzella and Meuwly, 2019; Christianson and Garrod, 2021), hydrogen (Hama and Watanabe, 2013), carbonaceous (Minissale et al., 2013b; Minissale et al., 2016a; Qasim et al., 2020; Molpeceres et al., 2021) and nitrogen-containing (Minissale et al., 2014) species and helps maintaining those species on (Minissale et al., 2019) or inside ASW. (Minissale et al., 2016b; Tsuge et al., 2020). This increases the probability for the reaction partners to diffuse to locations for collisions and association reactions to occur. As the diffusivity of individual atoms and small molecules has been established from both, experiments and simulations, (Minissale et al., 2013b; Lee and Meuwly, 2014; Pezzella et al., 2018), this is a likely scenario for formation of molecules on and within ASW.

Earlier thermoluminescence experiments suggested that the O(3P)+CO(1Σ+) reaction with both reaction partners in their electronic ground state yields excited CO2* which, after emission of a photon, leads to formation of CO2. (Fournier et al., 1979). Such a process has also been proposed to occur on interstellar grains (Ruffle and Herbst, 2001) and has been confirmed experimentally (Minissale et al., 2013b) with an estimated entrance barrier of 0.014–0.103 eV for the process on ASW, compared with a value of 0.3 eV from high-level electronic structure calculations. (Veliz et al., 2021). The surrounding water matrix provides the necessary coupling (Roser et al., 2001) to facilitate relaxation of the 3A’ or 3A ″ states of CO2 to the 1A’ ground state (correlating with linear Σg+1 ). The presence of an entrance barrier for the O(3P)+CO(1Σ+) reaction has one led to consider the alternative CO + OH pathway for CO2 formation. (Watanabe and Kouchi, 2002; Ioppolo et al., 2011b). This was, however, reconsidered to yield the HOCO intermediate in such environments in more recent experiments. (Qasim et al., 2019). Also, the reaction products of the CO + OH reaction have been found to depend on the experimental conditions. (Oba et al., 2010; Noble et al., 2011). As the CO + OH reaction also appears to have a barrier in the entrance channel, (Noble et al., 2011), attention has recently shifted to the HOCO + H reaction for CO2 formation. (Qasim et al., 2019).

For adsorbed species to react on ASW they need to be able to diffuse. This has been demonstrated from MD simulations with diffusion coefficients and desorption energies consistent with experiments. (Lee and Meuwly, 2014; Ghesquière et al., 2015). Atomic oxygen (Pezzella et al., 2018) on ASW experiences diffusional barriers between E dif = 0.2 kcal/mol and 2 kcal/mol (100–1000 K) compared with values of Edif=990-360+530 K determined from experiments. (Minissale et al., 2016b). For CO, MD simulations reported (Pezzella and Meuwly, 2019) desorption energies between 3.1 and 4.0 kcal/mol (1560–2012 K or 130–170 meV), compared with 120 meV from experiments. (Karssemeijer et al., 2013). It was also found that the CO desorption energy from ASW depends on CO coverage with ranges from E des = 1700 K for low to 1000 K for high coverage (He et al., 2016) which is consistent with the simulations. (Pezzella and Meuwly, 2019). On non-porous and crystalline water surfaces submonolayer desorption energies for CO are 1307 and 1330 K ( ∼115 meV), respectively. (Noble et al., 2012). Experimental diffusional barriers range from 350 ± 50 K (Kouchi et al., 2020) to 490 ± 12 K. (He et al., 2018).

As such association reactions are in general exothermic, the energy released needs to be transferred to environmental degrees of freedom for the reaction products to stabilize. This is the quest of the present work which investigates the time scale and degrees of freedom to receive the energy liberated for the O(1D)+CO(1Σ+) reaction to form ground state CO2( Σg+1 ). The chemical precursors for formation of CO2 are believed to be carbon monoxide and atomic oxygen and the CO + O reaction has been proposed as a non-energetic pathway, close to conditions in interstellar environments, for CO2 formation 20 years ago from experiments involving a water-ice cap on top of CO and O deposited on a copper surface. (Roser et al., 2001). Formation of CO2( Σg+1 ) from ground state CO(1Σ+) and electronically excited O(1D) is barrierless. The excited atomic oxygen species can, for example, be generated from photolysis of H2O (Stief et al., 1975) which has a radiative lifetime of 110 min (Garstang, 1951). An alternative pathway proceeds via electron-induced neutral dissociation of water into H2 + O(1D). (Schmidt et al., 2019). In the presence of CO formation of CO2 in cryogenic CO/H2O films was observed. (Schmidt et al., 2019).

After recombination O(1D)+CO(1Σ+) → CO2( Σg+1 ) the product is in a highly vibrationally excited state. For it to stabilize, excess internal energy needs to be channeled into the environment which is the ASW. The present work characterizes and quantifies energy relaxation of the CO2( Σg+1 ) product into internal and translational degrees of freedom of the surrounding water matrix. First, the methods used are described. Then, results are presented and discussed. Finally, conclusions are drawn.

2 Computational Methods

All molecular dynamics (MD) simulations were carried out using the CHARMM suite of programs (Brooks et al., 2009) with provisions for bond forming reactions through multi state adiabatic reactive MD (MS-ARMD). (Nagy and Yosa Reyes, 2014). The simulation system, Figure 1, consisted of an equilibrated cubic box of amorphous solid water with dimension 31 × 31 × 31 Å3 containing 1000 water molecules. As all bonds and angles are flexible, the simulations were run with a time step of Δt = 0.1 fs and the non-bonded cutoff was at 13 Å. Simulations were started from an existing, equilibrated ASW structure (Pezzella et al., 2018; Pezzella and Meuwly, 2019; Upadhyay et al., 2021) by adding COA and OB inside (Figure 1A) or on top of (Figure 1B) ASW.

FIGURE 1.

FIGURE 1

Open in a new tab

The simulation system for studying the O(1D)+CO(1Σ+)→ CO2( Σg+1 ) recombination reaction. (A): CO and O trapped inside a cavity of ASW; (B): CO and O on the top of the ASW surface.

In the following, the coordinates are the CO stretch r, the separation R between the center of mass of COA and OB and θ is the OACOB angle. In addition, the C–OB separation will be considered where appropriate. Initial conditions were generated for a grid of angles θ and separations R and simulations were carried out to obtain initial coordinates and velocities for each of the grid points. With constrained CO and O position, first 750 steps of steepest descent and 100 steps Adopted Basis Newton-Raphson minimization were carried out, followed by 50 ps heating dynamics to 50 K. Then, 100 ps equilibration dynamics was carried out. From each of the runs coordinates and velocities were saved regularly to obtain initial conditions for each combination of angle and distance. Production simulations 500 ps or 6 ns in length were then run from saved coordinates and velocities in the NVE ensemble. Data (energies, coordinates and velocities) were saved every 1000 steps for subsequent analysis.

Water was described by a reparametrized, (Burnham et al., 1997; Plattner and Meuwly, 2008), flexible KKY (Kumagai, Kawamura, Yokokawa) model. (Kumagai et al., 1994). The typical water modes that couple in the ∼2000cm−1 region relevant in the present work are the water bend (1,600 cm−1) and the framework rotation (600 cm−1) as was also found for the vibrational relaxation of cyanide in water. (Lee and Meuwly, 2011). To describe COA + OB recombination to form CO2 the Morse-Morse-Harmonic (MMH) parametrization was employed. (Upadhyay et al., 2021). This model treats the two CO bonds with a Morse potential and the OCO bend as a harmonic function. MMH is a computationally efficient model (fitted to MRCI/aug-cc-pVTZ data), which yields results for recombination probabilities on ASW comparable to a more elaborate reproducing kernel Hilbert space (RKHS) representation with an exothermicity of –7.27 eV (Veliz et al., 2021; Upadhyay et al., 2021).

For CO2, the partial charges were q O = − 0.3e and q C = 0.6e with standard van der Waals parameters from CHARMM. These charges are consistent with those obtained from B3LYP/6-31G (d,p) calculations snapshots from the MD simulations with CO2 adsorbed to a small water cluster (H2O)10 which yield q C = 0.73e and q O = − 0.35e. This compares with charges of q C = 0.22e and q O = − 0.21e for the CO molecule and q O = − 0.1e for an oxygen atom adsorbed to (H2O)10. To assess the dependence of the results on the partial charges used, additional reactive MD simulations using the MMH parametrization were carried out with q O = − 0.1e and q C = 0.2e (i.e., q CO = 0.1e) and with q O = − 0.2e and q C = 0.4e (i.e., q CO = 0.2e). In all cases, recombination was found to speed up compared with q CO = 0.3e and q O = − 0.3e due to the increased mobility of the CO molecule and the O atom on the ASW when reduced partial charges are used.

The main focus of the present work is to study energy redistribution within the system following recombination of COA + OB to form CO2. For this, the average total, translational and internal energy of the water molecules is analyzed for recombination on top of and inside ASW. Both, the time scale and amount of energy dissipated into translational and internal degrees of freedom was determined. The translational energy for the water molecules at each timestep was determined by first calculating the magnitude of the linear momentum of each water molecule from the stored velocities. From this the translational energy contribution is calculated for each water molecule considered and the total translational energy is accumulated. The internal energy for each water molecule is determined from the difference of the total kinetic energy and the translational energy.

3 Results and Discussion

In the following, the energy distribution in the water matrix of the ASW is separately discussed on the ∼100 ps and on the nanosecond time scale. Next, the energy flow away from the recombination site is analyzed and, finally, the energy redistribution to neighboring water molecules surrounding the recombination site is considered.

3.1 Recombination on the 100 ps Time Scale

A typical trajectory for COA + OB recombination inside the ASW cavity is shown in Figure 2 (left column). Initially, the C–OB separation is ∼6 Å (Figure 2A). Within 150 ps recombination takes place and angular distortions lead to exploration of angles θ OCO ∼ 90° (Figure 2B). Relaxation of the angle occurs within the following 50 ps and the CO2 molecule remains in an internally excited state on much longer time scales, see Figure 2C. (Upadhyay et al., 2021). Concomitantly, the average internal energy of the surrounding water molecules increases by about 10%, see black, red and green traces in Figures 2D,E. The translational (phononic) modes (green) acquire approximately 1/3 of the additional energy whereas the internal energy (red) increases by the remaining 2/3.

FIGURE 2.

FIGURE 2

Open in a new tab

Recombination of O (1D)+CO(1Σ+) to form ground state CO2(Σg1) in the ASW cavity (left column) and on top of the ASW surface (right column). Initially, R = 6.0 Å and θ = 180°. (A): OB–CCO separation (red) and COA separation (black); (B): the O-C-O angle θ; (C–E): the average total (black), internal (red), and translational (green) energies for the CO2 molecule (C), the average per water molecule (D), and the magnitude of the average per water molecule relative to the energy before recombination (E).

Figure 2 (right column) reports a recombination trajectory on top of ASW. In this case, recombination takes place after ∼35 ps and wide angular excursions extend out to 100 ps The amount of energy picked up by the water matrix is larger compared to recombination inside ASW (Figures 2D,E). The average total energy per water molecule increases by close to 20% and the amount that goes into internal degrees of freedom is considerably larger. For the translational modes, the energy after recombination is comparable to that for recombination within the cavity.

From a set of 70 recombination trajectories for the reaction within the cavity and on top of the ASW surface, the averaged energy contents in translational, internal and all degrees of freedom of the water molecules were determined (Figure 3). For this analysis, the time of reaction was set to zero (t = 0) to align all reactive trajectories and all energies are reported relative to the averages before recombination. The translational contribution for recombination within and on top of ASW re-equilibrates on the ∼ 25 ps time scale after which no change in the phononic degrees of freedom is observed. Contrary to that, the internal degrees of freedom (red traces) show temporal evolution on two time scales: a rapid phase on the picosecond time scale, followed by a slow, long increase in the internal energies. This is also reflected in the averaged total energy (black).

FIGURE 3.

FIGURE 3

Open in a new tab

Average total (black), internal (red) and translational (green) energies for water over 70 independent runs relative to the average before recombination. The time of reaction for all trajectories is shifted to t =0 and defined by the first instance at which rC-OB<1.6 Å. (A): recombination within the ASW cavity. (B): recombination for COA + OB on the top of the ASW surface. The blue solid line is a fit to an empirical expression ϵ=a0e−t/a1+a2t+a3 , see text.

As for the single trajectories, the amount of energy released from the recombination reaction into the translational degrees of freedom is similar for the reaction inside the cavity and on top of the ASW surface. For the internal degrees of freedom, however, recombination on top of the ASW surface leads on an average increase per water molecule by 0.075 kcal/mol within 400 ps (Figure 3B) compared with 0.06 kcal/mol for the process inside the cavity. Also, there is a characteristic decrease in the internal contribution for recombination on the surface after 15 ps which is even present when averaging over 70 independent runs. This feature is not found for recombination within ASW.

To estimate approximate time scales for the different processes involved, the average energies were fitted to an empirical expression ϵ=a0e−t/a1+a2t+a3 where ϵ is any of the energies considered. Such a functional form was chosen after inspection of the data in Figure 3 and accounts for the rapid initial increase in the three energies together with the slow variation of the internal energy on longer times. This parametrization is not able to model the dip around 15 ps for recombination on to of the surface, though. The time scales a 1 for total, internal, and translational energies are (4.8, 2.9, 7.1) ps for recombination inside the cavity and speed up to (3.9, 1.9, 6.1) ps for the process on the ASW surface. It is of interest to note that the rapid time scale for the internal energy is considerably faster than the kinetics of the translational degrees of freedom for both types of recombinations. The parameter a 2 which describes the slow increase of internal energy has a value of a 2 = 4.3 × 10–2 (kcal/mol)/ns for recombination in the cavity and a 2 = 3.8 × 10–2 (kcal/mol)/ns for the reaction on the surface, and is vanishingly small for the translational energy.

Average internal energies from representative independent runs for recombination inside the cavity and on top of the ASW surface are shown in Supplementary Figures S1, S2. For recombination inside the cavity (Supplementary Figure S1) the results confirm that the energy content in the internal degrees of freedom increases considerably faster than for the translation. Also, it is found that the amount of energy transferred to translation after recombination is smaller than that partitioned into internal degrees of freedom. For recombination on the ASW surface the same observations are made. In addition, the pronounced maximum after ∼5 ps is present in all examples shown in Supplementary Figure S2. To provide a molecularly resolved interpretation of this feature the HOH angle time series θ(t) was analyzed for a trajectory in which CO + O recombination occurred after 35 ps, see Supplementary Figure S3B. At the time of reaction the water bending angle decreases from its average equilibrium value by ⟨Δθ⟩∼ 1° over the next 70 ps after which it relaxes back to the original value. The signature in the internal energy extends over ∼30 ps, see Supplementary Figure S2. Hence, it is possible that changes in the average water geometry following CO + O recombination are responsible for the overshooting and subsequent relaxation of the internal energy for the reaction on the surface. The HOH angle for a simulation within ASW in Supplementary Figure S3A also shows a slight adjustment of the valence angle after recombination. Contrary to the situation on the surface, the average angle does not relax to the value before recombination, though.

3.2 Recombination Dynamics on Longer Time Scales

It is also of interest to analyze the energy redistribution on the multi-nanosecond time scale. Figure 4A demonstrates that the average total kinetic energy per water molecule continuously increases even on the nanosecond time scale. Most of this increase is due to the internal degrees of freedom although the translational component also shows a continuous slow increase on the nanosecond time scale.

FIGURE 4.

FIGURE 4

Open in a new tab

Total (black), internal (red), and translational (green) energies for water molecules (A) and for the recombined CO2 molecule (B) form a 6 ns rebinding trajectory on the top of the ASW surface. The time of reaction is shifted to t = 0 and defined by the first instance at which rC-OB<1.6 Å. CO2 continues to relax and the energy in the ASW further increases beyond the maximum simulation time of 6 ns after CO2 recombination.

The relaxation of the CO2 internal energy is reported in Figure 4B. Within the first few picoseconds (inset) the internal energy is quenched to ∼10  kcal/mol after which two relaxations are observed. A first phase during 1 nanosecond following recombination and a second, slower phase extending out to 6 ns and beyond. By the end of the simulation the average internal energy of the CO2 molecule has decreased to ∼2.5  kcal/mol on average. Hence, it is expected that energy transfer to the surrounding water continues but slows down considerably on the 10 ns time scale and longer.

3.3 Energy Migration Around the Recombination Site

For a positionally resolved picture of energy flow the simulation system was separated in voxels with dimension 31 × 1 × 1 Å3. The kinetic energy of all water molecules within one such voxel was averaged along the trajectory and projected onto the (y, z) − plane. Which water molecules belong to a particular voxel was decided based on the water-oxygen atom coordinates. Figure 5A reports the distribution of total kinetic energy distribution before recombination. The recombination site is at (y = 2, z = 2) Å and marked as a large cross. Within the first 5 ps after recombination the kinetic energy of water molecules within ∼ 10 Å of the recombination site increases considerably, by up to a factor of 4. Following this, energy redistributes continuously across the entire surface on the 200 ps time scale, see panels C to F.

FIGURE 5.

FIGURE 5

Open in a new tab

Kinetic energy of water molecules projected onto the (y, z) − plane averaged over seven independent simulations on the top layer of the ASW surface. Recombination of CO and O takes place at the location labelled with “X”. Before recombination (A): − 5 ≤ t ≤ 0 ps) and after recombination (B): 0 ≤ t ≤ 5 ps, (C): 5 ≤ t ≤ 10 ps, (D): 10 ≤ t ≤ 50 ps, (E): 50 ≤ t ≤ 100 ps and (F): 100 ≤ t ≤ 200 ps). The average position of CO is indicated by a large black cross. Noteworthy regions are labelled I to III and surrounded by solid lines. Region I is cool at early times and gradually warms up. Region II remains cool for most of the simulation time and region III alternates between cool and warm. For results within 10 Å of the surface, see Supplementary Figure S4.

Certain regions that are initially “cold” (blue)—e.g., the region labelled “I” at (y = 5, z = − 5) Å in Figure 5—warm up as energy transfer from CO2 to the water molecules occurs. Conversely, other regions remain “cool”, such as region “II” around (y = 5, z = 5) Å for which the color code remains blue until 200 ps. Yet for other regions, such as “III”, the total kinetic energy oscillates between cooler and warmer. It is also instructive to include only the first few ASW layers in this analysis which was done in Supplementary Figure S4. Here, the voxels have sizes 1 × 1 × 1 A3. For one, the cool regions are more extended before recombination. After recombination energy transfer occurs in a similar fashion as for the full system. However, the warm regions are less extended. This suggests that energy transfer also occurs to a considerable extent into the bulk rather than across the surface of the ASW even for recombination on top of ASW.

3.4 Energy Flow to Nearby Water Molecules

Finally, individual water molecules in immediate proximity of the recombination site are analyzed. For one trajectory with recombination on the ASW surface the average total, internal, and translational energies for the 5 water molecules closest to the recombination site are reported in Figure 6. During the first 10 ps after recombination the average total kinetic energy increases by up to 0.6 kcal/mol per water molecule. Conversely, the translational energy contribution fluctuates around zero which indicates that the local structure of ASW remains intact and most of the energy flows into internal degrees of freedom.

FIGURE 6.

FIGURE 6

Open in a new tab

Average total (black), internal (red), and translational (green) energies for 5 water molecules (A) closest to the CO2, and the CO2 molecule (B) formed from a recombination trajectory on the ASW surface. The time of reaction is shifted to t = 0 and defined by the first instance at which rC-OB<1.6 Å. The initial time scale for energy redistribution from the relaxing CO2 molecule to internal and translational degrees of freedom of the ASW occurs on the 10 ps time scale with slow gradual relaxation on the ∼100 ps to ns time scale. On the 500 ps time scale the 5 water molecules slightly cool compared with the kinetic energy before recombination.

After this initial increase, cooling of these few nearby water molecules takes place with a long-time average of −0.1 kcal/mol per water molecule in the internal degrees of freedom. On the 500 ps no noticeable change in the translational energy content is observed. For the CO2 molecule (Figure 6B) the translational energy remains small throughout the trajectory whereas the internal energy decreases rapidly within the first 5 ps following recombination. Subsequently, slow gradual cooling on the 100 ps to nanosecond time scale takes place as was already found earlier, Figure 4.

4 Discussion and Conclusion

The present work reports on the energy redistribution across ASW following O(1D)+CO(1Σ+) recombination to form CO2(Σg+1) on the surface and in a cavity. It is found that energy distribution occurs in two phases, one on the picosecond and one on the nanosecond time scale for both locations. Although the time dependence of the processes is similar for the two different recombination sites (inside vs. on top), the dynamics differs in a number of ways. Firstly, recombination on the surface leads to excess internal energy on the picosecond time scale which subsequently relaxes and additional energy transfer into water modes occurs on longer time scales. Secondly, recombination within the cavity considered here leads to smaller magnitude (∼ 15%) of energy transferred per water molecule compared with the process on the surface (∼ 25%). A possible reason for this is that within a sufficiently large cavity the recombined CO2 molecule exchanges energy with the surrounding through direct collision whereas on the surface CO2 is always in contact with the ASW. In other words, the coupling between CO2 and water differs for recombination within ASW and on top of it. Finally, heating of the water molecules occurs on the 10 ps time scale following the recombination reaction. Consistent with earlier work, (Fredon et al., 2021; Upadhyay et al., 2021), no CO2 desorption is found from the simulations carried out here.

It is of interest to note that - ultimately - energy redistribution in such systems follows quantum mechanical principles. The present results suggest that the local energy generated from CO + O recombination is probably sufficient to excite internal modes of individual water molecules surrounding the recombination site. Hence, after CO + O recombination the ASW will be in a state characterized by a few internally and vibrationally excited water molecules embedded into a matrix of water molecules in the ground state. Earlier work on a related problem—the vibrational relaxation of a quantum oscillator coupled to oscillators of a biomolecule (Stock, 2009)—found that using classical mechanics leads to qualitatively correct results compared with a full quantum treatment. For the relaxation times a moderate factor of two for the difference between classical and rigorous quantum simulations was reported. Hence, for the present problem it is also expected that similar conclusions apply and that the nonequilibrium relaxation dynamics of individual vibrationally excited water molecules surrounded by vibrationally cold water molecules can be captured qualitatively from using classical dynamics.

In summary, the present work demonstrates that O(1D)+CO(1Σ+) recombination to form CO2(Σg+1) leads to excitation of both, phononic and internal modes of the water molecules that constitute the ASW. The time scales for this are on the pico- and nano-second and lead to warming the water matrix. Water molecules in direct proximity of the recombination site may become vibrationally excited and the time scale for their relaxation back to the ground state will depend on the coupling to the immediate environment. Full relaxation of the CO2 molecule is expected to require several 10–100 nanoseconds.

Acknowledgments

The authors gratefully acknowledge financial support from the Swiss National Science Foundation through grant 200 021-117 810 and to the NCCR-MUST.

Data Availability Statement

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation. Data on the simulations and the PESs is available at: https://github.com/MMunibas/co2.asw and https://github.com/MMunibas/CO2-PESs.

Author Contributions

MM designed research, MU carried out simulations, MU and MM analyzed results, MU and MM wrote manuscript.

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fchem.2021.827085/full#supplementary-material

Click here for additional data file. (1.5MB, pdf)

References

  1. Bar-Nun A., Dror J., Kochavi E., Laufer D. (1987). Amorphous Water Ice and its Ability to Trap Gases. Phys. Rev. B 35, 2427–2435. 10.1103/physrevb.35.2427 [DOI] [PubMed] [Google Scholar]
  2. Bossa J.-B., Isokoski K., de Valois M. S., Linnartz H. (2012). Thermal Collapse of Porous Interstellar Ice. A&A 545, A82. 10.1051/0004-6361/201219340 [DOI] [Google Scholar]
  3. Bossa J.-B., Isokoski K., Paardekooper D. M., Bonnin M., van der Linden E. P., Triemstra T., et al. (2014). Porosity Measurements of Interstellar Ice Mixtures Using Optical Laser Interference and Extended Effective Medium Approximations. A&A 561, A136. 10.1051/0004-6361/201322549 [DOI] [Google Scholar]
  4. Bossa J.-B., Maté B., Fransen C., Cazaux S., Pilling S., Rocha W. R. M., et al. (2015). Porosity and Band-Strength Measurements of Multi-phase Composite Ices. ApJ 814, 47. 10.1088/0004-637x/814/1/47 [DOI] [Google Scholar]
  5. Brooks B. R., Mackerell A. D., Nilsson L., Petrella R. J., Roux B., Won Y., et al. (2009). CHARMM: the Biomolecular Simulation Program. J. Comput. Chem. 30, 1545–1614. 10.1002/jcc.21287 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Burnham C. J., Li J.-C., Leslie M. (1997). Molecular Dynamics Calculations for Ice Ih. J. Phys. Chem. B 101, 6192–6195. 10.1021/jp9632596 [DOI] [Google Scholar]
  7. Cazaux S., Bossa J.-B., Linnartz H., Tielens A. G. G. M. (2015). Pore Evolution in Interstellar Ice Analogues. A&A 573, A16. 10.1051/0004-6361/201424466 [DOI] [Google Scholar]
  8. Chaabouni H., Minissale M., Manicò G., Congiu E., Noble J. A., Baouche S., et al. (2012). Water Formation through O2 + D Pathway on Cold Silicate and Amorphous Water Ice Surfaces of Interstellar Interest. J. Chem. Phys. 137, 234706–706. 10.1063/1.4771663 [DOI] [PubMed] [Google Scholar]
  9. Christianson D. A., Garrod R. T. (2021). Chemical Kinetics Simulations of Ice Chemistry on Porous versus Non-porous Dust Grains. Front. Astro. Space Sci. 8, 21. 10.3389/fspas.2021.643297 [DOI] [Google Scholar]
  10. Dulieu F., Minissale M., Bockelée-Morvan D. (2017). Production of O2 through Dismutation of H2O2 during Water Ice Desorption: a Key to Understanding Comet O2 Abundances. A&A 597, A56. 10.1051/0004-6361/201628445 [DOI] [Google Scholar]
  11. Fournier J., Deson J., Vermeil C., Pimentel G. C. (1979). Fluorescence and Thermoluminescence of N2O, CO, and CO2 in an Argon Matrix at Low Temperature. J. Chem. Phys. 70, 5726–5730. 10.1063/1.437399 [DOI] [Google Scholar]
  12. Fredon A., Radchenko A. K., Cuppen H. M. (2021). Quantification of the Role of Chemical Desorption in Molecular Clouds. Acc. Chem. Res. 54, 745–753. 10.1021/acs.accounts.0c00636 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Garstang R. H. (1951). Energy Levels and Transition Probabilities in P2 and P4 Configurations. Monthly Notices R. Astronomical Soc. 111, 115–124. 10.1093/mnras/111.1.115 [DOI] [Google Scholar]
  14. Ghesquière P., Mineva T., Talbi D., Theulé P., Noble J. A., Chiavassa T. (2015). Diffusion of Molecules in the Bulk of a Low Density Amorphous Ice from Molecular Dynamics Simulations. Phys. Chem. Chem. Phys. 17, 11455–11468. 10.1039/c5cp00558b [DOI] [PubMed] [Google Scholar]
  15. Hagen W., Tielens A. G. G. M., Greenberg J. M. (1981). The Infrared Spectra of Amorphous Solid Water and Ice Ic between 10 and 140 K. Chem. Phys. 56, 367–379. 10.1016/0301-0104(81)80158-9 [DOI] [Google Scholar]
  16. Hama T., Watanabe N. (2013). Surface Processes on Interstellar Amorphous Solid Water: Adsorption, Diffusion, Tunneling Reactions, and Nuclear-Spin Conversion. Chem. Rev. 113, 8783–8839. 10.1021/cr4000978 [DOI] [PubMed] [Google Scholar]
  17. He J., Acharyya K., Vidali G. (2016). Binding Energy of Molecules on Water Ice: Laboratory Measurements and Modeling. ApJ 825, 89. 10.3847/0004-637x/825/2/89 [DOI] [Google Scholar]
  18. He J., Emtiaz S., Vidali G. (2018). Measurements of Diffusion of Volatiles in Amorphous Solid Water: Application to Interstellar Medium Environments. ApJ 863, 156. 10.3847/1538-4357/aad227 [DOI] [Google Scholar]
  19. Ioppolo S., Cuppen H. M., Linnartz H. (2011). Surface Formation Routes of Interstellar Molecules: Hydrogenation Reactions in Simple Ices. Rend. Fis. Acc. Lincei 22, 211–224. 10.1007/s12210-011-0135-3 [DOI] [Google Scholar]
  20. Ioppolo S., Van Boheemen Y., Cuppen H. M., Van Dishoeck E. F., Linnartz H. (2011). Surface Formation of CO2 Ice at Low Temperatures. Mon. Not. R. Astron. Soc. 413, 2281–2287. 10.1111/j.1365-2966.2011.18306.x [DOI] [Google Scholar]
  21. Jenniskens P., Blake D. F. (1994). Structural Transitions in Amorphous Water Ice and Astrophysical Implications. Science 265, 753–756. 10.1126/science.11539186 [DOI] [PubMed] [Google Scholar]
  22. Karssemeijer L., Ioppolo S., van Hemert M., van der Avoird A., Allodi M., Blake G., et al. (2013). Dynamics of CO in Amorphous Water-Ice Environments. Astrophys. J. 781, 1–15. 10.1088/0004-637x/781/1/16 [DOI] [Google Scholar]
  23. Keane J. V., Tielens A. G. G. M., Boogert A. C. A., Schutte W. A., Whittet D. C. B. (2001). Ice Absorption Features in the 5-8μm Region toward Embedded Protostars. A&A 376, 254–270. 10.1051/0004-6361:20010936 [DOI] [Google Scholar]
  24. Kouchi A., Furuya K., Hama T., Chigai T., Kozasa T., Watanabe N. (2020). Direct Measurements of Activation Energies for Surface Diffusion of CO and CO2 on Amorphous Solid Water Using In Situ Transmission Electron Microscopy. ApJL 891, L22. 10.3847/2041-8213/ab78a2 [DOI] [Google Scholar]
  25. Kouchi A., Tsuge M., Hama T., Oba Y., Okuzumi S., Sirono S.-i., et al. (2021). Transmission Electron Microscopy Study of the Morphology of Ices Composed of H2O, CO2, and CO on Refractory Grains. ApJ 918, 45. 10.3847/1538-4357/ac0ae6 [DOI] [Google Scholar]
  26. Kumagai N., Kawamura K., Yokokawa T. (1994). An Interatomic Potential Model for H2O: Applications to Water and Ice Polymorphs. Mol. Simulation 12, 177–186. 10.1080/08927029408023028 [DOI] [Google Scholar]
  27. Lee M. W., Meuwly M. (2014). Diffusion of Atomic Oxygen Relevant to Water Formation in Amorphous Interstellar Ices. Faraday Discuss. 168, 205–222. 10.1039/c3fd00160a [DOI] [PubMed] [Google Scholar]
  28. Lee M. W., Meuwly M. (2011). On the Role of Nonbonded Interactions in Vibrational Energy Relaxation of Cyanide in Water. J. Phys. Chem. A. 115, 5053–5061. 10.1021/jp202503m [DOI] [PubMed] [Google Scholar]
  29. Minissale M., Congiu E., Baouche S., Chaabouni H., Moudens A., Dulieu F., et al. (2013). Quantum Tunneling of Oxygen Atoms on Very Cold Surfaces. Phys. Rev. Lett. 111, 053201. 10.1103/PhysRevLett.111.053201 [DOI] [PubMed] [Google Scholar]
  30. Minissale M., Congiu E., Dulieu F. (2016). Direct Measurement of Desorption and Diffusion Energies of O and N Atoms Physisorbed on Amorphous Surfaces. A&A 585, A146. 10.1051/0004-6361/201526702 [DOI] [Google Scholar]
  31. Minissale M., Congiu E., Manicò G., Pirronello V., Dulieu F. (2013). CO2 formation on Interstellar Dust Grains: a Detailed Study of the Barrier of the CO + O Channel. A&A 559, A49. 10.1051/0004-6361/201321453 [DOI] [Google Scholar]
  32. Minissale M., Fedoseev G., Congiu E., Ioppolo S., Dulieu F., Linnartz H. (2014). Solid State Chemistry of Nitrogen Oxides - Part I: Surface Consumption of NO. Phys. Chem. Chem. Phys. 16, 8257–8269. 10.1039/c3cp54917h [DOI] [PubMed] [Google Scholar]
  33. Minissale M., Moudens A., Baouche S., Chaabouni H., Dulieu F. (2016). Hydrogenation of CO-bearing Species on Grains: Unexpected Chemical Desorption of CO. Mon. Not. R. Astron. Soc. 458, 2953–2961. 10.1093/mnras/stw373 [DOI] [Google Scholar]
  34. Minissale M., Nguyen T., Dulieu F. (2019). Experimental Study of the Penetration of Oxygen and Deuterium Atoms into Porous Water Ice. A&A 622, A148. 10.1051/0004-6361/201833830 [DOI] [Google Scholar]
  35. Molpeceres G., Kästner J., Fedoseev G., Qasim D., Schömig R., Linnartz H., et al. (2021). Carbon Atom Reactivity with Amorphous Solid Water: H2O-Catalyzed Formation of H2CO. J. Phys. Chem. Lett. 12, 10854–10860. 10.1021/acs.jpclett.1c02760 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Nagy T., Yosa Reyes J. (2014). Multisurface Adiabatic Reactive Molecular Dynamics. J. Chem. Theor. Comput. 10, 1366–1375. 10.1021/ct400953f [DOI] [PubMed] [Google Scholar]
  37. Noble J. A., Dulieu F., Congiu E., Fraser H. J. (2011). CO2 Formation in Quiescent Clouds: An Experimental Study of the CO + OH Pathway. ApJ 735, 121. 10.1088/0004-637x/735/2/121 [DOI] [Google Scholar]
  38. Noble J., Congiu E., Dulieu F., Fraser H. (2012). Thermal Desorption Characteristics of CO, O2 and CO2 on Non-porous Water, Crystalline Water and Silicate Surfaces at Submonolayer and Multilayer Coverages. Mon. Not. R. Astron. Soc. 421, 768–779. [Google Scholar]
  39. Oba Y., Miyauchi N., Hidaka H., Chigai T., Watanabe N., Kouchi A. (2009). Formation of Compact Amorphous H2O Ice by Codeposition of Hydrogen Atoms with Oxygen Molecules on Grain Surfaces. ApJ 701, 464–470. 10.1088/0004-637x/701/1/464 [DOI] [Google Scholar]
  40. Oba Y., Watanabe N., Kouchi A., Hama T., Pirronello V. (2010). Experimental Study of CO2 Formation by Surface Reactions of Non-energetic OH Radicals with CO Molecules. ApJ 712, L174–L178. 10.1088/2041-8205/712/2/l174 [DOI] [Google Scholar]
  41. Pezzella M., Meuwly M. (2019). O2 Formation in Cold Environments. Phys. Chem. Chem. Phys. 21, 6247–6255. 10.1039/c8cp07474g [DOI] [PubMed] [Google Scholar]
  42. Pezzella M., Unke O. T., Meuwly M. (2018). Molecular Oxygen Formation in Interstellar Ices Does Not Require Tunneling. J. Phys. Chem. Lett. 9, 1822–1826. 10.1021/acs.jpclett.8b00328 [DOI] [PubMed] [Google Scholar]
  43. Plattner N., Meuwly M. (2008). Atomistic Simulations of CO Vibrations in Ices Relevant to Astrochemistry. ChemPhysChem 9, 1271–1277. 10.1002/cphc.200800030 [DOI] [PubMed] [Google Scholar]
  44. Qasim D., Fedoseev G., Chuang K.-J., He J., Ioppolo S., van Dishoeck E. F., et al. (2020). An Experimental Study of the Surface Formation of Methane in Interstellar Molecular Clouds. Nat. Astron. 4, 781–785. 10.1038/s41550-020-1054-y [DOI] [Google Scholar]
  45. Qasim D., Lamberts T., He J., Chuang K.-J., Fedoseev G., Ioppolo S., et al. (2019). Extension of the HCOOH and CO2 Solid-State Reaction Network during the CO Freeze-Out Stage: Inclusion of H2CO. A&A 626, A118. 10.1051/0004-6361/201935068 [DOI] [Google Scholar]
  46. Romanzin C., Ioppolo S., Cuppen H. M., van Dishoeck E. F., Linnartz H. (2011). Water Formation by Surface O3 Hydrogenation. J. Chem. Phys. 134, 084504–504. 10.1063/1.3532087 [DOI] [PubMed] [Google Scholar]
  47. Roser J. E., Vidali G., Manicò G., Pirronello V. (2001). Formation of Carbon Dioxide by Surface Reactions on Ices in the Interstellar Medium. Astrophys. J. Lett. 555, L61–L64. 10.1086/321732 [DOI] [Google Scholar]
  48. Ruffle D. P., Herbst E. (2001). New Models of Interstellar Gas-Grain Chemistry - III. Solid CO2 . Mon. Not. R. Astron. Soc. 324, 1054–1062. 10.1046/j.1365-8711.2001.04394.x [DOI] [Google Scholar]
  49. Schmidt F., Swiderek P., Bredehöft J. H. (2019). Formation of Formic Acid, Formaldehyde, and Carbon Dioxide by Electron-Induced Chemistry in Ices of Water and Carbon Monoxide. ACS Earth Space Chem. 3, 1974–1986. 10.1021/acsearthspacechem.9b00168 [DOI] [Google Scholar]
  50. Stief L. J., Payne W. A., Klemm R. B. (1975). A Flasch Photolysis-Resonance Fluorescence Study of the Formation of O(1D) in the Photolysis of Water and the Reaction of O(1D) with H2, Ar, and He −. J. Chem. Phys. 62, 4000–4008. 10.1063/1.430323 [DOI] [Google Scholar]
  51. Stock G. (2009). Classical Simulation of Quantum Energy Flow in Biomolecules. Phys. Rev. Lett. 102, 118301. 10.1103/physrevlett.102.118301 [DOI] [PubMed] [Google Scholar]
  52. Tsuge M., Hidaka H., Kouchi A., Watanabe N. (2020). Diffusive Hydrogenation Reactions of CO Embedded in Amorphous Solid Water at Elevated Temperatures ∼70 K. ApJ 900, 187. 10.3847/1538-4357/abab9b [DOI] [Google Scholar]
  53. Upadhyay M., Pezzella M., Meuwly M. (2021). Genesis of Polyatomic Molecules in Dark Clouds: CO2 Formation on Cold Amorphous Solid Water. J. Phys. Chem. Lett. 12, 6781–6787. 10.1021/acs.jpclett.1c01810 [DOI] [PubMed] [Google Scholar]
  54. Veliz J. C. S. V., Koner D., Schwilk M., Bemish R. J., Meuwly M. (2021). The C(3P) + O2( Σg−3 ) ↔ CO2 ↔ CO(1Σ+)+ O(1D)/O(3P) Reaction: Thermal and Vibrational Relaxation Rates from 15 K to 20000K. Phys. Chem. Chem. Phys. 23, 251–263. 10.1039/d1cp01101d [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Watanabe N., Kouchi A. (2002). Measurements of Conversion Rates of CO to CO2 in Ultraviolet‐induced Reaction of D2O(H2O)/CO Amorphous Ice. ApJ 567, 651–655. 10.1086/338491 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file. (1.5MB, pdf)

Data Availability Statement

The raw data supporting the conclusion of this article will be made available by the authors, without undue reservation. Data on the simulations and the PESs is available at: https://github.com/MMunibas/co2.asw and https://github.com/MMunibas/CO2-PESs.

Articles from Frontiers in Chemistry are provided here courtesy of Frontiers Media SA

RESOURCES