The optimized structures of intermediates and transition states are shown in Figure 2. The zero-point corrected PES for the ^•CH₂NH₂ + O₂ reaction is depicted in Figure 3, and enthalpies values are given in Table 1. In the current reaction system, the O₂ molecule attacks the radical carbon atom, which leads to the formation of the intermediate ^•OO-CH₂NH₂ (INT1). Several conformational isomers of INT1 were observed, and for simplicity, we have considered the lowest energy conformer in our calculation. The calculated stabilization energy for INT1 is −31.7 kcal mol^-1, which is in very good agreement with the reported values by Rissanen et al. (2014) and Zhang et al. (2020) This value is also in very good agreement with its isoelectronic reactions, i.e., the O₂ + ^•CH₂OH value (−31.9 kcal/mol) (Dash and Ali, 2022). The unpaired electron in INT1 resides at the terminal O-atom, which can be decomposed differently. The lowest energy channel is an isomerization process where the terminal O-atom attacks the H-atoms of the NH₂ group via five- and six-membered cyclic transition states (TS5 and TS1), leading to hydrogen-bonded five- and six-membered cyclic complexes, i.e., INT2 and QOOH, respectively. The calculated barrier heights for TS5 and TS1 are 21.9 and 36.1 kcal mol^-1, respectively, with respect to INT1, indicating that the isomerization reaction going through TS5, leading to the formation of INT2, is energetically more favorable than that going through TS1 to QOOH. Supplementary Figure S1 provides an IRC scan that confirms the connectivity of TS5 with INT2 and CH₂NH + HO₂ at the M06-2X/6-311++G (3df, 3pd). It is noted that the influence of the formation of hydrogen-bonded cyclic complexes may change the energetics and kinetics of the reaction system. In INT2, two strong hydrogen bonds are formed between the H-atom of the HO₂ and the N-atom of the CH₂NH (1.72 Å) and O-atom of the HO₂ radical and the H-atom of the CH₂NH (2.55 Å), which leads to the formation of a stable six-membered ring planar cyclic structure. In QOOH, a five-membered ring cyclic structure with one hydrogen bond is formed between the terminal O and H-atoms (2.47 Å). Therefore, INT2 is energetically 6.6 kcal mol^-1 more stable than QOOH. QOOH can further dissociate via TS2 and TS4 to form HO₂ + CH₂NH and OH + OCH₂NH, respectively. Because the barrier height for the formation of QOOH is very high, the formation of CH₂NH and OCH₂NH via QOOH may be negligible under tropospheric conditions. The terminal O-atom of INT1 can also attack the H-atom of a nearby C-atom, leading to the formation of a hydrogen-bonded six-membered cyclic intermediate, INT3, via a four-membered ring transition state (TS3) and subsequently dissociating to form OH + NH₂CHO. In INT3, two strong hydrogen bonds are formed between the H-atom of the OH radical and the O-atom of the NH₂CHO (1.91 Å) and the O-atom of the OH radical and the H-atom of the NH₂CHO (2.16 Å). INT3 is energetically the most stable structure in the PES, with a stabilization energy of −78.9 kcal mol^-1 from the reactants. The barrier height of this reaction channel is 41 kcal mol^-1, which is 20 kcal mol^-1 higher than that of the TS5 and may not contribute to the overall reaction kinetics (vide infra).

The other reaction channel, such as the conversion from QOOH to INT3 via H-atom shift (C to N) through a three-membered transition state, TS6 (41 kcal mol⁻¹), is expected to have a negligible impact on the total rate coefficient due to its high energy barriers.

The enthalpies of reaction (ΔH_rxn (0 K) for ^•CH₂NH₂ + O₂ → CH₂NH + ^•HO₂ (−10.9 kcal mol^-1) are in very good agreement with those found in the most accurate active thermochemical database (ATcT) (−11.42 kcal mol^-1) (Ruscic et al., 2004; Ruscic and Bross, 2020) and in good agreement with the theoretically calculated value in Rissanen et al. (2014) (−12.2 kcal mol^-1). The computed PES for the ^•CH₂NH₂ + O₂ reaction is also consistent with its isoelectronic analogous reaction system, i.e., ^•CH₂OH + O₂ reported by Dash and Ali (2022) using CCSD(T)//ωB97XD/6-311++G (3df, 3pd) level of theory. The enthalpy values obtained in their calculations are also given in Table 1. The reaction energies for CH₃NH^• + O₂ → CH₂NH + HO₂ (−17.3 kcal/mol) are also calculated and found to be in very good agreement with that in the ATcT (−17.9 kcal/mol) (Ruscic et al., 2004; Ruscic and Bross, 2020). The energies obtained for most of the structures of the ^•CH₂OH + O₂ system (Ali, 2020) along the reaction paths are very close to those of the current system, indicating the reliability of the data presented here. However, the barrier heights for the isomerization pathways Int-1→TS-1 (−7.4 kcal/mol)→QOOH and Int-1 →TS-5 (−18.5 kcal/mol) Int-2 in the CH₂OH + O₂ system (Dash and Ali, 2022) are quite low and more stable compared to those in the same pathways in the current system with respect to the reactant’s energy (Table 1). Moreover, the barrier height for the reaction proceeding from HO···OCH₂O to OCH₂O + OH (Dash and Ali, 2022) is 24 kcal/mol (TS-4), and the stabilization energy for the products is −15 kcal/mol, whereas in the current system, the corresponding energies are 8.3 and 12.3 kcal/mol less stable than the former ones, respectively. These differences in barrier energies can affect the overall rate coefficients between the two systems.

When a single ammonia molecule is introduced in ^•CH₂NH₂ + O₂, the simultaneous collision between ^•CH₂NH₂, O₂, and NH₃ is very unlikely to occur; therefore, the probability of a trimolecular reaction is very small under real conditions. Hence, the first step is the formation of a CH₂NH₂···NH₃ complex, followed by collision with O₂. The CH₂NH₂···NH₃ complex (−2.2 kcal/mol) is assumed to be more important than CH₂NH₂···O₂ and NH₃···O₂ due to its lower binding energy (<1 kcal/mol). As discussed in our previous work, we have also used a similar approach for ammonia-assisted reactions (Ali et al., 2019; Ali, 2020; Ali and Balaganesh, 2022; Dash and Ali, 2022). The geometrical changes in ammonia-assisted intermediates and transition states are shown in Figure 4, and the zero-point corrected PES for the ammonia-assisted ^•CH₂NH₂ + O₂ reaction is given in Figure 5. The energy of all the stationary points, i.e., reactants, intermediates (INTs), and transition states, is tabulated in Table 2.

Figure 5 shows that the effect of the ammonia reaction proceeds via similar reaction pathways as a free reaction. For simplicity, only the most stable structures are shown in the PES. As shown in Figure 5, O₂ attacks the bimolecular complex ^• CH₂NH₂···NH₃ to form a trimolecular hydrogen-bonded complex (INT1n) (Figure 4), whose stabilization energy is −36.1 kcal mol^-1. The resulting ammonia-assisted intermediate (INT1n) is 4.4 kcal mol^-1 more stable than the corresponding ammonia-free intermediate, i.e., INT1. This is due to the formation of strong hydrogen bonds between the terminal O-atom of H₂NCH₂OO and one of the H-atoms of NH₃ (2.20 Å) and the H-atom of H₂NCH₂OO (2.08 Å) with the N-atom of NH₃, whereas no such effect is observed in INT1. On the other hand, Table 2 shows that INT1 is entropically more favorable than INT1n with respect to reactants. This is due to the fact that the hydrogen-bonded complex decreases the entropy of the system. Similar to uncatalyzed reaction pathways, the terminal O-atom intra-molecularly attacks the H-atoms in the NH₂ group in the presence of NH₃, leading to the formation of INT2n and QOOH-n via five-membered cyclic transition states (TS5n/TS7n and TS1n, respectively). The difference in the barrier heights between two isomeric transition states, TS5n and TS7n, is 2.1 kcal mol^-1. TS5n seems to be more stable than TS7n because, in the case of TS5n, all three hydrogen atoms of ammonia face toward the molecular center, leading to the formation of a six-membered ring hydrogen-bonded cyclic structure, whereas in the case of TS7n, hydrogen atoms of ammonia are away from the molecular center, leading to the formation of a similar six-membered ring hydrogen-bonded cyclic structure. Entropy data also support that TS5n is more disordered than TS7n.

Supplementary Figure S2 provides an IRC scan at the same level that confirms that TS5n bridges the OCH₂C(O)OOH radical (INT2n) and CH₂NH + HO₂ +NH₃. The IRC scan confirms that the only stationary point between INT2h and the trimolecular products is that associated. The barrier height of TS1n is 2.5 kcal mol^-1 lower than that of the corresponding ammonia-free transition state TS1. This is due to the formation of two strong hydrogen bonds (H-atoms of ammonia with N and O atoms of the cyclic ring) in TS1n (1.93 and 2.14 Å). On the other hand, the barrier height of TS5n is almost similar to that of TS5, although hydrogen bonds are present in TS5n. The differences in the barrier height can be explained by the formation of two adjacent cyclic ring structures (five- and six-membered), as aforementioned in TS5n, making it sterically hindered compared to only one ring structure in TS5. The stabilization energies of ammonia-assisted QOOH and INT2 are calculated to be −23.3 and −25.5 kcal mol^-1, respectively, which are 8.4 and 4.1 kcal mol^-1 lower than those of the corresponding free-ammonia structures. This can be explained similarly by comparing the presence of hydrogen bonds in the respective structures. In QOOH-n, the formation of two strong hydrogen bonds between the H-atom of ammonia with N atom QOOH (2.26 Å) and N atom NH₃ and terminal H atom of QOOH (1.76 Å) leads to a seven-membered ring-like structure rather than only one hydrogen bond in the case of free QOOH. Similarly, the stability of INT2n can be explained by the formation of an eight-membered ring with three hydrogen bonds, H of HO₂ and N of CH₂NH (1.64 Å), O of HO₂ and N of NH₃ (2.1 Å), and H of CH₂NH and N of NH₃ (2.24 Å) compared to a six-membered ring with two hydrogen bonds in the case of INT2. On the other hand, the structures of QOOH-n and INT2n are entropically less favorable compared to those of uncatalyzed QOOH and INT2n.

QOOH-n further dissociates to INT2n (via TS2n) and INT3n (via TS6n) and then subsequently forms HO₂+CH₂NH + NH₃ and OH + OCH₂NH + NH₃ via TS4n and OH + NH₂CHO + NH₃. INT2n and INT3n are eight-membered ring hydrogen-bonded structures, and their stabilization energies are 4.1 kcal mol^-1 and 7.4 of kcal mol^-1 lower than those of the corresponding uncatalyzed intermediates. Ammonia-assisted intermediates are more stable than ammonia-free ones because of the formation of an eight-membered ring structure with three strong hydrogen bonds between HO₂···NH₃ (2.1 Å), NH₃···CH₂NH (2.24 Å), and HO₂···CH₂NH (1.64 Å) in INT2n and OH···NH₃ (2.04 Å), NH₃···NH₂CHO (1.87 Å), and HO···NH₂CHO (1.72 Å) in INT3n. The barrier heights of TS2n (−5.6 kcal mol^-1), TS3n (2.5 kcal mol^-1), TS4n (9.7 kcal mol^-1), and TS6n (24.2 kcal mol^-1) were also consistently lower than those of the corresponding ammonia-free transition states due to similar hydrogen bonding interactions. Overall, the reaction in the presence of an ammonia-assisted intermediate is thermodynamically more favorable than the free reaction, and vice versa entropically.

As previously discussed in the case of ammonia reactions, we have also employed a similar approach for water reactions. When a single H₂O molecule is added to ^•CH₂NH₂ + O₂, the first step is the formation of a CH₂NH₂···H₂O complex, followed by collision with O₂. The CH₂NH₂···H₂O (-2.7 kcal/mol) is assumed to be more important than CH₂NH₂···O₂ and H₂O···O₂ (<1 kcal/mol) due to lower binding energy. The geometrical changes in water-assisted intermediates and transition states are shown in Figure 6 and the Cartesian coordinates of all the optimized geometries are given in Supplementary Table S1.

The zero-point-corrected PES for the water-assisted ^•CH₂NH₂ + O₂ reaction is given in Figure 7, and the energy of all the stationary points, i.e., reactants, INTs, and TSs, is tabulated in Table 3. As shown in Figure 6 and Figure 7, the O₂ molecule attacks the bimolecular complex ^•CH₂NH₂···H₂O to form a trimolecular hydrogen-bonded complex (INT1h) with a stabilization energy of −36.1 kcal mol^-1. The INT1h is 4.4 kcal mol^-1 lower than that of the water-free intermediate (INT1) and identical to the energy of ammonia-assisted INT1n. The result indicates water- and ammonia-assisted reactions are energetically more favorable than free reactions.

In INT1h, two strong hydrogen bonds are observed between the terminal O-atom of H₂NCH₂OO and one of the H-atoms of H₂O (1.91 Å) and O-atom H₂O and terminal H-atom H₂NCH₂OO (2.12 Å) (see Figure 6). On the other hand, Table 3 and Table 1 show that INT1h is entropically least favored than INT1n and INT1. Similar to the free reaction, the terminal O-atom of INT1h intra-molecularly attacks the H-atoms in the NH₂ group, leading to the formation of cyclic structures, i.e., INT2h and QOOH-h, via five-membered cyclic transition states, i.e., TS5h and TS1h, respectively. The barrier heights of TS5h and TS1h are 2 kcal mol^-1 and 3 kcal mol^-1 lower than those of water-free transition states TS5 and TS1, respectively. In a similar manner, TS5h and TS1h are 1.8 kcal mol^-1 and 0.4 kcal mol^-1 lower than TS5n and TS1n, respectively. These differences in barrier heights indicate that water-assisted transition states are energetically more stable than ammonia-assisted and water-free species. The stabilization energies of water-assisted QOOH-h and INT2h are calculated to be −22.3 kcal mol^-1 and −27.1 kcal mol^-1, which are 7.8 kcal mol^-1 and 5.7 kcal mol^-1 lower than those of the corresponding free structures. This can be understood by comparing the presence of hydrogen bonds in the respective structures. Supplementary Figure S3 provides an IRC scan at the same level that confirms TS5h bridges the OCH₂C(O)OOH radical (INT2h) and CH₂NH + H₂O + HO₂ products. The IRC scan confirms that the only stationary point between INT2h and the trimolecular products is that associated. In QOOH-h, the formation of two strong hydrogen bonds between the H-atom of water with the N atom of QOOH (1.95 Å) and the O atom of H₂O and the terminal H atom of QOOH (1.82 Å) led to the formation of a seven-membered ring-like structure. Similarly, the stability of INT2h can be found by the formation of an eight-membered cyclic ring with three hydrogen bonds, H of HO₂ and N of CH₂NH (1.65 Å), O of HO₂ and H of H₂O (1.87 Å), and H of CH₂NH and O of H₂O (2.13 Å), which are compared to that of a six-membered ring with two hydrogen bonds, as indicated in the case of INT2. Between water- and ammonia-catalyzed QOOH and INT2, QOOH-n is more stable than QOOH-h and INT2h is more stable than INT2n.

In general, the water-free pathways are entropically more favorable than water-free ones. The QOOH further dissociates to INT2h (via TS2h) and INT3h (via TS6h) and then subsequently forms HO₂+ CH₂NH + H₂O, OH + OCH₂NH + H₂O via TS4h, and OH + NH₂CHO + H₂O. In general, the water-assisted reaction channels are thermodynamically more favorable and entropically less favorable than the free reaction. It is also clear from Table 3 that all other pathways are thermodynamically less important compared to R + O₂→INT1h→TS5h→INT2h→ CH₂NH + HO₂+H₂O, whose barrier height is the lowest with respect to reactants. Therefore, the other reaction channels may have less contribution under tropospheric conditions.

To obtain the rate coefficients for ^•CH₂NH₂ + O₂ in temperatures between 200 K and 400 K and pressures from 0.000001 bar to 1000 bar, the RRKM/ME simulation has been used. The rate coefficients as a function of the temperature at 1 bar pressure are shown in Figure 8. The rate coefficients for the formation of O₂-CH₂NH and CH₂NH are observed to be pressure-dependent and negative temperature-dependent. This result is consistent with that of the previous reports by Rissanen et al. (2014) and a similar reaction system, i.e., ^•CH₂OH + O₂ (Dash and Ali, 2022). The calculated rate coefficient at 300 K (1.6 × 10⁻¹¹ cm³ molecule⁻¹ s⁻¹) is a factor of ∼2 lower than the experimentally measured ones (3.2 × 10⁻¹¹ cm³ molecule⁻¹ s⁻¹) (Rissanen et al., 2014). The calculated value is in good agreement with the lower temperature range <250 K when the positive error is considered. In light of the expected errors in computed thermochemistry, which may be up to 1 kcal/mol, we believe that this level of accuracy is sufficient for these purposes. A recent study by Ali et al. (2023) indicates that the barrier heights of similar reactions are very sensitive to quantum chemical calculations. The computed rate coefficients are compared with previously reported values and also with its isoelectronic analogous reaction system, i.e., ^•CH₂OH + O₂ (see Figure 8) (Dash and Ali, 2022). The calculated values at 300 K are a factor of 2 lower than the theoretically calculated values (Dash and Ali, 2022). The presence of the N atom in CH₂NH and the O atom ^•CH₂OH can explain this, leading to the development of various chemical kinetics conclusions.

Figure 9 reports the rate coefficients in the fall-off regions for the ^•CH₂NH₂ + O₂ reaction at different temperatures. The rate coefficients increase as pressure increases, and the HPL condition is observed at ∼100 bar. As shown in Figure 9, the largest difference between the two limits occurs at about 250 K, which reaches a factor of 2. However, at 200 K, the difference between the two regimes is about a factor of 2. To provide more detailed insights, the relative branching fractions of these channels were determined at 200 K, 300 K, and 400 K and shown in Supplementary Figure S4. For simplicity, the branching fraction at different pressures and at 300 K is shown in Figure 10. At all temperatures and pressures, the branching fractions of QOOH are almost negligible; therefore, they are not shown in Figure 10. The branching fraction for the formation CH₂NH/HO₂ contributes 40% at 300 K and increases as temperature increases to 400 K. The formation of CH₂NH decreases to 0% as pressure increases to 100 bar. At the same time, the formation of CH₂NH₂OO increases as pressure increases to 100 bar (100%). The plot shows that backward reaction to regenerate reactants is prominent at <0.01 bar. These results are also consistent with the previously reported branching ratio by Rissanen et al. (2014). At all the pressure rates, it is more advantageous to lose HO₂ via the formation of CH₂NH rather than through OH loss via formamide formation due to an energetically favorable pathway (see Figure 10). As temperature increases from 300 K to 400 K, the product branching ratio increases (see Supplementary Figure S4). The result is due to the fact that the relevant stationary points on the respective potential energy surfaces have a dominant entropy factor over the enthalpy factor.

As discussed in the previous section, only the entry channel ^•CH₂NH ···NH₃ + O₂ is considered for the rate coefficient calculations. CH 2 NH 2 + O 2 + NH 3 ⇌ k − 1 n k 1 n CH 2 NH 2 --- NH 3 + O 2 ⇌ k − 2 n k 2 n O 2 − CH 2 NH 2 --- NH 3 → k uni − n CH 2 NH + HO 2 + NH 3

The equation to calculate the effective pressure-dependent rate coefficients k e f f b i m o l T , M is as follows: k e f f b i m o l T , M = K e q − 1 n × N H 3 × K e q − 2 n × k ∞ u n i 1 − f • C H 2 N H 2 … N H 3 + O 2 , (4)

where K e q − 1 h = k 1 n k − 1 n   a n d and K e q − 2 n = k 2 n k − 2 n are equilibrium constants of each pathway involved in a reaction, [NH₃] is the concentration, and f is the branching fraction for the reaction proceeding to the reactant. The ammonia concentration used at 10 ppbv is based on the observations from previous studies (Ali et al., 2021). The rate coefficients for the ammonia-assisted reaction are almost similar to those of ^•CH₂NH₂ + O₂ in the temperature range of 200 K–400 K (see Table 4). In fact, some lower values were obtained at higher temperatures. The catalytic behavior does not take place if step 0 is not included in the reaction mechanism. The result could be due to the lower entropy change in the reaction. The total effective rate coefficient for ^•CH₂NH₂ + O₂ (2.7 × 10⁻²¹ cm³ molecule⁻¹ s^-1 at 300 K) is ∼12 orders of magnitude lower than that for the ^•CH₂NH + O₂ reaction (8.8 × 10⁻¹³ cm³ molecule⁻¹ s^-1). This result is due to the fact that the ammonia-assisted pathway depends on the ammonia concentration (see Table 4).

The relative branching fractions of these channels determined at 200 K, 300 K, and 400 K are shown in Supplementary Figure S5. For simplicity, the branching fraction at 300 K is shown in Figure 11. The branching fraction for the formation of CH₂NH/HO₂ contributes 50% at 300 K, and almost the same temperature increases to 400 K (see Supplementary Figure S5). At the same time, the formation of CH₂NH₂OO … NH₃ increases as the pressure increases to 100 bar (100%). The plot shows the back-reaction that regenerates ^•CH₂NH₂ … NH₃ + O₂. When the results are compared with those of the free reaction, it is easier to lose HO₂ via the formation of CH₂NH than through OH loss via formamide formation. We can say that the effect of ammonia has a negligible impact on the product branching ratios, and the results are almost similar to those of a free reaction, except at very low pressure (Supplementary Figure S5).

The scheme for the formation of INT1h, CH₂NH, and HO₂ from ^•CH₂NH₂ + O₂ reactions with the effect of a water can be written as follows. CH 2 NH 2 + O 2 + H 2 O ⇌ k − 1 h k 1 h CH 2 NH 2 --- H 2 O + O 2 ⇌ k − 2 h k 2 h O 2 − CH 2 NH 2 ––– H 2 O → k uni − h CH 2 NH + HO 2 + H 2 O

The equation to calculate the effective pressure-dependent rate coefficients k e f f b i m o l exp − B / T is as follows: k e f f b i m o l T , M = K e q − 1 h × H 2 O × K e q − 2 h × k ∞ u n i 1 − f • C H 2 N H 2 … H 2 O + O 2 , (5)

where K e q 1 h = k 1 h k − 1 h and K e q − 2 h = k 2 h k − 2 h are the equilibrium constants of each reaction pathway involved in equation (iii), [H₂O] is the concentration, and f is the branching fraction for the reaction proceeding to the reactants. The [H₂O] is calculated using a typical humidity concentration, as discussed in the previous paper (Dash and Ali, 2022). The rate coefficients were also calculated using different water concentrations, as shown in Supplementary Figure S6. The effect of relative humidity from 20% to 100% on calculated rate coefficients is a factor of 10 difference. The effective rate coefficient calculated based on Eq. 4 (2.04 × 10⁻¹⁷ cm³ molecule⁻¹ s^-1 at 298 K) is ∼six to seven orders of magnitude lower than that of the water-free ^•CH₂NH₂ + O₂ reaction (∼2.2 × 10⁻¹¹ cm³ molecule⁻¹ s^-1 at 298 K). This is due to the fact that the water-assisted pathway depends parametrically on water concentration and entropy reduces the rate coefficients. Our results are also consistent with the previously reported values for similar reaction systems (Ali et al., 2022; Dash and Ali, 2022).

Our calculation shows that the total effective rate coefficients for systems ^•CH₂NH₂ + O₂ (+NH₃) (∼10^–11 order) and ^•CH₂NH₂ + O₂ (+H₂O) (6 order) are smaller than that of the free reaction (see Figure 12).

It is clear that the geometries of INT and TSs are different in ^•CH₂NH₂ + O₂ (+NH₃) reaction systems compared to their isoelectronic analogous ^•CH₂NH₂ + O₂ (+H₂O) reactions, resulting in different computed enthalpies and rate coefficients. Because of this, the kinetics of ^•CH₂NH₂ + O₂ (+NH₃) is quite different from those of ^•CH₂NH₂ + O₂ (+H₂O) reaction systems. In the case of free reactions and ammonia, the rate coefficients exhibit negative temperature dependence, whereas in the case of water, positive temperature dependence was observed. This may be due to the fact that water concentration is highly dependent on temperature, and ammonia concentration is nearly independent of temperature. The branching fractions for the formation of ^•OOCH₂NH₂ and ^•CH₂NH and the reaction going back to ^•CH₂NH₂ + O₂ with the effect of a single water molecule at 300 K and pressure range 0.001 bar–1000 bar are shown in Figure 13, and a comparison of branching fractions for ^•CH₂NH₂ + O₂, ^•CH₂NH₂ + O₂ (+NH₃), and ^•CH₂NH₂ + O₂ (+H2O) at 200 K, 300 K, and 400 K is shown in Supplementary Figure S7.

As previously discussed in water- and ammonia-free reactions, the product branching ratios for the formation of CH₂NH + HO₂ decrease, and the reaction goes back to the reactants, i.e., ^•CH₂NH₂ + O₂, when the pressure increases from 0.1 bar. When a water molecule is added to the reaction, the product branching ratio changes significantly (∼80%), and a single water reaction favors the formation of CH₂NH + HO₂ at a temperature of <300 K; however, the effect of ammonia favors only ∼10%. Despite the slower water reaction, our ME calculations indicate that a favorable CH₂NH + HO₂ formation is observed under tropospheric conditions.