Decomposition of HCN during Experimental Impacts in Dry and Wet Planetary Atmospheres

Hydrogen cyanide (HCN), a key molecule of significant importance in contemporary perspectives on prebiotic chemistry, originates in planetary atmospheres from various processes, such as photochemistry, thermochemistry, and impact chemistry, as well as from delivery by impacts. The resilience of HCN during periods of heavy bombardment, a phenomenon caused by an influx of material on unstable trajectories after accretion, remains relatively understudied. This study extensively investigates the stability of HCN under impact conditions simulated using a laboratory Nd:YAG laser in the ELISE experimental setup. High-resolution infrared spectroscopy was employed to monitor the gas phase composition during these simulations. Impact chemistry was simulated in bulk nitrogen atmospheres with varying mixing ratios of HCN and water vapor. The probed range of compositions spans from ∼0 to 1.8% of HCN and 0 to 2.7% of H2O in a ∼1 bar nitrogen atmosphere. The primary decomposition products of HCN are CO and CO2 in the presence of water and unidentified solid phase products in dry conditions. Our experiments revealed a range of initial HCN decomposition rates between 2.43 × 1015 and 5.17 × 1017 molec J–1 of input energy depending on the initial composition. Notably, it is shown that the decomposition process induced by the laser spark simulating the impact plasma is nonlinear, with the duration of the irradiation markedly affecting the decomposition rate. These findings underscore the necessity for careful consideration and allowance for margins when applying these rates to chemical models of molecular synthesis and decomposition in planetary atmospheres.


Mechanism and rates
This section is a broadened description of the section Mechanism and rates in the main paper.
The main work of this paper was in total nine experiments, each with a different starting composition, but analyzed in the exact same manner.The data obtained from the experiments constitute FTIR spectra of the gas phase products measured at set intervals during the laser irradiation.The spectra were analyzed with the spectr library (as described in the main paper) and partial pressures were obtained as the output from the analysis.The time-dependent partial pressures of HCN, H2O, CO and CO2 were used to measure the rate of HCN decomposition.
The system of the main reactions is reasonably well described by the following equations: In this set of equations, XX and YY represent unobserved products.Those may be solid phase products as well as gas phase products without dipole moments (impossible to observe with FTIR, such as H2 or O2).
Eq. (1) describes the formation of CO.Then, as shown in Eq. (2), CO is further oxidized to CO2.Aside from these reactions, both HCN and H2O are themselves decomposed, as observed from the experimental data.We did not identify any products of this process, but the most likely products are H2 and O2 from the decomposition of water and solid phase material (tholins, carbon, etc.) from the decomposition of HCN.The formation of solid products was indeed observed during the experiment.
In order to describe the kinetics of the process, the following rate equations were fit to the data: For the purposes of the fit, we take: 2  =   2    2  (10) where Ax is a factor which accounts for adsorption.
Adsorption plays a very important role in the experiment.The experiment was constructed to minimize the amount of material that the gases come into contact with and currently, the only surfaces available to the gas are the glass walls of the apparatus, the Teflon fan and rubber o-rings, which are part of Ultra-Torr vacuum fittings (Swagelok®, USA).All present gases adsorb on the surface, which ideally could be experimentally estimated by injecting a known molar amount of gas and comparing the expected partial pressure (from the ideal gas equation) and the observed partial pressure.The issue with this approach is that the adsorption depends on the nature of the specific gas, the available surface, but also temperature, the composition of the rest of the mixture and the formation of solid phase products, which may also selectively enhance adsorption.
During the experiment, when HCN and H2O decompose in the gas phase, they are supplemented from the adsorbed layers, so when fitting the rate equations, one must consider the total amount of available molecules and not just the amount available in the gas phase.For this reason, two factors, Ax, were employed as multipliers to the observed partial pressures of HCN and H2O gas (Eqs.( 9) and ( 10)).The value of these multipliers is unknown with the lower bound of 1 (no adsorption) and experimental estimate around 2-3.For the cases of CO and CO2, the adsorption is in this case negligible, because both HCN and H2O adsorb much more efficiently.The value of these multipliers is unknown with the lower bound of 1 (no adsorption) and experimental estimate around 2-3.
To get the rate coefficients of the reactions (5), ( 6), ( 7) and (8), we fit the constants of the above model (as a set of ordinary differential equations) using the python symfit package with set values of the multipliers AHCN and AH2O.Eight fixed values of each multiplier were chosen (1, 2, 3, 4, 5, 6, 7, 8) for each compound.This means, that each of the nine experiments was fit 64×, with a different pair of multipliers each time.Each of these fits returned the values for six rate coefficients (k1,f, k1,r, k2,f, k2,r, k3 and k4) along with standard deviations for each value.As a goodness of fit qualifier, we used the R 2 value, which was stored as well.All this data are shown in files: However, comparison of the R 2 of each fit did not show any optimal values of AHCN and AH2O.Rather, the fit behaves so that when the multipliers increase from 1, the R 2 increases, and the fit improves up to a certain R 2 value.This is because at low (close to 1) values of the multipliers, there is not enough carbon containing compounds in the data to even explain the formation of CO and CO2.In fact, due to adsorption, the sum of the partial pressures of C-containing compounds is for most experiments less than the sum of partial pressures of products formed in the experiment.After escaping this problem by increasing the multipliers, any further increase is compensated by the increase in the rate coefficients (often k1,f and k1,r for experiments with lower initial amounts of water, and k3 and k4 for experiments with higher amounts of water).Therefore, the R 2 is a good indicator for incorrectly low multipliers but becomes ambiguous for their higher values.
Then, upon assumption that the values of rate coefficients k1,f, k1,r, k2,f and k2,r converge to the correct value, the mean value of R 2 and its standard deviation were thus calculated for each experiment.Combinations of multiplier pairs whose R 2 value differed from the mean by more than double the standard deviation of the mean of R 2 were discarded.This effectively separated out the incorrectly low values of multipliers.After calculating the covariances and verifying the calculation in this way, we calculated the mean of each coefficient across the nine experiments along with the standard deviations of the mean.
The final rate coefficients are shown in Table S1 below (identical to Table 3 in the main paper).
Table S1: Rate coefficients obtained from the fit of all our data.This is a copy of Table 3 in the main paper.

Rate constant Value k1,f
2.6×10 -4 ± 2.4×10 -4 Torr s -1 k1,r 6.4×10 -4 ± 3.6×10 -4 Torr s -1 k2,f 1.2×10 -3 ± 3.8×10 -4 Torr s -1 k2,r 4.9×10 -3 ± 9.9×10 -4 Torr s -1 k3 8.1×10 -4 ± 6.0×10 -4 Torr s -1 k4 2.4×10 -4 ± 5.5×10 -5 Torr s -1 As discussed above, it is possible that a weak dependence on the input parameters exists, but since we used a wide range of parameters (the minimum value of the multipliers is given, and we set the maximum as ~3× the value estimated from the experiment) and obtained reasonably small standard deviations and covariances, we do not expect such would-be dependencies to alter our final results.One possible source of error is that the exact radiative transfer model and mechanism in the plasma are not known.It is possible that other minor reaction channels or bottleneck reactions could be present, whose incorporating would help lower the standard deviation.Another possible source of error could be the fact that the multipliers are constant values.Both HCN and H2O adsorb on the walls of the apparatus and likely compete for the adsorption sites.Change of the amount of reactants in the apparatus during the experiment likely causes a change in the adsorption coefficient/multiplier.Better description of the adsorption mechanism and accounting for the adsorption in the process could also help with reducing the standard deviation values.
The rate coefficients obtained from this fit can be directly implemented into planetary atmospheric models.A possible way to treat the standard deviations is to include them in the planetary atmospheric model or to perform sensitivity analysis within the model.

Applicability range
This section contains additional information and lists of files for the Applicability range section of the main paper.
The section in the main paper discusses that a range of experiments with additional initial partial pressures of HCN and H2O was performed.For each of these experiments, we measured the composition of the gas phase by FTIR (in the same way as with the other experiments) and fit the spectra to obtain partial pressures of all the four principal components of the experiment: HCN, H2O, CO and CO2.We then applied the final rate coefficients (Table 3 in the main paper) to the data and using the conditions at 0 s (prior to laser irradiation), attempted to predict the result at 420 s.This was compared to the actual measured result after 420 s of irradiation.The plots which show this comparison for each experiment are contained in files: 1. HCN(0.5Torr)+N2+H2O(0.5Torr).pdf 2. HCN(0.5Torr)+N2+H2O(1.4Torr).pdf 3. HCN(0.5Torr)+N2+H2O(6.2Torr).pdf 4. HCN(0.5Torr)+N2+H2O(21.3Torr).pdf 5. HCN(0.5Torr)+N2+H2O(nika).pdf The irradiation cell for gas phase experiments is roughly spherical in shape with four 1-inch-in-diameter cylindrical tubes welded to it.The cylinders are used for inlet and outlet of gas, entrance of the laser radiation and UV-Vis inspection window.Glass beads were placed on the bottom of the cell to scatter the passing radiation (otherwise, the bottom of the cell would be damaged by the laser beam).The irradiation cell for solid phase experiments is cylindrical.The upper part is similar in design to the cell for the gas phase irradiation.The lower part contains a sample holder on which the solid sample can be placed.Underneath is a rotary shaft with a stepper motor nested in a glass tube.The glass tube is connected to the cell with a vacuum-tight joint.Both parts (the shaft and the sample holder) contain a magnet.The rotation of the stepper motor therefore rotates the sample and ensures that the laser does not ablate a single point in the sample.This is important a) because if the sample is inhomogeneous, ablation of more spots provides a more average result, and b) if the laser fired in one spot only, the sample would quickly be ablated away, and a hole would be created.

DCN and the reverse reaction
To prove the fact that the reaction of HCN decomposition is reversible, we performed experiments with D2 as the source of deuterium and observed the formation of DCN in our experiment.A FTIR spectrum showing the ν1 (2360 cm -1 ) band of DCN is shown in the main paper.Here in the Supplementary Information, we show the relative abundances of HCN and DCN in the experiment.Figure S3 shows the relative abundance of HCN and Figure S4 shows the relative abundance of DCN.

Composition of the gas phase
In this section, we show some visualizations of the gas phase composition after 420 s of irradiation.The following plots were created from the experimental data and not the model.For each of the plots, we define Yi, which is the relative amount of carbon locked in a species i in the gas phase relative to other species in the gas phase.Yi is defined as: where pi is the partial pressure of specie i and ni is the number of carbon atoms in the species.
Figure S5 shows YCO with respect to the initial partial pressures of HCN and H2O.
Figure S5: YCO with respect to the initial partial pressures of HCN and H2O.
This figure shows that the gas phase in experiments with roughly equal amounts of HCN and H2O is the highest.Higher initial amounts of H2O relative to HCN lead to faster consumption of HCN and formation of CO and subsequently CO2.After 420 s of irradiation, then, the relative amount of CO in the gas phase decreases for experiments with high pH2O, ini/ pHCN, ini.At the same time, experiments with high initial content of HCN and little to no H2O do not contain much CO either, mostly because there is little water for its production in the first place.
Figure S6 shows the same property, the YCO2 for CO2.

S15
Figure S6: YCO2 with respect to the initial partial pressures of HCN and H2O.
This figure shows the relative amount of CO2 in the gas phase after 420 s of irradiation and clearly, the amounts are much lower than in the case of CO, with the exception of experiments with very little HCN at the beginning and relatively high amounts of H2O.Notably, in experiments with little to no H2O, little to no CO2 is formed.Even though CO is present in these experiments (as some residual water is always present in the experiment), it seems to be so little that all water is consumed for the formation of CO and none is left for the formation of CO2.
Figure S7 shows the same property, the YC2H2 for C2H2.This figure shows the relative amount of carbon in the gas phase locked in acetylene.The formation of acetylene was observed only in three experiments.In these experiments, almost no water was initially present and the amounts of HCN were the highest of all the experiments.It can be expected the if even higher partial pressures of HCN were used, more acetylene could be formed.The actual observed amounts of acetylene were very small, but in the experiments where it was detected, little CO and CO2 was formed as well.For this reason, in the one experiment with least water and most HCN at the beginning shows that up to 75% of the gas phase is acetylene.Most of the decomposition products of HCN in this experiment are solid products, which are not shown here but are shown in the main paper.

UV-Vis spectra
We observed the laser-induced dielectric breakdown in our experiments in hopes to reveal hints about the reaction mechanism.The spectra of the laser induced dielectric breakdown plasma in the main experiments were observed with a UV-ViS spectrometer (Aryelle Butterfly, LTB Lasertechnik Berlin, Germany) 10 ns after the pulse with a gate width of 1000 ns.The spectra were measured in two ranges (187--425 nm and 409--763 nm), as shown in panels A and B of Figure S8, respectively.Twenty measurements were averaged for each spectrum.The spectra revealed signals of the .CN B 2Σ+ u -X 2Σ+ g Δυ = 0-0 violet 1 st , :C2 Swan D 3Π g -A 2Π u Δυ = 1-0 and Δυ = 0-0 transitions and the Hα line of atomic hydrogen.We also observed a signal at ~745 nm, which was not assigned.This measurement did not provide sufficient data to elucidate the reaction mechanism, but showed that the .CN radical, along with :C2 and Hα are all present in significant amounts in the discharge.Interestingly, we did not observe any O:, .OH, N:. or N2 emission bands at this or other gate delays.

Solid phase
The main focus of this study is the gas phase chemistry.Nevertheless, solid phase products form during the experiment as well.This manifests by forming a brownish layer on the walls of the irradiation cell.
Figure S9: The relative proportion of gas phase products in the product mixture with respect to the initial amounts of H2O and HCN.
Figure S9 visualizes the carbon distribution in the mixture in the form of its ratio in the gas phase vs. the solid phase.
This figure shows the relative proportion of carbon in the gas phase products in the product mixture, defined as: where pi are the partial pressures of gas phase products CO, CO2 and C2H2 and ni is the amount of carbon atoms in the molecule.
The contribution of C2H2 is very small, but was included here nonetheless.This figure was created using the experimental data rather than the model output.It again shows that the contribution of direct HCN and H2O decomposition is greater with the increasing amount of water at the start of the experiment.
It also shows that higher initial amounts of HCN decrease the relative amounts of gas phase products.The reason for this in terms of mechanism can be the higher ratio of the densities of .CN and .OH in the plasma.This higher relative density increases the likelihood of collisions between two .CN radicals (likely producing solid polycyanes), while decreasing the probability of collisions between .CN and .OH (likely producing gas-phase radicals).Such behavior is expected from the model and confirms that the model and the experiment behave in the same manner.
The solid phase products were collected by washing the cell with methanol and then evaporating the solution.
The solid phase was analyzed by scanning electron microscopy.However, the majority of the evaporated sample constituted of Teflon TM bits scraped off the fan during use and of NaCl probably resulting from contamination of the sample during manipulation.Carbon signals expected from the experiment were also observed, but only in small amounts, which could be attributed both to the reaction and contamination.Any results from this solid phase analysis are therefore ambiguous.Based on the color and on existing literature, the main components of the solid phase should be tholins, soot and refractive carbon.We believe, however, that the solid phase would merit a thorough investigation.For this, the experiments would have to be optimized, such as carried out for longer times, with higher initial pressures.

NO x es
The mechanism proposed in this study is the dominant pathway that governs the formation of the CO-CO2 product mixture, which is supported by the kinetics of the process described above.However, there exist side processes which lead to different products.The most efficient of those side processes are the formation of NO and N2O.Both NO and N2O were observed in two experiments (experiments 8 and 9 as described in the main text).Figure S11 shows FTIR spectrum of the gas phase of experiment 8 after 2520 s of irradiation.This experiment initially contained ~1.25 Torr of HCN, ~13.25 Torr of H2O and ~706 Torr of N2.These two experiments, 8 and 9, are the experiments performed with the highest initial amount of water.It is likely, therefore, that the formation of the NOxes includes H2O in this mechanism.These molecules were, however, observed only in these two experiments and their partial pressure did not exceed 0.1 Torr.For this reason, we did not consider those products in the main mechanism.
Figure S11: A FTIR spectrum of experiment 9 in the main table after 2520 s of laser irradiation.Shown in red are the observed bands of NO and the ν3 band of N2O.

S21
Figure S12: The apparatus and synthetic process of HCN used for this paper.
HCN was necessary as a reactant for the experiments shown in this paper.We developed a new protocol for its synthesis (shown in Figure S12): • Preheat CaCl2 traps to avoid condensation of HCN.
• Purge the apparatus with N2 to remove air.
• Keep a small flow of N2.
• Add 35 wt.% KCN dropwise to a constantly stirred 50 wt.%H2SO4.Add ~1 drop per second -the goal is to add KCN slow enough to keep the solution cool enough so that the HCN does not polymerize (indicated by a yellow brown color).Carefully collect all HCN vapors produced.• After all KCN is added, heat the solution to 70-80 C to evaporate the leftover HCN from the solution.
• Let cool to room temperature.
• Stop the flow of N2 and evacuate the central part of the apparatus.
• Close off the central part of the apparatus, let the HCN evaporate and then freeze out in liquid nitrogen in the storage flask.• Flush the whole apparatus thoroughly with N2.
• Add NaOCl to the solution to neutralize any remaining NaCN.

Decomposition fluxes for planets
Table S2 below shows decomposition fluxes calculated based on our model for the minimum and maximum HCN decomposition rates at given time and composition.The maximal values correspond to the initial amount of 13 Torr HCN and 20 Torr H2O, while the minimal ones correspond to 0.5 Torr HCN without any H2O.The given time intervals (number of pulses/energy delivered to the system) were chosen as 0.1 s (1 pulse), 106 s as same delivered energy as (McKay and Borucki, 1997, 10.1126/science.276.5311.390),420 s (4200 pulses) as approximate HCN half-life, 2500 s (25 000 pulses) as the time when derivations of maximal and minimal curves begin to be similar and 7000 s (70 000 pulses) s the end of experiments.This purpose of this table is to illustratively show outcomes of the model in its boundary conditions.Similar to the main text, based on the results of our experiments at the given HCN and H2O ranges, we present in Table S3 a maximum flux of HCN decomposition during LHB for Earth, Venus, and Mars with values of 1.63e+13, 4.52e+12 and 3.40e+12 cm -2 s -1 respectively.These were obtained from a model atmosphere containing 13.0 Torr HCN (~1.7 vol.%) and 20.0 Torr H2O (2.8 vol.%) and correspond to one laser pulse's energy (0.1 s, at 10 Hz laser repetition) delivered to the model gas mixture.At these conditions, the above values are the highest ones to be reached.
The minimum HCN decomposition fluxes were determined from experiment containing 0.5 Torr HCN (~0.07 vol.%) and almost no water (~0.01Torr) with values of 7.41e+10, 2.05e+10 and 1.54e+10 cm -2 s -1 for Earth, Venus and Mars respectively.This experiment, chosen as 0.5 Torr HCN and no H2O, was the one with the lowest initial amount of HCN.Even if our model can extrapolate any inlet values to 0, we cannot guarantee reaching viable results below the experimental limit of 0.5 Torr HCN.Any other values that lay in our HCN vs H2O range can be calculated using our system rate coefficients.This extended table includes calculations for points x1 -x9.Similarly, Table S4 below shows HCN decompositions rates calculated from our model for maximum and minimum partial pressures compositions (12.5 Torr HCN + 20 Torr H2O and 0.5 Torr HCN and 0 Torr H2O) at different times.

Applicability range
As outlined above and detailed in Table 1, we conducted 21 additional single-point experiments with a broader range of initial HCN and H2O concentrations.This expansion of the experimental parameter space enhances the applicability of the global rate coefficients for investigating plasma chemistry in planetary atmospheres.The initial composition at the start of each experiment was used to predict the gas phase composition after 420 seconds of irradiation.The computed predictions were subsequently compared with the corresponding experimental results.Given that each molecule's partial pressure was represented by a single data point from the 420-second irradiation interval, we introduced a modified version of the R 2 descriptor to evaluate the fit's accuracy.
We define this modified metric as R 2 -like', which is described as follows: where pi,obs (Torr) is the experimentally observed partial pressure of compound i and pi,calc (Torr) is the calculated counterpart.The R 2 -like indicator was calculated for each species (HCN, H2O, CO and CO2) and then averaged.The result is shown in Figure S13 and the individual R 2 -like value are shown in the Supplementary Information.Our model correctly predicts HCN and H2O partial pressures at 420 s in the whole range of initial conditions (i.e., HCN initial partial pressures <13 Torr and H2O initial pressures <20 Torr).The accuracy of the fit slightly, but reasonably, decreases when either of these components' initial pressure exceeds ~10 Torr.This decrease is likely attributable to adsorption or co-adsorption, as previously described.
For CO, the prediction is sufficient for all tested partial pressures with a slightly worse result for initial partial pressures of HCN and H2O <1 Torr.The partial pressure of CO2 is well predicted for initial partial pressures of HCN and H2O ~1-5 Torr.For other pressures, the R 2 -like indicator is negative, which would imply a wrong prediction.However, upon reviewing the data, it is evident that the absolute differences between the predicted and observed partial pressures of CO2 do not exceed ~0.1 Torr.Due to the nominal values of the partial pressures of CO2 being very small and the measurement error remaining relatively constant, the relative errors are high.Therefore, for CO2, this R 2 -like indicator is not a good descriptor of the fit.
In summary, our experiments covered a parameter space of initial partial pressures ranging from approximately 0 to 13 Torr (0-1.8%) for HCN and from 0 to 20 Torr (0--2.7%)for H2O.The data are therefore suitable for atmospheric plasma chemistry models with partial pressures of HCN and H2O within this range.Furthermore, since the range covers water vapor pressures up to the saturated vapor pressure and HCN content up to 1.8%, this range should be sufficient to cover the vast majority of known planetary systems.
The model enables calculation of the composition of the experimental mixture at any point in time and any initial composition within the established experimental parameter space ~0-13 Torr (0-1.8%) of HCN and ~0--20 Torr (0-2.7%) of H2O.Calculated results of the impact HCN decomposition are shown in Figure S14.There, partial pressures and the amount of carbon deposited to solid phase are depicted for all initial partial pressures at four selected times -0.1 s, 420 s, 900 s, and 2500 s.The last row (C in solid phase) illustrates that the drier the atmospheric mixture, the more carbon ends up in the solid phase.The fraction of this solid phase was calculated as follows:

Figure S1 :
Figure S1: Irradiation cell for gas phase experiments.

Figure S2 :
Figure S2: Irradiation cell for solid phase experiments.

Figure S3 :
Figure S3: The normalized relative abundance of HCN in the experiment.

Figure S4 :
Figure S4: The normalized relative abundance of DCN in the experiment.The experiment at the beginning contained HCN and D2.During the irradiation, the HCN is decomposed.In the first step, HCN presumably decomposes to H• and •CN.At the same time, D2 decomposes to two D• radicals.The D• and •CN radicals may then react during the afterglow of the laser-generated plasma to form DCN. FigureS3shows the relative decrease in the partial pressure of HCN during the experiment.FigureS4then shows the formation of DCN.Data for both figures were normalized and show relative changes in the abundances.Interestingly, there is a moment when DCN abundance reaches a maximum value and then decreases.At the start, the abundance of DCN increases as a lot of HCN is decomposed and no DCN is present.After some time, however, there is already enough DCN and small enough amount of HCN that the decomposition of DCN begins to dominate over its formation from HCN.The DCN FTIR spectrum in the main paper is the spectrum at the time of the maximum abundance of DCN in the sample.

Figure S7 :
Figure S7: YC2H2 with respect to the initial partial pressures of HCN and H2O.

Figure S8 :
Figure S8: UV-ViS spectrum of the laser induced dielectric breakdown induced in a mixture of HCN (1.3 Torr) and H2O (0.3 Torr) by the Nd:YAG laser (as described in the experimental section of the main paper body.)

Figure
Figure S10 below shows two photographs of the irradiation cell -before (A) and after irradiation (B), showing the solid phase.

Figure S10 :
Figure S10: Photograph of the measurement cell before (A) and after irradiation (B).Solid phase formed during the experiment is clearly visible.

Figure
Figure S13: R 2 -like indicator of the goodness of fit for each of the additional 420 s (1890 J) experiments plotted versus the initial contents of HCN and H2O.

Figure S14 :
Figure S14: Model results of the HCN decomposition.Each column represents a fixed irradiation time indicated at the top.Each row then shows results for a given reactant or products.Colorbars for the presented results are depicted on the right.

Figure S15 :
Figure S15: HCN decomposition rate calculated from our model.Panel A shows results for four selected times, panel B shows results for selected compositions.

Figure S16 :
Figure S16: HCN partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for selected compositions.

Figure S20 :
Figure S20: CO + CO2 partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for selected compositions.

Figure S22 :
Figure S22: Amount of solid phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Figure S23 :
Figure S23: Amount of solid phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Figure S24 :
Figure S24: Amount of solid phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Figure S25 :
Figure S25: Amount of solid phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Figure S26 :
Figure S26: Amount of gas phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for selected compositions.

Figure S27 :
Figure S27: Amount of gas phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Figure S28 :
Figure S28: Amount of gas phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Figure S29 :
Figure S29: Amount of gas phase carbon partial pressure calculated from our model.Panel A shows results for four selected times, panel B shows results for different selected compositions.

Table S2 :
Minimum and maximum HCN decomposition values based on our model calculated from our model at given times for Venus, Earth and Mars.

Table S3 :
Maximum and minimum HCN decomposition fluxes (cm-2 s-1) for Venus, Earth and Mars calculated from our model.The calculations were performed at 0.1 s of irradiation (1 pulse) for different compositions (indicated).

Table S4 :
HCN decomposition rates calculated for maximum and minimum partial pressure compositions and various irradiation times.