Infrared Spectra and Vapor Pressures of Crystalline C2N2, with Comparisons to Crystalline HCN

In a continuation of our work on nitriles, we have examined cyanogen (C2N2) as a crystalline solid, comparing the results to our recent work on crystalline hydrogen cyanide (HCN). A density and refractive index for C2N2 ice were measured and used to prepare solid samples from which infrared (IR) spectra, band strengths, and optical constants were measured. The vapor pressures (sublimation pressures) of both C2N2 and HCN ices were determined with a quartz-crystal microbalance at temperatures relevant to Titan's atmosphere and much lower than those in the literature. Comparisons of the newly measured low-temperature vapor pressures to values extrapolated from higher temperatures revealed differences on the order of 45 and 130%. The enthalpies of sublimation of C2N2 and HCN in the 120 K region were measured and found to be similar to those at higher temperatures. Comparisons were made to previous spectroscopic work, but hindered in the case of IR intensities by the lack of published details.


Introduction
Nitriles are molecules with a −C≡N: functional group, and are found within and beyond the solar system.The cyano radical (CN) can be taken as the simplest nitrile, and has long been observed as a component of cometary comae.Among the simpler nitriles is cyanogen (C 2 N 2 ), which has been identified in comets along with hydrogen cyanide (HCN), acetonitrile (CH 3 CN), and cyanoacetylene (HCC-CN); see Hänni et al. (2021).Cyanogen also is known in the interstellar medium (ISM) through its conjugate acid, HC 2 N 2 + (Agúndez et al. 2015).Members of the homologous series H-(C≡C) x -C≡N:, with x = 0-5, have been identified in the ISM by radio observations in the gas phase (e.g., Loomis et al. 2021), as have various unsaturated nitriles, nitrile isomers, the nitrile ion OCN − , known as cyanate, and two aromatic nitriles (Soifer et al. 1979;Lovas et al. 2006;McGuire et al. 2018McGuire et al. , 2020;;Sita et al. 2022).
The N 2 + CH 4 atmosphere of Titan is relatively rich in nitriles, with evidence for at least seven in the literature (i.e., HCN, C 2 N 2 , CH 3 CN, CH 3 CH 2 CN, HCCCN, C 2 H 3 CN, and C 4 N 2 ), with identifications coming from both ground-and space-based observations as well as in situ measurements (e.g., Kunde et al. 1981;Khanna et al. 1987;Lai et al. 2017).Such identifications and associated analyses rely heavily on laboratory measurements, to which we have contributed.For example, in Moore et al. (2010) we examined HCN, C 2 N 2 , CH 3 CN, C 2 H 5 CN, and HC 3 N, returning with new data to HCN, CH 3 CN, and C 2 H 5 CN in later papers (Hudson 2020;Gerakines et al. 2022).
In this paper, we focus on crystalline C 2 N 2 .We describe measurements of infrared (IR) intensities, comparing and contrasting with our recent work on HCN.For this, we required determinations of a reference density and refractive index for C 2 N 2 ice, which we also report here.Finally, we describe the first vapor-pressure measurements of solid C 2 N 2 and solid HCN in the 100-140 K region, lower than in previous papers and specifically chosen for applications to Titanʼs atmosphere.

Laboratory Procedures
Details of our experimental procedures and equipment are in print, and so only a summary is given here; see, for example, Moore et al. (2010) or Hudson et al. (2020) for additional information.
Our C 2 N 2 was from Matheson and contained ∼ 0.2% HCN, along with trace amounts of H 2 O and CO 2 (< ∼0.01%).No attempt was made to remove these impurities as the levels involved were insufficient to cause detectable changes in either vapor pressures or IR band strengths within the accuracy and sensitivity of our measurements.Our HCN was synthesized by the reaction of potassium cyanide and stearic acid as in earlier papers (Gerakines et al. 2004(Gerakines et al. , 2022)).Caution: Both HCN and C 2 N 2 are highly toxic and should be used only by experienced personnel.
Crystalline ices were prepared by gas-phase deposition of either C 2 N 2 or HCN onto a precooled CsI substrate in a vacuum chamber (∼10 −8 Torr).The condensation rate of the room-temperature gases was such as to give an increase in the resulting iceʼs thickness of 3-4 μm hr −1 .Spectra were recorded with a Thermo iS50 spectrometer at resolutions from 2 up to 0.125 cm −1 , but mainly at 1 and 0.5 cm −1 , over about 3000-500 cm −1 for C 2 N 2 and about 5000-500 cm −1 for HCN with 200 accumulations per spectrum.
Earlier laboratory work suggested that a temperature of 90 K was sufficient to crystallize amorphous cyanogen on warming from lower temperatures, and that 100 K was a reasonable estimate for C 2 N 2 condensation in Titanʼs atmosphere (Coustenis et al. 1999); see also later work on condensation by Lai et al. (2017), Anderson et al. (2018), andBézard et al. (2018).Therefore, our cyanogen ices were grown at 100 K and held there for recording spectra.The IR spectra of such ices were essentially identical to those of crystalline C 2 N 2 found in the literature (vide infra).
Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 licence.Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
Infrared band strengths, denoted A′ (in cm molecule −1 ) were obtained in the usual manner by integrating an IR band of interest in the spectrum of an ice of known thickness (Hollenberg & Dows 1961).Several such integrations with ices of different thicknesses were plotted to give a Beerʼs law curve in accord with Equation (1): In Equation (1), h is the ice thickness in cm, ln(10) converts the absorbance spectrum (common logarithm base) to an optical-depth scale (natural logarithm base), and ρ N is the number density (molecules cm −3 ).The latter was calculated from ice densities as ρ N = ρ (N A /M), where M is the molar mass (g mole −1 ) of the compound studied, ρ (in g cm −3 ) is mass density, and N A is Avogadroʼs constant (molecules mole −1 ).Each plot of the left-hand side of Equation (1) as a function of h possessed a slope (ρ N A′/ln(10)) from which A′ was calculated.All such Beerʼs law plots had correlation coefficients of 0.998 and higher; see our earlier papers for examples of the same method (e.g., Gerakines & Hudson 2015).
Our usual method for determining ice thickness is to record interference fringes during the growth of an ice; see, for example, our previous work with HCN (Gerakines et al. 2022).However, the IR features of C 2 N 2 were so weak that the initially uniform pattern of fringes was lost before ices were sufficiently thick to show IR bands intense enough for reliable integration.Therefore, thicknesses for crystalline cyanogen were found from the smooth, regular pattern of channel fringes in the iceʼs IR spectrum.For N fr channel fringes observed over a wavenumber interval  n D , the sampleʼs thickness was just N fr /(2 n 670  n D ); see, as examples, either Harrick (1971), Griffiths & de Haseth (1986), or Ospina et al. (1988).In all cases, a refractive index (n 670 ) was needed to determine an ice thickness and an ice density was required to calculate a band strength with Equation (1).The required refractive indices and ice densities (ρ) were measured in an ultra-high vacuum (UHV) chamber (∼10 −10 Torr).Two-laser interferometry gave n 670 and microgravimetry with a quartzcrystal microbalance (QCM) gave ρ (Tempelmeyer & Mills 1968;Lu & Lewis 1972).
Again, the experimental techniques employed resembled those used in our laboratory and elsewhere for many years.Readers unfamiliar with these methods might wish to consult some of our earlier papers such as Moore et al. (2010), Hudson et al. (2020), or Gerakines et al. (2022), and references therein.

Density and Refractive Index
Our measurements of the refractive index and density of crystalline HCN at 120 K were reported in Gerakines et al. (2022).We were unable to find comparable values for cyanogen, so they were measured in our UHV chamber.The result for crystalline C 2 N 2 at 100 K was n 670 = 1.462 ± 0.003 and ρ = 1.186 ± 0.005 g cm −3 as averages and standard errors from five determinations.The angles of the two lasers in our UHV chamber are known to three significant figures, but an extra number has been retained in these n 670 and ρ results.Rounding can be done as desired.We return to these n 670 and ρ values in our Discussion (see Section 4).

Infrared Spectra of Ices
Our recent results for HCN (Gerakines et al. 2022) provide a convenient comparison to the new work on C 2 N 2 presented here, and are so used.The IR spectra of hydrogen cyanide and cyanogen are quite different.With three atoms in a linear arrangement, HCN has 3N-5 = 3(3)-5 = 4 fundamental vibrations, each being both IR and Raman active.All can be seen in the upper trace of Figure 1, with the IR feature near 826 cm −1 being a doubly degenerate bending mode.This spectrum is of an ice with a column density of about 1.1 × 10 18 HCN molecules cm −2 .Table 1 gives band positions and intensities (i.e., A′) from Gerakines et al. (2022).
Cyanogen is a four-atom linear molecule, but its center of symmetry means that it is nonpolar and that there are no vibrations that are both IR and Raman active in the gas and crystalline phases.Its 3N-5 = 3(4)-5 = 7 fundamental vibrations include just three that are IR active, an asymmetric C≡N stretching mode near 2165 cm −1 and a doubly degenerate bending near 240 cm −1 , which is beyond the range we studied.Table 2 summarizes the activity and positions of the fundamental vibrations of crystalline C 2 N 2 (Andrews et al. 1984).Note that our description of the ν 2 vibration differs from that in the compilation of Shimanouchi (1972) due to a typographical error in the latter.
Crystalline cyanogen is unusual in being composed of polyatomic molecules with but one fundamental vibration in the mid-infrared (mid-IR) region.The spectrum of crystalline C 2 N 2 ice at 100 K is shown in Figure 2, with the three labeled transitions being the strongest bands from 5000 to 300 cm −1 , and the main cyanogen features examined for this study.Each of the three bands of Figure 2 consists of a small and large component, as expected from the theory of vibrations in crystalline solids (e.g., Verderame et al. 1963).As for intensities, even a casual comparison of the ice thicknesses and peak heights in Figures 1 and 2 shows that C 2 N 2 is a particularly weak IR absorber compared to HCN.
With Equation (1) in mind, the IR spectra of nine crystalline C 2 N 2 ices at 100 K with thicknesses from about 0.7 × 10 −4 to 5 × 10 −4 cm (0.7 to 5 μm) were recorded and the three bands in Figure 2 integrated for each ice sample.Plots of the integrals as a function of thickness were linear (correlation coefficients >0.998) and gave slopes from which band strengths (A′) were calculated.Table 3 summarizes the results.Note that a resolution of 1 cm −1 was used to record the IR spectra of Figures 1 and 2, and for the data leading to Tables 1 and 3, higher resolutions (e.g., 0.5 cm −1 ) giving essentially the same results.Our resolutions bracketed the 0.6 cm −1 used by Ospina et al. (1988).
The uncertainties in our A′ values are mainly due to the determination of cyanogen ice thicknesses, which depended on the n 670 we measured.The fact that we were able to measure n 670 to a reasonably high accuracy, coupled with the relative sharpness of the three IR bands examined and the lack of interference from overlapping features, enabled thickness uncertainties to be lowered from the ∼25% of Ospina et al. (1988) to about 2%.A propagation-of-errors approach and a least-squares regression analysis with uncertainties in both the x-and y-axes of Beerʼs law plots (Irvin & Quickenden 1983;Press et al. 1992) was used to calculate an uncertainty of about 8% in A′ for the ν 3 + ν 4 and ν 3 bands, and about 2% for the stronger ν 4 + ν 6 feature.
Having new IR spectra of C 2 N 2 for ices of known thickness, it seemed reasonable to use these same spectra to calculate optical constants n( n) and k( n) for cyanogen at 100 K. Our method was the same as in Gerakines & Hudson (2020) using the open-source software described in that paper, which was developed by one of us (P.A.G.) and remains the only such free, open software for the planetary-science community.Infrared spectra used for these calculations were for crystalline C 2 N 2 made, and spectra recorded, at 100 K with 1 cm −1 resolution.Figure 3 shows the results for the same three regions of our Figure 2; see https://science.gsfc.nasa.gov/691/cosmicice/constants.html or the Zenodo repository (doi:10.5281/zenodo.8335802)for our optical constants of crystalline cyanogen at 100 K.

Vapor Pressures of Ices
Vapor pressures are needed to determine the condensation height of atmospheric components of Titan and other worlds (e.g., Sagan & Thompson 1984), but the relevant laboratory data for both HCN and C 2 N 2 are very limited.Lewis & Schutz (1934) reported vapor pressures for solid HCN at six temperatures, followed by Appleton & Van Hook (1982), who published a vapor-pressure curve for solid HCN at 237-256 K, based on 26 points.To these measurements can be added about 20 data points for solid C 2 N 2 found in refereed publications, covering roughly 200-240 K (Perry & Bardwell 1925;Ruehrwein & Giauque 1939).Figure 4 summarizes the literature results for HCN and C 2 N 2 and shows that the available vapor pressures are far above the ∼100-150 K region at which nitriles are expected to condense in Titanʼs atmosphere.The altitude of ∼200 km for 150 K in Figure 1 is from Anderson et al. (2018), but values closer to 150 km (Lai et al. 2017) or even 100 km (Bézard et al. 2018) can be found in the literature.
The lack of low-temperature vapor-pressure data for HCN and C 2 N 2 has motivated extrapolations by several researchers to cover the temperature gap down to ∼120 K and lower, with accompanying pressure extrapolations over about five orders of magnitude; see Fray & Schmitt (2009) for a discussion of proposed extrapolations and curve fits.Here, we address this paucity of laboratory data by reporting vapor pressures of solid HCN and C 2 N 2 between about 120 and 140 K for HCN and 105 and 120 K for C 2 N 2 .
Our method and equipment for vapor-pressure measurements have been described in two previous publications, one on propanal (Yarnall et al. 2020) and the other on benzene (Hudson et al. 2022), and resemble the approach of Luna et al. (2012).Briefly, we grew ices on the precooled gold-coated surface of a QCM held in a UHV chamber.The increase in the microbalanceʼs oscillator frequency was recorded as the resulting ice was slowly warmed (at 1 K minute −1 ) to sublimation.At each time step, comparison of the observed frequency to the known frequency of a blank substrate at the same temperature gave the extent to which the observed changes were due to nitrile-ice sublimation.More specifically, with f 0 being the QCMʼs frequency (in hertz) with no sample present and f being the oscillator frequency with an ice on the Note.
a Positions are from Andrews et al. (1984).Splittings and substructure are seen for all but the ν 2 vibration.The ν 4 and ν 5 vibrations are doubly degenerate.substrate, then gave the iceʼs mass per unit area (μ, in grams per square centimeter) at each temperature during a warming sequence, with κ = 4.417 × 10 5 Hz g cm −2 (Lu & Lewis 1972).The flux (F) of subliming nitrile molecules was calculated from the time derivative of μ, and then vapor pressures were found from Equation (3), where m is the mass of a molecule of the nitrile ice: Readers unfamiliar with the use of Equation (3) might wish to consult a standard textbook of physical chemistry, such as that of either Chang (2000) or Atkins & de Paula (2006).
The procedure just described was followed for four HCN ices and the resulting vapor pressures averaged at temperatures (±0.1 K) from 121 to 139 K at 1 K intervals.A Clausius-Clapeyron plot (i.e., ln P as a function of 1/T) gave a line with a slope, obtained from a least-squares fit, from which an enthalpy of sublimation, ΔH subl , was calculated.The result was Equation (4) below, with pressure in Torr (i.e., a standard state of 1 Torr): Units of Torr are used here to match our previous work and the older literature, but conversions to other pressure units can be made with 1 Torr = 133.322Pa and 1 bar = 10 5 Pa; see also Tables 6 and 7 in the Appendix.
Work similar to that just described was carried out with four cyanogen ices.The Clausius-Clapeyron equation obtained was Equation (5) below, from which ΔH subl = 36.02± 0.03 kJ mol −1 was derived.The literature result for C 2 N 2 ice is 33.6 kJ mol −1 from measurements at ∼180-250 K (Stull 1947): We note that the difference in ΔH subl values for our nitriles is about 2 kJ mol −1 , with the HCN value being the higher, compared to the same order, the same difference, and similar values in the compilation of Stull (1947).
Our vapor pressures for HCN and C 2 N 2 are given in Tables 4 and 5 and in the lower-left corner of each panel of Figure 5.We have connected our low-temperature results to the highertemperature values in the literature for HCN and C 2 N 2 by using   Ruehrwein (1939).
The vapor pressures in the second column of Tables 4 and 5 were found from the Clausius-Clapeyron fits to the data for HCN and C 2 N 2 (Equations ( 4) and (5), respectively).A variation of ±0.1 K for our measured temperatures can lead to systematic uncertainties in these pressures, which we evaluated by finding the maximum and minimum pressures in the Clausius-Clapeyron fits for a 0.2 K window centered on the temperature for which each pressure is reported.The final vapor-pressure uncertainties are 3%-4% for the temperatures listed in Tables 4 and 5, with the higher uncertainty at the lower temperatures of Table 5 for C 2 N 2 .

Densities and Refractive Indices
The density and refractive index of crystalline HCN have been discussed previously (Gerakines et al. 2022), so here we consider only results for C 2 N 2 .Suffice it to say that our density of ρ = 1.186 g cm −3 compares favorably with an earlier value of 1.25 g cm −3 from a diffraction measurement, with the crystalline ice grown by a different method (Parkes & Hughes 1963).Our n 670 = 1.462 is slightly higher than that of Moore et al. (2010), 1.42 at 90 K.A value of ∼1.37 ± 0.07 was estimated by Ospina et al. (1988), but few details were provided, such as the temperature and the number of measurements.
Another way to evaluate our n and ρ results for C 2 N 2 is to calculate the moleculeʼs polarizability (α) using Equation (7): Our n and ρ values for cyanogen give α = 4.85 Å 3 , within about 3% of the 5.02 Å 3 value of Hirschfelder et al. (1954).In contrast, combining the n values of Moore et al. (2010) and Ospina et al. (1988) with the ρ from diffraction work gives differences of about 17% and 26%, respectively, this same 5.02 Å 3 .

Infrared Spectra of Ices
As with the n 670 and ρ of crystalline HCN, that iceʼs IR spectrum was discussed by Gerakines et al. (2022), so here we focus on C 2 N 2 .The weakness of cyanogenʼs mid-IR features already has been mentioned, an observation made earlier by Verderame et al. (1963).Those authors, and later Ospina et al. (1988) and Dello Russo & Khanna (1996), published IR spectra of crystalline C 2 N 2 , but on such small scales that direct visual comparisons are difficult.Qualitatively, our spectrum of crystalline C 2 N 2 resembles that of Verderame et al. (1963) at 80 K. Peak positions in those same authors' Table 1 are within about 1 cm −1 of our own and clearly indicate a small side peak for each of the three large bands of our Figure 2. Verderame et al. (1963) also show the same two features that we found for the band near 2165 cm −1 , and with the same separation within those authors' stated uncertainty of 0.5 cm −1 with a dispersive spectrometer.We also observed weak isotopic features (e.g., 2148 cm −1 for 15 NCCN and 2139 cm −1 for NC 13 CN) reported by Verderame et al. (1963).With our highest resolution, 0.125 cm −1 , we were just able to discern an additional splitting in the ν 3 peak of Figure 2, confirming the small splitting reported by Verderame & Nixon (1965), apparently for the first time.We conclude from these largely qualitative observations that the crystalline C 2 N 2 ices we studied were essentially identical in nature to those in the relevant spectroscopic literature.
Of the previous publications on IR spectra of crystalline C 2 N 2 , only the papers of Ospina et al. (1988) and Dello Russo & Khanna (1996) provide ice thickness, numerical scales, band strengths, and spectral resolution needed for quantitative comparisons.However, a difference in refractive indices to calculate ice thicknesses makes a direct comparison difficult.Ospina et al. (1988) reported an estimate of n, but few details were provided.From the estimated n, ice thicknesses were determined with channel fringes, but the spectrum shown possesses a very irregular fringe pattern, which introduces considerable uncertainty into thickness measurements and thus band strengths.
A second obstacle to making quantitative intensity comparisons to the papers of Ospina et al. (1988) and Dello Russo & Khanna (1996) is those papers' lack of integration ranges.Without such limits it is impossible to know if both the large and small components in our Figure 2 were integrated in the  earlier papers or if only the areas the larger bands were measured.
Little can be said in relation to the earlier work of Ospina et al. (1988) and Dello Russo & Khanna (1996) except that the ν 3 + ν 4 and ν 3 IR bands in their spectra have about the same integrated intensities, as do ours, and that the ν 4 + ν 5 feature is about 5 times stronger, in rough agreement with our work.These are not comparisons on an absolute scale, yet relative intensities are still valuable as they permit scaling to unmeasured IR features.For example, the far-IR band of C 2 N 2 near 244 cm −1 was found by Khanna and colleagues to be about 28 times more intense than the strong ν 4 + ν 5 feature we measured at 745 cm −1 .From our Table 3, we can estimate that A′(244 We know of only two other sets of IR optical constants for cyanogen with which to compare our n( n) and k( n) results of Figure 3.Our value of k(745 cm −1 ) at 100 K and 1 cm −1 resolution is just over twice that of Moore et al. (2010), who used a lower resolution (2 cm −1 ) and temperature (95 K), and similar comments apply to other IR peaks.The C 2 N 2 ice of Moore et al. (2010) was made at 20 K as an amorphous solid and then warmed to induce crystallization, which might explain some of the differences in spectra and optical constants.There is also the possibility that the crystallization was incomplete after that warming, although none of the nominally forbidden IR bands seem to be present (i.e., the Raman allowed features of our Table 2).Little more can be said without additional work.Ospina et al. (1988) showed mid-IR optical constants for crystalline C 2 N 2 , but the figure published is so small that precise comparisons to our work are difficult.The tallest mid-IR peak has k(745 cm −1 ) ≈ 0.040 at 70 K, while we find k(745 cm −1 ) ≈ 0.065 at 100 K.It is possible, as with the case of Moore et al. (2010), that amorphous ice is present in the C 2 N 2 sample of Ospina et al. (1988) due to the lower deposition temperature, but more work is needed to test this possibility.Differences in computational methods also might explain the differences in k(745 cm −1 ).Differences in n( n) seem mainly due to the different choices of a reference refractive index.

Vapor Pressures of Ices
There are no published vapor pressures available for comparison to our results for HCN and C 2 N 2 ices at ∼100 to 140 K.However, comparisons can be made to results extrapolated from higher temperatures.Fray & Schmitt (2009) reviewed vapor-pressure data in the literature for our two compounds and suggested equations for extrapolating.The comparisons are easily summarized: 1.For crystalline HCN, the five-term equation of Fray & Schmitt (2009) leads to vapor pressures from about 120 to 140 K that are about 130% too high when compared to our measured values.Our version of the Cox equation, seen in the upper panel of Figure 5, leads to an average deviation of about 4%. 2. For crystalline C 2 N 2 , the two-term Clapeyron equation of Fray & Schmitt (2009) leads to vapor pressures from 105 to 120 K that are about 45% too high when compared to our measured values, while our version of the Cox equation, seen in the lower panel of Figure 5, leads to an average deviation of about 1%.b Pressures in this column represent the average over a range covering ±0.1 K of the temperature listed.For example, the pressure listed at 121 K covers data from 120.9 to 121.1 K. c Vapor pressures of HCN in this column were calculated with the equation proposed by Fray & Schmitt (2009).We recommend that the equations we have HCN and C 2 N 2 vapor pressures be used in the future unless percent differences on the order of 50%-100% can be tolerated.

Applications and Future Work
Perhaps the parts of our work that are most directly applicable to planetary science is that (i) we have measured HCN and C 2 N 2 vapor pressures at lower temperatures than before, and (ii) we have found those pressures to be slightly lower than expected when compared to extrapolations of published results.This implies that the condensation of these two gases to make crystalline ices in Titanʼs atmosphere will occur at slightly higher altitudes than stated in the literature, although the difference probably is not great.Extensions to other compounds of the vapor-pressure measurements described here are desirable.
Beyond vapor pressures, our new n and ρ values for crystalline C 2 N 2 will be useful in future work on this compound, such as the preparation of ices of known composition.We recently published examples of how n, ρ, and band strengths can be useful in making ice mixtures of very precise composition (Yarnall & Hudson et al. 2022), which could be extended to mixtures with cyanogen.Similarly, our band strengths can be used to make cyanogen ices of known column density (or thickness) so that quantitative measurements of photolytic or radiolytic yields, for example, can be carried out.
Qualitatively, our IR spectra of HCN and C 2 N 2 are as expected from earlier work on these compounds.What is new is that we have reported band strengths with specific integration limits, have checked for resolution limitations, and have measured both n and ρ for our ices.All of these tasks lead to the best mid-IR band strengths for crystalline HCN and C 2 N 2 now available for these compounds, A′ values that can be adopted as reference values for studies at temperatures and wavenumbers (or wavelengths) other than those we have used.It is easy to conceive of future investigations that cover these variables.
As a final comment, we note that cyanogen (:N≡C-C≡N:) has the same NCC arrangement of atoms as amino acids.One can envision the possibility of reduction of one of the moleculeʼs C≡N triple bonds to give aminoacetonitrile, followed by hydrolysis to give glycine, as in the sequence shown below: We are not aware of laboratory investigations of these changes in ices.

Conclusions
Infrared band strengths for crystalline C 2 N 2 have been newly measured and the results presented, along with new measurements of the underlying physical properties, n and ρ, on which band strength determinations rest.The IR spectra previously published for solid C 2 N 2 have been verified qualitatively, even the small solid-state splittings and isotopic features.There is at least rough quantitative agreement with previous reports of IR intensities, but the lack of integration limits and uncertain n and ρ values in earlier work hinders close comparisons.
Vapor pressures for crystalline HCN and C 2 N 2 are reported here for lower temperatures than in earlier work, temperatures relevant to Titanʼs atmosphere.The lower values of the new vapor pressures, compared to those obtained from extrapolations of higher-temperature work, shift the condensation of these compounds to slightly higher Titan altitudes.Future work will determine the size of the change.Notes.
a Vapor pressures rounded to two decimal places.Values at temperatures not listed can be found by using Equation (6).Vapor-pressure uncertainties are 3%-4%.b Pressures in this column represent the average over a range covering ±0.1 K of the temperature listed.For example, the pressure listed at 121 K covers data from 120.9 to 121.1 K. c Vapor pressures calculated from column (2) using 1 Torr = 133.322Pa and 1 bar = 10 5 Pa.

Figure 2 .
Figure 2. The three strongest features in the mid-IR spectrum of crystalline C 2 N 2 deposited and recorded at 100 K. Small peaks near 2148 and 2139 cm −1 are from 15 NCCN and NC 13 CN, respectively.A weak, broad band near 820 cm −1 is from HCN.The ice thickness was 4.7 μm = 4.7 × 10 −4 cm (channel fringes removed for clarity).

Figure 3 .
Figure 3. Optical constants for crystalline cyanogen at 100 K.The three regions are the same as in Figure 2.

Figure 5 .
Figure 5. Vapor pressures of crystalline HCN and crystalline C 2 N 2 from the literature results and from the present work.The high-and low-temperature data in each panel are connected with a smooth curve, the Cox equation.

Figure 4 .
Figure 4. Vapor pressures of crystalline HCN and crystalline C 2 N 2 from the literature.In the upper panel, six data points of Lewis & Schutz (1934) have been plotted, along with values at six temperatures from the vapor-pressure curve of Appleton & Van Hook (1982); see the text for sources of data for C 2 N 2 .

Table 1
Fundamental Vibrations of Crystalline HCN at 120 K a Assignment Description  n /cm −1 A′/10 −18 cm molecule −1 a Positions and intensities are from Gerakines et al. (2022).All vibrations are both IR and Raman active.

Table 2
Fundamental Vibrations of Crystalline C 2 N 2 at 20 K a

Table 4
Vapor Pressures of Crystalline HCN a

Table 6
Vapor Pressures of Crystalline HCN