Icarus 177 (2005) 116–121 www.elsevier.com/locate/icarus
Condensed species in Titan’s atmosphere: Identification of crystalline propionitrile (C2H5CN, CH3CH2C≡N) based on laboratory infrared data R.K. Khanna ∗ Department of Chemistry and Biochemistry, University of Maryland, College Park, MD 20742, USA Received 8 October 2004; revised 15 February 2005 Available online 12 April 2005
Abstract We report the results of infrared studies of crystalline C2 H5 CN at several temperatures between 15 and 160 K. A case is made for the identification of crystalline C2 H5 CN in the stratosphere from the Voyager IRIS spectrum of Titan. 2005 Elsevier Inc. All rights reserved. Keywords: Atmospheres composition; Propionitrile; Titan; Spectroscopy; Infrared
1. Introduction It has been more than two decades since Voyager IRIS spectrometer on board provided a goldmine of infrared spectral data from various locations in Titan’s atmosphere. The IRIS instrument revealed the presence of several emission peaks in the 1500–200 cm−1 region, which were identified with seven hydrocarbon gases (methane—CH4 , ethane— C2 H6 , propane—C3 H8 , ethylene—C2 H4 , acetylene—C2 H2 , methylacetylene—C3 H4 , diacetylene—C4 H2 ) and three nitriles (hydrogen cyanide—HCN, cyanogen—C2 N2 , cyanoacetylene—HC3 N) also in the gas phases. These findings were reported in special issues of Nature (vol. 292, 1981) and Science (vol. 212, 1981) by several authors on the Voyager team. Carbon monoxide (CO) and carbon dioxide (CO2 ) were also identified by Lutz et al. (1983) and Samuelson et al. (1982), respectively. Subsequent ground based work identified acetonitrile (CH3 CN) gas in the upper stratosphere (Bèzard et al., 1993). A recent report by Marten et al. (2002) contains detailed observational results on nitriles in Titan’s stratosphere. With CO and CO2 as sources of oxygen, water is expected to be a possible photoproduct. * Fax: +1-301-314-9121.
E-mail address: [emailprotected]. 0019-1035/$ – see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.icarus.2005.02.014
A tentative identification of water vapor on Titan was reported by Coustenis et al. (1998). These findings have been incorporated into a working photokinetic model of Titan’s atmosphere Yung et al. (1984) and Yung (1987). It is argued that some of the photoproducts, particularly the nitrile derivatives with very low vapor pressures, would condense in the cold regions near the polar hood. Radiation induced polymerized species would provide the necessary nuclei for condensation. In Fig. 1, an average spectrum obtained from Voyager IRIS measurements at the north limb (RTSAT3) shows features tentatively associated with species in the solid phase, i.e., at 477 cm−1 (C4 N2 ), 505 cm−1 and 760 cm−1 (HC3 N), 700 cm−1 (unassigned), and 825 cm−1 (HCN). The region 200–240 cm−1 is practically accounted for by the gaseous C4 H2 and C2 N2 with, perhaps, a residue at 227 cm−1 , although, an average spectrum corresponding to the nadir viewing geometry over the north polar hood (NP50N) shows intense broad emission centered at 221 cm−1 , characteristic of a condensed species (Fig. 2). With the exception of C4 N2 , the identifications of solid species mentioned above, present some drawbacks. For example, the broad band centered at 221 cm−1 in Titan’s spectra does not agree with the laboratory data for H2 O ice at any temperature between 70 and 150 K (Coustenis et al., 1999).
Possible identification of propionitrile on Titan
117
Fig. 1. Voyager spectrum of Titan’s north limb (RTSAT3). Vertical bars show noise levels in different frequency ranges.
Fig. 2. Voyager spectrum of Titan’s north polar hood (NP50N). Vertical bars show noise levels in different frequency ranges.
Laboratory spectra of C2 H5 CN at 35 and 95 K (DelloRusso and Khanna, 1996) exhibit a strong peak at 226 cm−1 that is shifted 5 cm−1 downward from the observed peak in the Voyager spectra. Similarly, the width of the 505 cm−1 peak, inferred by subtracting the modeled gas HC3 N contribution from the total signal, is reported to be somewhat smaller than that in the laboratory spectrum (Coustenis et al., 1999). Some process(s) other than scattering must be involved; particulate scattering can only broaden the absorption/emission bands. Additionally, the broad feature at 760 cm−1 , obtained by subtracting modeled gas HC3 N contribution from the total signal, cannot be accounted for by condensed HC3 N alone. In the generally accepted model of Titan’s atmosphere, the photochemistry in the stratosphere involves molecules
and radicals from hydrocarbons and N2 (Toublanc et al., 1995). As the product species trickle down to lower altitudes (lower temperatures) condensation will occur at levels where the partial pressures exceed the saturation vapor pressures of the species. The condensation sites are possibly provided by the polymeric species resulting from irradiation of gases in its atmosphere. In Khare et al. ((1984) and references there in) work on irradiation of Titan mixtures is suggestive of polymeric material with characteristics similar to those of polyhydrogen cyanide as part of the visible haze on Titan. Nitrogen (N2 ), methane (CH4 ) and ethane (C2 H6 ), which are among the major constituents, cannot condense in Titan’s stratosphere and, hence, cannot provide sites for species like C2 H5 CN to condense on or to solvate them. Also, different photochemical product species cannot, in general, co-
118
R.K. Khanna / Icarus 177 (2005) 116–121
condense and produce mixed ices, unless they are present at the same cloud level. There are further symmetry restrictions on the formation of mixed ices; the crystal structures of the individual species must be the same. If the concentrations of the photochemical products species are of similar order of magnitude, they may phase separate during the annealing process. At low concentrations (1 part solute to 1000 or more parts solvent) results in matrix isolation and then the peak positions are drastically shifted. A recent report on the spectra of mixed ices containing nitriles and H2 O (Hudson and Moore, 2004) reveals no significant shifts in the C≡N band positions on mixing with H2 O. We can, therefore, argue that the condensed species on Titan can be considered as independently contributing to the emission flux. It is further argued that the surrounding temperatures can affect the positions and widths of the emission bands of the condensed species. In this manuscript we report results of detailed studies of the infrared spectra of solid C2 H5 CN at several temperatures between 15 and 160 K. The results indicate that annealed crystalline C2 H5 CN has a strong band at 221 cm−1 with a 7–9 cm−1 full width at half maximum (FWHM) between the temperatures 15 and 160 K and this agrees with the 221 cm−1 peak in the Voyager spectrum in several details. We discuss here possible contribution of solid C2 H5 CN in the 200–240 cm−1 region in Titan’s emission spectrum Of course, quantitative evaluation of the observed Voyager data will require redetermination of optical constants for the annealed samples, to be incorporated into the scattering program similar to the one employed for condensed C4 N2 (Samuelson et al., 1997).
2. Experimental Propionitrile (99%) was purchased from Aldrich and was used without further purification. The infrared spectra were obtained on a Perkin–Elmer GX instrument under 2 cm−1 resolution in the region 4000–370 cm−1 (KBr beam splitter) and under 4 cm−1 resolution in the 710–200 cm−1 region (Mylar beam splitter). The integrated intensities of the bands are already published in the literature (DelloRusso and Khanna, 1996); the current work is confined to temperature dependence of the band positions and bandwidths. The spectra were recorded at several temperatures from 15 to 160 K; the temperature of the sample was controlled with an externally powered heater wound around the sample holder which was attached to the cold head of a cryogenic cooler (Air Product Model CS 202). The temperature of the cold finger of the cryocooler was measured with a Gold-Chromel thermocouple which was calibrated at room temperature (295 K), ice point (273 K), dry ice (206 K) and liquid N2 (77 K) and a smooth graph was constructed by extrapolating the data between 15 and 295 K. The spectral traces with self-explanatory captions are given in Section 3.
3. Discussion The samples of C2 H5 CN in DelloRusso and Khanna’s (1996) work were deposited on a substrate window kept at 15 K. The samples were then warmed to 140 K and then cooled back to 35 and 95 K before recording their spectra. The strongest absorption peak for this compound below 1000 cm−1 is around 225–226 cm−1 , which is a poor match to the broad emission peak at 221 cm−1 in the Voyager spectrum. It is known that organics containing CH2 /CH3 groups have low barriers to rotation around single bonds. Consequently, samples quenched at low temperatures produce several possible orientations of these groups in the solid phase, which may be mostly amorphous. Annealing at higher temperatures is required to attain the most stable crystalline phase. Depending upon the sample deposition temperature and the time for annealing, the spectral peak positions and their relative intensities vary drastically. The sample may undergo some phase transitions as well. In order to obtain comparison with Titan’s spectral data, samples must be annealed as close to completion as possible. During this work we found that samples deposited at 15 and warmed to 160 K produced drastic changes in spectral details. Depositing the samples at 40–120 K gave different results. Keeping this sample at 120 K for about 4–5 h resulted in a spectrum that did not change subsequently until at 150 K when the sample started to sublime. We believe we obtained a reasonably complete annealing of crystalline C2 H5 CN when a sample deposited around 60–80 K was annealed at 100–120 K for about 4–5 h. The spectra of the annealed sample at several temperatures are reproduced in Fig. 3. Increasing intensities of the bands at 3300 and 800 cm−1 are due to H2 O (from the vacuum manifold) condensing on the substrate. Recycling the annealed sample to 15 K did not indicate any significant spectral changes except for minor changes in band widths and positions. The only exception noted was the 3000 and 1450 cm−1 regions corresponding to the CH3 /CH2 modes; these bands show different intensities and widths for samples deposited at different temperatures. In that sense, the CH3 /CH2 groups are still disordered to some extent. The differences between the spectra reported earlier (DelloRusso and Khanna, 1996) and the current data are as follows: First, this work shows an absorption feature at 221 cm−1 that is shifted from the 225–226 cm−1 band reported earlier. This coincides with the Voyager peak at 221 cm−1 . The width of the Voyager peak at 221 cm1 is 12 cm−1 ; the width of the corresponding laboratory obtained peak changes from 7 cm−1 at 15 K to 9 cm−1 at 150 K (Fig. 4). Second, earlier work (DelloRusso and Khanna, 1996) reported a doublet at 550 cm−1 . This work reports a feature at 547 cm−1 in the spectrum of the initial deposited sample at 15 K. As the sample is warmed, the band splits into 2 components and annealing at 130 K for 4–5 h again produces a sharper peak at 546 cm−1 . Third, the feature at 780 cm−1 is also differ-
Possible identification of propionitrile on Titan
119
Fig. 3. Infrared spectra of C2 H5 CN deposited at 60 K, annealed at 120 K for 4 h and cooled back to (a) 15, (b) 50, (c) 80, (d) 120, (e) 150, (f) 160 K.
Fig. 4. Infrared spectra of annealed crystalline C2 H5 CN in the region 200–600 cm−1 . (a) 15, (b) 50, (c) 80, (d) 120, (e) 150, (f) 160 K. The spectra are displaced vertically for clarity.
ent from the one reported in the earlier work. Although all bands show some differences from the earlier results, only the bands below 1000 cm−1 are of interest for the analysis of the Voyager data. The spectra in the region 200–600 cm−1 , shown in Fig. 4, give an idea of the temperature dependence of the widths of the 221, 393 and 546 cm−1 bands, respectively. 3.1. Identification of C2 H5 CN on Titan As discussed above, nitriles are the better choices among the candidates for condensed species. Previous studies report definitive identification of solids C4 N2 and tentative identifications of HC3 N and C2 H5 CN. In terms of production of these species several pathways are possible. For exam-
ple, among the pathways for the production of HC3 N, the following reactions may be among the major contributors (Toublanc et al., 1995): C2 H2 + CN → HC3 N + H, HCN + C2 H → HC3 N + H. It is to be noted that C2 H2 and HCN are among the most abundant photochemical product species from CH4 + N2 reactions. The published work by Hass et al. (1994) reports photoproduction of cyanobutadiyne (HC5 N) from a similar reaction: HC3 N + C2 H → HC5 N + H (this molecule has not yet been identified on Titan).
120
R.K. Khanna / Icarus 177 (2005) 116–121
The abundance of C2 H6 is somewhat larger than that of C2 H2 (a factor 4); its reaction with CN can be a source of C2 H5 CN: C2 H6 + CN → C2 H5 CN + H, C2 H5 + HCN → C2 H5 CN + H. We are not aware of any consideration on the reactions consuming C2 H5 CN (sinks) for this molecule. If the sinks are similar to those for other nitriles, it is expected that the abundance of C2 H5 CN is approximately the same as HC3 N within an order of magnitude. In terms of the Voyager data, in the north limb region (Fig. 1) there is a weak residual contribution at 227 cm−1 , possibly due to condensed HC3 N (Coustenis et al., 1999). For other geometries (Fig. 2), the 221 cm−1 band shows up strongly. Since H2 O ice and HC3 N can be ruled out as the major contributors to the emission at 221 cm−1 , C2 H5 CN is a more likely possibility and other gaseous species may be involved. The increased width of the Voyager peak at 221 cm−1 can be explained, at least partially, by particle scattering and lower resolution of the IRIS instrument, but contribution from other species cannot be ignored. The bandwidths of the 221, 393, and 546 cm−1 bands can be estimated from the spectral traces in the region 200–600 cm−1 in Fig. 4. Gaseous C2 H5 CN has strong absorption in the 206– 215 cm−1 region (Cerceau et al., 1985) but it is not completely resolved from strong and broad feature due to solid C2 H5 CN in the Voyager spectrum. Dicyanoacetylene also shows the same behavior, which was explained (Yung, 1987) in terms of its stability against photodecomposition in the cold environment; similar reasoning may apply for C2 H5 CN. That C2 H5 CN is likely to condense in the cold environment of Titan’s north-polar region is shown from thermodynamic considerations. From the published data (NIST website: http://webbook.nist.gov) o = 39.16 kj mol−1 , Hvap
Tbp = 370 K,
Hfus = 15.5 kj mol−1 ,
Tmp = 180 K = Ttp ,
one can estimate the vapor pressure of C2 H5 CN for the solid C2 H5 CN. The results are: vapor pressure at 100 K is 3 × 10−15 Torr, vapor pressure at 120 K is 9 × 10−11 Torr, vapor pressure at 140 K is 1.5 × 10−7 Torr. 140 K corresponds to approximately 10 Torr total pressure on Titan; therefore, mixing ratio of C2 H5 CN at the saturation level is 1.5 × 10−8 which is of the same order of magnitude as HC3 N. Thus, C2 H5 CN is likely to be completely condensed in the atmosphere at 100–120 K. In addition to the 221 cm−1 band, C2 H5 CN has characteristic bands at 393, 546, and 780 cm−1 but these are not observed in the Voyager spectrum. The emission flux at
120 K for the four bands is proportional to the product of the Planck function and the integrated band intensity. On a relative scale, their expected fluxes are 1, 0.003, 0.0021, 0.05 for the 221, 393, 546, and 780 cm−1 bands, respectively. Except for the 221 cm−1 band, others fall within the noise level of the instrument. Therefore, we propose that the major contributor to the 221 cm−1 region of the Voyager spectrum of Titan is C2 H5 CN with minor contribution from HC3 N and may be H2 O ices. It is to be mentioned that rigorous radiativetransfer calculations on the emission flux from a C2 H5 CN cloud requires optical constants data at several temperatures, ideas on the particle size distribution and concentration profile as well as the temperature profile of Titan’s clouds. We have the laboratory data on transmission spectra of crystalline C2 H5 CN at several temperatures which can be utilized to obtain the complex refractive indices. With Cassini mission ready to provide additional (and possibly more sensitive) data, this report will be of additional utility for the analysis of the Titan infrared data.
Acknowledgments This work was supported by a grant for NASA’s Planetary Atmospheres Program. The P.I. thanks Dr. Robert Samuelson fore useful discussions concerning these results. Mr. James Royer and Mr. Nilesh Bhumralkar participated in this project as undergraduate interns.
References Bèzard, B., Marten, A., Paubert, G., 1993. Detection of CH3 CN on Titan. Bull. Am. Astron. Soc. 25, 1100. Cerceau, F.F., Raulin, R., Gautier, D., 1985. Infrared spectra of gaseous mononitriles: application to atmosphere of Titan. Icarus 62, 207–220. Coustenis, A., Salama, A., Lellouch, E., Encrenaz, Th., Bjoraker, G., Samuelson, R.E., deGraauw, Th., Feuchtgruber, H., Kessler, M.F., 1998. Evidence for water vapor in Titan’s atmosphere from ISO/SWS data. Astron. Astrophys 336, L85–L89. Coustenis, A., Schmitt, B., Khanna, R.K., Trotta, F., 1999. Plausible condensates in Titan’s stratosphere from Voyager infrared spectra. Planet. Space Sci. 47, 1305–1329. DelloRusso, N., Khanna, R.K., 1996. Laboratory infrared spectroscopic studies of crystalline nitriles with relevance to outer planetary systems. Icarus 123, 366–395. Hass, S., Winnewisser, G., Yamada, M.T., 1994. The high resolution infrared spectrum of cyanobutadiyne: the ν7 band centered at 642 cm−1 . Can. J. Phys. 72, 1165–1178. Hudson, R.L., Moore, M.H., 2004. Reactions of nitriles in ices relevant to Titan, comets and interstellar medium: formation of cyanate ion, ketenimines and isonitriles. Icarus 172, 466–478. Khare, B.N., Sagan, C., Arakawa, E.T., Suit, F., Callcott, T.A., Williams, M.W., 1984. Optical constants of organic tholins produced in a simulated Titanian atmosphere: from soft X-ray to microwave frequencies. Icarus 60, 127–137. Lutz, B.L., deBergh, C., Owen, T., 1983. Titan: discovery of CO in its atmosphere. Science 220, 1374–1375.
Possible identification of propionitrile on Titan
Marten, A., Hydayat, T., Biraud, Y., Moreno, R., 2002. Vertical distributions of nitriles HCN, HC3 N, CH3 CN and the isotopic ratio 15 N/14 N in its atmosphere. Icarus 58, 532–544. Samuelson, R.E., Maguire, W.C., Hanel, R.A., Kunde, V.G., Jennings, D.E., Yung, Y.L., Aikin, A.C., 1982. CO2 on Titan. J. Geophys. Res. 88, 8709–8715. Samuelson, R.E., Mayo, L.A., Knuckles, M.A., Khanna, R.K., 1997. C4 N2 ice in Titan’s north polar stratosphere. Planet. Space Sci. 45, 941–948.
121
Toublanc, D., Parisot, J.P., Brillet, J., Gautier, D., Raulin, F., McKay, C.P., 1995. Photochemical modeling of Titan’s atmosphere. Icarus 113, 2– 26. Yung, Y.L., 1987. An update of nitrile photochemistry on Titan. Icarus 72, 468–472. Yung, Y.L., Allen, M., Pinto, J.P., 1984. Photochemistry of the atmosphere of Titan: comparison between model and observations. Astrophys. J. 55, 465–506.