PHOTODISSOCIATION DYNAMICS OF METHYLAMINE CATION AND ITS RELEVANCE TO TITAN'S IONOSPHERE

Titan's dense atmosphere is composed primarily of nitrogen and methane, while less abundant hydrocarbons and nitriles (<ppm) are generated from electron impact and photochemistry involving these major components (Kuiper 1944; Hanel et al. 1981; Yung & Demore 1999). The neutral hydrocarbons and nitriles are then ionized by solar radiation and plasma to create the ionosphere at an altitude of 800 km (Keller et al. 1992) and above. The ionosphere not only plays a crucial role in the chemical and thermal balance of the upper atmosphere, but also establishes a link between Saturn's magnetosphere and neutral atmosphere of Titan (Wilson & Atreya 2004). It may play a central role in the formation of Titan's notorious aerosol haze (Wilson & Atreya 2003).

The Voyager 1 mission in 1980 (Bird et al. 1997) was the first attempt to experimentally study the ionosphere in Titan, more recently extended by the Cassini/Huygens mission in 2005 (Wahuland et al. 2005; Waite et al. 2005). Cassini has performed the first in situ measurements of the environment of Titan's ionosphere with the ion and neutral mass spectrometer (INMS) experiment. Cassini also reported the surprising observation of very large anions in the upper atmosphere at an altitude of 900 km and above, but this is not the subject of the present investigation. Apart from these measurements, there are many other laboratory-based studies (Anicich et al. 2003; Anicich & McEwan 2006) and models (Carrasco et al. 2007, 2008; Vervack et al. 2004; Galand et al. 1999) that have been reported in the literature to explore the nature of the ionosphere. These studies indicate that a description of Titan's ion chemistry is crucial for our understanding not only of Titan's upper atmosphere but also of the aerosol formation process in general. In Titan's environment, major neutral species N₂ and CH₄ are ionized by solar radiation and plasma producing N₂⁺, N⁺, CH₄⁺, CH₃⁺, CH₂⁺, and CH⁺ ionic species, which further convert into higher mass compounds through complicated ion-neutral chemistry. Cassini's INMS shows the presence of a total of 45 different ions in Titan's ionosphere, with m/z = 28 as the most abundant ion. Further analysis of the data identified the mass 28 principally as HCNH⁺ ion, consistent with the theoretical model (Cravens et al. 2006). After these missions, several experimental and theoretical studies have been performed showing the importance of HCNH⁺ in many ion-neutral reactions, including those in Titan's ionosphere (Cui et al. 2009; Vuitton et al. 2007).

Apart from the m/z = 28 peak, there are several peaks such as m/z = 30, 18, and 66 that have also been found from the INMS for which assignment is still a matter of dispute. Recently, Vuitton et al. proposed that most of these new peaks are due to the reaction of the major species with several nitrile species whose abundance is low. They assigned the m/z = 30 peak as CH₂NH₂⁺ (Vuitton et al. 2006a, 2006b). CH₂NH₂⁺ is also an important species which has been found in the interstellar medium, and its role in astrobiology has been discussed in detail by the Kaiser group (Gu et al. 2009). Another important amine species is CH₃NH₂, which actually has been detected in the interstellar medium toward Sgr B2 and Ori A in its 5₁₅–5₀₆ (73.0 GHz), and 4₁₄–4₀₄ (86.1 GHz) transitions (Kaifu et al. 1975); however, the formation pathway has not been confirmed in laboratory studies.

These studies show the importance of HCNH⁺, CH₂NH₂⁺, and CH₃NH₂⁺ in Titan's ionosphere and in the interstellar medium, and suggest its potential role in many ion-neutral reactions and astrobiology. However, there are no laboratory studies to date which can give systematic information about the nature of the reaction dynamics and formation mechanisms for these species. It is surprising that HCNH⁺ and CH₂NH₂⁺ have been detected in Titan's environment, but until now no CH₂NH⁺ has been reported there. In order to understand the pathways involved in the formation of the HCNH⁺, we have studied the photodissociation of the CH₃NH₂⁺ cation using ab initio calculations and DC sliced ion imaging methods. It is worth mentioning here that although CH₃NH₂⁺ has not been detected in the Titan's ionosphere yet, it is a good source for the formation of HCNH⁺ and also provides the pathway to understand the ion–molecule reactions related to formation and reactions of HCNH⁺.

The detailed description of the experimental setup used in this study has been reported elsewhere (Leskiw et al. 2005), and only a brief sketch is given here. A pulsed molecular beam of methylamine (Linde, CP Grade, 99.5%) is made by supersonic jet expansion of a mixture of approximately 10% of methylamine in an argon carrier gas at a backing pressure of 2 bar from a pulsed nozzle valve (General Valve series 9) into the source chamber of the differentially pumped apparatus. After passing through a 1 mm diameter skimmer located 3.8 cm downstream of the pulsed nozzle orifice, the supersonic molecular beam enters into a velocity mapping ion optics assembly consisting of four electrodes. A laser beam tuned to a one-photon resonant excitation of methylamine perpendicularly intersects the molecular beam in the main chamber. The operating pressures were maintained at 1.0 × 10⁻⁵ torr in the source chamber and at 5.0 × 10⁻⁸ torr in the main chamber.

The ultraviolet light used here was generated by frequency doubling of the output of a tunable OPO laser (Spectra-Physics MOPO-HF) pumped by the third harmonic of a seeded neodymium-doped yttrium aluminum garnet (Nd:YAG) laser. The laser beam was vertically polarized, parallel to the detector surface, and then was focused with a 20 cm focal length lens into the interaction region. Typical output power for the pulsed laser beam entering into the chamber was 0.3–0.5 mJ pulse⁻¹. Wavelength calibration was achieved using a wavemeter (Coherent Wavemaster). In these experiments, a single color laser was used to ionize the CH₃NH₂ and to photodissociate the resulting CH₃NH₂⁺ ions. State selected CH₃NH₂⁺ were created using a one-color resonance-enhanced two-photon ionization technique at 41690 cm⁻¹ (150 kcal mol⁻¹) after which a third photon gives rise to photofragmentation. Both ionization and dissociation occurred from the same laser pulse. These experiments were performed at relatively low power, where it is assumed that photofragmentation takes place due to absorption of a single photon. However, the possibility of a multiphoton dissociation process in the cation is not ruled out. After ionization and dissociation, the photofragments were accelerated through the multi-lens velocity mapping assembly. Time of flight mass separation occurred along an 80 cm length flight tube after which the mass separated ions impacted a position sensitive dual MCP/P-47 phosphor screen of 120 mm diameter (Burle/Photonis). The resulting images were recorded using a CCD camera (Sony XT-ST50) in conjunction with the IMACQ megapixel acquisition software developed by our group (Li et al. 2005), which permitted real-time event counting and centroiding of the data. The central regions of the ion clouds were sliced by gating the detector on for 90 ns. The images were typically accumulated for 40,000 laser shots for each data set. Calibration of our images was achieved on the basis of the experimentally observed images of CO (ν = 0, J = 65) in the photodissociation of OCS at 230 nm. Total translational energy distributions for the different channels were obtained by integrating the images over angle at each radius, the converting to total translational energy using the IMAN program. The branching ratios between these channels were calculated by integrating the ion counts from their respective images for the same number of shots.

2.1. Theoretical Methods

To understand the pathways for photodissociation of CH₃NH₂⁺, the potential energies for photodissociation products of CH₃NH₂⁺ have been calculated using both the CBS-APNO and B3LYP/6–311G(d, p) levels of theory. Figure 1 shows the structures of the relevant minima as well as transition states at both the CBS-APNO and B3LYP/6–311G(d, p) level of theories. For the m/z = 30 (H loss) channel, both CH₃NH⁺ and CH₂NH₂⁺ products are accessible. For the m/z = 29 channel, there are two possible pathways: one in which there is direct H₂ molecule loss from the parent, and one in which there are two successive H atom losses. Similarly, the m/z = 28 product can arise via two general processes: one in which first one hydrogen atom is lost followed by H₂ molecule loss (H + H₂ loss), or alternatively, H₂ loss followed by H loss (H₂ + H loss). Successive loss of three H atoms is not energetically feasible given the available energy of ∼150 kcal mol⁻¹. In the case of the direct H + H₂ or H₂ + H loss from CH₃NH₂⁺ to form m/z = 28, two stable products HCNH⁺ and CNH₂⁺ are possible, while the other product, H₂CN⁺_, is a transition-state structure. Between the two stable m/z = 28 products, only HCNH⁺ is energetically feasible. For the hydrogen loss channel, both CH₃NH⁺ and CH₂NH₂⁺ products are barrierless processes. The lowest energy CH₃NH⁺ is a triplet species, and further dissociation from triplet CH₃NH⁺ to singlet HCNH⁺ is spin forbidden. CH₂NH₂⁺, another hydrogen loss product, can decompose further to HCNH⁺ via direct H₂ loss, which has a barrier of 118 kcal mol⁻¹. H₂ loss from CH₃NH₂⁺ can then proceed via two isomeric products CH₂NH⁺ and CHNH₂⁺; the former is 4 kcal mol⁻¹ higher in energy than the latter. The H₂ loss from the parent to form the CH₂NH⁺ product has a barrier of 86 kcal mol⁻¹; it can further dissociate to HCNH⁺ over a barrier of 34 kcal mol⁻¹. Efforts to elucidate the other pathway for direct H₂ loss from CH₃NH₂⁺ to produce CHNH₂⁺ resulted in the formation of CH₂NH₂⁺ + H.

Zoom In Zoom Out Reset image size

Methylamine's REMPI spectrum has mainly a v₉ amine wagging mode progression feature, and it has been extensively studied by Bernstein's group (Taylor & Bernstein 1995). Since methylamine undergoes a large geometry change upon excitation, no origin transition was observed in the REMPI spectrum (Baek et al. 2003a, 2003b). Photofragmentation of the CH₃NH₂⁺ at 41690 cm⁻¹ (5.17 eV) shows mass peaks at the 31, 30, 29, and 28 amu. These fragments are assigned, based on the theoretical calculations, as the parent ion CH₃NH₂⁺, CH₂NH₂⁺/CH₃NH⁺ from the H loss channel, CH₂NH⁺ from the H₂ loss channel, and HCNH⁺ from the H + H₂ loss channel, respectively.

To get more insight into the photodissociation dynamics of CH₃NH₂⁺, the DC sliced ion imaging method was used and images and translational energy release were recorded for the photodissociation products of CH₃NH₂⁺ at 41690 cm⁻¹. Figure 2 shows the sliced ion images for the H loss (mass 30), H₂ loss (mass 29) and H+H₂ loss (mass 28) produced from the photofragmenation of CH₃NH₂⁺ at 41690 cm⁻¹, while Figure 3 shows the total translational energy release for these channels derived from the respective images. The images are similar for all the channels except for the H₂ loss channel where the image has a "hole" inside it. The angular distributions of the fragments are isotropic, which implies of the long lifetime (ps–ns) prior to dissociation. The translational energy release for the hydrogen loss channel has a peak at 0.4 eV and extends to 2 eV, while for the H₂ loss channel, the peak is at 1 eV and it extends up to 3 eV. Figures 3(c) and (d) show two translational energy release plots for H+H₂ loss (mass 28), in which the first translational energy distribution is derived for cofragment value of 3 to account for the H+H₂ loss, while the second plot is for the translational energy of the HCNH⁺ species alone. The branching ratio for these photofragmention channels has been estimated from integration of their respective images and found to be 4.2:1:2.5 (m/z = 30:29:28). The branching ratio for these channels was measured on numerous other transitions from 41703 cm⁻¹ to 43768 cm⁻¹ (corresponding to a wavelength range from 240 to 228 nm), but we found no appreciable change in the branching. In addition, despite a thorough investigation on many transitions in this region, we found no evidence of mode selectivity on the photodissociation dynamics of these ions.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

For the hydrogen loss channel from CH₃NH₂⁺, there are two pathways: (1) CH₂NH₂⁺ + H and (2) CH₃NH⁺ + H. The translational energy release for hydrogen loss has maximum value around 0.4 eV and extends to 1.5 eV, which is consistent with the calculations for the CH₃NH⁺ + H channel, having maximum available energy of 1.5 eV. The CH₂NH₂⁺ product has the available energy of 5.1 eV; however, the translational energy release is found to be between 0.1—and 1.5 eV, which likely excludes this as a hydrogen loss product. Rather, the agreement between experimental and theoretical results suggests that CH₃NH⁺ is the product detected as the hydrogen loss channel. The other possible product, CH₂NH₂⁺, can further dissociate to HCNH⁺ by direct H₂ loss. This mechanism has a barrier of 4.03 eV, which is within the available energy of the primary CH₂NH₂⁺ product. This suggests that much of the CH₂NH₂⁺ dissociates further to form HCNH⁺ + H₂.

The direct H₂ loss from CH₃NH₂⁺ could have two pathways, one via CH₂NH⁺ and the other may be from CHNH₂⁺. However, theoretical calculation shows that only CH₂NH⁺ + H₂ is possible. The H₂ loss image, shown in Figure 2(b) has a hole inside it, and the translational energy release shows that it has maximum around 0.5–1.9 eV while it extends out to 3 eV. This hole in the H₂ loss image, implying the absence of slow fragments, can either correspond to an energy barrier for H₂ loss from CH₃NH₂⁺ or it can be because of the secondary decomposition of the primary products. The only possible channel for H₂ loss from CH₃NH₂⁺ is via CH₂NH⁺ product formation. This channel has a barrier of 3.7 eV from CH₃NH₂⁺. This transition state has maximum 2.8 eV available energy from this barrier, which is in nice agreement with the translation energy release plot for H₂ loss shown in Figure 3(b). This strongly suggests that the hole in H₂ loss image is due to the barrier for the H₂ loss from the CH₃NH₂⁺ to CH₂NH⁺. CH₂NH⁺ product can dissociate further to HCNH⁺ + H via a barrier of 1.5 eV. This energy barrier from CH₂NH⁺ to HCNH⁺ is less than the available energy of 2.8 eV, though the available energy depends upon the nature of the exit channel. Even if the decay does not allow the transfer of all excess energy to CH₂NH⁺, the 1.7 eV available energy in CH₂NH⁺ is enough to overcome the barrier for the formation of the HCNH⁺ product. Figure 3(c) shows the translational energy for H+H₂ loss, which has a maximum energy around 0.7 eV and extends to 3.0 eV, in good agreement with the maximum available energy for the HCNH⁺ + H + H₂ product of 3.1 eV. This channel is easily assigned to the HCNH⁺ product. Figure 3(d) shows the translational energy of HCNH⁺ alone, which is useful in envisioning the limiting momentum partitioning in this three-body decay process.

The above experimental and theoretical data clearly show that HCNH⁺ can be formed by two channels from methylamine cation; one is a direct loss of H₂ from CH₂NH₂⁺ and other is H loss from CH₂NH⁺. The H loss channel has two pathways, one is to form CH₃NH⁺ in the triplet state and the other product is CH₂NH₂⁺, which decomposes to HCNH⁺. Most of the H₂ loss product, CH₂NH⁺, finally goes to HCNH⁺, which is the reason that branching of the H loss is higher than the H₂ loss channels.

4.1. Implications for Titan's Atmosphere

In this paper, DC sliced ion imaging and theoretical studies of CH₃NH₂⁺ photodissociation have been performed to understand the pathways and branching ratios of different photodissociation product channels. CH₃NH₂⁺ has been studied as a source to understand the formation of HCNH⁺ which is the most abundant ion in Titan's upper atmosphere, and to characterize the potential energy surface for reactions leading to HCNH⁺. Our experimental and theoretical observations show that hydrogen loss from CH₃NH₂⁺ has two channels, one leading to triplet CH₃NH⁺, while the other product CH₂NH₂⁺ decomposes further, to form HCNH⁺. H₂ loss has only one channel, CH₂NH⁺ + H₂, which further loses a hydrogen atom to form HCNH⁺. The facile decomposition of CH₂NH⁺ to HCNH⁺ may be one of the reasons why no CH₂NH⁺ has been observed in Titan's ionosphere. The branching ratio of the m/z = 30:29:28 is found to be 4.2:1:2.5, which is justified by the fact that the H loss has one stable triplet product channel while most of the H₂ loss product converts to HCNH⁺.

This work was supported by the National Science Foundation under award numbers CHE-0627854 (AGS) and CHE-0910858 (HBS). We thank Stephanie Everhart for a critical reading of the manuscript.