ULTRA-LOW-TEMPERATURE REACTIONS OF CARBON ATOMS WITH HYDROGEN MOLECULES

Helium droplets with an estimated mean size of 10⁵ helium atoms were used. Helium droplets were first doped with dihydrogen and later in the second pick-up oven with carbon atoms. After arriving to the next differentially pumped vacuum chamber, the doped helium droplets are continuously ionized by electron impact using 70 eV electrons. The ions are pulse extracted to a commercial time-of-flight mass spectrometer equipped with a reflectron (Tofwerk AG, model HTOF). The implemented two-stage reflectron scheme allows us to achieve a high-mass resolution of m/Δm ∼ 3500 and the accuracy of the mass calibration better than 0.02 amu. Ions are detected by a microchannel plate detector that operates in an ion-counting mode. The residual gas pressures were 2 × 10⁻⁵ and 1 × 10⁻⁵ Pa in the pick-up and detector chambers, respectively. The presented mass spectra are the average of 1750 one-second measurements.

Helium droplets with an estimated mean size of 1.2 × 10⁴ helium atoms were used. These droplets were first doped with carbon atoms and later with dihydrogen. The residual gas pressures were 3 × 10⁻⁵ and 1.5 × 10⁻⁷ Pa in the pick-up and detector chambers, respectively. The lifetime of the helium droplet in the experimental setup is only a few milliseconds. Therefore, this pressure is sufficient to keep most of the droplets free from impurities. The total pumping speed of all pumps in the detector chambers was 600 l s⁻¹. Helium droplets arriving in the detector chamber are completely evaporated after collisions with the walls of the vacuum chamber. This causes the rise of the pressure in this chamber. The pressure in the detector chamber was recorded with a precision ion gauge (Varian UHV-24). To obtain the helium pressure in that chamber, the following procedure was performed. After each depletion measurement, the helium droplet beam was blocked by the shutter in the source chamber. Thirty seconds after closing the shutter, the residual gas pressure was measured and subtracted from the obtained values. Finally, the gas correction factor for helium (0.18) was applied. The depletion of the helium droplet beam flux due to the carbon incorporations were calculated as a ratio of the helium pressures in the detector chamber before and 12 s after doping the helium droplets with carbon atoms. This short time interval provides the high accuracy of the obtained depletion values despite the long-term instability of the source.

The energy released in the reaction inside helium droplets leads to the evaporation of the droplets and reduction of their sizes. As a result, a smaller amount of helium atoms arrive at the detector chamber. The evaporating helium droplets are the main source of helium in the detector chamber. The gaseous helium is continuously evacuated from this chamber with a constant speed. Therefore, the pressure in the detector chamber shows the flux of the helium droplet beam and the depletion of this beam flux shows the amount of the energy released in the reaction. Figure 1 displays the change of the helium pressure in the detector chamber as a function of time. The depletion peaks are due to the doping of helium droplets with carbon atoms in the first pick-up chamber. In the second pick-up chamber, the helium droplets can be doped with another reactant (Ar or H₂). For better accuracy, the experiment shown in Figure 1 was repeated several times using different doping efficiencies of carbon atoms and average values for the depletions were calculated. We obtained 21.2%, 20.9%, and 27.5% for the depletion of the beam of helium droplets, when the second pick-up chamber is empty, filled with Ar, and H₂, correspondingly. As can be seen, the depletion, which is due to the carbon doping was almost equal when the second pick-up chamber was empty or filled with Ar. Moreover, the depletion during the Ar doping was even a bit lower. The binding energy between Ar clusters and carbon atoms, is rather low and cannot be detected in this experiment at this level of sensitivity. During the Ar doping, the evaporation of helium droplets, which is due to the released cluster complex formation energy, is overcompensated by the fact that the carbon doped helium droplets have a lower chance of colliding with Ar atoms in the second pick-up cell. The same effect has already been observed when this calorimetric technique was employed to monitor the formation of van der Waals complexes (Krasnokutski & Huisken 2010a).

Zoom In Zoom Out Reset image size

The described effect is relatively weak and cannot prevent the detection of a large amount of energy released in the chemical reactions. When the second pick-up oven is filled with dihydrogen, a considerably stronger depletion can be detected. This demonstrates that, in the low-temperature reaction of carbon atoms with dihydrogen, the amount of energy released is much higher than that released in the formation of van der Waals complexes. The heat of evaporation of a single He atom is about 5 cm⁻¹ (Toennies & Vilesov 2004). Assuming an extra 7% depletion when Ar gas is replaced with dihydrogen, we can estimate the lower limit of the energy released in the droplets to be about 4200 cm⁻¹. The real reaction energy is expected to be much higher because not all helium droplets contain both reactants. Additionally, an ejection of the hot reaction products from the helium droplets, as well as a dissipation of the reaction energy by photon emission, are expected (Krasnokutski & Huisken 2010b).

This estimate is only consistent with the reaction leading to the formation of the HCH molecule becuase the formation of the CH₂ molecule would lead to the release of only a small amount of energy ∼600 cm⁻¹ (Harding et al. 1993). The high rate of this reaction found at ultra-low temperature, assumes no energy barrier in the reaction pathway. Therefore, our experiment is in line with the results of quantum chemical computations that predict the barrierless reaction C + H₂ → HCH (Harding et al. 1993).

In the case of doping of helium droplets by more than a single dihydrogen molecule, the formed HCH molecule can also further react with a H₂ molecule leading to the formation of methane. However, the quantum chemical computations predict the presence of a barrier for this reaction (Ge et al. 2010; Lu et al. 2010). Therefore, we expect that this reaction should not proceed inside the helium droplets and that the reactions of carbon atoms are terminated after the formation of HCH molecules.

In the next step, we applied mass spectrometry for the characterization of products of chemical reactions. In this experiment, larger helium droplets (up to 10⁵ He atoms) were used. The droplets were first doped with multiple dihydrogen molecules and, in the second pick-up chamber, one or a few carbon atoms were added. We used two different conditions for the doping with dihydrogen. The helium droplets picked up about 10 and 100 dihydrogen molecules on average at low and high doping conditions, correspondingly. Although, even at the high doping conditions, the C/H ratio is much higher than that present in the ISM, it is low enough to saturate the chemistry with hydrogen (i.e., more than five hydrogen atoms per carbon atom). The doped helium droplets were ionized by electron impact in the head of the TOF mass spectrometer.

The collision of a doped helium droplet with an electron, having energy larger than 25 eV, results in the formation of He⁺. This positive hole migrates within the droplet until it becomes localized either at a dopant or at a helium dimer to form ${{\mathrm{He}}_{n}}^{+}$ (Scheidemann et al. 1993). Therefore, the cations of dopants and helium clusters as well as their complexes are produced.

Figure 2 shows the mass spectra of helium droplets doped with dihydrogen alone and together with carbon atoms. Both mass spectra were recorded one after another using the same experimental conditions. As can be seen, the probability of the charge transfer to the dopant species is quite low. Therefore, the main intensity of the ion signal is detected on the masses of helium clusters. The efficient complex formation between hydrogen and helium leads to the formation of He_nH_m cations. Hydrogen clusters consisting of unbound protons in their nuclei are always a bit heavier than any other elements with the same number of nucleons. Therefore, pure hydrogen clusters or species with the high hydrogen content can be separated from other isobaric ions using high-resolution mass spectrometry. As can be seen from Figure 2, switching on the atomic carbon source leads to the reduction of the number of ions of pure hydrogen clusters. At the same time, there are new peaks appearing in the mass spectra. They were assigned to the different hydrocarbon molecules. Practically all hydrocarbon cations ${{\rm{C}}}_{m}{{{\rm{H}}}_{n}}^{+}$ with m = 1–4 and n = 1–15 are produced. The ion intensities on the masses of hydrocarbon products are summarized in Figure 3 for low and high H₂ pressures. The data point of ${{\rm{C}}}_{2}{{{\rm{H}}}_{2}}^{+}$ is missing in both panels because its mass (28.02126 amu) is too close to the mass of ${{{\rm{N}}}_{2}}^{+}$ (28.00615 amu), which is the most abundant species in the residual gas. The other peaks, caused by residual gas (H₂O, OH, and CO₂), do not interfere with our measurements.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

As can be seen in the figure, there is only a small cation signal for n = 1. This is in line with calorimetry measurements showing that the HCH molecule is formed before the ionization. Therefore, we expect that larger hydrocarbon cations are presumably produced due to ion-molecular reactions after the electron impact ionization.

The chemical and physical processes occurring after the electron impact ionization of doped helium droplets have some similarities to the processes occurring in the ISM. In both cases, a large amount of energy is inserted into a cold molecule. In the helium droplet experiment, the energy arises due to the charge transfer from the initially formed He⁺ to a dopant. This energy is the difference between ionization potentials (IPs) of helium and a dopant. The difference in IPs of He and C atoms is about 13.32 eV. The HCH molecule formed in helium droplets before the ionization is expected to have a lower IP than that of C atoms. In H i regions of the ISM, the energy is brought to a molecule by UV photons hν < 13.6 eV. The input of a large amount of energy leads to the breaking of weak bonds and, therefore, to the formation of the most stable species. Therefore, the appearance of the "magic" peaks in the mass spectra is particularly interesting.

Ionization of helium droplets doped with pure dihydrogen results in a distribution of ${{{\rm{H}}}_{n}}^{+}$ clusters with preferentially odd numbers of H atoms n (Jaksch et al. 2008) and some well-known intensity anomalies that can be explained by an ionic ${{{\rm{H}}}_{3}}^{+}$ core solvated by hydrogen molecules. With the addition of carbon atoms, new ions are observed. The intensity of ${{\rm{C}}}_{m}{{{\rm{H}}}_{n}}^{+}$ ions (m = 1, 2, 3) strongly depends on the number of the hydrogen atoms (see Figure 3). This can be easily understood, considering the covalent bonding in the formed hydrocarbon molecules.

At low dihydrogen doping, the most intense "magic" peaks are caused by the C₂H₃ cation. After increasing the number of picked up dihydrogens, the dominant formation of the CH₅ cation was observed. The number of produced ${{\mathrm{CH}}_{5}}^{+}$ cations was more than two times higher than that of CH₃ or C₂H₃ cations. In spite of the low ${{\mathrm{CH}}_{5}}^{+}\to {{\mathrm{CH}}_{3}}^{+}+{{\rm{H}}}_{2}$ dissociation energy, which was measured to be only 40 kcal mol⁻¹ (Hiraoka & Kebarle 1976) in our experiments, a high stability of ${{\mathrm{CH}}_{5}}^{+}$ was observed. Therefore, it implies that there is a fast pathway of energy dissipation by the CH₅ cation. This is in line with a high rate of the radiative association found for the ${{\mathrm{CH}}_{3}}^{+}+{{\rm{H}}}_{2}\to {{\mathrm{CH}}_{5}}^{+}$ + hv reaction, which was at least one order of magnitude larger than predicted taking into account only the relaxation via vibrational transitions (Barlow et al. 1984). A previous study on the ionization of helium droplets doped with methane clusters also shows the high probability of the ${{\mathrm{CH}}_{5}}^{+}$ formation. This demonstrates that the ${{\mathrm{CH}}_{5}}^{+}$ formation is not dependent on the types of the precursor molecules. Therefore, in the ISM, the ${{\mathrm{CH}}_{5}}^{+}$ can be produced not only by sequences of bottom-up reactions, but also by the top-down process when ice grains containing both carbon and hydrogen elements collide with the high energy particles. The obtained results suggest that the relative abundances of the ${{\mathrm{CH}}_{5}}^{+}$ among other hydrocarbon cations present in the ISM could be much higher than that considered previously (Sternberg & Dalgarno 1995). Due to the top-down formation, this could be particularly pronounced in the colder areas at extinctions A_V > 6.

The ${{\mathrm{CH}}_{5}}^{+}$ was first detected in 1952, and since then it has been a subject of intensive studies (Tal'roze & Lyubimova 1952; Sefcik et al. 1974; Semaniak et al. 1998; White et al. 1999; Thompson et al. 2005; Ivanov et al. 2010; Asvany et al. 2015). In spite of these thorough studies, only the IR spectrum of this cation is currently available (Asvany et al. 2005, 2015). At the same time, the high rate of ${{\mathrm{CH}}_{3}}^{+}$ + H₂ radiative association suggests the presence of electronic states of ${{\mathrm{CH}}_{5}}^{+}$ lying below or just a little above the ${{\mathrm{CH}}_{3}}^{+}$ + H₂ dissociation limit and, consequently, the presence of absorption bands in the NIR range. The attempt to locate such states by quantum chemical computation was unsuccessful (Talbi & Saxon 1992). However, considering an extreme complexity of this cation for the computations (Ivanov et al. 2010), further studies in this direction seem to be necessary. In particular, the spectral characterization of the cation in the optical and NIR ranges should help us to locate the possible excited electronic states and to understand its influence on the spectral properties of the ISM.