Spectroscopy of prospective interstellar ions and radicals isolated in para-hydrogen matrices

Masashi Tsuge *a, Chih-Yu Tseng a and Yuan-Pern Lee *ab
aDepartment of Applied Chemistry and Institute of Molecular Science, National Chiao Tung University, 1001 Ta-Hsueh Road, Hsinchu 30010, Taiwan. E-mail: yplee@nctu.edu.tw; Tel: +886-3-5131459
bInstitute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan

Received 20th August 2017 , Accepted 19th September 2017

First published on 19th September 2017

para-Hydrogen (p-H2) serves as a new host in matrix-isolation experiments for an investigation of species of astrochemical interest. Protonated and mono-hydrogenated species are produced upon electron bombardment during deposition of p-H2 containing a precursor in a small proportion. The applications of this novel technique to generate protonated polycyclic aromatic hydrocarbons (H+PAH), protonated polycyclic nitrogen heterocycles (H+PANH), and their neutral counterparts, which are important in the identification of interstellar unidentified infrared emission bands, demonstrate its superiority over other methods. The clean production with little fragmentation, ease of distinction between protonated and neutral species, narrow lines and reliable relative infrared intensities of the lines, and broad coverage of the spectral range associated with this method enable us to assign the isomers unambiguously. The application of this method to the protonation of small molecules is more complicated partly because of the feasible fragmentation and reactions, and partly because of the possible proton sharing between the species of interest and H2, but, with isotopic experiments and secondary photolysis, definitive assignments are practicable. Furthermore, the true relative infrared intensities are critical to a comparison of experimental results with data from theoretical calculations. The spectra of a proton-shared species in solid p-H2 might provide insight into a search for spectra of proton-bound species in interstellar media. Investigations of hydrogenated species involving the photolysis of Cl2 or precursors of OH complement those using electron bombardment and provide an improved ratio of signal to noise. With careful grouping of observed lines after secondary photolysis and a comparison with theoretical predictions, various isomers of these species have been determined. This photolytic technique has been applied in an investigation of hydrogenated PAH and PANH, and the hydrogenation reactions of small molecules, which are important in interstellar ice and the evolution of life. The electronic transitions of molecules in solid p-H2 have been little investigated. The matrix shift of the origins of transitions and the spectral width seem to be much smaller than those of noble-gas matrices; these features might facilitate a direct comparison of matrix spectra with diffuse interstellar bands, but further data are required to assess this possibility. The advantages and disadvantages of applying these techniques of p-H2 matrix isolation to astrochemical research and their future perspectives are discussed.

image file: c7cp05680j-p1.tif

Masashi Tsuge

Masashi Tsuge received his PhD in chemistry from Tokyo Institute of Technology in 2007 for his studies on spectroscopy of transient species. He has worked at Niigata University of Pharmacy and Applied Life Sciences, University of Helsinki, and National Chiao Tung University as a post-doctoral researcher, and is now an Appointed Assistant Professor at the Institute of Low Temperature Science, Hokkaido University. His research topics include spectroscopy of radicals, ions, and molecular complexes, which are important in atmospheric chemistry, astrochemistry and noble-gas chemistry. He is currently studying surface-reaction dynamics relevant to the reactions on interstellar grain surfaces.

image file: c7cp05680j-p2.tif

Yuan-Pern Lee

Yuan-Pern Lee received his PhD in chemistry from U. C. Berkeley in 1979 and has been a National Chair Professor in the Department of Applied Chemistry, Chiao Tung University, Taiwan, since 2004. His main research topics concern spectroscopy, kinetics, and dynamics of free radicals or unstable species that are important in atmospheric, combustion, or astrochemistry using various experimental methods including step-scan time-resolved FTIR (emission or absorption), matrix isolation using p-H2, cavity ringdown, IR-VUV photoionization/time-of-flight mass detection, and transient absorption using ultrafast lasers.

1. Introduction

Astrochemistry is important because it involves an investigation of the abundance of chemical species in the universe and the related reactions, including their interactions with radiation. The formation and evolution of clouds of molecular gases are of special interest because these clouds form solar systems. Laboratory spectroscopy is indispensable in astronomy and astrochemistry: on comparing astronomical observations with laboratory measurements, astrochemists can infer the chemical composition and temperatures of stars and interstellar clouds. Laboratory experiments mimicking interstellar conditions facilitate an understanding of the evolution of chemical species and the origin of life.

Although powerful, spectral measurements with radiation in various regions (radio-frequency, infrared, visible, or ultraviolet) have their own limitations; each method is able to detect species of particular types, depending on the structural properties of the molecules. Radio astronomy, mainly through microwave spectroscopy, has played a major role in the identification of interstellar chemical species including free radicals and ions. To date, more than 150 species have been identified in the interstellar medium (ISM)1 based on spectra obtained in laboratories, but this method is insensitive to molecules with small or no permanent dipole moment; it is more suitable for small gaseous species.

Infrared (IR) spectroscopy complements microwave spectroscopy; one can thereby detect species with no dipole moment such as methane,2 and complicated molecules3 such as fullerenes (C60 and C70).4 One can also infer the composition of solid materials in the ISM,5 including silicates, carbon-rich solids, and ices, because, unlike visible light that is scattered or absorbed by solid particles, IR radiation can penetrate the microscopic particles and reveal characteristic absorptions of the composition of the grains.

Infrared astronomy has also revealed that the interstellar medium contains complicated gaseous polycyclic aromatic hydrocarbons (PAHs). These molecules, composed primarily of carbon in fused rings, are believed to be the most common class of carbon compounds in the galaxy, meteorites, comets, and cosmic dust. These PAHs are thought to be produced in hot circumstellar environments around dying, carbon-rich, red-giant stars. Such IR observations also determined that, in dense clouds in which the particles attenuate the destructive UV radiation, thin ice layers coat the microscopic particles and permit some reactions to occur at low temperature. Because H2 is the most abundant molecule in the universe, hydrogenation becomes the most important reaction for most species in ice. Hydrogenation reactions of species containing C, N, and O might provide clues about the origin of life. Laboratory experiments on the photochemistry of simple interstellar ice analogues have produced amino acids.6

Protonated molecular hydrogen, H3+, is abundantly produced in the ISM, but its steady-state concentration is small because of its great reactivity.7 H3+ transfers a proton to other species to initiate ion–molecule chain reactions in the ISM. Most interstellar molecules can accept a proton from H3+ as their proton affinities are generally greater than that of H2 (422 kJ mol−1).8 Protonated species are hence thought to be abundant in the ISM. Several protonated polyatomic species have been identified in the ISM, including HCO+,9 HN2+,10,11 H3O+,12 HCNH+,13 and HOCO+.14–16 As PAHs are abundant in the ISM, protonated polycyclic aromatic hydrocarbons (H+PAH) have been postulated also to be present in the ISM.17–19 Large PAH and H+PAH are proposed to be the carriers of the unidentified infrared (UIR) emission bands, but laboratory measurements of IR spectra of large PAH and H+PAH are difficult.

Relatively broad absorption lines observed towards reddened stars are called diffuse interstellar bands (DIBs).20 Similar to those of the UIR bands, spectral carriers of DIBs are elusive. The finding of two DIBs observed at 963.2 nm and 957.7 nm21 was based on two electronic absorption bands of C60+ near 964.5 nm and 958.3 nm observed in a Ne matrix.22 However, because of the matrix shifts and broadening, an unequivocal confirmation of the carrier of these two bands could not be based on the matrix spectrum. Only recently were the assignments confirmed by the recording of the excitation spectra of cold gaseous C60+ by Campbell et al.23

The matrix-isolation technique coupled with IR absorption spectroscopy has been widely applied to investigate unstable species such as free radicals and ions.24–26 In this technique, a species of interest (termed as “guest”) is embedded in a small proportion within an inert solid (typically a noble gas or N2) at low temperature so that intermolecular reactions are diminished. The absence of rotational lines and hot bands yield IR spectra of matrix-isolated species much simpler than spectra of the gaseous phase. Furthermore, with Ar or Ne as a matrix host, matrix shifts for the vibrational wavenumbers of matrix-isolated species are typically less than 1 percent; the information thus obtained is hence applicable to gaseous systems.27 In contrast, for electronic transitions, including absorption and laser-induced fluorescence, matrix shifts >300 cm−1 of the origins and extensive phonon wings are typically observed.28

Recently, quantum solid para-hydrogen (p-H2) has emerged as a novel host for matrix experiments.29–35 One characteristic of solid p-H2 is its softness associated with its large amplitude zero-point lattice vibrations, which amount to a substantial fraction of the distance between adjacent H2 molecules. Associated with this softness is a diminished cage effect that allows the preparation of atoms or free radicals from photolysis in situ. For example, CH3O radicals were successfully produced from CH3ONO upon UV photolysis in solid p-H2in situ,36 whereas only H2CO and HNO were produced in noble-gas matrices because of the cage effect. A further advantage of the diminished cage effect is that bimolecular reactions can occur in solid p-H2; the application of such reactions consequently expands the variety of free radicals that can be produced via irradiation. For instance, photolysis of a matrix containing CH3I/SO2/p-H2 yielded CH3, which reacted readily with nearby SO2 to form CH3SO2, an important intermediate in the formation of acid rain.37

A variation of such bimolecular reaction is hydrogenation, which is a critical process in interstellar ice, as discussed above. Hydrogen atoms are produced in solid p-H2 on photolysis of Cl2 with light near 365 nm followed by irradiation of the matrix with infrared light of wavenumber >4000 cm−1; the Cl atom produced on UV photolysis reacts with H2 (v = 1), produced on IR irradiation, to form HCl + H. Anderson and co-workers38 extensively studied this process; Bahou et al.35 introduced its application to produce hydrogenated species. Alternatively, photolytically produced hydroxyl radical, OH, can react with H2 to produce H2O and H atoms;39 the application of these photolytic methods is discussed in Section 4.

Another novel method associated with p-H2 is the production of protonated and hydrogenated species on electron bombardment of a mixture of p-H2 and a guest molecule during deposition at 3.2 K.35,40 This method has been successfully applied to generate protonated and hydrogenated PAH to obtain their IR spectra in solid p-H2.40–44 The same method was applied to produce also smaller protonated species of astrochemical interest; for instance, the production and IR identification of protonated carbonyl sulfide HOCS+ and HSCO+ have been demonstrated.45 Recent results on the IR spectra of protonated species are presented in Section 3.

The softness of solid p-H2 might induce negligible matrix shifts for electronic transitions, but reports of electronic spectra of species isolated in solid p-H2 are scarce. In the laser-induced emission spectra of NO in solid p-H2 recently recorded, the matrix shift was much smaller than that in solid Ne;46 these results and their implications are introduced in Section 5.

2. Methods

The p-H2 matrix-isolation system has been described elsewhere.35,40 In brief, a gold-coated copper plate, cooled to 3.2 K with a closed-cycle helium refrigerator (Sumitomo Heavy Industries, RDK-415), serves as a substrate for the matrix sample and a mirror to reflect the incident IR beam to the detector. Infrared absorption spectra are measured with a Fourier-transform infrared (FTIR) spectrometer equipped with a KBr beam-splitter and a HgCdTe detector cooled to 77 K. For experiments to record electronic transitions, the laser-induced fluorescence was collected with an optical fiber, followed by dispersion with a monochromator (0.6 m, Andor, Shamrock 500i) and detection with an intensified CCD detector (Andor, DH320T-18U-73, 1024 × 255 array of pixels) with resolution ∼0.07 nm; the slit width was typically set at 100 μm. Fluorescence spectra were recorded typically 0–100 ns or 0–5 μs after irradiation. For measurements of temporal profiles, the emission was typically recorded at intervals of 0–12 ns or 0–2 μs, depending on the decay rate. To improve the ratio of signal to noise, we accumulated a fluorescence spectrum over 500 laser shots.

Protonated species are produced on electron bombardment of a mixture of p-H2 and the precursor of interest during deposition at 3.2 K. An electron gun (Kimball Physics, Model EFG-7) is employed to provide an electron beam. The voltage and current of the beam are adjustable to maximize the yield of the protonated species. The formation of ions can be confirmed through the appearance of the Stark-shifted Q1(0) transition of H2 in the region 3950–3980 cm−1,47 but the exact detailed mechanism and chemistry of charged species in solid p-H2 remain not fully understood and deserve further study.

For proper grouping of observed lines, we typically employ secondary photolysis. Typical sources for photolysis include an IR source (SiC rod), light-emitting diodes (LED) at varied wavelengths (365–520 nm), low-pressure and medium-pressure mercury lamps, a Nd:YAG laser (355 and 532 nm) and excimer lasers (308, 248, and 193 nm).

The conversion of normal-H2 to p-H2 is well established;48–50 our converter is described in ref. 33. In brief, conversion takes place in a copper coil filled with a catalyst (iron(III) oxide) cooled to 12–15 K with a closed-cycle helium refrigerator. Normal-H2 is passed through a liquid nitrogen trap before entering the convertor; the resultant p-H2 contains ortho-H2 at typically less than 100 ppm according to the Boltzmann distribution.

Spectral assignments of unknown species are achieved according to the expected chemistry in solid p-H2, the behavior on secondary photolysis, a comparison of observed and quantum-chemically predicted vibrational wavenumbers and IR intensities, and, if possible, isotope shifts. We generally use density-functional theory with hybrid functionals B3LYP or B3PW91 to predict IR spectra.51–53 When feasible, anharmonic vibrational analysis is also performed with second-order vibrational perturbation theory (VPT2).54

3. Protonated species

3.1. Protonated PAH

Unidentified infrared (UIR) emission bands and the PAH hypothesis. The IR emission from many astronomical objects is dominated by the unidentified infrared (UIR) emission bands, which show major features near 3.3, 6.2, 7.7, 8.6, and 11.2 μm;55,56 these wavelengths correspond approximately to 3000, 1600, 1300, 1150, and 900 cm−1, respectively. Because these bands are characteristic of the CH-stretching, CC-stretching, and CH-bending modes of aromatic compounds, PAH and their derivatives have been proposed to be responsible for these UIR bands upon UV excitation.57,58 The spectral pattern of UIR features varies slightly among galactic objects, indicating varied distributions of emitters. The UIR features near 6.2, 7.7, 8.6, and 11.2 μm have similar intensities, but the intensity of the feature of neutral PAH near 11.2 μm is significantly greater than that of other features. Because of non-negligible discrepancies between laboratory spectra and UIR emission bands, no PAH has been positively identified as a carrier of UIR emission bands. Theoretical investigations indicate that the intensities of features near 6.2, 7.7, and 8.6 μm are enhanced in ions of PAH, including PAH cations (designated as PAH+) and protonated PAH (designated as H+PAH); these species are consequently proposed to be prospective carriers of the UIR emission.17,19 Theoretical investigations indicated also that the electronic transitions of some H+PAH in the visible region might be responsible for particular DIB.59,60
Laboratory measurements of IR spectra of H+PAH. Measuring the IR spectra of large protonated molecules in a laboratory is challenging because preparing sufficient molecules for detection is difficult. Of two major techniques typically employed to record the IR spectra of H+PAH, one employs infrared multi-photon dissociation (IRMPD), with which protonated species, generally accumulated and trapped in an ion trap, are photo-dissociated with intense light from a free-electron laser or other source; the depletion of protonated species as a function of IR wavenumber is recorded.61 The IRMPD spectra of H+PAH such as protonated pyrene (C16H11+) and coronene (C24H13+), reported by Dopfer, Oomens, and co-workers,19,62 show some agreement with the UIR bands. This method entails mass selection, which is advantageous for detection in complicated systems, but it generally yields broad spectral bands that are significantly red-shifted from IR absorption bands because of the vibrational anharmonicity associated with the multi-photon excitation.

A second method employs infrared photodissociation (IRPD) of tagged protonated species;63,64 these species, produced typically in a discharged pulsed jet, are tagged with a noble-gas atom. The diminution of the tagged cation occurs when the wavenumber of the photolysis IR laser is in resonance with the vibrational transition of the tagged cation. Because of the small perturbation from the noble-gas atom, the action spectrum thus recorded is taken to serve as the IR absorption spectrum of the protonated species. Duncan and co-workers performed pioneering experiments on H+PAH and recorded the IRPD spectra of protonated benzene (C6H7+)65 and naphthalene (C10H9+)66 tagged with an Ar atom. The features observed for these species at 6.2 μm provided one of the first examples, in laboratory spectra, of an intense vibrational band at this wavelength. It arises from an “allyl-type” carbon ring distortion, which is a direct consequence of the perturbation of an aromatic ring from protonation in a σ-configuration. This method is more sensitive than IR absorption because it monitors the variation of an ion current; compared with the IRMPD method, it yields bands with narrow lines and without significant shifts. One difficulty of this method is to tag a large H+PAH with a noble-gas atom because the internal energy of the large H+PAH might exceed the binding energy between Ar and H+PAH. Because spectra obtained with these two techniques are action spectra, which record the variation of signals as the wavenumber of the IR irradiation is tuned and which depend on both the associated absorption coefficient and the dissociation yield, the relative intensities observed in these action spectra might not reflect the intrinsic IR absorption intensities.

Our novel method to produce protonated species and their neutral counterparts has demonstrated several advantages over the previous techniques. In this method, we apply an electron gun to bombard a p-H2 matrix containing PAH during deposition. Ionization of H2 on electron impact produces H2+, which subsequently transfers a proton to H2 to yield H + H3+.47 Because the proton affinity of PAH is much greater than that of H2, H3+ can readily transfer a proton to PAH to form H+PAH. The reaction of H atoms with PAH and neutralization of H+PAH with electrons yields mono-hydrogenated PAH (designated H1PAH). To distinguish H+PAH, H1PAH, and PAH, we typically maintain the electron-bombarded matrix in darkness for a prolonged period or irradiate this matrix with UV light to release trapped electrons; the intensities of lines of H+PAH and PAH consequently diminish, whereas those of H1PAH increase. Secondary photolysis at varied wavelengths is generally employed to distinguish further the groups of lines associated with disparate carriers, likely various isomers of H+PAH or H1PAH. A comparison of observed lines in each group with quantum-chemically predicted IR spectra of possible carriers provides the assignments of these species.

Using this technique, we have recorded the IR absorption spectra of H+PAH from protonated forms of benzene (C6H7+),40 naphthalene (1- and 2-C10H9+),41 pyrene (1-C16H11+),42 coronene (1-C24H13+),43 and ovalene (7-C32H15+).44 The advantages of our method are narrow lines, reliable relative IR intensities, wide spectral coverage, clean production, and high sensitivity. A comparison of the spectra obtained among various methods, when available, is illustrated in each paper.

The small widths and the reliable relative IR intensities of the spectra observed using this method enable us to assign the isomers unambiguously, as we list for each H+PAH in the preceding paragraph. For larger H+PAH, this feature is critical because many protonation sites are possible; we must distinguish them with spectroscopy. One example in which we distinguish clearly between isomers 1-C10H9+ and 2-C10H9+ is shown in Fig. 1. The spectrum of 1-C10H9+ recorded with the IRMPD method67 and that of Ar-tagged 1-C10H9+ (ref. 66) are shown in traces (A) and (B) of Fig. 1, respectively, for comparison with that of 1-C10H9+ in solid p-H2 and that from calculations, shown in traces (C) and (D) of Fig. 1, respectively. 2-C10H9+ was unstable and became converted to 1-C10H9+ in less than 30 min,41 but we observed its diminishing lines shown in Fig. 1(E) and assigned them unambiguously after comparison with calculated spectra of 2-C10H9+ and 4a-C10H9+, shown in Fig. 1(F) and (G), respectively. The most intense line of 2-C10H9+ at 1623.9 cm−1 lies close to a line of 1-C10H9+ at 1618.7 cm−1; these are readily discernible in solid p-H2, but not with the IRMPD or IRPD methods. Compared with the action spectra recorded with the IRMPD and IRPD methods, the absorption spectrum of H+PAH in solid p-H2 provides true relative IR intensities, which are important for a comparison with data from quantum-chemical calculations to derive the correct assignments. Furthermore, because of the great stability of PAH, H+PAH, and H1PAH, electron bombardment induces little fragmentation of these molecules; the species after bombardment are only PAH, H+PAH, and H1PAH, occasionally with anions of PAH in a small proportion; this feature makes possible the assignments of H+PAH and H1PAH. With a wide spectral coverage, many more lines could be determined and help to confirm the spectral assignments.

image file: c7cp05680j-f1.tif
Fig. 1 Comparison of experimental IR absorption spectra of isomers of C10H9+ with theoretical predictions. (A) Spectrum of C10H9+ recorded with the IRMPD method (ref. 67). (B) Spectrum of Ar-tagged C10H9+ recorded with the IRPD method (ref. 66). (C) Inverted difference spectrum after irradiation at 365 nm on an electron-bombarded C10H8/p-H2 matrix; lines of 1-C10H9+ (group A+) are indicated with arrows. (D) Stick spectrum of 1-C10H9+ simulated according to anharmonic vibrational wavenumbers and IR intensities calculated with the B3PW91/6-311++G(2d,2p) method. (E) Experimental spectrum of 2-C10H9+ (group B+). (F) Stick spectrum of 2-C10H9+, and (G) stick spectrum of 4a-C10H9+ simulated similarly as in (D). Lines marked with # are due to interference from C10H8; those marked with ? are uncertain. Reproduced from ref. 41 with permission from the PCCP Owner Societies.

One major disadvantage of our method is the lack of mass selection, which might prevent correct assignments in a complicated system, but, because little fragmentation occurs for PAH, the spectrum is contributed typically from only the parent PAH, H+PAH, and H1PAH. As discussed previously, lines of H+PAH and H1PAH become readily distinguishable on either maintaining the electron-bombarded matrix in darkness for a prolonged period or irradiating the matrix with UV light. Furthermore, secondary photolysis can generally help to distinguish groups of lines of various isomers. Another disadvantage is that the sensitivity of IR absorption is less than that of IRMPD or IRPD methods, but this condition can be compensated on depositing a thick matrix to increase the number of molecules in the light path so as to provide an excellent ratio of signal to noise. We typically deposited a PAH/p-H2 mixture for 7–8 h with a flow rate of 11–13 mmol h−1; the thickness of resultant matrices is 700–1000 μm (optical path length of 2–3 mm for IR light with a 45° angle of incidence and reflection). Because larger PAH have limited availability, isotopic experiments are rarely performed to support the assignments in all experimental methods.

H+PAH in solid p-H2 and the UIR bands. Fig. 2 presents a comparison of the UIR emission spectrum observed from the Iris nebula, NGC 7023,68 with laboratory p-H2 matrix-isolation IR spectra of protonated ovalene (7-C32H15+),44 coronene (1-C24H13+),43 pyrene (1-C16H11+),42 and naphthalene (1-C10H9+).41 For ease of comparison, the laboratory spectra are shown with sticks to represent relative intensities; spectra convoluted with a Gaussian profile with full width at half maximum (FWHM) of 70 cm−1 are shown also as black traces. All these protonated species have IR features near 6.2 μm, which is characteristic of the CC-stretching mode of H+PAH. These features shift insignificantly as the size increases, but the two combination bands of 7-C32H15+ (shown in light blue) at 6.06 and 6.08 μm, which are slightly blue-shifted from the fundamental bands near 6.2 μm, gain significant intensity. Intense features in the region 7.0–7.5 μm observed for 7-C32H15+, 1-C24H13+, and 1-C16H11+ shift to the red as their sizes increase; their wavelengths approach the UIR feature at 7.7 μm. These features are due to ring-deformation and CH2-scissoring modes; the latter is a characteristic feature of H+PAH protonated at an unfused edge. The features originating from the in-plane CH-bending modes (in black) also shift to the red as the size increases, and might be responsible for the 8.6 μm feature of the UIR bands. Intense features near 11–12 μm associated with the out-of-plane CH-bending modes (in pink) seem to shift to the blue as the size increases. In 7-C32H15+, two lines at 11.0 and 11.7 μm, assigned to the out-of-plane CH-bending modes involving both solo and duo hydrogens, are slightly blue-shifted from the UIR bands at 11.2 and 12.0 μm, respectively. As shown in Fig. 2, the best agreement with the UIR bands is the IR spectrum of 7-C32H15+. Moreover, if we take into account the red shifts (10–30 cm−1) associated with converting absorption spectra to emission spectra,69,70 the agreement improves further. Our experimental data hence seem to indicate that protonated ovalene, 7-C32H15+, might contribute to the interstellar PAH inventory.44 Although pyrene and coronene are considered to be too small to survive in the interstellar radiation field, ovalene lies at the lower end of the interstellar distribution of PAH size; protonated ovalene might well be one carrier of the UIR bands.57
image file: c7cp05680j-f2.tif
Fig. 2 Comparison of the UIR bands with observed spectra of H+PAH. (A) UIR spectrum observed from the Iris nebula, NGC 7023,68 and experimental stick spectra of (B) protonated ovalene (7-C32H15+),44 (C) protonated coronene (1-C24H13+),43 (D) protonated pyrene (1-C16H11+),42 and (E) protonated naphthalene (1-C10H9+).41 Each type of mode is indicated with a different color: pink, out-of-plane CH bend; black, in-plane CH bend; green, CH2 scissor; red, ring deformation; blue, CC stretch; light blue, combination. Solid black lines represent spectra convoluted with Gaussian profiles of FWHM 70 cm−1.
Non-planar H+PAH: protonated corannulene. We extended our investigation to non-planar protonated species; we here present some preliminary experimental results on protonated corannulene (C20H11+).

In corannulene (C20H10), five benzene rings are fused with a pentagonal ring (see the inset of Fig. 3). This molecule has a non-planar bowl shape and is considered to be a fragment of fullerene C60; as C60 is known to exist in the ISM,4 this species might also be present. Fig. 3 shows partial IR spectra of an electron-bombarded corannulene/p-H2 matrix. The black trace in Fig. 3(A) is the spectrum recorded after deposition, with lines of corannulene stripped according to a separate IR spectrum of corannulene/p-H2 matrix without electron bombardment; lines in this spectrum are hence all produced on electron bombardment. To differentiate lines due to protonated and hydrogenated corannulene, the matrix was maintained in darkness for 27 h; in the difference spectrum, shown as a red trace in Fig. 3(B), lines pointing upward indicate production and those pointing downward indicate destruction. The intensities of lines marked Y and Z increased after 27 h, implying that these lines are due to hydrogenated corannulene C20H11. These lines were separated into groups Y and Z based on their variations in intensity ratio; the intensities of lines in group Y (at 826.9, 807.8, 788.2, and 767.8 cm−1) increased by 20% and those in group Z (at 831.3, 822.8 and 777.9 cm−1) increased by 50% after 27 h. The intensities of three lines at 900.9, 878.3, and 854.1 cm−1 (marked as X+) decreased significantly after 27 h; as discussed previously, such a decrease is characteristic of protonated species. The lines in group X+ are hence likely due to protonated corannulene C20H11+.

image file: c7cp05680j-f3.tif
Fig. 3 Partial infrared spectra of an electron-bombarded corannulene (C20H10)/p-H2 matrix. (A) Spectrum recorded after deposition; lines of parent corannulene were stripped. (B) Difference spectrum showing the variations after maintaining the matrix in darkness for 27 h; lines pointing upward indicate production and those pointing downward indicate destruction. The spectral regions which suffer from interference from parent corannulene are shaded in gray. The inset shows the corannulene molecule and its distinct protonation sites. X+: C20H11+; Y: hub-C20H11, and Z: rim-C20H11.

Corannulene has three distinct protonation sites, at the hub, rim, and spoke, named according to a wheel (see Fig. 3). Protonation at the hub site yields the most stable isomer of C20H11+; isomers protonated at the rim and spoke sites have greater energies of 5.6 and 55.6 kJ mol−1, respectively, according to calculations with the B3PW91/6-311++G(2d,2p) method. The inverted difference spectrum (after 27 h in darkness) in the spectral region 1550–700 cm−1 is compared with the IRMPD spectrum of protonated corannulene71 in Fig. 4. When lines in group X+ are compared with scaled harmonic IR spectra predicted for hub-, rim-, and spoke-protonated corannulene, shown in Fig. 4(C)–(E), respectively, the best agreement is with the hub-protonated species (hub-C20H11+); we hence tentatively assign these lines to hub-C20H11+. We similarly assigned lines in groups Y and Z to hub-C20H11 and rim-C20H11, respectively, to be discussed elsewhere; hub-C20H11 is higher in energy by 17 kJ mol−1 than rim-C20H11. The IRPMD spectrum of gaseous C20H11+, shown in Fig. 4(A), was attributed, however, to the rim-protonated species (rim-C20H11+).71 For the first time distinct isomers of protonated PAH have been reported from the IRMPD and the matrix-isolation methods. To confirm the spectral assignments and to understand the discrepancies, further experiments and calculations are in progress.

image file: c7cp05680j-f4.tif
Fig. 4 Comparison of experimental and predicted spectra of protonated corannulene. (A) The IRMPD spectrum.71 (B) Inverted difference spectrum showing the variations after maintenance of the electron-bombarded corannulene/p-H2 matrix in darkness for 27 h; lines pointing upward indicate destruction. The spectral region which suffers from interference from parent corannulene is shaded in gray color. The lines in group X+ are indicated by arrows. (C–E) Predicted spectra of hub-, rim-, and spoke-protonated corannulene simulated according to the scaled harmonic wavenumber and IR intensity calculated with the B3PW91/6-311++G(2d,2p) method.

3.2. Protonated PANH (H+PANH)

Nitrogen-containing PAH, known as polycyclic aromatic nitrogen heterocycles (PANH), have been suggested to be partially responsible for the 6.2 μm feature of UIR bands because the N atom induces a blue shift of the CC-stretching modes of PAH near 6.3 μm.72 Moreover, the presence of a nitrogen atom within the PAH skeleton disrupts the electronic distribution within the entire PANH molecule, resulting in a doubled intensity in the region 1000–1600 cm−1.73 The existence of protonated PANH (designated H+PANH) in the ISM is expected because of the large proton affinity at the nitrogen site of PANH; for example, the proton affinity of quinoline (C9H7N) is 953 kJ mol−1, greater than 803 kJ mol−1 for naphthalene (C10H8).8 As a first step to study H+PANH, we applied our method to produce protonated quinoline (C9H7NH+) in solid p-H2.74 Vibronic spectra of protonated quinoline have been reported;75,76 IRMPD spectra of some H+PANH including protonated quinoline are also reported.77 The IRMPD spectra were recorded in the spectral region of only 600–1600 cm−1, likely limited by the tuning range of the free-electron laser employed. Our experiment involves an FTIR spectrometer to cover the full IR spectral region (500–4000 cm−1) so that the CH- and NH-stretching modes can confirm the preferred protonation sites.

The preliminary IR spectra of protonated quinoline (C9H7NH+) in solid p-H2 are characterized by features associated with the NH-stretching (3385.0 cm−1) and CH(NH)-bending (1298.0 cm−1) modes, confirming the protonation at the N atom; observed new features are thus assigned to 1-quinolinium (1-C9H7NH+). Even though we expect lines of 1-C9H7NH+ not to fit with the UIR bands because its size is too small to survive the cosmic radiation, this work clearly demonstrates the blue shifts for lines in the region of the CC-stretching mode, relative to those of H+PAH. As shown in Fig. 5, the bands of 1-C9H7NH+ at 1641.4, 1598.4, and 1562.0 cm−1 associated with the CC-stretching mode are blue-shifted from those at 1618.7, 1580.8, 1556.7, and 1510.0 cm−1 of the corresponding protonated PAH (C10H9+).

image file: c7cp05680j-f5.tif
Fig. 5 Comparison of partial IR spectra of (A) protonated quinoline (1-quinolinium, 1-C9H7NH+) and (B) protonated naphthalene (1-C10H9+) in solid p-H2.41,74 The experimental spectra are shown with sticks to represent relative peak intensities and those assigned to the CC stretching modes are indicated with red arrows. The blue shifts of the CC-stretching bands of protonated quinoline from those of protonated naphthalene are observed.

3.3. Protonation of small molecules

Protonated carbon dioxide, HOCO+, has been identified in the ISM.14–16 In 1908, its millimeter-wave lines observed from the Sgr B2 and Orion A were tentatively assigned by Thaddeus et al.,14 and subsequently confirmed by Bogey and co-workers with laboratory tests.15 Isovalent species HOCS+, HSCO+, and HSCS+ are yet to be identified despite their positive predictions. According to Fock and McAllister,78 the estimated number densities n(HOCS+ and HSCO+) are comparable with, or even greater than, n(HOCO+). The estimated ratio of number densities, n(HSCS+)/n(CS2) = 3 × 10−4, is a hundred times that of n(HOCO+)/n(CO2) because the proton affinity of CS2 (682 kJ mol−1) is greater than that of CO2 (541 kJ mol−1). The n(CS2) is unknown because nonpolar CS2 is undetectable with microwave spectroscopy; it is thus unclear whether HSCS+ exists in the ISM in a detectable amount.

Infrared spectra of HOCO+ in the gaseous phase,79,80 in solid Ne,81 and tagged with noble-gas atoms (He, Ne, and Ar)82 have been reported, but for protonated OCS only the OH-stretching fundamental mode of the HOCS+ isomer was reported.83 The SH-stretching fundamental mode of the HSCO+ isomer, which is more stable than HOCS+, has been extensively sought, but unsuccessfully,83,84 likely because of the method of protonation.85 We applied our electron bombardment method to mixtures of p-H2 with CO2, OCS, and CS2, in series and recorded IR spectra of HOCO+,86 HOCS+, HSCO+,45 and HSCS+ (ref. 87) in solid p-H2.

Of particular interest is the significant red shift of wavenumbers associated with the OH-stretching vibrational modes of HOCO+ and HOCS+ in solid p-H2 compared with those in the gaseous phase. In the case of HOCO+, the OH-stretching band at 3375.4 cm−1 in the gaseous phase79,80 is red-shifted to 2403.5 cm−1 in solid p-H2, shown in Fig. 6(A);86 this shift, amounting to −29%, is significant. We attribute this large red shift to proton sharing between CO2 and surrounding p-H2 molecules because the proton affinity of CO2 (541 kJ mol−1) is only 127 kJ mol−1 greater than that (422 kJ mol−1) of H2. The magnitude of the shift is smaller for HOCS+ (−14%, from 3435.2 to 2945.9 cm−1), shown in Fig. 6(B),45,83 because the proton affinity of OCS (protonation at the oxygen site, 610 kJ mol−1) is greater than that of CO2 and H2. Such correlations have been recognized for a proton-bound complex AH+-L (L is a ligand, such as a noble-gas atom).63 We found, however, that proton sharing is small in HSCO+ even though the proton affinity for the protonation at the sulfur site of OCS is 632 kJ mol−1, which is only slightly greater than that, 610 kJ mol−1, of protonation at the oxygen site. The proton affinity is hence not the sole parameter that determines the red shift of the vibrational wavenumbers of the mode involved; we must instead consider the interaction between H2 and the protonated species. The extent of protonic character of the H atom (i.e., the partial charge on the H atom) determines the strength of the interaction. The BSSE-corrected interaction energies of H2·HOCS+ and H2·HSCO+ are −17.3 and −4.7 kJ mol−1, respectively. The vibrational wavenumber of the OH-stretching mode of HOCS+ is consequently red-shifted by 604 cm−1 upon complexation, much greater than the corresponding red shift, 186 cm−1, for the SH-stretching mode of HSCO+, according to calculations with the B3LYP/aug-cc-pVTZ method. The SH-stretching modes of HSCO+ and HSCS+ were indeed observed at 2506.9 and 2477.2 cm−1, respectively, similar to the anharmonic vibrational wavenumbers of 2408 and 2424 cm−1 predicted for gaseous HSCO+ and HSCS+, respectively, as illustrated in Fig. 6(C) and (D).

image file: c7cp05680j-f6.tif
Fig. 6 Stick diagram representing the OH stretching mode of (A) HOCO+ (ref. 86) and (B) HOCS+ (ref. 45) and the SH stretching mode of (C) HSCO+ (ref. 45) and (D) HSCS+.87 Gaseous phase (black), p-H2 (blue), and anharmonic wavenumbers predicted at the B3LYP/aug-cc-pVTZ level with the VPT2 method (red).

As discussed in Section 3, the electron bombardment of PAH-containing p-H2 matrices during deposition produced H+PAH and H1PAH with negligible fragmentation or further hydrogenation. In contrast, more varied products were identified in experiments with small molecules (CO2, OCS, and CS2);45,86,87 beyond the protonated and mono-hydrogenated species, di-hydrogenated species, anions, and hydrogenated anions were identified. The production of di-hydrogenated species implies that the number of H atoms produced might be larger than in experiments with PAH. Although the formation of multiple products complicates the analysis of the IR spectra, the experimental steps after deposition (i.e., keeping the matrix in darkness or secondary photolysis) allow us to assign products. One advantage of the matrix-isolation method is the requirement of the sample in only a minute amount, typically less than 1 Torr in a 3 L flask, which enables us to perform experiments with isotopically substituted samples; these isotopic experiments provide definitive support of the assignments.

Proton sharing with a H2 molecule is of great importance as a collision of molecule X with H3+ might lead to the formation of H2·H+X if the excess energy becomes dissipated via intramolecular redistribution of the vibrational energy, radiative cooling, or collision with other species. The spectra of these proton-shared species in solid p-H2 might provide insight for the search of spectra of such H2·H+X complexes.

3.4. Proton-bound dimers

Protonated species such as HN2+ and HCO+ are known to be abundant in the ISM; a microwave transition of HN2+ is commonly used to estimate the abundance of N2.88 The existence of proton-bound dimers such as N2–H+–N2 in the ISM has been proposed.89,90 The formation of proton-bound dimers is exothermic; e.g., ΔH of the reaction HN2+ + N2 → N2–H+–N2 has been determined experimentally to be 67 kJ mol−1.91 Proton-bound species, represented as X–H+–Y, play also an important role in spectroscopy because these species are regarded as intermediate in the proton-transfer reaction XH+ + Y → [X–H+–Y] → X + HY+; the strong anharmonicity associated with the proton motion of these species makes an understanding of their spectra challenging. For example, the IR spectrum of the proton-bound water dimer H5O2+ has been investigated extensively both experimentally and theoretically; full-dimensional vibrational analyses are required to interpret the spectra.92 Although the spectral shifts due to the tagged atom are thought to be small, the breaking of symmetry for the symmetric proton-bound dimers by the tagged atom might affect the spectra significantly.93,94 We recently found that Ar-tagging of centrosymmetric species N2–H+–N2 broke its symmetry and activated additional IR transitions.95
Proton-bound rare-gas dimers. To assess the applicability of our method for the preparation of proton-bound dimers, we initiated an investigation with proton-bound rare-gas dimers Rg–H+–Rg′ (Rg/Rg′ = Kr or Xe). Fridgen and Parnis reported the IR spectra of Rg–H+–Rg′ (Rg and Rg′ are distinct atoms of Ar, Kr, and Xe) in noble-gas matrices,96 but Lundel pointed out that, based on theoretical predictions, those spectra assigned to mixed dimers Rg–H+–Rg′ by Fridgen and Parnis pertain to symmetric dimers Rg′–H+–Rg′ in a mostly Rg environment.97 The proton-bound rare-gas dimers Xe–H+–Xe, Kr–H+–Kr, and Kr–H+–Xe were successfully produced in solid p-H2 using the electron-bombardment method.98 The IR spectrum of the mixed rare-gas dimer Kr–H+–Xe was identified; a line at 1284 cm−1 was assigned to the antisymmetric stretching (ν3) mode, of which the wavenumber is much greater than 847 and 871 cm−1 for the corresponding modes of Xe–H+–Xe and Kr–H+–Kr, respectively, and those previously assigned erroneously at 765 cm−1 for Kr–H+–Xe.96 The reason is that the proton was equally shared between the two noble-gas atoms in Xe–H+–Xe and Kr–H+–Kr, but was more strongly bound to Xe (proton affinity 496 kJ mol−1) than to Kr (proton affinity 425 kJ cm−1) in Kr–H+–Xe. This deduction is supported by quantum-chemical calculations that predict for Kr–H+–Xe a Xe–H bond of length 1.70 Å, significantly smaller than 1.85 Å predicted for Xe–H+–Xe, and a Kr–H bond of length 1.88 Å, significantly greater than 1.64 Å predicted for Kr–H+–Kr.
Proton-bound nitrogen dimer. Although the existence of a proton-bound nitrogen dimer, N2–H+–N2, in the ISM has been proposed, its zero dipole moment hampers detection with rotational spectroscopy. A high-resolution IR spectrum of the antisymmetric NN/NN-stretching mode of N2–H+–N2 was reported by Verdes et al.,99 who concluded that N2–H+–N2 has a linear centrosymmetric structure. The IRPD spectrum of Ar-tagged N2–H+–N2 in the spectral range 700–4000 cm−1 was subsequently reported by Ricks et al.,100 as shown in Fig. 7(A). An intense line associated with the NH+N-antisymmetric stretching (or proton-rattling) mode was observed at 743 cm−1. Additional lines were assigned to various combination modes that derived their intensity via anharmonic couplings with the proton motion. Because the argon atom apparently affects the spectrum of bare N2–H+–N2, we recorded the IR spectrum of N2–H+–N2 in solid p-H2 for comparison; we performed reduced-dimensional anharmonic calculations using a discrete-variable representation (DVR) to account for the anharmonic coupling in this system.95 An IR spectrum of N2–H+–N2 in solid p-H2 is shown in Fig. 7(B); lines assigned to N2–H+–N2 are indicated with red arrows. The corresponding spectrum predicted with the DVR method and the respective assignments are shown in Fig. 7(C); the agreement between p-H2 experiments and calculations is satisfactory (see Table 3 of ref. 95 for comparison of integrated intensities). Beyond lines of fundamental modes at 715.0 (NH+N-antisymmetric stretching, ν4), 1129.6 (NH+N-bending, ν6), and 2352.7 (antisymmetric NN/NN-stretching, ν3) cm−1, we observed combination bands in a series 2 + ν4, in which ν2 is the N2⋯N2 stretching mode, which confirms the strong coupling between these modes. The IR spectrum of N2–H+–N2 in solid p-H2 shows fewer lines than that of Ar–(N2)2H+; it also shows true relative absorption intensities that facilitate a direct comparison with data from calculations. The additional lines of Ar–(N2)2H+ are likely induced by the tagging Ar. The five-dimensional DVR calculations involving (ν2, ν4, ν6, ν6′, ν11) of Ar-tagged N2–H+–N2 indicated that, in addition to the series 2 + ν4, a combination band of 2ν2 + ν11 and two ν6 bands predicted at 790 and 1123/1146 cm−1 were activated by Ar, in agreement with the observed lines at 780 and 1144 cm−1.
image file: c7cp05680j-f7.tif
Fig. 7 Comparison of experimental and predicted spectra of N2–H+–N2. (A) IRPD spectrum tagged with Ar.100 (B) IR spectrum in solid p-H2;95 lines indicated with red arrows are due to N2–H+–N2 and two additional lines indicated with blue arrows might be induced by perturbation from the third N2 molecule nearby. Lines marked with * are likely to be due to N2H+. (C) IR spectrum predicted with the four-dimensional DVR calculations.95

One advantage of our method to investigate the proton-bound dimer is the ability to adjust independently the ratio of the two components in the mixture. The successful preparation of Kr–H+–Xe in solid p-H2 demonstrates this advantage, compared with methods using noble-gas matrices.98 In proton-bound dimers X–H+–X, the XH+X antisymmetric stretching mode (or proton-rattling) is the most characteristic vibrational mode because of its large intensity and anharmonicity. This intense band might facilitate astronomical detection of these species. The significant anharmonicity of this mode is related to mode coupling and is of fundamental importance in spectroscopy.

A disadvantage of our method is that we are unable to prepare proton-bound species with atoms or molecules that have a proton affinity less than that of H2 (422 kJ mol−1). For example, Ar–H+–Ar could not be prepared with our method because the proton affinity of Ar, 371 kJ mol−1, is smaller than that of H2; proton transfer from H3+ to Ar is endothermic. Recently, McDonald et al. produced H+Arn (n = 3–7) in a pulsed-discharge supersonic jet and observed their IRPD spectra;101 they concluded that the core proton-bound Ar dimeric cation is responsible for the observed infrared bands. Their method is more suitable for the preparation of light astrochemically relevant species such as Ne–H+–Ar and Ne–H+–Ne.102

4. Hydrogenated species

As a byproduct of protonated species, mono-hydrogenated species are invariably present in experiments with solid p-H2 after electron bombardment because of the neutralization of associated cations and the reaction of H with the parent. Alternatively, without electron bombardment, hydrogen atoms can be produced on UV photolysis of Cl2 followed by IR irradiation to induce the reaction Cl + H2 (v = 1) → HCl + H,35,38 or a reaction of photolytically produced hydroxyl radicals with H2. These methods have been applied to produce hydrogenated species in solid p-H2.

4.1. Hydrogenated PAH

Molecular hydrogen H2 is the most abundant molecular species in the ISM103–105 and plays important roles in many ion–molecule and neutral–neutral reactions;106 the formation and evolution of H2 in the ISM are thus important astrochemical processes. It is generally assumed that the combination of hydrogen atoms on grain surfaces is the dominant route for interstellar H2, but the large rate for the formation of H2 in the photodissociation regions is inexplicable with only reactions on grains.107 Recently, PAHs were suggested to serve as catalysts for the formation of H2;107–109 the proposed model involves the formation of hydrogenated PAH (designated HPAH) that begins with H1PAH; these H1PAH might thus be important intermediates in the reactions to form H2. The detection of H1PAH in the ISM consequently provides information about the spatial activity of H2 formation.

In a laboratory, recording an IR spectrum of gaseous mono-hydrogenated PAH (designated H1PAH) is challenging with conventional techniques because it is difficult to prepare sufficient gaseous H1PAH without triggering further hydrogenation. The electron bombardment and the photolysis methods for p-H2 provide excellent means to produce sufficient H1PAH; spectral signatures characteristic of H1PAH, identified in solid p-H2, might facilitate observations to probe them in the ISM. To date, we have reported IR spectra of hydrogenated benzene (C6H7),40 naphthalene (C10H9),41 pyrene (C16H11),42 coronene (C24H13),43 and ovalene (C32H15)110 since these H1PAH were produced when we generated H+PAH with electron bombardment. Hydrogenated PAHs are produced in our experiments via reactions H + PAH → H1PAH or H+PAH + e → H1PAH. According to quantitative analyses,110 the former is a major process during the maintenance of deposited matrices in darkness at 3.2 K, indicating that hydrogenation of PAH can occur under astronomical conditions at low temperature via quantum-mechanical tunneling of H atoms. One advantage of our method is that the formation of H1PAH is dominant because a small mixing ratio of PAH in p-H2 is used and only a few H atoms are produced; the reaction of H1PAH + H is unlikely to occur in our experiments even though it has no barrier. In experimental conditions such as the bombardment of solid PAH with hydrogen, fully hydrogenated PAH were reported to be produced.108

In the case of the hydrogenation of a planar PAH as large as ovalene, only the most stable isomer of H1PAH was observed because of the lowest barrier associated with tunneling reaction H + PAH. In the case of corannulene (C20H10, see Fig. 3 for the structure and possible hydrogenation site), however, we observed lines in two groups (Y and Z in Fig. 3) that might originate from hydrogenated corannulene C20H11. According to our preliminary calculations, the barriers associated with the formation of rim- and hub-C20H11 are similar, 10.9 and 10.7 kJ mol−1, respectively, whereas the energy of the hub-C20H11 is greater by 17 kJ mol−1 than that of the rim-C20H11. Thus, the preferred site of hydrogenation in PAHs might hence be predictable based on the potential-energy barriers associated with the H + PAH reactions. Multiple isomers were also identified in the case of quinoline, to be discussed in Section 4.2, and other aromatic molecules.

4.2. Hydrogenated PANH

As discussed above, hydrogenation can be also initiated photolytically on photolysis of Cl2 or precursors of OH to generate Cl or OH, respectively, to generate H + HCl or H + H2O; the former reaction requires also IR irradiation to induce the reaction with p-H2. The H atoms thus produced are extremely mobile as they can tunnel through the p-H2 matrix.111–113 In this respect, the spectra of H1PAH obtained with this method are expected to have ratios of signal to noise improved over those from electron-bombardment experiments.

This photolytic method was first applied to the production of pyridinyl radicals C5H5NH and 4-C5H6N to confirm the spectral assignments of these species produced from electron bombardment.114 Recently, we have employed this method to produce hydrogenated quinoline (quinolinyl radicals, C9H8N); the preliminary results show that, in addition to N-hydrogenated species 1-C9H7NH, the spectra of four C-hydrogenated species 3-, 4-, 7-, and 8-C9H8N, as shown in Fig. 8(A), (C) and (F), are identifiable; the numbering for the prefix is indicated in the inset of Fig. 8(A). The assignments of each isomer were derived on grouping the lines according to their behavior on secondary photolysis at 450, 405, 365, 313, and 230 nm, followed by comparison of quantum-chemical calculations with the B3LYP/6-311++G(d,p) method, as shown in Fig. 8(B), (D), (E), (G) and (H) for 1-C9H7NH, 3-, 7-, 8-, and 4-C9H8N, respectively. This example is excellent to show the advantage of this photolytic method; the electron-bombardment method provided definitive identification of only 1-C9H7NH. Although the spectra were complicated and various isomers were produced, with a careful analysis of these narrow lines upon secondary photolysis at varied wavelengths, we identified these isomers without ambiguity.

image file: c7cp05680j-f8.tif
Fig. 8 Comparison of experimental spectra with theoretically predicted spectra of various isomers of hydrogenated quinoline produced from irradiation of a quinoline(C9H7N)/Cl2/p-H2 matrix. (A) Inverted difference spectrum after secondary photolysis at 365 nm; lines of group 1 are indicated. (B) Calculated stick spectrum of 1-C9H7NH. (C) Inverted difference spectrum after secondary photolysis at 365 nm; lines of groups 3 and 7 are indicated. (D) Calculated stick spectrum of 3-C9H7N. (E) Calculated stick spectrum of 7-C9H7N. (F) Difference spectrum after secondary photolysis at 365 nm; lines of groups 8 and 4 are indicated. (G) Calculated stick spectrum of 8-C9H7N. (H) Calculated stick spectrum of 4-C9H7N. All calculated stick spectra are simulated according to anharmonic vibrational wavenumbers and IR intensities (in km mol−1) predicted with the B3LYP/6-311++G(d,p) method. The inset in (A) shows the quinoline molecule and its hydrogenation sites.

4.3. Hydrogenated HONO

Hydrogenation involving species containing nitrogen and oxygen atoms is important in the formation of interstellar organic molecules and for the origin of life. Several studies focused on hydrogenation reactions of NO and NO2,115–118 but an intermediate, nitrous acid (HONO), has not yet been explored. If HONO were formed from reaction H + NO2 or OH + NO, the hydrogenation of HONO would be expected to be important. According to theoretical calculations on H + HONO by Hsu et al.,119 paths for the addition of H to HONO via two stable radical intermediates N(OH)2 and ONH(OH) yielding end products HNO + OH and NO + H2O, respectively, are dominant.

Using the photolytic method on the HONO/p-H2 matrix, we found that HONO reacts with H atoms at 3.3 K to form mainly ONH(OH), which slowly decomposes to form H2O and NO.39 H atoms were produced from the reaction of H2 with OH, which was produced from photolysis of HONO at 365 nm. The ν1 (OH-stretching) mode observed at 3549.6 cm−1 was predicted at 3542 cm−1 (anharmonic vibrational wavenumber). Lines observed at 1465.0 and 1372.2 cm−1 were predicted at 1458 (ν3) and 1374 (ν4) cm−1. The most intense line predicted for ν6 at 896 cm−1 was observed as a broad doublet at 898.5/895.6 cm−1; this mode involves mainly stretching of the central NO bond with some contributions from the out-of-plane deformation and ONO-bending modes. Experiments with DONO or HO15NO replacing HONO provided support for the assignments of these lines to ONH(OH). Two additional lines observed at 3603.4 and 991.0 cm−1 agree satisfactorily with the two most intense lines predicted for N(OH)2, with anharmonic vibrational wavenumbers 3618 (ν1) and 1014 cm−1 (ν5), but agree poorly with those predicted for HN(OH)2.

Our identification of intermediate ONH(OH) indicates that the hydrogenation of HONO should be considered in astrochemical models. The hydrogenation of HONO might explain that HONO remains unobserved even in the dark regions of the interstellar media.

5. Electronic transitions

The current spectral characterization of species in solid p-H2 has been applied mainly through IR absorption; spectral investigations of electronic transitions of species isolated in solid p-H2 are rare. Electronic transitions of only B atoms (absorption and fluorescence),120 C3 (absorption),121 and C60 (absorption)122 have been reported. We have recently performed experiments on the electronic transitions of NO in solid p-H2 to explore the advantages and disadvantages of investigating electronic transitions of species isolated in solid p-H2.46

The Rydberg state A2Σ+ of NO is known for its formation of a bubble-like electronically excited state; in a matrix, its absorption wavelength is significantly blue-shifted from that of gaseous NO and also from the emitting wavelengths of the A2Σ+ state of NO in a relaxed noble-gas matrix cage.123–125 We observed, distinct from all preceding reports, emission of NO in solid p-H2 from states A2Σ+ (v = 1), A2Σ+ (v = 0), B2Π (v = 0), and a4Π (v = 0). For Rydberg state A2Σ+, the T00 value (44[thin space (1/6-em)]102 ± 20 cm−1) of NO emission in solid p-H2 deviates only ∼21 cm−1 from the value of 44080.5 cm−1 for the gaseous phase,126 much smaller than the deviations of ∼1456, 2297, 119, and 1253 cm−1 for matrices Ne, Ar, Kr, and Xe, respectively. The p-H2 matrix clearly imposes a smaller perturbation of the electronic transitions than other matrices, especially for emission from relaxed Rydberg “bubble” state A2Σ+. For valence state B2Π, even though the matrix shifts of 178 and 521 cm−1 for Ar and Ne matrices, respectively, from the gaseous-phase value126T00 = 45[thin space (1/6-em)]392 cm−1 are much smaller than those for Rydberg state A2Σ+, the matrix shift <70 cm−1 in p-H2 is still smaller than the shifts for Ar and Ne matrices. The full width at half maximum (FWHM) intensity of the emission line for A2Σ+ (v = 0) → X2Π (v) is also smaller (∼400 cm−1) for p-H2 than for noble-gas matrices (∼645 cm−1). The FWHM of emission bands B2Σ+ (v = 0) → X2Π (v) is the smallest (∼30 cm−1) for p-H2, compared with 170–250 cm−1 for noble-gas matrices.

The small matrix shift of T00 values of NO, especially for Rydberg state A2Σ+, for p-H2 relative to noble-gas matrices, might be an advantage in the use of p-H2 to investigate electronic transitions of a guest. The small line width and the ability to produce free radicals via photolysis in situ are certainly benefits of using p-H2 as a host to investigate electronic transitions of free radicals or protonated species. Further experiments are required to assess these characteristics.

6. Future perspective

The application of the p-H2 matrix isolation technique to record IR and fluorescence spectra of species of astrochemical interest has clear advantages in many aspects, including the clean production of H+PAH, H1PAH, and proton-bound dimers. IR absorptions of species isolated in solid p-H2 at low temperature exhibit narrow lines and accurate relative IR intensities, enabling a close comparison with theoretical predictions and a clear identification of various conformers or isomers. Even lacking mass selectivity, by careful consideration of the chemistry of production and reaction, the behavior on secondary photolysis, and a comparison with quantum-chemical calculations, spectral assignments of various isomers in complicated systems are feasible. For electronic transitions, the matrix shift of the transition origin and the spectral width seem to be much smaller than those for noble-gas matrices, especially for upper electronic states that have disparate electronic distributions. These features might facilitate a direct comparison of matrix spectra with diffuse interstellar bands, but further data are required for an assessment.

For H+PAH as possible carriers of the UIR emission bands, the spectrum of protonated ovalene has attained the positions of UIR bands, consistent with an expectation that the PAH must be larger than ovalene to survive cosmic radiation. It would be exciting to explore the IR spectra of H+PAH of even larger circumference and of varied structures to discover whether their IR bands match the UIR bands. The challenges that we face are the availability of such precursors and the smaller vapor pressures associated with the greater masses. A new method to evaporate these compounds might have to be designed. Previous reports already indicate that the agreements between experimental vibrational wavenumbers and quantum-chemically predicted values became less satisfactory as the size of H+PAH increases;43,44 the combination and overtone modes begin to play a significant role in these spectra. More sophisticated theoretical treatments are required to assist the experimental spectral assignments. So far, most effort has been focused on highly symmetric H+PAH; other potential carriers of UIR bands, such as H+PANH, H+PAH and H+PANH with varied structures, or fragments of fullerenes, deserve investigation. The UV-excitation IR-emission experiments on these potential carriers must be performed to verify that the emission bands match the UIR bands.

A collision of a smaller molecule X with H3+ might lead to the formation of H2·H+X complexes in the ISM. The absorption intensity of the H+X-stretching mode is significantly enhanced upon complexation, which might allow astronomical observations. Therefore, the spectra of these proton-shared species in solid p-H2 might provide insight for the search of the spectra of such H2·H+X complexes. The formation of such complexes also induces red shift of the H+X stretching mode; however, an accurate method to predict the magnitude of these shifts is still lacking. Further experimental and theoretical studies are needed.

The challenges for investigation of H1PAH are similar to those of H+PAH. From our preliminary results, it appears that more sites are feasible for hydrogenation than for protonation; the spectra of the products are consequently more complicated and require careful grouping and analysis of observed lines to derive correct assignments. The photolytic method complements the electron-bombardment method and significantly improves the ability to identify varied isomers. The mechanisms and theoretical base of selective production of isomers of H1PAH and H1PANH are yet to be established.

The photolytic method is useful also in the investigation of hydrogenation reactions. Furthermore, hydrogenated radical species must be identified to complete our understanding of the mechanism of the hydrogenation network in ice. The dynamics and mechanism of the tunneling reaction of H atoms remain unclear; further efforts should be exerted on these fundamental issues. Some reactions involving deuterium might play important roles in the H/D ratios in interstellar media.

The investigation of electronic spectra of H+PAH, H1PAH, PAH+, and similar species of fullerenes in solid p-H2 is still in its infancy. Many data await collection to assess the advantages of these experiments using p-H2. The worst scenario would be that at least some H+PAH or H1PAH, which are difficult to produce with other techniques, can be produced in solid p-H2, so that their electronic spectra can be investigated. If indeed the spectra of species such as C60+ in solid p-H2 matches well the cold gaseous-phase spectrum recently published23 and the diffuse interstellar bands in terms of spectral width and band position, this technique might provide a simple yet efficient way to identify the carriers of these diffuse interstellar bands.

Conflicts of interest

There is no conflict of interest to declare.


The Ministry of Science and Technology of Taiwan (grant MOST106-2745-M-009-001-ASP) and Ministry of Education, Taiwan (“Aim for the Top University Plan” of National Chiao Tung University) supported this work. National Center for High-performance Computing, Taiwan, provided computer time. We thank Prof. Masaaki Baba for providing corannulene sample.


  1. H. S. P. Müller, in Encyclopedia of Astrobiology, ed. M. Gargaud, et al., Springer, Berlin and Heidelberg Germany, 2011, pp. 1080–1086 Search PubMed.
  2. J. H. Lacy, J. S. Carr, N. J. Evans II, F. Baas, J. M. Achetermann and J. F. Arens, Astrophys. J., 1991, 376, 556–560 CrossRef CAS.
  3. E. Herbst and E. F. van Dishoeck, Annu. Rev. Astron. Astrophys., 2009, 47, 427–480 CrossRef CAS.
  4. J. Cami, J. Bernard-Salas, E. Peeters and S. E. Malek, Science, 2010, 329, 1180–1182 CrossRef CAS PubMed.
  5. L. J. Allamandola, in Galactic and Extra-galactic Infrared Spectroscopy, XVIth ESLAB Symposium, ed. M. F. Kessler and J. P. Phillips, Reidel, Dordrecht Netherlands, 1984, pp. 5–35 Search PubMed.
  6. G. M. Muñoz Caro, U. J. Meierhenrich, W. A. Schutte, B. Barbier, A. Arcones Segovia, H. Rosenbauer, W. H.-P. Thiemann, A. Brack and J. M. Greenberg, Nature, 2002, 416, 403–406 CrossRef PubMed.
  7. T. Oka, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 12235–12242 CrossRef CAS PubMed.
  8. E. P. L. Hunter and S. G. Lias, J. Phys. Chem. Ref. Data, 1998, 27, 413–656 CrossRef CAS.
  9. D. Buhl and L. E. Snyder, Nature, 1970, 228, 267–269 CrossRef CAS PubMed.
  10. B. E. Turner, Astrophys. J., 1974, 193, L83–L87 CrossRef CAS.
  11. S. Green, J. A. Montgomery Jr. and P. Thaddeus, Astrophys. J., 1974, 193, L89–L91 CrossRef CAS.
  12. A. Wootten, J. G. Mangum, B. E. Turner, M. Bogey, F. Boulanger, F. Combes, P. J. Encrenaz and M. Gerin, Astrophys. J., 1991, 380, L79–L83 CrossRef CAS.
  13. L. M. Ziurys, A. J. Apponi and J. T. Yoder, Astrophys. J., 1992, 397, L123–L126 CrossRef CAS.
  14. P. Thaddeus, M. Guélin and R. A. Linke, Astrophys. J., 1981, 246, L41–L45 CrossRef CAS.
  15. S. Petrie and D. K. Bohme, Mass Spectrom. Rev., 2007, 26, 258–280 CrossRef CAS PubMed.
  16. T. P. Snow and V. M. Bierbaum, Annu. Rev. Anal. Chem., 2008, 1, 229–259 CrossRef CAS PubMed.
  17. D. M. Hudgins, C. W. Bauschlicher Jr. and L. J. Allamandola, Spectrochim. Acta, Part A, 2001, 57, 907–930 CrossRef CAS.
  18. A. Pathak and P. J. Sarre, Mon. Not. R. Astron. Soc., 2008, 391, L10–L14 Search PubMed.
  19. O. Dopfer, in PAHs and the Universe, ed. C. Joblin and A. G. G. M. Tielens, EAD Sciences, Les Ulis France, 2011, pp. 103–108 Search PubMed.
  20. G. H. Herbig, Annu. Rev. Astron. Astrophys., 1995, 33, 19–73 CrossRef CAS.
  21. B. H. Folng and P. Ehrenfreund, Nature, 1994, 369, 296–298 CrossRef.
  22. J. Fulara, M. Jakobi and J. P. Maier, Chem. Phys. Lett., 1993, 211, 227–234 CrossRef CAS.
  23. E. K. Campbell, M. Holz, D. Gerlich and J. P. Maier, Nature, 2015, 523, 322–323 CrossRef CAS PubMed.
  24. H. E. Hallam, Vibrational Spectroscopy of Trapped Species, Wiley, New York, USA, 1973 Search PubMed.
  25. L. Andrews and M. Moskovits, Chemistry and Physics of Matrix-Isolated Species, North-Holland, Amsterdam, Netherlands, 1989 Search PubMed.
  26. M. J. Almond and A. J. Downs, Advances in Spectroscopy, Wiley, New York USA, 1989, vol. 17 Search PubMed.
  27. M. E. Jacox, J. Mol. Spectrosc., 1985, 113, 286–301 CrossRef CAS.
  28. B. Meyer, Low temperature Spectroscopy, Elsevier, New York, USA, 1971 Search PubMed.
  29. T. Oka, Annu. Rev. Phys. Chem., 1993, 44, 299–333 CrossRef CAS.
  30. T. Momose and T. Shida, Bull. Chem. Soc. Jpn., 1998, 71, 1–15 CrossRef CAS.
  31. T. Momose, M. Fushitani and H. Hoshina, Int. Rev. Phys. Chem., 2005, 24, 533–552 CrossRef CAS.
  32. K. Yoshioka, P. L. Raston and D. T. Anderson, Int. Rev. Phys. Chem., 2006, 25, 469–496 CrossRef CAS.
  33. M. Bahou, C.-W. Huang, Y.-L. Huang, J. Glatthaar and Y.-P. Lee, J. Chin. Chem. Soc., 2010, 57, 771–782 CrossRef CAS.
  34. M. E. Fajardo, in Physics and Chemistry at Low Temperatures, ed. L. Khriachtchev, Pan Stanford Publishing Pte. Ltd., Singapore, 2011, pp. 167–202 Search PubMed.
  35. M. Bahou, P. Das, Y.-F. Lee, Y.-J. Wu and Y.-P. Lee, Phys. Chem. Chem. Phys., 2014, 16, 2200–2210 RSC.
  36. Y.-F. Lee, W.-T. Chou, B. A. Johnson, D. P. Tabor, E. L. Sibert III and Y.-P. Lee, J. Mol. Spectrosc., 2015, 310, 57–67 CrossRef CAS.
  37. Y.-F. Lee and Y.-P. Lee, J. Chem. Phys., 2011, 134, 124314 CrossRef PubMed.
  38. P. L. Raston, S. C. Kettwich and D. T. Anderson, J. Mol. Spectrosc., 2015, 310, 72–83 CrossRef CAS and references therein.
  39. K. A. Haupa, A. G. G. M. Tielens and Y.-P. Lee, Phys. Chem. Chem. Phys., 2017, 19, 16169–16177 RSC.
  40. M. Bahou, Y.-J. Wu and Y.-P. Lee, J. Chem. Phys., 2012, 136, 154304 CrossRef PubMed.
  41. M. Bahou, Y.-J. Wu and Y.-P. Lee, Phys. Chem. Chem. Phys., 2013, 15, 1907–1917 RSC.
  42. M. Bahou, Y.-J. Wu and Y.-P. Lee, J. Phys. Chem. Lett., 2013, 4, 1989–1993 CrossRef CAS PubMed.
  43. M. Bahou, Y.-J. Wu and Y.-P. Lee, Angew. Chem., Int. Ed., 2014, 53, 1021–1024 CrossRef CAS PubMed.
  44. M. Tsuge, M. Bahou, Y.-J. Wu, L. Allamandola and Y.-P. Lee, Astrophys. J., 2016, 825, 96 CrossRef.
  45. M. Tsuge and Y.-P. Lee, J. Chem. Phys., 2016, 145, 164308 CrossRef PubMed.
  46. S.-C. Huang and Y.-P. Lee, Chem. Phys. Lett., 2016, 665, 53–58 CrossRef CAS.
  47. M.-C. Chan, M. Okumura and T. Oka, J. Phys. Chem. A, 2000, 104, 3775–3779 CrossRef CAS.
  48. S. Tam and M. E. Fajardo, Rev. Sci. Instrum., 1999, 70, 1926–1932 CrossRef CAS.
  49. B. A. Tom, S. Bhasker, Y. Miyamoto, T. Momose and B. J. McCall, Rev. Sci. Instrum., 2009, 80, 016108 CrossRef PubMed.
  50. L. Andrews and X. Wang, Rev. Sci. Instrum., 2004, 75, 3039–3044 CrossRef CAS.
  51. A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652 CrossRef CAS.
  52. C. T. Lee, W. T. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789 CrossRef CAS.
  53. J. P. Perdew, K. Burke and Y. Wang, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 16533–16539 CrossRef CAS.
  54. J. Bloino and V. Barone, J. Chem. Phys., 2012, 136, 124108 CrossRef PubMed.
  55. S. P. Willner, in Galactic and Extra-galactic Infrared Spectroscopy, XVIth ESLAB Symposium, ed. M. F. Kessler and J. P. Phillips, Reidel, Dordrecht Netherlands, 1984, pp. 37–57 Search PubMed.
  56. A. G. G. M. Tielens, in PAHs and the Universe, ed. C. Joblin and A. G. G. M. Tielens, EAD Sciences, Les Ulis France, 2011, pp. 3–10 Search PubMed.
  57. L. J. Allamandola, A. G. G. M. Tielens and J. R. Barker, Astrophys. J., Suppl. Ser., 1989, 71, 733–775 CrossRef CAS PubMed.
  58. J. L. Puget and A. Léger, Annu. Rev. Astron. Astrophys., 1989, 27, 161–198 CrossRef CAS.
  59. T. P. Snow, V. Le Page, Y. Keheyan and V. M. Bierbaum, Nature, 1998, 391, 259–260 CrossRef CAS PubMed.
  60. M. Hammonds, A. Pathak and P. J. Sarre, Phys. Chem. Chem. Phys., 2009, 11, 4458–4464 RSC.
  61. N. C. Polfer, Chem. Soc. Rev., 2011, 40, 2211–2221 RSC.
  62. H. Knorke, J. Langer, J. Oomens and O. Dopfer, Astrophys. J., Lett., 2009, 706, L66–L70 CrossRef CAS.
  63. E. J. Bieske and O. Dopfer, Chem. Rev., 2000, 100, 3963–3998 CrossRef CAS PubMed.
  64. M. A. Duncan, J. Phys. Chem. A, 2012, 116, 11477–11491 CrossRef CAS PubMed.
  65. G. E. Douberly, A. M. Ricks, P. v. R. Schleyer and M. A. Duncan, J. Phys. Chem. A, 2008, 112, 4869–4874 CrossRef CAS PubMed.
  66. A. M. Ricks, G. E. Douberly and M. A. Duncan, Astrophys. J., 2009, 702, 301–306 CrossRef CAS.
  67. U. J. Lorenz, N. Solcá, J. Lemaire, M. Maître and O. Dopfer, Angew. Chem., Int. Ed., 2007, 46, 6714–6716 CrossRef CAS PubMed.
  68. D. Cesarsky, J. Lequeux, A. Abergel, M. Perault, E. Palazzi, S. Madden and D. Tran, Astron. Astrophys., 1996, 315, L305–L308 CAS.
  69. J. D. Brenner and J. R. Barker, Astrophys. J., 1992, 388, L39–L43 CrossRef CAS.
  70. R. M. Williams and S. R. Leone, Astrophys. J., 1995, 443, 675–681 CrossRef CAS.
  71. H. A. Galué, C. A. Rice, J. D. Steill and J. Oomens, J. Chem. Phys., 2011, 134, 054310 CrossRef PubMed.
  72. E. Peeters, S. Hony, C. Van Kerckhoven, A. G. G. M. Tielens, L. J. Allamandola, D. M. Hudgins and C. W. Bauschlicher, Astron. Astrophys., 2002, 390, 1089–1113 CrossRef CAS.
  73. A. L. Mattioda, C. W. Bauschlicher Jr., A. Ricca, J. Bregman, D. M. Hudgins and L. J. Allamandola, Spectrochim. Acta, Part A, 2017, 181, 286–308 CrossRef CAS PubMed.
  74. C.-Y. Tseng and Y.-P. Lee, presented in part at 72nd meeting of International Symposium on Molecular Spectroscopy, Champaign-Urbana Illinois USA, June, 2017.
  75. C. S. Hansen, S. J. Blanksby and A. J. Trevitt, Phys. Chem. Chem. Phys., 2015, 17, 25882–25890 RSC.
  76. G. Féraud, L. Domenianni, E. Marceca, C. Dedonder-Lardeux and C. Jouvet, J. Phys. Chem. A, 2017, 121, 2580–2587 CrossRef PubMed.
  77. H. A. Galué, O. Pirali and J. Oomens, Astron. Astrophys., 2010, 517, A15 CrossRef.
  78. W. Fock and T. McAllister, Astrophys. J., 1982, 257, L99–L101 CrossRef CAS.
  79. T. Amano and K. Tanaka, J. Chem. Phys., 1985, 82, 1045–1046 CrossRef CAS.
  80. T. Amano and K. Tanaka, J. Chem. Phys., 1985, 83, 3721–3728 CrossRef CAS.
  81. M. E. Jacox and W. E. Thompson, J. Chem. Phys., 2003, 119, 10824–10831 CrossRef CAS.
  82. O. Dopfer, R. V. Olkhov, D. Roth and J. P. Maier, Chem. Phys. Lett., 1998, 296, 585591 CrossRef.
  83. T. Nakanaga and T. Amano, Mol. Phys., 1987, 61, 313–323 CrossRef CAS.
  84. S. Saebø, M. M. Sanz and S. C. Foster, Theor. Chem. Acc., 1997, 97, 271–276 CrossRef.
  85. M. C. McCarthy and P. Thaddeus, J. Chem. Phys., 2007, 127, 221104 CrossRef CAS PubMed.
  86. P. Das, M. Tsuge and Y.-P. Lee, J. Chem. Phys., 2016, 145, 014306 CrossRef PubMed.
  87. M. Tsuge and Y.-P. Lee, Phys. Chem. Chem. Phys., 2017, 19, 9641–9653 RSC.
  88. M. Womack, L. M. Ziurys and S. Wyckoff, Astrophys. J., 1992, 393, 188–192 CrossRef CAS.
  89. M. Mladenović and E. Roueff, Astron. Astrophys., 2014, 566, A144 CrossRef.
  90. Q. Yu, J. M. Bowman, R. C. Fortenberry, J. S. Mancini, T. J. Lee, T. D. Crawford, W. Klemperer and J. S. Francisco, J. Phys. Chem. A, 2015, 119, 11623–11631 CrossRef CAS PubMed.
  91. K. Hiraoka, P. P. S. Saluja and P. Kerable, Can. J. Chem., 1979, 57, 2159–2166 CrossRef CAS.
  92. O. Vendrell, M. Brill, F. Gatti, D. Lauvergnat and H.-D. Meyer, J. Chem. Phys., 2009, 130, 234305 CrossRef PubMed.
  93. J. M. Bakker, R. G. Satink, G. von Helden and G. Meijer, Phys. Chem. Chem. Phys., 2002, 4, 24–33 RSC.
  94. J. C. Bopp, J. R. Roscioli, M. A. Johnson, T. M. Miller, A. A. Viggiano, S. M. Villano, S. W. Wren and W. C. Lineberger, J. Phys. Chem. A, 2007, 111, 1214–1221 CrossRef CAS PubMed.
  95. H.-Y. Liao, M. Tsuge, J. A. Tan, J.-L. Kuo and Y.-P. Lee, Phys. Chem. Chem. Phys., 2017, 19, 20484–20492 RSC.
  96. T. D. Fridgen and J. M. Parnis, J. Chem. Phys., 1998, 109, 2155–2161 CrossRef CAS.
  97. J. Lundel, J. Mol. Struct., 1995, 355, 291–297 CrossRef.
  98. M. Tsuge, J. Kalinowski, R. B. Gerber and Y.-P. Lee, J. Phys. Chem. A, 2015, 119, 2651–2660 CrossRef CAS PubMed.
  99. D. Verdes, H. Linnartz, J. P. Maier, P. Botschwina, R. Oswald, P. Rosmus and P. J. Knowles, J. Chem. Phys., 1999, 111, 8400–8403 CrossRef CAS.
  100. A. M. Ricks, G. E. Douberly and M. A. Duncan, J. Chem. Phys., 2009, 131, 104312 CrossRef.
  101. D. C. McDonald II, D. T. Mauney, D. Leicht, J. H. Marks, J. A. Tan, J.-L. Kuo and M. A. Duncan, J. Chem. Phys., 2016, 145, 231101 CrossRef PubMed.
  102. R. C. Fortenberry, ACS Earth Space Chem., 2017, 1, 60–69 CrossRef CAS.
  103. J. M. Shull and S. Beckwith, Annu. Rev. Astron. Astrophys., 1982, 20, 163–190 CrossRef CAS.
  104. J. H. Black and E. F. van Dishoeck, Astrophys. J., 1987, 322, 412–449 CrossRef CAS.
  105. B. T. Draine and F. Bertoldi, Astrophys. J., 1996, 468, 269–289 CrossRef CAS.
  106. A. G. G. M. Tielens, The Physics and Chemistry of the Interstellar Medium, Cambridge University Press, Cambridge UK, 2005 Search PubMed.
  107. L. Boschman, S. Cazaux, M. Spaans, R. Hoekstra and T. Schlathölter, Astron. Astrophys., 2015, 579, A72 CrossRef.
  108. A. L. Skov, J. D. Thrower and L. Hornekar, Faraday Discuss., 2014, 168, 223–234 RSC.
  109. E. Rauls and L. Hornekær, Astrophys. J., 2008, 679, 531–536 CrossRef CAS.
  110. M. Tsuge, M. Bahou, Y.-J. Wu, L. Allamandola and Y.-P. Lee, Phys. Chem. Chem. Phys., 2016, 18, 28864–28871 RSC.
  111. F. M. Mutunga, S. E. Follett and D. T. Anderson, J. Chem. Phys., 2013, 139, 151104 CrossRef PubMed.
  112. L. O. Paulson, F. M. Mutunga, S. E. Follett and D. T. Anderson, J. Phys. Chem. A, 2014, 118, 7640–7652 CrossRef CAS PubMed.
  113. W. R. Wonderly and D. T. Anderson, J. Phys. Chem. A, 2014, 118, 7653–7662 CrossRef CAS PubMed.
  114. B. Golec, P. Das, M. Bahou and Y.-P. Lee, J. Phys. Chem. A, 2013, 117, 13680–13690 CrossRef CAS PubMed.
  115. V. Blagojevic, S. Petrie and D. K. Bohme, Mon. Not. R. Astron. Soc., 2003, 339, L7–L11 CrossRef CAS.
  116. G. Fedoseev, S. Ioppolo, T. Lamberts, J. F. Zhen, H. M. Cuppen and H. Linnartz, J. Chem. Phys., 2012, 137, 054714 CrossRef CAS PubMed.
  117. M. Minissale, G. Fedoseev, E. Congiu, S. Ioppolo, F. Dulieu and H. Linnartz, Phys. Chem. Chem. Phys., 2014, 16, 8257–8269 RSC.
  118. S. Ioppolo, G. Fedoseev, M. Minissale, E. Congiu, F. Dulieu and H. Linnartz, Phys. Chem. Chem. Phys., 2014, 16, 8270–8282 RSC.
  119. C.-C. Hsu, M. C. Lin, A. M. Mebel and C. F. Melius, J. Phys. Chem. A, 1997, 101, 60–66 CrossRef CAS.
  120. S. Tam, M. Macler, M. E. DeRose and M. E. Fajardo, J. Chem. Phys., 2000, 113, 9067–9078 CrossRef CAS.
  121. H. Hoshina, Y. Kato, Y. Morisawa, T. Wakabayashi and T. Momose, Chem. Phys., 2004, 300, 69–77 CrossRef CAS.
  122. N. Sogoshi, Y. Kato, T. Wakabayashi, T. Momose, S. Tam, M. E. DeRose and M. E. Fajardo, J. Phys. Chem. A, 2000, 104, 3733–3742 CrossRef CAS.
  123. M. Chergui, N. Schwentner, W. Böhmer and R. Haensel, Phys. Rev. A: At., Mol., Opt. Phys., 1985, 31, 527–529 CrossRef CAS.
  124. M. Chergui, N. Schwentner and W. Böhmer, J. Chem. Phys., 1986, 85, 2472–2482 CrossRef CAS.
  125. M. Chergui and N. Schwentner, J. Chem. Phys., 1992, 97, 2881–2890 CrossRef CAS.
  126. R. de Vivie and S. D. Peyerimhoff, J. Chem. Phys., 1988, 89, 3028–3043 CrossRef CAS.


Present address: Institute of Low Temperature Science, Hokkaido University, Sapporo, Hokkaido 060-0819, Japan. E-mail: tsuge@lowtem.hokudai.ac.jp

This journal is © the Owner Societies 2018