Laser-induced Fluorescence of Mass-selected Electrosprayed Large Protonated Polycyclic Aromatic Hydrocarbons in the Gas Phase and Isolated in Solid Ne
Carbon-containing molecules and ions play an important role in space. Polycyclic Aromatic Hydrocarbons (PAHs) are the best-known candidates to account for the infrared emission bands (UIR). PAHs are also thought to be among the carriers of the diffuse interstellar absorption bands (DIBs). The DIBs – weak unidentified absorption features in the near ultraviolet (UV), visible, and near-infrared (NIR) region of the electromagnetic spectrum – were first observed in the 1920s. These bands are observed on lines of sight containing sufficiently high column densities, such as those traversing diffuse interstellar clouds. Although neutral PAHs have been postulated to be the carriers of DIBs and UIRs, no lines corresponding to particular carriers have been identified. The development of “The PAH hypothesis”, which holds PAHs responsible for the UIRs due to UV-pumped IR fluorescence, was very quickly followed by a hypothesis suggesting that ionized PAHs were likely carriers of the DIBs. Both experimental and theoretical studies support the identification of PAHs, and PAH ions, as the carriers of the DIBs and UIRs. Models have shown that UV pumping, or even NIR absorption (in areas of low UV flux), can drive mid-IR emission from PAHs, effectively linking DIBs with UIRs. One of the remaining points of debate is the nature of the PAH ions: whether they are radical cations (PAH+), protonated species (H+PAHs) or dehydrogenated PAH anions.1
Laboratory studies have confirmed up to now only contribution of C60+ in DIBs.2 It may support the hypothesis that PAHs ions are the carriers of DIBs and UIRs.
Protonated molecular hydrogen, H3+, is abundantly produced in the ISM, but its steady-state concentration is
small because of its great reactivity. 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 greater than that of H2 (422 kJ mol−1). Protonated species are hence thought to be abundant in the ISM.
Several protonated polyatomic species have been identified in the ISM. As PAHs are believed to be abundant in the ISM, protonated polycyclic aromatic hydrocarbons (H+PAHs) have been postulated also to be present in the ISM. In another scenario UV radiation can ionize neutral PAHs in the ISM. Their ionization potentials are low (6-9 eV depending their size and geometry). That PAH+ may react with highly abundant atomic/molecular hydrogen to form H+PAHs.3
Although large PAHs and H+PAHs are good candidates for UIR and DBI bands carriers, their laboratory measurements are challenging. The difficulties in these studies appear in three aspects: i) difficulties in generation of H+PAHs in sufficient quantities for spectral interrogation, ii) complicated nature of the spectra, and iii) necessity to transfer the large molecules to the gas phase without their fragmentation.
Spectral data is available for a number of smaller molecules. Several studies have been performed in the gas phase using IRMPD and Ar-tagging.4 A number of protonated aromatic hydrocarbons from benzene to ovalene was produced via electron bombardment in solid para-hydrogen and characterized with FTIR spectroscopy.5 Electronic absorption and laser induced fluorescence studies were performed for the smaller H+PAHs (up to coronene) isolated in solid Ne by J. Maier’s group.6 However, a number of laboratory investigations focused on spectral characteristic of neutral and cationic PAHs in the gas phase spectral data on bigger H+PAHs is missing in the literature. Protonated ovalene is up to now the largest H+PAH molecule characterized by the FTIR matrix isolation technique. Molecules smaller than ovalene are considered not to survive the interstellar radiation field.7 Ovalene lies at the lower end of the interstellar PAH size and studies on larger PAHs are essential.
Proposal Karolina Haupa
The main goal of the project is to extend the spectral database of protonated PAHs to larger molecules. The project will contain two stages. In the first one, the approach utilizes the combination of nano-electrospray ionization (nano-ESI) and mass-selection with matrix isolation spectroscopy. Ions of interest will be deposited in solid Ne and investigated with laser-induced fluorescence (LIF) spectroscopy. In the next stage, the studies will be extended to trapped ion laser-induced fluorescence (TLIF) of H+PAHs produced with ESI. In parallel, attempts to improve the matrix isolation setup by increasing ion flux to enable FTIR experiments will be performed.
The results of this study can help to evaluate astrochemical observations. In the long term, this can help to identify the UIRs and DBIs carriers and give some new insights to the fundamental questions about the origin of life. The instrumental development may also lead to creation of new powerful techniques which in the future can be applied in many areas of chemistry, i.e. studies on structure and spectroscopy of bigger protonated biomolecules.
Matrix isolation spectroscopy (MIS) is an extremely useful technique for a first spectroscopic assessment of species with low vapor pressure. There are many advantages of MIS: Species whose vapor pressure is too low to carry out absorption measurements in supersonic jets may be studied since they can be accumulated.
Due to the cryogenic temperature, only the lowest energy levels of the molecules are populated, resulting in simpler spectra typical for cold molecules. Finally, because the molecules are isolated from each other, spectra are not affected by intermolecular interactions. On the other hand, interactions between the molecules and the atoms of the matrix give rise to broadened and shifted absorption bands relative to measurements in supersonic jets where the molecules are truly isolated, i.e. in a collision-free environment.
This means that spectra obtained by MIS usually cannot be directly compared with the spectra of interstellar molecules for the purpose of their identification. Despite this fact, the observation of two DIBs observed at 963.2 nm and 957.7 nm was based on two electronic absorption bands of C60+ originally observed near 964.5 nm and 958.3 nm in a Ne matrix and later confirmed by gas-phase measurements, making this a good example of the possible accuracy of that approach.2
Combining matrix isolation with mass-selection further enhances the capability of MIS. Ions of interest can be produced, transferred, mass-selected, and deposited into a matrix. The equipment and experience of the host laboratory is necessary to realize that project. The general scheme of the apparatus currently developed in Prof. M. Kappes’ laboratory8 is presented in Fig. 1.
The ions of interest will be produced using a nano-ESI source. Electrospray ionization is a commonly used technique in which ions are produced by a high voltage applied to a liquid to create an aerosol. The solvent from the droplets progressively evaporates, leaving them more and more charged. When the charge exceeds the limit, the droplet dissociates in the Coulomb explosion leaving a stream of charged ions. Electrospray has many advantages which can be helpful in H+PAHs production:
Figure 1 Scheme of the MIS apparatus.8
1) The sample is prepared as a low concentration solution from where they are transferred into the gas phase. It limits the amount of material usage and enables easy transfer of the large molecules into the gas phase.
2) ESI belongs to “soft ionization” methods. This means that fragmentation of even big molecules is unlikely.
3) Produces mostly protonated species.
PAHs are exclusively available as solid samples with low vapor pressure. At room temperature, they are soluble in alcohols. Proton affinities of PAHs are high, so they can be protonated under ESI conditions. The studies with the GS(LC)/ESI-MS/MS technique showed that H+PAHs are the dominant ions produced in the ion sources and their protonation efficiency increases with PAH molecular mass.9 ESI was successfully employed to obtain smaller protonated PAHs from methanol solutions in amounts sufficient for IRMPD gas phase studies. To enhance protonation efficiency, a small portion of ammonium acetate was added.10
In our preliminary experiment we produced protonated ovalene with ES ionization with sufficient signal. We also found that H+PAH is formed as dominant ion and the population of H+PAH and PAH+ can be controlled by ESI parameters.
Electrosprayed ions are transferred via an ion optical path passing through several vacuum stages (marked with gray lines in Fig. 1). At the first stage, ions are focused in the ion funnel. The ion funnel “empties” into a radio-frequency ion guide and further through cylindrical electrostatic Einzel lenses to an electrostatic quadrupole bender which deflects charged particles by 90° but allows neutral molecules to continue straight on the axis unaffected, thus preventing contamination of the cryomatrix sample by residual neutrals from the ion source. The deflected ions are focused into a quadrupole mass filter (QMS) to select the desired m/z ratio, thereby removing unwanted charged species. Following the QMS, several Einzel lenses focused the ion beam onto the cold (∼5K) target. Prior to deposition, the ion current can be monitored with a picoamperometer.
To prepare samples of matrix-isolated ions for spectroscopy, the ion beam is co-deposited together with an excess of Ne onto a cold Al or Au coated sapphire substrate cooled to ∼5 K by a closed-cycle helium cryostat. Using that setup, it is possible to carry out “clean” deposition for a very long time (up to days). Maximum ion current measured on the target is 10-300 pA. The concentration of the ions in the matrix can be controlled by the amount of Ne added. Typical concentrations sufficient for LIF spectroscopic measurements is are 1 : 500 000.8
Deposition of ions of only positive charge into an insulating matrix results in charge accumulation, and incoming cations start to be deflected. The charge in the matrix has to be at least partially compensated. It can be achieved by addition of anions. For this purpose, the matrix can be doped with electron scavengers like CCl4, CO2, O2 or X2. The same procedure can also prevent cations from neutralization due to electron capture. The experimental procedure in this project will also follow that approach. Chloroform (CH3Cl) has a very high electron affinity (~3eV) and efficiently captured electrons induce the CH3Cl → Cl-+ CH3• reaction. It can be used in very low concentrations, so it does not interfere with trapped cations. None of the CH3Cl, CH3•, or Cl- species has have absorptions in the 200-1100 nm spectral range. Moreover, chlorine anions reduce the space charge, so more efficient cation beam deposition is possible.6
Application of CH3Cl as an additional dopant has more advantages. UV irradiation with ~300 nm light (from a small LED diode) induces neutralization of the trapped cations due to reaction with electrons dissociated from the Cl- counter ions. This fact can be used to distinguish protonated PAHs from their neutral counterparts–hydrogenated species (HPAHs). The differential spectra recorded before and after irradiation can provide information about these astrochemically relevant group of molecules in one experiment. After deposition, the target with the collected matrix can be rotated for the spectra collection. For the LIF measurements, the trapped ions can be excited with the tunable parametric oscillator laser (OPO) pumped by Nd:YAG / dye laser in the range 400-700 nm and 800-1100 nm and CW diode laser emitting at 420 nm. The light reflected from the target is focused on a CCD camera (optical fiber) or Raman spectrometer. The excitation and emission is perpendicular to the substrate. Absorption and emission are recorded simultaneously. All necessary equipment is currently in the host laboratory.
LIF allows the selective excitation of a single species. H+PAHs generally absorb light in the range 200-400 nm (S0 → Sn ; n=2,3,4…). S0 → S1 is located in the visible range (e.g. 678.5 nm was recorded for protonated coranulene) and they are generally red-shifted with an increase of molecular mass. The fluorescence spectrum is extended to the near infrared (NIR). Some vibrational propagation which can provide a set of vibrational excitations can be also observed.11
Even a matrix doped with mass-selected ions may contain more than one isomer and some unwanted species. In some extreme situations, the excitation energy of two compounds is very close. In this case the absorption spectra can be used. One way to distinguish ions of interest from their neutral counterparts was described above. Another useful strategy in the spectral identification of dopants is irradiation of the matrix with selected wavelengths in order to selectively bleach certain species. The induced changes can be used to discern the difference spectra, which is very useful for line grouping.
The assignment of electronic transitions will also be supported by quantum-chemical calculations. Calculated optimized structures, harmonic and anharmonic vibrational frequencies, thermodynamic properties, and electronic transitions energies suffice for a full analysis of the species. The size of the molecules planned to study in that project is large (thousands of electrons) so the use of highly correlated quantum chemistry methods is challenging. In this project, the hybrid-DFT methods (B3LYP, B3PW91, M062x) with timedependent extension TD-DFT including Coulomb correction (CAM-TD-B3LYP etc.) will be applied. Those methods are based on a modeling of electron correlation via general functions and electron density with a portion of exact exchange from Hartree-Fock theory. That approach is widely applied for many “bigger” molecules. TD-DFT has been used to calculate electronic transitions and oscillator strengths of protonated coronene and ovalene, and the experimental studies confirmed their reliability. If there is a need for more accurate calculations, the second-order approximated coupled-cluster (CC2) method can be also performed. All necessary tools are already implemented in the TURBOMOLE software package and can be directly applied. The use of super-computers (access available in host laboratory) allows performing these calculations in a reasonable amount of time.
During the project, it is planned to use the described procedure to perform a set of experiments using bigger PAHs. In the first step, in order to test the usability of the method, protonated ovalene has been chosen as the first experimental target. Ovalene is commercially available, and a number of experimental data including FTIR spectra in solid p-H2 are available for comparison. In the next stage, the experiments will be extended to hexabenzocoronenes (HBCs): hexa-peri-hexabenzocoronene (C42H18), 1,2:3,4:5,6:7,8:9,10:11,12 -hexabenzocoronene (C48H24), circumcoronene (C54H18), as well as small fullerenes: C60 and C70. All these compounds are commercially available. The host laboratory also has a collaboration with Prof. Klaus Müllen and Dr. Akimitsu Narita from the Max Planck Institute for Polymer Research who are specialists in synthesis of larger PAHs and can provide samples of molecules containing up to 96 C-atoms. Although the absorption and emission data cannot be directly compared with the astrochemical spectra due to the matrix effects, they can serve as a basis for comparison with gas-phase studies which will be recorded in the second stage of studies. The TLIF experimental setup available in the host laboratory is presented in Fig. 2. The TLIF apparatus consists of an electrospray ion source, ion optics, and a quadrupole ion trap coupled to a fluorescence microscope. The ion beam is generated by a home-built ES ionization source. The ion beam is electrostatically steered and focused into a RT quadrupole ion trap. The ion trap has a hyperbolic profile comprising an inner ring that corresponds to a stretched quadrupole trap configuration. The ion trap serves as both an ion storage device during the fluorescence measurements and as a mass spectrometer for ion isolation and detection. The walls of the ion trap can be regulated to temperatures in the range 90-650 K.12 The spectral resolution obtained in that experiment is expected to be sufficient to obtain information about vibrational levels. The gas phase spectra can be directly compared with astrochemical data to verify relevance of H+PAHs to the UIRs. The set of data on the vibrations of H+PAHs obtained from fluorescence spectra can be desirable.
As it was mentioned in Objectives, the second parallel goal of the project is a development of the experimental setup and extension of the spectral characterization into the IR range. Three spectral methods have been employed previously to yield the IR spectra of H+PAHs. In the case of FTIR spectra, the matrix shifts usually are not bigger than 1% and they provide real intensities, which shows its significant superiority of the spectra thus recorded to those with the Ar-tagging and IRMPD methods.
FTIR measurements require higher concentrations. To reach reasonable FTIR signals for moderately absorbing H+PAHs, deposition of about 5 nA of selected current is necessary. It implies a necessity to increase the (ESI flux) x (sensitivity) by a factor of 20. There are three strategies which can help to improve the apparatus performance: (i) modification of the ESI ion source and/or (ii) reduce the detection limit by modification of spectrometer optical path.
To achieve a high-intensity ESI source for the MIS experiments, the setup can be modified to incorporate a dual ion funnel, which enables operation with a higher gas load through an expanded diameter heated inlet into the additional first region of differential pumping as described previously.13 Further modification will include optimization of vacuum interface to exploit the hydrodynamic drag of the background gas for collimation and the reduction of space charge repulsion.14 These modifications enable a reduction in ion loss and increase ion current up to 4-5 times. The easiest way to improve FTIR detection is by measuring the absorption through the length of the matrix instead of its thickness. An increase of the angle between the IR beam and a the target elongates the beam path and increases the number of molecules the beam passes through. That simple modification can be easily obtained and can increase the sensitivity by another factor of 4-5.
Figure 2 Scheme of the TLIF apparatus.12
1 A.G.G.M. Tielens, 25 years of PAH hypothesis, in: PAHs and the Universe; C. Joblin, A.G G.M. Tielens, Eds. EAS Publication Series
2011 46, 240-248.
2 E. K. Campbell et al., 2015 Nature 523, 322.
3 V. L. Le Page et al. , ApJ 2003 584, 316.
4 F. Salama, Spectroscopy of PAH molecules and ions: from laboratory to astochemical observations in: Molecules in Space & Laboratory; J.L. Lemaire and F. Combes 2007 1
5 M. Tsuge, C.-Y. Tseng, and Y.-P. Lee, PCCP 2018 20, 5344.
6 I. Garkusha, J. Fulara, J.P. Maier, J. Mol. Struct. 2012 1025, 147 and cited papers.
7 Tielens A. G. G. M., 2008 ARA&A 46 289.
8 B. Kern et al., Anal. Chem. 2015, 87, 11901.
9 E. Cha et.al., Anal. Chem. 2018, 90, 4203.
10 H. Knorke et al., ApJ 2009 706, L66.
11 T. Pino et al., Electronic spectroscopy of PAHs in: Molecules in Space & Laboratory; J.L. Lemaire and F. Combes, 2011 355-371.
12 M. Kordel et al., JPCA 2010 114, 5509.
13 K. Don D. Gunaratne et al., Analyst 2015 140, 2957.
14 M. Pauly, Analyst 2014 139, 1856.