溶媒和クラスターカチオンの水素結合組み換えに関する分光学的研究

(1)

九州大学学術情報リポジトリ

Kyushu University Institutional Repository

溶媒和クラスターカチオンの水素結合組み換えに関する分光学的研究

池田, 貴将

https://doi.org/10.15017/1654654

出版情報：Kyushu University, 2015, 博士（理学）, 課程博士バージョン：

権利関係：Fulltext available.

(2)

Spectroscopic study

on the rearrangement of hydrogen bonds in solvated cluster cations

Ph.D. Thesis by

Takamasa Ikeda

Graduate School of Science Kyushu University

2016

(3)

Chapter 1.

General introduction

(6)

1.1. General introduction

1.1.1. Hydrogen bond and hydration

A hydrogen bond (H-bond) is one of the crucial intra- and intermolecular interactions to characterize many fundamental physical, chemical, and biological phenomena such as phase transitions, proton transfer reactions, and protein folding. In general, H-bonds are formed between H-bond donor and acceptor sites. The H-bond donor consists of H atom which is covalently bound to an electronegative atom, while electronegative atoms and groups having lone pairs and/or -electrons can act as the H-bond acceptor. The physical properties of H-bonds such as interaction strength are characterized by the type of the donor and accepter sites.^1–3 For example, water forms H-bonds, in which the lone pair of an oxygen atom of water accepts the H atom which is donated from the OH group of water. The strength of a single H-bond in water dimer is approximately 20 kJ/mol, which is typically classified into the H-bond having the modest strength. Actually, the strength of H-bonds extends from fairly weak (~10 kJ/mol) to significantly strong (~ 40 kJ/mol), which is larger than van der Waals forces but weaker than that of covalent bonds. The H-bonds having the modest strength such as O-H...O and N-H...O H-bonds are favorable to constructing three dimensional structures of large molecules, which shows both of stiffness and softness. This property is significant for various chemical and biological processes.

In the aqueous solutions of hydrophilic molecules, water molecules are bound to the hydrophilic solute molecules as well as the other water molecules. The layer of water molecules which is directly bound to the solute molecule is called a first hydration shell. The H-bonds between the solute and water molecules are entirely different properties from those of bulk waters such as the binding energy, preferential

(7)

motifs, and orientational flexibility. The water molecules in the first hydration shell are believed to play an essential role in various chemical and biological processes such as protein folding and biomolecular recognition.^4–9 For example, in protein folding, the water molecules directly bound to amino acids retain the high-order structures (the third- and/or fourth-order structures) of the protein by fixing the relative orientation of the sub-structures formed in the early process of protein folding, which obviously shows the importance of the first hydration shell in the protein folding. Accordingly, it is highly required to reveal the local hydration structures and their dynamics at the molecular level for understanding chemical and biological processes in aqueous solutions.

1.1.2. Laser spectroscopy in the gas phase

Nowadays, the sophisticated laser spectroscopy combined with a supersonic jet expansion is one of the powerful tools to investigate the physical property and the structure of molecules and molecular clusters at the molecular level. The supersonic jet expansion was introduced in the field of the gas-phase laser spectroscopy in the late 1970's. Up to the present, many experimental and theoretical breakthroughs have been achieved in this field. In this section, we focus on three progresses having a significant impact on the gas-phase laser spectroscopy. That is; (i) how to obtain the experimental signatures to identify the structures of molecules and molecular clusters, (ii) how to predict the favorable (stable) structures of molecules and molecular clusters, (iii) how to vaporize thermally fragile molecules such as biomolecules.

The significant progress for (i) was achieved by introducing the infrared (IR) spectroscopy in the gas phase. Vibrational spectroscopies such as IR and Raman

(8)

spectroscopy are the promising way of providing information on the structures of molecules and molecular clusters. However, the concentration of molecules and molecular clusters is so low in the molecular beam that it is difficult to obtain their IR and Raman spectra by using the direct absorption and scattering. Page et al. have introduced IR spectroscopy in the gas phase for the first time in the middle of 1980's.

They measured IR spectra by monitoring the ion intensity produced by resonant two-photon ionization (R2PI), which is currently called IR-dip spectroscopy.^10–12 The IR-dip spectroscopy allows us to obtain the electronic-state and isomer selective IR spectra of various molecules and molecular clusters.^13,14 The emergence of tabletop tunable and powerful IR sources such as the optical parametric oscillator (OPO) system have also driven the progress of IR spectroscopy in the gas phase.¹⁵ Furthermore, the various spectroscopic techniques using the ultraviolet (UV) and IR lasers have been developed by some research groups, which has revealed the stable conformers and the potential energy landscapes of biomolecules.^16–23

The breakthrough for (ii) was achieved by the development of quantum chemical calculations. This development has been facilitated by the remarkable evolution of hardware ability of computers and the decline in the price of the hardware.

Furthermore, in 1990's, the density-functional theory (DFT) calculations became popular in the field of molecular science, which reduces the computational costs of calculating the stable structures and theoretical IR spectra of molecules and molecular clusters as compared with the traditional ab initio molecular orbital calculations.^24–26 The calculated IR spectra have allowed us to compare the experimental and theoretical results, which is helpful for the assignments of the complicated IR spectra of flexible molecules.

(9)

The breakthrough for (iii) was achieved by the development of the novel soft ionization techniques such as the electrospray ionization (ESI) method and the laser ablation method. In the early stage of the supersonic jet technique, the molecules were vaporized by thermal heating to obtain vapor pressure which is high enough to measure molecular spectra. However, some molecules such as amino acids and nucleic acids are frequently decomposed by thermal heating, which prevents us from measuring their spectra in the gas phase. The soft ionization techniques have been developed to introduce thermolabile molecules into vacuum, which is nowadays applied to various biomolecules such as amino acids, polypeptides and their hydrated clusters.^27–31 Note that the vaporization techniques by thermal heating are also improved by applying various new materials and constructions to pulsed valves.^32–34

These improvements have been achieved synergistically. IR spectroscopy in the gas phase combined with the newly developed vaporization techniques allows us to investigate the thermolabile and/or flexible molecules. Their structures can be determined by combining the experimental IR spectra with the powerful theoretical calculations. Recently, the cold ion trap technique, which was introduced by Nolting et al. in the field of spectroscopy on biomolecules for the first time,²⁹ is one of the current trends of the gas-phase spectroscopy. The conformations of large polypeptides and flexible biomolecules have been investigated via the cryogenically cooled ion trap technique.^35–37

1.1.3. Current status of gas-phase spectroscopy

The gas-phase laser spectroscopy has currently been applied to more "realistic"

systems. As mentioned above, the larger and more flexible (bio)molecules like proteins

(10)

are the current targets of the state-of-the-art gas-phase spectroscopy. The strong and/or weak intra- and intermolecular interactions, which may determine the complicated conformations of the large molecules, can be investigated now at the molecular level. In addition, the study on the solvation effect of the large number of solvent molecules is also fascinating, though the large clusters have already been studied in the previous works.^38,39 Bridging the gap between the gas phase and the condensed phase is one of the most important issues to understand more "realistic" systems at the molecular level.^40–43

Finally, we would like to describe the studies on the rearrangement of H-bonds in the gas-phase spectroscopy. The H-bonded clusters produced by the supersonic jet expansion are cooled down to extremely low temperature, which facilitates the study on the static nature of H-bonds. However, many (bio)chemical reactions proceeds at or above room temperature, in which thermal energy causes the rearrangement and/or fluctuation of H-bonds. Actually, some research groups reported the studies on H-bonds in solvated clusters having large internal energy (high temperature).^44–51 The previous studies mainly reported the "one-way" rearrangement of H-bonds, which can be regarded as the relaxation process toward the stable H-bonded structures. Obviously, study on the fluctuation of the H-bonded structures, in which the H-bonded clusters isomerize among their stable H-bonded structures, is also important for understanding the rearrangement of H-bonds in aqueous solution at the molecular level. Furthermore, the potential barrier height, which is quite significant to characterize the H-bond dynamics, has never been determined experimentally in hydrated cluster cations. In this thesis, therefore, we particularly focus on the structural fluctuation and the potential barrier height for the rearrangement of H-bonds.

(11)

1.2. Scope of chapters

The present thesis focuses on two topics: (i) experimental observation of the structural fluctuation of H-bonds, and (ii) experimental determination of the potential barrier height for the rearrangement of H-bonds.

In chapter 2, the experimental apparatus is described in detail. Principles of the spectroscopic techniques we used in this thesis, such as resonant two-photon ionization (R2PI), photoionization efficiency (PIE), IR-dip, IR-UV hole-burning and IR photodissociation (IRPD) spectroscopy, are also explained.

In chapter 3, IR spectra of monohydrated benzyl alcohol cluster cations ([BA-(H2O)1]⁺) in the D0 state are reported. We will discuss structural fluctuation in [BA-(H2O)1]⁺ based on the IR-dip spectra measured with the intense IR pulse.

In chapter 4, the rearrangement of H-bonds in monohydrated 5-hydroxyindole cluster cation ([5HI-(H2O)1]⁺) is discussed based on the IRPD spectra in the D0 state. In addition, the potential barrier height for the rearrangement of H-bonds in [5HI-(H2O)1]⁺ is also presented.

In chapter 5, IRPD spectrum of 5HI-(tert-butyl alcohol) (t-BuOH) 1:1 cluster cation ([5HI-(t-BuOH)1]⁺) is reported. The difference in the rearrangement of H-bonds between [5HI-(H2O)1]⁺ and [5HI-(t-BuOH)1]⁺ is discussed based on their IRPD spectra and potential barrier heights for the rearrangement of H-bonds.

(12)

References for chapter 1

(1) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford University Press: Oxford, 1997.

(2) Desiraju, G. R.; Steiner, T. The Weak Hydrogen Bond: In Structural Chemistry and Biology;

Oxford University Press: Oxford, 1999.

(3) Maréchal, Y. The Hydrogen Bond and the Water Molecule; Elsevier B.V.: Amsterdam, 2007.

(4) Daune, M. Molecular biophisics: Structures in motion; Oxford University Press: Oxford, 1999.

(5) Phillips, R.; Kondev, J.; Theriot, J. Physical biology of the Cell; Garland Science: New York and Abingdon, 2009.

(6) Zuckerman, D. M. Statistical Physics of Biomoledules; CRC Press: Florida, 2010.

(7) Pal, S. K.; Zewail, A. H. Chem. Rev. 2004, 104 (4), 2099–2124.

(8) Bagchi, B. Chem. Rev. 2005, 105 (9), 3197–3219.

(9) Csermely, P.; Palotai, R.; Nussinov, R. Trends Biochem. Sci. 2010, 35 (10), 539–546.

(10) Page, R. H.; Shen, Y. R.; Lee, Y. T. Phys. Rev. Lett. 1987, 59 (12), 1293–1296.

(11) Page, R. H.; Shen, Y. R.; Lee, Y. T. J. Chem. Phys. 1988, 88 (8), 4621.

(12) Page, R. H.; Shen, Y. R.; Lee, Y. T. J. Chem. Phys. 1988, 88 (9), 5362.

(13) Zwier, T. S. Annu. Rev. Phys. Chem. 1996, 47 (1), 205–241.

(14) Ebata, T.; Fujii, A.; Mikami, N. Int. Rev. Phys. Chem. 1998, 17 (3), 331–361.

(15) Bosenberg, W. R.; Guyer, D. R. J. Opt. Soc. Am. B 1993, 10 (9), 1716.

(16) Zwier, T. S. J. Phys. Chem. A 2001, 105 (39), 8827–8839.

(17) Robertson, E. G.; Simons, J. P. Phys. Chem. Chem. Phys. 2001, 3 (1), 1–18.

(18) Mons, M.; Dimicoli, I.; Piuzzi, F.; Tardivel, B.; Elhanine, M. J. Phys. Chem. A 2002, 106 (20), 5088–5094.

(19) Dian, B. C.; Longarte, A.; Zwier, T. S. Science 2002, 296 (5577), 2369–2373.

(20) Chin, W.; Piuzzi, F.; Dimicoli, I.; Mons, M. Phys. Chem. Chem. Phys. 2006, 8 (9), 1033–1048.

(13)

(21) Shubert, V. A.; Zwier, T. S. J. Phys. Chem. A 2007, 111 (51), 13283–13286.

(22) Garand, E.; Kamrath, M. Z.; Jordan, P. A; Wolk, A. B.; Leavitt, C. M.; McCoy, A. B.; Miller, S.

J.; Johnson, M. A. Science 2012, 335 (6069), 694–698.

(23) Nagornova, N. S.; Rizzo, T. R.; Boyarkin, O. V. Science 2012, 336 (6079), 320–323.

(24) Becke, A. D. Phys. Rev. A 1988, 38 (6), 3098–3100.

(25) Becke, A. D. J. Chem. Phys. 1993, 98 (7), 5648.

(26) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37 (2), 785–789.

(27) Cable, J. R.; Tubergen, M. J.; Levy, D. H. J. Am. Chem. Soc. 1989, 111 (25), 9032–9039.

(28) Nielsen, S. B.; Lapierre, A.; Andersen, J. U.; Pedersen, U. V.; Tomita, S.; Andersen, L. H. Phys.

Rev. Lett. 2001, 87 (22), 228102.

(29) Nolting, D.; Marian, C.; Weinkauf, R. Phys. Chem. Chem. Phys. 2004, 6 (10), 2633.

(30) Mercier, S. R.; Boyarkin, O. V; Kamariotis, A.; Guglielmi, M.; Tavernelli, I.; Cascella, M.;

Rothlisberger, U.; Rizzo, T. R. J. Am. Chem. Soc. 2006, 128 (51), 16938–16943.

(31) Iavarone, A. T.; Patriksson, A.; van der Spoel, D.; Parks, J. H. J. Am. Chem. Soc. 2007, 129 (21), 6726–6735.

(32) Rizzo, T. R.; Park, Y. D.; Peteanu, L. A.; Levy, D. H. J. Chem. Phys. 1986, 84 (5), 2534.

(33) Li, L.; Lubman, D. M. Rev. Sci. Instrum. 1989, 60 (3), 499.

(34) Ebata, T. Bull. Chem. Soc. Jpn. 2009, 82 (2), 127–151.

(35) Wolk, A. B.; Garand, E.; Jones, I. M.; Hamilton, A. D.; Johnson, M. A. J. Phys. Chem. A 2013, 117 (29), 5962–5969.

(36) Zabuga, A. V.; Rizzo, T. R. J. Phys. Chem. Lett. 2015, 6 (9), 1504–1508.

(37) Aseev, O.; Perez, M. A. S.; Rothlisberger, U.; Rizzo, T. R. J. Phys. Chem. Lett. 2015, 6 (13), 2524–2529.

(38) Miyazaki, M.; Fujii, A.; Ebata, T.; Mikami, N. Science 2004, 304 (5674), 1134–1137.

(14)

(39) Mizuse, K.; Mikami, N.; Fujii, A. Angew. Chem. Int. Ed. 2010, 49 (52), 10119–10122.

(40) Chang, T. M.; Cooper, R. J.; Williams, E. R. J. Am. Chem. Soc. 2013, 135 (39), 14821–14830.

(41) Miyazaki, M.; Kawanishi, A.; Nielsen, I.; Alata, I.; Ishiuchi, S.; Dedonder, C.; Jouvet, C.; Fujii, M. J. Phys. Chem. A 2013, 117 (7), 1522–1530.

(42) Chang, T. M.; Chakrabarty, S.; Williams, E. R. J. Am. Chem. Soc. 2014, 136 (29), 10440–10449.

(43) Fournier, J. A.; Wolke, C. T.; Johnson, M. A.; Odbadrakh, T. T.; Jordan, K. D.; Kathmann, S. M.;

Xantheas, S. S. J. Phys. Chem. A 2015, 119 (36), 9425–9440.

(44) Clarkson, J. R.; Baquero, E.; Shubert, V. A.; Myshakin, E. M.; Jordan, K. D.; Zwier, T. S.

Science 2005, 307 (5714), 1443–1446.

(45) Ishiuchi, S.; Sakai, M.; Tsuchida, Y.; Takeda, A.; Kawashima, Y.; Dopfer, O.; Müller-Dethlefs, K.; Fujii, M. J. Chem. Phys. 2007, 127 (11), 114307.

(46) Sakota, K.; Harada, S.; Shimazaki, Y.; Sekiya, H. J. Phys. Chem. A 2011, 115 (5), 626–630.

(47) Tanabe, K.; Miyazaki, M.; Schmies, M.; Patzer, A.; Schütz, M.; Sekiya, H.; Sakai, M.; Dopfer, O.; Fujii, M. Angew. Chem. Int. Ed. 2012, 51 (27), 6604–6607.

(48) Wohlgemuth, M.; Miyazaki, M.; Weiler, M.; Sakai, M.; Dopfer, O.; Fujii, M.; Mitrić, R. Angew.

Chem. Int. Ed. 2014, 53 (52), 14601–14604.

(49) Zischang, J.; Suhm, M. A. J. Chem. Phys. 2014, 140 (6), 064312.

(50) Miyazaki, M.; Yoshikawa, S.; Michels, F.; Misawa, K.; Ishiuchi, S.; Sakai, M.; Dopfer, O.;

Müller-Dethlefs, K.; Fujii, M. Phys. Chem. Chem. Phys. 2015, 17 (4), 2494–2503.

(51) Shimamori, T.; Fujii, A. J. Phys. Chem. A 2015, 119 (8), 1315–1322.

(15)

Chapter 2.

Methods

(16)

2.1. Experimental apparatus

2.1.1. Vacuum chamber

Figure 2.1 shows a picture of a vacuum chamber equipped with a linear time-of-flight mass spectrometer (TOF-MS). Figure 2.2 shows a picture of the whole experimental system including a vacuum chamber and lasers. The apparatus consists of three chambers; a source, ionization, and drift chambers. A source chamber was evacuated with a 10-inch diffusion pump (ANELVA, CDP-3700A, 3700 l/s (air)) backed by an oil rotary pump (ALCATEL, T2033A, 635 l/min). A water cooling baffle was mounted on the 10-inch diffusion pump in order to prevent a backflow of the oil from the diffusion pump. An ionization camber was evacuated with a 6-inch diffusion pump (ANELVA, CDP-1200, 1200 l/s) backed by an oil rotary pumps (ALCATEL, M2015SD, 250 l/min). A drift camber was also evacuated with a 6-inch diffusion pump (ANELVA, CDP-1200, 1200 l/s) backed by an oil rotary pumps (ALCATEL, M2015SD, 250 l/min). Both a liquid-nitrogen trap and a water cooling baffle were mounted on the 6-inch diffusion pumps for the ionization and drift chamber in order to achieve a high-vacuum condition which is appropriate for mass spectrometry.

The pressure of the vacuum chamber was monitored by an ionization gauge tube (ANELVA, UGD-1S) with a controller (Ionization chamber: ANELVA, MIG-061, Drift chamber: ANELVA, M-722HG). The background pressure of the drift chamber was maintained below ca. 2.5 × 10^-5 and ca. 5.0 × 10^-5 Pa when a pulse nozzle is operated with He and Ne as a carrier gas, respectively. In the beginning of the evacuation from an atmospheric pressure, the pressure of the source and ionization chambers was monitored by two Pirani gauge tubes (ULVAC, WP-02) with a controller (ULVAC, GP-2A).

(17)

2.1.2. Molecular beam source

A stainless steel tube was adopted to the nozzle housing of samples. The nozzle housing was rolled by a coiled heater. The applied voltage to a coiled heater was controlled to obtain adequate signal intensity. The temperature of the nozzle housing was monitored by a thermoelectric couple with a controller (Watlow, 935A). He or Ne was used for a carrier gas, whose backing pressure was controlled by a regulating valve.

The carrier gas was passed through a reservoir which contained solvent molecules. The temperature of the reservoir was controlled by a thermostat bath (AS ONE, LTB-250) from 268 K to ca. 320 K. The carrier gas containing the solvent molecules was mixed with the sample gas in the nozzle housing.

For expanding the mixture of the sample and carrier gas into the source chamber, we used the commercial pulsed valve of a single solenoid type (Parker Hannifin, Pulse Valves Series 9, 0.8 mm as an orifice diameter with a cone-type orifice shape) controlled by a commercial pulsed valve driver (Parker Hannifin, IOTA ONE).

The maximum operating temperature of the commercial pulsed valve is ca. 400 K. To heat the sample above 400 K, we used a homemade heat-resistant pulsed valve whose solenoid consists of a polytetrafluoroethylene (PTFE) coated wire (Junkosha).

The gas expanded with the pulsed valve was skimmed into the ionization chamber. A commercial conical skimmer (Beam dynamics, Model 1, the diameter of the orifice is 1.5 mm) was used for generating the molecular beam. The distance between the pulse valve and the skimmer was adjusted to be ca. 20 mm in order to obtain adequate ion signal intensity. The skimmed molecular beam was ionized in the ionization chamber by the resonant two-photon ionization (R2PI) (see chapter 2.2.1)

(18)

2.1.3. Time-of-flight mass spectrometer

The cluster cations generated by R2PI were mass-selected by using a Wiley-McLaren type linear time-of-flight mass spectrometer (TOF-MS).¹ Figure 2.3 displays the schematic diagram of TOF-MS. The ion signal was detected by micro-channel plates (MCP) (BURLE 18mm Detection) with a Z-Gap detector assembly (R. M. Jordan). The voltages applied to the MCP detector, acceleration electrodes, deflectors and a mass gate were controlled by a TOF power supply (R. M.

Jordan). The voltage applied to the MCP detector was increased up to − 2.80 kV.

In the ion-acceleration region, high positive voltages were applied to the first and second extraction grids (+ 2.5 kV and + 2.2 kV, respectively), whereas the final grid was held at ground potential. The distance among each grid is 15 mm. In order to guide the ions in the MCP detector efficiently, we introduced two plate-type ion deflectors which adjust the ion trajectories in the perpendicular and horizontal directions to an expansion axis. The applied voltages to the deflectors were typically between − 50 and + 50 V. In front of the MCP detector, a plate-type mass gate was introduced so as not to detect unnecessary species. The pulsed voltage applied to the mass gate was fixed at 200 V, whose trigger pulse was synchronized with the time sequence of the whole experimental system.

The ion signal detected with MCP was amplified by a preamplifier (Stanford research systems, SR445A), then time-averaged by two digital oscilloscopes (LeCroy, WaveRunner 64xi and 9350A). The averaged signals were fed into a personal computer (Dell, OptiPlex 3020) to record the signal intensity of each mass channel separately.

(19)

2.1.4. Laser system

Figure 2.4 illustrates a block diagram of the laser system and detection devices for the measurement of all spectra in the present study. A single UV laser system was used for one-color R2PI (1C-R2PI) experiment, while two different UV laser systems were used for two-color R2PI (2C-R2PI) experiment. For 1C-R2PI, a frequency-doubled dye laser (Sirah Cobra Stretch and Inrad Autotracker III) pumped by the second harmonic of a Nd³⁺:YAG laser (Spectra Physics INDI-40-20, 20 Hz, 50 mJ/

pulse) was used as a UV source. For 2C-R2PI, the UV laser system used for 1C-R2PI was operated as an excitation UV source. A frequency-doubled dye laser (Lumonics HD-300 and BBO crystal) pumped by the second harmonic of a Nd³⁺:YAG laser (Spectra Physics LAB-130, 10 Hz, 100 mJ/ pulse) was used for the ionization UV source. The two UV lasers were combined collinearly with a half mirror and simultaneously focused on the molecular beam with a plano-convex quartz lens (300 mm focal length).

For the measurement of IR spectra, an optical parametric oscillator (LaserVision) pumped by an injection-seeded Nd³⁺:YAG laser (Continuum Powerlite Precision II 8000, 580 mJ/pulse) was used as an IR source. The IR laser was focused on the molecular beam with a plano-convex CaF2 lens (300 mm focal length). The IR laser system was triggered at 10 Hz and 5 Hz for 1C- and 2C-R2PI, respectively. The IR intensity was reduced to be ~ 1 mJ/pulse in the frequency region of 3300 ~ 3800 cm^-1 in order to avoid unfavorable saturation of vibrational transitions. The ion signals with and without the IR pulse were stored separately to correct the artificial fluctuation of the spectral baseline.

All laser systems were synchronized with the trigger pulses for the pulsed

(20)

valve, the oscilloscopes and the mass gate by digital delay and pulse generators (Stanford Research Systems, DG-535).

(21)

2.2. Spectroscopy

2.2.1. Resonant two-photon ionization (R2PI) spectroscopy

R2PI^2–4 was used for measuring the vibronic transitions between the S1 and S0

states. When the frequency of the UV photon is resonant with a specific vibronic state in the S1 state, a molecule (or a cluster) is excited to the vibronic state (S1-S0 transition).

The excited molecule absorbs another UV photon within the lifetime of the excited vibronic state. This sequential absorption allows the excited molecule to be ionized (D0-S1 transition). In 1C-R2PI, two photons coming from the same laser are absorbed while in 2C-R2PI, the frequency of the ionization laser (2) is different from that of the excitation laser (1). Figure 2.5 shows the schematic diagram of 2C-R2PI. In general, two UV lights were focused on the molecular beam at the same position simultaneously.

The 1 is scanned in the frequency region where the S1-S0 transition occurs while the 2

is fixed at the frequency which is large enough to ionize the molecules. If the 1 is resonant with the S1-S0 transition, the excited molecule is ionized by absorbing the 2

photon, then the ion signal is detected on MCP. Therefore, R2PI spectra contain information on the vibronic levels of the molecule (or cluster) in the S1 state. Note that mass spectrometry can be applied to the molecular (cluster) cations, which allows us to obtain mass-selected vibronic spectra.

2.2.2. Photoionization efficiency (PIE) spectroscopy

PIE spectroscopy^5–9 was used to determine the adiabatic ionization energy (IE0) of a molecule or a cluster. Figure 2.6 shows a schematic diagram of the PIE spectroscopy. The 1 is fixed at the S1-S0 origin transition of each isomer observed in R2PI spectra, while the 2 is scanned. When the total photon energy (h1 + h2) exceeds

(22)

IE0, the excited molecule in the S1 state is ionized to the ionization continuum in which the vibrational states in the D0 state are buried. Accordingly, we obtain PIE spectra by monitoring the ion signal intensity as a function of the total photon energy. The Franck-Condon factors between the S1 and D0 states determine the vibrational states to which the excited molecules are accessible via photoionization. The step-like increase in the ion signal intensity is observed when the Franck-Condon region covers the zero-point energy level in the D0 state. The ionization to the vibrational excited states in the D0 state is observed as the increase in the ion signal intensity above IE0. On the other hand, if the Franck-Condon region is far from IE0, the ion signal intensity does not show the step-like increase but gradual increase at the beginning of the Franck-Condon region.^8,9 We note that the disappearance of the step-like increase in the ion signal is also observed when the internal rotation of a methyl group consists of dense low-frequency vibrational states as is observed in the PIE spectrum of indole-(MeOH)1.⁷

2.2.3. IR-dip spectroscopy

IR spectroscopy is one of the powerful tools for determining the conformations and structures of molecular clusters. In the gas phase spectroscopy, however, it is difficult to detect the direct IR absorption of the molecular clusters because the density of molecular clusters per unit volume is typically below the detection limit. Accordingly, we applied the population labeling spectroscopy in order to obtain IR spectra of the molecular clusters, which is called IR-dip spectroscopy.^10–13 Figure 2.7 displays a schematic diagram of IR-dip spectroscopy in the S0 state. The 1 is fixed at the vibronic band of a single isomer observed in R2PI spectra. The 2 is fixed at the same frequency

(23)

as when the R2PI spectra is measured. The two UV lasers are fired simultaneously while the irradiation of the IR laser precedes the UV lasers by 20 ns. When the IR photon is resonant with the vibrational transition of the monitored isomer in the S0 state, the populations of the isomer in the S0 state decrease, leading to the depletion of the ion signal. Therefore, IR-dip spectrum is obtained when the ion signal intensity is monitored as a function of the frequency of the IR laser (IR). In the present study, the

IR is scanned in the frequency region of ca. 3000 ~ 3800 cm^-1. The ion intensity with and without the IR pulse was stored separately to correct the artificial fluctuation of the spectral baseline.

IR-dip spectroscopy can be also available for obtaining the IR spectrum of molecular cluster cations in the D0 state.^14,15 When the IR laser is irradiated after R2PI, the IR-dip spectrum of the cluster cations is obtained as the depletion of the ion signal intensity due to photodissociation. It should be noted that the internal energy of the cluster cations after the absorption of an IR photon must be larger than their binding energy to bring about photodissociation. The IR spectrum in the D0 state is also obtained by monitoring the fragment ion signals, which is called the IR photodissociation (IRPD) spectroscopy (see chapter 2.2.4).

2.2.4. IR-UV hole-burning spectroscopy

R2PI spectrum typically contains the vibronic bands of multiple isomers.

IR-UV hole-burning spectroscopy allow us to separate the vibronic bands into those of each isomer.¹² Figure 2.8 shows the schematic diagram of IR-UV hole-burning spectroscopy. The basic strategy of IR-UV hole-burning spectroscopy is similar to that of IR-dip spectroscopy. The IR laser is fixed at the vibrational transition of a specific isomer, while the 1 is scanned. Absorbing the IR laser reduces the population of the

(24)

isomer in the S0 state. Accordingly, the vibronic bands of the isomer disappear in IR-UV hole-burning spectrum, whereas the vibronic bands of the other isomers still remain.

The remaining vibronic bands are also classified into those of the other isomers by changing IR.

When the IR overlaps with the vibrational transitions of two or more isomers, the vibronic transitions of each isomer cannot be distinguished by IR-UV hole-burning spectroscopy. Instead of the vibrational transition, the electronic transition (in the UV region) can also be available for reducing the population of the isomers in the S0 state.

This method is called UV-UV hole-burning spectroscopy. In the present study, however, we did not apply it.

2.2.5. IR photodissociation (IRPD) spectroscopy

IR spectra of molecular clusters in the D0 state can be obtained by applying not only IR-dip spectroscopy but IRPD spectroscopy.^16,17 Figure 2.9 shows a schematic diagram of IRPD spectroscopy. The 1 and 2 are fixed at the vibronic band of a single isomer observed in R2PI spectra. The IR laser, the frequency of which is scanned in the region of 3300 ~ 3800 cm^-1, is fired after the irradiation of the UV lasers by 700 ns.

When the IR is resonant with a vibrational transition of the monitored isomer in the D0

state, the fragment ion (the daughter ion) is observed due to the photodissociation of the cluster cations. Accordingly, IRPD spectra are obtained as the increase in the intensity of the fragment ion signal without the fluctuation of the baseline (so-called zero-background spectroscopy). The portion of the molecular beam to which the UV lasers are irradiated flows to the downstream during the delay time of 700 ns. Thus, the alignment of the UV and IR lasers was manually adjusted to obtain the maximum

(25)

intensity of the fragment ion signals.

IRPD spectrum of monohydrated clusters is obtained as the increase in the monomer ion signal. In some cases, however, the unfavorable monomer ion signals originating from UV photodissociation (UVPD) process and/or direct photoionization of the monomer also appear in the mass channel of the monomer ion. Figure 2.10 displays a schematic diagram of the method used for eliminating unfavorable monomer ion signals. In this protocol, the delay time between the UV and IR lasers is fixed at 700 ns.

After irradiating the UV lasers, the cluster cations and the unfavorable monomer cations (UVPD and/or direct photoionization) are generated in the acceleration region. The cluster cations are accelerated more slowly than the monomer cations due to the difference in their own mass. After 700 ns, the monomer cations precede the cluster cations in the acceleration region. Therefore, the unfavorable monomer ion signal appears in a TOF mass spectrum prior to the signal of the monomer cation via IRPD, which allows us to separate the IRPD signal from the unfavorable monomer ion signal.

(26)

Figure 2.1. Schematic diagram of vacuum chambers.

Acceleration Electrode MCP Water

Heated Sample

Mass Gate Liquid-nitrogen TrapDiffusion Pump 3.7 kl/s RotaryPump 635 l/s

Diffusion Pump 1.2 kl/s Rotary Pump 250 l/min

Ne or He

Deflector

Source chamberIonization chamberDrift chamber

(27)

Excit ation -Laser Ioniz at ion -La ser Va cu u m C h am b er

IR Laser

Figure 2.2. Picture of the whole experimental system including vacuum chambers and lasers.

(28)

Figure 2.3. Schematic diagram of Wiley-McLaren type linear time-of-flight mass spectrometer (TOF-MS) with detection devices.

15 m m

2.2 2.5 0

Electric fi eld

(kV )

200 V ~5 0 V A cc eler atio n De fl ec tor Mass g at e MC P

Skimmed molecular beam

Ion iz at io n ch am b er Dr if t ch am b er

Pre amp

Oscilloscope

PC

(29)

Figure 2.4.Brockdiagram of the laser systems and detection devices.

Oscilloscope (IR off) PC

Nd3+:YAG Laser Frequency Doubled Dye-Laser(Sirah)

5 or 10 Hz Optical parametric converter (Laservision) Nd3+:YAG Laser (SHG)Frequency Doubled Dye-Laser(HD-300)

Nd3+:YAG Laser (SHG)20 Hz 10 Hz

10 or 20 Hz Nozzle Driver TOF controller20 Hz HV pulser MCP Pre amp

5 or 10 Hz5 or 10 HzPlus Generator Oscilloscope (IR on)Mass gateDeflector

Accelerate electrodes Sample nozzleSkimmer

(30)

Figure 2.5. Schematic diagram of two-color resonant two photon ionization (2C- R2PI) spectroscopy. The ₁ is scanned in the region of the S₁-S₀transition while the

₂is fixed at the frequency which is large enough to cause the D₀-S₁transition. The 2C-R2PI spectrum obtained as the increase in the ion signal intensity includes information on the vibrational states in the S₁state.

𝜈

₁

(Scan)

𝜈

₂

(Fixed)

𝑆

₀

𝑆

₁

𝐷

₀

Ion intensity

W av enu mber (cm

-1

)

(31)

Figure 2.6. Schematic diagram of photoionization efficiency (PIE) spectroscopy.

The₁was fixed at the S₁-S₀ origin transition of a specific isomer observed in R2PI spectra, while the ₂ was scanned. The adiabatic ionization energy (IE₀) is obtained as the step-like increase in the ion signal intensity.

𝜈

₁

(Fixed)

𝜈

₂

(Scan)

𝑆

₀

𝑆

₁

𝐼𝐸

₀

Ion intensity

W av enum ber (cm

-1

)

𝐷

₀

(32)

Figure 2.7. Schematic diagram of IR-dip spectroscopy in the S₀state. The₁and₂ are fixed, while the_IRis scanned in the range of ca. 3000 ~ 3800 cm^-1. When the_IR is resonant with the vibrational transition in the S₀state, the ion intensity is reduced.

The IR-dip spectrum is obtained as the depletion of the ion intensity.

Ion intensity

W av enumber (cm

-1

)

20 ns 𝜈

_IR

(Scan)

𝑆

₀

𝑆

₁

𝐷

₀

𝜈

₁

(Fixed)

𝜈

₂

(Fixed)

(33)

Figure 2.8.Schematic diagram of IR-UV hole-burning spectroscopy. The_IRis fixed at the vibrational transition of a specific isomer, while the ₁ is scanned. The vibronic bands of the isomer do not appear in the IR-UV hole-burning spectrum, whereas the vibronic bands of the other isomers still remain.

Ion intensity

W av enumber (cm

-1

)

𝜈

₁

(Scan)

𝜈

₂

(Fixed) ^R

2 P I

IR -UV h ol eb u rnin g

𝑆

₀

𝑆

₁

𝐷

₀

Isomer A 𝜈

_IR

(Fixed)

Isomer B 𝑆

₀

𝑆

₁

𝐷

₀

(34)

Figure 2.9. Schematic diagram of IR photodissociation (IRPD) spectroscopy. The

_IRin the region of 3300-3800 cm^-1 is fired after the irradiation of the UV lasers by 700 ns. The fragment ions derived from IRPD are selectively detected as a function of the IR frequency.

700 ns

Fragment Ion

W av enumber (cm

-1

)

𝜈

₁

(Fixed)

𝜈

₂

(Fixed)

𝑆

₀

𝑆

₁

𝐷

₀

𝜈

_IR

(Scan)

Dissociation

(35)

Figure 2.10. Schematic diagram of the method used for eliminating unfavorable monomer ion signals derived from UVPD or photoionization of the monomer. The delay between UV and IR lasers is fixed at 700 ns. The cluster cations are accelerated more slowly than the monomer cations. IRPD signal is separated from the signal originating from UVPD and/or photoionization of monomer.

UV

IR t = 0 ns

t = 700 ns

2.2 kV 2.5 kV

Ion i n tensit y

(m/s) Monomer

&

UVPD signal

IRPD signal

Parent ion Daughter ion

UVPD

IRPD

Monomer

Cluster

Monomer

(36)

References for chapter 2.

(1) Wiley, W. C.; McLaren, I. H. Rev. Sci. Instrum. 1955, 26 (12), 1150.

(2) Johnson, P. M. J. Chem. Phys. 1975, 62 (11), 4562–4563.

(3) Murakami, J.; Kaya, K.; Ito, M. J. Chem. Phys. 1980, 72 (5), 3263.

(4) Ichimura, T.; Shinohara, H.; Nishi, N. Chem. Phys. Lett. 1988, 146 (1), 83–88.

(5) Duncan, M. A.; Dietz, T. G.; Smalley, R. E. J. Chem. Phys. 1981, 75 (5), 2118.

(6) Smith, M. A.; Hager, J. W.; Wallace, S. C. J. Chem. Phys. 1984, 80 (7), 3097.

(7) Hager, J.; Ivanco, M.; Smith, M. A.; Wallace, S. C. Chem. Phys. 1986, 105 (3), 397–416.

(8) Gu, Q.; Knee, J. L. J. Chem. Phys. 2012, 137 (10), 104312.

(9) Nakamura, T.; Schmies, M.; Patzer, A.; Miyazaki, M.; Ishiuchi, S.; Weiler, M.; Dopfer, O.; Fujii, M. Chem. - A Eur. J. 2014, 20 (7), 2031–2039.

(10) Page, R. H.; Shen, Y. R.; Lee, Y. T. Phys. Rev. Lett. 1987, 59 (12), 1293–1296.

(11) Pribble, R. N.; Zwier, T. S. Science 1994, 265 (5168), 75–79.

(12) Robertson, E. G.; Simons, J. P. Phys. Chem. Chem. Phys. 2001, 3 (1), 1–18.

(13) Zwier, T. S. Annu. Rev. Phys. Chem. 1996, 47 (1), 205–241.

(14) Kleinermanns, K.; Janzen, C.; Spangenberg, D.; Gerhards, M. J. Phys. Chem. A 1999, 103 (27), 5232–5239.

(15) Miyazaki, M.; Fujii, A.; Ebata, T.; Mikami, N. Phys. Chem. Chem. Phys. 2003, 5 (6), 1137–1148.

(16) Ebata, T.; Fujii, A.; Mikami, N. Int. J. Mass Spectrom. Ion Process. 1996, 159 (1-3), 111–124.

(17) Sawamura, T.; Fujii, A.; Sato, S.; Ebata, T.; Mikami, N. J. Phys. Chem. 1996, 100 (20), 8131–8138.

(37)

Chapter 3.

Structural fluctuation

of hydrated benzyl alcohol cluster cations

(38)

3.1. Introduction

The hydrogen bond (H-bond) is one of the most important interactions in aqueous solutions. Many chemical and biological reactions in aqueous solutions proceed around room temperature. In general, the H-bonds in aqueous solutions frequently break and re-form repeatedly. This phenomenon is regarded as a rearrangement and/or fluctuation of the H-bonds. In particular, the dynamics of water molecules in the first solvation shell may be significantly different from those of bulk water molecules. Hydrated structures and their dynamics in the first solvation shell have attracted attention for the importance of understanding chemical and biological processes such as protein folding and molecular recognition.^1–7 Since there are many bulk water molecules in aqueous solutions, it is difficult to distinguish the behavior of water molecules in the first solvation shell on the chemical and biological processes from that in bulk. Furthermore, experiments in aqueous solutions are hardly able to focus on a particular water molecule. Although theoretical approaches^6–8 such as molecular dynamics (MD) simulation allow us to discuss the H-bonding dynamics at the atomic resolution, experimental approaches at the molecular level are also required for comprehending chemical and biological processes in aqueous solutions in detail.

A supersonic jet expansion combined with various spectroscopic techniques is one of the most powerful tools for the investigation of hydration structures, because the jet-cooled hydrated molecular clusters, which consist of limited number of water molecules, are free from the disturbance of the bulk water. Furthermore, the supersonic jet expansion typically simplifies the spectral pattern of the hydrated molecular clusters due to the elimination of thermal energy. This is a great advantage to determine the structure of the hydrated clusters precisely. For example, many hydrated structures of

(39)

biomolecules such as amino acids and nucleobases have been identified in the gas phase by applying the combination of IR spectroscopy in the 3 m region and theoretical calculations.^9–11

In general, the jet-cooled hydrated clusters are located at local potential energy minima. However, the hydrated clusters storing sufficient internal energy to overcome the potential energy barriers may interconvert among various potential minima, which is equivalent to the rearrangement and fluctuation of the H-bonds. Recently, resonant two-photon ionization (R2PI) has been applied to hydrated clusters to investigate the dynamics of the H-bonds in the D0 state.^12–22 In some cases, cluster cations produced by R2PI store a large amount of internal energy due to their large binding energy and the large structural displacement between the S1 and D0 states. For instance, the monohydrated trans-acetanilide cluster cation ([AA-(H2O)1]⁺), in which a water molecule is bound to the CO group, exhibits a migration of a water molecule from the CO group to the NH group after R2PI via the S1-S0 origin.^14,19,20 Similarly, in the monohydrated tryptamine cluster cation ([TRA-(H2O)1]⁺), a water molecule migrates from the amino group to the NH group of the indole ring after photoionization.^17,18 These cluster cations, however, show no migration in the opposite direction (i.e., a water molecule does not migrate from the NH group to the CO group in [AA-(H2O)1]⁺ and from the NH group to the amino group in [TRA-(H2O)1]⁺), indicating that these cluster cations show the “one-way” rearrangement of the H-bonds.

On the other hand, the fluctuation of hydration structures has been observed experimentally in the monohydrated 2-phenylethanol cluster cation ([PEAL-(H2O)1]⁺), in which the “consecutive” rearrangements of H-bonds occur.²² In this case, all of [PEAL-(H2O)1]⁺ fluctuate among the hydration structure of [PEAL(OH)-(H2O)1]⁺ and

(40)

[PEAL(Free)-(H2O)1]⁺, where a water molecule is bound to the OH group of [PEAL]⁺ or not. The internal energy of [PEAL-(H2O)1]⁺ produced by R2PI completely exceeds the potential energy barriers among of the structural isomers, since R2PI induces a large structural displacement including the configurational change of the PEAL side chain.

On the other hand, the monohydrated phenol cluster cation ([Phenol-(H2O)1]⁺) having no side chain does not show the rearrangement of a water molecule even after R2PI.^23,24 Accordingly, the flexibility and/or the length of the side chain may affect the rearrangement of H-bonds.

Compared with PEAL, Benzyl alcohol (BA) has a shorter side chain (hydroxymethyl group). Actually, the side chain of BA is a middle length between those of phenol and PEAL. Owing to the different length of the side chain in BA, the rearrangement of H-bonds in the monohydrated BA cluster cation ([BA-(H2O)1]⁺) may exhibit different features from those of [Phenol-(H2O)1]⁺ and [PEAL-(H2O)1]⁺. In this chapter, we report the H-bonded structure of [BA-(H2O)1]⁺ in the D0 state by using IR spectroscopy. IR-dip spectroscopy using high power IR laser successfully shows the structural fluctuation in [BA-(H2O)1]⁺.

(41)

3.2. Experimental and computational methods

The experimental setup used in this study has been described in chapter 2. A commercially available BA was purchased from Wako Pure Chem. Ind. and used without further purification. BA was introduced in a stainless steel tube without heating.

The vaporized BA molecules were mixed with neon carrier gas which passed through a reservoir containing water cooled down to 278 K. A typical stagnation pressure was 2 atm. The mixture gas was expanded into a vacuum chamber by using a pulsed valve (General Valve, series 9, 0.8 mm as an orifice diameter) operated at 20 Hz. The supersonic expansion was skimmed into an ion-source chamber. The skimmed hydrated clusters were photoionized by a UV laser for mass selection. The produced ions were analyzed with a linear time-of-flight mass spectrometer for all measurements.

For the measurements of 1C-R2PI spectra, a frequency-doubled dye laser (Sirah Cobra Stretch and Inrad Autotracker III) pumped by the second harmonic of an Nd³⁺:YAG laser (Spectra Physics INDI-40-20, 20 Hz, 50 mJ/ pulse) was used as the UV source. The UV light was focused on the molecular beam with a plano-convex lens (300 mm focal length). The UV laser was scanned in the frequency region of 37400-38000 cm^-1. For the measurements of IR spectra, an optical parametric oscillator (LaserVision) pumped by an injection-seeded Nd³⁺:YAG laser (Continuum Powerlite Precision II 8000, 5 Hz, 580 mJ/pulse) was used as an IR source. The IR energy used for the measurements of IR-dip spectra was reduced to be ~ 2 mJ/pulse at 3300 ~ 3800 cm^-1 in order to avoid unfavorable saturation of vibrational transitions, whereas the IR energy was increased up to ~ 10 mJ for measuring high power IR-dip spectra. The repetition rate of the UV laser was 20 Hz, whereas that of the IR laser was 10 Hz. The IR and UV lasers were spatially overlapped with one another. The ion signals with and without the

(42)

IR pulse were stored separately to correct the artificial fluctuation of the spectral baseline. For the measurement of IR spectra in the S0 state, the IR laser in the frequency region of 3300 ~ 3800 cm^-1 preceded the UV laser by ~ 20 ns, while the UV laser was followed by the IR laser with the delay time of ~ 20 ns for measuring IR-dip spectra in the D0 state (see chapter 2.2).

M06-2X²⁵/aug-cc-pVTZ calculations were performed to obtain the stable structures, stabilization energies, harmonic vibrational frequencies and IR intensities.

The calculated harmonic vibrational frequencies in the S0 and D0 states were scaled by 0.9459. The basis set superposition error was corrected by a counterpoise method. All quantum chemical calculations were performed by GAUSSIAN 09 program package.²⁶ The computations were carried out using the computer facilities at Research Institute for Information Technology, Kyushu University.

(43)

3.3. Results and discussions

Figure 3.1a shows the 1C-R2PI spectrum of the jet-cooled BA monomer. In the previous study, the vibronic band at 37528 cm^-1 was assigned to the origin transition of BA.^27–33 Furthermore, three vibronic bands at 37579 (32528 + 51), 37625 (32528 + 97) and 37650 (32528 + 122) cm^-1 in Figure 3.1a were also assigned to the vibronic transitions of a single conformer observed at 37582 cm^-1 on the basis of isotope shift.²⁷ Although the configuration of the side chain in BA has been debated, the origin band at 37528 cm^-1 was assigned to that of the gauche-cis conformer.^29–33 The 1C-R2PI spectra of the jet-cooled hydrated BA clusters obtained by monitoring at [BA-(H2O)1]⁺ and [BA-(H2O)2]⁺ mass channels are shown in Figure 3.1b, c, respectively. A prominent band at 37583 cm^-1 was previously assigned to the origin transition of BA-(H2O)1 where a water molecule forms a bridge between the OH group and the -ring of the BA moiety.^28,29,32 In addition, the band at 37605 (37583 + 22) cm^-1 was also assigned to the vibronic transition of BA-(H2O)1 observed at 37583 cm^-1.^28,29,32 On the other hand, the weak transition at 37529 cm^-1 in Figure 3.1b was previously assigned to the origin transition of BA-(H2O)2.^28,29 In the R2PI spectrum obtained by monitoring the [BA-(H2O)2]⁺ mass channel, however, the origin band of BA-(H2O)2 does not appear (Figure 3.1c). This result is consistent with the previous study.²⁸ The disappearance of the band at 37529 cm^-1 in Figure 3.1c indicates that a single water molecule completely dissociates from BA-(H2O)2 after R2PI via the origin transition of BA-(H2O)2. The vibronic band at 37684 cm^-1 in Figure 3.1c was also assigned to the origin transition of BA-(H2O)4 in the previous study.²⁸

In order to confirm the assignment of each origin transition in Figure 3.1b, we performed IR-dip spectroscopy for BA-(H2O)1 and BA-(H2O)2. Figure 3.2a, b show the

(44)

IR-dip spectra obtained by probing the origin band at 37583 and 37529 cm^-1, respectively. The stick spectra in Figure 3.2a, b are the theoretical IR spectra of BA-(H2O)1 and BA-(H2O)2, respectively. The calculated stable structures of BA-(H2O)1

and BA-(H2O)2 are also shown in Figure 3.3a, b. In Figure 3.2a, three vibrational bands are observed at 3727 (free OH stretch), 3616 (-bonded OH stretch) and 3562 cm^-1 (-bonded OH stretch), which are in good agreement with the previous assignments based on the IR spectroscopy.^29,32 We have confirmed that the IR-dip spectrum obtained by probing at 37605 cm^-1 is completely identical to the IR-dip spectra shown in Figure 3.2a. The IR-dip spectrum in Figure 3.2b shows four prominent bands. They can be assigned to a free OH stretch (3724 cm^-1), a -bonded OH stretch (3597 cm^-1) and two H-bonded OH stretches (3504 and 3464 cm^-1) based on the theoretical prediction. These assignments are consistent with the previous result obtained by measuring the fluorescence detected IR (FDIR) spectroscopy.²⁹

The experimental protocol of the IR-dip spectroscopy we used does not provide the IR spectrum of the bare [BA]⁺. Therefore, we performed an Ar-tagging technique.

Figure 3.4a displays the IR-dip spectrum of [BA-(Ar)1]⁺ in the D0 state, which was measured to obtain information on the free OH stretching vibration of the bare [BA]⁺. [BA-(Ar)1]⁺ was produced by the fragmentation of Ar atoms after R2PI of the neutral BA-(Ar)n. In general, the vibrational frequency of the OH group to which an Ar atom is bound is red-shifted from that of [BA]⁺. For example, the H-bonded OH stretching vibration of [Phenol-(Ar)2]⁺, where an Ar atom is bound to the OH group in the [Phenol]⁺ moiety, is red-shifted by ~200 cm^-1 from the free OH stretching vibration of the bare phenol in the S0 state (3657 cm^-1).^34–37 In Figure 3.4a, the observed vibrational band at 3666 cm^-1 is slightly blue-shifted from that of the bare BA in the S0 state,

溶媒和クラスターカチオンの水素結合組み換えに関 する分光学的研究